Cancer Neurology in Clinical Practice 2nd Edition - Neurologic Complications of Cancer and Its Treatment (Current Clinical Oncology)

CANCER NEUROLOGY

IN

CLINICAL PRACTICE

SECOND EDITION

CURRENT CLINICAL ONCOLOGY Maurie Markman,

MD,

SERIES EDITOR

Cancer Neurology in Clinical Practice: Neurologic Complications of Cancer and Its Treatment, edited by DAVID SCHIFF, SANTOSH KESARI AND PATRICK Y. WEN, 2008 Integrative Oncology: Incorporating Complementary Medicine into Conventional Cancer Care, edited by LORENZO COHEN AND MAURIE MARKMAN, 2008 Prostate Cancer: Signaling Networks, Genetics, and New Treatment Strategies, edited by RICHARD G. PESTELL AND MARJA T. NEVALAINEN, 2008 Intraperitoneal Cancer Therapy, edited by WILLIAM C. HELM AND ROBERT EDWARDS, 2007 Molecular Pathology of Gynecologic Cancer, edited by ANTONIO GIORDANO, ALESSANDRO BOVICELLI, AND ROBERT KURMAN, 2007 Colorectal Cancer: Evidence-Based Chemotherapy Strategies, edited by LEONARD B. SALTZ, 2007 High-Grade Gliomas: Diagnosis and Treatment, edited by GENE H. BARNETT, 2006 Cancer in the Spine: Comprehensive Care, edited by ROBERT F. MCLAIN, KAI-UWE LEWANDROWSKI, MAURIE MARKMAN, RONALD M. BUKOWSKI, ROGER MACKLIS, AND EDWARD C. BENZEL, 2006 Squamous Cell Head and Neck Cancer, edited by DAVID J. ADELSTEIN, 2005 Hepatocellular Cancer: Diagnosis and Treatment, edited by BRIAN I. CARR, 2005 Biology and Management of Multiple Myeloma, edited by JAMES R. BERENSON, 2004 Cancer Immunotherapy at the Crossroads: How Tumors Evade Immunity and What Can Be Done, edited by JAMES H. FINKE AND RONALD M. BUKOWSKI, 2004 Treatment of Acute Leukemias: New Directions for Clinical Research, edited by CHING-HON PUI, 2003 Allogeneic Stem Cell Transplantation: Clinical Research and Practice, edited by MARY J. LAUGHLIN AND HILLARD M. LAZARUS, 2003 Chronic Leukemias and Lymphomas: Biology, Pathophysiology, and Clinical Management, edited by GARY J. SCHILLER, 2003 Colorectal Cancer: Multimodality Management, edited by LEONARD SALTZ, 2002 Breast Cancer: A Guide to Detection and Multidisciplinary Therapy, edited by MICHAEL H. TOROSIAN, 2002 Melanoma: Biologically Targeted Therapeutics, edited by ERNEST C. BORDEN, 2002 Cancer of the Lung: From Molecular Biology to Treatment Guidelines, edited by ALAN B. WEITBERG, 2001 Renal Cell Carcinoma: Molecular Biology, Immunology, and Clinical Management, edited by RONALD M. BUKOWSKI AND ANDREW NOVICK, 2000 Current Controversies in Bone Marrow Transplantation, edited by BRIAN J. BOLWELL, 2000 Regional Chemotherapy: Clinical Research and Practice, edited by MAURIE MARKMAN, 2000 Intraoperative Irradiation: Techniques and Results, edited by L. L. GUNDERSON, C. G. WILLETT, L. B. HARRISON, AND F. A. CALVO, 1999

CANCER NEUROLOGY IN CLINICAL PRACTICE Neurologic Complications of Cancer and Its Treatment Second Edition

Edited by

DAVID SCHIFF, MD School of Medicine, University of Virginia, Charlottesville, VA

SANTOSH KESARI, MD, PhD Dana-Farber/Brigham and Women’s Cancer Center, Boston, MA

PATRICK Y. WEN, MD Dana-Farber/Brigham and Women’s Cancer Center, Boston, MA

Editors David Schiff University of Virginia School of Medicine Charlottesville, VA

Santosh Kesari Center for Neuro-Oncology Dana-Farber/Brigham and Women’s Cancer Center and Department of Neurology Brigham and Women’s Hospital Boston, MA

Patrick Y. Wen Center for Neuro-Oncology Dana-Farber/Brigham and Women’s Cancer Center and Department of Neurology Brigham and Women’s Hospital Boston, MA

Series Editor Maurie Markman MD Anderson Cancer Center University of Texas Houston, TX

ISBN 978-1-58829-983-3

e-ISBN 978-1-59745-412-4

Library of Congress Control Number: 2007942812 ©2008 Humana Press, a part of Springer Science+Business Media, LLC All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, 999 Riverview Drive, Suite 208, Totowa, NJ 07512 USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Cover illustration: Cancer Neurology Composite Cover Image- composite created by Santosh Kesari, MD, PhD Background: Fluorescence immunostaining showing antibodies reacting to antigens on the cell surface and dendrites of neurons in a patient with paraneoplastic limbic encephalitis and carcinoma of the thymus (Dr. Dalmau). Upper left: Histology of lymphoma invading a blood vessel in the brain. Upper middle: MRI showing numerous brain metastases from breast carcinoma, upper right: CSF cytology showing malignant glioma cells from a patient with high grade glioma and leptomeningeal spread; lower left: PET and MRI images showing an enhancing mass in a patient treated with prior chemoradiation with PET findings of FDG avid disease consistent with viable tumor; lower right: cervical spine MRI showing leptomeningeal spread of lung carcinoma. Printed on acid-free paper 987654321 springer.com

Foreword In the Foreword of the first edition of this book, Jerome Posner, one of the founding fathers of the field of neuro-oncology, observed that in the years prior to the publication of the book interest in neuro-oncology had increased. Since then, interest has only further increased. Both the North American and the European Society for Neuro-Oncology are flourishing, and the number of manuscripts devoted to neuro-oncologic topics is increasing. This development is a good thing. Many cancer patients suffer from neurological signs and symptoms, and all too often these signs and symptoms have a tremendous impact on the quality of life for these patients. Adequate treatment and palliation begins with the prompt recognition and diagnosis of these complications, and expeditious treatment if indicated. Because the nervous system has a limited potential for recovery, the sooner treatment is initiated the better. Dedicated neurologists with ample knowledge of both cancer and neurology are required for that task. This book will help them in their work. Jerome Posner asked himself: Why this book, when there are already many neuro-oncologic textbooks available? To that question we may add: Why a second edition? First of all, this one is special. Most textbooks on neurooncology focus on primary brain tumors. There is a paucity of textbooks covering neurological complications of systemic cancer and its treatment, which is exactly the scope of this book. One should realize that most neurologists spend more time caring for patients with systemic cancer and neurological complications than for primary brain tumor patients. At present there is no other book that provides such complete coverage of the neurological complications of systemic cancer. And although progress may be slow, we are making progress; thus, an update of this book is appropriate. New drugs are being utilized in the management of cancer patients; the role of surgery in the management of spinal metastases is better defined; novel radiotherapy techniques are being used; and novel imaging modalities are becoming more widely available. Therefore, I welcome this second edition, in which some chapters have been completely rewritten and others were added to cover additional topics. The structure of this book meets the requirements of the daily practice of neuro-oncology. The differential diagnosis of signs and symptoms in a cancer patient depends not only on those signs and symptoms, but also on which cancer the patient is suffering from and what treatment has been given. The distinction between more general chapters on metastases and on complications of treatment on the one hand, and tumor-specific chapters on the other, allows the reader to approach a clinical problem from these different points of view. The separate chapters on seizures and steroids reflect the great importance of these topics in the management of cancer patients. Seizures have a great impact on the quality of life of patients. Steroids are important in the management of many neuro-oncologic conditions, but they come with significant side effects. This book will help physicians to manage these. The decision to devote a separate section of the book to neurological complications of cancer treatment is well considered: many cancer patients suffer from treatment-related neurological morbidity. And again, adequate treatment starts with the recognition of possible treatment-related side effects. The chapter on pain focuses on an area that is too often forgotten by neurologists. Many neurological complications of cancer are accompanied by pain, and neurologists taking care of cancer patients need to know how to alleviate pain. Moreover, neurologists have a special role in the management of pain in cancer patient because they are trained in localizing the anatomical origin of signs and symptoms. Pain control is not just about anti-analgesic drugs: if a local cause of the pain is discovered this may very well have treatment implications such as focal radiotherapy. Neurologists should therefore participate in cancer pain management teams. The second edition of Cancer Neurology in Clinical Practice will serve the needs of both residents and experienced physicians. The editors deserve to be congratulated for the many experts they found willing to contribute to this book, coming from various medical disciplines. This also reflects the multidisciplinary character v

vi

Foreword

of neuro-oncology. Because of this combination of outstanding collaborators and its structure, the first edition of this book was already considered to be the new classic textbook on neurological complications of systemic cancer. With this second edition, the editors have not only fulfilled the promise of the first edition, they now also have the obligation to continue their work in future editions. Martin J. van den Bent, MD Neuro-Oncology Unit Daniel den Hoed Cancer Center Rotterdam, The Netherlands

Preface Neuro-oncology has evolved substantially as a clinical and research discipline over the past few decades. Initially the province of isolated devotees, it has become a well-recognized subspecialty of neurology, oncology, and neurosurgery. The Society for Neuro-Oncology, founded in 1995, now has 900 members. Most tertiary care hospitals have staff physicians who consider themselves neuro-oncologists. These physicians typically are involved in evaluation and management of neurologic complications of systemic cancer and its treatment as well as primary brain tumors. Neurologic complications occur in a substantial proportion of cancer patients, often present complex diagnostic and management problems, and commonly have a major impact on quality of life. As cancer patients live longer due to improvements in local and systemic therapies there will continue to be a rise in late neurological complications such as brain metastases and neurotoxicity from these treatments. Simultaneously, improvements in and the widespread availability of diagnostic studies such as magnetic resonance scanning, as well as market forces, have resulted in most patients with neurologic complications of cancer receiving treatment outside of the tertiary care setting. Many medical and radiation oncologists have little formal training in the evaluation of neurologic symptoms. Conversely, most neurologists see relatively few cancer patients and do not have time to keep abreast of technological and pharmacological advances in cancer management. These circumstances contribute to the risk that some patients may not have access to a desirable level of expertise. The principal aim of this work is to provide clinicians from various backgrounds and levels of training with a reference to help focus the differential diagnosis, diagnostic strategy, and treatment plan for the cancer patient with neurologic symptoms and findings. The volume begins with an overview of the field of neuro-oncology and a review of the role of neuroimaging in diagnosis of neuro-oncologic disease. Several chapters on interpretation and management of common neuro-oncologic symptoms follow. Subsequent sections contain chapters on the direct and indirect neurologic complications of cancer as well as complications of therapy. The final section focuses on the spectrum and management of neurologic disease in patients with cancer of specific organs. Although there is necessarily some overlap between these chapters and the earlier, more general chapters, this approach allows the reader the flexibility of approaching a clinical problem either from the symptoms or in the context of the patient’s known diagnosis of malignancy. Our great hope is that in broadening and deepening the familiarity of clinicians with the range and management of neuro-oncologic diagnoses we may improve the quality of care for cancer patients. The increased confidence and competence of the treating physicians in securing a diagnosis, selecting a treatment plan, and communicating prognosis should translate into increased peace of mind and quality of life for patients and their loved ones. David Schiff, MD Santosh Kesari, MD, PhD Patrick Y. Wen, MD

vii

Acknowledgments The warm reception that greeted the first edition of this book reflects the expertise and hard work of the individual chapter authors; similarly, any success of the current edition owes credit to these same individuals. The addition of Santosh Kesari as co-editor infused this project with fresh energy and helpful perspective in addition to acumen; his leadership was critical to the book’s completion. Patrick Wen remains a valued friend and role model of an academic physician; his wisdom, generosity, and good humor have provided me with important sustenance over the years. Most importantly, I would like to thank my patients, Tanya Nezzer, M.D., and Harold Schiff for all they have taught me and for allowing me to do what I love. David Schiff, MD I would also like to thank all the individual chapter authors for their expert contribution. I have learned a great deal from reviewing and editing these chapters, and will refer to them often. I am especially thankful for my clinical and academic mentors, Patrick Wen and David Schiff, for involving me in this revised edition and for all the support and advice over the years. I thank Ainsley Ross for providing secretarial assistance with good cheer and competence. Finally, I dedicate this book to my parents Sriramloo Kesari, M.D. and Sarojini Kesari, and to my family, Jyothsna, Sneha and Pranav for all their love and support. Santosh Kesari, MD, PhD I would like to thank the authors for their willingness to contribute to this book, despite their busy schedules; the editors for their support and encouragement; and Ainsley Ross for administrative and secretarial assistance. I am especially grateful for all the work that David Schiff did to make this book possible and for his friendship over the years. I would like to dedicate this book to my parents Hsiang-Lai Wen, M.D., and Grace Wen, and to my family, May, Katherine, and Jessica, and to Santosh Kesari for all his hard work on this book. Patrick Y. Wen, MD

ix

Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xiii

List of Color Plates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xv

Part I:

Overview

1. The Prevalence and Impact of Neurologic Disease in Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Evert C. A. Kaal and Charles J. Vecht

Part II:

Diagnostic Studies

2. Imaging Neurologic Manifestations of Oncologic Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Arastoo Vossough, R. Gilberto Gonzalez and John W. Henson

Part III:

15

Neurologic Symptoms

3. Seizures and Anti-Epileptic Drugs in Neuro-Oncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael J. Glantz and Julia Batten 4. Corticosteroids in Neuro-Oncology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Santosh Kesari, Nina A. Paleologos and Nicholas A. Vick 5. Headache . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Robert Cavaliere 6. Confusion and Delirium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Augusto Caraceni, Marco Bosisio, Jane M. Ingham 7. Cognitive Dysfunction, Mood Disorders, and Fatigue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Elana Farace and Zarui Melikyan 8. Cancer Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sebastiano Mercadante

Part IV:

3

33 47 57 65 91 113

Direct Complications of Cancer

9. Brain Metastases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ahmir H. Khan and Lawrence Recht 10. Skull and Dural Metastases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Herbert B. Newton 11. Spinal Metastases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jonathan H. Sherman, Dawit G. Aregawi, Mark E. Shaffrey, and David Schiff 12. Leptomeningeal Metastases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ayman I. Omar and Warren P. Mason 13. Peripheral Nervous System Metastases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nicholas Butowski

xi

131 145 163 181 203

xii

Part V:

Contents

Indirect Complications of Cancer

14. Cerebrovascular Complications of Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lisa R. Rogers, Megan C. Leary, and Jeffrey L. Saver 15. Paraneoplastic Syndromes of the Nervous System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Myrna R. Rosenfeld and Josep Dalmau

Part VI:

237

Complications of Cancer Therapy

16. Neurologic Complications of Radiation Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Daisy Chi, Anthony Béhin, and Jean-Yves Delattre 17. Neurologic Complications of Chemotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jörg Dietrich and Patrick Y. Wen 18. Neurological Complications of Hematopoietic Stem Cell Transplantation . . . . . . . . . . . . . . . . . . . . . . . . . . Eudocia Quant and Patrick Y. Wen 19. Central Nervous System Infections in Cancer Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amy A. Pruitt

Part VII:

215

259 287 327 353

Neurologic Complications of Specific Malignancies

20. Neurological Complications of Primary Brain Tumors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tracy T. Batchelor and Thomas N. Byrne 21. Neurologic Complications of Lung Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Suriya A. Jeyapalan and Anand Mahadevan 22. Neurologic Complications of Breast Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Willem Boogerd 23. Neurologic Complications of Female Reproductive Tract Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lauren E. Abrey 24. Neurologic Complications of Genitourinary Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . David E Traul and David Schiff 25. Neurologic Complications of Gastrointestinal Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jeffrey Raizer and Jeffrey Cilley 26. Neurologic Complications of Sarcoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Santosh Kesari and Lara J. Kunschner 27. Neurologic Complications of Head and Neck Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Katherine B. Peters and David Schiff 28. Neurologic Complications of Melanoma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Denise M. Damek 29. Neurologic Complications of Leukemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marc C. Chamberlain 30. Neurological Complications of Lymphomas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Brian P. O’Neill 31. Neurologic Complications of Plasma Cell Dyscrasias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . John J. Kelly 32. Neurologic Complications of Pediatric Systemic Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nicole J. Ullrich and Scott L. Pomeroy Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

381 397 423 449 459 481 495 507 523 555 567 591 607 621

Contributors Lauren E. Abrey, md, Department of Neurology, Memorial Sloan-Kettering Cancer Center, New York, NY Dawit G. Aregawi, md, Department of Neurology, University of Virginia, Charlottesville, VA Tracy T. Batchelor, md, mph, Stephen E. and Catherine Pappas Center for Neuro-Oncology, Massachusetts General Hospital, Boston, MA Julia Batten, aprn, mph, Department of Oncology, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT Anthony Béhin, md, Service de Neurologie Mazarin, Groupe Hospitalier Pitié-Salpêtrière, Paris, France Willem Boogerd, md, phd, Department of Neuro-Oncology, The Netherlands Cancer Institute/Antoni van Leeuwenhoekhuis and Slotervaart Hospital, Amsterdam, The Netherlands Marco Bosisio, phd, Psychology Unit, National Cancer Institute, Milan, Italy Nicholas Butowski, md, Neuro-Oncology Service, Department of Neurological Surgery, University of California, San Francisco, San Francisco, CA Thomas N. Byrne, md, Stephen E. and Catherine Pappas Center for Neuro-Oncology, Massachusetts General Hospital, Boston, MA Augusto Caraceni, md, Neurology, Rehabilitation, and Palliative Care Unit, “Virgilio Floriani” Hospice, National Cancer Institute, Milan, Italy Robert Cavaliere, md, Department of Neurology, The Ohio State University, Columbus, OH Marc C. Chamberlain, md, Department of Neurology and Neurological Surgery, University of Washington, Fred Hutchinson Cancer Research Center, Seattle, WA Daisy Chi, md, Service de Neurologie Mazarin, Groupe Hospitalier Pitié-Salpêtrière, Paris, France Jeffrey Cilley, md, Department of Medicine, Hematology/Oncology Section, Northwestern University, Feinberg School of Medicine, Chicago, IL Josep Dalmau, md, phd, Department of Neurology, Division of Neuro-Oncology, University of Pennsylvania, Philadelphia, PA Denise M. Damek, md, Department of Neurology, University of Colorado, Denver, CO Jean-Yves Delattre, md, Service de Neurologie Mazarin, Groupe Hospitalier Pitié-Salpêtrière, Paris, France Jörg Dietrich, md, phd, Division of Neuro-Oncology, Department of Neurology, Brigham and Women’s Hospital, Boston, MA Elana Farace, phd, Department of Neurosurgery, Pennsylvania State University Hershey Medical Center, Hershey, PA Michael J. Glantz, md, Department of Oncology, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT R. Gilberto Gonzalez, md, phd, Division of Neuroradiology, Massachusetts General Hospital, Boston, MA John W. Henson, md, Pappas Brain Tumor Imaging Research Program, Division of Neuroradiology, Massachusetts General Hospital, Boston, Massachusetts Jane M. Ingham mbbs, fracp, fachpm, Palliative Medicine, University of New South Wales, St Vincents Hospital, Darlinghurst 2010, NSW, Australia Suriya A. Jeyapalan, md, ma Brain Tumor Center, Beth Israel Deaconess Medical Center, Boston, MA Evert C. A. Kaal, md, phd, Department of Neurology, Medisch Centrum Rijnmond-Zuid, Rotterdam, The Netherlands John J. Kelly, md, Department of Neurology, The George Washington University Medical Center, Washington, D.C. Santosh Kesari, md, phd, Center for Neuro-Oncology, Dana-Farber/Brigham and Women’s Cancer Center, and Division of Cancer Neurology, Department of Neurology, Brigham and Women’s Hospital, Boston, MA xiii

xiv

Contributors

Ahmir H. Khan, md, phd, Department of Neurology, Stanford University Medical Center Stanford, CA Lara J. Kunschner, md, Department of Neurology, Drexel University College of Medicine, Pittsburgh, PA Megan C. Leary, md, Department of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, CA Anand Mahadevan, md, frcs, Brain Tumor Center, Beth Israel Deaconess Medical Center, Boston, MA Warren P. Mason, md, frcpc, Department of Medicine, Princess Margaret Hospital, University of Toronto, Toronto, Canada Zarui Melikyan, phd, Department of Neurosurgery, Pennsylvania State University Hershey Medical Center, Hershey, PA Sebastiano Mercadante, md, Department of Anesthesia and Intensive Care and Pain Relief and Palliative Care, University of Palermo, La Maddalena Cancer Center, Palermo, Italy Herbert B. Newton, md, faan Division of Neuro-Oncology, Dardinger Neuro-Oncology Center, Columbus, OH Ayman I. Omar, md, phd, Department of Medicine, Princess Margaret Hospital, University of Toronto, Toronto, Canada Brian P. O’Neill, md, Department of Neurology, Mayo Medical School, Rochester, MN Nina A. Paleologos, md, Department of Neurology, Evanston Northwestern Healthcare, Evanston, IL Katherine B. Peters, md, phd, Department of Neurology, Johns Hopkins Hospital, Baltimore, MD Scott L. Pomeroy, md, phd, Department of Neurology, Children’s Hospital Boston, Boston, MA Amy A. Pruitt, md, Department of Neurology, University of Pennsylvania School of Medicine, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania Eudocia Quant, md, Department of Neurology, Brigham and Women’s Hospital, Boston, MA Jeffrey J. Raizer, md, Department of Neurology, Northwestern University, Feinberg School of Medicine, Chicago, IL Lawrence Recht, md, Department of Neurology, Stanford University Medical Center, Stanford, CA Lisa R. Rogers, do, Department of Neurology, University of Michigan Medical Center, Ann Arbor, MI Myrna R. Rosenfeld, md, phd, Department of Neurology, Division of Neuro-Oncology, University of Pennsylvania, Philadelphia, PA Jeffrey L. Saver, md, Department of Neurology, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, CA David Schiff, md, Departments of Neurology, Neurological Surgery, and Medicine (Hematology/Oncology), University of Virginia, Charlottesville, VA Mark E. Shaffrey, md, Department of Neurological Surgery, University of Virginia, Charlottesville, VA Jonathan H. Sherman, md, Department of Neurological Surgery, University of Virginia, Charlottesville, VA David E. Traul, md, phd, Department of Neurology, University of Virginia, Charlottesville, VA Nicole J. Ullrich, md, phd, Department of Neurology, Children’s Hospital Boston, Boston, MA Charles J. Vecht, md, phd, Department of Neurology, Neuro-Oncology Unit, Medical Center, The Hague, The Netherlands Nicholas A. Vick, md, Department of Neurology, Evanston Northwestern Healthcare, Evanston, IL Arastoo Vossough, md, phd, Division of Neuroradiology, Massachusetts General Hospital, Boston, MA Patrick Y. Wen, md, Center for Neuro-Oncology, Dana-Farber/Brigham and Women’s Cancer Center, and Division of Cancer Neurology, Department of Neurology, Brigham and Women’s Hospital, Boston, MA

List of Color Plates

The following color illustrations are printed in the insert. Chapter 10 Fig. 5:

Fig. 7:

MRI scans of a 60-year-old female with stage 4 breast carcinoma and complaints of headache and abnormal vision. (A) Non-enhanced mid-sagittal T1-weighted image, demonstrating an isointense, metastatic tumor within the clivus, anterior to the lower brainstem. (B) Enhanced axial T1-weighted image, demonstrating the enhancing mass within the clivus. (C) Low power view of surgical specimen of metastatic medullary breast carcinoma removed from the skull base, revealing nests of tumor cells within and alongside regions of bone (hematoxylin and eosin @100x). (D) High-power view of the same specimen (hematoxylin and eosin @200x). Contrast-enhanced T1-weighted MRI scans of a 40-year-old female with a history of stage 4 breast carcinoma and complaints of headache and somnolence. The axial (A) and coronal (B) images both demonstrate the presence of a large, enhancing dural-based mass in the right frontal region, with compression of the underlying brain and regional edema. (C) Pathological materials from an unrelated patient with a dural metastasis from renal cell carcinoma, in whom surgical resection was necessary because of severe mass effect. Note the clumps of renal cell carcinoma cells in and around the fibrous layers of the dura (hematoxylin and eosin @100x).

Chapter 15 Fig. 2:

Fig. 3:

Fig. 4:

Antibodies to cell surface antigens in a patient with paraneoplastic limbic encephalitis and carcinoma of the thymus. Rat hippocampal neurons immunolabeled with the patient’s CSF. Note the intense reactivity of patient’s antibodies with the cell surface and dendrites of neurons (antigen unknown). CSF from control subjects does not produce any reactivity (not shown). Immunofluorescence x800. Studies in a patient with anti-Yo associated cerebellar degeneration. Panel A corresponds to a normal FDG-PET obtained when the patient had anti-Yo associated cerebellar degeneration for 2 years. Panel B is an FDG-PET obtained 3 years later (5 years after neurologic symptom presentation) showing a hypermetabolic abnormality in the right axillary lymph nodes (arrow). Panel C shows the patient’s anti-Yo antibodies immunolabeling a section of rat cerebellum (note the characteristic Purkinje cell reactivity of the anti-Yo antibodies). Panel D shows that the neoplastic cells from the lymph node (identified by PET) react with the anti-Yo antibodies of the patient (dark cells). (C and D, avidin–biotin–peroxidase, x400.) Studies in a patient with paraneoplastic anti-NMDAR encephalitis and ovarian teratoma. Panel A shows the CSF reactivity of a patient with anti-NMDAR antibodies with a sagittal section of rat hippocampus; the immunolabeling is mainly concentrated in the inner aspect of the molecular layer adjacent to the dentate gyrus (arrow). Panel B shows that the antibody reactivity is with the cell surface and dendrites of neurons (the picture corresponds to a culture of rat hippocampal neurons immunolabeled with the patient’s antibodies). Panel C shows that the teratoma of the patient contains immature neurons; these are demonstrated with MAP2 labeling, a specific marker of neurons and dendrites. (A, avidin–biotin–peroxidase, x50; B, immunofluorescence x800; C, avidin–biotin–peroxidase, x400.)

xv

xvi

Fig. 5:

List of Color Plates

Severe neurogenic muscle atrophy in a patient with SCLC and anti-Hu-associated myelitis. The initial neurologic symptom of this patient was flaccid motor weakness selectively involving the upper extremities and neck extension. After 8 weeks he was unable to move the upper extremities. These symptoms associated with fasciculations and loss of reflexes in the arms. Cranial nerves and strength and reflexes in the lower extremities were normal. The picture demonstrates widespread atrophy of the muscles of the neck and shoulder.

Chapter 17 Fig. 7:

This 78-year-old woman with an 8-year history of CLL developed slowly progressive right hemiparesis and hand twitching over several months. She had received rituximab and fludarabine within one year of onset of these symptoms. FLAIR MR sequences demonstrated multiple hyperintense lesions (A, B). Brain biopsy revealed prominent gliosis with scattered cells with enlarged nuclei and stippled, rim-like chromatin pattern suggestive of viral inclusions (C). Immunohistochemistry for JC virus showed positive staining of these cells (D). (Generously provided by Dr. David Schiff, University of Virginia, Charlottessville, VA.)

Chapter 20 Fig. 2:

Normal blood–brain barrier demonstrating tight junctions between endothelial cells; a normal basement membrane and adjacent astrocyte foot processes. (Adapted with permission.)

Chapter 30 Fig. 2:

Fig. 5:

Fig. 6:

Fig. 8:

Intravascular lymphoma: Multiple punctate and confluent areas of T2 signal intensity in the cerebral white matter (A) with spotty contrast enhancement (B) for unusual processes such as vasculitis, lymphoma, and infection. Biopsy of the lesion showed intravascular lymphomatosis with immunoperoxidase stains which support diagnosis of a large cell lymphoma, B-cell phenotype (C). Histologic features of dural marginal zone B-cell lymphomas. (A) Lymphoid infiltrate in the leptomeninges extending into the Virchow–Robin space. (B) Small- to medium-sized neoplastic cells with a distinct monocytoid appearance and frequent plasmacytoid/plasma cells (arrow). (C) Reactive lymphoid follicles with germinal centers in some patients. (D) Scattered areas with increased mitotic figures (arrowheads) were seen in two patients. Role of positron emission tomography in diagnosis and management lymphoma. Sixtytwo-year-old female with PCNSL involving cauda equina, conus and left retina, status post L2, L3 laminectomy, and cauda equina biopsy. PET is negative in the eyes, but positive in the conus (thin arrow) and cauda equina (thick arrow). (A) 3-D maximum intensity projection (MIP) image, (B) sagittal fused PET-CT image, (C) transaxial images of CT(I), PET(II), and fused PET/CT. Root entrapment at cranial and spinal nerve foramens. A simple yet helpful clue is the “winking owl” sign (A) of a destroyed pedicle ipsilateral to the radiculopathy as shown on the left in this case.

Chapter 32 Fig. 1:

Distribution of cancer types among children 0–19 years old, 1988–2001 (2).

I

Overview

1

The Prevalence and Impact of Neurologic Disease in Cancer Evert C. A. Kaal,

MD,

and Charles J. Vecht,

MD, PHD

CONTENTS Introduction Back Pain Encephalopathy Polyneuropathy Cerebrovascular Complications Brain Metastases Quality of Life Conclusion References

Summary Up to 45% of patients with cancer require evaluation of a neurologic problem. The most common reasons for referral to a neurology service are pain, altered mental status, motor weakness and headache. After consultation, the most frequent diagnoses made are metabolic and toxic encephalopathy, polyneuropathy, cerebrovascular diseases, vertebral metastasis, and brain metastasis. Early recognition of a neurologic problem and institution of therapy may prevent further neurologic deterioration and contribute to maintaining and improving quality of life in cancer patients. Moreover, adequate pain control should be part of the neurologic and psychological care in cancer patients. Key Words: spinal cord compression, encephalopathy, brain metastasis, polyneuropathy, stroke, pain

1. INTRODUCTION One of the most frequent and debilitating complications of cancer is the involvement of the nervous system. Up to 45% of patients with solid cancer eventually require evaluation of a neurologic problem, and neurologic complications are the most common reason for emergency admissions of patients with cancer (1–3). Studies on referrals to neurology service in cancer centers have provided insight into the frequency of neurologic signs and symptoms in cancer patients and associated diagnoses (1,3–5) (Tables 1 and 2). However, vertebral metastases are noted earlier in cancer patients with the wide availability and frequent use of imaging modalities, so that spinal cord compression is less common as an emergency diagnosis than in previous decades and referrals are made directly to radiation therapists. Among cancer patients undergoing neurologic consultation, back pain was the indication in 18–34% of the cases, motor weakness in 25%, altered mental status in 17–22%, and headache in 15%. The most common neurologic diagnoses were brain metastasis (15–21%), spinal cord compression (8–18%), encephalopathy (10–14%), and From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

3

4

Part I / Overview

Table 1 Reasons for consultation to the neurology service* Pain Altered mental status Motor weakness Headache ∗

18–34% 17–22% 25% 15%

Adapted from Refs. (1,3–5).

Table 2 Diagnosis after consultation* Spinal cord compression Brain metastasis Metabolic/toxic encephalopathy Stroke Leptomeningeal metastasis Plexus lesion Paraneoplastic ∗

8–18% 15–21% 10–14% 7% 5% 5% 1%

Adapted from Refs. (1,3–5).

stroke (7%). Although symptoms such as painful paresthesias or walking difficulties due to a polyneuropathy can be seriously disabling, they less often prompt referral for neurologic consultation. This also holds true for pain, with estimates that 50–80% of patients with cancer receive inadequate pain therapy (6). The impact of neurologic symptoms and diagnosis on improving or maintaining the functional status and quality of life of cancer patients has not been investigated systemically. However, one may safely assume that disabling symptoms impair the activities of daily life, and prevention and therapy of neurologic dysfunction and of pain is thereby of great value. In this chapter, we will discuss the most common neurologic symptoms and diagnoses that require neurologic consultation in cancer patients. In addition, the treatment of pain as a means of improving quality of life is discussed.

2. BACK PAIN In contrast to people without cancer, new-onset backache in cancer should be regarded with suspicion, as it can be the first sign of vertebral metastasis. The most common malignancies that metastasize to the vertebral column are cancers of the prostate, breast, kidney, and lungs as well as malignant melanoma, and are together responsible for 75% of cases of epidural spinal cord compression (ESCC). ESCC may be the first sign of a malignancy in 20% of all cancer patients. Data from retrospective studies on referrals to neurology services in cancer hospitals indicate that back pain is the most frequent reason for neurologic consultation. In more than two-thirds of cases, this is due to a vertebral metastasis with associated risks of development of ESCC (3,4,7). In fact, 30–38% of patients evaluated for back pain and cancer already had signs of ESCC (3–5). Often there is a substantial delay between the onset of back pain and the time that a diagnosis of ESCC is made. A prospective audit for establishing a diagnosis of malignant spinal cord compression demonstrated that most patients developed pain approximately three months earlier, and started to complain to their general practitioner two months before a diagnosis of ESCC was eventually made (8). The pain was severe in 84% of the cases, and at diagnosis, about 80% of patients were not able to walk without assistance (8–11). Warning signs of vertebral metastasis are new-onset back pain, pain in the thoracic spinal column (“between the shoulder blades”), pain worsening in the supine position (“at night”), weakness, and autonomic dysfunction. Almost 80% of patients with vertebral column metastasis will develop radicular pain due to compression of spinal nerve roots (8). Symptoms include uni- or bilateral pain in the arms, trunk, or legs.

Chapter 1 / The Prevalence and Impact of Neurologic Disease in Cancer

5

On physical examination, the site of the local pain on percussion may indicate which vertebral body is involved. Plain radiographs are rarely sufficient: they may predict the level of spinal cord compression in only 21% of cases and display pathological signal only when 30–50% of the bone tissue has been destroyed (8,12). At present, magnetic resonance imaging (MRI) is the only investigation that shows both the presence of vertebral metastasis and the exact level of epidural spinal cord compression in a single image (8) . However, because in 30% of cases the spinal cord is compressed at more than one level, patients with both suspicion of vertebral metastasis and signs of ESCC (i.e., weakness, sensory disturbances, autonomic failure) should receive a MRI of the whole spine to visualize all potential sites of cord compression (13). An expedited work-up for back pain is important because the neurologic function prior to the start of treatment predicts the outcome: approximately 80% of patients who are ambulatory before treatment remain so after either radiation therapy or surgery, and only 5-10% of all patients who are paraplegic before treatment will ever become ambulatory afterward (14,15).

3. ENCEPHALOPATHY Encephalopathy accounts for 10–14% of neurologic diagnoses in patients with cancer and is the second most common cause for neurologic consultation in cancer centers (3–5,16). The clinical spectrum of an encephalopathy ranges from mild cognitive dysfunction to delirium with disturbances in awareness, attention, consciousness, and the development of asterixis, lethargy, and, finally, coma. Delirium is associated with high morbidity: delirious cancer patients are hospitalized more than three times longer than cancer patients without delirium (16). The nature of the encephalopathy can be attributed to a few causes in two-thirds of patients, but is often multifactorial. Most commonly it is due to metabolic dysfunction, diagnosed in 61% of patients consulted for altered mental status, followed by brain metastasis (15%), and chemotherapy toxicity (11%) (3,16). These three major causes of encephalopathy will be discussed in the next section.

3.1. Metabolic Encephalopathy A metabolic encephalopathy in cancer patients can be linked to various causes and in the majority of patients more than one cause for confusion can be identified (17). The main cause of encephalopathy varies: In 32–64% of cases, it is treatment with opioids; in 54%, some form of organ dysfunction; in 43–46%, an infectious disorder; and in 43%, both hypoxia and electrolyte disturbances (16,17). Metastasis to the liver, occurring in at least one-third of cancer patients, may lead to hepatic failure in terminal patients or—irrespective of the extent of liver malfunction—to an acute cytokine-mediated bile duct destruction. Increased levels of ammonia lead to an encephalopathy and ultimately to coma. Also, electrolyte disturbances such as hyponatremia, hypomagnesemia, and hypercalcemia may induce an encephalopathy. The latter, “hypercalcemia of malignancy,” is caused by a parathyroid hormone-related peptide, resulting in generalized bone resorption and reabsorption of calcium by the kidney tubules. An inadequate diet or impaired gastrointestinal uptake of nutrients is also a risk factor for encephalopathy in cancer patients. The most prominent example is Wernicke’s encephalopathy due to insufficient uptake of thiamine. Wernicke’s encephalopathy is potentially preventable and usually reversible if treated immediately (18). These examples illustrate that metabolic encephalopathy may be due to multiple factors in patients with cancer. Therefore, an extensive work-up is essential, including critical evaluation of prescribed drugs (e.g., opioids, corticosteroids), a search for an infectious origin, evaluation of organ functions and electrolytes, and assessment of nutritional and vitamin status.

3.2. Encephalopathy Associated with Brain Metastasis Brain metastasis is common in patients with cancer, and may occasionally present with encephalopathic rather than with focal neurologic symptoms: For example, frontal lobe brain metastases may produce a syndrome with changes in behavior, affect, confusion, and cognitive dysfunction. Encephalopathy in patients with brain metastasis may be subtle and present in patients still able to perform normal daily activities, i.e., those patients with a Karnofsky Performance Score (KPS) of 70 or more. On systematically evaluating cognitive function in these patients, one will often find brain dysfunction: more than

6

Part I / Overview

one-third of patients with a KPS ≥ 70 scored below the 10th percentile on a trail-making test, which measures visual-motor and executive mental functions (19). Impaired cognitive function was present in 65% of patients with brain metastases prior to treatment, and an approximately 10% decline on Mini-Mental State Examination (MMSE) was found with 15% of patients having an MMSE of 23, indicating dementia (20,21). Thus, only a thorough exam can show signs of encephalopathy in patients with an apparently normal cognitive function. The impact of this observation is not known, but one can imagine that a small cognitive decline may already have substantial effects on social interactions or on the performance of complex tasks, especially in executive functioning. Whether cognitive training in patients with a cerebral neoplasm may be beneficial following antineoplastic therapy is currently under investigation (22).

3.3. Drug-Induced Encephalopathy A toxic encephalopathy secondary to chemotherapy or immunosuppressive agents or corticosteroids may be regarded as a separate entity, as its occurrence is often predictable and symptoms are often drug-specific and dose-dependent. A toxic encephalopathy secondary to chemotherapy is rare with a few notable exceptions. The alkylating agent ifosfamide is often used in patients with germ cell tumors and soft tissue sarcomas and induces an encephalopathy in approximately 15% of patients, with confusion as the most common manifestation (23). The encephalopathy is attributed to a reversible inhibition of mitochondrial respiration by ifosfamide metabolites, and symptoms usually clear 3–4 days after cessation of treatment. Another example of a specific drug-induced encephalopathy is a cisplatin encephalopathy, which exists in two forms: (1) a posterior leukoencephalopathy syndrome with cortical blindness, visible on MRI, and occurring at the end of IV cisplatin administration, and (2) an encephalopathy with focal neurologic signs and no abnormalities on MRI, starting within hours until up to 3 months after cisplatin treatment. Seizures and impaired consciousness are associated with both types of cisplatin encephalopathy. In most patients symptoms disappear without permanent neurologic deficits (24). Apart from treatment with antineoplastic drugs, treatment with immunosuppressive agents such as the calcineurin inhibitors cyclosporin and tacrolimus may cause an encephalopathy. These agents may induce a variety of neurologic symptoms including tremor, headache, or even more severe symptoms such as seizures and impaired consciousness (25,26). The immunosuppressant dexamethasone may occasionally induce an encephalopathy, ranging from a subtle manic mood to psychosis and delirium. Reduction or cessation of steroid treatment may be necessary (27).

4. POLYNEUROPATHY Although polyneuropathies are not a common reason for neurologic consultation, painful paresthesias and gait disturbances as a result of a polyneuropathy may overshadow the effect of successful chemotherapy. Development of a polyneuropathy depends on the type of malignancy, type, and dosing of chemotherapeutic agent, co-morbidities (diabetes, age, etc.), and amount of systemic disease. At least four types of polyneuropathies are associated with cancer: (1) toxic polyneuropathy, (2) paraneoplastic (immune-mediated) polyneuropathy, (3) polyneuropathy due to the cancer cachexia syndrome (weight loss, anorexia, asthenia, and anemia), or (4) polyneuropathy due to compression or infiltration of peripheral nerves.

4.1. Toxic Neuropathy A toxic polyneuropathy can either be sensory, motor, or mixed and the type is often drug-specific. One of the best-described neurotoxic drugs is cisplatin. This agent is used for treatment of patients with a variety of cancer including ovarian cancer, lung cancer, testicular tumors, and head and neck cancer. Cisplatin affects thick myelinated nerve fibers in a dose-dependent and cumulative fashion, leading to a sensory polyneuropathy. Symptoms can range from paresthesias to a severe disabling sensory ataxia. Withdrawal of cisplatin will not halt the development of a polyneuropathy; patients may still experience an increase of their symptoms up to 3 months after finishing treatment. The incidence of polyneuropathy is largely dependent on the cumulative dose: after a cumulative dose of 600 mg/m2 , a disabling neuropathy occurs in 10% of patients (28). Applying less intensive regimens (20 instead of 59 mg/m2 /week) does not result in less toxicity, and neuroprotective therapies have not

Chapter 1 / The Prevalence and Impact of Neurologic Disease in Cancer

7

been successful in preventing toxicity (28). The neurotoxic side effects of cisplatin have led to the increased use of other platinum-based therapies such as carboplatin with lesser neurotoxic effects and a similar clinical effect (29). An example of another chemotherapeutic drug causing a polyneuropathy is docetaxel, with 20% of patients developing a sensory neuropathy at a cumulative dose of 600 mg/m2 (30). Also vincristin, thalidomide and bortezomib may induce mono- or polyneuropathies.

4.2. Paraneoplastic Neuropathy Paraneoplastic peripheral neuropathies are relatively uncommon and represent approximately 5% in a series of 422 patients with peripheral neuropathy (31). In contrast to the slowly evolving toxic neuropathy caused by chemotherapy, the clinical presentation can be more acute, sometimes even presenting as a polyradiculitis (e.g., Guillain–Barre syndrome) with tetraparesis. Several types of paraneoplastic polyneuropathies can be distinguished, varying from a sensory anti-Hu-associated polyneuropathy in patients with small cell lung cancer to a chronic inflammatory demyelinating polyneuropathy associated with various hematologic and solid malignancies (32). Symptoms may diminish when the cancer is treated or after immune suppression. Unfortunately, many patients, especially those with a sensory polyneuropathy, will still suffer from their nerve damage months after treatment of their malignancy.

4.3. Cachexia-Associated Neuropathy A less well known cause of a polyneuropathy is severe weight loss, a feature commonly present in patients with cancer (33). Apart from a easily recognizable cause such as a vitamin B12 deficiency after gastric (oncologic) surgery, nutritional polyneuropathy is probably the result of multiple factors including reduced nutrional intake, catabolic disease state, sometimes malabsorption, and increased vulnerability of peripheral nerves to metabolic stress. A prospective neurophysiological study of 71 patients with small cell lung cancer revealed no increased incidence of peripheral neuropathy at the initial stages of disease. However, all patients developed a so-called “neuro(myo)pathy” by the time they had lost 15% of their body weight (33). Treating or preventing the cachexia syndrome with energy- and vitamin-enriched nutritional supplements may prevent further development of a polyneuropathy.

4.4. Plexopathy/Mononeuropathy Apart from the systemic effects of cancer and its therapy leading to a polyneuropathy, local tumor growth may lead to a mono- or oligoneuropathy or to involvement of the brachial or lumbar/sacral nerve plexus (plexopathy). Involvement of the brachial plexus is a common and painful complication of systemic cancer. A plexopathy can be the result of compression of the brachial plexus by lung cancer (Pancoast’s tumor) or as a result of axillary lymph node enlargement in metastic breast cancer. Additionally, radiation therapy of the apical lung (e.g., lung cancer) or the axilla (e.g., breast cancer) can induce a plexopathy. Two separate types of radiation plexopathy have been identified: an often reversible early delayed plexopathy, occurring at a median of 4 months, and an often irreversible delayed plexopathy, developing 1 year after radiotherapy. In the latter group, radiotherapy has induced fibrotic changes of the brachial plexus (34). In one consecutive series of 38 patients with breast cancer, 17 had a lesion of the brachial plexus, of whom 8 had tumor involvement, 5 had radiation fibrosis, 1 had lymphedema entrapment, and 3 had a probable transient neuritis of the plexus. Post-axillary dissection pain seems the most frequent type of postsurgical pain in breast cancer (35). Growth or spread of urogenital or colorectal cancers often induces a lumbar plexopathy. A less well-known complication of cancer is infiltration of peripheral nerves. For example, nasopharyngeal tumors can infiltrate cranial nerves resulting in signs of cranial nerve dysfunction such as diplopia or facial palsy. A classic complication of a metastasis to the skull base is the numb chin syndrome, which is the result of infiltration of the mentalis nerve or the inferior alveolar nerve (part of the trigeminal nerve) resulting in unilateral numbness of the chin (36). Lymphomas may lead to infiltration of the brachial or lumbar plexus as well as of individual peripheral nerves.

5. CEREBROVASCULAR COMPLICATIONS Cerebrovascular disease is the second most common cause of pathologically definable CNS disease in patients with cancer and is an important cause of morbidity and mortality (37,38). Cerebrovascular complications include

8

Part I / Overview

intracranial arterial or venous thrombosis or hemorrhage. Furthermore, the spread of tumor cells in cerebral arteries may lead to thromboembolic complications such as a transient ischemic attack (“transient tumor attack”) or ischemic stroke. The various mechanisms of cerebrovascular complications are presented.

5.1. Ischemic Stroke Patients may present with a TIA or ischemic stroke. In patients with cancer and suspected embolic tumor infarctions, a search for lung or cardiac tumor should be performed. Nonbacterial thrombotic endocarditis commonly produces multiple strokes and neurologic deterioration (39). In a series of consecutive patients with cancer and cerebrovascular events referred to a cardiologist for transesophageal echocardiography, 18% had marantic vegetations and 47% had a definite source for embolism (40). Under these circumstances, early anticoagulation to prevent recurrent strokes is necessary.

5.2. Cerebral Venous Thrombosis In patients with cancer, activated coagulation pathways result in systemic thrombosis in up to 20% of patients during the course of their disease (41). A minority develop cerebral venous thrombosis, with severe headache, papiledema, and focal neurologic signs or seizures as symptoms. The incidence of venous thrombosis is unknown, but probably increases in patients with brain metastasis and reaches percentages of up to 20% after surgical treatment of a cerebral metastasis (42,43). Incidence rates in a general oncologic population provide smaller figures; from approximately 7000 patients seen in neurologic consultation in a cancer center, only 20 (0.3%) were diagnosed with cerebral sinus thrombosis. Half of these patients had a hematologic malignancy (41). However, application of current angio- and venographic techniques may reveal substantially higher figures in the near future. Cancer therapy itself may contribute to a coagulopathy; for example, treatment of breast cancer patients with the estrogen-receptor modulator tamoxifen results in an 82% increased risk of thrombotic complications (44). Another example is induction therapy with l-asparaginase resulting in thrombosis of the superior sagittal sinus (45).

5.3. Cerebral Hemorrhage The occurrence of cerebral hemorrhage in patients with cancer largely depends on the type of cancer. An autopsy study of patients with cancer showed that in patients with leukemia almost 75% of cerebrovascular events are hemorrhages, whereas carcinoma leads to ischemic cerebral events in 54% of cases (37). Apart from the type of primary tumor, impaired coagulation in patients with liver metastases, sepsis, or with thrombocytopenia induced by bone marrow infiltration or disseminated intravascular coagulation may lead to a potentially fatal hemorrhage.

6. BRAIN METASTASES Brain metastases occur in 15–25% of patients with cancer, and in general the prognosis is poor. Symptoms of brain metastasis include headache, which is the presenting symptom in 40–50% of patients, seizures (15–25%), or focal neurologic deficits (40%). Therefore, imaging of the brain—preferably contrast-enhanced MR—is recommended in patients with cancer with a new headache, seizures, or focal deficits. Although prognosis is generally dismal, some patients with a relatively good prognosis may benefit from focal interventions such as surgery or radiosurgery. Recursive partitioning analysis of data from three RTOG trials, including 1200 patients, have identified three prognostic classes: • Class 1: KPS ≥ 70, age < 65 years, a controlled primary tumor and no extracranial disease. • Class 2: KPS ≥ 70, and age ≥ 65 years and/or active extracranial disease. • Class 3: KPS < 70.

Median survival times for Class 1, Class 2, and Class 3 are 7.1, 4.2, and 2.3 months, respectively. Neurologic deficits due to brain metastasis can be treated with corticosteroids. When possible, corticosteroids should be used in a low dose (e.g., 4 mg dexamethasone daily) to avoid serious side effects such as myopathy or diabetes. Higher doses of dexamethasone (16 mg/day or more), sometimes together with osmotherapy (e.g., mannitol, glycerol) or surgery may be applied in emergency situations (46).

Chapter 1 / The Prevalence and Impact of Neurologic Disease in Cancer

9

Patients with symptomatic epilepsy should be treated with anticonvulsants, starting after the first seizure. Prophylactic treatment with anticonvulsants does not prevent seizures in patients with brain metastases and is not recommended (47).

7. QUALITY OF LIFE When regarding the efficacy of a palliative treatment from a patient’s point of view, a treatment should be considered to be beneficial when a patient eventually “feels better” with the treatment rather than without. Therefore, and ideally, quality of life should be a major endpoint in clinical trials. However, quality of life has not been a standard endpoint in controlled studies with patients with neurologic complications from cancer. Often, the benefit has to be extrapolated from endpoints of survival, duration of functional independence, or from radiological endpoints such as time to recurrence in the case of brain metastasis. Thus far, recommendations for improving quality of life in patients with neurologic complications from cancer have not been based on evidence-based data. However, even without the help of such data, one may assume that certain palliative measures are beneficial. These include social support, physiotherapy, nutritional support, and so on; all of these likely contribute to improved well-being. Reduction of pain is considered to be beneficial, and optimal pain treatment should be part of a standard palliative care program. Patients with neurologic complications from cancer may experience specific kinds of pain, in particular neuropathic pain. The treatment of pain as a means to improve quality of life is discussed in the next section.

7.1. Pain Control as a Means to Improve Quality of Life As mentioned earlier, estimates are that 50–80% of patients with cancer pain receive sub-optimal therapy (6). Commonly, fear of side effects, fear of addiction to analgesics, or treatment with inappropriate dosages or drugs leads to inadequate pain control and may impair the patient’s quality of life. Pain as a neurologic complications of cancer can be generated in various organs or structures: e.g., in the head or skull secondary to increased intracranial pressure or meningeal irritation, in the spine as a result of epidural spinal cord metastases, or in the peripheral nerves secondary to surgical dissection of the axillary, cervical, or inguinal nodes as in breast and head and neck cancer or melanoma. Irrespective of the site of pain generation, a commonly accepted model for pain is the division of pain in nociceptive and non-nociceptive pain (48). One assumes that nociceptive pain or inflammatory pain is the result of involvement of the surrounding epineurium of the nerve root or plexus. This kind of pain often responds to analgesics as non-steroidal anti-inflammatory drugs (NSAIDs) and to opioids. Nonnociceptive pain (or neuropathic pain) particularly responds to treatment with anticonvulsants or antidepressants. The incidence of these different types of pain has been determined in patients with cancer of the head and neck. In a series of 25 patients, nineteen patients (75%) had nociceptive pain, which was secondary to tumor recurrence in 16 patients and secondary to benign inflammation in three patients. The six cases of non-nociceptive pain were all diagnosed as neuropathic pain associated with nerve damage associated with a neck dissection. In practice, both types of pain may be present in the same patient (49).

8. CONCLUSION Cancer is increasing in prevalence, and as treatments improve and result in increased patient survival, the incidence of neurologic problems will likely continue to increase. Neurologists, neuro-oncologists, oncologists, neurosurgeons, radiation therapists, and pain and palliative care physicians will continue to manage the complex neurologic problems that these patients face. Clearly, now more then ever, there is a great need to more completely recognize and understand the breadth of neurologic disease in cancer patients and to institute appropriate treatments early.

REFERENCES 1. Gilbert MR, Grossman SA. Incidence and nature of neurologic problems in patients with solid tumors. Am J Med 1986; 81(6):951–954. 2. Sculier JP, Feld R, Evans WK et al. Neurologic disorders in patients with small cell lung cancer. Cancer 1987; 60(9):2275–2283. 3. Clouston PD, DeAngelis LM, Posner JB. The spectrum of neurologic disease in patients with systemic cancer. Ann Neurol 1992; 31(3):268–273.

10

Part I / Overview

4. Porta–Etessam J, Dalmau J. Analysis of the neurologic consultations in an oncologic hospital: contributions of neuro-oncology. Neurologia 1999; 14(6):266–274. 5. Heimans JJ, Taphoorn MJ, van Kooten B. Neurologic consultation in cancer patients. Ned Tijdschr Geneeskd 1993; 137(52):2705–2709. 6. World Health Organization. Cancer pain relief: with a guide to opioid availability. 2nd ed. Geneva: WHO Office of Publications; 1996. 7. Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 1978; 3(1):40–51. 8. Levack P, Graham J, Collie D et al. Don’t wait for a sensory level—-listen to the symptoms: a prospective audit of the delays in diagnosis of malignant cord compression. Clin Oncol (R Coll Radiol) 2002; 14(6):472–480. 9. Helweg-Larsen S, Sorensen PS. Symptoms and signs in metastatic spinal cord compression: a study of progression from first symptom until diagnosis in 153 patients. Eur J Cancer 1994; 30A(3):396–398. 10. Greenberg HS, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: results with a new treatment protocol. Ann Neurol 1980; 8(4):361–366. 11. Bach F, Larsen BH, Rohde K et al. Metastatic spinal cord compression: occurrence, symptoms, clinical presentations and prognosis in 398 patients with spinal cord compression. Acta Neurochir (Wien) 1990; 107(1–2):37–43. 12. Schiff D, Batchelor T, Wen PY. Neurologic emergencies in cancer patients. Neurol Clin 1998; 16(2):449–483. 13. Quinn JA, DeAngelis LM. Neurologic emergencies in the cancer patient. Semin Oncol 2000; 27(3):311–321. 14. Byrne TN. Spinal cord compression from epidural metastases. N Engl J Med 1992; 327(9):614–619. 15. Bach F, Agerlin N, Sorensen JB et al. Metastatic spinal cord compression secondary to lung cancer. J Clin Oncol 1992; 10(11): 1781–1787. 16. Tuma R, DeAngelis LM. Altered mental status in patients with cancer. Arch Neurol 2000; 57(12):1727–1731. 17. Doriath V, Paesmans M, Catteau G, Hildebrand J. Acute confusion in patients with systemic cancer. J Neurooncol 2007. 18. Yae S, Okuno S, Onishi H, Kawanishi C. Development of Wernicke encephalopathy in a terminally ill cancer patient consuming an adequate diet: a case report and review of the literature. Palliat Support Care 2005; 3(4):333–335. 19. Herman MA, Tremont-Lukats I, Meyers CA et al. Neurocognitive and functional assessment of patients with brain metastases: a pilot study. Am J Clin Oncol 2003; 26(3):273–279. 20. Regine WF, Scott C, Murray K, Curran W. Neurocognitive outcome in brain metastases patients treated with accelerated-fractionation vs. accelerated-hyperfractionated radiotherapy: an analysis from Radiation Therapy Oncology Group Study 91–04. Int J Radiat Oncol Biol Phys 2001; 51(3):711–717. 21. Mehta MP, Rodrigus P, Terhaard CH et al. Survival and neurologic outcomes in a randomized trial of motexafin gadolinium and whole–brain radiation therapy in brain metastases. J Clin Oncol 2003; 21(13):2529–2536. 22. Taphoorn MJ, Sitskoorn MM, Aaronson NK. Cognitive rehabilitation of glioma patients: A prospective, randomized study. 2007. Unpublished work. 23. David KA, Picus J. Evaluating risk factors for the development of ifosfamide encephalopathy. Am J Clin Oncol 2005; 28(3):277–280. 24. Steeghs N, de Jongh FE, Sillevis Smitt PA et al. Cisplatin-induced encephalopathy and seizures. Anticancer Drugs 2003; 14(6):443–446. 25. Gijtenbeek JM, van den Bent MJ, Vecht CJ. Cyclosporine neurotoxicity: a review. J Neurol 1999; 246(5):339–346. 26. Kiemeneij IM, de Leeuw FE, Ramos LM et al. Acute headache as a presenting symptom of tacrolimus encephalopathy. J Neurol Neurosurg Psychiatry 2003; 74(8):1126–1127. 27. Stiefel FC, Breitbart WS, Holland JC. Corticosteroids in cancer: neuropsychiatric complications. Cancer Invest 1989; 7(5):479–491. 28. Hilkens PH, van der Burg ME, Moll JW et al. Neurotoxicity is not enhanced by increased dose intensities of cisplatin administration. Eur J Cancer 1995; 31A(5):678–681. 29. Screnci D, McKeage MJ, Galettis P et al. Relationships between hydrophobicity, reactivity, accumulation and peripheral nerve toxicity of a series of platinum drugs. Br J Cancer 2000; 82(4):966–972. 30. Hilkens PH, Verweij J, Stoter G et al. Peripheral neurotoxicity induced by docetaxel. Neurology 1996; 46(1):104–108. 31. Antoine JC, Mosnier JF, Absi L et al. Carcinoma associated paraneoplastic peripheral neuropathies in patients with and without anti-onconeural antibodies. J Neurol Neurosurg Psychiatry 1999; 67(1):7–14. 32. Darnell RB, Posner JB. Paraneoplastic syndromes affecting the nervous system. Semin Oncol 2006; 33(3):270–298. 33. Hawley RJ, Cohen MH, Saini N et al. The carcinomatous neuromyopathy of oat cell lung cancer. Ann Neurol 1980; 7(1):65–72. 34. Rubin DI, Schomberg PJ, Shepherd RF et al. Arteritis and brachial plexus neuropathy as delayed complications of radiation therapy. Mayo Clin Proc 2001; 76(8):849–852. 35. Vecht CJ. Arm pain in the patient with breast cancer. J Pain Symptom Manage 1990; 5(2):109–117. 36. Kapa BR, Ramanarayanan K, Smith M. Numb chin syndrome: a reflection of systemic malignancy. World J Surg Oncol 2006; 4(1):52. 37. Graus F, Rogers LR, Posner JB. Cerebrovascular complications in patients with cancer. Medicine (Baltimore) 1985; 64(1):16–35. 38. Rogers LR. Cerebrovascular complications in cancer patients. Neurol Clin 2003; 21(1):167–192. 39. Royter V, Cohen SN. Recurrent embolic strokes and cardiac valvular disease in a patient with non–small cell adenocarcinoma of lung. J Neurol Sci 2006; 241(1–2):99–101. 40. Dutta T, Karas MG, Segal AZ et al. Yield of transesophageal echocardiography for nonbacterial thrombotic endocarditis and other cardiac sources of embolism in cancer patients with cerebral ischemia. Am J Cardiol 2006; 97(6):894–898. 41. Raizer JJ, DeAngelis LM. Cerebral sinus thrombosis diagnosed by MRI and MR venography in cancer patients. Neurology 2000; 54(6):1222–1226. 42. Levi AD, Wallace MC, Bernstein M et al. Venous thromboembolism after brain tumor surgery: a retrospective review. Neurosurgery 1991; 28(6):859–863. 43. Sawaya R, Zuccarello M, Elkalliny M et al. Postoperative venous thromboembolism and brain tumors: Part I. Clinical profile. J Neurooncol 1992; 14(2):119–125.

Chapter 1 / The Prevalence and Impact of Neurologic Disease in Cancer

11

44. Bushnell CD, Goldstein LB. Risk of ischemic stroke with tamoxifen treatment for breast cancer: a meta-analysis. Neurology 2004; 63(7):1230–1233. 45. Lee JH, Kim SW, Sung KJ. Sagittal sinus thrombosis associated with transient free protein S deficiency after l-asparaginase treatment: case report and review of the literature. Clin Neurol Neurosurg 2000; 102(1):33–36. 46. Kaal EC, Vecht CJ. The management of brain edema in brain tumors. Curr Opin Oncol 2004; 16(6):593–600. 47. Glantz MJ, Cole BF, Forsyth PA et al. Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2000; 54(10):1886–1893. 48. Asbury AK, Fields HL. Pain due to peripheral nerve damage: an hypothesis. Neurology 1984; 34(12):1587–1590. 49. Vecht CJ, Hoff AM, Kansen PJ et al. Types and causes of pain in cancer of the head and neck. Cancer 1992; 70(1):178–184.

II

Diagnostic Studies

2

Imaging Neurologic Manifestations of Oncologic Disease Arastoo Vossough, MD, PHD, R. Gilberto Gonzalez, and John W. Henson, MD

MD, PHD,

CONTENTS Introduction Imaging Features of Brain Metastasis Meningeal Metastases Epidural Spinal Cord Compression (ESCC) Paraneoplastic Neurologic Syndromes Conclusion Acknowledgments References

Summary This chapter presents an up-to-date overview of the imaging manifestations of major neurological conditions that occur in patients with systemic cancer. Conventional and advanced imaging techniques are discussed. Pitfalls in the neuroimaging of brain metastasis are presented, and an imaging approach to the patient with a newly detected brain mass is provided. Key Words: imaging, magnetic resonance imaging (MRI), computed tomography, diffusion, perfusion, spectroscopy, PET, SPECT

1. INTRODUCTION Systemic cancers usually involve the central and peripheral nervous system through direct metastasis or by compression of neural tissue by metastatic disease in adjacent structures, but other mechanisms, including ischemic stroke, venous thrombosis, infection, paraneoplastic neurologic syndromes, metabolic derangements, and side effects of therapy may also produce neurologic complications. The majority of these neurologic complications have important imaging manifestations, and this chapter seeks to outline the most common of these manifestations.

2. IMAGING FEATURES OF BRAIN METASTASIS As many as 170,000 new cases of brain metastasis are diagnosed in the United States each year (1,2), making this the most common metastatic neurologic complication of cancer. Primary tumors of lung, breast, melanoma, carcinoma of unknown primary (CUP), kidney, and colon underlie the majority of cases of brain metastasis. Among individual primaries, metastatic malignant melanoma is associated with the highest frequency From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

15

16

Part II / Diagnostic Studies

of brain metastases. Approximately 80% of metastases occur supratentorially, with 15% in the cerebellum and 5% in brainstem structures, reflecting relative percentages of perfusion to these structures (3,4). However, pelvic malignancies, including gastrointestinal, uterine cancers, and possibly some subtypes of breast cancer demonstrate a predilection to metastasize to the posterior fossa (Fig. 1) (4). Multiple metastatic lesions are seen in 60–75% of all cases as determined with gadolinium-enhanced MR imaging (1,5). Melanoma and small cell lung carcinoma (SCLC) metastases are most likely to be multiple. Metastases are often located at gray-white matter junctions, but any brain site is at risk. It is thought that tumor emboli lodge in arterioles, where changes in vessel caliber trap the cells. New capillary formation, breakdown of the blood–brain barrier and increased vascular permeability lead to characteristic enhancement and surrounding vasogenic edema. Metastases often have significant surrounding vasogenic edema, but this is not a universal finding, particularly with lesions below 10 mm in diameter. On noncontrast-enhanced head CT scans, metastases are typically hypodense to isodense, and lesions may have low attenuation central regions associated with the presence of necrosis or cystic change. Vasogenic edema is seen as an area of low attenuation in the surrounding white matter. Less commonly, metastases are hyperdense before contrast administration, a finding that can indicate the presence of intratumoral hemorrhage, high cell density

Fig. 1. A woman with metastatic ovarian adenocarcinoma had 3 weeks of occipital headache, gait ataxia, and vomiting. A gadolinium-enhanced T1W image shows a septated, cystic–appearing lesion with a posterior mural nodule. There is mass effect on the fourth ventricle and medulla. The T2W image demonstrates a hyperintense signal within the cystic portion of the lesion, whereas the solid portion of the lesion is isointense. A FLAIR image shows more extensive vasogenic edema than does the T2W image. Pelvic malignancies often metastasize to the cerebellum. Cerebellar metastases are notorious for producing only mild symptoms of headache and ataxia. Gadolinium-enhanced T1W axial (upper left) and coronal (upper right) images, axial T2W (lower left) and FLAIR (lower right) images.

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

17

(seen with lymphoma, germinoma, and small cell lung cancer), melanin, or rarely the presence of calcification or of mucinous adenocarcinoma. The tumors most likely to undergo spontaneous hemorrhage are melanoma, choriocarcinoma, renal cell carcinoma, and thyroid carcinoma. The presence of calcification in a brain metastasis implies an indolent lesion (6). Most metastases enhance intensely with iodinated contrast because of the absence of a blood–brain barrier in tumor capillaries. Enhancement can be of ring configuration or, especially with smaller lesions, homogeneous. High-dose contrast with delayed imaging at 1 hr may detect additional lesions in a significant number of patients (7). CT has a lower sensitivity for posterior fossa lesions and isodense hemorrhagic metastases. Lesions missed on CT are almost always less than 10 mm in size. Although not as sensitive as MR imaging, contrast enhanced CT remains very useful for the detection of parenchymal metastases, and is still preferred as a screening test by some oncologists because it can be performed at the same time as CT of other body areas and because of its lower cost compared to MRI. On MRI, most brain metastases are slightly hypointense to isointense with respect to gray matter on T1weighted (T1W) images, with surrounding hypointense edema. Central hypointensity is characteristic of necrotic or cystic tumors. On T2-weighted (T2W) images the more densely cellular (i.e., homogeneously enhancing) component may demonstrate relative hypointensity to brain, but hyperintensity is more common, especially in the cystic or necrotic portions and in the surrounding vasogenic edema. Metastases almost always enhance following intravenous gadolinium contrast injection secondary to tumor angiogenesis and the absence of a blood–brain barrier. Gadolinium-enhanced MR imaging is superior to doubledose delayed CT for lesion detection, anatomic localization, and differentiation of solitary versus multiple metastases (7). Metastases may enhance in a solid or ringlike configuration; the wall of the ring is usually thick and irregular. Contrast enhancement may be strikingly decreased by steroid therapy.

2.1. Pitfalls in Imaging Newly Detected Brain Masses There are a number of unique imaging features that are occasionally seen with brain metastasis. These findings can be very useful in the differential diagnosis of enhancing brain masses. Hemorrhagic metastases demonstrate variable signal intensities depending on the stage of hemorrhage, with the most remarkable feature being intrinsic T1W hyperintensity seen with intracellular methemoglobin. Evolution of signal intensity is much slower in tumoral hematomas than with spontaneous hemorrhage. T2W hypointensity can be seen in some tumors due to hemorrhage, dense cellularity (such as metastatic lymphomas, small cell lung cancer, amelanotic melanomas), mucinous adenocarcinomas (especially of GI, ovarian, and occasionally breast primaries), and rare instances of dystrophic calcification in metastasis (8). Melanoma has unique imaging features (Fig. 2). Approximately one–half of melanotic melanoma metastases are hyperintense on T1-weighted images prior to the administration of gadolinium (9). Melanin itself may lead to T1 shortening (10,11) and melanoma metastases have a propensity for hemorrhage, with methemoglobin also producing T1 shortening. Blood products produce susceptibility effect on T2*-weighted (T2*W) images. T1W hyperintensity and T2*W susceptibility are each five times more common in melanoma metastases than in lung cancer metastases (9). T2*W images can increase the sensitivity of lesion detection, since in one study 7% of melanoma lesions were detected principally on this sequence (9). These findings underline the value of the T2*W sequence in patients with known melanoma who are undergoing CNS staging and in patients with suspected metastasis in whom the primary is not known. The presence of an enhancing lesion of typical appearance in a patient with known advancing metastatic disease permits a reasonable degree of confidence in the diagnosis of brain metastasis. In patients with a newly detected mass or in a patient with a distant history of cancer or an indolent primary, other diagnostic possibilities must be entertained. Primary brain tumors, abscesses, demyelinating lesions, subacute infarcts, and resolving hematomas can be difficult to distinguish from brain metastasis. Recent data have demonstrated that malignant gliomas are multifocal in one–third of patients at the time of diagnosis, and that there are two patterns in patients with multifocal lesions that are helpful in differentiating primary and metastatic tumors. Brain metastases are five times less likely than malignant gliomas to produce a pattern of two or more discrete foci of enhancement within a contiguous area of T2W hyperintensity. One pattern

18

Part II / Diagnostic Studies

Fig. 2. Unique imaging findings are often present with metastatic melanoma, where intrinsic T1W hyperintensity is seen due to the presence of melanin or blood products (T1W image before gadolinium administration, upper left). Susceptibility effect is produced by the presence of blood products. Combined susceptibility effect and T1W hyperintensity are seen in only one quarter of lesions, but this combination is 16 times more likely with melanoma than with lung cancer metastases. Axial T1W precontrast (upper left) and post contrast (upper right), axial FLAIR (lower left) and axial T2*W (GRE, lower right) images.

of multifocality, in which a focus of nonenhancing T2W hyperintensity is distinct from an enhancing mass lesion, is seen only with malignant gliomas (Fig. 3). The ability to distinguish cerebral abscess from neoplasm represents one of the most important advances in the imaging of newly detected brain masses. Diffusion-weighted imaging (DWI) demonstrates restricted diffusion in both neoplasms and abscesses, but the pattern allows the neuroradiologist to differentiate the lesion with a high degree of confidence. Specifically, in neoplasms the restricted diffusion results from dense cellularity, whereas in cerebral abscess the restricted diffusion results from mucopurulent contents of the lesion. Therefore, in brain tumors the restricted diffusion tends to colocalize with the area of nodular contrast enhancement and the central nonenhancing portion of the lesion shows elevated diffusion (Fig. 4). In abscess, the central nonenhancing component shows restricted diffusion. These findings apply to abscesses caused by bacterial and fungal organisms, but not to those from toxoplasmosis, where restricted diffusion is not common. Identification of a neoplasm as the underlying lesion in a parenchymal hematoma is a common and challenging problem in neuroimaging. The most important finding suggestive of tumor is the presence of one or more areas of nodular enhancement adjacent to the hematoma. Other features in favor of hemorrhagic neoplasms are an incomplete hemosiderin ring, the presence of edema that seems more extensive than expected for the age of the hemorrhage, and delay in the expected evolution of signal characteristics in the hematoma (Fig. 5). Serial imaging every 2–3 weeks may be required to be more certain regarding the nature of the lesion.

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

19

Fig. 3. Multifocality occurs with 60% of brain metastasis but is also seen in 33% of malignant gliomas. One pattern of multifocality that is seen with malignant gliomas but not brain metastasis is that of a separate focus of nonenhancing T2W hyperintensity (arrow). Axial FLAIR (upper and lower left) and T1W post–contrast (upper right) images.

Tumefactive demyelinating lesions can present a confusing diagnostic dilemma. However, they most often occur in younger women, which is an age group in which both malignant gliomas and brain metastasis are unusual, and they usually have an acute to subacute presentation. The “open ring sign” is an important feature suggesting the presence of a demyelinating lesion (Fig. 6). This finding is five times more common in demyelinating lesions than in brain tumors, and is 17 times more common than with abscess. When a tumefactive demyelinating lesion is suspected, the appropriate follow up diagnostic studies can be undertaken to evaluate for lesion evolution. Subacute infarcts typically are suggested by the sudden onset of symptoms and the presence, on imaging, of a territorial area of restricted diffusion. About 7–10 days following infarction the restricted diffusion may “pseudonormalize” (i.e., the area of infarction may become isointense on ADC maps) and contrast enhancement appears in reperfused areas of infarction. The territorial pattern of the lesion may be of diagnostic help, but smaller embolic infarcts can be more difficult to distinguish from metastases based on their location. Small metastases from SCLC may mimic a shower of embolic infarcts since they are multiple, small, enhancing, and show homogenously restricted diffusion (Fig. 7). Venous infarcts, which are typically located in the posterior temporal lobe due to thrombosis of the vein of Labbe, are unusual in their appearance, with heterogeneity of diffusion restriction and the not infrequent presence of blood products (Fig. 8). Peritumoral infarcts can be seen following resection of brain metastasis, and they often demonstrate transient contrast enhancement that can be confused with recurrent tumor (12).

20

Part II / Diagnostic Studies

Fig. 4. In tumors, restricted diffusion is present in the densely cellular portion of the lesion, which usually shows nodular enhancement (upper two images), whereas abscesses have restricted diffusion in the central nonenhancing area due to the presence of mucopurulent material (lower two images). Axial T1W images after administration of gadolinium (upper and lower left) and axial diffusion weighted images (upper and lower right).

Fig. 5. The presence of a tumor underlying a parenchymal brain hemorrhage can be suspected in the presence of an “incomplete hemosiderin ring” (arrow), when the presence of vasogenic edema that is more extensive than expected, when areas of nodular enhancement are present adjacent to or within the hematoma, or when the lesion enlarges over a short time interval (right image was taken four weeks after the left and shows a new nodular area of enhancement that could not be ascribed to enhancing hematoma). Axial T1W images (left and right) and T2*W (GRE) images (middle).

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

21

Fig. 6. The “open ring sign” (arrows) is suggestive of a tumefactive demyelinating lesion. The enhancement is in the subcortical white matter and does not cross sulci, producing the open ring. Axial T1W images after administration of gadolinium.

2.2. Strategies to Improve the Detection of Metastases by MRI 2.2.1. Double- and Triple-Dose Contrast There are several strategies to further improve the detection of metastases by MRI. A useful technique for increasing lesion detection is increasing the contrast dose beyond the standard 0.1mmol/kg of body weight. There is a relatively linear relationship between lesion contrast ratio and intravenous contrast dose over the range of 0.05 mmol/kg to 0.3 mmol/kg (13). In a large series specifically evaluating brain metastasis, it was shown that triple–dose gadolinium MRI increases mean lesion contrast, increases qualitative conspicuity of lesions, detects additional small lesions, and in a subset of patients would have changed choice of therapy (14). Increased contrastto-noise ratio and visual assessment ratings were also shown to be superior in a randomized trial (15). Lesions greater than 10 mm will be seen at all dose levels. The utility of higher-dose gadolinium mainly lies in the ability to reveal smaller lesions, especially less than 5 mm (14,16). 2.2.2. Delayed Imaging Several studies of standard-dose gadolinium-enhanced MR imaging also looked at delayed imaging and found marginal benefit. Although delayed imaging improves detection of small metastases, high-dose (0.3 mmol/kg) imaging shows the most lesions (14). One study of the effective time window for scanning concluded that the

Fig. 7. Diffusion–weighted imaging in small densely cellular metastatic small cell lung cancer lesions can mimic embolic infarcts. Axial T1W image after contrast administration (left), diffusion–weighted image (middle) and apparent diffusion coefficient (ADC) map (right).

22

Part II / Diagnostic Studies

Fig. 8. Venous infarcts can produce a confusing appearance with irregular enhancement, heterogeneous restricted diffusion, and areas of hemorrhage. The posterior temporal region is the most common site for venous infarction, due to thrombosis of the transverse sinus (missing on the left, lower right image) and the vein of Labbe. Axial T1W image after administration of gadolinium (upper left), FLAIR (upper right), diffusion-weighted image (lower left) and 3D reconstructed MR venogram (lower right).

post-contrast scan should be started with a 2- to 5-minute delay after injection of the contrast medium, and that there is no major advantage in waiting longer (17). 2.2.3. Magnetization Transfer (MT) Magnetization transfer MR imaging is a technique that increases the contrast of enhancing and nonenhancing lesions by suppressing background signal. It is achieved by selectively saturating the signal from the immobile water protons. MT can be used to improve contrast of gadolinium-enhanced brain lesions on T1-weighted spinecho images. In one study, using MT resulted in a 108% improvement in the contrast–to–noise compared with conventional T1-weighted gadolinium-enhanced images (18). Some studies have suggested that use of standard gadolinium dose with MT are equivalent to those reported for triple-dose gadolinium-enhanced MR imaging with conventional spin–echo techniques (18,19). While use of triple–dose with MT further increases contrast, it does not result in detection of additional tumors (19). 2.2.4. Newer Contrast Agents The sensitivity of brain metastasis detection is improved with MRI contrast agents that possess higher T1 relaxivity. Gadobenate dimeglumine, for instance, significantly increased the sensitivity compared to gadopentetate dimeglumine, gadodiamide, and gadoterate meglumine at similar doses (20). Gadobenate dimeglumine has also been shown to have similar sensitivity at reduced doses when compared to standard dose gadodiamide (21).

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

23

2.2.5. Higher Field MRI Magnet strengths above 1.5 Tesla are now widely available. It has been shown that 3.0 T imaging increases both the subjective assessment of brain metastases and also produces significantly higher signal–to–noise and contrast– to–noise ratios when compared to 1.5 T (22). Gadolinium administration produces higher contrast between tumor and normal brain on 3.0 T than on 1.5 T (23), resulting in better detection of brain metastases and leptomeningeal involvement (22). The relatively improved detection rate by using higher contrast doses is maintained at high field.

2.3. Advanced Techniques in Brain Metastasis Imaging 2.3.1. Magnetic Resonance Spectroscopy (MRS) Consistent with the lack of neuroglial elements, N-acetyl aspartate (NAA) is low or absent in metastases, (24–26). Nodular enhancing portions of metastatic lesions may show an elevated choline/creatine ratio, reflecting high cell density and phospholipid turnover, as do gliomas (24,27,28). Metastatic lesions were initially thought to be distinguishable from high-grade gliomas by the presence of lipid and lactate peaks, lack of creatine, and increased lipid/creatine ratio (29,30). However, in areas of central necrosis in high-grade gliomas, there is often the presence of lactate and lipid, as well as a lack of creatine, therefore decreasing the specificity of the above findings (31). The finding of high choline in the peritumoral region of a lesion is more likely to represent a glioma than metastasis, since as noted earlier, the peritumoral region of metastases is pure vasogenic edema, whereas in gliomas, it represents a combination of tumor cell infiltration and vasogenic edema (31). Complicating this analysis is the fact that the resolution of small voxels is poor, leading to contamination of signal from adjacent voxels. Increased choline/creatine ratio can sometimes distinguish areas of tumor recurrence from radiation necrosis. 2.3.2. Diffusion-Weighted Imaging (DWI) Metastatic tumors with dense cell packing or high nuclear–to–cytoplasmic ratio, such as lymphoma and small cell lung cancer, can show restricted diffusion on DWI, as discussed above (see Fig. 4). Diffusion imaging aids in the differentiation of cystic neoplasms and abscesses, and can play a role in delineating cystic from solid components in metastases (32). Fractional anisotropy (FA) values of either the enhancing or nonenhancing peritumoral portions of high–grade gliomas and solitary brain metastases do not reveal significant differences (33,34). However, when peritumoral FA values were expressed as a differential from values “expected” in bland edema, there was a significant difference between metastases and gliomas. Expected FA values were predicted from a linear regression of FA onto ADC for a series of bland edema cases; the difference between expected and observed FA was called the “tumor infiltration index” (35). Also, the peritumoral mean diffusivity (MD) of metastatic lesions measured significantly greater than that of gliomas (33,35). Displacement of subcortical white matter fibers by subjective visual assessment of diffusion tensor tractography images (as opposed to tract disruption or invasion), is also more commonly seen in metastases compared to high-grade gliomas (34). Diffusion anisotropy is highly sensitive to microstructural changes that do not appear on conventional imaging. This high sensitivity is accompanied by relatively low specificity such that useful DTI-based tissue characterization may ultimately require the use of more sophisticated approaches. The major eigenvalue of the diffusion tensor (reflecting diffusivity in the longitudinal direction) was found to be significantly lower in the peritumoral white matter surrounding high-grade gliomas than metastases, even when the anisotropy showed no difference (36). 2.3.3. Perfusion-Weighted Imaging (PWI) Tumor angiogenesis is a critical feature in the regulation of tumor growth. Metastatic lesions induce neovascularization as they grow and expand. Neovessels resemble those of the primary systemic tumor with fenestrated membranes and open endothelial junctions, in contrast to the highly effective blood–brain barrier of normal brain capillaries (37). In one study PWI could not reliably differentiate metastatic tumors from high–grade gliomas when regional cerebral blood volume (rCBV) was measured in the enhancing portion of the tumors, because both are highly vascular tumors and demonstrate increased rCBV (38). However, significantly higher rCBV was detected within the peritumoral region of primary gliomas compared to metastatic lesions in some studies (39,40). Within individual patients these findings can be difficult to interpret, since the variability is quite high between patients and between lesions.

24

Part II / Diagnostic Studies

2.3.4. Nuclear Medicine Techniques In single-photon emission computed tomography (SPECT) gamma photons emitted from a patient following intravenous injection with a radiotracer are detected with a rotating gamma camera. Thallium-201 is a potassium analogue tracer that localizes in tumors. SPECT imaging with thallium 201 has been used chiefly in the evaluation of primary tumors, but also shows accumulation in metastatic tumors (41,42). Delayed thallium imaging may have some utility in differentiation of high-grade glioma and metastases, according to one group (42). The reported sensitivity of both thallium-201 and technetium-99m MIBI SPECT in the detection of cerebral metastases is 70%. SPECT thallium scanning is not useful for following tumors that do not demonstrate thallium uptake and cannot reliably be used for screening. Positron emission tomography (PET) has been used extensively in the evaluation and follow-up of primary brain tumors, especially for distinguishing between treatment–related necrosis and recurrent tumor. Residual tumor usually demonstrates increased FDG uptake, whereas necrosis from radiation or chemotherapy usually demonstrates isometabolic or decreased FDG uptake. In one study, FDG-PET detected only 50–68% of metastases from nonprimary CNS neoplasms (43) and in another study, only 61% of metastatic lesions in the brain were identified at PET (44). Co-registration of PET with MRI improved the sensitivity for detecting metastatic recurrence from 65% to 86% (45). Most clinical whole–body FDG PET studies performed in cancer patients specifically exclude the brain from the images. Newer tracers are continuously being developed for PET imaging, which may further increase the utility of this modality in evaluation of brain metastases.

2.4. Approach to Evaluation of Patients with Newly Diagnosed Brain Masses The diagnostic approach to patients with a newly detected brain mass has recently been critically evaluated. Patients present to their physician’s office or to the emergency ward with neurologic symptoms and a brain mass is discovered on neuroimaging. In the absence of a history of systemic cancer, a series of radiological studies usually follows, including CT imaging of the chest, abdomen, and pelvis, and whole–body FDG-PET imaging. A site for biopsy is then selected. Because the first step in the management of these patients is to obtain a histopathologic diagnosis, the search for a systemic tumor can be seen as an approach to selecting a biopsy site. An extended search for the extent of systemic tumor prior to establishing a histopathologic diagnosis can lead to an inappropriate expenditure of resources, including unnecessary imaging studies and longer interval to diagnostic biopsy. Staging examinations are appropriate for patients with a systemic malignancy but are inappropriate for most patients with primary brain tumors. In the ideal situation, a decision regarding biopsy site selection would be made quickly after the identification of a brain mass based on a rationally chosen set of diagnostic studies, and the procedure scheduled immediately thereafter. A retrospective analysis of this well-defined clinical situation compared the presenting features and diagnostic workup of patients with neurologic symptoms as the presenting manifestation of cancer to patients ultimately found to have a primary brain tumor (46). Notable patterns were seen in this group of patients. The features of the clinical presentation in patients with metastatic and primary tumors are presented in Table 1. Patients with brain metastasis were more likely to have an acute presentation, as assessed by the likelihood of the first evaluation occurring in an emergency room setting, and the erythrocyte sedimentation rates were higher in patients with brain metastasis. Lung cancer was markedly over–represented as a primary systemic malignancy in patients who present with neurologic symptoms from brain metastasis. Breast cancer, abdominal malignancies, and melanoma were uncommon primaries in this population. The propensity of lung cancers to produce early brain and other organ metastasis is well–documented (47). Early systemic metastasis likely reflects a unique aspect of lung cancer, such as its relative proximity to the systemic arterial circulation. The diagnostic imaging features of the patients with metastatic and primary tumors is presented in Table 2. The high frequency of the lung as the primary site of cancer would indicate that imaging of the chest should be a major focus in the search for a systemic malignancy. Brain MRI and chest CT identified the ultimate site of diagnostic biopsy in 98% of patients who presented with a newly detected brain mass. CXR is substantially less

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

25

Table 1 Features of Patients with a Newly Detected Brain Mass in the Absence of Known Cancer (46) Patients with a brain metastasis Patient number Age (average in yr) M:F Emergency ward presentation Average duration of symptoms (days) Percentage with lung cancer Presenting neurologic symptoms**: Headache Altered mental status Weakness Seizure ESR (mean)***

Patients with a primary brain tumor*

88 60.3 1:1 60% 48 82%

88 59.2 (p = 0.59) 1.8:1 (p = 0.07) 32% (p < 0.01) 54 (p = 0.56) 0%

32 29 26 12 49

30 38 25 19 23

(p (p (p (p (p

= = = = <

0.87) 0.21) 1) 0.23) 0.01)

* p values are given in parentheses. ** Total exceeds 88 because patients had more than one presenting symptom. *** Erythrocyte sedimentation rate in mm/hr.

Table 2 Imaging of Patients with Newly Detected Brain Masses (46) Patients with a brain metastasis Patient number Single mass CXR suggested tumor CT chest suggested tumor CT abdomen/pelvis suggested tumor Bone scan positive Chest CT revealed site of biopsy

88 42% 60% 97% 29% 20% 93%

(37/88) (53/88)* (83/86)* (20/69) (13/66) (41/44)***

Patients with a primary brain tumor 88 75% (66/88) (p < 0.01) 1% (1/88) 15% (4/27)** 9% (2/23) 0% (0/8) 0%

* Comparison of the sensitivity of chest CT and CXR (p < 0.01). ** Findings included one or more small lung nodules and an adrenal adenoma. *** Of 44 patients who had a chest CT and biopsy other than brain.

sensitive as a diagnostic test in the detection of a systemic malignancy (48). These clinical and imaging features can be used to optimize the diagnostic workup of patients with newly detected brain masses.

3. MENINGEAL METASTASES Leptomeningeal metastasis (LM) is a less common form of metastatic disease than brain metastasis, with breast carcinoma, lung carcinoma, melanoma, lymphoma, and leukemia being the most common underlying tumors. Most frequently, metastatic disease spreads to the meninges hematogenously via small meningeal vessels. Superficial parenchymal lesions and skull metastases may, however, directly invade the meninges. Pachymeningeal metastases are usually associated with calvarial metastases, but isolated dural metastases can be seen in breast cancer, lymphoma, leukemia, prostate cancer, and neuroblastoma. Metastases involving the dura may result in subdural effusions and can also lead to venous sinus thrombosis, either by direct invasion or by compression. Noncontrast-enhanced CT is insensitive in detecting the direct lesions of LM, but can demonstrate the presence of hydrocephalus when present. Contrast-enhanced CT is more sensitive than nonenhanced T1- and T2-weighted MR images, but inferior to contrast MR imaging. MR imaging with gadolinium is most commonly used to detect pachymeningeal and leptomeningeal metastases (Fig. 9), but FLAIR imaging is also quite sensitive in

26

Part II / Diagnostic Studies

Fig. 9. Extensive leptomeningeal and pial enhancement over the surface of the brain and cervical cord can be seen with leptomeningeal metastasis. Imaging is a valuable adjunct to diagnosis when CSF cytology is negative or when lumbar puncture cannot be safely obtained. Imaging can also be helpful in choosing radiation therapy fields for bulky LM. Sagittal T1W images after gadolinium administration through the head (left) and upper spine (right).

detection of leptomeningeal disease, showing hyperintensity in the subarachnoid spaces. Patchy enhancement of the leptomeninges is the most common MR finding. Occasionally, meningeal metastases invade the underlying parenchyma with parenchymal T2W hyperintensity, swelling, and contrast enhancement. Dural metastases are distinguished by the fact that the lesions do not extend into sulci along the subarachnoid pathways. Although MR imaging with gadolinium is relatively more sensitive for detecting meningeal disease, it is not specific. Diffuse linear or focal linear can be seen in both neoplastic and inflammatory lesions. Intracranial hypotension produces diffuse symmetrical dural enhancement in about 10% of patients following diagnostic lumbar puncture (LP) and this finding can be seen within 24 hours of LP (unpublished observations). In patients with mass lesions causing obstructive hydrocephalus, apparent leptomeningeal enhancement may be secondary to vascular stasis rather than to carcinomatosis. It is not uncommon for the neuro-oncologist to use imaging evidence for diagnosis of LM in patients where CSF cytopathologic findings are nondiagnostic or when patients are unable to undergo LP. In this case, the entire neuraxis should be imaged with MRI after the administration of gadolinium. Detection of nodular enhancement along the cauda equina may be particularly helpful in this situation. Radionuclide cisternography is often employed to assess the patency of the CSF flow prior to instillation of intrathecal chemotherapy in order to increase the likelihood that the therapeutic agent will reach the entire subarachnoid space (49). If a site of blockage is demonstrated, then focal irradiation can be administered to areas of bulky disease to re-establish CSF flow.

4. EPIDURAL SPINAL CORD COMPRESSION (ESCC) The incidence of clinically diagnosed ESCC in patients with systemic cancer is approximately 5%. There are approximately 18,000 cases of symptomatic ESCC in the United States each year. ESCC is the initial presentation of cancer in up to 20% of patients with this syndrome. Patients with lung cancer, hematological malignancies, and cancer of unknown primary are especially likely to present with symptoms and signs of ESCC (50). About 60% of vertebral metastases producing ESCC occur in the thoracic spine—25% in the lumbar spine and the remainder in the cervical spine. A large retrospective study found that 30% of patients with ESCC have multiple, synchronous epidural lesions.

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

27

Fig. 10. Epidural spinal cord compression in the midthoracic spine. In this patient the mass lesion in the spinal canal is arising from posterior bony elements (i.e., lamina) but there is edema in the posterior aspect of the vertebral body that represents contiguous bony metastatic disease. The most common pattern is direct extension from the body posteriorly into the spinal canal. Lymphomas often enter the spinal canal via neural foramina. Sagittal FSE T2-weighted image.

MRI is the most sensitive and specific test for spinal metastases (Fig. 10). Because MRI is sensitive to changes in the bone marrow, this technique is able to detect the presence of metastatic disease within the vertebral bodies. Gadolinium can mask bony involvement because T1W hypointense metastases may become isointense after contrast administration. Because synchronous epidural metastases occur at multiple levels in approximately 20–30% of cases, the entire vertebral column should be imaged. Plain X-rays may be useful for identifying vertebral body metastases. Abnormalities include erosion of a pedicle and collapse of the vertebral body. Plain X-rays are suggestive of epidural extension in the presence of greater than 50% vertebral body collapse or pedicle erosion. CT scans are more sensitive than plain X-rays or bone scans for identifying vertebral metastases, and like MRI, can image paravertebral disease. MRI has replaced myelography as the definitive radiographic test for epidural metastases. However, if the patient is unable to tolerate MRI because of a contraindication, claustrophobia or severe pain, or if there is a strong clinical suspicion of cord compression despite a negative MRI of poor quality, CT myelography can be useful. The entire spinal canal should be imaged.

5. PARANEOPLASTIC NEUROLOGIC SYNDROMES The paraneoplastic neurologic syndromes constitute a group of rare disorders that may affect the nervous system at all levels, and not uncommonly involve multiple sites within a single individual (51). These syndromes share several common features. They often precede the diagnosis of the primary cancer. The primary cancer is usually of limited stage and may not be detectable at all. The presence of a known primary cancer often confounds the diagnosis of the paraneoplastic syndrome, as the neurologic symptoms are often attributed to undiagnosed metastatic disease. SCLC appears to have the highest incidence among all cancers of associated paraneoplastic neurologic syndromes, but clinically significant paraneoplastic neurologic syndromes probably occur in less than 1% of all cancer patients. Paraneoplastic neurologic syndromes can occur in any part of the nervous system ranging from cortical neurons to muscle fibers. Imaging manifestation are most relevant and patients with paraneoplastic encephalomyelitis and paraneoplastic cerebellar degeneration.

5.1. Encephalomyelitis Paraneoplastic limbic encephalomyelitis (PLE) is characterized by memory loss, confusion, personality changes, and hallucinations (52). Less commonly, involvement of the brainstem results in cranial nerve symptoms (such as

28

Part II / Diagnostic Studies

Fig. 11. Paraneoplastic encephalitis is most often seen as a nonenhancing area of T2W signal hyperintensity in the medial temporal lobe. This 62-year-old man had relapsing limbic encephalitis and anti-voltage-gated potassium channel antibodies in his serum. The signal abnormality in the medial aspect of the left temporal lobe showed near complete resolution over time. Axial FLAIR image.

deafness, vertigo, and diplopia), weakness, central respiratory failure, or involvement of the autonomic system. Neuropathological findings show multifocal inflammatory infiltrates. In a recent review of 50 patients with PLE (52), lung cancer was the most commonly associated malignancy, comprising 50% of the cases. In 60% of patients the neurologic symptoms preceded the diagnosis of cancer, by a median interval of 3.5 months. One-half of MRIs showed changes in the limbic system. MRI abnormalities consist of T2–weighted changes in the medial temporal lobes and brainstem that do not enhance (Fig. 11). The most important differential diagnosis to be considered with this radiologic picture is herpes simplex encephalitis.

5.2. Paraneoplastic Cerebellar Degeneration Symptoms of PCD begin as mild truncal ataxia, evolving over the course of several weeks to months to include the limbs and trunk. Dysarthria, nystagmus, vertigo, diplopia, and oscillopsia are common symptoms. After a period of subacute progression, the disease usually stabilizes, leaving the patient severely disabled. Signs are bilateral, though one side can be more affected than the other. Although signs and symptoms are primarily confined to the cerebellar system, other areas of the nervous system may be affected, producing altered mental status, extrapyramidal signs, hearing loss, hyperreflexia, and peripheral neuropathy. In patients with SCLC, this syndrome may exist in combination with a widespread encephalomyeloneuritis. Detection of anti-Hu or anti-Yo antibodies in serum or CSF can be useful to confirm the clinical suspicion and to aid the search for a primary tumor. Over time, neuroimaging shows cerebellar atrophy, reflecting the specific loss of Purkinje cells. Other conditions that can produce selective cerebellar atrophy include phenytoin intoxication or chronic phenytoin use, cerebellar degenerative conditions such as olivopontocerebellar atrophy (OPCA) and metabolic conditions such as Wernicke–Korsakoff syndrome resulting from thiamine deficiency.

6. CONCLUSION High–resolution MR imaging with intravenous gadolinium has greatly enhanced the ability to detect and characterize metastases as well as to differentiate them from other intracranial processes. Combination of conventional MRI with more advanced techniques of DWI, PWI, MRS, and PET can be useful in the evaluation and management of patients with intracranial metastases. Nevertheless, additional research is needed to further determine which imaging modality is most appropriate for specific patient populations and to determine whether the newer imaging techniques are useful in terms of patient outcome.

Chapter 2 / Imaging Neurologic Manifestations of Oncologic Disease

29

ACKNOWLEDGMENTS This work was supported by the Pappas Brain Tumor Imaging Research Program at Massachusetts General Hospital in Boston.

REFERENCES 1. Hutter A, Schwetye KE, Bierhals AJ et al. Brain neoplasms: epidemiology, diagnosis, and prospects for cost-effective imaging. Neuroimaging Clin N Am 2003;13(2):237–50, x–xi. 2. Johnson JD, Young B. Demographics of brain metastasis. Neurosurg Clin N Am 1996;7(3):337–344. 3. Delattre J-Y, Krol G, Thaler HT et al. Distribution of brain metastases. Arch Neurol 1988;45:741–744. 4. Posner JB. Management of brain metastases. Rev Neurol (Paris) 1992;148(6–7):477–487. 5. Sze G, Milano E, Johnson C, Heier L. Detection of brain metastases: comparison of contrast–enhanced MR with unenhanced MR and enhanced CT. AJNR 1990;11:785–791. 6. Ohmoto Y, Nishizaki T, Kajiwara K et al. Calcified metastatic brain tumor: two case reports. Neurol Med Chir (Tokyo) 2002;42(6): 264–267. 7. Schaefer PW, Budzik RF, Jr., Gonzalez RG. Imaging of cerebral metastases. Neurosurg Clin N Am 1996;7(3):393–423. 8. Ricke J, Baum K, Hosten N. Calcified brain metastases from ovarian carcinoma. Neuroradiology 1996;38(5):460–461. 9. Gaviani P, Mullins ME, Braga TA et al. Improved detection of metastatic melanoma by T2*-weighted imaging. AJNR Am J Neuroradiol 2006;27(3):605–608. 10. Atlas SW, Braffman BH, LoBrutto R. Human malignant melanoma with varying melanin content in nude mice: MR imaging, histopathology, and electron paramagnetic resonance. J Comput Assist Tomogr 1990;14:547–554. 11. Enochs WS, Petherick P, Bogdanova A et al. Paramagnetic metal scavenging by melanin: MR imaging. Radiology 1997;204(2): 417–423. 12. Ulmer S, Braga TA, Barker FG et al. Clinical and radiographic features of peritumoral infarction following resection of glioblastoma. Neurology 2006;67:1668–1670. 13. Yuh WT, Fisher DJ, Mayr–Yuh NA et al. Review of the use of high-dose gadoteridol in the magnetic resonance evaluation of central nervous system tumors. Invest Radiol 1992;27 Suppl 1:S39–S44. 14. Yuh WTC, Tali ET, Nguyen HD et al.. The effect of contrast dose, imaging time, and lesion size in the MR detection of intracerebral metastasis. AJNR 1995;16:373–380. 15. Haustein J, Laniado M, Niendorf H et al. Triple-dose versus standard-dose gadopentetate dimeglumine: a randomized study in 199 patients. Radiology 1993;186:855–860. 16. Akeson P, Larsson EM, Kristoffersen DT et al. Brain metastases: comparison of gadodiamide injection-enhanced MR imaging at standard and high-dose, contrast-enhanced CT and non-contrast-enhanced MR imaging. Acta Radiol 1995;36(3):300–306. 17. Akeson P, Nordstrom CH, Holtas S. Time-dependency in brain lesion enhancement with gadodiamide injection. Acta Radiol 1997;38(1):19–24. 18. Finelli DA, Hurst GC, Gullapali RP et al.. Improved contrast of enhancing brain lesions on postgadolinium, T1-weighted spin-echo images with use of magnetization transfer. Radiology 1994;190(2):553–559. 19. Knauth M, Forsting M, Hartmann M et al. Enhancement of brain lesions: increased contrast dose compared with magnetization transfer. AJNR Am J Neuroradiol 1996;17(10):1853–1859. 20. Colosimo C, Ruscalleda J, Korves M et al. Detection of intracranial metastases: a multicenter, intrapatient comparison of gadobenate dimeglumine-enhanced MRI with routinely used contrast agents at equal dosage. Invest Radiol 2001;36(2):72–81. 21. Runge VM, Parker JR, Donovan M. Double-blind, efficacy evaluation of gadobenate dimeglumine, a gadolinium chelate with enhanced relaxivity, in malignant lesions of the brain. Invest Radiol 2002;37(5):269–280. 22. Ba–Ssalamah A, Nobauer–Huhmann IM, Pinker K et al. Effect of contrast dose and field strength in the magnetic resonance detection of brain metastases. Invest Radiol 2003;38(7):415–422. 23. Nobauer–Huhmann IM, Ba–Ssalamah A, Mlynarik V et al. Magnetic resonance imaging contrast enhancement of brain tumors at 3 tesla versus 1.5 tesla. Invest Radiol 2002;37(3):114–119. 24. Bruhn H, Frahm J, Gyngell ML et al. Noninvasive differentiation of tumors with use of localized H-1 MR spectroscopy in vivo: initial experience in patients with cerebral tumors. Radiology 1989;172(2):541–548. 25. Kwock L, Smith JK, Castillo M et al. Clinical applications of proton MR spectroscopy in oncology. Technol Cancer Res Treat 2002;1(1):17–28. 26. Poptani H, Gupta RK, Roy R et al. Characterization of intracranial mass lesions with in vivo proton MR spectroscopy. AJNR Am J Neuroradiol 1995;16(8):1593–1603. 27. Castillo M, Kwock L, Mukherji SK. Clinical applications of proton MR spectroscopy. AJNR Am J Neuroradiol 1996;17(1):1–15. 28. Sijens PE, Knopp MV, Brunetti A et al. 1H MR spectroscopy in patients with metastatic brain tumors: a multicenter study. Magn Reson Med 1995;33(6):818–826. 29. Bulakbasi N, Kocaoglu M, Ors F et al. Combination of single-voxel proton MR spectroscopy and apparent diffusion coefficient calculation in the evaluation of common brain tumors. AJNR Am J Neuroradiol 2003;24(2):225–233. 30. Ishimaru H, Morikawa M, Iwanaga S et al. Differentiation between high-grade glioma and metastatic brain tumor using single-voxel proton MR spectroscopy. Eur Radiol 2001;11(9):1784–1791. 31. Law M. MR spectroscopy of brain tumors. Top Magn Reson Imaging 2004;15(5):291–313. 32. Bukte Y, Paksoy Y, Genc E et al. Role of diffusion-weighted MR in differential diagnosis of intracranial cystic lesions. Clin Radiol 2005;60(3):375–383.

30

Part II / Diagnostic Studies

33. Lu S, Ahn D, Johnson G, Cha S. Peritumoral diffusion tensor imaging of high-grade gliomas and metastatic brain tumors. AJNR Am J Neuroradiol 2003;24(5):937–941. 34. Tsuchiya K, Fujikawa A, Nakajima M et al. Differentiation between solitary brain metastasis and high-grade glioma by diffusion tensor imaging. Br J Radiol 2005;78(930):533–537. 35. Lu S, Ahn D, Johnson G et al. Diffusion tensor MR imaging of intracranial neoplasia and associated peritumoral edema: introduction of the tumor infiltration index. Radiology 2004;232(1):221–228. 36. Wiegell MR, Henson JW, Tuch DS et al. Diffusion tensor imaging shows potential to differentiate infiltrating from noninfiltrating tumors. Proceedings of the International Society for Magnetic Resonance in Medicine (ISMRM) 11th Scientific Meeting 2003:2075. 37. Groothuis DR. The blood–brain and blood–tumor barriers: a review of strategies for increasing drug delivery. Neuro-oncol 2000;2(1):45–59. 38. Law M, Cha S, Knopp EA et al. High-grade gliomas and solitary metastases: differentiation by using perfusion and proton spectroscopic MR imaging. Radiology 2002;222(3):715–721. 39. Bertossi M, Virgintino D, Maiorano E et al. Ultrastructural and morphometric investigation of human brain capillaries in normal and peritumoral tissues. Ultrastruct Pathol 1997;21(1):41–49. 40. Cha S. Perfusion MR imaging of brain tumors. Top Magn Reson Imaging 2004;15(5):279–289. 41. Dierckx RA, Martin JJ, Dobbeleir A et al. Sensitivity and specificity of thallium-201 single-photon emission tomography in the functional detection and differential diagnosis of brain tumours. Eur J Nucl Med 1994;21(7):621–633. 42. Kojima Y, Kuwana N, Noji M et al. Differentiation of malignant glioma and metastatic brain tumor by thallium-201 single-photon emission-computed tomography. Neurol Med Chir (Tokyo) 1994;34(9):588–592. 43. Griffeth LK, Rich KM, Dehdashti F et al. Brain metastases from noncentral nervous system tumors: evaluation with PET. Radiology 1993;186(1):37–44. 44. Rohren EM, Provenzale JM, Barboriak DP. Screening for cerebral metastases with FDG PET in patients undergoing whole–body staging of noncentral nervous system malignancy. Radiology 2003;226(1):181–187. 45. Chao ST, Suh JH, Raja S et al. The sensitivity and specificity of FDG PET in distinguishing recurrent brain tumor from radionecrosis in patients treated with stereotactic radiosurgery. Int J Cancer 2001;96(3):191–197. 46. Mavrakis AN, Halperin EF, Barker FG et al. Diagnostic evaluation of a brain mass as the presenting manifestation of cancer. Neurology 2005;65:908–911. 47. Agazzi S, Pampallona S, Pica A et al. The origin of brain metastases in patients with an undiagnosed primary tumour. Acta Neurochir (Wien) 2004;146(2):153–157. 48. Altorki N, Kent M, Pasmantier M. Detection of early-stage lung cancer: computed tomographic scan or chest radiograph? J Thorac Cardiovasc Surg 2001;121(6):1053–1057. 49. Chamberlain MC, Kormanik PA. Prognostic significance of 111indium–DTPA CSF flow studies in leptomeningeal metastases. Neurology 1996;46(6):1674–1677. 50. Schiff D, O’Neill BP, Suman VJ. Spinal epidural metastasis as the initial manifestation of malignancy: clinical features and diagnostic approach. Neurology 1997:49:452–456. 51. Posner J. Side effects of radiation therapy. In: Posner J, ed. Neurologic Complications of Cancer. Philadelphia: FA Davis; 1995. 52. Gultekin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurologic symptoms, immunological findings and tumour association in 50 patients. Brain 2000;123 (Pt 7):1481–1494.

III

Neurologic Symptoms

3

Seizures and Anti-Epileptic Drugs in Neuro-Oncology Michael J. Glantz,

MD,

and Julia Batten,

APRN, MPH

CONTENTS Introduction Epidemiology of Seizures in Patients with Brain Tumors Clinical Features of Tumor-Associated Seizures Evaluation of Seizures in Patients with Known Cancer Treatment of Tumor-Associated Seizures: Drug Therapy Treatment of Tumor-Associated Seizures: Radiation and Surgery Outcome in Tumor-Associated Epilepsy Anticonvulsant Prophylaxis in Patients with Brain Tumors Special Issues in Patients with Tumor-Associated Seizures References

Summary Seizures are common in patients with central nervous system cancer, although the exact frequency of seizures depends considerably on the location, growth rate, and histology of the tumor. The occurrence of a first seizure in an adult mandates a neuroimaging study of the brain, preferably an enhanced MRI scan with FLAIR sequences. Focal seizures in children, particularly in the presence of post-ictal or intra-ictal deficits, may also merit a similar evaluation. In patients with known cancer, an MRI scan with FLAIR images and contrast enhancement is the optimum neuroimaging test. Extensive studies to evaluate infectious, metabolic, and drug-related etiologies are also critical, and in many cases a cerebrospinal fluid examination is indicated. While surgical therapy of tumorassociated seizures holds promise, particularly for patients with low-grade primary brain tumors, intraoperative electrocorticography may be required for optimum seizure control, and post-operative anticonvulsant medication is usually necessary. Anticonvulsant medications are associated with more frequent and often more severe side effects in patients with cancer than in patients with other causes for their seizures. These side effects include important interactions with corticosteroids and chemotherapeutic agents, and should influence the choice of anticonvulsant agents and the monitoring schedule of patients under treatment. Prophylactic anticonvulsants are not effective, and should not be used routinely. Key Words: seizure, epilepsy, anticonvulsant therapy, brain tumors, brain metastases, anticonvulsant prophylaxis

1. INTRODUCTION Despite an increasing incidence and a corresponding increase in basic and clinical research, the overall survival of patients with primary or metastatic central nervous system cancer has not changed substantially over the last 20 years (1,2). In these diseases, where cures are rare and symptom management is often the most vital role the From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

33

34

Part III / Neurologic Symptoms

physician can play, treatment of seizures assumes major importance. Seizures are common in patients with brain tumors, and a single seizure, or even the fear of a seizure, can profoundly impair quality of life. In addition to the potential for injury from seizures, seizures can lead to the forfeiture of driving privileges and employment. Seizures and their sequelae can mimic tumor progression and prompt unwarranted diagnostic interventions. Anticonvulsant therapy and the monitoring which therapy necessitates can be inconvenient and uncomfortable, can produce side effects that mimic or exacerbate disease-related symptoms and reduce the effectiveness of therapy for the underlying tumor. For all of these reasons, an understanding of the epidemiology, clinical manifestations, and available treatments for patients with brain tumors and seizures is essential.

2. EPIDEMIOLOGY OF SEIZURES IN PATIENTS WITH BRAIN TUMORS In 2005, more than 43,800 patients were diagnosed with primary brain tumors, and more than 170,000 developed brain metastases (1–5). Many of these patients will develop seizures. Those seizures typically occur early in the clinical course (Table 1), and are often a presenting feature of central nervous system cancer (2,6–22). Overall, 20–40% of adults with primary brain tumors experience at least one seizure prior to the diagnosis of their tumor, and another 20–45% will develop seizures at some point following diagnosis. Both numbers vary appreciably however, depending on the histology of the underlying tumor (Table 1, Fig. 1). In general, slowgrowing, histologically low-grade primary brain tumors are associated with the highest incidence of seizures, and this incidence falls with increasing grade (6,22–31). As a result, 50–80% of children with supratentorial tumors ultimately develop seizures (32,33). Similarly, tumors, typically low grade, are discovered in roughly one-third of patients undergoing temporal lobe surgery for medically intractable epilepsy (34–38), and in 28% of those patients undergoing surgery for seizures with extratemporal foci (39). In contrast, approximately 20% of patients with brain metastases present with seizures, and an additional 20% develop seizures later in their course. Multiple metastases and hemorrhagic metastases may be associated with a higher incidence of seizures. Patients Table 1 Frequency of Seizures by Tumor Histology Incidence of Seizures Tumor Histology Glioblastoma multiforme Anaplastic astrocytoma Astrocytoma Oligodendroglioma Meningioma Medulloblastoma Brain metastasis Neoplastic meningitis

At Presentation

Overall

30 − 40% 40 − 50% 60 − 95% 75% 30 − 40% 1 − 2% 20% 14%

40–60% 60–70% — 85% 50–60% 2–3% 25–40% 19%

Fig. 1. Overall frequency of seizures related to tumor growth rate. Astro-astrocytoma, Oligo-oligodendroglioma, AA-anaplastic astrocytoma, GBM-glioblastoma, PCNSL-primary CNS lymphoma.

Chapter 3 / Seizures and Anti-Epileptic Drugs in Neuro-Oncology

35

Fig. 2. Frequency of tumor as the etiology of first seizures (by age).

Table 2 Frequency of Different Tumor Histologies as the Cause of First Seizures at Different Ages Ag e at Fi rst Sei zu re Occurrenc e 0 – 24 years > 65 years 25 – 44 years Astrocytoma

Ma li g n an t g l i o m a

Metastasis

Oligode ndrog lio ma

M et a s t a s i s

Maligna nt g lioma

M ali g n a n t g l i om a

Meningioma

M eningi oma

DNET1

O l i g o d e n d r o g l i o m a C NS ly mp hom a

Increasing Frequency

1. Dysembryoplastic neuroepithelial tumor

with neoplastic meningitis alone present with seizures in 15% of cases, and patients with both parenchymal brain metastases and neoplastic meningitis may be especially likely to experience seizures (40–42). Regardless of tumor or patient characteristics, seizures are much more common with supratentorial than with infratentorial tumors, and among supratentorial tumors, much more common with superficial and cortical lesions (63% of cases) than with tumors located within the basal ganglia or entirely within the white matter (29% of cases) (9,22,25,31,43). For cortical tumors, seizure frequency increases with increasing proximity to the Rolandic fissure, and early onset seizures are more frequent with increasing proximity to the central sulcus (44,45). In contrast, tumors located in the occipital lobes are considerably less likely to be associated with seizures. From a diagnostic standpoint, tumors are an increasingly common cause of seizures with increasing age (Fig. 2), and often provide the first indication of an underlying tumor (Table 2). The type of tumor most likely to underlie new onset seizures varies according to patient age. The likelihood of a tumor etiology for new onset seizures also increases when the seizures are complex-partial, and when abnormalities on the neurologic examination are present (45–47). Thus neuroimaging studies are rarely abnormal and in many centers are not routinely performed in children with clinically typical primary generalized or febrile seizures (48). In contrast, new onset seizures in adults are more frequently associated with intracranial tumors (1.3–16% of cases) and always necessitate a brain imaging study, especially when the seizure itself or the post-ictal examination suggests a focal onset, or when the neurologic examination is abnormal (49–54). The inclusion of fluid attenuated inversion recovery (FLAIR) sequences and gadolinium enhancement significantly enhances the diagnostic yield (55–57). Even in patients with normal neurologic examinations and very longstanding epilepsy, low-grade gliomas are being identified with increasing frequency as the routine use of MR scanning becomes more widespread (58,59).

3. CLINICAL FEATURES OF TUMOR-ASSOCIATED SEIZURES As with other symptomatic seizures, the ictal and post-ictal characteristics of tumor-associated seizures depend on the location of the seizure focus and are rarely “false localizing” (60). Perhaps reflecting the propensity of malignant gliomas for the temporal lobes, several authors have emphasized the frequent association of olfactory

36

Part III / Neurologic Symptoms

or gustatory hallucinations with tumor (61,62). Increasingly frequent seizures, and variability in seizure phenotype between attacks, may also be more characteristic of tumor-related seizures than of other seizure types. When focal seizures occur in the setting of an underlying brain tumor, associated focal neurologic findings are also generally present (11,49–54). Conversely, prolonged or permanent focal deficits can occur in patients with brain tumors following seizures of focal onset (22,63). Similarly, prolonged or irreversible cognitive decline may occur following generalized seizures in brain tumor patients. Seizure-induced increases in intracranial pressure, disruption of the blood–brain barrier, and elevations in excitotoxic neurotransmitter levels superimposed upon pre-existing abnormalities in intracranial pressure, blood flow, and the peritumoral microenvironment may provide a physiologic explanation for these observations (64,65). Table 3 Differential Diagnosis of Seizures in Patients with Cancer Etiology Infection

Comment Meningitis Brain abscess Sepsis Encephalitis

Metabolic

Hypomagnesemia Hyponatremia

Hypocalcemia Drug-induced

Cisplatin Vincristine, Etoposide Busulfan, Methotrexate IL-2, Interferon- Ifosfamide Thienamycin antibiotics (imipenem and others)

Listeria is common; ventricular reservoirs or shunts predispose Usually in the setting of high fever, hypoxia, or hypotension Particularly HSV, CMV, and progressive multifocal leukoencephalopathy (JC virus) Common with cisplatin use, often 3–8 days after therapy; also after biphosphonate administration Common following neurosurgery; also vincristine, carbamazepine, oxcarbazepine, SIADH, cyclophosphamide, or chemotherapy requiring co-administration of large fluid volumes Following cisplatinum or biphosphonate administration Possibly related to electrolyte disturbances; may be delayed up to 2 weeks after completion of treatment Uncommon Particularly during high-dose therapy Especially in patients with renal failure or hypoalbuminemia Up to 6% of all patients

Radiation-related

Radiation necrosis (67)

MRI may be indistinguishable from recurrent tumor

Intracranial hemorrhage

Thrombocytopenia Coagulopathy

Disease-, chemotherapy-, or drug- (e.g.. heparin) related Disseminated intravascular coagulation

Tumor-Related

New or progressive disease

Including neoplastic meningitis and brain metastases

Paraneoplastic

Encephalitis (68,157)

Increasingly recognized; most common with small cell lung cancer and ovarian teratoma

Other

Posterior Reversible Leukoencephalopathy Syndrome (158)

Seen in multiple settings, including bone marrow transplant, immunosuppressive therapy with cyclosporin, tacrolimus, interferon- and others. Abrupt increases in blood pressure may predispose.

Chapter 3 / Seizures and Anti-Epileptic Drugs in Neuro-Oncology

37

4. EVALUATION OF SEIZURES IN PATIENTS WITH KNOWN CANCER The differential diagnosis of seizures occurring in patients with known cancer is considerably different in many respects from that of patients without an underlying malignancy (Table 3). Although seizures may assume a natural history independent of the underlying tumor, new seizures or a change in the frequency or phenotype of seizures in patients with known central nervous system cancer often herald tumor progression or recurrence (6). Therefore, a neuroimaging study of the brain, preferably an enhanced MR scan with FLAIR sequences, is mandatory. If the MR scan provides insufficient explanation for the new seizure pattern, a lumbar puncture with a large volume cytologic examination is appropriate (66). First or recurrent seizures in patients with cancer but without known central nervous system involvement require an identical evaluation: enhanced MRI of the brain and cerebrospinal fluid evaluation. Even when an apparent neuroradiographic or cytologic explanation for the seizure is identified, a more extensive evaluation is still required for both groups of patients, as suggested by Table 3. Screening for metabolic derangements, infection, and potentially offending drugs (including missed or inadequate doses of prescribed anticonvulsants) is particularly important.

5. TREATMENT OF TUMOR-ASSOCIATED SEIZURES: DRUG THERAPY The potentially irreversible neurologic decline sometimes seen in patients with central nervous system cancer and seizures, in addition to the impairment in quality of life that seizures can produce in any patient, constitute a strong rationale for the aggressive treatment of seizures when they occur. The basic principles of seizure therapy, including optimum monotherapy, meticulous clinical monitoring for evidence of drug-related toxicity, and detailed questioning to detect the occurrence of subtle seizures, are the same for patients with brain tumors as for those with idiopathic or other seizure types. Although there is no unequivocal evidence suggesting the superior efficacy of any particular anticonvulsant in the setting of tumor-associated seizures, some reasonable guidelines are possible. Anticonvulsant side effects, including sedation, cognitive impairment, psychomotor slowing, cutaneous reactions, hepatic toxicity, and bone marrow suppression are more common in patients with brain tumors than in those without nervous system cancer, and occur at lower drug doses (15,69–77). Symptoms severe enough to warrant discontinuation of medication occur in nearly 25% of patients (range: 5–38%) (8,9,11,12,14,20,70). In addition, the drugspecific side effects of certain anticonvulsants make those agents less preferable choices for seizure control in patients with cancer. For example, the language dysfunction seen in 10% of patients receiving topirimate (78); “shoulder–hand syndrome” which occurs in nearly 20% of brain tumor patients receiving phenobarbital (79); and the encephalopathy and parkinsonian symptoms that develop rarely patients receiving valproic acid (80,81) may mimic the signs and symptoms of tumor progression, or exacerbate tumor-related neurologic deficits. Rashes (occasionally progressing to Stevens–Johnson syndrome) occur in 14–25% of patients. Most cases have been reported with the use of carbamazepine, phenytoin, lamotrigine, and phenobarbital, although this may in part reflect longer experience with these agents. Cranial irradiation may be implicated; and although the precise mechanism is uncertain, and patients are frequently receiving decreasing doses of corticosteroids at the time the rash develops. The additional risks of immunosuppression with phenytoin and carbamazepine (82–85), leukopenia with carbamazepine, and thrombocytopenia with valproic acid are considerations whose clinical importance is uncertain. Perhaps the most concerning problem with the use of anticonvulsants in patients with cancer is the interaction of anticonvulsants metabolized by the cytochrome P450 system (“enzyme-inducing anticonvulsants”; Table 4) with corticosteroids (86–90) and with many common antineoplastic agents, including cisplatinum, taxanes, irinotecan, topotecan, vinca alkaloids, cyclophosphamide, methotrexate, doxorubicin, nitrosoureas, and many targeted agents such as erlotinib and bevacizumab (91–103). Because of these interactions, a given dose of dexamethasone may be less effective in a patient receiving an enzyme-inducing anticonvulsant, and the exposure of a patient’s tumor to an antineoplastic agent (“area under the curve”) is reduced, often dramatically (104). While the evidence remains inconclusive, some investigators have reported that the concurrent use of enzyme-inducing antiepileptic drugs and P450-metabolized chemotherapeutic agents may even decrease survival (105–107). Conversely, anticonvulsant levels are often reduced in the setting of concurrent chemotherapy despite seemingly adequate doses (108–117). In addition to P450-mediated interactions, antineoplastic agents may decrease anticonvulsant absorption by damaging

55% <5% 15–41%

Potent

Enzyme inhibitor No No

Modest No

Limited∗

Modest Modest

Phenobarbital

Valproic acid Gabapentin Pregabalin

Lamotrigine Levetiracetam

Topiramate

Tiagabine Zonisamide

7-9 > 60

21

25 6-8

9–16 5–7 6

100

12–17 8–10** 22

Half-life (hours)

* Induces catabolism of ethinyl estradiol but inhibits metabolism of phenytoin. ** For the major active (10-monohydroxy) metabolite (MHD). *** Both higher and lower doses are commonly used based on clinical response.

96% 40–60%

90% <5% <5%

40–60%

76% 40%** 90%

Potent Modest Potent

Carbamazepine Oxcarbazepine Phenytoin

Protein Binding

Enzyme Inducer

Agent

NMDA inhibition, blocks voltage-gated sodium channels, GABA inhibition GABA augmentation Voltage-gated sodium and calcium channel blockade; carbonic anhydrase inhibition

Blocks voltage-gated calcium channels GABA analogue GABA analogue; binding to a2 -d subunit of voltage-gated calcium channel Blocks voltage-gated sodium channels Synaptic vesicle protein binding

GABA inhibition

Blocks voltage-gated sodium channels Blocks voltage-gated sodium channels Blocks voltage-gated sodium channels

Proposed Mechanisms of Antiepileptic Action

Typical Effective Dose and Schedule∗∗∗

32–56 mg daily, divided bid to qid 100–400 mg, given once or twice daily

200–600 mg, divided big 1000–4000 mg daily, divided bid or tid 200–600 mg given once or twice daily

800–2400 mg, divided bid or tid 1200–3000 mg divided bid 300–600 mg, given once or twice daily 90–150 mg, given once or twice daily 1000–3000 mg, divided bid or tid 1200–3600 mg, divided bid or tid 150–600 mg/d, divided bid or tid

Table 4 Commonly Used Enzyme-Inducing and Nonenzyme-Inducing Anticonvulsants

Chapter 3 / Seizures and Anti-Epileptic Drugs in Neuro-Oncology

39

the mucosal lining of the gastrointestinal tract, and can displace anticonvulsants from plasma proteins, transiently increasing free drug concentrations, but ultimately decreasing the steady-state concentration due to increased anticonvulsant clearance (Table 4) (35,39,40,118). As a result, more frequent and higher drug doses, and more frequent monitoring of drug levels, are often necessary, and the risks of under- and over-dosing, with consequent seizures or toxicity, are more common. To further complicate patient management, both the cognitive side effects of overdosing and seizures resulting from underdosing may be interpreted as evidence of tumor progression, leading to unnecessary diagnostic or therapeutic interventions. For all of these reasons, some of the newer, nonenzyme-inducing anticonvulsants (e.g., gabapentin, pregabalin, levetiracetam, and zonisamide) (119–121) may prove to be the most attractive agents for patients with cancer, although limited experience to date precludes firm recommendations.

6. TREATMENT OF TUMOR-ASSOCIATED SEIZURES: RADIATION AND SURGERY While no formal studies have been conducted, elimination of seizures or reduction in seizure frequency has been reported following cranial irradiation (122) and occasionally in patients whose tumors have responded completely or partially to chemotherapy. Similar benefits have been reported following stereotactic radiosurgery and interstitial radiation implants (123,124). Randomized, controlled trials are also lacking in the area of tumor surgery, but numerous case series suggest that gross total resection of tumors may lead to improved seizure control or elimination of seizures. Benefits are more common and more durable in patients with histologically low-grade tumors, in whom 65–80% become seizure-free, and approximately one-third of all patients can be weaned off anticonvulsant medication entirely (33,125–133). According to some (but not all (131)) authors, this number increases to more than 90% of all patients if an extensive resection including lobectomy and removal of medial temporal and frontal lobe structures is performed (126,127,134). The duration and number of preoperative seizures, epileptiform activity on the post-operative electroencephalogram, increased signal around the tumor resection cavity on the post-operative T2-weighted MRI scan, and the absence of a focal lesion identified on the pre-operative MRI scan may be predictors of poorer post-operative seizure outcome (125,128,131,135). Even patients with high-grade gliomas become seizure-free (on anticonvulsant medication) in 30–60% of cases following surgery (13,33,132,134). Similarly, in modern series more than 60% of patients with meningiomas are seizure-free post-operatively (136,137). There are several explanations for the failure of surgery to eliminate seizures in all patients. Incomplete tumor resection is the most obvious reason. The development of post-surgical seizure foci related to reactive changes and scarring, or to tumor recurrence, represent additional common explanations. Another possibility for the persistence of seizures despite a gross total tumor resection is that the offending epileptogenic focus or foci are frequently located at considerable distance from the tumor, in completely tumor-free areas of brain (138). In such patients, these tumor-free epileptogenic foci have been found to contain significantly lower concentrations of GABA and somatostatin neurons than in adjacent normal brain (139–141). This apparent tumor-induced deafferentiation and loss of inhibitory neurotransmitters in areas of brain remote from the tumor may account for many of the apparent failures to control seizures in patients undergoing gross total tumor resections. These observations have led some to advocate performing both a tumor resection and additional resections of seizure foci identified by intraoperative electrocorticography (35,142–145).

7. OUTCOME IN TUMOR-ASSOCIATED EPILEPSY The prognosis for complete seizure control in patients with tumor-associated epilepsy is relatively poor. The necessity of incomplete tumor resection in many patients, the presence of seizure foci remote from the identified tumor, difficulties in achieving and maintaining therapeutic anticonvulsant levels in patients with cancer due to multiple drug interactions and increased sensitivity to anticonvulsant side effects, and the multiple alternative causes of seizures in the cancer population all conspire against optimum seizure control. In addition, there is some evidence to suggest that the relative treatment-resistance of brain tumor-associated epilepsy is due to the presence of epileptogenic mechanisms in these patients that are not well targeted by currently available antiepileptic drugs. These mechanisms include peritumoral alterations in pH, cytokine and glutamate concentrations, localized immunological interactions (possibly involving glutamate decarboxylase), and the progressive nature of most

40

Part III / Neurologic Symptoms

brain tumors (119,146). As a consequence, overall more than 50% of patients with gliomas experience recurrent seizures during the course of their disease, while 11% of patients with brain metastases and 19% of patients with neoplastic meningitis suffer recurrent seizures during their substantially shorter lifetimes (6,22). Low-grade tumors may be more refractory than higher-grade lesions (6). Seizures that develop late in the clinical course are typically more responsive to treatment than those that occur early (9). Seizures that are initially controlled but return at the time of tumor recurrence are also relatively refractory to treatment (147). Importantly, anticonvulsant therapy does appear to substantially decrease the proportion of seizures that are generalized (6). In some studies, presentation with a seizure constitutes a favorable prognostic sign for survival in patients with gliomas (148–150). In part this observation reflects the overrepresentation of patients with lower-grade tumors in the population of glioma patients presenting with seizures. Even among patients with malignant gliomas, presentation with a seizure may constitute a favorable prognostic sign, perhaps because seizures typically result in more expeditious detection and treatment of the underlying tumor, and because the underlying tumor is more commonly cortically based and potentially more amenable to resection than a deep-seated lesion.

8. ANTICONVULSANT PROPHYLAXIS IN PATIENTS WITH BRAIN TUMORS Although definitive studies are lacking, anticonvulsant therapy in patients with tumor-associated epilepsy does appear to reduce frequency of recurrent seizures and the percentage of generalized seizures in some patients (6,119). Anticonvulsant prophylaxis (i.e., the use of anticonvulsants to prevent seizures in patients who have never had a seizure) constitutes an important issue about which there has been much debate and substantial variation in practice both between and within the subspecialties of neurology, neurosurgery, oncology, and radiation oncology (2,9,119,151). Because seizures are common in patients with brain tumors; because seizures add to the fear and limitations with which cancer patients already contend; and because of the risk of permanent neurologic injury following a seizure in a patient with a brain tumor, prevention of seizures is a desirable goal. If drugs that effectively prevented first seizures and were associated with a modest risk of side effects were available, the prophylactic use of anticonvulsants would be very attractive. Unfortunately, the potential side effects of anticonvulsants in cancer patients are neither uncommon nor insubstantial. Moreover, the consensus from four randomized, controlled clinical trials, multiple retrospective studies, and three independent meta-analyses is that anticonvulsant prophylaxis is not effective in preventing first seizures beyond the immediate (7-day) post-operative period (70,151,152). This conclusion is the same whether the underlying tumor is primary or metastatic, and regardless of tumor histology. Pretreatment patient characteristics such as age, performance status, number of brain metastases, brain surgery for the tumor, tumor location, or EEG features also have no influence on the success of prophylactic therapy. A number of issues remain unresolved, including the possibility of a very small prophylactic effect in certain subsets of patients, and the potential benefit of newer anticonvulsants. Disagreement also persists regarding the benefit of anticonvulsant prophylaxis in the immediate (within 7 days) post-operative period in patients undergoing surgery for their brain tumors, with some studies and one meta-analysis suggesting a benefit for phenytoin and oxcarbazepine (152,153) while other meta-analyses have found no advantage (70,154). Table 5 Physician Practices for Prescribing Prophylactic Anticonvulsants Before and After Publication of AAN Guidelines Study Glantz (1996)∗ Siomin (2005)∗∗∗

Number of Physicians

Subspecialty

Response Rate

Number (%) Prescribing Anticonvulsant Prophylaxis

115 299

Multiple Neurosurgeons

100% 12%

62 (53.9)∗∗ 288 (96.3)∗∗∗∗

* Study conducted prior to publication of the AAN Practice Parameter. ** Includes 21/42 (50%) of oncologists, 20/38 (53%) of neurologists, 17/21 (81%) of neurosurgeons, and 4/12 (33%) of radiation oncologists *** Study conducted five years after publication of the AAN Practice Parameter. **** Sixty-nine neurosurgeons (24%) initiated anticonvulsant therapy prior to craniotomy, but, in the absence of seizures, discontinued anticonvulsants one week following surgery, in accordance with AAN guidelines.

Chapter 3 / Seizures and Anti-Epileptic Drugs in Neuro-Oncology

41

Table 6 Use of Prophylactic Anticonvulsants in Patients with Primary and Metastatic Brain Tumors

Study Chang (2005) Hildebrand (2005) Mehta (2003) Total

Number of Patients

Number (%) Presenting with a Seizure

565 234 401 1200

180 (31.9) 158 (67.5) 68 (17) 406 (33.8)

Number (%) at Risk for a Seizure 354 51 327 732

(62.7)∗ (21.8)∗∗ (81.5)∗∗∗ (61)

Number (%) Receiving Anticonvulsant Prophylaxis 279 22 231 532

(78.8) (43.1) (70.6) (72.7)

* Excludes 31 patients with missing data. ** Excludes 25 patients with subsequent seizures for whom timing of anticonvulsant administration is not available. *** Excludes 6 patients with missing data.

In summary, the currently available evidence is sufficient to support a clear recommendation against the routine use of prophylactic anticonvulsants in patients with brain tumors beyond the immediate (one week) postoperative period. Despite the preponderance of evidence supporting this approach, and the publication of a formal practice guideline by the American Academy of Neurology (151) the percentage of physicians recommending long-term anticonvulsant prophylaxis, and the percentage of patients actually receiving long term anticonvulsant prophylaxis does not appear to have decreased in the six years since publication of the guideline (Tables 5 and 6) (6,7,9,155,156).

9. SPECIAL ISSUES IN PATIENTS WITH TUMOR-ASSOCIATED SEIZURES The issue of anticonvulsant withdrawal arises frequently in patients with brain tumors who have been placed on an anticonvulsant at the time of diagnosis or prior to craniotomy, but who have never experienced a seizure. Once again, there have been no formal studies to guide clinical practice. Despite compelling evidence against the use of anticonvulsants in this setting, many clinicians are reluctant to withdraw anticonvulsants from stable patients. While this reluctance is understandable, our own practice has been to carefully taper and ultimately discontinue anticonvulsants after the immediate post-operative period in any patient who is receiving truly prophylactic anticonvulsants and has had drug-related side effects. We define “side effects” broadly and include subtle cognitive deficits, fatigue, anorexia or other gastrointestinal symptoms, and the risk of interactions with corticosteroids or chemotherapeutic agents when taking enzyme-inducing anticonvulsants. This approach requires an exhaustive clinical evaluation to exclude the possibility of subtle seizures, which are frequently unreported by patients and unrecognized by their physicians. When even a single, very brief, partial seizure has occurred, we believe that prolonged if not life-long anticonvulsant therapy is necessary except in the exceptional circumstances of tumor cure or, possibly, complete resection of a low-grade tumor and associated epileptogenic foci. This approach also requires a careful review of the potential risks and benefits of anticonvulsant withdrawal with the patient and patient’s caregivers in light of current knowledge. A second common concern in patients with tumor-associated epilepsy is whether driving is permissible. Frequently the greatest barrier to driving in patients with central nervous system cancer is not the potential for seizures, but the neurologic and non-neurologic deficits associated with their disease and its treatment. When seizures are the sole barrier to operating a motor vehicle, decisions must be guided by state-specific legal statutes, and the specific features of each patient’s clinical situation. In patients who have never had a seizure, however, anticonvulsant prophylaxis does not reduce the risk of first seizures, and should not be a requirement for driving.

REFERENCES 1. Prados MD, Berger MS, Wilson CB. Primary central nervous system tumors: Advances in knowledge and treatment. CA Cancer J Clin 1998;48:331–360. 2. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis Company, 1995. 3. CBTRUS (2005). Statistical report: primary brain tumors in the United States, 1998–2002. Published by the Central Brain Tumor Registry of the United States, 2006.

42

Part III / Neurologic Symptoms

4. Barnholtz-Sloan JS, Sloan AE, Davis FG et al. Incidence proportions of brain metastases in patients diagnosed (1973–2001) in the metropolitan Detroit cancer surveillance system. J Clin Oncol 2004;22:2865–2872. 5. Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43–66. 6. Hildebrand J, Lecaille C, Perennes J et al. Epileptic seizures during follow-up of patients treated for primary brain tumors. Neurology 2005;65:212–215. 7. Chang SM, Parney IF, Huang W et al. Patterns of care for adults with newly diagnosed malignant glioma. JAMA 2005;293:557–564. 8. Forsyth PA, Weaver S, Fulton D et al. Prophylactic anticonvulsants in patients with brain tumour. Can J Neurol Sci 2003;30:89–90. 9. Glantz MJ, Cole BF, Friedberg MH et al. A randomized, blinded, placebo-controlled trial of divalproex sodium prophylaxis in adults with newly diagnosed brain tumors. Neurology 1996;46:985–991. 10. Dent S, Bociek G. Prophylactic anticonvulsants for cancer patients with newly diagnosed brain metastases. Proc Am Soc Clin Oncol 1996;15:529, 1996 [abstract]. 11. Moots PL, Maciunas RJ, Eisert DR et al. The course of seizure disorders in patients with malignant gliomas. Arch Neurol 1995;52: 717–724. 12. Hung S, Hilsenbeck S, Feun L. Seizure prophylaxis with phenytoin in patients with brain metastases. Proc Am Soc Clin Oncol 10:A1151, 1991 [abstract]. 13. Franceschetti S, Binelli S, Casazza M et al. Influence of surgery and antiepileptic drugs on seizures symptomatic of cerebral tumours. Acta Neurochirurgica 1990;103:47–51. 14. Hagen NA, Cirrincione C, Thaler HT et al. The role of radiation therapy following resection of single brain metastasis from melanoma. Neurology 1990;40:158–160. 15. Cohen N, Strauss G, Lew R et al. Should prophylactic anticonvulsant be administered to patients with newly diagnosed cerebral metastases? A retrospective analysis. J Clin Oncol 1988;6:1621–1624. 16. Gilbert MR, Grossman SA. Incidence and nature of neurologic problems in patients with solid tumors. Am J Med 1986;81:951–954. 17. Boarini DJ, Beck DW, VanGilder JC. Post-operative prophylactic anticonvulsant therapy in cerebral gliomas. Neurosurgery 1985;16:290–292. 18. North JB, Penhall RK, Hanieh A et al. Phenytoin and postoperative epilepsy: a double-blind study. J Neurosurg 1983;58:672–677. 19. Byrne TN, Cascino TL, Posner JB. Brain metastasis from melanoma. J Neurooncol 1983;1:313–317. 20. Mahaley MS, Dudka L. The role of anticonvulsant medications in the management of patients with anaplastic gliomas. Surg Neurol 1981;16:399–401. 21. Zimm S, Wampler GL, Stablein D et al. Intracerebral metastases in solid-tumor patients: natural history and results of treatment. Cancer 1981;48:384–394. 22. Cobb JL, Lekos A, Cole BF et al. Natural history of seizures and the role of anticonvulsants in patients with primary and metastatic central nervous system cancer. Neurology 1997;48:A19 [abstract]. 23. Olson JD, Riedel E, DeAngelis LM. Long-term outcome of low-grade oligodendroglioma and mixed glioma. Neurology. 2000;54: 1442–1448. 24. Daumas-Duport C. Dysembryoplastic neuroepithelial tumours. Brain Pathol 1993;3:283–295. 25. Gilles FH, Sobell E, Leviton A et al. Epidemiology of seizures in children with brain tumors. J Neurooncol 1992;12:53–68. 26. Daumas-Duport C, Scheithauer BW, Chodkiewicz JP et al. Dysembryoplastic neuroepithelial tumor: a surgically curable tumor of young patents with intractable partial seizures: report of thirty-nine cases. Neurosurgery 1988;23:545–556. 27. Mathieson G. Pathologic aspects of epilepsy with special reference to the surgical pathology of focal cerebral seizures. Adv Neurol 1975;8:107–138. 28. Rasmussen T. Cortical resection in the treatment of focal epilepsy. Adv Neurol 1975;8:139–154. 29. Rasmussen T. Surgery of epilepsy associated with brain tumors. Adv Neurol 1975;8:227–239. 30. Backus RE, Millichap JG. The seizure as a manifestation of intracranial tumor in childhood. Pediatrics 1962;29:978. 31. Penfield W, Erickson TC, Tarlov I. Relation of intracranial tumors and symptomatic epilepsy. Arch Neurol Psychiatr 1940;44:300–315. 32. Iannelli A, Guzzetta F, Battaglia D et al. Surgical treatment of temporal tumors associated with epilepsy in children. Pediatr Neurosurg 2000 32:248–254. 33. Hirsch JF, Saint Rose C, Pierre-Kahn A et al. Benign astrocytic and oligodendrocytic tumors of the cerebral hemispheres in children. J Neurosurg 1989;70:568–572. 34. Janszky J, Jokeit H, Schulz R et al. EEG predicts surgical outcome in lesional frontal lobe epilepsy. Neurology 2000;54:1470–1476. 35. Diehl B, Luders HO. Temporal lobe epilepsy: when are invasive recordings needed? Epilepsia 2000;41 Suppl 3:S61–S74. 36. Zentner J, Hufnagel A, Wolf HK et al. Surgical treatment of temporal lobe epilepsy: clinical, radiological, and histopathological findings in 178 patients. J Neurol Neurosurg Psychiatr 1995;58:666–673. 37. Plate KH, Wieser HG, Yasargil MG et al. Neuropathologic finding in 224 patients with temporal lobe epilepsy. Acta Neuropathol 1993;86:433–438. 38. Wolf HK, Campos MG, Zentner J et al. Surgical pathology of temporal lobe epilepsy.: experience with 216 cases. J Neuropathol Exp Neurol 1993;52:499–506. 39. Frater JL, Prayson RA, Morris III HH et al. Surgical pathologic findings of extratemporal-based intractable epilepsy: a study of 133 consecutive resections. Arch Pathol Lab Med 2000;124:545–549. 40. Jacobs M, Phuphanich S. Seizures in brain metastasis and meningeal carcinomatosis. Proc Am Soc Clin Oncol 1990;9:96 [abstract]. 41. Olson ME, Chernik NL, Posner JB. Infiltration of the leptomeninges by systemic cancer: a clinical and pathologic study. Arch Neurol 1974;30:122–137. 42. Wasserstrom WR, Schwartz MK, Fleisher M et al. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49:759–772.

Chapter 3 / Seizures and Anti-Epileptic Drugs in Neuro-Oncology

43

43. Smith DF, Hutton JL, Sandermann D et al. The prognosis of primary intracerebral tumours presenting with epilepsy: the outcome of medical and surgical management. J Neurol Neurosurg Psychiatr 1991;54:915–920. 44. Penfield W, Jasper H. Epilepsy and the Functional Anatomy of the Human Brain. Little, Brown & Company, Boston, MA, 1954. 45. White JC, Liu CT, Mixter WJ. Focal epilepsy: a statistical study of its causes and the results of surgical treatment. I. Epilepsy secondary to intracranial tumors. N Engl J Med 1948;438:891–899. 46. Suarez JC, Sfaello ZM, Guerrero A et al. Epilepsy and brain tumors in infancy and adolescence. Child’s Nerv Syst 1986;2:169–174. 47. Blume WT, Girvin JP, Kaufmann JC. Childhood brain tumors presenting as chronic uncontrolled seizure disorders. Ann Neurol 1982;12:538–541. 48. Hirtz D, Ashwal S, Berg A et al. Practice parameter: evaluating a first nonfebrile seizure in children: report of the quality standards subcommittee of the American Academy of Neurology, The Child Neurology Society, and The American Epilepsy Society. Neurology. 2000;55:616–23. 49. Commission on Neuroimaging of the International League Against Epilepsy. Recommendations for neuroimaging of patients with epilepsy. Epilepsia 1997;38:1255–1256. 50. Schoenenberger RA, Heim SM. Indication for computed tomography of the brain in patients with first uncomplicated generalized seizure. BMJ 1994;309:986–989. 51. Hopkins A, Garman A, Clarke C. The first seizure in adult life: value of clinical features, electroencephalography, and computerized tomgraphic scanning in prediction of seizure recurrence. Lancet 1988;1:721–726. 52. Daras M, Tuchman AJ, Strobos RJ. Computed tomography in adult-onset epileptic seizures in a city hospital population. Epilepsia. 1987 Sep–Oct;28:519–522. 53. Ramirez-Lassepas M, Cipolle RJ, Morillo LR et al. Value of computed tomographic scan in the evaluation of adult patients after their first seizure. Ann Neurol 1984;15:536–543. 54. Russo LS, Goldstein KH. The diagnostic assessment of single seizures: is cranial computed tomography necessary? Arch Neurol 1983;40:744–746. 55. Bradley WG, Shey RB. MR imaging evaluation of seizures. Radiology 2000;214:651–656. 56. Taillibert S, Oppenheim C, Baulac M et al. Yield of fluid-attenuated inversion recovery in drug-resistant focal epilepsy with noninformative conventional magnetic resonance imaging. Eur Neurol 1999;41:64–72. 57. Bergin PS, Fish DR, Shorvon SD et al. Magnetic resonance imaging in partial epilepsy: additional abnormalities shown with the fluid attenuated inversion recovery (FLAIR) pulse sequence. J Neurol Neurosurg Psychiatry 1995;58:439–443. 58. Fried I, Kim JH, Spencer DD. Limbic and neocortical gliomas associated with intractable seizures: A distinct clinicopathological group. Neurosurgery 1994;34:815–824. 59. Recht LD, Smith TW, Lew RA. Suspected low-grade glioma: how safe is deferring treatment? Ann Neurol 1992;31:431–436. 60. Arseni C, Maretsis M. Focal epileptic seizures ipsilateral to the tumour. Acta Neurochir 1979;49:47–60. 61. Howe JG, Gibbon JD. Uncinate seizures and tumors, a myth reexamined. Ann Neurol. 1982;12:227. 62. Lund M. Epilepsy in association with intracranial tumour. Acta Psychiatr 1952;81(suppl). 63. Meyer JS, Portnoy HD. Post-epileptic paralysis.: a clinical and experimental study. Brain 1959;82(part II):162–185. 64. Cornford EM. Epilepsy and the blood–brain barrier: endothelial cell responses to seizures. Adv Neurol 1999;79:845–862. 65. Nitsch C, Klatzo J. Regional patterns of blood–brain barrier breakdown during epileptiform seizures induced by various convulsive agents. J Neurol Sci 1983;59:305–322. 66. Glantz MJ, Cole BF, Glantz LK et al. Cerebrospinal fluid cytology in patients with cancer: how to minimize false negative results. Cancer 1998;82:733–739. 67. Cross NE, Glantz MJ. Neurologic complications of radiation therapy. In: Wen PY, Glantz MJ, eds. Complications of Systemic Cancer. W.B. Saunders Co., Philadelphia, 2003, pp. 249–277. 68. Gultekin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain 2000;123 (Pt 7):1481–1494. 69. Lehmann DF, Hurteau TE, Newman N et al. Anticonvulsant usage is associated with an increased risk of procarbazine hypersensitivity reactions in patients with brain tumors. Clin Pharmacol Therapeutics 1997;62:225–229. 70. Sirven JI, Wingerchuk DM, Drazkowski JF et al. Seizure prophylaxis in patients with brain tumors: a metaanalysis. Mayo Clin Proc 2004;79:1489–1494. 71. Taphoorn MJ. Neurocognitive sequelae in the treatment of low-grade gliomas. Semin Oncol 2003;(6 suppl 19):45–48. 72. Ahmed I, Reichenberg J, Lucas A et al. Erythema multiforme associated with phenytoin and cranial radiation therapy: a report of three patients and review of the literature. Int J Dermatol 2004;43:67–73. 73. Cockey GH, Amann ST, Reents SB et al. Stevens–Johnson syndrome resulting from whole-brain radiation and phenytoin. Am J Clin Oncol 1996;19:32–34. 74. Borg MF, Probert JC, Zwi LJ. Is phenytoin contraindicated in patients receiving cranial irradiation? Australasian Radiol 1995;39:42–46. 75. Khe HX, Delattre JY, Poisson M. Stevens–Johnson syndrome in a patient receiving cranial irradiation and carbamazepine. Neurology 1990;40:1144–1145. 76. Taylor LP, Posner JB: Phenobarbital rheumatism in patients with brain tumor. Ann Neurol 1989;25:92–94. 77. Delattre J, Safai B, Posner JB. Erythema multiforme and Stevens–Johnson syndrome in patients receiving cranial irradiation and phenytoin. Neurology 38:194–198, 1988. 78. Crawford P. An audit of topiramate use in a general neurology clinic. Seizure 1998;7:207–211. 79. Taylor LP, Posner JB. Phenobarbital rheumatism in patients with brain tumor. Ann Neurol 1989;25:92–94. 80. Sasso E, Delsoldato S, Negrotti A et al. Reversible valproate-induced extrapyramidal disorders. Epilepsia 1994;35:391–393. 81. Sackellares JC, Lee SI, Dreifuss FE. Stupor following administration of valproic acid to patients receiving other antiepileptic drugs. Epilepsia 1979;20:697–703.

44

Part III / Neurologic Symptoms

82. Kikuchi K, McCormick CI, Neuwelt EA. Immunosuppression by phenytoin: implications for altered immune competence in brain tumor patients. J Neurosurg 61:1085–1090, 1984. 83. Bardana EJ, Gabourel JD, Davies GH et al: Effects of phenytoin on man’s immunity: evaluation of changes in serum immunoglobulins, complement, and anti-nuclear antibody. Am J Med 74:289–296, 1983. 84. Neuwelt EA, Kikuchi K, Hill S et al. Immune responses in patients with brain tumors. Factors such as anti-convulsants that may contribute to impaired cell-mediated immunity. Cancer 51:248–255, 1983. 85. Sorrell TC, Forbes IJ. Depression of immune competence by phenytoin and carbamazepine: studies in vivo and in vitro. Clin Exp Immunol 1975;20:273–285. 86. Dropcho EJ, Soong S. Steroid-induced weakness in patients with primary brain tumors. Neurology 41:1235–1239, 1991. 87. Chalk JB, Ridgeway K, Brophy T et al. Phenytoin impairs the bioavailability of dexamethasone in neurological and neurosurgical patients. J Neurol Neursurg Psychiatry 1984;47:187–190. 88. Gambertoglio JG, Holford NH, Kapusnik JE, et al. Disposition of total unbound prednisolone in renal transplant patients receiving anticonvulsants. Kidney Int 25:119–123, 1984. 89. Wassner SJ, Malekzadem MH, Pennisi AJ et al. Allograft survival in patients receiving anticonvulsant medications. Clin Nephrol 8:293–297, 1977. 90. Haque N, Thrasher K, Werk E. Studies on dexamethasone metabolism in man: effect of diphenylhydantoin. J Clin Endocrinol Metab 34:44–50, 1972. 91. Gilbert MR, Supko JG, Batchelor T et al. Phase I clinical and pharmacokinetic study of irinotecan in adults with recurrent malignant glioma. Clin Cancer Res 2003;9:2940–2949. 92. Chang SM, Kuhn JG, Rizzo J et al. Phase I study of paclitaxel in patients with recurrent malignant gliomas: a North American Brain Tumor Consortium report. J Clin Oncol 16:2188–2194, 1998. 93. Friedman AH, Ashley DM, Kerby T et al. Topotecan treatment of adults with primary malignant glioma. Proc Am Soc Clin Oncol 17:390a, 1998 [abstract]. 94. Grossman SA, Hochberg F, Fisher J et al. Increased 9-aminocamptothecin dose requirements in patients on anticonvulsants. Cancer Chemother Pharmacol 42:118–126, 1998. 95. Fetell MR, Grossman SA, Fisher JD et al. Preirradiation paclitaxel in glioblastoma multiforme: efficacy, pharmacology, and drug interactions. J Clin Oncol 15:3121–3128, 1997. 96. Chang TK, Chen G, Waxman DJ. Modulation of thiotepa antitumor activity in vivo by alteration of liver cytochrome P450-catalyzed drug metabolism. J Pharmacol Exp Therapeutics 274:270–275, 1995. 97. Baker DK, Relling MV, Pui C-H et al. Increased teniposide clearance with concomitant anticonvulsant therapy. J Clin Oncol 10:311–315, 1992. 98. Fitzsimmons WE, Ghalie R, Kaizer H. The effect of hepatic enzyme inducers on busulfan neurotoxicity. Cancer Chemother Pharmacol 27:226–228, 1990. 99. Workman P, Bleehen NM, Wiltshire CR. Phenytoin shortens the half-life of the hypoxic cell radiosensitizer misonidazole in man: implications for possible reduced toxicity. Br J Cancer 41:302–304, 1980. 100. Muller PJ, Tator CH, Bloom M. The effect of phenobarbital on the toxicity and tumoricidal activity of CCNU in a murine brain tumor model. J Neurosurg 52:359–366, 1980. 101. Levin VA, Stearns J, Byrd A et al. The effect of phenobarbital pretreatment on the antitumor activity of 1,3-bis(2-chloroethyl)-1nitrosourea (BCNU), 1-(2–chlorethyl)-3-cyclohexyl-1-nitrosourea (CCNU) and 1-(2-chlorethyl)-3-(2,6-dioxo-3-piperidyl-1-nitrosourea (PCNU), on the plasma pharmacokinetics and biotransformation of BCNU. J Pharmacol Exp Ther 208:1–6, 1979. 102. Warren RD, Bender AR. Drug interactions with antineoplastic agents. Cancer Treat Rep 61:1231–1241, 1977. 103. Reich SD, Bachur NR. Alterations in adriamycin efficacy by phenobarbital. Cancer Res 36:3803–3806, 1976. 104. Glantz MJ, Kim L, Choy H, Akerley W. Concurrent chemotherapy and radiotherapy in patients with brain tumors. Oncology 1999;13:78–82. 105. Oberndorfer S, Piribauer M, Marosi C et al. P450 enzyme inducing and nonenzyme inducing antiepileptics in glioblastoma patients treated with standard chemotherapy. J Neurooncol 2005;72:255–260. 106. Relling MV, Pui C–H, Sandlund JT et al. Adverse effect of anticonvulsants on efficacy of chemotherapy for acute lymphoblastic leukaemia. Lancet 2000;356:285–290. 107. Jaeckle KA, Ballman K, Uhm J et al. Comparison of survival endpoints in glioblastoma patients receiving or not receiving enzymeinducing anticonvulsants in NCCTG Trials. J Clin Oncol, 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition). Vol 22, No 14S (July 15 Supplement), 2004: 1525. 108. Gattis WA, May DB. Possible interaction involving phenytoin, dexamethasone, and antineoplastic agents: a case report and review. Ann Pharmacotherapy 30:520–526, 1996. 109. Rabinowicz AL, Hinton DR, Dyck P et al. High-dose tamoxifen in treatment of brain tumors: interaction with antiepileptic drugs. Epilepsia 36:513–515, 1995. 110. Ghosh C, Lazarus HM, Hewlett JS et al. Fluctuation of serum phenytoin concentrations during autologous bone marrow transplant for primary central nervous system tumors. J Neuro-Oncology 12:25–32, 1992. 111. Lackner TE: Interaction of dexamethasone with phenytoin. Pharmacotherapy 11:344–347, 1991. 112. Grossman SA, Sheilder VR, Gilbert MR: Decreased phenytoin levels in patients receiving chemotherapy. Am J Med 87:505–510, 1989. 113. Jarosinski PF, Moslow JA, Alexander MS et al. Altered phenytoin clearance during intensive treatment for acute lymphoblastic leukemia. J Pediatr 112:996–999, 1988. 114. Neef C, Voogd–van der Straaten I. An interaction between cytostatic and anticonvulsant drugs. Clin Pharmacol Ther 43:372–375, 1988.

Chapter 3 / Seizures and Anti-Epileptic Drugs in Neuro-Oncology

45

115. Sylvester RK, Lewis FB, Caldwell KC et al: Impaired phenytoin bioavailability secondary to cisplatinum, vinblastine and bleomycin. Ther Drug Monit 6:302–305, 1984. 116. Perucca E: Pharmacokinetic interactions with antiepileptic drugs. Clin Pharmacokinet 7:57–84, 1982. 117. Finchum RW, Schottelius DD: Decreased phenytoin levels in antineoplastic therapy. Ther Drug Monit 1:277–283, 1979. 118. Dofferhoff ASM, Berendsen HH, Naalt J et al. Decreased phenytoin level after carboplatin treatment. Am J Med 1990;89:247–248. 119. Vecht CJ, van Breemen M. Optimizing therapy of seizures in patients with brain tumors. Neurology 2006;76(suppl 4):S10–S13. 120. Newton HB. J Neurooncol 78:99–102, 2006 121. White HS. Comparative anticonvulsant and mechanistic profile of the established and newer antiepileptic drugs. Epilepsia 1999;40(Suppl5):S2–S10. 122. Rogers LR, Morris HH, Lupica K. Effect of cranial irradiation on seizure frequency in adults with low-grade astrocytoma and medically intractable epilepsy. Neurology 1993;43:1599–1601. 123. Regis Y, Roberts DW. Gamma knife radiosurgery relative to microsurgery: epilepsy. Stereotact Funct Neurosurg 1999;72 Suppl 1:11–21. 124. Warnke PC, Berlis A, Weyerbrock A et al. Significant reduction of seizure incidence and increase of benzodiazepine receptor density after interstitial radiosurgery in low-grade gliomas. Acta Neurochir Suppl (Wien) 1997;68:90–92. 125. Schiller Y, Cascino GD, So EL et al. Discontinuation of antiepileptic drugs after successful epilepsy surgery. Neurology 2000 54:346–349. 126. Khajavi K, Comair YG, Wyllie E et al.. Surgical management of pediatric tumor-associated epilepsy. J Child Neurol 1999;14:15–25. 127. Salanova V, Markand O, Worth R. Longitudinal follow-up in 145 patients with medically refractory temporal lobe epilepsy treated surgically between 1984 and 1995. Epilepsia 1999;40:1417–23. 128. Eliashiv SD, Dewar S, Wainwright I et al. Long–term follow-up after temporal lobe resection for lesions associated with chronic seizures. Neurology 1997;48:1383–1388. 129. Zentner J, Hufnagel A, Wolf HK et al. Surgical treatment of neoplasms associated with medically intractable epilepsy. Neurosurgery 1997;41:378–386. 130. Britton JW, Cascino GD, Sharbrough FW et al. Low-grade glial neoplasms and intractable partial epilepsy: efficacy of surgical treatment. Epilepsia 1994;35:1130–1135. 131. Kirkpatrick PJ, Honavar M, Janota I et al. Control of temporal lobe epilepsy following en bloc resection of low-grade tumors. J Neurosurg 1993;78:19–25. 132. Cascino GD, Kelly PJ, Hirschorn KA, et al. Stereotactic resection of intra-axial cerebral lesions in partial epilepsy. Mayo Clin Proc 1990;65:1053–1060. 133. Khan RB, Boop FA, Onar A et al. Seizures in children with low-grade tumors: outcome after tumor resection and risk factors for uncontrolled seizures. J Neurosurg 2006;104(6 Suppl Pediatrics):377–382. 134. Spencer DD, Spencer SS, Mattson RH et al. Intracerebral masses in patients with intractable partial epilepsy. Neurology 1984;34: 432–436. 135. Awad IA, Rosenfeld J, Ahl J et al. Intractable epilepsy and structural lesions of the brain: mapping, resection strategies, and seizure outcome. Epilepsia 1991;32:179–186. 136. Lieu AS, Howng SL. Intracranial meningiomas and epilepsy: incidence, prognosis and influencing factors. Epilepsy Res. 2000;38: 45–52. 137. Chozick BS, Reinert SE, Greenblatt SH. Incidence of seizures after surgery for supratentorial meningiomas: a modern analysis. J Neurosurg 1996;84:382–386. 138. Berger MS, Ghatan S, Haglund MM et al. Low-grade gliomas associated with intractable epilepsy: seizure outcome utilizing electrocorticography during tumor resection. J Neurosurg 1993;79:62–69. 139. Hagland MM, Berger MS, Kunkel DD et al. Changes in gamma-aminobutyric acid and somatostatin in epileptic cortex associated with low-grade gliomas. J Neurosurg 1992;77:209–216. 140. Strowbridge BW, Bean AJ, Spencer DD et al.. Low levels of somatostatin-like immunoreactivity in neocortex resected from presumed seizure foci in epileptic patients. Brain Res 1992;587:164–168. 141. Bateman DE, Hardy JA, McDermott JR et al. Amino acid neurotransmitter levels in gliomas and their relationship to the incidence of epilepsy. Neurol Res 1988;10:112–114. 142. Berger MS. Functional mapping–guided resection of low-grade gliomas. Clin Neurosurg 1995;42:437–452. 143. Kirkpatrick PJ, Honavar M, Janota I et al. Control of temporal lobe epilepsy following en bloc resection of low-grade tumors. J Neurosurg 1993;78:19–25. 144. Pilcher WH, Silbergeld DL, Berger MS et al.. Intraoperative electrocorticography during tumor resection: impact on seizure outcome in patients with gangliogliomas. J Neurosurg 1993 Jun;78(6):891–902. 145. Berger MS, Ghatan S, Geyer JR et al.. Seizure outcome in children with hemispheric tumors and associated intractable epilepsy: the role of tumor removal combined with seizure foci resection. Pediatr Neurosurg 1991–92;17:185–191. 146. Schaller B, Ruegg SJ. Brain tumor and seizures: pathophysiology and its implication for treatment revisited [published retraction appears in Epilepsia 44:1463, 2003]. Epilepsia 44:1223–1232, 2003. 147. Whittle IR, Beaumont A. Seizures in patients with supratentorial oligodendroglial tumours.: clinicopathological features and management considerations. Acta Neurochir 1995;135:19–24. 148. Smith DF, Hutton JL, Sandermann D, et al. The prognosis of primary intracerebral tumours presenting with epilepsy: the outcome of medical and surgical management. J Neurol Neurosurg Psychiatry 1991;54:915–920. 149. MRC Brain Tumour Working Party. Prognostic factors for high-grade malignant gliomas: development of a prognostic index. J Neurooncol 1990;9:47–55. 150. Scott GM, Gibberd FB. Epilepsy and other factors in the prognosis of gliomas. Acta Neurol Scand 1980;61:227–239.

46

Part III / Neurologic Symptoms

151. Glantz MJ, Cole BF, Forsyth PA, et al. Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Neurology 2000;54:1886–1893. 152. Temkin NR. Antiepileptogenesis and seizure prevention trials with antiepileptic drugs: meta-analysis of controlled trials. Epilepsia 42:515–524, 2001. 153. Mauro AM, Bomprezzi C, Morresi S et al. Prevention of early post-operative seizures in patients with primary brain tumors: preliminary experience with oxcarbazepine. J Neurooncol 2007;81:279–285. 154. Kuijlen JMA, Teernstra OPM, Kessels AGH, et al. Effectiveness of antiepileptic prophylaxis used with supratentorial craniotomies: a meta-analysis. Seizure 1996;5:291–298. 155. Siomin V, Angelov L, Li L et al.. Results of a survey of neurosurgical practice patterns regarding the prophylactic use of anti-epilepsy drugs in patients with brain tumors. J Neurooncol 74:211–215, 2005. 156. Mehta MP, Rodrigus P, Terhaard CH et al. Survival and neurologic outcomes in a randomized trial of motexafin gadolinium and whole-brain radiation therapy in brain metastases. J Clin Oncol 2003;21:2529–2536. 157. Dalmau J, Tuzun E, Wu H-y et al. Paraneoplastic anti-N-methyl-d-aspartate receptor encephalitis associated with ovarian teratoma. Ann Neurol 2007;61:25–36. 158. Garg RK. Posterior leukoencephalopathy syndrome. Postgrad Med J 2001;77:24–28.

4

Corticosteroids in Neuro-Oncology Santosh Kesari, MD, PHD, Nina A. Paleologos, and Nicholas A. Vick, MD

MD,

CONTENTS Introduction Clinical Neuro-Oncologic Usage of Steroids Toxicity Drug Interaction Mechanisms of Action Novel Anti-Edema Agents Conclusion References

Summary This chapter will review the clinical aspects of the use of corticosteroids in neuro-oncology and the known biological basis of the effects of this class of drugs on peritumoral brain edema. Despite their importance, there are relatively few studies specifically addressing these issues. In the majority of circumstances, corticosteroids are used at supraphysiologic and pharmacologic doses to reduce cerebral edema. Thus, corticosteriods are notorious for their adverse effects, which include gastrointestinal complications, myopathy, opportunistic infections, osteoporosis, and mood disturbances and must be used with caution. The emerging role of steroid sparing agents such as angiogenesis inhibitors and human corticotropin-releasing factor will be discussed. Key Words: corticosteroids, brain tumor, neurotoxicity, myopathy, VEGF inhibitors, corticotropin-releasing factor

1. INTRODUCTION Corticosteriods have been employed in the management of a wide variety of conditions, including peritumoral edema. This chapter will review the clinical aspects of the use of corticosteroids in neuro-oncology and the known biological basis of the remarkable effects of this class of drugs on peritumoral brain edema. Unfortunately, despite decades of work, we still know very little. Most of what we do clinically with corticosteroids is empirical. The absence of a full understanding of their mechanism of action has blocked the search for drugs with similar benefit but that lack the troublesome side effects until recently. Nonetheless, this deficiency in our understanding of how steroids work does not detract from the tremendous impact that this class of drugs has had on the care of patients with brain tumors since they were introduced for control of tumor-induced edema nearly 50 years ago. Ingraham pioneered the use of cortisone to treat postoperative cerebral edema in neurosurgical patients in 1952 (1) (a finding that has since revolutionized the practice of neurosurgery), and Korfman first used prednisone for peritumoral edema from brain metastases in 1957 (2). But only after 1961, when Galicich From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

47

48

Part III / Neurologic Symptoms

et al. showed that dexamethasone effectively alleviates cerebral edema due to brain tumors (3), was there an immediate revolution in brain tumor care. It is of some historical interest to note that the rationale for Galicich’s work was not to control cerebral edema in brain tumor patients. Instead, he and his colleagues were after a chemotherapeutic response. They had some laboratory evidence that large doses of corticosteroids would inhibit the growth of experimental brain tumors. They gave high doses of corticosteroids to 14 brain tumor patients in conjunction with angiography and promptly observed that within 24 hrs the patients were remarkably improved in a way that could not be explained by any chemotherapeutic effect. They used dexamethasone, which had first been synthesized in 1958, because it had a low index of sodium and water retention compared to the other corticosteroids then available and is currently the favored drug in clinical practice.

2. CLINICAL NEURO-ONCOLOGIC USAGE OF STEROIDS Prednisone, prednisolone, dexamethasone, and methylprednisolone are active antineoplastic agents chiefly against hematologic malignancies. In addition to the usage of dexamethasone for peritumoral edema in primary and secondary brain tumors, glucocorticoids are also used to control pain, edema, nausea, and vomiting, as well as to improve appetite. They also may be effective in controlling pain in patients with carcinomatous meningitis and are effective antineoplastic agents in primary central nervous system (CNS) lymphoma. Dexamethasone has become the drug of choice in neuro-oncology, in part due to its long half-life, low mineralocorticoid activity (Table 1), and a relatively low tendency to induce psychosis. There has been only one small prospective clinical trial to determine dexamethasone’s best dosage in neurooncology (4), but nearly four decades of experience has led to fairly standardized usage. Several important points deserve emphasis. Large doses of dexamethasone must be given intravenously when a patient presents with acute neurological symptoms or signs suggestive of the presence of a brain tumor or a spinal cord lesion. If the signs and symptoms are severe with potentially dangerous increased intracranial pressure, 10–20 mg intravenously (IV) is required before starting maintenance dosage. In situations when the tempo of evolution or degree of symptoms is less critical there is no need for intravenous administration. Oral dexamethasone at a dosage of 4–24 mg in divided doses is usually adequate after an oral loading dose. If focal neurologic symptoms are due to peritumoral vasogenic edema, dexamethasone induces improvement within 48 hrs and usually sooner. If there is no benefit, the neurologic symptoms are likely to be due to damage of the brain tissue by the tumor and not to edema. Dexamethasone is well absorbed orally and has a remarkably long biological half-life; its action by that route is almost as rapid as when given intravenously (Table 1). It is therefore unnecessary to dose patients more than twice daily. Because it is conventional inpatient neurosurgical practice to use an IV dosing schedule of every 4–6 hrs, many of these patients are discharged on the same schedule for oral medication. This is very difficult for patients to comply with at home and is unnecessary given what is known about dexamethasone’s long biological half-life. The efficacy of steroids in reducing the edema associated with tumors is well confirmed (5–7). More than 70 percent of patients with cerebral metastases symptomatically improve after starting steroids. In general, symptoms reflecting generalized neurologic dysfunction or brain edema improve more consistently than do focal symptoms Table 1 Glucocorticoid Comparison Glucocorticoid Cortisone Hydrocortisone Prednisolone Prednisone Methylprednisolone Dexamethasone

Approximate Equivalent Dose

Biologic Half-life (hrs)

Relative MineraloCorticoid Activity

25 mg 20 mg 5 mg 5 mg 4 mg 0.75 mg

8–12 8–12 18–36 18–36 18–36 36–54

++ ++ + + 0 0

Chapter 4 / Corticosteroids in Neuro-Oncology

49

Fig. 1. (1a): MRI (T1 post-gadolinium) of a patient with recurrent glioblastoma, on a low-dose (4 mg twice a day) of dexamethasone, whose symptoms were worsening. (1b). MRI (T1 post-gadolinium) of the same patient 8 days after increasing the dexamethasone to 20 mg twice a day. There were no other changes in management.

50

Part III / Neurologic Symptoms

such as hemiparesis. The effect can be manifest within hours, but is usually insufficiently rapid to be relevant for intra-operative care unless preoperative dosing was provided. Administration of steroids beginning 48 hrs prior to an elective surgical procedure has the potential to reduce edema formation and to improve clinical condition by the time of the craniotomy (8). Although clinical response (as measured by a reduction in the number of intracranial pressure plateau waves) can occur within 24 hrs, the pressure may not be measurably less for 2–4 days after a pulse dose (5,9).

2.1. Imaging Neuroimaging studies may show a dramatic improvement in peritumoral edema and mass effect (Fig. 1), although they may lag behind clinical improvement and not show a decrease in edema for at least one week (10). Nonfocal deficits such as headache and lethargy tend to respond better than focal deficits. Dexamethasone can always be tapered down or even discontinued in a few days if the initial vigorous treatment proves unnecessary. In patients who have evidence of spinal cord compression we usually use high doses of dexamethasone at 50–100 mg initially. While no controlled study in tumor patients documents that 50–100 mg is better than smaller doses, high-dose steroids have been shown in traumatic cord injury to be of benefit, and moreover no clear additional patient risk occurs from using larger doses as compared to smaller doses in the acute setting. There is little to lose and much to be gained by treating patients with known cancer who complain of sensory loss or weakness in the legs before definitive diagnosis. If myelopathy due to cord compression is excluded with appropriate neuroimaging, then the dexamethasone can be discontinued. If primary CNS lymphoma is a possible diagnostic consideration, dexamethasone should be withheld unless cerebral edema and mass effect are clinically significant. The reason for this is the prompt antineoplastic effect of dexamethasone. In some cases the response can be so dramatic as to impact on the ability to biopsy the tumor. In only a few days the enhancing tumor can improve to a degree that it may not appear on the localizing scan done prior to biopsy! In patients with brain tumors who have had gross total resections or large subtotal resections, dexamethasone can usually be tapered down and discontinued within a week or two after surgery. In that patient population there is usually no need for dexamethasone through radiotherapy. In patients who are not able to undergo surgical resection, or who have significant residual tumor, dexamethasone can be useful in controlling edema during radiation therapy. Dexamethasone should be maintained throughout radiation therapy in patients being treated for spinal cord compression. The dose can then usually be tapered over the following 2–3 weeks depending on stability or improvement in neurological function.

3. TOXICITY The side effects and complications that can occur with all of the glucocorticoids may be bothersome and reversible or cause significant disability and may be fatal. While a detailed review of the many side effects of steroids (Table 2) is beyond the scope of this chapter, a few merit discussion.

3.1. Gastrointestinal Bleeding Most patients receiving corticosteroids are treated routinely with medications (usually histamine H2 antagonists) to reduce the risk of gastric ulcer and hemorrhage. An association between steroid usage and peptic ulceration has not been clearly shown. No prospective trials have been done, but in retrospective analyses of clinical trials evaluating corticosteroid usage in a number of diseases, no significant association between steroid usage and gastrointestinal bleeding could be found (11–13) In practice, however, it may be prudent to use H2 antagonists in patients on chronic steroids, especially in the immediate post-operative period and in those patients receiving unusually high doses of corticosteroids. In most patients, twice daily corticosteroid dosing with meals may reduce the risk of possible stomach irritation and spare patients the added side effects and expense of H2 antagonists. Other rare problems associated with corticosteroid use include pancreatitis, small bowel perforation, and fatty liver.

Chapter 4 / Corticosteroids in Neuro-Oncology

51

Table 2 Complications of Corticosteroid Therapy Common Neurologic

Dermatologic

Gastrointestinal

Rheumatologic

Ophthalmologic

Endocrine/metabolic

Urogenial Hematologic Miscellaneous

Uncommon

Behavioral change Insomnia Myopathy Tremor Reduced taste and smell Thin, fragile skin Purpura Ecchymoses Striae, acne Inhibition of wound healing Hirsutism Increased appetite Bloating

Osteoporosis Fracture Growth retardation in children Visual blurring Cataract Hyperglycemia Hypokalemia Hypernatremia Hyperlipidemia Redistribution of body fat Amenorrhea Polyuria Neutrophilia Lymphopenia Night sweats Candidiasis Infection

Psychosis, seizures Dependence Paraparesis due to epidural lipomatosis

Kaposi’s sarcoma

Gastrointestinal bleeding Pancreatitis Liver hypertrophy Perforation Avascular necrosis Tendinous rupture Glaucoma, uveitis Exopthalmos Arrhythmia (with IV push)

Genital burning (with IV push)

Some opportunistic infections (PCP, aspergillosis etc.)

3.2. Myopathy Steroid myopathy is common and troublesome at times, impacting significantly on the quality of function and life of patients with cancer. It is the most common side effect of dexamethasone in patients with primary brain tumors and occurs in as many as 10.6% of those patients (14,15) The majority of patients develop weakness between the ninth and twelfth weeks of treatment. Individual susceptibility varies dramatically. Some patients develop significant weakness after taking a low dose of steroids for only a few weeks, while other patients receive large doses for months to years and never develop symptoms. Those who do not develop myopathy also seem to be patients who do not become cushingoid. It is thought that fluorinated glucocorticoids, such as dexamethasone, are more likely to produce muscle weakness and atrophy than nonfluorinated glucocorticoids such as hydrocortisone or prednisone (16–19). While there may be improvement in steroid myopathy with substitution of nonfluorinated glucocorticoids such as prednisone for dexamethasone, prednisone may not be so effective in controlling brain edema. One of us (N.V.) attempted to use high-dose alternate day prednisone in place of daily dexamethasone in six patients. Only in one

52

Part III / Neurologic Symptoms

of those was alternate day prednisone as effective as the previous equivalent dosage of dexamethasone. One of the six did so poorly that the switch may have contributed to his decline and death. The precise pathophysiology of steroid myopathy is unknown. It is likely that steroids exert their effects through inhibition of protein synthesis (partly through inhibition of peptide initiation), increased protein catabolism, and possible induction of glutamine synthetase activity (20,21). Steroid myopathy may improve if the drug can be withdrawn or the dose reduced. Recovery can take a few months after discontinuation of the steroid. Improvement in patients treated at reduced doses can take even longer. In animal models, muscle activity may reduce steroidinduced muscle wasting, suggesting the possibility that exercise or a physical therapy program may help reduce the severity of myopathy in patients receiving steroids chronically (22).

3.3. Infections Use of moderate to high doses of glucocorticoids can result in clinically significant suppression of the immune system and vulnerability to opportunistic infections. Pneumocystis jirovecii (previously known as Pneumocystis carinii) is a archiascomycetous fungus capable of causing life-threatening pneumonitis (PJP) in immunocompromised patients (23,24). PJP is relatively rare in patients with solid tumors (25), but there is increasing evidence that patients with brain tumors receiving corticosteroids may be at increased risk of PJP(26–28).

3.4. Osteoporosis Patients receiving chronic corticosteroids are at increased risk of developing osteoporosis. In the past this has not been considered a significant problem due to the limited prognosis of most patients with malignant brain tumors. However, as the survival of these patients improves, an increasing number are developing complications of osteoporosis such as fractures on the lumbar spine and hip. The mechanism of bone loss is multifactorial, but the most important effects are due to direct actions of glucocorticoids in skeletal cells and include reduction in calcium absorption, with secondary hyperparathyroidism and decreased gonadal hormones. Molecular mechanisms such as reduction in insulin-like growth factor-1 and prostaglandin E2, both of which stimulate bone growth, are also implicated. Patients receiving chronic corticosteroid therapy should be given calcium supplements (1500 mg/day) with vitamin D (800 IU daily) or an activated form of vitamin D (e.g., alfacalcidiol at 1 microgram/day or calcitriol at 0.5 microgram/day) (29). In addition, bisphosphonates such as etidronate, alendronate, risedronate, and zoledonate (29,30) should be considered. For patients who develop severe pain from compression fractures, kyphoplasty may have a role.

3.5. Mood Disturbance Mild neuropsychiatric effects of steroids such as anxiety, insomnia, and emotional lability are probably the most common and pervasive; however, the more dramatic euphoric and psychotic presentations are the more memorable and significant (31). Patients with a history of psychiatric problems are at greatest risk. Seizures are more unusual features of steroid effect, seen at high doses, and are usually limited to patients with a history of a seizure disorder. Distinguishing these steroid complications from manifestations of gliomas, cerebral irradiation, or changing intracranial pressure can be difficult. Management of steroid-induced neuropsychiatric side effects involves discontinuing or tapering the implicated agent as soon as is practical. Neuroleptics or lithium may be considered in consultation with a psychiatrist. Tricyclic antidepressants should be avoided as they may confound the problem.

3.6. Other Effects Dexamethasone has pleiotrophic effects that result in various side effects and complications. Well-known side effects of corticosteroids include the development of varying manifestations of Cushing’s syndrome, hypothalamicpituitary-adrenal insufficiency, impaired wound healing, hyperglycemia, and hypertension. More recently, dexamethasone was thought to have caused a posterior reversible encephalopathy syndrome (32). Dexamethasone can also inhibit proliferation of neural stem cells and astroglial differentiation, thereby causing cognitive dysfunction (33,34).

Chapter 4 / Corticosteroids in Neuro-Oncology

53

4. DRUG INTERACTION Phenytoin increases the metabolic clearance rate of cortisol and dexamethasone and reduces the plasma half-life of dexamethasone by up to 50% (35–37) It has been postulated that this may be a mechanism for a protective effect of phenytoin in reducing the risk of development of steroid myopathy (19). Carbamazepine and phenobarbital may also induce the hepatic metabolism of dexamethasone (38). Patients with brain tumors on anticonvulsants may therefore need a higher dose of dexamethasone to control brain edema.

5. MECHANISMS OF ACTION 5.1. Cerebral Edema In 1967, in a paper entitled “Neuropathological Aspects of Brain Edema,” (39) Klatzo proposed that his topic should be divided conceptually into two major and usually distinct types, vasogenic and cytotoxic. The earlier literature on the subject was very confusing; this classic article proved to be extremely important for future thinking about the subject. Earlier in the electron microscopic era, it was thought that edema around brain tumors was confined to the intracellular space and specifically to the glial processes. It soon became known, however, that artifacts caused by hypertonic fixatives delayed full recognition of an even more important space in which fluid accumulates adjacent to brain tumors—the intercellular space. Intercellular edema is obvious in white matter. In lightly myelinated gray matter structures, such as the thalamus, intercellular swelling is very inconspicuous. This correlates quite well with what is seen clinically as revealed by contrast-enhanced CT and MRI: generally, it is tumors involving white matter that have consequential mass effect due to edema. Such instances provide as good an example of the clinical reality of vasogenic edema as one can see. In Klatzo’s words, vasogenic edema is the type “in which the starting point of the edema is related to the injury of the walls of cerebral vessels leading to the escape of water and plasma constituents into the surrounding parenchyma.” Cytotoxic edema, on the other hand, is the type “in which a noxious factor directly effects the structural elements of the parenchyma producing intracellular swelling, vascular permeability remaining relatively undisturbed.” Cytotoxic edema, such as occurs in ischemia and infarction, has been repeatedly shown to be resistant to the effects of corticosteroids. Many studies have shown no benefit of corticosteroids for patients with strokes (40). The recognition of the extraordinary difference between the brain swelling due to tumors and that due to strokes is part of our daily working knowledge. What needs to be understood, then, is what is special about vasogenic edema and the pathophysiology of peritumoral brain edema. Simply stated, what biological mechanisms underlay the remarkable response of brain tumor edema to corticosteroids? The structural background underlying the problem was worked out in the1960s and 1970s, and little has been added since (41–44) In the normal mammalian brain, except for some small special regions mentioned below, the endothelium is continuous. It is held together by continuous belts of tight or pentalaminar junctions, and has few microvesicles in the cytoplasm. Gaps or fenestrations are conspicuously lacking. This specialized endothelium is the anatomic locus for the blood–brain barrier (BBB) as defined almost a century ago with the use of cationic dyes and in the 1970s by various electron dense tracers suitable for ultrastructural investigation. Regions of the brain that have been known for years to lie outside the blood–brain-barrier, such as the choroid plexus and area postrema, are freely permeable to cationic dyes and tracers in current usage because in these regions the capillaries are fenestrated. In these brain regions, there is complete access to the plasma and its constituents, such as the contrast agents in use for CT and MRI. While such contrast agents and other test substances are normally excluded from the brain by its specialized continuous endothelium, the endothelium in brain tumors is altered structurally and resembles fenestrated endothelium. The mechanism by which tumors cause such abnormal endothelium is unknown. We do know that leakage of contrast is quite variable from one tumor type to another, and even within the same tumor. This variability is the basis for many diagnostic neuroradiological criteria. Without contrast agents, CT or MRI for brain tumors is very limiting and, by current standards, simply not acceptable. Solute transport across normal brain capillaries is confined to two mechanisms. Solutes may either dissolve in and diffuse through the endothelial membranes or be transported to the cytoplasm of the endothelial cell.

54

Part III / Neurologic Symptoms

Blood-to-brain transfer of substances depends directly on the substance’s membrane (lipid) solubility and, inversely on molecular size. Transport through the cytoplasm of endothelial cells is by carrier-mediated transfer processes which are highly stereo-specific, saturable, and often competitive. This mechanism, unlike diffusion, is relatively independent of lipid solubility and molecular size. Bulk flow of water is very slow in the normal brain due to the low hydraulic conductivity of normal brain vessels. The osmotic pressure of plasma proteins at the endothelial cell membrane of normal brain capillaries balances the hydrostatic pressure gradient between intravascular and extravascular tissue compartments. In addition, because normal brain capillaries are relatively impermeable to the main ionic constituents of plasma, such as sodium and chloride, there is an additional osmotic buffering capacity that slows the passage of water into the brain. The small, narrow channels that comprise the extracellular space of the brain also restrict the flow of extracellular fluid. They are, in effect, high-resistance pathways. These phenomena that regulate the normal blood–brain barrier are overwhelmed in brain tumors by an increase in the hydraulic conductivity of the endothelium due to the presence of fenestrations and gaps of the endothelial wall. These defects permit bulk flow of fluid from the vascular compartment into the brain tumor and adjacent brain. The rate and extent of the development of edema depends on hydrostatic pressure gradients in the tissue, the hydraulic conductivity of fluid flow through the tissue through the extracellular space, and the rate at which the edema fluid is removed. The flow of plasma derived fluid into tumor and adjacent edematous brain is, in fact, exactly what Klatzo described as vasogenic edema.

5.2. Vasogenic Edema and Dexamethasone The vasogenic edema surrounding brain tumors contributes significantly to the morbidity experienced by patients. This edema results from the flow of fluid into the extracellular space of the brain parenchyma through an incompetent BBB (45). In high-grade gliomas and brain metastases, the BBB is disrupted, allowing passage of fluid into the extracellular space. The leakiness of the BBB is due primarily to opening of the interendothelial tight junctions, but also to increased endothelial pinocytosis and endothelial fenestrations (46). Defective endothelial tight junctions result from deficiency of normal astrocytes (which produce factors required for the formation of a normal BBB) and the production by tumor cells of factors such as vascular endothelial growth factor (VEGF) (47) and scatter factor/hepatocyte growth factor (48) (which increase the permeability of tumor vessels) (49). As noted above, the mechanism of action of corticosteroids is not well understood, but they may produce their anti-edema effect by reducing the permeability of tumor capillaries (46,50,51). Corticosteroids diffuse through the plasma membrane and bind to the cytoplasmic receptor, allowing the steroid-receptor complex to move to the nucleus where it affects transcription of genes and also interacts with other transcription factors such as NF-B (52). Corticosteroids decrease endothelial permeability in part by causing dephosphorylation of the tight junction component proteins occludin and ZO1 (46).

6. NOVEL ANTI-EDEMA AGENTS The large number of complications associated with corticosteroid therapy has led to the search for alternative therapies for peritumoral edema. Corticotropin-releasing factor (CRF) (53–55) reduces peritumoral edema by a direct effect on blood vessels through CRF 1 and 2 receptors, independent of the release of adrenal steroids, and has been effective in animal models (54). Phase I/II trials of this agent suggested that it is relatively well tolerated (53,55). Several phase III trials are currently in progress examining the efficacy of this drug in the treatment of acute and chronic peritumoral edema. Recently, preliminary studies suggest that cyclooxygenase-2 (COX-2) inhibitors might be effective in treating cerebral edema (56,57). Clinical studies using COX-2 inhibitors for peritumoral edema are planned, although the cardiac complications of this class of drugs have delayed these studies. Since VEGF plays an important role in the pathogenesis of peritumoral edema, it is possible that inhibitors of VEGF, such as VEGF antibodies (e.g., bevacizumab [Avastin]) or inhibitors of VEGF receptors, such as sorafenib (Nexavar) and sunitinib (Sutent) may be helpful in reducing peritumoral edema. It is hoped that some of these agents will prove to be more effective and less toxic alternatives to corticosteroids.

Chapter 4 / Corticosteroids in Neuro-Oncology

55

CONCLUSION In recent years, interest in and understanding of the blood–brain barrier and cerebral edema has increased and use of dexamethasone on peritumoral brain edema will be looked at more critically. As evidence for the benefit of novel anti-edema agents increases, the use of corticosteriods will likely decrease since these novel agents may have similar or improved benefits compared to dexamethasone without its side effects.

REFERENCES 1. Ingraham FD, Matson DD, McLaurin RL. Cortisone and ACTH as an adjunct to the surgery of craniopharyngiomas. N Engl J Med 1952;246(15):568–571. 2. Kofman S, Garvin JS, Nagamani D et al. Treatment of cerebral metastases from breast carcinoma with prednisolone. J Am Med Assoc 1957;163(16):1473–1476. 3. Galicich JH, French LA, Melby JC. Use of dexamethasone in treatment of cerebral edema associated with brain tumors. J Lancet 1961;81:46–53. 4. Vecht CJ, Hovestadt A, Verbiest HB et al. Dose–effect relationship of dexamethasone on Karnofsky performance in metastatic brain tumors: a randomized study of doses of 4, 8, and 16 mg per day. Neurology 1994;44(4):675–680. 5. Miller JD, Leech P. Effects of mannitol and steroid therapy on intracranial volume–pressure relationships in patients. J Neurosurg 1975;42(3):274–281. 6. Miller JD, Sakalas R, Ward JD et al. Methylprednisolone treatment in patients with brain tumors. Neurosurgery 1977;1(2):114–117. 7. Yeung WT, Lee TY, Del Maestro RF et al. Effect of steroids on iopamidol blood–brain transfer constant and plasma volume in brain tumors measured with X-ray computed tomography. J Neurooncol 1994;18(1):53–60. 8. Bell BA, Smith MA, Kean DM et al. Brain water measured by magnetic resonance imaging: correlation with direct estimation and changes after mannitol and dexamethasone. Lancet 1987;1(8524):66–69. 9. Gutin PH. Corticosteroid therapy in patients with cerebral tumors: benefits, mechanisms, problems, practicalities. Semin Oncol 1975;2(1):49–56. 10. Vecht CJ, Verbiest HBC. Use of glucocorticoids in neuro-oncology. In: RG Wiley, ed. Neurologic Complications of Cancer. New York: Marcel Dekker; 1995:199–218. 11. Carson JL, Strom BL, Schinnar R et al. The low risk of upper gastrointestinal bleeding in patients dispensed corticosteroids. Am J Med 1991;91(3):223–228. 12. Conn HO, Blitzer BL. Nonassociation of adrenocorticosteroid therapy and peptic ulcer. N Engl J Med 1976;294(9):473–479. 13. Conn HO, Poynard T. Adrenocorticosteroid administration and peptic ulcer: a critical analysis. J Chronic Dis 1985;38(6):457–468. 14. Dropcho EJ, Soong SJ. Steroid-induced weakness in patients with primary brain tumors. Neurology 1991;41(8):1235–1239. 15. Vick NA. Steroid toxicity. J Neurooncol 1988;6(2):199. 16. Askari A, Vignos PJ, Jr., Moskowitz RW. Steroid myopathy in connective tissue disease. Am J Med 1976;61(4):485–492. 17. Kelly FJ, McGrath JA, Goldspink DF et al. A morphological/biochemical study on the actions of corticosteroids on rat skeletal muscle. Muscle Nerve 1986;9(1):1–10. 18. Koski CL, Rifenberick DH, Max SR. Oxidative metabolism of skeletal muscle in steroid atrophy. Arch Neurol 1974;31(6):407–410. 19. Ruff RL. Endocrine myopathies. In: AG. Engle, BQ Banker, eds. Myology: Basic and Clinical. New York: McGraw–Hill; 1986: 1871–1879. 20. Kimura K, Kanda F, Okuda S et al. Insulin-like growth factor 1 inhibits glucocorticoid-induced glutamine synthetase activity in cultured L6 rat skeletal muscle cells. Neurosci Lett 2001;302(2–3):154–156. 21. Owczarek J, Jasinska M, Orszulak-Michalak D. Drug-induced myopathies: an overview of the possible mechanisms. Pharmacol Rep 2005;57(1):23–34. 22. Layzer RB. Neuromuscular Manifestations of Systemic Disease. Philadelphia: F.A. Davis 1985. 23. Edman JC, Kovacs JA, Masur H et al. Ribosomal RNA sequence shows Pneumocystis carinii to be a member of the fungi. Nature 1988;334(6182):519–522. 24. Thomas CF, Jr., Limper AH. Pneumocystis pneumonia. N Engl J Med 2004;350(24):2487–2498. 25. Henson JW, Jalaj JK, Walker RW et al. Pneumocystis carinii pneumonia in patients with primary brain tumors. Arch Neurol 1991;48(4):406–409. 26. Mathew BS, Grossman SA. Pneumocystis carinii pneumonia prophylaxis in HIV negative patients with primary CNS lymphoma. Cancer Treat Rev 2003;29(2):105–119. 27. Schiff D. Pneumocystis pneumonia in brain tumor patients: risk factors and clinical features. J Neurooncol 1996;27(3):235–240. 28. Sepkowitz KA, Brown AE, Telzak EE et al. Pneumocystis carinii pneumonia among patients without AIDS at a cancer hospital. JAMA 1992;267(6):832–837. 29. Osteoporosis, ACoRAHCoG-I. Recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis: 2001 update. American College of Rheumatology Ad Hoc Committee on Glucocorticoid-Induced Osteoporosis. Arthritis Rheum 2001;44(7):1496–1503. 30. Lipton A. New therapeutic agents for the treatment of bone diseases. Expert Opin Biol Ther 2005;5(6):817–832. 31. Truhan AP, Ahmed AR. Corticosteroids: a review with emphasis on complications of prolonged systemic therapy. Ann Allergy 1989;62(5):375–391. 32. Irvin W, MacDonald G, Smith JK et al. Dexamethasone-induced posterior reversible encephalopathy syndrome. J Clin Oncol 2007;25(17):2484–2486.

56

Part III / Neurologic Symptoms

33. Kim JB, Ju JY, Kim JH et al. Dexamethasone inhibits proliferation of adult hippocampal neurogenesis in vivo and in vitro. Brain Res 2004;1027(1–2):1–10. 34. Sabolek M, Herborg A, Schwarz J et al. Dexamethasone blocks astroglial differentiation from neural precursor cells. Neuroreport 2006;17(16):1719–1723. 35. Chalk JB, Ridgeway K, Brophy T et al. Phenytoin impairs the bioavailability of dexamethasone in neurological and neurosurgical patients. J Neurol Neurosurg Psychiatry 1984;47(10):1087–1090. 36. Choi Y, Thrasher K, Werk EE et al. Effect of diphenylhydantoin on cortisol kinetics in humans. J Pharmacol Exp Ther 1971;176(1): 27–34. 37. Haque N, Thrasher K, Werk EE et al. Studies on dexamethasone metabolism in man: effect of diphenylhydantoin. J Clin Endocrinol Metab 1972;34(1):44–50. 38. Penry JK, Newmark ME. The use of antiepileptic drugs. Ann Intern Med 1979;90(2):207–218. 39. Klatzo I. Presidental address: neuropathological aspects of brain edema. J Neuropathol Exp Neurol 1967;26(1):1–14. 40. Fishman RA. Steroids in the treatment of brain edema. N Engl J Med 1982;306(6):359–360. 41. Brightman MW, Reese TS, Vick NA et al. A mechanism underlying the lack of a blood–brain barrier to peroxidase in virally induced brain tumors. J Neuropathol Exp Neurol 1971;30(1):139–140. 42. Reese TS, Karnovsky MJ. Fine structural localization of a blood–brain barrier to exogenous peroxidase. J Cell Biol 1967;34(1):207–217. 43. Vick NA. Brain tumor microvasculature. In: L Weiss, HA Gilbert, JG Posner, eds. Brain Metastasis. Boston: G.K. Hall & Company; 1980:115–133. 44. Vick NA, Bigner DD. Microvascular abnormalities in virally induced canine brain tumors: structural bases for altered blood–brain barrier function. J Neurol Sci 1972;17(1):29–39. 45. Wen PY, Schiff D, Kesari S et al. Medical management of patients with brain tumors. J Neuro-oncol 2006;80(3):313–332. 46. Papadopoulos MC, Saadoun S, Binder DK et al. S. Molecular mechanisms of brain tumor edema. Neuroscience 2004;129(4):1011– 1020. 47. Machein MR, Plate KH. VEGF in brain tumors. J Neuro-oncol 2000;50(1–2):109–120. 48. Lamszus K, Laterra J, Westphal M et al. Scatter factor/hepatocyte growth factor (SF/HGF) content and function in human gliomas. Int J Dev Neurosci 1999;17(5–6):517–530. 49. Cloughesy TF, Black KL. Peritumoral edema. In: MS. Berger, CB Wilson, eds. The Gliomas. Philadelphia W.B. Saunders; 1999: 107–114. 50. Hedley-Whyte ET, Hsu DW. Effect of dexamethasone on blood–brain barrier in the normal mouse. Ann Neurol 1986;19(4):373–7. 51. Heiss JD, Papavassiliou E, Merrill MJ et al. Mechanism of dexamethasone suppression of brain tumor-associated vascular permeability in rats: involvement of the glucocorticoid receptor and vascular permeability factor. J Clin Invest 1996;98(6):1400–1408. 52. Barnes PJ. Molecular mechanisms and cellular effects of glucocorticosteroids. Immunol Allergy Clin North Am 2005;25(3):451–468. 53. Hariharan S, Shapiro W, Chang S et al. Phase II randomized dose-ranging trial of human corticotrophin releasing factor in symptomatic brain tumor patients. In: American Academy of Neurology; 2000; p. A12; S06.001. 54. Tjuvajev J, Uehara H, Desai R et al. Corticotropin-releasing factor decreases vasogenic brain edema. Cancer Res 1996;56(6): 1352–1360. 55. Villalona-Calero MA, Eckardt J, Burris H et al. A phase I trial of human corticotropin-releasing factor (hCRF) in patients with peritumoral brain edema. Ann Oncol 1998;9(1):71–77. 56. Nathoo N, Barnett GH, Golubic M. The eicosanoid cascade: possible role in gliomas and meningiomas. J Clin Pathol 2004;57(1):6–13. 57. Portnow J, Suleman S, Grossman SA et al. A cyclooxygenase-2 (COX–2) inhibitor compared with dexamethasone in a survival study of rats with intracerebral 9L gliosarcomas. Neuro Oncol 2002;4(1):22–25.

5

Headache Robert Cavaliere,

MD

CONTENTS Introduction The Brain Tumor Headache Incidence Pathophysiology of Headache Related to Brain Tumors Headaches in Children with Brain Tumors Headache in Patients with Systemic Cancer Sellar Tumors Other Causes of Headcahes Headache Management Summary References

Summary Headaches are a common problem among the general population. The prevalence of headaches in patients with brain tumors has been inadequately studied, as has been the impact on patient quality of life. The purpose of this chapter is to describe the available epidemiological data on headaches, common characteristics of headaches in brain tumor patients, and the pathophysiology of tumor-related headaches. The chapter will conclude with a brief discussion of the management of headaches. Key Words: brain tumor, glioma, glioblastoma, cerebral metastases, pituitary tumor, pituitary adenoma, headache, pain

1. INTRODUCTION Headaches are a common problem in the general population. Perhaps the greatest concern among those who suffer from headaches is an underlying causative brain tumor. The reality is that brain tumors account for only a small minority of headaches. Nonetheless, headaches are a common symptom among patients with brain tumors. Headache data, which are derived primarily from retrospective analysis, are limited, and few studies have focused exclusively on this symptom. All tumor types, including primary and metastatic brain tumors, have been associated with headaches, although this symptom may be more common with rapidly growing and infratentorial lesions. While the “classic” brain tumor headache is recognized by most clinicians, in many circumstances the head pain is nonspecific. Typically patients have other associated neurological signs and symptoms and only rarely does the headache occur in isolation. Headaches can usually be palliated with conservative measures, although opioids and surgical resection may be required. The purpose of this chapter is to describe the common characteristics of headaches in brain tumor patients. The pathophysiology of tumor-related headaches as well as the available epidemiological data will be reviewed. The discussion will conclude with a brief discussion of headache management. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

57

58

Part III / Neurologic Symptoms

2. THE BRAIN TUMOR HEADACHE Headaches associated with brain tumors are classically described as occurring at night, often awakening the patient from sleep, or upon awakening in the morning (Table 1). They may be exacerbated by valsalva maneuvers, such as lifting heavy objects, sneezing, or coughing. Headaches may be positional and exacerbated when the head is placed in the dependant position such as when bending over (1–3). Alternatively, headaches may occur upon standing after lying recumbent for a period of time. In reality, however, fewer than half of all patients describe these type of symptoms. Often, headaches may be indistinguishable from benign tension or migraine headaches. Not uncommonly, patients are managed medically for benign headaches prior to diagnosis of the underlying causative tumor. Tumor-related headaches are usually intermittent in nature, generally lasting hours (1–3). Progressive headaches have been associated with more rapid growing tumors, such as glioblastoma and cerebral metastases, the presence of hydrocephalus, midline tumor location, and the extent of edema. Less often, tumors cause chronic headaches. Acute onset headaches may occur with intratumoral hemorrhage or cerebrospinal fluid obstruction. The headache may localize to the same side of the tumor or be generalized. The headaches are usually of moderate to severe intensity. In the presence of elevated intracranial pressure, headaches tend to be more severe and constant. In a prospective study of tumor-related headaches, among patients with hemispheric tumors, the pain localized to the side of the tumor in 41% of cases (3). Similarly, Forsyth et al. noted that bilateral headaches occurred in patients with elevated intracranial pressure with midline or bilateral tumors (2). In the absence of these factors, only 17% of patients with supratentorial tumors had bilateral headaches. Furthermore, among patients with unilateral headaches, the tumor was always on the ipsilateral side.

3. INCIDENCE The incidence of headaches in patients with brain tumors is unknown. At presentation, 37–62% of patients have headaches, with 60–90% of patients developing headaches over the course of their illness (4). Usually, headaches are associated with other neurological deficits and only rarely are they an isolated symptom (Table 2) (5–8). Nausea and vomiting frequently accompany tumor-related headaches (1,3,5,8). Posterior fossa and midline tumors, which may disturb cerebrospinal fluid mechanics, are more often associated with headache (Table 3) (1–3,8). Other factors possibly associated with tumor-related headaches include the size of the tumor (2), the presence of midline shift (2,3), edema (3), and hydrocephalus (2,3). Papilledema, present in 33–48% (1,2,5,9,10) Table 1 Headache Characteristics in Brain Tumors (Including Low and High-Grade Tumors) Description Timing Duration Location Intensity

Associated symptoms

“Tension type,” “dull ache,” “pressure-like,” and “sinus-like headache” in 77 % of patients “Migrainous” in 9–26 % of cases Intermittent, develops and resolves over several hours Worse with cough, valsalva maneuver, and bending in 23 % Interferes with sleep in 32 % Less than 1 mo: 29 % 1–6 mo: 26 % Greater than 6 mo: 45 % 72 % bilateral and 25 % unilateral Frontal in 68 % May be mild, moderate, or severe. Mean intensity of 8.5/10 when associated with increased intracranial pressure and 6.5/10 if no evidence of increased intracranial pressure Nausea and vomiting (38 %) Visual disturbance (40 %) Seizures (50 %)

Chapter 5 / Headache

59

Table 2 Clinical Factors That Suggest a Structural Cause for Headache Any change in pre-existing headache pattern Headaches unresponsive to previously effective therapies Any focal symptom or sign Papilledema Change in behavior, personality, or mentation Vomiting Awake patient from sleep or worse upon awakening in morning Worse with bending over, coughing, sneezing, or valsalva

Table 3 Factors Associated with an Increased Risk of Headache in Patients with Brain Tumors Elevated intracranial pressure Midline or posterior fossa tumors Leptomeningeal involvement Larger tumors Edema Midline shift Hydrocephalus Previous history of headaches

of tumor patients with headache, and hydrocephalus (2,3) are indicators of elevated intracranial pressure and commonly present among tumor patients with headaches. In such cases, pain is more diffuse in distribution, severe in intensity, refractory to common analgesics, and associated with nausea and vomiting (2). Patients with a history of headache report headaches as a tumor-related symptom more often than those without such a history. In many such instances, headaches are similar to those patients had previously experienced although they may be more frequent or intense (2). Slow-growing tumors, such as meningiomas, are less likely to cause headaches than rapid-growing lesions, such as metastases and glioblastomas (1–3). Presumably this is related to continuous adaptation of pain-sensitive structures as the tumor grows. Metastatic and primary brain tumors are equally likely to cause headaches (1,2) although metastatic tumors in one study were more likely to cause headaches (3).

4. PATHOPHYSIOLOGY OF HEADACHE RELATED TO BRAIN TUMORS Only a limited number of anatomic structures on the head are pain-sensitive, including the skin and subcutaneous tissue, muscles, periosteum of the skull, cerebral arteries, venous sinuses, dura at the base of the brain, and intracranial portions of the trigeminal, glossopharyngeal, vagus, and upper cervical nerves (4). Stimulation of these structures will generate pain. Alternatively, the brain parenchyma, pia-arachnoid, ependyma, and choroid plexus are insensitive to pain. Several mechanisms have been postulated to explain how tumors cause headaches and focus on traction or displacement of the pain-sensitive structures by the tumor. Intraparenchymal tumors distant from pain-sensitive structures may impact these areas indirectly as a consequence of edema and mass effect or hydrocephalus. Alternatively, a tumor may directly place pressure on or invade these structures. Although elevated intracranial pressure is common among tumor patients with headaches, its contribution to the development of headaches remains uncertain. Presumably, the pain is caused by traction of pain-sensitive structures by displaced brain.

60

Part III / Neurologic Symptoms

5. HEADACHES IN CHILDREN WITH BRAIN TUMORS Given the higher incidence of posterior fossa tumors in children, one might speculate that children may be at greater risk of headaches. In a recently published, single-institution, retrospective review of 200 children with brain tumors, headache was the most common presenting symptom, occurring in 41% and 56% of patients at presentation and over the course of illness, respectively. The median duration of the headaches prior to diagnosis was 2 months (5). In a retrospective review of the Childhood Brain Tumor Consortium, 62% of children with tumors presented with a headache, including 70% with infratentorial and 58% of supratentorial lesions (8). Irrespective of the duration of the headache, other symptoms were also present at diagnosis in all cases. Vomiting, vision changes, unsteadiness and educational/behavioral problems were the most common associated symptoms (5,8).

6. HEADACHE IN PATIENTS WITH SYSTEMIC CANCER Headache is a common symptom among patients with systemic cancer, including cerebral (11,12) and leptomeningeal metastases (13). Headaches are caused by both direct involvement of CNS by tumor (metastases) and indirect causes (treatment factors, psychological, exacerbation of pre-existing headache syndrome) (Table 4) (14).

6.1. Headache in Patients with Cerebral Metastases Cerebral metastases occur in approximately 25% of patients with systemic cancer. Approximately 24–38% of patients with cerebral metastases present with headaches (11,12). In a prospective study of patients with systemic cancer without known brain lesions referred for evaluation of headache, 32.4% of patients were found to have new cerebral metastases on magnetic resonance imaging. Patients found to have cerebral metastases had more abnormalities on neurological examination (lower mini-mental status score, coordination disturbance, Babinski sign) and associated symptoms (emesis, diplopia). In addition, the proportion of patients with headache duration of less than 10 weeks was greater among those with newly diagnosed cerebral metastases (10). Similar results were reported by Argyriou et al. in their prospective study of patients with systemic cancer presenting with new onset headaches or with a change in pattern of chronic headaches (9). Fifty-four percent of patients were subsequently found to have cerebral metastases. Among this group, emesis, gait instability, and Babinski sign were more often identified relative to those who were not found to have new brain lesions. Approximately 50% of patients had a single lesion. In both studies, headaches present in patients with cerebral metastases were distinct from tension headaches as defined by the International Headache Society.

6.2. Headache in Patients with Leptomeningeal Metastases Leptomeningeal metastases are estimated to occur in 15% of patients with solid and hematological tumors. Headaches, which are typically nonspecific, are common and afflict 44–75% of patients with this complication (13,15,16). Headaches may result from hydrocephalus (which occurs in approximately 30% of cases) or nerve root infiltration. The multifocal nature of leptomeningeal metastases is such that any level of the nervous system axis may be impacted including the cerebral hemispheres, cranial nerves, or spinal cord. Consequently, other symptoms and signs are present including alteration in mental status, nausea and vomiting, cranial nerve palsies, ataxia, lateralizing limb weakness or paraparesis, bowel and bladder dysfunction, or radiculopathic pain. Imaging may demonstrate leptomeningeal or cranial nerve enhancement or hydrocephalus. While headaches are common among patients with cancer, several factors should raise a clinician’s suspicions for an intracranial lesion. As stated earlier, headaches associated with vomiting, exertion or valsalva, focal neurological symptoms or signs, or alterations in personality or mentation may predict underlying central nervous system metastases. Subsequent work-up should include contrast-enhanced cerebral imaging. MRI is preferable to CT scan as it is more sensitive. Thin cuts may further increase the sensitivity of the study. Attention should be paid not only to mass lesions and hydrocephalus, but also to pathological enhancement of the leptomeninges and cranial nerves, which may be indicative of carcinomatous meningitis. Other factors that may contribute to headache include depression, anxiety, and treatment-related factors.

Chapter 5 / Headache

61

Table 4 Causes of Headache in Cancer Patients Tumor-related causes Acute

Chronic

Nontumor-related causes Treatment-related causes Chemotherapy

Radiotherapy

Supportive therapies

Surgery Other causes of headache Acute

Chronic

Intratumoral hemorrhage Acute venous sinus thrombosis CSF obstruction with resulting increase in intracranial pressure Pressure wave headache Persistent or new tumor growth New metastatic lesion involving skull, meninges, brain, skull base, sinuses, orbits, etc. Invasion of tumor into calvarium, skull base, meninges, leptomeninges. Increased intracranial pressure with midline shift causing traction on veins, arteries, nerves, etc.

Hormones (e.g., tamoxifen) Differentiation agents (retinoic acids) Antibiotics Reverse transcriptase inhibitors (e.g., AZT, DDI) Conventional agents (e.g., L-asparaginase, procarbazine, PCNU, fludarabine, fazarabine, caracemide, gallium nitrate) Cytokines (e.g., tumor necrosis factor, OKT3 , interferons, interleukins, levamisole, GM-CSF) Intrathecal therapy (e.g., methotrexate, Ara-C) Acute cerebral edema (early onset during radiotherapy) Radionecrosis (late onset commonly months to years after radiotherapy) Radiation-induced neoplasm (late: years after radiotherapy) Radiation-induced atherosclerosis causing stroke Corticosteroids, cimetidine, ondansetron, narcotics (withdrawal), metoclopramide, anticoagulants (intratumoral hemorrhage), dipyridamole, ibuprofen (aseptic meningitis) Hemorrhage, vascular injury, perioperative stroke, cerebrospinal fluid leak Cerebral infarcts Fever Infection (abscess, meningitis) Metabolic (hypoxemia, hypercarbia, and hypoglycemia) Referred pain from extracranial structures (cervical metastases, lung tumors, etc.) Postlumbar-puncture headache

7. SELLAR TUMORS Tumors of the sellar region, including pituitary tumors and craniopharyngiomas, are often associated with headaches. Close proximity of the tumor to the optic nerves and chiasm and hypothalamus often results in visual disturbances and endocrinological dysfunction, respectively (17). Consequently, patients may have concomitant visual field cuts or blindness or various systemic manifestations of pituitary dysfunction (e.g., growth disorder, sexual dysfunction, diabetes insipidus, testicular atrophy, galactorrhea, or menstrual disorders). In addition, close proximity to the cavernous sinus, in which cranial nerves III, IV, V, and VI course, may lead to dysconjugate gaze with diplopia and facial numbness. A severe, sudden-onset headache associated with visual disturbances and vomiting may occur with hemorrhage into a pituitary adenoma (pituitary apoplexy) (18). Levy et al. studied 84 patients presenting to a neurosurgical center with pituitary tumors and troublesome headaches (19). The most common location of headaches was in the frontal and orbital/retro-orbital regions.

62

Part III / Neurologic Symptoms

In 71% of patients, the trigeminal territory was exclusively involved. The most common symptoms included photophobia and nausea. Autonomic symptoms, such as lacrimation and conjunctival injection, were present in 50% of patients. Cranial nerve dysfunction was not reported in any patients. Headaches in 76% of patients fit the criteria of migraine. Interestingly, 49% of patients had a family history of headaches. The authors suggest that the presence of a pituitary tumor may have lowered the threshold for attacks in predisposed migraineurs (19). Other authors have also reported an association of brain tumors and migraines (20). Alternatively, the pituitary tumor may have been an incidental discovery although >50% responded to surgical or medical tumor-directed treatment. Cavernous sinus invasion was present in only 21% of patients (19). The pituitary size (21) and extent of cavernous sinus invasion (21,22) did not correlate with the occurrence of headache in patients with pituitary tumors. It has been suggested that pituitary-related tumor headache has a biochemical–neuroendocrine basis (21). Alternatively, the presence of headache correlated with median intrasellar pressure measured at the time of surgery. Furthermore, intrasellar pressure did not correlate with pituitary size (23). The most common presenting symptom among 93 patients with craniopharyngioma was headache with visual and/or endocrinological related symptoms often coexisting (17).

8. OTHER CAUSES OF HEADCAHES Other, indirect causes of headaches in cancer patients must also be considered. Chemotherapeutics, such as retinoids and cytokines, and supportive medications including 5-HT3 receptor antagonists, commonly cause headaches (24). Radiotherapy may acutely cause an increase in tumor related edema and chronically incite radiation necrosis, both of which may be managed with dexamethasone. Neurosurgical procedures may also cause pain, usually within the post-operative period (25). Cerebrospinal fluid leaks may also lead to chronic, severe positional headaches. Infection, including meningitis and abscess, must be considered, especially in immunosuppressed patients who have undergone neurosurgical procedures. Intrathecal chemotherapy, particularly with liposomal cytarabine (Depocyt), may result in chemical meningitis. Prophylactic corticosteroids minimize this side effect (26). Venous sinus thrombosis may occur as a complication of tumor invasion or compression. Venous sinus thrombosis has also been attributed to treatment; classically this has been described among patients treated with l-asparaginase, which may induce a hypercoaguable state (27,28). Among patients with underlying brain tumors, headaches may re-emerge as corticosteroids are withdrawn.

9. HEADACHE MANAGEMENT Simple analgesics may be sufficient to palliate patients with tumor-related headaches. Such interventions were found to be effective in 42–58% of patients (1,2). These agents are inexpensive and are possibly associated with fewer side effects than opioids. If headaches are refractory to such measures, opioids may be indicated. Typically, these agents are combined with analgesics. Options include morphine, hydromorphone, oxycodone, and fentanyl. Patients should be instructed to maintain a medication diary. Dosages should be titrated until acceptable palliation is achieved, and conversion to extended release agents should be considered depending on patient consumption. The nonopioid constituent rather than the opioid limits dosing of combination agents. In fact, there is no ceiling for opioids. Patients should be educated regarding side effects including constipation and fatigue, and appropriate prophylactic measures should be taken. Addiction, often a concern among patients and families, is rare. Dexamethasone may be indicated for patients with a large disease burden particularly if there is significant edema. The optimal dose is uncertain. Typical starting doses of 4 mg two times per day are not unreasonable, although higher doses may be necessary. After palliation has been achieved, every attempt to taper the dexamethasone dose should be made. The lowest dose necessary to palliate the patient’s symptoms effectively should be sought. Clinicians should be attentive to side effects including hyperglycemia and hypertension, dyspepsia, insomnia and, more rarely, psychosis, excessive weight gain, and osteopenia, among others. Prophylactic measures may be required. Tumor-directed treatments are effective at controlling tumor-related headaches. Perhaps the most effective intervention is surgical resection that rapidly decompresses the brain, thereby reducing intracranial pressure. Other measures, including radiotherapy, may also improve pain control. Data on specific symptom relief are lacking.

Chapter 5 / Headache

63

10. SUMMARY Headaches are a common symptom of brain tumors. They may occur in brain tumors of all types but affect patients with infratentorial and more rapid growing lesions most often. Traction of pain-sensitive structures is the presumed mechanism of pain. Elevated intracranial pressure is frequently present. Although the pain may be nonspecific, some characteristics may suggest an underlying structural lesion. Headaches are usually subacute and intermittent with moderate to severe intensity. The pain can usually be palliated with conservative measures such as simple analgesics. More aggressive interventions such as corticosteroids, opiods, and surgical resection may be required. Clinicians must recall, however, that not all headaches are directly related to the underlying tumor. Treatment-related and psychological factors may also play a role and thus should be considered when evaluating such patients. Also, benign headache syndromes, such as migraines and tension headaches, which commonly affect the general population, may still impact tumor patients and in fact be exacerbated by the underlying illness.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Suwanwela N, Phanthumchinda K, Kaoropthum S. Headache in brain tumor: a cross–sectional study. Headache 1994;34(7):435–438. Forsyth PA, Posner JB. Headaches in patients with brain tumors: a study of 111 patients. Neurology 1993;43(9):1678–1683. Pfund Z, Szapary L, Jaszberenyi O et al. Headache in intracranial tumors. Cephalalgia 1999;19(9):787–790; discussion 65. Purdy RA, Kirby S. Headaches and brain tumors. Neurol Clin 2004;22(1):39–53. Wilne SH, Ferris RC, Nathwani A et al. The presenting features of brain tumors: a review of 200 cases. Arch Dis Child 2006;91(6): 502–506. Boiardi A, Salmaggi A, Eoli M et al. Headache in brain tumors: a symptom to reappraise critically. Neurol Sci 2004;25 Suppl 3:S143–S147. Vazquez-Barquero A, Ibanez FJ, Herrera S et al. Isolated headache as the presenting clinical manifestation of intracranial tumors: a prospective study. Cephalalgia 1994;14(4):270–272. Childhood Brain Tumor Consortium, Gilles FH. The epidemiology of headache among children with brain tumor: headache in children with brain tumors. J Neurooncol 1991;10(1):31–46. Argyriou AA, Chroni E, Polychronopoulos P et al. Headache characteristics and brain metastases prediction in cancer patients. Eur J Cancer Care (Engl) 2006;15(1):90–95. Christiaans MH, Kelder JC, Arnoldus EP et al. Prediction of intracranial metastases in cancer patients with headache. Cancer 2002;94(7):2063–2068. Nussbaum ES, Djalilian HR, Cho KH et al. Brain metastases: histology, multiplicity, surgery, and survival. Cancer 1996;78(8): 1781–1788. Zimm S, Wampler GL, Stablein D et al. Intracerebral metastases in solid-tumor patients: natural history and results of treatment. Cancer 1981;48(2):384–394. Balm M, Hammack J. Leptomeningeal carcinomatosis: presenting features and prognostic factors. Arch Neurol 1996;53(7):626–632. Clouston PD, DeAngelis LM, Posner JB. The spectrum of neurological disease in patients with systemic cancer. Ann Neurol 1992;31(3):268–273. Fizazi K, Asselain B, Vincent–Salomon A et al. Meningeal carcinomatosis in patients with breast carcinoma: clinical features, prognostic factors, and results of a high–dose intrathecal methotrexate regimen. Cancer 1996;77(7):1315–1323. Chamberlain MC. Leptomeningeal metastases: a review of evaluation and treatment. J Neurooncol 1998;37(3):271–284. Larijani B, Bastanhagh MH, Pajouhi M et al. Presentation and outcome of 93 cases of craniopharyngioma. Eur J Cancer Care (Engl) 2004;13(1):11–15. Randeva HS, Schoebel J, Byrne J et al. Classical pituitary apoplexy: clinical features, management, and outcome. Clin Endocrinol (Oxf) 1999;51(2):181–188. Levy MJ, Matharu MS, Meeran K et al. The clinical characteristics of headache in patients with pituitary tumors. Brain 2005;128(Pt 8):1921–1930. Schlake HP, Grotemeyer KH, Husstedt IW et al. “Symptomatic migraine”: intracranial lesions mimicking migrainous headache: a report of three cases. Headache 1991;31(10):661–5. Levy MJ, Jager HR, Powell M et al. Pituitary volume and headache: size is not everything. Arch Neurol 2004;61(5):721–725. Abe T, Matsumoto K, Kuwazawa J et al. Headache associated with pituitary adenomas. Headache 1998;38(10):782–786. Arafah BM, Prunty D, Ybarra J et al. The dominant role of increased intrasellar pressure in the pathogenesis of hypopituitarism, hyperprolactinemia, and headaches in patients with pituitary adenomas. J Clin Endocrinol Metab 2000;85(5):1789–1793. Einhorn LH, Nagy C, Werner K et al. Ondansetron: a new antiemetic for patients receiving cisplatin chemotherapy. J Clin Oncol 1990;8(4):731–735. Gee JR, Ishaq Y, Vijayan N. Postcraniotomy headache. Headache 2003;43(3):276–278. Glantz MJ, Jaeckle KA, Chamberlain MC et al. A randomized controlled trial comparing intrathecal sustained-release cytarabine (DepoCyt) to intrathecal methotrexate in patients with neoplastic meningitis from solid tumors. Clin Cancer Res 1999;5(11):3394–3402. Priest JR, Ramsay NK, Bennett AJ et al. The effect of l-asparaginase on antithrombin, plasminogen, and plasma coagulation during therapy for acute lymphoblastic leukemia. J Pediatr 1982;100(6):990–995. Priest JR, Ramsay NK, Steinherz PG et al. A syndrome of thrombosis and hemorrhage complicating l-asparaginase therapy for childhood acute lymphoblastic leukemia. J Pediatr 1982;100(6):984–989.

6

Confusion and Delirium Augusto Caraceni MD, Marco Bosisio PHD, and Jane M. Ingham MB, BS, FRACP, FACHPM CONTENTS Introduction History and Terminology Pathophysiology Clinical Features Diagnostic Criteria and Classifications for Delirium Differential Diagnosis of Delirium Diagnostic Tools and Instruments for Assessment of Delirium Etiology and Risk Factors Delirium Treatment Prognosis Conclusions References

Summary Delirium is a highly prevalent disorder among the medically ill. The cancer patient is at risk for developing delirium as a consequence of several general and numerous specific factors. Delirium can be responsive to interventions that address its etiologic factors, or it may be an irreversible event characterizing the final evolution of terminal illness. Neurologists involved in the assessment of oncologic patients need to have an understanding of both oncology and palliative care if they are to be able to conduct a comprehensive assessment of the patient and embark upon an appropriate treatment plan. The breadth of the etiologic factors and the risks of the condition highlight the frequent need for multidisciplinary interaction with other medical specialties and health professionals both to treat the condition and minimize risk. Key Words: delirium, confusion, cancer, palliative care, encephalopathy, psychosis

1. INTRODUCTION Delirium is one of the most frequent neurologic complications occurring in patients with cancer, ranking second after pain among the reasons for requesting a neurologic consult at a tertiary cancer center (1). Inappropriate behavior, cognitive disturbance, and lack of judgment can be distressing for patients and families, and can impact on attempts to deliver optimal medical care. In addition, delirium is usually a sign of the presence of significant medical complications and is associated with increased mortality. The signs and symptoms of this disorder can be diverse and are sometimes mistaken for other psychiatric disorders including mood and anxiety disorders. Different terms have been used to denote delirium over the years, including confusion, acute confusional states, acute brain failure, acute dementia, acute organic syndrome, cerebral insufficiency, metabolic encephalopathy, From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

65

66

Part III / Neurologic Symptoms

organic brain syndrome, reversible toxic psychosis, and intensive care unit psychosis (2). This inconsistent terminology has served to contribute to the insufficient understanding of delirium in clinical and research settings (2). This chapter will describe the terminology used for delirium, the symptoms that characterize the condition and its diagnosis, etiologic factors, treatment, and prognosis.

2. HISTORY AND TERMINOLOGY Delirium has been defined as a transient organic brain syndrome characterized by acute disturbances in attention, cognition, psychomotor behavior, and perception (2). The term delirium is derived etymologically from the Latin: de meaning “down” or “away from” and lira meaning a “furrow or track in the fields.” It thus means to be “off the track” (2). Historically, the term was first used by Celso in the first century as being synonymous with phrenitis, a term previously used by Hippocrates (who is known to have well-described the syndrome of an agitated state with compromised cognition occurring in the course of fevers) (2,3). This state was opposed to lethargus, a term used to describe a state that today may be described as “somnolent hypoactive delirium.” While the history of the concept developed over the centuries (2), it was only in the nineteenth century that the difference between symptoms due to altered perception and cognition in the course of fevers (and other organic conditions including surgical interventions, poisoning, and alcohol withdrawal) was clearly distinguished from similar symptoms due to other psychiatric conditions. This development led to the present use of the word delirium as a separate term from delusion. The contribution of French literature to the evolution of psychiatric and neurologic classifications was fundamental in that it highlighted the concepts of an “acute confusional state,” introduced by Chaslin, and an “oneiric consciousness,” used by Regis (4). This led to the distinction between the symptoms of acute confusion and agitation when occurring in the context of acute medical illness and those same symptoms occurring in the psychopathological state (4). During this same period, Hughling Jackson proposed a model to explain confusional states based on a hierarchical structure of the central nervous system (CNS) (5). Mental confusion in his model was seen as a separate syndrome from other psychiatric conditions. In this model the occurrence of confusion was attributed to the loss of voluntary control and intellectual functioning with the associated clinical findings deriving from psychological automatisms of lower brain functions “released” as a consequence of lack of control of higher brain functions. Recent views have generally considered the global failure of cognitive processes to be at the core of the etiology of delirium, occurring in an “acute” form opposed to the chronic progressive form of cognitive dysfunction that is found in dementia. Two other recent theories have been proposed to explain the pathophysiology of delirium. Some authors, rather than adhering to the theory of a “global failure of cognition,” have considered delirium to be the consequence of a selective failure of attention (6). Others, given the compromise that occurs during delirium in vigilance and level of consciousness, have described it as an abnormality in the level of consciousness that occurs on the continuum between normal wakefulness and coma (7). It is known that acutely altered mental states, despite an array of different etiologic triggers, can be characterized by a common “spectrum” of fluctuating clinical phenomena. In addition, symptoms can fluctuate and the whole “range” of symptoms may or may not be present in one individual. As a consequence of this clinical observation and the history described above, there has been persistence in the literature of two terms or categories: “delirium” and “acute confusional state (or encephalopathy).” Some authoritative neurologists have preferred to keep these states separate, with “delirium” being used to describe acute mental status changes characterized by agitated hallucinated phenomena—delirium tremens (DT) being the prototype—and “acute mental confusion” or “encephalopathy” to describe those states characterized by confusion but not characterized by agitated phenomena (8). Although the terms delirium, acute confusional state, and encephalopathy persist, Engel and Romano have demonstrated that the electroencephalographic (EEG) correlates of most acute confusional states and deliria are very similar (although it is worth noting that the EEG findings specific to DT are somewhat different, with unique and characterizing features) (9). Their work has set the scientific basis for a unified concept of delirium, a concept that has been further developed by Lipowski (2). This work has evolved and now underlies the definitions of delirium found within the Diagnostic Statistical Manual of Mental Disorders (DSM) (10–12). The latter is a document produced by the American Psychiatric Association that aims to provide a standard classification for psychiatric conditions. In addition, the International

Chapter 6 / Confusion and Delirium

67

Classification of Diseases (ICD-10) also provides a definition of delirium (13). Both definitions recognize the spectrum of phenomena that can occur among and within cases of this condition, and both recognize that it is a transient syndrome characterized by acute disturbances in attention, cognition, psychomotor behavior, and perception (see details below for each definition). Although the terminology utilized to describe states characterized by acute confusion continues to be inconsistent in the medical literature, the developments to date have set the stage for the adoption of a unitary concept of delirium—a condition characterized nonetheless by a spectrum of clinical phenomena.

3. PATHOPHYSIOLOGY The pathophysiology of delirium has not been fully identified, but an understanding of several anatomical functions and neurotransmitter substrate disturbances is important if we are to interpret the clinical findings. It is difficult to attribute the abnormal functioning seen in delirious states to discrete cerebral structures. Indeed, numerous areas of neurologic dysfunction have been identified as potential contributors to this condition. One debated interpretation is that the syndrome is caused by the ability of different etiological factors to impact on a final common pathway producing stereotyped clinical consequences (14,15). One fundamental neurologic concept must be recognized if the pathophysiology of delirium is to be understood: the functional distinction, on both clinical and neuroanatomical grounds, of arousal and alertness from cognition and awareness. (Arousal and alertness are generally considered to be synonyms reflective of level of consciousness or wakefulness, and therefore the term arousal will be utilized from herein). In simplistic terms arousal may be considered to reflect a general activation of the brain cortical functions by the action of subcortical structures linked also with the regulation of the sleep-wake cycle. Arousal can be present despite cognitive failure but arousal is a prerequisite for clear cognition. For example, arousal can be preserved in patients with profound cognitive failure triggered by dementia, and even more dramatically in patients without any sign of active cognition or awareness, such as those in a vegetative state. Another clinical observation that must be acknowledged in considering the potential pathophysiologic explanations for delirium is that cases of delirium in which the cause cannot be removed frequently evolve to stupor and coma. While this clinical course may reflect the underlying pathology triggering the delirium, it can also be viewed as lending support to an interpretation of delirium pathophysiology that considers the cerebral structures involved in the modulation of arousal to be dysfunctional in delirious states (15). It has been speculated that selective altered arousal may be the trigger of all or most of the other disordered brain functions in delirious states. This concept requires that it be possible to distinguish brain structures (or functions of specific structures) that are responsible for a basic form of brain arousal (synonyms: activation, level of consciousness, alertness, wakefulness) from those structures responsible for clear cognition manifest by the “higher” cognitive processes such as emotion, memory, and the ability to accurately interpret sensory phenomena. As mentioned above, arousal is considered to be a prerequisite for such clear cognition. The pioneering work by Moruzzi and Magoun has described a primary role of brainstem structures in regulating arousal and sleep-wake cycle and indeed provides an anatomical basis upon which the theory of delirium being based on a disturbance of these structures can be put forward (16–18). Fukutani et al. reported three cases of patients with Joseph disease in which delirium occurred during the course of illness (19). Two necropsies in this series showed degeneration of the reticular formation—raphe nuclei and locus ceruleus in the brainstem tegmentum (19). In addition to evidence that focal structures can control arousal, it is also known that any type of lesion diffusely affecting the brain will impact on the level of consciousness. It has also been show that integrative brain functions are lost proportionally to the amount of brain matter lost and also in relation to the rapidity of the onset of the pathological insult (7). The selective deficits in neurologic function that can occur during the onset of delirium, in the absence of arousal disturbance, suggest the presence of a more diffuse, multifocal process, affecting numerous structures other than the brainstem. A failure of selective attention, which is the ability to select in the environment significant stimuli and to focus attention for a protracted time, is found in delirium and its very frequent presence has led to the suggestion that this be considered to be an essential feature of the syndrome (12). Selective lesions of cortical association areas in the right, nondominant, cerebral hemisphere can produce acute confusional states (20). Other symptoms typical of delirium such as language and memory alterations, disruption

68

Part III / Neurologic Symptoms

of the wake-sleep cycle, or hallucinations can be early findings in the course of the syndrome and also isolated findings of partial syndromes. These presentations suggest focal origins for a diffuse global syndrome triggered by metabolic derangements that first affect the most sensitive cerebral structures causing focal symptoms that subsequently combine with other symptoms and signs to be manifest as the full delirium syndrome. In this latter situation arousal and consciousness can also become impaired (15). One recent pathophysiologic interpretation suggests that delirium can be operationally defined by the impairment of functions regulating arousal, temporal orientation, and attention (21). This hypothesis appears to be supported by the recent findings of a study of over 500 patients with delirium that showed that among the 20 items of the Mini-Mental State Examination (MMSE) only four items (temporal orientation to year and to date, spelling a word backwards, and copying a design) were needed to screen for cases of delirium with reasonable accuracy (22,23). Many neurotransmitters have been implicated in the pathogenesis of delirium. Abnormalities in cholinergic neurotransmission, overstimulation or understimulation of the gamma-aminobutyric acid (GABA) system, Nmethyl-d-aspartate (NMDA) receptor blockade, serotonin antagonism, and overstimulation of dopaminergic pathways have all been described in association with etiologically diverse forms of delirium (15). Currently, the neurotransmitter system that is most commonly cited in relation to its involvement in delirium pathophysiology is the cholinergic system (18). The cholinergic projection pathways from the brainstem to the thalamus, cortex, and hypothalamus are implicated in the sedative properties of many drugs, and their dysfunction has been demonstrated in dementing illnesses such as Alzheimer’s disease and Lewy body dementia (18). Further, in some cases, the use of cholinesterase inhibitors seems to improve some of the symptoms of dementia that are also found in delirium (24). It has been known for some time that anticholinergic drugs can cause delirium while many other drugs that also have anticholinergic activity have been associated with delirium. In addition, hypoglycemia, hypoxia, ischemia, and other toxic metabolic and nutritionally induced insults impact on the function of neurotransmitters in the cholinergic system, primarily on acetylcholine (25). The cellular links among neurotransmitter abnormalities, altered brain metabolism, and the clinical manifestations of delirium are most likely based on alterations of the second messenger system involving calcium, cyclic GMP, and/or the phosphatidylinositol cascade (25,26). One other proposed theory for the cerebral effect of infectious states or altered immune reactions is that neurotransmitter synthesis and neurotransmission may be directly or indirectly influenced by the effect of cytokines. Although the association between cytokine release and delirium remains to be fully determined (27,28), this hypothesis could provide an explanation for why immunomodulation therapy with interferon- and interleukin-2 has a relatively high rate of associated cognitive, emotional, and behavioral disturbances. Although delirium may occur as a consequence of a variety of pathophysiologic processes or events, most of these processes and events interfere with oxygen–glucose metabolism at the cerebral level, which in turn affects the function of groups of neurons that are necessary for normal arousal and cognitive activities. In many individuals there are not only multiple potential causes of delirium but also some predisposing factors that may compound the clinical manifestations. For example, the degeneration of the cholinergic system—typical of brain aging and Alzheimer’s disease—may explain why both of these conditions are recognized risk factors for developing delirium. Other metabolic factors and impairments in oxidative metabolism can also affect cholinergic transmission. An increase in serum anticholinergic activity has been associated with the onset of delirious episodes due to many different exogenous toxic or pharmacological factors and this has also been observed in course of fevers (29). Another example of a potentially concurrent causal factor is the presence of thiamine deficiency, a condition that is more common in the elderly and in medically ill institutionalized patients (30–32). These observations have practical value in assessing the role of potential causal factors in precipitating delirium and support a treatment approach that focuses on the minimization of any potentially reversible cause of delirium (including assessing the need for each medication that is being administered to the patient with respect to its potential to precipitate anticholinergic effects or result in other neurologic dysfunction). The presence of concurrent etiologic factors with the potential to impact on differing aspects of neural functioning may create in an individual a particular sensitivity to an insult that, alone, may not have been associated with severe alterations in functioning (33). In summary, in a particular individual, multiple potential causes of

Chapter 6 / Confusion and Delirium

69

delirium may be present. Alone, each of these etiologic factors may not be sufficient to cause the syndrome, but the clinical manifestations and the individual reaction to the etiologic insult may be modulated as a consequence of host biological variables (general predisposing factors) in play together with environmental and psychological factors. Regarding general predisposing factors, it is hoped that further studies will shed light on the detail of these factors. As an example, an interesting recent study sought to explore the relative risk for delirium related to pre-operative impairment in memory and executive function in patients undergoing cardiac surgery (34). This study found that the risk for delirium was linked with preoperative mental status and within that context, with the pre-morbid executive functioning impairment and not with pre-existing memory impairment (34). This type of study may help not only to clarify risk factors for delirium but to shed light on pathophysiologic mechanisms that may be linked to delirium. Clearly there is a need for a greater understanding of the pathophysiology of delirium, especially at the cellular and neurotransmitter levels. As such an understanding evolves, it may facilitate the development of treatment strategies aimed at specific aberrations in second messenger system, neurotransmitters, and the like (35).

4. CLINICAL FEATURES 4.1. Epidemiological Aspects Epidemiological data have led to a greater understanding of the magnitude of the problem of delirium in different settings of care. It has been estimated that the prevalence of delirium in acute general hospitals is around 10% (2). The prevalence is higher in intensive care, cardiac surgery, geriatric, and burns units where it is often reported as being around 30–40% with some reports of it being of even higher prevalence (2,36,37). In the general oncology population, a recent prospective, observational study found that delirium occurred during the course of hospital admission in 18% of patients (38). The prevalence of delirium clearly varies greatly depending on a variety of factors including, among others, the severity of illness and the age of the patient population. For example, in the advanced phases of cancer the prevalence of delirium is around 30% and can reach as high as 70–90% prior to death (39–42). One important factor that has facilitated a more accurate description of the epidemiology of delirium is the fact that the syndrome can now be diagnosed through the use of specific diagnostic criteria (43). With the use of these more accurate criteria, it has become apparent that under-reporting of delirium has been a problem. One study, in an emergency care unit setting, examined delirium screening scores collected by a screening nurse independently of the emergency room physicians (44). This study demonstrated that only 25% of cases of delirium in the elderly were correctly identified by physicians and fewer (7%) were identified when the delirium screening score was reflective of “probable” delirium (44).

4.2. Description The clinical features of delirium include a variety of neuropsychiatric symptoms that are common in depression, dementia, and psychoses and these features have been described in detail in the past by Lipowski and reviewed recently by Caraceni (2,45). The onset of delirium is often preceded by subtle mood or personality changes that may go unnoticed by professional caregivers but be observed and reported by family. This prodromal phase may be characterized by one or more symptoms including fatigue, irritability, insomnia, daytime somnolence, malaise, headache, hypersensitivity to visual and auditory stimuli, illusions, vivid dreams, nightmares, low concentration and attention, and mild disorientation (2). The patient may be aware of these changes but may discount their significance, or may even masquerade the symptoms (2,45). Variability in clinical findings is typical with a wide range in the severity and nature of symptoms exhibited (2,45). It is common for symptoms to be present through the course of the day with nocturnal exacerbations. It is not uncommon for the first symptoms of delirium to occur at night. Reactions manifesting as anxious, aggressive, or depressive features are frequently noted at the onset of cognitive failure, but other less “dramatic” features including, among others, apathy and affective inhibition may also be present (2,45). From this early stage the patient may proceed into remission or progress to more severe symptoms. The variations in symptoms have led to a classification of delirium as hyperactive, hypoactive, and mixed type.

70

Part III / Neurologic Symptoms

The onset of delirium is, by definition, acute or subacute—within hours to days—and fluctuations in symptoms are common (2,45). The time course is protracted and it may be difficult to recognize an acute phase. Attention failure is characteristically present and manifests as easy distractibility, difficulty in focusing and maintaining attention, and frequent shifts in the focus of attention (2,45). Perseveration is common. Memory deficits are frequent. Orientation to time is usually lost before orientation to place and person (2,45). Clarity of thought is altered and speech may be incoherent, rambling, and fragmented. Poorly systematized delusions may be found (2,45). Delusions, when they occur, are often persecutory, influenced by elements of the environment, but are commonly not recalled by the patient (46). Perception is abnormal at times with illusions and hallucinations, the latter frequently being visual rather than auditory (46). Erroneous identifications are common. In the hypoactive variant of delirium the patient can be somnolent—in contrast to the hyperactive form in which the patient can appear excited or hyperalert. In the latter case the patient is often excessively reactive to external stimuli. Light and sound may be perceived by the patient as being unpleasant stimuli. Illusions, hallucinations, and incorrect interpretations are common (2). In the hypoactive form the patient is usually lethargic with little interest in external stimuli. Although hallucinations and delusions may be present, the patient may not manifest or report them. Very frequently the hypo- and hyperactive forms are not distinct subtypes but indeed may be manifestations of delirium that can occur in an individual patient with fluctuations or evolutions in the clinical findings over time. These “mixed hypo-hyperactive” forms have generally been considered to be the most frequent delirium variant (14,47), but in some populations this may be different. It is worth noting that it is possible that the clinical features noted at presentation may not reflect the course of a case of delirium. A recent study in a palliative care population found that between 78% and 86% of cases in this population had hypoactive features at presentation (48). A recent study in a medical ICU study found the mixed type of delirium to be the most common type overall and that elderly patients were more prone than younger patients to hypoactive delirium (37). In contrast, another ICU study in a surgical and trauma population found hypoactive delirium was more prevalent than the mixed and the hyperactive delirium (49). Finally, and importantly from a palliative care perspective, recent work by Breitbart et al. has explored the recalled distress of patients with delirium in a cancer population (50). Over half of the patients (53.5%) who recovered from delirium recalled the experience and, of these, 80% reported that the experience had been associated with severe distress (50). Patients were less likely to recall the delirium when it was associated with severe short-term memory impairment and the presence of perceptual disturbances. Interestingly, patients with “hypoactive” delirium were likely to be just as distressed as patients with “hyperactive” delirium (50). Further studies will be needed to clarify the impact of various treatment strategies on distress.

4.3. Other Neurologic Signs As discussed above, delirium is primarily characterized by cognitive impairment, but other neurologic signs may also occur. For example, tremors and myoclonus can be associated with delirium. Asterixis is typically found in hepatic encephalopathy but may also occur in other types of delirium. This latter sign can be seen best when the patient is asked to hold arms and hands extended for at least 30 seconds. Another sensitive alternative maneuver in the fatigued patient consists of asking him or her to keep the index finger extended with the hand prone on a surface. Multifocal myoclonus and asterixis are signs that may occur early in cases of opiate toxicity and in other metabolic conditions. Frontal release reflexes (palmomental, glabella, snout, grasping) are also potentially useful findings. These signs, in young patients, are more indicative of an acute underlying neurologic problem, but in some patients (especially in the elderly) these signs may have been present prior to the onset of delirium, in which case they are likely to have another etiology. Impaired writing and design abilities are known to be very sensitive indicators of dysfunction of the higher integrative brain function, and screening for these can assist in the detection of early symptoms of delirium. One of the following—the classic clock drawing test, reproduction of a geometric figure as required by the Mini-Mental State Examination, or writing a complete sentence—should be a standard part of the neurologic examination in a confused patient (51,52).

Chapter 6 / Confusion and Delirium

71

5. DIAGNOSTIC CRITERIA AND CLASSIFICATIONS FOR DELIRIUM A well-defined diagnostic system is very important to allow standardized assessment and to ensure accuracy in clinical reporting and communication. Such a system is also crucial for research in this field. The diagnosis of delirium has, recently, relied on the classification of the DSM-IV (12) and much of the recent research on this subject has used this, or the earlier, DMS-III-R definition (11). In 2000, the DSM-IV-TR was published with revisions to some psychiatric diagnoses although the classification of delirium was not changed significantly (53). Both the DSM-IV and the DSM-IV TR include delirium in the same chapter as dementia and amnestic disorders, based on the fact that impairment of cognitive functions is typical of each of these syndromes. There is overlap of the psychopathological features in delirium and many other conditions, but through adherence to the DSM criteria diagnostic accuracy can be facilitated and will allow the distinction to be made between this and other psychiatric conditions. The main diagnostic systems in use, in addition to the DSM IV, are the DSM III-R and the ICD-10 (11–13,53). Although the DSM III-R is not the most recent version of the DSM, many recently published reports reflect work done using this classification criteria. These criteria, however, are slightly different from those described in the more recent DSM IV and DSM IV-TR. • DSM III-R (11)—In DSM III-R, delirium is considered to be an organic brain syndrome and the criteria for diagnosis are listed in Table 1. The differential diagnosis from psychiatric conditions and the biological correlates of delirium are not considered specifically. • DSM IV (12)—In this version of the DSM delirium is considered as a single nosological entity, the term organic brain syndrome is abandoned, and the diagnostic criteria are simplified with a focus on two main areas of brain function derangement—consciousness and attention, and cognition (Table 2). Specific symptoms and secondary manifestations are not included. It is worth noting that in the DSM IV the impairment of the level of consciousness (a concept that can overlap with arousal) is a primary characteristic of the syndrome. Of note, there are subclassifications of delirium within the DSM-IV and DSM-IV-TR that relate to substance intoxication and substance withdrawal delirium and delirium due to a medical condition or due to multiple etiologies. (Of note, the latter is extremely common in patients with advanced malignant disease.) • ICD-10 (13)—The International Classification of diseases classifies delirium that is not due to alcohol or other psychoactive substances, among the “organic psychic syndromes and disturbances including the symptomatic ones” Table 1 DSM IIIR Criteria for Diagnosing Delirium A. Reduced ability to maintain attention to external stimuli (e.g. questions must be repeated because attention wanders) and to appropriately shift attention to new external stimuli (e.g., perseverates answer to previous question). B. Disorganized thinking as indicated by rambling, irrelevant or incoherent speech. C. At least two of the following: 1. Reduced level of consciousness, e.g., difficulty keeping awake during examination; 2. Perceptual disturbances: misinterpretations, illusions, or hallucinations; 3. Disturbances of sleep–wake cycle with insomnia or daytime somnolence; 4. Increased or decreased psychomotor activity; 5. Disorientation to time, place or person; 6. Memory impairment, e.g., inability to learn new material such as the names of several unrelated objects after 5 minutes, or to remember past events, such as history of the present disease. D. Clinical features develop over a short period of time (usually hours to days) and tend to fluctuate over the course of a day E. Either (1) or (2): 1. Evidence from the history, physical examination or laboratory tests of a specific organic factor (or factors) judged to be etiologically related to the disturbance; 2. In the absence if such evidence, an etiological organic factor can be presumed if the disturbance cannot be accounted for by any nonorganic mental disorder, e.g., manic episode, accounting for agitation and sleep disturbance. Printed with permission from the •Diagnostic and Statistical Manual of Mental Disorders. •American Psychiatric Association. Copyright 1987 (11).

72

Part III / Neurologic Symptoms

Table 2 DSM IV Criteria for Diagnosing Delirium Due to a General Medical Condition A. Disturbance of consciousness with reduced ability to focus, sustain and shift attention. B. Change in cognition (such as memory deficit, disorientation, language disturbances) or the development of a perceptual disturbance not better explained by a preexisting stabilized or evolving dementia. C. The disturbance develops over a short period of time and tends to fluctuate during the course of the day. D. There is evidence from the history, physical examination or laboratory findings that the disturbance is caused by the direct physiological consequences of a general medical condition. Printed with permission from the •Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association. Copyright 1994 (12).

Table 3 ICD-10 Criteria for Diagnosing Delirium, Not Induced by Alcohol and Other Psychoactive Substances [Includes Acute Brain Syndrome, Acute Confusional State (Nonalcoholic), Acute Infective Psychosis, Acute Organic Reaction, Acute Psychoorganic Syndrome] For a definite diagnosis, symptoms, mild or severe, should be present in the following areas: a. Impairment of consciousness and attention (on a continuum from clouding to coma; reduced ability to direct, focus, sustain, and shift attention); b. Global disturbance of cognition (perceptual distortions, illusions and hallucinations—most often visual; impairment of abstract thinking and comprehension, with or without transient delusions, but typically with some degree of incoherence; impairment of immediate recall and of recent memory but with relatively intact remote memory; disorientation for time as well as, in more severe cases, for place and person); c. Psychomotor disturbances (hypo- or hyperactivity and unpredictable shifts from one to the other; increased reaction time; increased or decreased flow of speech; enhanced startle reaction); d. Disturbance of the sleep–wake cycle (insomnia or, in severe cases, total sleep loss or reversal of the sleep–wake cycle; daytime drowsiness; nocturnal worsening of symptoms; disturbing dreams or nightmares, which may continue as hallucinations after awakening); e. Emotional disturbances, e.g. depression, anxiety or fear, irritability, euphoria, apathy, or wondering perplexity. Reproduced with permission from the World Health Organization. The ICD-10 Classification of Mental and Behavioral Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva: WHO Publications; 1992 (13).

(Table 3). In this classification delirium is considered to be a syndrome that is characterized by disturbances of consciousness, cognition, behavior and sleep-wakefulness cycle, rapid onset, transient course, and fluctuation of symptoms. The etiology of the delirium does not necessarily need to be specified for an ICD-10 diagnosis. If this classification is used the diagnosis requires that all areas of dysfunction described on Table 3 are present even if in a mild form (13).

6. DIFFERENTIAL DIAGNOSIS OF DELIRIUM When assessing patients for delirium it is important to recognize that almost any of the symptoms that occur in delirium can occur also in other psychiatric and neurologic conditions, with the common differential diagnosis being dementia and acute psychosis, but with the differential diagnosis encompassing many other psychiatric disorders and neurologic entities. Many of the psychiatric symptoms that are common in cancer patients can occur as manifestations of delirium and include anxiety, tearfulness, nervousness, depression, and irritability. Given this observation in the medically ill cancer patient, it is always important to examine the patient’s mental status. When there is evidence of impaired cognition, the potential diagnoses include delirium, dementia of various types, acute neurologic conditions including epilepsy, amnestic disorders, and drug-induced psychoses. The features that can be helpful in distinguishing the conditions responsible for cognitive impairment include the timing of the onset of the disorder, alterations in the level of consciousness, the systematization of delusions, evidence of psychomotor disturbances, and deviations from normal speech patterns.

Chapter 6 / Confusion and Delirium

73

In contrast to delirium, which is characterized by the onset of an acute change in cognition, the dementias are typically characterized by progressive cognitive impairment. A history of chronic decline without alteration in the level of arousal can assist in distinguishing syndromes of dementia from delirium. The diagnosis of delirium can be especially difficult when delirium and dementia occur concurrently. Patients with dementia are particularly sensitive to drug toxicity, metabolic abnormalities and physical stress, all of which have clear associations with the onset of delirium. This is especially relevant in elderly or otherwise debilitated patients. A history of the patient’s baseline mental status, level of cognitive impairment and behavioral pattern is most important in assessing mental status changes in cases where dementia is also present. The term pseudo-deliria has been used to describe acute psychotic episodes that resemble delirium. Most of these episodes can be distinguished from delirium on the basis of a history of psychosis and by certain clinical signs manifest on mental status examination. In delirium, for example, vigilance is not usually preserved and there is clear impairment of arousal usually manifesting by concentration deficits and fluctuating levels of alertness. In psychotic episodes these latter features are not present, and delusions are most frequently sustained and systematized. It is also important to keep in mind that, as can occur in cases of dementia, delirium and psychosis can occur concurrently and a known history of psychiatric disease should not eliminate the need for careful screening to assess for other causes of altered mental status. Here again, the presence or absence of decreased arousal can be a diagnostic clue as can the EEG. The excessive slow-wave activity that most frequently characterizes the EEG in cases of delirium is not present in most cases of acute psychosis although it is important to be aware the EEG abnormalities can occur in association with the use of some antipsychotic medications and these must be taken into account when interpreting EEG findings (54). The possibility of a diagnosis of catatonia—also known as psychogenic unresponsiveness—should be considered in patients who present with a history of depression or other psychiatric disease (55). Some of the symptoms of delirium may be seen in mood disorders. The latter are common, particularly among the elderly and among cancer patients. Depression, for example, is common in the cancer population and depressive symptoms are often an associated feature of delirium. Because of the high prevalence of both mood disorders and delirium in the medically ill population it is extremely important that evaluation of a seemingly depressed patient include a comprehensive assessment that incorporates neurologic, psychiatric, and cognitive evaluations. It is not uncommon when a patient is clearly confused for a clinician to consider a diagnosis of delirium. In contrast, if an ill patient is anxious it is common for subtle mental status changes to go undetected in that setting and the appropriate diagnosis—delirium—may be overlooked. This provides further support for the need for thorough mental state examinations in all patients with psychiatric symptoms. In summary, the process of making a diagnosis of delirium must include careful examination of the mental state of the patient, often repeated over time in cases where symptoms may be fluctuating. An accurate diagnosis of delirium may also be facilitated by the use of an instrument or diagnostic tool (see below). The other key part of diagnosis involves the careful examination of the patient with a clinical work-up aimed at defining reversible etiologic factors. The intensity and detail of this aspect of the diagnostic process will depend on the clinical situation and goals of care.

7. DIAGNOSTIC TOOLS AND INSTRUMENTS FOR ASSESSMENT OF DELIRIUM Although the clinical psychiatric interview using the DSM criteria remains the gold standard for the evaluation and diagnosis of delirium, several instruments have been developed to facilitate the diagnosis of this syndrome. The ideal instrument to assess and diagnose delirium both in research and in clinical practice should have the following characteristics: 1. 2. 3. 4.

Be validated specifically for diagnosing delirium according to accepted criteria (content validity); Be simple and administrable by different staff members (inter-rater reliability); Discriminate delirium from dementia and psychoses (discriminant validity); Allow for repeated administrations and be sensitive to time fluctuations of symptoms and to therapeutic intervention (i.e., it should be sensitive to clinically relevant changes); 5. Identify all the clinically relevant aspects of the syndrome.

To date, there is no instrument among those available that satisfies all of these requirements (45). For diagnosis in the clinical setting we suggest that consideration be given to the integration of a diagnostic system

74

Part III / Neurologic Symptoms

based on clinical criteria with an assessment instrument that has some utility in the quantification of severity, symptoms, and course. The diagnosis can be based on the previously mentioned criteria (DSM IV, ICD 10). Many neuropsychological assessment tools can be adapted to assess the patient with delirium. Readers are referred to Caraceni and Grassi for a comprehensive review of those instruments specifically designed to assess this syndrome (45). Instrument selection for use in research will depend on the purpose of the research, but the above criteria should also be considered in this process. Certainly the consistent use of appropriate scales will be useful in addressing the gaps that exist in the description and quantification of the clinical phenomena of delirium that, to date, have often been reported in idiosyncratic forms and open to variable interpretations. In this section of the chapter we will briefly review the Confusion Assessment Method (56,57), the Delirium Rating Scale (58), and the Memorial Delirium Assessment Scale (59). While we also recognize that other scales or methods are also available (45,60–62), we elected to review those instruments that are more recent, practical, respond to most of the criteria required for a clinically useful assessment tool and have data available about their validity in the field. • Confusion Assessment Method (CAM) (56,57)— This is a useful tool for ensuring that all the criteria for the DSM III-R are met. It should be noted, however, that the instrument was developed using these criteria and not those in the DSM IV. The instrument is short and assesses four areas—acute change in mental status, failure of attention observed by the interviewer, disorganization of thought, and altered state of consciousness. The CAM had 94–100% sensitivity and 90–95% specificity in the original validation study (56). Similar results have been subsequently reproduced (57), but it is important to note that the instrument is designed for use by trained medical personnel (63). • Delirium Rating Scale (DRS) (58)—The DRS is an instrument that contains a list of symptoms that describe typical clinical aspects of delirium and rates their relevance in the phenomenology of delirium. Ten clinical aspects or symptoms of delirium are scored. The maximum score is 30, and scores from 12 to 30 are found in patients with delirium. The DRS has been assessed as a diagnostic tool using different cutoffs of scores between 8 and 12, but the results obtained by using these different thresholds are not without some problems (64,65). The DRS is intended to reflect a 24-hrs assessment period. The scores are designed to differentiate the patient with delirium from patients with dementia and psychiatric disorders, but certainly there exist areas of overlap. The revised version (DRS-R-98) (66) is designed to improve the characteristics of DRS by operationalizing most items; it is longer than the original and with a cut-off score of 17.75 has a 92% sensitivity and 95% specificity. • Memorial Delirium Assessment Scale (MDAS) (59)—The MDAS is also a 10-item instrument. It quantifies the severity of symptoms that are found in delirium and is based on criteria that are included in both the DSM-IIIR and the DSM-IV. Scores from the MDAS and DRS are significantly correlated and also correlate with a global clinical judgment of delirium severity (59,65). The MDAS is intended for repeated administrations over a short time period and it is hoped that this new instrument may prove to be useful to capture short-term fluctuations in delirium. The MDAS has been well-tested in cancer patients.

In addition to diagnostic tools based on clinical assessment it is important to address the role of the electroencephalogram (EEG). The EEG can be a useful tool in assessing the diagnoses of delirium, psychosis, and dementia (67–69). The main EEG finding in the course of delirium is a progressive slowing of the EEG dominant frequencies with reduction of alpha rhythm, as well as the onset of an increase in delta and theta frequencies (9). In metabolic encephalopathies, EEG changes can also evolve into burst suppression patterns and epileptiform activity. These latter changes are found both as an effect of general anesthetics and of metabolic encephalopathies, while triphasic waves are typical of metabolic encephalopathies alone. A recent classification of metabolic comas can be used to assess these conditions (70). The EEG is especially important in differential diagnosis. For example, nonconvulsive status epilepticus is a condition that results in altered consciousness and may lead to a clinical picture of delirium. The latter diagnosis can be confirmed by EEG. In oncology, complex metabolic situations can lead to partial status, and this condition is also typical of ifosfamide encephalopathy (71). In such instances a diagnosis of delirium may still be present with its etiology being a seizure disorder.

8. ETIOLOGY AND RISK FACTORS Having established a diagnosis of delirium, defining the etiology is key to the implementation of appropriate therapy. There are innumerable causes of delirium with the etiologic categories of delirium being arbitrarily grouped into primary CNS diseases, systemic diseases with secondary CNS effects, exogenous intoxications, and alcohol or drug withdrawal. As discussed above, multiple etiologic factors can occur concurrently in an

Chapter 6 / Confusion and Delirium

75

Table 4 Diagnostic Tests* Temperature Bedside screen of medication profile Pulse oxymetry Blood glucose Serum electrolytes (Na, K, CL, Mg, Ca) Urea, creatinine Liver function tests Ammonia Leukocyte count Red cell count Coagulation profile Urinalysis Blood/urine and other cultures for infection screen Urine or blood drug screening Blood gases and acid-base balance CSF examination: complete blood count, glucose, protein, cytology, culture B12 levels Thyroid hormone and TSH Adrenal function Brain CT or MRI EEG** * The clinical situation and goals of care will influence decisions regarding investigations. On occasions, proceeding with an investigation may not change the treatment strategy in an individual case and therefore it may be concluded that proceeding with such an investigation would have little role. ** The EEG has a special role in the investigation of delirium and its place in last position on this list should not be read to imply that it has limited usefulness. The EEG has an important role in the differentiation of different pathological conditions resulting in cognitive impairment. It should be considered when there is a specific factor in the history or clinical situation to suggest its utility.

individual who may have few, or many, predisposing characteristics. All etiologic and predisposing factors should be considered in defining potential factors amenable to treatment. Table 4 provides a list from simple to more complex tests and laboratory or imaging procedure that can be useful in assessing the etiology of delirium. Table 5 provides a list of causes of delirium which are discussed below.

8.1. Primary CNS Diseases • Cerebrovascular Diseases—Cerebral infarcts and transient ischemic attacks (TIAs) can present with delirium. Cases of vascular insufficiency in which small focal lesions have been found to be the cause of delirium have been described due to the involvement of the territory of the middle cerebral artery of the nondominant hemisphere. (20,72,73). Henon et al. conducted a systematic prospective survey that showed the most relevant factor associated with the onset of delirium after ischemic stroke is the prestroke cognitive status (74). Other vascular causes such as brain hemorrhage or hypertensive encephalopathy can globally affect brain function and level of vigilance by producing brain edema or herniation. • Trauma—Delirium is often the only sign of brain contusion, concussion, or the occurrence of hematoma. • Epilepsy—Postictal states are a common cause of persistent cognitive failure. The possibility of a partial status epileptics or of partial complex seizures is an important differential diagnosis that can be revealed by EEG. As mentioned above, in oncology this condition has been well recognized in association with ifosfamide toxicity (71). • Infections—Meningitis and encephalitis are commonly associated with altered sensorium and cognition. • Intracranial Tumor—Consciousness and cognition are frequently affected by intracranial tumors. Intracranial hypertension and involvement of brainstem functions are obvious causes of impaired cognition. Frequently, the presence of cerebral or meningeal metastases, in the absence of focal neurologic deficits, results in mental changes and neuropsychological symptoms that can be detected only after careful examination. Indeed, with careful detection techniques, mental changes are the most frequent sign of brain dysfunction due to intracranial and meningeal

76

Part III / Neurologic Symptoms

Table 5 Causes of Delirium in Cancer Patients Primary CNS tumor Secondary CNS tumor Brain metastases Meningeal metastases Nonmetastatic complications of cancer Metabolic encephalopathy due to hepatic, renal or pulmonary failure Infections Electrolyte abnormalities Glucose abnormalities Hematological abnormalities Paraneoplastic neurologic syndromes Nutritional deficiency (including thiamine, folic acid, vitamin B12 deficiency) Toxicity of antineoplastic therapies Chemotherapy toxicity (see Table 6) Radiation to brain Toxicity of other medications (see Table 6) Other diseases/conditions not related to neoplasm CNS diseases (including cerebrovascular disease, vasculitis or trauma) Cardiac disease Lung disease Endocrinopathy Alcohol or drug abuse or withdrawal Other

metastases (75–77). Case studies show that microscopic meningeal infiltration can cause mental status changes without focal signs and/or symptoms (78). The mechanism of encephalopathy due to meningeal metastases is uncertain. Tumor cell seeding to the meninges can interfere with CSF formation and re-absorption, compete with the brain parenchyma for essential nutrients, and/or produce ischemic damage by infiltrating the Virchow–Robin spaces. Intracranial pressure changes can be present early in course of leptomeningeal disease due to altered dynamics of the CSF and present without the classic findings of papilledema, severe headaches or meningismus (79). Fluctuations of intracranial pressure (plateau waves) can be responsible for unexpected, reversible, acute changes of mental status (76). In the advanced phases of cancer brain metastases are associated with an increased risk of developing delirium. In this setting patients frequently also have other problems present, such as hypoxia, anorexia, and poor performance status, that place them at additional risk for complications of cancer (and for delirium) (41). • Degenerative Diseases—Dementia is a well-known predisposing condition for delirium. There are no studies that evaluate the degree to which other primary neurologic diseases and systemic diseases with secondary CNS effects predisposed to delirium although it is clear that they do have this effect.

8.2. Systemic Diseases with CNS Effects Delirium is probably the most frequent neurologic complication of non-neurologic general medical illness. Many of the etiologic factors for delirium occur commonly in the patient with cancer (76). Fever and infection have long been described as triggers of delirium; indeed, this was described by Hippocrates (3). Liver, renal, pancreatic, or pulmonary failure can cause brain dysfunction and give rise to delirium. Brain hypoperfusion and consequent hypoxia can be due to any cardiac or extracardiac cause. Vitamin, cofactor and glucose deficiency affect cerebral metabolism directly and can be responsible of acute confusion. Electrolytes imbalances influence brain water content and can result in osmotic neuronal damage triggering delirium, stupor, or coma. In addition, most endocrine organ dysfunction can manifest as delirium (panhypopituitarism, hypoparathyroidism, hypo- or hyperthyroidism).

Chapter 6 / Confusion and Delirium

77

8.3. Exogenous Intoxications and Alcohol or Drug Withdrawal Drug toxicity is frequently implicated in the etiology of acute confusional states, especially in the elderly and in patients with other predisposing factors such as severe illness or organ failure. Drugs with primarily anticholinergic activity are commonly implicated in delirious states, but many other drugs can also trigger delirium although the mechanism by which they trigger brain dysfunction remains unknown. Excessive cholinergic stimulation can also cause delirium (80). The elderly population is at risk for delirium due to common alterations in pharmacokinetic and pharmacodynamic parameters that occur with aging. Table 6 provides a classification of the drugs that have been associated with delirium (or in some studies “encephalopathy”) and includes drugs that are commonly prescribed in oncology. In the oncology setting the role of opioids in delirium deserves a brief summary as these medications are commonly used for the treatment of pain. Opioids are sedative, with at least part of this sedation resulting from their anticholinergic activity. It is well known that opioid-related somnolence is usually dose-related. The impact of opioids on alertness can be documented by specific testing especially after a dose increase or in healthy volunteers (81,82). Most commonly, however, the cognitive impairment in patients with cancer is a complex multifactorial phenomenon that is explained by a decline in general performance, severe pain and concurrent illness, and in many instances cognitive change is not necessarily caused by the effect of opioids alone (83,84). This stated opioid exposure during an admission for advanced cancer has been associated with an increased risk of delirium (85). Certainly, opioid toxicity can manifest as delirium with idiosyncratic patient reactions occurring with any opioid at any dose (86,87). This probably occurs due to the anticholinergic effects of opioids, Table 6 Examples of Drugs That Have Been Associated with Reports of Delirium or Confusion* Prototypical anticholinergic medications (Examples: belladonna alkaloids, scopolamine, atropine, hyoscine, biperiden) Medications for psychiatric conditions, including anxiolytics/hypnotics/sedatives (Examples: Barbiturates, benzodiazepines, chloral hydrate, chlorpromazine**, lithium carbonate, tricyclic antidepressants**, SSRIs) Pain medications (Examples: Opioids, NSAIDs) Antihistamines** Chemotherapeutic agents (Examples: More frequently cited—asparaginase, ifosfamide; rarely reported associations, usually reported in patients in unusual circumstances or with other complications and/or medications present—cisplatin, cytosine arabinoside, etoposide (at high doses), 5-fluorouracil, methotrexate, nitrosourea (at high doses or via arterial route), procarbazine, vincristine) Immunosuppressants and immunomodulatory agents (Examples: Steroids, interferons and interleukins, cyclosporin, tacrolimus) Antibiotics and antivirals (Examples: More frequently cited—quinolones, acyclovir, gancyclovir. Numerous other antibiotics, including, e.g., imipenem and aztreonam have reported associations, usually reported in patients with other complications and/or medications present) Medications for gastrointestinal conditions (Examples: H2-antagonists (including cimetidine, ranitidine, famotidine), omeprazole) Medications for CNS conditions (Examples: Anticonvulsants, levodopa) Medications for cardiovascular conditions (Examples: Beta-blockers, digitalis) * The purpose of this table is to illustrate some of the common medications used in oncology settings that have been associated (rarely in some cases) with the occurrence of confusion or delirium and to demonstrate the spectrum of medications that have been associated with this condition. Examples are not inclusive of all medications in the category that may be associated with confusion or delirium. Reports of confusion with one medication often document the occurrence of this problem in ill patients who are taking more than one medication and, consequently, the exact etiology of the confusion can be difficult to define. **Medications with established anticholinergic activity

78

Part III / Neurologic Symptoms

but also in certain instances may occur as a consequence of excitatory CNS phenomena that have been reported with high-doses of opioids (88,89). Hallucinations, delirium, myoclonus, hyperalgesia, and seizures have each been described as part of this syndrome. Recent data confirm the increased risk of developing delirium in cancer patients associated with the use of high doses of opioids, benzodiazepines and corticosteroids (90).

8.4. Risk and Precipitating Factors Not only is it more frequent to find multiple etiologies of delirium than to find a single identifiable cause but it has also been suggested that the patient’s underlying condition can modify susceptibility to external factors that might not, in less susceptible individuals, be pathogenic (2,7), Furthermore, in some instances, the rate of change of a metabolic or toxic factor may be more important than the absolute value of the variation (2,7). An example would be a sudden, versus a slow, change in a metabolic parameter such as sodium. Several studies undertaken in the elderly have identified risk factors including age, previous cognitive failure (dementia), severity of illness, infections, renal failure, use of psychotropics, and electrolyte abnormalities (91–93). In the specific case of alcohol withdrawal triggering delirium tremens the risk factors have been shown to include intercurrent febrile illnesses, trauma, reduced food intake and gastrointestinal disturbances (94). Work by Inouye and colleagues has resulted in a validated multifactorial model for the development of delirium (93,95,96). These authors distinguish baseline vulnerability factors—factors already present at the time of hospitalization—from precipitating factors that occurred at or after hospitalization. Baseline vulnerability in their model was associated with age, cognitive failure, visual impairment, and gravity of disease. Precipitating factors included, among others, the use of psychotropics, dehydration, and polypharmacy. The true pathogenic role of each of these latter factors in inciting delirium was not clear within these studies because the investigators utilized statistical methods that explored the specific events and diagnoses rather than known etiologic triggers that were documented during hospitalization. Nonetheless, hypotheses related to potential risk factors for delirium can be made. Further, by modifying environmental and biological conditions, including, for example, using reorientation methods and patient counseling, the same authors were able to show a decrease in the cases of delirium in an elderly hospitalized population (96). In the oncologic population many of the risk factors that have been studied in other populations are common. In addition, some other specific factors exist as previously listed in Table 5. As alluded to above, these may be risk factors that increase baseline vulnerability or precipitating factors that appear to initiate an episode of delirium. In a study of 140 cancer patients assessed for altered mental status (not specifically defined in this series as “delirium”) in a cancer center, a single cause of the altered mental status was present in 33% of patients, and 67% had multiple causes (84). Drugs were considered to be associated with altered mental status in 64% of cases and opioids were commonly linked with the syndrome. Metabolic abnormalities were present in 53% of cases infection in 46%, and recent surgery in 32%. A structural brain lesion was considered to be the sole cause of the change in mental status in 15% of patients (84). In another group of patients admitted to a palliative care unit for the control of difficult symptoms due to advanced cancer, 71 delirium episodes were documented (40). In the 71 episodes, 158 probable and 70 possible precipitating factors were identified and included psychoactive medications (mainly opioids), dehydration, nonrespiratory infections, alcohol or drug withdrawal, intracranial causes, hypoxia from respiratory infections and lung cancer, metabolic, and hematologic causes. In this study, reversibility of delirium was associated with opioid toxicity while irreversibility was associated with hypoxia due to lung cancer or respiratory infection. One recent prospective study in cancer patients showed that at the time of admission, advanced age, cognitive impairment, low albumin, bone metastases and the presence of hematological malignancies increased the risk of developing delirium (38).

8.5. Common Clinical Situations in Which Delirium Occurs Although delirium can occur in many settings there are three specific clinical situations in which delirium is frequent and these warrant specific mention: delirium tremens, post-operative delirium, and delirium in the elderly patient. A fourth common condition is “terminal” delirium, a condition that is common at the end of life and also warrants specific mention.

Chapter 6 / Confusion and Delirium

79

• Delirium Tremens—The symptoms of acute alcohol withdrawal are usually grouped into three different syndromes: alcoholic hallucinosis, alcohol withdrawal syndrome, and delirium tremens (DT). When consideration is given to the criteria for these three conditions it becomes apparent that DT is a relatively uncommon complication of alcoholism, reported in 5% of alcoholic patients admitted to a general hospital (94,97). Alcohol withdrawal syndrome is more common. This syndrome occurs usually in patients who have had 5–15 years of excessive alcohol consumption, in subjects 30–40 years old, and is more common in males (94,97). DT is the most severe clinical manifestation of alcohol withdrawal and usually occurs 72–96 hrs after the cessation of alcohol intake in those patients prone to withdrawal. This syndrome is the most classic form of the hyperactive delirium. Disorientation to time and space is associated with visual and auditory hallucinations, agitation, a hypervigilant state, tremors, and autonomic hyperactivity (i.e., tachycardia, sweating, arterial hypertension, at times fever). Visual hallucinations are reported by most patients and can consist of microzoopsias (visions of small animals), but are usually of a variable nature and are not necessarily frightening. Auditory hallucinations can include menacing voices but also neutral sounds or music. Interestingly, recent work has suggested, with regard to predisposing factors, that a genetic base, relating to the regulation of dopamine transmission, for DT is plausible (98). • Postoperative Delirium—Delirium can follow awakening immediately after surgery but may also be delayed 3–5 days (99,100). The pathogenesis of this complication is thought to be associated with numerous factors including, sleep deprivation, sensory deprivation, pain, metabolic factors, anesthetic drugs, anticholinergic medications, analgesics, hypoxia, fever, and blood loss. The incidence of this condition is high after open heart surgery, and following orthopedic surgery in the elderly patient (100–103). A predictive model for this condition has been validated and can be used to assess the relative risk of the individual patients. This model is based on some intraoperative stress factors (i.e., type of surgery) and some preoperative predisposing factors (including for example, age > 70 years, preexisting cognitive impairment, history of alcohol abuse, poor functional status, marked abnormalities of serum sodium, potassium or glucose) (102,104,105). • Delirium in the Elderly—The elderly population is at a high risk for developing delirium when hospitalized. In this setting delirium impacts on morbidity, functional recovery, time of hospitalization and most probably also on survival (106). In the elderly, delirium may be a complication of a known pre-existing dementia or a dementia that has been subclinical. The recovery from an acute episode can be protracted and often incomplete, such that the distinction between an acute and chronic condition can be difficult or impossible (91,107). In the elderly, hypoactive and depressive features of delirium are common. Depression is a common erroneous diagnosis in elderly patients who are suffering from delirium (108). In elderly patients with delirium it is not uncommon to find neurologic evidence of brain aging and other concomitant diseases. The presence of tremors, dysarthria, gait ataxia, and incontinence should alert the clinician to the possibility of the clinical condition having another underlying cause. • Terminal Delirium—Delirium is frequent in cancer patients with advanced disease and is an independent predictor of shortened life prognosis (41,109). In the patient with advanced disease delirium can be the consequence of an irreversible organ or multi-organ failure leading to death. It is important to note that in many cases, despite far advanced disease, reversible etiologic factors are present. For example, a recent study demonstrated that almost 50% of the episodes of delirium occurring in patients with far advanced cancer were reversible (41). This highlights the importance of an approach to delirium management in the terminal setting that considers addressing reversible contributory factors while concurrently addressing distress and safety concerns (110).

9. DELIRIUM TREATMENT The therapy of delirium is based on interventions directed towards symptoms and towards etiologic and, if possible, predisposing factors. The overall clinical condition will dictate the goals of care in each case, which will in turn impact on the diagnostic and therapeutic interventions. Approaches include the removal of etiologic triggers, environmental changes, and pharmacological interventions. Each of these approaches will be briefly reviewed below.

9.1. Etiological Interventions As mentioned previously, often the identification of a single etiologic, risk or precipitating factor is not easy. Attention to fluid and electrolyte balance, vitamin and glucose support, oxygenation, and treatment of infections should be considered as baseline primary interventions. A review should be undertaken to consider the possible withdrawal of any drug that is not considered strictly necessary (although keeping in mind that the withdrawal of certain drugs can also aggravate delirium). Consideration should be given to the role of optimal hydration.

80

Part III / Neurologic Symptoms

For patients for whom opioids are an essential component of the treatment plan, the role of substitution of one opioid for another or dose reduction should be considered among potential interventions (111).

9.2. Environmental Interventions In the setting of delirium it is vital that patient safety be ensured (36,112). Where there is evidence of impaired cognition, and/or delusions and hallucinations, patients should be assessed for risks of falls, wandering, self-harm and risk to others, with appropriate measures being taken to ensure safety. Appropriate measures may include medications (as detailed below), and increased surveillance and supervision including, where needed, measures such as a sitter in the hospital room. In addition, environmental hazards should be avoided (e.g., high beds and objects that can be a danger to the patient) (36). In addition to ensuring safety, it has long been considered important to foster a tranquil quiet environment and, if feasible, to consider the role of nonintrusive sensory and cognitive reorientation protocols (e.g., clock, calendar, familiar people and objects, audiovisual aids). One study has demonstrated that environmental changes and reorientation protocols were effective in reducing the incidence of delirium in the elderly at-risk population (96). Further work by this same group has demonstrated that adherence to the nonpharmacologic protocols is important if their impact is to be maintained (113). Of note, patient awareness is usually compromised in delirium but is not always lost and the perception of cognitive failure can be an important source of suffering (50). For this reason reorientation should, where possible, be approached in a manner that seeks to minimize patient distress. Psychological support for the patient and family is recommended (36). This can serve to address patient and family distress, enhance communication with the staff, and clarify questions about causes, clinical fluctuations, reversibility, and prognosis (50,114–116). Explanations about the nature of the condition and about the efforts being directed towards reversing the problem can be reassuring for patients and carers. Family counseling may also begin to address what can become a vicious cycle of incorrect treatments for distress. For example, requests for pain medication may be made in the setting of agitation, a situation in which careful consideration must be given to the appropriate role of pain medications and medications or interventions directed towards the delirium (96,117,118).

9.3. Pharmacological Interventions Pharmacological interventions for delirium have been recommended for the treatment of significant symptoms and distress. At this time no study has been undertaken that provides data to clarify the role for these interventions in mild, asymptomatic cases of delirium; this is an area in need of research. Nonetheless, it is clear that agitated delirium is difficult for both patients and their family caregivers and can also be dangerous. In such cases a pharmacological intervention directed at symptoms and distress is considered to be justified (36,112,119). The American Psychiatric Association has developed guidelines, based on the available research, for the assessment and treatment of delirium (36,112), as has the Royal College of Physicians in London (119). The British guidelines provide three reasons for the use of pharmacologic treatments: (1) in order to carry out essential investigations or treatments, (2) to prevent a patient from endangering themselves or others, and (3) to relieve distress in a highly agitated or hallucinating patient (119). Of note, however, even in the absence of these reasons, there are certain types of delirium that may warrant specific early intervention (e.g., delirium tremens). When pharmacologic treatments are implemented, continuous monitoring and frequent assessment are necessary and are integral part of treatment. A problem in this field is that studies that explore the pharmacological treatments of delirium are rare and, for example, there is only one study that has provided insight into the comparative role of benzodiazepines and antipsychotic medications (120). This study, undertaken in a HIV-infected population, was a randomized controlled clinical trial that compared the use of lorazepam, haloperidol, and chlorpromazine in treating delirium and showed that while haloperidol and chlorpromazine produced an objective improvement in symptoms, lorazepam worsened confusion (120). In the absence of large numbers of controlled studies, clinical experience, case reports, and uncontrolled trials have guided the development of practice guidelines (36,112,119). Antipsychotic medications are generally recommended as first-line treatment in cases of delirium with agitation or distress (36,112,119). The exceptions to this are in cases of alcohol or benzodiazepine withdrawal and in situations where there are contraindications to

Chapter 6 / Confusion and Delirium

81

the use of antipsychotic medications, for example, in Parkinson’s disease and Lewy body dementia (119). When antipsychotic medications are indicated, haloperidol is generally recommended as the initial drug to consider in the majority of cases because of its limited anticholinergic, hypotensive, and sedative actions (36). That stated, there are also significant concerns about the use of this medication (see below). Although more data is needed, data are accumulating suggesting that the pharmacologic treatment options may expand with the introduction of second-generation antipsychotic medications (see below) (36,112). Prior to the initiation of any pharmacological intervention the potential side effects and risks for the individual patient should be considered. Side effects of antipsychotic medications include sedation, anticholinergic effects and alpha-adrenergic blocking effects (and thus hypotension) (36). These side effects are less common with haloperidol, a butyrophenone, than with the phenothiazines (36). In addition, extrapyramidal reactions, tardive dyskinesia, and neuroleptic malignant syndrome can occur with antipsychotic medications (36,121,122). Furthermore, there are numerous other side effects of these medications that include lowering of seizure threshold, withdrawal movement disorders, abnormal liver function and others (36). A significant potential side effect of haloperidol and other neuroleptic anti-psychotics is cardiac dysrythmia related to prolongation of the QT interval in the EKG (123–128). This can lead to torsades de points and sudden death (128). These events have been reported with higher doses but, importantly, also with low-dose intravenous and oral haloperidol and well as with other antipsychotic medications (36,128). EKG monitoring may be useful when using IV infusions and in patients with baseline cardiac conduction abnormalities who are, as a consequence of these abnormalities, considered to be at enhanced risk. Readers are advised to review a recent advisory from the United States Food and Drug Administration (FDA) that addresses these risks – risks that are considered to be significantly enhanced when haloperidol is administered intravenously and in higher doses. (http://www.fda.gov/cder/drug/InfoSheets/HCP/haloperidol.htm) This advisory strengthened warnings in the haloperidol labeling, addressed risk factors for cardiac problems, advised regarding dosing and monitoring, and reiterated that haloperidol was not approved for IV use in the United States despite evidence from the medical literature that the intravenous administration of haloperidol is a relatively common “off-label” clinical practice especially in ICU settings. A baseline EKG for assessing for the presence of a prolonged QT interval is considered mandatory before initiating neuroleptic therapy by some healthcare authorities (128). Whether the cardiac side effects will be less common with the newer generation of antipsychotic medications is yet to be established, but certainly there are reports of prolonged QT intervals and torsades de points with the newer medications (128). Of note, female gender and family history are linked with increased risk for cardiac problems with these medications (128–130). Where a prolonged QT interval exists at the baseline or when the patient has heart disease, hypokalemia, or has been prescribed another agent known to lengthen QT interval, the antipsychotic drugs should be avoided where possible. In these circumstances, if it is considered necessary that these medications be used, the physician should address monitoring for arrhythmias within the context of the treatment plan and goals of care (128–130). While monitoring may be considered very important in some settings, priorities may alter in the context of care in the last days of life. For all cases of delirium, it is recommended that when medication is prescribed, the lowest dose of medication needed to achieve an effect be used and that caution be undertaken with close monitoring for side effects. When treatment with haloperidol is implemented it is usually initiated with oral, subcutaneous (SC), or intravenous (IV) medication. Of note, although there are many reports of the use of intravenous haloperidol, and guidelines suggesting its use (36), the United States Food and Drug Administration (FDA) has not approved this medication for intravenous administration. When haloperidol is administered it is usually with small doses in the mild cases (0.5–2 mg daily) and higher doses in more severe cases (2–5 mg daily) (36,119). In the elderly, in frail medically ill patients, and in patients with low performance scores, lower initial doses are recommended (0.25–0.50 mg daily) (36,131). These doses reflect the total daily oral doses but most usually these doses are administered in divided prn doses. Dosing guidelines recommend that the treatment commence with a dose every 2–4 hrs as needed although some authors have discussed a “loading dose” with the administration of a dose of haloperidol initially every 30–60 minutes in cases involving significant agitation or distress, with medication then given on an “as needed” or “prn” basis or, in more severe cases, with the establishment of an “around the clock” regimen (36,132).

82

Part III / Neurologic Symptoms

Severely agitated patients may prompt clinicians to consider either titration of antipsychotic medication to higher doses or the addition of another type of medication (see below). If titration is considered, while high intravenous doses of haloperidol, both via bolus dosing and continuous infusions, have been reported in monitored intensive care unit (ICU) settings, with total daily doses of more than 100 mg/day, this is by far the exception, associated with increased risk. Common doses are in the realm of 1–10mg/day of haloperidol (133–135). In settings of severe agitation, it has been suggested that continuous infusion of haloperidol may minimize the risk of hypotension associated with bolus dosing and be associated with minimal effects on heart rate, respiratory rate, blood pressure, and pulmonary artery pressure and few extrapyramidal side effects (134,135). As was discussed above, significant side effects and potentially severe adverse reactions have been reported; however, with haloperidol (and other antipsychotics), these reactions have been reported more commonly with higher doses, suggesting the need for caution and close monitoring always, but especially when higher doses are prescribed (128,136–140). Other neuroleptics, in particular chlorpromazine and droperidol, have been used for delirium and have been, to date, selected where more profound sedation is necessary to control symptoms (141). For example, in the ICU setting continuous infusions of droperidol have been utilized (142), but this medication has been reported to be associated with more sedation and hypotension than haloperidol (36,142), and, like other antipsychotics, has risks of prolongation of the QT interval and arrhythmias (36,128). Regarding the use of the newer, atypical antipsychotics for delirium treatment, a number of case reports and series have been published reporting on this newer generation of atypical neuroleptics with activity on dopamine, serotonin, and histamine receptors (risperidone, olanzapine, clozapine, quetiapine, and ziprasidone) (36,45,112,136,143–146). A recent Cochrane review sought to explore whether there was evidence that the standard drug for delirium, haloperidol, should be replaced by atypical antipsychotics such as risperidone, olanzapine, or quetiapine, and whether these medications were as effective as haloperidol in controlling delirium and/or had a lower incidence of extrapyramidal adverse effects (136). The conclusions of this review must be interpreted in the light of the fact that few studies were available; indeed only three studies satisfied selection criteria and these studies compared haloperidol with risperidone, olanzapine, and placebo. The authors concluded, from the limited data, that there was no evidence that low-dose haloperidol has a different efficacy compared with olanzapine and risperidone in the management of delirium or a greater frequency of adverse drug effects than these drugs. Reported in one study was a higher incidence of extrapyramidal side effects when high dose haloperidol (> 4.5 mg per day) was compared with olanzapine, but of note the authors also commented about the lack of studies relating to this question (136). In summary, these medications have not been widely used to date for delirium but future studies will be important to clarify their role (112). Benzodiazepines are not generally recommended as first line interventions in delirium, the exceptions being in cases of alcohol or benzodiazepine withdrawal (36,112). In alcohol withdrawal the use of benzodiazepines is claimed to control hyperarousal and neurovegetative activation (147,148). A full discussion of the treatment of alcohol withdrawal is beyond the scope of this chapter but early recognition of alcohol withdrawal is very important so that treatment can be implemented promptly and its very serious potential complications minimized. Benzodiazepine protocols are the mainstay of treatment for alcohol withdrawal but, as the manifestations of DT include those related to autonomic hyperactivity, the use of clonidine has been considered by some due to its effects in inhibiting noradrenergic central transmission resulting in both sedative and antihypertensive effects (147,148). The general recommendation is that benzodiazepines be avoided in the treatment of delirium except in certain circumstances (36,112,119). In addition to the roles mentioned above, the circumstances in which treatment with benzodiazepines may be considered includes when there is a need for a medication that can raise the seizure threshold (unlike antipsychotics, benzodiazepines raise the seizure threshold), when anticholinergic or other side effects of antipsychotics exist, or when the patient has a contraindication to the use of antipsychotic medications (36,112,119). Although studies are lacking, benzodiazepines have also been used for sedation in cases of agitated delirium in which symptoms are not controlled with the use of neuroleptics alone especially in the ICU and palliative care settings (149–154). Finally, benzodiazepines may have a role in patients for whom their anxiolytic effect is desired, including, for example, in some cases of terminal delirium (152,155,156).

Chapter 6 / Confusion and Delirium

83

As with the antipsychotic medications the side effects and risks of benzodiazepines should always be considered when a treatment decision is being made. Side effects of benzodiazepines include sedation, confusion, behavioral disinhibition, sleep disturbances, anxiety, increased fall risk, respiratory depression, withdrawal syndrome on cessation, and others (36). Of note, the use of benzodiazepines in patients with hepatic encephalopathy and in the elderly is associated with frequent adverse reactions and thus if possible it is best to avoid these medications in these populations. When benzodiazepines are to be utilized, drugs with short half-life and no active metabolites are generally the preferred approach, with lorazepam, for example, meeting these criteria (36,119). Low doses of lorazepam can be used orally or sublingually (0.5–1 mg daily in divided prn doses.). Intravenous lorazepam is available in some countries, but of note, high doses of intravenous lorazepam have been associated with significant toxicity manifest by renal dysfunction and metabolic acidosis related to a propylene glycol vehicle used in the IV formulation (112,149,157–160). Due to this problem this drug is not recommended for high-dose or prolonged use intravenously and midazolam is more commonly the benzodiazepine of choice in ICU settings (149). Midazolam has been frequently used in ICU settings (149), and also in the palliative care of advanced cancer patients, especially when antipsychotic medications alone do not control agitation or distress (150–154). Midazolam has the advantage of being well absorbed after SC administration, has very fast onset of action, and can be used also for short-term reversible sedation with starting of doses of 0.5–1 mg for induction, with the dose being repeated as needed or, if needed, escalated sometimes 1–2 hourly. Achieving prolonged sedation with midazolam can be difficult, as this medication often requires frequent dose adjustment. Daily doses of this medication commonly range from 10 to 60 mg with the occasional use of higher doses needed to achieve comfort (150–154). The pharmacokinetics of midazolam are such that, just as it can be difficult to establish prolonged sedation, when sedation has been achieved it can be slow to reverse. This is especially relevant in cases where the patient may be improving medically and the delirium resolving. As noted above, benzodiazepines may be used when antipsychotic medications alone do not control agitation or distress. In treating delirium, in the ICU, at the end of life or in other settings, it is recommended that the goals of any pharmacologic intervention should be to optimize patient safety, minimize agitation and distress, and, where possible, facilitate the patient’s communication with his or her family (36,110). In occasional circumstances, for example, when a delirious patient requires mechanical ventilation, a deeper level of sedation may be considered a necessary goal of delirium treatment. In the advanced cancer population there is a significant proportion of patients who will have reversible pathology contributing to delirium which can be addressed (115). In the last days of life it is not uncommon for antipsychotic medications alone to fail to control agitation or distress despite attention to reversible factors. For example, investigators aiming to treat delirium and provide symptom palliation undertook a study of 39 advanced cancer patients with delirium and found that 60% were managed with haloperidol alone and, in the remaining 16 patients other psychotropic drugs had to be added (lorazepam, chlorpromazine, or methotrimeprazine) (152). In 10 patients (26%) symptoms of delirium were reported to be controllable only by sedation—usually achieved in this study through the use of midazolam. It should be noted that there are many factors present at the end of life that may impact delirium and level of consciousness (110). Sedation may be a consequence of the underlying disease, the delirium process, and/or the delirium treatments, but in treating delirium deep sedation is not the goal, per se, of pharmacologic interventions. Although sedation may be unavoidable in some cases—especially at the very end of life—the dosing of medications for delirium should aim to facilitate, where possible, the patient’s ability to communicate while also minimizing distress and promoting safety. Several other categories of medication have been used in the treatment of delirium. The recent introduction of another group of cholinesterase inhibitors for the treatment of dementia has raised suggestions about their potential usefulness to ameliorate symptoms of delirium. Some clinical observations regarding the use of donepezil in this regard are interesting, and have resulted in a few case reports that provide only preliminary data about reversal of hallucinations and delusions and improved alertness (21,161). In cases of hypoactive delirium the use of psychostimulants has been suggested but, to date, clear guidelines for these indications are lacking and clinical experience has been very limited (162,163). One small series that included only 14 patients described a beneficial effect on cognitive function associated with the use of methylphenidate

84

Part III / Neurologic Symptoms

in patients with advanced cancer experiencing hypoactive deliria with no apparent reversible cause; however, to further elucidate the role of this class of medication, and the role of treating hypoactive delirium, further work will be needed (164). Finally, in the treatment of delirium induced by anticholinergic drugs, physostigmine has been used (2,165). This should be used only if anticholinergic toxicity is well-proven, with careful monitoring for the occurrence of side effects due to cholinergic hyperstimulation and with attention to contraindications such as a history of heart disease, asthma, diabetes, peptic ulcer, and bladder or bowel obstruction. It must also be administered very cautiously so as to avoid seizures and cardiac arrhythmias. On a final treatment note, a recent randomized study demonstrated that low-dose haloperidol as prophylactic treatment for elderly patients undergoing hip surgery had no impact on incidence of post-operative delirium but did decrease the severity and duration of episodes of delirium (166). Research evidence on the effectiveness of any pharmacologic interventions to prevent delirium is sparse and further work is needed. (167)

9.4. Summary of Therapy Guidelines • • • • • • •

•

Interventions directed towards reversible etiological factors in delirium are optimal. Environmental interventions should be implemented to minimize risk of harm and distress. Symptomatic treatment is recommended to control and minimize distress and suffering and to prevent harm. In cases of mild to moderate delirium of the hyperactive type neuroleptic medication is the first line of treatment with haloperidol generally considered as the medication of first choice, with other neuroleptics as possible alternative choices (e.g., risperidone, olanzapine, quetiapine, ziprasidone). Benzodiazepines are recommended as the first-line treatment for delirium related to alcohol or benzodiazepine withdrawal. Severe hyperactive delirium may not respond to oral haloperidol alone and regimens of IV, SC, or IM haloperidol in association with lorazepam or midazolam may be needed to control acute symptoms. Hypoactive deliria can be tentatively treated with haloperidol in low oral doses if disturbing hallucinations are suspected; otherwise, currently there is no strong evidence for pharmacological treatment although the role of psychostimulants is unclear but being actively studied and these medications may have a role in some cases. When medication administration for delirium is considered careful consideration should be given to the potential for side effects, risks and the role monitoring. Counseling and support for patients, family members, and caregivers are important interventions to minimize distress and ensure safety.

10. PROGNOSIS The classic perception of delirium is that it represents a transitory state that is brief and usually resolves promptly. Providing a contrast to this perception a study of elderly delirious patients found the average duration of an episode to be longer than 2 weeks (168). In this group the most common etiologies for delirium were stroke, infections, and metabolic disorders; coexistent structural brain disease was also present in a significant number of patients (81%) (168). Levkoff et al. evaluated a group of acutely hospitalized elderly patients and demonstrated that, among those with delirium, resolution of symptoms was often incomplete with only 4% experiencing resolution of all new symptoms before hospital discharge, with 20.8% taking 3 months to achieve symptom resolution and 17.7% taking 6 months (169). These data point to delirium being less transient than was previously thought and also point to the likely presence of incomplete manifestations of the condition. To complicate this issue further, as discussed above, a significant number of dying patients experience delirium in the terminal phases of illness and have no resolution of symptoms prior to death. In some cases sudden failure of cognition develops in the setting of advanced disease with no specifically definable reversible cause, and this condition persists for days to weeks through the rest of the clinical course of disease until death. The mortality associated with delirium varies depending on the population studied, the time frame studied (in-hospital or post-discharge), and other risk factors for mortality. Very varied mortality figures between 10% and 65% have been reported (38,41,169–171). In a study of ill hospitalised oncology patients by Tuma and DeAngelis,

Chapter 6 / Confusion and Delirium

85

although delirium improved in 67% of patients, it was a poor prognostic factor for overall outcome—the 30-day mortality was 25%, and 44% of patients died within 6 months, usually from progression of the underlying cancer (84). Delirium is more frequent in patients with multiple medical problems and it is likely that these processes and their complications, rather than the delirium itself, contribute to the high mortality rate. Nonetheless, delirium has been shown to be an indicator of poor prognosis. For example, in a study conducted in the elderly, the presence of delirium identified those patients at risk for prolonged hospitalization, loss of independent community living, and future cognitive debility (91). In this group, mortality was not significantly associated with delirium after adjusting for the severity of comorbidity (91). In patients with advanced cancer delirium is considered to be an independent predictor of worse prognosis and has been used in a model with other clinical and laboratory data to establish a short-term prognosis (39,41,91). In another study in advanced cancer patients reversibility of the delirious episode was associated with drug toxicity and irreversibility with hypoxia (due to pulmonary infection or tumor) (38), but the relationship, if any, of these problems with mortality was not reported. It can be concluded that delirium-associated mortality may be reflective of reduced brain functional reserve in the elderly; of the severity of irreversible organic processes that may lead to death; or of the toxicity of a reversible cause.

11. CONCLUSIONS Delirium is a highly prevalent disorder among the medically ill. In Baht and Rokwood’s words, it is “the failure of a high-order function that is close to system failure” (21). The cancer patient is at risk for developing delirium as a consequence of several general and numerous specific factors. Delirium can be responsive to interventions that address its etiologic factors, or be an irreversible event characterizing the final evolution of terminal illness. Neurologists involved in the assessment of oncologic patients need to have an understanding of both oncology and palliative care if they are to be able to conduct a comprehensive assessment of the patient and embark upon appropriate treatment plan. The breadth of the etiologic factors and the risks of the condition highlight the frequent need for multidisciplinary interaction with other medical specialties and health professionals to both treat the condition and minimize risk. Certainly the responsibility of the consulting neurologist must go beyond the provision of an elegant neurologic diagnosis. Neurologists treating delirium must also consider the importance of care that utilizes diagnostic and therapeutic decisions within the context of the patient’s goals of care, and addresses needs for treatment of physical and psychological symptoms, for minimizing the risks of the condition, and for patient and family support.

REFERENCES 1. Clouston PD, DeAngelis LM, Posner JB. The spectrum of neurologic disease in patients with systemic cancer. Ann Neurol 1992;31(3):268–273. 2. Lipowski ZJ. Delirium: Acute Confusional States. New York: Oxford University Press; 1990. 3. Jones W. Hippocates. English translation. London: William Heinemann; 1931. 4. Berrios GE. Delirium and confusion in the 19th century: a conceptual history. Br J Psychiatry 1981;139:439–449. 5. Steinberg D. What modern neuroscience can learn from Hughling Jackson. In: Clifford Rose F, ed. A Short History of Neurology. Oxford: Butterworth Heinemann; 1999:165–177. 6. Mesulam MM. Attention, confusional states and neglect. In: Mesulam MM, ed. Principles of Behavioral Neurology. Philadelphia: F.A. Davis; 1985:125–168. 7. Plum F, Posner JB. The Diagnosis of Stupor and Coma. Philadelphia: F.A. Davis; 1980. 8. Adams R, Victor M. Principles of Neurology. 6th ed. New York: McGraw-Hill; 1997. 9. Engel GL, Romano J. Delirium, a syndrome of cerebral insufficiency. J Chronic Dis 1959;9(3):260–277. 10. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington DC: American Psychiatric Association; 2000:147. 11. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 3rd ed. revised. Washington DC: American Psychiatric Association; 1987. 12. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Fourth Edition. Washington DC: American Psychiatric Association; 1994. 13. World Health Organization. The ICD-10 classification of mental and behavioral disorders. Geneva: WHO Publications; 1992. 14. Ross CA. Etiological models and their phenomenological variants. CNS arousal systems: possible role in delirium. Int Psychogeriatr 1991;3(2):353–371.

86

Part III / Neurologic Symptoms

15. Trzepacz PT. Update on the neuropathogenesis of delirium. Dement Geriatr Cogn Disord 1999;10:330–334. 16. Moruzzi G, Magoun HW. Brain stem reticular formation and activation of the EEG. Electroenceph Clin Neurophysiol 1949;1:455–473. 17. Moruzzi G, Magoun HW. Brain stem reticular formation and activation of the EEG. 1949. J Neuropsychiatry Clin Neurosci 1995;7(2):251–267. 18. Perry E, Walker M, Grace J, Perry R. Acetylcholine in mind: a neurotrasmitter correlate of consciousness. Trends Neurosci 1999;22(6):273–280. 19. Fukutani Y, Katsukawa K, Matsubara R et al. Delirium associated with Joseph disease. J Neurol Neurosurg Psychiatry 1993;56(11):1207–1212. 20. Mesulam MM, Waxman SG, Geschwind N et al. Acute confusional states with right middle cerebral artery infarctions. J Neurol Neurosurg Psychiatry 1976;39(1):84–89. 21. Baht R, Rockwood K. Delirium as a disorder of consciousness. J Neurol Neurosurg Psychiat 2007. 22. Folstein M, Folstein S, McHugh P. Mini-mental state. J Psychiatr Res 1975;12:189–198. 23. Fayers PM, Hjermstad MJ, Ranhoff AH et al. Which Mini-Mental State Exam items can be used to screen for delirium and cognitive impairment? J Pain Symptom Manage 2005;30(1):41–50. 24. Kaufer D, Catt K, Lopez O, De KS. Dementia with Lewy bodies: response of delirium-like features to donepezil. Neurology 1998;51(5):1512. 25. Blass JP, Gibson GE. Cerebrometabolic aspects of delirium in relationship to dementia. Dement Geriatr Cogn Disord 1999;10(5): 335–338. 26. Gibson GE, Blass JP, Huang HM et al. The cellular basis of delirium and its relevance to age-related disorders including Alzheimer’s disease. Int Psychogeriatr 1991;3(2):373–395. 27. Stefano GB, Bilfinger TV, Fricchione GL. The immune neuro-link and the macrophage: postcardiotomy delirium, HIV associate dementia and psychiatry. Prog Neurobiol 1994;42(4):475–488. 28. Beloosesky Y, Hendel D, Weiss A et al. Cytokines and C-reactive protein production in hip-fracture-operated elderly patients. J Gerontol A Biol Sci Med Sci 2007;62(4):420–426. 29. Flacker JM, Lipsitz LA. Serum anticholinergic activity changes with acute illness in elderly medical patients. J Gerontol MS 1999;54A:M12–M6. 30. Barbato M, Rodriguez PJ. Thiamine deficiency in patients admitted to a palliative care unit. Pall Medicine 1994;8:320–324. 31. O’Keeffe, Tormey WP, Glasgow R et al. Thiamine deficiency in hospitalized elderly patients. Gronyology 1994;40(1):18–24. 32. Papersack T, Garbusinski J, Robberecht J et al. Clinical relevance of thiamine status amongst hospitalized elderly patients. Gerontology 1999;45:96–101. 33. Flacker JM, Lipsitz LA. Neural mechanisms of delirium: current hypotheses and evolving concepts. J Gerontol Biol Sci 1999;54A:B239–B46. 34. Rudolph JL, Jones RN, Grande LJ et al. Impaired executive function is associated with delirium after coronary artery bypass graft surgery. J Am Geriatr Soc 2006;54(6):937–941. 35. Trzepacz PT. The neuropathogenesis of delirium: a need to focus our research. Psychosomatics 1994;35:374–91. 36. Practice guideline for the treatment of patients with delirium. American Psychiatric Association. Am J Psychiatry 1999;156(5 Suppl): 1–20. 37. Peterson JF, Pun BT, Dittus RS et al. Delirium and its motoric subtypes: a study of 614 critically ill patients. J Am Geriatr Soc 2006;54(3):479–484. 38. Ljubisavljevic V, Kelly B. Risk factors for development of delirium among oncology patients. Gen Hosp Psychiatry 2003;25(5): 345–352. 39. Minagawa H, Yosuke U, Yamawaki S, Ishitani K. Psychiatric morbidity in terminally ill cancer patients. Cancer 1996;78:1131–1137. 40. Lawlor PG, Gagnon B, Mancini IL et al. Occurrence, causes, and outcome of delirium in patients with advanced cancer. Arch Intern Med 2000;160:786–794. 41. Caraceni A, Nanni O, Maltoni M et al. Impact of delirium on the short-term prognosis of advanced cancer patients. Italian Multicenter Study Group on Palliative Care. Cancer 2000;89(5):1145–1149. 42. Gagnon P, Allard P, Masse B, DeSerres M. Delirium in terminal cancer: a prospective study using daily screening, early diagnosis and continuous monitoring. J Pain Symptom Manage 2000;19(6):412–426. 43. Liptzin B, Levkoff SE, Cleary PD et al. An empirical study of diagnostic criteria for delirium. Am J Psychiatry 1991;148(4):454–457. 44. Lewis LM, Miller DK, Morley JE, et al. Unrecognized delirium in ED geriatric patients. Am J Emerg Med 1995;13(2):142–145. 45. Caraceni A, Grassi L. Delirium: Acute Confusional States in Palliative Medicine. Oxford: Oxford University Press; 2003. 46. Wolf HG, Curran D. Nature of delirium and allied states: the disergastic reaction. Arch Neurol Psychiatry 1935;33:1175–1215. 47. Liptzin B, Levkoff SE. An empyrical study of delirium subtypes. Br J Psychiatry 1992;161:843–845. 48. Spiller JA, Keen JC. Hypoactive delirium: assessing the extent of the problem for inpatient specialist palliative care. Palliat Med 2006;20(1):17–23. 49. Pandharipande P, Cotton BA, Shintani A et al. Motoric subtypes of delirium in mechanically ventilated surgical and trauma intensive care unit patients. Intensive Care Med 2007;online. 50. Breitbart W, Gibson C, Tremblay A. The delirium experience: delirium recall and delirium–related distress in hospitalized patients with cancer, their spouses/caregivers, and their nurses. Psychosomatics 2002;43(3):183–194. 51. Chedru F, Geschwind N. Writing disturbances in acute confusional state. Neuropsychologia 1972;10:343–353. 52. Macleod AD, Whitehead LE. Dysgraphia in terminal delirium. Palliat Med 1997;11:127–132. 53. DSM–IV TFo. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 2000. 54. Pisani F, Oteri G, Costa C et al. Effects of psychotropic drugs on seizure threshold. Drug Saf 2002;25(2):91–110.

Chapter 6 / Confusion and Delirium

87

55. Taylor MA, Fink M. Catatonia in psychiatric classification: a home of its own. Am J Psychiatry 2003;160(7):1233–1241. 56. Inouye SK, van Dyck CH, Alessi CA et al. Clarifying confusion: the confusion assessment method—-a new method for detection of delirium. Ann Intern Med 1990;113(12):941–948. 57. Zou Y, Cole M, Primeau F et al. Detection and diagnosis of delirium in the elderly: psychiatrist diagnosis, confusion assessment method, or consensus diagnosis? Int Psychogeriatr 1998;10(3): 303–308. 58. Trzepacz PT, Baker RW, Greenhouse J. A symptom rating scale for delirium. Psychiatry Res 1988;23(1):89–97. 59. Breitbart W, Rosenfeld B, Roth A et al. The memorial delirium assessment scale. J Pain Symptom Manage 1997;13(3):128–137. 60. Smith MJ, Breitbart WS, Platt MM. A critique of instruments and methods to detect, diagnose, and rate delirium. J Pain Symptom Manage 1994;10:35–77. 61. Hjermstad M, Loge JH, Kaasa S. Methods for assessment of cognitive failure and delirium in palliative care patients: implications for practice and research. Palliat Med 2004;18(6):494–506. 62. Gaudreau JD, Gagnon P, Harel F et al. Impact on delirium detection of using a sensitive instrument integrated into clinical practice. Gen Hosp Psychiatry 2005;27(3):194–199. 63. Pompei P, Foreman M, Cassel C et al. Detecting delirium among hospitalized older patients. Arch Intern Med 1995;155:301–307. 64. Rockwood K, Gooodman J, Flynn M et al. Cross-validation of the delirium rating scale in older patients. J Am Geriatr Soc 1996;44:839–842. 65. Grassi L, Caraceni A, Beltrami E et al. Assessing delirium in cancer patients: the Italian versions of the delirium rating scale and the memorial delirium assessment scale. J Pain Symptom Manage 2001;21(1):59–68. 66. Trzepacz PT, Mittal D, Torres R et al. Validation of the delirium rating scale—-revised–98: comparison with the delirium rating scale and the cognitive test for delirium. J Neuropsychiatry Clin Neurosci 2001;13(2):229–242. 67. Trzepacz PT, Brenner RP, Coffman GC et al. Delirium in liver transplantation candidates: discriminant analysis of multiple test variables. Biol Psychiatry 1988;24:3–15. 68. Jacobson SA, Leuchter AF, Walter DO. Conventional and quantitative EEG in the diagnosis of delirium among the elderly. J Neurol Neurosurg Psychiatry 1993;56:153–158. 69. Jacobson S, Jerrier H. EEG in delirium. Semin Clin Neuropsychiatry 2000;5(2):86–92. 70. Young GB, McLachlan RS, Kreeft JH et al. An electroencephalographic classification system for coma. Can J Neurol Sci 1997;24: 329–325. 71. Wengs WJ, Talwar D, Bernard J. Ifosfamide–induced nonconvulsive status epilepticus. Arch Neurol 1993;50:1104–1105. 72. Schmidley JW, Messing RO. Agitated confusional states in patients with right hemisphere infarctions. Stroke 1984;5:883–885. 73. Mori E, Yamadori A. Acute confusional state and acute agitated delirium: occurrence after infarction in the right middle cerebral artery territory. Arch Neurol 1987;44:1139–1143. 74. Henon H, Lebert F, Durieu I et al. Confusional state in stroke: relation to preexisting dementia, patient characteristics, and outcome. Stroke 1999;30:773–779. 75. Wasserstrom WR, Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49(4):759–772. 76. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. 77. Formaglio F, Caraceni A. Meningeal metastases: clinical aspects and diagnosis. Ital J Neurol Sci 1998;19(3):133–149. 78. Weitzener MA, Olofson SM, Forman AD. Patients with malignant meningitis presenting with neuropsychiatric manifestations. Cancer 1995;76:1804–1808. 79. Grossman SA, Trump DL, Chen DC et al. Cerebrospinal fluid flow abnormalities in patients with neoplastic meningitis: an evaluation using 111indium-DTPA ventriculography. Am J Med 1982;73(5):641–647. 80. Trzepacz P, Ho V, Mallavarapu H. Cholinergic delirium and neurotoxicity associated with tacrine for Alzheimer’s dementia. Psychosomatics 1996;37(3):299–301. 81. Bruera E, Macmillan K, Hanson J et al. The cognitive effects of the administration of narcotic analgesics in patients with cancer pain. Pain 1989;39(1):13–16. 82. Zacny J. A review of the effects of opioids on psychomotor function and cognitive functioning in humans. Exp Clin Psychopharmacol 1995;3(4):432–466. 83. Sjogren P, Olsen AK, Thomsen AB et al. Neuropsychological performance in cancer patients: the role of oral opioids, pain, and performance status. Pain 2000;86(3):237–245. 84. Tuma R, DeAngelis LM. Altered mental status in patients with cancer. Arch Neurol 2000;57(12):1727–1731. 85. Gaudreau JD, Gagnon P, Roy MA et al. Opioid medications and longitudinal risk of delirium in hospitalized cancer patients. Cancer 2007;109(11):2365–2673. 86. Jellema JG. Hallucination during sustained-release morphine and methadone administration. Lancet 1987;2(8555):392. 87. Caraceni A, Martini C, De Conno F et al. Organic brain syndromes and opioid administration for cancer pain. J Pain Symptom Manage 1994;9(8):527–533. 88. Hagen N, Swanson R. Strychnine-like multifocal myoclonus and seizures in extremely high-dose opioid administration: treatment strategies. J Pain Symptom Manage 1997;14:51–58. 89. Sjogren P, Jonsson T, Jensen NH et al. Hyperalgesia and myoclonus in terminal cancer patients treated with continuous intravenous morphine. Pain 1993;55(1):93–97. 90. Gaudreau JD, Gagnon P, Harel F et al. Psychoactive medications and risk of delirium in hospitalized cancer patients. J Clin Oncol 2005;23(27):6712–6718. 91. Francis J, Kapoor WN. Prognosis after hospital discharge of older medical patients with delirium. J Am Geriatr Soc 1992;40:601–606. 92. Schor JD, Levkoff SE, Lipsitz LA et al. Risk factors for delirium in hospitalized elderly. JAMA 1992;267(6):827–831.

88

Part III / Neurologic Symptoms

93. Inouye SK, Viscoli CM, Horwitz RI et al. A predictive model for delirium in hospitalized elderly medical patients based on admission characteristics. Ann Intern Med 1993;119(6):474–481. 94. Ferguson JA, Suelzer CJ, Eckert GJ et al. Risk factors for delirium tremens development. J Gen Intern Med 1996;11(7):410–414. 95. Inouye SK, Charpentier PA. Precipitating factors for delirium in hospitalized elderly persons: predictive model and interrelationship with baseline vulnerability. JAMA 1996;275(11):852–857. 96. Inouye SK, Bogardus ST, Charpentier PA et al. A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med 1999;340(9):669–676. 97. Turner RC, Lichstein PR, Peden JG, Jr. et al. Alcohol withdrawal syndromes: a review of pathophysiology, clinical presentation, and treatment. J Gen Intern Med 1989;4(5):432–444. 98. van Munster BC, Korevaar JC, de Rooij SE et al. Genetic polymorphisms related to delirium tremens: a systematic review. Alcohol Clin Exp Res 2007;31(2):177–184. 99. Seibert CP. Recognition, management, and prevention of neuropsychological dysfunction after operation. Int Anesthesiol Clin 1986;24(4):39–58. 100. Smith LW, Dimsdale JE. Postcardiotomy delirium: conclusions after 25 years? Am J Psychiatry 1989;146(4):452–458. 101. Berggren D, Gustafson Y, Eriksson B et al. Postoperative confusion after anesthesia in elderly patients with femoral neck fractures. Anesth Analg 1987;66(6):497–504. 102. Marcantonio ER, Goldman L, Mangione CM et al. A clinical prediction rule for delirium after elective noncardiac surgery. JAMA 1994;271(2):134–139. 103. O’Keeffe ST, Chonchubhair A. Postoperative delirium in the elderly. Br J Anaesth 1995;73(5):673–687. 104. Marcantonio ER, Juarez G, Goldman L et al. The relationship of postoperative delirium with psychoactive medications. JAMA 1994;272(19):1518–1522. 105. Marcantonio ER, Goldman L, Orav EJ et al. The association of intraoperative factors with the development of postoperative delirium. Am J Med 1998;105(5):380–384. 106. O’Keeffe ST. Clinical subtypes of delirium in the elderly. Dement Geriatr Cogn Disord 1999;10:380–385. 107. Carlson LA, Gottfries CG, Winbland B et al. (eds.). Delirium in the elderly: epidemiological, pathogenetic, and treatment aspects. Dement Geriatr Cog Disord 1999;10(5):306–429. 108. Farrell KR, Ganzini L. Misdiagnosing delirium as depression in medically ill elderly patients. Arch Intern Med 1995;155:2459–2464. 109. Morita T, Tsunoda J, Inoue S et al. The palliative prognostic index: a scoring system for survival prediction of terminally ill cancer patients. Support Care Cancer 1999;7(3):128–133. 110. Casarett DJ, Inouye SK. Diagnosis and management of delirium near the end of life. Ann Intern Med 2001;135(1):32–40. 111. Ripamonti C, Bruera E. CNS adverse effects of opioids in cancer patients: guidelines for treatment. CNS Drugs 1997;8(1):21–37. 112. Guideline Watch: Practice guideline for the treatment of patients with delirium. 2006. 113. Inouye SK, Bogardus ST, Jr., Williams CS et al. The role of adherence on the effectiveness of nonpharmacologic interventions: evidence from the delirium prevention trial. Arch Intern Med 2003;163(8):958–964. 114. Borreani C, Caraceni A, Tamburini M. Counselling for the confused patient and the family. In: Portenoy RK, Bruera E, eds. Topics in Palliative Care; 1997:45–54. 115. Centeno C, Sanz A, Bruera E. Delirium in advanced cancer patients. Palliat Med 2004;18(3):184–194. 116. Morita T, Hirai K, Sakaguchi Y et al. Family-perceived distress from delirium–related symptoms of terminally ill cancer patients. Psychosomatics 2004;45(2):107–113. 117. Bruera E, Fainsinger RL, Miller MJ et al. The assessment of pain intensity in patients with cognitive failure: a preliminary report. J Pain Symptom Manage 1992;7(5):267–270. 118. Coyle N, Breitbart W, Weaver S et al. Delirium as a contributing factor to “crescendo” pain: three case reports. J Pain Symptom Manage 1994;9(1):44–47. 119. British Geriatrics Society and Royal College of Physicians. Guidelines for the prevention, diagnosis, and management of delirium in older people: concise guidance to good practice series. No 6. London: Royal College of Physicians; 2006. 120. Breitbart W, Marotta R, Platt MM et al. A double-blind trial of haloperidol, chlorpromazine, and lorazepam in the treatment of delirium in hospitalized AIDS patients. Am J Psychiatry 1996;153(2):231–237. 121. Wirshing WC. Movement disorders associated with neuroleptic treatment. J Clin Psychiatry 2001;62 Suppl 21:15–18. 122. Strawn JR, Keck PE, Jr., Caroff SN. Neuroleptic malignant syndrome. Am J Psychiatry 2007;164(6):870–876. 123. Wilt JL, Minnema AM, Johnson RF et al. Torsade de pointes associated with the use of intravenous haloperidol. Ann Intern Med 1993;119(5):391–4. 124. Hunt N, Stern TA. The association between intravenous haloperidol and torsades de pointes. Psychosomatics 1995;36(6):541–9. 125. DiSalvo TG, O’Gara PT. Torsade de pointes caused by high–dose intravenous halpoeridol in cardiac patients. Clin Cardiol 1995;18(5):285–290. 126. Jackson T, Ditmanson L, Phibbs B. Torsade de pointes and low–dose oral haloperidol. Arch Intern Med 1997;157(17):2013–2015. 127. Tei Y, Morita T, Inoue S et al. Torsades de pointes caused by a small dose of risperidone in a terminally ill cancer patient. Psychosomatics 2004;45(5):450–451. 128. Stollberger C, Huber JO, Finsterer J. Antipsychotic drugs and QT prolongation. Int Clin Psychopharmacol 2005;20(5):243–251. 129. Justo D, Prokhorov V, Heller K et al. Torsade de pointes induced by psychotropic drugs and the prevalence of its risk factors. Acta Psychiatr Scand 2005;111(3):171–176. 130. Zeltser D, Justo D, Halkin A et al. Torsade de pointes due to noncardiac drugs: most patients have easily identifiable risk factors. Medicine (Baltimore) 2003;82(4):282–290. 131. Liptzin B. Delirium. In: Sadavoy J, Lazarus L, Jarvik L et al. (eds.). Comprehensive Review of Geriatric Psychiatry. 2nd ed. Washington, DC: American Psychiatric Press; 1996:479–495.

Chapter 6 / Confusion and Delirium

89

132. Shapiro BA, Warren J, Egol AB et al. Practice parameters for intravenous analgesia and sedation for adult patients in the intensive care unit: an executive summary. Society of Critical Care Medicine. Crit Care Med 1995;23(9):1596–1600. 133. Tesar GE, Murray GB, Cassem NH. Use of high-dose intravenous haloperidol in the treatment of agitated cardiac patients. J Clin Psychopharmacol 1985;5(6):344–347. 134. Riker RR, Fraser GL, Cox PM. Continuous infusion of haloperidol controls agitation in critically ill patients. Crit Care Med 1994;22(3):433–440. 135. Levenson JL. High-dose intravenous haloperidol for agitated delirium following lung transplantation. Psychosomatics 1995;36(1): 66–68. 136. Lonergan E, Britton AM, Luxenberg J et al. Antipsychotics for delirium. Cochrane Database Syst Rev 2007(2):CD005594. 137. Riker RR, Fraser GL. Adverse events associated with sedatives, analgesics, and other drugs that provide patient comfort in the intensive care unit. Pharmacotherapy 2005;25(5 Pt 2):8S–18S. 138. O’Brien JM, Rockwood RP, Suh KI. Haloperidol-induced torsade de pointes. Ann Pharmacother 1999;33(10):1046–1050. 139. Riker RR, Fraser GL, Richen P. Movement disorders associated with withdrawal from high–dose intravenous haloperidol therapy in delirious ICU patients. Chest 1997;111(6):1778–1781. 140. Di Salvo TG, O’Gara PT. Torsade de pointes caused by high–dose intravenous haloperidol in cardiac patients. Clin Cardiol 1995;18(5):285–290. 141. Thomas H, Schwartz E, Petrilli R. Droperidol versus haloperidol for chemical restraint of agitated and combative patients. Ann Emerg Med 1992;21:407–413. 142. Frye MA, Coudreaut MF, Hakeman SM et al. Continuous droperidol infusion for management of agitated delirium in an intensive care unit. Psychosomatics 1995;36(3):301–305. 143. Sipahimalani A, Masand P. Use of risperidone in delirium: case reports. Ann Clin Psychiatry 1997;9(2):105–107. 144. Breitbart W, Tremblay A, Gibson C. An open trial of olanzapine for the treatment of delirium in hospitalized cancer patients. Psychosomatics 2002;43(3):175–182. 145. Boettger S, Breitbart W. Atypical antipsychotics in the management of delirium: a review of the empirical literature. Palliat Support Care 2005;3(3):227–237. 146. Lacasse H, Perreault MM, Williamson DR. Systematic review of antipsychotics for the treatment of hospital–associated delirium in medically or surgically ill patients. Ann Pharmacother 2006;40(11):1966–1973. 147. Mayo–Smith MF, Beecher LH, Fischer TL et al. Management of alcohol withdrawal delirium: an evidence-based practice guideline. Arch Intern Med 2004;164(13):1405–1412. 148. Stanley KM, Worrall CL, Lunsford SL et al. Experience with an adult alcohol withdrawal syndrome practice guideline in internal medicine patients. Pharmacotherapy 2005;25(8):1073–1083. 149. Arcangeli A, Antonelli M, Mignani V et al. Sedation in PACU: the role of benzodiazepines. Curr Drug Targets 2005;6(7):745–748. 150. Bottomley DM, Hanks GW. Subcutaneous midazolam infusion in palliative care. J Pain Symptom Manage 1990;5(4):259–261. 151. Burke AL, Diamond PL, Hulbert J et al. Terminal restlessness: its management and the role of midazolam. Med J Aust 1991;155(7): 485–487. 152. Stiefel F, Fainsinger R, Bruera E. Acute confusional states in patients with advanced cancer. J Pain Symptom Manage 1992;7(2):94–98. 153. Burke AL. Palliative care: an update on “terminal restlessness. Med J Aust 1997;166:39–42. 154. Morita T, Tei Y, Inoue S. Correlation of the dose of midazolam for symptom control with administration periods: the possibility of tolerance. J Pain Symptom Manage 2003;25(4):369–375. 155. Fernandez F, Levy JK, Mansell PW. Management of delirium in terminally ill AIDS patients. Int J Psychiatry Med 1989;19(2):165–172. 156. Fernandez F, Adams F, Levy JK et al. Cognitive impairment due to AIDS-related complex and its response to psychostimulants. Psychosomatics 1988;29(1):38–46. 157. Arbour RB. Propylene glycol toxicity related to high-dose lorazepam infusion: case report and discussion. Am J Crit Care 1999;8(1):499–506. 158. Cawley MJ. Short-term lorazepam infusion and concern for propylene glycol toxicity: case report and review. Pharmacotherapy 2001;21(9):1140–1144. 159. Parker MG, Fraser GL, Watson DM et al. Removal of propylene glycol and correction of increased osmolar gap by hemodialysis in a patient on high-dose lorazepam infusion therapy. Intensive Care Med 2002;28(1):81–84. 160. Mullins ME, Barnes BJ. Hyperosmolar metabolic acidosis and intravenous Lorazepam. N Engl J Med 2002;347(11):857–858; author reply –8. 161. Slatkin NE, Rhiner M, Bolton TM. Donepezil in the treatment of opioid-induced sedation: report of six cases. J Pain Symptom Manage 2001;21(5):425–438. 162. Levenson JA. Should psychostimulants be used to treat delirious patients with depressed mood. J Clin Psychiatry 1992;53(2):69. 163. Morita T, Otani H, Tsunoda J et al. Succesful palliation of hypoactive delirium due to multiorgan failure by oral methylphenidate. Supp Care Cancer 2000;8(2):134–137. 164. Gagnon B, Low G, Schreier G. Methylphenidate hydrochloride improves cognitive function in patients with advanced cancer and hypoactive delirium: a prospective clinical study. J Psychiatry Neurosci 2005;30(2):100–107. 165. Burns MJ, Linden CH, Graudins A, Brown RM, Fletcher KE. A comparison of physostigmine and benzodiazepines for the treatment of anticholinergic poisoning. Ann Emerg Med 2000;35(4):374–381. 166. Kalisvaart KJ, de Jonghe JF, Bogaards MJ et al. Haloperidol prophylaxis for elderly hip-surgery patients at risk for delirium: a randomized placebo-controlled study. J Am Geriatr Soc 2005;53(10):1658–1666. 167. Siddiqi N, Stockdale R, Britton AM et al. Interventions for preventing delirium in hospitalised patients. Cochrane Database Syst Rev 2007(2):CD005563.

90

Part III / Neurologic Symptoms

168. Koponen HJ, Riekkinen PJ. A prospective study of delirium in elderly patients admitted to a psychiatric hospital. Psychol Med 1993;23:103–109. 169. Levkoff SE, Evans DA, Liptzin B et al. Delirium: the occurrence and persistence of symptoms among elderly hospitalized patients. Arch Intern Med 1992;152(2):334–340. 170. van Hemert AM, van der Mast RC, Hengeveld MW et al. Excess mortality in general hospital patients with delirium: a 5-year follow-up of 519 patients seen in psychiatric consultation. J Psychosom Res 1994;38(4):339–346. 171. Leslie DL, Zhang Y, Holford TR et al. Premature death associated with delirium at 1-year follow-up. Arch Intern Med 2005;165(14):1657–1662.

7

Cognitive Dysfunction, Mood Disorders, and Fatigue Elana Farace,

PHD,

and Zarui Melikyan,

PHD

CONTENTS Introduction Cognitive Dysfunction in Cancer Patients Mood and Psychiatric Disorders in Cancer Patients Fatigue in Cancer Patients Conclusion References

Summary Changes in neurocognition, such as in memory, language, or speed of thinking, are among the most frequent complaints in cancer survivors. The degree of cognitive deficits may vary from very subtle changes in areas unimportant to the patient to very pronounced difficulties in several cognitive domains preventing the person from independent functioning or even requiring hospitalization. Neurocognitive deficits have a significant effect on patients’ quality of life (QOL) with more severe impairment leading to lower QOL. Cancer treatments such as radiation and chemotherapy have the potential to interact with neurocognition. Additionally, changes in mood and other psychiatric symptoms, and fatigue, can have significant impact on quality of life. Key Words: quality of life, neurocognition, neuropsychology, depression, fatigue

1. INTRODUCTION The concept of quality of life (QOL), roughly defined as “one’s overall enjoyment of life,” has been less studied than curative therapy in cancer. The recent Institute of Medicine publication on cancer survivorship (1) shows that there have been more than 30,000 times more publications with a curative focus than there have been publications regarding improving cancer survivorship. Survival is, of course, the primary goal of cancer patients and healthcare providers, but emphasis on quality of life and symptom management has become increasingly important. Increased survival along with good QOL is the ultimate goal for patients and their caregivers. QOL is considered to be composed of several components: physical, such as the ability to perform activities of daily living; psychological, such as cognitive functioning and emotional status; and social, which includes a person’s roles and functions and relationships with others (2). In this chapter, we will discuss the impact of cancer on patients’ cognitive functioning, the impact of different treatments on cognitive functioning, and the importance of assessment of cognitive functioning to improve adjustment to cognitive problems in cancer patients. We then will discuss mood disorders and fatigue in cancer patients and how these symptoms contribute to cognitive functioning and overall perception of quality of life. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

91

92

Part III / Neurologic Symptoms

2. COGNITIVE DYSFUNCTION IN CANCER PATIENTS Multiple factors contribute to the development of neurocognitive deficits in cancer patients. Interacting factors include surgery, chemotherapy, or radiation therapy (or any combinations of those), stage of disease, and speed of the disease progression. Each of these will be discussed in turn below, with the emphasis on neurocognitive function (NCF) in patients with structural brain lesions, supplemented by discussions of neurocognition in patients with other types of cancer. Finally, both depression and fatigue significantly affect neurocognitive functioning; these factors will be discussed below in Sections 3 and 4.

2.1. Neurocognitive Functioning in Patients with Brain Tumors For both metastatic and primary brain tumors, tumor size and location play a critical role in neurocognitive symptom development (3). The larger the tumor size, the more likely it is to affect multiple brain regions and cause deficits in multiple modalities (3). Cognitive impairment in patients with brain tumors that precedes surgery, radiotherapy, or chemotherapy is generally related to lesion location (3–5). The cognitive consequences of tumors involving cortical structures will be described next. As might be expected, tumors in the frontal lobes may cause executive function disorders: difficulty planning and regulating one’s own activities, acting inappropriately for the situation, and displaying flat or inappropriate affect (6–9). Depending on the severity of the deficit, these patients may need to be closely supervised in their everyday lives, particularly in those cases where patients have agnosagnosia, in that they do not realize the extent of their problems (10,11). Difficulties building up phrases and whole speech passages, troubles initiating speech, and echolalias may be a hallmark of patients who have posterior frontal lesions (8,9,12,13). Additionally, the tumor itself and associated edema may damage the rich afferent and efferent connections that the frontal region has with other brain regions, which can interrupt modulatory frontal influences on brain function (14,15). Thus, even if a patient does not have a frontal brain lesion, many patients exhibit impairment of executive frontal lobe function manifested by apathy, lack of motivation, lack of spontaneity, impaired attention, impaired working memory, and difficulty shifting mental set (16). Patients with temporal lobe lesions may demonstrate a variety of language problems. Patients with dominant hemisphere temporal lesions usually manifest disturbances of verbal speech components, and patients with nondominant hemisphere temporal lesions demonstrate nonverbal speech disturbances. Patients with dominant hemisphere temporal lesions may have difficulties with phonemic analysis (distinguishing between semantically meaningful phoneme characteristics, e.g., length of the sounds), difficulties naming objects or with word-finding problems, and difficulties comprehending ideas from speech or written text especially if complicated grammatical constructions and long and complicated sentences are used. They also may have troubles with verbal learning and memory, verbal reasoning, and contralateral motor dexterity (6,9,17). Nondominant hemisphere temporal lobe tumor location may cause some intonation and prosodic speech components changes, difficulties in perceiving emotions in the speech of others, or expressing emotions in one’s own speech. They may also have difficulties with visual-perceptual skills and left-sided motor dexterity (6,9,17). Parietal lobe tumors may cause spatial orientation problems and difficulties with the expression of any kind of inter-relationships between objects and/or events. Patients with right parietal lobe tumors may develop agnosagnosia, a condition of being unaware of personal deficits. Finally, patients with occipital lobe tumors may have difficulties with visual perception of objects by either misinterpreting the images or requiring more time and effort to perceive them correctly (6,7,9). Noncortical tumors lead to a variety of symptoms (18). Cerebellar posterior-lateral regions seem to be relevant for cognition, while vermian lesions seem to be associated with changes in affect (19). However, the results remain controversial: several authors have found neuropsychological deficits in patients with a variety of cerebellar diseases, while others have reduced the deficits to motor disabilities. In the former group are Schmahmann and Sherman (20), who postulated a “cerebellar cognitive affective syndrome” which implies an impairment of executive functions as well as disturbances in spatial cognition, language deficits, and personality changes. Such deficits have been attributed to the disruption of neural circuits linking prefrontal, temporal, posterior parietal, and limbic cortices with the cerebellum. Right cerebellar lesions lead to verbal deficits because of the crossed pathways, whereas in left cerebellar lesions spatial deficits are most prominent. This hypothesis has been

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

93

confirmed for children with cerebellar lesions arising from tumor resection (21,22). Other studies have shown deficits resulting from cerebellar lesions in the areas of verbal fluency (23,24), error detection (25), planning (26), effortful memory (27), nonmotor associative learning (28), spatial attention (29), and shifting attention (30). However, other researchers have failed to replicate these results (31,32) or doubt the cognitive function of the cerebellum, instead explaining possible deficits by motor impairment or methodological problems (33). Patients with third ventricle tumors are at risk for developing impairments in memory, executive function, and fine motor speed and dexterity, which are domains associated with frontal subcortical functions. In one study (34), neuropsychological outcomes of 33 patients with primary brain tumors in the third ventricular region were explored. A retrospective analysis of neuropsychological data of patients—26 of whom underwent surgery and 7 of whom were not treated by surgical intervention—showed that different types of treatment were not associated with differential cognitive sequelae, and surgical intervention did not account for cognitive deficits. However, tumor pathology does not seem to be predictive of neurocognitive function. In a retrospective study (35) the relationship between tumor histology and post-surgical cognitive function in patients diagnosed with malignant brain tumors was investigated. Results showed that tumor histology is not clearly predictive of cognitive performance in adults with anaplastic astrocytoma and glioblastoma multiforme.

2.2. Importance of Neuropsychological Assessment Although the involvement of neuropsychological science in the study of brain tumor patients is relatively recent, there are similarities between the neuropsychology of neuro-oncology and that of dementia and acquired brain injuries (2). To that end, long-term cognitive impairment in patients with low-grade gliomas has been compared to that seen in patients with mild-to-moderate traumatic brain injury (36). Neurocognitive and mental status changes are among the most common presenting symptoms of primary and metastatic brain tumors, occurring in approximately 30% of cases in one careful report of presenting symptoms of a group of 729 patients with intracranial metastases (37). This can be compared to the frequency of other presenting symptoms: headache (31%), weakness (24%), seizure (19%), ataxia (11%), visual change (5%), nausea or vomiting (4%), bulbar symptoms, dizziness and syncope (4%), sensory change (2%), papilledema (0.5%), and no symptoms or signs (9%) (38). Accurate measurement of neurocognitive function is somewhat different from the use of standard QOL measures, and the two should be considered to be separate constructs (39). For example, cognitive deterioration has been demonstrated to be a reliable predictor of radiographic disease progression in a study of 56 patients with recurrent brain tumors (40) where cognitive deterioration occurred on average 6 weeks prior to radiographic failure. In the same study, however, measures of QOL and activities of daily living were not strongly tied to cognitive decline or to time to tumor progression, suggesting that these QOL measures may not be sufficiently sensitive to change in patient function, although they are important measures in terms of patient care (40). Most QOL measures, a standard physician exam, or other methods of informal assessment of neuropsychological function depend on the patient’s report of their neurocognitive symptoms. There is a great deal of evidence, however, suggesting that cancer patients’ cognitive complaints usually indicate feelings of anxiety, depression, or fatigue rather than neurocognitive dysfunction (41,42). Moreover, a dissociation between objective cognitive test results and self-reported cognitive function holds true especially for patients with brain cancer whose judgment may be severely impaired by the tumor (43). Recently, a brief neurocognitive battery comprised of tests measuring memory, visual motor function, executive function, motor dexterity, verbal fluency, and functional independence has been incorporated into more and more research including several recent cooperative group trials through the Radiation Therapy Oncology Group and North Central Cancer Treatment Group. This battery (Table 1) has been shown to be quick and easy to administer by nonphysicians with very good compliance and motivation of patients with brain metastases. Several studies have indicated that a multifaceted assessment of cognition, QOL, and patient function is practical for brain tumor patients in clinical trials and can provide information regarding the relative risks versus benefits of new treatment regimens that supplements the information from the usual clinical variables (44). Therefore. neurocognitive testing can be added prospectively to clinical trials that compare outcomes of different therapeutic interventions for patients with metastatic brain tumors (45).

94

Part III / Neurologic Symptoms

Table 1 Brief Neurocognitive Battery Test

Domain Measured

Outcome

Trail Making Test A

Visual scanning speed

Number of seconds to complete (0–300)

Trail Making Test B

Divided attention

Number of seconds to complete (0–300)

Controlled Oral Word Association

Verbal fluency

Age and sex-adjusted raw score (range 0–no upper limit)

Hopkins Verbal Learning Test

Verbal memory

Immediate memory of word list rehearsed three times (maximum score = 36). After 20–30 min delay, number of words correctly recalled (maximum score = 12). Recognition is number of words recognized from a longer list (maximum score = 12).

Digit Symbol Subtest of the WAIS-III

Psychomotor speed

Age-corrected subtest score (0–20).

Grooved Pegboard Test

Fine motor control for dominant and nondominant hands

Number of seconds to complete (0–300)

Awareness that QOL is at least as important as survival to patients is increasing in neuro-oncology (46,47). Neurocognitive impairment has been demonstrated to be a significant predictor of survival in numerous other neurological disorders such as multiple sclerosis (48) and dementia (49). Both depression and neurocognitive impairment have been shown to be independent predictors of mortality in medically ill older adults (50). As compared to standard neurological evaluation, neuropsychological assessment has proved to be more precise and reliable diagnostic tool for cognitive disorders in a study of 35 patients with low-grade glioma (51). An improved understanding of neurocognitive sequelae has the potential to help physicians, patients, and families make informed choices bearing upon QOL and survivorship, perhaps even including length of survival. Ideally, elucidation of these factors will improve decision-making by patients, families, and the clinicians who care for them. The ultimate goal is to maximize both survival and QOL (39).

2.3. Neurocognitive Side Effects of Cancer Treatment Recent scientific and technologic advances have permitted substantial advances in cancer management. However, aggressive treatments with combinations of surgery, radiation, chemotherapy, and immunotherapy may produce significant disruption of the CNS, resulting in disturbances of everyday and cognitive functioning. The limited data available regarding interaction of treatments and neurocognitive function are discussed below. 2.3.1. Surgery Brain tumor resection has been shown to have more favorable cognitive outcomes than biopsy, although this is at least partially due to the fact that biopsies are the preferred method for large or deeply located tumors (43). Although surgery has the potential to induce deficits through focal damage to surrounding tissue, increased risk

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

95

of hemorrhage and ischemia, and so on, it may also improve function, perhaps through resolution of mass effect, relief of intracranial pressure, or even a benefit of debulking. Benefit of resection was shown in 27 adult patients with various brain tumor pathologies (e.g., craniopharyngiomas, meningiomas, PNETs, central neurocytomas, and all grades of glioma), in a wide range of locations (left, right, callosal, skull base, etc.) (52). Pre- and post-operative neurocognitive testing including measures of attention, executive functioning, verbal memory, psychomotor speed, and fine motor coordination revealed that 84% of patients showed improvement of at least one standard deviation in at least one domain at 6 weeks post-surgery. Thirty-six percent worsened by at least one standard deviation in at least one neurocognitive domain (percentages overlap due to some patients who improved in one domain and worsened in another). Significant improvement was seen for psychomotor speed (52). Among 23 patients with low-grade glioma in language areas who underwent awake tumor resection, preoperatively 91% of patients had verbal working memory disorders, and immediately after surgery 96% had verbal working memory worsening. However, among the eight patients examined 3 months later, five fully recovered their preoperative verbal working memory score and three improved significantly (53). Significant adult brain plasticity was shown in a study of 77 right-handed patients with low-grade gliomas invading primary and/or secondary sensorimotor and/or language areas (as shown anatomically by preoperative MRI and intraoperatively by electrical brain stimulation and cortico-subcortical mapping). The absence of clinical deficits prior to surgery despite the growth of tumor in eloquent regions, occurrence of a deficit immediately after surgery, and almost complete recovery 3 months after surgery may be explained by dynamic spatio-temporal functional reorganization in peritumoral brain and recruitment of compensatory areas with long-term perilesional functional reshaping (54). In the treatment of brain metastases, resection is an important modality for patients with a single lesion, particularly when favorable prognostic factors and systemic disease control are present. Resection is almost always accompanied by adjuvant radiation (38). The benefit of surgery has been assessed in three prospective randomized trials. In the first study, 48 patients were randomly assigned to surgery followed within 14 days by Whole Brain Radiotherapy (WBRT) (36 Gy in 12 fractions) or to WBRT alone (55). Surgery significantly reduced brain metastasis recurrence (20% vs. 52%) and prolonged median survival (40 vs. 15 weeks). Similarly, in a study of 63 patients, those randomly assigned to surgery and WBRT (40 Gy in 2-Gy fractions twice daily) had significantly longer median survival than those who received WBRT alone (10 vs. 6 months) (56). In these studies, patients who underwent surgery were better able to maintain their quality of life as assessed either by KPS or WHO performance status. In a third study, however, there was no difference in quality of life or survival between the group randomly assigned to surgery plus WBRT or to WBRT alone (5.6 vs. 6.3 months; p = .24) (57). Compared with the other two studies, patients in the third study had the highest percentage of disseminated disease (45% vs. 38% and 32% in the first and second studies, respectively). The role of surgery in patients with multiple brain metastases is controversial. In a retrospective review, the median postoperative survival was 6 months for 30 patients who did not have all brain metastases surgically resected, and 14 months for 26 patients who had resection of all brain metastases (58). In comparison, a matched group of patients with single brain metastasis who underwent surgery had a median survival of 14 months. Survival of the group that did not have resection of all lesions was significantly shorter than that of the other two groups (p = .003 and p = .012, respectively). This retrospective analysis, however, should be regarded as hypothesis-generating; no level 1 evidence exists to support the routine use of surgery in patients with multiple brain metastases. 2.3.2. Radiation Therapy Radiotherapy produces a predictable pattern of neurocognitive and neurobehavioral changes, and the time and course of their development are related to treatment parameters, adjuvant therapy, and patient characteristics. The occurrence of radiation-induced neurocognitive changes is best studied in whole brain radiotherapy (WBRT), while the effects of stereotactic radiosurgery and intensity-modulated radiotherapy have not been so thoroughly studied yet. Cranial radiotherapy remains the single greatest advance in the treatment of high-grade gliomas to date, with a doubling of median survival in randomized trials compared with patients not receiving radiotherapy (59,60).

96

Part III / Neurologic Symptoms

Additionally, the use of combined-modality therapy (e.g., radiation plus temozolomide) has allowed the majority of patients to achieve a survival of greater than 1 year, with more than 25% of patients on a multicenter study surviving greater than 2 years (61). Consequently, concerns regarding the potential late cognitive toxicities of cranial radiotherapy for the treatment of brain tumors have increased recently and are now relevant even for patients with high-grade gliomas. The belief that the late effects of treatment account for a significant proportion of cognitive deficits in brain tumor patients (and therefore that modification of treatment may be necessary to decrease this risk) has dominated discussions during the last couple of decades and continues even to this day (62–65). Risk factors for developing radiation-induced cognitive dysfunction are age > 60, fraction size > 2 Gy, higher total dose, greater volume of brain irradiated, hyperfractionated schedules, shorter overall treatment time, concomitant or subsequent use of chemotherapy, and presence of comorbid vascular risk factors (66). WBRT offers tumor shrinkage and palliation in many cases, but some speculate that adverse effects on neurocognitive function may outweigh these benefits. In a study of 208 patients with brain metastases it was shown that WBRT-induced tumor shrinkage, assessed at 4 and 15 months, correlated with better survival and NCF preservation (significant for executive function and fine motor skills). Neurocognitive function was stable or improved in long-term survivors. Tumor progression adversely affected neurocognitive function more than WBRT, thus making enhancement of radiation response a worthwhile aim in this patient population (67). Several studies have focused on enhancing the efficacy of WBRT and thus increasing time to neurologic failure. This has been shown with two radiosensitizers (efaproxiral and motexafin gadolinium), as well as with new systemic agents that synergize or have additive effect with radiation by potentially crossing blood brain barrier (temozolomide, lapatinib, MPC 6827). These agents might also be used potentially to treat micrometastatic disease in a so-called “prophylactic” setting in high-risk patients, reducing tumor development and symptom manifestation. In a phase III randomized trial evaluating survival and neurologic and neurocognitive function in patients with brain metastases from solid tumors receiving WBRT 30 Gy in 10 fractions with or without motexafin gadolinium (MGd) (68), 401 patients were enrolled (251 with non-small-cell lung cancer, 75 with breast cancer, and 75 with other cancers); 90.5% patients had impairment of one or more neurocognitive tests at baseline. Neurocognitive test scores of memory, fine motor speed, executive function, and global neurocognitive impairment at baseline were correlated with brain tumor volume and predictive of survival. There was no statistically significant difference between treatment arms in time to neurocognitive progression. Patients with lung cancer (but not other types of cancer) receiving MGd tended to have improved memory and executive function and improved neurologic function as assessed by a blinded events review committee. The mechanism of radiation-induced neurotoxicity has been attributed to demyelination or vasculopathy (69–72). According to the time of appearance cerebral irradiation injury has been classified into: acute, appearing during the course of radiation; early delayed, occurring a few weeks to a few months after irradiation; and late, occuring a few months to years after. Late effects are usually irreversible, progressive, and sometimes fatal. Essential factors are focal or diffuse necrosis and demyelination of white matter. In support of vasculopathy are characteristic atypical endothelial cells and relatively unique fibrinoid necrosis of small arterial vessels (73). Magnetic resonance imaging (MRI) and pathological studies, however, often do not demonstrate such findings despite significant clinical neurocognitive deficits. The hippocampus has been implicated on clinical and experimental grounds as a principal site of injury (74,75). Several studies suggest that clinical deficits may be secondary to the effect of radiotherapy on neurogenesis in the hippocampus, which has been shown to occur in adult humans (76,77). Radiotherapy primarily affects white matter tracts and cerebral vasculature (78). The relative density of white matter in frontal and subcortical areas contributes to the frequency of deficits found in executive functioning, including impaired processing speed, attention, learning, memory and executive function as well as fine motor function (17,79–81). Motor slowing can occur bilaterally and irrespective of tumor location. These changes may occur even in patients with brain tumors with no evidence of disease recurrence (82). The development of neuropsychological and neurological dysfunction is often the major dose-limiting factor of radiotherapy (79). In some cases, radiation therapy may produce dementia or lead directly to the patient’s death (83). However, not all studies find evidence of neurocognitive impairment that can be linked clearly to radiation, and toxicities are not the same for all patients. For example, when early radiotherapy is given for low-grade gliomas (as opposed to delayed radiotherapy), the data show wide variation in treatment-related neurotoxicity. Worse

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

97

outcomes in this setting are associated with older age at diagnoses, pre-existing neurological disease, radiation dose and fraction, volume of brain irradiated, and concomitant antiepileptic drugs (43). Apart from radiotherapy variables such as overall dose and treatment volume, the size of the fraction dose is largely responsible for the development of late neurotoxicity. Radiotherapy given in daily fractions exceeding 2 Gy appears to be harmful for the normal surrounding brain with respect to memory, whereas smaller daily fractions are less harmful (43). In another study of 34 adults with primary cerebral neoplasias (cerebral lymphomas, glioblastoma multiforme, and lower-grade gliomas treated surgically and with RT), the consequences of RT proved to be slowness of executive functions and profound alterations of frontal functions, including attention focusing, mentation control, analogical judgment and insight, similar to those obtained by the patients suffering from subcortical vascular dementia (84). In one study, eight patients with high-grade tumors receiving limited-field radiotherapy following tumor resection did not show improvement in all the tested cognitive domains (intellectual functions, executive functions, memory, word processing, and perception), whereas six patients with low-grade and nonmalignant tumors did. This may reflect other factors such as higher radiation dose for high-grade tumor patients, more rapid tumor progression in high-grade tumors, and difference in surgical procedures (gross total vs. subtotal resection) (85). The effect of timing of cognitive decline following radiotherapy is also uncertain. In one study of 160 low-grade glioma patients, those undergoing radiotherapy in addition to resection showed greater cognitive impairment then those with resection alone (86). Other studies (87) reported initial deterioration of the cognitive functioning after radiation therapy that then improved. Armstrong et al. studied 20 patients with low-grade primary brain tumors treated with radiation therapy. Verbal and visual memory were tested after the surgery but prior to the beginning of radiation therapy and again at 3, 6, and 12 months post-radiotherapy. Patients showed normal verbal memory functioning at the baseline, deficit following radiation therapy, and rebound later. They showed impaired visual memory at the baseline, followed by recovery within a year long period following radiation treatment. In a 3-year long follow-up study of 20 adult patients with supratentorial low-grade glioma, assessed prior to WBRT treatment (50.4 Gy (10 patients) or 64.8 Gy (10 patients)) and at 18-months intervals, no significant cognitive changes were demonstrated. Baseline test scores were below average compared with age-specific norms. At the second evaluation, the groups’ mean test scores were higher than their initial performances on all psychometric measures, although the improvement was not statistically significant. No changes in cognitive performance were seen during the evaluation period when test scores were analyzed by age, treatment, tumor location, tumor type, or extent of resection (88). Pituitary tumors are often treated with radiotherapy, which may cause cognitive impairment when given in high doses. It is assumed that current regimens do not cause damage, but this has not been established. One study examined a retrospective comparison of outcomes in a total of 71 patients from two groups: patients with pituitary tumors who had undergone radiotherapy and surgery and patients with pituitary tumors who had surgery alone. A decrease in cognitive function was found regardless of treatment type. The decrease seemed to be greater in the radiotherapy group and was mainly in executive function which could affect daily life (89). This study is difficult to interpret, however, because patients treated for pituitary disease may have cognitive impairment prior to treatment. 2.3.3. Effects of Stereotactic Radiosurgery Alone or with Whole-Brain Radiation Whole-brain radiation therapy (WBRT) has historically been considered the mainstay of treatment for metastatic brain disease. During the past 15 years, significant advances have been made through randomized clinical trials showing that focal treatments with either surgery or stereotactic radiosurgery (SRS) result in improved survival in patients with a single brain metastasis (90). The role of WBRT has recently been called into question because of its perceived potential to cause neurocognitive dysfunction (91). Under investigation is whether to administer or omit adjuvant WBRT in conjunction with SRS in the initial management of patients with one to three newly diagnosed brain metastases. The advocates of adjuvant WBRT emphasize its merit in reducing the risk for subsequent brain metastases. Furthermore, it has been argued that subsequent brain metastases are best prevented because they are more likely to cause neurological sequelae than WBRT (92–94). Detractors argue that WBRT should be avoided and that, instead, a strategy of close follow-up and neuroimaging to detect recurrences treatable by focal means should be used. It is postulated that withholding WBRT from the initial management of brain metastases with SRS may delay the onset of neurocognitive dysfunction (92,95). The neurocognitive outcome of patients with 1–3 new brain metastases undergoing this strategy of SRS without initial WBRT is not yet well defined.

98

Part III / Neurologic Symptoms

The RTOG conducted the largest randomized trial comparing SRS plus WBRT versus WBRT alone (96). In this trial, 333 patients with 1–3 brain metastases were randomly assigned to WBRT (37.5 Gy in 15 fractions) or WBRT plus SRS. The overall trial demonstrated no significant improvement in survival (median survival of 5.7 vs 6.5 months; p = .13). A subgroup of patients with a single metastasis, however, experienced significant improvement in median survival (6.5 vs. 4.9 months; p = .05). The Japanese Radiosurgery Oncology Group (JROSG) 99-1 (97) performed a randomized comparison of SRS alone versus SRS plus WBRT for patients with up to four brain metastases. For the 132 randomly assigned patients, overall median survival was 7.9 vs. 7.6 months, respectively. The functional preservation rate (defined as KPS≥ 70) at 1 year was 27% vs. 32%. Preliminary data of one of the prospective studies (98) of 15 patients with 1–3 newly diagnosed brain metastases treated with SRS alone showed that at baseline 67% of patients had impairment on one or more tests of neurocognitive functions. The domains most frequently impaired at baseline were executive function, motor dexterity, and learning/memory with an incidence of 50%, 40%, and 27%, respectively. Brain metastasis volume measured at the time of initial SRS treatment was associated with worse performance on attention (p < .05). At 1 month, declines in the learning/memory and motor dexterity domains were most common. In a subgroup of five patients still alive 200 days after enrollment, four patients (80%) demonstrated stable or improved learning/memory, three (60%) demonstrated stable or improved executive function, and three (60%) demonstrated stable or improved motor dexterity relative to their baseline evaluation. Thus, although two-thirds of the brain metastasis patients had impaired neurocognitive function at baseline, the majority of five long-term survivors had stable or improved neurocognitive function across executive function, learning/memory, and motor dexterity. The effects of prophylactic cranial irradiation (PCI) have principally been studied using the models of small and non-small cell lung cancer, revealing comparable results. The effects of prophylactic and therapeutic cranial irradiation were studied in patients with small cell lung cancer with and without brain metastases. Results showed that although in all the patients cognitive performance was below average range prior to irradiation, whole brain irradiation did not induce a significant decline of cognitive functions in patients with prophylactic or therapeutic cranial irradiation. A decline with longer follow-up nevertheless seems possible (99). In one thorough prospective study of prophylactic cranial irradiation in patients with limited stage small cell lung cancer who responded well to their systemic therapy, 97% (29 out 30) of patients had evidence of cognitive dysfunction prior to PCI. The most frequent impairment was verbal memory, followed by frontal lobe dysfunction and fine motor incoordination. Of the patients with no prior neurologic or substance abuse history, 20 out of 21 (95%) had impairments on neuropsychological assessment. This neurologically normal group was just as impaired as the group with such a history with respect to delayed verbal memory and frontal lobe executive function. Eleven patients had neuropsychological testing 6-20 months after PCI; no significant differences were found from their pretreatment tests (100). The effect of prophylactic cranial irradiation was studied in 11 patients with small cell lung cancer who had chemotherapy and radiation treatment for their lung disease and were prescribed PCI with total dose up to 30 Gy in 10 fractions. Pre- and post-PCI measurements of the auditory event-related potentials (ERPs) (P50, N100, P300, and N400) during digit-span Wechsler test performance were obtained as well as measures of depression and anxiety. No significant difference was noticed pre- and post-radiotherapy for both the latencies and the amplitudes of ERP auditory components. Additionally, no changes were found with regard to behavioral performance (memory recall), depression symptomatology and state anxiety, according to pre- and post-radiation measurements. However, the self-reported depression symptomatology showed that the patients presented moderate depression. The absence of short-term psychophysiological neurotoxicity of this PCI schedule using these instruments lends additional support to the benefit of this PCI schedule for patients with SCLC (101). 2.3.4. Chemotherapy Adjuvant chemotherapy has been associated with a decline in neurocognitive and neurobehavioral functioning during the acute treatment phase, but it is not yet known whether or not, or to what degree, these symptoms persist. Potential neurological complications of chemotherapy include: acute encephalopathy (confusional state, insomnia, agitation), chronic encephalopathy (cognitive dysfunction consistent with “subcortical dementia,” incontinence, gait disturbance), stroke-like episodes, cerebellar symptoms, and peripheral neuropathies (79).

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

99

Effects of treatment of nonbrain cancers on cognitive functioning represents a relatively new area of study. In one study, 33% of women with breast cancer exhibited cognitive impairment. At the short-term postchemotherapy time point, 61% of the cohort exhibited a decline relative to baseline in one or more domains of cognitive functioning and reported greater difficulty in maintaining their ability to work. The most common domains of cognitive dysfunction were related to attention, learning, and processing speed. At the long-term post-chemotherapy time point, approximately 50% of patients who experienced declines in cognitive function demonstrated improvement, whereas 50% remained stable. Self-reported ability to perform work-related activities also improved over this interval. Neither impairment at baseline nor subsequent treatment-related cognitive decline exhibited any statistically significant correlation with affective well-being or with demographic or clinical characteristics (102). In a study of longitudinal cognitive functioning of 28 older women (ages 65–84) receiving adjuvant chemotherapy for breast cancer who were tested before and 6 months after chemotherapy: 14 (50%) had no change, 11 (39%) worsened, and three (11%) improved (p = .05). Seven patients (25%) experienced a decline in cognitive function, defined as a 1SD decline from pre- to post-testing in two or more neuropsychological domains. Exploratory analyses revealed no significant difference between functional status, comorbidity, and depression scale scores and change in overall QOL scores before and after chemotherapy (103). One study of 23 patients with primary CNS lymphoma treated with methotrexate-based chemotherapy assessed neurocognitive outcomes a median of 44 months after completion of treatment. Twenty-two patients (95%) showed either preserved or improved cognitive functions as compared with pre-treatment and immediate post-treatment baseline assessment. One patient showed an isolated decline in psychomotor speed. Eleven (48%) of 23 patients displayed at least mild cognitive deficits at long-term follow-up not related to therapy. Nineteen (83%) of 23 patients reported a good QOL. MRI revealed confluent white matter abnormalities in eight patients that were not associated with cognitive decline (104). Supportive nonchemotherapeutic medications such as steroids, immunosuppressive agents, anticonvulsants can alter cognitive functioning. Glucocorticoids may induce euphoria, mania, insomnia, restlessness, and increased motor activity; they may also contribute to memory dysfunction. Likewise, anticonvulsants sometimes produce adverse cognitive effects. Pain medications are associated with sedation and associated decrease in neurocognitive functioning (79). 2.3.5. Interaction Between Radiotherapy and Chemotherapy In the setting of primary brain tumors, radiation therapy is now rarely done without concomitant chemotherapy, and it is often difficult to separate effects of these two treatments that are likely synergistic. Treatment-related neurocognitive sequelae have recently become increasingly recognized in adult neuro-oncology (71,74,84). Wholebrain radiation in combination with chemotherapy was shown to prolong survival but is associated with a risk of long-term neurotoxicity. Treatment toxicity produces a diffuse pattern of deficits (including neurocognitive deficits) that is likely to emerge several months after the end of the treatment. Several studies suggest the combination of radiation and chemotherapy leads to worse cognitive outcomes then chemotherapy alone (105). This concern may be particularly pertinent to the management of low-grade gliomas and primary central nervous system lymphoma (PCNSL)—diseases with which many patients live long enough to experience delayed treatment neurotoxicity. Whereas with gliomas or brain metastases it is difficult to distinguish the effects of the residual tumor on cognitive function from the effects of therapy, much of the neurocognitive sequelae following PCNSL treatment is often attributed to therapy because the tumor often completely responds on imaging (106). One of the prior standard treatments for PCNSL involves high-dose methotrexate-based (MTX) chemotherapy and WBRT. This combined regimen prolongs patient survival, but also carries a substantial risk for delayed neurotoxicity. One study of 28 PCNSL patients treated either with WBRT ± MTX-based chemotherapy or chemotherapy alone in disease remission performed neurocognitive testing post-treatment and 8 months following completion of the treatment. Patients displayed mild to moderate impairments across several cognitive domains. These were of sufficient severity to reduce QOL in half of the patient sample. Comparisons according to treatment type revealed more pronounced cognitive impairment, particularly in the memory and attention/executive domains, among patients treated receiving WBRT. Extent of white matter disease correlated with attention/executive, memory, and language impairment (107). Harder et al. (108) evaluated the

100

Part III / Neurologic Symptoms

QOL and cognitive status in 19 PCNSL patients with median age 44 treated in a prospective European study. All patients were in complete remission following intravenous and intrathecal methotrexate followed by WBRT. The results were compared with a cohort of matched control patients with systemic hematologic malignancies treated with systemic chemotherapy or non-CNS radiotherapy. Cognitive impairment was found in 63% of PCNSL patients compared with 11% of control patients (p <.002). Only 42% of PCNSL patients returned to work compared with 81% of control subjects. The authors concluded that the effects of residual tumor were negligible because all of the patients had a complete tumor response. Several studies have shown that radiation and chemotherapy for brain tumors cause white matter changes (109,110). In one study it was shown that low-grade glioma patients who received no therapy had no or minimal white matter changes on MRI studies, whereas 62% of patients who received radiation and chemotherapy had white matter changes. The amount of white matter changes positively correlated with low executive function performance and the tendency toward lower verbal memory performance (105). Cognitive dysfunction in lowgrade glioma patients is more consistently seen several years post-treatment (105). It has been shown that radiation and chemotherapy together decrease performance on psychomotor tests, nonverbal memory as well as tendency to lower performance on attention, executive functions, and verbal memory tests. Patients who received radiation in addition to chemotherapy performed worse on the above-mentioned cognitive domains than patients who received only chemotherapy. Longer time from diagnosis is associated with lower scores on psychomotor tests and nonverbal memory (105). 2.3.6. Role of Disease Progression There is growing recognition that the primary cause of cognitive deficits in adult patients with primary brain tumors is the tumor itself and, more significantly, tumor progression. In high-grade glioma patients several studies found that the progression of brain tumors are the predominant causes of cognitive deterioration in this population (43,68,111) and Taylor et al. (112) found no clear trend to cognitive worsening from treatment. In one study (113) of 1,244 high-grade brain tumor patients in eight consecutive North Central Cancer Treatment Group treatment trials that used radiation and nitrosourea-based chemotherapy the effects of tumor progression (verified by imaging data) on the cognitive performance (Folstein Mini-Mental State Examination (MMSE)) at baseline, 6, 12, 18, and 24 months were analyzed. The proportion of patients without tumor progression who experienced clinically significant cognitive deterioration compared with baseline was stable at 6, 12, 18, and 24 months (18%, 16%, 14%, and 13%, respectively). In patients without radiographic evidence of progression, clinically significant deterioration in MMSE scores was a strong predictor of a more rapid time to tumor progression and death. At evaluations preceding interval radiographic evidence of progression, there was significant deterioration in MMSE scores for patients who were to experience radiographic progression, whereas the scores remained stable for the patients who did not have tumor progression. Thus, although other factors may contribute to cognitive decline, the predominant cause of cognitive decline seems to be subclinical tumor progression that precedes radiographic changes. 2.3.7. Therapies to Improve or Preserve Neurocognition At the present time, there are no proven treatments for cognitive impairment following brain cancer and subsequent cranial irradiation, nor are there any known effective preventive strategies. In this section we will discuss some efforts to evaluate different treatment modalities and additional interventions aimed at prevention of neurocognitive deterioration in brain tumor patients. Imaging findings in radiation-induced injury cerebral white-matter changes, decreased cerebral perfusion, decreased cerebral metabolism, and decreased cerebral N-acetyl aspartate (114, 115). Alzheimer’s dementia (AD) is characterized by regionally specific hypometabolism and grey matter atrophy with secondary white matter pathology. Among the most widely studied drugs to reduce the cognitive impairment and improve behavioral functioning in AD patients are those enhancing cholinergic neurotransmission. In a prospective, open-label phase II study (116) the effect of donepezil, a reversible acetylcholinesterase inhibitor used to treat mild to moderate Alzheimer’s-type dementia, on improvement of cognitive functioning, mood, and QOL in irradiated brain tumor patients was studied. Twenty-four patients with primary brain tumors (mostly low-grade gliomas) received donepezil 5 mg/d for 6 weeks, then 10 mg/d for 18 weeks, followed by a washout period of 6 weeks. Outcomes

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

101

were assessed at baseline, 12, 24 (end of treatment), and 30 weeks (end of wash-out). Patient performance improved significantly between baseline (pretreatment) and week 24 on attention/concentration, verbal memory, and figural memory and a trend for verbal fluency (all p < .05). Confused mood also improved from baseline to 24 weeks (p = .004), with a trend for fatigue and anger (all p < .05). Health-related QOL improved significantly from baseline to 24 weeks, particularly, for brain-specific concerns with a trend for improvement in emotional and social functioning (all p < .05). Toxicities were minimal. Psychostimulants such as methylphenidate have been used to treat many symptoms associated with advanced cancer, including cancer-related fatigue, opioid-induced sedation, depression, and cognitive dysfunction associated with malignancies. These uses for psychostimulants came after approval for treatment of disorders such as attention deficit disorder. Modafinil, a new psychostimulant, is following a similar path after its approval for use in attention deficit disorder in 1998. Modafinil has been used to treat fatigue associated with neurodegenerative disorders such as multiple sclerosis and amyotrophic lateral sclerosis, and is now increasingly used for cancer-related symptoms. Preliminary evidence suggests that modafinil is efficacious in improving opioid-induced sedation, cancer-related fatigue, and depression. There is no evidence to support its use in the treatment of cognitive dysfunction related to cancer or to support its having analgesic properties. Well-designed, randomized, controlled clinical trials are still needed to further elucidate the role of this drug in cancer patient care (117). Patients with malignant glioma develop progressive neurobehavioral deficits over the course of their illness caused both by the effects of the disease and the effects of radiation and chemotherapy. The role of methylphenidate in improving neurobehavioral functioning despite expected neurologic deterioration has been studied (118). Thirty patients with primary brain tumors underwent neuropsychologic assessment before and during treatment with methylphenidate. Ability to function in activities of daily living and MRI findings were also documented. Patients were assessed on 10, 20, and 30 mg of methylphenidate twice daily. Significant improvements in cognitive function were observed on the 10-mg twice-daily dose. Functional improvements included improved gait, increased stamina and motivation to perform activities, and in one case, increased bladder control. Adverse effects were minimal and immediately resolved when treatment was discontinued. There was no increase in seizure frequency, and the majority of patients on glucocorticoid therapy were able to decrease their dose. Gains in cognitive function and ability to perform activities were observed in the setting of progressive neurologic injury documented by MRI in half of the subjects. Many brain tumor patients may develop seizure disorder in the course of their disease; antiepileptic drugs are commonly used in this group of patients. Studies have shown that patients on antiepileptic monotherapy perform significantly better on cognitive tests than patients on polytherapy (105). In many cases antiepileptic drugs are given prophylactically, and prospective and retrospective studies of patients with primary and metastatic disease treated prophylactically with antiepileptic drugs did not show any benefit (119,120). In a recent meta-analysis of 12 studies in patients with brain tumors, 10 of which included patients with brain metastases, none supported a role for prophylactic anticonvulsants. Prophylactic anticonvulsants did not protect against subsequent seizures; furthermore, antiepileptic drugs frequently are associated with side effects or drug interactions (38). In another study, patients on antiepileptics had poorer objectively measured and self reported cognitive function than did those without such treatment (43). Thus, the prophylactic use of these agents is generally best avoided.

3. MOOD AND PSYCHIATRIC DISORDERS IN CANCER PATIENTS Brain tumors may produce a variety of psychiatric symptoms. Rarely they present without any localizing signs but with psychiatric symptoms, and often psychiatric or mood changes are part of the initial presentation of brain tumor. Therefore, it is important for clinicians to have an index of suspicion of brain tumor in patients with newonset psychiatric symptoms, atypical presentations and treatment resistance and therefore consider neuroimaging for early detection of brain tumors (121). Likewise during treatment for cancer, symptoms may develop that may be attributed to direct treatment side effects rather then a underlying mood disorder, which would clearly dictate a different management approach. For example, the necessity for individual psychological support during radiation therapy in head and neck cancer was demonstrated in a follow-up study where a significant deterioration in composite QOL scores and a significant increase in depression without any change in anxiety levels was shown (122).

102

Part III / Neurologic Symptoms

In a retrospective (5 years) study by Gupta et al. (123), fifteen (21%) of 72 meningioma patients presented with initial psychiatric symptoms in the absence of neurological symptoms. Affective disorders were a common presentation. There was no correlation between brain laterality and the psychiatric comorbidity. In one case report (124) a person initially treated for post-traumatic stress disorder and borderline personality traits, who developed depressive symptoms and memory difficulties, was later found to have a left thalamic tumor eventually confirmed as glioblastoma multiforme. Psychiatric and mood disorders in brain tumor patients did not show any specific correlations with localization of the tumors; however, it is possible to describe some typical patterns. Cummings and Benson reported that frontal lobe tumors produce mental status and personality changes in 90% of cases and frank dementia in 70% of cases, but few neurological signs (125). Characteristic manifestations of frontal lobe involvement include apathy, disinhibition or impulsivity. Additional symptoms may include euphoria or depression, dementia, lack of concern, poor judgment, disorientation and poor attention (126). Belyi suggests that patients with right-frontal tumors tend to display more euphoria, whereas left-frontal tumor patients may display more depression and abulia (127). Due to significant association of temporal lobe with the frontal lobe, symptoms of temporal lobe tumors tend to be similar to those of frontal lobe tumors, such as depressed mood with apathy and irritability or euphoria and mania. Personality change and anxiety have also been noted (128). Patients with parietal lobe tumors may present with a lack of awareness of their symptoms/neglect syndromes including disregard of grooming and personal care of the contralateral side of the body (126). Psychiatric symptoms are not generally observed with tumors of this region; however, some cases have been reported (128). Tumors of the corpus callosum are frequently associated with behavioral symptoms, including personality changes, psychosis, and most commonly affective symptoms (128). Behavioral symptoms have been noted in 90% of patients with tumors of the corpus callosum (129). A small case series comparison of five patients with corpus callosal tumors and eight patients with other locations of tumors noted significantly more patients suffering from depression in the corpus callosum group (129). Tumors of the basal ganglia may result in personality changes, and depression (126). The hypothalamus and thalamus are sites of tumor formation prone to psychiatric and behavioral manifestations. Thalamic tumors may result in memory loss, confusion, and emotional lability (126). Hypothalamic tumors have been associated with eating disorders and hypersomnia (128). Tumors of the pituitary gland, which is closely associated with the hypothalamus, can cause neuropsychiatric symptoms such as affective lability, depression, and psychotic symptoms because of endocrine changes (128). Even cerebellar and brain stem tumors may be associated with affective disorders, paranoid delusions and personality change (128). Cerebellar lesions may present with the cerebellar cognitive affective syndrome, characterized by impairments in executive, visuospatial and linguistic abilities, and affective disturbance ranging from emotional blunting and depression to disinhibition and psychotic features (130). One of the difficulties of studying mood disorders in cancer patients is the finding that somatic symptoms are often a function of the physical disease process and/or side effects of therapy. Occasionally, organic factors (e.g., pain or treatment with interferon-) may account for emotional symptoms of mood disorders. Mood disorders due to medical conditions (including cancer) are recognized entities in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (131). Somatic symptoms relevant to the diagnosis of depression are so prevalent in oncology that many investigators consider them of little use in trying to diagnose major depression in cancer patients. Instead, emphasis is placed on psychological symptoms (132–134). However, a significant correlation has been noted between depression and survival. In a study (135) of 18 low-grade glioma patients followed up at 1 and 5 years post-resection with the Beck Depression Inventory (BDI) measure of depression, depressed low-grade glioma patients had a significantly shorter survival time (3.3–5.8 years) compared to nondepressed low-grade glioma patients (10.0-11.7 years) (42). Depressed patients with primary brain tumors have lower reported quality of life as compared to nondepressed ones. Clinically significant depression is a common complication in patients with cancer. The word depression may refer to an emotion common in everyday experience (sadness), or a frequently encountered reaction to severe stress (i.e., the diagnosis of cancer and its sequelae). The term also applies to a mood disorder (major depression) with defined emotional (e.g., dysphoria, anhedonia) and physical symptoms (e.g., fatigue, decreased concentration, anorexia, and changes in sleep patterns) (136). The prevalence of depression in cancer patients varies according to the setting. A well-known Psychosocial Oncology Group study found that major depression was present in

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

103

6% of patients and accounted for 13% of all psychiatric diagnoses (137). Rates of depression are higher in more severely ill cancer patients (138). The overall rate of major depression in cancer patients may be two to three times that of the general population (139). Pringle and colleagues (140) studied the effects of hemispheric location of the tumor, tumor type, and gender of patients with solitary intracranial neoplasm on the development of anxiety and depression. They found a 30% and 16% respective incidence of anxiety and depression in 109 patients with a solitary intracranial neoplasm who completed the Hospital Anxiety and Depression Scale (HAD) before and after biopsy or resection. Females with a left hemisphere tumor reported higher levels of emotional disturbance than any other group of patients; the relationships between dysphasia and levels of anxiety or depression were not significant. Patients with a meningioma had higher levels of anxiety and depression as measured by the HAD than those with any other tumor types. Levels of both anxiety and depression were significantly lower after tumor surgery. There were no significant differences in HAD scores between left and right hemispheric tumor groups, and the tumor and control group of patients who had lumbar spine surgery. Patients with pituitary disease in many ways resemble patients with other malignant brain tumors. Weitzner et al. (141) present a series of cases in which patients with pituitary disease were diagnosed and treated for depression and showed little response to the treatment for depression. When the diagnosis of apathy syndrome was considered and treatment implemented, the patients’ condition improved. Brain tumor patients experience a whole spectrum of emotions immediately after receiving the diagnosis. Edvardsson et al. have attempted to describe adult patients’ experiences of falling ill and being diagnosed with low-grade glioma (142). Interviews of 27 adult patients were analyzed using inductive content analysis. Most of the patients in the study experienced the illness onset as stressful. The salient negative life-situation consequences included a lack of social support and attitudes expressing a lack of understanding. However, to some extent, positive experiences also emerged in the interviews concerning healthcare and life situation despite the onset of the illness. Affective disorders among patients with brain tumors must be considered immediately after surgery, especially in persons with a depression history and in those with a coincident physical disability. In a study of functional status and depression in 77 patients with a solitary primary brain tumor treated surgically (143), the authors found that prior to surgery 35% of patients indicated symptoms of depression as measured by the Beck Depression Inventory (BDI); these scores were significantly higher in patients with a history of depression and in those with a lower functional outcome as measured by Karnofsky Performance Status (KPS). In the entire study sample the severity of depression decreased significantly at 3 months post-surgery. A lower functional status (KPS ≤ 70) in patients was significantly associated with high depression scores at the 3-month and 1-year assessments. The decrease in the level of depression was significant in patients with an anterior tumor and those with a pituitary adenoma. Despite the very common finding of depression in brain tumor patients, there are discrepancies between patientreported and physician-reported depression. In the early post-operative period, physicians reported depression in 15% of patients, whereas 93% of patients reported symptoms consistent with depression. The incidence of patient self-reported depression remained similar at 3- and 6-month follow-up, whereas physician-reported depression increased from 15% in the early post-operative period to 22% at both 3- and 6-month follow-up. Concordance between physician recognition of depression and treatment of depression was low initially (33%) and increased at 3 and 6 months (51 and 60%, respectively). Antidepressant therapy was provided for only 15% of patients who reported symptoms of depression. As compared to patients who were not depressed, survival was shorter and complications more common among depressed patients. Symptoms of depression were common immediately after glioma surgery, and they increased throughout the 6-month period after surgery (144). One more confounding variable that contributes to the accuracy of patient-reported symptoms is alexithymia (problems identifying, describing, and working with one’s own feelings, difficulty distinguishing between feelings and the bodily sensations of emotional arousal, confusion of physical sensations often associated with emotions), which is not uncommon in cancer patients. Yet this phenomenon is of utmost importance for correct diagnosis and management of symptoms during the initial diagnosis and in the later stages of the disease. In one of the few studies that explored this problem in cancer patients, Mantani et al. (145) in a cross-sectional study of 46 post-surgical ambulatory women with breast cancer and their husbands found that alexithymia and

104

Part III / Neurologic Symptoms

family functioning are associated with anxiety and depression, respectively, in both women with breast cancer and in their husbands. A high degree of alexithymia in patients correlated with a high degree of patient anxiety. Patient perceptions of inappropriate affective responsiveness among family members correlated with a high degree of depression. Among husbands, a high degree of anxiety was correlated with their own high level of alexithymia or low level of education, and with adjuvant therapy in their spouse. Husband perceptions of inappropriate sharing of roles among family members, their own low education level, and a large number of family members correlated with high degrees of depression. Interferons (IFN; alpha, beta, and gamma) are a family of proteins that produce a regression or control over disease processes in more than a dozen cancers. However, the neurotoxicity of IFNs including fatigue, apathy, confusion, and impaired concentration, may overshadow their benefits. In particular, until recently physicians have underestimated the occurrence of depression as a side effect of IFN therapy. Case reports suggest that various medications including selective serotonin reuptake inhibitors (SSRIs), tricyclics, and opioid antagonists effectively ameliorate the CNS side effects of IFN-alpha treatment (146). Trials with IFN-alpha have provided contradictory findings regarding the presence of affective side effects. The development of depression in some patients also raises questions about whether cognitive dysfunction might be secondary to an organic, IFN-induced mood disorder. For example, in a study of patients with chronic myelogenous leukemia, increased depressive symptoms and declines in information processing and executive functions were observed, but depression alone could not account for cognitive dysfunction. There was some evidence suggesting that exposure to chemotherapy and higher cumulative IFN-alpha dose may contribute to cognitive impairment (147). Comparison of the effectiveness of three different types of supportive therapies for women with breast cancer showed that telephone interpersonal counseling combined (TIP-C) with cancer education and self-managed exercise (regular, low-impact exercise) significantly decreased anxiety and depression both in women and their partners as compared to attention control (printed information about breast cancer and very brief weekly phone calls without counseling or encouragement to exercise) which did not have such a significant effect. The TIP-C intervention is used to target the social support behaviors of both cancer survivors and their partners using interpersonal communication techniques. Improvement is achieved largely through interpersonal communication that allows the person to work through the affective reaction to the stressor and to marshal instrumental support for tangible assistance with roles and functions, informational support for advice or suggestions, and appraisal support for gauging and adjusting to the stressor. Among cancer patients, such behaviors can make the cancer experience seem less overwhelming and can aid in adjusting to the illness (148). Similarly, coping skills training (CST) in African-American patients with prostate cancer have shown that an intervention developed to enhance coping with treatment side effects resulted in a moderate to large treatment effects for QOL related to bowel, urinary, sexual, and hormonal symptoms. Partners who underwent CST reported less caregiver strain, depression, and fatigue, and more vigor (149).

4. FATIGUE IN CANCER PATIENTS Fatigue is a very common side effect of cancer and its treatment, with prevalence exceeding 60% in many studies (150–152). It occurs in the majority of patients who are in active therapy and persists in a substantial proportion of patients after treatment is discontinued. It is among the most distressing symptoms associated with cancer and cancer treatments because it substantially disturbs patients’ QOL and ability to function optimally on a daily basis. Despite the identification of risk factors for the development of fatigue factors, its etiology remains poorly understood. Important elements to include in any definition of cancer-related fatigue include its pervasiveness, persistence, detrimental effect on quality of life, and its inability to be relieved by rest or sleep. Fatigue may be physical (e.g., associated with muscle weakness or lack of stamina), mental (e.g., involving reduced alertness and lack of motivation), or both. Several validated questionnaires are used to measure fatigue in patients with cancer, and research efforts are currently focused on ways to distinguish it from depression with which it shares many symptoms. All patients with cancer should be evaluated for fatigue, and treatment options should be considered for those who are experiencing excessive levels of fatigue. Assessment of cancer-related fatigue and contributing causes is difficult for several reasons. The first is inability to define the problem in question. There may be disagreement within medical specialties, and between

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

105

health care providers and patients, over what constitutes fatigue. Increasingly, there is agreement that fatigue is a multidimensional construct and that assessment instruments, if they are to have any utility in treatment, will have to be multidimensional as well. Another factor complicating assessment is that perceptions of subjective phenomena change with time (153). This has implications for possible inconsistencies between complaints of fatigue and actual performance, and how clinicians interpret those inconsistencies. The perception of cognitive impairment is also subjective (154). Knowledge of the specific impact of mental fatigue on the individual patient allows the health care provider to institute proper interventions. Interventions may be pharmacologic (e.g., stimulant therapy), behavioral (e.g., relaxation training to help develop sustained attention), or lifestyle-related (e.g., establishing a more distractionfree environment for working, or moving to a job that allows flexible working hours and emphasizes completing work properly and on time rather than the number of consecutive hours spent at the job). As with most secondary behavioral syndromes, effective intervention against fatigue associated with cognitive and mood impairment begins with identification and correction, if possible, of the underlying causes. Examples of this include treatment of anemia associated with malignancy, treatment of thyroid dysfunction, and discontinuation or dose reduction of interferon therapy. Patients experiencing fatigue due to these problems will often experience an improvement in mood and cognitive function simply because they feel better. In cases of thyroid dysfunction and interferon toxicity, the improvement may also be a function of relief from a neuroendocrine or neurochemical insult that is a primary cause of the symptoms experienced (154). Treatment for fatigue should be individualized according to the underlying pathology when a specific cause has been identified (e.g., anemia, sleep disorder, depression, or metabolic disorder). Nonspecific therapies may be useful in short- and long-term cancer-related fatigue management in many patients. In addition to older therapies, such as hematopoietic support, antidepressants, corticosteroids, and psychostimulants, the effectiveness of the new wake-promoting agent modafinil is currently being studied. A more thorough evaluation of the various therapeutic options is required to better define their efficacy and safety profiles in this patient population. Fatigue is known to prevent patients from maintaining a normal daily routine and quality of life, is correlated with poor survival (155), and is negatively associated with patients’ perception of hope (156). Previous studies show that distress caused by disease is a factor associated with cancer-related fatigue (157–159). Fatigue in cancer patients has also been associated with other symptoms that may cause distress, including pain (160), sleep disturbance (161), inability to concentrate, muscle weakness (162), dyspnea (162,163), mouth sores (162), lack of appetite, nausea and vomiting (164), and uncontrolled diarrhea (165). Another factor that has been related to patients’ fatigue is performance status (162,163). Hemoglobin levels also correlate with cancer patients’ fatigue (163,164). In addition to the distress caused by these physical factors, cancer patients experience psychological distress or mood disturbances, which have been related to fatigue (158,159,162,166,167). Fatigue has adverse effects on cognitive function and mood. However, impaired cognitive function and mood disorders may also cause fatigue (134). Interrelation of fatigue and other common symptoms accompanying brain tumors such as depression and sleep disturbance with patients’ reported QOL was studied in a cohort of 73 high-grade glioma patients, revealing that co-occurring symptoms of depression, fatigue, sleep disturbance, and cognitive impairment were significantly correlated with each other and explained 29% of the variance in QOL. Depression, fatigue, sleep disturbance, cognitive impairment, and pain were significantly correlated with each other and explained 62% of the variance in functional status (41). The fact that a relationship exists between depression and fatigue in cancer is generally acknowledged (153,160), but the causal relationship between them is under-appreciated. Visser and Smets failed to find a significant cause-effect relationship between depression and fatigue in patients treated with radiation therapy, although both symptoms predicted decreased QOL (168). Broeckel et al. found that breast cancer patients were more likely to perceive fatigue as a significant problem post-chemotherapy if they experienced a concurrent mood or anxiety disorder; however, pretreatment psychiatric disorder did not predict more severe fatigue post-chemotherapy (167). Patients who suffer from mental fatigue often report that they are easily overwhelmed, have difficulty being organized and efficient in their daily activities, and have difficulty meeting deadlines. In addition, patients report that activities that once were automatic now require more effort (169). Patients also report being easily tired from performing normal routine tasks, having multitasking difficulties and becoming easily overwhelmed when doing

106

Part III / Neurologic Symptoms

several tasks at a time, difficulties sustaining attention for a prolonged period of time, and generalized slowing of cognitive processes (154). Trajectory of changes in fatigue following radiation therapy in 60 adult patients with glioblastoma (GBM) showed an increase in fatigue after radiation therapy that was associated with decreases in almost all aspects of patients’ QOL (170). Low psychological distress and a low level of fatigue may cause a greater cancer resistance or may reflect underlying mental and physical robustness. In patients with breast cancer, low levels of psychological distress and low fatigue independently predicted longer recurrence-free and overall survival, controlling for biological factors. Lack of anxiety also predicted longer recurrence-free survival. When added to the biological model, fatigue remained a significant predictor of recurrence-free survival and emotional function remained a significant predictor of overall survival (171). Antidepressant therapy is appropriate for patients who meet criteria for primary or secondary depression. If fatigue is strictly a function of mood disorder (often not the case in cancer patients), it should improve as with resolution of other depressive symptoms. At present, there is no gold standard antidepressant for use in oncology. The choice of antidepressants is often based on the side effect profile. Antidepressants with more activating characteristics (i.e., fluoxetine, bupropion) may be good choices for depressed patients with significant fatigue. If insomnia is a prominent symptom of the mood disorder and is contributing to daytime fatigue, then a more sedating antidepressant (e.g., some tricyclic antidepressants, mirtazapine) may be a better choice for nighttime use (154). With regard to pharmacologic intervention, the other major class of medications to consider in the setting of fatigue with depression or cognitive impairment is psychostimulants. These medications (e.g., methylphenidate, d-amphetamine, pemoline) have an established role in the treatment of depression in the medically ill (172). While not studied in large, controlled trials, they have been used to counter sedation associated with opioid administration in cancer patients with fatigue (173) and cognitive impairment and psychomotor slowing in primary brain tumor patients (118).

5. CONCLUSION The review detailed in this chapter shows the importance of and ease with which neurocognitive function measurement can be a part of the care of cancer patients. Neurocognitive impairment, depression, and other psychiatric variables, as well as fatigue significantly interact with quality of life. Mediators of these systems include a number of individual variables, including disease, treatment, and individual difference variables. Research in these areas of cancer survivorship is important to improve global quality of life.

REFERENCES 1. Committee on Cancer Survivorship: Improving Care and Quality of Life. Hewitt M, Greenfield S, Stovall E (eds.). From Cancer Patient to Cancer Survivor: Lost in Transition. Washington, D.C.: National Academies Press; 2005. 2. Farace E, Shaffrey M.E. Neuropsychological issues. In: Schiff D. O’Neill BP, ed. Principles of Neuro-oncology. New York: McGrawHill; 2005:201–217. 3. Meyers CA, Kayl A.E. Neurocognitive function In: Levin VA, ed. Cancer in the Nervous System. 2nd ed. New York: Oxford University Press; 2002. 4. Kayl AE, Meyer CA. Neuropsychological impact of brain metastases and its treatment. In: Sawaya R, ed. Intracranial Metastases. Malden, MA: Blackwell Futura; 2004. 5. Kayl AE, Meyer CA. Neuropsychological complications in patients with brain tumors. In: Booth S. Bruera E, Oliver D, eds. Palliative Care Consultations in Primary and Metasatic Brain Tumors. Oxford: Oxford University Press; 2004. 6. Martin GN. Human Neuropsychology. Englewood Cliffs, NJ: Pearson/Prentice Hall, 2006. 7. Lezak MD. Neuropsychological Assessment. 3rd ed. New York: Oxford University Press; 1995:650–685. 8. Halligan PW, Kischka U, Marshall JC, ed. Handbook of Clinical Neuropsychology. New York: Oxford University Press; 2003. 9. Heilman KM, Valentine E, ed. Clinical Neuropsychology. 4th ed. New York: Oxford University Press; 2003. 10. Weitzner MA, Meyers CA. Quality of life and neurobehavioral functioning in patients with malignant gliomas. In: Yung WKA, ed. Ballieres Clinical Neurology: Cereberal Gliomas. London: Bailliere Tindall; 1996:425–439. 11. Meyers CA. Quality of life of brain tumor patients. In: Benstein M. BM, ed. Essential Neuro-oncology. New York: Thieme; 2000: 466–472. 12. Coltheart M. The Cognitive Neuropsychology of Language. London: Erlbaum; 1987. 13. Hillis AE. The Handbook of Adult Language Disorders: Integrating Cognitive Neuropsychology, Neurology, and Rehabilitation. Philadelphia: Psychology Press; 2002.

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

107

14. Tucha O, Smely C, Preier M et al. Cognitive deficits before treatment among patients with brain tumors. Neurosurgery 2000;47: 324–334. 15. Packer RJ, Miller DC, Shaffrey M et al. Intracranial neoplasms. In: Rosenberg RN, Pleasure DE, ed. Comprehensive Neurology. New York: John Wiley; 1998:187–243. 16. Lilja A, Smith GJ, Salford LG. Microprocesses in perception and personality. J Nerv Ment Dis 1992;180(2):82–88. 17. Scheibel RS, Meyers CA, Levin VA. Cognitive dysfunction following surgery for intracerebral glioma: influence of histopathology, lesion location, and treatment. J Neurooncol 1996;30(1):61–69. 18. Gottwald B, Wilde B, Mihajlovic Z et al. Evidence for distinct cognitive deficits after focal cerebellar lesions. J Neurol Neurosurg Psychiatry 2004;75(11):1524–1531. 19. Renning C,. Sundet K, Due-Tonnessen B, et al. Persistent cognitive dysfunction secondary to cerebellar injury in patients treated for posterior fossa tumors in childhood. Pediatr Neurosurgery 2005;41(1):15–21. 20. Schmahman JD, Sherman JC. The cerebellar cognitive affective syndrome. Brain 1998;121:561–579. 21. Riva D, Giorgi C. The cerebellum contributes to higher functions during development: evidence from a series of children surgically treated for posterior fossa tumors. Brain 2000;123:1051–1061. 22. Levinsohn L, Cronin-GolombA, Schmahmann JD. Neuropsychological consequences of cerebellar tumour resection in children: cerebellar cognitive affective syndrome in a paediatric population. Brain 2000;123:1041–1050. 23. Molinari M, Leggio MG, Silveri MC Verbal fluency and agrammatism. In: Schmahmann JD ed. The Cerebellum and Cognition: International Review of Neurobiology. San Diego: Academic Press; 1997:325–339. 24. Leggio MG, Silveri MC, Petrosini L, et al. Phonological grouping is specifically affected in cerebellar patients: a verbal fluency study. J Neurol Neurosurg Psychiatry 2000;69:102–106. 25. Fiez JA, Petersen SE, Cheney MK, et al. Impaired nonmotor learning and error detection associated with cerebellar damage. Brain 1992;115:155–178. 26. Grafman J, Litvan I, Massaquoi S et al. Cognitive planning deficit in patients with cerebellar atrophy. Neurology 1992;42:1493–1496. 27. Appollonio IM Grafman J, Schwartz V et al. Memory in patients with cerebellar degeneration. Neurology 1993;43:1536–1544. 28. Drepper J, Timmann D, Kolb FP et al. Nonmotor associative learning in patients with isolated degenerative cerebellar disease. Brain 1999;122:87–97. 29. Townsend J, Courchesne E, Covington J et al. Spatial attention deficits in patients with acquired or developmental cerebellar abnormality. J Neurosci 1999;19:5632–5643. 30. Courchesne E, Townsend J, Akshoomoff NA et al. Impairment in shifting attention in autistic and cerebellar patients. Behavioral Neurosci 1994;108:848–865. 31. Gomez Beldarrain M, Garcia-Monco JC, Quintana JM et al. Diaschisis and neuropsychological performance after cerebellar stroke. Eur Neurol 1997;37:82–89. 32. Helmuth LL, Ivry RB, Shimizu N. Preserved performance by cerebellar patients on tests of word generation, discrimination learning, and attention. Learning and Memory 1997;3:456–474. 33. Daum I. AH. Neuropsychological abnormalities in cerebellar syndromes: fact or fiction? In: Schmahmann J.D, ed. The Cerebellum and Cognition. International Review of Neurobiology. San Diego: Academic Press; 1997:61–83. 34. Friedman MA, Meyers CA, Sawaya R. Neuropsychological effects of third ventricle tumor surgery. Neurosurgery 2003;52(4):791–798; discussion 8. 35. Kayl AE MC. Does brain tumor histology influence cognitive function? Neuro-oncol 2003(5):255–260. 36. Klein M, Houx PJ, Jolles J. Long-term persisting cognitive sequeale traumatic brain injury and the effect of age. J Neuropathol Mental Disorders 1996;184:459–467. 37. Nussbaun ES, Djalilian HR, Cho KH et al. Brain metastases: histology, multiplicity, surgery and survival. Cancer 1996;78(8): 1781–1788. 38. Lassman AB, DeAngelis LM. Brain metastases. Neurol Clin 2003;21(1):1–23, vii. 39. Farace E, Shaffrey M.E. Neuropsychological issues. In: Schiff D. O’Neill BP, ed. Principles of Neuro-oncology. New York: McGraw– Hill; 2005. 40. Meyers CA, Hess KR. Multifaceted end points in brain tumor clinical trials: cognitive deterioration precedes MRI progression. Neuro-oncol 2003;5(2):89–95. 41. Fox SW, Lyon D, Farace E. Symptom clusters in patients with high-grade glioma. J Nurs Scholarsh 2007;39(1):61–67. 42. Mainio A, Tuunanen S, Hakko H et al. Decreased quality of life and depression as predictors for shorter survival among patients with low-grade gliomas: a follow-up from 1990 to 2003. Eur Arch Psychiatry Clin Neurosci 2006;256(8):516–521. 43. Klein M, Heimans JJ, Aaronson NK et al. Effect of radiotherapy and other treatment-related factors on mid-term to long-term cognitive sequelae in low-grade gliomas: a comparative study. Lancet 2002;360(9343):1361–1368. 44. Meyers CA, Hess KR, Yung WKA et al. Cognitive function as a predictor of survival in patients with recurrent malignant glioma. J Clini Oncol 2000;18(3):646. 45. Herman MA, Tremont-Lukats I, Meyers CA et al. Neurocognitive and functional assessment of patients with brain metastases: a pilot study. Am J Clin Oncol 2003;26(3):273–279. 46. Laws ER. The decade of the brain: 1990–2000. Neurosurgery 2000;47(6):1257–1260. 47. Laws E, Shaffrey M. Surgical management of intracranial gliomas: biopsy resection or watchful waiting. Clin Neurosurg 2001;48: 37–45. 48. Peyser JM, Rao SM, LaRocca NG et al. Guidelines for neuropsychological research in multiple sclerosis. Arch Neurol 1990;47(1): 94–97. 49. Jelic V, Johansson SE, Almkvist O et al. Quantitative electroencephalography in mild cognitive impairment: longitudinal changes and possible prediction of Alzheimer’s disease. Neurobiol Aging 2000;21(4):533–540.

108

Part III / Neurologic Symptoms

50. Arfken CL, Lichtenberg PA, Tancer ME. Cognitive impairment and depression predict mortality in medically ill older adults. J Gerontol Ser A Biol Sci Med Sci 1999;54(3):M152–M156. 51. Pahlson A, Ek L, Ahlstrom G, Smits A. Pitfalls in the assessment of disability in individuals with low-grade gliomas. J Neuro-Oncol 2003;65(2):149–158. 52. Farace E., Sheehan J. M. Majority of brain tumor patients show neurocognitive improvement after initial surgical resection In: Society of Neurooncology 12th Meeting; 2007; Dallas, TX; 2007. 53. Teixidor P, Gatignol P, Leroy M et al. Assessment of verbal working memory before and after surgery for low-grade glioma. J Neurooncol 2007;81(3):305–313. 54. Duffau H, Capelle L, Denvil D, et al. Functional recovery after surgical resection of low-grade gliomas in eloquent brain: hypothesis of brain compensation. J Neurol Neurosurg Psychiatry 2003;74(7):901–907. 55. Patchell RA, Tibbs PA, Walsh JW et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322(8):494–500. 56. Vecht CJ, Haaxma-Reiche H, Noordijk EM et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993;33(6):583–590. 57. Mintz AH, Kestle J, Rathbone MP et al. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996;78(7):1470–14776. 58. Bindal RK, Sawaya R, Leavens ME et al. Surgical treatment of multiple brain metastases. J Neurosurg 1993;79(2):210–216. 59. Walker MD, Alexander E, Hunt WE et al. Evaluation of BCNU and/or radiotherapy in the treatment of anaplastic gliomas: a cooperative clinical trial. J Neurosurg 1978;49:333–343. 60. Kristiansen K, Hagen S, Kollevold T et al. Combined modality therapy of operated astrocytomas grade III and IV: confirmation of the value of post-operative irradiation and lack of potentiation of bleomycin on survival time: a prospective multicenter trial of the Scandinavian Glioblastoma Study Group. Cancer 1981;47(4):649–652. 61. Stupp R, Mason W, van den Bent MJ et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987–996. 62. Sneed PK, Suh JH, Goetsch SJ et al. A multi-institutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases. Int J Rad Oncol Biol Phys 2002;53:519–526. 63. Crossen JR Garwood D, Glatstein E et al. Neurobehavioral sequale of cranial irradiation in adults: a review of radiation-induced encephalopathy. J Clin Oncol 1994;12:627–642. 64. Surma-aho O, Niemela M, Vilkki J et al. Adverse long-term effects of brain radiotherapy in adult low-grade glioma patients. Neurology 2001;56(10):1285–1290. 65. DeAngelis LM, Delattre JY, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology 1989;56: 1285–1290. 66. Lee AW, Kwong DL, Leung SW et al. Factors affecting risk of symptomatic temporal lobe necrosis: significant of fractional dose and treatment time. Int J Rad Oncol Biol Phys 2002(53):75–85. 67. Li J, Bentzen SM, Renschler M, Mehta MP. Regression after whole-brain radiation therapy for brain metastases correlates with survival and improved neurocognitive function. J Clin Oncol 2007;25(10):1260–1266. 68. Meyers CA, Smith JA, Bezjak A et al. Neurocognitive function and progression in patients with brain metastases treated with whole-brain radiation and motexafin gadolinium: results of a randomized phase III trial. J Clin Oncol 2004;22(1):157–165. 69. Nordal RA, Nagy A, Pintilie M et al. Hypoxia and hypoxia-inducible factor-1 target genes in central nervous system radiation injury: a role for vascular endothelial growth factor. Clin Cancer Res 2004;10(10):3342–3353. 70. Abayomi OK. Pathogenesis of cognitive decline following therapeutic irradiation for head and neck tumors. Acta Oncol 2002;41(4): 346–351. 71. Taphoorn MJ, Klein M. Cognitive deficits in adult patients with brain tumours. Lancet Neurol 2004;3(3):159–168. 72. Nordal RA, Wong CS. Intercellular adhesion molecule-1 and blood–spinal cord barrier disruption in central nervous system radiation injury. J Neuropathol Exp Neurol 2004;63(5):474–483. 73. Fujii O, Tsujino K, Soejima T et al. White matter changes on magnetic resonance imaging following whole-brain radiotherapy for brain metastases. Radiat Med 2006;24(5):345–350. 74. Sarkissian V. The sequelae of cranial irradiation on human cognition. Neurosci Lett 2005;382(1–2):118–123. 75. Armstrong CL, Gyato K, Awadalla AW et al. A critical review of the clinical effects of therapeutic irradiation damage to the brain: the roots of controversy. Neuropsychol Rev 2004;14(1):65–86. 76. Eriksson PS, Perfilieva E, Bjork–Eriksson T et al. Neurogenesis in the adult human hippocampus. Nat Med 1998;4(11):1313–1317. 77. Monje ML, Palmer T. Radiation injury and neurogenesis. Curr Opin Neurol 2003;16(2):129–134. 78. Tsuruda JS, Kortman KE, Bradley WG et al. Radiation effects on cerebral white matter: MR evaluation. AJR Am J Roentgenol 1987;149(1):165–171. 79. Wefel JS, Kayl AE, Meyers CA. Neuropsychological dysfunction associated with cancer and cancer therapies: a conceptual review of an emerging target. Br J Cancer 2004;90(9):1691–1696. 80. Hochberg FH, Slotnick B. Neuropsychologic impairment in astrocytoma survivors. Neurology 1980;30(2):172–177. 81. Taphoorn MJ, Heimans JJ, Snoek FJ et al. Assessment of quality of life in patients treated for low-grade glioma: a preliminary report. J Neurol Neurosurg Psychiatry 1992;55(5):372–376. 82. Grant R, Slattery J, Gregor A, Whittle IR. Recording neurological impairment in clinical trials of glioma. J Neurooncol 1994;19(1): 37–49. 83. DeAngelis LM, Mandell LR, Thaler HT et al. The role of post-operative radiotherapy after resection of single brain metastases. Neurosurgery 1989;24(6):798–805.

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

109

84. Moretti R, Torre P, Antonello RM et al. Neuropsychological evaluation of late-onset post-radiotherapy encephalopathy: a comparison with vascular dementia. J Neurol Sci 2005;229–230:195–200. 85. Costello A, Shallice T, Gullan R et al. The early effects of radiotherapy on intellectual and cognitive functioning in patients with frontal brain tumours: the use of a new neuropsychological methodology. J Neurooncol 2004;67(3):351–359. 86. Surma-aho O, Niemela M, Vilkki J et al. Adverse long-term effects of brain radiotherapy in adult low-grade glioma patients. Neurology 2001;56(10):1285–1290. 87. Armstrong CL, Corn BW, Ruffer JE et al. Radiotherapeutic effects on brain function: double dissociation of memory systems. Neuropsychiatry Neuropsychol Behav Neurol 2000;13(2):101–111. 88. Laack NN, Brown PD, Ivnik RJ et al. Cognitive function after radiotherapy for supratentorial low-grade glioma: a North Central Cancer Treatment Group prospective study. Int J Radiat Oncol Biol Phys 2005;63(4):1175–1183. 89. Noad R, Narayanan KR, Howlett T et al. Evaluation of the effect of radiotherapy for pituitary tumours on cognitive function and quality of life. Clin Oncol (R Coll Radiol) 2004;16(4):233–237. 90. Andrews DW, Scott CB, Sperduto PW et al. Whole-brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363(9422):1665–1672. 91. Sneed PK, Lamborn KR, Forstner JM et al. Radiosurgery for brain metastases: is whole brain radiotherapy necessary? Int J Radiat Oncol Biol Phys 1999;43(3):549–558. 92. Patchell RA, Regine WF. The rationale for adjuvant whole brain radiation therapy with radiosurgery in the treatment of single brain metastases. Technol Cancer Res Treat 2003;2(2):111–115. 93. Regine WF, Huhn JL, Patchell RA et al. Risk of symptomatic brain tumor recurrence and neurologic deficit after radiosurgery alone in patients with newly diagnosed brain metastases: results and implications. Int J Radiat Oncol Biol Phys 2002;52(2):333–338. 94. Regine WF, Scott C, Murray K et al. Neurocognitive outcome in brain metastases patients treated with accelerated-fractionation vs. accelerated-hyperfractionated radiotherapy: an analysis from Radiation Therapy Oncology Group Study 91–04. Int J Radiat Oncol Biol Phys 2001;51(3):711–717. 95. Lo SS, Chang EL, Suh JH. Stereotactic radiosurgery with and without whole-brain radiotherapy for newly diagnosed brain metastases. Expert Rev Neurother 2005;5(4):487–495. 96. Andrews DW, Scott C, Sperduto PW et al. Phase III randomized trial comparing whole-brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: eesults of the RTOG 9508 trial. Lancet 2004;363: 1665–1673. 97. Aoyama H, Shirato H, Nakagawa K et al. Interim report of the JRSOG99–1 multi-institutional randomized trial, comparing radiosurgery alone vs. radiosurgery plus whole brain irradiation for 1–4 brain metastases. Proc Amer Soc Clin Oncol 2004;23:108. 98. Chang EL, Wefel JS, Maor MH et al. A pilot study of neurocognitive function in patients with one to three new brain metastases initially treated with stereotactic radiosurgery alone. Neurosurgery 2007;60(2):277–283; discussion 83–84. 99. Penitzka S, Steinvorth S, Sehlleier S et al. Assessment of cognitive function after preventive and therapeutic whole-brain irradiation using neuropsychological testing. Strahlenther Onkol 2002;178(5):252–258. 100. Komaki R, Meyers CA, Shin DM et al. Evaluation of cognitive function in patients with limited small cell lung cancer prior to and shortly following prophylactic cranial irradiation. Int J Radiat Oncol Biol Phys 1995;33(1):179–182. 101. Parageorgiou C, Dardoufas C, Kouloulias V et al. Psychophysiological evaluation of short-term neurotoxicity after prophylactic brain irradiation in patients with small cell lung cancer: a study of event related potentials. J Neurooncol 2000;50(3):275–285. 102. Wefel JS, Lenzi R, Theriault RL et al. The cognitive sequelae of standard-dose adjuvant chemotherapy in women with breast carcinoma: results of a prospective, randomized, longitudinal trial. Cancer 2004;100(11):2292–2299. 103. Hurria A, Rosen C, Hudis C et al. Cognitive function of older patients receiving adjuvant chemotherapy for breast cancer: a pilot prospective longitudinal study. J Am Geriatr Soc 2006;54(6):925–931. 104. Fliessbach K, Helmstaedter C, Urbach H et al. Neuropsychological outcome after chemotherapy for primary CNS lymphoma: a prospective study. Neurology 2005;64(7):1184–1188. 105. Correa DD, Deangelis LM, Shi W et al. Cognitive functions in low-grade gliomas: disease and treatment effects. J Neurooncol 2007;81(2):175–184. 106. O’Neill BP. Neurocognitive outcomes in primary CNS lymphoma (PCNSL). Neurology 2004;62(4):532–533. 107. Correa DD, DeAngelis LM, Shi W et al. Cognitive functions in survivors of primary central nervous system lymphoma. Neurology 2004;62(4):548–555. 108. Harder H, Holtel H, Bromberg JE et al. Cognitive status and quality of life after treatment for primary CNS lymphoma. Neurology 2004;62(4):544–547. 109. Butler JM, Rapp SR, Shaw EG. Managing the cognitive effects of brain tumor radiation therapy. Curr Treat Options Oncol 2006;7(6):517–523. 110. Byrne TN. Cognitive sequelae of brain tumor treatment. Curr Opin Neurol 2005;18(6):662–666. 111. Torres IJ, Mundt AJ, Sweeney PJ et al. A longitudinal neuropsychological study of partial brain radiation in adults with brain tumors. Neurology 2003;60(7):1113–1118. 112. Taylor BV, Buckner JC, Cascino TL et al. Effects of radiation and chemotherapy on cognitive function in patients with high-grade glioma, J Clin Oncol 1998;16(6):2195–2201. 113. Brown PD, Jensen AW, Felten SJ et al. Detrimental effects of tumor progression on cognitive function of patients with high-grade glioma. J Clin Oncol 2006;24(34):5427–5433. 114. Frytak S, Shaw JN, O’Neill BP et al. Leukoencephalopathy in small cell lung cancer patients receiving prophylactic cranial irradiation. Am J Clin Oncol 1989;12(1):27–33. 115. Dimberg Y, Vazquez M, Soderstrom S et al. Effects of X-irradiation on nerve growth factor in the developing mouse brain. Toxicol Lett 1997;90(1):35–43.

110

Part III / Neurologic Symptoms

116. Shaw EG, Rosdhal R, D’Agostino RB, Jr. et al. Phase II study of donepezil in irradiated brain tumor patients: effect on cognitive function, mood, and quality of life. J Clin Oncol 2006;24(9):1415–1420. 117. Prommer E. Modafinil: is it ready for prime time? J Opioid Manag 2006;2(3):130–136. 118. Meyers CA, Weitzner MA, Valentine AD et al. Methylphenidate therapy improves cognition, mood, and function of brain tumor patients. J Clin Oncol 1998;16(7):2522–2527. 119. Cohen N, Strauss G, Lew R et al. Should prophylactic anticonvulsants be administered to patients with newly diagnosed cerebral metastases? A retrospective analysis. J Clin Oncol 1988(6):1621–1624. 120. Forsyth PA, Weaver S, Fulton D et al. Prophylactic anticonvulsants in patients with brain tumour. Can J Neurol Sci 2003;30(2):106–112. 121. Madhusoodanan S, Danan D, Moise D. Psychiatric manifestations of brain tumors: diagnostic implications. Expert Rev Neurother 2007;7(4):343–349. 122. Kelly C, Paleri V, Downs C et al. Deterioration in quality of life and depressive symptoms during radiation therapy for head and neck cancer. Otolaryngol Head Neck Surg 2007;136(1):108–111. 123. Gupta RK, Kumar R. Benign brain tumors and psychiatric morbidity: a 5-year retrospective data analysis. Austral New Zealand J Psychiatry 2004;38(5):316–319. 124. Moise D, Madhusoodanan S. Psychiatric symptoms associated with brain tumors: a clinical enigma. CNS Spectr 2006;11(1):28–31. 125. Cummings JL. Dementia: A Clinical Approach. Stoneham, MA: Butterworth-Heinemann; 1992. 126. Scharre. Neoplastic, demyelinating, infectious, and inflammatory brain disorders. In: Coffey CE CJ, ed. Textbook of Geriatric Neuropsychiatry. Washington, D.C.: American Psychiatric Publishing; 2000. 127. Belyi BI. Mental impairment in unilateral frontal tumours: role of the laterality of the lesion. Int J Neurosci 1987;32(3–4):799–810. 128. Price T, Lovell MR. Neuropsychiatric aspects of brain tumors. In: Yudofsky SC, Hales RE, ed. Neuropsychiatry and Clinical Neurosciences. Washington, D.C.: American Psychiatric Publishing; 2002. 129. Nasrallah HA, McChesney CM. Psychopathology of corpus callosum tumors. Biol Psychiatry 1981;16(7):663–669. 130. Schmahmann JD. Disorders of the cerebellum: ataxia, dysmetria of thought and the cerebellar cognitive affective syndrome. J Neuropsychiatry Clin Neurosci 2004(16):367–378. 131. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Washington, D.C.; 1994. 132. Meyers CA. Neurocognitive dysfunction in cancer patients. Oncology (Huntington) 2000;14(1):75–79. 133. Endicott J. Measurement of depression in patients with cancer. Cancer 1984;53:2243–2248. 134. Valentine AD, Meyers CA. Cognitive and mood disturbance as causes and symptoms of fatigue in cancer patients. Cancer 2001;92(6):1694–1698. 135. Mainio A, Hakko H, Niemela A et al. Gender difference in relation to depression and quality of life among patients with a primary brain tumor. Eur Psychiatry 2006;21(3):194–199. 136. Massie MJ. Depressive disorders. In: Holland JC, ed. Psycho-oncology. New York: Oxford University Press; 1998:518–540. 137. Derogatis LR, Morrow GR, Fetting J. The prevalence of psychiatric disorders among cancer patients. JAMA 1983(249):751–757. 138. Bukberg J, Penman D, Holland JC. Depression in hospitalized cancer patients. Psychosomatic Medicine 1984(46):199–212. 139. Pirl WF, Roth AJ. Diagnosis and treatment of depression in cancer patients. Oncology (Huntington) 1999(13):1293–1306. 140. Pringle AM, Taylor R, Whittle IR. Anxiety and depression in patients with an intracranial neoplasm before and after tumour surgery. Br J Neurosurg 1999;13(1):46–51. 141. Weitzner MA, Kanfer S, Booth–Jones M. Apathy and pituitary disease: it has nothing to do with depression. J Neuropsychiatry Clin Neurosci 2005;17(2):159–166. 142. Edvardsson T, Pahlson A, Ahlstrom G. Experiences of onset and diagnosis of low-grade glioma from the patient’s perspective. Cancer Nurs 2006;29(5):415–422. 143. Mainio A, Hakko H, Niemela A et al. Depression and functional outcome in patients with brain tumors: a population-based 1-year follow-up study. J Neurosurg 2005;103(5):841–847. 144. Litofsky NS, Farace E, Anderson F, Jr. et al. Depression in patients with high-grade glioma: results of the Glioma Outcomes Project. Neurosurgery 2004;54(2):358–366. 145. Mantani T, Saeki T, Inoue S et al. Factors related to anxiety and depression in women with breast cancer and their husbands: role of alexithymia and family functioning. Support Care Cancer 2007(10). 146. Hauser P, Soler R, Reed S et al. Prophylactic treatment of depression induced by interferon-alpha. Psychosomatics 2000;41(5):439–441. 147. Scheibel RS, Valentine AD, O’Brien S et al. Cognitive dysfunction and depression during treatment with interferon-alpha and chemotherapy. J Neuropsychiatry Clin Neurosci 2004;16(2):185–191. 148. Badger CT, Dorros SM, Meek P et al Depression and anxiety in women with breast cancer and their partners. Nursing Research 2007;56(1):44–53. 149. Campbell LC, Keefe FJ, Scipio C et al. Facilitating research participation and improving quality of life for African-American prostate cancer survivors and their intimate partners : a pilot study of telephone-based coping skills training. Cancer 2007;109(15):414–424. 150. Pelletier G, Verhoef MJ, Khatri N et al. Quality of life in brain tumor patients: the relative contributions of depression, fatigue, emotional distress, and existential issues. J Neuro-oncol 2002;57(1):41–49. 151. Osoba D, Aaronson NK, Muller M et al. Effect of neurological dysfunction on health-related quality of life in patients with high-grade glioma. J Neurooncol 1997;34(3):263–278. 152. Osoba D, Brada M, Prados MD et al. Effect of disease burden on health-related quality of life in patients with malignant gliomas. Neuro-oncol 2000;2(4):221–228. 153. Dimeo F, Stieglitz RD, Novelli-Fischer U et al. Correlation between physical performance and fatigue in cancer patients. Ann Oncol 1997(8):1251–1255. 154. Valentine AD, Meyers CA. Cognitive and mood disturbance as causes and symptoms of fatigue in cancer patients. Cancer 2001(92):1694–1698.

Chapter 7 / Cognitive Dysfunction, Mood Disorders, and Fatigue

111

155. Portenoy RK, Itri LM. Cancer-related fatigue: guidelines for evaluation and management. Oncologist 1999;4(1):1–10. 156. Lai YH, Chang JT, Keefe FJ et al. Symptom distress, catastrophic thinking, and hope in nasopharyngeal carcinoma patients. Cancer Nurs 2003;26(6):485–493. 157. Berger AM, Higginbotham P. Correlates of fatigue during and following adjuvant breast cancer chemotherapy: a pilot study. Oncol Nurs Forum 2000;27(9):1443–1448. 158. Irvine DM, Vincent L, Graydon JE et al. Fatigue in women with breast cancer receiving radiation therapy. Cancer Nurs 1998;21(2): 127–135. 159. Okuyama T, Tanaka K, Akechi T et al. Fatigue in ambulatory patients with advanced lung cancer: prevalence, correlated factors, and screening. J Pain Symptom Manage 2001;22(1):554–564. 160. Blesch KS, Paice JA, Wickham R et al. Correlates of fatigue in people with breast or lung cancer. Oncol Nurs Forum 1991;18(1):81–87. 161. Miaskowski C, Lee KA. Pain, fatigue, and sleep disturbances in oncology outpatients receiving radiation therapy for bone metastasis: a pilot study. J Pain Symptom Manage 1999;17(5):320–332. 162. Jacobsen PB, Hann DM, Azzarello LM et al. Fatigue in women receiving adjuvant chemotherapy for breast cancer: characteristics, course, and correlates. J Pain Symptom Manage 1999;18(4):233–242. 163. Yellen SB, Cella DF, Webster K et al. Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage 1997;13(2):63–74. 164. Wang XS GS, Mendoza TR et al. Clinical factors associated with cancer-related fatigue in patients being treated for leukemia and non-Hodgkin’s lymphoma. J Clin Oncol 2000(20):1319–1328. 165. Wang XS JN, Johnson BA et al. Fatigue during preoperative chemoradiation for respectable rectal cancer. Cancer 2001;92:1725–1732. 166. Hwang SS, Chang VT, Rue M et al. Multidimensional independent predictors of cancer-related fatigue. J Pain Symptom Manage 2003;26(1):604–614. 167. Broeckel JA, Jacobsen PB, Horton J et al. Characteristics and correlates of fatigue after adjuvant chemotherapy for breast cancer. J Clin Oncol 1998;16(5):1689–1696. 168. Visser MR, Smets EM. Fatigue, depression, and quality of life in cancer patients: how are they related? Support Care Cancer 1998;6(2):101–108. 169. Van Zandvoort MJ, Kappelle LJ, Algra A et al. Decreased capacity for mental effort after single supratentorial lacunar infarct may affect performance in everyday life. J Neurol Neurosurg Psychiatry 1998;65(5):697–702. 170. Lovely MP, Miaskowski C, Dodd M. Relationship between fatigue and quality of life in patients with glioblastoma multiformae. Oncol Nurs Forum 1999;26(5):921–925. 171. Groenvold M, Petersen MA, Idler E et al. Psychological distress and fatigue predicted recurrence and survival in primary breast cancer patients. Breast Cancer Research and Treatment 2007 (3). 172. Masand PS, Tesar GE. Use of stimulants in the medically ill. Psychiatry Clinics North America 1996(19):515–547. 173. Bruera E, Miller MJ, Macmillan K et al. Neuropsychological effects of methylphenidate in patients receiving a continuous infusion of narcotics for cancer pain. Pain 1992;48(2):163–166.

8

Cancer Pain Sebastiano Mercadante,

MD

CONTENTS Introduction Pain Mechanisms Assessment Treatment Breakthrough Pain Adjuvants Interventional Procedures Conclusion References

Summary Pain is a significant problem in cancer and there are many barriers to adequate pain control. Cancer-related pain is common and has a destructive impact on a patient’s quality of life. Physicians need to understand better the appropriate use of opioid and nonopioid analgesics and to consider other therapeutic options when appropriate. This chapter discusses the mechanisms of pain underlying cancer, and principles of management, and treatment of such pain. Key Words: pain, opioids, neuropathic pain, bone pain, headache

1. INTRODUCTION It has been estimated that the prevalence of chronic pain is about 30–50% among patients with cancer who are undergoing active treatment for a solid tumor and 70–90% among those with advanced disease. Pain is consistently one of the most feared consequences of cancer for both patients and families, and has a major destructive impact on quality of life. Availability of guidelines and accumulated experience have greatly improved the possibility of satisfactory pain control for most patients with advanced cancer. Nonetheless, pain frequently remains less than optimally controlled, due to physicians’ poor knowledge of and negative attitudes toward pain management as well as patient-related barriers and pain pathophysiology (1). In the United States, where medical care and opioids are readily available, it has been estimated that only 50% of patients receive adequate pain relief (2). There are many barriers to adequate pain control. Poor communication may exist between the patient and caregiver regarding the severity of the pain and the impact on quality of life. Many patients are reluctant to complain of pain or may not realize that adequate analgesia is available. Medical professionals often fail to assess the severity of cancer pain and have a poor knowledge of the proper use of analgesics in treating such pain, reflecting a lack of training in symptom management in most medical and nursing schools. Although attitudes are changing with improved education, many patients and medical professionals still harbor an unrealistic fear of addiction when opioids are used, despite overwhelming data that this is a very small risk in cancer patients. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

113

114

Part III / Neurologic Symptoms

Table 1 Pain Syndromes Associated with Oncologic Treatments Post-chemotherapy syndromes: • Chemoembolization, intraperitoneal, pleural, intrathecal infusions, hyperthermia • Mucositis • Polyneuropathy • Bone pain • Aseptic necrosis of bone • Steroid pseudorheumatism Post-surgical neuropathic syndromes: • Post-mastectomy • Post-axillary dissection • Post-thoracotomy • Post-radical neck dissection • Post-limb amputation • Post-rectal amputation • Stump pain Post-radiotherapy syndromes: • Enteritis • Dermatitis • Mucositis • Osteonecrosis • Myelopathy • Peripheral neuropathy • Radiation plexopathy

Table 2 Tumor-Related Pain Syndromes Nociceptive pain syndromes: Bone, joint, soft tissue pain syndromes Multifocal bone metastases Skull base metastases Vertebral syndromes Pelvis and hip metastases Joint and soft tissue involvement Paraneoplastic syndromes Osteoarthropathy Gynecomastia Visceral involvement Hepatic distension Retroperitoneal syndrome Carcinomatosis and intestinal obstruction Pelvic and perineal syndrome Ureteral obstruction Neuropathic syndromes: Peripheral mono- and polyneuropathies Plexopathies Radiculopathies Epidural compression

Chapter 8 / Cancer Pain

115

In approximately two-thirds of patients with cancer, pain is directly related to the presence of primary cancer or metastases. About one-third of patients develop pain syndromes because of oncologic treatments, which are increasing considerably (Table 1). Other related causes include comorbidities due to osteoporosis, immobility, and infections. The distinction is not always simple and requires a careful evaluation, including the use of imaging techniques and an expert neurologic assessment. Establishment of a pain mechanism is relevant for appropriate management. Pain related to malignant disease is commonly classified as nociceptive (somatic and visceral, respectively) and neuropathic in type (Table 2). Psychological factors may also have an important influence on these mechanisms. The term idiopathic is used when pain is perceived to be excessive for the extent of an organic pathology, and a psychological pathogenesis predominates.

2. PAIN MECHANISMS Cancer pain is a complex issue. It is initially a signal of ongoing injury associated with the onset or recrudescence of disease or may be caused by some diagnostic procedures. Commonly it subsides after oncologic treatments. In different stages of the disease, however, the causes cannot be adequately eliminated and symptoms persist. At this point cancer pain serves no biological purpose in alerting the organism to the presence of harmful stimuli but assumes the status of a chronic disease and is characterized by alterations in mood and pain behavior. The differences between acute and chronic pain include peripheral responses of the organism and central nervous system modifications induced by the chronic afferent volley of nociceptor activity (3). There are different clinical pictures with pathophysiological mechanisms that depend on the characteristics and progression of disease and the preferential sites of metastases. In order to prevent the clinical course and develop rational strategies in clinical practice, it is necessary to understand the basic mechanisms underlying cancer pain syndromes. Pain related to malignant disease can be classified as nociceptive (somatic and visceral, respectively) or neuropathic in type. Somatic and visceral pains involve direct activation of nociceptors, and are often complications of tumor infiltration of tissue or injury of tissues as a consequence of oncologic treatments. Neuropathic pain may be a complication of injury to the peripheral or central nervous system. This type of pain is often poorly tolerated and difficult to control.

2.1. Nociceptive Pain Nociceptive pain is usually subdivided into somatic and visceral types and involves direct activation of nociceptors. Nociceptive pain is associated with identifiable somatic or visceral lesion and is presumed to be related to ongoing activation of primary afferent neurons responsive to noxious stimuli. Nociceptors are widely distributed in skin, muscle, connective tissue, and viscera. No specific histological structure acts as a nociceptive receptor. A-delta and C nociceptors have been clearly identified in fibers innervating somatic structures. A repeated and intense stimulus induces the release of several inflammatory mediators that reduces the threshold for activation, increases the response to a given stimulus, or induces the appearance of spontaneous activity. Sustained stimuli or damage to the nerve can alter the profile of several peptides such as substance P contained within primary afferents. Substance P is able to induce the production of nitrous oxide, a vasodilatator, and the degranulation of mast cells with a further vasodilatation and subsequent extravasation and release of bradykinin. Bradykinin is a powerful algogenic substance that also sensitizes nociceptors by means of prostaglandins E2. Other factors, such as cytokines, are released after tissue damage under the influence of bradykinin. These substances have an important role in inflammatory processes. The concerted effects of these mediators at the site of tissue damage underlie peripheral hyperalgesia, which accounts for much of the peripheral sensitization of nociceptors (4). Deep pain originating from bone and visceral structures is more common than cutaneous pain. Muscle and visceral nociceptors exist in almost all organs, and appear to have different properties than do cutaneous nociceptors, including the property of referred pain. Experiments using distension most likely measure the effects of a mixture of visceral stimuli, including the stimulation of stretch receptors in the intestinal wall and a concomitant excitation of mesenteric mechanoreceptors

116

Part III / Neurologic Symptoms

(5). There is a good correlation between the size of the response with stimulus intensity. Traction on mesentery is another suitable method to investigate acute visceral nociception.

2.2. Neuropathic Pain Neuropathic pain most commonly occurs as a consequence of tumor compression or infiltration of peripheral nerves or the spinal cord. Trauma and chemical or radiation-induced injury as a result of surgery, chemotherapy, or radiotherapy may also result in this type of pain. This type of pain is increasing in patients who are long-term survivors—a population that will continue to increase as our treatments improve. Neuropathic pain may be a complication of injury to the peripheral or central nervous system, and is sustained by aberrant somatosensory processing. The pain cannot be explained by ongoing tissue injury, but rather is ascribed to foci of disease in the peripheral and central nervous system. It results from changes in the physiological response of neurons in the central or peripheral somatosensory system due to persistent stimulation or to a lesion of the nervous tissue caused by tumor itself, surgery, or chemotherapy. Neuropathic pain is most strongly suggested when dysesthesias (abnormal and unfamiliar pain sensations) occur in an area of motor, sensory or autonomic dysfunction attributable to a neurologic lesion. Subjective perception often includes burning or stabbing sensations associated with negative signs, such as hypoesthesia, and positive findings, such as dysesthesia, allodynia, and hyperalgesia (6). Clinical implications of a diagnosis of neuropathic pain are important as usually this type of pain is less responsive to opioids or requires higher doses (7). However, the diagnosis may be challenging because associated symptoms may be missed. Moreover, the labelling of a pain according to its inferred pathophysiology is a simplification of very complex processes that can involve multiple interacting mechanisms that can evolve and change over time. Although nociceptive and neuropathic pains depend on separate peripheral mechanisms, they are both significantly influenced by changes in central nervous function, due to a prolonged stimulation at high intensity. It is now clear that cancer pain is a more complex entity, particularly regarding the response to analgesics, where numerous factors play a role. While there is general acceptance of classification by inferred pathophysiology, the clinical utility in cancer pain syndromes cannot be precisely determined. Nevertheless specific pain related phenomena may yield information that has direct relevance to patient care, as some syndromes are predictably less responsive to treatment and may suggest the need to improve clinical monitoring and provide a more careful evaluation and possible alternative treatments.

3. ASSESSMENT Early recognition of pain is critical to successful treatment. Unfortunately, many studies have demonstrated the failure of medical professionals to identify the presence of pain and quantify it. Indeed, investigating the source of pain in patients with cancer may thus have important implications in the management of not just the pain but also the underlying malignancy. From a temporal perspective, acute pain is associated with a well-defined onset and may subside after anticancer treatment. Chronic pain is usually insidious and progressive, also characterized by fluctuations in intensity. Pain assessment should be included at each routine visit and particularly before, during, and after an oncologic treatment because it is possible to reduce the pain. A careful chronological review of the previous medical and oncologic history may help in placing the pain complaint in context. If the patient is either unable or unwilling to describe the pain, a family member may need to be questioned. It is important that caregivers routinely inquire about the presence and severity of pain and take responsibility for pain treatment. Responses to previous disease-modifying and analgesic treatments should be evaluated. Every type of pain should be considered, as multiple pain problems are common, and each of them should be assessed independently. Consequences of pain, including impairment of daily living activities, psychological and social dysfunction, decreased appetite, sleep disorders, and possible symptoms associated with pain have to be assessed. Psychological status, the level of anxiety, depression, and pervasive attitudes can be detected through a careful questioning of patient and caregivers. The physical examination, including a detailed neurologic exam, follows data collection. Identification of the underlying etiology of the pain problem and the relationship of the pain complaint to the disease are necessary parts of the initial pain assessment. Careful review of previous imaging

Chapter 8 / Cancer Pain

117

studies can provide important information about the cause of the pain and the extent of the underlying disease. Additional investigations may be required to clarify areas of uncertainty in a way proportional to the patient’s general status and goals of care. Pain intensity is commonly assessed by using simple validated methods, such as numerical scales from 0 to 10, verbal descriptors, or visual analogue scales, which may be repeated over time for an appropriate evaluation of the effects of an analgesic treatment. Older patients may prefer verbal scales and should be asked if their pain is mild, moderate, or severe (8). The severity of the pain will at least partially dictate the selection of analgesic. Pain ratings of 1–4 generally correspond to “mild pain.” A rating of 5–7 generally is considered “moderate,” and 8–10 is “severe.” In most patients, a pain rating over 4 will substantially interfere with daily activities and level of function (9), and requires further interventions. The topographic distribution of a specific pain has implications for diagnosis and treatment and often clarifies the relationship to the underlying organic lesion. “Referred pain” is a term applied when the distribution is remote to the lesion, different from focal pain, which is in the region of the underlying lesion. Discovering the underlying cause and mechanisms of pain is equally important, and imaging studies may be able to identify the source of pain. The principal pain syndromes are illustrated in Table 2. Transitory exacerbations of severe pain over a baseline of moderate pain have been described as breakthrough or episodic pain. These exacerbations may be precipitated by volitional actions, such as movement, or nonvolitional events. In some cases no identifiable precipitant is identifiable. This type of pain represents a negative prognostic indicator and poses complex therapeutic problems (vide infra). Symptoms such as anorexia, nausea, weakness, and drowsiness are invariably present in patients with cancer pain. These symptoms may affect the expression of pain and may be aggravated by the pain. On the other hand, these symptoms may occur as a consequence of the use of analgesic drugs. Thus, symptom monitoring is mandatory for the evaluation of a pharmacological pain treatment (10).

4. TREATMENT Anticancer therapy, including surgery and chemotherapy, may provide some analgesia, which is difficult to quantify. Radiotherapy provides an effective symptomatic treatment for local bone pain. As protracted courses of radiotherapy are difficult to perform in advanced cancer patients with a limited life expectancy and poor performance status, treatment with single fractions may be more convenient and equally effective in terms of pain relief. Orthopedic intervention may be indicated to restore mobility in bed-bound patients. Impending fractures require surgical stabilization using fixation devices or prosthetic reconstruction.

4.1. Analgesics: Pharmacological Treatment Analgesic drug therapy is the cornerstone of cancer pain therapy. Current treatment is based on the World Health Organization (WHO) analgesic ladder (11), which involves a stepwise approach to the use of analgesic drugs and is essentially a framework of principles rather than a rigid protocol. Despite the large experience with this approach, there is a paucity of controlled studies confirming that this is the best therapeutic option (12). However, apart from its educational importance, adherence to this basic approach will provide adequate analgesia in 80–90% of patients experiencing cancer-related pain (13). Step 1 is for patients with mild to moderate pain and involves the use of nonopioid analgesics, including acetaminophen and nonsteroidal anti-inflammatory drugs (NSAIDs). Step 2 analgesics include low-potency opioids including codeine, tramadol, and propoxyphene. Step 2 drugs are indicated for those patients with mild to moderate pain who do not obtain relief with nonopioid analgesics. Step 3 drugs are high-potency opioids, including morphine, oxycodone, hydromorphone, methadone, fentanyl, and buprenorphine. Step 3 is indicated for patients with severe pain or for those who fail to obtain adequate relief with a low-potency opioid. An “adjuvant” analgesic may be co-administered with drugs in Steps 1, 2, and 3. An adjuvant drug is defined as a medication that may not be considered analgesic by itself, but which may enhance the effectiveness of opioids, relieve opioid-related side effects, or act independently as an analgesic only or some specific types of pain (11). It is important to select the appropriate drug, dosage, and route of administration for the individual patient as well as to know how to titrate the dosage according to the analgesic response. Drug side effects should be

118

Part III / Neurologic Symptoms

anticipated and managed. A sequential trial of drugs may be appropriate if one medication is ineffective or the side effects unmanageable (vide infra). 4.1.1. Step 1 NSAIDs are drugs commonly used as the first step of the analgesic ladder (Table 3). The precise mechanism of analgesia is unknown. Acetaminophen is a weak cyclooxygenase inhibitor. It lacks the gastric irritant effects associated with NSAIDs and will not affect platelet aggregation. The maximum recommended dose is 3,000-4,000 mg/day in divided doses. Hepatotoxicity is the main dose-limiting side effect of acetaminophen. NSAIDs work peripherally by blocking the enzyme cyclooxygenase and inhibiting the formation of prostaglandins. Prostaglandins are potent mediators of inflammation that sensitize peripheral nociceptors. Adverse effects include GI bleeding, antiplatelet effects, and renal insufficiency. NSAIDs should not be coadministered with anticoagulants or corticosteroids, both of which enhance the risk of GI bleeding. Omeprazole, misoprostol, and H2-histamine blocking agents lessen the risk of GI bleeding and gastric upset when given with NSAIDs. NSAIDs have been claimed to have a major role in the management of some specific cancer pain syndromes, including pain from bone metastasis, soft-tissue infiltration, arthritis, and recent surgery. However, NSAIDs may be equally effective both for somatic and visceral pain as the first step of the analgesic ladder, and are also useful in combination with opioids, regardless of the pain mechanism involved (14). This class of drugs has been shown to have relevant opioid-sparing effects in long-term controlled studies and to maintain their effect for prolonged periods of time. Their principal drawback is the potential toxicity, particularly with the prolonged use, in the elderly, with comorbid conditions, and with drug interactions. No clear guidelines exist on the use of NSAIDS for prolonged periods, especially when opioids are given. In a pharmacoeconomic study, NSAIDs have been found more useful in patients already receiving opioids, as they still improve opioid-induced analgesia with a

Table 3 Non-Opioid Analgesics Drug

Half-Life (hrs)

Average Dose (mg)

Maximum Daily Dose (mg)

Comments

Acetaminophen

4

500 mg Q4-6 hrs PO or PR

4,000 mg

Avoid in patients with hepatic disease Avoid in children Avoid in renal disease Avoid in bleeding disorders Avoid in renal disease Avoid in bleeding disorders Less GI upset Avoid in renal disease

Aspirin

3–6

325-mg Q4-6 hrs PO or PR

2,000 mg

Ibuprofen

2–3

400 mg Q4-6 hrs

2,000 mg

Ketorolac

4–6

10 mg Q4-6 hrs PO 30 mg Q6 hrs IM or IV

60 mg (PO) 120 mg (IM or IV)

Celecoxib

11

200 mg QD PO

200 mg

Less GI upset Avoid in renal/cardiac disease

Rofecoxib

17

50 mg QD

50 mg

Similar to celecoxib

Chapter 8 / Cancer Pain

119

smaller financial impact. This approach may be more practical in avoiding prolonged administration of NSAIDs and selecting patients who have a consistent analgesic response which may limit opioid escalation (15). 4.1.2. Step 2 Opioids for moderate pain are largely used in patients who do not obtain adequate relief with a nonopioid analgesic. Step 2 drugs include codeine, propoxyphene, and tramadol (Table 4). High potency opioids can be used at relatively lower doses for the same purposes (16). The use of the opioid/nonopioid combination limits the maximum dose that can be administered, due to adverse effects from high doses of acetaminophen or aspirin. All of the low-potency opioids are approximately equally efficacious. Codeine has been the most popular opioid in this category. The analgesic activity seems to be dependent on the transformation to its metabolite, which is morphine, with large interindividual differences due to genetic polymorphism. About 10% of subjects are poor metabolizers. As most effects are derived from transformation to morphine, and that morphine is then metabolized to further substances, the clinical course may depend on the accumulation of these compounds in the presence of renal failure. The major metabolite of propoxyphene is norpropoxyphene. Norpropoxyphene has a 30-hrs half-life, and tends to accumulate in patients with reduced renal function. Propoxyphene should be avoided in elderly patients or those with a reduced renal function. Tramadol is a centrally acting analgesic with two mechanisms of action: weak opioid agonist activity and inhibition of monoamine uptake. Eighty-five per cent of tramadol is metabolized in the liver to one active metabolite, O-demethyl tramadol, and 90% is excreted by the kidneys. In patients with impaired hepatic or renal function the elimination half-life is increased approximately twofold. Consequently, multiple administrations of tramadol require increased dosage intervals (17). Notwithstanding, tramadol is the most popular drug of step 2. 4.1.3. Step 3 When cancer patients experience severe pain, strong opioids are the mainstay of therapy (Table 5). There is a large variety of options for the delivery of opioids in the management of cancer pain. In some instances, there are clear indications for using one preparation or delivery system over another, according to the patient’s ability to use a specific type of delivery system, the efficacy of that system to deliver acceptable analgesia, the ease of use for the patient and their family, and the potential or actual complications associated with that system. Cost is another important consideration for patients who must purchase their own medications. 4.1.3.1. Oral Route. The oral route is the most common, least invasive, least expensive, and easiest route for opioid administration for most patients with cancer pain. As most opioids are available in an oral formulation and the bioavailability by this route is relatively good, in patients who can take oral medications this route should be considered first (18).

Table 4 Low-Potency Opioid Analgesics Drug

Half-Life (hrs)

Starting Dose (mg)

Comments

Codeine

3–5

30-60 mg Q4-6H PO or IV

Often combined with non-narcotics

Propoxyphene

8–12

65 mg Q4H PO

Often combined with non-narcotics Norpropoxyphene may cause seizures Avoid in renal disease and elderly

Tramadol

6–8

50 mg Q4-6H PO

120

Part III / Neurologic Symptoms

Table 5 High-Potency Opioid Analgesics Drug

Half-Life (hrs)

Equianalgesic Dose (mg)

Comments

Morphine

3–5

10

Available in immediate and sustained release form

Hydromorphone

3–5

2

Available in immediate and sustained release form

Methadone

15–24

1–2

Drug accumulates after several days favored by hepatic dysfunction, potential interactions

Fentanyl

–

0.1

Available in transdermal patch Transmucosal formulation for breakthrough pain

Buprenorphine

–

0.14

Available in transdermal patches

Oxycodone

2–3

7

Available in immediate and sustained release form

The main problem with the oral route is the first-pass biotransformation of opioids in the liver. All opioids given orally are absorbed via the gastric and duodenal mucosa and then transported to the liver via the portal venous system. In the liver, these medications undergo “first-pass metabolism” before entering the systemic circulation. This has a major impact on the systemic plasma concentrations of drugs. Bioavailability is defined as the percentage of administered medication that reaches the systemic circulation. For example, the oral morphine dose for a patient with cancer pain must be three times the intravenous or intramuscular dose. Oxycodone, while roughly equipotent to morphine if given parenterally, appears to be approximately twice as potent as morphine when given orally because of less first-pass metabolism. Morphine, the most commonly used medication in the world to treat cancer pain, has a terminal elimination plasma half-life of about 3 hrs. To provide longer-lasting analgesia, several preparations have become available. Bioavailability of these slow-release preparations is the same as that of immediate-release preparations, but time to peak plasma drug concentrations is longer, and peak plasma concentrations is decreased. These preparations are recommended by the manufacturer to be administered every 12 hrs. Clinicians occasionally use an 8-hrs schedule, if necessary, to provide adequate analgesia. Other preparations, such as a morphine pellet coated with a polymer, are manufactured to be administered once every 24 hrs. If additional analgesia is needed for “breakthrough” pain, doses of a fast-onset, short-acting opioid preparation should be available to the patient (18). However immediaterelease oral opioid preparations usually require approximately 30 min to onset of analgesic action when taken on an empty stomach, and faster routes may be required. Morphine is highly extracted by the liver and is metabolized to two major metabolites, morphine-3-glucuronide (M3G) and morphine-6-glucuronide (M6G), which account for approximately 40% and 10%, respectively, of the morphine dose recovered in the urine. M6G formed from morphine in the liver accumulates in blood and penetrates the BBB, binding with strong affinity to opioid receptors, exerting a strong analgesic effect. While M6G is a more potent analgesic than morphine, M3G has been shown to antagonize effects of both morphine and M6G, although it is devoid of any binding to opioid receptors. Different routes of administration seem to be associated with different ratios of metabolites. Patients receiving oral morphine had higher plasma concentration ratios of glucuronide/morphine than those receiving subcutaneous therapy, presumably due to first-pass glucuronidation. While hepatic impairment was reported to be an insignificant factor influencing pharmacokinetics of morphine, probably since there is a relatively large hepatic reserve for glucuronidation, renal failure is recognized to have profound effects on the behavior of the glucuronide metabolites of morphine. While morphine itself remains largely unaffected by renal failure, accumulation of both the metabolites has

Chapter 8 / Cancer Pain

121

been reported. As M6G is excreted by the kidney, its concentration rises in renal failure and can lead to severe toxicity (17). Oxycodone, methadone, and hydromorphone are possible alternative to oral morphine. Oxycodone is a semisynthetic opioid agonist that can be used as an alternative to morphine in controlling chronic pain. Noroxycodone, oxymorphone and conjugated forms of oxycodone are the major metabolites—oxymorphone contributing minimally in clinical activity, despite its analgesic potency, because it accounts for only 10% of oxycodone metabolites and has little passage through the blood–brain barrier. Most metabolites, as polar substances produced by hepatic biotransformation, are principally eliminated by the kidneys. The equivalence ratio with morphine is about 1:1.5 (17). Hydromorphone is more potent and soluble than morphine and, like morphine, can be tailored to the patient’s needs. Hydromorphone is transformed in the liver to glucuronides, H3G and H6G, respectively, which are polar substances undergoing renal excretion. Hydromorphone and its metabolites accumulate in renal failure, recapitulating the problems associated with morphine and its metabolite elimination in such circumstances. The equivalence ratio with morphine is about 1:5 (17). Methadone is a low-extraction drug and is metabolized, mainly by N-methylation in the liver, to inactive metabolite. Its pharmacokinetics are highly variable, due to pronounced interindividual differences in the proportion of the metabolite versus parent drug that is excreted renally. This makes the use of the drug more complex, because of the difficult predictability and the risk of accumulation. Unlike the previous drugs, methadone plasma concentrations do not change significantly in patients with abnormal renal function (17). Many patients will develop tolerance to most of the undesirable side effects of opioids (such as nausea/vomiting or sedation) over a period of several days. However, certain patients may not be able to tolerate oral medications because of esophageal motility problems or gastrointestinal obstruction (e.g., head and neck or esophageal cancer, bowel obstruction) or may have persistent nausea and vomiting, limiting the utility of the oral route. Finally, some patients are unable to swallow due to the site of their cancer or because they are neurologically impaired. Alternative routes, including the intravenous and subcutaneous, as well as the transdermal ones, have been advocated in such circumstances. 4.1.3.2. Parenteral Route. The intravenous route of administration is indicated for patients whose pain cannot be controlled by a less invasive route or who already have central venous access (18). The major disadvantage of this route is that it is more complex to manage, especially at home, and requires some expertise. On the other hand, this route is the fastest, allowing for an immediate effect in emergency conditions, and providing the best conditions for rapid opioid dose titration (19). For patients with severe pain, the intravenous route is the fastest way to obtain analgesia and determine the patient’s opioid requirements. Once the pain is controlled, the opioid is easily converted into an equianalgesic oral dosage (Table 5). Different opiods are available in intravenous solution in the majority of countries, including morphine, hydromorphone, fentanyl, alfentanil, sufentanil, and methadone. Fentanyl is approximately 80–100 times more potent than morphine, and sufentanil is approximately 1,000 times more potent. The main drawback to their use in the practice of oncology is their high cost compared with the cost of morphine. For patients requiring parenteral opioids who do not have indwelling intravenous access, the subcutaneous route can be used. This simple method of parenteral administration is quite popular in the palliative care setting. The limiting factor is the volume of fluid that can be injected per hour, often requiring more concentrated solutions (20). Most drugs used by intravenous route can also be used by subcutaneous infusion, except methadone because of local toxicity. The main advantages of a subcutaneous over an intravenous route is that there is no need for vascular access, changing sites can be easily accomplished, and problems associated with indwelling intravenous catheters can be avoided. The oral/parenteral ratio for morphine is 2:1 or 3:1 (18). Intravenous or subcutaneous opioid infusions can be given as continuous infusions or by a patient-controlled analgesia (PCA) device, which provides continuous infusion plus on-demand boluses. The PCA device should be set initially to deliver a continuous infusion with bolus and lock-out intervals depending on clinical needs. Patients unable to operate a PCA device due to cognitive impairment or overwhelmed by the technical aspects of care, and patients with drug-seeking behavior, would be inappropriate for PCA. Other drawbacks of conventional parenteral PCA include its invasiveness and the complications inherent in the long-term use of subcutaneous needles or intravenous lines.

122

Part III / Neurologic Symptoms

4.1.3.3. Transdermal Route. For patients unable to take oral medications, the transdermal route is a noninvasive option of maintaining relatively constant plasma concentrations of opioids. Transdermal delivery systems have several advantages over traditional routes of administration in the management of chronic pain. These advantages include noninvasive administration and rate-controlled release; the active compound diffuses passively into the systemic circulation, allowing therapeutic serum levels to be maintained 48–72 hours following a single application (21). Fentanyl and buprenorphine are available transdermally because of their favorable potency and lipophilicity. Upon initial application of the patch, a subcutaneous “depot” is formed as opioids saturate the subcutaneous fat beneath the patch. After approximately 12 hrs, steady-state plasma opioid concentrations are reached, which are maintained for about 48–72 hrs. Fentanyl is metabolized in the liver to compounds that are both inactive and nontoxic, which are excreted in the urine. Less than 10% of the drug is excreted unchanged in the urine. Fentanyl may be an alternative to morphine in patients with renal impairment, due to lack of accumulation of important metabolites. Fentanyl patches are currently available in 25, 50, 75, and 100 μg/hr dosages. The bioavailability of transdermal fentanyl is very high, approximately 90%. Because of the slow depot formation and slow rise in plasma concentrations, this system is not suitable for rapid titration against pain, and is best suited for patients with stable pain in whom the 24-hrs opioid requirement has already been determined. Problems arise from conversion to transdermal fentanyl, as no clear protocols have been established. It has been suggested to use a conversion fentanyl-morphine ratio of 1:70–100. However, patients may still be under- or overdosed. Buprenorphine is metabolized in the liver to produce inactive and weakly active metabolites, which are principally excreted via the biliary system. Pharmacokinetics of buprenorphine change minimally in patients with renal failure, and patients with renal impairment may safely receive buprenorphine. Buprenorphine patches are available in different sizes, 35 mcg/hr corresponding to about 0.8 mg/day. The equianalgesic ratio with oral morphine is approximately 1:70. Buprenorphine should be considered as a strong opioid although lower doses can be effective in naïve patients. Globally, fentanyl and buprenorphine are less influenced from an abnormal renal function, in comparison with morphine and hydromorphone, and possibly have a less constipating effect (17). 4.1.3.4. Transmucosal/Sublingual Route. The sublingual administration of opioids is particularly beneficial in patients who are unable to tolerate oral administration because of nausea/vomiting or dysphagia. It may also be attractive in patients who cannot receive parenteral opioids because of the lack of venous access or with contraindications for subcutaneous drug administration. Because sublingual venous drainage is systemic rather than portal, hepatic first-pass elimination can be avoided. Furthermore, the transmucosal or sublingual route also offers the potential for more rapid absorption and onset of action relative to the oral route. This is particularly useful for treating breakthrough pain. Lipophilic drugs are better absorbed than are hydrophilic drugs. Transmucosal fentanyl is the only medication that has been found to be a very useful tool in the management of breakthrough pain in cancer patients in several controlled studies.

4.2. Opioid-Related Adverse Effects All opioids have common adverse effects. The clinician must understand the mechanism and management of these effects in order to use opioids safely and effectively. The concomitant use of adjuvant drugs may minimize these effects in many circumstances. The most common opioid side effects are constipation, sedation, dry mouth, delirium, nausea, respiratory depression, pruritus, convulsions, and myoclonus. All opioid drugs have similar side effects, although an individual patient may tolerate one opioid better than another. Opioid adverse effects are preventable or treatable, and subsequently patients will develop some degree of tolerance to them (22). Precise guidelines about the symptomatic treatment of these side effects have not been produced (23). Constipation is the most common and often refractory opioid-related adverse effect. Opioids produce increased resting tone in smooth muscle of the small and large intestine and reduce peristalsis. Stool softeners should be started when opioids are first prescribed. A combination of senna and docusate may be useful. If constipation has already developed, stimulant laxatives or osmotic agents should be added. Respiratory depression is the most feared opioid adverse effects. However, it most commonly occurs in opioidnaïve patients who receive high initial doses of opioids and it is unlikely to occur in patients receiving therapeutic doses regularly. Opioids may produce sedation, particularly when starting the medication in an opioid-naïve

Chapter 8 / Cancer Pain

123

individual or when titrating the dosage upwards. Generally the sedation is not severe and tolerance will develop within a period of days. If the tolerance is incomplete or the patient is not willing to wait, then a stimulant may be administered. Small doses of methylphenidate may mitigate sedation. Opioid-induced nausea and vomiting may have multiple mechanisms, as opioids activate receptors in the chemoreceptor trigger zone of the dorsal medulla, reduce gastric emptying by increasing smooth muscle tone in the gastroduodenal sphincter and other intestinal sites, and sensitize the vestibular apparatus in the inner ear, particularly in young patients. Phenothiazines, metoclopramide or haloperidol, and scopolamine may be effective according to the presumed prevalent mechanism. Any opioid may produce myoclonus, although meperidine is particularly prone to producing this adverse effect. Meperidine may produce convulsions even at relatively low dosages with chronic administration, particularly in those patients with reduced renal function, probably due to its toxic metabolite accumulation. A decrease in opioid dose or adding clonazepam or gabapentin may be helpful. When opioids produce pruritus, it is usually from direct histamine release. Antihistamines may often be sufficient. If symptomatic treatments fail, however, switching the patient to a different opioid may remedy the situation.

4.3. Opioid Switching As opioids have differential effects on selective subsets of opioid receptors in the central nervous system and cross tolerance between opioids is incomplete, a shift from one opioid to another is a useful option when the side effect-analgesic relationship is inconvenient and also because it allows for elimination of possible toxic metabolites. Patients who experience dose-limiting adverse effects during opioid escalation may benefit from a trial of an alternative opioid. In a switch from one opioid to another, the latter drug is often observed to be relatively more potent than would be anticipated, given published estimates. The first large series published on opioid rotation included 80 patients undergoing 111 episodes of opioid rotation. Most patients were receiving morphine and generally were switched to hydromorphone. Principal symptoms improved in more than 70% of cases, and after the switch the median equivalent daily dose of parenteral morphine decreased from 577 to 336 mg, associated with a significant decrease in pain intensity. This observation was particularly prominent with switching to methadone, with a decrease in parenteral equivalent morphine dose of about 36% (24). Hydromorphone or oxycodone is often used. Common conversion ratios with morphine for these drugs are 5 and 1.5, respectively. No correlation between the dose of the previous opioid and the final ratio between hydromorphone and morphine has been found, demonstrating partial cross-tolerance between these two substances (25). Methadone’s potency may be much greater than expected when a switch is made from another drug because tolerance is reversed, probably due to its anti-NMDA effect, and a strict surveillance is necessary when converting patients who are taking high doses of opioids. Methadone has been reported to be useful in restoring opioid responsiveness in patients whose pain ceases to be controlled by morphine, hydromorphine, or diamorphine at doses much lower than those suggested by the opioid conversion charts. The clinical benefit will depend on the degree to which cross-tolerance exists to analgesia as well as to side effects. As the degree of cross-tolerance may change as opioid doses are escalated, it is advisable to proceed with caution when switching from any opioid to another one in patients receiving very high opioid doses. Data suggest that switching to methadone using current proposed ratios may lead to severe toxicity. A strongly positive correlation between dose ratio and previous morphine dose suggests the need for a highly individualized and cautious approach when rotating from morphine to methadone in patients with cancer pain on high doses of morphine (25–27). Other authors suggest an initial dose of methadone of one-tenth of the total daily morphine dose, but not greater than 30 mg at intervals determined by the patients, but no more frequently than every 3 hrs, and then to divide the steady-state total daily dose (after 6 days) into a twice-daily regime (28). If long-acting opioid metabolites are considered responsible for adverse effects, a rapid switching using a morphine/methadone ratio of 5:1, and a monitoring of the dosage in the following days may be useful (and is currently used at our institution). The subsequent titration process should take into account the characteristics of the pain syndromes and the individual clinical situation (29).

124

Part III / Neurologic Symptoms

Due to the slow onset of transdermal analgesia, switching to fentanyl due to poor pain control is not advisable, unless other methods of analgesia are provided in the meantime. Patients are more commonly switched from morphine to fentanyl for convenience or adverse effects. An initial oral morphine/transdermal fentanyl of 100:1 seems to be acceptable, although the final ratio after switching may be lower (70:1) (30).

5. BREAKTHROUGH PAIN Breakthrough pain is a transitory flare of pain superimposed on an otherwise stable pain pattern in patients treated with opioids, and is normally severe in intensity, has a rapid onset, and is considered a negative prognostic factor. Although different mechanisms, often unknown, may precipitate a flare, the most common form of breakthrough pain is incident pain, due to movement and commonly associated with bone metastases. A rescue dose of opioid can provide relief in patients already stabilized on a baseline opioid regimen. The use of a short half-life opioid, such as immediate-release morphine or hydromorphone, is suggested. The size of the most effective dose remain unknown, although clinicians suggest a dose roughly equivalent to about 15% of the total opioid dose administered as needed every 2–3 hrs. Titration of the rescue dose according to the characteristics of breakthrough pain should be attempted in an individual way to identify the most appropriate dose, as the approach to opioid supplemental dosing has been based solely on anecdotal experience (31). However, the onset of action of an oral dose may be too slow (more than 30 min) and better results may be obtained with a parenteral rescue dose. Although the intravenous route is the fastest, subcutaneous administration is associated with an acceptable onset of effect and should be considered equivalent in terms of efficacy. Oral transmucosal dosing is a recent noninvasive approach to the rapid onset of analgesia. Highly lipophilic agents may pass rapidly through the oral mucosa, avoiding the first-pass metabolism and achieving active plasma concentrations within minutes. Fentanyl, incorporated in a hard matrix on a handle, is rapidly absorbed. It has been shown to have an onset of pain relief similar to intravenous morphine within 10–15 min. When the fentanyl matrix dissolves, approximately 25% of the total fentanyl concentration crosses the buccal mucosa and enters the circulation. The remaining amount is swallowed and about one-third of this part is absorbed, thus achieving a total bioavailability of 50%. Different controlled studies have shown the effectiveness of oral transmucosal fentanyl for treating episodes of breakthrough pain. Of importance, the effective dose was not correlated with the basal analgesic regimen, stressing the need to individualize the dose (32–34).

6. ADJUVANTS One approach to the patient with pain that is poorly responsive to opioids is the co-administration of a nonopioid analgesic. There are a very large number of options (35). Antidepressants may improve depression, enhance sleep and provide decreases in perception of pain. The analgesic efficacy of the tricyclic antidepressants has been established in many painful disorders. The evidence supporting analgesic effects is particularly strong for amitriptyline. The analgesic effect of tricyclics is not directly related to antidepressant activity. The analgesic response is usually observed within 5 days. Alternative drugs with lower incidences of side effects should be considered in patients predisposed to the sedative, anticholinergic or hypotensive effects of amitriptyline. Despite the frequent use of amitriptyline in neuropathic cancer pain, its effectiveness has not been demonstrated appropriately in this context. Adverse effects are of concern. Common side effects of tricyclic compounds include antimuscarinic effects, such as dry mouth, impaired visual accomodation, urinary retention, and constipation, antihistaminic effects (sedation), and antialpha-adrenergic effects (orthostatic hypotension). The potential benefits of amitriptyline are associated with a high rate of adverse effects, particularly in advanced disease. It has been suggested that an anomaly in ion channels plays a role in the molecular mechanism of neuropathic pain. Sodium channel-blocking agents, including systemic local anesthetics (e.g., mexiletine), carbamazepine, and phenytoin are useful for the management of chronic neuropathic pain due to a presumed inhibition of sodium channels of hyperactive and depolarized nerves while not interfering with normal sensory function. No conclusive clinical study, however, has statistically verified these observations in cancer pain and open sodium channel blockers have been reported to relieve neuropathic pain states. Anticonvulsants, such as carbamazepine, phenytoin, valproate, and clonazepam, have been reported to relieve pain in numerous peripheral and central neuropathic

Chapter 8 / Cancer Pain

125

pain conditions, although contradictory results have been found. The efficacy of these drugs can be explained by the inhibitory effects exerted on NMDA receptors, and other mechanisms, including sodium channel blockade for some. No measurable differences in the analgesic benefit of anticonvulsants and antidepressants were evidenced in a systematic review of available trials in neuropathic pain. Gabapentin and pregabalin are increasingly used as adjuvants to opioid analgesia for neuropathic cancer pain. The addition of gabapentin in the therapeutic regimen of cancer patients decreased the pain score and some typical sensations associated with neuropathic pain (36). A number of studies have documented the positive effects of corticosteroids on various cancer-related symptoms, including pain, appetite, energy level, food consumption, general well-being, and depression. Although analgesia in diverse pain syndromes has been reported, most of the evidence for these effects is anecdotal. The bisphosphonates are pyrophosphate analogues, characterized by a P—C—P bond that renders them resistant to hydrolysis by phosphatates. Bisphosphonates potentiate the effects of analgesics in metastatic bone pain. Incident pain due to bone metastases may potentially benefit directly or through prevention of pathological fractures (37). Pamidronate, a potent bisphosphate, significantly reduced morbidity caused by bone metastases, including a 30–50% reduction in pain, impending pathological fractures, and the need for radiotherapy. Best results are obtained with doses of 60 mg or 90 mg pamidronate. This treatment is generally well-tolerated; side effects include transient low-grade fever, nausea, myalgia, and mild infusion site reactions. Intravenous application has been preferred by most investigators, mostly because of gastrointestinal adverse events and the urgency of the situation. The subsequent generation, including zoledronate and particularly ibandronate, seem to be even more effective than pamidronate.

7. INTERVENTIONAL PROCEDURES Numerous different chemical substances and anesthesiological techniques have been applied to elements of the central and peripheral nervous systems in efforts to disrupt or modulate pain transmission.

7.1. Spinal Route A small number of patients may still fail to obtain adequate analgesia despite large systemic opioid doses, or they may suffer from uncontrollable side effects such as nausea, vomiting, or over sedation. These patients may be candidates for the administration of a combination of opioids, local anaesthetics, and clonidine via the spinal (epidural or intrathecal) route. The goal of spinal opioid therapy is to place a small dose of an opioid and/or local anesthetic close to the spinal opioid receptors located in the dorsal horn of the spinal cord to enhance analgesia and reduce systemic side effects by decreasing the total daily opioid dose. Use of this route to deliver opioids requires placing a catheter into the epidural or intrathecal space and using an external or implantable infusion pump to deliver the medications. Deciding between epidural versus intrathecal placement or external versus implantable pumps to deliver the opioid is based on multiple factors including duration of therapy, type and location of the pain, disease extent and central nervous system involvement, opioid requirement, and individual experience. The daily epidural opioid requirement is approximately 10 times that of intrathecal administration. Intrathecal opioid administration has the advantage of allowing a higher concentration of drug to be localized at the receptor site while minimizing systemic absorption, thus possibly decreasing drug-related side effects. Morphine remains the drug of choice for the spinal route because of its relatively low lipid solubility. It has a slow onset of action, but a long duration of analgesia when given via intermittent bolus. The starting dose is quite difficult to calculate and should take into consideration various factors, including the previous opioid dose, age, and the pain mechanism. Adding a local anesthetic (bupivacaine or ropivacaine) to morphine via the spinal route has been successful in providing good analgesia in patients whose pain was resistant to epidural morphine alone, despite high doses. Further clinical studies and trials are still required to judge the safety, efficacy, and extended role of the spinal route in chronic cancer pain and, more importantly, to define in which patients this technique is indicated. Clonidine, an alpha-adrenergic agonist that acts at the dorsal horn of the spinal cord to produce analgesia, has been used in cancer patients in combination with epidural (or intrathecal) morphine infusions. There is

126

Part III / Neurologic Symptoms

some evidence to suggest that neuropathic pain may be somewhat more responsive to the combination of clonidine/morphine than to morphine alone, although orthostatic hypotension is of concern. Procedural and surgical complications, system malfunction, and pharmacological adverse effects are the main categories of complications associated with spinal drug delivery (38).

7.2. Cordotomy and Pituitary Ablation A percutaneous cordotomy is the interruption of the ascending spinothalamic tract, usually at the cervical level. A percutaneous cervical cordotomy by radiofrequency ablation has been utilized in patients with unilateral bone pain below the C5 dermatome. Cordotomy may be indicated in a selected group of patients with refractory breakthrough pain due to bone metastases, for example, in femoral neck fractures, but it is not indicated in the presence of neuropathic lesions. The risk of serious complications, including mirror pain, general fatigue or hemiparesis, and respiratory failure, with performance status deterioration, is high. Pituitary ablation has a potential role in patients with widely disseminated pain of bony metastatic origin, and in patients with hormonally responsive primary tumor. Although the success rate has been quoted as 74–94%, long-term follow-up to death has not been carried out successfully. High mortality rate (2–6%) has been reported with transient morbidity regarding rhinorrea, meningitis, visual disturbances, diabetes insipidus, headache, and hypothalamic disturbances (39).

7.3. Sympathetic Blocks Superior hypogastric plexus block has been claimed to be highly effective to control pelvic pain syndromes (40). However, criticisms regarding the frequent occurrence of mixed pain syndrome in pelvic cancers due to overlapping of neuropathic and somatic pain mechanisms reducing the advantages of the block, have been raised. The less favorable results obtained with the hypogastric plexus block may be due to the greater tendency of pelvic tumors to infiltrate somatic structures and nerves compared to pancreatic tumors where the celiac plexus block is usually used. Other than with visceral pain, pelvic cancers are more often associated with myofascial involvement, and as a consequence, with a somatic pain mechanism. Thus, pelvic cancer pain can arise, among other reasons, from visceral and peritoneal involvement as well as from the involvement of muscles in the perineal floor with nerve entrapment and muscle necrosis. Due to the location of the piriformis muscle deep in the pelvic floor, pain may be a complication of a primary tumor or relapse, surgery, and radiotherapy to the pelvis. Buttock or rectal pain with or without posterior thigh pain, aggravated by sitting or activity, may be due to the compression of piriformis muscle. As a consequence, it may be associated with referred/localized pain of somatic origin. Motor reflexes may lead to muscle spasm and additional component of somatic pain. The clinical picture is often mixed. In pelvic cancer lumbar pain may also be due to iliopsoas muscle involvement. Nerve trunk pain often radiating to lower limbs may be observed, due either to the involvement of lumbosacral plexus involvement in the presacral area or radiculopathy related to retroperitoneal spread. As a consequence, candidates for hypogastric plexus block, a sympathetic block performed for visceral pelvic pain, have to be strictly selected. On the other hand, retroperitoneal invasion may result in a limited spread of neurolytic solution and may be another reason for a lack of success. In a sample of advanced cancer patients with a mean survival of 6 weeks, pelvic pain syndromes were found to present multiple mechanisms. Fewer than half of patients had a pure visceral component, which should be expected to respond optimally to superior hypogastric plexus block. The choice of performing such a block should be based on patient preference after careful explanation of the possible advantages and complications, considering it as an adjuvant treatment and not as a last resort, when the advantages of the block seem to be minimal. The neurolytic celiac plexus block (NCPB) is an established, well-developed procedure and the most widely applicable of all the neurolytic pain blocks available (41). It is the least hazardous means of palliative treatment for cancer of the upper abdominal viscera. Unlike other nerve blocks, motor and sensory complications are rare as the celiac ganglia carries only sympathetic and parasympathetic supply to the upper abdominal viscera. Pain transmitted via the celiac plexus is primarily from the upper abdomen, including pancreas, diaphragm, liver,

Chapter 8 / Cancer Pain

127

spleen, stomach, small bowel, ascending and proximal transverse colon, adrenal glands, kidneys, abdominal aorta, and mesentery. Several techniques have been proposed differing significantly in the site of approach, solution injected, instruments used, and radiological guidance methods. Substantially, there is a clear distinction between the retrocrural and transcrural techniques, and anterior or posterior approaches. The NCPB is claimed to be an effective treatment, particularly for pancreatic cancer pain. The NCPB made it possible to maintain a reduced opioid consumption and fewer problems related to treatment when compared to the use of analgesics only. Some authors hypothesized that NCPB is more effective if performed early after pain onset, when the pain is still only or mainly of a visceral type and responds to anti-inflammatory drugs. However, this effect will be limited in any case where the tumor grows diffusely as complex abdominal disease, regardless of the early advantages due to a better anatomical situation when the block is performed. Thus, NCPB should be considered as an adjuvant measure in the context of the management of abdominal cancer pain with the aim of improving quality of life, reducing pain intensity and allowing a reduction in opioid dosing, and consequently limiting the adverse effects associated with their administration, in the context of comprehensive palliative care treatment.

8. CONCLUSION Pain is a prevalent and important symptom associated with all stages of cancer. For the patient with cancer pain, the oral route of opioid delivery should be the first choice. If the oral route cannot be used because of gastrointestinal obstruction and/or severe nausea/vomiting, a noninvasive alternative to the oral route is the transdermal route, which at present is available for administration of fentanyl or buprenorphine. For treatment of breakthrough pain a transmucosal preparation of fentanyl is available, producing a more rapid effect than oral opioids. For those patients in whom oral or transdermal opioids are not appropriate, intravenous or subcutaneous administration is effective, the latter route being relatively easier to administer. The spinal route can be attempted when the oral and other parenteral routes have been unsuccessful. This route may be most successful when opioids and local anaesthetics and/or clonidine are used in combination. Whatever route is used, administration of opioids to manage cancer pain requires knowledge of potency relative to morphine and bioavailability of the route chosen. Dose-equivalent tables are only close approximations and substantial interpatient variability is often observed. Therefore, patients should be closely followed and doses titrated to minimize side effects whenever the opioid, route, or dose is changed. Hypogastric plexus block and NCPB may help in reducing the visceral component of pain in an attempt to reduce analgesic consumption. The role of neurolytic procedures has not still substantiated in controlled studies. When the pain is complex or intractable, an early and comprehensive consultation with the pain and palliative care service will ultimately benefit the patient and improve his or her quality of life.

REFERENCES 1. Cleeland CS, Gonin R, Hatfield AK et al. Pain and its treatment in outpatients with metastatic cancer. N Engl J Med 1994;330:592–596. 2. Von Roenn JH, Cleeland CS, Gonin R et al. Physician attitudes and practice in cancer pain management. Ann Int Med 1993;119: 121–126. 3. Besson JM. The neurobiology of pain. Lancet 1999;353:1610–1615. 4. Dickenson AH. Pharmacology of pain transmission and control. In: Pain 1996: An Updated Review, JN Campbell ed. Seattle: IASP Press, 1996;113–121. 5. Cervero F, Laird JMA. Visceral pain. Lancet 1999;353:2145–2148. 6. Portenoy RK, Conn M. Cancer pain syndromes. In : Cancer Pain, Assessment and Management, Bruera E, Portenoy RK, eds. New York: Cambridge University Press, 2003;38–50. 7. Portenoy RK, Foley KM, Inturrisi CE. The nature of opioid responsiveness and its implications for neuropathic pain : new hypotheses derived from studies of opioid infusions. Pain 1990;43:273–286. 8. Caraceni A, Brunelli C, Martini C et al. Cancer pain assessment in clinical trials: a review of the literature (1999–2002). J Pain Symptom Manage. 2005;29:507–519. 9. Serlin RC, Mendoza TR, Nakamura Y et al. When is cancer pain mild, moderate, or severe? Grading pain severity by its interference with function. Pain. 1995;61:277–284. 10. Mercadante S, Villari P, Ferrera P et al. Opioid-induced or pain relief–reduced symptoms in advanced cancer patients? Eur J Pain. 2006;10:153–159. 11. World Health Organization. Cancer Pain Relief and Palliative Care. Geneva: WHO 1990. 12. Jadad AR, Browman GP. The WHO analgesic ladder for cancer pain management. JAMA 1995;274:1870–1873.

128

Part III / Neurologic Symptoms

13. Mercadante S. Pain treatment and outcome in advanced cancer patients followed at home. Cancer 1999, 85:1849–1858. 14. Eisenberg E, Berkey C, Carr DB et al. Efficacy and safety of nonsteroidal antinflammatory drugs for cancer pain: a meta-analysis. J Clin Oncol 1994;12:2756–2765. 15. Mercadante S, Fulfaro F, Casuccio A. A randomised controlled study on the use of anti-inflammatory drugs in patients with cancer on morphine therapy: effect on dose-escalation and pharmacoeconomic analysis. Eur J Cancer 2002;38:1358–1363. 16. Mercadante S, Porzio G, Ferrera P et al. Low morphine doses in opioid-naive cancer patients with pain. J Pain Symptom Manage 2006;31:242–247. 17. Mercadante S, Arcuri E. Opioids and renal function. J Pain. 2004;5:2–19. 18. Hanks GW and Expert Working Group of the Research Network of the EAPC. Morphine and alternative opioids in cancer pain: the EAPC recommendations. Br J Cancer 2001;84:587–593. 19. Mercadante S, Villari P, Ferrera P et al. Rapid titration with intravenous morphine for severe cancer pain and immediate oral conversion. Cancer 2002;95:203–208. 20. Nelson KA, Glare PA, Walsh D et al. A prospective, within-patient, crossover study of continuous intravenous and subcutaneous morphine for chronic cancer pain. J Pain Symptom Manage. 1997;13:262–267. 21. Mercadante S, Fulfaro F. Alternatives to oral morphine in cancer pain. Oncology. 1999;13:215–225. 22. Glare P, Walsh D, Sheehan D. The adverse effects of morphine: a prospective survey of common symptoms during repeated dosing for chronic cancer pain. Am J Hosp Palliat Care. 2006;23:229–235 23. Cherny N, Ripamonti C, Pereira J et al. Strategies to manage the adverse effects of oral morphine: an evidence-based report. J Clin Oncol. 2001;19:2542–2554. 24. Mercadante S. Opioid rotation in cancer pain. rationale and clinical aspects. Cancer 1999;86:1856–1866. 25. Bruera E, Pereira J, Watanabe S et al. Opioid rotation in patients with cancer pain: a retrospective comparison of dose ratios between methadone, hydromorphone, and morphine. Cancer 1996;78;852–857. 26. Lawlor P, Turner K, Hanson J et al. Dose ratio between morphine and methadone in patients with cancer pain: a retrospective study. Cancer 1998;82 :1167–1173. 27. Ripamonti C, Groff L, Brunelli C et al. Switching from morphine to oral methadone in treating cancer pain: what is the equianalgesic dose ratio? J Clin Oncol, 1998;16:3216–3221. 28. Morley JS, Makin MK. Comments on Ripamonti et al. Pain, 1997;70:109–115. 29. Mercadante S, Casuccio A, Calderone L. Rapid switching from morphine to methadone in cancer patients with poor response to morphine. J Clin Oncol 1999;17:3307–3312. 30. Donner B, Zenz M, Tryba M et al. Direct conversion from oral morphine to transdermal fentanyl: a multicenter study in patients with cancer pain. Pain 1996;64:527–534. 31. Portenoy RK, Hagen NA. Breakthrough pain: definition, prevalence, and characteristics. Pain 1990;41:273–281. 32. Portenoy RK, Payne R, Coluzzi P et al. Oral transmucosal fentanyl citrate (OTFC) for the treatment of breakthrough pain in cancer patients: a controlled-dose titration study. Pain 1999;79:303–312. 33. Christie JM, Simmonds M, Patt R et al. Dose-titration, multicenter study of oral transmucosal fentanyl citrate for the treatment of breakthrough pain in cancer patients using transdermal fentanyl for persistent pain. J Clin Oncol 1998;16:3238–3245. 34. Farrar JT, Cleary J, Rauck R et al. Oral transmucosal fentanyl citrate: randomised, double-blinded, placebo-controlled trial for treatment of breakthrough pain in cancer patients. J Natl Cancer Inst 1998;90:611–616. 35. Portenoy RK, Rowe G. Adjuvant analgesic drugs. In: Cancer Pain, Assessment and Management, Bruera E, Portenoy RK, eds. New York: Cambridge University Press 2003;188–198. 36. Caraceni A, Zecca E, Bonezzi C et al. Gabapentin for neuropathic cancer pain: a randomized controlled trial from the Gabapentin Cancer Pain Study Group. J Clin Oncol. 2004;22:2909–2917. 37. Mercadante S. Malignant bone pain: physiopathology, assessment and treatment. Pain, 1997;69:1–18. 38. Mercadante S. Problems of long-term spinal opioid treatment in advanced cancer patients. Pain 1999;79:1–13. 39. Tasker RR. Neurosurgical and neuroaugmentative intervention. In: Patt RB, ed. Cancer Pain. Philadelphia: JB Lippincott Co, 1993;471–500. 40. Plancarte R, de Leon-Casasola OA, El–Helaly M et al. Neurolytic superior hypogastric plexus block for chronic pelvic pain associated with cancer. Reg Anesth, 1997;22:562–568. 41. Mercadante S, Nicosia F. Celiac plexus block: a reappraisal. Reg Anesth Pain Med 1998;21:407–413.

IV

Direct Complications of Cancer

9

Brain Metastases Ahmir H. Khan

MD, PhD,

and Lawrence Recht,

MD

CONTENTS Epidemiology Pathophysiology Presentation and Diagnosis Prevention of Brain Metastases Treatment of Brain Metastases Conclusions and Recommendations References

Summary Brain metastases are increasing in frequency as patients with cancer live longer as a result of improved local control of disease. Surgery for accessible lesions and either focused or whole-brain radiation therapy are currently the most utilized treatment modalities. For patients with good prognostic factors and a single metastasis, surgical resection or focused radiosurgery with or without whole-brain irradiation is recommended. However, the management of patients with multiple metastases, poor prognostic factors, or unresectable lesions is controversial. The role for chemotherapeutics that cross the blood–brain barrier and for novel targeted molecular agents remains to be elucidated. Key Words: brain metastases, stereotactic radiosurgery, whole-brain irradiation, cancer palliative treatments

1. EPIDEMIOLOGY Approximately 170,000 new cases of intracranial metastasis (ICMs) occur in the United States each year, making it by far the most common type of brain tumor (1). One recent retrospective population-based study from the Netherlands found that 8.5% of a cohort of over 2,500 cancer patients developed brain metastases over the course of their disease (2). This incidence was highest in small cell lung carcinoma (SCLC) (29.7%), followed by non-small cell lung carcinoma (NSCLC) (12.6%), renal cell carcinoma (9.8%), melanoma (7.4%), breast cancer (5.0%), and colorectal carcinoma (1.2%). Conversely, the most common identified primary sites that give rise to ICMs are (in order of decreasing frequency) lung (40–60% of all ICM), followed by breast, melanoma, colorectal, genitourinary tract, and sarcoma. Despite the availability of modern imaging techniques, however, the source of newly identified brain metastases still remained unclear in 26% of cases in a recent series (3). A recent study of patients presenting with ICM as the initial manifestation of malignancy found that 82% had lung carcinomas. The combination of brain MR and chest CT scan identified an appropriate biopsy site in almost all patients (4). From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

131

132

Part IV / Direct Complications of Cancer

2. PATHOPHYSIOLOGY Brain metastases spread hematogenously via the arterial circulation, settling in the “watershed” areas between vascular distributions in the brain (e.g., between posterior cerebral and middle cerebral arterial territories). As the tumor satellite cells pass through the circulation, they aggregate in capillaries that lie at the gray–white junction. This is not a random process, but rather involves a complex interplay between cell surface molecules from the tumor cell interacting with the endothelium and eventually leads to its extravasation into the brain parenchyma. Aberrant vascular proliferation induced by the new metastases leads to a permeable blood–brain barrier that allows for edema to surround the growing tumor mass and contrast agents to highlight the tumor (5). Because brain metastases are distributed relative to blood flow, 80% of ICMs are found in the cerebral hemispheres, 15% in the cerebellum, and only 5% in the brainstem (6). For unclear reasons, pelvic (prostate and uterus) and gastrointestinal tumors have a predilection to metastasize to the posterior fossa (6).

3. PRESENTATION AND DIAGNOSIS While up to one-third of patients with brain metastases may be asymptomatic, most patients have symptoms that can be split as generalized and focal categories. Generalized symptoms often relate to increased intracranial pressure and vasogenic edema, and include headaches, lethargy, nausea, vomiting, and confusion. Focal symptoms and signs relate to localized neurologic dysfunction from tissue destruction or compression; common examples are hemiparesis, visual field deficits, seizures (see below), aphasia, and ataxia. Metastases from certain tumor types, typically melanoma, choriocarcinoma, renal cell, and thyroid carcinoma, have a propensity for intratumoral bleeding, which may result in a more serious or acute neurologic presentation including stroke, coma, and death. Brain imaging typically is diagnostic. A noncontrast head CT usually reveals a hypodense lesion at the gray– white junction, often with surrounding edema, mass effect, or hemorrhage. Further imaging with contrast CT or MRI will reveal a ring-enhancing lesion. Unenhanced T2-weighted MRI is as sensitive as contrast-enhanced CT, and the addition of gadolinium contrast to MRI renders MRI considerably superior to CT. One study of patients with single brain metastases on contrast CT scans found that 31% had multiple metastases with gadoliniumenhanced MRI (7). With the use of MRI, roughly 75% of patients with ICM are found to have multiple ICMs. CT scan is particularly unreliable for posterior fossa metastases. In 80% of cases, metastases are detected in patients with a history of a known primary tumor (8). Patients with untreated brain metastases have an astonishingly poor prognosis. Untreated, the median survival is about one month, and death is usually due to the complications of the metastases itself (8). However, various treatment modalities can significantly improve both quantity and quality of life. The current availability of a number of high-quality randomized controlled studies (RCTs) allow the clinician to construct evidence-based treatment guidelines for many clinically relevant situations.

4. PREVENTION OF BRAIN METASTASES Certain cancers have a particularly high rate of cerebral metastases. For example, the incidence of cerebral metastases is as high as 50% at 2 years in SCLC patients receiving systemic treatment only. This alarming rate of CNS failure has prompted a number of studies assessing the value of prophylactic cranial irradiation (PCI). A meta-analysis of 7 RCTs in patients with histologically diagnosed SCLC and a complete response to induction therapy examined patients randomized to PCI vs. observation and revealed a significant absolute 5.4% increase in survival at 3 years, an 8.8% absolute increase in disease-free survival at 3 years, and a 54% relative reduction in the incidence of new brain metastases with PCI treatment (9). Therefore, PCI in SCLC not only delays but also likely prevents the development of ICMs. The benefits of PCI must be weighed against the potential risks, particularly since higher doses of PCI protect better against ICM development. The principal long-term concern is the development of delayed leukoencephalopathy. This risk is relatively low with the small fraction sizes typically used with PCI (2 Gy/day to a total dose of 30–36 Gy), thus justifying the routine use of PCI in the treatment of SCLC. A recent study additionally supports the use of PCI in patients with extensive SCLC at initial diagnosis with any response to chemotherapy

Chapter 9 / Brain Metastases

133

(i.e., even those patients with less than a complete response). In this population, PCI significantly decreased the risk of developing symptomatic brain metastases and improved disease-free and overall survival (10). In contrast, three RCTs investigating the value of PCI for patients with non-small cell lung cancers (NSCLC) demonstrate no improvement in survival, although there was evidence of a significant decrease in the incidence of ICMs (11). It is unclear whether this difference in outcome between the two tumor types reflects a better response of chemotherapy in SCLC, the higher likelihood that patients with NSCLC have more advanced cancer, or other factors. At the time of this writing, however, the use of PCI in NCSLC remains uncertain and the evidence does not rise to a level whereby it should be considered standard of care. At present, there are no RCTs examining ICM prophylaxis from other cancer types and such an approach cannot be advocated strongly for any cancer other than SCLC.

5. TREATMENT OF BRAIN METASTASES Optimal management of ICMs involves both symptomatic management of neurologic dysfunction as well as more definitive treatment intended to reduce the size of or eradicate the metastases itself. The symptomatic treatments are titrated to their effectiveness in dealing with the symptoms at hand, and given that they do not alter the course of the ICM, they have no effect on long-term prognosis. Definitive treatment, on the other hand, is intended to eradicate CNS cancer, thus improving patient survival as well as symptoms. In the evaluation of treatment for ICMs, several primary endpoints must be considered. Thus, while survival is the prototypical primary endpoint in many studies examining definitive treatment, it is affected by many factors other than the response of the ICM to treatment. For example, age, pretreatment performance status, the degree of extracranial disease, and/or response to prior treatment may impact survival, thus potentially obscuring a positive response to therapy. To counter this, recent studies have used alternative primary outcomes such as time to neurological progression to measure therapeutic efficacy and have been better able to detect differences in outcome between treatment groups (12,13).

5.1. Symptomatic Treatment The direct effect of ICMs on brain function worsens quality of life through the creation of new neurologic deficits, including hemiparesis, aphasia, headache, or seizures. Symptomatic treatment may provide relief for the patient in the short term, but will have no effect on survival or prevent progression of the disease. However, its utility cannot be underestimated and for many patients, especially those with advanced disease, may be the only therapy that produces any clinical benefit (14). 5.1.1. Seizures Approximately 20% of ICM patients present with seizures and another 10% will develop seizures later in their course (15). Prevention of further seizures is therefore an important goal of symptomatic therapies and usually requires administration of one or more antiepileptic drugs (AEDs) (1). The choice of agent is particularly important in the cancer patient as the frequency of side effects is increased. For instance, the incidence of serious rash including erythemia multiforme and Stevens–Johnson syndrome is higher in cancer patients placed on phenytoin or phenobarbital, an effect that may be enhanced with concurrent radiation or chemotherapy (16). Additionally, the conventionally used AEDs (phenytoin, phenobarbital, and carbamazepine) induce the cytochrome p450 system, which affects the drug levels of both dexamethasone and chemotherapeutic agents (14). AEDs with fewer side effects and less of an effect on hepatic metabolism including lamotrigine, topiramate, gabapentin, and levitracetam seem more appropriate for these patients. Although a formal systematic analysis has not been carried out to distinguish which agent best prevents further seizures in these patients, we prefer levetiratacem as a first line therapy because of its lack of effect on the p450 system and excellent side effect profile (17). Although AEDs decrease the frequency of subsequent seizures in patients with ICMs, several trials of prophylactic administration of AEDs failed to demonstrate a reduction in frequency of new-onset seizure in these patients (15,18,19). Therefore, especially in light of their potential side effects, the prescription of prophylactic AEDs is not necessary in patients with newly discovered brain metastases without clinical evidence for seizures

134

Part IV / Direct Complications of Cancer

(20). Nevertheless, several investigators still recommend prophylactic AEDs be prescribed in certain high-risk situations such as the presence of multiple, hemorrhagic metastases. 5.1.2. Cerebral Edema ICMs increase brain volume both by their additional mass as well as the induction of vasogenic edema and lead to increased intracranial pressure; if untreated, this may eventually result in fatal consequences. Symptoms of mass effect induced by ICMs include aphasia, weakness, somnolence, seizures and coma, and the presence of increased intracranial pressure constitutes a neuro-oncologic emergency. Acute treatment of brain edema is the same as in other conditions and includes measures to reduce brain volume including surgery, hyperventilation, and osmotic diuretics. Cerebral edema from ICMs is somewhat distinguished from other causes of brain edema by its relative sensitivity to corticosteroid administration (14). Dexamethasone, a potent glucocorticoid given at a dosage of 4–16 mg daily, is often quite effective at alleviating symptoms. This can be lifesaving until definitive treatments are administered and the effect persists for months. Unfortunately, it also has several deleterious side effects including personality changes (as severe as the induction of psychosis), elevated blood sugar, weight gain, thinning of the skin, gastrointestinal bleeding, increased susceptibility to infections, and osteopenia, the combination of which can completely counterbalance its ameliorative effects (21). Therefore, the benefits must carefully be weighed against the risks, especially with long-term treatment. Nevertheless, although the search continues for a less toxic medication that is equally efficacious, steroids remain the best treatment of chronic brain edema from ICMs. One randomized study found that dexamethasone doses of 4 mg daily were as effective as doses of 16 mg daily in patients not at risk of impending herniation and carried fewer side effects (22). Preliminary data with agents blocking vascular endothelial growth factor and its receptors may be salutary in management of peritumoral and radiation-associated edema (23–26). Additionally, studies of corticorelin for vasogenic edema are ongoing. This topic is discussed further in Chapter 4.

5.2. Definitive Therapy The goal of definitive therapy is to improve survival and control symptoms via cytoreductive therapies. Because several important RCTs (Table 1) have been completed that address various specific treatment aspects, data are now available from which to formulate evidence-based recommendations clarifying the use of surgery and the various forms of radiation therapy (i.e., whole-brain and focused radiosurgery). Furthermore, evidence is mounting (although not quite to the level of these primary treatments) that there may also be a role for chemotherapy or radiation enhancers in certain situations. 5.2.1. Whole-Brain Radiation Therapy (WBRT) Radiation to the entire brain provides coverage against surgically inaccessible metastases, metastases involving eloquent cortex, multiple metastases, and micrometastases that remain undetectable with current techniques. This may afford definitive symptom relief, with the capability of controlling or even reversing neurologic dysfunction caused by the mass and its edema (Fig. 1). For these reasons, it forms an important treatment modality and can be used either alone or in association with other more focused therapies. Without definitive treatment, the median survival after diagnosis of brain metastases is 1 month. While there are no RCTs comparing WBRT to placebo, WBRT is the standard of care for palliative treatment of multiple brain metastases based on a number of early important studies performed by the RTOG in the 1970s that established survival could be improved, albeit modestly, from 2.4 to 4.8 months. Furthermore, these studies demonstrated that this improvement was for the most part independent of the particular fractionation scheme applied (27). Therefore, in large part to minimize treatment time, the typical regimen for WBRT of cerebral metastases is one that delivers a total dose of 30 Gy in ten 3 Gy fractions. In addition to demonstrating the benefits of WBRT, these studies also provided the basis from which to develop a useful classification schema that allows patients to be grouped according to their prognosis (28). This Recursive Partitioning Analysis (RPA) examined a number of disease characteristics and determined that age, Karnofsky Performance Score (KPS), and the presence of extracranial metastases allowed patients to be subdivided into three categories based on their median survival.

Chapter 9 / Brain Metastases

135

Table 1 Summary Slide of RCTs Comparing WBRT, SRS, and Surgery for Patients with ICMs Number and Characteristics of Study Particpants

Primary Endpoint

Test Group

Standard

Result

Reference

Surgery + WBRT vs WBRT alone 48 patients with a single brain metastasis, KPS > 70

Median survival

40 weeks

15 weeks

Significant (p < 0.01)

34

66 patients with a single brain metastasis, estimated life expectancy of 6 months, WHO performance status < 2 84 patients with a single brain metastasis

Median survival

10 months

6 months

Significant (p = 0.04)

35

Median survival

5.6 months

6.3 months

Not significant (p = 0.24)

37

Median survival

11 months

7.5 months

Not significant (p = 0.22)

68

Median survival

5.7 months

6.5 months

Not signficant (p = 0.14)

38

Median survival

7.5 months

8.0 months

Not significant (p = 0.42)

43

SRS + WBRT vs WBRT alone 27 patients with 2–4 brain metastases, KPS > 70, known histology 333 patients with 1–3 brain metastases, KPS > 70 WBRT + SRS vs SRS alone 132 patients with 1–4 brain metastases

The RPA class I patients, defined as those with a KPS ≥ 70 (i.e., capable of handling daily activities of living), age less than 65 years, no extracranial disease, and controlled primary tumor had a median survival of 7.1 months with WBRT. An RPA class III, defined as a KPS < 70 on the other hand, had a much worse median survival of 2.3 months. All other patients not in class I or III were defined as class II and had a median survival of 4.2 months. This analysis has proven very useful for investigators interested in treatment of ICMs because it allows prognostic grouping based on some very simple quantifiable factors, even though the majority (64%) of patients will still fall into the broad and loosely defined class II category. The most serious complications of WBRT occur anywhere from six months to decades after treatment, and manifest clinically as some degree of cognitive impairment, ranging from mild to severe dementia. Damage to the vascular endothelium and glia, manifesting pathologically as endothelial abnormalities (including fibrinoid necrosis) and demyelination, causes these changes. Retrospective studies suggest that dementia occurs in over 10% of patients with metastatic brain cancer controlled with WBRT, and at least some cognitive impairment may be present in anywhere from 20% to 80% of patients (29). Risk factors for progression include older age (> 55 years) and higher doses of radiation. This topic is discussed comprehensively in Chapter 16. To minimize the risk of delayed neurotoxicity, we commonly treat patients with good prognosis with relatively small fractions over a longer period (e.g., 20 fractions of 200 cGy over four weeks). WBRT alone provides local and distant CNS control in approximately 70–90% of cancer patients with ICMs (30). It remains the best treatment for patients with more than four brain metastases or for whom progressive disease is anticipated to markedly curtail survival. This would include all patients in the RPA class III, as they may benefit more from the amelioration of their symptoms rather than the relatively small increase in survival. Furthermore, although fears about its long-term side effects, especially cognitive decline, has led to its deferment

136

Part IV / Direct Complications of Cancer

Fig. 1. (a) A T1-weighted, gadolinium-enhanced image of a patient with breast cancer and brain metastases. A larger left occipital lesion is accompanied by two bilateral perisylvian lesions (arrows). (b) After WBRT, this T1-weighted, gadolinium-enhanced image of the same patient reveals a reduction in the size of the left occipital lesion, while the two perisylvian lesions have disappeared.

in good prognosis patients, a recent prospective study of 26 patients that noted only minimal dysfunction six years after WBRT suggests that the likelihood of this impairment may not be as high as previously reported (31). Recent reports suggest that tumor recurrence is by far the more frequent cause of cognitive decline in ICM patients (32). Therefore, we believe that its administration is still warranted for most ICM patients at this time, either alone or as an adjunct to more focused treatment. An exception may be patients with oligometastatic brain metastases from highly radioresistant primaries such as melanoma and renal cell carcinoma, in which the benefits of fractionated radiotherapy may be particularly modest at best. 5.2.2. Surgical Resection Plus WBRT Surgical removal of ICMs is often utilized to immediately decrease tumor burden. While there are reports of removing multiple metastases (33), in most situations its use is probably better restricted to the instance where there is only one detectable mass or when one mass is dominant. Its addition to WBRT has been demonstrated to improve survival in a selected patient population based on the results of three RCTs. The first RCT compared surgery plus WBRT to WBRT alone in a total of 48 patients with a single brain metastasis and KPS > 70 and showed both a significant increase in survival (40 weeks vs. 15 weeks) as well as a significant improvement in neurologic function (median time with KPS > 70 of 38 weeks vs. 8 weeks for controls) (34). A second study performed by a Dutch group made a similar comparison in 63 patients with a single brain metastasis and drew a similar conclusion, demonstrating a significant increase in survival (10 months with both treatments vs. 6 months

Chapter 9 / Brain Metastases

137

with a WBRT alone). This study also noted that those with inactive extracranial disease had a longer survival regardless of treatment (12 months with surgery, 7 months with WBRT only) while those with active disease displayed equivalent survival (5 months) regardless of treatment (35,36). Similar to the RPA analysis, age < 60 was a significant predictor of improved survival. A third Canadian study examined the same question in an RCT but did not reach the same conclusions (37). This study randomized 84 patients with a single brain metastasis to surgery alone or surgery with WBRT and showed no significant difference in survival (5.6 months with WBRT only vs. 6.3 month in patients receiving both treatments). In this study, even neurologic function measured by median days with a KPS higher than 70 did not differ between the two treatments. While it is not clear what accounts for the differences in the results among these three studies, the patient populations did differ. Seventy-four percent of the patients in the Canadian study group had uncontrolled primary disease or extracranial metastases, compared to 32% of the patients in the Dutch study group. The results of these studies suggest that good-prognosis patients with a single, resectable ICM benefit from surgical removal followed by WBRT. This is in retrospect what one would infer from the RPA analysis, namely that those with inactive extracranial disease, age < 60 and good neurologic function will survive long enough to benefit from surgery. Patients with uncontrolled primary disease have poorer survival outcomes regardless of the treatment given, as they will likely die from the consequences of their extracranial disease. 5.2.3. Stereotactic Radiosurgery Stereotactic radiosurgery (SRS) represents a noninvasive alternative to surgery, providing local control of brain metastases without the morbidity of surgical treatment. SRS has the potential advantage of providing direct, specific control of metastases in patients who are in categories that do not clearly benefit from surgical resection, including those with multiple metastases, active extracranial disease and age > 60. Additionally, SRS also allows for treatment of brain metastases that are not surgically accessible (Fig. 2). Stereotactic radiosurgery uses multiple noncoplanar beams that intersect within the target to produce highly conformal dose distribution while reducing the dose to normal tissues. Traditional SRS systems require a stereotactic head frame both to create a three-dimensional coordinate system and to provide rigid fixation for precise targeting. A high-resolution MR or CT scan is performed with the head frame on for target localization. The two most common systems for delivering SRS are the Gamma Knife and linear accelerator (Linac). The former consists of a helmet containing 201 radioactive cobalt sources focused at the center of the helmet; by varying head position within the helmet as well as collimators the delivery of a high dose of localized radiation with rapid drop-off is achieved. Linac-based systems utilize high-energy photons; beams are shaped and the linear accelerator moved while the patient’s head remains immobilized to achieve the desired dosimetry. Several commerical Linac-based radiosurgery systems have been devised. One of the more popular ones is the CyberKnife® , which incorporates a linear accelerator on a robotic arm and incorporates image guidance technology to deliver radiosurgery to metastases in the brain and elsewhere in the body. SRS provides the ability to deliver higher doses of radiation to a targeted lesion, thereby theoretically minimizing the effect of radiation on surrounding normal brain tissue. It is an outpatient procedure that can be performed without the cost and morbidity of surgery. Complications from SRS are generally limited to acute edema that can develop almost immediately after SRS ablation of a lesion. SRS has gained wide usage in most centers that take care of patients with ICMs and its efficacy has been assumed, but not proven, to be at least equivalent to surgery. It seems unlikely therefore that a RCT will ever be done comparing surgery with SRS. The role of SRS as an adjunct to WBRT for has been examined in a large RCT (RTOG 9508). In this prospective study, 333 patients with 1–3 brain metastases and a known primary were randomized to SRS plus WBRT or WBRT alone (38). No significant survival advantage was noted with the addition of SRS to WBRT (5.7 months with WBRT and SRS vs. 6.5 months with WBRT alone). There was, however, a slight improvement in KPS status, and a significant decrease in the amount of steroids used in patients with the addition of SRS. In contrast to the entire cohort, subgroup analysis did reveal a significantly improved survival (6.5 months) with SRS vs. WBRT alone (4.9 months) in patients with a single metastasis. Although these differences can hardly be considered substantial, the authors concluded that the combination was superior to WBRT alone in this subgroup of ICM patients. Additionally, multivariate analysis revealed a significant survival benefit in RPA class I patients

138

Part IV / Direct Complications of Cancer

Fig. 2. (a) A T1-weighted, gadolinium-enhanced image of a patient with renal cell carncinoma and a single left occipital metastases. (b) After SRS, this T1-weighted, gadolinium-enhanced image reveals complete radiographic resolution of the metastases.

(Karnofsky performance score ≥ 70, age < 65, a controlled primary tumor, and no other sites of metastatic disease). Because SRS is relatively easy to administer repeated times, a practice option has emerged in which SRS is administered without WBRT. Although anecdotally stated that this is safe and effective, whether this approach appropriately balances the risk of missing micrometastases with the prevention of the long-term deleterious effects of WBRT (which are presumed to be worse than SRS) is unknown. This risk and benefit assessment is further emphasized in light of an earlier RCT that noted a significant increase in distant CNS failure in post-surgical patients in whom WBRT was deferred (39). Proponents of WBRT note that distant CNS recurrence is often symptomatic and impairs quality of life (40). Additionally, tumor progression has a larger negative impact on neurocognitive function than does WBRT (32). Thus, attempts to minimize distant brain failure are worthwhile. Moreover, they note that radiation-associated dementia has principally been associated with treatment schedules that administered radiation in daily fractions > 300 cGy (41); such schedules are presently avoided. Proponents of SRS alone for the management of oligometastatic brain disease argue that with surveillance imaging new brain metastases are generally detectable in the presymptomatic phase. In addition to avoiding the risk of delayed leukoencephalopathy, the avoidance of fatigue and hair loss associated with WBRT are advantages for SRS. Furthermore, despite a dearth of studies following the neurocognitive function in patients treated with SRS, pilot data suggest that successful local control of ICM with SRS is associated with stable to improved neurocognitive function (42).

Chapter 9 / Brain Metastases

139

A recent Japanese study addressed this question by randomizing 132 patients with 1–4 brain metastases to treatment with SRS and WBRT or SRS only. Median survival with either treatment was not significantly different (7.5 months with both treatments, 8 months with SRS only), even though the addition of WBRT did reduce distant CNS recurrence (46.8% of patients receiving both therapies with recurrence at 12 months vs. 76.4% with SRS) (43). As with other studies, stable extracranial disease, high KPS and fewer brain metastases were predictive of an improved survival. Neurologic function was no different between the groups when measured by KPS, rates of neurologic preservation at 12 months, and death by neurologic cause. Although the authors concluded that patients with 1–4 ICMs treated with SRS may not benefit from additional WBRT, an alternative conclusion that the high rate of distant failures warrants its inclusion in the treatment plan seems equally valid, especially if CNS recurrence is associated with cognitive decline (as has been suggested by other studies (32)). An ongoing phase III study in the United States through the North Central Cancer Treatment Group is currently randomizing patients with 1–3 brain metastases to SRS ± WBRT. This study, NCCTG N0574, has extensive neurocognitive and quality-of-life assessments built in and, when completed, may provide further insight into this contentious topic. 5.2.4. Radiation Enhancers The poor outlook for patients with ICMs treated with conventional measures has spurred interest in discovering ways of enhancing treatment response. Although many agents have been tested, the results are for the most part negative. Recent work on two agents holds potential promise, however. Efaproxiral (RSR13) is an allosteric modifier of hemoglobin that reduces oxygen affinity, thus improving its release and the effectiveness of radiation on killing malignant cells. A phase III study of 538 RPA class I or II patients with brain metastases and inactive extracranial disease revealed no difference in the mean survival time for patients treated with efaproxiral over WBRT (5.4 vs. 4.4 months) (4.4 months). However, subgroup analysis did suggest a benefit in patients with metastases from breast cancer, as patients receiving efaproxiral had a median survival of 8.7 months compared to 4.6 months with WBRT alone (44). Efaproxiral is currently undergoing further evaluation in a phase III study limited to patients with ICM from breast cancer. Another sensitizer that has spurred recent interest is motexafin gadolinium (MGd), an intravenously administered metalloporphyrin taken up preferentially by tumor cells. It catalyzes the oxidation of intracellular reducing metabolites, thereby generating reactive oxygen species that make tumor cells more susceptible to radiation. A phase III study randomizing patients with ICM to WBRT ± MGd failed to demonstrate improved survival or time to neurologic progression (TTNP) with MGd, though subgroup analysis demonstrated that patients with NSCLC had improved TTNP (45). A follow-up phase III trial restricted to patients with NSCLC again failed to meet the primary endpoint of prolonged TTNP, although patients initiating treatment within three weeks of ICM diagnosis (which occurred more commonly in North America than other sites) did have significantly prolonged TTNP with the addition of MGd (12). As such, FDA approval and future development of MGd remain uncertain. In a paradoxical twist to their original intent, this agent may also hold promise as a neuroprotective agent against the effects of radiation or radiation-induced tumor death on normal brain. These studies underscore a fundamental difficulty for clinical trials of brain metastases: the great majority of patients succumb to systemic disease. Thus, the traditional historic endpoint of overall survival may be unsatisfactory for demonstrating patient benefit. Successful future trials will necessarily incorporate robust measures of neurocognitive function and quality of life. 5.2.5. Chemotherapy The use of chemotherapy to treat ICM has both obvious appeal and drawbacks (46,47). Most patients with ICM have active systemic tumor, which for many patients will be the ultimate cause of death. Thus, an approach to target both ICM and systemic cancer is desirable. Approximately 75% of patients with ICM have multiple brain metastases; like WBRT, systemic chemotherapy provides a global strategy for intracranial tumor control. Moreover, while potential subtle neurocognitive sequelae of common systemic chemotherapies are debated, these are arguably insignificant compared to those associated with WBRT (48). The obvious hurdles for chemotherapy include the blood–brain barrier and the lack of durably effective systemic chemotherapies for the common systemic tumors that produce ICM. Tight intercellular junctions limit the passage

140

Part IV / Direct Complications of Cancer

of drugs into brain parenchyma, though small lipophilic agents do cross. Additionally, carrier mechanisms like P-glycoprotein help to exclude agents including taxanes, anthracyclines, vinca alkaloids, and podophyllotoxins. However, the fact that by the time they are visible on MR scan metastases virtually always enhance with intravenous contrast indicates that the barrier is leaky when tumors reach a certain size. Furthermore, analysis of brain tumors resected following exposure to systemic chemotherapy often demonstrates high concentrations of chemotherapy that would not have been predicted based on CSF/plasma ratios. Temozolomide, nitrosoureas, irinotecan, and topotecan all cross the intact blood–brain barrier well, while high doses of methotrexate, cytarabine, idarubicin, and etoposide achieve tumoricidal CSF levels. Nonetheless, the development of brain metastases in patients with SCLC who are responding to systemic chemotherapy or breast cancer receiving trastuzumab or paclitaxel highlights the role of the blood–brain barrier in providing a sanctuary for micrometastases to develop into clinically detectable tumors. Numerous published reports have assessed the role of chemotherapy in ICM; however, these reports are often retrospective, contain small patient numbers, or include patients heterogenous in tumor type and prior therapies. These factors, as well as variability of methodologies, render interpretation difficult. This topic has been recently reviewed (49). For NSCLC, generally resistant to systemic chemotherapy, platinum-based regimens may be active as initial therapy but are less often active in patients progressing post-radiotherapy. The oral methylating agent temozolomide has been extensively evaluated. Single-agent response rates are < 10%, but the combination of temozolomide with fractionated radiotherapy is promising. A phase III study randomizing NSCLC patients with ICM to RT ± temozolomide reported improved radiographic response rate (53% vs. 33%) and a suggestion of improved median survival (8.3 vs 6.3 months), although these results remain available only in abstract form (50). Another randomized phase II trial similarly randomized patients with ICM from a variety of tumor types, about half of whom had NSCLC, and reported improved 3-month tumor control rates and neurologic survival despite no improvement in response rates or overall survival from the addition of temozolomide (51). Further studies are required to determine the contribution of temozolomide to whole-brain RT in these patients. Small-molecule tyrosine kinase inhibitors targeting the EGF receptor, such as erlotinib and gefitinib, have systemic activity in selected patients with NSCLC and have been studied in ICM. These agents are known to have higher systemic activity in Asian populations, who are more likely to harbor a favorable EGFR mutation. Several Asian studies suggest high response rates with EGFR TK inhibitors for ICM (52–54). Response rates in non-Asian patients with ICM appear to be substantially lower (55), although patients who respond systemically typically respond in the CNS as well. The development of ICM in patients responding systemically to gefitinib suggests that the intact blood–brain barrier may not allow meaningful brain levels of these agents to attack micrometastatic disease (56). Brain metastases from SCLC are often exquisitely radiosensitive, and the role of chemotherapy is typically limited to post-radiotherapy relapse. Topotecan has modest activity in this setting (57,58). Breast cancer is the second-leading cause of brain metastases and a relatively chemosensitive tumor compared to NSCLC. Thus, it is unsurprising that accumulating literature addresses the role of chemotherapy in ICM (this topic is also discussed in Chapter 22). In brief, traditional cytotoxic drugs that cross the intact blood–brain barrier poorly but with activity against systemic breast cancer are still sometimes active against breast cancer ICMs (49). In contrast, temozolomide has essentially no single-agent activity. Case reports suggest the oral 5fluorouracil prodrug capecitabine has occasional activity (59). The oral human epidermal growth factor receptor (HER-2) tyrosine kinase inhibitor lapatinib induced responses in several patients with HER-2-positive ICMs and is undergoing further evaluation (60). Melanoma represents a particularly neurotropic tumor that frequently causes brain metastases. The role of chemotherapy in melanoma ICM is exhaustively covered in Chapter 28. Modest evidence suggests that temozolomide-containing systemic regimens may help decrease the incidence of ICM (61,62); however, singleagent temozolomide rarely induces meaningful responses in established melanoma ICM (63,64). Whether temozolomide-based combinations are more effective requires further study. Germ cell tumors, while an uncommon cause of ICM, are highly chemosensitive. Case series document the responsiveness of germ cell ICM, generally in patients receiving cisplatin-based regimens. Most long-term

Chapter 9 / Brain Metastases

141

responders are patients who presented with cerebral metastases (65,66); treatment of CNS relapse in patients previously managed with platinum-based regimens is less effective (67).

6. CONCLUSIONS AND RECOMMENDATIONS The recent completion and publication of several RCTs has provided a solid basis with which to make treatment recommendations for patients with ICMs (Fig. 3). These trials all demonstrate that both prognosis and response to treatment depend on age, current neurologic function, the number of cerebral metastases, and the extent of extracranial disease, all factors that must be taken into consideration in any treatment plan. Future investigations should address quality of life and neurocognitive outcomes, in addition to traditional outcome measures such as recurrence and survival rates.

6.1. Prophylactic Treatment Recommendations Based on several RCTs, there is enough data to justify prophylactic cranial irradiation only in patients with SCLC. At the current time, although there is a suggestion that a similar approach may also be effective for NSCLC, there is no other tumor type for which this approach can be justified.

6.2. RPA Class I with a Single Brain Metastasis In patients younger than 65 with good neurologic function, no extracranial disease and a single, surgically accessible brain metastasis there is solid evidence that surgical resection, followed by WBRT, improves survival and prevents both local and distant CNS recurrences. There is a suggestion that the addition of SRS to WBRT is beneficial in patients with one metastasis based only on subgroup analysis of patients with a single metastasis. While withholding WBRT and administering SRS alone to ICMs does not adversely impact survival, it does increase the risk of distant CNS failure. At present, therefore, patients in this RPA class can be treated with either SRS or WBRT, pending further studies. When surgery is administered, subsequent WBRT is recommended.

6.3. RPA Class I with Multiple Metastases. There is no clear evidence regarding whether SRS or WBRT is superior in the treatment of multiple metastases. RCTs suggest that the use of both modalities is not superior to a single modality in terms of improving survival in good risk patients, While the addition of SRS to WBRT does not improve survival, it does decrease local recurrence. On the other hand, the addition of WBRT to SRS also does not improve survival, but decreases distant recurrence. Whether local control (i.e., SRS) is adequate depends on an as yet undetermined risk/benefit analysis of the long-term effects of WBRT versus the morbidities associated with distant CNS recurrence. It is

Fig. 3. Treatment algorithm for brain metastases.

142

Part IV / Direct Complications of Cancer

our current recommendation to combine both treatments based on the fact that tumor recurrence is much more likely to impact cognition than long-term effects of radiation.

6.4. RPA Class II These patients, who have a KPS > 70 but have either age > 65, uncontrolled primary site, or other metastatic sites, may benefit from surgical/SRS ablation of single brain metastasis, followed by WBRT. In the presence of multiple brain metastases, the situation is identical to patients in the RPA class I as described in Section 6.3.

6.5. RPA Class III These patients have a KPS < 70. Since their prognosis from systemic disease is very poor, one should aim for palliation only. In symptomatic patients, WBRT provides effective amelioration of their symptoms. Surgery is not indicated because it requires valuable time for hospitalization, but SRS can be used to treat symptomatic lesions that are refractory to treatment.

REFERENCES 1. Kuo T, Recht L. Optimizing therapy for patients with brain metastases. Semin Oncol 2006;33:299–306. 2. Schouten LJ, Rutten J, Huveneers HA et al. Incidence of brain metastases in a cohort of patients with carcinoma of the breast, colon, kidney, and lung and melanoma. Cancer 2002;94:2698–2705. 3. Agazzi S, Pampallona S, Pica A et al. The origin of brain metastases in patients with an undiagnosed primary tumour. Acta Neurochir (Wien) 2004;146:153–157. 4. Mavrakis AN, Halpern EF, Barker FG, 2nd et al. Diagnostic evaluation of patients with a brain mass as the presenting manifestation of cancer. Neurology 2005;65:908–911. 5. Gavrilovic IT, Posner JB. Brain metastases: epidemiology and pathophysiology. J Neurooncol 2005;75:5–14. 6. Delattre JY, Krol G, Thaler HT et al. Distribution of brain metastases. Arch Neurol 1988;45:741–744. 7. Schellinger PD, Meinck HM, Thron A. Diagnostic accuracy of MRI compared to CCT in patients with brain metastases. J Neurooncol 1999;44:275–281. 8. Tosoni A, Ermani M, Brandes AA. The pathogenesis and treatment of brain metastases: a comprehensive review. Crit Rev Oncol Hematol 2004;52:199–215. 9. Auperin A, Arriagada R, Pignon JP et al. Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. Prophylactic Cranial Irradiation Overview Collaborative Group. [see comments.]. 1999;341:476. 10. Slotman B, Faivre-Finn C, Kramer G et al. A randomized trial of prophylactic cranial irradiation (PCI) versus no PCI in extensive disease small cell lung cancer after a response to chemotherapy (EORTC 08993–22993). J Clin Oncol (Meeting Abstracts) 2007;25:4. 11. Gore EM. Prophylactic cranial irradiation for patients with locally advanced non-small cell lung cancer. Oncology (Williston Park) 2003;17:775–779; discussion 779–780, 784, 787 passim. 12. Mehta MP, Gervais R, Chabot P et al. Motexafin gadolinium (MGd) combined with prompt whole-brain radiation therapy (RT) prolongs time to neurologic progression in non-small cell lung cancer (NSCLC) patients with brain metastases: results of a phase III trial. J Clin Oncol 2006;24:7014. 13. Renschler MF, Smith JA, Shapiro WR. Validation of blinded events review committee (ERC): determined time to neurologic progression (TTNP) demonstrates correlation with survival, radiologic progression, and functional endpoints. In: Society for Neurooncology; 2005. 14. El Kamar FG, Posner JB. Brain metastases. Semin Neurol 2004;24:347–362. 15. Cohen N, Strauss G, Lew R et al. Should prophylactic anticonvulsants be administered to patients with newly diagnosed cerebral metastases? A retrospective analysis. J Clin Oncol 1988;6:1621. 16. Delattre JY, Safai B, Posner JB. Erythema multiforme and Stevens–Johnson syndrome in patients receiving cranial irradiation and phenytoin. Neurology 1988;38:194. 17. Newton HB, Goldlust SA, Pearl D. Retrospective analysis of the efficacy and tolerability of levetiracetam in brain tumor patients. J Neurooncol 2006;78:99–102. 18. Glantz MJ, Cole BF, Friedberg MH et al. A randomized, blinded, placebo-controlled trial of divalproex sodium prophylaxis in adults with newly diagnosed brain tumors. Neurology 1996;46:985–991. 19. Forsyth PA, Weaver S, Fulton D et al. Prophylactic anticonvulsants in patients with brain tumour. Can J Neurol Sci 2003;30:106–112. 20. Glantz MJ, Cole BF, Forsyth PA et al. Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2000;54:1886. 21. Wen PY, Schiff D, Kesari S et al. Medical management of patients with brain tumors. J Neuro-oncol 2006;80:313–332. 22. Vecht CJ, Hovestadt A, Verbiest HB et al. Dose–effect relationship of dexamethasone on Karnofsky performance in metastatic brain tumors: a randomized study of doses of 4, 8, and 16 mg per day. Neurology 1994;44:675. 23. Pope WB, Lai A, Nghiemphu P et al. MRI in patients with high–grade gliomas treated with bevacizumab and chemotherapy. Neurology 2006;66:1258–1260. 24. Vredenburgh JJ, Desjardins A, Herndon JE, 2nd et al. Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 2007;13:1253–1259.

Chapter 9 / Brain Metastases

143

25. Batchelor TT, Sorensen AG, di Tomaso E et al. AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer Cell 2007;11:83–95. 26. Gonzalez J, Kumar AJ, Conrad CA, Levin VA. Effect of bevacizumab on radiation necrosis of the brain. Int J Radiat Oncol Biol Phys 2007;67:323–326. 27. Borgelt B, Gelber R, Kramer S et al. The palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1980;6:1–9. 28. Gaspar L, Scott C, Rotman M et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Rad Oncol Biol Phys 1997;37:745. 29. Crossen JR, Garwood D, Glatstein E et al. Neurobehavioral sequelae of cranial irradiation in adults: a review of radiation-induced encephalopathy. J Clin Oncol 1994;12:627–642. 30. Hoegler D. Radiotherapy for palliation of symptoms in incurable cancer. Curr Probl Cancer 1997;21:129–183. 31. Armstrong CL, Hunter JV, Ledakis GE et al. Late cognitive and radiographic changes related to radiotherapy: initial prospective findings. Neurology 2002;59:40–48. 32. Li J, Bentzen SM, Renschler M, Mehta MP. Regression after whole-brain radiation therapy for brain metastases correlates with survival and improved neurocognitive function. J Clin Oncol 2007;25:1260–1266. 33. Bindal RK, Sawaya R, Leavens ME et al. Surgical treatment of multiple brain metastases. J Neurosurg 1993;79:210–216. 34. Patchell RA, Tibbs PA, Walsh JW et al. A randomized trial of surgery in the treatment of single metastases to the brain [see comments]. New Engl J Med 1990;322:494. 35. Noordijk EM, Vecht CJ, Haaxma-Reiche H et al. The choice of treatment of single brain metastasis should be based on extracranial tumor activity and age [see comments]. Int J Rad Onco Biol Phys 1994;29:711. 36. Vecht CJ, Haaxma-Reiche H, Noordijk EM et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993;33:583. 37. Mintz AH, Kestle J, Rathbone MP et al. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996;78:1470. 38. Andrews DW, Scott CB, Sperduto PW et al. Whole-brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363:1665–1672. 39. Patchell RA, Tibbs PA, Regine WF et al. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial [see comments.]. JAMA 1998;280:1485. 40. Regine WF, Huhn JL, Patchell RA et al. Risk of symptomatic brain tumor recurrence and neurologic deficit after radiosurgery alone in patients with newly diagnosed brain metastases: results and implications. Int J Radiat Oncol Biol Phys 2002;52:333–338. 41. DeAngelis LM, Delattre JY, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology 1989;39:789. 42. Chang EL, Wefel JS, Maor MH et al. A pilot study of neurocognitive function in patients with one to three new brain metastases initially treated with stereotactic radiosurgery alone. Neurosurgery 2007;60:277–283; discussion 283–274. 43. Aoyama H, Shirato H, Tago M et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs. stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295:2483–2491. 44. Suh JH, Stea B, Nabid A et al. Phase III study of efaproxiral as an adjunct to whole-brain radiation therapy for brain metastases. J Clin Oncol 2006;24:106–114. 45. Mehta MP, Rodrigus P, Terhaard CH et al. Survival and neurologic outcomes in a randomized trial of motexafin gadolinium and whole-brain radiation therapy in brain metastases. J Clin Oncol 2003;21:2529–2536. 46. Shaffrey ME, Mut M, Asher AL et al. Brain metastases. Curr Probl Surg 2004;41:665–741. 47. van den Bent MJ. The role of chemotherapy in brain metastases. Eur J Cancer 2003;39:2114–2120. 48. Vardy J, Rourke S, Tannock IF. Evaluation of cognitive function associated with chemotherapy: a review of published studies and recommendations for future research. J Clin Oncol 2007;25:2455–2463. 49. Cavaliere R, Schiff D. Chemotherapy and cerebral metastases: misperception or reality? Neurosurg Focus 2007;22:E6. 50. Antonadou D, Coliarakis N, Pareskavaidis M et al. Whole-brain radiotherapy alone or in combination with temzolomide for brain metastase: a phase III study. Int J Radiat Oncol Biol Phys 2002;54:93–94. 51. Verger E, Gil M, Yaya R et al. Temozolomide and concomitant whole-brain radiotherapy in patients with brain metastases: a phase II randomized trial. Int J Radiat Oncol Biol Phys 2005;61:185–191. 52. Chiu CH, Tsai CM, Chen YM et al. Gefitinib is active in patients with brain metastases from non-small cell lung cancer and response is related to skin toxicity. Lung Cancer 2005;47:129–138. 53. Hotta K, Kiura K, Ueoka H et al. Effect of gefitinib (“Iressa,” ZD1839) on brain metastases in patients with advanced non-small cell lung cancer. Lung Cancer 2004;46:255–261. 54. Namba Y, Kijima T, Yokota S et al. Gefitinib in patients with brain metastases from non-small cell lung cancer: review of 15 clinical cases. Clin Lung Cancer 2004;6:123–128. 55. Ceresoli GL, Cappuzzo F, Gregorc V et al. Gefitinib in patients with brain metastases from non-small cell lung cancer: a prospective trial. Ann Oncol 2004;15:1042–1047. 56. Omuro AM, Kris MG, Miller VA et al. High incidence of disease recurrence in the brain and leptomeninges in patients with non-small cell lung carcinoma after response to gefitinib. Cancer 2005;103:2344–2348. 57. Korfel A, Oehm C, von Pawel J et al. Response to topotecan of symptomatic brain metastases of small-cell lung cancer also after whole-brain irradiation. a multicentre phase II study. Eur J Cancer 2002;38:1724–1729. 58. von Pawel J. The role of topotecan in treating small cell lung cancer: second-line treatment. Lung Cancer 2003;41 Suppl 4:S3–8. 59. Ekenel M, Hormigo AM, Peak S et al. Capecitabine therapy of central nervous system metastases from breast cancer. J Neuro-oncol 2007;8(2):223–227.

144

Part IV / Direct Complications of Cancer

60. Lin NU, Dieras V, Paul Det al. EGF105084, a phase II study of lapatinib for brain metastases in patients (pts) with HER2+ breast cancer following trastuzumab (H)-based systemic therapy and cranial radiotherapy (RT). J Clin Oncol 2007;25:1012. 61. Paul MJ, Summers Y, Calvert AH et al. Effect of temozolomide on central nervous system relapse in patients with advanced melanoma. Melanoma Res 2002;12:175–178. 62. Weber RW, O’Day S, Rose M et al. Low-dose outpatient chemobiotherapy with temozolomide, granulocyte-macrophage colony stimulating factor, interferon-alpha2b, and recombinant interleukin-2 for the treatment of metastatic melanoma. J Clin Oncol 2005;23: 8992–9000. 63. Agarwala SS, Kirkwood JM, Gore M et al. Temozolomide for the treatment of brain metastases associated with metastatic melanoma: a phase II study. J Clin Oncol 2004;22:2101–2107. 64. Schadendorf D, Hauschild A, Ugurel S et al. Dose-intensified bi-weekly temozolomide in patients with asymptomatic brain metastases from malignant melanoma: a phase II DeCOG/ADO study. Ann Oncol 2006;17:1592–1597. 65. Kollmannsberger C, Nichols C, Bamberg M et al. First-line high-dose chemotherapy +/– radiation therapy in patients with metastatic germ-cell cancer and brain metastases. Ann Oncol 2000;11:553–559. 66. Fossa SD, Bokemeyer C, Gerl A et al. Treatment outcome of patients with brain metastases from malignant germ cell tumors. Cancer 1999;85:988–997. 67. Bokemeyer C, Nowak P, Haupt A et al. Treatment of brain metastases in patients with testicular cancer. J Clin Oncol 1997;15: 1449–1454. 68. Kondziolka D, Patel A, Lunsford LD et al. Stereotactic radiosurgery plus whole-brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys 1999;45:427–434.

10

Skull and Dural Metastases Herbert B. Newton,

MD, FAAN

CONTENTS Introduction Calvarial Metastases Skull Base Metastases Dural Metastases Conclusion Acknowledgments References

Summary Metastases of the skull base, dura, and calvarium are common complications of systemic cancer. In many patients, the tumors can remain asymptomatic until enlargement causes symptoms such as headache, lethargy, cranial nerve palsies, and focal deficits. Skull base metastases can present with specific clinical syndromes, including the orbital, parasellar, middle fossa, jugular foramen, occipital condyle, temporal bone, and sellar varieties. MRI is the most sensitive neuroimaging technique for the diagnosis of skull and dural metastases. The most common form of treatment is conventional radiation therapy, with doses ranging from 30 to 35 Gy. Surgical resection is appropriate for selected patients with accessible, solitary lesions, especially if there is mass effect. Chemotherapy may be beneficial, alone or in combination with irradiation, for patients with chemosensitive tumors. Key Words: dura, skull, metastases

1. INTRODUCTION Metastatic tumors from systemic malignancies frequently affect the central nervous system (CNS) (1). The most common clinically significant metastases are to the brain and spinal cord region, affecting between 170,000 and 200,000 new patients each year in the United States. However, metastases to the skull (calvarium, skull base) and dura have a similar, and possibly even greater, incidence, but are often more clinically silent (1,2). Due to recent improvements in the treatment of systemic tumors and longer overall survival, as well as from advances in neuro-imaging, the incidence of skull and dural metastases appears to be increasing (2). Although many metastates to the skull and dura are initially asymptomatic, and may even be found serendipitously, further enlargement will often lead to progressive neurologic morbidity. Therefore, it is important to detect these lesions while they are still small, localized, and more responsive to treatment. This chapter will review the epidemiology, clinical presentation, syndromes, neuroimaging, and treatment of metastases to the skull and dura. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

145

146

Part IV / Direct Complications of Cancer

2. CALVARIAL METASTASES The calvarium consists of fused portions of the frontal, temporal, parietal, and occipital bones, as well as the associated periosteum and soft tissue coverings, and forms the convexity of the skull (3). Embryologically, the calvarial bones develop during the sixth fetal week from primitive connective tissue membranes, unlike the bones of the skull base, which are derived from cartilage (4).

2.1. Epidemiology The incidence of calvarial metastases remains unknown. In patients without known cancer, the incidence appears to be quite low. In a series of trauma patients evaluated by x-ray or computed tomography (CT), the incidence of asymptomatic calvarial lesions was less than 0.8% (5). For patients with known cancer being staged for surgery with CT or magnetic resonance imaging (MRI), the incidence of solitary calvarial metastases was much higher – 6% to 11% (5–7). The most likely primary tumors to metastasize to the calvarium, in descending order, are from breast, lung, prostate, renal, thyroid, and melanoma.

2.2. Pathophysiology Metastatic deposits usually reach the calvarium via arterial hematogenous spread from the primary neoplasm (2). Less often, tumor cells can reach the calvarium from Batson’s venous plexus or by direct extension from regional structures (2). In the case of spread from Batson’s plexus, this can occur when venous blood is shunted rostrally due to increases in intrathoracic and intraabdominal pressure. This mechanism explains the relatively high incidence of calvarial metastases from gastrointestinal and urogenital neoplasms.

2.3. Presentation Calvarial metastases are often clinically silent and may not become symptomatic until there is significant bone destruction and enlargement of the mass (5,8). Localized pain usually occurs because of irritation to the pain-sensitive periostium. This is more likely when there is expansion of the inner and outer tables of the calvarial bones. Pain can also develop if there is outward growth, with irritation or damage to delicate structures of the scalp. At this stage the patient may notice the presence of an enlarging, tender lump under the scalp. In other cases, the mass may remain painless, but become quite large and disfiguring (9). If the mass extends inward and erodes through the inner table of the skull, it can cause direct brain and/or neurovascular compression, with secondary ischemia and edema. In this setting, symptoms include headache, seizures, and focal findings such as weakness (hemiparesis, monoparesis), sensory loss, visual field defects, and gait abnormalities. In addition, inward or lateral growth of the mass may involve the sagittal sinus or one of the lateral sinuses, causing compression and impaired venous blood flow. Resulting symptoms and signs would include increased intracranial pressure, headaches, and papilledema. Less often, tumor growth can lead to invasion of the sagittal sinus and secondary hemorrhagic infarction, with severe headache, seizure activity, focal findings (e.g., hemiparesis, dysphasia, hemianopsia), and mental status changes. In children with neuroblastoma, calvarial metastases can occur, usually in the orbito-frontal region (10). Rarely, the metastases can be more diffuse and lead to splitting of the cranial sutures and expansion of the skull, similar to the presentation of severe hydrocephalus. However, the size of the brain and ventricular system are normal. The general physical and neurologic examinations are usually normal in patients with asymptomatic calvarial metastases (5,8). The head and scalp should be examined thoroughly for any sign of a calvarial mass, and whether the mass is tender to palpation. Patients with a large mass and underlying brain compression may have focal deficits that should be documented, including hemiparesis, sensory loss, hemianopsia, dysphasia, reflex asymmetries, and gait abnormalities. If the calvarial metastasis is the result of an unknown primary tumor, a laboratory evaluation is necessary and should include a CBC, calcium and phosphate levels, sedimentation rate, CEA, serum electrophoresis, urine calcium and Bence–Jones proteins, and a hemoccult (5). In addition, all patients should undergo a screening chest x-ray and female patients should receive a mammogram.

Chapter 10 / Skull and Dural Metastases

147

2.4. Imaging Calvarial metastases can be readily detected on plain films, but are most clearly delineated using CT and MRI (5,8,11,12). Plain x-rays typically demonstrate a focal osteolytic or osteoblastic lesion that affects the inner and outer table of the involved bone. On CT, metastatic deposits are most often osteolytic, with ragged margins and an associated soft tissue mass (Fig. 1). T2-weighted MR images demonstrate a region of hyperintensity replacing the normal diploetic space and cortical bone, with an associated soft tissue component. On T1-weighted precontrast MR images, the lesions appear as a region of low signal intensity, surrounded by or replacing the signal of the normal fat-containing bone marrow. After the administration of gadolinium, calvarial metastases almost universally demonstrate dense enhancement throughout the mass. Contrast-enhanced MRI is now considered the most sensitive imaging technique for calvarial metastases, as shown in a study by West and colleagues (Fig. 2) (13). They studied 14 patients with calvarial metastases using MRI and CT, with and without contrast, and compared the findings with CT scans from 60 normal controls. Gadolinium-enhanced MRI was superior to CT with bone windows and non-enhanced MRI in the detection of calvarial lesions. In addition, MRI has the advantage of visualization of the lesion in three planes and the ability to discern involvement of the dural sinuses, including compression and occlusion. Radionuclide bone scans are also sensitive to the presence of metastases in the calvarium (8,12). However, bone scans are rarely used in the initial diagnostic evaluation of these patients.

Fig. 1. CT scan with contrast administration in a 46-year-old female patient with stage 4 breast carcinoma and right-sided head pain. In Figures 1(A) and 1(B), note the region of calvarial bone loss associated with a large soft tissue mass that extends into the nearby brain, causing regional edema. Also note the ragged edges of the damaged calvarial bone. Surgical resection of the mass revealed metastatic breast carcinoma.

148

Part IV / Direct Complications of Cancer

Fig. 2. MRI scan of the same patient as Fig. 1. (2A) Midsagittal, non-contrast T1-weighted image demonstrating the large calvarial metastasis, with an extensive extra-axial component compressing the underlying brain. (2B) Coronal, contrast enhanced T1-weighted image revealing the densely enhancing calvarial mass.

2.5. Treatment After the diagnosis of calvarial metastasis has been confirmed, the initial approach to treatment will depend on the severity of symptoms and the size and extent of involvement of the lesion (e.g., compression of a dural sinus) (5,8). Specific treatment is not required for small, asymptomatic lesions of the calvarium; clinical follow-up and serial neuroimaging will be sufficient. If the only symptom is mild to moderate pain, analgesics and biphosphonates may provide relief. For more symptomatic or extensive metastases, surgical resection is a viable option, but only in carefully selected patients. Surgery should be considered for patients with an equivocal histological diagnosis, for calvarial tumors that are rapidly enlarging and unresponsive to irradiation and/or chemotherapy, and for solitary lesions that are the only residual evidence of systemic malignancy (5). In a review of 27 patients with calvarial metastases (with and without dural sinus involvement) treated with surgical resection, Michael and colleagues concluded that surgery was a relatively safe and effective treatment option (14). For the cohort with sinus involvement, the tumor could be removed en bloc or using piecemeal technique, with similar results. Patients with sinus involvement were more likely to have transient or permanent postoperative deficits (23% vs. 0%). The overall actuarial median survival for the entire cohort was 16.5 months. Median survival was similar between the groups of patients with and without dural sinus involvement. External beam radiation therapy will be the mainstay of treatment for the majority of patients with calvarial metastases (5,8). The most commonly used treatment schedule is 30 Gy of conformal electron beam irradiation,

Chapter 10 / Skull and Dural Metastases

149

delivered over the course of two weeks (5,15). Using this approach, objective tumor shrinkage is noted in 60% or more of patients. For patients suspected of having calvarial metastases from a thyroid primary, radioactive iodine can also be an effective mode of treatment (16). Chemotherapy can be used as the primary treatment modality in carefully selected patients with chemosensitive tumors (e.g., breast, lymphoma).

3. SKULL BASE METASTASES The skull base consists of fused portions of the ethmoid bone, the sphenoid bone (including the rostrum, lesser, and greater wings), the petrous and tympanic portions of the temporal bone (excluding the squama), the clivus, dorsum sella, and occipital bone (17). Intracranially, the surface of the skull base is divided into the anterior, middle, and posterior cranial fossae. During embryogenesis of the skull, the bones that comprise the skull base develop during the sixth fetal week from cartilaginous precursors (4).

3.1. Epidemiology The incidence of skull base metastases is quite variable depending on whether or not the data is antemortem or autopsy based, with an overall estimate of approximately 4% (5,8,18–20). Antemortem, population-based studies of patients with cranial neuropathy and parasellar or cavernous sinus metastases suggest an incidence of less than .05% (21–23). However, autopsy based studies indicate a higher incidence, ranging from 3% to 24%. Several authors have reviewed the incidence of metastasis into the temporal bones of patients with various forms of cancer. In a series of 249 temporal bones, Jung and co-workers noted an incidence of metastasis in approximately 24% of cases (24). Similarly, Gloria-Cruz noted an incidence of metastasis in 22.2% of 415 temporal bones analyzed from a cohort of 212 patients with primary nondisseminated primary neoplasms (25). Reese reviewed orbital skull base metastases in a series of 877 consecutive cases of orbital neoplasms from the Institute of Ophthalmology in New York (26). The majority of tumors were of primary origin, with an incidence of orbital metastases of only 3%. In surgical series of skull base surgery, metastases tend to be relatively uncommon, with an incidence of less than 8% (27). The most common primary neoplasms to metastasize to the skull base are breast, lung, and prostate, accounting for 40%, 14%, and 12% of cases, respectively (18). However, many other tumors have the potential to disseminate to this location, especially at the end stages of disease, including colorectal, renal, thyroid, lymphoma, melanoma, and neuroblastoma. Breast cancer is the most common cause of skull base metastases in female patients, while prostate cancer is most frequently diagnosed in male patients (19). Although skull base metastases tend to occur at late stages of disease dissemination, it can be the first sign of cancer in a significant percentage of patients (5,19).

3.2. Pathophysiology Skull base metastases most often arise via hematogenous spread from a systemic primary tumor (5,18–20). Arterial tumor emboli gain access to the skull base through small anastomotic vessels at the neural foramina. Similar to calvarial metastases, Batson’s venous plexus can also function as a conduit for tumor emboli in conditions of increased intra-abdominal and intrathoracic pressure (2). Another mode of metastatic spread involves invasion into the skull base from regional neoplasms, including tumors of the nasopharynx, paranasal sinuses, external ear, and salivary glands. Tumor cells typically gain access to the skull base through retrograde growth along cranial nerves. This is especially prevalent with nasopharyngeal carcinoma, noted in 15–35% of all patients (20,28). Other potential pathways for tumor spread include the muscles attached to the skull base, the fibrofatty spaces in between the skull base musculature, and along the perineural spaces of these tissues.

3.3. Presentation In many patients, skull base metastases can remain silent for extended periods of time, until further enlargement causes pain or neurologic impairment (5,8,18–20). For the majority of patients, craniofacial pain will be the first sign of skull base metastasis, as tumor invades and destroys bone. The pain becomes more severe over time and typically precedes neurologic involvement by several weeks. Unlike calvarial metastases, the onset of neurologic signs and symptoms in a patient with skull base metastasis can often have localizing value for the

150

Part IV / Direct Complications of Cancer

Table 1 Manifestations of Metastatic Skull Base Syndromes Location

Syndrome

Anatomy

Manifestations

Anterior skull base

Orbital

Extraocular muscles; CNs III, IV, V1 , and VI

Supraorbital and orbital pain; blurred vision; proptosis; no diplopia; external ophthalmoplegia; enophthalmus; periorbital swelling and tenderness

Middle skull base

Parasellar

CNs III, IV, V1 , and VI

Gasserian ganglion

CN V (all branches, sensory and motor); possibly CNs III, IV, VI, VII Middle ear CNs VII, VIII

Unilateral frontal HA No proptosis No visual loss Ocular paresis Periorbital edema Diplopia Facial pain and numbness V12 Numbness, paresthesias, and pain along V23 ; unilateral weakness of muscles of mastication; abducens weakness

Temporal bone Posterior skull base

Jugular foramen

CNs IX, X, XI, XII

Occipital condyle

CN XII

Hearing loss, otalgia, periauricular swelling, facial paresis Unilateral occipital & Postauricular pain; Dysphagia Hoarseness Palatal weakness Vocal cord paralysis Horner’s syndrome SCM/trapezius weakness and atrophy Occipital pain Neck stiffness/pain Dysarthria and dysphagia Ipsilateral tongue weakness and atrophy

Abbreviations: CN: cranial nerve, HA: headache, SCM: sternocleidomastoid muscle; Adapted from Refs. (18–20)

clinician (5,19,20). This was first described by Greenberg and colleagues, in their seminal description of skull base metastases, wherein they identified five clinical syndromes (Table 1): orbital, parasellar, middle fossa, jugular foramen, and occipital condyle (18). 3.3.1. Orbital Syndrome The orbital syndrome occurs with metastatic involvement of the bony orbit; metastases into the orbital soft tissues are relatively uncommon. The incidence of orbital metastases is variable, ranging from 6.9% in the Greenberg study up to 11.9% in a review of 235 patients with metastases to the eye (18,29). The most common tumors to metastasize to this region are from breast, lung, prostate, and lymphoma. Clinically, the orbital syndrome presents with a progressive, dull headache in the supraorbital region of the affected eye (18–20). The presence of persistent and progressive pain in the supraorbital region is a common finding in bony orbital metastases, unlike the majority of primary orbital neoplasms. After the onset of headache, other symptoms typically arise, including blurred vision, diplopia, periorbital swelling and tenderness, and numbness of the periorbital region. On neurologic examination, proptosis is present in virtually all patients, along with a variable degree of external ophthalmoplegia. Reduced sensation consistent with dysfunction of the first division of the trigeminal nerve may be noted. In some patients, periorbital swelling may be present, and the tumor may be palpable during examination of the orbit. Decreased vision due to involvement of the optic nerve may occur, but is usually a late finding.

Chapter 10 / Skull and Dural Metastases

151

3.3.2. Parasellar Syndrome The parasellar syndrome, also described as the sphenocavernous syndrome, results from metastatic involvement of either the petrous apex or the sella turcica, with secondary invasion of the cavernous sinus (5,18–20,30). Less often it can result from a direct metastasis into the cavernous sinus. The incidence of parasellar region metastases generally ranges from 7% to 16%, but may be higher if sellar metastases are included (e.g., 29% in the report by Laigle-Donadey et al.) (18,20,30). The most common tumors to metastasize into the parasellar region are breast, prostate, lung, and lymphoma. In fact, some authors feel that lymphoma has a particular tropism for the parasellar region (31). As the parasellar syndrome develops, tumor causes compression of the ocular motor nerves (i.e., CN III, IV, VI) as they traverse the cavernous sinus, as well as the ophthalmic and maxillary divisions of the trigeminal nerve (32). Clinical symptoms usually manifest as a unilateral, progressive, frontal supraorbital headache (in 80–85% of patients), associated with eye movement abnormalities and diplopia. Patients may also complain of facial numbness, hypesthesia, or dysesthesia, in a distribution consistent with the ophthalmic division of the trigeminal nerve. Less often, facial sensory complaints involve the maxillary division of the trigeminal nerve. On neurologic examination, patients will have complete or partial oculoparesis and periorbital swelling, without significant proptosis. In rare cases, the ophthalmoplegia can be bilateral. Vision remains intact until the very late stages of tumor growth. Diminished or abnormal sensation in the distribution of the ophthalmic and/or maxillary divisions of the trigeminal nerve may be noted. 3.3.3. Middle Fossa Syndrome The middle fossa syndrome, also described as the Gasserian ganglion syndrome, occurs when metastatic tumor deposits involve the middle fossa, within the Gasserian ganglion and/or proximal branches of the trigeminal nerve (18–20). Although the exact incidence is unknown, the syndrome appears to be relatively common, comprising 35% of the cohort reviewed by Greenberg and co-workers (18). The most common symptoms include a slowly progressive, dull, aching pain in the cheek, jaw or forehead and numbness or sensory loss in the lower face and jaw. The sensory changes usually begin close to the midline of the upper lip or chin and progress laterally towards the ear. Sometimes the pain can have a “lightning-like” quality, similar to trigeminal neuralgia. Headaches are relatively uncommon (25–30% of patients) in comparison to the orbital and parasellar syndromes. Tumor spread beyond the confines of the gasserion ganglion may lead to involvement of other cranial nerves and symptoms such as diplopia and facial weakness. The neurologic examination typically reveals loss of sensation in the territory of the maxillary and mandibular divisions of the trigeminal nerve. Mild weakness of the muscles of mastication may be present. However, electromyography may be necessary in some cases to demonstrate dysfunction (i.e., denervation) of these muscle groups. In 20–35% of patients, other cranial nerve palsies may be noted (alone or in combination), including facial weakness, abducens nerve palsy, or more extensive impairment of extraocular movements. 3.3.4. Jugular Foramen Syndrome The jugular foramen syndrome occurs when metastatic tumor involves the region of the skull base around the jugular foramen, affecting multiple cranial nerves (18–20,33–35). It appears to be relatively common and accounted for 21% of the cohort described by Greenberg and colleagues (18). The typical clinical presentation includes unilateral occipital or postauricular pain, along with progressive hoarseness and dysphagia. In some cases, the pain may be consistent with glossopharyngeal neuralgia and associated with syncope (18,36). On neurologic examination, patients usually have evidence of cranial nerve palsies affecting CNs IX, X, and XI, with weakness or paralysis of the palate, vocal cords, ipsilateral sternocleidomastoid muscle, and upper part of the trapezius muscle and, occasionally, a Horner’s syndrome. Papilledema can be noted in some patients, as a result of tumor-related compression of the sigmoid sinus or jugular vein. If there is tumor spread to the hypoglossal canal, resulting in ipsilateral weakness and atrophy of the tongue, then the syndrome may become a variant of the jugular foramen syndrome—the Collet–Sicard Syndrome—in which the four lowest cranial nerves become involved (19,37).

152

Part IV / Direct Complications of Cancer

3.3.5. Occipital Condyle Syndrome The occipital condyle syndrome can arise when metastatic deposits affect the condylar region of the occipital bone (18–20,38). Although the occipital condyle syndrome accounted for 21% of the series reported by Greenberg and co-workers, some authors still suggest that the syndrome is generally underdiagnosed (18,38). In the series of 11 patients reported by Capobianco and colleagues, breast carcinoma was most common in women, while prostate carcinoma was most prevalent in men (38). The clinical picture is very stereotyped and consists of progressive, continuous, severe pain localized to the occipital region, in combination with ipsilateral atrophy and paresis of the tongue (i.e., CN XII palsy). Neck pain is usually the first symptom and precedes tongue involvement by several weeks. The neck pain is often accompanied by neck stiffness and is typically exacerbated by rotational movements. Radiation of the pain anteriorly, towards the temporal region and orbit, is noted in some patients. In over half the patients, dysarthria, dysphagia, or a combination of both symptoms will develop. Both symptoms are caused by weakness and immobility of the tongue. On neurologic examination, all patients will have tenderness to palpation in the affected occipital region, along with some degree of neck stiffness. Rotational maneuvers or flexion of the neck will increase the pain. The ipsilateral tongue will be weak and atrophic, and may exhibit fasciculations. 3.3.6. Other Syndromes Other clinical syndromes can also occur in the setting of skull base metastases, including the temporal bone syndrome, sellar syndrome, and invasion by regional tumors such as esthesioneuroblastoma (ENB) and nasopharyngeal carcinoma (19,20). The temporal bone syndrome occurs when metastatic deposits involve the temporal bone and auditory apparatus (20,39,40). Conductive hearing loss is the most common symptom associated with temporal bone metastases, and is noted in 30–40% of symptomatic patients. The mechanism is usually dysfunction of the eustachian tube, with secondary serous otitis media. Less often, sensorineural hearing loss occurs as a result of damage to cochlear fibers in the internal auditory meatus. Other common manifestations include otalgia, periauricular swelling, and facial nerve paresis. The sellar syndrome occurs when metastatic deposits directly involve the sella turcica and pituitary gland (19,41,42). Pituitary metastases are relatively common, with an incidence ranging from 3.5% to 26% in autopsy series. The most common primary tumors to metastasize to the pituitary gland are breast, lung, and bladder carcinomas. If metastatic tumor mostly involves the anterior pituitary gland (13%), the patient may remain asymptomatic until late in the course of the disease, when hypopituitarism and visual loss develop (42). More significant involvement of the posterior pituitary gland (57%) often leads to early onset of symptoms from diabetes insipidus, which is a common presentation of the sellar syndrome. Involvement of both lobes is noted in 12% of patients with sellar metastases. Lateral extension of tumor into the cavernous sinus will lead to cranial neuropathies and symptoms of the parasellar syndrome. In fact, many reports do not differentiate patients with the sellar syndrome, and just include them with the cohort that has parasellar involvement. Esthesioneuroblastoma is an uncommon tumor that arises from the sensory epithelium of the nasal cavity and has the capacity to invade the skull base and extend intracranially (5,8,43,44). Metastases are common in patients with ENB, and involve the central nervous system in 25–30% of cases. The tumor typically affects the skull base in the region of the cribiform plate, with further invasion allowing access to the leptomeninges and brain. After the tumor has entered the intracranial cavity, it can manifest as leptomeningeal disease or an intraparenchymal metastasis. Common symptoms include diplopia, decreased vision, facial weakness, headache, personality changes, and extremity weakness (44). Nasopharyngeal carcinomas usually arise from the fossa of Rosenmüller and present in a very similar fashion to ENB, with perineural invasion of cranial nerves, retrograde growth into the skull base, and extension into the brain (5,8,45–47). Skull base erosion and invasion is noted in 35–50% of patients with nasopharyngeal tumors, with actual intracranial spread occurring in 12–31%. However, in some patients there is minimal erosion of the skull base. In these cases, tumor cells gain access to the intracranial cavity via retrograde growth along cranial nerves through the foramen lacerum and foramen ovale (46,47). After skull invasion and perineural spread occurs, patients develop progressive craniofacial pain and cranial neuropathies (46,48). The overall incidence of cranial nerve involvement in patients with nasopharyngeal carcinoma ranges from 12% to 30%. The most commonly affected cranial nerves are the trigeminal and the abducens; however, any combination of the middle and lower nerves can be involved. Patients

Chapter 10 / Skull and Dural Metastases

153

with trigeminal nerve infiltration typically present with facial paresthesias or dysesthesias, in the distribution of the V2 and V3 divisions.

3.4. Imaging Neuroimaging is critical to confirm the diagnosis of skull base metastases. Plain x-rays are somewhat insensitive to the presence of metastases in the skull base, although they may show some evidence of bone erosion. In general, tumor deposits are more clearly delineated with the use of CT and MRI scans (19,20,49–51). CT scans with high resolution and bone windowing can detect subtle osseous destruction or sclerosis of the skull base (Fig. 3). In addition, CT with soft tissue windows and contrast administration can often reveal enhancing masses within the bones of the skull base. Three-dimensionally reconstructed CT scans may be able to provide a complete outline of tumor involvement in the skull base. Overall, CT is more sensitive to the presence of tumor-related calcification and the outlines of lytic bone destruction than MRI (20,49). However, despite these advantages, CT is generally inferior to MRI in the detection and delineation of metastatic tumors of the skull base. On T2 and FLAIR images, the tumor will be clearly demonstrated as a high signal mass within the bones of the skull (49,50). On T1 non-enhanced images, the tumor will appear hypointense, as it replaces the hyperintense appearing fat signal within the marrow of the skull bones. After the administration of contrast, the mass will usually exhibit dense enhancement on T1weighted images (Figs. 4 and 5 and Color Plate 1). Evaluation of the mass on axial, coronal, and sagittal images will allow complete delineation of tumor margins and involvement of other skull base structures, especially the cranial nerves. In addition, MRI is very sensitive to invasion of the skull base by regional tumors such as ENB and nasopharyngeal carcinoma, and can clearly demonstrate perineural invasion and retrograde growth through the skull base (Fig. 6) (28,45,46,49,50,52,53). In many cases, the involved cranial nerves will exhibit contrast enhancement and the common foramina of passage through the skull base (i.e., lacerum, ovale) will become enlarged. Nuclear medicine studies such as bone scans and SPECT imaging are also sensitive to the presence of metastases within the skull base and can complement CT and MRI, in selected cases (8,12,54).

3.5. Treatment Surgical forms of treatment will only be appropriate in carefully selected patients with skull base metastases (5,8,19,20). In rare cases, a biopsy may be necessary to confirm that a given mass is a metastatic tumor and not some other diagnostic entity, such as a primary tumor (e.g., chordoma, dermoid) or non-neoplastic lesion (e.g., fibrous dysplasia, eosinophilic granuloma) (8). Surgical resection appears to be most appropriate in patients with solitary metastases, particularly those localized to the sellar and parasellar regions (5,30).

Fig. 3. Bone windowed CT scan of an 83 year-old male patient with stage 4 prostate carcinoma and a history of frontal and orbital pain, and a left sided cranial nerve III palsy with ptosis. Note the mass centered in the left lateral sphenoid and clivus region, with erosion of involved bone and extension into the pituitary fossa.

154

Part IV / Direct Complications of Cancer

Fig. 4. MRI of the same patient as in Fig. 3. Axial contrast enhanced T1-weighted images demonstrating the enhancing mass involving the left cavernous sinus, clivus, and pituitary gland.

There are numerous surgical techniques available for skull base surgery of metastatic lesions, including the transfacial, posterolateral, transpetrosal, translabyrinthine, transsphenoidal, and transcochlear approaches, as well as various combinations (5,20,55). The surgeon will have to determine the specific approach and technique based on the location and size of the metastatic tumor. In general, the correct approach should minimize brain retraction, provide adequate tumor exposure and control of the proximal and distal internal carotid artery, and allow for proper vascular, cranial nerve, and cranial base reconstruction to prevent CSF leakage and infection (5,55). Using aggressive surgical techniques, Jia and colleagues reported that 13 of 15 patients with skull base metastases were able to undergo a radical resection (56). However, there were some significant complications with this approach, including cases of CSF leakage and meningitis. Overall, the metastatic tumor types most appropriate for surgical

 Fig. 5. MRI scans of a 60-year-old female with stage 4 breast carcinoma and complaints of headache and abnormal vision. (A) Non-enhanced mid-sagittal T1-weighted image, demonstrating an isointense, metastatic tumor within the clivus, anterior to the lower brainstem. (B) Enhanced axial T1-weighted image, demonstrating the enhancing mass within the clivus. (C) Low power view of surgical specimen of metastatic medullary breast carcinoma removed from the skull base, revealing nests of tumor cells within and alongside regions of bone (hematoxylin and eosin @100x). (D) High-power view of the same specimen (hematoxylin and eosin @200x) (see Color Plate 1).

Chapter 10 / Skull and Dural Metastases

155

Fig. 5.

156

Part IV / Direct Complications of Cancer

Fig. 6. MRI scans of a 55-year-old male with proptosis and complaints of deteriorating vision. Surgical resection revealed nasopharyngeal carcinoma. (A) Nonenhanced sagittal T1-weighted image demonstrating a large nasopharyngeal mass with erosion into the anterior skull base below the frontal lobes. (B) Enhanced axial T1-weighted image of the mass invading through the anterior skull base and displacing the left orbital contents.

resection are those with poor responsiveness to irradiation and chemotherapy, such as renal cell carcinoma, melanoma, and various forms of sarcoma (5,20). The most frequently used mode of therapy for patients with skull base metastases is standard, fractionated, external beam radiotherapy (5,8,18–20). In most reports, between 70% and 80% of patients receive this form of irradiation as the main treatment modality (alone or in combination with chemotherapy). The typical dosing regimen is 30–35 Gy of conformal treatment, delivered in 10–15 fractions over 2–3 weeks. Overall, radiotherapy improves neurologic symptoms, including pain, in 75–80% of patients (20,57). In many patients, the symptom relief is persistent and durable. Cranial neuropathies are also responsive to irradiation and can demonstrate improvement in up to 90% of cases (19). The potential for responsiveness to radiotherapy and neurologic improvement appears to be dependent on the timing of treatment (57). When irradiation was initiated within 4 weeks of symptom onset, 87% of patients improved. In contrast, only 25% of patients demonstrated improvement when radiotherapy was initiated 12 weeks or more after symptom onset. Although it remains somewhat controversial, some authors recommend administering larger doses of radiation (36–50 Gy) to patients with well-controlled systemic disease and a long life expectancy (20). A newer mode of radiation therapy that has been recently applied to skull base metastases is radiosurgery, including gamma knife and linear accelerator-based techniques (19,20,58–60). With both methods, the tumor is treated to a very high dose in a single fraction or in multiple, smaller fractions. Radiosurgery is an excellent choice

Chapter 10 / Skull and Dural Metastases

157

of treatment for patients who have already failed more conventional, fractionated modes of irradiation. Typical doses range from 12 to 16 Gy, with symptomatic control noted in 60–70% of patients (58–60). Local tumor control has been documented in 65–70% of patients, with acceptable toxicity. Additionally, cranial neuropathies can improve in up to 60% of patients following radiosurgery (59). Chemotherapy has only a limited role in the treatment of skull base metastases (5,8,18–20). It is generally ineffective when used as the solitary modality of treatment, and is unlikely to reverse or improve neurologic deficits. When used in combination with conventional radiation therapy, chemotherapy may improve response rates in sensitive tumor types such as breast carcinoma, small cell lung carcinoma, and lymphoma.

4. DURAL METASTASES Three membranes cover the brain; dura, arachnoid, and pia. The dura consists of a tough, fibrous sheath that extends from the cranial convexity to the level of the second sacral vertebra, where it ends as a blind sac. The dura is separated from the thin arachnoid by a potential space (subdural space) with little CSF whereas the subarachnoid space separates the arachnoid from pia and contains the majority of the CSF and major arteries. The dura has an inner (meningeal) and an outer (periosteal) layer, and between these layer lie the dural sinuses.

4.1. Epidemiology The incidence of dural metastases is approximately 20% in autopsy studies (62). In the report by Posner and Chernik, 25% of this cohort had involvement only affecting the subdural space. Antemortem diagnosis of dural metastases occurs much less frequently and, in some cases, can be a serendipitous finding. The tumors most likely to metastasize to the dura are lung carcinoma, prostatic adenocarcinoma, breast carcinoma, and melanoma (2,62). Other tumor types with potential to involve the dura include hepatocellular and pancreatic carcinoma, neuroblastoma, lymphoma, leukemia, osteogenic and Ewing’s sarcoma, carcinoid tumors, and plasmacytoma.

4.2. Pathophysiology Dural metastases can involve the subdural space, epidural space, or both, and can be focal or more diffuse (5,12). Tumor cells usually gain access to the dura via arterial hematogenous spread from the systemic primary neoplasm, since the dura and subdural membranes contain a rich vascular capillary bed (2). Less often, tumor cells can reach the dura through the venous system or by contiguous spread from calvarial metastases or tumors invading through the skull base. In addition, subdural metastases can develop via spread into a previously existing chronic subdural fluid collection (61).

4.3. Presentation The most common presenting symptoms of dural metastases include localized headache, lethargy, confusion, and focal findings such as contralateral hemiparesis or sensory loss (5,12). Typically, symptoms arise in a progressive fashion, but can be abrupt in some patients. A more acute presentation of focal deficits and seizures suggests that tumors cells have begun to infiltrate the brain parenchyma through the Virchow–Robin spaces. Acute onset of focal symptoms can also arise from tumor-related hemorrhage or venous sinus occlusion. Similar to calvarial metastases, dural metastases can invade and occlude the sagittal or lateral venous sinuses, causing increased intracranial pressure, headache, visual changes, and mental status changes.

4.4. Imaging Dural metastases are readily demonstrated by MR imaging (Fig. 7 and Color Plate 2) (5,12,63). On T2 and FLAIR images, the masses appear as iso- to hyperintense soft tissue lesions associated with the dura in the epidural and/or subdural spaces. The masses typically demonstrate dense enhancement after contrast administration. MRI is also sensitive to the presence of leptomeningeal metastases and impaired flow within the venous system. Obstruction of a venous sinus is revealed as a flow void within the affected vessel. Further delineation of vascular compromise can be made with MR angiography. CT with contrast can sometimes demonstrate an enhancing dural metastasis. However, CT is generally inferior to MRI in terms of sensitivity and the ability to reveal

158

Part IV / Direct Complications of Cancer

Fig. 7. Contrast-enhanced T1-weighted MRI scans of a 40-year-old female with a history of stage 4 breast carcinoma and complaints of headache and somnolence. The axial (A) and coronal (B) images both demonstrate the presence of a large, enhancing dural-based mass in the right frontal region, with compression of the underlying brain and regional edema. (C) Pathological materials from an unrelated patient with a dural metastasis from renal cell carcinoma, in whom surgical resection was necessary because of severe mass effect. Note the clumps of renal cell carcinoma cells in and around the fibrous layers of the dura (hematoxylin and eosin @100x) (see Color Plate 2).

small metastatic deposits involving the dura. The presence of dural enhancement in a cancer patient is not pathognomonic for metastatic involvement. Unrelated non-neoplastic and neoplastic processes that can induce dural enhancement include CSF leakage, dural sinus thrombosis, meningitis, and benign primary CNS tumors (e.g., meningioma) (63).

Chapter 10 / Skull and Dural Metastases

159

4.5. Treatment Surgical resection is a reasonable treatment option for patients with a large, solitary dural metastasis, especially if there is significant underlying mass effect (5). After recovery from surgery, these patients would also require postoperative irradiation. For most other patients, external beam radiation therapy would be the primary mode of treatment (5,15,64). The most common regimen consists of 30 Gy in 10 fractions over 2 weeks, delivered to the whole brain or the involved region of dura. Following irradiation, between 60% and 75% of patients will note significant clinical improvement. Neuroimaging studies with CT and MRI note objective responses (i.e., 50% tumor shrinkage) in more than 60% of patients. In addition, metastatic venous obstruction will frequently improve after treatment, with the return of venous blood flow in some cases (15). Chemotherapy can be considered as a treatment option, either alone or in combination with irradiation, for patients with dural metastases from chemosensitive tumors, such as lymphoma, breast carcinoma, small cell carcinoma, and germ cell tumors.

5. CONCLUSION Although skull and dural metastases are relatively silent until they are large in size, they are increasingly recognized in asymptomatic stages due to better and more frequent neuroimaging and prolonged survival of cancer patients. Thus, early recognition of such lesions and knowing that they will eventually lead to clinical symptoms allows us to treat early and maintain patients’ quality of life.

6. ACKNOWLEDGMENTS The author would like to thank Dr. Abhik Ray-Chaudhury for providing the neuropathological materials, Drs. Wayne Slone and Eric Bourekas for assistance with the neuroimaging studies, and Julia Shekunov for research assistance. Dr. Newton was supported in part by National Cancer Institute grant CA 16058 and the Dardinger Neuro-Oncology Center Endowment Fund.

REFERENCES 1. Newton HB. Neurologic complications of systemic cancer. Am Fam Phys 1999;59:878–886. 2. Posner JB (ed.). Pathophysiology of metastases to the nervous system. In: Neurologic Complications of Cancer. Philadelphia: F.A. Davis, 1995:15–36. 3. Fournier HD, Delliere V, Gourraud JB et al. Surgical anatomy of calvarial skin and bones: with particular reference to neurosurgical approaches. Adv Tech Stand Neurosurg 2006;31:253–271. 4. Couly GF, Coltey PM, Le Douarin NM. The triple origin of skull in higher vertebrates: a study in quail-chick chimeras. Develop 1993;117:409–429. 5. Rodas RA, Greenberg HS. Dural, calvarial and skull base metastasis. In: Vecht CJ (Ed.). Handbook of Clinical Neurology, Vol. 25 (69): Amsterdam: Elsevier Science, 1997;8:123–134. 6. Jacobs L, Kinkel WR, Vincent RG. Silent brain metastases from lung carcinoma determined by computerized tomography. Arch Neurol 1977;34:690–693. 7. Salbeck R, Grau HC, Artman H. Cerebral tumor staging in patients with bronchial carcinoma by computed tomography. Cancer 1990;66:2007–2011. 8. Stark AM, Eichmann T, Mehdorn HM. Skull metastases: clinical features, differential diagnosis, and review of the literature. Surg Neurol 2003;60:219–226. 9. Stark RJ, Henson RA. Cerebral compression by myeloma. J Neurol Neurosurg Psych 1981;44:833–836. 10. Edema OT, Oviawe O, Akenzua GI. Orbitocephalic metastases from neuroblastoma: report of three cases. West Afr J Med 1998;17: 286–289. 11. Arana E, Marti-Bonmati L. CT and MR imaging of focal calvarial lesions. Am J Radiol 1999;172:1683–1688. 12. Maroldi R, Ambrosi C, Farina D. Metastatic disease of the brain: extra-axial metastases (skull, dura, leptomeningeal) and tumour spread. Eur Radiol 2005;15:617–626. 13. West MS, Russell EJ, Breit R et al. Calvarial and skull base metastases: comparison of nonenhanced and Gd-DTPA-enhanced MR images. Radiol 1990;174:85–91. 14. Michael CB, Gokaslan ZL, DeMonte F et al. Surgical resection of calvarial metastases overlying dural sinuses. Neurosurg 2001;48: 745–755. 15. Harwood AR, Simpson WJ. Radiation therapy of cerebral metasases: a randomized prospective clinical trial. Int J Rad Oncol Biol Phys 1977;2:1091–1094. 16. Greco FA, Hainsworth JD. Cancer of unknown primary site. In: DeVita VT, Hellman S, Rosenberg SA (Eds.). Cancer Principles and Practice of Oncology. 4th ed. Philadelphia: J.B. Lippincott 1993:2072–2093. 17. Glasscock ME, Miller GW, Drake FD. Surgery of the skull base. Laryngoscope 1978;88:905–923.

160

Part IV / Direct Complications of Cancer

18. Greenberg HS, Deck MDF, Vikram B et al. Metastasis to the skull: clinical findings in 43 patients. Neurol 1981;31:530–537. 19. Laigle-Donadey F, Taillibert S, Martin–Duverneuil N et al. Skull base metastases. J Neuro–Oncol 2005;75:63–69. 20. DeMonte F, Hanbali F, Ballo MT. Skull base metasasis. In: Berger MS, Prados MD (Eds.). Textbook of Neuro–Oncology. Philadelphia: Elsevier Medical Publishers, 2005;61:466–475. 21. Hall SM, Buzdar AU, Blumenschein GR. Cranial nerve palsies in metastatic breast cancer due to osseous metastasis without intracranial involvement. Cancer 1983;52:180–184. 22. Bitoh S, Hasegawa H, Ohtsuki H et al. Parasellar metastases: four autopsied cases. Surg Neurol 1985;23:41–48. 23. Ahmad K, Kim YH, Post MJ et al. Involvement of the cavernous sinus region by malignant neoplasms: report of 5 cases. J Am Osteopath Assoc 1987;87:504/91–508/95. 24. Jung TT, Jun BH, Shea D et al. Primary and secondary tumors of the facial nerve: a temporal bone study. Arch Otolaryngol Head Neck Surg 1986;112:1269–1273. 25. Gloria–Cruz TI, Schachern PA, Paparella MM et al. Metastases to temporal bones from primary nonsystemic malignant neoplasms. Arch Otolaryngol Head Neck Surg 2000;126:209–214. 26. Reese A. Tumors of the Eye. 2nd ed. New York: Harper & Row, 1962. 27. Jackson CG, Netterville JL, Glasscock ME, et al. Defect reconstruction and cerebrospinal fluid management in neurotologic skull base tumors with intracranial extension. Laryngoscope 1992;102;1205–1214. 28. Chong VFH, Fan YF, Khoo JBK. Nasopharyngeal carcinoma with intracranial spread: CT and MR characteristics. J Comput Assist Tomogr 1996;20:563–569. 29. Font RL, Ferry AP. Carcinoma metastatic to the eye and orbit III: a clinicopathologic study of 28 cases metastatic to the orbit. Cancer 1976;38:1326–1335. 30. Yi HJ, Kim CH, Bak KH et al. Metastatic tumors in the sellar and parasellar regions: clinical review of four cases. J Korean Med Sci 2000;15:363–365. 31. Julien J, Ferrer X, Drouillard J et al. Cavernous sinus syndrome due to lymphoma. J Neurol Neurosurg Psych 1984;47:558–560. 32. Johnston JL. Parasellar syndromes. Curr Neurol Neurosci Rep 2002;2:423–431. 33. Wilson H, Johnson DH. Jugular foramen syndrome as a complication of metastatic cancer of the prostate. South Med J 1984;77:92–93. 34. Boileau MA, Grotta JC, Borit A et al. Metastatic renal cell carcinoma simulating glomus jugulare tumor. J Surg Oncol 1987;35:201–203. 35. Schhweinfurth JM, Johnson JT, Weissman J. Jugular foramen syndrome as a complication of metastatic melanoma. Am J Otolaryngol 1993;14:168–174. 36. Ferrante L, Artico M, Nardacci B et al. Glossopharyngeal neuralgia with cardiac syncope. Neurosurg 1995;36:58–63. 37. Chacon G, Alexandraki I, Palacio C. Collet–Sicard syndrome: an uncommon manifestation of metastatic prostate cancer. South Med J 2006;99:898–899. 38. Capobianco DJ, Brazis PW, Rubino FA et al. Occipital condyle syndrome. Headache 2002;42:142–146. 39. Berlinger NT, Koutroupas S, Adams G et al. Patterns of involvement of the temporal bone in metastatic and systemic malignancy. Laryngoscope 1980;90:619–627. 40. Gloria-Cruz TI, Schachern PA, Paparella MM et al. Metastases to temporal bones from primary nonsystemic malignant neoplasms. Arch Otolaryngol Head Neck Surg 2000;126:209–214. 41. Mayr NA, Yuh WTC, Muhonen MG et al. Pituitary metastases: MR findings. J Comput Ass Tomogr 1993;17:432–437. 42. Fassett DR, Couldwell WT. Metastases to the pituitary gland. Neurosurg Focus 2004;16:1–4. 43. Meneses MS, Thurel C, Mikol J, et al. Esthesioneuroblastoma with intracranial extension. Neurosurg 1990;27:813–820. 44. Chamberlain MC. Treatment of intracranial metastatic esthesioneuroblastoma. Cancer 2002;95:243–248. 45. Su CY, Lui CC. Perineural invasion of the trigeminal nerve in patients with nasopharyngeal carcinoma: imaging and clinical correlations. Cancer 1996;78:2063–2069. 46. Chong VFH, Fan YF, Khoo JBK. Nasopharyngeal carcinoma with intracranial spread: CT and MR characteristics. J Comp Assist Tomogr 1996;20:563–569. 47. Sham JSY, Cheung YK, Choy D et al. Cranial nerve involvement and base of skull erosion in nasopharyngeal carcinoma. Cancer 1991;68:422–426. 48. Leung SF, Tsao SY, Teo P et al. Cranial nerve involvement by nasopharyngeal carcinoma: response to treatment and clinical significance. Clin Oncol 1990;2:138–141. 49. Ginsberg LE. Neoplastic disease affecting the central skull base: CT and MR imaging. Am J Radiol 1992;159:581–589. 50. Curtin HD, Hirsch WL. Base of the skull. In: Atlas SW (Ed.). Magnetic Resonance Imaging of the Brain and Spine. New York: Raven Press, 1991:669–707. 51. Kumar AJ, Lee YY, Zinreich SJ et al. Imaging features of skull base tumors. Neuroimag Clin N Am 1993;3715–734. 52. Ginsberg LE. MR imaging of perineural tumor spread. Magn Reson Imag Clin N Am 2002;10:511–525. 53. Ishidea H, Mohri M, Amatsu M. Invasion of the skull base by carcinomas: histopathologically evidenced findings with CT and MRI. Eur Arch Otorhinolaryngol 2002;259:535–539. 54. Jansen B, Pillay M, De Bruin H, et al. Tc99m–SPECT in the diagnosis of skull base metastases. Neurol 1997;48:1326–1330. 55. Rothbart D, Lawton MT, Spetzler RF. Cranial base surgery for invasive carcinomas. In: Maciunas RJ (Ed.). Advanced Techniques in Central Nervous System Metastases. Park Ridge: American Association of Neurologic Surgeons, 1998;5:69–94. 56. Jia G, Zhang J, Wu Z. Diagnosis and treatment of skull base metastasis. Zhonghua Yi Xue Za Zhi 1998;78:761–762. 57. Vikram B, Chu FC. Radiation therapy for metastases to the base of the skull. Radiol 1979;130:465–468. 58. Cmelak AJ, Cox RS, Adler JR et al. Radiosurgery for skull base malignancies and nasopharyngeal carcinoma. Int J Rad Oncol Biol Phys 1997;37:997–1003. 59. Iwai Y, Yamanaka K. Gamma knife radiosurgery for skull base metastasis and invasion. Stereotact Funct Neurosurg 1999;72:81–87.

Chapter 10 / Skull and Dural Metastases

161

60. Miller RC, Foote RL, Coffey RJ et al. The role of stereotactic radiosurgery in the treatment of malignant skull base tumors. Int J Rad Oncol Biol Phys 1997;39:977–981. 61. Rouch E, Goodman FC, Harper RL. Acute subdural hemorrhage and metastatic seminoma. Neurol 1986;36:418–420. 62. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol 1978;19:575–587. 63. Sze G, Soletzky S, Bronen R et al. MR imaging of the cranial meninges with emphasis on contrast enhancement and meningeal carcinomatosis. Am J Neuroradiol 1989;10:965–975. 64. Tallibert S, Hildebrand J. Treatment of central nervous system metasastases: parenchymal, epidural, and leptomeningeal. Curr Opin Oncol 2006;18:637–643.

11

Spinal Metastases Jonathan H. Sherman, MD, Dawit G. Aregawi, Mark E. Shaffrey, MD, and David Schiff, MD

MD,

CONTENTS Introduction Epidemiology Pathogenesis and Pathophysiology Presentation Diagnostic Work-Up Prognosis Treatments Intradural Intramedullary Spinal Cord Metastasis Conclusions References

Summary Spinal cord metastases are a common complication of systemic malignancy, most commonly from lung, breast, and prostate cancer. Patients may present with a variety of symptoms, most notably pain and weakness. The ultimate goals in managing these patients include maximizing neurologic function, length of survival, and quality of life. These goals can best be reached via early, accurate diagnosis followed by the appropriate treatment for a particular patient. Pharmacotherapy plays an important role in treatment for these patients not only for analgesia but also for treatment of edema with corticosteroids and adjuvant treatment with chemotherapy. The patient’s prognosis defines the appropriate treatment for spinal metastasis, with the goal of maintaining quality of life. Radiation therapy continues to be a primary treatment option and a variety of new techniques are now available to maximize the radiation dose to the tumor while minimizing the dose to the spinal cord. Surgical resection and spinal stabilization also have critical roles in the treatment armamentarium. The combination of these modalities will continue to be a vital component in the treatment of metastatic spinal cord disease. Key Words: spinal metastases, epidural spinal cord compression, radiation, corticosteroids, surgery, spine stabilization

1. INTRODUCTION Metastatic spinal cord disease represents a common complication of systemic cancer and a major cause of morbidity in cancer patients (1,2). Since the first report of spine metastasis by Dr. William Spiller in 1925 (3), this disease has proven to be a challenge with regard to both the diagnosis and management. Symptomatic spine metastasis is seen in 5–10% of patients with cancer and such lesions must be caught early and treated in an effective manner in order to salvage residual neurologic function and to prevent new neurologic deficits (4,5). From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

163

164

Part IV / Direct Complications of Cancer

These patients typically present with such signs and symptoms as pain, weakness, autonomic dysfunction, sensory loss, and ataxia (6). Several prognostic factors have been evaluated in order to adequately assess the appropriate treatment options for these patients, including extent of metastatic disease, aggressiveness of the cancer, and pre-operative function, among others. By assessing the prognosis for the patient, the appropriate treatment options that minimize additional morbidity and maximize the patient’s quality of life may be selected. These treatment options range from palliative measures, such as radiation therapy, to curative resection, and the type of treatment must be individualized for each patient (7,8).

2. EPIDEMIOLOGY Nearly 1,400,000 new cases of cancer were diagnosed in 2006 (9). Sixty to seventy percent of patients with a previous diagnosis of cancer harbor systemic neoplasia at the time of death (10). Post-mortem studies display the prevalence of skeletal metastasis in cancer patients to range from 7% to 27%. The prevalence is similar for both men and women (11). Of patients with skeletal metastasis, 36–70% have lesions to the spine (12–14). Metastatic disease to the spine can present in a variety of ways and causes significant morbidity in these patients. Metastatic lesions can present as intradural intramedullary, intradural extramedullary, or extradural in location. Extradural disease can either be isolated in the bony spine or an epidural component can be present with or without compression of the spinal cord or thecal sac. Approximately 94–98% of patients with metastasis to the spine have either vertebral or epidural involvement (11). On the other hand, intradural extramedullary and intradural intramedullary seeding are only seen in 5–6% and 0.9–2.1% of patients, respectively (15). Intradural extramedullary lesions of metastatic origin typically arise via seeding of the spinal subarachnoid space (e.g., lymphoma). This topic will primarily be discussed in chapter 12. Epidural spinal metastases (ESM) are most commonly seen in the thoracic spine (70% of cases). Disease is also seen in the lumbar spine (20% of cases) and less commonly the cervical spine (10% of cases) (12,16). Despite the incidence of metastasis, symptomatic ESMs develop in only 5–10% of patients with cancer (17–19). The incidence of metastatic spine disease varies among different tumor types. Subsequently, a higher index of suspicion must be maintained with particular cancer patients in an attempt to retain and possibly restore neurologic function. The most common malignancies to result in symptomatic ESM include breast cancer, lung cancer, and prostate cancer. Lymphoma, sarcoma, and the kidney also display a high prevalence as the primary site of neoplasia in this area. In addition, the gastrointestinal tract, melanoma, and myeloma can provide a source for metastatic disease. An unknown primary site can also play a significant role in ESM (19–21). Table 1 displays the incidence of each of these primary sites as a source of metastasis. Loblaw et al. analyzed the prevalence of ESM at diagnosis and the cumulative incidence of ESM in the 5 years preceding death among different cancer types. The overall incidence is 2.5% for all cancer with a range of 0.2% in pancreatic carcinoma Table 1 Site of Primary Tumor with Epidural Spinal Metastasis Epidural Spinal Metastasis Primary Breast Lung Prostate Lymphoma Sarcoma Kidney Myeloma Gastrointestinal Melanoma Unknown

Incidence (%) 13–22 15–19 10–18 8–10 7.5–9 6–7 4.5–5 4–5 2–4 4–11

Chapter 11 / Spinal Metastases

165

Table 2 Evaluation of ESM in Ontario Primary Myeloma Prostate Nasopharynx Breast Kidney Melanoma Small cell lung Lymphoma Non-small cell lung Cervix Unknown primary Uterus Bladder Head and neck Colorectal Stomach Ovary Leukemia Pancreas Other

Prevalence at diagnosis (%)

Cumulative Incidence (%)

1.95 0.20 0.00 0.11 0.38 0.04 0.73 0.42 0.53 0.03 0.77 0.01 0.02 0.00 0.02 0.07 0.02 0.07 0.02 0.3

7.91 7.24 6.50 5.52 4.98 3.46 3.36 2.64 2.56 2.51 1.69 1.37 0.98 0.87 0.78 0.58 0.42 0.30 0.22 1.39

to 7.9% in myeloma (Table 2) (22). This figure likely represents a significant underestimate, as only hospitalized patients with radiographically proven ESM were captured in the study.

3. PATHOGENESIS AND PATHOPHYSIOLOGY Batson, through his cadaveric experiments, identified the valveless vertebral–venous plexus. This plexus was shown to be a low-pressure system that extends from epidural and perivertebral veins to veins of the thoracoabdominal wall as well as to veins of the head and neck. Venous blood can bypass the portal, caval, and pulmonary veins via valsalva or venous obstruction, secondary to a tumor, or increasing intra-thoracic and intra-abdominal pressure resulting in flow inversion to the vertebral–venous plexus. Subsequently, a pathway is available for distant organs to spread disease (23,24). An alternative route for metastases to the bony spine and epidural space is via arterial emboli. Metastases are believed to embolize to the rich vascular network of the body spine. The vertebral body has a significantly more vascular network than the posterior elements, and consequently spinal lesions often arise more often from the former as compared to the latter (5). However, isolated involvement of the vertebral bony is observed only in 3.8% of cases. In fact, 75% of patients with ESM involve the vertebral body, pedicle and posterior elements (25). These tumors can either grow within the anterior or posterior bony elements or spread to the epidural space via venous drainage (26). In addition, metastases to the spine can spread via direct extension from the paraspinal region to the nerve roots through the neural foramina (1). The pathophysiology by which spine metastasis causes neurologic injury is a matter of some debate. Spinal cord compression is associated with endogenous neurochemical changes that lead to neuronal injury. This compression was initially thought to result in arterial ischemia. Subsequently, animal and human studies have demonstrated that compression and obstruction of the vertebral–venous plexus result in vasogenic spinal cord edema, venous hemorrhage, and ischemia (20,27). In addition to venous obstruction, spinal auto-regulatory mechanisms induce arteriolar dilatation and increased edema via induction of such enzymes as nitric oxide synthase. Cytokine production, e.g., PGF2, IL-1, IL-6, locally promotes an inflammatory response with vasodilatation and increased edema formation. In addition, animal studies display myelin loss secondary to ischemia and compression (28–31).

166

Part IV / Direct Complications of Cancer

4. PRESENTATION Patients with spinal metastases can present in a variety of ways. Symptoms secondary to metastases are not uncommonly the symptoms by which a primary malignancy is discovered. In fact, 20% of patients with cancer have signs and symptoms of ESM as the initial manifestation of their disease (32). Patients with bony spine metastasis, with or without an epidural component, most commonly present with a prolonged period of persistent back pain with a median time course of 8 weeks. Unfortunately, back pain is a frequent complaint among the general populace and the physician at least must consider spine metastases in the differential diagnosis. This is especially true in older patients and patients with pain at the level of the thoracic spine, as pain at this level is uncommon in such conditions as disc herniation. Even with the diagnostic modalities available to the physician, patients are being diagnosed very late in the course of their disease. Levack et al. performed a prospective observational study of 319 patients and found that 82% of patients at diagnosis of ESM were either unable to walk or only able to do so with help. Ninety-four percent of these patients reported approximately a 3-month history of axial spine pain (33). It is important to note that patients with compression secondary to epidural disease can still present with isolated axial spine pain without neurologic deficit or radicular symptoms. Epidural spinal compression cannot simply be excluded because a patient with back pain does not manifest myelopathy or radiculopathy. Pain may occur for a variety of reasons including pathologic fracture, local compression resulting in axial spine pain, or via nerve root impingement resulting in radicular pain. Radicular pain affecting the upper or lower extremities is seen in cervical and lumbar disease, respectively. Thoracic cord lesions associated with radicular pain present with bilateral pain radiating around the chest or upper abdomen (34). The location of both local and radicular pain can help localize the lesion within one to two vertebral segments. However, pain can be a false localizing sign and may not always correlate directly with the location of the metastasis. The differential must also include more common entities such as herniated disk disease. Such patients typically present with a history of trauma or other inciting event with an acute onset of their pain, rather than the more common insidious onset of pain symptoms seen in spine metastasis (35,36). Weakness is typically the second or third most common complaint of patients with spine metastasis and actually is both a symptom and sign of disease. Subjective weakness can actually be a manifestation of axial or radicular pain without true weakness evident on examination and is present in 76% of patients with ESM. Objective weakness is seen in 84% of patients with compressive ESM. At the time of diagnosis, approximately 50% of patients are ambulatory, 35% are paraparetic, and 15% are paraplegic. Rapid diagnosis and treatment are critical in these patients as 30% of those individuals presenting with weakness become paraplegic within one week (21). Patients with spinal metastases commonly present with numbness and paresthesias so that 51% have subjective sensory symptoms on presentation. On neurologic examination, 78% of patients actually have sensory deficits (21). These deficits can assist in localizing the metastatic lesion. However, dermatomal sensory loss or reflex loss is more predictive than a sensory level, as the level can actually be present between one and four levels below the level of disease. Patients with cervical and thoracic disease can also present with Lhermitte’s sign (37). Bowel or bladder dysfunction is seen in as many as 57% of patients. Urinary retention is the most common form of dysfunction, more common than both urinary and fecal incontinence. The degree of autonomic abnormality often correlates with the severity of motor and sensory deficits and is considered a late finding. In addition to the aforementioned presenting signs and symptoms, patients may also present with other forms of autonomic dysfunction such as absence of sweating below the lesion level and Horner’s syndrome as well as ataxia, spasticity, and syringomyelia (20,21). In assessing patients with ESM, it is important to differentiate between lesions causing myelopathy from spinal cord compression and those causing deficits from cauda equina syndrome (CES). Patients with either lesion can present with back pain, weakness, sensory deficits, or bowel and bladder dysfunction. However, the former results in upper motor neuron signs such as clonus, Babinski sign, and hyperactive reflexes. On the other hand, the latter displays unique sensory deficits such as saddle anesthesia as well as lower motor neuron signs such as hypoactive reflexes and muscle wasting (38,39).

Chapter 11 / Spinal Metastases

167

5. DIAGNOSTIC WORK-UP The diagnosis of spine metastasis is continually evolving as the diagnostic tools available to the physician continue to improve. Plain x-rays are a valuable tool in analyzing the bony spine. Plain radiographs detect bony erosion better in cortical bone than in cancellous bone. The pedicle is primarily composed of cortical bone as compared to the vertebral body. Consequently, metastasis to the pedicle is typically identified first on plain radiographs despite the higher degree of involvement in the vertebral body (16). Despite the fact that detection on plain x-ray requires 50% local bony destruction, plain radiographs display abnormalities in up to 90% of patients. Metastatic tumors are commonly lytic lesions that may present as vertebral body pathologic compression (Fig. 1). Plain radiographs also show paraspinal soft-tissue shadows and pathological fracture-dislocation (40,41). Despite these advantages, false-negative plain radiographs occur in 10–17% of patients with ESM (18). Tumors can also be osteoblastic as seen in prostate and breast metastases and difficult to identify on plain radiographs. The clinically relevant lesion can be difficult to discern with metastatic involvement of multiple vertebrae. In addition, paraspinal tumors invading the neural foramen may not show a radiographic abnormality. Bone scans are more sensitive in assessing metastatic disease than plain radiographs. Bone scans use technetium diphosphonate, a radioactive substance, to identify diseased bone that present as “hot spots” (Fig. 2). This diagnostic method also has the advantage of providing a survey of the entire skeleton. Degenerative changes seen in elderly patients can complicate the diagnosis using this method (42,43). As an alternative to a conventional bone scan, whole body positron emission tomography (PET) can be used to assess bony metastases. This imaging modality has been shown to have equal sensitivity and improved accuracy in detecting metastatic bone lesions as compared to a bone scan (44). The improved accuracy is thought to be partly secondary to the mechanism by which the modalities detect tumor involvement. The former relies upon the osteoblastic bone response to tumor, while the latter measures glucose uptake in the tumor itself by the use of a radiotracer (45–48). Consequently, PET scans are more likely to detect tumors that are at an earlier stage of growth, while bone scans are less likely to detect osteolytic and slow growing metastases (48,49). Myelography, as first brought forward by Jean Athanase Sicard, has been an important diagnostic technique in the evaluation of spinal metastasis (50). Prior to the advent of MRI, myelography was the gold standard for evaluation of these tumors (51). Myelography can be used to identify the site and extent of metastasis when MRI is not readily available, a patient is unable to tolerate MRI, or MRI is contraindicated as in patients with ferromagnetic implants (41,52). The relationship of the metastasis to the spinal cord, dura, and nerve roots can also be discerned. Typically myelography is performed via a lumbar injection of a radio-opaque dye. As some metastasis can present in multiple locations a complete spinal block can prevent possible identification of an additional metastasis. Consequently, a cisternal injection is required to complete the evaluation (41). Computed tomography (CT) can be used either as a separate modality or in combination with myelography. CT imaging is primarily useful in assessing the bony elements surrounding the spinal cord (Fig. 3). CT in combination

Fig. 1. Plain Radiographic Image. A 63-year-old female with uterine cancer presented with a complaint of isolated back pain. A pathologic compression fracture is identified via this lateral plain radiograph. In this case, no epidural component is present.

168

Part IV / Direct Complications of Cancer

Fig. 2. Bone Scan Image. This type of radionucleotide scan can detect bony metastasis (black arrow) but cannot confirm spinal cord compression. The scan detects increased signal only in the pathologic compression fracture located at L3.

Fig. 3. Computed Tomography (CT). This imaging modality displays excellent bony imaging of this pathologic compression fracture both in the sagittal (A) and axial (B) plains.

Chapter 11 / Spinal Metastases

169

Fig. 4. Magnetic Resonance Imaging (MRI). A 72-year-old male with a history of prostate cancer presented with a 2-day history of back pain and lower extremity numbness and weakness. This imaging modality displays ESM at the T4 level with spinal cord compression. (A) Sagittal T2 displaying pathologic compression fracture (black arrow). (B) Axial T2 displaying spinal cord compression (black arrow).

with myelography can greatly improve the data available from each study alone and can actually provide better anatomical detail of the spinal axis. Consequently, the total extent of the neoplasm can be demonstrated both inside and outside the spinal canal (41,53). While the aforementioned modalities can be of value in assessing spinal metastasis, magnetic resonance imaging (MRI) is now considered the modality of choice. MRI provides multiplanar imaging of the spine that is noninvasive. In addition, paravertebral soft tissue masses and bone marrow involvement can also be detected (51). MRI has been shown to be equivalent if not superior to CT myelography in detecting cord compression in ESM as well as cord atrophy (Fig. 4) (54). Signs and symptoms of the patient can be misleading with regard to tumor location. As stated previously, sensory levels can be up to three to four levels above or below the level of the lesion. Multiple levels of the spinal cord can also be involved. MRI has been shown to be the most accurate and noninvasive method to assess the entire spinal axis so that the appropriate treatment modality can be initiated (55). In addition to evaluation of the spine itself, laboratory and additional radiographic assessment are important in assessing systemic disease. In assessing the nutritional status and immunological status of the patient, a basic metabolic profile, complete blood count, and prealbumin can be obtained. Renal dysfunction either via metastasis or primary disease can be assessed via blood urea nitrogen and creatinine, while liver dysfunction, most commonly via metastatic disease, can be assessed with liver function tests (41). Tumor markers such as prostate specific antigen for prostate cancer, urine Bence Jones proteins for myeloma, CA 125 for ovarian cancer, and CEA for colorectal cancer can assist with diagnosis (56). Urinalysis, chest radiography, abdominal ultrasound, and CT imaging of the chest, abdomen, and pelvis can also be useful screening methods for malignancy (41).

6. PROGNOSIS In discussing the various treatment options of spinal metastasis, multiple factors must be considered first in order to determine the patient’s prognosis. Prognosis can be a key item in patient assessment that can influence how aggressive the treatment is for a particular patient. Tokuhashi generated such a system for assessing prognosis that is useful in predicting length of survival (Table 3). This system includes such items as general condition, number of extraspinal metastasis, the number of spinal metastasis, the extent of metastasis to internal organs, the primary site of the tumor, and the degree of spinal cord injury. This scoring system has been correlated with prognosis such that patients with a score of 9–12 are predicted to survive greater than 12 months, while patients with a score of 0–5 are predicted to survive less than 3 months (57).

170

Part IV / Direct Complications of Cancer

Table 3 Tokuhashi’s Evaluation for Prognosis Score Symptoms Karnofsky score Extraspinal metastasis Internal organ metastasis Primary site of tumor Spinal metastasis Spinal cord injury

0

1

2

10–40 >3 Unresectable Lung, stomach >3 Complete

50–70 1–2 Resectable Kidney, liver, uterus 2 Incomplete

80–100 0 No metastasis Thyroid, prostate, breast, rectum 1 None

In general, the median time of survival after diagnosis of ESM is approximately 6 months. Patients who are ambulatory at the time of diagnosis display a median survival of 8–10 months as compared to 2–4 months for nonambulatory patients. In addition, patients with slow-growing cancers such as breast and prostate cancer tend to live longer than those with fast-growing cancers such as lung cancer. The former has a median survival of 9–10 months, while the latter has a median survival of approximately 3 months (58,59).

7. TREATMENTS The treatment options for spinal metastasis are divided into three categories. These options include pharmacologic therapy, radiation therapy, and surgical resection with or without fusion. Treatment of the patient’s symptoms and adjuvant therapy, for example, corticosteroids or chemotherapy, are two roles of pharmacologic therapy. The physician typically combines medications for analgesia as well as for the control of neuropathic pain with radiation therapy and/or surgical intervention. Opiates are the primary treatment for analgesia, while amitriptyline and gabapentin are effective treatment options for neuropathic pain. These medications are a key component to palliative therapy (34).

7.1. Pharmacotherapy Pain develops not only from bony infiltration but also from pathologic fracture with or without an epidural component. In addition to pain, such fractures can lead to spinal instability requiring surgical intervention. Pharmacotherapy directed at bone turnover can provide a method in preventing pathologic compression fractures in this patient population. Bisphosphonates inhibit osteoclast activity and subsequently inhibit bony resorption, decreasing the risk of pathologic fracture. Such therapy has shown a benefit in patients with bony metastasis from multiple myeloma and breast cancer (60,61). Corticosteroids have been shown in experimental models to reduce peritumoral vasogenic spinal cord edema and transiently improve neurologic function. After initiation of steroid therapy, patients have shown a significant improvement in pain symptoms (34,41,62). A randomized trial by Sorensen et al. compared outcome in patients receiving high-dose dexamethasone and radiation therapy to those receiving radiation therapy alone. In the former group, 81% of patients were ambulatory after treatment and 59% of patients remained ambulatory after 6 months. In contrast, only 63% of patients in the latter group were ambulatory after treatment and 33% of patients remained ambulatory after 6 months. These differences displayed statistical significance identifying the importance of corticosteroids as adjuvant treatment in patients with ESM (63). Studies have also focused on the effect of high-dose bolus dexamethasone (100 mg) versus moderate-dose bolus dexamethasone (10 mg) versus no corticosteroid treatment. These studies displayed equivalent efficacy between doses with regard to improvement in pain, ambulatory status, and bladder function. The physician must consider the potential cost of corticosteroids, especially at higher dosages. Significant adverse side effects include severe psychoses, gastric ulcers, rectal bleeding, and gastrointestinal perforations (63–66). Heimdal et al. performed a retrospective study of patients who received radiation therapy in combination with corticosteroids. All patients received pretreatment with either antacids or H2 blockers prior to high-dose

Chapter 11 / Spinal Metastases

171

corticosteroid therapy. Despite preventive measures, two patients developed gastric perforations and two patients developed gastrointestinal bleeding, one of which proved fatal. A subsequent cohort of patients received a lower dose corticosteroid regimen of 16 mg tapered over two weeks. These patients did not experience serious side effects and the ambulatory outcome was similar to those patients receiving the high-dose corticosteroid regimen (66). In addition, the use of corticosteroids has been analyzed in patients with less severe metastatic disease. Maranzano et al. analyzed 20 consecutive patients with ESM causing less than 50% narrowing of the spinal canal and no neurologic deficit in a phase II trial. Patients treated with corticosteroids and radiation therapy showed no additional survival benefit and equivalent return of neurologic function in comparison to patients treated with radiation therapy alone (67).

7.2. Chemotherapy Chemotherapeutic agents can be a valuable treatment option in ESM. The primary use of these agents is dependent on the chemosensitivity or chemoresistance of the particular tumor. Treatment must be designed to maintain neurologic function and maximize quality of life. Consequently, chemotherapy is typically used as adjuvant therapy along with radiation therapy and/or surgical resection in tumors with uncertain or limited chemosensitivity. On the other hand, the role of chemotherapy in chemosensitive tumors has been a matter of debate. Patients with symptomatic chemosensitive metastases have most often been given chemotherapy in combination with other therapeutic modalities. However, patients with chemosensitive tumors have shown good neurologic improvement with chemotherapy alone. Especially early in the course of the disease, chemosensitive tumors are likely to respond to chemotherapy. These tumors include germ cell tumors and hematological malignancies, such as lymphoma (68,69). In addition, chemotherapy can be considered as a single mode of treatment for patients who have previously received radiation or surgery and are not candidates for further treatment (34).

7.3. Radiation Therapy Historically, decompression via laminectomy was considered the primary treatment for spinal metastasis. Studies were subsequently conducted in the 1970s and 1980s comparing radiation therapy alone to laminectomy followed by adjuvant radiation therapy. These studies displayed similar rates of neurologic improvement. Consequently, radiation therapy became the standard as primary treatment, while surgery was reserved for patients who deteriorated during or failed to improve after radiation therapy (70–72). As more advanced surgical techniques for resection and stabilization have been developed; the role of radiotherapy has also been modified. The prognosis for patients with spinal metastases receiving radiotherapy depends highly on the radiosensitivity of the primary tumor. The most radiosensitive tumors that commonly metastasize to the spine include breast cancer, small cell lung cancer, prostate cancer, myeloma, and lymphoma. Patients with these tumors tend to show improved functional recovery and better tumor control rates as compared to patients with radioresistent tumors such as melanoma and renal cell carcinoma. Patients with radioresistent tumors can, however, still obtain significant pain control from radiotherapy (73,74). A variety of techniques are available for the effective delivery of radiotherapy. Such techniques include fractionated radiotherapy, stereotactic radiotherapy, and intensity-modulated radiotherapy. In general, the use of radiotherapy is limited by the level of radiation tolerance of the spinal cord. Such factors as hypertension, advanced age, prior spinal cord pathology, combination chemotherapy, and immunocompromised status may lower the spinal cord tolerance to radiation (75). This tolerance level is not completely understood, limiting the radiation dose that can be provided to the spinal cord in order to prevent such serious complications as radiation myelitis or myelopathy. The diagnosis of radiation myelopathy requires the combination of clinical presentation as well as radiographic imaging. Patients usually present with progressive rostral spread of sensorimotor symptoms within months of radiation therapy. Within one year of symptomatic onset, MR imaging typically displays cord swelling as high signal intensity within the spinal cord on T2-weighted images resembling that of syringomyelia. T1-weighted images often display post-gadolinium enhancement. The final clinical level usually corresponds to a level well within the boundaries of the length of abnormal signal. The combination of clinical presentation and abnormalities on T1- and T2-weighted imaging is used to differentiate between radiation toxicity and tumor recurrence (76–79).

172

Part IV / Direct Complications of Cancer

Conventional external-beam radiation therapy (EBRT) is an excellent treatment option for pain associated with bony spinal metastases. As stated previously, bone metastases are commonly seen in lung, breast, and prostate cancer. Patients presenting with spinal metastases without an epidural component can display significant axial spine pain from bony infiltration of tumor. Local irradiation has shown an overall response rate of approximately 85% with complete relief of pain in 50% of patients (80). Various studies have looked at the role of radiotherapy with or without the presence of spinal cord compression. Such studies have identified compression as a negative predictor for ambulation with radiotherapy alone (81,82). The pain associated with bone metastases can be treated with various dose schedules ranging from 8 Gy in one fraction to 40 Gy in 20 fractions. A comparison of patients with isolated bony metastases treated with EBRT of 8 Gy in one fraction and 30 Gy in 10 fractions did not display a significant difference in pain control (83,84). A variety of studies have analyzed these dose schedules in patients with metastasis with epidural extension. In a retrospective analysis by Rades et al., 252 patients with metastasis from non-small cell lung cancer were given either a short course of EBRT (8 Gy in 1 fraction or 20 Gy in 5 fractions) or a long course of EBRT (30 Gy in 10 fractions or 37.5 Gy in 15 fractions). In comparing these groups, no significant difference in functional outcome was seen between the short and long term treatment arms nor was a difference seen between the different short term fractionation regimens (85). In another retrospective analysis of 1300 patients compared five different treatment regimens between 8 Gy in 1 fraction and 40 Gy in 20 fractions. Functional outcome was similar between all groups; however, the more protracted regimens were associated with a lower rate of local recurrence (86). A protracted regimen is subsequently advantageous in patients with a longer life expectancy. In contrast, a randomized trial of 300 patients with a life expectancy of less than 6 months treated patients with regimens of either 16 Gy in 2 fractions or 30 Gy in 8 fractions. No significant difference in either efficacy or toxicity was seen in comparing the two groups (87). Patients with a relatively short life expectancy experience similar palliation with short courses of radiation without the inconvenience of more protracted radiation courses. As stated previously, one goal of therapy is to minimize the dose to normal spinal cord in order to prevent complications. However, up to 25% of patients can develop recurrence after radiotherapy and 64% of patients with recurrence within 3 months of treatment will have this recurrence within two vertebral bodies of the initial disease. Initial treatment, therefore, should be extended to include one to two vertebral bodies adjacent to the metastasis (88). EBRT can subject the spinal cord to significant radiation exposure. Stereotactic radiotherapy provides an alternative modality for focused high-dose radiation to the tumor while minimizing radiation to the adjacent spinal cord. Accurate targeting requires localization of multiple radiation beams to converge on the lesion of interest at a high dose. The typical dose ranges from 8 to18 Gy. This treatment can be administered between one to three sessions on an outpatient basis. Current available systems include Helical Tomotherapy® (TomoTherapy Incorporated, WI, USA), Novalis® (BrainLAB, Heinstetten, Germany) and CyberKnife® (Accuray Incorporated, Sunnyvale, CA, USA) (89). Only a few studies have assessed the utility of stereotactic radiotherapy in ESM. This modality has produced significant improvement in pain in a recent study with the use of CyberKnife. In addition, the tumor control rate was 100% in lesions without previous irradiation (90). Radiosurgery is considered safer for recurrent tumors than traditional methods, as repeat fractionated radiotherapy poses a significant risk at surpassing the radiation tolerance of the spinal cord. However, the optimal dose for radiosurgery and the tolerance of the spinal cord to single fractions are not known. Radiosurgery, like other forms of radiotherapy, does not address the issue of spinal instability. Intensity-modulated radiotherapy (IMRT) offers an alternative to the aforementioned treatment modalities. Helical TomoTherapy, Novalis, or CyberKnife delivers IMRT. While radiosurgery provides radiation beams of equal intensity, IMRT individualizes the intensity of each beam in order to minimize the radiation exposure to the surrounding structures (91). This modality, like radiosurgery, can be a valuable tool in the re-treatment of spinal metastasis. A study by Milker-Zabel et al. analyzed the role of IMRT for tumor recurrence after previous fractionated radiotherapy. The group assessed eighteen patients with 19 lesions and each received IMRT. The study displayed an overall local control rate of 94.7%. Thirteen of 16 patients experienced significant pain relief, while 5 of 12 patients displayed neurologic improvement. Patients in the study did not show radiation toxicity of clinical significance (92).

Chapter 11 / Spinal Metastases

173

7.4. Surgical Management The role of surgery in the treatment of spine metastasis has changed as the techniques available for spinal reconstruction have improved. Despite the variety of options in a surgeon’s armamentarium, the ability to maintain a patient’s quality of life must be of utmost importance. A more extensive reconstruction correlates with a longer and more painful recovery period, in addition to the recovery required for treatment of the primary disease. Consequently, prognosis is a key factor in deciding the aggressiveness of treatment for a particular patient. Tomita et al. clarified the correlation between length of survival and surgical treatment goals. In this study, they analyzed the growth rate of primary tumor, the presence of visceral metastasis, and the presence and number of bone metastasis. Slow-growth tumors such as breast and thyroid cancer equated to 1 point; moderate-growth tumors such as renal cell carcinoma equated to 2 points; rapid-growth tumors such as lung and gastrointestinal cancer equated to 4 points. Patients with visceral metastasis equated to 2 points if they were treatable, while those with untreatable lesions equated to 4 points. Finally, patients with solitary or isolated bony metastasis equated to 1 point, while those with multiple metastases equated to 2 points. The group then separated patients into prognostic scores from 2 to 10. Patients with a prognostic score of 2–3 had a treatment goal of long-term local control via wide or marginal excision, for example, en bloc spondylectomy, with a mean survival time of 38.2 months. Patients with a prognostic score of 4–5 had a treatment goal of middle term local control via intralesional excision with a mean survival time of 21.5 months. Patients with a prognostic score of 6–7 had a treatment goal of short-term palliation via simple decompression and stabilization with a mean survival time of 10.1 months. Finally, patients with a prognostic score of 8–10 had supportive care only with a mean survival time of 5.3 months (7). The indications for surgery as compared to radiation therapy remain controversial. The initial surgical treatment described for these tumors was laminectomy for the purpose of decompression. The typical technique involves removal of the posterior arch of the spinal canal from one level above to one level below the level of the metastasis (34). However, a randomized trial comparing laminectomy plus radiation therapy versus radiation therapy alone displayed no difference in outcome. In addition, laminectomy plus tumor induced bony deconstruction can lead to instability (93). A variety of techniques are now available to effectively decompress the spinal cord and provide stability when necessary. The best surgical treatment options depend not only on the level of the disease, that is, cervical, thoracic, or lumbar, but also on the location of the disease, that is, anterior or posterior. Metastasis to the vertebral body can be destructive both to cortical as well as cancellous bone. Pathologic compression fractures can subsequently develop, resulting in significant pain to the patient. Percutaneous vertebroplasty and kyphoplasty are valuable options for treating pain from lesions either isolated to the vertebral body or lesions with a mild epidural component. In this treatment, the surgeon injects polymethylmethacrylate (PMMA) either directly into the vertebral body in vertebroplasty or after expansion of the collapsed vertebral body with a balloon in kyphoplasty. The surgeon usually injects bone cement via a transpedicular route; however, an anterolateral, intercostovertebral, and a posterolateral route are used in the cervical, thoracic, and lumbar regions, respectively. Patients can receive pain relief within 24–48 hours following therapy and have shown to maintain improved pain control upon 2-year follow-up examination (94,95). As stated previously, the vertebral body is the most frequent portion of the bony spine affected by spine metastasis. Therefore, laminectomy is most often not the appropriate operation because it provides inadequate surgical exposure for tumor resection as the tumor bulk is most often not primarily in the posterior spinal elements. Attempting to produce adequate exposure may also worsen spinal instability and create deformity (12). Lesions in the vertebral body not amenable to percutaneous vertebroplasty or kyphoplasty, therefore, require a more appropriate avenue by which the spinal cord can be decompressed. The surgeon typically performs this decompression with a corpectomy via an anterior approach. The anterior approach requires access through the neck musculature with cervical lesions, transthoracic with thoracic lesions, and retroperitoneal with lumbar lesions. Metastatic disease to the thoracic spine can be a surgical challenge with anterior column disease, as a direct anterior transthoracic approach would pose many surgical risks. An anterolateral approach is therefore used such that the patient is positioned in a lateral decubitus position and the chest is entered between the ribs laterally (96). For lumbar lesions, this approach is also used to obtain a retroperitoneal plane to the spinal column; however, the iliac crest may limit the ability to reach the lower lumbar vertebrae.

174

Part IV / Direct Complications of Cancer

Following a corpectomy/vertebrectomy, the space filled by the vertebral body must be reconstructed to the appropriate height. The typical material in degenerative spine disease is autologous bone or bone allograft. However, most patients with metastatic lesions will or have already received radiation therapy that decreases the rate of a bony fusion. PMMA is an alternative for reconstruction in the cervical or thoracic spine, while a titanium expandable cage can provide an effective alternative in the thoracic or lumbar spine. The former requires the addition of a plate and screw construct to aid in stability (97). Sawaya et al. recently studied 72 patients with primarily thoracic spine anterior column disease that underwent a vertebrectomy and fusion procedure; 76% of these patients displayed improvement in neurologic function. In addition, 77% of patients nonambulatory prior to surgery regained ambulatory capacity post-operatively (12). While the anterior approach to metastatic disease can be effective in treating the majority of patients, the extent of a patient’s primary disease may not warrant such an extensive tumor resection. A posterior approach for decompressive laminectomy and fusion with pedicle screw instrumentation provides a palliative surgical option. This approach is also an effective treatment modality for tumors involving only the posterior elements of the spinal canal. In addition, pain secondary to a pathologic compression fracture of one or multiple levels with associated instability can be effectively treated via a posterior alone approach. In a retrospective study by Oda et al., 32 patients with extensive metastatic disease in the cervico-thoracic spine underwent posterior decompression and fusion alone. Ninety-four percent of these patients maintained pain relief, neurologic function, and spinal stability throughout the survival period (98–100). In the appropriate patient, a combination of both anterior and posterior approaches is used to treat metastatic disease invading both the anterior and posterior vertebral columns (Fig. 5). These surgical approaches are performed as two separate operations or as a two-staged single operation. Tomita et al. first described total en bloc spondylectomy with complete removal of the vertebrae from a posterior alone approach (101,102). This technique, however, poses a risk of injury to major vessels or visceral organs as these structures are not directly visualized. The combined approach, on the other hand, provides direct visualization of the entire extent of the tumor. A study by Fourney et al. assessed 24 patients who underwent a combined anterior and posterior tumor resection. Pain symptoms improved in 96% of patients while 62% exhibited neurologic improvement (103). Patchell et al. recently compared 50 patients treated by the surgical procedure appropriate to the site of the metastasis followed by radiation therapy with 51 patients treated with radiation alone. In this randomized, multi-institutional, nonblinded trial, the treatment group randomly assigned patients with metastatic spinal cord

Fig. 5. Anterior and Posterior (A/P) Surgical Intervention. Metastatic disease involving both the anterior and posterior bony spinal columns requires a combined anterior and posterior surgical intervention. These plain radiographs display both lateral (A) and A/P (B) views following this surgical procedure.

Chapter 11 / Spinal Metastases

175

compression to two different treatment arms. In comparing the two groups after treatment, 84% of patients were ambulatory in the former group while only 57% were ambulatory in the latter group. Thirty-two patients were nonambulatory prior to treatment, 50% in each group. Of these patients, 62% of the former group were ambulatory after treatment, while only 19% of the latter group were ambulatory after treatment. Finally, patients in the former group were able to retain the ability to walk for a mean of 122 days, while patients in the latter group were only able to retain the ability to walk for a mean of 13 days. This study clearly showed that direct surgical decompression, and fusion where appropriate, followed by radiation therapy was superior to radiation therapy alone. However, a limitation of the study is patient selection bias as most patients with ESM do not meet the eligibility criteria for the study, and in fact it required 10 years for the multi-center study to enroll 101 patients. Additionally, the results of this trial cannot be used to justify surgery in all patients with ESM and apply only to patients comparable to those included in the study (104).

8. INTRADURAL INTRAMEDULLARY SPINAL CORD METASTASIS Intradural intramedullary spinal cord metastasis (ISCM) is limited in frequency to only 0.9–2.1% of cancer patients (105–107). Approximately 50% of ISCM arise from lung carcinoma and the majority of these cases are small cell carcinoma. Breast cancer, lymphoma, kidney cancer, melanoma, gastrointestinal cancer, ovarian cancer, and tumors of unknown primary are other causes of ISCM (107–112). Table 4 displays the incidence of each of these primary sites as a source of metastases with regard to ISCM. Most ISCM are thought to spread via emboli through a secondary capillary network to penetrating arteries of the spinal cord (23,110,112). Alternatively, ISCM may originate via direct extension from leptomeningeal disease and subsequently spread to the cord parenchyma (107,112). Pain is a common presenting sign in ISCM. In addition, patients may present with weakness and display a similar manor of rapid progression to paraplegia like ESM (114). However, true motor weakness typically follows sensory disturbances in ISCM as they are most commonly located in the posterior cord (115). The presence of a Brown-Sequard syndrome can also be a common initial finding and help differentiate between ISCM and ESM (115). MRI is the primary imaging modality for detecting cord enlargement in ISCM. As in ESM, opiates for analgesia and gabapentin for neuropathic pain are commonly used. Patients have shown significant relief of pain symptoms as well as transient improvement in neurologic function with the use of corticosteroids (107). Treatment of ISCM has primarily been based on anecdotal experience and case series, as no prospective trials on treatment have been performed. EBRT with or without corticosteroids has been the most effective treatment of ISCM (112). Clinical response primarily depends on the duration of symptoms, the degree of neurologic deficit, and the radiosensitivity of the tumor. Standard management consists of fractionated radiotherapy to the ISCM. As pathologic evidence indicates that ISCM are multifocal in as many as 30% of patients, radiation therapy to the entire spinal cord could be a treatment option (116,117). However, the consequences of bone marrow suppression associated with total spinal irradiation has limited this extensive treatment modality (111,116). Stereotactic radiosurgery has a potential role in treating these lesions; however, this modality has only been shown Table 4 Site of Primary Tumor with Intradural Intramedullary Spinal Metastasis Primary Lung Breast Lymphoma Kidney Melanoma Gastrointestinal Ovarian Unknown

Incidence (%) 47–54 11–14.5 4–12 4–9 3.6–9 3–7.3 0.8–1.1 1.8–6

176

Part IV / Direct Complications of Cancer

in the literature to be effective in primary vascular tumors (118). The true issue is discerning the proximity of tumor to functioning spinal cord and limiting radiation exposure to this tissue. The role of surgery in ISCM remains a matter of debate. Seventy-five percent of patients with ISCM display a one month time period of progression from the first symptom of disease to paraplegia. However, patients with rapidly progressive neurologic deficit have shown improved neurologic outcome with early surgical management (107). In such patients, the objective of surgery is maximal removal of the lesion via microsurgical resection with preservation of existing neurologic function (114). Focal radiation can then be applied to the involved area, especially in patents with evidence of residual disease (109).

9. CONCLUSIONS Spinal cord metastases are a common complication of systemic malignancy. ESM most commonly stem from lung, breast, and prostate cancer, while greater than 50% of ISCM stem from lung cancer alone. Patients may present with a variety of symptoms, most notably pain and weakness. The ultimate goals in managing these patients include maximizing both length of survival and quality of life. These goals can best be reached via early, accurate diagnosis followed by the appropriate treatment for a particular patient. As imaging modalities have improved, delineating the exact location and extent of disease has become significantly more accurate. Despite the advantages of MR imaging, other imaging modalities such as CT and plain radiographs still play a valuable role in diagnosis. Pharmacotherapy plays an important role in treatment for these patients not only for analgesia but also for treatment of edema with corticosteroids and adjuvant treatment with chemotherapy. The patient’s prognosis defines the appropriate treatment for spinal metastasis, with the goal of maintaining that patient’s quality of life. Radiation therapy continues to be a primary treatment option and a variety of new techniques are now available to maximize the radiation dose to the tumor while minimizing the dose to the spinal cord. Surgical resection and spinal stabilization also have critical roles in the treatment armamentarium. The combination of these different modalities will certainly continue to be a vital component in the treatment of metastatic spinal cord disease.

REFERENCES 1. Schiff D. Spinal cord compression. Neurol Clin N Am. 2003;21:67–87. 2. Mut M, Schiff D, Shaffrey ME. Metastasis to nervous system: spinal epidural and intramedullary metastases. J Neurooncol. 2005 Oct;75(1):43–56. 3. Spiller WG, Rapidly developing paraplegia associated with carcinoma. Arch Neurol Psychiatry. 1925;13:471. 4. Bilsky MH, Shannon FJ, Sheppard SS et al. Diagnosis and management of a metastatic tumor in the atlantoaxial spine. Spine. 2002;27(10):1062–1069. 5. Barron KD, Hirano A, Araki S. Experiences with metastatic neoplasms involving the spinal cord. Neurology. 1959;9:91–106. 6. Townsend CM. Intraspinal tumors. In: Sabiston Textbook of Surgery, 17th ed. Philadelphia: Saunders, 2004:2150–2152. 7. Tomita K, Kawahara N, Kobayashi T et al. Surgical strategy for spinal metastasis. Spine. 2001;26(3):298–306. 8. Weigel B, Maghsudi M, Neumann C et al. Surgical management of symptomatic spinal metastasis. Spine. 1999;24(21):2240–2246. 9. American Cancer Society. Cancer Facts and Figures 2006. Atlanta: American Cancer Society;2006. 10. Wise JJ, Fischgrund JS, Herkowitz HN et al. Complication, survival rates, and risk factors of surgery for metastatic disease of the spine. Spine. 1999;24(18):1943–1951. 11. Tubiana-Hulin M. Incidence, prevalence and distribution of bone metastasis. Bone. 1991;12(Suppl 1):9–10. 12. Perrin RG McBroom RJ: Anterior versus posterior decompression for symptomatic spinal metastasis. Can J Neurol Sci. 1987;14:75–80. 13. Raj PP. Cancer pain syndromes. In: Textbook of Regional Anesthesia, 1st ed. Lubbock, Churchill Livingstone, 2002: 565–566. 14. Browner B. Metastatic disease of the spine. In: Skeletal Trauma: Basic Science, Management, and Reconstruction, 3rd ed. Philadelphia: Saunders, 2003: 405–419. 15. Goetz CG. Direct metastatic disease. In: Textbook of Clinical Neurology, 2nd ed. Philadelphia: Saunders, 2003:1042–1051. 16. Gokaslan ZL, York JE, Walsh GL et al. Transthoracic vertebrectomy for metastatic spinal tumors. J Neurosurg. 1998;89:599–609. 17. Kim RY. Extradural spinal cord compression from metastatic tumor. Ala Med. 1990;60:10–15. 18. Bach F, Larsen BH, Rohde K et al. Metastatic spinal cord compression: occurrence, symptoms, clinical presentations and prognosis in 398 patients with spinal cord compression. Acta Neurochir (Wien) 1990;107:37–43. 19. Sorenson PS, Borgeson SE, Rasmussen B et al. Metastatic epidural spinal cord compression: results of treatment and survival. Cancer. 1990;65(7):1502–1508. 20. Gabriel K, Schiff D. Metastatic spinal cord compression by solid tumors. Semin Neurol. 2004;24(4):375–383. 21. Grant R, Papadopoulos SM, Greenberg HS. Metastatic epidural spinal cord compression. Neurol Clin. 1991;9:825–841. 22. Loblaw DA, Laperriere NJ, Mackillop. A population-based study of malignant spinal cord compression in Ontario. Clin Oncol. 2003;15:211217.

Chapter 11 / Spinal Metastases

177

23. Batson OV: The role of the vertebral veins in metastatic process. Ann Intern Med. 1942;16:38. 24. Abeloff MD. Bone metastasis. In: Clinical Oncology, 3rd ed. Churchill Livingstone, New York 2004:1091–1123. 25. Khaw FM, Worthy SA, Gibson MJ et al. The appearance on MRI of vertebrae in acute compression of the spinal cord due to metastases. J Bone Joint Surg Br. 1999;81-B(5):830–834. 26. Arguello F, Baggs LB, Duerst RE et al: Pathogenesis of vertebral metastasis and epidural spinal cord compression. Cancer. 1990;65: 98–106. 27. Byrne TN. Spinal cord compression from epidural metastasis. N Engl J Med. 1992;327(9):614–619. 28. Siegal T. Spinal cord compression: from laboratory to clinic. Eur J Cancer. 1995;31A(11):1748–1753. 29. Siegal T, Siegal TZ, Sandback U et al. Experimental neoplastic spinal cord compression: evoked potentials, edema, prostaglandins, and light and electron microscopy. Spine. 1987;12(5):440–448. 30. Ishikawa M, Sekizuka E, Krischek Bet al. Role if nitric oxide in the regulation of spinal arteriolar tone. Neurosurgery. 2002;50(2): 371–377. 31. Kato A, Ushio Y, Hayakawa T et al. Circulatory disturbance of the spinal cord with epidural neoplasm in rats. J Neurosurg. 1985;63(2):260–265. 32. Schiff D, O’Neil BP, Suman VJ. Spinal epidural metastasis as the initial manifestation of malignancy: clinical features and diagnostic approach. Neurology. 1997;49:452–456. 33. Levack P, Graham J, Collie D et al. Don’t wait for a sensory level: listen to the symptoms: a prospective audit of the delays in diagnosis of malignant cord compression. Clin Oncol (R Coll Radiol). 2002;14(6):474–480. 34. Spinazzé S, Caraceni A, Schrijvers D. Epidural spinal cord compression. Crit Rev Oncol Hematol. 2005;56(3):397–506. 35. Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol. 1978;3:40–51. 36. Love JG. The differential diagnosis of intraspinal tumors and protruded intervertebral disks and their surgical treatment. J Neurosurg. 1944;1:275–290. 37. Prasad D, Schiff D. Malignant spinal cord compression. Lancet Oncol 2005;6:15–24. 38. Maxwell M, Chir MB, Borges LF et al. Renal cell carcinoma: a rare source of cauda equina metastasis. J Neurosurg (Spine 1). 1999;90:129–132. 39. Shapiro S. Medical realities of cauda equina syndrome secondary to lumbar disk herniation. Spine. 2000;25(3):348–351. 40. Brice J, McKissock WS. Surgical treatment of malignant extradural spinal tumors. BMJ. 1965;i:1341–1344. 41. Jacobs WD, Perrin RG. Evaluation and treatment of spinal metastasis: an overview. Neurosurg Focus. 2001;11(6):1–11. 42. Buckwalter JA, Brandser EA. Metastatic disease of the spine. Am Fam Physician. 1997;55:1761–1768. 43. Ayyathurai R, Mahaptra R, Rajasudaram R et al. A study on staging bone scans in newly diagnosed prostate cancer. Urol Int. 2006;76:209–212. 44. Hsia TC, Shen YY, Yen RF et al. Neoplasma. 2002;49(4):267–271. 45. Thrupkaew AK, Henkin RE, Quinn IL. False negative bone scans in disseminated metastatic disease. Radiology. 1974;113:383–386. 46. Gulenchyn KY, Papoff W. Technetium-99m MDP scintigraphy; an insensitive tool for the detection of bone marrow metastasis. Clin Nucl Med. 1987;12:45–46. 47. Moog F, Bangerter M, Kotzerke J et al. 18-F-fluorodeoxyglucose–positron emission tomography as a new approach to detect lymphomatous bone marrow. J Clin Oncol. 1998;16:603–609. 48. Wu HC, Yen RF, Shen YY et al. Comparing whole-body 18F-2-deoxyglucose positron emission tomography and technetium-99m methylene diphosphate bone scan to detect bone metastasis in patients with renal cell carcinomas: a preliminary report. J Cancer Res Clin Ocol. 2002;128:503–506. 49. Jacobson AF. Bone scanning in metastatic disease. In: Collier BD Jr, Fogelman I, Rosenthall L, eds. Skeletal Nuclear Medicine. St. Louis:Mosby, 1996: 87–123. 50. Sicard JA, Forestier J. Méthode radiographique d’expolration de la cavité épidurale par le lipiodol. Rev Neurol. 1921;37:1264–1266. 51. Husband DJ, Grant KA, Romaniuk CS. MRI and the diagnosis and treatment of suspected malignant spinal cord compression. Br J Radiol. 2001;74:15–23. 52. Stark RJ, Henson RA, Evans SJ. Spinal metastasis: a retrospective survey from a general hospital. 1981;105(1):189–213. 53. Resjó M, Harwood-Nash DC, Fitz CR et al. CT metrizamide myelography for intraspinal and paraspinal neoplasms in infants and children. AJR. 1979;132:367–372. 54. Karnaze MG, Gado MH, Sartor KJ et al. Comparison of MR and CT myelography in imaging the cervical and thoracic spine. AJR AM J Roentgenol. 1988;150(2):397–403. 55. Cook AM, Lau TN, Tomlinson MJ et al. Magnetic resonance imaging of the whole spine in suspected malignant spinal cord compression: impact on management. Clin Oncol (R Coll Radiol). 1998;10(1):39–43. 56. Perkins GL, Slater ED, Sanders GK. Serum tumor markers. Am Fam Physician. 2003;68(6):1075–1082. 57. Tokuhashi Y, Mastsuzaki H, Toriyama S et al. Scoring system for the perioperative evaluation of metastatic spine tumor prognosis. Spine. 1990;15:1110–1113. 58. Marazano E, Latini P, Beneventi S, et al. Comparison of two different radiotherapy schedules for spinal cord compression in prostate cancer. Tumori. 1998;84:472–477. 59. Maranzano E, Latini P. Effectiveness of radiation therapy without surgery in metastatic spinal cord compression: final results from a prospective trial. Int J Radiation Oncol Biol Phys. 1995;32:959–967. 60. Berenson JR, Lichtenstein A, Porter L et al. Long-term pamidronate treatment for advanced multiple myeloma patients reduces skeletal events. Myeloma Aredia Study Group. J Clin Oncol. 1998;16:593–602. 61. Hortobagyi GN, Theriault RL, Lipton A et al. Long-term prevention of skeletal complications of metastatic breast cancer with pamidronate. Protocol 19 Aredia Breast Cancer Study Group. J Clin Oncol. 1998;16:2038–2044.

178

Part IV / Direct Complications of Cancer

62. Ushio Y, Posner R, Kim JH et al. Experimental spinal cord compression by epidural neoplasm. Neurology. 1977;27422–429. 63. Sorensen S, Helweg-Larsen S, Mouridsen H et al. Effect of high-dose dexamethasone in carcinomatous metastatic spinal cord compression treated with radiotherapy: a randomized trial. Eur J Cancer. 1994;30A:22–27. 64. Loblaw DA, Perry J, Chambers A et al. Systematic review of diagnosis and management of malignant extradural spinal cord compression: the cancer care Ontario practice guidelines initiative’s neuro–oncology disease site group. J Neurooncol. 2005;23(9):2028–2037. 65. Vecht CJ, Haaxma-Reiche H, van Putten WL et al. Initial bolus of conventional versus high-dose dexamethasone in metastatic spinal compression. Neurology. 1989;39:1255–1257. 66. Heimdal K, Hirschberg H, Slettebo H et al. High incidence of serious side effects of high-dose dexamethasone treatment in patients with epidural spinal cord compression. J Neurooncol. 1992;12:141–144. 67. Maranzano E, Latini P, Beneventi S et al. Radiotherapy without steroids in selected metastatic spinal cord compression patients: a phase II trial. Am J Clin Oncol. 1996;19:179–183. 68. Siegal T. Spinal epidural involvement in haematological tumors: clinical features and therapeutic options. Leuk Lymphoma. 1991;5: 101–110. 69. Cooper K, Bajorin D, Shapiro W et al. Decompression of epidural metastasis from germ cell tumors with chemotherapy. J Neurooncol. 1990;8(3):275–280. 70. Barcena A, Lobato RD, Rivas JJ et al. Spinal metastatic disease: analysis of factors determining functional prognosis and the choice of treatment. Neurosurgery. 1984;15:820–827. 71. Klimo P, Jr, Kestle JR, Schmidt MH. Clinical trials and evidence-based medicine for metastatic spinal disease. Neurosurg Clin N Am. 2004;15:549–564. 72. Wu AS, Fourney DR. Evolution of treatment for metastatic spine disease. Neurosurg Clin N Am. 2004;15:401–411. 73. Helweg-Larsen S, Sorensen PS, Kreiner S. Prognostic factors in metastatic spinal cord compression: a prospective study using multivariate analysis of variables influencing survival and gait function in 153 patients. Int J Radiat Oncol Biol I. 2000;46:1163–1169. 74. Helwig-Larsen S, Rasmussen B, Sorensen PS. Recovery of gait after radiotherapy in paralytic patients with metastatic epidural spinal cord compression. Neurology. 1990;40:1234–1236. 75. Schultheiss TE. Spinal cord radiation tolerance. Int J Radiat Oncol Biol Phys. 1994;30:735–736. 76. Murakami H, Kawahara N, Yahata T et al. Radiation myelopathy after radioactive iodine therapy for spine metastasis. Br J Radiol. 2006;79:e45–e49. 77. Maddison P, Southern P, Johnson M. Clinical and MRI discordance in a case of delayed radiation myelopathy. J Neurol Neurosurg Psychiatry. 2000;69:563–564. 78. Wang PY, Shen WC, Jan JS. MR imaging in radiation myelopathy. Am J Neuroradiol.1992;13:1049–1055. 79. Michikawa M, Wada Y, Sano M, et al. Radiation myelopathy: significance of gadolinium–DTPA enhancement in the diagnosis. Neuroradiology. 1991;33:286–289. 80. Hoskin PJ. Scientific and clinical aspects of radiotherapy in the relief of bone pain. Cancer Surv. 1988;7:69–86. 81. Maranzano E, Latini P. Effectiveness of radiation therapy without surgery in metastatic spinal cord compression: final results from a prospective trial. Int J Radiol Oncol Biol Phys. 1995;32:959–967. 82. Leviov M, Dale J, Stein M et al. The management of metastatic spinal cord compression: a radiotherapeutic success ceiling. Int J Radiol Oncol Biol Phys. 1993;27:231–234. 83. Chow E, Danjoux C, Wong R et al. Palliation of bone metastasis: a survey of patterns of practice among Canadian oncologists. Radiother Oncol. 2000;56:305–314. 84. Wu JS, Wong RK, Lloyd NS et al. Radiotherapy fractionation for the palliation of uncomplicated painful bone metastasis: an evidence-based practice guideline. BMC Cancer. 2004;4:71. 85. Rades D, Stalpers LJ, Schulte R et al. Defining the appropriate radiotherapy regimen for metastatic spinal cord compression in non-small cell lung cancer patients. Eur J Cancer. 2006;42(8):1052–6. 86. Rades D, Stalpers LJ, Veninga T et al. Evaluation of five radiation schedules and prognostic factors for metastatic spinal cord compression. J Clin Oncol. 2005;23:3366. 87. Marazano E, Bellavita R, Rossi R et al. Short-course versus split-course radiotherapy in metastatic spinal cord compression: results of a phase III, randomized, multicenter trial. J Cln Oncol 2005;23:3358. 88. Kaminski HJ, Diwan VG, Ruff RL. Second occurrence of spinal epidural metastasis. Neurology. 1991;41:744–746. 89. Witham TF, Khavkin YA, Gallia GL et al. Surgery Insight: current management of epidural spinal cord compression from metastatic spine disease. Nat Clin Pract Neurol. 2006;2(2):87–94. 90. Degen JW, Gagnon GJ, Voyadzis JM et al. CyberKnife stereotactic radiosurgical treatment of spinal tumors for pain control and quality of life. J Neurosurg. 2005; 2: 540–549. 91. Klimo P, Schmidt MH. Surgical management of spinal metastasis. Oncologist. 2004;9:188–196. 92. Milker-Zabel S, Zabel A, Thilmann C et al. Clinical results of re-treatment of vertebral bone metastases by stereotactic conformal radiotherapy and intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys. 2003;55:162–167. 93. Young RF, Post EM, King GA. Treatment of spinal epidural metastasis: randomized prospective comparison of laminectomy and radiotherapy. J Neurosurg. 1980;53:741–748. 94. Purkayastha S, Gupta AK, Kapilamoorthy TR et al. Percutaneous vertebroplasty in the management of vertebral lesions. Neurol India. 2005;53(2):167–173. 95. Sabharwal T, Salter R, Adam A et al. Image-guided therapies in orthopedic oncology. Orthop Clin N Am. 2006;37:105–112. 96. Huang TJ, Hsu RWW, Li YY et al. Minimal access spinal surgery (MASS) in treating thoracic spine metastasis. Spine. 2006;31(16):1860–1863.

Chapter 11 / Spinal Metastases

179

97. Miller DJ, Lang FF, Walsh, GL et al. Coaxial double-lumen methylmethacrylate reconstruction in the anterior cervical and upper thoracic spine after tumor resection. J Neurosurg (Spine 2). 2000;92:181–190. 98. Fourney DR, Abi-Said D, Lang FF et al. Use of pedicle screw fixation in the management of malignant spinal disease: experience in 100 consecutive patients. J Neurosurg (Spine 1). 2001;94:25–37. 99. Oda I, Abumi K, Ito M et al. Palliative spinal reconstruction using pedicle screws for metastatic lesions of the spine: a retrospective analysis of 32 cases. Spine. 2006;31(13):1439–1444. 100. Mazel C, Hoffmann E, Antonietti P et al. Posterior cervicothoracic instrumentation in spine tumors. Spine. 2004;29(11):1246–1253. 101. Tomita K, Kawahara N, Baba H et al. Total en bloc spondylectomy: a new surgical technique for primary malignant vertebral tumors. Spine. 1997;22:324–3333. 102. Tomita K, Kawahara N, Baba H et al. Total en bloc spondylectomy for solitary spinal metastasis. Int Orthop. 1994;18:291–298. 103. Fourney DR, Abi-Said D, Rhines LD et al. Simultaneous anterior–posterior approach to the thoracic and lumbar. J Neurosurg. 2001;94(2 Suppl):232–244. 104. Patchell DA, Tibbs PA, Regine WF et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomized trial. Lancet. 2005;366(9486):643–648. 105. Chason JL, Walker FB, Landers JW. Metastatic carcinoma in the central nervous system and dorsal root ganglia. Cancer. 1963;16: 781–787. 106. Costigan DA, Winkelman MD. Intramedullary spinal cord metastasis. A clinicopathological study of 13 cases. J Neurosurg. 1985;62:227–233. 107. Kalayci M, Çagavi F, Güi S et al. Intramedullary spinal cord metastasis: diagnosis and treatment and illustrated review. Acta Neurochir (Wien). 2004;146:1347–1354. 108. Connolly S, Winfree C, McCormick P et al. Intramedullary spinal cord metastasis: report of three cases and review of the literature. Surg Neurol. 1996;46:329–339. 109. Schiff D, O’Neil BP. Intramedullary spinal cord metastasis: clinical features and treatment outcomes. Neurology. 1996;47:906–912. 110. Edelson RN, Deck MDF, Posner JB. Intramedullary spinal cord metastasis. Neurology. 1972;22:1222–1231. 111. Grem JL, Burgess J, Trump D. Clinical features and natural history of intramedullary spinal cord metastasis. Cancer. 1985;56: 2305–2314. 112. Villegas AE, Guthrie TH. Intramedullary spinal cord metastasis in breast cancer: clinical features, diagnosis, and therapeutic consideration. Breast J. 2004;10(6):532–535. 113. Sutter B, Arthur A, Laurent J et al. Treatment options and time course for intramedullary spinal cord metastasis: report of three cases and review of the literature. Neurosurg Focus. 1998;4(5): Article 3. 114. Conill C, Sanchez M, Puig S et al. Intramedullary spinal cord metastasis and melanoma. Melanoma Res. 2004;14:431–433. 115. Aryan HE, Azadeh F, Nakaji P et al. Intramedullary spinal cord metastasis of lung adenocarcinoma presenting as Brown–Sequard syndrome. Surg Neurol. 2004;61(1):72–76. 116. Winkelman MD, Adelstein DJ, Karling NL. Intramedullary spinal cord metastasis: diagnosis and therapeutic considerations. Arch Neurol. 1987;44:526–531. 117. Siegal T, Siegal T. Spinal metastasis. In: Principles of Neuro-oncology, New York: McGraw-Hill, 2005:581–606. 118. Jawahar A, Kondziolka D, Graces YI. Stereotactic radiosurgery for hemangioblastomas of the brain. Acta Neurochir 2000;142:641–644.

12

Leptomeningeal Metastases Ayman I. Omar, MD, PHD, and Warren P. Mason, MD,

FRCPC

CONTENTS Introduction Incidence Clinical Findings Pathology and Pathophysiology Laboratory Investigations Differential Diagnosis Staging Treatment Prognosis Conclusions References

Summary Leptomeningeal metastases have become a well-described and increasingly recognized complication of cancer. Recent advances in diagnostic radiology have increased the ability of the clinician to make this diagnosis and to make therapeutic decisions. The incidence of this disorder will continue to rise as treatments for systemic cancer become more effective. Unfortunately, the prognosis for this illness remains grim even for patients who receive aggressive multimodal therapy. Reasons for our failure to make a significant impact on the course of this disease are multiple. Further clinical and laboratory research is sorely needed to improve the outcome of this devastating complication of cancer. Key Words: leptomeningeal metastases, CSF, intrathecal chemotherapy

1. INTRODUCTION Cancer cells can invade the cerebrospinal fluid (CSF) and seed the leptomeninges in a diffuse and multifocal manner, producing a complication known as leptomeningeal metastases or seeding. These cancer cells can remain confined to the meninges or penetrate the brain, spinal cord, or nerve roots, leading to a variety of symptoms and neurological signs. The multiplicity of clinical findings associated with leptomeningeal metastases has made the diagnosis particularly challenging for the clinician. This devastating complication was first described by Eberth in 1870 (1), and later named “meningitis carcinomatosa” by Siefert in 1902 (2). Once considered uncommon and described usually as a finding at autopsy, leptomeningeal metastases has been diagnosed with increasing frequency in recent decades. For this reason, and because of the severe and devastating symptoms caused by this disorder, leptomeningeal metastases has become a common problem in neuro-oncology. Early diagnosis of leptomeningeal metastases is important because recent studies have suggested that intervention From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

181

182

Part IV / Direct Complications of Cancer

is most beneficial for patients who have minimal symptoms, high performance status, and low leptomeningeal tumor burden.

2. INCIDENCE The incidence of leptomeningeal metastases is difficult to know with certainty, but it appears to be increasing for most cancers (3,4). An exception is childhood acute lymphocytic leukemia, where recognition of isolated central nervous system (CNS) relapse and the development of appropriate and effective prophylaxis, have reduced the incidence of this complication from as high as 66% to the present level of approximately 5% (5). This apparent increase in the incidence of leptomeningeal metastases is due to improved diagnostic modalities, prolonged survival of patients with systemic cancers, and increased awareness of this disorder. Although leptomeningeal metastases can be the initial presentation of an underlying cancer, most patients who develop this disorder do so late in the course of their illness (6–8). Not uncommonly, patients with breast carcinoma and leukemia may develop leptomeningeal metastases many years after the initial diagnosis and treatment of their cancers. Most patients who develop leptomeningeal metastases have been heavily pretreated and have widespread metastatic and progressive cancer (9–12). Leptomeningeal metastases are rarely the only intracranial site of disease. Posner and Chernik found isolated leptomeningeal metastases in only 1.9% of 2088 patients with intracranial metastases and cancer excluding leukemia (13). Coincident parenchymal brain metastases, dural or bony tumors are usual findings associated with leptomeningeal metastases. For this reason, clinicians may attribute new neurological symptoms and signs to parenchymal CNS or systemic metastases or nonmetastatic complications of cancer and not search for leptomeningeal metastases. Moreover, because many patients who develop leptomeningeal metastases are terminally ill, the clinician may choose a palliative approach and not investigate for this complication. Consequently, leptomeningeal metastases are often underdiagnosed by the clinician. Recent autopsy studies have suggested that leptomeningeal metastases occur in approximately 3–8% of all cancer patients (13,14). Postmortem underdiagnosis, however, is also suspected because leptomeningeal metastases are a multifocal and often microscopic disorder, and unless the pathologist samples many sites, the diagnosis may be overlooked. A report published by the Memorial Sloan-Kettering Cancer Center documented an 8% incidence of leptomeningeal metastases in 2375 patients with cancer and a postmortem neuropathologic evaluation (13). A study conducted by the National Cancer Institute reported an antemortem diagnosis of this condition in 11% and a postmortem confirmation of 25% of patients with small cell lung cancer (7). Autopsy reports from patients with breast cancer have identified leptomeningeal metastases in 3–40% of patients (15,16). Other autopsy studies have revealed that as many as 19% of patients with neurological signs and symptoms have pathological evidence of carcinomatous infiltration of the meninges at postmortem examination (17). Any cancer can potentially infiltrate the leptomeninges. Clinical reports have suggested that leptomeningeal metastases occur in 4–15% of carcinomas, 4–15% of non-Hodgkin’s lymphomas, and 5–15% of acute nonlymphocytic leukemia (3,6,11,18–21). Among solid tumors, the most common systemic cancers to spread in this way include breast cancer, small cell lung cancer, and melanoma where approximately 5%, 9–25%, and 23% of patients, respectively, will develop this complication (7,8,22). Occasionally, uncommon tumors such as multiple myeloma, thyroid cancer, chronic lymphocytic leukemia, and Hodgkin’s disease can cause leptomeningeal metastases. Gastric carcinoma, reported in the past as a tumor that commonly infiltrates the meninges, is today rarely a cause of this disorder. Between 1% and 7% of leptomeningeal metastases are attributed to cancers of unknown primary sites (6,10,11,19). Primary brain tumors can also disseminate through the leptomeninges, with recent reports suggesting that this complication occurs between 10% and 32% of cases (13,19,23–25). Occasionally tumors can arise as primary malignancies of the leptomeninges. Melanomas, rhabdomyosarcomas, and lymphomas can be confined to the leptomeninges. Most cases of primary CNS lymphomas, however, are parenchymal, with leptomeningeal seeding common and often asymptomatic (26).

3. CLINICAL FINDINGS The clinical manifestations of leptomeningeal metastases are numerous because multifocal involvement of the brain, spinal cord, or nerve roots is characteristic (6,18,27–29). Leptomeningeal metastases should be suspected whenever a patient presents with multiple neurological signs and symptoms. Typically, signs are more widespread

Chapter 12 / Leptomeningeal Metastases

183

than symptoms, but if the diagnosis is considered early during the course of the illness, there is greater likelihood of uncovering subtle or unifocal neurological findings. Thus, a high index of suspicion in an appropriate clinical context is required if this diagnosis is to be made during the initial stages of the illness. The most common site of involvement of the neuraxis is the spinal region, with lower motor neuron weakness the most common finding, being present in 80% of patients at diagnosis. Other common findings on examination are reflex loss or asymmetry and dermatomal sensory loss. Signs suggestive of leptomeningeal irritation, such as nuchal rigidity or pain on straight leg raising, are uncommon and detected in only 15% of patients (10). Findings such as aphasia, hemiparesis or visual field loss usually indicate advanced disease, and are usually attributed to concomitant parenchymal metastases. As the symptoms and signs of leptomeningeal metastases are pleomorphic, it is best to conceptualize the potential clinical abnormalities as being localized to three compartments: the cerebral hemispheres, the cranial nerves, and the spinal cord and roots.

3.1. Cerebral Symptoms and Signs Common cerebral signs and symptoms are presented in Table 1. Headache is among the most common symptoms but is usually nonspecific and may be associated with nausea and vomiting. Severe episodic headache with disturbance of consciousness may be attributed to plateau waves associated with increased intracranial pressure. Gait abnormalities such as gait apraxia may be related to increased intracranial pressure; however, gait impairment is due to multiple causes including cerebellar dysfunction and lower motor neuron/radicular involvement. Other common complaints related to cerebral involvement include memory and concentration disturbance, seizures, either focal or generalized, and vertigo or lightheadedness. Common signs caused by cerebral leptomeningeal involvement include papilledema, mental status changes including confusion and dementia, seizures, and extensor plantar responses. Some uncommon signs attributable to cerebral dysfunction include central diabetes insipidus due to involvement of the posterior pituitary or hemiparesis due to parenchymal dysfunction (30). However, significant hemispheric signs usually suggest the coexistence of CNS metastases in addition to leptomeningeal disease.

3.2. Cranial Nerve Symptoms and Signs Common cranial nerve symptoms and signs are listed in Table 2. Cranial nerve abnormalities usually become more prominent as leptomeningeal metastases progress and become more severe. Common findings include abnormalities of ocular movement manifesting as diplopia, facial weakness or asymmetry, hearing loss or tinnitus, and abnormalities of facial sensation. Although diplopia is the most common complaint due to cranial nerve infiltration, examination of ocular motility is often normal. When palsies become apparent, involvement of the abducens nerve is more common than oculomotor or trochlear nerve dysfunction. The development of ophthalmoplegia usually indicates infiltration of the cavernous sinus by tumor. Blindness is a common finding as well, and is due to infiltration of the optic nerve, chiasm, or tracts by tumor. The occurrence of multiple cranial nerve palsies, particularly if they are bilateral, should suggest the diagnosis of leptomeningeal metastases rather than pathology involving the dura or skull base.

Table 1 Cerebral Symptoms and Signs Symptoms Gait difficulty Headache Confusion Nausea and vomiting Loss of consciousness Dizziness Dysphagia

Signs Mental status change Papilledema Seizures Hemiparesis Diabetes insipidus

184

Part IV / Direct Complications of Cancer

Table 2 Cranial Nerve Symptoms and Signs Symptoms

Signs

Diplopia Visual loss Hearing loss Facial numbness Tinnitus Hoarseness Dysphagia

Ocular muscle paresis Optic neuropathy Facial weakness Hearing loss Trigeminal neuropathy Hypoglossal neuropathy Decreased gag reflex

Table 3 Spinal Cord and Root Symptoms and Signs Symptoms Back/neck pain Radicular pain Weakness Paresthesias Bowel/bladder dysfunction

Signs Lower motor neuron weakness Reflex asymmetry Dermatomal sensory loss Positive straight leg raise Nuchal rigidity

3.3. Spinal Cord and Root Symptoms and Signs Spinal symptoms and signs are the most common manifestations of leptomeningeal metastases. Symptoms are referable to either the leptomeninges or nerve roots (Table 3). Symptoms and signs related to spinal meningeal involvement include neck or back pain that may be associated with nuchal rigidity. Lumbar puncture with the detection of malignant cells in the CSF distinguishes leptomeningeal metastases from an infectious process. Infiltration of nerve roots usually present as weakness, radicular sensory dysfunction, or bladder and bowel dysfunction. Absent deep tendon reflexes are the most common neurological sign in patients with leptomeningeal metastases, and a cauda equina syndrome with leg weakness, foot numbness, and bladder or bowel impairment is frequently encountered in patients with this complication.

4. PATHOLOGY AND PATHOPHYSIOLOGY Malignant cells can invade the leptomeninges and gain access to the subarachnoid space by a number of mechanisms including: 1. Hematogenous dissemination, by which tumor cells circulating in the blood invade the meninges by first penetrating arachnoid vessels and choroid plexus. Furthermore, tumor cells from metastatic deposits adjacent to the nervous system can also gain access to the subarachnoid space by invading the venous plexus of Batson and by perivenous spread from the bone marrow (31,32). This mechanism has been best described in autopsy cases of patients with leukemia, where tumor cells have been observed in the walls of superficial arachnoid veins, the surrounding adventitia, the CSF and the Virchow–Robin spaces (33). 2. Direct invasion of the meninges and CSF by tumor cells from metastases that are in contact with the meninges (dural, subdural, bony, or intraparenchymal metastases) (4). Such cancer cells can gain access to the subarachnoid space by penetrating the pia or ependyma or by tracking along perivascular spaces. 3. Perineural migration of tumor cells from systemic metastases to the subarachnoid space. This mechanism has been described as a means by which tumors of the head and neck can produce leptomeningeal metastases. 4. Iatrogenic seeding of the meninges during surgical extirpation of CNS metastases. In particular, this complication has been frequently observed in patients undergoing posterior fossa surgery for resection of cerebellar metastases (34).

Chapter 12 / Leptomeningeal Metastases

185

The two most common mechanisms by which tumor cells seed the leptomeninges are by hematogenous dissemination or by direct extension from metastatic deposits. Most patients with leptomeningeal metastases have widely metastatic disease and the meninges are but one site of infiltration. In fact, between 33% and 75% of patients with leptomeningeal metastases have concurrent brain metastases and between 16% and 37% have dural metastases (10,35). When malignant cells enter the subarachnoid space, they can disseminate by CSF flow and form distant meningeal tumor deposits at multiple sites within the neuraxis. The most common sites of tumor deposition include the basal cisterns, the dorsal surface of the cerebral hemispheres, the dorsal surface of the spinal cord, and the cauda equina, where the force of gravity and sluggish CSF flow promote the deposition of suspended tumor cells. After tumor cells have anchored themselves onto the surface of the meninges, they grow and invade the parenchyma of the brain, spinal cord, cranial, or spinal nerve roots, forming bulky subarachnoid deposits. Leptomeningeal metastases can cause symptoms and signs by several mechanisms. CNS dysfunction secondary to the development of hydrocephalus is a common occurrence. Tumor cells that gain access to the subarachnoid space can cause disruption of normal CSF dynamics by causing obstruction of the fourth ventricle foramina or other sites such as the arachnoid villi. Ventricular pressure is usually elevated, but this may not occur if hydrocephalus develops slowly and causes progressive ventriculomegaly with a trivial increase of CSF pressure. Even in the absence of hydrocephalus, CSF flow dynamics as measured by radioisotope flow studies are abnormal in as many as 70% of patients with leptomeningeal metastases (36). Additionally tumor cells can invade the brain, spinal cord, and roots to cause neurological symptoms such as seizures, weakness, or sensory dysfunction. The formation of tumor deposits in the Virchow–Robin spaces can cause arterial and arteriolar compression with consequent ischemia to the underlying parenchyma (10,37). This phenomenon has been observed in cerebral arteriograms of patients with leptomeningeal metastases and probably accounts for strokes, TIAs and diffuse white matter changes that occur in these patients (10,37). Finally, the idea that tumor cells consume vital nutrients, thereby starving surrounding neurons, has given rise to a hypothesis of metabolic competition between tumor and neuronal cells as an explanation for nervous system dysfunction (38). Hypoglycorrhacia is a typical CSF finding in leptomeningeal metastases, but this may not be due entirely to consumption of CSF glucose by metastatic tumor cells and deposits. However, the phenomenon whereby infiltrating tumor cells disrupt the normal functioning of adjacent neurons by consumption of glucose explains the occurrence in hypothalamic leukemia of patients with leukemic infiltration of the posterior pituitary gaining weight as their only manifestation of hypothalamic–pituitary dysfunction. In this unusual situation, it is believed that tumor cells deprive sugar-sensitive hypothalamic neurons of this vital nutrient, thereby setting into motion a complicated behavioral response that increases food consumption.

5. LABORATORY INVESTIGATIONS The two investigations that are useful for establishing the diagnosis of leptomeningeal metastases are CSF analysis and MR scans of the brain and spine.

5.1. CSF Analysis CSF analysis, usually from a sample obtained by lumbar puncture, is abnormal in the vast majority of patients with leptomeningeal metastases (3,6,10,11,19,28,29,35). An opening pressure should be obtained and CSF should be submitted for protein, glucose, cell count, and cytology. Opening pressure is elevated in approximately 50% of patients with leptomeningeal metastases because of impaired CSF dynamics and attendant hydrocephalus. CSF protein concentrations are often elevated as well, due to disruption of the blood–brain barrier (BBB) by this disorder, and the breakdown of white blood cells as well as tumor cells extruded into the CSF. A depressed CSF glucose relative to concurrent serum levels is observed in 30–40% of patients with leptomeningeal metastases. Reasons for this relate to impaired glucose transport across the blood–CSF or BBB and consumption of this nutrient by proliferating tumor cells, white blood cells, and reactive pial cells. The identification of neoplastic cells in CSF is diagnostic of leptomeningeal metastases. A positive CSF cytology is found on the initial lumbar puncture in approximately 50% of patients with leptomeningeal metastases. The yield of this diagnostic procedure increases to approximately 90% of patients if a total of three high-volume lumbar punctures are obtained (10). Diagnostic yield improves if at least 10 cc of CSF are withdrawn. Adding a

186

Part IV / Direct Complications of Cancer

fixative to the CSF is not necessary, but the specimen should be delivered to the laboratory expeditiously. The specimen should also be processed immediately in the laboratory (39). Patients with extensive leptomeningeal infiltration by neoplastic cells are more likely to have a positive cytologic examination than patients with scant and focal involvement of the meninges. For patients with suspected leptomeningeal metastases and an apparent negative CSF cytology, there is little benefit in performing multiple lumbar punctures after the initial three procedures. In a series of patients with leptomeningeal metastases due to systemic solid tumors reported by Wasserstrom, approximately 5% of patients had a positive CSF cytology only from samples obtained from the cerebral ventricles or cisterna magna (10). Hence there is a subset of patients with leptomeningeal metastases who will have persistently false-negative CSF cytologic examinations unless CSF is obtained from rostral sites of the neuraxis. Occasionally, patients with extensive infiltration of the meninges by tumor have persistent negative CSF cytologic tests. In this situation, it is believed that tumor cells are highly adherent to the membranes and rarely exfoliate into the CSF. Glass and associates have emphasized this point in a series of patients who underwent postmortem examinations. In their study, 41% of patients with autopsy-proven leptomeningeal metastases had negative antemortem CSF cytologic examination (35). Some investigators have suggested that CSF is more likely to be positive for malignant cells when the underlying tumor is a carcinoma, but Kaplan and colleagues have reported no difference in CSF cytologic results between patients with carcinomas, leukemias or lymphomas (28). Recently, Chamberlain and colleagues have suggested that the site from which CSF is sampled can increase the sensitivity of detecting leptomeningeal disease. In patients with cranial nerve involvement, ventricular CSF cytology was more likely to be positive when compared to lumbar CSF cytology (odds ratio = 2.71; 95% confidence interval, 0.76–9.71). Conversely, patients with spinal signs and symptoms, lumbar CSF cytology was more likely to produce positive results when compared to ventricular CSF sampling (odds ratio = 2.86; 95% confidence interval, 0.86–9.56). The study concludes that CSF should be sampled from the site where clinical and radiographic disease is present (40). A false-positive cytology is an uncommon finding, but can occur in rare situations. Occasionally, cerebral metastases can infiltrate focally adjacent leptomeninges. In this situation, tumor cells can be shed into the CSF and thereby produce a positive result. It may therefore be prudent in such instances to treat these patients as if they have active leptomeningeal disease. Similarly, rare patients with systemic cancers and no neurologic findings have malignant cells identified in the CSF. These patients, although asymptomatic, should probably also receive treatment directed at the leptomeninges. Finally, occasionally reactive or degenerated lymphocytes can be mistaken for neoplastic cells by the cytopathologist (35). In difficult cases, a number of ancillary diagnostic techniques available to the cytopathologist can be useful. The use of monoclonal antibodies (MAbs) has been combined with routine cytology to increase diagnostic yield (41–43). The use of immunocytochemistry has been employed to improve diagnostic certainty in cases of suspected CNS lymphoma, where the identification of a monoclonal cell population by lymphocyte surface markers can assist in the diagnosis of a tumor. However, the demonstration of a polyclonal population of lymphocytes in the CSF does not exclude this diagnosis, as a malignant CNS lymphoma will often stimulate the migration and proliferation of reactive polyclonal lymphocytes. A variety of tumor markers have been studied in the CSF of patients with suspected and confirmed leptomeningeal metastases as a means of improving the yield of a CSF specimen (Table 4) (44–46). Unfortunately, poor sensitivity and specificity generally limit the use of these markers in the CSF. Tumor markers can be separated into specific and nonspecific. Specific markers such as carcinoembryonic antigen (CEA), -fetoprotein (AFP), ß-human chorionic gonadotrophin (ßHCG), and monoclonal immunoglobulins elaborated from multiple myeloma are diagnostic of tumor invasion into the CSF compartment when they are identified in a CSF specimen. Occasionally, when serum levels of these specific markers are markedly elevated, false elevations in the CSF can be caused by passage of a tumor marker across a normal or partially disrupted BBB. In this situation, determination of a CSF-serum ratio for a specific marker will help determine whether the CSF concentration is abnormally elevated. CSF levels of specific tumor markers can be useful for determining response to therapy, as they tend to decline with successful treatment, and sometimes rebound at the time of relapse before other findings are apparent.

Chapter 12 / Leptomeningeal Metastases

187

Table 4 CSF Tumor Markers Specific Carcinoembryonic antigen (Carcinoma) Human chorionic gonadotrophin (Choriocarcinoma, embryonal carcinoma, germ-cell tumor) Alpha-fetoprotein (Teratocarcinoma, yolk sac tumor, embryonal carcinoma, endodermal sinus tumor) CA-125 (Ovarian carcinoma) CA 15-3 (Breast carcinoma) Prostate-specific antigen (Prostate carcinoma) Melanin (Melanoma)

Nonspecific -microglobulin -glucoronidase Lactate dehydrogenase

Nonspecific tumor markers such as ß-glucoronidase, ß2-microglobulin, and isoenzyme V of lactate dehydrogenase are frequently elevated in the CSF of patients with leptomeningeal metastases as well as a variety of acute and chronic meningitic processes. The tumor marker ß2-microglobulin is elevated particularly in leukemias and lymphomas as opposed to solid tumors. If other causes for elevations of these markers, such as infectious or inflammatory meningitis, can be excluded, marked elevations of nonspecific tumor markers can signify the presence of leptomeningeal metastases in an appropriate clinical context. While nonspecific tumor markers are often in the normal range in cases of definite leptomeningeal metastases, these markers can be useful clinically in situations where they are elevated and serial determinations can be used as adjunctive measures of response to treatment. The evaluation of DNA abnormalities by flow cytometry has the potential to identify tumor cells that may be missed by conventional cytology. Flow cytometry detects aneuploid cells as well as estimates the proportion of cells in the S phase and G2–M phases of the cell cycle, and thus can identify populations of malignant cells (47). This technique appears to be most useful for diagnosing meningeal leukemia and lymphoma where distinguishing malignant cells from reactive ones can be challenging. More recently, investigators have explored the role of fluorescent in situ hybridization (FISH) and polymerase chain reaction (PCR) techniques to identify malignant cells in CSF (48–51). Although a positive CSF cytology is considered the gold standard for the diagnosis of leptomeningeal metastases, it has a rather low negative predictive value and sensitivity. A recent study has shown that CSF cytology carries only 75% sensitivity in diagnosing leptomeningeal metastases but is 100% specific. The use of gadoliniumenhanced MRI in the diagnosis of leptomeningeal metastases is 76% sensitive and only 77% specific. The study concludes that a positive gadolinium-enhanced MRI in patients with cancer provides strong support for the diagnosis of leptomeningeal dissemination despite a negative CSF cytology (52). It is apparent that more sensitive and specific markers for diagnosing leptomeningeal dissemination need to be added to the currently available techniques. A recent study examined CSF protein profiling using multiplex immunoassay (MIA). The authors identified several proteins associated with leptomeningeal metastases including interleukin-8 (IL-8), Pulmonary and Activation Regulated Chemokine (PARC) and interferon-gamma inducible protein (IP-10) (53). Interestingly, in a separate article, Brandsma et al. found that an elevated CSF IL-8 level correlated negatively with survival and therefore has a prognostic as well as a diagnostic value (54). The addition of CSF protein profiling may therefore be useful in diagnosing leptomeningeal dissemination. Yet another group of biomarkers reportedly achieved 100% sensitivity in diagnosing leptomeningeal metastases. A study investigating the role of various biomarker concentrations by enzyme-linked immunosorbent assay in

188

Part IV / Direct Complications of Cancer

CSF found that endothelial growth factor (VEGF) and tissue-type plasminogen activator (tPA) detected in CSF can be highly sensitive in diagnosing leptomeningeal metastases. The indices of these biomarkers (CSF/serum value relative to CSF/serum albumin ratios) were calculated. The authors found that an elevated VEGF index and a low tPA index in the CSF yielded 100% sensitivity in diagnosing leptomeningeal metastases (55). Additionally, several biomarkers may be associated with leptomeningeal disease from specific primary cancers involving the breast or esophagus. Okumura et al., reported the early detection of leptomeningeal dissemination of a basaloid esophageal carcinoma using quantitative RT-PCR for carcinoembryonic antigen messenger ribonucleic acid despite negative neuroimaging (56). On the other hand, the use of proteomic profiling by mass spectrometry has recently been shown to predict the presence of leptomeningeal metastases from breast cancer. Dekker et al. used matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) for detecting several peaks representing differential expression of peptides in CSF of patients with leptomeningeal dissemination of breast cancer (57). These newer techniques, although not widely available, may some day improve the diagnostic accuracy of leptomeningeal metastases. Other variables such as albumin CSF/serum ratio, glucose CSF/serum ratio and CSF LDH are much less specific albeit useful in diagnosing leptomeningeal metastases. For example, an elevated albumin CSF/serum ratio can be found in bacterial, cryptococcal, and tuberculous meningitis, leptomeningeal metastases, as well as acute and chronic demyelinating polyneuropathies. A low glucose CSF/serum ratio and/or elevated CSF LDH may be seen with leptomeningeal metastases as well as bacterial and fungal meningitic disease (45,58). To increase the probability of accurately diagnosing leptomeningeal metastases, it may therefore be prudent to use a combination of the above approaches. Presently, these technically challenging and resource intensive tools remain investigational and complementary to conventional cytology and are available only at a few research-oriented medical centers.

5.2. Imaging Neuroradiographic techniques are assuming increasing importance in the diagnosis of leptomeningeal metastases (59–65). The tools currently employed to assist in the diagnosis of leptomeningeal metastases are cranial computed tomography (CT), magnetic resonance imaging (MRI) of the brain and spine, myelography and radionuclide CSF flow studies. Contrast-enhanced CT or MR scans of the brain should be obtained in all patients with suspected leptomeningeal metastases and an intracranial mass lesion before lumbar puncture to estimate the risk of cerebral herniation following this procedure. Moreover, scans should be performed of regions of the neuraxis potentially implicated in patient’s symptoms. 5.2.1. CT Imaging CT scans are abnormal in 25–56% of patients with leptomeningeal metastases (62). Abnormalities suggestive of leptomeningeal metastases by tumor include enhancement of the meninges, ventricular ependyma, basal cisterns, tentorium, or cauda equina, hydrocephalus and effacement of sulci or cisterns. Enhancement may be linear or nodular. Significantly, 30–60% of patients with leptomeningeal metastases will have concurrent brain metastases. Contrast-enhanced MR scans are more sensitive and specific than contrast-enhanced CT scans in revealing abnormalities consistent with leptomeningeal metastases, and have become the diagnostic imaging modality of choice for this disease (Fig. 1). False-negative results in patients with suspected leptomeningeal metastases can be minimized by repeating MR scans with increased doses of gadolinium (66). These scans require meticulous interpretation by experienced individuals, however, because some enhancement of normal meninges and nerve roots is expected (67). 5.2.2. MR Imaging The use of high-field 3.0 T MR units especially when combined with cumulative triple-dose MR contrast agent administration has been shown to produce the best resolution, and therefore has higher diagnostic accuracy for leptomeningeal metastases (68). Additionally, further refinement and choice of specific MR imaging sequences can result in better detection of leptomeningeal metastases. The addition of gadolinium-enhanced fast fluid-attenuated inversion-recovery (gad-enhanced fast FLAIR) MR imaging to the usual gadolinium-enhanced T1 sequence was shown to be useful in detecting superficial meningeal disease compared with gadolinium-enhanced T1. This is

Chapter 12 / Leptomeningeal Metastases

189

Fig. 1. Leptomeningeal metastases as seen on MR brain after gadolinium–DTPA contrast administration. This patient presented with confusion and dysarthria due to leptomeningeal carcinomatosis from a recurrent serous adenocarcinoma of the ovary. Coronal (a) and axial (b) T1-weighted MR images reveal abnormal leptomeningeal enhancement consistent with meningeal carcinomatosis.

because gad-enhanced fast FLAIR images do not result in contrast enhancement of blood vessels with low flow and because adjacent CSF is suppressed in such sequences allowing for better resolution of contrast enhanced areas. However, although useful, gad-enhanced fast FLAIR was associated with sensitivity and specificity of only 41% and 88%, respectively for detecting leptomeningeal metastases. On the other hand, contrast-enhanced T1 sequences were associated with higher sensitivity and specificity values of 59% and 93%, respectively (69–71). It is therefore apparent that the clinician should carefully review all such sequences to improve the chances of detecting leptomeningeal disease. Newer imaging modalities such as fluorine-18 fluorodeoxyglucose (FDG) and carbon 11–labelled methionine (Cmet) positron emission tomography (PET) are also useful in the diagnosis of leptomeningeal metastases (72,73). Although currently not widely utilized, the combination of such imaging modalities with conventional MR sequences may increase the specificity of diagnosing leptomeningeal metastases. 5.2.3. Spine Imaging Contrast-enhanced spinal MRI and computerized tomographic myelography are both capable of detecting abnormalities suggestive of leptomeningeal metastases such as nerve root or cord thickening, subarachnoid nodules, scalloping of the subarachnoid membranes, and spinal canal block. A study comparing these two imaging modalities for patients with leptomeningeal metastases by Chamberlain and associates revealed no striking differences in the sensitivity and specificity of these two techniques (74). Consequently, MRI of the spine has become the preferred spine-imaging modality for detecting leptomeningeal metastases at this site because of its noninvasive nature, minimal morbidity and patient preference (Fig. 2). Moreover, the expanding familiarity with characteristic MR features of leptomeningeal metastases has led some authors to suggest that MR scans can be diagnostic in patients with known cancer, multifocal neurological findings, and characteristic imaging abnormalities (59). Recently, radionuclide CSF flow studies have been used increasingly for the evaluation of patients with leptomeningeal metastases prior to the administration of regional chemotherapy (19,36,75–79). Radionuclide CSF flow studies provide a reasonable radiographic assessment of CSF flow dynamics and are abnormal in approximately 30–40% of patients with leptomeningeal metastases (Fig. 3). Failure of radionuclide to migrate in a predictable manner following lumbar or ventricular administration of radioimmunoconjugate is defined as CSF flow block. Typical sites of CSF obstruction include the cisterna magna, spinal canal, and cerebral convexities. Sites of CSF obstruction often but not always correlate with bulky leptomeningeal disease on CT or MR imaging. Chamberlain and associates, who have extensive experience using CSF flow studies in patients with leptomeningeal metastases, have suggested that a significant proportion of patients with no obvious CSF block on conventional imaging will have CSF flow abnormalities that can often be reversed by involved-field radiotherapy (75). In their series, patients who did not experience flow reversal appeared to benefit minimally from intrathecal chemotherapy and had a disproportionately poor survival. While CSF flow studies do predict the distribution of intrathecally administered drug, the role of radionuclide flow studies in the management of patients with

190

Part IV / Direct Complications of Cancer

Fig. 2. Drop metastases from a diffuse brainstem glioma causing cord compression. The MR images shown are of a 27-year-old male who presented initially with progressive diplopia and hemibody paresthesias. He subsequently received focal radiotherapy and adjuvant temozolomide chemotherapy for a diffuse brainstem glioma. Two years later, he developed a T5 sensory level, paraparesis, and sphincter dysfunction. A T2-weighted sagittal MR scan of the spine revealed extensive bulky leptomeningeal metastases with cord compression at multiple levels (a). Following spinal radiotherapy, significant tumor regression was noted on subsequent imaging (b).

leptomeningeal metastases is unclear. It is possible that CSF flow studies are of prognostic significance by virtue of their ability to serve as a sensitive measure of the disease burden for patients with leptomeningeal metastases. CSF flow studies may provide the most information for patients without evidence of bulky and extensive leptomeningeal metastases, where abnormal CSF flow studies might predict a poor response to regional chemotherapy and associated risks (79).

6. DIFFERENTIAL DIAGNOSIS Characteristic findings suggestive of leptomeningeal metastases in cancer patients are multifocal symptoms and signs as well as evidence of meningeal irritation. The differential diagnosis usually is limited to multiple parenchymal metastases of the brain or cord, extensive epidural disease, or an inflammatory or meningitic process (80). Metastatic bulky disease to the brain or spinal cord can be diagnosed by MR or CT scans and a lumbar puncture with appropriate analysis of CSF, including cultures, which can help resolve the possibility of infections. Typically patients with infections have more meningitic symptoms and signs than fixed neurological findings. The situation is reversed for patients with leptomeningeal metastases. Occasionally, a meningeal biopsy is required to make a definitive diagnosis and exclude other common meningitic processes (67). To maximize the yield of this invasive procedure, the biopsy site should correspond to areas of abnormal meningeal enhancement seen on a MR scan.

7. STAGING Patients with suspected leptomeningeal metastases should undergo the following investigations: 1. MRI with gadolinium injection of the area of maximal symptomatology. If leptomeningeal metastases is suspected and an MR scan of the brain does not reveal leptomeningeal enhancement, an MR scan with gadolinium of the spine should be performed to search for asymptomatic spinal or root involvement. Imaging of the entire neuraxis should be considered and is useful for documenting bulky leptomeningeal disease and parenchymal metastases that are often best treated by focal radiotherapy.

Chapter 12 / Leptomeningeal Metastases

191

Fig. 3. 111 Indium-DTPA CSF flow study of a patient with communicating hydrocephalus due to leptomeningeal carcinomatosis. This 55-year-old woman with metastatic breast cancer presented with headaches. MR scan of the brain revealed communicating hydrocephalus as shown in this T1-weighted image (a). Radionuclide ventriculography with 111 indium-DTPA showed ventricular retention of the radionuclide at 4 hrs (b) and 24 hrs (c) following injection via an Ommaya reservoir. Following extensive treatment with intraventricular methotrexate, she developed progressive dementia and ataxia. T2-weighted MR scans before (d) and after methotrexate treatment (e) revealed the development of extensive white matter changes consistent with methotrexate-induced leukoencephalopathy.

2. Lumbar puncture for CSF analysis. If the initial lumbar puncture is negative for malignant cells and leptomeningeal metastases is suspected, this procedure should be repeated up to three times over subsequent days or weeks to search for malignant cells. If the CSF remains negative for malignant cells, treatment can be initiated if characteristic MR features of leptomeningeal metastases are noted and not explained by other processes such as meningeal infection or unquestionably elevated levels of specific tumor markers are found in the CSF. 3. Radionuclide CSF flow studies to establish the presence or absence of CSF flow delays or obstruction should be considered, particularly if intrathecal chemotherapy is planned. If abnormal flow is detected, MRI of sites of CSF flow obstruction should be performed to search for bulky leptomeningeal disease, parenchymal metastases, or epidural tumor. Moreover, radionuclide imaging may be useful for determining prognosis and to ensure that normal CSF flow dynamics are present before initiating intrathecal chemotherapy. 4. In cases where the diagnosis of leptomeningeal metastases is suspected but all investigations are negative, it may be useful to repeat imaging or CSF analysis several weeks or months later if symptoms and signs persist. Occasionally, asymptomatic patients may have malignant cells identified in a CSF specimen collected for unrelated reasons. These patients are probably best treated with intrathecal chemotherapy alone. Rarely, patients with no known cancer are found to have focal meningeal enhancement. Such patients, and the rare patient with the rare history of cancer who is found to have an area of meningeal enhancement, should be considered for meningeal biopsy. In this instance, the biopsy should be of a site of meningeal enhancement to increase the likelihood of establishing a firm diagnosis. 5. Serial CSF analyses are useful and samples should be obtained prior to each intrathecal therapy and periodically after treatment is completed if remission is achieved. Serial CSF cytologic evaluations are useful for documenting response to therapy and for detecting cytologic progression or relapse.

192

Part IV / Direct Complications of Cancer

8. TREATMENT The treatment of leptomeningeal metastases must be directed at the entire subarachnoid space including the cerebral ventricles, basal cisterns and spinal subarachnoid space (3,6,10,11,28,29). The prognosis for patients with untreated leptomeningeal metastases is approximately 4–6 weeks, and death usually results from progressive neurological dysfunction. If only areas of symptomatic involvement are treated, progression elsewhere or relapse at treated sites will develop rapidly due to continued growth and ongoing CSF dissemination of cancer cells from untreated tumor deposits. The difficulties associated with delivering treatment to the subarachnoid compartment and the scarcity of effective chemotherapeutic agents available for treatment of this disorder frequently result in remissions of short durations and therapeutic failures. Moreover, because patients with leptomeningeal metastases often have widespread and progressive systemic disease resistant to further treatment, the treatment of leptomeningeal metastases is generally considered palliative. Typically, fixed neurological deficits such as cranial nerve palsies and radicular weakness do not resolve with therapy but encephalopathic symptoms can improve, particularly if the underlying cause is hydrocephalus that can be reversed by CSF shunting procedures. Despite the grim prognosis associated with this complication, aggressive multimodal treatment can result in effective palliation and prolonged survival for a minority of patients (81). Patients who are the best candidates for intensive treatments are those with limited and chemosensitive tumors, absent or minimal fixed neurological deficits, controlled or no systemic disease and a high performance status (27,82–84). While there are no standard approaches to the therapy of this disorder, the major modalities of cancer treatment, chemotherapy, radiotherapy and to a lesser degree surgery, play important roles in the management of this illness.

8.1. Radiotherapy Radiotherapy is the most effective treatment for leptomeningeal metastases and often produces symptomatic palliation (63). Pain appears to be the symptom most responsive to this modality. External-beam radiotherapy is often administered to symptomatic sites and to areas of bulky leptomeningeal metastases even if they are asymptomatic because intrathecal chemotherapy does not penetrate beyond two to three millimeters from the tumor-CSF interface (85). Patients with leptomeningeal metastases and significant symptoms from an area without radiographic evidence of bulky disease should have these areas irradiated to relieve symptoms. Thus, patients with back pain and radicular leg weakness should receive focal radiotherapy to the lumbosacral spine, while patients with cranial neuropathies should receive radiation treatments to the base of the brain. Whole-brain radiotherapy is usually reserved for patients with symptomatic hydrocephalus. While treatment schedules vary, patients receiving focal palliative radiotherapy for leptomeningeal metastases usually receive a total of 3000 cGy in 10 fractions. Leptomeningeal metastases is a disseminated disease, and for this reason occasional patients receive entire neuraxis radiotherapy. This approach, while reasonable from a theoretical perspective, is not curative, often highly morbid, and may cause profound myelosuppression to the extent where the prescribed course of radiotherapy cannot be completed or further systemic chemotherapy for the underlying cancer cannot be contemplated. Furthermore, many patients who develop leptomeningeal metastases cannot receive neuraxis radiotherapy because of previous treatment to various regions of the CNS for other cancer-related complications. Soluble radioconjugates have been injected into the CSF as a means of delivering neuraxis radiotherapy, but this approach is constrained by the limitations of intrathecal therapy, including impaired CSF distribution and limited parenchymal penetration (86,87). For these reasons, neuraxis radiotherapy is impractical and should not be administered to most patients with leptomeningeal metastases. Focal radiotherapy usually does not result in neurologic recovery. Exceptional patients are often those with radiosensitive cancers such as leptomeningeal leukemia or lymphoma or less frequently patients with breast carcinoma. However, even in situations where radiotherapy is effective in eradicating circulating and adherent tumor cells, axonal injury and demyelination caused by tumor infiltration are generally irreversible and limit the extent of clinical recovery. For this reason, early diagnosis and treatment of leptomeningeal metastases is crucial.

Chapter 12 / Leptomeningeal Metastases

193

8.2. Regional Chemotherapy Intrathecal chemotherapy is the principal medical therapy for patients with leptomeningeal metastases. This approach enhances exposure of leptomeningeal tumors to a drug while minimizing systemic toxicity. The subarachnoid compartment can be accessed by lumbar puncture or by the placement of a subcutaneous catheter and reservoir (Ommaya reservoir) into the lateral ventricle. The insertion of an Ommaya reservoir is a minor surgical procedure associated with minimal morbidity when performed by an experienced neurosurgeon (88,89). The most common complications include misplacement of the catheter tip, infections, and rarely intracranial hemorrhage. A postoperative CT scan can confirm the correct placement of a catheter, and the risk of hemorrhage can be minimized by careful attention to preoperative coagulation parameters. Infections associated with Ommaya reservoirs are usually due to skin flora and primarily Staphylococcus epidermidis, and can usually be managed with intravenous and intraventricular antibiotics, thereby obviating the need to remove the device. Most clinicians prefer to instill intrathecal chemotherapy via an Ommaya reservoir for a variety of reasons: 1. Repeated lumbar punctures are uncomfortable and cannot be performed safely when thrombocytopenia or a coagulation disorder is present. Ommaya reservoirs can usually be accessed even if a bleeding tendency is present. Repeated lumbar punctures increases the risk of post-LP headache and intracranial hypotension. The latter can result in diffuse meningeal enhancement, which can be mistaken radiographically for progression of leptomeningeal disease. 2. Administration of drug into the subarachnoid space is assured when an Ommaya reservoir is used, but as many as 10–15% of lumbar punctures do not succeed in injecting drug into the lumbar subarachnoid space (90). Repeated lumbar punctures often result in the formation of subdural and epidural fluid collections, which can interfere with the accurate instillation of drug if continued lumbar punctures are performed (91). Additionally, fluid sampled from these iatrogenic fluid collections is not representative of circulating CSF and thereby cannot be used to monitor treatment response. 3. Even when lumbar punctures succeed in the accurate injection of drug into the subarachnoid space, the common occurrence of impaired CSF flow and obstruction found in this disorder may result in incomplete distribution of drug throughout the subarachnoid compartment (36,77,79). Furthermore, the use of an Ommaya reservoir ensures that the ventricular system receives adequate concentration of drug, which may not occur following lumbar injection, even if CSF dynamics are normal (92). Consideration should be given to the use of radionuclide flow studies prior to the commencement of intrathecal chemotherapy because sites of impaired flow can often be corrected by focal radiotherapy. 4. There is evidence to suggest that intrathecal chemotherapy delivered by an Ommaya reservoir is more effective than by lumbar puncture (93).

The agents widely available for intrathecal injection are methotrexate, cytarabine, and thio-TEPA. Methotrexate and cytarabine have activity against leukemia and lymphoma while thio-TEPA and methotrexate have activity against breast carcinoma, but other common solid tumors such as melanoma and lung cancer are resistant to these agents. Consequently, the limited spectrum of activity conferred by these drugs account in part for the modest therapeutic efficacy of intrathecal chemotherapy. A number of investigational agents such as temozolomide, topotecan, 4-hydroperoxycyclophosphamide, interferon (IFN) and interleukin-2 (IL-2) have been evaluated but only in research settings and are thus not widely available (93). Methotrexate is the most frequently used agent and a therapeutic level of at least 1 μM can be achieved for 48–72 hrs in the CSF by administering a dose of 12 mg (3,90,94,95). There is no need to reduce this dose for children over age 4 because the subarachnoid space reaches its adult volume by this age (96). Most clinicians administer drug initially on a twice-weekly schedule for 3–4 weeks, followed by a reduction in frequency over a total treatment period of 3–6 months. The most effective duration of treatment is not established and some patients appear to benefit from indefinite treatment. Alternative treatment schedules have been devised, most notably a “concentration times time” schedule where methotrexate is administered initially at a dosage of 2 mg for five consecutive days every two weeks (97–99). While this approach has theoretical advantage of increasing the duration over which drug levels in the CSF are therapeutic, it is very labor-intensive and has not been demonstrated to be superior to conventional treatment schedules. Methotrexate is absorbed into the systemic circulation by the bulk flow of CSF or by the choroid plexus. Administering folinic acid on the day following treatment for a total of three days can minimize common

194

Part IV / Direct Complications of Cancer

systemic toxicities such as myelosuppression and mucositis. Because folinic acid does not penetrate the BBB to a significant degree, the administration of this antidote does not interfere with the efficacy of methotrexate in the CSF. Cytarabine is a common alternative to methotrexate, particularly for patients with hematogenous malignancies and is used commonly following methotrexate failure (100,101). This drug is administered initially at a dosage of 50 mg twice weekly and tapered in a manner similar to that of methotrexate. The recent availability of a liposome-encapsulated cytarabine formulation (DepoCyt) that can be administered once every two weeks may simplify treatment for patients receiving this drug (102). Cytarabine is rapidly metabolized in the systemic circulation, but in the CSF its half-life is prolonged due to the absence of cytidine deaminase in the CNS. DepoCyt produced a 71% cytologic response rate in patients with leptomeningeal lymphoma as compared to 15% for cytarabine (103). Additionally, treatment with DepoCyt was associated with a statistically significant improvement in KPS. However, despite this impressive difference, improvements in time to neurological progression and survival were less impressive and not statistically significant. A study describing the experience in solid tumor leptomeningeal disease (LMD) suggested DepoCyt significantly prolonged time to neurological progression but not overall median survival (104). The fewer administrations required with DepoCyt provide an advantage over conventional cytarabine treatment schedules but further studies of efficacy and cost-effectiveness with clinically relevant endpoints are needed. Some clinicians prefer to alternate methotrexate with cytarabine or alternatively use a combination regimen. The benefit or superiority of combination versus single agent regimens is still the subject of debate (105–107). Hitchins et al. reported a prospective randomized trial involving 44 patients with leptomeningeal metastases. Patients were randomized to methotrexate treatment versus methotrexate with Ara-C. Most patients also reportedly received hydrocortisone to minimize arachnoiditis. The results of this trial suggested that single-agent intrathecal methotrexate was better than the combination regimen with methotrexate and Ara-C although the difference did not reach statistical significance. There was also a trend towards increased median overall survival in the methotrexate-treated group versus the methotrexate/Ara-C combination regimen, although once again this did not reach statistical significance (108). Dae-Young Kim et al reported an opposite finding where a retrospective analysis of patients with leptomeningeal metastases treated with the combination of Ara-C/methotrexate and hydrocortisone was found to be superior to intrathecal methotrexate alone (109). Several important findings can be derived from such trials despite their opposing results. First, the prospective randomized trial by Hitchins et al. showed that patients with small cell lung cancer responded more favorably to the combination regimen when compared with the single-agent regimen. This is in agreement with the retrospective analysis by Kim et al. There was an overall 86% response to methotrexate/Ara-C compared to 50% response to methotrexate alone in patients with small cell lung cancer. Kim et al. reported an overall response rate of 80% of patients with lung cancer. Interestingly, the Hitchins et al. trial had only 29% of its patients with small cell lung cancer as opposed to 60% in the retrospective analysis reported by Kim. This difference in patient demographics may well account for the difference in conclusions drawn from either trials. Most notably however, Kim et al. reported an overall survival benefit that reached statistical significance in the combination treatment arm although neurological improvement was not significantly different. The disparities between these two trials suggests that further trials with similar patient demographics are needed to evaluate the best approach for the treatment of leptomeningeal metastases. Careful subgroup analyses in future trials may result in the identification of tailored treatment regimens for leptomeningeal metastases based on the underlying primary malignancy. Thio-TEPA is rapidly cleared from the CSF after a standard dose of 10 mg, usually administered twice a week (73,95). For this reason, and because this prodrug requires hepatic activation, its efficacy is questioned. Nonetheless, responses have been documented and this drug is a common treatment, particularly for patients with leptomeningeal metastases form breast cancer. Occasionally, patients receiving intrathecal chemotherapy develop aseptic chemical meningitis characterized by fever, headache, neck pain, nausea, vomiting, and photophobia following treatment. This side effect can be ameliorated by the use of dexamethasone for a few days following drug administration. Late neurotoxicity manifested by the development of a leukoencephalopathy is a risk for patients receiving methotrexate and cytarabine, particularly if whole brain irradiation is also administered (110–112). It

Chapter 12 / Leptomeningeal Metastases

195

develops months after the initiation of therapy, is irreversible and manifested by progressive dementia, ataxia, and sphincter incontinence. The efficacy of intrathecal chemotherapy remains the subject of debate. Recent trials evaluating a variety of intrathecal chemotherapy regimens have failed to show clinical or radiographic responses. In a recent study, 13 patients with leptomeningeal disease from breast cancer received intrathecal combination chemotherapy with cytarabine, thiotepa, methotrexate, and hydrocortisone. None of the evaluable patients showed any objective response or relief of clinical symptoms (113). Another study examined the role of intrathecal thiotepa in children with leptomeningeal metastases. In this study, the data from 15 patients were analyzed. The patient population included those with leptomeningeal dissemination from medulloblastoma, anaplastic astrocytoma, glioblastoma, retinoblastoma, neuroblastoma, rhabdomyosarcoma, non-Hodgkin lymphoma, acute lymphoblastic leukemia, and acute myelogenous leukemia. Although 4 of 15 patients showed cytologic responses in the form of clearance of malignant cells from the CSF, none of the patients showed radiographic response. In addition, median survival was only 15.1 weeks (114). Although such negative results call into question the role of intrathecal chemotherapy, these studies are limited by small sample sizes as well as the heterogeneity of primary cancers from which leptomeningeal metastases originated. Further refinements in the choice of chemotherapeutic agents administered based on the primary malignancy may result in improved responses.

8.3. Systemic Chemotherapy The management of leptomeningeal metastases rarely involves systemic chemotherapy because most agents do not penetrate the BBB in sufficient concentration when administered systemically. Furthermore, the interval between systemic treatments necessary for bone marrow recovery usually results in inadequate exposure interval in the CSF. However, in patients with leptomeningeal metastases, the BBB is disrupted and adequate blood levels within the CSF can sometimes be attained after systemic administration (115–120). A potential advantage of systemic chemotherapy is the ability of this approach to reach all regions of the subarachnoid space. Furthermore, bulky leptomeningeal metastases that would not be treated effectively by intrathecal chemotherapy may respond to drugs that can penetrate the tumor via the circulation. Agents that can be used systemically for leptomeningeal metastases include lipophilic drugs such as thioTEPA or chemotherapies such as methotrexate or Ara-C which can safely be administered systemically at high concentrations such that sufficient drug penetrates the subarachnoid space to create therapeutic CSF levels. Meningeal leukemia and lymphoma have been treated effectively in this way but managing leptomeningeal metastases from solid tumors has proven more challenging, in part because of drug resistance and the availability of few chemotherapeutic choices. However, Siegal and associates have had promising preliminary results substituting systemic for intrathecal chemotherapy and thereby avoiding complications of intrathecal therapy for selected patients with leptomeningeal metastases from solid tumors, suggesting that this approach is worthy of further investigation (121,122). Recent reports indicate that directed systemic therapies can also result in stabilization of leptomeningeal disease in some patients. A recent case report describes a patient with advanced breast cancer presenting with leptomeningeal as well as parenchymal metastases treated with capecitabine. The patient has reportedly been stable for 3.7 years despite a persistently positive CSF cytology. This patient was also treated with whole brain radiation therapy (123). Additionally, leptomeningeal metastases from breast cancer may potentially be responsive to hormonal manipulation. A recent report suggests stabilization of leptomeningeal disease in a breast cancer patient treated with letrozole (124). Letrozole is a nonsteroidal aromatase inhibitor that inhibits the aromatase enzyme, which catalyzes the conversion of androgens to estrogens. Other targeted approaches may be useful in the treatment of leptomeningeal metastases. For example, targeting the tyrosine kinase receptors with small molecule inhibitors such as gefitinib and erlotinib is increasingly playing a role in the management of non-small cell lung cancer, especially those with specific epidermal growth factor receptor mutations. While there is no conclusive evidence for the use of such tyrosine kinase inhibitors in the management of leptomeningeal dissemination from lung cancer, a recent report showed improvement in neurologic symptomatology in a patient with leptomeningeal metastases from lung cancer following gefitinib use (125). In the future, the molecular profiling of malignant

196

Part IV / Direct Complications of Cancer

cells in the CSF may guide the choice of therapeutic agents and result in improved outcome following treatment of leptomeningeal metastases.

8.4. Surgery Surgery plays a very limited role in the management of leptomeningeal metastases. The most common surgical procedure performed in this group of patients is the placement of an intraventricular catheter with a subgaleal reservoir (Ommaya reservoir) to facilitate the administration of intraventricular chemotherapy. This procedure is usually uncomplicated and this device can usually be accessed frequently with rare complications. Patients with symptomatic hydrocephalus are occasionally candidates for CSF diversion via a ventriculoperitoneal shunt. Symptomatic hydrocephalus includes patients with intractable headaches, papilledema with threatened visual loss, impaired mentation or consciousness and recurrent plateau waves. Such patients can sometimes be managed with steroids and whole brain or skull base radiotherapy, but many eventually require shunt placement. The placement of a shunt is one of the most effective palliative treatments for patients with leptomeningeal metastases and results in rapid resolution or improvement of symptoms due to increased intracranial pressure. Following shunting, intrathecal chemotherapy becomes problematic and can be administered only if a catheter with an on–off valve has been inserted. This device allows the instillation of drug into the ventricle when the valve is in the off position, thereby preventing reflux of injected drug into the catheter and eventually the peritoneal cavity. However, in patients with hydrocephalus, CSF flow is markedly disturbed and drug injected when the valve is in the off position usually remains in the ventricle and does not reach most of the remaining subarachnoid space. Furthermore, drug often penetrates the ventricular ependyma, eventually causing neurotoxicity if the administration is repetitive. Finally, many patients with symptomatic hydrocephalus cannot tolerate having the shunt turned off for several hours. In addition, shunts with on–off valves are often cumbersome to use and frequently malfunction. For these reasons, it may be best to avoid intrathecal chemotherapy for patients who require CSF diversion.

9. PROGNOSIS The prognosis for most patients with leptomeningeal metastases is poor. Untreated patients usually succumb to their illness within 6–8 weeks from their initial diagnosis (6,7). Despite an extensive literature documenting treatment approaches in usually small series of often-disparate patients with leptomeningeal metastases, there is no standard approach to management. Even with aggressive treatment, the average survival for most patients with leptomeningeal metastases from solid tumors is in the range of 4–6 months (126). Therapy is unlikely to improve neurologic disability and symptoms other than pain. However, treatment can stabilize or improve up to 75% of patients for a period of time and for this reason treatment is offered (27,127,128). Furthermore, treatment can often preserve quality of life for weeks to months for these patients. Treatment can also result in improvement of CSF abnormalities, with disappearance of malignant cells, reduction in tumor marker levels, and normalization of CSF glucose and protein concentrations. Selection of appropriate patients for aggressive therapy is often based on a number of prognostic factors that appear to predict survival. Favorable prognostic factors include good performance status, minimal neurological symptoms and signs (77), the absence of bulky leptomeningeal disease, parenchymal CNS metastases, epidural cord compression, limited systemic disease, and a chemosensitive tumor (27,82,126,129). Recently, radionuclide CSF flow studies have been used to predict response to therapy and survival and should be considered for patients who are to receive intrathecal chemotherapy (77,79). Patients with leptomeningeal leukemia and lymphoma have the best prognosis and should be treated aggressively (93,130,131). Intrathecal methotrexate and cytarabine can eradicate these tumor cells from the CSF. The recent DepoCyt studies suggested a cytologic response rate of approximately 71% compared to only 20% for LMD from solid tumors (103,104). Patients with hematologic malignancies producing LMD can remain in a sustained remission for months to years following early and intensive therapy. Of solid tumors, breast cancer appears to have the best prognosis, with Ongerboer de Visser and associates reporting a 1-year survival of 25% for patients receiving intensive intraventricular chemotherapy (128). Moreover, a minority of patients with leptomeningeal

Chapter 12 / Leptomeningeal Metastases

197

metastases from breast carcinoma can have indolent disease that can be controlled with therapy for two years or more. Patients with leptomeningeal metastases from small cell lung cancer sometimes respond to therapy and also should be considered for treatment (132,133). The prognosis for patients with leptomeningeal metastases from non-small cell lung cancer, other adenocarcinomas, and melanoma is so poor that many investigators have seen little justification for aggressive therapy (126).

10. CONCLUSIONS Leptomeningeal metastases have become a well-described and increasingly recognized complication of cancer. Recent advances in diagnostic radiology have increased the ability of the clinician to make this diagnosis and to make therapeutic decisions. The incidence of this disorder will continue to rise as treatments for systemic cancer become more effective. Unfortunately, the prognosis for this illness remains grim even for patients who receive aggressive multimodal therapy. Reasons for our failure to make a significant impact on the course of this disease are multiple and include limited therapeutic choices, the challenge of having to treat the entire subarachnoid compartment, the development of drug resistance, progressive systemic and CNS parenchymal disease, the limited capacity of patients to tolerate further treatment, and the toxicity associated with the currently available therapies. Although leptomeningeal metastases have been the focus of numerous clinical studies, no therapy has emerged as a standard treatment for this disorder. Indeed, the overall efficacy of common treatments such as intrathecal methotrexate has been called into question in recent years, and several investigators have suggested that intrathecal chemotherapy has no convincing role to play in this disorder (134). Provocative reports suggesting that systemic chemotherapy can result in clinical outcomes no different from treatments using intrathecal chemotherapies suggest that systemic chemotherapy may have a more important role in this disorder. For consensus to be reached on several key therapeutic dilemmas, future studies will need to select carefully appropriate and homogeneous study populations and define rigorous response criteria. Given the limited efficacy of therapeutic choices at present, future studies should focus on identifying subsets of patients who could benefit from the limited available therapies. Recent evidence suggests that patients with epidural cord compression, CNS metastases and bulky leptomeningeal disease have poor outcomes, but more work is required to identify reliable prognostic factors. Radionuclide ventriculography appears to have promise here as a clinical tool for evaluating prognosis. Further clinical and laboratory research is sorely needed to improve the outcome of this devastating complication of cancer. There is a desperate need for new and improved therapies. Agents in development for possible intrathecal administration include temozolomide, mafosfamide, and topotecan (135). It is likely that liposomal cytarabine will be of increasing use for patients with leptomeningeal leukemia and lymphoma. Furthermore, novel approaches using immunotoxins and gene therapy, although in clinical or early clinical development, have shown encouraging results in animal models and hold promise as emerging future therapeutic options (136). Further research is required on the mechanisms whereby conventional therapies cause neurotoxicity. The mechanisms whereby intrathecal chemotherapy and radiotherapy cause neurologic dysfunction remain poorly understood (137). A better understanding of how treatments cause nervous system damage may yield improved and less toxic therapies. Unfortunately, most patients with leptomeningeal disease die quickly of progressive leptomeningeal or systemic cancer; significant advances in the prognosis of this disorder await the development of more effective and less toxic drugs and improved drug-delivery systems.

REFERENCES 1. 2. 3. 4.

Eberth CJ. Zur Entwickelung des Epithelioms (cholesteatoms) der Pia und der Lunge. Virchows Arch 1870;49:51–63. Siefert E. Uber die multiple Karzinomatose des Zentralnerven-systems. Munchener Medizinische Wochenschrift 1902;49:826–828. Shapiro WR, Posner JB, Ushio Y et al. Treatment of meningeal neoplasms. Cancer Treat Rep 1977;61(4):733–743. Nugent JL, Bunn PA, Jr., Matthews MJ et al. CNS metastases in small cell bronchogenic carcinoma: increasing frequency and changing pattern with lengthening survival. Cancer 1979;44(5):1885–1893. 5. Bleyer WA, Poplack DG. Prophylaxis and treatment of leukemia in the central nervous system and other sanctuaries. Sem Oncol 1985;12(2):131–148. 6. Olson ME, Chernik NL, Posner JB. Infiltration of the leptomeninges by systemic cancer: a clinical and pathologic study. Arch Neurol 1974;30(2):122–137.

198

Part IV / Direct Complications of Cancer

7. Rosen ST, Aisner J, Makuch RW et al. Carcinomatous leptomeningitis in small cell lung cancer: a clinicopathologic review of the National Cancer Institute experience. Medicine 1982;61(1):45–53. 8. Yap HY, Yap BS, Tashima CK et al. Meningeal carcinomatosis in breast cancer. Cancer 1978;42(1):283–286. 9. Gonzalez-Vitale JC, Garcia-Bunuel R. Meningeal carcinomatosis. Cancer 1976;37(6):2906–2911. 10. Wasserstrom WR, Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49(4):759–772. 11. Little JR, Dale AJ, Okazaki H. Meningeal carcinomatosis: clinical manifestations. Arch Neurol 1974;30(2):138–143. 12. Boyle R, Thomas M, Adams JH. Diffuse involvement of the leptomeninges by tumour: a clinical and pathological study of 63 cases. Postgrad Med J 1980;56(653):149–158. 13. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol 1978;19:579–592. 14. Takakura K, Sano K, Hojo Sea. Metastatic Tumors of the Central Nervous System. Tokyo: Igaku-Shoin Ltd.; 1982. 15. Lee YT. Breast carcinoma: pattern of metastasis at autopsy. J Surg Oncol 1983;23(3):175–180. 16. Tsukada Y, Fouad A, Pickren JW et al. Central nervous system metastasis from breast carcinoma: autopsy study. I 1983;52(12): 2349–2354. 17. Garson JA, Brownell B, Allan PM et al. Malignant cells in cerebrospinal fluid (CSF): the meaning of a positive CSF cytology. Lancet 1984;1(8386):1095–1098. 18. Kesari S, Batchelor TT. Leptomeningeal metastases. Neurol Clin 2003;21(1):25–66. 19. Chamberlain MC, Corey-Bloom J. Leptomeningeal metastases: 111 indium-DTPA CSF flow studies. Neurology 1991;41(11):1765–1769. 20. Dekker AW, Elderson A, Punt K et al. Meningeal involvement in patients with acute nonlymphocytic leukemia: incidence, management, and predictive factors. Cancer 1985;56(8):2078–2082. 21. Ersboll J, Schultz HB, Thomsen BL et al. Meningeal involvement in non-Hodgkin’s lymphoma: symptoms, incidence, risk factors and treatment. Scand J Haematol 1985;35(5):487–496. 22. Amer MH, Al-Sarraf M, Baker LH et al. Malignant melanoma and central nervous system metastases: incidence, diagnosis, treatment and survival. Cancer 1978;42(2):660–668. 23. Awad I, Bay JW, Rogers L. Leptomeningeal metastasis from supratentorial malignant gliomas. Neurosurgery 1986;19(2):247–251. 24. Yung WA, Horten BC, Shapiro WR. Meningeal gliomatosis: a review of 12 cases. Ann Neurol 1980;8(6):605–608. 25. Packer RJ, Siegel KR, Sutton LN et al. Leptomeningeal dissemination of primary central nervous system tumors of childhood. Ann Neurol 1985;18(2):217–221. 26. DeAngelis LM. Primary central nervous system lymphoma: a new clinical challenge. Neurology 1991;41(5):619–621. 27. Balm M, Hammack J. Leptomeningeal carcinomatosis: resenting features and prognostic factors. Arch Neurol 1996;53:626–632. 28. Kaplan JG, DeSouza TG, Farkash A et al. Leptomeningeal metastases: comparison of clinical features and laboratory data of solid tumors, lymphomas and leukemias. J Neuro-oncol 1990;9(3):225–229. 29. Theodore WH, Gendelman S. Meningeal carcinomatosis. Arch Neurol 1981;38(11):696–699. 30. Yap HY, Tashima CK, Blumenschein GR et al. Diabetes insipidus and breast cancer. Arch Internal Med 1979;139:1009–1011. 31. Lyster MT, Kies MS, Kuzel TM. Neurologic complications of patients with small cell prostate carcinoma: report of two cases. Cancer 1994;74(12):3159–3163. 32. MacKintosh FR, Colby TV, Podolsky WJ et al. Central nervous system involvement in non-Hodgkin’s lymphoma: an analysis of 105 cases. Cancer 1982;49(3):586–595. 33. Price RA, Johnson WW. The central nervous system in childhood leukemia. I. The arachnoid. Cancer 1973;31(3):520–533. 34. Norris LK, Grossman SA, Olivi A. Neoplastic meningitis following surgical resection of isolated cerebellar metastasis: a potentially preventable complication. J Neuro-Oncol 1997;32(3):215–223. 35. Chernik NL, Kim JH, Posner JB et al. Malignant cells in cerebrospinal fluid (CSF): the meaning of a positive CSF cytology. Neurology 1979;29(9 Pt 1):1195–1202. 36. Grossman SA, Trump DL, Chen DC et al. Cerebrospinal fluid flow abnormalities in patients with neoplastic meningitis: an evaluation using 111 indium-DTPA ventriculography. Amer J Med 1982;73(5):641–647. 37. Klein P, Haley EC, Wooten GF et al. Focal cerebral infarctions associated with perivascular tumor infiltrates in carcinomatous leptomeningeal metastases. Arch Neurol 1989;46(10):1149–1152. 38. Hiesiger EM, Picco-Del Bo A, Lipschutz LE et al. Experimental meningeal carcinomatosis selectively depresses local cerebral glucose utilization in rat brain. Neurology 1989;39(1):90–95. 39. Glantz MJ, Cole BF, Glantz LK et al. Cerebrospinal fluid cytology in patients with cancer: minimizing false-negative results. Cancer 1998;82(4):733–739. 40. Chamberlain MC, Kormanik PA, Glantz MJ. A comparison between ventricular and lumbar cerebrospinal fluid cytology in adult patients with leptomeningeal metastases. Neuro-oncol 2001;3(1):42–45. 41. Boogerd W, Vroom TM, van Heerde P et al. CSF cytology versus immunocytochemistry in meningeal carcinomatosis. J Neurol Neurosurg Psychiatry 1988;51(1):142–145. 42. Coakham HB, Garson JA, Brownell B et al. Use of monoclonal antibody panel to identify malignant cells in cerebrospinal fluid. Lancet 1984;1(8386):1095–1098. 43. Jorda M, Ganjei-Azar P, Nadji M. Cytologic characteristics of meningeal carcinomatosis: increased diagnostic accuracy using carcinoembryonic antigen and epithelial membrane antigen immunocytochemistry. Arch Neurol 1998;55(2):181–184. 44. Schold SC, Wasserstrom WR, Fleisher M et al. Cerebrospinal fluid biochemical markers of central nervous system metastases. Ann Neurol 1980;8(6):597–604. 45. Fleisher M, Wasserstrom WR, Schold SC et al. Lactic dehydrogenase isoenzymes in the cerebrospinal fluid of patients with systemic cancer. Cancer 1981;47(11):2654–2659.

Chapter 12 / Leptomeningeal Metastases

199

46. Malkin MG, Posner JB. Cerebrospinal fluid tumor markers for the diagnosis and management of leptomeningeal metastases. Eur J Cancer Clin Oncol 1987;23(1):1–4. 47. Cibas ES, Malkin MG, Posner JB et al. Detection of DNA abnormalities by flow cytometry in cells from cerebrospinal fluid. Amer J Clin Pathol 1987;88(5):570–577. 48. van Oostenbrugge RJ, Hopman AH, Lenders MH et al. Detection of malignant cells in cerebrospinal fluid using fluorescence in situ hybridization. J Neuropathol Exp Neurol 1997;56(6):743–748. 49. van Oostenbrugge RJ, Hopman AH, Ramaekers FC et al. In situ hybridization: a possible diagnostic aid in leptomeningeal metastasis. J Neuro-oncol 1998;38(2–3):127–33. 50. van Oostenbrugge RJ, Hopman AH, Arends JW et al. Treatment of leptomeningeal metastases evaluated by interphase cytogenetics. J Clin Oncol 2000;18(10):2053–2058. 51. Rhodes CH, Glantz MJ, Glantz L et al. A comparison of polymerase chain reaction examination of cerebrospinal fluid and conventional cytology in the diagnosis of lymphomatous meningitis. Cancer 1996;77(3):543–548. 52. Straathof CS, de Bruin HG, Dippel DW et al. The diagnostic accuracy of magnetic resonance imaging and cerebrospinal fluid cytology in leptomeningeal metastasis. J Neurol 1999;246(9):810–814. 53. Brandsma D, Voest EE, de Jager W et al. CSF protein profiling using multiplex immuno-assay: a potential new diagnostic tool for leptomeningeal metastases. J Neurol 2006;253(9):1177–1184. 54. Brandsma D, Taphoorn MJ, de Jager W et al. Interleukin-8 CSF levels predict survival in patients with leptomeningeal metastases. Neurology 2006;66(2):243–246. 55. van de Langerijt B, Gijtenbeek JM, de Reus HP et al. CSF levels of growth factors and plasminogen activators in leptomeningeal metastases. Neurology 2006;67(1):114–119. 56. Okumura H, Natsugoe S, Yokomakura N et al. A case of leptomeningeal carcinomatosis from esophageal basaloid carcinoma diagnosed by quantitative reverse transcription-polymerase chain reaction for carcinoembryonic antigen. J Gastroenterol 2005;40(1):87–93. 57. Dekker LJ, Boogerd W, Stockhammer G et al. MALDI-TOF mass spectrometry analysis of cerebrospinal fluid tryptic peptide profiles to diagnose leptomeningeal metastases in patients with breast cancer. Mol Cell Proteomics 2005;4(9):1341–1349. 58. Hinckley L, Teh BS, Elledge R et al. Guidelines on routine cerebrospinal fluid analysis: report from an EFNS task force. Clinical Breast Cancer 2006;7(2):164–166. 59. Freilich RJ, Krol G, DeAngelis LM. Neuroimaging and cerebrospinal fluid cytology in the diagnosis of leptomeningeal metastasis. Ann Neurol 1995;38(1):51–57. 60. Gomori JM, Heching N, Siegal T. Leptomeningeal metastases: evaluation by gadolinium-enhanced spinal magnetic resonance imaging. J Neuro-oncol 1998;36(1):55–60. 61. Kros JM, Avezaat CJ, Valerio D et al. Presenting features and value of diagnostic procedures in leptomeningeal metastases. Human Gene Therapy 1999;10(14):2347–2354. 62. Chamberlain MC, Sandy AD, Press GA. Leptomeningeal metastasis: a comparison of gadolinium-enhanced MR and contrast-enhanced CT of the brain. Neurology 1990;40(3 Pt 1):435–438. 63. Jaeckle KA, Krol G, Posner JB. Evolution of computed tomographic abnormalities in leptomeningeal metastases. Ann Neurol 1985;17(1):85–89. 64. Lee YY, Glass JP, Geoffray A, Wallace S. Cranial computed tomographic abnormalities in leptomeningeal metastasis. AJR 1984;143(5):1035–1039. 65. Sze G, Abramson A, Krol G, Liu D et al. Gadolinium-DTPA in the evaluation of intradural extramedullary spinal disease. AJR. 1988;150(4):911–921. 66. Kallmes DF, Gray L, Glass JP. High-dose gadolinium-enhanced MRI for diagnosis of meningeal metastases. Neuroradiology 1998;40(1):23–26. 67. River Y, Schwartz A, Gomori JM et al. Clinical significance of diffuse dural enhancement detected by magnetic resonance imaging. J Neurosurg 1996;85(5):777–783. 68. Ba-Ssalamah A, Nobauer-Huhmann IM, Pinker K et al. Effect of contrast dose and field strength in the magnetic resonance detection of brain metastases. Invest Radiol 2003;38(7):415–422. 69. Mathews VP. Brain: gadolinium-enhanced fast fluid-attenuated inversion-recovery MR imaging. [see comment]. 70. Ercan N, Gultekin S, Celik H et al. Diagnostic value of contrast-enhanced fluid-attenuated inversion recovery MR imaging of intracranial metastases. AJNR Am J Neuroradiol 2004;25(5):761–765. 71. Singh SK, Leeds NE, Ginsberg LE. MR imaging of leptomeningeal metastases: comparison of three sequences. AJNR Am J Neuroradiol 2002;23:817–821. 72. Padma MV, Jacobs M, Kraus G et al. 11C-methionine PET imaging of leptomeningeal metastases from primary breast cancer: a case report. J Neuro-oncol 2001;55(1):39–44. 73. Komori T, Delbeke D. Leptomeningeal carcinomatosis and intramedullary spinal cord metastases from lung cancer: detection with FDG positron emission tomography. Clin Nucl Med 2001;26(11):905–907. 74. Chamberlain MC. Comparative spine imaging in leptomeningeal metastases. J Neuro-oncol 1995;23(3):233–238. 75. Chamberlain MC, Kormanik PA. Prognostic significance of 111 indium-DTPA CSF flow studies in leptomeningeal metastases. Neurology 1996;46(6):1674–1677. 76. Chamberlain MC. Current concepts in leptomeningeal metastasis. Curr Opin Oncol 1992;4(3):533–539. 77. Chamberlain MC. Radioisotope CSF flow studies in leptomeningeal metastases. J Neuro-oncol 1998;38(2–3):135–140. 78. Glantz MJ, Hall WA, Cole BF et al. Diagnosis, management, and survival of patients with leptomeningeal cancer based on cerebrospinal fluid-flow status. Cancer 1995;75(12):2919–2931. 79. Mason WP, Yeh SDJ, DeAngelis LM. 111 Indium-diethylenetriamine pentaacetic acid cerebrospinal fluid flow studies predict distribution of intrathecally administered chemotherapy and outcome in patients with leptomeningeal metastases. Neurology 1998;50:438–443.

200

Part IV / Direct Complications of Cancer

80. Rosenblum MK, Papadopoulos E, Childs BH et al. Sarcoidosis of the cauda equina mimicking leptomeningeal malignancy. J Neurooncol 1999;44(2):147–153. 81. Chamberlain MC. New approaches to and current treatment of leptomeningeal metastases. Curr Opin Neurol 1994;7(6):492–500. 82. Chamberlain MC, Kormanik PR. Carcinomatous meningitis secondary to breast cancer: predictors of response to combined modality therapy. J Neurooncol 1997;35(1):55–64. 83. Boogerd W, Hart AA, van der Sande JJ et al. Meningeal carcinomatosis in breast cancer: prognostic factors and influence of treatment. Cancer 1991;67(6):1685–1695. 84. Clamon G, Doebbeling B. Meningeal carcinomatosis from breast cancer: spinal cord vs. brain involvement. Breast Cancer Res Treat 1987;9(3):213–217. 85. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. 86. Brown MT, Coleman RE, Friedman AH et al. Intrathecal 131I–labeled antitenascin monoclonal antibody 81C6 treatment of patients with leptomeningeal neoplasms or primary brain tumor resection cavities with subarachnoid communication: phase I trial results. Clin Cancer Res 1996;2(6):963–972. 87. Coakham HB, Kemshead JT. Treatment of neoplastic meningitis by targeted radiation using (131)I-radiolabelled monoclonal antibodies: results of responses and long term follow-up in 40 patients. J Neuro-oncol 1998;38(2–3):225–232. 88. Chamberlain MC, Kormanik PA, Barba D. Complications associated with intraventricular chemotherapy in patients with leptomeningeal metastases. J Neurosurg 1997;87:694–699. 89. Obbens EA, Leavens ME, Beal JW et al. Ommaya reservoirs in 387 cancer patients: a 15-year experience. Neurology 1985;35(9): 1274–1278. 90. Shapiro WR, Young DF, Mehta BM. Methotrexate: distribution in cerebrospinal fluid after intravenous, ventricular and lumbar injections. N Engl J Med 1975;293(4):161–166. 91. Larson SM, Schall GL, Di Chiro G. The influence of previous lumbar puncture and pneumoencephalography on the incidence of unsuccessful radioisotope cisternography. J Nucl Med 1971;12(8):555–557. 92. Nabors M, Grossman S, Burch P et al. Concentrations of chemotherapeutic agents in brain following lumbar (lum) and ventricular (ven) administration. Proc Am Soc Clin Oncol; 1989; 366. 93. Bleyer WA, Poplack DG. Intraventricular versus intralumbar methotrexate for central nervous system leukemia: prolonged remission with the Ommaya reservoir. Med Ped Oncol 1979;6(3):207–213. 94. Ettinger LJ, Chervinsky DS, Freeman AI et al. Pharmacokinetics of methotrexate following intravenous and intraventricular administration in acute lymphocytic leukemia and non-Hodgkin’s lymphoma. Cancer 1982;50(9):1676–1682. 95. Trump DL, Grossman SA, Thompson G et al. Treatment of neoplastic meningitis with intraventricular thiotepa and methotrexate. Cancer Treat Rep 1982;66(7):1549–1551. 96. Bleyer AW. Clinical pharmacology of intrathecal methotrexate. II. An improved dosage regimen derived from age-related pharmacokinetics. Cancer Treat Rep 1977;61(8):1419–1425. 97. Bleyer WA, Poplack DG, Simon RM. “Concentration x time” methotrexate via a subcutaneous reservoir: a less toxic regimen for intraventricular chemotherapy of central nervous system neoplasms. Blood 1978;51(5):835–842. 98. Moser AM, Adamson PC, Gillespie AJ et al. Intraventricular concentration times time (C x T) methotrexate and cytarabine for patients with recurrent meningeal leukemia and lymphoma. Cancer 1999;85(2):511–516. 99. Petit T, Dufour P, Korganov AS et al. Continuous intrathecal perfusion of methotrexate for carcinomatous meningitis with pharmacokinetic studies: two case studies. Clin Oncol (Roy College Radiol) 1997;9(3):189–190. 100. Zimm S, Collins JM, Miser J et al. Cytosine arabinoside cerebrospinal fluid kinetics. Clin Pharmacol Therapeutics 1984;35(6):826–830. 101. Yap HY, Yap BS, Rasmussen S et al. Treatment for meningeal carcinomatosis in breast cancer. Cancer 1982;50(2):219–222. 102. Kim S, Chatelut E, Kim JC et al. Extended CSF cytarabine exposure following intrathecal administration of DTC 101. J Clin Oncol 1993;11(11):2186–2193. 103. Glantz MJ, LaFollette S, Jaeckle KA et al. Randomized trial of a slow-release versus a standard formulation of cytarabine for the intrathecal treatment of lymphomatous meningitis. [see comment]. J Clin Oncol 1999;17(10):3110–3116. 104. Glantz MJ, Jaeckle KA, Chamberlain MC et al. A randomized controlled trial comparing intrathecal sustained-release cytarabine (DepoCyt) to intrathecal methotrexate in patients with neoplastic meningitis from solid tumors. Clin Cancer Res 1999;5(11):3394–3402. 105. Stewart DJ, Maroun JA, Hugenholtz H et al. Combined intraommaya methotrexate, cytosine arabinoside, hydrocortisone and thio-TEPA for meningeal involvement by malignancies. J Neuro-oncol 1987;5(4):315–322. 106. Hitchins RN, Bell DR, Woods RL et al. A prospective randomized trial of single-agent versus combination chemotherapy in meningeal carcinomatosis. J Clin Oncol 1987;5(10):1655–1662. 107. Giannone L, Greco FA, Hainsworth JD. Combination intraventricular chemotherapy for meningeal neoplasia. J Clin Oncol 1986;4(1):68–73. 108. Hitchins RN, Bell DR, Woods RL, Levi JA. A prospective randomized trial of single-agent versus combination chemotherapy in meningeal carcinomatosis. J Clin Oncol 1987;5(10):1655–1662. 109. Kim DY, Lee KW, Yun T et al. Comparison of intrathecal chemotherapy for leptomeningeal carcinomatosis of a solid tumor: methotrexate alone versus methotrexate in combination with cytosine arabinoside and hydrocortisone. Jap J Clin Oncol 2003;33(12):608–12. 110. Bleyer WA, Drake JC, Chabner BA. Neurotoxicity and elevated cerebrospinal-fluid methotrexate concentration in meningeal leukemia. N Engl J Med 1973;289(15):770–773. 111. Sause WT, Crowley J, Eyre HJ et al. Whole brain irradiation and intrathecal methotrexate in the treatment of solid tumor leptomeningeal metastases: a Southwest Oncology Group study. J Neuro-oncol 1988;6(2):107–112. 112. Bleyer WA. Current status of intrathecal chemotherapy for human meningeal neoplasms. Natl Cancer Inst Monogr 1977;46:171–178.

Chapter 12 / Leptomeningeal Metastases

201

113. Orlando L, Curigliano G, Colleoni M et al. Intrathecal chemotherapy in carcinomatous meningitis from breast cancer. Anticancer Res 2002;22(5):3057–3059. 114. Fisher PG, Kadan-Lottick NS, Korones DN. Intrathecal thiotepa: reappraisal of an established therapy. J Ped Hematol/Oncol 2002;24(4):274–278. 115. Borsi JD, Sagen E, Romslo I et al. Comparative study on the pharmacokinetics of 7-hydroxy-methotrexate after administration of methotrexate in the dose range of 0.5–33.6 g/m2 to children with acute lymphoblastic leukemia. Med Ped Oncol 1990;18(3):217–224. 116. Borsi JD, Moe PJ. A comparative study on the pharmacokinetics of methotrexate in a dose range of 0.5 g to 33.6 g/m2 in children with acute lymphoblastic leukemia. Cancer 1987;60(1):5–13. 117. DeAngelis LM, Kreis W, Chan K et al. Pharmacokinetics of ara-C and ara-U in plasma and CSF after high-dose administration of cytosine arabinoside. Cancer Chemotherapy Pharmacol 1992;29(3):173–177. 118. Cortes J, O’Brien SM, Pierce S et al. The value of high-dose systemic chemotherapy and intrathecal therapy for central nervous system prophylaxis in different risk groups of adult acute lymphoblastic leukemia. Blood 1995;86(6):2091–2097. 119. Evans WE, Crom WR, Abromowitch M et al. Clinical pharmacodynamics of high-dose methotrexate in acute lymphocytic leukemia: identification of a relation between concentration and effect. N Engl J Med 1986;314(8):471–477. 120. Kantarjian H, Barlogie B, Plunkett W et al. High-dose cytosine arabinoside in non-Hodgkin’s lymphoma. J Clin Oncol 1983;1(11): 689–694. 121. Siegal T. Leptomeningeal metastases: rationale for systemic chemotherapy or what is the role of intra-CSF-chemotherapy? J Neurooncol 1998;38(2–3):151–157. 122. Siegal T. Leptomeningeal metastases: analysis of 31 patients with sustained off-therapy response following combined-modality therapy. Neurology 1994;44(8):1463–1469. 123. Tham YL, Hinckley L, Teh BS et al. Long–term clinical response in leptomeningeal metastases from breast cancer treated with capecitabine monotherapy: a case report. Clin Breast Cancer 2006;7(2):164–166. 124. Ozdogan M, Samur M, Bozcuk HS et al. Durable remission of leptomeningeal metastasis of breast cancer with letrozole: a case report and implications of biomarkers on treatment selection. Jap J Clin Oncol 2003;33(5):229–231. 125. Sakai M, Ishikawa S, Ito H et al. Carcinomatous meningitis from non-small cell lung cancer responding to gefitinib. Int J Clin Oncol 2006;11(3):243–245. 126. Grant R, Naylor B, Greenberg HS et al. Clinical outcome in aggressively treated meningeal carcinomatosis. Arch Neurol 1994;51(5):457–461. 127. Chamberlain MC, Kormanik PA. Prognostic significance of 111 indium-DTPA CSF flow studies in leptomeningeal metastases. Neurology 1996;46(6):1674–1677. 128. Ongerboer de Visser BW, Somers R, Nooyen WH et al. Intraventricular methotrexate therapy of leptomeningeal metastasis from breast carcinoma. Neurology 1983;33(12):1565–1572. 129. Chamberlain MC, Kormanik PA. Prognostic significance of coexistent bulky metastatic central nervous system disease in patients with leptomeningeal metastases. Arch Neurol 1997;54(11):1364–1368. 130. Bleyer WA, Poplack DG. Prophylaxis and treatment of leukemia in the central nervous system and other sanctuaries. Sem Oncol 1985;12(2):131–148. 131. Steinherz P, Jereb B, Galicich J. Therapy of CNS leukemia with intraventricular chemotherapy and low-dose neuraxis radiotherapy. J Clin Oncol 1985;3(9):1217–1226. 132. Aroney RS, Dalley DN, Chan WK et al. Meningeal carcinomatosis in small cell carcinoma of the lung. Amer J Med 1981;71(1):26–32. 133. Balducci L, Little DD, Khansur T et al. Carcinomatous meningitis in small cell lung cancer. Amer J Med Sci 1984;287(1):31–33. 134. Siegal T. Leptomeningeal metastases: rationale for systemic chemotherapy or what is the role of intra-CSF-chemotherapy? J Neurooncol 1998;38(2–3):151–157. 135. Blaney SM, Poplack DG. New cytotoxic drugs for intrathecal administration. J Neuro-oncol 1998;38(2–3):219–223. 136. Herrlinger U, Weller M, Schabet M. New aspects of immunotherapy of leptomeningeal metastasis. J Neuro-oncol 1998;38(2–3): 233–239. 137. Gilbert MR, Harding BL, Grossman SA. Methotrexate neurotoxicity: in vitro studies using cerebellar explants from rats. Cancer Research 1989;49(9):2502–2505.

13

Peripheral Nervous System Metastases Nicholas Butowski,

MD

CONTENTS Introduction Metastases to Metastases to Metastases to Metastases to Metastases to Conclusions References

Peripheral Nerves Muscles the Cervical Plexus the Brachial Plexus the Lumbosacral Plexus

Summary Cancer is an important differential diagnosis when evaluating dysfunction of the peripheral nervous system (PNS). While the toxic effects of anticancer drugs and paraneoplastic disorders are to be kept in mind, one should also consider multilevel metastases to the PNS resulting in dysfunction by infiltration or direct compression. These metastases can result in severe, unrelenting pain. Later, weakness and focal sensory disturbances occur in the distribution of the nerve(s) involved. In previously treated patients, the main differential diagnostic consideration is radiation-induced plexopathy. Treatment of metastases to the PNS is generally palliative and includes radiotherapy, chemotherapy, and pain management. The response to therapy is variable and commonly difficult to maintain. Every effort should be made on behalf of the patient to provide satisfactory pain control and to maximize neurologic function and quality of life. Key Words: peripheral nervous system, cancer, metastasis

1. INTRODUCTION Approximately 20% of cancer patients experience neurologic dysfunction at some point in their care. Cancer can cause neurologic dysfunction by metastasis; metabolic effects of systemic organ dysfunction; vascular disorders, including thrombosis, embolism, and hemorrhage; and paraneoplastic disorders. The toxic effects of anticancer agents also may cause neurologic dysfunction. In regard to the peripheral nervous system (PNS), the most frequent neurologic complications are due to the toxic side effects of treatment, either chemotherapy or radiation. Conversely and less commonly, cancer may affect the PNS by infiltration of nerves or muscles or by local compression. The purpose of this chapter is to review the incidence and types of metastases to the PNS and to review their clinical presentation, differential diagnosis, and treatment.

2. METASTASES TO PERIPHERAL NERVES The tendency of cancer to invade or compress cranial and peripheral nerves is rare in comparison with its tendency to invade the central nervous system. However, cancer can affect peripheral nerve function in many From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

203

204

Part IV / Direct Complications of Cancer

ways. One such way is for cancer to metastasize to bone and compress a peripheral nerve that is in close proximity. For example, a tumor at the axilla or elbow may result in ulnar nerve dysfunction by direct compression; similarly, intercostal nerves can be affected by metastasis to ribs and the sciatic nerve can be compressed at several places along its course. Metastases to the base of the skull may result in several syndromes by compression of surrounding nerves as well (1). Cancer of the breast, lung, and prostate tend to spread to the skull base more often than other types of cancer. Clinical syndromes correlate with the anatomic location of the metastasis and resulting nerves involved. The occipital condyle syndrome results in ipsalateral neck and posterior head pain along with dysarthria, dysphagia, and tongue weakness from hypoglossal nerve involvement. The jugular foramen syndrome results in hoarseness and dysphagia from involvement of the glossopharyngeal and vagus nerves; unilateral vocal cord paralysis may also be seen. The parasellar syndrome consists of unilateral frontal headache and extraocular palsies from infiltration of the cavernous sinus and cranial nerves III, IV, and V1, and VI. The orbital syndrome results in pain, blurred vision, and diplopia; opthalmoplegia and proptosis may also be present. All of these syndromes require prompt clinical confirmation by computed tomography (CT) with bone windows or magnetic resonance imaging (MRI). Patients may experience symptomatic relief after skull-based radiotherapy. However, corticosteroids may be warranted prior to radiotherapy to avoid vision loss secondary to radiation-induced edema.

2.1. Compression of Nerves Metastases to the mandible or base of the skull may cause dysfunction of the inferior alveolar nerve, which results in a set of symptoms known as “numb chin syndrome” (2,3). A retrospective study revealed that breast cancer and lymphoproliferative neoplasms are the most common types of cancer to cause this syndrome, which consists of numbness of the chin, lower lip, and gum. Median survival after diagnosis of this syndrome is poor: 5 months when due to bone metastases and 12 months if associated with leptomeningeal seeding (2). Recognizing this syndrome is important as it may be the first sign that a patient has cancer. Depending on the underlying cancer, treatment with radiotherapy or chemotherapy may alleviate some symptoms (3,4). In contradistinction to bone metastasis with resulting compression, cancer may also spread directly to cranial or peripheral nerves and result in dysfunction. For example, metastasis to the trigeminal ganglion may produce facial parethesias and cheek or jaw pain. These symptoms generally quickly progress to numbness and weakness of the motor branch of the trigeminal nerve (1,5). There are many other reported cases of cancer-induced cranial and peripheral neuropathy by direct metastases to peripheral nerves with expected neuro-anatomical dysfunction (6,7). Of course, cancer also may cause neurologic dysfunction by local extension such as when lung cancer causes vocal cord paralysis through invasion of the recurrent laryngeal nerve through invasion of the mediastinal lymph nodes. Local extension may also affect the phrenic nerve or even the cervical sympathetic chain.

2.2. Intraneural Metastases Intraneural metastases are rare. However, there have been cases of intraneural metastasis by several types of cancer including synovial sarcoma and carcinoid, which may even result in multiple mononeuropathies, resembling mononeuritis multiplex (8–10). Lymphoma may also infiltrate the PNS, and is referred to as neurolymphomatosis (11–14). 2.2.1. Neurolymphomatosis Neurolymphomatosis may be the first sign of systemic lymphoma. Different types of lymphoma may be associated with this disorder though the most common type is large B-cell lymphoma that arises from systemic or central nervous system (CNS) lymphoma. A review of 72 patients with neurolymphomatosis identified four clinical presentations: (i) painful involvement of multiple peripheral nerves or roots, (ii) cranial neuropathy with or without pain, (iii) painless involvement of peripheral nerves, and (iv) painful or painless involvement of a single peripheral nerve (14). Symptom presentation is usually subacute but may be relapsing/remitting and chronic. Demonstrating the diagnosis of neurolymphomatosis is difficult and often cannot be done until the time of autopsy. Making the correct diagnosis requires a high index of suspicion and the appropriate work-up, which generally includes

Chapter 13 / Peripheral Nervous System Metastases

205

biopsy of the relevant nerve, cerebrospinal fluid (CSF) studies, imaging, and electrophysiological tests. Nerve biopsy is not always successful and CSF studies may vary and lead one to incorrect diagnoses such as Guillain– Barré syndrome. MRI can be used to demonstrate nerve enhancement, and positron emission tomography (PET) imaging may also show the lesions extending along the course of the nerve, but such findings are also seen with other disorders (13,15). Treatment entails systemic chemotherapy, which can reach multiple potential sites of involvement. However, even when treated, neurolymphomatosis carries a poor prognosis similar to primary CNS lymphoma. 2.2.2. Neuroleukemia Leukemia may also infiltrate peripheral nerves. The incidence of any type of leukemia metastasizing to peripheral nerves is extremely low, but there are several reported cases (16–19). Patients may present before diagnosis or while in remission with any number of neurologic problems including weakness, loss of sensation, or pain (Fig. 1). While rare, the complication should clearly be suspected in leukemic patients with signs of peripheral nerve dysfunction. 2.2.3. Other PNS Tumors The PNS may also be affected by tumors that arise from the peripheral nerves themselves. For example, malignant peripheral nerve sheath tumors (MPNSTs) are cancers that account for approximately 5–10% of all soft tissue sarcomas and histologically resemble fibrosarcomas (20). They occur either sporadically, in association with neurofibromatosis type 1 (NF1), or subsequent to radiation therapy. At the molecular level, loss of the NF1 gene and high levels of Ras activity are hallmarks. MPNST share similar prognostic factors with patients who have other soft tissue sarcomas and have some of the worst clinical outcomes (21). In a large Italian series of MPNST patients the presence of NF1 did not affect survival, but patients with NF1 were more likely to have larger tumors. Most MPNSTs are schwannian in nature, but a recent study characterized the pathologic features of MPNST with perineural cell differentiation and concluded that 4% of MPNST show perineural cell differentiation with an uncertain association with NF1 (22). Structurally the perineural MPNST possessed an immunophenotype that indicated ultrastructural perineural differentiation. The prognosis of perineural MPNST appeared to be more favorable than that of conventional MPNST. MRI is the most helpful imaging technique to identify the extent of MPNST (Fig. 2). The presence of heterogeneity with evidence of necrosis and hemorrhage on MRI and increased uptake on PET scan may prove

Fig. 1. A 31-year-old patient with acute promyelocytic leukemia, in remission, presented with right-sided sciatica. Lumbar puncture was negative for tumor cells. MRI of the lumbar spine showed an enlarged S1 nerve root lesion (arrow) extending from the S1 foramen to pelvic end of the foramen thought to be a schwannoma. Biopsy of the lesion confirmed leukemia. (Courtesy of Dr. Santosh Kesari, Dana-Farber/Brigham and Women’s Cancer Center, Boston, MA.).

206

Part IV / Direct Complications of Cancer

Fig. 2. A 29-year-old patient with history of NF-1 presented with rapidly enlarging mass on left leg. MRI showed a large plexiform mass arising from anterior iliac wing, exiting out along the course of femoral nerve in the left leg (A) with atrophy of left leg muscles. The lesion enhanced avidly (B) with mixed signal on T1-weighted images. Pathology revealed an atypical neurofibroma with features concerning for MPNST. (Courtesy of Dr. Santosh Kesari, Dana-Farber/Brigham and Women’s Cancer Center, Boston, MA.).

helpful in detecting malignant changes. MPNSTs are treated by a multidisciplinary team, with the aim of complete surgical removal of the lesion with wide margins followed by radiotherapy (23).

3. METASTASES TO MUSCLES The diagnosis of cancer metastasis to skeletal muscle should be considered in patients with systemic cancer who develop a soft tissue mass within muscle that may or may not be painful. Other presenting symptoms obviously depend on the muscle(s) involved. Metastases to muscle from malignant tumors have a reported incidence that varies among autopsy studies from 1% to nearly 18% (24,25). As such, a broad differential diagnosis (trauma, abscess, hemorrhage) should be considered along with potential metastases. Clinical reports describe numerous types of cancers that may spread to muscle though lung cancer, hematological cancers, and melanoma are among the more common (26–29). When metastases to muscle do occur, any muscle may be affected. One study found that the muscles most commonly involved were the diaphragm and the iliopsoas (25,26).

3.1. Pathophysiology The low frequency of clinically detected muscular metastases contrasts with the large volume of the musculature and its abundant blood supply. However, it appears that muscle is resistant both to primary and to metastatic cancer. This fact may be explained by muscle’s lactic acid production, which has been shown to inhibit growth of intramuscular metastases (30). Other investigators have suggested that blood flow in skeletal muscle is variable owing to -adrenergic receptors and therefore subject to varying degrees of tissue pressure that may inhibit tumor growth (31).

Chapter 13 / Peripheral Nervous System Metastases

207

3.2. Imaging Various imaging studies can identify muscular metastases; however, none of them are specific in differentiating among carcinoma, sarcoma, and other muscular disorders (31). Ultrasound can be useful in differentiating solid from cystic masses. CT can demonstrate muscle enlargement with well-defined or ill-defined masses containing low density, loss of normal fat, and vascular planes. The most common appearance of a metastasis on contrast-enhanced helical CT is a rim-enhancing intramuscular lesion with central hypoattenuation (32). MR imaging is thought to be superior to CT in its ability to detect muscle metastases (33–36); the typical appearance includes low signal intensity on T1-weighted images and high signal intensity on T2-weighted scans. Intramuscular metastases from malignant melanoma may show high signal intensity on T1-weighted images and mixed signal intensities containing high and low signals on T2-weighted images. The signal intensity on T1-weighted images, which is due to the paramagnetic effect of melanin, is a characteristic MR finding of this entity (34). It should be noted that primary soft tissue sarcomas often show identical signal characteristics to melanoma, making differentiation between entities difficult.

3.3. Treatment The fact that imaging modalities are not able to prove metastatic muscular involvement emphasizes the importance of a thorough histological examination. Imaging may be used to obtain tissue. Percutaneous CT-guided biopsy can be valuable in obtaining tissue (33). Therapy in patients with metastases to muscle includes surgical resection but only in localized disease. Extensive disease or residual tumor after incomplete resection warrants the use of radiation or chemotherapy. Despite treatment, the prognosis in patients with clinically evident metastases to muscles remains poor, with a median survival of less than 12 months (25,26,33,37).

4. METASTASES TO THE CERVICAL PLEXUS The cervical plexus consists of the ventral rami of the first four cervical spinal nerves (C1–C4) and is deep to the sternocleidomastoid. The plexus is located lateral to the transverse processes between prevertebral muscles from the medial side and vertebral muscles (m. scalenus, m. levator scapulae, m. splenius cervicis) from the lateral side. The cervical plexus carries sensory information from the scalp and neck and supplies motor innervation to the diaphragm, sternocleidomastoid, and trapezius muscles, middle scalene, and other muscles of neck flexion.

4.1. Presentation Metastasis to cervical region lymph nodes from cancer of the head and neck, lung cancer, or breast cancer are the most frequent causes of cancer related cervical plexopathies (38). Patients generally present with pain and stiffness in the neck or shoulder. Depending on the nerves involved there may be loss of sensation over the scalp or motor weakness of neck flexors or hemidiaphragm. Often, the metastatic disease will also involve the brachial plexus or base of the skull (see relevant sections for a more detailed discussion). The spinal cord may also be involved due to its proximity to the cervical plexus. Neuroimaging with CT or MRI can establish the diagnosis. Spinal fluid should be examined if leptomeningeal disease is suspected.

4.2. Imaging Radiation-induced plexopathy may also be the cause of similar symptoms in those patients with a previous history of radiation (39). MRI may assist in differentiating between radiation plexopathy and tumor plexopathy. Tumor plexopathy appears hypointense on T1-weighted images and hyperintense on T2 images while radiation plexopathy appears hypointense on T1- and T2-weighted images. However, fine needle aspiration may be required to differentiate between these two causes with confidence, especially in those patients who have symptoms without abnormal imaging. In fact, exploratory surgery may be needed in cases where biopsy is nondiagnostic.

4.3. Treatment Radiotherapy or chemotherapy should be considered in the context of the histopathological diagnosis and the patient’s functional status.

208

Part IV / Direct Complications of Cancer

5. METASTASES TO THE BRACHIAL PLEXUS The brachial plexus is an arrangement of nerve fibers running from the spine (C5–T1 roots), through the neck, the axilla and into the arm. The brachial plexus is responsible for cutaneous and muscular innervation of most of the upper limb. The ventral rami of these roots form three trunks (superior, medial, and inferior) which in turn form dorsal and ventral divisions. The dorsal divisions form the posterior cord which gives rise to the thoracodorsal, subscapular, axillary, and radial nerves. The ventral divisions form the lateral and medial cords which then form the musculocutaneous, median, and ulnar nerves.

5.1. Presentation Brachial plexus dysfunction is a recognized complication of cancer. Approximately 1 out of 100 patients with cancer will have metastatic involvement of their brachial plexi (40). Brachial plexopathy from metastases most commonly occurs in carcinoma of the breast and lung and is often characterized by severe, unrelenting pain. Later on, weakness and sensory disturbances occur in the anatomical distribution of the part of the plexus involved. The onset of symptoms may be subtle and pain may precede other symptoms by months. A quarter of patients will develop an ipsilateral Horner’s syndrome (disruption of the sympathetic innervation of the eye, resulting in ptosis, pupillary miosis, and facial anhidrosis) due to involvement of stellate ganglion near the T1 nerve root. There also may be associated upper extremity lymphedema and a palpable mass within the axilla. Metastatic cancers most often cause symptoms in the distribution of the lower trunk due to the proximity of this trunk to the lateral group of axillary lymph nodes. The upper trunk and its divisions are less involved due to a relatively greater distance from lymph nodes. In previously treated patients, the main differential diagnostic consideration is radiation-induced plexopathy (41). Characteristic temporal, clinical, radiographic, and electrophysiologic features distinguish these causes, but thorough and recurrent evaluations are often necessary to establish a diagnosis. Radiation plexopathy, which occurs in approximately 1–9% of patients, presents with symptoms that occur after a latent period of several months to years after the completion of therapy. Radiation damage is only probable when the dose has exceeded a threshold the magnitude of which depends on the length of the nerve irradiated. For the brachial plexus the focal dose needed to induce damage is thought to be 4500 Gy in 4 weeks (42). Radiation plexopathy has a tendency to be painless with upper trunk dysfunction (the lower trunk is somewhat protected from radiation by the clavicle) and lymphedema. As stated above, the pain with metastatic lesions is usually more severe and more difficult to treat and results in lower trunk dysfunction with a possible concomitant Horner’s syndrome.

5.2. Imaging The distinction between metastatic and radiation plexopathy based on clinical examination alone can be difficult due to overlapping symptoms and therefore requires further investigation. MRI is the test of choice in evaluating brachial plexopathy. A recent study of patients with brachial plexopathy (most of whom had been previously radiated) revealed that MRI for detection of tumor yielded a sensitivity of 96%, specificity of 95%, positive predictive value of 96%, and negative predictive value of 95% (43). MRI findings of tumor usually show a discrete mass with hypointense T1 findings and hyperintense T2 images with or without enhancement. MRI findings of radiation plexopathy include thickening and diffuse enhancement of the brachial plexus without a focal mass and soft-tissue changes with low signal intensity on both T1- and T2-weighted images (44). Recent data also suggest that 18 FDG-PET scans are a useful tool in evaluation of patients with suspected metastatic plexopathy, principally if other imaging studies are normal (45). FDG-PET usually shows increased uptake in the involved region of metastatic plexopathy while there is little uptake in radiation plexopathy. Electrophysiological testing may also be helpful in differentiating metastatic plexopathy from radiation induced plexopathy. Myokymia (continuous, involuntary small muscle discharges that affect a bundle of muscles, thought due to innervation of the muscles) is a common finding on the EMG of radiation-induced plexopathy and is generally not present in metastatic plexopathy (46).

Chapter 13 / Peripheral Nervous System Metastases

209

5.3. Treatment Treatment of brachial plexopathy secondary to metastatic cancer is predominantly symptom-based. The first endeavor of treatment should be to eliminate the tumor with radiation or chemotherapy (41). Usually, significant pain remains after such treatment and requires any number of pain management strategies including narcotic analgesics, paravertebral nerve block, and dorsal rhizotomy. High-quality physical therapy, tricyclics, antiarrhythmics, anticonvulsants, nonsteroidal anti-inflammatory drugs and steroids may also be helpful. Radiation plexopathy is treated with supportive measures. Transdermal electrical nerve stimulation, dorsal column stimulators, neurolysis, and neurolysis with omentoplasty have been tried successfully (47). Preventive measures such as using the lowest effective radiation dose should also be utilized.

6. METASTASES TO THE LUMBOSACRAL PLEXUS The lumbar plexus is derived from the ventral rami of T1–L4 while the sacral plexus is derived from the ventral rami of S1–S4. Each plexus divides to form anterior and posterior branches; in general, anterior branches supply flexor muscles of the lower limb and posterior branches supply extensor and abductor muscles.

6.1. Presentation Metastatic lumbosacral plexopathy is often due to direct extension of abdominal and pelvic region tumors. The most frequent types include cervical, colorectal, retroperitoneal sarcomas, and metastasis from breast and prostate (48–52). More often than not, patients present with pain followed by weakness or sensory deficits related to the anatomical location involved. In a retrospective study of 85 patients with lumbosacral plexopathy due to cancer, 70% of patients experienced the insidious onset of pelvic or radicular leg pain, followed weeks to months later by sensory symptoms and weakness (48). Ipsilateral leg edema may occur in roughly half of patients as well. Patients can also develop a “warm and dry foot” due to involvement of the ipsilateral sympathetic nerves located close to the vertebral bodies (53). If the upper lumbar plexus is involved, patients experience anterior thigh muscle weakness and diminished patellar reflexes. Lower plexus dysfunction presents with ipsalateral foot drop with diminished ankle reflex. Sacral reflex dysfunction results in anal sphincter weakness and loss of anal wink reflexes. As with brachial plexopathy, the main differential in previously treated patients is radiation-induced plexopathy. Other considerations include hip metastasis, vertebral compression fractures, and avascular necrosis of the hip. Characteristic temporal, clinical, radiographic, electrophysiological, and laboratory features distinguish between these etiologies, but recurrent evaluations may be necessary to establish a diagnosis (54). Post-irradiation lumbosacral plexopathy presents with a predominantly bilateral motor disorder affecting both legs after a variable latency. This bilateral lower extremity weakness can be accompanied by pain but it is of no more than mild to moderate intensity and may even be painless. Nearly all patients also develop sensory features on prolonged follow-up. Tumor-induced plexopathy typically presents with unilateral weakness involving more than one root, intense and unrelenting pain and patchy sensory findings (48,55,56).

6.2. Imaging MRI is the test of choice to diagnose metastatic lumbosacral plexopathy and is thought to be more sensitive than CT (57). The findings for either tumor or radiation-induced plexopathy are similar to that discussed in the section regarding brachial plexopathy. Also, as discussed above, when MRI or CT are unrevealing FDG-PET may be useful to differentiate between metastatic and radiation induced plexopathy though the normal elimination of radioisotope by the kidney and bladder may limit test interpretation (24). EMG testing will show myokymia in radiation-induced plexopathy while this finding is absent in metastatic plexopathy.

6.3. Treatment Treatment of lumbosacral plexopathy in patients with metastatic cancer is similar to the treatment of metastatic brachial plexopathy, with the initial step attempts to eliminate the tumor with radiation or chemotherapy. Pain management usually requires a mutli-modality approach including physical therapy, abortive and chronic pain

210

Part IV / Direct Complications of Cancer

medicines, and possible referral to a pain management service (41,48,54). There is a report indicating that hyperbaric oxygen therapy may be helpful in those patients with signs and symptoms of sacral plexopathy thought due to osteoradionecrosis of the sacrum from previous radiation (58).

7. CONCLUSIONS Though relatively uncommon, cancer may affect the PNS by infiltration of nerves or muscles, or by local compression. A patient’s clinical presentation depends on anatomic location of the metastases, but in general metastases to the PNS present with considerable pain in possible conjunction with motor or sensory deficits. After a thorough history and examination a physician should consider appropriate differential diagnoses and promptly proceed with pertinent imaging and diagnostic tests. Diagnosis and relevant treatment should be initiated without delay with the aim of alleviating pain, eliminating tumor when possible, and preventing if not reversing neurologic injury.

REFERENCES 1. Greenberg HS, Deck MD, Vikram B et al. Metastasis to the base of the skull: clinical findings in 43 patients. Neurology 1981;31(5): 530–537. 2. Lossos A, Siegal T. Numb chin syndrome in cancer patients: etiology, response to treatment, and prognostic significance. Neurology 1992;42(6):1181–1184. 3. Maillefert JF, Gazet-Maillefert MP, Tavernier C et al. Numb chin syndrome. Joint Bone Spine 2000;67(2):86–93. 4. Sweet JM. The numb chin syndrome: a critical sign for primary care physicians. Arch Intern Med 2004;164(12):1347–1348. 5. Ampil FL, Burton GV, Hardjasudarma M et al. Breast cancer metastasis to the gasserian ganglion. South Med J 2006;99(1):93–94. 6. Keane JR. Multiple cranial nerve palsies: analysis of 979 cases. Arch Neurol 2005;62(11):1714–1717. 7. Recht L, Mrugala M. Neurologic complications of hematologic neoplasms. Neurol Clin 2003;21(1):87–105. 8. Grisold W, Piza–Katzer H, Jahn R et al. Intraneural nerve metastasis with multiple mononeuropathies. J Peripher Nerv Syst 2000;5(3):163–167. 9. Cantone G, Rath SA, Richter HP. Intraneural metastasis in a peripheral nerve. Acta Neurochir (Wien) 2000;142(6):719–720. 10. Matsumine A, Kusuzaki K, Hirata H et al. Intraneural metastasis of a synovial sarcoma to a peripheral nerve. J Bone Joint Surg Br 2005;87(11):1553–1555. 11. Baehring J, Cooper D. Neurolymphomatosis. J Neurooncol 2004;68(3):243–244. 12. Bi HY, Na J, Zang GM et al. Neurolymphomatosis: a case report. Beijing Da Xue Xue Bao 2005;37(3):331. 13. Kim JH, Jang JH, Koh SB. A case of neurolymphomatosis involving cranial nerves: MRI and fusion PET-CT findings. J Neuro-oncol 2006;80(2):209–210. 14. Baehring JM, Damek D, Martin EC et al. Neurolymphomatosis. Neuro-oncol 2003;5(2):104–115. 15. Rosso SM, de Bruin HG, Wu KL et al. Diagnosis of neurolymphomatosis with FDG PET. Neurology 2006;67(4):722–723. 16. Juhn YJ, Inoue S. Facial nerve palsy as an early manifestation of relapse in T-cell acute lymphoblastic leukemia. Ear Nose Throat J 1996;75(3):157–160. 17. Kishimoto N, Shimada H, Adachi M et al. Infiltrative peripheral neuropathy of acute monoblastic leukemia during hematologic remission. Rinsho Ketsueki 1994;35(9):876–880. 18. Billstrom R, Lundquist A. Acute myelomonocytic leukaemia with infiltrative peripheral neuropathy. J Intern Med 1992;232(2): 193–194. 19. Nishi Y, Yufu Y, Shinomiya S et al. Polyneuropathy in acute megakaryoblastic leukemia. Cancer 1991;68(9):2033–2036. 20. Fuchs B, Spinner RJ, Rock MG. Malignant peripheral nerve sheath tumors: an update. J Surg Orthop Adv 2005;14(4):168–174. 21. Anghileri M, Miceli R, Fiore M et al. Malignant peripheral nerve sheath tumors: prognostic factors and survival in a series of patients treated at a single institution. Cancer 2006;107(5):1065–1074. 22. Hirose T, Scheithauer BW, Sano T. Perineurial malignant peripheral nerve sheath tumor (MPNST): a clinicopathologic, immunohistochemical, and ultrastructural study of seven cases. Am J Surg Pathol 1998;22(11):1368–1378. 23. Wanebo JE, Malik JM, VandenBerg SR et al. Malignant peripheral nerve sheath tumors: a clinicopathologic study of 28 cases. Cancer 1993;71(4):1247–1253. 24. Ramchandren S, Dalmau J. Metastases to the peripheral nervous system. J Neurooncol 2005;75(1):101–110. 25. Acinas Garcia O, Fernandez FA, Satue EG et al. Metastasis of malignant neoplasms to skeletal muscle. Rev Esp Oncol 1984;31(1): 57–67. 26. Menard O, Parache RM. Muscle metastases of cancers. Ann Med Interne (Paris) 1991;142(6):423–428. 27. Kondo S, Onodera H, Kan S et al. Intramuscular metastasis from gastric cancer. Gastric Cancer 2002;5(2):107–111. 28. Foldi M, Randelzhofer B, Gitsch G. Symptomatic skeletal muscle metastasis and elephantiastic lymphedema in a patient with recurrent ovarian carcinoma. Gynecol Oncol 2003;90(2):471–3. 29. Bese NS, Ozguroglu M, Dervisoglu S et al. Skeletal muscle: an unusual site of distant metastasis in gastric carcinoma. Radiat Med 2006;24(2):150–153. 30. Seely S. Possible reasons for the high resistance of muscle to cancer. Med Hypotheses 1980;6(2):133–137.

Chapter 13 / Peripheral Nervous System Metastases

211

31. Herring CL, Jr., Harrelson JM, Scully SP. Metastatic carcinoma to skeletal muscle: a report of 15 patients. Clin Orthop Relat Res 1998(355):272–281. 32. Pretorius ES, Fishman EK. Helical CT of skeletal muscle metastases from primary carcinomas. AJR 2000;174(2):401–404. 33. Heyer CM, Rduch GJ, Zgoura P et al. Metastasis to skeletal muscle from esophageal adenocarcinoma. Scand J Gastroenterol 2005;40(8):1000–1004. 34. Yoshioka H, Itai Y, Niitsu M et al. Intramuscular metastasis from malignant melanoma: MR findings. Skeletal Radiol 1999;28(12): 714–716. 35. Berquist TH. Magnetic resonance imaging of musculoskeletal neoplasms. Clin Orthop Relat Res 1989(244):101–118. 36. Schlemmer HP, Schafer J, Pfannenberg C et al. Fast whole-body assessment of metastatic disease using a novel magnetic resonance imaging system: initial experiences. Invest Radiol 2005;40(2):64–71. 37. Hundt W, Braunschweig R, Reiser M. Diffuse metastatic infiltration of a carcinoma into skeletal muscle. Eur Radiol 1999;9(2):208–210. 38. Jaeckle KA. Nerve plexus metastases. Neurol Clin 1991;9(4):857–866. 39. Cross NE, Glantz MJ. Neurologic complications of radiation therapy. Neurol Clin 2003;21(1):249–277. 40. Kori SH, Foley KM, Posner JB. Brachial plexus lesions in patients with cancer: 100 cases. Neurology 1981;31(1):45–50. 41. Jaeckle KA. Neurologic manifestations of neoplastic and radiation-induced plexopathies. Semin Neurol 2004;24(4):385–393. 42. Hughes RA, Britton T, Richards M. Effects of lymphoma on the peripheral nervous system. J R Soc Med 1994;87(9):526–530. 43. Qayyum A, MacVicar AD, Padhani AR et al. Symptomatic brachial plexopathy following treatment for breast cancer: utility of MR imaging with surface-coil techniques. Radiology 2000;214(3):837–842. 44. Wittenberg KH, Adkins MC. MR imaging of nontraumatic brachial plexopathies: frequency and spectrum of findings. Radiographics 2000;20(4):1023–1032. 45. Ahmad A, Barrington S, Maisey M et al. Use of positron emission tomography in evaluation of brachial plexopathy in breast cancer patients. Br J Cancer 1999;79(3–4):478–482. 46. Krarup C, Crone C. Neurophysiological studies in malignant disease with particular reference to involvement of peripheral nerves. J Neurol 2002;249(6):651–661. 47. Killer HE, Hess K. Natural history of radiation-induced brachial plexopathy compared with surgically treated patients. J Neurol 1990;237(4):247–250. 48. Jaeckle KA, Young DF, Foley KM. The natural history of lumbosacral plexopathy in cancer. Neurology 1985;35(1):8–15. 49. Agar M, Broadbent A, Chye R. The management of malignant psoas syndrome: case reports and literature review. J Pain Symptom Manage 2004;28(3):282–293. 50. Ladha SS, Spinner RJ, Suarez GA et al. Neoplastic lumbosacral radiculoplexopathy in prostate cancer by direct perineural spread: an unusual entity. Muscle Nerve 2006;34(5):659–665. 51. Felice KJ, Donaldson JO. Lumbosacral plexopathy due to benign uterine leiomyoma. Neurology 1995;45(10):1943–1944. 52. Saphner T, Gallion HH, Van Nagell JR et al. Neurologic complications of cervical cancer: a review of 2261 cases. Cancer 1989;64(5):1147–1151. 53. Lopez Martin JA, Benito Urbina S, Gila Useros L et al. Hot foot, as the first manifestation of lumbosacral plexopathy. Med Clin (Barc) 1990;95(7):276. 54. Pettigrew LC, Glass JP, Maor M et al. Diagnosis and treatment of lumbosacral plexopathies in patients with cancer. Arch Neurol 1984;41(12):1282–1285. 55. Bowen J, Gregory R, Squier M et al. The post-irradiation lower motor neuron syndrome neuronopathy or radiculopathy? Brain 1996;119 (Pt 5):1429–1439. 56. Thomas JE, Cascino TL, Earle JD. Differential diagnosis between radiation and tumor plexopathy of the pelvis. Neurology 1985;35(1):1–7. 57. Taylor BV, Kimmel DW, Krecke KN et al. Magnetic resonance imaging in cancer-related lumbosacral plexopathy. Mayo Clin Proc 1997;72(9):823–829. 58. Videtic GM, Venkatesan VM. Hyperbaric oxygen corrects sacral plexopathy due to osteoradionecrosis appearing 15 years after pelvic irradiation. Clin Oncol (R Coll Radiol) 1999;11(3):198–1999.

V

Indirect Complications of Cancer

14

Cerebrovascular Complications of Cancer Lisa R. Rogers, DO, Megan C. Leary, and Jeffrey L. Saver, MD

MD

CONTENTS Introduction Stroke Due to Central Nervous System Tumor Stroke Due to Remote Effects of Tumor: Hyper- and Hypocoagulopathies Traditional Stroke Mechanisms Appearing in the Setting of Tumor Sequelae of Cancer Diagnostic Tests and Treatment Conclusion References

Summary A wide variety of cerebrovascular disorders can complicate the clinical course of cancer patients and, in rare instances, be the presenting sign of cancer. Central nervous system (CNS) hemorrhagic or ischemic events are typically symptomatic. Hemorrhage can occur into the parenchymal, subdural, epidural, or subarachnoid compartments. Cerebral infarction may cause focal signs presenting as a transient ischemic attack or as a completed infarction, or encephalopathy secondary to multiple small infarctions. The most common mechanisms of cerebrovascular disease in the cancer patient are the presence of CNS tumor or coagulopathy. The coagulopathy is a direct effect of the neoplasm in many cases; in other cases it develops from cancer treatment, most commonly chemotherapy. Radiation therapy to extracranial or cranial sites can also result in vessel stenosis or thrombosis. It is important to identify cerebrovascular disease as the cause of neurologic symptoms in the cancer patient. Laboratory studies of coagulation function and CNS imaging techniques are useful in identifying the precise etiology of the cerebrovascular disorder. Prompt diagnosis leads to appropriate therapy, which in many patients can improve the clinical neurologic condition as well as prevent additional vascular episodes. Key Words: cancer, stroke, vascular disease, embolic events, hypercoaguable states, cerebrovascular disease

1. INTRODUCTION Cancer and stroke are the second and third leading causes of mortality in the United States, respectively, so it is not surprising to encounter patients with these conditions concomitantly. A wide variety of cerebrovascular disorders can occur within the oncologic population, complicating the cancer patient’s overall clinical course, treatment, and long-term outcome. Unfortunately, the incidence of stroke within this population is substantial. Graus and colleagues’ 1985 autopsy series of 3,426 systemic cancer patients revealed that stroke was second only to metastases as the most common central nervous system lesion, occurring in 14.6% of patients (1). Hemorrhages and ischemic lesions were present in equal numbers; however, hemorrhages were more frequently symptomatic. Of the patients with pathologically defined stroke, greater than half had significant clinical symptomatology due to their cerebrovascular injury. In From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

215

216

Part V / Indirect Complications of Cancer

addition, Posner and Chernik’s earlier 1978 postmortem study suggests the incidence of stroke in the oncologic population may actually be as high as 30% (2). Certainly stroke and cancer co-exist far more than would be expected by chance, and frequently enough to render knowledge of the cerebrovascular complications of malignancy essential to the care of the oncologic patient. A detailed investigation and precise diagnosis of cerebrovascular disorders in cancer patients is critical for several reasons. Early recognition of acute stroke may allow the cancer patient access to surgical or medical intervention to improve overall patient outcome. In addition, secondary stroke prevention therapies are guided by the etiology of the initial cerebrovascular event. Lastly, the diagnostic evaluation in young or cryptogenic stroke patients may lead to the first recognition of an underlying malignancy. In determining the etiology of stroke in the cancer patient, various factors must be considered. Traditional cerebrovascular risk factors seen in the general population such as age, hypertension, coronary artery disease, hypercholesterolemia, tobacco use, diabetes, and family history of stroke should be assessed, but cancer patients frequently have stroke as a cancer-related event, in which the malignancy directly or indirectly contributes to the cerebrovascular insult. Thus, additional consideration must be given to the causes of stroke that are unique to the cancer patient. This chapter will explore the etiologies of stroke within the oncologic population and discuss their diagnosis and management.

2. STROKE DUE TO CENTRAL NERVOUS SYSTEM TUMOR 2.1. Intratumoral Parenchymal Hemorrhage Intracranial hemorrhage into brain tumors, both metastatic and primary, is a relatively frequent occurrence, reported to account for 1.7–9.6% of all intracranial hemorrhages (3–6). Metastatic tumors are more often associated with hemorrhage than are primary tumors (7). Intracranial hemorrhage is the initial manifestation of primary or metastatic tumor in up to 3% of oncologic patients (3,6,8–11). Brain metastases are common in the oncologic population, affecting 20–40% of all systemic cancer patients. Among metastatic malignancies, two of the most common brain tumors associated with intratumoral hemorrhage are melanoma and lung cancer (6,7). Other common underlying tumors are renal, thyroid, and germ cell tumors (12). Bleeding into brain metastases has also been reported in patients with almost any other histology. The most common clinical symptoms in patients with intratumoral brain hemorrhage are headache, nausea, vomiting, obtundation, and seizures (6,7,9). Focal neurological deficits are also frequently present. Bleeding may be spontaneous or associated with predisposing factors such as head trauma, hypertension, coagulopathy, shunting procedures, and anticoagulation (3,7,11,13). Various pathophysiologic processes contribute to the pathogenesis of intratumoral hemorrhage, including overexpression of vascular endothelial growth factor and matrix metalloproteinases, endothelial proliferation with vascular obliteration, vessel compression and distortion secondary to tumor growth, vessel necrosis, invasion of vessel walls by tumor, and an increase in venous pressure due to elevated intracranial pressure (5,14–19). Histologic factors associated with intratumoral hemorrhage include rapid tumor growth, tumor necrosis, vessel thrombosis, the presence of multiple thin-walled vessels, and tumor invasion of adjacent cerebral vessels. The most common primary central nervous system malignancies associated with intratumoral hemorrhage are glial tumors and germ cell tumors. Macroscopic, clinically significant, hemorrhage is more common in lowgrade astrocytoma than in glioblastoma, but bleeding with both neoplasms has been reported (Fig. 1). Among astrocytoma patients, the highest hemorrhage rate has been noted in cases of mixed gliomas (oligoastrocytomas), with gross bleeding reported in up to one-third of cases. Other primary central nervous system malignancies have been infrequently associated with intratumoral hemorrhage. Hemorrhage into other tumor types such as meningioma is relatively uncommon but has been reported (20). Lazaro and colleagues noted that endotheliomatous meningiomas were most likely to bleed, accounting for 50% of meningioma-associated hemorrhages (21). Cases of meningioma-associated bleeding have been reported with transitional and fibroblastic meningiomas as well. Hemorrhages into other tumor types such as medulloblastoma, choroid plexus papilloma, ependymoma, pineal region tumors, and primary intracranial sarcoma have also been reported.

Chapter 14 / Cerebrovascular Complications of Cancer

217

Fig. 1. Glioblastoma multiforme with intratumoral hemorrhage. (A) MRI gradient refocused echo sequence shows large areas of hypointensity reflecting intratumoral hemorrhage (white arrows). (B) T2-weighted MR sequence demonstrates mixed signal intensities (white arrow).

Hemorrhage can develop in metastatic tumors in any location in the cerebral hemispheres, brain stem, and cerebellum. Single or multiple hemorrhages can be present. Hemorrhage into tumor must be differentiated from other etiologies of intracerebral hemorrhage, including cerebral aneurysm, vascular malformations, amyloid angiopathy, and hypertension. Unusual locations for hemorrhage should suggest an underlying etiology of tumor or vascular abnormality. CT findings suggestive of tumoral hemorrhage include early edema and an indentation appearing on the hematoma surface on noncontrast studies, which demonstrates after contrast injection enhancement. In one series, computed tomography (CT) scanning was able to differentiate the hemorrhage etiology in 61% of oncologic patients (6). MR findings suggestive of tumoral hemorrhage are signal heterogeneity, evidence of nonhemorrhagic tumor mass, delayed hemorrhage evolution, hemosiderin deposition, and early edema (22). In one series, MR imaging revealed a tumoral etiology for intracerebral hemorrhage unrecognized by CT in 4% of patients studied with both modalities (23). Patient outcome after intratumoral hemorrhage is notably related to the specific histological malignancy of the tumor. Additionally, there appears to be a higher risk of recurrent hemorrhage if the tumor is incompletely excised or if metastases recur. Rebleeding is associated with graver prognosis.

2.2. Neoplastic Subdural Hemorrhage In cancer patients with cerebrovascular disease, subdural hematomas and subdural fluid collections are common, comprising 12.6% of all strokes and 25.8% of hemorrhagic lesions within this population (1). Subdural hematomas have been reported in association with a wide variety of malignancies. In patients with solid tumors, subdural hematomas tend to occur with dural tumor metastases (24). Neoplastic infiltration of the dura results either from hematogenous spread of tumor via the dural vessels or from direct extension of skull metastases. Histologically, the tumors most frequently associated with subdural hematoma include gastric carcinoma, prostate carcinoma, breast carcinoma, leukemia, and lymphoma (25,26). Subdural hematomas have also been reported with primary brain tumors, including meningioma and glioblastoma (27). Clinical manifestations of subdural hematoma in the oncologic population differ little from the general population. Patients may have an acute, subacute, chronic, or asymptomatic initial course. Graus and colleagues found that one-quarter of their 53 autopsied subdural hematoma patients with cancer were symptomatic (1). The most common clinical symptoms are altered mental status, headache, and lethargy. Focal neurological deficits and seizures may also be present (24,28).

218

Part V / Indirect Complications of Cancer

In one retrospective review of 70 patients with systemic cancer and subdural hematoma, dural metastatic lesions were the most common cause of subdural hematoma in patients with no predisposing risk factors (28). Of their patients with predisposing risk factors, trauma and anticoagulants were the major predisposing risk factors. Proposed mechanisms for the occurrence of subdural hematoma secondary to dural metastases include hemorrhage directly into the dural tumor, hemorrhage secondary to dilatation and rupture of inner dural layer capillaries/venules/veins due to outer vessel layer obstruction by tumor, and in rare cases, dural tumor production of a hemorrhagic effusion. Acute or chronic subdural hematomas and skull metastases are generally easily visualized with both CT and MRI (29). Isodense subdural hematomas, especially when present bilaterally, may be missed. Contrast studies are helpful in revealing dural enhancement suggestive of dural metastases. Histologic examination of the dura with biopsy or cytologic studies of the subdural fluid is necessary to confirm the tumoral origin of the subdural hematoma. Treatment of dural metastasis-associated hemorrhage is palliative and includes drainage of subdural fluid and radiation therapy.

2.3. Neoplastic Infiltration of Cerebrae Vessels 2.3.1. Venous Infiltration Thrombosis of cerebral veins or dural sinuses is a rare event in any patient population, including the oncologic population. In a recent series, cerebral venous thrombosis from any cause accounted for 0.3% of neurologic consultations at a large cancer center (30). When obstruction of cerebral venous drainage in cancer patients occurs, one culprit is invasion or compression of cortical veins or dural sinuses by tumor in the adjacent dura or skull. Rarely, bulky leptomeningeal metastases can cause cerebral sinus occlusion (31). The most common sinus affected by metastases is the superior sagittal sinus. A variety of malignancies have been reported in association with sinus thrombosis, including leukemia, lymphoma, neuroblastoma, breast carcinoma, lung carcinoma, cervical carcinoma, gallbladder carcinoma, Ewing’s sarcoma, and myeloma (30,32,33). The typical clinical presentation of cerebral venous thrombosis is headache, vomiting, papilledema, and seizures. Focal neurological signs and encephalopathy may also be seen, particularly in association with venous infarction due to sinus or cortical vein occlusion (34). Occasionally patients present with an isolated intracranial hypertension syndrome, with headache and papilledema as the only manifestations (“tumoral pseudotumor cerebri”) (33,35). Radiologic studies helpful in diagnosis include CT scanning with contrast dye to visualize the lack of contrast within the sagittal sinus, a finding known as the “empty delta sign” (36). Conventional MRI and MR venography are more reliable than CT, and are an excellent method to diagnose and follow venous thrombosis. (Fig. 2) The sensitivity of MRI alone in detecting cerebral venous thrombosis is 90%; however, small case series suggest that sensitivity increases close to 100% with the concomitant use of MRA and MRV (30,37). CT angiography/venography correlates well with MRV. Two randomized clinical trials have studied the benefits and risks of anticoagulation in patients with sinus thrombosis without cancer. A small trial of intravenous unfractionated heparin found a benefit; a larger trial of low molecular weight heparin only a nonsignificant trend to benefit (38,39). Recent series in the literature have reported success in treating venous cerebral thrombosis with selective dural catheterization and local thrombolysis (40,41). Endovascular mechanical rheolytic thrombectomy followed by low-dose local thrombolysis (for persistent cortical vein thrombosis) is an additional emerging treatment modality that has been reported to accelerate recanalization of occluded dural sinuses (42). The safety and efficacy of these treatments in the oncologic population has not been established. Radiation therapy or surgical intervention should generally be pursued in patients with cerebral sinus occlusion due to metastases. 2.3.2. Arterial Infiltration Neoplastic infiltration of arterial vessels has been reported to cause both hemorrhagic and ischemic strokes. Neoplastic infiltration of arteries due to tumor embolization can result in aneurysm or pseudoaneurysm formation and subsequent aneurysm rupture produces intracerebral and/or subarachnoid hemorrhage (Fig. 3) (43). Cardiac myxomas account for two-thirds of neoplastic aneurysms and choriocarcinoma for one-quarter, with rarer cases

Chapter 14 / Cerebrovascular Complications of Cancer

219

Fig. 2. Cerebral venous thrombosis in a patient with ALL. (A) Gradient refocused echo imaging and (B) FLAIR imaging show multiple small areas of cortical and subcortical hemorrhage (black arrows). (C) Magnetic resonance venography reveals occlusion at the anterior and posterior portions of the sagittal sinus and near occlusion of the right transverse sinus at its origin (white arrows).

due to bronchogenic carcinoma and undifferentiated carcinoma (1,43,44). Neoplastic aneurysms are typically small in size and are often located in distal cerebral arterial branches, in contrast to saccular aneurysms which typically arise in proximal cerebral arteries. A second, less common, mechanism of aneurysm formatia is secondary invasion of nearby vessels by parenchymal brain metastases. Ischemic stroke has also been associated with tumor infiltration of arteries. Autopsy studies of cancer patients with leptomeningeal metastases have documented ischemic strokes due to neoplastic arterial wall infiltration (45, 46). Patients experiencing stroke secondary to leptomeningeal metastases present with abrupt, focal neurological deficits alone or in addition to the typical features of leptomeningeal metastases. There are two proposed mechanisms for cerebral infarction due to leptomeningeal metastases: infarction due to infiltration of arterial walls by tumor cells and infarction due to pial vessel vasospasm. Pathologically, multifocal mural invasion by tumor cells with occlusion of the arterial vessel is accompanied by a variable perivascular inflammatory reaction and intraluminal thrombosis. Angiography may reveal focal arteriolar narrowing at the base of the brain, over

Fig. 3. Intracerebral hemorrhages due to metastatic choriocarcinoma with pseudoaneurysm formation (reprinted with permission from reference 43). (A). Computed tomography on day 20 shows right frontal (large arrow) and right parietal hemorrhages with extension to the lateral ventricle (small arrows). (B). Magnetic resonance imaging on day 25 shows a new left posterior frontal hemorrhage (large black arrow) and left parieto-occipito-temporal subdural hematoma (large white arrow) and old right frontal and parietal hemorrhages (small arrows). (C) Computed tomography on day 29 shows increase in the left frontal hemorrhage and new occipital hemorrhages (large arrows) and old right temporal and parietal hemorrhages (small arrows).

220

Part V / Indirect Complications of Cancer

the cerebral convexities, or both (47). In addition, cerebral infarction adjacent to a cerebral glioblastoma is a rare event (48) and is presumably due to arterial invasion or compression by the tumor. 2.3.3. Hematologic Malignancies Proliferation of blood elements can lead to ischemic stroke both by hyperviscous obstruction of arterial vessels by neoplastic cells and by an induced procoagulant state. Thrombocytosis in polycythemia vera is linked with transient ischemic attacks and cerebral infarction (49). Ischemic stroke accounted for approximately 33% of 119 neurologic events reported in 443 patients with polycythemia vera followed by the Polycythemia Vera Study Group (49,50). Cytoreductive treatment of blood hyperviscosity by phlebotomy or chemotherapy substantially reduces thrombotic events and improves survival. The rate of stroke in one cohort of aggressively treated patients was reported at 0.3% annually (51). Patients with essential thrombocythemia have an increased risk of transient ischemic attack and ischemic stroke (52). Cytoreductive therapy and low-dose aspirin reduce the risk (53). Intravascular lymphomatosis is an uncommon variant of non-Hodgkin’s lymphoma characterized by a proliferation of lymphoma cells within small caliber blood vessels, with a predilection for the cerebral circulation (54,55). Patients most commonly present with fever accompanied by subacute progressive multifocal cerebral infarcts and/or a rapidly progressive encephalopathy. Anemia and elevated erythrocyte sedimentation rate may be present. Neuroimaging frequently shows multiple infarcts and parenchymal and meningeal enhancement (56). Noninvasive and catheter angiography may demonstrate a vasculitis-like appearance. Chemotherapy or radiotherapy can initially stabilize the clinical course, but average survival from symptom onset is only a few months, in part due to a delay in diagnosis and treatment.

2.4. Tumor Embolus Ischemic stroke directly secondary to tumor embolism is rare. In Graus and colleagues’ large review of patients with systemic cancer and stroke, only two patients had tumor emboli identified as the etiology of their infarcts (1). Most reported cases of ischemic stroke due to tumor emboli have resulted from intracardiac tumors. In general, half of these tumors are myxomas, with the other half being papillary fibroelastomas (papillomas), hamartomas, teratomas, and other less common malignant tumors (57–59). Neurological complications most commonly occur from myxomas in the left atrium, especially from polypoid lesions with soft, irregular shapes and a mobile surface (60). Of non-myxomatous benign primary cardiac tumors, papillomas are the most frequent cause of emboli to the cerebrovascular circulation (Fig. 4) (59). Among the primary malignant cardiac tumors, sarcomas are the most common to result in cerebral embolic infarction. Cerebral infarction may also occur from tumor emboli arising from tumors that have metastasized to the heart. The most common malignant neoplasms metastasizing to the heart include lung carcinoma and breast carcinoma, followed by melanoma, lymphoma, leukemia, and sarcoma (61,62). Among noncardiac tumors, cerebral infarction due to tumor embolization has been reported most frequently with choriocarcinoma and lung carcinoma, and has been reported in a variety of other tumors (1,57,63–65). The identification of a cerebral infarction following lung tumor resection should raise the possibility of a tumor embolus dislodged at surgery. Finally, although rare, aortic arch tumor–associated embolic stroke has also been reported (66). Ischemic stroke secondary to tumor emboli may affect the anterior and/or posterior circulations. Transient ischemic attacks may precede the cerebral infarction. Symptoms tend to occur suddenly, and patients frequently exhibit focal neurological deficits on examination. History and physical examination evidence of cardiac dysfunction are helpful in identifying a cardiac tumor as a potential source, with dyspnea, peripheral edema, and precordial murmurs the most common cardiac clinical manifestations (58,64). When noncardiac tumor emboli produce cerebral infarcts, the most likely mechanism is an embolus originating from a primary or metastatic pulmonary tumor: the embolus accesses the pulmonary venous system, passes through the left heart chambers, and travels to and occludes a vessel in the intracranial arterial circulation. Diagnostically, echocardiography is a reliable means of noninvasively diagnosing intracardiac tumors (67). Both transthoracic and transesophageal approaches are helpful in diagnosing the presence of an intracardiac tumor, and in suggesting tumor type by evaluating the area of tumor attachment, tumor size, and degree of mobility. Transesophageal echocardiogram is superior to transthoracic in evaluating atrial tumors. False negative

Chapter 14 / Cerebrovascular Complications of Cancer

221

Fig. 4. Papillary fibroelastoma of the aortic valve (reprinted with permission from reference 58). A 72-year-old man with headache, left hemiparesis, and left hemineglect. Diffusion weighted MRI (not shown) revealed an acute right middle cerebral artery stroke. (A) Papillary fibroelastoma of the aortic valve illustrated with transesophageal echocardiogram (white arrow), with (B) “sea anemone” appearance on gross examination (white arrow), and (C) multiple branching fronds noted on microscopy.

echocardiograms have been reported in cardiac neoplasm patients. MRI and CT may also contribute to the diagnosis of intracardiac tumor, and are useful in delineating the extent of tumor involvement in the great vessels and mediastinum (66,68,69). Tissue diagnosis involves obtaining a pathological specimen through endomyocardial biopsy or through surgical resection of the tumor (60,70). The rapidity of tumor progression, cardiac complication rate, and recurrent stroke rate are unpredictable in cardiac tumors. Thus, the treatment of choice in these patients is prompt surgical resection. In addition, some authors suggest that cerebral aneurysms may improve after resection of the primary cardiac tumor (57,71).

2.5. Pituitary Apoplexy Pituitary apoplexy is characterized by pituitary hemorrhage, infarction, or both in pituitary adenomas and occasionally into nontumorous pituitary glands (72). The classical clinical presentation is sudden retro-orbital headache, vomiting, visual impairment, oculomotor paresis, and meningismus. Partial or complete hypopituitarism follows. In different series, the incidence of pituitary adenomas presenting with apoplexy ranges from 1.5–27%. Proposed mechanisms include tumor growth, which outstrips the blood supply, leading to ischemic necrosis, or compresses vessels causing ischemia, rupture of tumor-associated intracranial aneurysms, and an intrinsic bleeding tendency of pituitary adenomas. MRI is superior to CT for radiologic diagnosis. Transphenoidal decompression of apoplectic tumors is commonly pursued, although many cases treated with conservative medical therapy recover well (73).

3. STROKE DUE TO REMOTE EFFECTS OF TUMOR: HYPER- AND HYPOCOAGULOPATHIES 3.1. Hypercoagulability and Thrombosis Coagulation disorders resulting in thrombosis and/or hemorrhage commonly complicate the natural history of cancer and its treatment. Hypercoagulability in cancer patients was first described by Trousseau in 1865, who

222

Part V / Indirect Complications of Cancer

reported accelerated bleeding times and thrombophlebitis in greater than 60% of the oncologic patients he observed (74). Modern studies confirm that abnormalities of the coagulation system are very common in the cancer patient. Hemostatic abnormalities on blood tests are found in more than 90% of oncologic patients, and it is estimated that venous thromboembolism and disseminated intravascular coagulation complicate the course in 15%. Increases of coagulation factors V, VIII, IX, and XI are often documented in malignancy. Markers of coagulation activation are frequently elevated, including prothrombin fragment 1.2, thrombin antithrombin complex, fibrin degradation products, and D-dimers. Also consistent with a consumptive coagulopathy is the frequent finding of increased fibrinogen and platelet turnover. The pathogenesis of cancer-related thrombophilia is complex and multifactorial. The interaction of neoplastic cells with the hemostatic system includes direct activation of the coagulation and fibrinolytic systems, release of inflammatory cytokines, perturbation of the vascular endothelium, activation of monocytes and platelets, and promotion of blood stasis (75–77). Tumor cells release multiple procoagulant substances, among which the best characterized are tissue factor and cancer procoagulant. Most tumor cells express on their surface all the major proteins that regulate the fibrinolytic pathway, including urokinase- and tissue-type plasminogen activator and plasminogen activator inhibitors 1 and 2. Among the proinflammatory cytokines released by tumor cells, several can impair the normal anticoagulant activity of the vascular endothelium, including tumor necrosis factor alpha (TNF-alpha) and interleukin (IL)-1 beta. Tumor cells also interact with vascular endothelium through direct adherence by membrane adhesion molecules, including integrins and selectins. Malignant cells attached to vessel walls promote localized clotting activation and thrombus formation. Tumor cells activate platelets by several distinct mechanisms, including platelet adhesion to tumor cell surface and malignant cell release of proaggregratory molecules, including adenosine diphosphatase and a cathepsin B–like cysteine proteinase. Tumor cells also activate the monocyte–macrophage system and induce the expression of tissue factor.

3.2. Venous Occlusions Cerebral venous thrombosis can occur not only from direct tumor invasion of the dural sinuses (reviewed above), but also from a hypercoagulable state induced by a remote neoplasm or by chemotherapy. A systemic coagulopathy is the most common cause of cerebral venous thrombosis in patients with hematologic malignancies (30). Imaging characteristics are reviewed above, but in the instance of coagulation-related venous thrombosis, no skull or dural tumor is identified. Treatment options, as reviewed above, include anticoagulation, local catheteradministered thrombolysis, and emerging endovascular mechanical recanalization techniques. Brain radiation is not indicated. In addition to the general hypercoagulable state of malignancy, several molecular abnormalities in the hemostatic system have been particularly associated with venous thrombosis in the oncologic population. Acquired protein S deficiency has been reported in association with leukemia, multiple myeloma, and pancreatic adenocarcinoma (78–80). Total protein S concentrations may be normal to high, but free protein S concentrations low, in patients with advanced cancer. An increase in the C4b binding protein by the neoplasm-induced free protein S deficiency contributes to the predilection for venous thromboses. As warfarin treatment may result in an additional decrease of free protein S, unfractionated heparin and low molecular weight heparin are preferred anticoagulants if recurrent thrombosis occurs on warfarin. Treatment of the underlying malignancy can also be beneficial. Venous thrombosis due to marked protein C deficiency has been reported in association with some cancers (81). The presence of a second hypercoagulable risk factor, such as heterozygosity for the factor V Leiden mutation, may increase risk for venous thrombosis (82). Activated protein C resistance is most commonly due to an Arg506-Gln point mutation in the factor V gene. The factor V Leiden mutation may interact with malignancyinduced or chemotherapy-induced hypercoagulability to cause thrombosis. In a prospective study of 65 children with leukemia and 65 controls, three children in each group had the factor V Leiden gene mutation. The three children in the leukemic group all had venous thromboembolic events, whereas the three children in the control group did not (83). Activated protein C resistance may occur as an acquired condition, independent of the factor V Leiden mutation, in oncologic patients. In a prospective study of 113 adult cancer patients with and without venous thromboembolism and 110 control patients with and without thromboembolic events, cancer patients with thromboembolism

Chapter 14 / Cerebrovascular Complications of Cancer

223

had a significantly greater prevalence of acquired activated protein C resistance than cancer patients without thromboembolism or both control groups. Control patients with thromboembolism had a significantly higher prevalence of factor V Leiden mutation than did the cancer patients (84).

3.3. Arterial Occlusions Cerebral arterial occlusions constitute a major source of morbidity in cancer patients. Three common culprits responsible for arterial occlusions in cancer-associated thrombophilia are nonbacterial thrombotic endocarditis (NBTE), mucin-positive adenocarcinoma-associated hypercoagulability, and hypercoagulability secondary to an antiphospholipid syndrome. In 1938, Sproul described widespread venous thrombosis, multiple arterial infarcts, and NBTE in patients with pancreatic cancers (85). Since that time, a variety of studies have linked NBTE with cancer and with embolic phenomena. In NBTE, an underlying coagulopathy results in a predisposition for sterile platelets and fibrin to deposit on cardiac valves. Vegetations are most commonly located on the aortic and mitral valves. An autopsy series found that NBTE is significantly more common in cancer patients than in patients without malignancy (1.25% vs. 0.2%) (86). NBTE is a common cause of ischemic stroke in the oncologic population (1). Although NBTE is associated with a variety of neoplasms, clinical and pathologic studies suggest that it most commonly occurs in adenocarcinoma patients (86–88). In a large autopsy study, the frequency of NBTE in the setting of adenocarcinoma was 2.7% versus 0.5% with other malignancies. Of the adenocarcinomas themselves, NBTE was most strongly associated with pancreatic cancer in comparison with other forms of adenocarcinoma (10.3% vs. 1.6% risk) (88). Several other studies have found that mucin-producing adenocarcinomas, of which many are pancreatic, are strongly associated with NBTE (87,89). MacDonald and Robbins’ autopsy series in 78 cancer patients with NBTE noted that although the sites of the primary tumors differed significantly, all the neoplasms were well-differentiated and mucinous (90). Histochemical studies on the valvular vegetations and thrombi in NBTE patients with mucin-producing adenocarcinomas have revealed that histochemically stainable mucin was an integral part of the vegetations and thrombi (89). Systemic thromboembolism may be a clue to an underlying NBTE. One-third of patients, however, have only neurologic symptomatology. NBTE can result in either focal, diffuse encephalopathic, or mixed neurological deficits when there is embolization from the affected cardiac valve to the brain or in situ cerebrovascular thrombosis. Spinal cord infarction has also been reported (91). Focal neurological signs in NBTE-associated embolism begin abruptly and may be preceded by transient ischemic attacks. A small proportion of patients have a concomitant acute DIC syndrome as evidenced by laboratory testing; however, the majority of patients with NBTE have only mildly abnormal coagulation parameters (87,92). Although NBTE typically occurs in disseminated cancers, NBTE can precede the diagnosis of malignancy and thus may be the initial sign of occult neoplasm. The diagnosis of NBTE is most often rendered in vivo by echocardiographic detection of valvular vegetations. Transesophageal echocardiography is more sensitive than transthoracic (93). Both neuroimaging and autopsy studies show that cerebral infarcts may be multiple and sometimes have a hemorrhagic component (87,94). Cerebral angiography typically discloses multiple abrupt vessel branch occlusions (87) and may also show changes suggestive of vasculitis (94). Appropriate treatment of NBTE includes treatment directed to the underlying etiology of the coagulation disorder, such as the neoplasm or sepsis. There are no prospective studies of anticoagulation in these patients; however, authors of individual case reports and retrospective cases series suggest that anticoagulation with heparin appears to reduce ischemic symptomatology in some patients (87). The potential benefits of anticoagulation should be weighed cautiously against the potential risks of hemorrhage in the cancer patient with NBTE-associated stroke.

3.4. Mucin-Positive Adenocarcinoma-Associated Hypercoaguability Mucin-producing adenocarcinomas have also been associated with arterial ischemic stroke both in association with, and independently of, NBTE. A 1989 study by Amico and colleagues examined oncologic patients with mucinous adenocarcinoma and systemic and cerebral infarcts (95). Widely disseminated metastases were present

224

Part V / Indirect Complications of Cancer

in all cases. Varying sizes of infarcts were found, including disseminated microinfarcts in all patients and large or small/moderate sized infarcts in most. Nervous system compromise was widespread, and included the cerebral hemispheres, cerebellum, brainstem, basal ganglia, spinal cord, and dorsal spinal roots. Petechial and small hemorrhages were also relatively common. In all cases in the Amico series, intravascular mucin was noted within central nervous system capillaries and small arteries on pathological examination. The mechanism of hypercoagulability in mucin-secreting adenocarcinoma is still not clearly understood. Mucin itself may be prothrombotic, the mucin-producing tumor cells may be prothrombotic, or both (95,96). At present, the diagnosis of mucin-positivity is only reliably made pathologically. Treatment of the underlying malignancy is the only known method to reduce further cerebrovascular events. Cautious anticoagulation has been suggested in the setting of ischemic stroke and mucin-positive neoplasm, although this therapy is unproven.

3.5. Antiphospholipid Antibodies Antiphospholipid antibodies increase the risk of both arterial and venous cerebral thrombosis (97). An antiphospholipid antibody may be present in up to 17% of patients with active underlying malignancies and in a smaller percentage of patients in clinical remission (98,99). There are multiple subspecies of antiphospholipid antibodies, including anticardiolipin antibodies, the lupus anticoagulant, and anti-beta 2-glycoprotein I antibodies. Non-Hodgkin’s lymphoma has been associated with both positive anticardiolipin and antibeta 2-glycoprotein I antibodies (100). Positive lupus anticoagulant and anticardiolipin antibodies have also been associated with a variety of other tumors, often in the setting of transient ischemic attack and stroke (101–103). Treatment of the underlying malignancy may result in resolution of the antiphospholipid syndrome. One report of a chronic myeloid leukemic patient with positive lupus anticoagulant antibodies described disappearance of the antibodies after allogenic bone marrow transplant (104). Hyperfibrinogenemia is common in cancer patients: a prospective laboratory study of 108 oncologic patients demonstrated hyperfibrinogenemia in 46% (105). Hyperfibrinogenemia in the setting of malignancy has been associated with ischemic stroke.

3.6. Combined Hypercoaguability/Bleeding Diathesis Normal physiologic hemostasis involves a balance between thrombus formation and thrombolysis. Disseminated intravascular coagulation (DIC) is characterized by widespread activation of coagulation, with resulting acceleration of fibrin clot formation and thrombotic occlusion of small and medium size vessels. Concurrently, consumption and depletion of platelets and coagulation proteins may induce severe bleeding. Patients may present with symptoms and signs of excessive hypercoaguability, uncontrolled hemorrhage, or both simultaneously (75,106). Disseminated intravascular coagulation in cancer patients has been associated with both ischemic stroke as well as intracerebral hemorrhage (107,108). The chronic, or compensated, form of DIC is typically seen in patients with solid tumors, commonly with adenocarcinoma of gastrointestinal, lung, or breast origin. Chronic DIC more often manifests as thrombosis, rather than bleeding, although either or both hematologic dyscrasias are possible. Chronic DIC has been reported in the clinical settings of NBTE and mucin-positive adenocarcinoma-associated thrombosis. Patients in chronic DIC typically present with deep venous thrombosis and pulmonary thromboembolism, although some may also develop arterial ischemia. Acute, or uncompensated, DIC occurs most frequently with hematogenous malignancies such as the acute leukemias, and is less frequent with solid tumors. Acute DIC typically presents with clinically significant bleeding with concomitant thrombosis. Bleeding from venipuncture sites and surgical wounds may be seen, as well as diffuse mucosal, skin, or retroperitoneal hemorrhage. Central nervous system hemorrhage is a significant and potentially fatal complication, especially in acute promyelocytic leukemia (APML) (109). Routine coagulation tests, such as the prothrombin time (PT) and the partial thromboplastin time (PTT), may be normal or only mildly prolonged in the chronic variety of DIC. Fibrinogen levels may be low due to consumption, but often are normal because of a co-existing acute phase response. A platelet count less than 100,000, or one that is rapidly declining, is suggestive. Additional molecular markers of hemostatic system activity that

Chapter 14 / Cerebrovascular Complications of Cancer

225

frequently provide laboratory support for the diagnosis include elevated serum levels of D-dimer, fibrinopeptide A, thrombin-antithrombin complex, plasmin–plasmin inhibitor complex, and soluble fibrin monomer (107,110,111). No prospective clinical trials exist regarding the optimal treatment for chronic or acute DIC that is associated with symptomatic thrombosis. Treatment of the precipitating cause, in this case the malignancy, is fundamental to successful long-term therapy (107,112). Anticoagulants, by interrupting the coagulation cascade, are of theoretical benefit. Unfractionated and low molecular weight heparins have appeared beneficial in small, uncontrolled cohorts but have not been definitively evaluated in controlled clinical trials (112,113). Even in patients with bleeding, heparin appears to be relatively safe. A low starting dose of heparin is often recommended, for example continuous infusion of unfractionated heparin at 5–10 units/kg/hr, adjusting the dose as needed to stabilize and normalize platelet count and fibrinogen level. Direct thrombin inhibitors are promising but unvalidated in controlled trials. Administration of antithrombin III concentrate at supraphysiologic doses has shown modest benefit in controlled clinical trials (106,112). In patients with severely low levels of platelets and coagulation factors, treatment with platelet transfusion and plasma is beneficial.

3.7. Bleeding Diathesis/Hemorrhage 3.7.1. Primary Fibrinolysis Primary fibrinolysis is characterized by systemic activation of plasmin or direct fibrinogen degradation. Intracranial hemorrhage may result. Both DIC and systemic primary fibrinolysis may co-exist in patients with APML and in patients with prostate cancer. Primary fibrinolysis alone has been observed in leukemias and solid tumors (107). Neoplasms may precipitate primary fibrinolysis through several potential mechanisms, including direct tumor production of tissue plasminogen activator (t-PA), direct tumor production of urokinase plasminogen activator (u-PA), and direct tumor production of a protease that digests fibrinogen. Laboratory findings suggesting the diagnosis of primary fibrinolysis include a normal platelet count in the setting of hemorrhage associated with hypofibrinogenemia, elevated fibrinogen degradation products, and negative tests of markers of activation of the coagulation system. Factors V and VIII may be diminished due to plasmin digestion. Treatment consists of administering cryoprecipitate or fresh frozen plasma. Epsilon-aminocaproic acid or tranexamic acid may also be given (114). 3.7.2. Hyperleukocytic Syndrome The primary neurological manifestation of the hyperleukocytic syndrome is intracranial hemorrhage (115). The hyperleukocytic syndrome is a distinct entity that typically affects patients with acute myelogenous leukemia (AML), although it can also occur in chronic lymphocytic leukemia (CLL). The increased number of abnormal white blood cells, or myeloblasts, elevates the patient’s white blood cell count. Circulating myeloblasts, especially in AML, tend to be less deformable than normal leukocytes. These abnormal calls increase blood viscosity, which in turn can precipitate “sludging” and vessel occlusion with aggregation of white cell thrombi in the vasculature. Additionally, myeloblasts can be invasive and directly damage blood vessels walls. Vessels in the central nervous system and pulmonary circulations are most vulnerable to invasion (116). An AML patient with a total white blood cell count greater than 100,000/mm3 or a CLL patient with an absolute lymphocyte count greater than 250,000/mm3 should be considered at risk for this event. Treatment for CNS leukostasis can include one or two high dose fractions of radiation therapy, combined allupurinol/hydroxyurea therapy, exchange transfusions or leukapheresis. The effectiveness of these treatments is unclear, especially in the setting of intracranial hemorrhage (117). If intracranial hemorrhage occurs, the survival rate is extremely low. 3.7.3. Thrombocytopenia Thrombocytopenia is not uncommon in the cancer population and poses a risk for intracranial hemorrhage (118–120). Thrombocytopenia-associated cerebral hemorrhage in oncology patients can be secondary to extensive marrow infiltration by tumor, peripheral destruction of platelets due to tumor-associated hypersplenism, underproduction of platelets due to radiation- or chemotherapy-induced toxicity, DIC, autoimmune dysfunction, and/or microangiopathic hemolytic anemia. Immune-mediated peripheral platelet destruction is rarely seen with solid tumors, but has been reported with lymphoproliferative disorders such as Hodgkin’s disease, CLL, and low-grade lymphoma (121,122). Diagnosis is difficult but may be supported by the acute onset of thrombocytopenia, large

226

Part V / Indirect Complications of Cancer

platelet size, elevated megakaryocyte count, and increased platelet-associated immunoglobulin. Treatment may include corticosteroids, immunoglobulin infusions, plasmapheresis, antineoplastic therapy directed at the specific underlying malignancy, vincristine, danazol, and immunoabsorption with staphylococcal protein A. Thrombotic thrombocytopenic purpura (TTP) is a syndrome of target organ dysfunction due to marked platelet aggregation in the microcirculation that can be induced both by cancer and by chemotherapeutic treatment. Thrombotic thrombocytopenic purpura is characterized by severe thrombocytopenia, a microangiopathic hemolytic anemia, and renal failure (hemolytic-uremic syndrome) (123–125). Recent pathophysiologic investigations suggest that endothelial cell perturbation and apoptosis caused by as yet unidentified plasma factors lead to release of an abnormal von Willebrand factor that facilitates the deposition of platelet microthrombi. Intracerebral hemorrhage and cerebral infarction are potentially disastrous events that may complicate the course of patients with TTP (126). Platelet aggregates in TTP most commonly occlude the arterioles and capillaries in the brain, heart, kidneys, and adrenal glands. Clinically, purpuric rash, fever, and neurological and renal symptoms are common. Laboratory studies demonstrate severe hemolytic anemia, thrombocytopenia, and schistocytosis. Thrombotic thrombocytopenic purpura can be differentiated from DIC by the absence of a coagulopathy. In cancer patients, TTP is most commonly seen with gastric adenocarcinoma, followed by breast, colon, and small cell lung carcinoma. Treatment options include corticosteroids, plasma exchange, immunoabsorption with staphylococcal protein A, platelet inhibitor drugs, vincristine, and splenectomy. Platelet transfusions are reserved for situations of documented bleeding. Mortality in TTP without treatment is 90–100%. With appropriate treatment mortality decreases to 10%.

4. TRADITIONAL STROKE MECHANISMS APPEARING IN THE SETTING OF TUMOR Patients with cancer frequently have risk factors for stroke independent of their neoplasm, such as hypertension, atrial fibrillation, hypercholesterolemia, diabetes, coronary artery disease, and tobacco use (127,128). Accordingly, cancer patients may experience ischemic stroke through mechanisms unrelated to oncologic factors. It is possible, but unproven, that the risk of these standard mechanisms may be potentiated by neoplasm-associated thrombophilia. Determining whether and to what extent neoplasm-associated hypercoagulability contributed to an infarct in an individual patient is often difficult. Similarly, cancer patients may develop intracerebral hemorrhages due to risk factors that have been identified in other patient groups, such as hypertension, amyloid angiopathy, coagulopathy, or head trauma, or an interaction of standard risk factors and a neoplasm-associated bleeding diathesis.

5. SEQUELAE OF CANCER DIAGNOSTIC TESTS AND TREATMENT In addition to cancer and its associated coagulopathies as a cause of stroke, ischemic or hemorrhagic strokes can result from certain diagnostic procedures, radiation therapy, surgical therapy, endovascular treatments, or chemotherapy.

5.1. Diagnostic Procedures 5.1.1. Lumbar Puncture Spinal subdural hematoma is a rare sequela of lumbar puncture in cancer patients with severe thrombocytopenia or other coagulopathies (129,130). Back pain, myelopathic symptoms from spinal cord compression, or radiculopathy from nerve root compression typically develop within 24 hrs of the procedure. Early surgical treatment is indicated for patients with symptoms of cord or nerve root compression, and may yield good recovery of function. To prevent lumbar puncture–associated subdural hematoma, preprocedural platelet transfusions should be considered if the procedure is mandatory despite severe thrombocytopenia. 5.1.2. Lymphangiography Lymphangiography is rarely associated with ischemic stroke due to lipid embolism. Typically the clinical picture is an encephalopathy appearing within hours of the procedure, although mild focal deficits have also been noted. Pulmonary embolism and skin petechiae may occur as well.

Chapter 14 / Cerebrovascular Complications of Cancer

227

5.2. Cancer Therapy – Radiation 5.2.1. Radiation-Induced Vasculopathy There are a variety of delayed vasculopathies that may complicate neck or brain radiation. An occlusive vasculopathy with accelerated atherosclerosis can develop cranial vessels within the radiated field. Pathologic studies demonstrate that radiotherapy produces a sequence of changes in arteries characterized by initial damage to endothelial cells, thickening of the intimal layer caused by smooth muscle cell proliferation, cellular degeneration, and hyaline transformation. In two series, large vessel vasculopathy developed intracranially in 19% of children treated with cranial irradiation. Risk factors for vasculopathy were increased radiation dose to the circle of Willis and major cerebral arteries and possibly younger age (131,132). Intracranial vasculopathy may appear after wholebrain radiation, gamma knife, or other focused cranial radiation, and radiation brachytherapy (133). Stenosis and occlusion of medium and large vessels leading to ischemic infarction is the most common sequelae, but lacunar infarction, primary intracerebral hemorrhage, moyamoya changes with hemorrhage or ischemia, formation of cerebral aneurysms with subarachnoid or intraparenchymal hemorrhage, and formation of vascular malformations have also been reported (134–138). The cervical internal and common carotid arteries are a particularly common site for radiation-induced accelerated atherosclerosis following radiotherapy for head and neck cancer. In recent large series of ultrasound studies, the frequency of common or internal carotid artery stenosis after external irradiation for head and neck malignancy ranged from 12–60% (139,140). Symptomatic clinical presentation is typically delayed by months to years following irradiation (141). Hemispheric TIAs, hemispheric strokes, amaurosis fugax, and/or seizures are the typical neurological presentation in patients with carotid disease. Transient ischemic attack and infarction have also been observed in patients with vertebral stenoses (142). On physical exam, affected patients may exhibit extensive post-radiation skin atrophy and fibrosis of the tissues overlying the diseased vascular segment in the neck, and radiation-induced necrosis of the mandible. Several authors have recommended routine surveillance noninvasive imaging of the carotid artery following head and neck radiation (143,144). For symptomatic post-radiation carotid stenosis, and possibly for asymptomatic severe post-radiation stenosis, revascularization procedures may be indicated. Though technically more challenging than in standard cases, carotid endarterectomy is feasible in radiation-induced carotid stenosis, and long-term patency rates similar to that obtained in the absence of radiation therapy have been reported (142,145). Other variations of carotid artery repair, such as external carotid endarterectomy, carotid patch angioplasty alone, aorto-carotid bypass grafting, subclavian-carotid bypass grafting, and carotid interposition grafting have also been reported to be safe in this patient population. Carotid angioplasty and stenting has been a safe and effective technique for high-risk surgical candidates, although the long-term patency after stenting is not yet fully characterized (146). In head and neck cancer patients, arterial injury due to surgery and radiation therapy may also present as arterial rupture. Post-radiation rupture of the carotid artery typically occurs within 2–16 weeks after radical neck surgery and radiation therapy (147). Of note, in the series reported by McCready and colleagues, all patients with carotid rupture were infected, with sloughing of the skin flaps and development of orocutaneous fistulas (147). Nd:YAG laser bronchoscopy treatment of patients with unresectable pulmonary cancer has been associated with stroke due to air embolism to the cerebral circulation (148).

5.3. Cancer Therapy – Effects of Surgery 5.3.1. Direct Effects of Surgery Bronchoscopic biopsies and pneumonectomies for pulmonary cancer have been associated with perioperative and post-operative stroke secondary to tumor emboli. Surgical manipulation of the lung promotes release of emboli, especially in the setting of tumor invasion into pulmonary vasculature (64,65). Stroke tends to occur within the first 48 hrs post-pneumonectomy (Fig. 5). Stroke is also a risk associated with resections of advanced head and neck cancer when the neoplasm encroaches upon, encases, or invades the cervical carotid artery. Surgical options for this challenging management dilemma include carotid ligation, “shaving” the tumor from the carotid, or en bloc resection/replacement of the carotid artery by polytetrafluoroethylene, vein graft, or superficial femoral artery graft (149–151). Surgical ligation of the carotid in unselected patients has been associated with stroke rates and mortality of up to 50%. Presurgical endovascular

228

Part V / Indirect Complications of Cancer

Fig. 5. Bronchoscopy procedure–induced tumor emboli. A 70-year-old woman with non-small cell lung adenocarcinoma underwent fiberoptic bronchoscopy and mediastinoscopy with tracheobronchial lymph node biopsy. Two hours after the procedure, she developed left faciobrachial weakness. (A) Diffusion weighted MRI shows right frontal, right parietal, and right occipital infarcts (white arrows). (B) Lower slice from diffusion MR study demonstrates additional acute bilateral cerebellar infarcts (white arrows).

test occlusion of the carotid artery permits identification of patients with adequate circle of Willis collaterals who will tolerate permanent carotid occlusion, and of patients with inadequate Willisian collaterals who will require extracranial–intracranial bypass surgery or resections that spare or reconstruct the carotid artery (151,152). In patients with adequate collaterals, permanent balloon occlusion may be undertaken, and may decrease the risk of perioperative carotid rupture. With regard to oncologic neurosurgical procedures, cerebral venous thrombosis has been reported as a rare complication of craniotomies and craniectomies (153,154). Anticoagulation, lumboperitoneal shunting, or ventriculoperitoneal shunting may be appropriate treatments in this situation. Symptomatic cerebral vasospasm is a rare complication of surgery in patients undergoing cranial base tumor resection (155). Standard vasospasm treatment is indicated, including hypervolemia, hypertensive therapy, intraarterial angioplasty, and intraarterial papaverine. In recent years, the significance of immediate post-operative imaging changes suggesting tissue injury and infarction after the resection of gliomas has been recognized. Ulmer (156) and Smith (157) report the largest series of cases, describing restricted diffusion in up to 70% of patients on early post-operative MRI scans. Some of these patients present with new post-operative neurological signs. The regions of restricted diffusion typically develop into encephalomalacia within 90 days, suggesting that the imaging changes signify infarction associated with surgery. The later scans also often demonstrate enhancement, which can be misinterpreted as tumor progression. Immediate post-operative areas of brain ischemia can also be identified outside the area of a resected brain tumor (158) and the mechanism for this is not known. Hemorrhage in the area of brain tumor biopsy or resection is a rare complication of brain tumor surgery that is described most commonly in patients undergoing surgery for a glioma. The hemorrhage may be immediate or delayed by several days (159). 5.3.2. Hypercoagulability Related to Surgery Post-operative hypercoaguability has been reported in patients both with and without cancer. Protein C levels have been noted to decrease within 72 hrs in patients undergoing minor and major surgeries. In cancer patients, however, the decrease in protein C occurs more rapidly, typically within the first 24 hrs (160). One prospective study of 10 patients undergoing free flap microvascular reconstruction of cancer-related defects in the head and neck noted protein C deficiency in 70% of all patients within the first 72 hrs post-operatively (161). In addition to protein C abnormalities, antithrombin III and plasminogen levels have been noted to decrease within the first 48–96 hrs post-operatively (162). Lastly, DIC can develop after surgery.

Chapter 14 / Cerebrovascular Complications of Cancer

229

5.3.3. Endovascular Treatment–Associated Stroke Selective intra-arterial infusion of blood–brain barrier disruption and antineoplastic agents is a treatment approach for selected cerebral malignancies (163). Ischemic stroke is a reported complication of these procedures (164). Superselective catheter-administered embolization to occlude the vascular supply to meningiomas has been associated with post-procedural peritumoral hemorrhage (165). Ischemic stroke has been reported during bone marrow transplantation infusion. A patient with a patent foramen ovale, permitting right-to-left shunting, developed multiple cerebral emboli (166). Positioning the infusion catheter tip in the main pulmonary artery and reducing the volume of marrow infused are steps that can prevent this complication.

5.4. Cancer Therapy – Chemotherapy and other Antineoplastics 5.4.1. Hypercoagulability and Thrombocytopenia Antineoplastic chemotherapy, including single or multiagent chemotherapy, hormonal therapy, and hematopoietic growth factors can produce a hypercoagulable state in cancer patients and contribute to cerebral arterial and venous thrombosis. Physiologic investigations in patients treated with chemotherapeutic agents have documented activation of the coagulation pathway, suppression of natural anticoagulants, suppression of natural fibrinolysis, and injury to vascular endothelium (167,168). Thrombocytopenia, TTP, DIC, and microangiopathic hemolytic anemia have all been linked to chemotherapeutic agents (120,169,170). Postulated mechanisms for antineoplastic drug-related thrombophilia include release of procoagulants and cytokines from injured tumor cells, direct drug toxicity to vascular endothelium, direct induction of monocyte or malignant cell tissue factor, and decrease in physiological anticoagulants (167). Among individual chemotherapeutic agents associated with stroke, l-asparaginase is one of the most wellknown (118,168,171–173). L-asparaginase is an enzymatic inhibitor of protein synthesis that is used in combination with other chemotherapeutic agents in the treatment of acute lymphoblastic leukemia and some other lymphoid malignancies. The reduction in protein synthesis produced by l-asparaginase not only inhibits growth of leukemic neoplasms, but also decreases liver production of multiple plasma proteins involved in hemostasis. Strokes associated with l-asparaginase induction therapy may present as dural sinus thrombosis, cortical or capsular infarction, or intracerebral hemorrhage. Venous thrombosis is most common. The incidence of stroke in patients treated with l-asparaginase induction therapy has ranged in different series from 0.9% to 2.9%. Stroke tends to occur shortly after induction treatment. The clinical presentation varies depending on the location and type of stroke. The exact mechanism for l-asparaginase-associated thrombosis stroke is unclear, although l-asparaginase has been shown to diminish antithrombin III, protein C, protein S, factor XI, factor IX, and fibrinogen, and to increase PT/PTT and platelet aggregability. Coagulation factors return to normal within 7–10 days after therapy. Therapies for l-asparaginase-associated strokes, depending on the clinical situation, vary widely and may include fresh frozen plasma, heparin, cryoprecipitate, platelet transfusion, aspirin, and surgery for hematoma drainage. Stroke has also been associated with a variety of other chemotherapeutic agents. 5-fluorouracil therapy, alone and in combination with cisplatin, methotrexate, and cyclophosphamide, has been associated with acquired protein C deficiency and stroke (174,175). Acute stroke and acquired protein C deficiency has also been reported following cisplatin therapy without 5-fluorouracil (176,177). Stroke has been associated with paclitaxel, shortly after administration (178). The antiestrogen tamoxifen induces a mild hypercoagulable state and may increase the incidence of stroke in women over age 50 (179). Granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colonystimulating factor (GM-CSF) have been associated with venous and arterial thrombosis, possibly by enhancing aggregation and binding of neutrophils to vascular endothelium. A meta-analysis of 52 reported series found an incidence of venous and arterial thrombosis of 4.2% with GM-CSF and 1.2% with G-CSF (180). Bevacizumab, a monoclonal antibody that binds to and inhibits the biologic activity of vascular endothelial growth factor (VEGF), when given in combination with chemotherapy, results in improved survival than chemotherapy alone for nonsmall cell lung cancer and colon cancer. The most significant toxicities of this agent are systemic thrombosis or hemorrhage, including deep venous thrombosis, myocardial infarction, and hemoptysis. Central nervous system thrombosis or hemorrhage, including subarachnoid hemorrhage, can also occur (181).

230

Part V / Indirect Complications of Cancer

Based on a CNS hemorrhage occurring in one patient with lung cancer brain metastasis in an early trial of bevacizumab, patients with brain metastasis have been excluded from subsequent controlled clinical trials of this agent in solid tumors, so the exact incidence of brain hemorrhage in these patients is not known. There are rare instances of brain hemorrhage resulting from the hypertension induced by bevacizumab. The combination of bevacizumab and irinotecan is currently being used as treatment for recurrent malignant gliomas. In a recent report of this treatment administered to 32 such patients, there were no instances of brain hemorrhage, but there was one instance of cerebral infarction (182). 5.4.2. Cardiomyopathy Cardiomyopathy is a well-known complication of anthracycline chemotherapy with agents such as doxirubicin and epirubicin, occurring in up to 20% or more of patients (183). Other chemotherapeutic agents less commonly associated with cardiotoxicity include cyclophosphamide, ifosfamide, cisplatin, carmustine, busulfan, and mitomycin. Severe cardiomyopathy with reduced flow in cardiac chambers permits thrombus formation and cardioembolic stroke. In patients presenting with cerebral cardioembolism, long-term anticoagulation may be recommended for secondary prevention. 5.4.3. Infection and Stroke Patients with immunosuppression from the effects of antineoplastic therapy or directly from hematologic malignancy are at increased risk for infection-related stroke through several mechansims, including sepsis-induced DIC, bacterial endocarditis, and angioinvasive microorganisms (1,184,185). Immunosuppressed leukemic patients may experience cerebral infarcts secondary to vessel infiltration with fungi such as Mucor, Aspergillus, and Candida. In Graus’s autopsy series, the majority of patients with septic infarction were symptomatic with seizures, focal neurological deficits, and encephalopathy (1). 5.4.4. Anticoagulation-Induced Hemorrhage in the Cancer Patient In cancer patients who develop deep venous thrombosis, treatment options include long-term warfarin anticoagulation or placement of filters in the inferior vena cava. Patients with malignancy placed on warfarin therapy are at increased risk for major hemorrhages, with an incidence rate of 13.3 per 100 patient years. Patients with brain metastases and deep venous thrombosis may be at particularly increased risk for intracerebral hemorrhage on warfarin therapy. Among 42 patients who received anticoagulation for a mean of 100 days, the incidence of symptomatic intracerebral hemorrhage was 7%, generally in the setting of supratherapeutic anticoagulation (186). This risk must be balanced against data suggesting greater effectiveness of anticoagulation than filter placement in preventing recurrent deep venous thrombosis and pulmonary embolism. 5.4.5. Bone Marrow Transplantation A thrombotic microangiopathy, which may affect the CNS, occurs in up to 6% of patients following bone marrow transplantation (187,188). Contributing factors include the administration of cyclosporin A, graft-versushost disease, irradiation, intensive conditioning chemotherapy, and infection. Subdural hematoma and other sites of intracranial hemorrhage occur with prolonged thrombocytopenia and coagulopathy as predisposing factors (189–192). Intraparenchymal hemorrhage has also been identified at autopsy following bone marrow transplantation. Cases of NBTE-associated stroke have also been reported following bone marrow transplantation (193). Central nervous system angiitis is a rare complication of graft versus host disease (194).

6. CONCLUSION A diverse array of pathophysiologic processes increases the risk of stroke in patients with malignancies. A systematic evaluation will often disclose the type, location, and proximate cause of stroke, allowing patient classification among the specific etiologies of cerebral infarction and hemorrhage reviewed in this chapter. Accurate diagnosis will guide acute intervention and secondary prevention treatment. All physicians who encounter patients with cancer should be cognizant of the risk of cerebrovascular disease within the oncologic population, and include stroke in the differential diagnosis of any alteration in central nervous system function.

Chapter 14 / Cerebrovascular Complications of Cancer

231

REFERENCES 1. Graus F, Rogers LR, Posner JB. Cerebrovascular complications in patients with cancer. Medicine 1985;64:16–35. 2. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. In BS Schoenberg (ed.): Advances in Neurology. Vol 19. New York: Raven Press, 1978: 579–592. 3. Kothbauer P, Jellinger K, Flament H. Primary brain tumour presenting as spontaneous intracerebral hemorrhage. Acta Neurochir 1979;49:35–45. 4. Little JR, Dial B, Belanger G et al. Brain hemorrhage from intracranial tumor. Stroke 1979;10:283–288. 5. Kondziolka D, Bernstein M, Resch L et al. Significance of hemorrhage into brain tumors: clinopathological study. J Neurosurg 1987;67:852–857. 6. Iwama T, Ohkuma A, Miwa Y et al. Brain tumors manifesting as intracranial hemorrhage. Neurol Med Chir 1992;32:130–135. 7. Nutt SH, Patchell RA. Intracranial hemorrhage associated with primary and secondary tumors. Neurosurg Clin N Am 1992;3:591–599. 8. Nataf F, Emery E, Kherli P, Seigneuret E. Neurosurgical multicenter study of cerebral metastases. Neurochirurgie 1999;45:369–374 9. Barth H, Fritsch G, Haaks T. Intracerebral hematoma as an acute manifestation of intracranial tumors. Nervenarzt 1994;65:854–858. 10. Otsuka S, Nakatsu S, Matsumoto S et al. Brain tumors associated with hemorrhage from tumors as their first manifestation. Nippon Geka Hokan 1989;58:147–154. 11. Wakai S, Yamakawa K, Manaka S et al. Spontaneous intracranial hemorrhage caused by brain tumor: its incidence and clinical significance. Neurosurgery 1982;10:437–444. 12. Isoda H, Takahashi M, Arai T et al. Multiple haemorrhagic brain metastases from papillary thyroid cancer. Neuroradiology 1997;39:198–202. 13. Mandybur TI. Intracranial hemorrhage caused by metastatic tumors. Neurology 1977;27:650–655. 14. Scatliff JH, Radcliff WB, Pitman HH et al. Vascular structure of glioblastomas. Am J Radiol 1969;105:795–805. 15. Weller RO, Foy M, Cos S. The development and ultrastructure of the microvasculature in malignant gliomas. Neuropathol Appl Neurobiol 1977:3:307–322. 16. Cheng SY, Nagane M Huang HS et al. Intracerebral tumor–associated hemorrhage caused by overexpression of the vascular endothelial growth factor isoforms VEGF121 and VEGF165 but not VEGF189. Proc Natl Acad Sci USA 1997;94:12081–12087. 17. Hirano A, Matsui T. Vascular structures in brain tumors. Hum Pathol 1975;6:611–621. 18. Jung S, Moon KS, Jung TY et al. Possible pathophysiological role of vascular endothelial growth factor (VEGF) and matrix metalloproteinases (MMPs) in metastatic brain tumor–associated intracerebral hemorrhage. J Neuro-oncol 2006;76:257–263. 19. Scott M. Spontaneous intracranial hemorrhage caused by cerebral neoplasms. J Neurosurg 1975;10:437–444. 20. Bosnjak R, Berham C, Popovic M et al. Spontaneous intracranial meningioma bleeding: clinicopathological features and outcome. J Neurosurg 2005;103:473–84. 21. Lazaro RP, Messer HD, Brinker RA. Intracranial hemorrhage associated with meningioma. Neurosurgery 1981;8:96–101. 22. Atlas SW, Grossman RI, Gomori JM et al. Hemorrhagic intracranial malignant neoplasms: spin-echo MR imaging. Radiology 1987;164:71–77. 23. Dylewski DA, Demchuk AM, Morgenstern LB. Utility of magnetic resonance in acute intracerebral hemorrhage. J Neuroimaging 2000;10:78–83. 24. Tseng SH, Liao CC, Lin SM et al. Dural metastasis in patients with malignant neoplasm and chronic subdural hematoma. Acta Neurol Scand 2003;108: 43–46. 25. Bromberg JE, Vandertop WP, Jansen GH. Recurrent subdural haematoma as the primary and sole manifestation of chronic lymphocytic leukemia. Br J Neurosurg 1998;12:373–376. 26. Laigle-Donadey F, Taillibert S, Mokhtari K et al. Dural metastases. J Neurooncol 2005;75:57–61. 27. Sunada I, Nakabayashi H, Matsusaka Y et al. Meningioma associated with acute subdural hematoma: case report. Radiat Med 1998;16: 483–486. 28. Minette SE, Kimmel DW. Subdural hematoma in patients with systemic cancer. Mayo Clin Proc 1989;64:637–642. 29. Wang AM, Chinwuba CE, O’Reilly GV et al. Subdural hematoma in patients with brain tumor: CT evaluation. J Comput Assist Tomogr 1985; 9:511–513. 30. Raizer JJ, DeAngelis LM. Cerebral sinus thrombosis diagnosed by MRI and MR venography in cancer patients. Neurology 2000;54:1222–1226. 31. Brown MT, Friedman HS, Oakes WJ et al. Sagittal sinus thrombosis and leptomeningeal medulloblastoma. Neurology 1991;41: 455–456. 32. Reddinghaus RE, Patte C, Couanet D et al. Dural sinus thrombosis in children with cancer. Med Pediatri Oncol 1997;29:296–302. 33. Kim AW, Trobe JD. Syndrome simulating pseudotumor cerebri caused by partial transverse venous sinus obstruction in metastatic prostate cancer. Am J Ophthalmol 2000;129:254–256. 34. Bousser MG, Ferro JM. Cerebral venous thrombosis: an update. Lancet 2007;6:162–170. 35. Goldsmith P, Burn DJ, Coulthard A et al. Extrinsic cerebral venous sinus obstruction resulting in intracranial hypertension. Postgrad Med J 1999;75:550–551. 36. Zilkha A, Diaz AS. Computed tomography in the diagnosis of superior sagittal sinus thrombosis. J Comput Assist Tomogr 1980;4: 124–126. 37. Lafitte F, Boukobza M, Guichard JP et al. MRI and MRA for diagnosis and follow-up of cerebral venous thrombosis. Clin Radiol 1997;52:672–679. 38. Einhaupl KM, Villringer A, Meister W et al. Heparin treatment in sinus venous thrombosis. Lancet 1991;338:597–600. 39. De Bruijn SF, Stam J. Randomized, placebo-controlled trial of anticoagulant treatment with low-molecular-weight heparin for cerebral sinus thrombosis. Stroke 1999;304:484–488.

232

Part V / Indirect Complications of Cancer

40. Buccino G, Scoditti U, Pini M et al. Loco-regional thrombolysis in the treatment of cerebral venous and sinus thrombosis: report of two cases. Acta Neurol Scand 2001;103:59–63. 41. Philips MF, Bagley LJ, Sinson GP et al. Endovascular thrombolysis for symptomatic cerebral venous thrombosis. J Neurosurg 1999;90:65–71. 42. Chow K, Gobin YP, Saver J et al. Endovascular treatment of dural sinus thrombosis with rheolytic thrombectomy and intra-arterial thrombolysis. Stroke 2000;31:1420–1425. 43. Ho KL. Neoplastic aneurysm and intracranial hemorrhage. Cancer 1982;50:2935–2940. 44. Kalafut M, Vinuela F, Saver JL et al. Multiple cerebral pseudoaneurysms and hemorrhages: the expanding spectrum of metastatic cerebral choriocarcinoma. J Neuroimaging 1998;8:44–47. 45. Guttman DH, Cantor CR, Piacente GJ et al. Cerebral vasculopathy and infarction in a woman with carcinomatous meningitis. J Neuro-oncol 1990;9:183. 46. Klein P, Haley C, Wooten GF et al. Focal cerebral infarctions associated with perivascular tumor infiltrates in carcinomatous leptomeningeal metastases. Arch Neurol 1989;46:1149–1152. 47. Latchaw RE, Gabrielsen TO, Seeger JF. Cerebral angiography in meningeal sarcomatosis and carcinomatosis. Neuroradiology 1974;8:131. 48. Rojas-Marcos I, Martin-Duverneuil N, Laigle-Donadey F et al. Ischemic stroke in patients with glioblastoma. J Neurol 2005;252: 488–489. 49. Charache S, Adams RJ. Neurologic complications associated with disorders of red blood cells. In Handbook of Clinical Neurology Vol. 19: Systemic Diseases, Part I. Baltimore: Elsevier Science Publishers, 1993, 249–269. 50. Berk PD, Goldberg JD, Donovan PB et al. Therapeutic recommendations in polycythemia vera based on the Polycythemia Vera Study Group protocols. Semin Hematol 1986;23:132–143. 51. Gruppo Italiano Studio Policitemia. Polycythemia vera: the natural history of 1213 patients followed for 20 years. Ann Intern Med 1995;123:656–664. 52. Arboix A, Besses C, Acin P et al. Ischemic stroke as first manifestation of essential thrombocythemia: report of six cases. Stroke 1995;26:1463–1468. 53. Michiels JJ. Aspirin and platelet-lowering agents for the prevention of vascular complications in essential thrombocythemia. Clin Appli Thromb Hemost 1999;5:247–251. 54. Arboix A, Costa I, Besses C. Intravascular lymphomatosis: a rare etiology of recurrent cerebral ischemia. Rev Neurol 2000;30: 1188–1190. 55. Calamia KT, Miller A, Shuster EA et al. Intravascular lymphomatosis: a report of ten patients with central nervous system involvement and a review of the literature. Adv Exp Med Biol 1999;455:249–265. 56. Liow K, Asmar P, Liow M et al. Intravascular lymphomatosis: contribution of cerebral MRI findings to diagnosis. J Neuroimaging 2000;10:116–118. 57. Roeltgen DP, Syna DR. Neurological complications of cardiac tumors. In: Goetz CG, Tanner CM, Aminoff MJ (eds.). Handbook of Clinical Neurology: Systemic Diseases, Part I. New York: Elsevier Science Publishers, 1993, 93–109. 58. Brebenc ML, Rosado de Christenson ML, Burke AP et al. Primary cardiac and pericardial neoplasms: radiologic–pathologic correlation. Radiographics 2000;20:1073–1103. 59. Liebeskind DS, Buljubasic N, Saver JL. Cardioembolic stroke due to papillary fibroelastoma. J Stroke Cerebrovasc Dis 2001;10:94–95. 60. Schaff HV, Mullany CJ. Surgery for cardiac myxomas. Semin Thorac Cardiovasc Surg 2000;12:77–88. 61. Klatt EC, Heitz DR. Cardiac metastases. Cancer 1990;65:1456–1459. 62. Gibbs P, Cebon JS, Calafiore P et al. Cardiac metastases from malignant melanoma. Cancer 1999;85:78–84. 63. Weir B, MacDonald, Mielke B. Intracranial vascular complications of choriocarcinoma. Neurosurgery 1978;2:138–142. 64. O’Neill BP, Dinapoli RP, Okazaki H. Cerebral infarction as a result of tumor emboli. Cancer 1987;60:90–95. 65. Lefkovitz NW, Roessmann UR, Kori SH. Major cerebral infarction from tumor embolus. Stroke 1986;17:555–557. 66. Urrutia V, Jichici D, Thomas CE et al. Embolic stroke secondary to an aortic arch tumor: a case report. Angiology 2000;51:959–962. 67. Alam M, Rosman HS, Grullon C. Transesophageal echocardiography in evaluation of atrial masses. Angiology 1995;46:123–128. 68. Araoz PA, Mulvagh SL, Tazelaar HD et al. CT and MR imaging of benign primary cardiac neoplasms with echocardiographic correlation. Radiographics 2000;20:1303–1319. 69. Wintersperger BJ, Becker CR, Gulbins H et al. Tumors of the cardiac valves: imaging findings in magnetic resonance imaging, electron beam computed tomography, and echocardiography. Eur Radiol 2000;10:443–449. 70. Ravikumar E, Pawar N, Gnanamuthu R et al. Minimal access approach for surgical management of cardiac tumors. Ann Thorac Surg 2000;70:1077–1079. 71. Knepper LE, Biller J, Adams HP et al. Neurological manifestations of atrial myxoma: a 12-year experience and review. Stroke 1988;19:1435–1440. 72. Nielsen EH, Lindholm J, Bjerre P et al. Frequent occurrence of pituitary apoplexy in patients with nonfunctioning pituitary adenoma. Clin Endocrinol 2006;64:319–322. 73. Da Motta LACR, De Mello PA, De Lacerda CM et al. Pituitary apoplexy: clinical course, endocrine evaluations, and treatment analysis. J Neurosurg Sci 1999;43:25–36. 74. Trousseau A. Phlegmasia alba dolens. In: Clinique Medicale de L’Hotel Dieu de Paris 3, 2nd ed. Paris: Balliere, 1865. 75. Falanga A, Rickles FR. Pathophysiology of the thrombophilic state in the cancer patient. Sem Thromb Hemostasis 1999;25:173–182. 76. Schwartz JD, Simantov R. Thrombosis and malignancy: pathogenesis and prevention. In Vivo 1998;12:619–624. 77. Arkel YS. Thrombosis and cancer. Semin Oncol 2000; 27:362–374. 78. Miljic P, Milosevic-Jovicic N, Antunovic P et al. Recurrent venous thrombosis in a patient with chronic lymphocytic leukemia and acquired protein S deficiency. Haematologia 2000;30:51–54.

Chapter 14 / Cerebrovascular Complications of Cancer

233

79. Deitcher SR, Erban JK, Limentani SA. Acquired free protein S deficiency associate with multiple myeloma: a case report. Am J Hematol 1996;51:319–323. 80. Odegaard OR, Lindahl AK, Try K et al. Recurrent venous thrombosis during warfarin treatment related to acquired protein S deficiency. Thromb Res 1992;66:729–734. 81. Conlan MG, Mosher DF. Concomitant chronic lymphocytic leukemia, acute myeloid leukemia, and thrombosis with protein C deficiency: case report and review of the literature. Cancer 1989;63:1398–1401. 82. Gurgey A, Buyukpamukeu M, Baskut C et al. Portal vein thrombosis in association with factor V Leiden mutation in a patient with hepatocellular carcinoma. Med Pediatr Oncol 1997;29:224–225. 83. Nowak-Gottl U, Aschka I, Koch HG et al. Resistance to activated protein C (APCR) in children with acute lymphoblastic leukaemia: the need for a prospective multicenter study. Blood Coagul Fibrinolysis 1995;6:761–764. 84. Haim N, Lanir N, Hoffman R et al. Acquired activated protein C resistance is common in cancer patients and is associated with venous thromboembolism. Am J Med 2001;110:91–96. 85. Sproul EE. Carcinoma and venous thrombosis: the frequency of association of carcinoma in the body or tail of the pancreas with multiple venous thrombus. Am J Cancer 1938;34:566–585. 86. Bedikian A, Valdivieso M, Luna M et al. Nonbacterial thrombotic endocarditis in cancer patients: comparison of characteristics of patients with and without concomitant disseminated intravascular coagulation. Med Pediatri Oncol 1978;4:149–157. 87. Rogers LR, Cho E, Kempin S et al. Cerebral infarction from nonbacterial thrombotic endocarditis. Am J Med 1987;83:746–758. 88. Gonzalez Quintela A, Candela MJ, Vidal C et al. Nonbacterial thrombotic endocarditis in cancer patients. Acta Cardiol 1991;46:1–9. 89. Min KW, Gyorkey F, Sato C. Mucin-producing adenocarcinomas and nonbacterial thrombotic endocarditis: pathogenic role of tumor mucin. Cancer 1980;45:2374–2382. 90. MacDonald RA, Robbins SL. The significance of nonbacterial thrombotic endocarditis: an autopsy and clinical study of 78 cases. Ann Intern Med 1957;46:255–273. 91. Ojeda VJ, Frost F, Mastaglia FL. Non-bacterial thrombotic endocarditis associated with malignant disease: a clinicopathological study of 16 cases. Med J Aust 1985;142:629–631. 92. Biller J, Challa VR, Toole JF et al. Nonbacterial thrombotic endocarditis: a neurologic perspective of clinicopathologic correlations of 99 patients. Arch Neurol 1982;39:95–98. 93. Le Ber I, Auzou P, Derumeaux G et al. Cerebrovascular complication in nonbacterial thrombotic endocarditis: value of cardiac transesophageal ultrasonography. Presse Med 1997;26:756–758. 94. Vassallo R, Remstein ED, Parisi JE et al. Multiple cerebral infarctions from nonbacterial thrombotic endocarditis mimicking cerebral vasculitis. Mayo Clin Proc 1999;74:798–802. 95. Amico L, Caplan LR, Thomas C. Cerebrovascular complications of mucinous cancers. Neurology 1989;39:522–526. 96. Towfighi J, Simmonds MA, Davidson EA. Mucin and fat emboli in mucinous adenocarcinomas. Arch Path Lab Med 1983;107:646–649. 97. Saver JL. Emerging risk factors for stroke: patent foramen ovale, proximal aortic atherosclerosis, antiphospholipid antibodies, and activated protein C resistance. J Stroke Cerebrovasc Dis 1997;6:167–172. 98. Shapiro B, Jude J, Goudemand I et al. Lupus anticoagulant: a clinical and laboratory study of 100 cases. Clin Lab Haematol 1988;10:41–51. 99. Schved JF, Dupuy-Fona C, Biron C et al. A prospective epidemiological study on the occurrence of antiphospholipid antibody: The Montpelier Antiphospholipid (MAP) Study. Haemostasis 1994;24:175–182. 100. Genvresse I, Buttgereit F, Spath-Schwalbe E et al. Arterial thrombosis associated with anticardiolipin and anti-beta 2-glycoprotein I antibodies in patients with non-Hodgkin’s lymphoma: a report of two cases. Eur J Haematol 2000;65:344–347. 101. Donner M, Bekassy NA, Garcwicz S et al. Cerebral infarction in a girl who developed anticardiolipin syndrome after acute lymphoblastic leukemia. Pediatr Hematol Oncol 1992;9:377–379. 102. Mouas H, Lortholary O, Eclache V et al. Antiphospholipid syndrome during acute monocytic leukaemia. Eur J Haematol 1994;53: 59–60. 103. Ruffatti A, Aversa S, Del Ross T et al. Antiphospholipid antibody syndrome associated with ovarian cancer: anew paraneoplastic syndrome? J Rheumatol 1994;21:2162–2163. 104. Olalla JI, Ortin M, Hermida G et al. Disappearance of lupus anticoagulant after allogenic bone marrow transplantation. Bone Marrow Transplant 1999;23:83–85. 105. Sun NC, McAfee WM, Hum GJ et al. Hemostatic abnormalities in malignancy: a prospective study of 108 patients. Part I. Coagulation studies. Am J Clin Pathol 1979;71:10–16. 106. Levi M, ten Cate H. Disseminated intravascular coagulation. N Engl J Med 1999;341:586–592. 107. Rosen PJ. Bleeding problems in the cancer patient. Hematol Oncol Clin North Am 1992;6:1315–1328. 108. Re G, Lanzarini C, Visani G et al. Latent acute promyelocytic leukaemia in a case of ischaemic stroke underlines the importance of prompt diagnostic confirmation prior to acute care. Eur J Emerg Med 1995;2:102–104. 109. Goldberg MA, Ginsburg D, Mayer RJ et al. Is heparin administration necessary during induction chemotherapy for patients with acute promyelocytic leukemia? Blood 1987;69:187–191. 110. Wada H, Gabazza E, Nakasaki T et al. Diagnosis of disseminated intravascular coagulation by hemostatic molecular markers. Semin Thromb Hemost 2000;26:17–21. 111. Carey MJ, Rodgers GM. Disseminated intravascular coagulation: clinical and laboratory aspects. Am J Hematol 1998;56:65–73. 112. Levi M, de Jonge E, van der Poll T et al. Novel approaches to the management of disseminated intravascular coagulation. Crit Care Med 2000;28(suppl):S20–S24. 113. Sakuragawa N, Hasegawa H, Maki M et al. Clinical evaluation of low-molecular-weight–heparin (FR-860) on disseminated intravascular coagulation (DIC): a multicenter co-operative, double-blind trial in comparison with heparin. Thromb Res 1993;72:475–500.

234

Part V / Indirect Complications of Cancer

114. Ablin AR. Supportive care for children with cancer: Guidelines of the Childrens Cancer Study Group: managing the problem of hyperleukocytosis in acute leukemia. Am J Pediatr Hematol Oncol 1984;6:287–290. 115. Avvisati G, ten Cate JW, Buller HR et al. Tranexamic acid for control of hemorrhage in acute promyelocytic leukaemia. Lancet 1989;2:122–124. 116. Lichtman MA, Rowe JM. Hyperleukotic leukemias: rheological, clinical, and therapeutic considerations. Blood 1982;60:279–283. 117. Nelson SC, Bruggers CS, Kurtzberg J et al. Management of hyperleukosis with hydration, urinary alkalinization, and allopurinol: are cranial irradiation and invasive cytoreduction necessary? Am J Pediatr Hematol Oncol 1993;15: 351–355. 118. Deichmann M, Helmke B, Bock M et al. Massive lethal cerebral bleeding in a patient with melanoma without intracranial metastasis. Clin Oncol 1998;10:272–273. 119. Packer RJ, Rorke LB, Lange BJ et al. Cerebrovascular accidents in children with cancer. Pediatrics 1985;76:194–201. 120. Dutcher JP, Schiffer CA, Aisner J et al. Incidence of throbocytopenia and serious hemorrhage among patients with solid tumors. Cancer 1984;53:557–562. 121. Kim HD, Boggs DR. A syndrome resembling idiopathic thrombocytopenic purpura in 10 patients with diverse forms of cancer. Am J Med 1979;67:371–377. 122. Kaden BR, Rosse WF, Hauch TW. Immune thrombocytopenia in lymphoproliferative diseases. Blood 1979;53:545–551. 123. Eldor A. Thrombotic thrombocytopenic purpura: diagnosis, pathogenesis, and modern therapy. Baillieres Clin Haematol 1998;11: 475–495. 124. Gordon LI, Kwaan HC. Thrombotic microangiopathy manifesting as thrombotic thrombocytopenic purpura/hemolytic uremic syndrome in the cancer patient. Semin Thromb Hemost 1999;25:217–221. 125. Horn M, Steiner T, Gunther T et al. Reversible cerebral MRI findings in acute microangiopathic hemolytic anemia. Nervenarzt 1996;67:502–505. 126. Guzzini F, Conti A, Esposito F. Simultaneous ischemic and hemorrhagic lesions of the brain detected by CT scan in a patient with thrombotic thrombocytopenic purpura. Haematolojica 1998;83:280. 127. Chaturvedi S, Ansell J, Recht L. Should cerebral ischemic events in cancer patients be considered a manifestation of hypercoagulability? Stroke 1994;25:1215–1218. 128. Cestari DM, Weine DM, Panageas KS et al. Stroke in patients with cancer: incidence and etiology. Neurology 2004;62:2025–2030. 129. Domenicucci M, Ramieri A, Ciapetta P et al. Nontraumatic acute spinal subdural hematoma; report of five cases and review of the literature. J Neurosurg 1999;9:65–73. 130. Howard SC, Gajjar A, Riberio RC et al. Safety of lumbar puncture for children with acute lymphoblastic leukemia and thrombocytopenia. JAMA 2000;284:2222–2224. 131. Omura M, Aida N, Sakido K et al. Large intracranial vessel occlusive vasculopathy after radiation therapy in children: clinical features and usefulness of magnetic resonance imaging. Int J Radiat Oncol Biol Phys 1997;38:241–249. 132. Grill J, Couanet D, Cappelli C et al. Radiation-induced cerebral vasculopathy in children with neurofibromatosis and optic pathway glioma. Ann Neurol 1999;45:393–396. 133. Bernstein M, Lumley M, Davidson G et al. Intracranial arterial occlusion associated with high-activity iodine-125 brachytherapy for glioblastoma. J Neurooncol 1993;17:253–260. 134. Fouladi M, Langston J, Mulhern R et al. Silent lacunar lesions detected by magnetic resonance imaging of children with brain tumors: a late sequela of therapy. J Clin Oncol 2000;18:824–831. 135. Casey AT, Marsh HT, Uttley D. Intracranial aneurysm formation following radiotherapy. Br J Neurosurg 1993;7:575–579. 136. Epstein MA, Packer RJ, Rorke LB et al. Vascular malformation with radiation vasculopathy after treatment of chiasmatic/hypothalamic glioma. Cancer 1992;70:887–893. 137. Poussaint TY, Siffert J, Barnes PD et al. Hemorrhagic vasculopathy after treatment of central nervous system neoplasia in childhood: diagnosis and follow-up. Am J Neuroradiol 1995;16:693–699. 138. Larson JJ, Ball WS, Bove KE et al. Formation of intracerebral cavernous malformations after radiation treatment for central nervous system neoplasia in children. J Neurosurg 1998;88:51–56. 139. Cheng SW, Wu LL, Ting AC et al. Irradiation-induced extracranial carotid stenosis in patients with head and neck malignancies. Am J Surg 1999;178:323–328. 140. Dubec JJ, Munk PL, Tsang V et al. Carotid artery stenosis in patients who have undergone radiation therapy for head and neck malignancy. Br J Radiol 1998;71:872–875. 141. Murros KE, Toole J. The effect of radiation on carotid arteries. Arch Neurol 1989;46:449–455. 142. Andros G, Schneider PA, Harris RW et al. Management of arterial occlusive disease following radiation therapy. Cardiovasc Surg 1996;4:135–142. 143. Atkinson JLD, Sundt TM, Dale AJD et al. Radiation-associated atheromatous disease of the cervical carotid artery: report of seven cases and review of the literature. Neurosurgery 1989;24:171–178. 144. Carmody BJ, Arora S, Avena R et al. Accelerated carotid artery disease after high-dose head and neck radiotherapy: is there a role for routine carotid duplex surveillance? J Vasc Surg 1999;30:1045–1051. 145. Kashyap VS, Moore WS, Quinones-Baldrich WJ. Carotid artery repair for radiation-associated atherosclerosis is a safe and durable procedure. J Vasc Surg 1999;29:97–99 146. Al-Mubarak N, Roubin GS, Iyer SS et al. Carotid stenting for severe radiation-induced extracranal carotid artery occlusive disease. J Endovasc Ther 2000;7:36–40. 147. McCready RA, Hyde GL, Bivins BA et al: Radiation-induced arterial injuries. Surgery 1983;93:306–312. 148. Lang NP, Wait GM, Read RR. Cardio-cerebrovascular complications from Nd:YAG laser treatment of lung cancer. Am J Surg 1991;162:629–632.

Chapter 14 / Cerebrovascular Complications of Cancer

235

149. Sessa CN, Morasch MD, Berguer R et al. Carotid resection and replacement with autogenous arterial graft during operation for neck malignancy. Ann Vasc Surg 1998;12:229–235. 150. Brisman MH, Sen C, Catalano P. Results of surgery for head and neck tumors that involve the carotid artery at the skull base. J Neurosurg 1997;86:787–792. 151. Adams GL, Madison M, Remley K et al. Preoperative permanent balloon occlusion of internal carotid artery in patients with advanced head and neck squamous cell carcinoma. Laryngoscope 1999;109:460–466. 152. Lorberboym M, Pandit N, Machac J et al. Brain perfusion imaging during preoperative temporary balloon occlusion of the internal carotid artery. J Nucl Med 1996;37:415–419. 153. Leonetti JP, Reichman OH, Silberman SJ et al. Venous infarction following translabyrinthine access to the cerebellopontine angle. Am J Otol 1994;15:723–727. 154. Keiper GL, Sherman JD, Tomsick TA et al. Dural sinus thrombosis and pseudotumor cerebri: unexpected complications of suboccitpital craniotomy and translabyrinthine craniectomy. J Neurosurg 1999;91:192–197. 155. Bejjani GK, Sekhar LN, Yost AM et al. Vasospasm after cranial base tumor resection: pathogenesis, diagnosis, and therapy. Surg Neurol 1999;52:571–583. 156. Ulmer S, Braga TA, Barker II FG et al. Clinical and radiographic features of peritumoral infarction following resection of glioblastoma. Neurology 2006;67:1668–1670. 157. Smith JS, Cha S, Mayo MC et al. Serial diffusion–weighted magnetic resonance imaging in cases of glioma: distinguishing tumor recurrence from postresection injury. J Neurosurg 2005;103:428–438. 158. Catsman-Berrevoets CE, van Breemen M, van Veelen ML et al. Supratentorial arterial ischemic stroke following cerebellar tumor resection in two children. Pediatr Neurosurg 2005;41:206–211. 159. Kranz R, Gliemroth J, Gaebel C et al. Atypical delayed intracranial haematoma following stereotactic biopsy of a right parietal anaplastic oligodendroglioma. Clin Neurol Neurosurg 2003;105:188–192. 160. Mari G, Marassi A, Di Carlo V. Protein C: a new plasma protein related to post-operative hypercoagulability. Ital J Surg Sci 1984;14:9–12. 161. Ayala C, Blackwell KE. Protein C deficiency in microvascular head and neck reconstruction. Laryngoscope 1999;109:259–265. 162. Caporale A, Tirindelli MC, Aurello P et al. Evaluation of post-operative blood coagulation changes in elderly patients undergoing major surgery. Ital J Surg Sci 1989;19:51–56. 163. Chow KL, Gobin YP, Cloughesy T et al. Prognostic factors in recurrent glioblastoma multiforme and anaplastic astrocytoma treated with selective intra-arterial chemotherapy. Am J Neuroradiol 2000;21:471–478. 164. Larner JM, Phillips CD, Dion JE et al. A phase 1–2 trial of superselective carboplatin, low-dose infusional 5-fluorouracil and concurrent radiation for high-grade gliomas. Am J Clin Oncol 1995;18:1–7. 165. Kallmes DF, Evans AJ, Kaptain GJ et al. Hemorrhagic complications in embolization of a meningioma: case report and review of the literature. Neuroradiology 1997;39:877–880. 166. Moore TB, Chow VJ, Ferry D et al. Intracardiac right-to-left shunting and the risk of stroke during bone marrow infusion. J Clin Oncol 1997;15:2510–2517. 167. Falanga A. Mechanisms of hypercoagulation in malignancy and during chemotherapy. Haemostasis 1998;28(suppl 3):50–60. 168. Lee AY, Levine MN. The thrombophilic state induced by therapeutic agents in the cancer patient. Semin Thromb Hemost 1999;25: 137–145. 169. Nordstrom B, Strang P. Microangiopathic hemolytic anemias (MAHA) in cancer: a case report and review. Anticancer Res 1993;13:1845–1849. 170. Schultz A, Sitzler G, Scheid C et al. A case of thrombotic thrombocytopenic purpura in an adult treated with vincristine. Ann Hematol 1999;78:39–42. 171. Feinberg WM, Swenson MR. Cerebrovascular complications of l–asparaginase therapy. Neurology 1988;38:127–133. 172. Bushman JE, Palmierir D, Whinna HC et al. Insight into the mechanism of asparaginase-induced depletion of antithrombin III in treatment of childhood acute lymphoblastic leukemia. Leuk Res 2000;24:559–565. 173. Gugliotta L, Mazzucconi MG, Leone G et al. Incidence of thrombotic complications in adult patients with acute lymphoblastic leukaemia receiving l–aspariginase during induction therapy: a retrospective study. Eur J Haematol 1992;49:63–66. 174. Matsushita K, Kuriyama Y, Sawada T et al. Cerebral infarction associated with protein C deficiency. Stroke 1992;23:108–111. 175. Serrano-Castro PJ, Guardado-Santervas P, Olivares-Romero J. Ischemic stroke following cisplatin and 5-fluorouracil therapy: a transcranial Doppler study. Eur Neurol 2000;44:63–64. 176. Gamble GE, Tyrrell P. Acute stroke following cisplatin therapy. Clin Oncol 1998;10:274–275. 177. Taher A, Shamsseddine A, Saghir N et al. Acquired protein C deficiency following cisplatinum–navelbine administration for locally advanced breast cancer: case report. Eur J Gynaecol Oncol 1999;20:323–324. 178. Chan AT, Yeo W, Leung WT et al. Thromboembolic events with paclitaxel. Clin Oncol 1996;8:133. 179. Fisher B, Constantino JP, Wickerham et al. Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 1998;90:1371–1388. 180. Barbui T, Finazzi G, Grassi A et al. Thrombosis in cancer patients treated with hematopoietic growth factor: a meta-analysis. Thromb Haemost 1996;75:368–371. 181. Kabbinavar FF, Schulz J, McCleod M et al. Addition of bevacizumab to bolus fluorouracil and leucovorin in first-line metastatic colorectal cancer: results of a randomized phase II trial. J Clin Oncol 2005;23:3697–3705. 182. Vrendenburgh JJ, Desjardins A, Herndon II J et al. Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 2007;13:1253–1259. 183. Pai VB, Nahata MC. Cardiotoxicity of chemotherapeutic agents: incidence, treatment and prevention. Drug Saf 2000;22:263–302.

236

Part V / Indirect Complications of Cancer

184. Mathur SC, Friedman HD, Kende AI et al. Cryptic Mucor infection leading to massive cerebral infarction at initiation of antileukemic chemotherapy. Ann Hematol 1999;78:241–245. 185. Mammen EF. The haematological manifestations of sepsis. J Antimicrob Chemother 1998;41(suppl A):17–24. 186. Schiff D, DeAngelis LM. Therapy of venous thromboembolism in patients with brain metastasis. Cancer 1994;73:493–498. 187. Paquette RL, Tran L, Landaw EM. Thrombotic microangiopathy following allogeneic bone marrow transplantation is associated with intensive graft-versus-host disease prophylaxis. Bone Marrow Transplant 1998;22:351–357. 188. Pettitt AR, Clark RE. Thrombotic microangiopathy following bone marrow transplantation. Bone Marrow Transplant 1994; 14: 495–504. 189. Colosimo M, McCarthy N, Jayasinghe R et al. Diagnosis and management of subdural haematoma complicating bone marrow transplantation. Bone Marrow Transplant 2000;25:549–552. 190. Graus F, Saiz A, Sierra J et al. Neurologic complications of autologous and allogeneic bone marrow transplantation in patients with leukemia: a comparative study. Neurology 1996;46:1004–1009. 191. Coplin WM, Cochran MS, Levine SR et al. Stroke after bone marrow transplantation: frequency, aetiology and outcome. Brain 2001;124:1043–1051. 192. Bleggi-Torres LF, Werner B, Gasparetto EL et al. Intracranial hemorrhage following bone marrow transplantation: an autopsy study of 58 patients. Bone Marrow Transplant 2002;29:29–32. 193. Patchell RA, White CL, Clark AW et al. Nonbacterial thrombotic endocarditis in bone marrow transplant patients. Cancer 1985;55: 631–635. 194. Ma M, Barnes G, Pulliam J et al. CNS angiitis in graft vs. host disease. Neurology 2005;59:1994–1997.

15

Paraneoplastic Syndromes of the Nervous System Myrna R. Rosenfeld

MD, PHD,

and Josep Dalmau,

MD, PHD

CONTENTS Introduction Diagnosis of Paraneoplastic Neurologic Disorders: General Concepts Paraneoplastic Limbic Encephalitis Paraneoplastic Cerebellar Degeneration Paraneoplastic Encephalomyelitis Paraneoplastic Sensory Neuronopathy Paraneoplastic Opsoclonus–Myoclonus Paraneoplastic Stiff-Man Syndrome Motor Neuron Disease Paraneoplastic Sensorimotor Neuropathy Paraneoplastic Vasculitis of Nerve and Muscle Sensorimotor Peripheral Neuropathy Associated with Malignant Monoclonal Gammopathies Paraneoplastic Neuromyotonia Paraneoplastic Autonomic Dysfunction Lambert–Eaton Myasthenic Syndrome Polymyositis and Dermatomyositis Acute Necrotizing Myopathy Conclusion References

Summary Paraneoplastic disorders may affect any part of the central and peripheral nervous system and can be confused with other neurologic complications of cancer. Because paraneoplastic syndromes usually develop before the presence of a cancer is known, it is important for the clinician to have a high index of suspicion that a patient’s symptoms may be paraneoplastic. For many of these syndromes there is increasing evidence that early treatment of the tumor and immunotherapy may result in neurologic stabilization or improvement. Key Words: paraneoplastic syndrome, cerebellar degeneration, neuropathy, cancers

1. INTRODUCTION In patients with cancer the development of neurological symptoms usually represents metastatic involvement of the nervous system or complications secondary to coagulopathy, infection, metabolic and nutritional deficits, and toxic effects of cancer therapy (1). From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

237

238

Part V / Indirect Complications of Cancer

Table 1 Paraneoplastic Syndromes of the Nervous System Central nervous system Limbic encephalitis Cerebellar degeneration Encephalomyelitis Sensory neuronopathy Opsoclonus–myoclonus Stiff-man syndrome Motor neuron syndrome and motor neuronopathy Necrotizing myelopathy (*) Peripheral nervous system Chronic sensorimotor neuropathy Acute sensorimotor neuropathy (Guillain–Barré, Plexitis) (*) Vasculitis of the nerve and muscle Neuropathy associated with malignant monoclonal gammopathies Neuromyotonia Autonomic neuropathy Lambert–Eaton myasthenic syndrome Myasthenia gravis (*) Polymyositis/dermatomyositis Acute necrotizing myopathy (*) Not included in this chapter; reviewed in reference (150).

A neurologic disorder is defined as paraneoplastic when none of the above causes are detected or when specific cancer-related immunological mechanisms are involved. Paraneoplastic neurologic disorders are important for several reasons. They may affect any part of the central and peripheral nervous system (Table 1) and mimic any other neurologic complications of cancer. The paraneoplastic disorder usually develops before the presence of a cancer is known and its prompt recognition may help to uncover the neoplasm. The neurologic symptoms are often severe and can result in the patient’s death. Furthermore, there is increasing evidence that early intervention with oncologic and immunotherapy may result in stabilization or improvement of neurologic symptoms (2,3). The potential for improvement depends on the type of syndrome and associated immune responses; for example, several recently identified immune mediated encephalitides are treatment-responsive and patients fully recover if appropriately diagnosed and treated (4).

1.1. Frequency It is estimated that fewer than 1% of patients with cancer develop clinically symptomatic paraneoplastic neurologic syndromes, but the frequency varies with the type of cancer. For example, while 10–30% of patients with plasma cell dyscrasias or thymoma develop paraneoplastic neurologic symptoms, far fewer than 1% of patients with breast or ovarian cancer develop paraneoplastic neurologic disorders.

1.2. Pathogenesis Most paraneoplastic neurologic disorders appear to be mediated by immunological mechanisms (5). The occurrence of serum and CSF antibodies that target proteins selectively expressed by the tumor and nervous system has suggested a mechanism whereby the tumor expression of neuronal proteins triggers an antitumor immune response that cross-reacts with the nervous system. In general, the efficacy of the immune response against the tumor is limited or not sustained enough to control its growth, but the effects on the nervous system are prominent (6). In some antibody-associated paraneoplastic syndromes, the accompanying cytotoxic T-cell mechanisms appear to be the main pathogenic effectors (7,8). The autoantigens of these disorders are usually intracellular, and the

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

239

associated antibodies are detectable in serum and CSF of the patients. Although the pathogenic role of these antibodies is unclear, their detection forms part of the diagnostic tests that confirm the paraneoplastic nature of the neurologic disorder and direct the search of the underlying tumor. In contrast, there are antibody-associated paraneoplastic syndromes in which the humoral immune response plays a dominant pathogenic role. Until recently it was believed that these syndromes only affected the neuromuscular junction and peripheral nerves (i.e., myasthenic syndromes, neuromyotonia), but there is increasing evidence that several paraneoplastic encephalitides are also mediated by antibodies (9). The autoantigens of these disorders are usually in the cell membrane, and the related syndromes are more responsive to immunotherapies than the syndromes associated with antibodies and cytotoxic T-cell responses against intracellular antigens (4). Several other paraneoplastic disorders, including inflammatory neuropathies or myopathies, may associate with infiltrates of mononuclear cells, cytotoxic T-cells, or deposits of IgG and complement in the involved nerve or muscle (10). These disorders are likely immune-mediated, but the target autoantigens have not been identified, and there are no specific circulating specific antibodies that can be used as surrogate markers of paraneoplasia. In addition to immune-mediated mechanisms, there are other paraneoplastic causes of neurologic dysfunction, including competition between the tumor and nervous system for substrates (i.e., glucose), and inappropriate secretion of hormones or cytokines by tumor cells (i.e., antidiuretic hormone), among others. This chapter focuses on the syndromes that occur in association with immune-mediated mechanisms.

2. DIAGNOSIS OF PARANEOPLASTIC NEUROLOGIC DISORDERS: GENERAL CONCEPTS There are four clinical features that complicate the diagnosis of most paraneoplastic neurologic disorders: (i) the frequent presentation of neurologic symptoms before the diagnosis of the cancer; (ii) the occurrence of similar syndromes without a cancer association; (iii) the absence of paraneoplastic antibodies in a variable proportion of patients; and (iv) the small size of the associated tumors, which are usually difficult to demonstrate at the time of neurologic symptom presentation (11).

2.1. Presentation of Symptoms The majority of paraneoplastic neurologic syndromes develop rapidly in a matter of days or weeks and subsequently stabilize or continue to progress until the patient’s death. Patients who develop syndromes of the central nervous system often describe prodromic gastrointestinal or upper respiratory tract symptoms resembling a viral illness, which is followed by the neurologic symptoms. Some syndromes such as the cerebellar degeneration associated with anti-Yo antibodies or the encephalitis associated with anti-NMDA receptor antibodies appear to be more frequently preceded by these prodromic symptoms than others (12). Most patients with syndromes affecting the central nervous system have CSF abnormalities such as pleocytosis, increased protein concentration, oligoclonal bands and elevated IgG index that suggest an inflammatory or immune mediated process. In about 60% of patients, the paraneoplastic neurologic disorder develops before the presence of a tumor is known. There are a few exceptions such as the paraneoplastic retinopathy that affects patients with melanoma (“melanoma-associated retinopathy”); patients with this disorder usually have a history of metastatic melanoma (13).

2.2. Syndromes Similar to Paraneoplastic Disorders May Occur Without a Cancer Association Although some syndromes are more frequently associated with cancer than others, all paraneoplastic syndromes have a counterpart that may occur without a cancer association. The disorders that are frequent paraneoplastic manifestations of cancer include limbic encephalitis, opsoclonus–myoclonus, subacute cerebellar degeneration of the elderly, Lambert–Eaton myasthenic syndrome (LEMS), encephalomyelitis, cancer-associated retinopathy, and melanoma-associated retinopathy (11).

2.3. Paraneoplastic Antibodies The term paraneoplastic antibodies refers to a group of antibodies that when detected almost always indicates that the neurologic disorder is a paraneoplastic manifestation of cancer (Table 2). Other antibodies that can occur with or without a cancer association, such as the antibodies against voltage-gated calcium channels (VGCC) in

240

Part V / Indirect Complications of Cancer

Table 2 Antibodies Associated with Paraneoplastic Neurologic Disorders Antibody

Associated Cancer

Syndrome

Antibodies That Are Markers of Paraneoplasia Anti-Hu Anti-Yo Anti-Ri Anti-Tr Anti-CV2/CRMP5

SCLC, other Gynecological, breast Breast, gynecological, SCLC Hodgkin’s lymphoma SCLC, thymoma, other

Anti-Ma proteins(*)

Testicular germ cell tumors and other neoplasms Breast, SCLC Teratoma of the ovary or mediastinum

Anti-amphiphysin Anti-NMDAR

Encephalomyelitis, sensory neuronopathy Cerebellar degeneration Cerebellar ataxia, opsoclonus, brasintem encephalitis Cerebellar degeneration Encephalomyelitis, striatal encephalitis (chorea), cerebellar degeneration, uveitis, peripheral neuropathy Limbic, diencephalic (hypothalamic) and upper brainstem encephalitis; rarely cerebellar degeneration Stiff-man syndrome, encephalomyelitis Encephalitis with predominant psychiatric symptoms, autonomic dysfunction, dyskinesias, hypoventilation. Infrequently presents as classical limbic encephalitis

Antibodies That Are Not Markers of Paraneoplasia (associated with the indicated neurologic syndromes whether they are paraneoplastic or not) Anti-VGKC

Thymoma, SCLC

Anti-VGCC Anti-AChR (muscle) Anti-AChR (neuronal)

SCLC Thymoma SCLC and other

Neuromyotonia, limbic encephalitis, Morvan’s syndrome LEMS, cerebellar degeneration Myasthenia gravis Autonomic neuropathy

(*) Antibodies limited to Ma2 (also called anti-Ta antibodies) usually associate with limbic and brainstem encephalitis and germ-cell tumors. Antibodies directed at Ma1 and Ma2 usually associate with brainstem encephalitis, cerebellar degeneration and several types of cancer (lung, breast, ovary, among others). Antibodies in italics denote that are directed against cell membrane antigens. The other antibodies are directed against intracellular antigens. NMDAR: N-methyl-d-aspartate receptor; VGKC: voltage-gated potassium channels; VGCC: voltage-gated calcium channels; AChR: acetylcholine receptor.

LEMS, acetylcholine receptors in myasthenia gravis, or voltage-gated potassium channels (VGKC) in neuromyotonia or limbic encephalitis are not considered paraneoplastic antibodies (14). Paraneoplastic antibodies are identified in approximately 60% of patients with syndromes of the central nervous system. Similar syndromes may occur without antibodies, and yet be a paraneoplastic manifestation of cancer. The probability of detecting an antibody characteristic of a paraneoplastic syndrome depends on the type of cancer association. For example, older women with a paraneoplastic cerebellar degeneration associated with breast or gynecologic cancers almost always harbor anti-Yo antibodies, but if another tumor is involved other paraneoplastic antibodies (or no antibodies) will be identified. Most paraneoplastic syndromes of the peripheral nervous system do not associate with paraneoplastic antibodies. Only a few sensorimotor neuropathies develop in association with anti-CV2/CRMP5 antibodies or anti-Hu, the latter suggesting a neuronopathy caused by dorsal root ganglia dysfunction, rather than a peripheral neuropathy (15,16).

2.4. Tumors Associated with Paraneoplastic Disorders At the time of neurologic symptom presentation the tumors of most patients are usually small and confined to a single organ or to the regional lymph nodes. Therefore, the demonstration of the tumor can be difficult. The combined use of CT and PET imaging uncovers occult tumors in approximately 80% of cases; the remaining 20% require close follow-up with repeat studies (17,18). Eventually, 90% of all tumors associated with paraneoplastic disorders are diagnosed within the first year of neurologic symptom development. The detection of specific paraneoplastic antibodies often helps in the selection of diagnostic tests, directing the search of the tumor to a few organs. For some syndromes, such as anti-Ma2 encephalitis associated with

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

241

germ cell neoplasms of the testis, CT and PET studies can be negative, but ultrasound often reveals the neoplasm or abnormalities that associate with the neoplasm (i.e., microcalcifications). In men younger than 50 with antiMa2-associated encephalitis but without a clinically detectable tumor, a study showed that in all instances a microscopic tumor was present at orchiectomy (19). Until recently the prevailing concept was that the tumors involved in antibody-associated paraneoplastic neurologic syndromes were always malignant, either carcinomas or less frequently lymphomas and leukemias. This concept has changed after the discovery that benign tumors such as mature cystic teratomas (dermoid cysts) can also trigger paraneoplastic neurologic syndromes by mechanisms similar to those occurring in malignant neoplasms (20). It is believed that the expression of mature or immature nervous tissue by the teratomas, along with other unknown factors, triggers the immune response against the antigens (NMDA receptors) also expressed in the nervous system. The optimal studies to reveal ovarian teratomas are CT and pelvic ultrasound; PET studies are often negative. The use of vaginal ultrasound often clarifies whether a “benign-appearing cyst” contains solid tissue, suggesting a teratoma (12). Patients with peripheral neuropathies of unclear etiology should undergo a skeletal survey to determine the presence of osteolytic or sclerotic lesions associated with myeloma, and blood and urine tests for monoclonal M proteins that may associate with plasma cell dyscrasias or lymphomas (21).

3. PARANEOPLASTIC LIMBIC ENCEPHALITIS This disorder is characterized by mood disturbances, insomnia, seizures, and short-term memory loss (22). The outcome is variable. The disorder may stabilize, leaving the patient with severe anterograde memory deficits; it may progress, causing profound deficits of behavior and cognition, leading to frank dementia; or it may resolve. In two-thirds of the patients, the CSF shows mild pleocytosis, increased proteins, intrathecal synthesis of IgG, and oligoclonal bands. The typical MRI findings include uni- or bilateral mesial temporal lobe abnormalities that are best seen on fluid attenuated inversion recovery (FLAIR) and T2-weighted images, and sometimes contrast enhance (Fig. 1) (23). However, the MRI may be normal despite evidence of temporal lobe dysfunction demonstrated by EEG or hypermetabolism sometimes detected by PET (9). The EEG may reveal that patients with unexplained low level of consciousness are in status epilepticus.

Fig. 1. MRI findings in limbic encephalitis. Axial (A) and coronal (B) MRI fluid-attenuated inversion recovery (FLAIR) sequences from a patient with anti-Hu associated limbic encephalitis. Note the bilateral medial temporal lobe hyperintensities that are considered characteristic of most cases of limbic encephalitis. Similar findings occur with other immune mediated limbic encephalitis, paraneoplastic or not, and with some viral encephalitis.

242

Part V / Indirect Complications of Cancer

Fig. 2. Antibodies to cell surface antigens in a patient with paraneoplastic limbic encephalitis and carcinoma of the thymus. Rat hippocampal neurons immunolabeled with the patient’s CSF. Note the intense reactivity of patient’s antibodies with the cell surface and dendrites of neurons (antigen unknown). CSF from control subjects does not produce any reactivity (not shown). Immunofluorescence x800. (see Color Plate 3).

Neurological symptoms usually precede the diagnosis of the tumor. The tumor most frequently involved is lung cancer, usually small cell lung cancer (SCLC). Other tumors include germ cell neoplasms of the testis, breast cancer, Hodgkin’s lymphoma, thymoma, and teratomas of the ovary (22). The pathological findings in most paraneoplastic limbic encephalitis include perivascular and interstitial inflammatory infiltrates, neuronal loss, and microglial proliferation that predominate in the limbic system (i.e., hippocampus, amygdala, hypothalamus, and insular and cingulate cortex). In addition, the majority of patients had variable involvement of other areas of the nervous system, mainly the brainstem (24). Patients with antiNMDAR-associated encephalitis (which may present as typical limbic encephalitis) have prominent microglial proliferation in the limbic system and rare or infrequent inflammatory infiltrates (25). There are several paraneoplastic immune responses that often associate with typical clinical features of limbic encephalitis; they include anti-Hu, anti-Ma2, and anti-NMDAR (Table 2) (2,12,26). Each of these immune responses may also associate with a clinical picture of multifocal encephalitis in which other brain regions are involved in addition to the limbic system (see encephalomyelitis). Other immune responses (i.e., anti-CV2/CRMP5, amphiphysin, or Ri) associate less frequently with a classical syndrome of limbic encephalitis. Detection of antibodies to VGKC has been considered characteristic of nonparaneoplastic limbic encephalitis; however, 15–20% of patients with these antibodies and either limbic encephalitis or Morvan’s syndrome (a combination of peripheral nerve hyperexcitability and encephalopathy) have an underlying tumor, usually thymoma or lung cancer (27–29). A recent study showed that many patients with subacute limbic encephalitis without paraneoplastic or VGKC antibodies harbor antibodies that are not detectable by currently available commercial tests (30). These antibodies appear to target several cell surface autoantigens (not yet characterized) and their detection can occur with or without the presence of an underlying cancer (Fig. 2 and Color Plate 3). Regardless of the identity of the antigens, the detection of antibodies that react with cell surface antigens usually indicates that the disorder will respond to immunotherapy (corticosteroids and IVIg or plasma exchange) (4). The tumors that may associate with these antibodies are thymomas and less frequently, SCLC and Hodgkin’s lymphoma.

4. PARANEOPLASTIC CEREBELLAR DEGENERATION The presenting symptoms of this disorder are dizziness, nausea, blurry or double vision, oscillopsia, and gait difficulties. Associated with these symptoms, or occurring after a few days, the patient develops truncal and limb ataxia, dysarthria, and dysphagia. On examination, patients usually have down-beating nystagmus (31). This clinical picture is similar for most types of paraneoplastic cerebellar degeneration, irrespective of the type of cancer or antibody association, although the course of the disease may be different depending

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

243

upon the associated immune response (32). In general, neurologic symptoms precede the tumor diagnosis. The CSF usually shows pleocytosis, increased protein, intrathecal synthesis of IgG and oligoclonal bands. In the early stages of the disease, brain MRI is usually normal, but after several months may show global cerebellar atrophy. There is a strong association between the development of certain antineuronal antibodies and the type of tumor associated with the paraneoplastic cerebellar disorder (Fig. 3 and Color Plate 4). These include SCLC and anti-Hu antibodies (33,34), ovarian and breast cancer and anti-Yo antibodies (35), Hodgkin’s lymphoma and anti-Tr antibodies (36), and breast cancer and anti-Ri antibodies (Table 2) (37,38). Furthermore, the presence of anti-Hu antibodies is usually associated with symptoms indicating involvement of other areas of the nervous system (i.e., encephalomyelitis and sensory neuronopathy) (39). The presence of anti-Ri antibodies is associated with opsoclonus or other abnormalities of ocular motility, including nystagmus, abnormal visual tracking, and abnormal vestibulo-ocular reflexes in 70% of the patients. Symptoms of paraneoplastic cerebellar dysfunction may occur without the presence of antineuronal antibodies. In this case, the tumors more frequently involved are non-Hodgkin’s lymphoma and lung cancer (non-SCLC and SCLC) (33,40). A subset of patients with SCLC without anti-Hu antibodies develops antibodies against VGCC (41). These antibodies are similar to those associated with LEMS and some patients develop symptoms of both cerebellar dysfunction and LEMS. Pathological studies show diffuse loss of Purkinje cells accompanied by degeneration of the dentate and olivary nuclei, and long tracts of the spinal cord. These findings can be associated with mild or prominent lymphocytic infiltrates (42,43). When present, the inflammatory infiltrates usually involve the deep cerebellar nuclei in addition

Fig. 3. Studies in a patient with anti-Yo associated cerebellar degeneration. Panel A corresponds to a normal FDG-PET obtained when the patient had anti-Yo associated cerebellar degeneration for 2 years. Panel B is an FDG-PET obtained 3 years later (5 years after neurologic symptom presentation) showing a hypermetabolic abnormality in the right axillary lymph nodes (arrow). Panel C shows the patient’s anti-Yo antibodies immunolabeling a section of rat cerebellum (note the characteristic Purkinje cell reactivity of the anti-Yo antibodies). Panel D shows that the neoplastic cells from the lymph node (identified by PET) react with the anti-Yo antibodies of the patient (dark cells). (C and D, avidin–biotin–peroxidase, x400.) (see Color Plate 4).

244

Part V / Indirect Complications of Cancer

to the brainstem and other areas of the nervous system, suggesting that the cerebellum is the main target of a multifocal encephalomyelitis. Treatment of the tumor and immunosuppressants do not usually affect the course of the cerebellar disorder (44). However, neurological improvement can occur in patients with anti-Ri and anti-Tr antibodies (45,46).

5. PARANEOPLASTIC ENCEPHALOMYELITIS This disorder describes patients with cancer who develop multifocal neurological deficits and signs of inflammation involving two or more areas of the nervous system, including brain, cerebellum, brainstem, spinal cord, dorsal root ganglia and autonomic ganglia (42). This gives rise to a mixture of symptoms derived from limbic encephalitis, cerebellar degeneration, brainstem encephalitis, and myelitis along with sensory deficits and autonomic dysfunction. Symptoms of paraneoplastic brainstem encephalitis can include diplopia, dysarthria, dysphagia, internuclear or supranuclear gaze abnormalities, facial numbness, and subacute hearing loss. The spinal cord symptoms usually result from an inflammatory degeneration of the lower motor neurons (47). Symptoms of autonomic dysfunction may include gastrointestinal paresis and pseudo-obstruction, orthostatic hypotension and cardiac arrhythmias among others (see section 14). Several paraneoplastic immunities can associate with encephalomyelitis or multifocal encephalitis (Table 2), as described in the following sections.

5.1. Anti-Hu Patients with anti-Hu associated encephalomyelitis frequently develop sensory neuronopathy secondary to dorsal root ganglia involvement; the tumor most frequently involved is SCLC (34,48–50). Other antineuronal antibodies that occur less frequently (sometimes in combination with anti-Hu) include, anti-CV2/CRMP5 and anti-amphiphysin (51,52).

5.2. Anti-CV2/CRMP5 The encephalomyelitis associated with anti-CV2/CRMP5 antibodies may affect any of the areas indicated above along with the striatum (chorea), uvea (uveitis), and peripheral nerves, resulting in a mixed axonal demyelinating sensorimotor neuropathy (15,53,54). The tumors more frequently associated are SCLC and thymoma. Paraneoplastic disorders associated with anti-CV2/CRMP5 or anti-Hu are in general poorly responsive to treatment of the tumor or immunotherapies, including plasma exchange, IVIg, or cyclophosphamide. However, successful treatment of the tumor and prompt immunotherapy directed at the cytotoxic T-cell response (e.g., IVIg and cyclophosphamide) may result in stabilization or partial improvement of symptoms (3).

5.3. Anti-Ma Proteins The encephalitis of patients with immunity to Ma proteins is more restricted to limbic system, hypothalamus, brainstem, and cerebellum than the encephalomyelitis associated with other antibodies (55). These patients may present with classical limbic encephalitis or severe hypokinesis and hypophonesis (pseudomutism) with relative preservation of cognitive functions (56). Supranuclear ocular paresis is common, usually affecting vertical gaze more than horizontal gaze. The tumors more frequently associated are germ cell neoplasms of the testis and nonSCLC (19). About 35% of patients with anti-Ma2 associated encephalitis respond to treatment; the neurological improvement is usually partial and predominates in young men with successfully treated testicular neoplasms (2).

5.4. Anti-NMDAR The encephalitis of patients with anti-NMDAR antibodies affects young women with ovarian teratomas and usually presents with prominent psychiatric symptoms, including behavioral and personality change, agitation, delusional thoughts, and sometimes catatonia (12,20). Patients usually develop autonomic dysfunction and several types of dyskinesias, and often require ventilatory support. Epileptic seizures are frequent, but in many instances the EEG only shows general slowing while the patient has facial twitching, limb automatisms, or rhythmic

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

245

Fig. 4. Studies in a patient with paraneoplastic anti-NMDAR encephalitis and ovarian teratoma. Panel A shows the CSF reactivity of a patient with anti-NMDAR antibodies with a sagittal section of rat hippocampus; the immunolabeling is mainly concentrated in the inner aspect of the molecular layer adjacent to the dentate gyrus (arrow). Panel B shows that the antibody reactivity is with the cell surface and dendrites of neurons (the picture corresponds to a culture of rat hippocampal neurons immunolabeled with the patient’s antibodies). Panel C shows that the teratoma of the patient contains immature neurons; these are demonstrated with MAP2 labeling, a specific marker of neurons and dendrites. (A, avidin–biotin–peroxidase, x50; B, immunofluorescence x800; C, avidin–biotin–peroxidase, x400.) (see Color Plate 5).

contraction of the abdominal muscles. Different from other paraneoplastic encephalitis, this disorder appears to be mediated by antibodies (Fig. 4 and Color Plate 5) and often responds to removal of the tumor and IgGdepleting therapies (plasma exchange or IVIg). Patients who do not respond to these treatments may respond to cyclophosphamide (12).

6. PARANEOPLASTIC SENSORY NEURONOPATHY This disorder is characterized by progressive sensory loss involving lower and upper extremities, trunk, and face. The sensory deficits are frequently accompanied by painful paresthesias and dysesthesias. This and the frequent asymmetric presentation of symptoms may lead to the diagnosis of radiculopathy or multineuropathy (57,58). At presentation, vibration and joint position sensations may be more affected than nociceptive sensation. The sensory loss causes disorganization of movement resulting in sensory ataxia and pseudoathetoid movements. Some patients develop sensorineural hearing loss. Paraneoplastic sensory neuronopathy frequently develops in association with encephalomyelitis and autonomic dysfunction (see paraneoplastic encephalomyelitis, section 5). In more than 80% of patients, the sensory neuronopathy precedes the diagnosis of the tumor, usually a SCLC (39). Nerve conduction studies demonstrate low amplitude or absent sensory nerve action potentials. Motor nerve and F-wave studies are usually normal, with no signs of denervation unless there is involvement of the spinal motor neurons in the setting of encephalomyelitis. Some patients develop motor conduction abnormalities as a result of a mixed axonal and demyelinating neuropathy that accompanies the degeneration of dorsal root ganglia neurons (59,60). Pathological studies show an inflammatory, probably immune-mediated degeneration of the neurons of the dorsal root ganglia and equivalent ganglia of cranial nerves (e.g., Gasserian ganglia) (61). Other findings include

246

Part V / Indirect Complications of Cancer

atrophy of the posterior nerve roots, axonal degeneration, and secondary degeneration of the posterior columns of the spinal cord. Mild inflammatory infiltrates can be found in peripheral nerves and sometimes muscle (57). Paraneoplastic sensory neuronopathy rarely responds to immunotherapies, including plasma exchange, intravenous IgG, and immunosuppressants (34). In some patients the use of steroids may result in partial improvement of symptoms (62). Efforts should be directed towards prompt identification and treatment of the tumor.

7. PARANEOPLASTIC OPSOCLONUS–MYOCLONUS Paraneoplastic opsoclonus–myoclonus (POM) usually affects children younger than 4 years of age (median age, 18 months) and often presents with staggering and falling along with body jerks, ataxia, refusal to walk or sit, opsoclonus, irritability, and sleep problems that may contribute to episodes of rage (63,64). Nearly 50% of children with POM have neuroblastoma, and about 2% of children with this tumor develop opsoclonus. Neurologic symptoms may precede or develop after the diagnosis of neuroblastoma. POM frequently responds to treatment of the tumor and immunotherapy that may includes corticosteroids, adrenocorticotrophic hormone, IVIg, and rituximab (65). The sleep disturbances and episodes of rage often respond to trazodone; however, residual psychomotor deficits and behavioral and sleep disturbances are common (63,64,66). Patients with POM have a better tumor prognosis than patients without paraneoplastic symptoms. In adults, POM develops in association with truncal ataxia resulting in gait difficulty and frequent falls. In more than half of patients, POM precedes the diagnosis of the tumor, usually a SCLC (67). Patients with breast cancer may harbor anti-Ri antibodies (see paraneoplastic cerebellar degeneration) (37). The clinical course of paraneoplastic opsoclonus is worse than that of idiopathic opsoclonus. Paraneoplastic opsoclonus may respond to immunotherapy or IVIg, but symptom improvement depends on successful treatment of the tumor (68). If the tumor is not treated, neurologic symptoms often progress to a severe encephalopathy resulting in the patient’s death. In addition to treatment of the tumor and IVIg, there are reported clinical responses to depletion of serum IgG using protein-A columns, clonazepam, valproic acid, and thiamine (69).

8. PARANEOPLASTIC STIFF-MAN SYNDROME This disorder is characterized by fluctuating rigidity of the axial musculature with superimposed spasms. Rigidity primarily affects the lower trunk and legs, but it can extend to the shoulders, upper limbs, neck, and less frequently muscles of the face (70). Symptoms may be limited to one extremity (stiff-limb syndrome) (71). Spasms are often precipitated by voluntary movement, emotional upset, and auditory and somesthetic stimuli. Typically, the rigidity disappears during sleep or following local or general anesthesia, suggesting dysfunction at the spinal or supraspinal level. Electrophysiologic studies show continuous activity of motor units in the stiffened muscles that considerably improve after treatment with diazepam. The disorder can occur as a paraneoplastic manifestation of cancer (usually breast or SCLC) or, more frequently, without a cancer association (72). The serum and CSF of patients with paraneoplastic stiff-man syndrome may contain antibodies to amphiphysin (73,74). In patients without cancer the major autoantigen is glutamic acid decarboxylase (GAD) and 70% of these patients develop type I diabetes and other autoimmune diseases (75). These two autoantigens are expressed in the spinal cord inhibitory interneurons that control motor neuron activity and co-secrete both GABA and glycine. Histopathological abnormalities found in stiff-man syndrome include mild perivascular lymphocyte infiltration and loss of motor neurons and interneurons in the anterior horn of the spinal cord (76–78). Improvement can be obtained with IVIg and GABA-enhancing drugs, such as diazepam, clonazepam, gabapentin, or baclofen, but sustained improvement usually requires treatment of the tumor and steroids (73,79). In addition to stiff-man syndrome, paraneoplastic rigidity and spasms can occur in patients with extensive encephalomyelitis or focal myelitis (80). Some of these patients harbor anti-Ri antibodies (81).

9. MOTOR NEURON DISEASE The occurrence of motor neuron disease (MND) as a paraneoplastic syndrome is controversial. Two systematic reviews of the literature (82,83) concluded that there was not an increased incidence of cancer among patients with amyotrophic lateral sclerosis (ALS). However, there is evidence to suggest that MND does rarely occur as

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

247

Fig. 5. Severe neurogenic muscle atrophy in a patient with SCLC and anti-Hu-associated myelitis. The initial neurologic symptom of this patient was flaccid motor weakness selectively involving the upper extremities and neck extension. After 8 weeks he was unable to move the upper extremities. These symptoms associated with fasciculations and loss of reflexes in the arms. Cranial nerves and strength and reflexes in the lower extremities were normal. The picture demonstrates widespread atrophy of the muscles of the neck and shoulder. (see Color Plate 6).

a paraneoplastic phenomenon. This includes reports of patients with typical MND (i.e., without unusual features such as sensory loss) who have improvement following treatment of their malignancy (84). There are three situations in which a PND may manifest with symptoms of MND. First, some patients with encephalomyelitis and sensory neuronopathy may develop symptoms of MND due to predominant involvement of the spinal cord (myelitis) (Fig. 5 and Color Plate 6). Second, patients with lymphoma may develop upper and lower motor neuron dysfunction, which progresses in a fashion typical of ALS (85,86). Third, an association between primarily lateral sclerosis and breast cancer was suggested in one study; no paraneoplastic markers were identified and the development of both disorders could have been coincidental (47).

10. PARANEOPLASTIC SENSORIMOTOR NEUROPATHY Many patients with advanced malignancy develop a peripheral neuropathy, which is usually mild, with little impact on quality of life (87). The cause of these neuropathies is multifactorial, including metabolic, nutritional deficits, and toxic effects of chemotherapy, which may include cisplatin, paclitaxel, docetaxel, vinca alkaloids, and thalidomide among others (88). There is a group of sensorimotor neuropathies that develop before or by the time the malignancy is discovered and have a major impact on quality of life (89). Symptoms may present in a subacute or acute fashion and are usually progressive. Pathological studies usually show axonal degeneration with frequent inflammatory infiltrates, and some patients have predominant demyelinating findings. The latter are more likely to respond to corticosteroids and IVIg than the axonal neuropathies.

248

Part V / Indirect Complications of Cancer

11. PARANEOPLASTIC VASCULITIS OF NERVE AND MUSCLE This disorder is a nonsystemic vasculitic neuropathy, which involves nerve, muscle, or both. The tumors more frequently involved are SCLC and lymphoma (90). The disorder usually affects older men and the neuropathy is subacute and progressive. Patients develop a painful symmetric or asymmetric sensorimotor polyneuropathy, and less frequently a multiple mononeuropathy. The erythrocyte sedimentation rate is usually elevated and the CSF shows a high protein content. Electrophysiological studies show axonal degeneration equally involving motor and sensory nerves. Nerve biopsy studies show intraneural and perivascular inflammatory infiltrates, usually without necrotizing vasculitis. The inflammatory infiltrates are mainly composed of CD8+ T cells (91,92). Paraneoplastic vasculitis of nerve and muscle often responds to treatment of the tumor, corticosteroids and cyclophosphamide (90).

12. SENSORIMOTOR PERIPHERAL NEUROPATHY ASSOCIATED WITH MALIGNANT MONOCLONAL GAMMOPATHIES The malignancies associated with monoclonal gammopathies or M proteins include multiple myeloma and sclerotic myeloma, which are typically associated with IgG or IgA M proteins, and Waldenström’s macroglobulinemia, B-cell lymphoma, and chronic B-cell lymphocytic leukemia, which are associated with IgM M proteins.

12.1. Multiple Myeloma One-third of patients with multiple myeloma have electrophysiologic signs of peripheral neuropathy, but only 5–10% develop clinical symptoms (93). Neurologic symptoms usually precede the diagnosis of myeloma. Patients may develop a mild sensorimotor axonal neuropathy, a pure sensory neuropathy, or a subacute monophasic or relapsing and remitting neuropathy with evidence of demyelination on electrophysiologic and morphologic studies (94). Amyloid deposition occurs in 20–40% of myeloma patients with peripheral neuropathy (21,95). In these patients, symptoms are similar to those with distal axonal sensorimotor neuropathy, but frequently include atypical features, such as carpal tunnel syndrome, a clinical picture of multiple mononeuropathy, and autonomic dysfunction. There is no specific treatment for these neuropathies. Treatment of the myeloma rarely improves the neuropathy.

12.2. Osteosclerotic Myeloma This is an unusual form of myeloma characterized by a single or multiple plasmacytomas that manifest as sclerotic bone lesions. These lesions involve ribs, vertebrae, pelvic bones, and proximal long bones, and usually spare skull and distal extremities (96). More than 50% of patients with sclerotic myeloma develop a peripheral neuropathy, which resembles a chronic demyelinating polyradiculoneuropathy with motor predominance and high CSF protein content. All or some features of the POEMS syndrome (polyneuropathy, organomegaly, endocrinopathy, M component, and skin changes) may be present. Focal treatment of the sclerotic lesions with resection or radiation therapy, or with prednisone plus/minus melphalan often results in neurologic improvement (97). Patients with rapidly progressive neuropathy may respond to peripheral blood stem cell transplant (98).

12.3. Waldenström’s Macroglobulinemia Subtle symptoms of peripheral neuropathy occur in 45% of patients with this condition, and in approximately 10% the deficits are severe (99). The neuropathy may result from antibody activity of the IgM M-protein against myelin-associated glycoprotein (MAG), sulphatide, or various gangliosides (100). The neuropathy associated with IgM anti-MAG is characterized by progressive distal sensorimotor deficits, with predominant involvement of vibration sense, sometimes with postural tremor, pseudoathetosis, and progressive gait dysfunction. Electrophysiologic studies demonstrate slow conduction velocities and prolonged distal motor and sensory latencies, compatible with a demyelinating neuropathy. Pathology studies show widening between lamellae of myelin sheaths due to intercalation of anti-MAG antibodies (101). The neuropathy associated with anti-sulphatide antibodies is predominantly axonal. Treatment should be directed at the Waldenström’s macroglobulinemia. Patients with demyelinating neuropathy and IgM anti-MAG M proteins may respond to plasma exchange, IVIg, or rituximab (102). However, most

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

249

patients require aggressive treatment with chemotherapeutic agents such as chlorambucil, cyclophosphamide, or fludarabine (103,104).

13. PARANEOPLASTIC NEUROMYOTONIA This disorder, also called Isaacs’ syndrome, is characterized by spontaneous and continuous muscle fiber activity of peripheral nerve origin (105). Symptoms include muscle cramps, weakness, and sometimes excessive sweating. The involved muscles show undulating myokymia and may be hypertrophic (106). EMG shows fibrillation, fasciculation, and doublet, triplet or multiplet single unit discharges that have a high intraburst frequency. This abnormal activity continues during sleep and general anesthesia, is abolished by curare, and may be reduced or abolished by peripheral nerve block (105,107). Neuromyotonia is associated with antibodies to VGKC, which increase the release of quanta of acetylcholine and prolong the action potential (108,109). The tumors more frequently involved are lung cancer and thymoma; patients with thymoma may have additional symptoms of myasthenia gravis (110). Plasma exchange has been shown to be more effective that IVIg (111). Symptomatic treatments include anticonvulsants that block sodium channels such as diphenylhydantoin and carbamazepine (107). Approximately 25% of patients with neuromyotonia develop central nervous system dysfunction characterized by changes in mood, irritability, hallucinations, delusions and sleep dysfunction. This disorder is called Morvan’s syndrome, and it can also occur as a paraneoplastic manifestations of cancer. The treatments are similar to those indicated above.

14. PARANEOPLASTIC AUTONOMIC DYSFUNCTION This disorder usually develops in association with other paraneoplastic syndromes, such as encephalomyelitis or LEMS. Symptoms often precede the detection of the tumor, usually a SCLC. The autonomic dysfunction may result from adrenergic or cholinergic nerve dysfunction at the pre- or, most frequently, postganglionic level (112,113). There are three disorders that can be life-threatening: (i) esophageal and gastrointestinal dysmotility with intestinal pseudoobstruction, (ii) cardiac dysrhythmias, and (iii) orthostatic hypotension. Other accompanying symptoms may include dry mouth, erectile and sphincter dysfunction. Because autonomic dysfunction can be the presentation of encephalomyelitis, testing for anti-Hu antibodies should be considered in some patients (114,115). A subgroup of patients with autonomic neuropathy develop antibodies to the ganglionic acetylcholine receptor (116); these antibodies may occur in patients with or without cancer, but their detection suggests that symptoms may improve with immunotherapy (117).

15. LAMBERT–EATON MYASTHENIC SYNDROME LEMS is a disorder of the neuromuscular junction characterized by impaired acetylcholine release from the presynaptic motor terminal (118). Symptoms include fatigue, leg weakness, muscle aches, and vague paresthesias. Dry mouth and other symptoms of autonomic dysfunction are common (119). Cranial nerve involvement tends to be mild and transient, usually described as transient diplopia. Neurologic examination shows proximal weakness in the legs more than the arms and depressed reflexes, sometimes accompanied by eyelid ptosis and sluggishly reactive pupils (120). After brief muscle contraction, reflexes may potentiate. Similarly, after brief exercise strength may improve. LEMS is associated with cancer in 50–70% of patients, most commonly SCLC. Neurologic symptoms typically precede or coincide with the diagnosis of the tumor. LEMS can also occur in conjunction with other paraneoplastic syndromes, such as paraneoplastic cerebellar degeneration and encephalomyelitis (33,121,122). Routine nerve conduction studies show small amplitude compound muscle action potentials (CMAP) (123). At slow rates of repetitive nerve stimulation (2–5 Hz), a decremental response of greater than 10% is seen. At fast rates (20 Hz or greater) or after maximal voluntary muscle contraction, facilitation occurs and there is an incremental response of at least 100% (124,125).

250

Part V / Indirect Complications of Cancer

LEMS results from an immunological attack against the presynaptic VGCC, interfering with the release of acetylcholine vesicles. The transfer of IgG from patients with LEMS into mice reproduces the clinical and electrophysiologic features of the disease (126,127). The detection of antibodies to P/Q-type VGCC is used as a serologic test for LEMS (128). Therapies for LEMS include treatment of the associated cancer, medication to increase the release of acetylcholine, and immunotherapy (125,129). The majority of patients have neurological improvement with combined treatment of the cancer and therapy specific for LEMS. Drugs that increase the release of acetylcholine include 3,4-diaminopyridine that results in improvement of 80% of patients, and the combination of pyridostigmine and guanidine (130). Immunomodulation with IVIg or plasma exchange is usually effective but the benefits are short lasting (131–133). Long-term immunosuppression with prednisone alone or combined with azathioprine or cyclosporine should be considered in patients who remain symptomatic after controlling the tumor or using 3,4-diaminopyridine (133,134).

16. POLYMYOSITIS AND DERMATOMYOSITIS Polymyositis (PM) and dermatomyositis (DM) are inflammatory disorders of the muscle and are likely autoimmune in nature. The association of PM/DM with cancer is rare, and the existence of paraneoplastic PM controversial. However, a number of studies support the view that patients with DM are at higher risk for cancer (135–140). In women the most common tumors are ovarian and breast cancer, and in men, lung and gastrointestinal cancer. An association with cancer has not been demonstrated in childhood DM. Patients with PM/DM typically present with proximal muscle weakness of subacute onset, elevated levels of serum creatine kinase, and electromyographic evidence of myopathy. Neck flexors and pharyngeal and respiratory muscles are commonly involved; their dysfunction may result in aspiration and hypoventilation and contribute to death. Reflexes and sensory exam are normal. In DM the classic skin manifestations include purplish discoloration of the eyelids (heliotrope rash) with edema, and erythematous, scaly lesions over the knuckles. Necrotic skin ulcerations and pruritus are more frequently associated with paraneoplastic dermatomyositis (141). Clinical, electromyographic, and pathological findings of PM/DM are similar in patients with and without cancer. In some patients, the serum creatine kinase levels are normal. Patients with interstitial lung disease may harbor antibodies to histidyl-tRNA synthetase (anti-Jo-1) (142). There are no specific markers indicative of the paraneoplastic origin of DM. Different immune mechanisms appear to be involved in PM and DM. While PM results from cellmediated cytotoxic mechanisms, DM results from a humoral immune-mediated vasculopathy leading to ischemia, muscle fiber necrosis, and perifascicular atrophy. Some patients develop cutaneous involvement without myopathy (amyopathic dermatomyositis) in association with cancer; MRI studies may show subclinical muscle involvement (143). Corticosteroids, IVIg, and other immunotherapies (azathioprine, cyclophosphamide, methotrexate, cyclosporine) have been used successfully in paraneoplastic and non-paraneoplastic PM/DM (144,145). Patients with graft versus host disease (GVHD) may develop symptoms of PM. Some of these patients also have skin abnormalities secondary to GVHD which may resemble those associated with DM. Treatment with cyclosporine or tacrolimus in association with corticosteroids often results in improvement (146,147).

17. ACUTE NECROTIZING MYOPATHY Patients with this disorder develop muscle pain and proximal weakness, which are associated with high levels of serum creatine kinase. The disorder evolves rapidly to generalized weakness, which involves pharyngeal and respiratory muscles, often leading to death in a few weeks. Several types of tumors are involved, including SCLC, cancer of the gastrointestinal tract (stomach, colon, gall bladder, pancreas), breast, kidney and prostate (148). Muscle biopsy shows prominent necrosis with little or absent inflammation. There is alkaline phosphatase staining of connective tissue, and some muscle fibers are immunolabeled by antibodies to terminal components

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

251

of the complement cascade (C5–C9). Treatment of the tumor may result in neurological improvement (148). In cancer patients, the differential diagnosis should include chemotherapy and cytokine-induced (IL-2, interferon-) rhabdomyolysis (149).

18. CONCLUSION Paraneoplastic syndromes of the nervous system comprise an extensive group of disorders caused by diverse pathogenic mechanisms, among which the immunologic mechanisms appear to be predominant. The main concern of the clinician should be to rule out other etiologies and to uncover the presence of the associated neoplasm. The best approach to treat immune-mediated paraneoplastic syndromes is to treatment the tumor and to use immunotherapy at early stages of the neurologic disorder. In patients with stable neurologic symptoms, care should be based upon symptomatic treatment and physical therapy.

REFERENCES 1. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A.Davis, 1995. 2. Dalmau J, Graus F, Villarejo A et al. Clinical analysis of anti-Ma2-associated encephalitis. Brain 2004; 127:1831–1844. 3. Vernino S, O’Neill BP, Marks RS et al. Immunomodulatory treatment trial for paraneoplastic neurological disorders. Neuro-oncol 2004; 6:55–62. 4. Bataller L, Kleopa KA, Wu GF et al. Autoimmune limbic encephalitis in 39 patients: immunophenotypes and outcomes. J Neurol Neurosurg Psychiatry, 2007;78:381–5. 5. Bataller L, Dalmau JO. Paraneoplastic disorders of the central nervous system: update on diagnostic criteria and treatment. Semin Neurol 2004; 24:461–471. 6. Mathew RM, Cohen AB, Galetta SL et al. Paraneoplastic cerebellar degeneration: Yo-expressing tumor revealed after a 5-year follow-up with FDG-PET. J Neurol Sci 2006; 250:153–155. 7. Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med 2003; 349:1543–1554. 8. Rousseau A, Benyahia B, Dalmau J et al. T-cell response to Hu-D peptides in patients with anti-Hu syndrome. J Neuro-oncol 2005; 71:231–236. 9. Ances BM, Vitaliani R, Taylor RA et al. Treatment-responsive limbic encephalitis identified by neuropil antibodies: MRI and PET correlates. Brain 2005; 128:1764–1777. 10. Dalakas MC. Polymyositis, dermatomyositis and inclusion-body myositis. N Engl J Med 1991; 325:1487–1498. 11. Graus F, Delattre JY, Antoine JC et al. Recommended diagnostic criteria for paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry 2004; 75:1135–1140. 12. Dalmau J, Tüzün E, Wu H–Y et al. Paraneoplastic anti-NMDA receptor encephalitis associated with ovarian teratoma. Ann Neurol, 2007;6:25–36. 13. Keltner JL, Thirkill CE, Yip PT. Clinical and immunologic characteristics of melanoma-associated retinopathy syndrome: eleven new cases and a review of 51 previously published cases. J Neuroophthalmol 2001; 21:173–187. 14. Dalmau J, Rosenfeld MR. Paraneoplastic neurologic syndromes. In: Kasper DL, Braunwald E, Fauci AS et al. (eds.). Harrison’s Principles of Internal Medicine. New York: McGraw–Hill, 2005:571–575. 15. Antoine JC, Honnorat J, Camdessanche JP et al. Paraneoplastic anti-CV2 antibodies react with peripheral nerve and are associated with a mixed axonal and demyelinating peripheral neuropathy. Ann Neurol 2001; 49:214–221. 16. Kuntzer T, Antoine JC, Steck AJ. Clinical features and pathophysiological basis of sensory neuronopathies (ganglionopathies). Muscle Nerve 2004; 30:255–268. 17. Younes-Mhenni S, Janier MF, Cinotti L et al. FDG-PET improves tumour detection in patients with paraneoplastic neurological syndromes. Brain 2004; 127:2331–2338. 18. Rees JH, Hain SF, Johnson MR et al. The role of [18F]fluoro-2-deoxyglucose-PET scanning in the diagnosis of paraneoplastic neurological disorders. Brain 2001; 124:2223–2231. 19. Mathew RM, Vandenberghe R, Garcia-Merino A et al. Orchiectomy for suspected microscopic tumor in patients with anti-Ma2associated encephalitis. Neurology, 2007;68:900–5. 20. Vitaliani R, Mason W, Ances B et al. Paraneoplastic encephalitis, psychiatric symptoms, and hypoventilation in ovarian teratoma. Ann Neurol 2005; 58:594–604. 21. Ropper AH, Gorson KC. Neuropathies associated with paraproteinemia. N Engl J Med 1998; 338:1601–1607. 22. Gultekin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain 2000; 123:1481–1494. 23. Lawn ND, Westmoreland BF, Kiely MJ et al. Clinical, magnetic resonance imaging, and electroencephalographic findings in paraneoplastic limbic encephalitis. Mayo Clin Proc 2003; 78:1363–1368. 24. Corsellis JA, Goldberg GJ, Norton AR. “Limbic encephalitis” and its association with carcinoma. Brain 1968; 91:481–496. 25. Stein-Wexler R, Wootton-Gorges SL, Greco CM et al. Paraneoplastic limbic encephalitis in a teenage girl with an immature ovarian teratoma. Pediatr Radiol 2005; 35:694–697. 26. Alamowitch S, Graus F, Uchuya M et al. Limbic encephalitis and small cell lung cancer: clinical and immunological features. Brain 1997; 120 :923–928.

252

Part V / Indirect Complications of Cancer

27. Pozo-Rosich P, Clover L, Saiz A et al. Voltage-gated potassium channel antibodies in limbic encephalitis. Ann Neurol 2003; 54: 530–533. 28. Buckley C, Oger J, Clover L et al. Potassium channel antibodies in two patients with reversible limbic encephalitis. Ann Neurol 2001; 50:73–78. 29. Liguori R, Vincent A, Clover L et al. Morvan’s syndrome: peripheral and central nervous system and cardiac involvement with antibodies to voltage-gated potassium channels. Brain 2001; 124:2417–2426. 30. Bataller L, Dalmau J. Clinical and immunological diversity of limbic encephalitis: a model for paraneoplastic disorders. Hematol Oncol Clin North Am 2006; 20:1319–1335. 31. Peterson K, Rosenblum MK, Kotanides H et al. Paraneoplastic cerebellar degeneration. I. A clinical analysis of 55 anti-Yo antibodypositive patients. Neurology 1992; 42:1931–1937. 32. Shams’ili S, Grefkens J, De Leeuw B et al. Paraneoplastic cerebellar degeneration associated with antineuronal antibodies: analysis of 50 patients. Brain 2003; 126:1409–1418. 33. Mason WP, Graus F, Lang B et al. Small cell lung cancer, paraneoplastic cerebellar degeneration and the Lambert–Eaton myasthenic syndrome. Brain 1997; 120:1279–1300. 34. Graus F, Keime-Guibert F, Rene R et al. Anti-Hu–associated paraneoplastic encephalomyelitis: analysis of 200 patients. Brain 2001; 124:1138–1148. 35. Rojas-Marcos I, Rousseau A, Keime-Guibert F et al. Spectrum of paraneoplastic neurologic disorders in women with breast and gynecologic cancer. Medicine (Baltimore) 2003; 82:216–223. 36. Bernal F, Shams’ili S, Rojas I, et al. Anti-Tr antibodies as markers of paraneoplastic cerebellar degeneration and Hodgkin’s disease. Neurology 2003; 60:230–234. 37. Luque FA, Furneaux HM, Ferziger R et al. Anti-Ri: an antibody associated with paraneoplastic opsoclonus and breast cancer. Ann Neurol 1991; 29:241–251. 38. Sutton IJ, Barnett MH, Watson JD et al. Paraneoplastic brainstem encephalitis and anti-Ri antibodies. J Neurol 2002; 249:1597–1598. 39. Dalmau J, Graus F, Rosenblum MK et al. Anti-Hu–associated paraneoplastic encephalomyelitis/sensory neuronopathy: a clinical study of 71 patients. Medicine (Baltimore) 1992; 71:59–72. 40. Sabater L, Bataller L, Carpentier AF et al. Protein kinase C gamma autoimmunity in paraneoplastic cerebellar degeneration and non-small cell lung cancer. J Neurol Neurosurg Psychiatry 2006; 77:1359–1362. 41. Graus F, Lang B, Pozo-Rosich P et al. P/Q-type calcium-channel antibodies in paraneoplastic cerebellar degeneration with lung cancer. Neurology 2002; 59:764–766. 42. Henson RA, Hoffman HL, Urich H. Encephalomyelitis with carcinoma. Brain 1965; 88:449–464. 43. Verschuuren J, Chuang L, Rosenblum MK et al. Inflammatory infiltrates and complete absence of Purkinje cells in anti-Yo–associated paraneoplastic cerebellar degeneration. Acta Neuropathol (Berl) 1996; 91:519–525. 44. Rojas I, Graus F, Keime-Guibert F et al. Long-term clinical outcome of paraneoplastic cerebellar degeneration and anti-Yo antibodies. Neurology 2000; 55:713–715. 45. Dropcho EJ, Kline LB, Riser J. Antineuronal (anti-Ri) antibodies in a patient with steroid-responsive opsoclonus–myoclonus. Neurology 1993; 43:207–211. 46. Graus F, Dalmau J, Valldeoriola F et al. Immunological characterization of a neuronal antibody (anti-Tr) associated with paraneoplastic cerebellar degeneration and Hodgkin’s disease. J Neuroimmunol 1997; 74:55–61. 47. Forsyth PA, Dalmau J, Graus F et al. Motor neuron syndromes in cancer patients. Ann Neurol 1997; 41:722–730. 48. Dalmau J, Graus F, Rosenblum MK et al. Anti-Hu–associated paraneoplastic encephalomyelitis/sensory neuronopathy: a clinical study of 71 patients. Medicine 1992; 71:59–72. 49. Sillevis Smitt P, Grefkens J, De Leeuw B et al. Survival and outcome in 73 anti-Hu positive patients with paraneoplastic encephalomyelitis/sensory neuronopathy. J Neurol 2002; 249:745–753. 50. Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and oncologic profiles of patients seropositive for type 1 antineuronal nuclear autoantibodies. Neurology 1998; 50:652–657. 51. Honnorat J, Antoine JC, Derrington E et al. Antibodies to a subpopulation of glial cells and a 66 kDa developmental protein in patients with paraneoplastic neurological syndromes. J Neurol Neurosurg Psychiatry 1996; 61:270–278. 52. Dropcho EJ. Antiamphiphysin antibodies with small cell lung carcinoma and paraneoplastic encephalomyelitis. Ann Neurol 1996; 39:659–667. 53. Yu Z, Kryzer TJ, Griesmann GE et al. CRMP-5 neuronal autoantibody: marker of lung cancer and thymoma-related autoimmunity. Ann Neurol 2001; 49:146–154. 54. Vernino S, Tuite P, Adler CH et al. Paraneoplastic chorea associated with CRMP-5 neuronal antibody and lung carcinoma. Ann Neurol 2002; 51:625–630. 55. Rosenfeld MR, Eichen JG, Wade DF et al. Molecular and clinical diversity in paraneoplastic immunity to Ma proteins. Ann Neurol 2001; 50:339–348. 56. Matsumoto L, Yamamoto T, Higashihara M et al. Severe hypokinesis caused by paraneoplastic anti-Ma2 encephalitis associated with bilateral intratubular germ-cell neoplasia. Mov Disord, 2007;22:728–31. 57. Denny-Brown D. Primary sensory neuropathy with muscular changes associated with carcinoma. J Neurol Neurosurg Psychiatry 1948; 11:73–87. 58. Chalk CH, Windebank AJ, Kimmel DW et al. The distinctive clinical features of paraneoplastic sensory neuronopathy. Can J Neurol Sci 1992; 19:346–351. 59. Oh SJ, Gurtekin Y, Dropcho EJ et al. Anti-Hu antibody neuropathy: a clinical, electrophysiological, and pathological study. Clin Neurophysiol 2005; 116:28–34.

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

253

60. Camdessanche JP, Antoine JC, Honnorat J et al. Paraneoplastic peripheral neuropathy associated with anti-Hu antibodies: a clinical and electrophysiological study of 20 patients. Brain 2002; 125:166–175. 61. Wanschitz J, Hainfellner JA, Kristoferitsch W et al. Ganglionitis in paraneoplastic subacute sensory neuronopathy: a morphologic study. Neurology 1997; 49:1156–1159. 62. Oh SJ, Dropcho EJ, Claussen GC. Anti-Hu–associated paraneoplastic sensory neuropathy responding to early aggressive immunotherapy: report of two cases and review of literature. Muscle Nerve 1997; 20:1576–1582. 63. Russo C, Cohn SL, Petruzzi MJ et al. Long-term neurologic outcome in children with opsoclonus–myoclonus associated with neuroblastoma: a report from the Pediatric Oncology Group. Med Pediatr Oncol 1997; 28:284–288. 64. Pranzatelli MR, Tate ED, Dukart WS et al. Sleep disturbance and rage attacks in opsoclonus–myoclonus syndrome: response to trazodone. J Pediatr 2005; 147:372–378. 65. Pranzatelli MR, Tate ED, Travelstead AL et al. Immunologic and clinical responses to rituximab in a child with opsoclonus–myoclonus syndrome. Pediatrics 2005; 115:e115–e119. 66. Koh PS, Raffensperger JG, Berry S et al. Long-term outcome in children with opsoclonus–myoclonus and ataxia and coincident neuroblastoma. J Pediatr 1994; 125:712–716. 67. Anderson NE, Budde-Steffen C, Rosenblum MK et al. Opsoclonus, myoclonus, ataxia, and encephalopathy in adults with cancer: a distinct paraneoplastic syndrome. Medicine (Baltimore) 1988; 67:100–109. 68. Bataller L, Graus F, Saiz A et al. Clinical outcome in adult onset idiopathic or paraneoplastic opsoclonus–myoclonus. Brain 2001; 124:437–443. 69. Batchelor TT, Platten M, Hochberg FH. Immunoadsorption therapy for paraneoplastic syndromes. J Neurooncol 1998; 40:131–136. 70. Moersch F, Woltman H. Progressive fluctuating muscular rigidity and spasm (“stiff-man syndrome”): report of a case and some observations in 13 other cases. Mayo Clin Proc 1956; 31:421–427. 71. Silverman IE. Paraneoplastic stiff-limb syndrome. J Neurol Neurosurg Psychiatry 1999; 67:126–127. 72. Brown P, Marsden CD. The stiff-man and stiff-man-plus syndromes. J Neurol 1999; 246:648–652. 73. Folli F, Solimena M, Cofiell R et al. Autoantibodies to a 128-kd synaptic protein in three women with the stiff-man syndrome and breast cancer. N Engl J Med 1993; 328:546–551. 74. De Camilli P, Thomas A, Cofiell R et al. The synaptic vesicle–associated protein amphiphysin is the 128-kD autoantigen of stiff-man syndrome with breast cancer. J Expt Med 1993; 178:2219–2223. 75. Solimena M, Folli F, Aparisi R et al. Autoantibodies to GABA-ergic neurons and pancreatic beta cells in stiff-man syndrome. N Engl J Med 1990; 322:1555–1560. 76. Ishizawa K, Komori T, Okayama K et al. Large motor neuron involvement in Stiff-man syndrome: a qualitative and quantitative study. Acta Neuropathol (Berl) 1999; 97:63–70. 77. Saiz A, Minguez A, Graus F et al. Stiff-man syndrome with vacuolar degeneration of anterior horn motor neurons. J Neurol 1999; 246:858–860. 78. Warich-Kirches M, Von Bossanyi P, Treuheit T et al. Stiff-man syndrome: possible autoimmune etiology targeted against GABA-ergic cells. Clin Neuropathol 1997; 16:214–219. 79. Dalakas MC, Fujii M, Li M et al. High-dose intravenous immune globulin for stiff-person syndrome. N Engl J Med 2001; 345: 1870–1876. 80. Ishii A, Hayashi A, Ohkoshi N et al. Progressive encephalomyelitis with rigidity associated with anti-amphiphysin antibodies. J Neurol Neurosurg Psychiatry 2004; 75:661–662. 81. Casado JL, Gil-Peralta A, Graus F et al. Anti-Ri antibodies associated with opsoclonus and progressive encephalomyelitis with rigidity. Neurology 1994; 44:1521–1522. 82. Barron KD, Rodichok LD. Cancer and disorders of motor neurons. In: Rowland LP, (ed.). Human Motor Neuron Diseases. New York: Raven Press, 1982:267–272. 83. Rosenfeld MR, Posner JB. Paraneoplastic motor neuron disease. In: Rowland LP, (ed.). Advances in Neurology, Volume 56: Amyotrophic Lateral Sclerosis and Other Motor Neuron Diseases. New York: Raven Press, 1991:445–459. 84. Forman D, Rae-Grant AD, Matchett SC et al. A reversible cause of hypercapnic respiratory failure: lower motor neuronopathy associated with renal cell carcinoma. Chest 1999; 115:899–901. 85. Younger DS, Rowland LP, Latov N et al. Motor neuron disease and amyotrophic lateral sclerosis: relation of high CSF protein content to paraproteinemia and clinical syndromes. Neurology 1990; 40:595–599. 86. Younger DS, Rowland LP, Latov N et al. Lymphoma, motor neuron diseases, and amyotrophic lateral sclerosis. Ann Neurol 1991; 29:78–86. 87. Croft PB, Urich H, Wilkinson M. Peripheral neuropathy of sensorimotor type associated with malignant disease. Brain 1967; 90:31–66. 88. Singhal S, Mehta J, Desikan R et al. Antitumor activity of thalidomide in refractory multiple myeloma. N Engl J Med 1999; 341:1565–1571. 89. Antoine JC, Mosnier JF, Absi L et al. Carcinoma-associated paraneoplastic peripheral neuropathies in patients with and without anti-onconeural antibodies. J Neurol Neurosurg Psychiatry 1999; 67:7–14. 90. Oh SJ. Paraneoplastic vasculitis of the peripheral nervous system. Neurol Clin 1997; 15:849–863. 91. Matsumuro K, Izumo S, Umehara F et al. Paraneoplastic vasculitic neuropathy: immunohistochemical studies on a biopsied nerve and post-mortem examination. J Intern Med 1994; 236:225–230. 92. Vincent D, Dubas F, Hauw JJ et al. Nerve and muscle microvasculitis in peripheral neuropathy: a remote effect of cancer? J Neurol Neurosurg Psychiatry 1986; 49:1007–1010. 93. Walsh JC. The neuropathy of multiple myeloma: an electrophysiological and histological study. Arch Neurol 1971; 25:404–414. 94. Kelly JJ, Jr. Peripheral neuropathies associated with monoclonal proteins: a clinical review. Muscle Nerve 1985; 8:138–150.

254

Part V / Indirect Complications of Cancer

95. Kelly JJ, Jr. The electrodiagnostic findings in peripheral neuropathy associated with monoclonal gammopathy. Muscle Nerve 1983; 6:504–509. 96. Kelly JJ, Jr., Kyle RA, Miles JM et al. Osteosclerotic myeloma and peripheral neuropathy. Neurology 1983; 33:202–210. 97. Dispenzieri A, Kyle RA, Lacy MQ et al. POEMS syndrome: definitions and long-term outcome. Blood 2003; 101:2496–2506. 98. Dispenzieri A, Moreno-Aspitia A, Suarez GA et al. Peripheral blood stem cell transplantation in 16 patients with POEMS syndrome, and a review of the literature. Blood 2004; 104:3400–3407. 99. Levine T, Pestronk A, Florence J et al. Peripheral neuropathies in Waldenstrom’s macroglobulinaemia. J Neurol Neurosurg Psychiatry 2006; 77:224–228. 100. Dimopoulos MA, Panayiotidis P, Moulopoulos LA et al. Waldenstrom’s macroglobulinemia: clinical features, complications, and management. J Clin Oncol 2000; 18:214–226. 101. Vital C, Vallat JM, Deminiere C et al. Peripheral nerve damage during multiple myeloma and Waldenstrom’s macroglobulinemia: an ultrastructural and immunopathologic study. Cancer 1982; 50:1491–1497. 102. Pestronk A, Florence J, Miller T et al. Treatment of IgM antibody–associated polyneuropathies using rituximab. J Neurol Neurosurg Psychiatry 2003; 74:485–489. 103. Latov N. Prognosis of neuropathy with monoclonal gammopathy. Muscle Nerve 2000; 23:150–152. 104. Weide R, Heymanns J, Koppler H. The polyneuropathy associated with Waldenstrom’s macroglobulinaemia can be treated effectively with chemotherapy and the anti-CD20 monoclonal antibody rituximab. Br J Haematol 2000; 109:838–841. 105. Isaacs H. A syndrome of continuous muscle-fibre activity. J Neurol Neurosurg Psychiatry 1961; 24:319–325. 106. Hart IK, Maddison P, Newsom-Davis J et al. Phenotypic variants of autoimmune peripheral nerve hyperexcitability. Brain 2002; 125:1887–1895. 107. Newsom-Davis J, Mills KR. Immunological associations of acquired neuromyotonia (Isaac’s syndrome): report of five cases and literature review. Brain 1993; 116:453–469. 108. Shillito P, Molenaar PC, Vincent A et al. Acquired neuromyotonia: evidence for autoantibodies directed against K+ channels of peripheral nerves. Ann Neurol 1995; 38:714–722. 109. Sinha S, Newsom-Davis J, Mills K et al. Autoimmune aetiology for acquired neuromyotonia (Isaac’s syndrome). Lancet 1991; 338:75–77. 110. Hart IK, Waters C, Vincent A et al. Autoantibodies detected to expressed K+ channels are implicated in neuromyotonia. Ann Neurol 1997; 41:238–246. 111. van den Berg JS, Van Engelen BG, Boerman RH et al. Acquired neuromyotonia: superiority of plasma exchange over high-dose intravenous human immunoglobulin. J Neurol 1999; 246:623–625. 112. Chiappa KH, Young RR. A case of paracarcinomatous pandyautonomia. Neurology 1973; 23:423. 113. Fagius J, Westerberg C-E, Olsson Y. Acute pandysautonomia and severe sensory deficit with poor clinical recovery: a clinical, neurophysiological and pathological case study. J Neurol Neurosurg Psychiatry 1983; 46:725–733. 114. Lennon VA, Sas DF, Busk MF et al. Enteric neuronal autoantibodies in pseudoobstruction with small cell lung carcinoma. Gastroenterology 1991; 100:137–142. 115. Condom E, Vidal A, Rota R et al. Paraneoplastic intestinal pseudo-obstruction associated with high titres of Hu autoantibodies. Virchows Archiv A Pathol Anat 1993; 423:507–511. 116. Vernino S, Adamski J, Kryzer TJ et al. Neuronal nicotinic ACh receptor antibody in subacute autonomic neuropathy and cancer-related syndromes. Neurology 1998; 50:1806–1813. 117. Schroeder C, Vernino S, Birkenfeld AL et al. Plasma exchange for primary autoimmune autonomic failure. N Engl J Med 2005; 353:1585–1590. 118. Elmqvist D, Lambert EH. Detailed analysis of neuromuscular transmission in a patient with the myasthenic syndrome sometimes associated with bronchogenic carcinoma. Mayo Clin Proc 1968; 43:689–713. 119. O’Suilleabhain P, Low PA, Lennon VA. Autonomic dysfunction in the Lambert–Eaton myasthenic syndrome: serologic and clinical correlates. Neurology 1998; 50:88–93. 120. Clark CV, Newsom-Davis J, Sanders MD. Ocular autonomic nerve function in Lambert–Eaton myasthenic syndrome. Eye 1990; 4:473–481. 121. Clouston PD, Saper CB, Arbizu T et al. Paraneoplastic cerebellar degeneration. III. Cerebellar degeneration, cancer, and the Lambert– Eaton myasthenic syndrome. Neurology 1992; 42:1944–1950. 122. Voltz R, Carpentier AF, Rosenfeld MR et al. P/Q-type voltage-gated calcium channel antibodies in patients with paraneoplastic disorders of the central nervous system. Muscle Nerve 1999; 22:119–122. 123. Eaton LM, Lambert EH. Electromyography and electrical stimulation of nerves in diseases of the motor unit: observations on a myasthenic syndrome associated with malignant tumors. JAMA 1957; 163:1117–1124. 124. Jablecki C. Lambert–Eaton myasthenic syndrome. Muscle Nerve 1984; 7:250–257. 125. Sanders DB. Lambert–Eaton myasthenic syndrome: clinical diagnosis, immune- mediated mechanisms, and update on therapies. Ann Neurol 1995; 37:S63–S73. 126. Lang B, Newsom-Davis J, Wray D et al. Autoimmune aetiology for myasthenic (Eaton–Lambert) syndrome. Lancet 1981; 2:224–226. 127. Lang B, Newsom-Davis J, Peers C et al. The effect of myasthenic syndrome antibody on presynaptic calcium channels in the mouse. J Physiol (Lond) 1987; 390:257–270. 128. Motomura M, Lang B, Johnston I et al. Incidence of serum anti-P/O-type and anti-N-type calcium channel autoantibodies in the Lambert–Eaton myasthenic syndrome. J Neurol Sci 1997; 147:35–42. 129. Newsom-Davis J. A treatment algorithm for Lambert–Eaton myasthenic syndrome. Ann N Y Acad Sci 1998; 841:817–822. 130. Oh SJ, Kim DS, Kwon KH et al. Wide spectrum of symptomatic treatment in Lambert–Eaton myasthenic syndrome. Ann N Y Acad Sci 1998; 841:827–831.

Chapter 15 / Paraneoplastic Syndromes of the Nervous System

255

131. Rich MM, Teener JW, Bird SJ. Treatment of Lambert–Eaton syndrome with intravenous immunoglobulin. Muscle and Nerve 1997; 20:614–615. 132. Bain PG, Motomura M, Newsom-Davis J et al. Effects of intravenous immunoglobulin on muscle weakness and calcium-channel autoantibodies in the Lambert–Eaton myasthenic syndrome. Neurology 1996; 47:678–683. 133. Newsom-Davis J. Lambert–Eaton myasthenic syndrome. Rev Neurol (Paris) 2004; 160:177–180. 134. Vedeler CA, Antoine JC, Giometto B et al. Management of paraneoplastic neurological syndromes: report of an EFNS Task Force. Eur J Neurol 2006; 13:682–690. 135. Sigurgeirsson B, Lindelof B, Edhag O et al. Risk of cancer in patients with dermatomyositis or polymyositis: a population-based study. N Engl J Med 1992; 326:363–367. 136. Maoz CR, Langevitz P, Livneh A et al. High incidence of malignancies in patients with dermatomyositis and polymyositis: an 11-year analysis. Semin Arthritis Rheum 1998; 27:319–324. 137. Garcia VE, Gutierrez GJ, Blanco GA. Report of 8 cases of dermatomyositis: does association of this entity and neoplasms exist? Rev Clin Esp 1998; 198:217–220. 138. Maugars YM, Berthelot JM, Abbas AA et al. Long-term prognosis of 69 patients with dermatomyositis or polymyositis. Clin Exp Rheumatol 1996; 14:263–274. 139. Leow YH, Goh CL. Malignancy in adult dermatomyositis. Int J Dermatol 1997; 36:904–907. 140. Davis MD, Ahmed I. Ovarian malignancy in patients with dermatomyositis and polymyositis: a retrospective analysis of fourteen cases. J Am Acad Dermatol 1997; 37:730–733. 141. Mahe E, Descamps V, Burnouf M et al. A helpful clinical sign predictive of cancer in adult dermatomyositis: cutaneous necrosis. Arch Dermatol 2003; 139:539. 142. Hochberg MC, Feldman D, Stevens MB et al. Antibody to Jo-1 in polymyositis/dermatomyositis: association with interstitial pulmonary disease. J Rheumatol 1984; 11:663–665. 143. Lam WW, Chan H, Chan YL et al. MR imaging in amyopathic dermatomyositis. Acta Radiol 1999; 40:69–72. 144. Dalakas MC, Illa I, Dambrosia JM et al. A controlled trial of high-dose intravenous immune globulin infusions as treatment for dermatomyositis. N Engl J Med 1993; 329:1993–2000. 145. Amato AA, Barohn RJ. Idiopathic inflammatory myopathies. Neurol Clin 1997; 15:615–648. 146. Couriel DR, Beguelin GZ, Giralt S et al. Chronic graft versus host disease manifesting as polymyositis: an uncommon presentation. Bone Marrow Transplant 2002; 30:543–546. 147. Arin MJ, Scheid C, Hubel K et al. Chronic graft versus host disease with skin signs suggestive of dermatomyositis. Clin Exp Dermatol 2006; 31:141–143. 148. Levin MI, Mozaffar T, Al Lozi MT et al. Paraneoplastic necrotizing myopathy: clinical and pathological features. Neurology 1998; 50:764–767. 149. Anderlini P, Buzaid AC, Legha SS. Acute rhabdomyolysis after concurrent administration of interleukin-2, interferon-alfa, and chemotherapy for metastatic melanoma. Cancer 1995; 76:678–679. 150. Rudnicki SA, Dalmau J. Paraneoplastic syndromes of the spinal cord, nerve and muscle. Muscle Nerve 2000; 23:1800–1818.

VI

Complications of Cancer Therapy

16

Neurologic Complications of Radiation Therapy Daisy Chi, MD, Anthony Béhin, and Jean-Yves Delattre, MD

MD,

CONTENTS Introduction Acute Sequelae of Radiation Therapy on the Brain Early-Delayed Complications of RT Late-Delayed Complications of RT Radiation-Induced Brain Tumors Radiation-Induced Vasculopathy Endocrine Dysfunction Sequelae of Radiotherapy to the Spinal Cord Sequelae of Radiotherapy on the Cranial Nerves Consequences of RT on the Peripheral Nervous System Lower Motor Neuron Syndrome Radiation-Induced Peripheral Nerve Sheath Tumors Conclusion References

Summary Radiation therapy (RT) plays a central role in current cancer treatment modalities. However, despite advances in our knowledge of the mechanisms of radiation-induced neurotoxicity and the subsequent development of safer procedures, radiotherapy still accounts for disabling and sometimes life-threatening conditions. In addition, the range of indications of RT is widening to nontumoral disorders such as vascular malformations or trigeminal neuralgia, a move spearheaded by the development of radiosurgery and other modern irradiation techniques. The effects of radiation therapy on the brain and spinal cord have been sufficiently studied over the past decades that the main clinical, radiological or neuropathologic features of radiation-induced syndromes are increasingly well known. However, the pathophysiology of cerebral and spinal radiation injury is not fully understood, and the role of stem cells in both the initial damage from radiation and subsequent recovery are of great interest. Key Words: radiation, radiosurgery, stem cells, neurotoxicity, leukoencephalopathy

1. INTRODUCTION Radiation therapy (RT) can be associated with significant, sometimes life-threatening neurotoxicity. The development of new techniques (such as radiosurgery or brachytherapy) has widened the indications of RT (as exemplified by trigeminal neuralgia or vascular malformations), and has increased rates of long-term survivors who may develop late toxicity. Thus, complications of RT have gained increasing interest. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

259

260

Part VI / Complications of Cancer Therapy

Table 1 Main Neurological Complications of Radiotherapy

Site

Acute Complications (< 4 Weeks After RT)

Early-Delayed Complications (1–6 Month After RT)

Brain

Acute encephalopathy

* Somnolence syndrome * Worsening of pre-existing symptoms * Transitory cognitive impairment * Subacute rhombencephalitis * Lhermitte’s phenomenon

Spinal cord

Cranial nerves

Peripheral nerves

* Hearing loss * Anosmia * Ageusia Paresthesias

* Brachial or lumbosacral transitory plexopathy

Late-Delayed Complications (> 6 Month After RT) * Focal brain radionecrosis * Cognitive impairment and leukoencephalopathy * Secondary brain tumors

* * * * * *

Focal spinal radionecrosis Progressive myelopathy Spinal hemorrhage Hearing loss Visual loss Lower cranial nerve paralysis

* Brachial or lumbosacral irreversible plexopathy * Malignant nerve sheath tumors * Lower motor neuron syndrome

It is well known that the old dogma of nervous tissue radioresistance is only relative (1). Tolerance depends on several currently well-defined factors such as volume, total dose, dose per fraction, and duration of irradiation. In addition, other factors that may increase the risk of radiation-induced toxicity are older age, concomitant diseases (such as diabetes), vascular disease, adjuvant chemotherapy, and probably genetic predisposition (2). The main adverse neurological effects of radiotherapy are usually classified according to their time of appearance after irradiation and include acute disorders (days to weeks), early-delayed complications (1–6 months) and late-delayed complications (more than 6 months) (Table 1). The central or peripheral nervous system is often damaged directly but sometimes the damage is secondary to vascular or endocrine lesions, or to the development of a radiation-induced tumor.

1.1. Pathophysiology The pathophysiology of white matter abnormalities potentially leading to necrosis are not fully understood, but several hypotheses have been proposed. Because histological examination usually shows vascular lesions and demyelination, vessels and glial cells have often been considered as the primary target of radiation injury. However, the situation can be more complicated: recently, the interactions between radiation and other cell types have been studied more closely, as well as the recuperative abilities of the CNS (3). A short discussion of the main aspects of the pathophysiology of RT-related injuries follows.

1.2. Vascular Damage Transient disruption of the blood–brain barrier, possibly initiated by sphingomyelinases-mediated endothelial apoptosis, is thought to be responsible for the acute or early delayed, steroid-responsive forms of radiation toxicity (4–6). A vascular theory has also been advocated to explain radiation necrosis. Necrosis would be a consequence of ischemia secondary to blood vessel damage. Vascular damage is indeed prominent in radionecrosis, including thrombosis, hemorrhage, fibrinous exudates, telangiectasias, vascular fibrosis/hyalinization with luminal stenosis, and fibrinoid vascular necrosis. Furthermore, a progressive endothelial cell loss and vessel rarefaction seems to precede white matter necrosis (5). Thus, a cascade (involving among others, signaling molecules) could be

Chapter 16 / Neurologic Complications of Radiation Therapy

261

initiated at the time of irradiation producing gradual cell loss throughout the clinically silent period (7). This progressive loss would eventually lead to overt necrosis (8). Despite these findings, several points argue against a purely ischemic model: first, neurons are very sensitive to ischemia and should be prominently damaged if the lesion was primarily vascular whereas neurons are in fact largely spared during radionecrosis (9). Second, vascular damage is not always present in radionecrosis (10). Other cellular mechanisms may be involved and perpetuate vascular damage or contribute to edema, gliosis, and demyelination in the brain, such as up-regulation of diverse adhesion molecules (11), production of cytokines (12,13) or cumulative oxidative stress in endothelial cells.

1.3. Oligodendrocytes The demyelinating lesions observed after RT underscore the putative role of oligodendrocytes as a target of radiation damage. Oligodendrocytes are responsible for the production of myelin sheath in the CNS and derive from progenitor cells such as O-2A. It has been shown that CNS radiation induces depletion of oligodendrocytes and suppresses, at least transiently, the production of oligodendrocytes progenitors (14–16), possibly through a p53-dependant pathway (17–19). However, the contribution of demyelination in tissue destruction remains questionable since severely demyelinating conditions such as multiple sclerosis do not lead to overt necrosis (9). Although this theory accounts for predominant white matter lesions, arguments clearly go against an exclusive explanation.

1.4. Radiation and Other CNS Cell Types Several studies over the past few years have focused on the potential role of neurons, astrocytes, and microglia in the development of radiation-induced lesions, either as primary targets or through alterations of their regulatory capacity (20–22). Particularly noteworthy is the possible role of microglia (23) that may enhance radiation injury through persistent oxidative stress (9,24). Radiation damage to the subventricular zone, a tissue containing glial and neural stem cells, was reported more than 25 years ago (25). Recent studies have found a dose-dependant reduction of neural stem cells of the subependyma after irradiation in this region (26,27). Their implication in radiation injury thus seems probable but remains to be elucidated. As a whole, the development of radiation injury is complex, and results from interactions between cells (vascular cells and oligodendrocytes, but probably other cell types such as neural stem cells) and the physiological responses of these cells to injury. A wide range of cell interactions mediated by cytokines and other molecules and the influence of indirect factors such as microglial proliferation or the production of oxidative products resulting from cell destruction may account for the wide variety of lesions observed in different patients.

2. ACUTE SEQUELAE OF RADIATION THERAPY ON THE BRAIN 2.1. Acute Encephalopathy Acute encephalopathy usually appears within 2 weeks after the beginning of cranial RT, often a few hours after the first fraction delivery. The patient presents with nausea and vomiting, drowsiness, headache, dysarthria and a worsening of pre-existing neurological deficits, sometimes associated with fever. The clinical course is usually favorable, but herniation and death are a threat in patients with large tumors or who already presented with intracranial hypertension at the beginning of irradiation (e.g., in multiple metastases, posterior fossa or intraventricular tumors). Large doses per fraction (usually > 3 Gy/fraction) are the main risk factor: Young et al. (28) reported acute disorders in 50% of patients with brain metastases treated with 15 Gy in two fractions, and Hindo et al. (29) reported 4 deaths within 48 hrs of a 10 Gy RT given in one fraction. As these large doses are no longer in use, the frequency of the severe forms of this syndrome has decreased. A minor form of this condition is seen in many patients, consisting of nausea and moderate headache occurring within hours following cranial irradiation. Steroids may help in preventing or limiting the consequences of acute encephalopathy, especially in patients with large primary or secondary brain tumors or with considerable edema particularly at risk of herniation. In such patients, daily doses of steroids of at least 16 mg dexamethasone should be prescribed 48–72 hrs before the

262

Part VI / Complications of Cancer Therapy

first fraction; a limitation of dose per fraction (2 Gy or less per fraction) is also recommended in this situation (30,31) The pathophysiology of acute complications supposedly results from radiation-induced blood-brain barrier (BBB) disruption, accounting for a rise in intracranial pressure (32).

3. EARLY-DELAYED COMPLICATIONS OF RT Several early-delayed clinical patterns have been described (Table 1). They occur 2 weeks to 6 months after RT. The pathophysiology is not entirely understood: a transitory demyelinating process triggered by BBB disruption and/or selective oligodendrocyte dysfunction is incriminated.

3.1. Somnolence Syndrome This condition was first described in the late 1920s in children receiving low-dose RT for scalp ringworm. Several studies reported cases of children without brain tumor who developed somnolence syndrome 5–8 weeks after prophylactic cranial RT for leukemia (33,34); other reports have shown that it also occurs in adults. The incidence of somnolence syndrome varies greatly (with figures ranging from 8% (34) to 84% (36)); this difference is related to various factors including tumor types, radiation dose, fractionation, and diagnostic criteria (35). Prominent symptoms include drowsiness and excessive sleep, nausea, and anorexia, along with a general feeling of being slow, unwell, and lethargic. Headache and/or fever may also be reported. Friends or family often notice irritability, attentional deficits, and short-term memory impairment. The clinical severity of this complication is variable, with extremes ranging from minimal disorders to sleeping over 20 hrs a day (34). MRI studies are not contributory. Electroencephalographic abnormalities include non-specific diffuse slow waves (36). Most studies report a monophasic course of symptoms, usually with a favorable outcome within a few weeks. However, a prospective study on 19 adults treated for primary brain tumors with cranial RT (45–55 Gy) reported a biphasic pattern of symptoms with two critical periods (from the 11th to the 21st day and from the 31st to the 35th day) following RT (37). Furthermore, in this study, accelerated fractionation led to significantly more severe drowsiness and fatigue than a conventional scheme. Some authors advocate the use of steroids during and after radiotherapy as a prophylactic or symptomatic treatment (38). A prospective double-blind randomized trial in leukemic children found that a dexamethasone dose of 4 mg/m2 during cranial radiotherapy reduced the incidence of somnolence syndrome (17.6% vs. 64.3%) as compared to a dose of 2 mg/m2 (39).

3.2. Worsening of Pre-existing Symptoms, or Tumor-Pseudoprogression In patients under treatment for brain tumors, a worsening of pre-existing neurological focal deficits can be an alarming situation, leading to concerns of tumor progression or recurrence, especially when observed in association with features of the somnolence syndrome and with transitory cognitive impairment. This complication arises within 6 weeks to 3 months after RT, and is clinically impossible to differentiate from recurrence or progression. Neuroimaging may be normal or show edema and contrast enhancement within the tumor bed, a situation that does not allow this syndrome to be differentiated from tumor recurrence and explains why inclusion of patients in experimental regimens for “recurrence” is not indicated during this period (40); it is worthwhile noting that this radiological pattern can also be associated with no clinical worsening. Improvement usually follows within a few weeks or months and a close follow-up of CT or MR scan will show a regression of these signs in 4–8 weeks. As in the somnolence syndrome, the treatment lies on supportive care; steroids are usually proposed in this situation. Improvements in imaging techniques such as magnetic resonance spectroscopy (MRS) may help clarify this, but currently in clinical practice residual tumor is often still present within the irradiated area.

3.3. Transitory Cognitive Decline A transitory cognitive decline can be observed within the first 6 months after cranial RT, mainly affecting attention and recent memory, and may sometimes be associated with a somnolence syndrome. Several studies have addressed the features and frequency of this disorder, showing that the impairment primarily affects recent

Chapter 16 / Neurologic Complications of Radiation Therapy

263

memory and attention. Armstrong et al. (41) prospectively followed 5 patients treated for primary brain tumors: memory impairment was conspicuous in all patients 1.5 months after focal cranial RT (43–63 Gy), but complete regression was observed after 2.5–10.5 months. In another prospective study by Vigliani et al. (42) comparing 17 patients treated with focal cranial RT (54 Gy) for good prognosis gliomas and 14 matched control patients who did not undergo RT, 36% of the patients had a significant early-delayed impairment of their reaction test, with a return to normal baseline performance 12 months after RT. This test result was correlated with their occupational status: 69% of the patients could not work at 6 months, whereas 73% had continued or resumed their job at 1 year. In our experience, informing patients about this possible difficulty to return to a normal life (particularly work) at least during the first 6 months following radiotherapy is useful. Although severe and in some cases persistent early-delayed symptoms have been occasionally reported (43), it is of note that transitory cognitive impairment does not appear to be a clearcut prognostic factor predicting the further development of long-term cognitive disorders.

3.4. Subacute Rhombencephalitis Distinct from brainstem radionecrosis, which occurs later, early-delayed subacute rhombencephalitis may be observed about 1–3 months after RT using portals involving the brainstem, as in ocular, pituitary, or head and neck tumors. The clinical picture includes ataxia, dysarthria, diplopia, and/or nystagmus as well as auditory loss. In some cases, the cerebrospinal fluid analysis shows inflammatory signs. MRI may demonstrate white matter abnormalities appearing as grossly round or more extensive T1-weighted hypointensities and T2-weighted hyperintensities affecting the brainstem and the cerebellar peduncles; the lesions may enhance after Gadolinium injection (44,45). The condition usually improves progressively over a few weeks to a few months, either spontaneously or with steroids, but coma and death have been reported in rare cases (46,47).

4. LATE-DELAYED COMPLICATIONS OF RT Two main delayed complications may follow RT, usually after more than 6 months: focal radionecrosis and mild to severe cognitive impairment associated with leukoencephalopathy. These adverse effects may occasionally be delayed by many years.

4.1. Focal Brain Radionecrosis Focal cerebral radionecrosis (Fig. 1) constitutes a challenging complication of radiation therapy because it often mimics tumor recurrence and its functional consequences can be devastating. This complication may occur not only in patients who received local RT for a primary or metastasic brain tumor but also in patients without brain lesions with a history of RT for extra-parenchymal lesions, in whom normal brain was included in the radiation field (head and neck or pituitary tumors, meningiomas, skull osteosarcomas). A classical example is the bilateral medial temporal lobe necrosis that results from RT for pituitary or nasopharyngeal tumors. This once frequent complication of conventional RT has become rarer over the last 20 years, as a consequence of the generalized use of safer irradiation protocols. It has been shown that the upper limits of a “safe dose” were defined by a total dose of 55–60 Gy administered to a focal field with fractions of 1.8–2 Gy per day. Vascular risk factors such as diabetes, old age, and associated chemotherapy may also favor the potential development of radionecrosis. However, patients without any particular risk factors may develop radiation necrosis, probably because of an individual, unpredictable sensitivity to irradiation (48). The focal delivery of a single large radiation fraction during “radiosurgery” may also lead to focal necrosis of the brain adjacent to the irradiated lesion. In arteriovenous malformations (AVM), the incidence of brain necrosis ranges from <5% to 20% of cases, with location and volume found to be the main risk factors (49–51). After standard RT, radiation necrosis generally occurs within 1 to 2 years (52), but it has also been observed after several decades. As described above, shorter latencies (as short as 3 months) have also been reported, especially in patients treated with interstitial brachytherapy (53) or radiosurgery. Patients may experience seizures (first symptom in about 50% of cases), intracranial hypertension and/or focal neurological deficits (54–56). Such symptoms closely mimic tumor recurrence or progression.

264

Part VI / Complications of Cancer Therapy

Fig. 1. Axial CT-scan in a 42-year-old man treated with external beam conventional RT for a right frontal anaplastic astrocytoma. (a) Radionecrosis focus appearing as an enhancing lesion in the tumor bed one year after RT; (b) isolated calcifications 4 years later.

Diagnosing radiation necrosis is often a challenge. In many cases, CT and MR scans show a tumor-like pattern, often indistinguishable from tumor progression or recurrence (57). The lesions are typically T1-hypointense and T2-hyperintense, generally manifesting gadolinium enhancement. As diagnostic assessment based on the sole clinical and standard neuroimaging data is difficult, many studies have tried to address the problem of noninvasive differentiation between tumor recurrence/progression and radionecrosis. Positron emission tomography (PET) with 18 F-fluorodeoxyglucose (58) or 11 C-methionine, single photon emission computed tomography (SPECT) with 201 thallium or marked methoxy-isobutyl-isonitrile (99m TcMIBI) (59) and more recently the use of 3-[(123)I]iodo-alpha-methyl-l-tyrosine (IMT) (60,61) have been proposed to assess the nature of the lesions. In typical cases, radionecrosis is characterized by hypometabolism and tumoral growth by hypermetabolism. The more recent development of magnetic resonance spectroscopy (MRS) seems promising (62): spectral analysis of necrosis areas shows an overall, harmonious decrease of metabolite peaks, and a possible increase of lipids corresponding to cellular necrosis; no lactate peak is observed (63,64). However, none of these techniques offers 100% sensitivity or specificity (65,66). One of the reasons for limitations is the frequent co-existence of radiation-induced necrosis and tumor tissue within the same area, a situation that obviously renders clearcut distinction impossible (30). In some cases, angiography may provide further information, with an avascular mass seen in patients with radionecrosis; however, the risks and benefits of this procedure must be carefully weighed. While neuropathological examination remains the diagnostic standard (67), even pathological analysis may be difficult because of the frequent mixture of both residual/recurrent tumor and radiation necrosis within the lesion (6). Resection of necrotic foci is often the best treatment in symptomatic cases. Steroids are generally used, with possible long-term improvement (30); dependance is, however, neither rare nor predictable, with possible relapse while steroid doses are being tapered. A few other treatments have been reported in this setting, but their efficacy has not yet been addressed in large studies: anticoagulants have been prescribed by Glantz et al. (68) in 8 glioma patients (7 with histological evidence of necrosis) after the failure of steroids, leading to improvement in 5 patients. In our experience, anticoagulants have been disappointing. Further assessment of this therapy is required to confirm those results.

Chapter 16 / Neurologic Complications of Radiation Therapy

265

Some authors have advocated the use of hyperbaric oxygen (HBO), with the rationale that HBO increases the tissue pO2 and enhances angiogenesis. Chuba et al. (69) treated 10 patients with CNS radionecrosis (proved by biopsy in 8 cases) with 100% oxygen at 2.0–2.4 atm for 90–120 minutes/session in at least 20 sessions. All patients were stabilized or improved initially, and the 6 surviving patients showed durable improvement after 3–36 months. However, most patients were given steroids, and the respective effect of each treatment is not clear. HBO treatment was also reported to improve radionecrosis due to radiosurgery (70,71). The possible role of HBO in radiation-induced neurotoxicity needs to be evaluated in prospective trials (72). Other drugs or combinations such as pentoxifylline, alpha-tocopherol (73), low iron diet, desferrioxamine and pentobarbital (1) have also been proposed occasionally without definite evidence of efficacy. The usefulness of radioprotective agents such as difluoromethylornithine (DMO) (74), U-74389G (a 21-aminosteroid (75)), or others (76–79), in reducing the risk of necrosis remains to be determined. Preclinical studies have suggested that embryonic stem cell–derived glial precursors could be used as myelinating transplants in demyelinated postradiation experimental lesions (80). Several assays have been developed to identify individuals at risk for radiation sensitivity such as analysis of the survival fraction of cultured skin fibroblasts after 2 Gy irradiation (48), study of radiation-induced chromosomal aberrations, search for the ataxia telangectasia mutation, and the G2 cell cycle phase delay analysis (1,81). However, to date, such assessments have not reached the clinical setting.

4.2. Cognitive Dysfunction and Leukoencephalopathy Late-delayed cognitive impairment covers a wide continuum of patterns ranging from mild dysfunction to severe (and sometimes fatal) dementia, and has become an important concern over the last twenty years. The impact (and awareness) of this complication has grown partly because of a better assessment of quality of life data in clinical practice, but also because of a growing population of long-term survivors with tumor stability or remission. Considering cognitive dysfunction as a consequence of RT alone would result in a considerable overestimation of the incidence of radiation-induced sequelae. Cognitive impairment may actually be the consequence of complex interactions (82) between pre-existing cognitive abnormalities (especially in case of brain tumors), brain tumor growth, concomitant treatments (such as chemotherapy, but also antiepileptic (83) or psychotropic drugs), paraneoplastic encephalomyelitis, and endocrine dysfunction (Table 2). The imputability of cognitive dysfunction to RT should thus be interpreted with caution. In practice, this should lead to a precise individual workup in patients with potential radiation-induced cognitive impairment, but also to a careful and critical interpretation of the higher cognitive dysfunction rates found in the literature. Interestingly, a recent study comparing 195 low-grade glioma patients (104 had undergone RT during the previous years) to lowgrade hematological patients and controls, showed that cognitive dysfunction was mostly tumor-related; patients treated with doses per fraction over 2 Gy were the only ones to develop RT-linked memory impairment (84). Recent studies seem to confirm that RT plays a limited part in cognitive decline when using modern irradiation procedures (85–87). Nevertheless, several factors have been clearly linked to an increased risk of leukoencephalopathy (Fig. 2): (i) Old age: several studies have demonstrated that demented patients were clearly older (55–60 years) than nondemented patients (< 45 year old) (88–90). (ii) Large radiation doses, particularly when fractions over 2 Gy are used. (iii) Large irradiated brain volume: this factor may also increase the risk of cognitive impairment; whole brain radiotherapy (WBRT) may eventuate in as high as a 50% rate of RT-induced cognitive dysfunction (89,91,92) whereas the precise incidence of cognitive impairment after focal brain radiotherapy is difficult to assess because of contradictory figures in the literature (93,94). (iv) Combined treatment: the incidence of dementia in patients treated with combined RT and chemotherapy ranges from 4 to 63%. Methotrexate (MTX) is clearly implicated in combined toxicity; neurotoxic by itself, it is responsible for frequent cognitive dysfunction when associated with RT. In children, most studies to date have concluded that the combination of cranial irradiation and intrathecal MTX was associated with declines in both IQ and achievement scores (95). In adults treated with WBRT (40 Gy + 14 Gy boost) and a combination of intravenous and intrathecal MTX for CNS lymphoma, the incidence of severe progressive cognitive impairment

266

Part VI / Complications of Cancer Therapy

Table 2 Main Differential Diagnoses of Cognitive Impairment in Cancer Patients Related to the tumor • Progression of a known brain tumor • Development of brain metastases from systemic cancer • Tumor hemorrhage • Paraneoplastic limbic encephalitis/encephalomyelitis • Non-convulsive status epilepticus Related to treatment • Drug-induced dysfunction – chemotherapy (foremost methotrexate) – antiepileptic agents – psychotropic drugs • Radiation-induced cognitive impairment Vascular complications • Cerebral thrombophlebitis • Post-traumatic subdural hematoma Infection • Progressive multifocal leukoencephalopathy • Brain abscess Others • Endocrine dysfunction • Depression

increases with age, reaching 83% in patients over 60 years (96,97). The timing of MTX chemotherapy is important because the cognitive dysfunction rate is higher whenever MTX is prescribed during or after RT. This drug should therefore be given before irradiation. Data about the neurotoxicity of other combinations are sparse, but agents such as nitrosoureas, cisplatin, etoposide, cytarabine, or actinomycin D are also suspected to increase

Fig. 2. Axial FLAIR brain MRI showing a diffuse bilateral hyperintensity of the hemispheric white matter compatible with severe radiation-induced leukoencephalopathy, in a 57-year-old male patient previously treated with whole-brain radiotherapy for brain metastases from pulmonary adenocarcinoma.

Chapter 16 / Neurologic Complications of Radiation Therapy

267

radiation-induced cognitive toxicity. Multidrug regimens or high-dose chemotherapy combined with WBRT are probably associated with a higher risk of neurotoxicity (98).

Although there is a progressive continuum between mild to moderate cognitive impairment and severe fatal dementia, we will consider the two conditions separately. 4.2.1. Radiation-Induced Mild to Moderate Cognitive Impairment A mild to moderate cognitive dysfunction is more frequent in long-term survivors than frank dementia. The features of this condition are not perfectly defined, as results greatly vary according to the studies, probably due to neuropsychological evaluation procedures, duration of follow-up and population discrepancies (99). Cognitive impairment affects mainly attention and short-term memory in most reported cases, while intellectual functions are generally preserved on neuropsychological evaluations. Nevertheless, most patients have to decrease or even discontinue their professional activities. CT scan may be abnormal, showing periventricular hypodensities, an increase in the normal interface between white and gray matter, and ventricular enlargement. However, there seems to be no correlation between CT-scan abnormalities and the degree of cognitive impairment. MRI shows variable degrees of T2-weighted hyperintensities in the white matter, with a gross correlation between neuropsychological status and white matter lesions (100) (Figs. 2, 3). The course of the disease is difficult to predict: some patients deteriorate slowly while the majority apparently remain stable. Progression to dementia is seldom reported. There is no recognized treatment for this syndrome although some authors have advocated the use of methylphenidate for symptomatic relief (101). More recently, anticholinesterase drugs have been used with encouraging results (102). Data from animal studies have also shown that the administration of erythropoietin (EPO) may prevent cognitive impairment (103). 4.2.2. Radiation-Induced Dementia The incidence of this devastating complication varies widely in the literature (from 0% to more than 60%) according to the series. In a large review of several studies comprising 748 adult patients, the incidence of severe cognitive impairment compatible with dementia was at least 12.3% (104). More recent stuides and clinical practice as well as more recent studies provide less cause for alarm (24,105). The clinical picture is characterized by a “subcortical dementia” pattern that probably reflects consequences of diffuse white matter injury, occuring within 2 years in 69% of patients developing this complication (89). Patients present with progressive memory and attention deficits, intellectual loss, gait abnormalities and fatigue (106). Subsequently, emotional lability and apathy enrich the symptoms. The absence of hallucinations or delirium

Fig. 3. Axial T2-weighted brain MRI showing an extensive hyperintensity of the left hemispheric white matter in a 56-year-old male patient treated with conventional radiotherapy for a left supratentorial anaplastic oligodendroglioma. This patient presented with severe cognitive impairment and improved with methylphenidate.

268

Part VI / Complications of Cancer Therapy

and the very unusual occurrence of aphasia, agnosia, or apraxia (deficits suggesting cortical involvement) are important clinical features for narrowing the differential diagnosis, especially in elderly patients. Depression is common but antidepressants do not improve cognitive function. Eventually, patients may develop gait ataxia, incontinence, and sometimes a picture of akinetic mutism. Non specific features such as seizures, pyramidal or extrapyramidal signs, or tremor are also frequently encountered in the course of the disease. Neuroimaging always shows diffuse white matter lesions, best seen on MRI as T2-weighted hyperintensities, associated with cortical and subcortical atrophy as well as ventricular enlargement. When performed, the lumbar puncture usually shows normal to moderately elevated (< 1 g/l) CSF protein levels. No specific treatment is currently able to cure radiation-induced dementia. However, as an association with normal pressure hydrocephalus is possible, some authors have advocated utilisation of a ventriculoperitoneal shunting; while one must admit that responses are incomplete and transient, this procedure may be considered in order to improve the quality of life in a few selected patients (24,107,108). Deterioration occurs in about 80% of cases, leading to the death of the patient; stabilization is possible (18% of cases). Lasting improvement is exceptional. Death generally occurs within 1–48 months after the onset of the disorder (94).

5. RADIATION-INDUCED BRAIN TUMORS The precise role of RT in the development of a tumor is difficult to determine and cannot be assessed with certainty, in great part because these tumors have no distinctive features in previously irradiated and unirradiated patients. However, data from animal and epidemiological studies indicate that irradiated patients or animals have a higher likelihood to develop a second brain tumor than would have been expected from the control data (109). The relative risk of developing a radiation-induced tumor has been studied in several large studies. A study by Ron et al. on 10,834 patients treated with low-dose cranial and cervical irradiation for Tinea capitis (mean dose to neural tissue: 1.5 Gy) showed a relative risk of developing a tumor of 6.9; the risk for glioma was 2.6 (111). In another study of 10,106 survivors of childhood cancers led by the British Childhood Cancer Research Group (112), the relative risk of developing a secondary CNS tumor was 7. Most other studies have found similar results (113,114). However, the risk is probably higher in patients treated for acute lymphoblastic leukemia (ALL): a large retrospective cohort study of 9720 children (115) found a relative risk of developing a nervous system tumor as high as 22. In another setting, a large study on second brain tumors in 426 patients with pituitary adenoma treated with surgery and radiotherapy showed a 2.4% risk at 20 years (114). Imputability criteria include: (a) a long time before the occurrence of the second tumor (the mean onset delay is 12 years, with cases ranging from 1 to 40 years); (b) tumor growth within the radiation portal or at its margins; (c) a different histological subtype. After stereotactic radiosurgery, radiation-induced neoplasms are extremely rare, with only four reported cases in the literature (116). Three types of tumors have been reported to be linked with cranial irradiation: meningioma in about 70% of cases, glioma in 20%, and sarcoma in fewer than 10%. More than 300 cases of radiation-induced meningiomas have been reported in the literature; female predominance is less prominent than in spontaneous meningiomas (117). The risk of occurrence correlates with radiation dose: low-dose RT induced a relative risk of 9.5 in one study (111), whereas high-dose RT was linked to a relative risk of 37 (113). The tumor emerges after a long latency period: a review found extremes ranging from 2 to 63 years (mean 18.7 years) after high-dose RT (118). Most patients (68%) had been irradiated during childhood, and the interval between RT and the onset of the tumor was shorter in younger patients; however, the radiation dose did not influence the latency. Radiation-induced meningiomas are often multiple and recurrent with malignant histological features (119,120). Unlike sporadic meningiomas, NF2 gene inactivation and chromosome 22q deletions seem to be less frequent in radiation-induced meningiomas while other chromosomal lesions, especially loss of 1p, possibly induced by irradiation, may be more important in the development of these tumors (121,122). Radiation-induced gliomas are much less frequent. Since 1960, about 120 cases have been reported in the English literature (123,124); fewer than half of them were glioblastomas. In the group of patients treated with RT for acute leukemia, multifocality was particularly frequent, reaching 20% of cases. The median delay of onset ranges from of

Chapter 16 / Neurologic Complications of Radiation Therapy

269

6 to 9 years (124). The prognosis of these tumors is obviously poor: intrinsic resistance to treatment as well as prior aggressive therapies considerably limits the applicable therapies. Molecular alterations are apparently the same in sporadic and radiation-induced gliomas (125). Fewer than 40 cases of sarcomas have been reported to date, including several histological types (e.g., gliosarcomas, meningiosarcomas, neurofibrosarcomas) (109).

6. RADIATION-INDUCED VASCULOPATHY Distinct from radionecrosis, in which severe lesions of the arterioles and capillaries constitute a cardinal feature, radiation can induce other types of vascular damage leading to stroke or hemorrhage.

6.1. Large and Medium Intra- and Extracranial Artery Injury An arteriopathy affecting the large cervical blood vessels, especially the carotid artery (126), may be a complication of cervical radiation therapy, usually administered for lymphomas or head and neck cancers. Intracranial vessels may also be affected. The main early-delayed vascular complication is carotid rupture (127,128), which usually follows an association of cervical RT and surgery for head and neck tumors by a few weeks. Associated skin lesions such as necrosis or wound infection are common. The outcome of this exceptional complication is of course very poor. Late-delayed complications are more frequent, and generally occur many years after RT (median time about 20 years for extracranial, 7 years for intracranial artery lesions). The lesions are similar to those induced by atherosclerosis, but are often located in unusual places for common atherosclerosis and occur in an accelerated fashion. It has been observed that the larger the diameter of an irradiated artery, the longer the latency between RT and the onset of vasculopathy, a fact that might explain the shorter latency of RT-induced vasculopathy in children. Shorter latencies have also been reported with interstitial radiotherapy (129). The dose required to induce vascular lesions usually exceeds 50 Gy, but the type of irradiation, fractionation, and portal differs greatly from one case to another. The lesions consist of one or several stenosis or occlusions on the arteries included within the radiation portal. The diagnosis, suspected when a cervical murmur is heard in the immediate vicinity of radiationinduced skin lesions, relies on magnetic resonance angiography, ultrasound examinations and arteriography. The treatment is similar to that of usual atherosclerous lesions; in the event of carotid lesions, endarterectomy may be appropriate. However, surgery may be more difficult than in unirradiated patients because of vascular fibrosis and skin lesions, with higher post-operative risk of infection or healing problems. In other patients, antiplatelet agents may be prescribed if there is no contraindication. Some authors have advocated lowering serum cholesterol levels to prevent such lesions in patients at risk (130).

6.2. Radiation-Induced Vasculopathy with Moyamoya Pattern Intracranial vasculopathy leading to a progressive occlusive disease and a moyamoya pattern (characterized by abnormal anastomoses and netlike blood vessels) accounting for focal seizures, strokes, or transient ischemic attacks may follow intracranial irradiation, especially in very young children. This complication is particularly frequent in children treated for optic chiasm glioma, a condition often associated with neurofibromatosis type 1 (NF-1, which is a risk factor for vasculopathy in itself). It may also occur with other tumors such as brainstem glioma and craniopharyngioma (131). In a recent series (132) of 69 children (11 with NF-1) treated for optic pathway glioma with RT (median dose 55 Gy), 13 (19%) developed clinical and radiological signs of vasculopathy after a median latency of 36 months. The strong association between NF-1 and moyamoya is one of the reasons why radiation has been replaced with chemotherapy in younger children (95). The treatment focuses on preventing further strokes through surgical revascularization techniques; calcium blockers such as flunarizine have been advocated by some authors (133–136). The role of anti-aggregants has not been defined in this setting.

6.3. Silent Lacunar Lesions A recent report (137) described a pattern of silent cerebral lacunes occurring in children treated for brain tumors. In this study reviewing 524 consecutive children, 5 of 421 treated with RT and chemotherapy had lacunes.

270

Part VI / Complications of Cancer Therapy

Fig. 4. Radiation-induced aneurysm of the left internal carotid artery in a 65-year-old patient, occuring 27 years after irradiation for a cavernous sinus recurrence of systemic lymphoma. (a) The native slices of the angio-MRI show the lesion as a grossly round hyperintensity. (b) Conventional angiography confirmed the diagnosis (arrows).

Patients were a median of 4.5-years old at the time of the diagnosis and RT, and developed lacunes after a median latency of 2 years (ranging from 0.26 to 6 years). This pattern was associated with no further clinical deficit or neuropsychological impairment when compared to patients without lacunes. This condition is probably linked to delayed radiation-induced capillary and small vessel lesions.

6.4. Radiation-Induced Cavernomas, Angiomatous Malformations, and Aneurysms Brain vascular malformations such as telangiectasias and cavernomas (117,138,139) have been rarely observed following RT. Ocular telangiectasia may also occur (140). When present, their main risk is intracranial bleeding. Several cases of multiple radiation-induced cavernous angiomas have also been reported (141), occurring 18 months to 23 years after RT. Finally, fewer than 15 cases of radiation-induced intracranial aneurysms (Fig. 4) have been described in the literature (142,143). The median age of the patients was 37.5 years (ranging from 11 to 65 years) and a latency of 10 months to 21 years, with no correlation between the onset of aneurysms and the radiation dose. This represents a rare but severe problem, as rupture is always possible; 6 of 9 aneurysm ruptures proved fatal. A growing aneurysm can also mimic tumor recurrence. Aneurysms are sometimes detected preclinically with usual imaging procedures for tumors (CT scan and MRI), as was stressed by Azzarelli et al. (144), and particular attention should be drawn to evaluating the onset of such lesions during the imaging follow-up. When an aneurysm detected on CT or MR scan, or if the clinical history strongly suggests its presence, cerebral angiography is required for delineation.

7. ENDOCRINE DYSFUNCTION Frequently underestimated (95,145), endocrine disorders can be the consequence of direct irradiation of a gland (e.g., the thyroid gland, with about 50% of patients developing hypothyroidism within 20 years following radiotherapy for Hodgkin’s disease or certain head and neck cancers) or result from a hypothalamic-pituitary dysfunction secondary to cranial irradiation (several authors believe that the hypothalamus is more radiosensitive than the pituitary gland) (146). We will focus on the second type of disorder, which can be induced by cerebral or nasopharyngeal tumor irradiation. There is a positive correlation between radiation dose and the incidence of endocrine complications. For instance, in a prospective study on 268 patients treated with different RT schemes involving the brain, Littley et al. found 5 years after RT a 9% incidence of TSH deficiency in patients treated with 20 Gy, 22% with 35–37 Gy, and 52% with 42–45 Gy (147). A hormonal deficit can appear at any time after RT, but this may arise more rapidly in patients treated with higher radiation doses (148).

Chapter 16 / Neurologic Complications of Radiation Therapy

271

In children, varied endocrinologic deficits may result from cranial RT (administered for brain tumors or during prophylactic irradiation in acute lymphoblastic leukemia). Growth hormone (GH) is usually the first and in many cases the only anterior pituitary deficit in those young patients. Bothersome because of its consequences on statural growth, this complication affects about 50% of children treated with prophylactic cranial RT for acute lymphoblastic leukemia (149). According to a recent Danish study of 73 children treated with RT for a primary brain tumor (not involving the hypothalamo-pituitary axis directly) and with a long follow-up (median 15 years), 80% of patients manifested growth hormone deficiency; the median biological effective dose (BED) in the hypothalamo-pituitary area was higher in GH deficient children than in patients without GH deficiency (150). Administration of GH is recommended in children with growth hormone deficiency but unfortunately it has no effects in vertebral bodies and long-term survivors acquire a typical “spiderlike” physical appearance with long extremities and short trunk (95). A subtle central hypothyroidism is common in children and should be treated with thyroxine replacement therapy in order to limit the potential for thyroid carcinoma as well as to improve longitudinal growth (151,152). In adults, a recent study (153) evaluating 31 long-term brain tumor survivors followed 1.5–11 years after RT with a mean total dose of 62.3 ± 2.8 Gy compared with 31 age- and sex-matched controls, found hypothalamic hypothyroidism in 26% of patients, hypothalamic hypogonadism in 32% of patients, hyperprolactinemia in 29% of patients and panhypopituitarism in one patient. Low adrenal hormone levels were found in most patients, but without apparent clinical consequence. In the control group, only 6% had a baseline hormonal concentration outside the normal range. None of the controls had two or more hormonal abnormalities, while 42% of the patients had multiple deficits. Only 23% of patients had normal thyroid, gonadal and adrenal baseline levels; this result is consistent with another earlier study by Taphoorn et al. reporting hypothalamic-pituitary dysfunction in 10 of 13 (77%) long-term survivors irradiated for supratentorial low-grade glioma (154). Another study of patients treated for nasopharyngeal cancer found secondary hypothyroidism in 27% of cases (of hypothalamic origin in 19% and pituitary origin in 8%) (155). The neurological consequences of severe hypothyroidism are well known, including encephalopathy, cerebellar ataxia, pseudo-myotonia, and sometimes peripheral neuropathy. Moderate elevation of CSF protein is also usual. All these abnormalities may be misleading if the correct diagnosis is not considered. Secondary hypogonadism is an important concern especially in male patients, responsible for a decrease in libido and sometimes impotence impacting negatively on quality of life. Hyperprolactinemia of hypothalamic origin is a notable concern in women who develop oligoamenorrhea and galactorrhea (156); in men, it may result in gynecomastia and a decrease in libido. The follow-up consultations are a good place for a regular clinical endocrine evaluation; the precise biological follow-up scheme is debated and is adapted according to the emerging deficits, but long-term assessment should be the rule. The treatment of hormonal deficits lies in replacement therapy, and usually leads to an improvement in the patient’s condition. Bromocriptine has been utilized with success in patients with symptomatic hyperprolactinemia (109).

8. SEQUELAE OF RADIOTHERAPY TO THE SPINAL CORD Damage to the spinal cord may be the consequence of RT administered for spinal cord tumors, Hodgkin’s disease, mediastinal or head and neck cancers. Early descriptions in the 1940s (157) were followed by numerous descriptions of post-radiation myelopathy delineating the main clinical patterns, that is, early-delayed myelopathy and several types of late-delayed complications including progressive myelopathy, lower motor neuron disorder, and spinal hemorrhage. There is no clear clinical or experimental evidence of acute spinal cord toxicity due to RT, and a sudden worsening during irradiation should lead to a search for intratumoral hemorrhage or tumor progression (30).

8.1. Early-Delayed (Transient) Radiation Myelopathy The onset of this complication occurs from 6 weeks to 6 months after RT, and improvement follows in most cases within 2 to 9 months (158), though persistence of the symptoms for a longer time is possible in rare

272

Part VI / Complications of Cancer Therapy

Table 3 Main Causes of Lhermitte’s Phenomenon in Cancer Patients • Chemotherapy – cisplatin – docetaxel • Radiotherapy • Spinal tumor • Vitamin B12 deficiency • Viral myelitis (in particular herpes zoster) • Multiple sclerosis

cases. It usually follows radiation to the cervical or thoracic spinal cord. After mantle RT for Hodgkin’s disease, early-delayed myelopathy occurred in 15% of cases (159). In another study, Fein et al. found a global incidence of 3.6% (40 cases among 1112 patients receiving 30 Gy or more). The incidence was 8% in the group of patients receiving 50 Gy or more, 3% after doses of 45–49.9 Gy, 4% after doses of 40–44.9 Gy, and 2% after doses of 30–39.9 Gy. The risk was also increased with a fraction size over 2 Gy (160). The clinical pattern first described by Esik et al. (161) generally consists of Lhermitte’s phenomenon, triggered by neck flexion, and is characterized by brief unpleasant sensations of numbness, tingling, and/or often electriclike feelings from the neck to the spine and extremities. There is no known MRI change associated with this condition. This symptom is nonspecific (Table 3), and other causes should be considered in a patient with cancer (162) including chemotherapy (cisplatin or docetaxel), spinal tumor, vitamin B12 deficiency, herpes zoster, or even multiple sclerosis (which may be aggravated by irradiation). The presumed pathophysiology of early-delayed myelopathy is transient demyelination, probably secondary to a loss of oligodendroglial cells following RT (163,164). There is no specific treatment for this condition, and none is required, as recovery occurs in most cases. Early-delayed spinal cord disorder is not predictive of evolution to the much more serious progressive myelopathy.

8.2. Late-Delayed Radiation-Induced Spinal Cord Disorders Spinal radionecrosis (Fig. 5) (with features similar to its cerebral counterpart), progressive myelopathy, and spinal hemorrhage have been described as late complications of spine radiation. 8.2.1. Progressive Myelopathy, or Delayed Radiation Myelopathy (DRM) This complication occurs 6 months to 10 years after exposure to RT. Risk factors include advancing age, large radiation doses and fractions, previous irradiation especially in childhood, and large portals involving thoracic or lumbar spinal cord (158). Chemotherapy may increase the risk of delayed radiation myelopathy (165), but data are still unclear on this point. The generally accepted tolerance for the spinal cord is 45 Gy in 22–25 daily fractions, with a risk < 1% for a dose of 50 Gy increasing to 5% for a dose of 60 Gy delivered in 1.8–2 Gy fractions (166). Delayed radiation myelopathy may begin abruptly or, more often, in a progressive way; patients complain of sensory and/or motor deficits leading to para- or tetraparesis. A typical initial clinical presentation is a Brown– Sequard syndrome, consisting of a motor deficit associated with ipsilateral sensory loss affecting tactile, vibration and passive movement sense on one side, and sensory loss affecting mainly temperature and pain sensory modalities on the other side. In some patients, a transverse myelopathy develops with bilateral leg weakness and sensory loss up to the irradiated region. Some patients also experience pain. Bladder and bowel sphincter as well as diaphragmatic dysfunction (in upper cervical spinal cord lesions) are possible. The evolution of delayed radiation myelopathy varies; in some patients the symptoms stabilize, while in others they progress to a complete deficit.

Chapter 16 / Neurologic Complications of Radiation Therapy

273

Fig. 5. Radiation myelopathy of the lower cervical spinal cord appearing as an MRI T1-weighted gadolinium-enhancing lesion, occuring one year after an irradiation for a cancer of the pharynx with cervical nodes in a 47-year-old woman.

The diagnosis of delayed radiation myelopathy implies—as was underlined as early as 1961 by Pallis et al. (167)—that the site of the main lesion is within the radiation-exposed area of the spinal cord and that all other potential causes of myelopathy have been carefully reviewed and eliminated. Spinal cord MRI is helpful, though nonspecific. The initial description of Wang et al. (168,169) has been confirmed in several subsequent studies (170–172): the initial MRI may be normal if performed during the first weeks of the disease, but a slightly delayed examination usually reveals a swollen cord with T1-weighted hypointensity and T2-weighted hyperintensity. Lesions enhance in about 50% of cases after gadolinium injection (173,174). In contrast, late examinations, performed years after the onset of the disease, may show spinal cord atrophy without any signal abnormality; a case of cystic formation in late delayed radiation myelopathy has also been reported. Moderately elevated protein is the most common finding in the CSF but lacks any specificity. If performed, somatosensory evoked potentials show changes correlated to the extent of the lesions, whereas spinal conduction velocity is decreased. Neuropathological findings include demyelination, focal necrosis, and axonal loss, together with vascular abnormalities including telangiectasias, endothelial swelling with fibrin exudation, hyaline degeneration, thickening and fibrinoid necrosis of the vessel walls with perivascular fibrosis and sometimes vasculitis. Corticosteroids may improve some patients, probably because of their action on the inflammatory and edematous components of the disorder; however, patients often become steroid-dependent and only a few experience longterm improvement. There is no current proven long-term treatment for delayed radiation myelopathy. However, Angibaud et al. have reported the efficacy of hyperbaric oxygen in stabilizing or improving 6 out of 9 patients with DRM (175), and Calabro et al. recently reported a similar case (176). Anticoagulation has also been tried, with improvement in one patient with myelopathy treated for > 3 months with full anticoagulation and stabilization in another treated with coumarin (68). 8.2.2. Late-Delayed Spinal Hematoma This rare complication has only been described in a few cases, following spinal radiotherapy by 6–30 years, and occurring within the radiation portal but outside the location of the primary tumor (177); acute onset leg weakness and back pain rapidly lead to para- or tetraparesis. The diagnosis relies on MRI demonstration of hemorrhage. Spontaneous symptom resolution is possible, but new episodes may occur later. There is no proven effective

274

Part VI / Complications of Cancer Therapy

treatment for this condition. Avoidance of aspirin or NSAIDs is prudent. Radiation-induced telangiectasias with secondary hemorrhage could explain this condition.

9. SEQUELAE OF RADIOTHERAPY ON THE CRANIAL NERVES Apart from acute reversible radiation toxicity, any of the cranial nerves may be involved in radiation-induced, late-delayed complications if included in the radiation portal. These complications are rare, probably arising in fewer than 1% of cases after conventional radiotherapy (60 Gy, 2 Gy per daily fraction). Large daily radiation fractions increase this risk. However, a recent study in brain tumor survivors found that 17% of patients developed neurosensory impairment and that RT exposure greater than 50 Gy to the posterior fossa was associated with a higher likelihood of developing any hearing impairment (178). The main complications are described below:

9.1. Olfactory Nerve Injury During cranial radiation, patients may describe reversible sensations of smelling an odor (179). This may be due to direct acute stimulation of the olfactory neurons. Anosmia has also been described in some patients (180,181), often associated with taste disorders (182).

9.2. Optic Neuropathy Probably facilitated by pre-existing lesions (e.g., in diabetic patients), optic neuropathy may occur 6 months to 14 years after radiation therapy for a tumor of the orbital, pituitary, or suprasellar regions (183). In one study, the incidence of retrobulbar optic neuropathy was 3.8% after a conventional radiation scheme for head and neck cancer (184). Optic neuropathy can also overshadow the prognosis of patients treated with high-energy electron beam therapy for age-related macular degeneration in up to 19% of cases (185). Proton beam irradiation, currently proposed in the treatment of several tumors including meningioma and choroidal melanoma, may also induce this complication (186). The classical pattern consists of progressive or sometimes acute onset visual loss, leading to monocular or binocular blindness with optic atrophy (187). This disorder is painless. In the case of anterior lesions, the ocular fundus usually shows papilledema and prepapillary and premacular hemorrhage, sometimes associated with radiation-induced retinal lesions. In contrast, fundoscopy may be normal if the lesions are posterior. In those cases, brain MRI may be useful, demonstrating enlargement of the optic nerve and chiasma, with T2-weighted hyperintensities and contrast enhancement (Fig. 6). Demyelination, axonal loss, gliosis, and modifications of the vessel walls characterize these lesions histologically; endothelial cell loss has recently been stressed in this setting, with more significant abnormalities in patients treated with high-dose (55–70 Gy) compared with those treated with low-dose (10 Gy or less) radiation therapy (141). These lesions are irreversible in many patients. Steroids and anticoagulants (189) have been advocated in chiasmatic lesions, but their efficacy is quite inconstant, while the use of hyperbaric oxygen in optic neuropathy

Fig. 6. MRI axial and coronal T1-weighted sequences showing enlargement and gadolinium enhancement of the prechiasmatic optic nerves after cranial RT in a 44-year-old man. The patient became progressively blind, and the follow-up scan revealed chiasmatic atrophy 4 years after RT.

Chapter 16 / Neurologic Complications of Radiation Therapy

275

remains controversial (190–192). Optic nerve sheath fenestration has been attempted with some success in a few patients (193).

9.3. Ocular Motor Nerve Injury Rarely reported, the involvement of ocular motor nerves may be associated with optic neuropathy. The most frequent of these palsies affect the abducens (VI) nerve (30). Transient ocular motor palsies have been reported following radiation schemes focused on the pituitary tract, but permanent palsies have also been described after radiation therapy of nasopharyngeal carcinoma. Possible regression suggests that a demyelinating process may be involved, rather than progressive fibrosis. Neuromyotonia is a late-delayed complication following RT to the sella turcica or the cavernous sinus region by several years, and characterized by spontaneous spasms of the eye muscles, responsible for episodes of transitory painless diplopia, usually lasting a few seconds; these episodes can occur up to several times an hour (194). Membrane stabilizers such as phenytoin or carbamazepine may improve this disorder. Radiation-induced hyperexcitability of the nerve fibers may underlie the pathophysiology.

9.4. Trigeminal Nerve Dysfunction Involvement of the trigeminal nerve is quite rare. Neuromyotonia is exceptionally encountered; treatment with carbamazepine is effective in this condition (195,196). After gamma knife radiosurgery for trigeminal neuralgia, the principal reported complication is mild facial numbness occuring in 2.7–14% of patients (197–199). Trigeminal neuropathy can also result from radiosurgery for vestibular schwannoma.

9.5. Facial Nerve Injury The different branches of the facial nerve are not equally affected by radiation. Taste dysfunction is usual and many patients complain of aguesia, a symptom that may be permanent in up to 50% of patients irradiated with 50–60 Gy for head and neck tumors (200). However, taste disturbances are a common feature in cancer patients, and chemotherapy may play a part (201). Motor deficit is almost never the consequence of fractionated RT and should lead to a search for tumoral invasion (30). In contrast, radiosurgery has been reported to account for possible facial palsy; however, the relatively high initial figures of facial weakness after treatment for vestibular schwannoma (202,203) seem to have much decreased, being currently under 5% of cases (199,204).

9.6. Acoustic Nerve Dysfunction Acute damage to the cochlea may be responsible for usually reversible complaints of high-frequency hearing loss and tinnitus. Otitis media can also be responsible for an early-delayed hearing loss. Following RT by a few weeks; this condition results from an obliteration of the eustachian tube by edema and mucosal vasodilatation. The diagnosis is often easy, as otoscopic examination reveals fluid behind the tympanic membrane. Otitis media usually regresses spontaneously, but can require myringotomy in some cases for symptom alleviation. In most cases, relief can be obtained by prescribing nasal vasoconstriction agents. Late-delayed hearing loss might result from lesions to the organ of Corti with subsequent acoustic nerve atrophy; however, a recent report underlines the relative resistance of the organ of Corti to radiation. The precise histological pattern of these disorders is not known; however, the labyrinth has been shown to be damaged in previous studies (205,206). The consequences of radiosurgery for vestibular schwannoma on hearing have been better assessed over the last few years (204,207–210). In a recent report (211), 14% of patients with measurable hearing before treatment became deaf after radiosurgery, and 42% of patients had an elevation of their pure tone threshold of 20 dB or more. The risk factors for hearing loss in this study included neurofibromatosis type 2 (NF-2), history of prior surgical resection, and tumor size. Several other studies showed varying complication rates (159–161). The contribution of RT, however, is not always easy to assess, as spontaneous growth of the tumor may also lead to deafness.

276

Part VI / Complications of Cancer Therapy

9.7. Lower Cranial Nerve Involvement These nerves (glossopharyngeal, vagus, spinal accessory, and hypoglossal nerves) can be damaged after head or neck radiation therapy with large doses. Complications occur earlier with a larger radiation dose and typically arise months to years after the treatment. The pathophysiology is likely radiation fibrosis. The hypoglossal nerve is the most commonly involved lower cranial nerve (212); the patient may present with unilateral, often asymptomatic tongue paralysis (213–215), or with bilateral and disabling paralysis. This complication may occur many years after RT (214). Longstanding paralysis is responsible for tongue atrophy, with asymmetry that may be associated with fatty infiltration or edema-like changes on MRI (216,217). Paralysis of the vagus nerve leads to unilateral paralysis of the vocal cord and of the palate, responsible for difficulties in swallowing (218). Horner’s syndrome may be associated with these disorders, resulting from injury to the sympathetic fibers (30). Lesions of the spinal accessory nerve lead to shoulder drop which is easily diagnosed during the clinical examination. Patients may also present with multiple lower cranial nerve palsies (219). Cranial nerve palsies may occur during skull base osteoradionecrosis after radiotherapy for nasopharyngeal carcinoma (220).

10. CONSEQUENCES OF RT ON THE PERIPHERAL NERVOUS SYSTEM Apart from rare peripheral nerve tumors, brachial and lumbosacral plexopathies are the main bothersome and disabling complications of RT in the peripheral nervous system. In this situation, differential diagnosis with tumor infiltration of the plexus is often a challenge, even with modern imaging techniques.

10.1. Dropped Head Syndrome Dropped head syndrome has been recently described as a potential late-delayed complication of RT (221). These patients usually develop weakness of neck extensors several years after irradiation involving the cervical region, for example, mantle irradiation for Hodgkin’s lymphoma. Clinical examination reveals an amyotrophic deficit of neck muscles, often extending to other muscles innervated by upper cervical roots, without any sensitive impairment. The differential diagnoses include myasthenia gravis, ALS, and inflammatory myopathies. The mechanism and precise location of the causative lesions are unclear.

10.2. Brachial Plexopathy Brachial plexopathy results from RT to the supraclavicular, infraclavicular or axillary regions, usually for lung or breast cancers and sometimes Hodgkin’s disease. In most cases, the main problem is to differentiate this condition from neoplastic invasion of the plexus. 10.2.1. Early-Delayed Brachial Plexopathy This complication occurs a median of 4.5 months after RT, with a range of 2–14 months. Its incidence is about 1–2% after irradiation for breast cancer (222,223). The clinical pattern includes paresthesia in the hand and a distal motor deficit; amyotrophy and fasciculations may be present at later stages. Axillary pain is reported in about 60% of cases; always moderate, it may be spontaneous or occur with movement. Although a progressive course towards paralysis of the brachial plexus is possible, complete improvement is the rule, in most cases after 3–6 months. Electrophysiological investigations show a decrease in nerve conduction velocities. The pathophysiology of this condition is not fully understood; a direct radiation toxicity on the Schwann cells inducing demyelination has been invoked (224). 10.2.2. Late-Delayed Progressive Brachial Plexopathy Delayed radiation-induced brachial plexopathy appears after a median time of 40 months (up to 20 years) (225). Its incidence varies widely in the literature, with figures ranging from 14% to 73% (227), the most important risk factors being total radiation dose (> 60 Gy) and fraction size (> 2 Gy). Overlap of radiation fields has also been incriminated as well as combined radio-chemotherapy. For example, Olsen et al. found a definite or probable radiation plexopathy in 42% of patients treated with chemotherapy (cyclophosphamide, tamoxifen, or a

Chapter 16 / Neurologic Complications of Radiation Therapy

277

combination of cyclophosphamide, methotrexate and 5-fluorouracil) and RT versus 26% in patients treated with RT alone (227). The pathophysiology is unclear and may have biphasic components: during the first phase, direct radiation damage to the nerves may cause electrophysiological and histochemical changes; later on, injury to the small vessels and fibrosis around the nerves may account for severe nerve injury (228). The disorder is usually progressive. Initial symptoms include distal paresthesias (typically, pins and needles or numbness of the thumb and first finger) and mild sensory deficit on clinical examination, often with no clear radicular topography. The other initial findings consist of some degree of amyotrophy and an early loss of reflexes. Proximal weakness is found in about a quarter of cases (229,230). Visible myokymia, when present, is quite suggestive of the diagnosis. The examination may also show local complications of radiotherapy, such as radiation dermatitis, painful induration of the axillary region, and/or lymphedema. During the later course of late-delayed plexopathy, a generally progressive motor deficit may be observed (in a few cases, an apoplectic onset has been reported, sometimes after physical effort). Pain is quite uncommon at diagnosis and is usually a relatively minor feature. The severity of the condition is variable, from a simple discomfort to an almost complete paralysis of the limb. The differential diagnosis must necessarily eliminate a neoplastic invasion of the brachial plexus. Some clinical signs may be important clues. In a large retrospective study on 100 cases of brachial plexus lesions, including 22 radiation plexopathies and 78 metastatic brachial plexopathies (34 in irradiated patients and 44 in nonirradiated patients), Kori et al. (229) found several factors as indicators of neoplastic invasion: (i) pain, especially when severe, is an important feature, present in 89% of irradiated patients with neoplastic infiltration of the plexus (versus 18% of patients with radiation plexopathy); (ii) Horner’s syndrome was present in 56% of patients with tumor infiltration (versus 14%). On the contrary, the following signs argue for a radiation-induced disorder: (i) dysesthesia, present in 55% of radiation plexopathies (versus 6% of plexopathies linked with tumor infiltration); (ii) lymphedema, reported in 73% of cases (versus 15%). These results are consistent with those of other authors (231). Motor conduction velocities are usually normal or slightly decreased in radiation plexopathies. Sensory conduction velocities are rarely altered. The F waves can be absent or delayed. Electromyography is always abnormal, with fasciculations, fibrillation and slow denervation potentials. The most important finding in favor of radiation-induced plexopathy is the presence of myokymic discharge, present in about two-thirds of patients; this feature is quite rare (< 5%) in patients with an infiltrating tumor. Myokymic discharges are often located in the abductor pollicis brevis and pronator quadratus muscles (232). The main aim of imaging is to differentiate between radiation plexopathy and neoplastic invasion. CT scan was the first noninvasive examination to be useful (233,234); this imaging technique may show a distortion of the tissue planes and fat or may be normal. MRI is superior to CT scan in this indication (235); furthermore, bone artifacts do not impair the interpretation of MRI, which also allows a study of the cervical spine in search for epidural or cervical root secondary lesions. Radiation fibrosis is responsible for a thickening of the components of the brachial plexus, sometimes with contrast enhancement (235). Tumor invasion is diagnosed when a mass lesion is visible along the roots of the branches of the brachial plexus. Nevertheless, a retrospective study at the Mayo Clinic of 71 patients with cancer and brachial plexopathy who had an MRI yielded a 21% (15 patients) discordance rate between imaging and eventual diagnosis (179,231). A recent study of 50 breast cancer patients (236) using an association of body-coil and surface-coil techniques suggests a major role for MRI to assess neoplastic recurrence. This technique allowed a correct diagnosis of tumor recurrence in 26 of 27 patients, directly related to brachial plexus in 17 of them, and associated with cervical spine degenerative lesions in seven cases. Exclusion of a malignant disease was accurate in 20 of 21 cases. These results corresponded to a sensitivity of the MR criteria for tumor detection of 96% and a specificity of 95%, with similar positive and negative predictive values. Positron emission tomography using 18-fluoro2-deoxyglucose (18 FDG-PET) may also be helpful to differentiate tumor infiltration from radiation-induced plexopathy (237). Sometimes, noninvasive investigations are inconclusive, and a biopsy may be indicated. In radiation plexopathy this reveals fibrosis and the absence of tumor infiltration. The treatment of pain is often challenging, and includes analgesic drugs, tricyclic antidepressants, and/or anticonvulsants. Steroids can also be helpful. Anticoagulants

278

Part VI / Complications of Cancer Therapy

have also been reported to be beneficial in radiation-induced neuropathies (238). Techniques such as transdermal electrical nerve stimulation and dorsal column roots stimulation have also been proposed. Neurolysis with or without omentoplasty has been performed in radiation-induced brachial plexopathy, but the benefit of this approach is questionable. Prudent physiotherapy may be indicated (230). 10.2.3. Ischemic Late-Delayed Brachial Plexopathy Sudden late-delayed plexopathy has rarely been reported following an occlusion of the subclavian artery (239).

10.3. Lumbosacral Plexopathy Far less common than brachial plexopathy, lumbosacral plexopathy may follow pelvic or lower abdomen cancer (uterus, ovary, testis, rectum, or lymphoma) irradiation. 10.3.1. Early-Delayed Lumbosacral Plexopathy As with brachial plexopathies, an early-delayed, generally transient lumbosacral plexopathy is possible. It usually begins a few months (median 4 months) after RT, with a typical pattern of distal bilateral paresthesias of the lower limbs. Clinical examination is normal in most cases, and improvement follows within 3–6 months. 10.3.2. Late-Delayed Lumbosacral Plexopathy This disorder shares similar features with brachial plexopathy but is much less frequently reported. The onset follows initial RT by 1–30 years (median: 5 years). The clinical pattern is characterized by a progressive, usually asymmetric, and bilateral motor deficit of the lower limbs associated with less-marked sensory deficits. As in brachial plexopathy, pain is generally mild or absent. The course of the disease leads to a slow worsening of the motor deficit. The patient may stabilize after several months or years (240). On electromyography (EMG), motor nerve conduction velocities are normal or moderately decreased in the leg. The saphenous nerve sensory potential is absent or has a decreased amplitude in about 50% of cases. EMG shows myokymias in the proximal muscles in 60% of cases. Fibrillation potentials in the paravertebral muscles are found in 50% of patients. Histology, when performed, may show fibrosis tightly hugging the plexus, and in some cases the cauda equina as well. Treatment of pain is identical to that of brachial plexopathy. Two patients treated with anticoagulation remained stable with an improvement in pain (68).

11. LOWER MOTOR NEURON SYNDROME A lower motor neuron syndrome can be a consequence of pelvic irradiation for testicular tumors, lumbosacral RT, or craniospinal RT for medulloblastoma, and begins 3 months to 25 years after RT. Maier et al. (241) reported 15 cases of this syndrome out of 343 patients who had undergone a lumboaortic irradiation scheme. About 30 cases have been reported since that period with various radiation schemes. The patient presents with a progressive proximal and distal, often bilateral, and more-or-less symmetrical weakness of the inferior limbs; muscle atrophy and fasciculations may be associated with this deficit. Physical examination confirms a flaccid motor deficit and areflexia, but no sensory loss appears during the early stages. Sensory deficit may appear after several years, as well as sphincter disturbance characterized by lack of bladder sensation and incontinence (242). Variable patterns of progression and associated disability can be seen. MRI may be normal, but contrast enhancement of the roots of the cauda equina has been described (242). The CSF is usually acellular, frequently showing high protein levels. On electromyography, different stages of denervation are identified while sural sensory nerve action potentials are usually preserved. It is unclear whether the lesions localize to the anterior horn cells of the spinal cord or the proximal part of the nerve roots. Some reports advocate an anterior horn cell disorder, as no sensory signs were reported and electrophysiological data were compatible with pure motor neuron syndrome (243). However, the few available neuropathological data support radicular lesions affecting the nerve roots of the cauda equina (193). In a study of 6 patients treated with RT (mean dose 45 Gy) for testicular cancer and including neuropathological examination of one case, Bowen et al. (242) found strong arguments favoring radiculopathy, including (i) the

Chapter 16 / Neurologic Complications of Radiation Therapy

279

presence of late-delayed sensory and sphincter disturbances, appearing 4–8 years after the motor symptoms; (ii) MRI abnormalities showing contrast enhancement of the lumbosacral roots of the cauda equina in 2 out of 3 patients; (iii) no lesion in the cord at necropsy but thickening of the roots of the cauda equina with focal areas of hemorrhagic discoloration, fibrosis and axonal loss; the roots included abnormal dilated vessels with thickened and hyalinized walls. Thus, Wohlgemuth et al. (244) suggest that a better term for this syndrome would be “post-irradiation cauda equina syndrome”. There is no recognized treatment of this condition. A patient has been reported to improve while on warfarin and steroids (245).

12. RADIATION-INDUCED PERIPHERAL NERVE SHEATH TUMORS A few dozen cases of radiation-induced nerve sheath tumors have been reported (246,247). Patients with neurofibromatosis type 1 (NF-1) have an increased risk of developing this complication. In a retrospective study on radiation-induced peripheral nerve sheath tumors, 3 patients out of 9 (33%) had familial and/or clinical signs of NF-1 (248); in this series, the latency between RT and the onset of the secondary tumor ranged from 4 to 41 years. Pain followed by the development of a sensorimotor deficit typifies the clinical presentation. The differential diagnosis of a local recurrence of the primary tumor generally requires biopsy. The treatment of these nerve sheath tumors relies principally on aggressive surgery with tumor-free margins; amputation of a limb, when performed, does not significantly change overall survival (249). Neither chemotherapy nor radiotherapy have shown any clear benefit yet in terms of survival in those tumors (250).

13. CONCLUSION Radiation therapy remains one of the most effective treatments of cancer and will probably become, through the development of new irradiation techniques, a standard option for treating some nonmalignant diseases. Familiarity with its potential risks is thus essential in order to prevent complications when possible as well as to be able to inform the patients of their possible onset. The development of RT-related neurotoxicity remains largely unpredictable, and seems to depend on yet to be discovered individual predisposition. As progress has been made in understanding the pathophysiology of radiation-induced injury and in determining “safe” doses over the past few decades, many complications have become rarer than a few years ago.

REFERENCES 1. New P. Radiation injury to the nervous system Curr Opin Neurol. 2001; 14:725–734. 2. Swennen MH. Delayed radiation toxicity after focal or whole brain radiotherapy for low-grade glioma J Neurooncol. 2004; 66:333–339. 3. Belka C, Budach W, Kortmann RD et al. Radiation-induced CNS toxicity: molecular and cellular mechanisms Br J Cancer. 2001; 85:1233–1239. 4. Li YQ, Chen P, Haimovitz–Friedman A et al. Endothelial apoptosis initiates acute blood–brain barrier disruption after ionizing radiation Cancer Res. 2003; 63:5950–5956. 5. Lyubimova N, Hopewell JW. Experimental evidence to support the hypothesis that damage to vascular endothelium plays the primary role in the development of late radiation-induced CNS injury Br J Radiol. 2004; 77:488–492. 6. Perry A, Schmidt RE. Cancer therapy–associated CNS neuropathology: an update and review of the literature Acta Neuropathol (Berl). 2006; 111:197–212. 7. Lai R, Abrey LE, Rosenblum MK et al. Treatment-induced leukoencephalopathy in primary CNS lymphoma: a clinical and autopsy study Neurology.2004; 62:451–456. 8. Coderre JA, Morris GM, Micca PL et al. Late effects of radiation on the central nervous system: role of vascular endothelial damage and glial stem cell survival Radiat Res. 2006;166:495–503. 9. Tofilon PJ, Fike JR. The radioresponse of the central nervous system: a dynamic process Radiat Res. 2000; 153: 357–370. 10. Schultheiss TE, Kun LE, Ang KK et al. Radiation response of the central nervous system Int J Radiat Oncol Biol Phys. 1995; 31: 1093–1112. 11. Quarmby S, Kumar P, Kumar S. Radiation-induced normal tissue injury: role of adhesion molecules in leukocyte–endothelial cell interactions Int J Cancer. 1999; 82:385–395. 12. Eissner G, Kohlhuber F, Grell M et al. Critical involvement of transmembrane tumor necrosis factor-alpha in endothelial programmed cell death mediated by ionizing radiation and bacterial endotoxin Blood. 1995; 86:4184–4193. 13. Daigle JL, Hong JH, Chiang CS et al. The role of tumor necrosis factor signaling pathways in the response of murine brain to irradiation Cancer Res. 2001; 61:8859–8865.

280

Part VI / Complications of Cancer Therapy

14. Van der Maazen RWM, Berhagen I, Kleiboer BJ et al. Radiosensitivity of glial progenitor cells of the perinatal and adult rat optic nerve studied by an in vitro clonogenic assay Radiother Oncol. 1991; 20:258–264. 15. Van der Maazen RWM, Kleiboer BJ, Berhagen I et al. Irradiation in vitro discriminates between different O-2A progenitor cell subpopulations in the perinatal central nervous system of rats Radiat Res. 1991; 128:64–72 [Abstract]. 16. Van der Maazen RWM, Kleiboer BJ, Berhagen I et al. Repair capacity of adult rat glial progenitor cells determined by an in vitro clonogenic assay after in vitro or in vivo fractionated irradiation Int J Radiat Biol. 1993; 63:661–666. 17. Chow, BM., Li, YQ, Wong, CS. Radiation-induced apoptosis in the adult central nervous system is p53-dependent Cell Death Differ. 2000; 7:712–720. 18. Atkinson SL, Li YQ, Wong CS. Apoptosis and proliferation of oligodendrocyte progenitor cells in the irradiated rodent spinal cord Int J Radiat Oncol Biol Phys. 2005; 62:535–544. 19. Hornsey S, Myers R, Coultas PG et al. Turnover of proliferative cells in the spinal cord after X-irradiation and its relation to time-dependent repair of radiation damage Br J Radiol.1981; 54:1081–1085. 20. Enokido Y, Araki T, Tanaka K et al. Involvement of p53 in DNA strand break-induced apoptosis in postmitotic CNS neurons Eur J Neurosci. 1996; 8:1812–1821. 21. Gobbel GT, Bellinzona M, Vogt AR et al. Response of postmitotic neurons to X-irradiation: implications for the role of DNA damage in neuronal apoptosis J Neurosci. 1998; 18: 147–155. 22. Chiang CS, McBride WH, and Withers HR. Radiation-induced astrocytic and microglial responses in mouse brain Radiother Oncol. 1993; 29:60–68. 23. Thomas WE. Brain macrophages: evaluation of microglia and their function Brain Res Rev. 1992. B17:61–74. 24. Omuro AM. Delayed neurotoxicity in primary central nervous system lymphoma Arch Neurol. 2005; 62:1595–1600. 25. Hopewell JW and Cavanagh JB. Effects of X-irradiation on the mitotic activity of the subependymal plate of rats Br J Radiol. 1972; 45:461–465. 26. Tada E, Yang E, Gobbel GT, Lamborn KR et al. Long-term impairment of subependymal repopulation following damage by ionizing irradiation Exp Neurol. 1999; 160:66–77. 27. Bellinzona M, Gobbel GT, Shinohara C et al. Apoptosis is induced in the subependyma of young adult rats by ionizing irradiation Neurosci Lett. 1996; 208:163–166. 28. Young DF, Posner JB, Chu F et al. Rapid-course radiation therapy of cerebral metastases: results and complications Cancer. 1974; 34:1069–1076. 29. Hindo WA, DeTrana FA, III, Lee MS et al. Large dose increment irradiation in treatment of cerebral metastases Cancer. 1970; 26:138–141. 30. Posner JB. Side effects of radiation therapy. In Posner JB (ed.). Neurologic Complications of Cancer. Philadelphia: F.A. Davis Company, 1995:311–337. 31. Keime–Guibert F, Napolitano M, Delattre JY. Neurological complications of radiotherapy and chemotherapy J Neurol. 1998; 245: 695–708. 32. Phillips PC, Delattre JY, Berger CA et al. Early and progressive increases in regional brain capillary permeability following singleand fractionated-dose cranial radiation in rat Neurology. 1987; 37(Suppl.1):301. 33. Freeman JE, Johnston PG, Voke JM. Somnolence after prophylactic cranial irradiation in children with acute lymphoblastic leukaemia Br Med J. 1973; 4:523–525. 34. Littman P, Rosenstock J, Gale G et al. The somnolence syndrome in leukemic children following reduced daily dose fractions of cranial radiation Int J Radiat Oncol Biol Phys. 1984; 10:1851–1853. 35. Chow E, Davis L, Holden L et al. Prospective assessment of patient-rated symptoms following whole brain radiotherapy for brain metastases J Pain Symptom Manage. 2005; 30:18–23. 36. Ch’ien LT, Aur RJ, Stagner S et al. Long-term neurological implications of somnolence syndrome in children with acute lymphocytic leukemia Ann Neurol. 1980; 8:273–277. 37. Faithfull S, Brada M. Somnolence syndrome in adults following cranial irradiation for primary brain tumours Clin Oncol. 1998; 10:250–254. 38. Mandell LR, Walker RW, Steinherz P et al. Reduced incidence of the somnolence syndrome in leukemic children with steroid coverage during prophylactic cranial radiation therapy: results of a pilot study Cancer. 1989; 63:1975–1978. 39. Uzal D, Ozyar E, Hayran M et al. Reduced incidence of the somnolence syndrome after prophylactic cranial irradiation in children with acute lymphoblastic leukemia. Radiother Oncol. 1998; 48:29–32. 40. De Wit MC, de Bruin HG, Eijkenboom W et al. Immediate post-radiotherapy changes in malignant glioma can mimic tumor progression Neurology. 2004; 63:535–537. 41. Armstrong C, Ruffer J, Corn B et al. Biphasic patterns of memory deficits following moderate-dose partial-brain irradiation: neuropsychologic outcome and proposed mechanisms J Clin Oncol. 1995; 13: 2263–2271. 42. Vigliani MC, Sichez N, Poisson M et al. A prospective study of cognitive functions following conventional radiotherapy for supratentorial gliomas in young adults: 4-year results Int J Radiat Oncol Biol Phys. 1996; 35:527–533. 43. Chak LY, Zatz LM, Wasserstein P et al. Neurologic dysfunction in patients treated for small cell carcinoma of the lung: a clinical and radiological study Int J Radiat Oncol Biol Phys. 1986; 12:385–389. 44. Creange A, Felten D, Kiesel I et al. Leucoencéphalopathie subaiguë du rhombencéphale après radiothérapie hypophysaire Rev Neurol (Paris) 1994; 150:704–708. 45. Song T, Liang BL, Huang SQ et al. Magnetic resonance imaging manifestations of radiation injury in brain stem and cervical spinal cord of nasopharyngeal carcinoma patients after radiotherapy Ai Zheng. 2005; 24:357–361[Abstract]. 46. Lampert P, Tom MI, Rider WD. Disseminated demyelination of the brain following Co60 radiation Arch Pathol. 1959; 68:322–330.

Chapter 16 / Neurologic Complications of Radiation Therapy

281

47. Ochi S, Takahashi Y, Yokoyama S. Fulminating midbrain irradiation injury of pediatric brain tumor No To Shinke.. 2005; 57:800–805 [Abstract]. 48. Malone S, Raaphorst GP, Gray R et al. Enhanced in vitro radiosensitivity of skin fibroblasts in two patients developing brain necrosis following AVM radiosurgery: a new risk factor with potential for a predictive assay Int J Radiat Oncol Biol Phys. 2000; 47:185–189. 49. Flickinger JC, Kondziolka D, Lunsford LD et al. Development of a model to predict permanent symptomatic postradiosurgery injury for arteriovenous malformation patients. Arteriovenous Malformation Radiosurgery Study Group Int J Radiat Oncol Biol Phys. 2000; 46:1143–1148. 50. Schlienger M, Atlan D, Lefkopoulos D et al. Linac radiosurgery for cerebral arteriovenous malformations: results in 169 patients Int J Radiat Oncol Biol Phys. 2000; 46:1135–1142. 51. Miyawaki L, Dowd C, Wara W et al. Five-year results of LINAC radiosurgery for arteriovenous malformations: outcome for large AVMS Int J Radiat Oncol Biol Phys. 1999; 44:1089–1106. 52. Morris GM, Coderre JA, Micca PL et al. Central nervous system tolerance to boron neutron capture therapy with p-boronophenylalanine Br J Cancer. 1997; 76:1623–1629. 53. Oppenheimer JH, Levy ML, Sinha U et al. Radionecrosis secondary to interstitial brachytherapy: correlation of magnetic resonance imaging and histopathology Neurosurgery. 1992; 31:336–343. 54. Van Effenterre R, Boch AL. Radionécrose du chiasma Neurochirurgie. 1993; 39:75–84. 55. Coghlan KM, Magennis P. Cerebral radionecrosis following the treatment of parotid tumours: a case report and review of the literature Int J Oral Maxillofac Surg. 1999; 28:50–52. 56. Cirafisi C, Verderame F. Radiation-induced rhombencephalopathy Ital J Neurol Sci. 1999; 20:55–58. 57. Kumar AJ, Leeds NE, Fuller GN et al. Malignant gliomas: MR imaging spectrum of radiation therapy–and chemotherapy-induced necrosis of the brain after treatment Radiology. 2000; 217:377–384. 58. Janus TJ, Kim EE, Tilbury R et al. Use of [18F]fluorodeoxyglucose positron emission tomography in patients with primary malignant brain tumors Ann Neurol. 1993; 33:540–548. 59. Lamy-Lhullier C, Dubois F, Blond S et al. Intérêt de la tomoscintigraphie cérébrale au sestamibi marqué au technétium dans le diagnostic différentiel récidive tumorale-radionécrose des tumeurs gliales sus-tentorielles de l’adulte Neurochirurgie. 1999; 45: 110–117. 60. Henze M, Mohammed A, Schlemmer H et al. Detection of tumour progression in the follow–up of irradiated low-grade astrocytomas: comparison of 3–[123I]iodo–alpha–methyl-l-tyrosine and 99mTc–MIBI SPET Eur J Nucl Med Mol Imaging. 2002; 29:1455–1461. 61. Matheja P, Weckesser M, Rickert Ch et al. I-123-lodo-alpha-methyl tyrosine SPECT in nonparenchymal brain tumours Nuklearmedizin. 2002; 41:191–196. 62. Galanaud D, Nicoli F, Figarella-Branger D et al. MR spectroscopy of brain tumors J Radiol. 2006; 87:822–832. 63. Schlemmer HP, Bachert P, Herfarth KK et al. Proton MR spectroscopic evaluation of suspicious brain lesions after stereotactic radiotherapy. AJNR Am J Neuroradiol. 2001; 22:1316–1324. 64. Lichy MP, Bachert P, Hamprecht F et al. Application of (1)H MR spectroscopic imaging in radiation oncology: choline as a marker for determining the relative probability of tumor progression after radiation of glial brain tumors Rofo. 2006; 178:627–633 [Abstract]. 65. Ricci PE, Karis JP, Heiserman JE et al. Differentiating recurrent tumor from radiation necrosis: time for re–evaluation of positron emission tomography? AJNR Am J Neuroradiol. 1998; 19:407–413. 66. Matheja P, Rickert C, Weckesser M et al. Scintigraphic pitfall: delayed radionecrosis: case illustration J Neurosurg. 2000; 92:732. 67. Forsyth PA, Kelly PJ, Cascino TL et al. Radiation necrosis or glioma recurrence: is computer-assisted stereotactic biopsy useful? J Neurosurg. 1995; 82:436–444. 68. Glantz MJ, Burger PC, Friedman AH et al. Treatment of radiation-induced nervous system injury with heparin and warfarin Neurology. 1994; 44:2020–2027. 69. Chuba PJ, Aronin P, Bhambhani K et al. Hyperbaric oxygen therapy for radiation-induced brain injury in children Cancer. 1997; 80:2005–2012. 70. Leber KA, Eder HG, Kovac H et al. Treatment of cerebral radionecrosis by hyperbaric oxygen therapy Stereotact Funct Neurosurg. 1998; 70:229–236 71. Kohshi K, Imada H, Nomoto S et al. Successful treatment of radiation-induced brain necrosis by hyperbaric oxygen therapy J Neurol Sci. 2003; 209:115–117. 72. Feldmeier JJ, Hampson NB. A systematic review of the literature reporting the application of hyperbaric oxygen prevention and treatment of delayed radiation injuries: an evidence based approach. Undersea Hyperb Med. 2002; 29:4–30[Abstract]. 73. Chan AS, Cheung MC, Law SC et al. Phase II study of alpha-tocopherol in improving the cognitive function of patients with temporal lobe radionecrosis Cancer. 2004; 100:398–404. 74. Fike JR, Gobbel GT, Marton LJ, et al: Radiation brain injury is reduced by the polyamine inhibitor alpha-difluoromethylornithine Radiat Res. 1994; 138:99–106. 75. Kondziolka D, Mori Y, Martinez AJ, et al: Beneficial effects of the radioprotectant 21-aminosteroid U-74389G in a radiosurgery rat malignant glioma model Int J Radiat Oncol Biol Phys. 1999; 44:179–184. 76. Guelman LR, Zorrilla Zubilete MA, Rios H et al. WR–2721 (amifostine, ethyol) prevents motor and morphological changes induced by neonatal X-irradiation Neurochem Int. 2003; 42:385–391. 77. Sasse AD, Clark LG, Sasse EC et al. Amifostine reduces side effects and improves complete response rate during radiotherapy: results of a meta-analysis Int J Radiat Oncol Biol Phys. 2006; 64:784–791. 78. Lyubimova N, Coultas P, Yuen K et al. In vivo radioprotection of mouse brain endothelial cells by Hoechst 33342 Br J Radiol. 2001; 74:77–82. 79. Guelman L, Zorilla Z, Rios H, et al. GM1 ganglioside treatment protects against long-term neurotoxic effects of neonatal X-irradiation on cerebellarcortex cytoarchitecture and motor functions Brain Res. 2000; 858:303–311.

282

Part VI / Complications of Cancer Therapy

80. Brustle O, Jones KN, Learish RD et al. Embryonic stem cell–derived glial precursors: a source of myelinating transplants Science. 1999; 285:650–651. 81. Christie D, Lavin M, Tan L. Clinical application of in vitro radiohypersensitivity testing Australas Radiol. 2000; 44:333–335. 82. Taphoorn MJ, Klein M. Cognitive deficits in adult patients with brain tumours Lancet Neurol. 2004; 3: 159–168. 83. Klein M, Engelberts NH, van der Ploeg HM et al. Epilepsy in low-grade gliomas: the impact on cognitive function and quality of life Ann Neurol. 2003; 54:514–520. 84. Klein M, Heimans JJ, Aaronson NK et al. Effect of radiotherapy and other treatment-related factors on mid-term to long-term cognitive sequelae in low-grade gliomas: a comparative study Lancet. 2002; 360:1361–1368. 85. Armstrong CL, Hunter JV, Ledakis GE et al. Late cognitive and radiographic changes related to radiotherapy: initial prospective findings Neurology. 2002; 59:40–48. 86. Torres IJ, Mundt AJ, Sweeney PJ et al. A longitudinal neuropsychological study of partial brain radiation in adults with brain tumors Neurology. 2003; 60: 1113–1118. 87. Klein M, Heimans JJ, Aaronson NK et al. Impaired cognitive functioning in low-grade glioma patients: relationship to tumor localisation, radiotherapy and the use of anticonvulsants J Clin Oncol. 2004; 22:966–967. 88. Asai A, Matsutani M, Kohno T et al. Subacute brain atrophy after radiation therapy for malignant brain tumor Cancer. 1989; 63:1962–1974. 89. Imperato JP, Paleologos NA, Vick NA. Effects of treatment on long-term survivors with malignant astrocytomas Ann Neurol. 1990; 28:818–822. 90. Vigliani MC, Duyckaerts C, Delattre JY. Radiation-induced cognitive dysfunction in adults. In Vecht CJ (ed.). Handbook of Clinical Neurology (Vol. 23). Elsevier Science, Amsterdam 1997: 371–388. 91. Fisher B, Seiferheld W, Schultz C et al. Secondary analysis of Radiation Therapy Oncology Group study (RTOG) 9310: an intergroup phase II combined modality treatment of primary central nervous system lymphoma J Neurooncol. 2005; 74:201–205. 92. Archibald YM, Lunn D, Ruttan LA et al. Cognitive functioning in long-term survivors of high-grade glioma J Neurosurg. 1994; 80:247–253. 93. Lee PW, Hung BK, Woo EK et al. Effects of radiation therapy on neuropsychological functioning in patients with nasopharyngeal carcinoma J Neurol Neurosurg Psychiatry. 1989; 52:488–492. 94. DeAngelis LM, Delattre JY, Posner JB. Radiation-induced dementia in patients cured of brain metastases Neurology. 1989; 39:789–796. 95. Duffner PK, 2004. Long-term effects of radiation on cognitive and endocrine function in children with leukemia and brain tumors Neurologist. 2004; 10:293–310. 96. DeAngelis LM, Yahalom J, Thaler HT et al. Combined modality therapy for primary CNS lymphoma J Clin Oncol. 1992; 10:635–643. 97. Abrey LE, Yahalom J, DeAngelis LM. Treatment for primary CNS lymphoma: the next step J Clin Oncol. 2000; 18:3144–3150. 98. Wassenberg MW, Bromberg JE, Witkamp TD et al. White matter lesions and encephalopathy in patients treated for primary central nervous system lymphoma J Neurooncol. 2001; 52:73–80. 99. Armstrong CL, Corn BW, Ruffer JE et al. Radiotherapeutic effects on brain function: double dissociation of memory systems Neuropsychiatry Neuropsychol Behav Neurol. 2000; 13:101–111. 100. Postma TJ, Klein M, Verstappen CC et al. Radiotherapy-induced cerebral abnormalities in patients with low-grade glioma Neurology. 2002; 59:121–123. 101. Meyers CA, Weitzner MA, Valentine AD et al. Methylphenidate therapy improves cognition, mood, and function of brain tumor patients J Clin Oncol. 1998; 16:2522–2527. 102. Shaw EG, Rosdhal R, D’Agostino RB Jr et al. Phase II study of donepezil in irradiated brain tumor patients: effect on cognitive function, mood, and quality of life J Clin Oncol. 2006; 24:1415–1420 [Abstract]. 103. Senzer N: Rationale for a phase III study of erythropoietin as a neurocognitive protectant in patients with lung cancer receiving prophylactic cranial irradiation Semin Oncol. 2002; 29:47–52. 104. Crossen JR, Garwood D, Glatstein E et al. Neurobehavioral sequelae of cranial irradiation in adults: a review of radiation-induced encephalopathy J Clin Oncol. 1994; 12:627–642. 105. Mulhern RK, Merchant TE, Gajjar A et al. Late neurocognitive sequelae in survivors of brain tumours in childhood Lancet Oncol. 2004; 5:399–408. 106. Brown PD, Buckner JC, O’Fallon JR et al. Effects of radiotherapy on cognitive function in patients with low-grade glioma measured by the Folstein Mini-Mental State Examination J Clin Oncol. 2003; 21: 2519–2524. 107. Thiessen B, DeAngelis LM. Hydrocephalus in radiation leukoencephalopathy: results of ventriculoperitoneal shunting Arch Neurol. 1998; 55:705–710. 108. Perrini P, Scollato A, Cioffi F et al. Radiation leukoencephalopathy associated with moderate hydrocephalus: intracranial pressure monitoring and results of ventriculoperitoneal shunting Neurol Sci. 2002; 23:237–241. 109. Kleinschmidt-Demasters BK, Kang JS, Lillehei KO. The burden of radiation-induced central nervous system tumors: a single institution s experience J Neuropathol Exp Neurol. 2006; 65:204–216. 110. Cahan WG, Woodard HQ, Higinbotham NL, Stewart FW, Coley BL. Sarcoma arising in irradiated bone: report of eleven cases. Cancer. 1998; Jan 1;82(1):8–34. 111. Ron E, Modan B, Boice JD, Jr. et al. Tumors of the brain and nervous system after radiotherapy in childhood N Engl J Med. 1988; 319:1033–1039. 112. Hawkins MM, Draper GJ, Kingston JE. Incidence of second primary tumours among childhood cancer survivors Br J Cancer. 1987; 56:339–347. 113. Brada M, Ford D, Ashley S et al. Risk of second brain tumour after conservative surgery and radiotherapy for pituitary adenoma BMJ. 1992; 304: 1343–1346.

Chapter 16 / Neurologic Complications of Radiation Therapy

283

114. Minniti G, Traish D, Ashley S et al. Risk of second brain tumor after conservative surgery and radiotherapy for pituitary adenoma: update after further 10 years J Clin Endocrinol Metab. 2005; 90:800–804. 115. Neglia JP, Meadows AT, Robison LL et al. Second neoplasms after acute lymphoblastic leukemia in childhood N Engl J Med. 1991; 325:1330–1336. 116. Muracciole X, Cowen D, Regis J. Radiosurgery and brain radio-induced carcinogenesis: update. Neurochirurgie. 2004; 50:414–420. 117. Amirjamshidi A, Abbassioun K. Radiation-induced tumors of the central nervous system occurring in childhood and adolescence: four unusual lesions in three patients and a review of the literature Childs Nerv Syst. 2000; 16:390–397. 118. Strojan P, Popovic M, Jereb B. Secondary intracranial meningiomas after high-dose cranial irradiation: report of five cases and review of the literature Int J Radiat Oncol Biol Phys. 2000; 48:65–73. 119. Musa BS, Pople IK, Cummins BH. Intracranial meningiomas following irradiation: a growing problem? Br J Neurosurg. 1995; 9:629–637. 120. De Tommasi A, Occhiogrosso M, De Tommasi C et al. Radiation-induced intracranial meningiomas: review of six operated cases Neurosurg Rev 2005; 28:104–114. 121. Shoshan Y, Chernova O, Juen SS et al. Radiation-induced meningioma: a distinct molecular genetic pattern? J Neuropathol Exp Neurol. 2000; 59:614–620. 122. Rajcan-Separovic E, Maguire J, Loukianova T et al. Loss of 1p and 7p in radiation-induced meningiomas identified by comparative genomic hybridization Cancer Genet Cytogene. 2003; 144:6–11. 123. Mack EE. Radiation-induced tumors. In Berger MS, Wilson CB (eds.). The Gliomas. WB Saunders, Philadelphia. 1999:724–735. 124. Salvati M, Frati A, Russo N et al. Radiation-induced gliomas: report of 10 cases and review of the literature Surg Neurol. 2003; 60:60–67. 125. Brat DJ, James CD, Jedlicka AE et al. Molecular genetic alterations in radiation-induced astrocytomas Am J Pathol. 1999; 154: 1431–1438. 126. Murros KE, Toole JF. The effect of radiation on carotid arteries: a review article Arch Neurol. 1989; 46: 449–455. 127. Gupta S. Radiation-induced carotid artery blow out: a case report Acta Chir Belg. 1994; 94:299–300. 128. McCready RA, Hyde GL, Bivins BA et al. Radiation-induced arterial injuries Surgery. 1983; 93:306–312. 129. Bernstein M, Lumley M, Davidson G et al. Intracranial arterial occlusion associated with high-activity iodine-125 brachytherapy for glioblastoma J Neurooncol. 1993; 17:253–260. 130. Werner MH, Burger PC, Heinz ER et al. Intracranial atherosclerosis following radiotherapy Neurology. 1988; 38:1158–1160. 131. Bitzer M, Topka H. Progressive cerebral occlusive disease after radiation therapy Stroke. 1995; 26:131–136. 132. Grill J, Couanet D, Cappelli C et al. Radiation-induced cerebral vasculopathy in children with neurofibromatosis and optic pathway glioma Ann Neurol. 1999; 45:393–396. 133. Suzuki Y, Negoro M, Shibuya M et al. Surgical treatment for pediatric moyamoya disease: use of the superficial temporal artery for both areas supplied by the anterior and middle cerebral arteries Neurosurgery. 1997; 40:324–329. 134. Dauser RC, Tuite GF, McCluggage CW. Dural inversion procedure for moyamoya disease: technical note J Neurosurg. 1997; 86:719–723. 135. Ross IB, Shevell MI, Montes JL et al. Encephaloduroarteriosynangiosis (EDAS) for the treatment of childhood moyamoya disease Pediatr Neurol. 1994; 10:199–204. 136. Kim SK, Wang KC, Kim IO et al. Combined encephaloduroarteriosynangiosis and bifrontal encephalogaleo(periosteal)synangiosis in pediatric moyamoya disease Neurosurgery. 2002; 50:88–96. 137. Fouladi M, Langston J, Mulhern R et al. Silent lacunar lesions detected by magnetic resonance imaging of children with brain tumors: a late sequela of therapy J Clin Onco. 2000; 18:824–831. 138. Baumgartner JE, Ater JL, Ha CS et al. Pathologically proven cavernous angiomas of the brain following radiation therapy for pediatric brain tumors Pediatr Neurosurg. 2003; 39:201–207. 139. Heckl S, Aschoff A, Kunze S. Radiation-induced cavernous hemangiomas of the brain: a late effect predominantly in children Cancer. 2002; 94:3285–3291. 140. Mauget-Faysse M, Vuillaume M, Quaranta M et al. Idiopathic and radiation-induced ocular telangiectasia: the involvement of the ATM gene Invest Ophthalmol Vis Sci. 2003; 44:3257–3262. 141. Novelli PM, Reigel DH, Langham GP et al. Multiple cavernous angiomas after high-dose whole-brain radiation therapy Pediatr Neurosurg. 1997; 26:322–325. 142. Jensen FK, Wagner A. Intracranial aneurysm following radiation therapy for medulloblastoma: a case report and review of the literature Acta Radiol. 1997; 38:37–42. 143. Louis E, Martin-Duverneuil N, Carpentier AF et al. Anévrysme post-radique de la carotide intra–caverneuse Rev Neurol (Paris). 2003; 159:319–322. 144. Azzarelli B, Moore J, Gilmor R et al. Multiple fusiform intracranial aneurysms following curative radiation therapy for suprasellar germinoma: case report J Neurosurg. 1984; 61:1141–1145. 145. Constine LS, Woolf PD, Cann D et al. Hypothalamic-pituitary dysfunction after radiation for brain tumors N Engl J Med. 1993; 328:87–94. 146. Littley MD, Shalet SM, Beardwell CG. Radiation and hypothalamic-pituitary function Baillieres Clin Endocrinol Metab. 1990; 4:147–175. 147. Littley MD, Shalet SM, Beardwell CG et al. Radiation-induced hypopituitarism is dose-dependent Clin Endocrinol (Oxf) 1989; 31:363–373. 148. Clayton PE, Shalet SM. Dose dependency of time of onset of radiation-induced growth hormone deficiency J Pediatr. 1991; 118: 226–228. 149. Rappaport R, Brauner R. Growth and endocrine disorders secondary to cranial irradiation Pediatr Res. 1989; 25:561–567.

284

Part VI / Complications of Cancer Therapy

150. Schmiegelow M, Lassen S, Poulsen HS et al. Cranial radiotherapy of childhood brain tumours: growth hormone deficiency and its relation to the biological effective dose of irradiation in a large population based study Clin Endocrinol (Oxf). 2000; 53:191–197. 151. Stevens G, Downes S, Ralston A. Thyroid dose in children undergoing prophylactic cranial irradiation Int J Radiat Oncol Biol Phys. 1998; 42:385–390. 152. Livesey EA, Hindmarsh PC, Brook CGD, et al. Endocrine disorders following treatment of childhood brain tumours Br J Cancer. 1990; 61622–61625. 153. Arlt W, Hove U, Muller B et al. Frequent and frequently overlooked: treatment-induced endocrine dysfunction in adult long-term survivors of primary brain tumors Neurology. 1997; 49:498–506. 154. Taphoorn MJ, Heimans JJ, van der Veen EA et al. Endocrine functions in long-term survivors of low-grade supratentorial glioma treated with radiation therapy J Neurooncol. 1995; 25:97–102. 155. Samaan NA, Vieto R, Schultz PN et al. Hypothalamic, pituitary and thyroid dysfunction after radiotherapy to the head and neck Int J Radiat Oncol Biol Phys. 1982; 8:1857–1867. 156. Petterson T, MacFarlane IA, Foy PM et al. Hyperprolactinemia and infertility following cranial irradiation for brain tumours: successful treatment with bromocriptine Br J Neurosurg. 1993; 7:571–574. 157. Ahlbom H. Results of radiotherapy of hypopharyngeal cancer at Radium–Hemmet, Stockholm. Acta Radiol. 1941; 22:155–171. 158. Rampling R, Symonds P. Radiation myelopathy Curr Opin Neurol. 1998; 11:627–632. 159. Word JA, Kalokhe UP, Aron BS et al. Transient radiation myelopathy (Lhermitte’s sign) in patients with Hodgkin’s disease treated by mantle irradiation Int J Radiat Oncol Biol Phys. 1980; 6:1731–1733. 160. Fein DA, Marcus RB, Jr., Parsons JT et al. Lhermitte’s sign: incidence and treatment variables influencing risk after irradiation of the cervical spinal cord Int J Radiat Oncol Biol Phys. 1993; 27: 1029–1033. 161. Esik O, Csere T, Stefanits K et al. A review on radiogenic Lhermitte’s sign Pathol Oncol Res. 2003; 9:115–120. 162. Lewanski CR, Sinclair JA, Stewart JS. Lhermitte’s sign following head and neck radiotherapy Clin Oncol (R Coll Radiol). 2000; 12:98–103. 163. Li YQ, Jay V, Wong CS. Oligodendrocytes in the adult rat spinal cord undergo radiation-induced apoptosis Cancer Res. 1996; 56:5417–5422. 164. Lengyel Z, Reko G, Majtenyi K et al. Autopsy verifies demyelination and lack of vascular damage in partially reversible radiation myelopathy Spinal Cord. 2003; 41:577–585. 165. Chao MW, Wirth A, Ryan G et al. Radiation myelopathy following transplantation and radiotherapy for non-Hodgkin’s lymphoma Int J Radiat Oncol Biol Phys. 1998; 41:1057–1061. 166. Schultheiss TE, Stephens LC. Invited review: permanent radiation myelopathy Br J Radiol. 1992; 65: 737–753. 167. Pallis CA, Louis S, Morgan RL. Radiation myelopathy Brain. 1961; 84:460–479. 168. Wang PY, Shen WC, Jan JS. MR imaging in radiation myelopathy AJNR Am J Neuroradiol. 1992; 13:1049–1055. 169. Wang PY, Shen WC, Jan JS. Serial MRI changes in radiation myelopathy Neuroradiology. 1995; 37:374–377. 170. Yasui T, Yagura H, Komiyama M et al. Significance of gadolinium-enhanced magnetic resonance imaging in differentiating spinal cord radiation myelopathy from tumor: case report J Neurosurg. 1992; 77:628–631. 171. Komachi H, Tsuchiya K, Ikeda M et al. Radiation myelopathy: a clinicopathological study with special reference to correlation between MRI findings and neuropathology J Neurol Sci. 1995; 132:228–232. 172. Koehler PJ, Verbiest H, Jager J et al. Delayed radiation myelopathy: serial MR-imaging and pathology. Clin Neurol Neurosurg. 1996; 98:197–201. 173. Michikawa M, Wada Y, Sano M et al. Radiation myelopathy: significance of gadolinium-DTPA enhancement in the diagnosis Neuroradiology. 1991; 33:286–289. 174. Alfonso ER, De Gregorio MA, Mateo P et al. Radiation myelopathy in over-irradiated patients: MR imaging findings Eur Radiol. 1997; 7:400–404. 175. Angibaud G, Ducasse JL, Baille G et al. Potential value of hyperbaric oxygenation in the treatment of post-radiation myelopathies Rev Neurol (Paris). 1995; 151:661–666. 176. Calabro F, Jinkins JR. MRI of radiation myelitis: a report of a case treated with hyperbaric oxygen Eur Radiol. 2000; 10:1079–1084. 177. Allen JC, Miller DC, Budzilovich GN et al. Brain and spinal cord hemorrhage in long-term survivors of malignant pediatric brain tumors: a possible late effect of therapy Neurology. 1991; 41:148–150. 178. Packer RJ, Gurney JG, Punyko JA et al. Long-term neurologic and neurosensory sequelae in adult survivors of a childhood brain tumor: childhood cancer survivor study J Clin Oncol. 2003; 21:3255–3261. 179. Sagar SM, Thomas RJ, Loverock LT et al. Olfactory sensations produced by high-energy photon irradiation of the olfactory receptor mucosa in humans Int J Radiat Oncol Biol Phys. 1991; 20:771–776. 180. Carmichael KA, Jennings AS, Doty RL. Reversible anosmia after pituitary irradiation Ann Intern Med. 1984; 100:532–533. 181. Ophir D, Guterman A, Gross-Isseroff R. Changes in smell acuity induced by radiation exposure of the olfactory mucosa Arch Otolaryngol Head Neck Surg. 1988; 114:853–855. 182. Qiu Q, Chen S, Meng C et al. Observation on the changes in nasopharyngeal carcinoma patients’ olfactory before and after radiotherapy Lin Chuang Er Bi Yan Hou Ke Za Zhi. 2001; 15:57–58 [Abstract]. 183. Lessell S. Friendly fire: neurogenic visual loss from radiation therapy J Neuroophthalmol. 2004; 24:243–250. 184. Parsons JT, Bova FJ, Fitzgerald CR et al. Radiation optic neuropathy after megavoltage external-beam irradiation: analysis of time-dose factors Int J Radiat Oncol Biol Phys. 1994; 30:755–763. 185. Valanconny C, Koenig F, Benchaboune M et al. Complications de la radiothérapie des néovaisseaux de la dégénérescence maculaire liée à l‘âge J Fr Ophtalmol. 2000; 23:151–157. 186. Meyer A, Levy C, Blondel J et al. Neuropathie optique après protonthérapie pour mélanome malin de la choroïde J Fr Ophtalmol. 2000; 23:543–553.

Chapter 16 / Neurologic Complications of Radiation Therapy

285

187. Piquemal R, Cottier JP, Arsene S et al. Radiation-induced optic neuropathy 4 years after radiation: report of a case followed up with MRI Neuroradiology. 1998; 40:439–441. 188. Levin LA, Gragoudas ES, Lessell S. Endothelial cell loss in irradiated optic nerves Ophthalmology. 2000; 107:370–374. 189. Danesh–Meyer HV, Savino PJ, Sergott RC. Visual loss despite anticoagulation in radiation-induced optic neuropathy Clin Experiment Ophthalmol. 2004; 32:333–335. 190. Borruat FX, Schatz NJ, Glaser JS et al. Visual recovery from radiation-induced optic neuropathy. The role of hyperbaric oxygen therapy J Clin Neuroophthalmol. 1993; 13:98–101. 191. Roden D, Bosley TM, Fowble B et al. Delayed radiation injury to the retrobulbar optic nerves and chiasm. Clinical syndrome and treatment with hyperbaric oxygen and corticosteroids Ophthalmology. 1990; 97:346–351. 192. Boschetti M, De Lucchi M, Giusti M et al. Partial visual recovery from radiation-induced optic neuropathy after hyperbaric oxygen therapy in a patient with Cushing disease Eur J Endocrinol. 2006; 154:813–818. 193. Mohamed IG, Roa W, Fulton D et al. Optic nerve sheath fenestration for a reversible optic neuropathy in radiation oncology Am J Clin Oncol. 2000; 23:401–405. 194. Lessell S, Lessell IM, Rizzo JF, III. Ocular neuromyotonia after radiation therapy Am J Ophthalmol. 1986; 102:766–770. 195. Marti-Fabregas J, Montero J, Lopez-Villegas D et al. Post-irradiation neuromyotonia in bilateral facial and trigeminal nerve distribution Neurology. 1997; 48:1107–1109. 196. Diaz JM, Urban ES, Schiffman JS et al. Post-irradiation neuromyotonia affecting trigeminal nerve distribution: an unusual presentation Neurology. 1992; 42:1102–1104. 197. Maesawa S, Salame C, Flickinger JC et al. Clinical outcomes after stereotactic radiosurgery for idiopathic trigeminal neuralgia J Neurosurg. 2001; 94:14–20. 198. Rogers CL, Shetter AG, Fiedler JA et al. Gamma knife radiosurgery for trigeminal neuralgia: the initial experience of The Barrow Neurological Institute Int J Radiat Oncol Biol Phys. 2000; 47:1013–1019. 199. Combs SE, Thilmann C, Debus J et al. Long-term outcome of stereotactic radiosurgery (SRS) in patients with acoustic neuromas Int J Radiat Oncol Biol Phys. 2006; 64:1341–1347. 200. Giese WL, Kinsella TJ. Radiation injury to peripheral and cranial nerves. In Gutin PH, Leibel SA, Sheline GE (eds.). Radiation Injury to the Nervous System. New York: Raven Press 1991: 383–403. 201. DeWys WD, Walters K. Abnormalities of taste sensation in cancer patients Cancer. 1975; 36:1888–1896. 202. Noren G, Greitz D, Hirsch A et al. Gamma knife surgery in acoustic tumours Acta Neurochir Suppl (Wien ). 1993; 58:104–107. 203. Foote RL, Coffey RJ, Swanson JW et al. Stereotactic radiosurgery using the gamma knife for acoustic neuromas Int J Radiat Oncol Biol Phys. 1995; 32:1153–1160. 204. Selch MT, Pedroso A, Lee SP et al. Stereotactic radiotherapy for the treatment of acoustic neuromas J Neurosurg 2004; 101:362–372. 205. Gibb AG, Loh KS. The role of radiation in delayed hearing loss in nasopharyngeal carcinoma J Laryngol Otol. 2000; 114:139–144. 206. McDonald LW, Donovo MP, Plantz RG. Radiosensitivity of the vestibular apparatus of the rabbit Radiat Res. 1966; 27:510–511. 207. Kondziolka D, Nathoo N, Flickinger JC et al. Long-term results after radiosurgery for benign intracranial tumors Neurosurgery. 2003; 53:815–821. 208. Kondziolka D, Lunsford LD, McLaughlin MR et al. Long-term outcomes after radiosurgery for acoustic neuromas N Engl J Med. 1998; 339:1426–1433. 209. Spiegelmann R, Lidar Z, Gofman J et al. Linear accelerator radiosurgery for vestibular schwannoma J Neurosurg. 2001; 94:7–13. 210. Niranjan A, Lunsford LD, Flickinger JC et al. Dose reduction improves hearing preservation rates after intracanalicular acoustic tumor radiosurgery Neurosurgery. 1999; 45:753–762. 211. Ito K, Shin M, Matsuzaki M et al. Risk factors for neurological complications after acoustic neurinoma radiosurgery: refinement from further experiences Int J Radiat Oncol Biol Phys. 2000; 48: 75–80. 212. Berger PS, Bataini JP. Radiation-induced cranial nerve palsy Cancer. 1977; 40:152–155. 213. Cheng VS, Schultz MD. Unilateral hypoglossal nerve atrophy as a late complication of radiation therapy ofhead and neck carcinoma: a report of four cases and a review of the literature on peripheral and cranial nerve damages after radiation therapy Cancer. 1975; 35:1537–1544. 214. Johnston EF, Hammond AJ, Cairncross JG. Bilateral hypoglossal palsies: a late complication of curative radiotherapy Can J Neurol Sci. 1989; 16:198–199. 215. Kang MY, Holland JM, Stevens KR, Jr. Cranial neuropathy following curative chemotherapy and radiotherapy for carcinoma of the nasopharynx J Laryngol Otol. 2000; 114:308–310. 216. King AD, Ahuja A, Leung SF et al. MR features of the denervated tongue in radiation-induced neuropathy Br J Radiol. 1999; 72:349–353. 217. King AD, Leung SF, Teo P et al. Hypoglossal nerve palsy in nasopharyngeal carcinoma Head Neck. 1999; 21:614–619. 218. Stern Y, Marshak G, Shpitzer T et al. Vocal cord palsy: possible late complication of radiotherapy for head and neck cancer Ann Otol Rhinol Laryngol. 1995; 104:294–296. 219. Takimoto T, Saito Y, Suzuki M et al. Radiation-induced cranial nerve palsy: hypoglossal nerve and vocal cord palsies J Laryngol Otol. 1991; 105:44–45. 220. Huang XM, Zheng YQ, Zhang XM et al. Diagnosis and management of skull base osteoradionecrosis after radiotherapy for nasopharyngeal carcinoma Laryngoscope. 2006; 116:1626–1631. 221. Rowin J, Cheng G, Lewis SL, Meriggioli MN. Late appearance of dropped head syndrome after radiotherapy for Hodgkin’s disease Muscle and Nerve. 2006; 34:666–669. 222. Salner AL, Botnick LE, Herzog AG et al. Reversible brachial plexopathy following primary radiation therapy for breast cancer Cancer Treat Rep. 1981; 65:797–802.

286

Part VI / Complications of Cancer Therapy

223. Pierce SM, Recht A, Lingos TI et al. Long-term radiation complications following conservative surgery (CS) and radiation therapy (RT) in patients with early stage breast cancer Int J Radiat Oncol Biol Phys. 1992; 23:915–923. 224. Vega F, Davila L, Delattre JY et al. Experimental carcinomatous plexopathy J Neurol. 1993; 240:54–58. 225. Pradat PF, Poisson M, Delattre JY. Neuropathies radiques Rev Neurol (Paris). 1994; 150:664–677. 226. Stoll BA, Andrews JT. Radiation-induced peripheral neuropathy BMJ. 1966; 11:834–837. 227. Olsen NK, Pfeiffer P, Mondrup K et al. Radiation-induced brachial plexus neuropathy in breast cancer patients Acta Oncol. 1990; 29:885–890. 228. Gillette EL, Mahler PA, Powers BE et al. Late radiation injury to muscle and peripheral nerves Int J Radiat Oncol Biol Phys. 1995; 31:1309–1318. 229. Kori SH, Foley KM, Posner JB. Brachial plexus lesions in patients with cancer: 100 cases Neurology. 1981; 31:45–50. 230. Kori SH. Diagnosis and management of brachial plexus lesions in cancer patients Oncology (Huntingt). 1995; 9:756–760. 231. Thyagarajan D, Cascino T, Harms G. Magnetic resonance imaging in brachial plexopathy of cancer. Neurology. 1995; 45:421–427. 232. Harper CM, Jr., Thomas JE, Cascino TL et al. Distinction between neoplastic and radiation-induced brachial plexopathy, with emphasis on the role of EMG Neurology. 1989; 39:502–506. 233. Cooke J, Powell S, Parsons C. The diagnosis by computed tomography of brachial plexus lesions following radiotherapy for carcinoma of the breast Clin Radiol. 1988; 39:602–606. 234. Fishman EK, Campbell JN, Kuhlman JE et al. Multiplanar CT evaluation of brachial plexopathy in breast cancer J Comput Assist Tomogr. 1991; 15:790–795. 235. Wouter van Es H, Engelen AM, Witkamp TD et al. Radiation-induced brachial plexopathy: MR imaging Skeletal Radiol. 1997; 26:284–288. 236. Qayyum A, MacVicar AD, Padhani AR et al. Symptomatic brachial plexopathy following treatment for breast cancer: utility of MR imaging with surface–coil techniques Radiology. 2000; 214:837–842. 237. Ahmad A, Barrington S, Maisey M et al. Use of positron emission tomography in evaluation of brachial plexopathy in breast cancer patients Br J Cancer. 1999; 79:478–482. 238. Soto O. Radiation-induced conduction block: resolution following anticoagulant therapy Muscle Nerve. 2005; 31:642–645. 239. Gerard JM, Franck N, Moussa Z et al. Acute ischemic brachial plexus neuropathy following radiation therapy Neurology. 1989; 39:450–451. 240. Thomas JE, Cascino TL, Earle JD. Differential diagnosis between radiation and tumor plexopathy of the pelvis Neurology. 1985; 35:1–7. 241. Maier JG, Perry RH, Saylor W et al. Radiation myelitis of the dorsolumbar spinal cord Radiology. 1969; 93:153–160. 242. Bowen J, Gregory R, Squier M et al. The post-irradiation lower motor neuron syndrome neuronopathy or radiculopathy? Brain. 1996; 119 (Pt 5):1429–1439. 243. Lamy C, Mas JL, Varet B et al. Post-radiation lower motor neuron syndrome presenting as monomelic amyotrophy J Neurol Neurosurg Psychiatry. 1991; 54:648–649. 244. Wohlgemuth WA, Rottach K, Jaenke G et al. Radiogenic amyotrophy: cauda equina lesion as a late radiation sequel Nervenarzt. 1998; 69:1061–1065 [Abstract]. 245. Anezaki T, Harada T, Kawachi I et al. A case of post-irradiation lumbosacral radiculopathy successfully treated with corticosteroid and warfarin Rinsho Shinkeigaku. 1999; 39: 825–829 [Abstract]. 246. Hussussian CJ, Mackinnon SE. Post-radiation neural sheath sarcoma of the brachial plexus: a case report Ann Plast Surg. 1999; 43: 313–317. 247. Adamson DC, Cummings TJ, Friedman AH. Malignant peripheral nerve sheath tumor of the spine after radiation therapy for Hodgkin’s lymphoma Clin Neuropathol. 2004; 23:245–255. 248. Foley KM, Woodruff JM, Ellis FT et al. Radiation-induced malignant and atypical peripheral nerve sheath tumors Ann Neurol. 1980; 7:311–318. 249. Shiu MH, Hilaris BS, Harrison LB et al. Brachytherapy and function-saving resection of soft tissue sarcoma arising in the limb Int J Radiat Oncol Biol Phys. 1991; 21:1485–1492. 250. Wanebo JE, Malik JM, VandenBerg SR et al. Malignant peripheral nerve sheath tumors: a clinicopathologic study of 28 cases Cancer. 1993; 71:1247–1253.

17

Neurologic Complications of Chemotherapy Jörg Dietrich,

MD PHD,

and Patrick Y. Wen,

MD

CONTENTS Introduction Drugs That Commonly Cause Neurotoxicity Drugs That Occasionally Cause Neurotoxicity Drugs That Less Frequently Cause Neurotoxicity Hormonal Therapy Biologic Agents Growth Factors Monoclonal Antibodies Small Molecule Inhibitors Other Agents Conclusion References

Summary Neurotoxicity associated with chemotherapy is a common complication of many anticancer agents and frequently constitutes a dose-limiting toxicity. There is a wide range of adverse effects including short-term and long-term complications. With the recent use of more aggressive treatment regimens and prolonged survival of cancer patients, neurologic complications of chemotherapy have been observed with increasing frequency. As a result, the nervous system may be more vulnerable than previously thought, and neurotoxicity may not always be reversible after discontinuation of treatment. Both the central and the peripheral nervous system can be affected. It is critical for the oncologist to recognize treatment-related side effects early and to distinguish these from symptoms related to the progression of the disease. This chapter will discuss the neurotoxic side effect profile of commonly used antineoplastic agents, including traditional cytotoxic agents, hormonal agents, monoclonal antibodies and small molecule inhibitors. Key Words: chemotherapy, toxicity, neurotoxicity, central nervous system, peripheral nervous system, adverse effects, neuropathy, leukoencephalopathy, cognitive function, seizures

1. INTRODUCTION Neurotoxicity of chemotherapy is a common problem of many antineoplastic agents and generally constitutes a dose-limiting side effect. With the recent use of more aggressive treatment regimens and prolonged survival of cancer patients, neurologic complications of chemotherapy have been observed with increasing frequency. Recent years have witnessed a wave of new chemotherapeutic agents introduced to treat cancer patients. In addition to traditional cytotoxic chemotherapeutic agents, there is now a wide range of different drugs From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

287

288

Part VI / Complications of Cancer Therapy

available, including hormonal agents, biologic agents, monoclonal antibodies, and small-molecule signal transduction inhibitors. Many of these agents, such as immune-modulating and biological drugs, are new enough that we have an insufficient understanding about their long-term effects on the central and peripheral nervous system. Cancer therapeutics may affect both the central and peripheral nervous systems, and neurotoxic adverse reactions may be recognized as both acute and delayed treatment complications. This is particularly of concern for long-term survivors, such as in children treated with high-dose chemotherapy for acute leukemia. Despite a large number of reports documenting both acute and prolonged neurotoxicity following chemotherapy, surprisingly little is known about the cellular mechanisms underlying the damage to the nervous system. While many conventional cytotoxic drugs may preferentially target rapidly dividing cells, such as glia cells and endothelial cells, recent studies suggest that the cause of neurotoxicity is far more complex than simply toxic effects on proliferating cells. Most recent data suggests that in addition to lineage-committed progenitor cell populations (including precursor cells for oligodendrocytes, astrocytes and neurons), the nondividing and postmitotic oligodendrocytes may be preferential targets of chemotherapeutic agents (1). Neurotoxic complications may result both from direct toxic effects of the drug on the cells of the nervous system or indirectly from metabolic abnormalities or cerebrovascular disorders induced by chemotherapeutic agents. While the presence of the blood–brain barrier may prevent many hydrophilic agents and large molecules from entering the nervous system, this barrier may be less protective when multiple agents are applied at the same time or sequentially, challenging the integrity of the blood–brain barrier itself. This protective barrier may also be breached by radiation treatment, or if the drug is administered directly into the cerebrospinal fluid or into the cerebral vasculature (e.g., intra-arterial chemotherapy for CNS tumors). As a result, the nervous system may be more vulnerable than previously thought, and neurotoxicity may not always be reversible after discontinuation of treatment. In support of this view is a large body of literature revealing long-term cognitive deficits in survivors of cancer treatment even many years after cessation of treatment. More importantly, neurotoxicity may be subtle and not readily detectable by conventional neuroimaging using computed tomography or magnetic resonance imaging. Some drugs, such as methotrexate, are associated with a relatively high frequency of neurotoxicity, which may be severe and progressive, especially if the drug is administered after radiation therapy. In general, the development of CNS or PNS complications and their severity are dependent on many factors. These include the dose, frequency, and route of administration of the drug; the age of the patient; the presence of other therapies such as prior or concurrent radiation therapy or chemotherapies (e.g., levamisole administered with 5-FU may be more likely to be associated with demyelination); organ dysfunction that may affect drug metabolism (e.g., patients with hepatic or renal dysfunction are at increased risk of CNS toxicity from high-dose cytosine arabinoside); preexisting CNS lesions or dysfunction; and concurrent or drug-induced metabolic abnormalities. Some drugs that do not penetrate into the CNS may still be able to produce CNS complications indirectly. For example, l-asparaginase produces a coagulopathy that may result in venous sinus thrombosis, while vincristine may cause confusion as a result of the syndrome of inappropriate antidiuretic hormone secretion (SIADH). In addition, underlying genetic factors may pose a higher risk of neurotoxicity in certain individuals. Recognition of neurologic complications is important, both because symptoms may be confused with metastatic disease, radiation neurotoxicity, paraneoplastic disorders, or opportunistic infections, and because discontinuation of the drugs may prevent irreversible injury. Patients must be informed of potential side effects affecting the central and peripheral nervous system, and clinicians must recognize signs and symptoms as they emerge in order to prevent disability and decreased quality of life in the absence of tumor progression. This chapter will discuss complications to the peripheral and central nervous system associated with more commonly used chemotherapeutic agents, hormones, biological response modifiers, and targeted molecular agents.

2. DRUGS THAT COMMONLY CAUSE NEUROTOXICITY The neurologic complications of the most common chemotherapeutic drugs are summarized in Table 1.

2.1. Cisplatin. (Cis-diamminedichloroplatinum (II), CDDP, DDP) Cisplatin acts similarly to an alkylating agent, producing its cytotoxic effects by forming DNA crosslinks, impairing DNA synthesis and transcription (2). It is used to treat ovarian, testicular, cervical, bladder, lung,

Chapter 17 / Neurologic Complications of Chemotherapy

289

Table 1 Neurologic Complications of Chemotherapy Acute encephalopathy • Asparaginase • 5-Azacytidine • BCNU • Chlorambucil • Cisplatin • Corticosteroids • Cyclophosphamide • Cytosine arabinoside • Dacarbazine • Docetaxel • Doxorubicin • Etoposide • Fludarabine Neuropathy • 5-Azacytidine • Bortezomib • Capecitabine • Carboplatin • Cisplatin • Cytosine arabinoside • Docetaxel • Etoposide • 5-FU • Fludarabine • Gemcitabine • Hexamethylmelamine • Ifosfamide Seizures • Amifostine • Asparaginase • BCNU • Busulfan • Chlorambucil • Cisplatin • Corticosteroids • Cyclophosphamide • Cytosine arabinoside • Dacarbazine • Docetaxel • Erythropoietin Headaches • Asparaginase • BCNU • Capcitabine • Cetuximab • Cisplatin • Corticosteroids • Cytosine arabinoside • Danazol • Erlotinib • Estramustine • Etoposide • Fludarabine

• • • • • • • • • • • •

5-FU Gemcitabine Hexamethylmelamine Hydroxyurea Ifosfamide Imatinib Interferons IL-1, IL-2 Levamisole Mechlorethamine Methotrexate Misonidazole

• • • • • • • • • • • •

Mitomycin C Nelarabine Paclitaxel Pentostatin Procarbazine Suramin Tamoxifen Thalidomide Thiotepa Tipifarnib TNF Vinca alkaloids

• • • • • • • • • •

Interferon-alpha Misonidazole Nelarabine Nitroimidazole Oprelvekin Oxaliplatin Paclitaxel Pemetrexed Procarbazine Purine analogs (fludarabine, cladribine, pentostatin)

• • • • • • • • •

Sorafenib Sunitinib malate Suramin Taxanes Teniposide (VM-26) Thalidomide Tipifarnib TNF Vinca alkaloids

• • • • • • • • • • •

Etanercept Etoposide 5-FU Gemcitabine Hexamethylamine Hydroxyurea Ifosfamide Interferon IL-2 Letrozole Leuprolide

• • • • • • • • • • •

Levamizole Mechlorethamine Methotrexate Octreotide Paclitaxel Pentostatin Suramin Temozolomide Teniposide Thalidomide Vinca alkaloids

• • • • • • • • • • • •

Gemtuzumab Hexamethylmelamine Hydroxyurea Ibritumomab Imatinib Interferons Interleukins Leuprolide Levamisole Mechlorethamine Methotrexate Mitomycin

• • • • • • • • • • • •

Oprelvekin Plicamycin Procarbazine Rituximab Retinoic acid SU-5416 Tamoxifen Temozolamide Thalidomide Thiotepa TNF Topotecan (Continued)

290

Part VI / Complications of Cancer Therapy

Table 1 (Continued) • 5-FU • Gefitinib • Gemcitabine Vasculopathy and Stroke • Asparaginase • BCNU • Bevacizumab • Bleomycin • Carboplatin • Cisplatin Myelopathy • Cisplatin • Cladarabine • Corticosteroids • Cytosine arabinoside • Docetaxel Ocular toxicity/Visual loss • BCNU • Carboplatin • Chlorambucil • Cisplatin • Cytosine arabinoside • Etanercept Cranial neuropathies • BCNU • Cisplatin • Cytosine arabinoside Posterior reversible leukoencephalopathy • Bevacizumab • Capecitabine • Cisplatin • Cytosine arabinoside • 5-FU Extrapyramidal syndrome • 5-FU • Hexamethylmelamine Aseptic meningitis • Cytosine arabinoside • Levamisole Dementia • BCNU • Carmofur • Cisplatin • Corticosteroids Syncope • Bevacizumab Chronic encephalopathy • BCNU • Cisplatin • Cytosine arabinoside

• Mitotane • Nelarabine • Octreotide

• Tositumomab • Trastuzumab

• • • • • •

Danazol Doxorubicin Erlotinib Erythropoietin Estramustine 5-FU

• • • • • •

• • • • •

Doxorubicin Fludarabine Interferon Methotrexate Mitoxantrone

• Taxotere • Thiotepa • Vincristine

• • • • • •

Etoposide Fludarabine 5-FU Interferon Interleukin Methotrexate

• • • • • •

Interleukin Imatinib mesylate Methotrexate Mitomycin Nelarabine Tamoxifen

Paclitaxel Pentostatin Retinoic acid Tamoxifen Suramin Vinca alkaloids

• Ifosfamide • Methotrexate • Nelarabine

• Vincristine

• • • • •

• Sorafenib • Sunitinib • Vinca alkaloids

Cyclosporin A G-CSF/GM-CSF Methotrexate Oxaliplatin Nelarabine

• Interferon • Ifosfamide

• Retinoic acid • Vincristine

• Methotrexate • Thiotepa • • • •

Cytosine arabinoside Dacarbazine Fludarabine 5-FU

• Interferon • Levamisole • Methotrexate

• Erlotinib

• Nelarabine

• Fludarabine • 5-FU • Ifosfamide

• Interferon • Levamisole • Methotrexate

Abbreviations: BCNU, camustine; IL, interleukin; TNF, tumor necrosis factor; 5-FU, fluorouracil, G-CSF, granulocytpe-colony stimulating factor; GM-CSF, granulocyte-macrophage colony stimulating factor.

Chapter 17 / Neurologic Complications of Chemotherapy

291

gastrointestinal, and head and neck cancers, lymphomas, and medulloblastomas. While it is found in high concentrations in the dorsal root ganglia and peripheral nerves, passage through the blood–brain barrier is considered low (3,4). CSF penetration may be higher, however, in patients with brain tumors, or when cisplatin is used in combination with other chemotherapeutic agents compromising the integrity of the blood–brain barrier. In one study, intravenous administration of cisplatin resulted in a peak CSF concentration as high as 40% of nonprotein bound cisplatin (5). Despite the overall poor penetration of cisplatin across the blood–brain barrier, several CNS complications may occur (4) (Table 2). The main neurologic complication of cisplatin is a neuropathy affecting predominantly large myelinated sensory fibers (6–9). Symptoms primarily result from injury to the dorsal root ganglion. The peripheral nerve may also be affected. The neuropathy is characterized by subacute development of numbness, paresthesias, and occasionally pain in the extremities. Symptoms usually begin distally in extremities and then spread proximally to affect both legs and arms. Proprioception is impaired and reflexes are frequently lost. Pinprick sensation, temperature sensation, and muscle strength are often spared. Nerve conduction studies typically show decreased amplitude of sensory action potentials and prolonged sensory latencies, compatible with a mainly sensory axonopathy. Sural nerve biopsy may show both demyelination and axonal loss. There appears to be a marked individual susceptibility to the development of cisplatin-induced neuropathies. Typically, neuropathies develop in patients following cumulative doses of cisplatin greater than 400 mg/m2 (6,10,11). Increased dose intensity of cisplatin administration does not appear to enhance the severity of the neuropathy. Patients with mild neuropathies can continue to receive full doses of cisplatin. After the neuropathy becomes more severe and begins to interfere with neurologic function, the clinician must decide whether to continue with therapy and risk potentially disabling neurotoxicity, reduce the dose of drug, or discontinue the drug and replace it with less neurotoxic agents. The most appropriate course of action varies with each patient and must take into account factors such as the severity of the neuropathy and the availability of less neurotoxic alternatives. After cessation of chemotherapy, the neuropathy continues to deteriorate for several months in 30% of patients (12). Most patients show improvement, although recovery may remain incomplete. There is no treatment for cisplatin neurotoxicity. Ethiofos (WR-2721) (13), amifostine (14), vitamin E (15), and the ACTH analogue, Org 2766 (16–23), partially protect peripheral nerves from cisplatin neurotoxicity. Arterial infusions of cisplatin in the extremities or neck may produce focal neuropathies. Autonomic neuropathies have also been rarely observed (19). 2.1.1. Cranial Neuropathies Cisplatin may cause ototoxicity, leading to high-frequency sensorineural hearing loss and tinnitus. The toxicity is due to peripheral receptor (hair) loss in the organ of Corti and is related to dose (24). Audiometric hearing loss is present in 74–88% of patients receiving cisplatin, and symptomatic hearing loss occurs in 16–20% of patients. Cranial irradiation probably increases the likelihood of significant hearing loss (25). Other risk factors include concurrent ototoxic drugs such as aminoglycosides, furosemide and ifosfamide (24,26). The hearing loss tends to be worse in children, although they have a slightly greater ability to improve after the drug has been stopped. Neurotrophin 4/5 enhances the survival of cultured spinal ganglion cells in vitro and may have therapeutic value in preventing cisplatin-induced ototoxicity (27). Genetic polymorphisms in glutathione S-transferase, one of the enzymes responsible for the metabolism of cisplatin, may contribute to the observed interindividual differences in the severity of hearing loss caused by cisplatin (28). Table 2 Neurologic Complications of Cisplatin Common • • • • •

Neuropathy Lhermitte’s sign Hearing loss Vestibular toxicity Tinnitus

Less common • • • • •

Encephalopathy Seizures Cerebral infarction Electroyte imbalance Ocular toxicity

292

Part VI / Complications of Cancer Therapy

Cisplatin may also cause a vestibulopathy, resulting in ataxia and vertigo. Previous use of aminoglycosides may exacerbate the vestibulopathy (24,29). Intra-arterial infusion of cisplatin for head and neck cancer produces cranial palsies in approximately 6% of patients (30). Intracarotid infusion of cisplatin may also cause ocular toxicity (31), although these complications may also rarely occur after intravenous administration of the drug. They include retinopathy, papilledema (32), optic neuritis (32) and disturbed color perception due to dysfunction of retinal cones (33). Other complications of intraarterial cisplatin include headaches, confusion and seizures (34). 2.1.2. Spinal Cord Involvement (Lhermitte’s Sign) This symptom, characterized by paresthesias in the back and extremities with neck flexion, is seen in 20–40% of patients receiving cisplatin. Patients tend to develop this after several weeks or months of treatment. Neurologic exam and MR scans are usually normal. Somatosensory evoked responses may suggest spinal cord involvement (12). Lhermitte’s sign usually resolves spontaneously several months after the drug has been discontinued (35,36). It is believed to result from transient demyelination of the posterior columns (26). Very rarely, a true myelopathy has been reported (17,35). 2.1.3. Other Complications Rarely, cisplatin produces an encephalopathy resulting in seizures and focal neurologic symptoms, including cortical blindness (19,37–39). The encephalopathy is associated with reversible abnormalities in the white matter of the occipital, parietal, and frontal lobes and clinically resembles a reversible posterior leukoencephalopathy syndrome (Fig. 1) (38). The encephalopathy tends to be more common after intra-arterial administration of the drug (40). It has to be distinguished from a metabolic encephalopathy that may result from water intoxication caused by prehydration (41), or from renal impairment, hypomagnesemia resulting from impaired magnesium absorption in the proximal renal tubule (42,43), hypocalcemia, and SIADH (44) that may follow treatment with cisplatin. Cisplatin can also cause vascular toxicity resulting in strokes (19,45–47). This usually occurs when cisplatin is administered in combination with other agents. The underlying pathophysiology remains obscure; however, both acute and long-term vascular toxicity may be associated with cisplatin administration (48,49). It is unclear whether this toxicity is related to a direct effect of the drug on endothelial cells, hypomagnesemia-induced vasospasm, coagulopathy, vasculitis, or to other agents that are given concomitantly, such as bleomycin (19,46,47,50). Another rare complication is loss of taste (19). Interestingly, cisplatin may cause long-term adverse effects on cognitive function (51). The exact mechanisms of cognitive dysfunction are unclear.

2.2. Cytosine Arabinoside (Ara-C, Cytarabine) Ara-C is a pyrimidine antimetabolite, which is converted by deoxycytidine kinase into its active metabolite Ara-CTP. Ara-CTP acts as a competitive inhibitor of DNA polymerase and incorporates itself into the DNA molecule resulting in premature chain termination. While its oral absorption is poor, Ara-C distributes well in CSF after intrathecal application with high CSF levels for at least 24 hrs. The half-life is much prolonged with use of liposomal Ara-C and cytotoxic CSF levels are maintained for up to two weeks (52). Blood–brain barrier penetration is good after intravenous application, and about 50% of plasma Ara-C levels are detected in CSF. It is used in the treatment of leukemias, lymphomas, and leptomeningeal metastases. This agent is considered to have mild neurotoxicity when used at conventional doses. High doses (1–3g/m2 every 12–24 hrs) cause an acute cerebellar syndrome in 10–25% of patients (53–55), presenting with ataxia, dysarthria, nystagmus, lethargy, and confusion. The symptoms may develop several days to a week after initiation of treatment (56). The degree of ataxia and dyscoordination is variable and can range from mild dysmetria to severe ataxia with inability to sit or walk. In addition to cerebellar syndromes, Ara-C may result in seizures (57). Patients above the age of 50 (58) with abnormal liver or renal function (59), underlying neurologic dysfunction, or receiving more than 30 g of the drug are especially likely to develop neurotoxicity. Neuroimaging studies may show T2/FLAIR hyperintensities, white matter abnormalities, and eventually cerebellar atrophy (Fig. 2). MRI in patients with acute toxicity may demonstrate multifocal T2/FLAIR hyperintensities involving both gray and white matter and may resemble a picture of reversible posterior leukoencephalopathy (60). Cerebrospinal fluid is usually unremarkable or may show a mild increase in cell count. EEG may show slowing. Pathologic changes are predominately seen in the cerebellum, where widespread loss of Purkinje cells can be

Chapter 17 / Neurologic Complications of Chemotherapy

293

Fig. 1. Acute cisplatin toxicity. The MRI with sagittal (A) and axial (B) T1-weighted and gadolinium enhanced sequences shows multiple small hemorrhages and small necrotic cysts with minor focal contrast enhancement and small edema restricted to the corpus callosum. The 38-year-old developed significant lethargy and confusion after the second cycle of cisplatin-based chemotherapy (cisplatin, ifosfamide, and etoposide) for testicular cancer.

found. No specific treatment is available, but the drug should be discontinued immediately. While the cerebellar syndrome resolves spontaneously in some patients, neurotoxicity may be irreversible in others. Avoidance of very high doses of the drug, especially in patients with renal impairment, has led to a decline in the incidence of this syndrome (61). The risk of neurotoxicity is significantly increased with higher doses and increased frequency of administration of cytosine arabinoside. In addition, increased frequency of neurotoxicity is frequently seen when other neurotoxic drugs are applied together with Ara-C, such as methotrexate. In addition to mental status changes, cerebellar toxicity and seizures (57), there have been reports on ocular toxicity (62), lateral rectus palsy, bulbar and pseudobulbar palsy (63), Horner’s syndrome, anosmia (64,65), papilledema (66), aseptic meningitis (19,67), locked-in syndrome (68), encephalitis (66), and extrapyramidal syndromes (69–71). Ara-C can cause myelopathy similar to that seen after intrathecal methotrexate (72,73). While CNS toxicity of Ara-C has been clinically well recognized, peripheral nervous system toxicity is rare. There have been reports of sensory neuropathy after high-dose applications (74–76) or when applied concomitantly with other neurotoxic drugs, such as daunorubicin, l-asparaginase, or fludarabine (77,78). Peripheral neuropathies can be demyelinating (64,79) or axonal (80), and fatal cases have been described (78,80).

294

Part VI / Complications of Cancer Therapy

Fig. 2. Leukoencephalopathy after high-dose Ara-C and indomethacin. The axial MRI reveals extensive and diffuse T2/FLAIR air hyperintensities affecting bilateral subcortical white matter consistent with diffuse leukoencephalopathy. Two years earlier, this 44-year-old patient received high-dose chemotherapy with cytosine arabinoside/indomethacin and bone marrow transplantation for acute myeloid leukemia. The patient became increasingly symptomatic with progressive cognitive impairment and recurrent falls.

Peripheral toxicity following Ara-C may also present as plexopathies (81). The mechanisms of peripheral toxicity are unclear and both direct toxicity to Schwann cells and autoimmune reactions have been discussed to play a role (79). The mechanism of Ara-C–associated central nervous system toxicity is also poorly understood. Ara-C may induce the formation of reactive oxygen species resulting in DNA damage and p53-dependent apoptosis (82). A redox-related cellular toxicity is supported by recent data showing that the potent anti-oxidant N-acetyl-cysteine may prevent Ara-C–induced neurotoxicity (J.D., unpublished observations; (83)). While it has been shown that Ara-C is preferentially toxic to cerebellar Purkinje cells and cerebellar granule neurons (55,84,85), more recent data suggests that Ara-C targets both lineage-committed progenitor cell populations and nondividing oligodendrocytes—the myelin-forming cells in the central nervous system (1). Thus, some of the neurotoxic adverse reactions and symptoms seen in patients may therefore be a direct consequence of both oligodendrocyte toxicity and impairment of progenitor self-renewal in the germinal zones of the CNS.

2.3. Ifosfamide Ifosfamide is an alkylating agent that is metabolically activated in the liver to ifosfamide mustard and acrolein. The ifosfamide mustard produces cross-linking of DNA strands. It is an analog of cyclophosphamide, with a similar

Chapter 17 / Neurologic Complications of Chemotherapy

295

systemic toxicity profile. Unlike cyclophosphamide, it produces an encephalopathy in 20–30% of patients (86–89). Encephalopathy may especially occur with single dose applications of greater than 2500 mg/m2 . Symptoms of CNS toxicity include lethargy, confusion, hallucinations, cerebellar dysfunction, seizures, cranial nerve palsies, extrapyramidal signs, and occasionally coma (90–93). Ifosfamide-induced encephalopathy may start hours or days after administration of the drug and usually resolves completely after several days (94). Electroencephalography initially shows mild slowing and then high voltage rhythmic delta (95). The development of encephalopathy is usually a relative contraindication to using the drug again, although some patients have been successfully retreated without recurrence of the encephalopathy (95). The encephalopathy is thought to result from accumulation of chloracetaldehyde, one of the breakdown products of ifosfamide. Patients at increased risk for the encephalopathy include those with renal dysfunction, low serum albumin (96,97), underlying brain disease (98), concurrent phenobarbital treatment (99), prior treatment with other chemotherapeutic agents such as cisplatin (88), and previous encephalopathy following ifosfamide treatment (86). There have been reports that methylene blue may be useful in preventing or treating ifosfamide encephalopathy by inhibiting monoamine oxidases (100–103). Benzodiazepines have also been reported to produce rapid clinical and electrographic improvement (104). For most patients no specific treatment is necessary and the encephalopathy usually improves over time. Rarely, however, neurological deficits are irreversible (94) and deaths have been reported (105,106).

2.4. Methotrexate (Amethopterin, MTX) Methotrexate is a dihydrofolate reductase inhibitor, preventing the conversion of folic acid to tetrahydrofolate, which is required for purine and thymidine synthesis. Through folate depletion, methotrexate inhibits DNA, RNA, and protein synthesis. Methotrexate is used in the treatment of a wide range of cancers, including leukemias, lymphomas (both Hodgkin and non-Hodgkin lymphomas), breast cancer, lung cancer, sarcomas, central nervous system lymphoma, and leptomeningeal metastases. Methotrexate crosses the blood–brain barrier relatively poorly and only about 5–10% of plasma levels are detected in cerebrospinal fluid after conventional dosages. However, significant CNS concentrations can be achieved when the drug is administered intrathecally or when high intravenous doses (> 500 mg/m2 ) are used (19). Intrathecal application can also result in reverse methotrexate diffusion into plasma with associated cytotoxic plasma levels. The degree of neurotoxicity is determined by the dosage, route of administration, and the use of other conconcomitant therapeutic modalities with overlapping neurotoxicities including other chemotherapeutic agents and irradiation. With a triphasic half-life pattern between 45 min (T1/2 ) and 10 hrs (T1/2 ), methotrexate is primarily excreted in the urine. Other drugs that compete with methotrexate for urinary excretion (e.g., salicylates, sulfonamides, phenytoin, and penicillin) may therefore impair its renal clearance and can result in increased toxicity. Both acute and chronic neurotoxicity following intrathecal, intravenous, and oral application have been clinically well described. 2.4.1. Intrathecal Methotrexate Toxicity Aseptic meningitis is the most common neurotoxicity associated with intrathecal methotrexate therapy (107–109). The incidence ranges from 10% to 50% of patients with evidence of cumulative toxicity following multiple rounds of intrathecal application (19,108,109). Symptoms and signs of CNS toxicity usually begin 2–4 hrs after the drug is injected and may last for several days. The syndrome is characterized by fever, headaches, nuchal rigidity, back pain, nausea, vomiting, and lethargy and is indistinguishable from other types of chemical meningitis. The CSF shows a lymphocytic or monocytic pleocytosis and an elevated protein. The symptoms are usually self-limited and require no specific treatment. While symptoms are self-limiting in most patients, there have been reports of delayed necrotizing leukoencephalopathy several months after treatment, especially in patients receiving high cumulative doses of intrathecal methotrexate combined with whole-brain radiotherapy (110). Aseptic meningitis can be prevented to some extent by injecting methotrexate with hydrocortisone or using oral corticosteroids. Some patients who developed aseptic meningitis have been subsequently retreated with methotrexate without problems.

296

Part VI / Complications of Cancer Therapy

Fig. 3. Acute methotrexate toxicity. The axial MRI shows bihemispheric T2/FLAIR hyperintensities with frontal and occipital accentuation after intrathecal methotrexate injection in a 19-year-old male with acute myeloid leukemia. The patient developed mental status changes and confusion 1–2 hours after methotrexate injection.

Transverse myelopathy, a much less common complication of intrathecal methotrexate, is characterized by back or leg pain followed by paraplegia, sensory loss, and sphincter dysfunction (19,111,112). The symptoms usually occur between 30 min and 48 hrs after treatment but may occur up to 2 weeks later. The majority of cases show clinical improvement, but the extent of recovery is variable (113). The exact mechanisms of metothrexate-related CNS toxicity are poorly understood, and both direct and indirect mechanisms due to folate depletion must be postulated. Pathologically there is vacuolar demyelination and necrosis in the spinal cord, without inflammatory or vascular changes (114). This complication is more common in patients receiving concurrent radiotherapy or frequent treatments of intrathecal methotrexate, suggesting cumulative neurotoxicity. Rarely, intrathecal methotrexate produces an acute encephalopathy (Fig. 3), especially if CSF outflow is obstructed (110,115), or the methotrexate is injected directly into cerebral white matter as a result of a misplaced ventricular catheter (116). Additional risk factors for methotrexate-induced encephalopathy and neurotoxicity have been suggested, including low pretreatment folate level, high plasma homocysteine levels, and the presence of MTHFR (methyl-tetrahydrofolate reductase) gene mutation (117). Methotrexate can also cause seizures (118), acute (119,120), and subacute focal neurologic deficits (121,122), cranial nerve palsies, radiculopathy (123), neurogenic pulmonary edema, and sudden death (70,109,124). Acute and subacute neurotoxicity and leukoencephalopathy with stroke-like focal deficits may be associated with abnormal MRI findings, such as diffusion-weighted imaging (125) hyperintensities that may not be confined to typical vascular territories. DWI abnormalities can be seen in subcortical or deep periventricular white matter, corpus callosum, cortex, cerebellum, and thalamus (119,122,126–128). It has been suggested that DWI changes in MTX-associated neurotoxicity represent reversible cerebral dysfunction with associated cytotoxic edema and metabolic derangement (122,128) rather than ischemic structural injury. Accidental overdoses of methotrexate (> 500 mg; with detectable methotrexate concentrations in CSF as high as 16.5 mM) usually result in severe myelopathy, seizures, encephalopathy, or death (129,130). The use of rapid CSF drainage (131), ventriculolumbar perfusion (132,133), carboxypeptidase G2 (125,130), highdose leucovorin (129,131), dextromethorphan (a noncompetitive antagonist of the N-methyl-1-aspartate (NMDA) receptor) (134), and alkaline diuresis have been used to antagonize the neurotoxic effects of methotrexate and have allowed some patients to survive. 2.4.2. Weekly Low-Dose Methotrexate Neurotoxicity Up to 20% of patients receiving weekly low-dose methotrexate may experience headaches, dizziness, dysphoria, and subtle cognitive impairment (135). Both renal insufficiency and older age have been reported as associated risk factors for neurotoxicity. Oral methotrexate may also result in acute focal neurological deficits (136) and abnormal imaging findings consistent with reversible posterior leukoencephalopathy (137). Symptoms usually resolve when the methotrexate is discontinued.

Chapter 17 / Neurologic Complications of Chemotherapy

297

2.4.3. High-dose Methotrexate Neurotoxicity High-dose methotrexate may cause acute, subacute, or chronic neurotoxicity, and has been clinically well described in patients treated for CNS lymphoma and children treated for acute leukemia. Acute high-dose methotrexate neurotoxicity is characterized by somnolence, confusion, and seizures within 24 hrs of treatment. Symptoms usually resolve spontaneously without sequelae and patients can often continue to receive this drug (19,109,138). In children, administration of high-dose methotrexate can produce a reversible acute confusional state and cortical blindness associated with increased T2 signal in the parietal-occipital lobes on MRI (139). These patients all had hypomagnesemia, raising the possibility that electrolyte abnormalities may have been a significant contributing factor. Weekly treatments with moderate- to high-dose methotrexate may produce a subacute “stroke-like” syndrome characterized by transient focal neurologic deficits, confusion, and occasionally seizures (140,141). Typically, the disorder develops several days after high-dose methotrexate, lasts between minutes and days, and resolves spontaneously without sequelae. Neuroimaging studies are usually normal, although nonenhancing hyperintense T2 lesions in the white matter have been reported (142). CSF is typically normal but EEG may show diffuse slowing. Methotrexate may be subsequently administered without the encephalopathy recurring. The pathogenesis of this syndrome is unknown, but may be related to reduced cerebral glucose metabolism (143) or reduced biogenic amine synthesis (144). 2.4.4. Leukoencephalopathy The major delayed complication of methotrexate therapy is a leukoencephalopathy (109,145–149) (Fig. 4). Although this syndrome may be produced by methotrexate alone, it is exacerbated by radiotherapy, especially if radiotherapy is administered before or during methotrexate therapy. The leukoencephalopathy usually occurs following repeated administration of intrathecal methotrexate or high-dose intravenous methotrexate, but has also been described after standard-dose intravenous methotrexate (145). While the development of acute methotrexate neurotoxicity does not predict subsequent leukoencephalopathy, there have been reports of leukoencephalopathy developing in patients with prior methotrexate induced aseptic

Fig. 4. Leukoencephalopathy. A 75-year old woman with primary central nervous system lymphoma was treated with CHOP, ten doses of intraventricular methotrexate, and fractionated whole brain radiotherapy (5040 cGy in 28 fractions). Her tumor responded and never recurred. Three years later, she noted moderate short-term memory deficits and gait unsteadiness. MRI (axial T2-weighted image) demonstrated extensive periventricular white matter changes. The patient’s dementia progressed, and she developed rigidity and mutism prior to her death one year later.

298

Part VI / Complications of Cancer Therapy

meningitis (110). The clinical features are characterized by the gradual development of cognitive impairment months or years after treatment with methotrexate. This ranges from mild learning disabilities to severe progressive dementia together with somnolence, seizures, ataxia, and hemiparesis (150). Children with leukemia treated with intrathecal methotrexate and radiation therapy or high-dose methotrexate show significant deterioration in IQ (19,151). CT and MRI scans reveal focal or diffuse white matter lesions that may progress to affect the entire white matter during serial follow up (152) (Fig. 5). The presence of contrast enhancement has been shown to correlate with areas of tissue necrosis and appears to consititute a poor prognostic factor (152,153). Leukoencephalopathy eventually is associated with cerebral atrophy and enlarged ventricular space (154,155). CSF may show increased myelin basic protein concentration as a result of myelin breakdown (156). Pathologic lesions range from loss of oligodendrocytes with focal or diffuse demyelination and reactive gliosis to a diffuse necrotizing leukoencephalopathy (145,157). There is demyelination, axonal swelling, dystrophic mineralization of axonal debris and fibrinoid necrosis of small blood vessels (19,145,154,157,158). There is usually a remarkable absence of inflammatory cellular response and paucity of macrophage reaction (145). Occasionally, children may have a mineralizing microangiopathy, characterized by calcification of capillaries and venules, especially in the basal ganglia (159). The clinical course is variable. While some patients stabilize, others show a progressive course that eventually may lead to death. No effective treatment is available. The cause of the leukoencephalopathy is unknown. Possibilities include disruption of myelin metabolism (160), depletion of reduced folates in the brain (161,162), injury to cerebral vascular endothelium possibly mediated through elevated homocysteine levels (117,162,163) with increasing blood–brain barrier permeability, inhibition of cerebral glucose or protein metabolism (164), or inhibition of catecholamine synthesis (19,143). Recent studies suggest that genetic polymorphisms for methionine metabolism, which is required for myelination, may constitute a risk factor for methotrexate-associated neurotoxicity (165). It is likely that cranial

Fig. 5. Disseminated necrotizing leukoencephalopathy. This 44-year old man with Burkitt’s lymphoma developed progressive lethargy and subsequent coma weeks after receiving fractionated radiotherapy to the skull base and eight doses of intraventricular methotrexate for lymphomatous involvement of the right cavernous sinus. Multiple CSF exams demonstrated no leptomeningeal lymphoma despite his neurologic deterioration. Persistent vegetative state ensued, and he died two months later from systemic relapse. MRI (coronal, T1 with gadolinium) demonstrated multiple scattered punctate foci of abnormal signal, particularly in the deep gray nuclei, with contrast enhancement. Post-mortem analysis revealed multiple discrete, microscopic foci of demyelination, axonal loss, and necrosis distributed in a random manner throughout the white matter and gray/white interface. The foci contained a moderate to large number of CD68-immunoreactive foamy macrophages and a scant number of perivascular lymphocytes. Although there was evidence of dural lymphoma, there was no leptomeningeal tumor identified.

Chapter 17 / Neurologic Complications of Chemotherapy

299

irradiation potentiates the toxic effects of methotrexate either by direct radiation-induced toxicity or by disrupting the blood-brain barrier, allowing higher concentrations of methotrexate to enter the brain.

2.5. Oxaliplatin Oxaliplatin is a third-generation platinum complex that has activity against cisplatin-resistant cancer (166). Its main use has been in the treatment of cisplatin-resistant colorectal cancer. Similar to cisplatin-based protocols, the occurrence of peripheral neuropathy is the dose-limiting toxicity (167,168), which may occur in up to 50–90% of treated patients (169–171). Transient acute paresthesias and dysesthesias are seen in the majority of patients during or immediately after the end of the infusion and may be associated with muscular contractions of the extremities or the jaw. Symptoms usually resolve within 24 hrs and calcium and magnesium supplementation have been shown to be effective in treating and reducing the severity of neuropathic symptoms (171–173). It has been suggested that functional alterations in membrane ion channels play a critical role in the acute neuropathic syndrome. A more serious adverse reaction occurs after long-term administration of oxaliplatin with sensory neuropathy and associated sensory ataxia and functional impairment, similar to the neuropathy seen with cisplatin. This type of neurotoxicity correlates with the cumulative dose of oxaliplatin (174). Both amifostine and carbamazepine have been shown to have benefit in treatment of oxaliplatin-associated neuropathy (171,175,176). Oxcarbazepine, a structural analog of carbamazepine, has been evaluated in a randomized trial, in which 32 patients were randomly assigned to receive oxaliplatin and fluorouracil plus leucovorin, with or without oxcarbazepine. The incidence of peripheral neuropathy was significantly decreased with oxcarbazepine (31 versus 75 percent without oxcarbazepine) (177). Other agents in studies include xaliproden (178), glutathione (179), and glutamine (180). Other reports on oxaliplatin-associated neurotoxicity include visual disturbances, papilledema (181), facial paresthesias (182), seizures, and posterior reversible leukoencephalopathy (183,184).

2.6. Taxanes: Paclitaxel and Docetaxel Taxanes are used to treat a variety of cancers including ovary, breast, and non-small cell lung cancers. They contain a plant alkaloid that inhibits microtubule function, leading to mitotic arrest (185,186). Paclitaxel is also a radiation sensitizer (187). Its main toxicity is a dose-limiting predominantly sensory peripheral neuropathy with numbness, tingling, and pain (188), which occurs in the majority (approximately 60%) of patients receiving 250 mg/m2 of the drug (189). The neuropathy is reversible in most cases over the course of several months in the absence of drug exposure. Some patients develop arthralgias and myalgias beginning 2–3 days after a course of paclitaxel and lasting 2–4 days. Less commonly, paclitaxel can result in motor neuropathies that predominantly affect proximal muscles (190). Vitamin E and N-acetyl carnitine may reduce the severity of the neuropathy. Neuropathies are less common with docetaxel, but some patients develop sensory and motor neuropathies similar to paclitaxel (190,191). Because taxanes do not cross the blood–brain barrier to any significant degree (192), CNS toxicities are rare. The most common CNS complications are visual disturbances, including transient scintillating scotomas during infusion of the drug, and visual loss. The visual loss may be due to involvement of the optic nerves as visual evoked potentials are abnormal. Most patients show good recovery, although visual loss can be permanent (193). In rare cases, taxanes may cause seizures, transient encephalopathies (194–199), facial numbness, Lhermitte’s sign (200), or phantom limb pain in patients with prior amputation (201). High-dose paclitaxel (> 600 mg/m2 ) can lead to an acute encephalopathy and death between 7 and 23 days after treatment (198,202).

2.7. Thalidomide Thalidomide was introduced in Europe in 1954 as a sedative/hypnotic agent, but was withdrawn in 1961 due to the high incidence of limb malformations in children of women who took the drug. In 1998, the FDA approved thalidomide for treatment of erythema nodosum leprosum. In preclinical studies, thalidomide has shown potent anti-angiogenic effects. Based on this property, it has been used in clinical trials for multiple myeloma (203), gliomas (204), Kaposi’s sarcoma (205), and breast cancer (206). The most common side effect is somnolence, affecting between 45% and 55% of patients. Many patients develop tachyphylaxis to this side effect with

300

Part VI / Complications of Cancer Therapy

improvement after 2–3 weeks. Seizures have occurred in a minority of patients with gliomas with a history of seizures in the past. Thalidomide is associated with significant neuropathy. A predominantly sensory axonal neuropathy occurs in 3–32% of patients with prolonged use (207). This may improve slightly with discontinuation of the medication.

2.8. Vinca Alkaloids: Vincristine, Vinblastine, Vindesine, and Vinorelbine Vinca alkaloids include both natural alkaloids, such as vincristine and vinblastine, and semi-synthetic compounds, such as vindesine and vinorelbine. These agents bind to tubulin and prevent microtubule formation, thereby arresting cells in metaphase. Vincristine is a vinca alkaloid derived from the periwinkle plant used to treat many cancers, including leukemia, lymphomas, sarcomas, and brain tumors. Its main neurotoxicity, which is doselimiting, is an axonal sensory-motor neuropathy presenting with paresthesias, weakness, and loss of deep tendon reflexes. Neuropathy is thought to result from disruption of the microtubules within axons and interference with axonal transport. Sural nerve biopsies may reveal both axonal degeneration and segmental demyelination (208). The neuropathy is usually reversible over the course of several months in the absence of the offending drug; however, severe cases with subsequent development of quadriparesis have been described (209). Unfortunately, no neuroprotective agent has been identified for clinical practice to treat or prevent vinca-alkaloid associated neuropathy. In addition to a symmetric peripheral neuropathy, mononeuropathies (including peroneal and femoral nerve) and cranial neuropathies may occasionally be caused by vincristine. The most common involved nerve is the oculomotor nerve, resulting in ptosis and ophthalmoplegia. Other nerves that may be involved include the optic nerve (210) with associated visual impairment, recurrent laryngeal nerve, facial nerve, and auditory nerve. Vincristine may also cause retinal damage and night blindness. Some patients may experience jaw and parotid pain. Autonomic neuropathy may be seen with vincristine and may present as orthostatic intolerance, impotence, and constipation. CNS complications are rare as vincristine only poorly penetrates the blood–brain barrier. Accidental administration of vincristine into the CSF has been shown to produce a rapidly ascending myelopathy, coma and death (211–214), and may only be survivable in rare cases with emergency neurosurgical intervention (215). Rarely, vincristine may cause SIADH, resulting in hyponatremia, confusion, and seizures (20). CNS complications unrelated to SIADH may also occur, including seizures (213), encephalopathy, transient cortical blindness (216), reversible posterior leukoencephalopathy (217,218), ataxia, athetosis, tremor, and parkinsonism (17,19,70). The related vinca alkaloids—vindesine, vinblastine, and vinorelbine—tend to have less neurotoxicity. This may be related to differences in lipid solubility, plasma clearance, terminal half-life, and sensitivities of axoplasmic transport (17,19). Vinorelbine, a semisynthetic analogue of vinblastine, is being used for patients with breast and lung cancer. Like vincristine, vinorelbine inhibits microtubule assembly but has less affinity for neural tissue and is less neurotoxic.

3. DRUGS THAT OCCASIONALLY CAUSE NEUROTOXICITY 3.1. 5-Fluorouracil 5-Fluorouracil (5-FU) is a fluorinated pyrimidine that disrupts pyrimidine synthesis, DNA synthesis, and protein synthesis by inhibiting thymidylate synthetase. It is used to treat many cancers, including gastrointestinal, breast, pancreas, prostate, renal, and head and neck cancers. While 5-FU readily crosses the blood–brain barrier, CNS toxicity is usually limited primarily to patients receiving high doses of the drug (> 15 mg/kg/week) (70). An acute cerebellar syndrome occurs in approximately 5% of patients (71,219). This usually begins weeks or months after treatment and is characterized by the acute onset of ataxia, dysmetria, dysarthria, and nystagmus. Neuroimaging and CSF studies are usually normal. The drug should be discontinued in any patient who develops a cerebellar syndrome. Over time these symptoms usually resolve completely. Retreatment may result in recurrence of the symptoms (220). The development of a cerebellar syndrome may be partly explained by the fact that 5-fluorouracil readily crosses the blood–brain barrier, with the highest concentrations found in the cerebellum. The etiology is unknown, but 5-FU appears to be preferentially toxic to Purkinje and granule cells in the cerebellum (219). 5-Fluorouracil can also produce acute and subacute encephalopathies (221,222). This is

Chapter 17 / Neurologic Complications of Chemotherapy

301

sometimes associated with hyperammonemia and tends to occur more commonly in patients with dehydration and infections (223). Other 5-fluorouracil-associated neurotoxic adverse effects include optic neuropathies (224), eye movement abnormalities (225), focal dystonias, cerebrovascular disorders (226), parkinsonian syndromes (17,227), and seizures (228). Intracarotid infusion of 5-FU may cause somnolence, ataxia, and upper motor neuron signs. Patients with decreased dihydropyrimidine dehydrogenase activity are at increased risk of developing severe neurological toxicity following 5-FU chemotherapy (229). The administration of 5-fluorouracil with other drugs may increase the incidence of neurotoxicity. The co-administration of 5-fluorouracil and allopurinol, N-phosphonoacetyl-laspartate (20), thymidine, doxifluridine, carmofur, or tegafur has been reported to cause encephalopathies and cerebellar syndromes (17,230). The combination of 5-fluorouracil and levamisole used to treat colon cancer has been occasionally associated with delayed multifocal and partially enhancing demyelinating lesions (231–235). Biopsy of these lesions shows multifocal demyelination, relative axonal sparing, and perivascular lymphocytic infiltration. A similar syndrome of multifocal leukoencephalopathy has also been reported after capecitabine, a 5-FU prodrug (236,237). MRI may show increased signals on FLAIR, T2, and DWI sequences in cerebral white matter tracts. The cause of these lesions is unknown, although an immune etiology has been suggested. These lesions usually improve with corticosteroids and discontinuation of the drugs. The importance of recognizing this demyelinating syndrome is that the cerebral lesions may be mistaken for brain metastases.

3.2. l-Asparaginase Asparaginase is an enzyme that hydrolyses l-asparagine to aspartic acid and ammonia, depleting the extracellular supply of l-asparagine, indirectly inhibiting protein synthesis in tumor cells dependent on asparagine. It is used mainly to treat acute lymphocytic leukemia. Neurotoxicity with asparaginase is rare, as it does not readily cross the blood–brain barrier. In early studies, when higher doses of asparaginase were frequently used, a reversible encephalopathy with lethargy and somnolence was observed in 33–60% of patients (70,238–241). The encephalopathy was possibly related to hepatic toxicity (240,241). l-asparaginase is currently used at much lower doses with the consequence that acute neuotoxicity, such as encephalopathy, is rarely seen. l-asparaginase interferes with coagulation homeostasis, rarely leading to thrombotic or hemorrhagic cerebrovascular complications (242,243). It depletes serum levels of several hemostatic factors, including antithrombin III, protein C and S, fibrinogen, factors IX and XI and fibrinolytic enzymes such as plasminogen (70). The most common cerebrovascular complication is venous sinus thrombosis with cerebral infarction (16,244,245) (Fig. 6). The sagittal sinus is usually involved but other sinuses may be affected. These complications typically occur after several weeks of treatment. Patients may present with headaches, seizures, and focal neurologic deficits. There may be papilledema as a result of increased intracranial pressure. MRI may show venous infarction, which is often hemorrhagic, and MR venography demonstrates decreased or absent flow in the affected sinus (246,247). Treatment is controversial but anticoagulation with heparin is generally recommended (70). Others have proposed treatment with fresh-frozen plasma (16). Steroids and opioids may be needed to relieve the headache. The prognosis for recovery is generally good, although recurrent episodes of venous thrombosis can occur with repeat treatment with l-asparaginase. Rarely, asparaginase can cause seizures (248). The related pegasparaginase has similar neurotoxicity. This drug can also cause lethargy and somnolence.

3.3. Pentostatin (2 -Deoxycoformycin) Pentostatin is used for the treatment of a variety of acute and chronic leukemias. The drug belongs to the family of purine analogs. Antineoplastic effects are medicated through inhibition of adenosine deaminase and depletion of nicotinamide adenine dinucleotide, which blocks DNA repair. Various drug interactions have been described that may result in increased toxicity (e.g., allopurinol). Neurotoxic adverse effect include lethargy and fatigue that are commonly seen. Higher doses can cause a severe encephalopathy, seizures, and coma (249,250). In addition, optic neuritis, photophobia, and tinnitus have been described as pentostatin-related complications (251).

302

Part VI / Complications of Cancer Therapy

Fig. 6. Acute l-asparaginase toxicity. Intracerebral hemorrhage after cortical vein thrombosis in a 41-year-old female during induction chemotherapy with l-asparaginase for acute lymphoic leukemia. The patient developed dizziness, left-sided weakness, and headaches. CT imaging (without contrast) revealed a right fronto-parietal hemorrhage. Additional imaging with CT angiography and CT venogram demonstrated the presence of a right cortical vein thrombosis and left vertebral artery occlusion (not shown). (Generously provided by Dr. Darren Volpe, Brigham and Women’s Hospital, Boston, MA.).

4. DRUGS THAT LESS FREQUENTLY CAUSE NEUROTOXICITY 4.1. Anthracycline Antibiotics: Doxorubicin, Adriamycin, Daunorubicin, Epirubicin, Idarubicin, and Mitoxantrone Doxorubicin is an anthracycline antibiotic used to treat a variety of cancers including hematopoetic malignancies, adenocarcinomas, melanoma, lung cancer, gastrointestinal cancer, and breast cancer. Doxorubicin is an inhibitor of topoisomerase I and II, and intercalates DNA. Anthracyclines can cause arrhythmias and cardiomyopathies, which in turn can result in cerebrovascular complications (252). Doxorubicin in combination with cyclosporine can lead to coma and death (249). Accidental intrathecal injection can cause a myelopathy and encephalopathy (19,253). Intracarotid administration of the drug can lead to necrosis and hemorrhagic infarction (254). Various drug–drug interactions have been described that may lead to increased anthracyclin-related neurotoxicity, including concomitant use of alpha interferon, ranitidine, verapamil, and cyclosporine (255). Idarubicin, epirubicin, and daunorubicin appear to be much less neurotoxic and nervous system toxicity may only be seen in high-dose applications or in combination with other neurotoxic agents. Mitoxantrone has no neurotoxicity when given intravenously, but may produce a radiculopathy and myelopathy when given intrathecally (256,257).

4.2. Azacitidine (5-Azacytidine) This drug is used for refractory acute myelogenous leukemia and myelodysplastic syndromes. Azacitidine acts as an antimetabolite and results in impairment of protein biosynthesis. It can cause weakness, myopathy, and lethargy (251,258).

4.3. Bleomycin Sulfate Bleomycin inhibits DNA synthesis by binding to guanosine and cytosine through intercalation and by production of free oxygen radicals. It has been used to treat lymphoma, Hodgkin’s disease, testicular cancer, and head and neck cancer. When used in combination with cisplatin, it can produce cerebral infarction (259,260).

Chapter 17 / Neurologic Complications of Chemotherapy

303

4.4. Busulfan Busulfan is a bifunctional alkylating agent used to treat chronic myelogenous leukemia, and as part of the conditioning regimen for some bone marrow transplantation regimens. The drug has little neurotoxicity at standard doses, but high-dose therapy can cause seizures (261).

4.5. Capecitabine Capecitabine is a prodrug for 5-fluorouracil (5-FU). It is metabolized to 5-FU by the enzyme thymidine phosphorylase and has mainly been used to treat breast cancer. Neurologic complications are relatively uncommon, but some patients experience paresthesias, headaches, dizziness, and insomnia. A syndrome of multifocal leukoencephalopathy has been reported after capecitabine, similar to the leukoencephalopathy seen after 5-fluorouracil (236,237). In affected patients MRI may show increased signals on FLAIR, T2, and DWI sequences in cerebral white matter tracts.

4.6. Carboplatin Carboplatin is a platinating agent used for ovarian, cervical, testicular, lung, and head and neck cancers. Unlike cisplatin, peripheral neuropathy and CNS toxicity occur only rarely at conventional doses. Intra-arterial carboplatin may produce stroke-like syndromes (262), cortical blindness (263), and retinal toxicity (264). Carpoplatin may be associated with peripheral nervous system toxicity, including ototoxicity and neuropathy. Toxicity may present as a pure sensory and painful neuropathy, as also seen with cisplatin and oxaliplatin (265). Neurotoxicity depends on the total cumulative dose and other predisposing factors, such as diabetes mellitus, alcohol, or inherited neuropathy. Overall, the incidence of carboplatin-related neuropathy is considered much less frequent as seen with cisplatin (266).

4.7. Chlorambucil Chlorambucil is an alkylating agent mainly used for the treatment of chronic lymphocytic leukemia, Hodgkin’s, and non-Hodgkin’s lymphoma. It usually has little neurotoxicity but encephalopathy, myoclonus (267,268), and seizures can occur with high and cumulative doses (269,270). Ocular toxicities, including keratitis, retinal edema and hemorrhages, have also been described following oral administration of chlorambucil (271).

4.8. Cladribine (2-chlorodeoxyadenosine) This drug inhibits DNA polymerase, DNA ligase, and ribonucleotide reductase, resulting in DNA strand breakage. It is used for hairy cell leukemia, low-grade non-Hodgkin’s lymphoma, chronic myelogenous leukemia, and Waldenstrom’s macroglobulinemia. It has little neurotoxicity at conventional doses but can produce a paraparesis at high doses (272).

4.9. Cyclophosphamide Cyclophosphamide is a nitrogen mustard alkylating agent mainly used to treat lymphoma, leukemia, and some solid tumors. Standard-dose cyclophosphamide has little neurotoxicity. High-dose cyclophosphamide has produced reversible visual blurring, dizziness and confusion (19). Its metabolite, 4-hydroperoxycyclophosphamide, has been used experimentally for intrathecal therapy of leptomeningeal metastases. At high doses it can cause lethargy and seizures (273).

4.10. Dacarbazine (DTIC) Dacarbazine has been used to treat melanoma and Hodgkin’s disease. The drug mainly acts through alkylation of RNA, DNA, and proteins. Neurotoxicity is very rare but seizures, encephalopathy and dementia have been reported (274).

4.11. Estramustine This is an estradiol molecule linked to a non-nitrogen mustard. It has estrogenic effects as well as causing dissociative effects on microtubules leading to metaphase arrest. It is used to treat advanced prostate carcinoma.

304

Part VI / Complications of Cancer Therapy

It has been associated with headaches and stroke (272). A more recent study estimates a relatively high rate of thromboembolic events (including strokes) in up to 25% of patients, which appears to be a dose-independent side effect (275).

4.12. Etoposide (VP-16) Etoposide is a topoisomerase II inhibitor used in the treatment of lung cancer, germ cell tumors, multiple myeloma, and lymphoma. Its effect is associated with a G2 cell cycle arrest and DNA breakage. It does not readily penetrate the blood–brain barrier and generally has little neurotoxicity, even at high doses. Rarely it can cause a peripheral neuropathy, mild disorientation, seizures, transient cortical blindness, or optic neuritis (17,276).

4.13. Fludarabine Fludarabine inhibits DNA polymerase and ribonucleotide reductase. It is used to treat chronic lymphocytic leukemia, macroglobulinemia, and indolent lymphomas. Neurotoxicity is uncommon, and appears to be dose related. Over one-third of patients receiving more than 90 mg/m2 /day of intravenous fludarabine develop severe neurotoxicity, while less than 0.5% of patients receiving standard doses of fludarabine (<40 mg/m2 /day) develop neurologic complications (250). At low doses fludarabine can cause headaches, somnolence, confusion, and paresthesias (19,250,277,278). Patients with mild neurologic complications usually improve when the drug is discontinued, but some patients continue to have permanent deficits (70). At high doses, fludarabine can cause a delayed progressive encephalopathy with visual loss, tremor, ataxia, seizures, paralysis, and coma (250,279,280). Some of these patients progress to a persistent vegetative state and occasionally death. Patients may also develop a severe myelopathy with quadriparesis. There have been a number of reports documenting severe leukoencephalopathy following fludarabine with lethal outcome (281,282). Fludarabine may increase the risk of JC-virus–associated multifocal leukoencephalopathy (283–286). MRI may show diffuse or multifocal areas of nonenhancing, increased T2/FLAIR signal in the white matter and brain stem (250,277) (Fig. 7 and Color Plate 7). Histopathological analyis from biopsy or autopsy cases have revealed multifocal demyelination and necrosis (278,287).

4.14. Gemcitabine Gemcitabine is a deoxycytidine analogue used for the treatment of pancreatic cancer, but it also has activity against other tumors, including breast cancer and lung cancer. The drug is a cell cycle–specific nucleoside analogue. Antineoplastic effects are mediated through G1/S-phase cell cycle arrest and DNA synthesis inhibition. Neurotoxicity occurs infrequently and usually involves the peripheral nervous system, especially when gemcitabine is administered in combination with other agents known to be toxic to the peripheral nervous system (e.g., oxaliplatin). The occurrence of a sensory neuropathy has been reported in up to 20% of gemcitabine-treated patients (288,289).

4.15. Hexamethylmelamine (Altretamine) This is probably an alkylating agent with activity against ovarian cancer, lymphoma, and lung cancer. It can produce headaches, encephalopathy, seizures, tremor, ataxia, parkinsonism, and paresthesias. These neurologic complications appear to be dose related and are usually reversible (19,70).

4.16. Hydroxyurea Hydroxyurea is an antimetabolite that inhibits DNA synthesis and conversion from RNA to DNA. It has been used to treat chronic myelogenous leukemia and certain solid tumors including melanoma, ovarian carcinoma,

Chapter 17 / Neurologic Complications of Chemotherapy

305

Fig. 7. This 78-year-old woman with an 8-year history of CLL developed slowly progressive right hemiparesis and hand twitching over several months. She had received rituximab and fludarabine within one year of onset of these symptoms. FLAIR MR sequences demonstrated multiple hyperintense lesions (A, B). Brain biopsy revealed prominent gliosis with scattered cells with enlarged nuclei and stippled, rim-like chromatin pattern suggestive of viral inclusions (C). Immunohistochemistry for JC virus showed positive staining of these cells (D). (Generously provided by Dr. David Schiff, University of Virginia, Charlottessville, VA.) (see Color Plate 7).

trophoblastic neoplasms, meningiomas, cervical cancer, head and neck cancer, and prostate cancer. Rarely, it causes encephalopathy with headaches, drowsiness, hallucinations, confusion, and seizures (19).

4.17. Irinotecan (CPT-11) Irinotecan is a prodrug that is converted to a topoisomerase I inhibitor used to treat refractory colon cancer, lung cancer, cervical and ovarian cancer, and pancreas and skin cancer. Severe neurologic toxicity has not been observed, but some patients experience transient visual disturbances and symptoms suggestive of cholinergic overactivity (249).

4.18. Levamisole Levamisole activates T-lymphocytes, and has been used in combination with 5-FU for patients with colon cancer. A metabolite, p-hydroxy-tetramisole, may enhance 5-FU activity by inhibiting tyrosine phosphatase. When used in combination with 5-FU it can cause multifocal leukoencephalopathy (231–233). Levamisole alone can cause headache, insomnia, dizziness, seizures, and aseptic meningitis (249), and in rare cases a clinical syndrome resembling acute disseminated encephalomyelitis responding to withdrawal of levamisole and steroid treatment (290–292).

306

Part VI / Complications of Cancer Therapy

4.19. Mechlorethamine (Nitrogen Mustard) This is an early generation bifunctional alkylating agent used to treat Hodgkin’s lymphoma and malignant pleural effusions. Rarely, it causes sleepiness, headaches, weakness, vertigo, hearing loss, and encephalopathy (19). At high doses used for bone marrow transplantation it has been reported to cause acute and delayed encephalopathy, including confusion and seizures (293,294). Intracarotid administration can produce uveitis and cerebral necrosis (19,70). Advanced age and concomitant use of cyclophosphamide or procarbazine are associated with an increased risk of neurotoxicity.

4.20. Mitomycin C This is an alkylating agent leading to DNA crosslinking and inhibition of DNA synthesis. It has been used to treat carcinomas of the gastrointestinal tract, breast cancer, and head and neck malignancies. It has been associated with an encephalopathy caused by thrombotic microangiopathy (259).

4.21. Nitrosoureas (BCNU, CCNU, PCNU, ACNU) The nitrosoureas are lipid-soluble alkylating agents that rapidly cross the blood–brain barrier and are used to treat brain tumors, melanoma, and lymphoma. These drugs are considered to have little neurotoxicity when used at conventional doses. Patients may, however, experience encephalopathy with confusion, lethargy, and ataxia even with conventional drug doses. In contrast, high-dose intravenous BCNU used in the setting of autologous bone marrow transplantation can cause delayed encephalomyelopathy developing over a period of weeks to months after the administration of the drug. Intra-arterial BCNU produces neurotoxicity and ocular toxicity in up to 50% of patients (295,296). Patients often complain of headache, as well as eye and facial pain. Retinopathy and blindness may occur. The neurotoxicity includes confusion, seizures, and progressive neurologic deficits associated with leukoencephalopathy (297). Concurrent radiotherapy increases the neurotoxicity of intracarotid BCNU (298). Occurrence of such leukoencephalopathy is not always predictable based on BCNU dosage and may be misinterpreted as tumor regrowth or tumor necrosis (299). Imaging and pathologic studies show findings similar to radiation necrosis (19,297,300). Injection of the drug above the origin of the ophthalmic artery may reduce the incidence of ocular toxicity but increases the neurotoxicity. Gliadel is a biodegradable polymer impregnated with BCNU, which is directly implanted into the resection cavity at the time of surgery for malignant gliomas. Local toxicity can occur and pharmacokinetic analysis indicates that tissue exposure to carmustine area under concentration-time curve achieved by polymeric delivery may be up to 1000 times higher than achieved with intravenous administrations (301,302). In a recent study, BCNU has been shown to be directly toxic to oligodendrocytes and neural progenitor cells. Sublethal drug exposure significantly impairs the key progenitor cell functions of proliferation and differentiation. In animal models, repetitive exposure was associated with long-term impairment of proliferation in the germinal zones of the central nervous system, including the dentate gyrus of the hippocampus and the subventricular zone (1). Consequently, it has been suggested that toxicity to progenitor cells and oligiodendrocytes constitutes the main cellular basis for CNS toxicity, including leukoencephalopathy and cognitive impairment (1,303).

4.22. Plicamycin (Mithramycin) This agent is used to treat refractory hypercalcemia and blast crisis of chronic myelogenous leukemia. Its mechanism of action includes inhibition of osteoclasts and parathyroid hormone and binding to DNA. It may cause headaches, fatigue, lethargy, and irritability. These side effects tend to be dose dependent.

4.23. Procarbazine-HCL (N-Methylhydrazine) Procarbazine inhibits DNA, RNA, and protein biosynthesis. It is used to treat Hodgkin’s and non-Hodgkin’s lymphoma, lung cancer, and brain tumors. The drug penetrates into CNS and peak levels occur within 60–90 min. At normal oral doses it can cause a mild reversible encephalopathy, and rarely psychosis and stupor (17,19,304). The incidence of encephalopathy may be increased in patients receiving high-dose procarbazine (> 300 mg/day), CCNU, and vincristine (PCV) chemotherapy for malignant gliomas (305). Other neurologic side effects include

Chapter 17 / Neurologic Complications of Chemotherapy

307

neuropathies, paresis, ataxia, seizures, and hallucinations. Procarbazine potentiates sedative effects of narcotics, phenothiazines, and barbiturates. Intravenous and intracarotid procarbazine may produce severe encephalopathy in some patients.

4.24. Pyrazolonacridine (PZA) This is a DNA intercalating agent with activity against a variety of solid tumor cells lines. Neurotoxicity is considered to be a dose-limiting factor. The complications seen included neuropsychiatric symptoms (restlessness, agitation, anxiety, hallucinations, personality changes, nightmares), motor symptoms, myoclonus, and dizziness (306,307).

4.25. Retinoic Acid All-trans-retinoic acid (Tretinoin) and 13-cis-retinoic acid (Isoretinoin) are vitamin A derivatives, which have been shown to induce differentiation in some tumors. They readily cross the blood–brain barrier. Headaches are a common complication (308). Retinoic acid can cause pseudotumor cerebri with visual field defects and compromised color vision (309,310), which is a reversible but a dose-limiting adverse reaction. Mononeuropathies have also been described following all-trans-retinoic acid (311). 13-cis-retinoic acid has been reported to cause oculogyric crisis (312). While it is well known that retinoic acid has teratogenic effects in the developing central nervous system by disrupting axonal outgrowth, migration of the neural crest, and specification of rostrocaudal position in the developing CNS (313,314), there is also some evidence that brain plasticity and cellular integrity may also be compromised in the postnatal and adult brain (315).

4.26. Suramin Suramin inhibits the binding of a number of growth factors to their receptors, including platelet-derived growth factor, basic fibroblast growth factor, and transforming growth factor beta. It also inhibits DNA polymerases and glycosaminoglycan catabolism. Suramin is used mainly for the treatment of refractory prostate cancer. It causes potentially severe peripheral neuropathy in approximately 10% of patients (316–318). Two patterns of peripheral neuropathies have been recognized, a distal axonal neuropathy and an inflammatory demyelinating neuropathy that is partially reversible (316). Close neurological monitoring has been suggested in patients receiving suramin for development of neuropathy. Visual changes have been reported in 7–9% of patients (319). Suramin can also cause mood changes, confusion, encephalopathy, ototoxicity, and seizures.

4.27. Temozolomide This is an alkylating agent related to imidazotetrazines with activity against malignant gliomas and melanoma. Up to 40% of patients receiving this drug may experience headaches, although serious neurologic complications are considered to be rare (320). Seizures and fatigue may occur in some patients (321).

4.28. Teniposide (VM-26) Teniposide is a topoisomerase II inhibitor used for acute lymphocytic leukemia, Kaposi’s sarcoma, and cutaneous T-cell lymphoma. Neurotoxicity is considered rare, but may present as paresthesias, fatigue, somnolence, and seizures (251,322).

4.29. Thioguanine (6-TG) Thioguanine is an antimetabolite is used to treat acute and chronic leukemia. It is an active nucleotide that inhibits DNA synthesis through substitution of purine bases. It shows poor or no blood–brain barrier penentration and generally has little neurotoxicity. Rarely, it causes ataxia or an encephalopathy secondary to hepatic toxicity (19).

308

Part VI / Complications of Cancer Therapy

4.30. Thiotepa Thiotepa is an alkylating agent occasionally used to treat leptomeningeal metastases, leukemia, lymphoma, ovarian, and breast cancer. Intrathecal thiotepa can cause aseptic meningitis, and very rarely, a myelopathy (323). Both thiotepa and its metabolite TEPA are lipid soluble and readily cross the blood–brain barrier. High intravenous doses of thiotepa can produce an encephalopathy that can be fatal (61,249). CNS toxicity can be seen in > 50% of patients who received prior cranial radiation (324).

4.31. Topotecan Topotecan is a topoisomerase I inhibitor used in ovarian cancer, non-small cell lung cancer, and Ewing’s sarcoma. It can occasionally cause headaches, fatigue, and parasthesias.

5. HORMONAL THERAPY 5.1. Aminoglutethimide Aminoglutethimide is a nonsteroidal inhibitor of corticoid biosynthesis. It results in depletion of endogenous estrogen, androgen, and cortisol. This hormonal drug has been used to treat breast carcinoma, adrenocortical carcinoma, and ectopic Cushing’s disease. It frequently is associated with fatigue and lethargy that can be severe in some cases. Rarely it may cause dizziness, vertigo and ataxia (251,325).

5.2. Anastrozole Anastrozole is an oral selective nonsteroidal aromatase inhibitor (inhibitor of estrogen) used for postmenopausal women with hormone receptor positive advanced breast cancer refractory to tamoxifen. It can be associated with weakness, headache, and back pain (326).

5.3. Corticosteroids Corticosteroids are frequently used in cancer patients. They reduce peritumoral edema in patients with primary and secondary brain tumors and spinal cord edema in patients with epidural spinal cord compression. Corticosteroids have a direct cytotoxic effect against neoplastic lymphocytes and are used in the treatment of leukemias and lymphomas. High-dose corticosteroids are frequently given with chemotherapy to reduce nausea and vomiting, whereas low doses are used to improve appetite and overall thrive. The side effects of prolonged steroid therapy are well known (327). The incidence of complications increases with higher doses and prolonged therapy, but individual susceptibility varies significantly. One of the most common complications of corticosteroids is steroid-induced myopathy (328,329), but CNS complications also occur frequently (Table 3). Corticosteroids often produce alterations in mood (330). An improved sense of thrive, anxiety, irritability, insomnia, difficulty concentrating, and depression are relatively common. Occasionally, patients may develop steroid psychosis (331). This usually takes the form of acute delirium, but the psychosis may resemble mania, depression, or schizophrenia. Other common neurologic complications of corticosteroids include tremors, visual blurring, reduced sense of taste and smell. Long-term effects may present as cognitive impairment and cerebral atrophy on neuroimaging studies. Rare complications include hiccups (332), seizures, and cord compression as a result of epidural lipomatosis (333). Steroid withdrawal can also produce a variety of symptoms, including headaches, lethargy, nausea, vomiting, anorexia, in addition to systemic symptoms such as myalgia, arthralgia, and postural hypotension. Rarely pseudotumor may occur.

5.4. Danazol Danazole suppresses the pituitary–ovarian axis. It also has weak antiandrogenic activity. It is rarely used to treat endometrial cancer. It can cause headaches, somnolence, and irritability.

Chapter 17 / Neurologic Complications of Chemotherapy

309

Table 3 Neurologic Complications of Corticosteroids Common • • • • • • •

Less common

Visual blurring Myopathy Behavioral changes Tremor Insomnia Change of taste and olfaction Cerebral atrophy

• • • • • • • •

Psychosis Hallucinations Hiccups Cognitive impairment and dementia Seizures Dependency Epidural lipomatosis Neuropathy

5.5. Goserelin This is an analogue of luteinizing hormone-releasing hormone used to treat prostate and breast cancer. When the drug is first used it can produce a tumor flare, resulting in bone pain and potentially exacerbating cord compression. In addition, complex pain syndromes and muscle weakness have been described (334).

5.6. Letrozole Letrozole is a third-generation nonsteroidal aromatase inhibitor (inhibitor of estrogen synthesis) used in breast cancer. It can cause headaches, insomnia, myalgias, and arthralgias.

5.7. Leuprolide Acetate This is a gonadotropin-releasing hormone analogue used to treat prostate cancer and refractory breast cancer. Neurologic complications are uncommon, but it can cause headaches, dizziness, and paresthesias. Bone pain and cord compression may initially worsen when the drug is first used. This flare can be prevented by concurrent use of an antiandrogen such as flutamide (19). Rare cases of seizures induced by leuprolide have been described; however, the presence of pre-existing brain damage in this report cannot be ruled out (335).

5.8. Mitotane (OP -DDD) Mitotane suppresses adrenocorticosteroid production and is cytotoxic to adrenal cortical cells. It is used to treat adrenocortical carcinoma. It produces lethargy, sedation, and dizziness in up to 40% of patients (336).

5.9. Octreotide Octreotide is a long-acting analogue of somatostatin used to treat carcinoid tumors, vasoactive intestinal peptide-secreting tumors, and certain pituitary adenomas. It can cause headaches, dizziness, and rarely seizures.

5.10. Tamoxifen This is an antiestrogen used to treat breast cancer and high-grade gliomas. The most common neurotoxicity is a reversible retinopathy and keratopathy (337–339). Affected patients experience decrease visual acuity. Examination may show retinal edema. Rarely, optic neuritis may occur (340). It can also produce an encephalopathy (341) and ataxia (342), especially when used in high doses (19,343). Mild headaches occur fairly frequently and it can exacerbate migraines (344). It can also cause depression, irritability, insomnia, poor concentration, or fatigue (19,345), and a radiation recall syndrome (344). Tamoxifen is associated with an increased incidence of systemic vascular thrombosis, but there is no definite evidence that it is associated with an increased risk of cerebrovascular disease (346). Other antiestrogens and antiandrogens such as leuprolide and flutamide are usually not associated with significant neurotoxicity, but some patients may experience headaches.

310

Part VI / Complications of Cancer Therapy

5.11. Toremifene Citrate This is a antiestrogen used to treat breast cancer. Neurologic complications are uncommon but patients may experience dizziness, depression, tremor, ataxia, and ocular toxicity (347).

6. BIOLOGIC AGENTS In recent years there has been increasing interest in the use of biologic agents for the treatment of cancers. Frequently, they are used in combination with conventional chemotherapeutic agents.

6.1. Interferon-Alpha This is a glycoprotein cytokine with antiviral, cytotoxic, and immunomodulatory activities. It is used therapeutically in a number of cancers including hairy cell leukemia, Kaposi’s sarcoma, chronic myelogenous leukemia, non-Hodgkin’s lymphoma, melanoma, renal cell carcinoma, and myeloma. Systemic toxicities include flu-like symptoms and myelosuppression. The flu-like symptoms, which include lethargy and headaches, tend to be worse at the onset of therapy and usually improve with time. Neurotoxicity tends to be dose-related. It is generally mild when low doses of alpha interferon are used as adjuvant therapy in patients with malignant melanoma (348). At higher doses, alpha interferon can cause headaches, confusion, lethargy, and seizures (349–352). These effects are more common in older patients (353). Neuroimaging studies are usually normal. EEG may show diffuse slowing (354,355) and rarely epileptiform activity (356). These neurotoxicities are usually reversible, but occasionally a permanent dementia or a persistent vegetative state may result (349,350). Rarely, alpha interferon has been associated with oculomotor palsy, tremor (357), visual hallucinations, retinopathy (358,359), parkinsonism (70), and spastic diplegia (360). A high incidence of neuropsychiatric toxicity has been noted in patients treated with recombinant interferonalpha-2b. In a study of 91 patients with chronic myelogenous leukemia, one-quarter experienced grade 3 or grade 4 neuropsychiatric toxicity that affected daily functioning. All patients recovered upon withdrawal of interferonalpha-2b. Patients with a psychiatric history were more likely to develop severe neuropsychiatric toxicity than patients without a psychiatric history (361). Other studies report significant incidence of depression and anxiety with alpha interferon (355,362). Intrathecal administration of alpha interferon has been evaluated for the treatment of meningeal and brain tumors, multiple sclerosis, amyotrophic lateral sclerosis, and progressive multifocal leukoencephalopathy. An acute reaction is usually seen within hours of the first injection and consists of headache, nausea, vomiting, fever, and dizziness. The symptoms usually resolve over the next 12–24 hrs. A severe encephalopathy develops in a significant number of patients within several days of the onset of treatment. This is dose dependent and tends to be worse in patients who have received cranial irradiation (349). The mechanism of interferon neurotoxicity is unknown, but may include induction of neurotoxic cytokines and competition with naturally occurring neurotrophic hormones and opioids (70).

6.2. Interferon-Beta and -Gamma These have neurotoxicities similar to interferon-alpha, although interferon-beta appears to be better tolerated (363).

6.3. Interleukin-1 (IL-1) Interleukin-1 can cause headache, encephalopathy, and seizures (249). The headaches and encephalopathy are usually mild and dose dependent.

6.4. Interleukin-2 (IL-2) This is a glycoprotein-released during T-cell activation. It has been used alone or in combination with lymphokine-activated killer (LAK) cells and tumor-infiltrating lymphocytes for the treatment of a number of cancers, including renal cell carcinoma and melanoma. Neuropsychiatric complications occur in 30–50% of patients (363–365). These include cognitive changes, delusions, hallucinations, and depression. Rarely, a severe

Chapter 17 / Neurologic Complications of Chemotherapy

311

encephalopathy and coma may result (366). In addition there have been reports of transient focal neurologic deficits. These include monocular blindness and homonymous hemianopia (367), aphasia, hemiparesis, hemisensory loss, seizures, and encephalopathy (368). MRI may show multiple cortical and subcortical areas of increased T2 signal, which may improve over time (368,369). CSF is usually normal (368). IL-2 can also cause a fatal acute leukoencephalopathy (370). Administration of IL-2 into the tumor bed for the treatment of gliomas can produce significant cerebral edema by increasing capillary permeability (371). Intraventricular administration of IL-2 for leptomeningeal metastases may cause a progressive subcortical dementia due to diffuse leukoencephalopathy (372). The MRI shows multiple white matter lesions. One case of grade 5 neurotoxicity has been reported in a patient treated with IL-2 and granulocyte-macrophagecolony stimulating factor. This patient experienced a fatal cerebral hemorrhage associated with thrombocytopenia, leading the authors to recommend extreme caution in using these agents together (373). The mechanism of neurotoxicty of IL-2 is unknown. It stimulates the production of a number of cytokines, including tumor-necrosis factor, which may be toxic to neurons and glia. It also stimulates the release of neuroendocrine hormones, which may also contribute to the encephalopathy (70).

6.5. Interleukin-4 (IL-4) IL-4 has been associated with headaches in as many as 70% of patients (374).

6.6. Tumor Necrosis Factor (TNF) TNF can cause encephalopathy, transient aphasia, or other focal deficits when administered systemically (249).

7. GROWTH FACTORS 7.1. Colony-Stimulating Factors: G-CSF, GM-CSF These are used to increase the granulocyte count and reduce the incidence of infections in patients with nonmyeloid tumors receiving chemotherapy. Rarely, they cause fatigue and headaches. Other rare adverse reactions include confusion and reversible posterior leukoencephalopathy (375).

7.2. Erythropoietin This growth factor is used to stimulate erythrocyte production. Some patients may experience fatigue, dizziness, and paresthesias, but serious neurologic complications have not been described.

7.3. Oprelvekin This is a recombinant human interleukin-11 used to prevent severe chemotherapy-induced thrombocytopenia. Neurologic complications are uncommon, but some patients complain of headaches, dizziness, insomnia, and parasthesias. Rare cases of optic disc edema have been desribed (376).

8. MONOCLONAL ANTIBODIES 8.1. Bevacizumab Bevacizumab is a humanized immunoglobulin G monoclonal antibody that binds to VEGF with high specificity, thereby blocking VEGF-mediated signalling pathways and angiogenesis. It has been used for colorectal cancer, breast, lung, and renal cell carcinoma and a variety of other cancers currently being evaluated in clinical trials. Bevacizumab is generally well-tolerated (377). Common side effects include headache (378). In addition, there is a higher risk of bleeding, hypertension (379), and thrombembolic events (380) with associated increased risk for transient ischemic events and stroke (381,382). There have also been recent reports describing the clinical picture of posterior reversible leukoencephalopathy associated with bevacizumab (383,384) (JD and PW, own observations). Other adverse effects include visual hallucinations and retinal pigment epithelium tearing after intravitreal injections (385–387).

312

Part VI / Complications of Cancer Therapy

8.2. Cetuximab Cetuximab is a human/mouse chimeric anti-EGF-receptor monoclonal antibody. The drug competitively inhibits EGF binding, and induces receptor dimerization and downregulation. Cetuximab has been active in various tumors, including colorectal, head and neck, pancreatic, and lung cancers. Cetuximab has received FDA approval in combination with irinotecan for the treatment of metastatic colorectal cancer in patients refractory to irinotecan, and for use as a single agent in the treatment of recurrent metastatic colorectal cancer in patients intolerant of irinotecan-based chemotherapy. Neurologic side effects appear to be mild, but headaches occur frequently and may be dose limiting (388).

8.3. Gemtuzumab Ozogamicin Gemtuzumab ozogamicin is an antibody-targeted chemotherapeutic agent used in patients with CD33+ acute myelogenous leukemia. It can cause headaches and dizziness. Severe thrombocytopenia is common and may result in intracranial hemorrhage.

8.4. Ibritumomab Tiuxetan This is a radiopharmaceutical that is used for patients with relapsed or refractory low-grade, follicular, or transformed B-cell NHL. It is a combination of monoclonal antibody ibritumomab bound to tiuxetan (chelated to Y-90 and In-111) and directed against the CD20 antigen on lymphoma cells. Neurologic adverse effects include headaches (12%), dizziness (10%), back pain (8%), insomnia (5%), and encephalopathy (less than 1%) (389).

8.5. Iodine131 -Tositumomab This is a radiolabeled immunoglobulin G-2a murine monoclonal antibody directed against the CD20 antigen. In addition to the cytotoxic effects induced by the antibody, the presence of iodine-131 results in focused targeting of beta radiation to the tumor and surrounding tissue. To date, it has been primarily used to treat relapsed or refractory non-Hodgkin lymphoma. Iodine-131 tositumomab is well tolerated. A minority of patients experience headache or myalgia and a few develop hypothyroidism (390,391).

8.6. Panitumumab This is an anti-EGFR human monoclonal antibody approved for treatment of colorectal cancer. While headaches occur there have been no reports on serious neurological side effects.

8.7. Rituximab Rituximab is a genetically engineered chimeric murine/human monoclonal antibody directed against the CD20 antigen found on the surface of normal and malignant B-lymphocytes. It is used for the treatment of low-grade or follicular B-cell lymphoma. Neurologic complications are uncommon but some patients complain of headaches, myalgia, dizziness, or paresthesias (392–394). More recently there have been reports suggesting an increased risk to develop leukoencephalopathy that may be reversible (217), or progressive and associtated with viral infections (395–397) following therapy with rituximab.

8.8. Trastuzumab Trastuzumab is a humanized anti-p185 (HER-2) monoclonal antibody used alone or in combination with chemotherapeutic agents in patients with HER-2-overexpressing metastatic breast cancer (398). Rarely, patients may experience headaches, dizziness and insomnia (399,400).

9. SMALL MOLECULE INHIBITORS 9.1. Bortezomib (PS-341) This is a proteosome inhibitor, which has shown promising activity in patients with multiple myeloma. A small fiber painful sensory axonal neuropathy occurs in approximately 35% of patients but appears reversible after dose

Chapter 17 / Neurologic Complications of Chemotherapy

313

reduction or discontinuation in the majority of cases. The incidence of severe neuropathies tended to be higher in patients with clinical or electrophysiological evidence of neuropathy prior to treatment (401).

9.2. Gefitinib (ZD-1839) Gefitinib is an orally active selective epidermal growth factor receptor tyrosine kinase inhibitor (402). It has been approved for the treatment of advanced/refractory non-small cell lung cancer. It may cause somnolence in up to 15% of patients and headaches (403,404).

9.3. Imatinib Mesylate (STI-571) Imatinib is a protein-tyrosine kinase inhibitor, which potently inhibits the Abl-tyrosine kinase and the receptors for platelet-derived growth factor (PDGF) and stem cell factor, c-kit. Imatinib has been shown to have significant activity in patients with chronic myelogenous leukemia (CML) (405,406) and c-kit expressing gastrointestinal stromal tumors (407). Neurotoxic side effects may consist of myalgias, cramps, fatigue, headache, confusion, and ocular toxicity (408–411). Recent studies have indicated a possible increased risk for intracranial, intratumoral and subdural hemorrhage with administration of imatinib (412,413).

9.4. Sorafenib This is a multikinase inhibitor with activity against Raf kinase and receptor tyrosine kinases. It has recently been approved for renal cancer. It has been associated with peripheral neuropathies in 10–15% of cases. Headaches and fatigue may occur in up to 10% of treated patients (414,415). In addition, a posterior reversible leukencephalopathylike syndrome has been reported (416).

9.5. Sunitinib Sunitinib (SU-011248) is an oral small molecular tyrosine kinase inhibitor that exhibits potent antiangiogenic and antitumor activity. Sunitinib inhibits other tyrosine kinases including KIT, FLT3, colony-stimulating factor 1 (CSF-1), and RET, which are involved in a number of cancers, including small-cell lung cancer, gastrointestinal, and breast cancer. It has been approved for the treatment of advanced renal cell carcinoma and in imatinibrefractory gastrointestinal stromal tumors. Headaches and fatigue are common, possibly associated with druginduced impairment of the thyroid function.

9.6. Temsirolimus (CCI-779) This is an intravenous analog of rapamycin. It presents immunosuppressive properties and also antiproliferative activity. Its principal target is the mTOR serine/threonine kinase which controls the initiation of the transcription of many genes implicated in carcinogenesis. It has been used in breast cancers, glioblastoma, and renal cell carcinoma. Neurotoxicity may consist of confusion and fatigue.

9.7. Tipifarnib This is a selective inhibitor of farnesyl transferase. Clinical trials in patients with high-risk acute leukemias and myelodysplastic syndromes have demonstrated good efficacy with tipifarnib. It can cause a peripheral neuropathy. CNS complications are uncommon but some patients experience lethargy, confusion, ataxia, and photophobia (417–419).

10. OTHER AGENTS 10.1. Amifostine Amifostine is a thiophosphate cytoprotectant agent, which is used to reduce renal toxicity associated with cisplatin. There is also evidence that it may also reduce the neurotoxicity of many chemotherapeutic agents, including cisplatin (420). Neurologic complications are uncommon but amifostine may cause hypotension and lead to syncope and there have been rare reports of seizures.

314

Part VI / Complications of Cancer Therapy

10.2. Denileukin Difitox This is a fusion toxin used to treat cutaneous T-cell lymphoma expressing the CD25 component for the IL-2 receptor. The most common complication is a vascular leak syndrome, but some patients experience myalgias, dizziness, paresthesias, nervousness, confusion, and insomnia.

10.3. Pamidronate This is a bisphosphonate used to treat hypercalcemia and bony metastases. Approximately 2% of patients experience insomnia, sleepiness, or abnormal vision (251).

10.4. Talotrexin (PT-523) Talotrexin is a potent nonpolyglutamatable antifolate agent that has been used in experimental trials to treat refractory solid tumors, including gastrointestinal and lung cancers, and leukemias. This drug targets the enzyme DHFR resulting in inhibition of DNA synthesis tumor growth. Preclinical studies suggest that talotrexin, as compared to methotrexate, enters into cells up to ten times more efficiently and demonstrates 10- to 100-fold more potency in overcoming polyglutamation, a well-established mechanism of antifolate resistance. While the drug is too novel to have sufficient experiences about the spectrum of neurotoxic adverse effects, neurotoxic complications may present as fatigue and hypoxia, which secondarily may impair cognitive function. At higher and cumulative doses this agent may produce a methotrexate-like leukoencephalopathy that can be fatal (J. Dietrich, unpublished observations).

10.5. Zoledronic Acid Zoledronic acid is a potent biphosphonate used to treat bony metastases and tumor-induced hypercalcemia. CNS complications are uncommon and similar to those of pamidronate.

11. CONCLUSION Cancer therapeutics may target both the central and peripheral nervous systems, and neurotoxic adverse reactions may be recognized as both acute and delayed treatment complications. With more aggressive treatment regimens and improved survival in cancer patients, long-term neurotoxicity, such as cognitive decline, may be detected more frequently. While some of the more traditional agents including methotrexate and cisplatin are well known to be associated with neurotoxic adverse effects, many novel agents, such as immune-modulating and biological drugs, have not been available long enough to have sufficient understanding of their long-term effects on the central and peripheral nervous systems. The mechanisms of chemotherapy-related cellular neurotoxicity have been studied in more detail in recent years, and it has become clear that the cause of neurotoxicity is far more complex than simply toxic effects on proliferating cells within the nervous system. Recognition of neurologic complications is critically important for any oncologist or neuro-oncologist in order to prevent irreversible injury, and to distinguish chemotherapy related complications from metastatic disease, radiation related toxicity, paraneoplastic disorders, or opportunistic infections.

REFERENCES 1. Dietrich J., Han R., Yang Y et al. CNS progenitor cells and oligodendrocytes are targets of chemotherapeutic agents in vitro and in vivo. J Biol 2006;5(7):22. 2. Johnson SW, Stevenson JP, O’Dwyer PJ. Cisplatin and its analogues. In: Devita VT, Hellman S, Rosenberg SA, (eds.). Cancer: Principles and Practice of Oncology. Philadelphia: Lippincott, Williams and Wilkins; 2001:376–388. 3. Gormley PE, Gangji D, Wood, JH et al. Pharmacokinetic study of cerebrospinal fluid penetration of cis-diamminedichloroplatinum (II). Cancer Chemother Pharmacol 1981;5(4):257–260. 4. Gregg RW, Molepo JM, Monpetit VJ et al. Cisplatin neurotoxicity: the relationship between dosage, time, and platinum concentration in neurologic tissues, and morphologic evidence of toxicity. J Clin Oncol 1992;10(5):795–803. 5. DeGregorio M, Wilbur B, King O et al. Peak cerebrospinal fluid platinum levels in a patient with ependymoma: evaluation of two different methods of cisplatin administration. Cancer Treat Rep 1986;70(12):1437–1438.

Chapter 17 / Neurologic Complications of Chemotherapy

315

6. Roelofs RI, Hrushesky W, Rogin J et al. Peripheral sensory neuropathy and cisplatin chemotherapy. Neurology 1984;34(7):934–938. 7. Thompson SW, Davis LE, Kornfeld M et al. Cisplatin neuropathy: clinical, electrophysiologic, morphologic, and toxicologic studies. Cancer 1984;54(7):1269–1275. 8. Allen JC. The neurotoxicity of cisplatin. In: Rottenberg, DA (ed.). Neurologic Complications of Cancer Treatment. Boston: ButterworthHeinemann; 1991:135–142. 9. Pietrangeli A, Leandri M, Terzoli E et al. Persistence of high-dose oxaliplatin-induced neuropathy at long-term follow-up. Eur Neurol 2006;56(1):13–16. 10. van der Hoop RG., van der Burg ME, ten Bokkel Huinink WW et al. Incidence of neuropathy in 395 patients with ovarian cancer treated with or without cisplatin. Cancer 1990;66(8):1697–1702. 11. van der Hoop RG, Vecht CJ, van der Burg ME et al. Prevention of cisplatin neurotoxicity with an ACTH(4–9) analogue in patients with ovarian cancer. N Engl J Med 1990;322(2):89–94. 12. Siegal T, Haim, N. Cisplatin-induced peripheral neuropathy: frequent off-therapy deterioration, demyelinating syndromes, and muscle cramps. Cancer 1990;66(6):1117–1123. 13. Mollman JE, Glover DJ, Hogan WM et al. Cisplatin neuropathy: risk factors, prognosis, and protection by WR-2721. Cancer 1988;61(11):2192–2195. 14. Kemp G, Rose P, Lurain J et al. Amifostine pretreatment for protection against cyclophosphamide-induced and cisplatin-induced toxicities: results of a randomized control trial in patients with advanced ovarian cancer. J Clin Oncol 1996;14(7):2101–2112. 15. Argyriou AA, Chroni E, Koutras A et al. A randomized controlled trial evaluating the efficacy and safety of vitamin E supplementation for protection against cisplatin-induced peripheral neuropathy: final results. Support Care Cancer 2006;14(11):1134–1140. 16. Feinberg WM, Swenson MR. Cerebrovascular complications of l-asparaginase therapy. Neurology 1988;38(1):127–133. 17. Forsyth PA, Cascino TL. Neurologic complications of chemotherapy. In: Dekker M (ed.). Neurologic Complications of Cancer. New York: Wiley; 1995:241–266. 18. Gilbert MR. The neurotoxicity of chemotherapy. Neurologist 1998;4:43–53. 19. Posner JB. Side effects of chemotherapy. In: Posner JB (ed.). Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995:282–310. 20. Robertson GL, Bhoopalam N, Zelkowitz LJ. Vincristine neurotoxicity and abnormal secretion of antidiuretic hormone. Arch Intern Med 1973;132(5):717–720. 21. Rottenberg DA. Neurological Complications of Cancer Treatment. Boston: Butterworth-Heinemann; 1991. 22. Gerritsen van der Hoop R, de Koning P, Boven E et al. Efficacy of the neuropeptide ORG.2766 in the prevention and treatment of cisplatin-induced neurotoxicity in rats. Eur J Cancer Clin Oncol 1988;24(4):637–642. 23. Gerritsen van der Hoop R, Hamers FP, Neijt JP et al. Protection against cisplatin-induced neurotoxicity by ORG 2766: histological and electrophysiological evidence. J Neurol Sci 1994;126(2):109–115. 24. Moroso MJ, Blair RL. A review of cis-platinum ototoxicity. J Otolaryngol 1983;12(6):365–369. 25. Schell MJ, McHaney VA, Green AA et al. Hearing loss in children and young adults receiving cisplatin with or without prior cranial irradiation. J Clin Oncol 1989;7(6):754–760. 26. Walsh TJ, Clark AW, Parhad IM et al. Neurotoxic effects of cisplatin therapy. Arch Neurol 1982;39(11):719–720. 27. Zheng JL, Stewart RR, and Gao WQ. Neurotrophin-4/5 enhances survival of cultured spiral ganglion neurons and protects them from cisplatin neurotoxicity. J Neurosci 1995;15(7 Pt 2):5079–5087. 28. Oldenburg J, Kraggerud SM, Cvancarova M et al. Cisplatin-induced long-term hearing impairment is associated with specific glutathione S-transferase genotypes in testicular cancer survivors. J Clin Oncol 2007;25(6):708–714. 29. Black FO., Myers EN, Schramm VL et al. Cisplatin vestibular ototoxicity: preliminary report. Laryngoscope 1982;92(12):1363–1368. 30. Frustaci S, Barzan L, Comoretto R et al. Local neurotoxicity after intra-arterial cisplatin in head and neck cancer. Cancer Treat Rep 1987;71(3):257–259. 31. Kupersmith MJ, Frohman LP, Choi IS et al. Visual system toxicity following intra-arterial chemotherapy. Neurology 1988;38(2): 284–289. 32. Ostrow S, Hahn D, Wiernik PH et al. Ophthalmologic toxicity after cis-dichlorodiammineplatinum(II) therapy. Cancer Treat Rep 1978;62(10):1591–1594. 33. Wilding G, Caruso R, Lawrence TS et al. Retinal toxicity after high-dose cisplatin therapy. J Clin Oncol 1985;3(12):1683–1689. 34. Tfayli A, Hentschel P, Madajewicz S et al. Toxicities related to intra-arterial infusion of cisplatin and etoposide in patients with brain tumors. J Neurooncol 1999;42(1):73–77. 35. List AF, Kummet TD. Spinal cord toxicity complicating treatment with cisplatin and etoposide. Am J Clin Oncol 1990;13(3):256–258. 36. Walther PJ, Rossitch E, Jr., Bullard DE. The development of Lhermitte’s sign during cisplatin chemotherapy: possible drug-induced toxicity causing spinal cord demyelination. Cancer 1987;60(9):2170–2172. 37. Berman IJ, Mann MP. Seizures and transient cortical blindness associated with cis-platinum (II) diamminedichloride (PDD) therapy in a thirty-year-old man. Cancer 1980;45(4):764–766. 38. Ito Y, Arahata Y, Goto, Y et al. Cisplatin neurotoxicity presenting as reversible posterior leukoencephalopathy syndrome. AJNR Am J Neuroradiol 1998;19(3):415–417. 39. Lyass O, Lossos A, Hubert A et al. Cisplatin-induced nonconvulsive encephalopathy. Anticancer Drugs 1998;9(1):100–104. 40. Newton HB, Page MA, Junck L et al. Intra-arterial cisplatin for the treatment of malignant gliomas. J Neuro-oncol 1989;7(1):39–45. 41. Walker RW, Cairncross JG, Posner JB. Cerebral herniation in patients receiving cisplatin. J Neurooncol 1988;6(1):61–65. 42. Bellin SL, Selim M. Cisplatin-induced hypomagnesemia with seizures: a case report and review of the literature. Gynecol Oncol 1988;30(1):104–113. 43. Schilsky RL, Anderson T. Hypomagnesemia and renal magnesium wasting in patients receiving cisplatin. Ann Intern Med 1979;90(6):929–931.

316

Part VI / Complications of Cancer Therapy

44. Ritch PS. Cis-dichlorodiammineplatinum II–induced syndrome of inappropriate secretion of antidiuretic hormone. Cancer 1988;61(3):448–450. 45. Dietrich J, Marienhagen J, Schalke B et al. Vascular neurotoxicity following chemotherapy with cisplatin, ifosfamide, and etoposide. Ann Pharmacother 2004;38(2):242–246. 46. Gerl A, Clemm C, Wilmanns W. Acute and late vascular complications following chemotherapy for germ cell tumors. Onkologie 1993;16:88–92. 47. Icli F, Karaoguz H, Dincol D et al. Severe vascular toxicity associated with cisplatin-based chemotherapy. Cancer 1993;72(2):587–593. 48. Gerl A. Vascular toxicity associated with chemotherapy for testicular cancer. Anticancer Drugs 1994;5(6):607–614. 49. Samuels BL, Vogelzang NJ, Kennedy BJ. Severe vascular toxicity associated with vinblastine, bleomycin, and cisplatin chemotherapy. Cancer Chemother Pharmacol 1987;19(3):253–256. 50. El Amrani M, Heinzlef O, Debroucker T et al. Brain infarction following 5-fluorouracil and cisplatin therapy. Neurology 1998;51(3):899–901. 51. Troy L, McFarland K, Littman-Power S et al. Cisplatin-based therapy: a neurological and neuropsychological review. Psycho-oncology 2000;9(1):29–39. 52. Glantz MJ, LaFollette S, Jaeckle KA et al. Randomized trial of a slow-release versus a standard formulation of cytarabine for the intrathecal treatment of lymphomatous meningitis. J Clin Oncol 1999;17(10):3110–3116. 53. Herzig RH, Hines JD, Herzig GP et al. Cerebellar toxicity with high-dose cytosine arabinoside. J Clin Oncol 1987;5(6):927–932. 54. Hwang TL, Yung WK, Estey EH et al. Central nervous system toxicity with high-dose Ara-C. Neurology 1985;35(10):1475–1479. 55. Winkelman MD, Hines JD. Cerebellar degeneration caused by high-dose cytosine arabinoside: a clinicopathological study. Ann Neurol 1983;14(5):520–527. 56. Lazarus HM, Herzig RH, Herzig GP et al. Central nervous system toxicity of high-dose systemic cytosine arabinoside. Cancer 1981;48(12):2577–2582. 57. Eden OB, Goldie W, Wood T et al. Seizures following intrathecal cytosine arabinoside in young children with acute lymphoblastic leukemia. Cancer 1978;42(1):53–58. 58. Gottlieb D, Bradstock K, Koutts J et al. The neurotoxicity of high-dose cytosine arabinoside is age-related. Cancer 1987;60(7): 1439–1441. 59. Damon LE, Mass R, Linker CA. The association between high-dose cytarabine neurotoxicity and renal insufficiency. J Clin Oncol 1989;7(10):1563–1568. 60. Saito B, Nakamaki T, Nakashima H et al. Reversible posterior leukoencephalopathy syndrome after repeat intermediate-dose cytarabine chemotherapy in a patient with acute myeloid leukemia. Am J Hematol 2007;82(4):304–306. 61. Smith GA, Damon LE, Rugo HS et al. High-dose cytarabine dose modification reduces the incidence of neurotoxicity in patients with renal insufficiency. J Clin Oncol 1997;15(2):833–839. 62. Ritch PS, Hansen RM, Heuer DK. Ocular toxicity from high-dose cytosine arabinoside. Cancer 1983;51(3):430–432. 63. Shaw PJ, Procopis PG, Menser MA et al. Bulbar and pseudobulbar palsy complicating therapy with high-dose cytosine arabinoside in children with leukemia. Med Pediatr Oncol 1991;19(2):122–125. 64. Nevill TJ, Benstead TJ, McCormick CW et al. A. Horner’s syndrome and demyelinating peripheral neuropathy caused by high-dose cytosine arabinoside. Am J Hematol 1989;32(4):314–315. 65. Hoffman DL, Howard JR, Jr., Sarma R et al. Encephalopathy, myelopathy, optic neuropathy, and anosmia associated with intravenous cytosine arabinoside. Clin Neuropharmacol 1993;16(3):258–262. 66. Jabbour E, O’Brien S, Kantarjian H et al. Neurological complications associated with intrathecal liposomal cytarabine given prophylactically in combination with high-dose methotrexate and cytarabine to patients with acute lymphocytic leukemia. Blood 2007;109(8):3214–3218. 67. van den Berg H, van der Flier M, van de Wetering MD. Cytarabine-induced aseptic meningitis. Leukemia 2001;15(4):697–699. 68. Kleinschmidt-DeMasters BK, Yeh M. “Locked–in syndrome” after intrathecal cytosine arabinoside therapy for malignant immunoblastic lymphoma. Cancer 1992;70(10):2504–2507. 69. Luque FA, Selhorst JB, Petruska P. Parkinsonism induced by high-dose cytosine arabinoside. Mov Disord 1987;2(3):219–222. 70. Hammack JE, Cascino TL. Chemotherapy and other common drug-induced toxicities of the central nervous system in patients with cancer. In: Vecht CJ, (ed.). Handbook of Clinical Neurology. Amsterdam: Elsevier Science; 1998:481–514. 71. Phillips PC, Reinhard CS. Antipyrimidine neurotoxicity: cytosine arabinoside and 5-fluorouracil. In: Rottenberg DA, (ed.) Neurologic Complications of Cancer Treatment. Boston: Butterworth-Heinemann; 1991:97–114. 72. Dunton SF, Nitschke R, Spruce WE et al. Progressive ascending paralysis following administration of intrathecal and intravenous cytosine arabinoside: a Pediatric Oncology Group study. Cancer 1986;57(6):1083–1088. 73. Resar LM, Phillips PC, Kastan MB et al. Acute neurotoxicity after intrathecal cytosine arabinoside in two adolescents with acute lymphoblastic leukemia of B-cell type. Cancer 1993;71(1):117–123. 74. Russell JA, Powles RL. Letter: neuropathy due to cytosine arabinoside. Br Med J 1974;4(5945):652–653. 75. Borgeat A, De Muralt B, Stalder M. Peripheral neuropathy associated with high-dose Ara-C therapy. Cancer 1986;58(4):852–854. 76. Saito T, Asai O, Dobashi N et al. Peripheral neuropathy caused by high-dose cytosine arabinoside treatment in a patient with acute myeloid leukemia. J Infect Chemother 2006;12(3):148–151. 77. Powell BL, Capizzi RL, Lyerly ES et al. Peripheral neuropathy after high-dose cytosine arabinoside, daunorubicin, and asparaginase consolidation for acute nonlymphocytic leukemia. J Clin Oncol 1986;4(1):95–97. 78. Osborne WL, Holyoake TL, McQuaker IG et al. Fatal peripheral neuropathy following FLA chemotherapy. Clin Lab Haematol 2004;26(4):295–296. 79. Openshaw H, Slatkin NE, Stein AS et al. Acute polyneuropathy after high-dose cytosine arabinoside in patients with leukemia. Cancer 1996;78(9):1899–1905.

Chapter 17 / Neurologic Complications of Chemotherapy

317

80. Paul M, Joshua D, Rahme N et al. Fatal peripheral neuropathy associated with axonal degeneration after high-dose cytosine arabinoside in acute leukaemia. Br J Haematol 1991;79(3):521–523. 81. Scherokman B, Filling-Katz MR, Tell D. Brachial plexus neuropathy following high-dose cytarabine in acute monoblastic leukemia. Cancer Treat Rep 1985;69(9):1005–1006. 82. Geller HM, Cheng KY, Goldsmith NK et al. Oxidative stress mediates neuronal DNA damage and apoptosis in response to cytosine arabinoside. J Neurochem 2001;78(2):265–275. 83. Koros C, Papalexi E, Anastasopoulos D et al. Effects of Ara-C treatment on motor coordination and cerebellar cytoarchitecture in the adult rat: a possible protective role of NAC. Neurotoxicology 2007;28(1):83–92. 84. Courtney MJ, Coffey ET. The mechanism of Ara-C-induced apoptosis of differentiating cerebellar granule neurons. Eur J Neurosci 1999;11(3):1073–1084. 85. Dworkin LA, Goldman RD, Zivin LS et al. Cerebellar toxicity following high-dose cytosine arabinoside. J Clin Oncol 1985;3(5): 613–616. 86. Meanwell CA, Blake AE, Kelly KA et al. Prediction of ifosfamide/mesna associated encephalopathy. Eur J Cancer Clin Oncol 1986;22(7):815–819. 87. Zalupski M, Baker LH. Ifosfamide. J Natl Cancer Inst 1988;80(8):556–566. 88. Pratt CB, Goren MP, Meyer WH et al. Ifosfamide neurotoxicity is related to previous cisplatin treatment for pediatric solid tumors. J Clin Oncol 1990;8(8):1399–1401. 89. Dechant KL, Brogden RN, Pilkington T et al. Ifosfamide/mesna: a review of its antineoplastic activity, pharmacokinetic properties, and therapeutic efficacy in cancer. Drugs 1991;42(3):428–467. 90. Cerny T, Kupfer A. The enigma of ifosfamide encephalopathy. Ann Oncol 1992;3(9):679–681. 91. Fleming RA. An overview of cyclophosphamide and ifosfamide pharmacology. Pharmacotherapy 1997;17(5 Pt 2):146S–154S. 92. Miller LJ, Eaton VE. Ifosfamide-induced neurotoxicity: a case report and review of the literature. Ann Pharmacother 1992;26(2): 183–187. 93. Primavera A, Audenino D, Cocito L. Ifosfamide encephalopathy and nonconvulsive status epilepticus. Can J Neurol Sci 2002;29(2):180–183. 94. Watkin SW, Husband DJ, Green JA et al. Ifosfamide encephalopathy: a reappraisal. Eur J Cancer Clin Oncol 1989;25(9):1303–1310. 95. Pratt CB, Green AA, Horowitz ME et al. Central nervous system toxicity following the treatment of pediatric patients with ifosfamide/mesna. J Clin Oncol 1986;4(8):1253–1261. 96. David KA, Picus J. Evaluating risk factors for the development of ifosfamide encephalopathy. Am J Clin Oncol 2005;28(3):277–280. 97. Curtin JP, Koonings PP, Gutierrez M et al. Ifosfamide-induced neurotoxicity. Gynecol Oncol 1991;42(3):193–196; discussion 1–2. 98. Chastagner P, Sommelet-Olive D, Kalifa C et al. Phase II study of ifosfamide in childhood brain tumors: a report by the French Society of Pediatric Oncology (SFOP). Med Pediatr Oncol 1993;21(1):49–53. 99. Ghosn M, Carde P, Leclerq B et al. Ifosfamide/mesna related encephalopathy: a case report with a possible role of phenobarbital in enhancing neurotoxicity. Bull Cancer 1988;75(4):391–392. 100. Aeschlimann C, Cerny T, Kupfer A. Inhibition of (mono)amine oxidase activity and prevention of ifosfamide encephalopathy by methylene blue. Drug Metab Dispos 1996;24(12):1336–1339. 101. Pelgrims J, De Vos F, Van den Brande J et al. Methylene blue in the treatment and prevention of ifosfamide-induced encephalopathy: report of 12 cases and a review of the literature. Br J Cancer 2000;82(2):291–294. 102. Turner AR, Duong CD, Good DJ. Methylene blue for the treatment and prophylaxis of ifosfamide-induced encephalopathy. Clin Oncol (R Coll Radiol) 2003;15(7):435–439. 103. Patel PN. Methylene blue for management of ifosfamide-induced encephalopathy. Ann Pharmacother 2006;40(2):299–303. 104. Simonian NA, Gilliam FG, Chiappa KH. Ifosfamide causes a diazepam-sensitive encephalopathy. Neurology 1993;43(12):2700–2702. 105. Verdeguer A, Castel V, Esquembre C et al. Fatal encephalopathy with ifosfamide/mesna. Pediatr Hematol Oncol 1989;6(4):383–385. 106. Shuper A, Stein J, Goshen J et al. Subacute central nervous system degeneration in a child: an unusual manifestation of ifosfamide intoxication. J Child Neurol 2000;15(7):481–483. 107. Geiser CF, Bishop Y, Jaffe N 3rd et al., Adverse effects of intrathecal methotrexate in children with acute leukemia in remission. Blood 1975;45(2):189–195. 108. Glantz MJ, Jaeckle KA, Chamberlain MC et al. A randomized controlled trial comparing intrathecal sustained-release cytarabine (DepoCyt) to intrathecal methotrexate in patients with neoplastic meningitis from solid tumors. Clin Cancer Res 1999;5(11):3394–3402. 109. Phillips J. Methotrexate toxicity. In: Rottenberg DA (ed.). Neurologic Complications of Cancer Treatment. Boston: ButterworthHeinemann; 1991:115–134. 110. Boogerd W, Sande JJ vd, Moffie D. Acute fever and delayed leukoencephalopathy following low-dose intraventricular methotrexate. J Neurol Neurosurg Psychiatry 1988;51(10):1277–1283. 111. Bates S, McKeever P, Masur H et al. Myelopathy following intrathecal chemotherapy in a patient with extensive Burkitt’s lymphoma and altered immune status. Am J Med 1985;78(4):697–702. 112. McLean DR, Clink HM, Ernst P et al. Myelopathy after intrathecal chemotherapy: a case report with unique magnetic resonance imaging changes. Cancer 1994;73(12):3037–3040. 113. Gagliano RG, Costanzi JJ. Paraplegia following intrathecal methotrexate: report of a case and review of the literature. Cancer 1976;37(4):1663–1668. 114. Clark AW, Cohen SR, Nissenblatt MJ et al. Paraplegia following intrathecal chemotherapy: neuropathologic findings and elevation of myelin basic protein. Cancer 1982;50(1):42–47. 115. Shapiro WR, Chernik NL, Posner JB. Necrotizing encephalopathy following intraventricular instillation of methotrexate. Arch Neurol 1973;28(2):96–102.

318

Part VI / Complications of Cancer Therapy

116. Lemann W, Wiley RG, Posner, JB. Leukoencephalopathy complicating intraventricular catheters: clinical, radiographic and pathologic study of 10 cases. J Neurooncol 1988;6(1):67–74. 117. Valik D, Sterba J, Bajciova V et al. Severe encephalopathy induced by the first but not the second course of high-dose methotrexate mirrored by plasma homocysteine elevations and preceded by extreme differences in pretreatment plasma folate. Oncology 2005;69(3):269–272. 118. Winick NJ, Bowman WP, Kamen BA et al. Unexpected acute neurologic toxicity in the treatment of children with acute lymphoblastic leukemia. J Natl Cancer Inst 1992;84(4):252–256. 119. Kuker W, Bader P, Herrlinger U et al. Transient encephalopathy after intrathekal methotrexate chemotherapy: diffusion-weighted MRI. J Neurooncol 2005;73(1):47–49. 120. Kinirons P, Fortune A, Enright H et al. Acute pseudobulbar palsy due to methotrexate with rapid response to intravenous immunoglobulin. J Neurol 2005;252(11):1401–1403. 121. Yim YS, Mahoney DH, Jr., Oshman DG. Hemiparesis and ischemic changes of the white matter after intrathecal therapy for children with acute lymphocytic leukemia. Cancer 1991;67(8):2058–2061. 122. Haykin ME, Gorman M, van Hoff J et al. Diffusion-weighted MRI correlates of subacute methotrexate-related neurotoxicity. J Neurooncol 2006;76(2):153–157. 123. Koh S, Nelson MD, Jr., Kovanlikaya A et al. Anterior lumbosacral radiculopathy after intrathecal methotrexate treatment. Pediatr Neurol 1999;21(2):576–578. 124. Dietrich J, Wen PY. Complications of chemotherapy. MedLink 2007. 125. Adamson PC, Balis FM, McCully CL et al. Rescue of experimental intrathecal methotrexate overdose with carboxypeptidase-G2. J Clin Oncol 1991;9(4):670–674. 126. Ziereisen F, Dan B, Azzi N et al. Reversible acute methotrexate leukoencephalopathy: atypical brain MR imaging features. Pediatr Radiol 2006;36(3):205–212. 127. Fisher MJ, Khademian ZP, Simon EM et al. Diffusion-weighted MR imaging of early methotrexate-related neurotoxicity in children. AJNR Am J Neuroradiol 2005;26(7):1686–1689. 128. Rollins N, Winick N, Bash R et al. Acute methotrexate neurotoxicity: findings on diffusion-weighted imaging and correlation with clinical outcome. AJNR Am J Neuroradiol 2004;25(10):1688–1695. 129. Ettinger LJ, Freeman AI, Creaven PJ. Intrathecal methotrexate overdose without neurotoxicity: case report and literature review. Cancer 1978;41(4):1270–1273. 130. Widemann BC, Balis FM, Shalabi A et al. Treatment of accidental intrathecal methotrexate overdose with intrathecal carboxypeptidase G2. J Natl Cancer Inst 2004;96(20):1557–1559. 131. Finkelstein Y, Zevin S, Heyd J et al. Emergency treatment of life-threatening intrathecal methotrexate overdose. Neurotoxicology 2004;25(3):407–410. 132. Jakobson AM, Kreuger A, Mortimer O et al. Cerebrospinal fluid exchange after intrathecal methotrexate overdose: a report of two cases. Acta Paediatr 1992;81(4):359–361. 133. Spiegel RJ, Cooper PR, Blum RH et al. Treatment of massive intrathecal methotrexate overdose by ventriculolumbar perfusion. N Engl J Med 1984;311(6):386–368. 134. Drachtman RA, Cole PD, Golden CB et al. Dextromethorphan is effective in the treatment of subacute methotrexate neurotoxicity. Pediatr Hematol Oncol 2002;19(5):319–327. 135. Wernick R, Smith DL. Central nervous system toxicity associated with weekly low-dose methotrexate treatment. Arthritis Rheum 1989;32(6):770–775. 136. Aplin CG, Russell-Jones R. Acute dysarthria induced by low-dose methotrexate therapy in a patient with erythrodermic cutaneous T-cell lymphoma: an unusual manifestation of neurotoxicity. Clin Exp Dermatol 1999;24(1):23–24. 137. Renard D, Westhovens R, Vandenbussche E et al. Reversible posterior leucoencephalopathy during oral treatment with methotrexate. J Neurol 2004;251(2):226–228. 138. Rubnitz JE, Relling MV, Harrison PL et al. Transient encephalopathy following high-dose methotrexate treatment in childhood acute lymphoblastic leukemia. Leukemia 1998;12(8):1176–1181. 139. Sanchez-Carpintero R, Narbona J, Lopez de Mesa R et al. Transient posterior encephalopathy induced by chemotherapy in children. Pediatr Neurol 2001;24(2):145–148. 140. Martino RL, Benson AB, 3rd, Merritt JA et al. Transient neurologic dysfunction following moderate-dose methotrexate for undifferentiated lymphoma. Cancer 1984;54(9):2003–2005. 141. Walker RW, Allen JC, Rosen G et al. Transient cerebral dysfunction secondary to high-dose methotrexate. J Clin Oncol 1986;4(12):1845–1850. 142. Borgna-Pignatti C, Battisti L, Marradi P et al. Transient neurologic disturbances in a child treated with moderate-dose methotrexate. Br J Haematol 1992;81(3):448. 143. Phillips PC, Dhawan V, Strother SC et al. Reduced cerebral glucose metabolism and increased brain capillary permeability following high-dose methotrexate chemotherapy: a positron emission tomographic study. Ann Neurol 1987;21(1):59–63. 144. Millot F, Dhondt JL, Hayte JM et al. Impairment of cerebral biogenic amine synthesis in a patient receiving high-dose methotrexate. Am J Pediatr Hematol Oncol 1992;14(3):276–278. 145. Rubinstein LJ, Herman MM, Long TF et al. Disseminated necrotizing leukoencephalopathy: a complication of treated central nervous system leukemia and lymphoma. Cancer 1975;35(2):291–305. 146. Pizzo PA, Poplack DG, Bleyer WA. Neurotoxicities of current leukemia therapy. Am J Pediatr Hematol Oncol 1979;1(2):127–140. 147. Lovblad K, Kelkar P, Ozdoba C et al. Pure methotrexate encephalopathy presenting with seizures: CT and MRI features. Pediatr Radiol 1998;28(2):86–91.

Chapter 17 / Neurologic Complications of Chemotherapy

319

148. Shuper A, Stark B, Kornreich L et al. Methotrexate-related neurotoxicity in the treatment of childhood acute lymphoblastic leukemia. Isr Med Assoc J 2002;4(11):1050–1053. 149. Montour-Proulx I, Kuehn SM, Keene DL et al. Cognitive changes in children treated for acute lymphoblastic leukemia with chemotherapy only according to the Pediatric Oncology Group 9605 protocol. J Child Neurol 2005;20(2):129–133. 150. Ch’ien LT, Aur RJ, Verzosa M. S et al. Progression of methotrexate-induced leukoencephalopathy in children with leukemia. Med Pediatr Oncol 1981;9(2):133–141. 151. Ochs J, Mulhern R, Fairclough D et al. Comparison of neuropsychologic functioning and clinical indicators of neurotoxicity in long-term survivors of childhood leukemia given cranial radiation or parenteral methotrexate: a prospective study. J Clin Oncol 1991;9(1):145–151. 152. Oka M, Terae S, Kobayashi R et al. MRI in methotrexate-related leukoencephalopathy: disseminated necrotising leukoencephalopathy in comparison with mild leukoencephalopathy. Neuroradiology 2003;45(7):493–497. 153. Laxmi SN, Takahashi S, Matsumoto K et al. Treatment-related disseminated necrotizing leukoencephalopathy with characteristic contrast enhancement of the white matter. Radiat Med 1996;14(6):303–307. 154. Fernandez-Bouzas A, Ramirez Jimenez H, Vazquez Zamudio J et al. Brain calcifications and dementia in children treated with radiotherapy and intrathecal methotrexate. J Neurosurg Sci 1992;36(4):211–214. 155. Hertzberg H, Huk WJ, Ueberall MA et al. CNS late effects after ALL therapy in childhood. Part I: Neuroradiological findings in long-term survivors of childhood ALL: an evaluation of the interferences between morphology and neuropsychological performance. The German Late Effects Working Group. Med Pediatr Oncol 1997;28(6):387–400. 156. Gangji D, Reaman GH, Cohen SR et al. Leukoencephalopathy and elevated levels of myelin basic protein in the cerebrospinal fluid of patients with acute lymphoblastic leukemia. N Engl J Med 1980;303(1):19–21. 157. Lai R, Abrey LE, Rosenblum MK et al. Treatment-induced leukoencephalopathy in primary CNS lymphoma: a clinical and autopsy study. Neurology 2004;62(3):451–456. 158. Sakamaki H, Onozawa Y, Yano Y et al. Disseminated necrotizing leukoencephalopathy following irradiation and methotrexate therapy for central nervous system infiltration of leukemia and lymphoma. Radiat Med 1993;11(4):146–153. 159. Price RA, Birdwell DA. The central nervous system in childhood leukemia. III. Mineralizing microangiopathy and dystrophic calcification. Cancer 1978;42(2):717–728. 160. Davidson A, Payne G, Leach MO et al. Proton magnetic resonance spectroscopy ((1)H-MRS) of the brain following high-dose methotrexate treatment for childhood cancer. Med Pediatr Oncol 2000;35(1):28–34. 161. Vezmar S, Becker A, Bode U, Jaehde U. Biochemical and clinical aspects of methotrexate neurotoxicity. Chemotherapy 2003; 49(1–2):92–104. 162. Quinn CT, Griener JC, Bottiglieri T et al. Effects of intraventricular methotrexate on folate, adenosine, and homocysteine metabolism in cerebrospinal fluid. J Pediatr Hematol Oncol 2004;26(6):386–388. 163. Quinn CT, Griener JC, Bottiglieri T et al. Elevation of homocysteine and excitatory amino acid neurotransmitters in the CSF of children who receive methotrexate for the treatment of cancer. J Clin Oncol 1997;15(8):2800–2806. 164. Miyatake S, Kikuchi H, Oda Y et al. A case of treatment-related leukoencephalopathy: sequential MRI, CT and PET findings. J Neurooncol 1992;14(2):143–149. 165. Linnebank M, Pels H, Kleczar N et al. MTX-induced white matter changes are associated with polymorphisms of methionine metabolism. Neurology 2005;64(5):912–913. 166. Stordal B, Pavlakis N, Davey R. Oxaliplatin for the treatment of cisplatin-resistant cancer: a systematic review. Cancer Treat Rev 2007;33(8):688–703. 167. Extra JM, Marty M, Brienza S et al. Pharmacokinetics and safety profile of oxaliplatin. Semin Oncol 1998;25(2 Suppl 5):13–22. 168. Giacchetti S, Perpoint B, Zidani R et al. Phase III multicenter randomized trial of oxaliplatin added to chronomodulated fluorouracil– leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000;18(1):136–147. 169. Krishnan AV, Goldstein D, Friedlander M et al. Oxaliplatin-induced neurotoxicity and the development of neuropathy. Muscle Nerve 2005;32(1):51–60. 170. Kiernan MC, Krishnan AV. The pathophysiology of oxaliplatin-induced neurotoxicity. Curr Med Chem 2006;13(24):2901–2907. 171. Cersosimo RJ. Oxaliplatin-associated neuropathy: a review. Ann Pharmacother 2005;39(1):128–135. 172. Gamelin L, Boisdron-Celle M, Delva R et al. Prevention of oxaliplatin-related neurotoxicity by calcium and magnesium infusions: a retrospective study of 161 patients receiving oxaliplatin combined with 5-Fluorouracil and leucovorin for advanced colorectal cancer. Clin Cancer Res 2004;10(12 Pt 1):4055–4061. 173. Hochster HS, Grothey A, Childs BH. Use of calcium and magnesium salts to reduce oxaliplatin-related neurotoxicity. J Clin Oncol 2007;25(25):4208–4209. 174. Pasetto LM, D’Andrea MR, Rossi E et al. Oxaliplatin-related neurotoxicity: how and why? Crit Rev Oncol Hematol 2006;59(2): 159–168. 175. Lersch C, Schmelz R, Eckel F et al. Prevention of oxaliplatin-induced peripheral sensory neuropathy by carbamazepine in patients with advanced colorectal cancer. Clin Colorectal Cancer 2002;2(1):54–58. 176. Penz M, Kornek GV, Raderer M et al. Subcutaneous administration of amifostine: a promising therapeutic option in patients with oxaliplatin-related peripheral sensitive neuropathy. Ann Oncol 2001;12(3):421–422. 177. Argyriou AA, Chroni E, Polychronopoulos P et al. Efficacy of oxcarbazepine for prophylaxis against cumulative oxaliplatin-induced neuropathy. Neurology 2006;67(12):2253–2255. 178. Cassidy J, Bjarnason GA, Hickish T et al. Randomized double-blind placebo-controlled phase III study assessing the efficacy of xaliproden in reduicng the cumulative peripheral sensory neuropathy induced by the oxaliplatin and 5-FU/LV combination (FOLFOX4) in first-line treatment of patients with metastatic colorectal cancer. J Clin Oncol 2006;24:147s.

320

Part VI / Complications of Cancer Therapy

179. Cascinu S, Catalano V, Cordella L et al. Neuroprotective effect of reduced glutathione on oxaliplatin-based chemotherapy in advanced colorectal cancer: a randomized, double-blind, placebo-controlled trial. J Clin Oncol 2002;20(16):3478–3483. 180. Wang WS, Lin JK, Lin TC et al. Oral glutamine is effective for preventing oxaliplatin-induced neuropathy in colorectal cancer patients. Oncologist 2007;12(3):312–319. 181. O’Dea D, Handy CM, Wexler A. Ocular changes with oxaliplatin. Clin J Oncol Nurs 2006;10(2):227–229. 182. Leonard GD, Wright MA, Quinn MG et al. Survey of oxaliplatin-associated neurotoxicity using an interview-based questionnaire in patients with metastatic colorectal cancer. BMC Cancer 2005;5:116. 183. Skelton MR, Goldberg RM, O’Neil BH. A case of oxaliplatin-related posterior reversible encephalopathy syndrome. Clin Colorectal Cancer 2007;6(5):386–388. 184. Moris G, Ribacoba R, Gonzalez C. Delayed posterior encephalopathy syndrome following chemotherapy with oxaliplatin and gemcitabine. J Neurol 2007. 185. Rowinsky EK, Cazenave LA, Donehower RC. Taxol: a novel investigational antimicrotubule agent. J Natl Cancer Inst 1990;82(15):1247–1259. 186. Rowinsky EK, Donehower RC. Paclitaxel (taxol). N Engl J Med 1995;332(15):1004–1014. 187. Choy H, Rodriguez FF, Koester S et al. Investigation of taxol as a potential radiation sensitizer. Cancer 1993;71(11):3774–3778. 188. Postma TJ, Vermorken JB, Liefting AJ et al. Paclitaxel-induced neuropathy. Ann Oncol 1995;6(5):489–94. 189. Lipton RB, Apfel SC, Dutcher JP et al. Taxol produces a predominantly sensory neuropathy. Neurology 1989;39(3):368–373. 190. Freilich RJ, Balmaceda C, Seidman AD et al. Motor neuropathy due to docetaxel and paclitaxel. Neurology 1996;47(1):115–118. 191. New PZ, Jackson CE, Rinaldi D et al. Peripheral neuropathy secondary to docetaxel (Taxotere). Neurology 1996;46(1):108–111. 192. Rowinsky EK, Chaudhry V, Cornblath DR et al. Neurotoxicity of Taxol. J Natl Cancer Inst Monogr 1993(15):107–115. 193. Capri G, Munzone E, Tarenzi E et al. Optic nerve disturbances: a new form of paclitaxel neurotoxicity. J Natl Cancer Inst 1994;86(14):1099–1101. 194. Chang SM, Kuhn JG, Rizzo J et al. Phase I study of paclitaxel in patients with recurrent malignant glioma: a North American Brain Tumor Consortium report. J Clin Oncol 1998;16(6):2188–2194. 195. Brown T, Havlin K, Weiss G et al. A phase I trial of taxol given by a 6-hour intravenous infusion. J Clin Oncol 1991;9(7):1261–1267. 196. McGuire WP, Rowinsky EK, Rosenshein NB et al. Taxol: a unique antineoplastic agent with significant activity in advanced ovarian epithelial neoplasms. Ann Intern Med 1989;111(4):273–279. 197. Perry JR, Warner, E. Transient encephalopathy after paclitaxel (Taxol) infusion. Neurology 1996;46(6):1596–1599. 198. Guglani S, Farrugia D, Elsdon M et al. Reversible life-threatening encephalopathy in the absence of hepatic failure following conventional doses of docetaxel. Clin Oncol (R Coll Radiol) 2003;15(3):160–161. 199. Ziske CG, Schottker B, Gorschluter M et al. Acute transient encephalopathy after paclitaxel infusion: report of three cases. Ann Oncol 2002;13(4):629–631. 200. van den Bent MJ, Hilkens PH, Sillevis Smitt PA et al. Lhermitte’s sign following chemotherapy with docetaxel. Neurology 1998;50(2):563–564. 201. Khattab J, Terebelo HR, Dabas, B. Phantom limb pain as a manifestation of paclitaxel neurotoxicity. Mayo Clin Proc 2000;75(7): 740–742. 202. Nieto Y, Cagnoni PJ, Bearman SI et al. Acute encephalopathy: a new toxicity associated with high-dose paclitaxel. Clin Cancer Res 1999;5(3):501–506. 203. Singhal S, Mehta J, Desikan R et al. Antitumor activity of thalidomide in refractory multiple myeloma. N Engl J Med 1999;341(21):1565–1571. 204. Fine HA, Figg WD, Jaeckle K et al. Phase II trial of the antiangiogenic agent thalidomide in patients with recurrent high-grade gliomas. J Clin Oncol 2000;18(4):708–715. 205. Little RF, Wyvill KM, Pluda JM et al. Activity of thalidomide in AIDS-related Kaposi’s sarcoma. J Clin Oncol 2000;18(13):2593–2602. 206. Baidas SM, Winer EP, Fleming GF et al. Phase II evaluation of thalidomide in patients with metastatic breast cancer. J Clin Oncol 2000;18(14):2710–2707. 207. Chaudhry V, Cornblath DR, Corse A et al. Thalidomide-induced neuropathy. Neurology 2002;59(12):1872–1875. 208. Bradley WG, Lassman LP, Pearce GW et al. The neuromyopathy of vincristine in man: clinical, electrophysiological, and pathological studies. J Neurol Sci 1970;10(2):107–131. 209. Moudgil SS, Riggs JE. Fulminant peripheral neuropathy with severe quadriparesis associated with vincristine therapy. Ann Pharmacother 2000;34(10):1136–1138. 210. Munier F, Perentes E, Herbort CP et al. Selective loss of optic nerve beta-tubulin in vincristine-induced blindness. Am J Med 1992;93(2):232–234. 211. Bain PG, Lantos PL, Djurovic V et al. Intrathecal vincristine: a fatal chemotherapeutic error with devastating central nervous system effects. J Neurol 1991;238(4):230–234. 212. Bleck TP, Jacobsen J. Prolonged survival following the inadvertent intrathecal administration of vincristine: clinical and electrophysiologic analyses. Clin Neuropharmacol 1991;14(5):457–462. 213. Hurwitz RL, Mahoney DH, Jr., Armstrong DL et al. Reversible encephalopathy and seizures as a result of conventional vincristine administration. Med Pediatr Oncol 1988;16(3):216–219. 214. Williams ME, Walker AN, Bracikowski JP et al. Ascending myeloencephalopathy due to intrathecal vincristine sulfate: a fatal chemotherapeutic error. Cancer 1983;51(11):2041–2047. 215. Qweider M, Gilsbach JM, Rohde V. Inadvertent intrathecal vincristine administration: a neurosurgical emergency: case report. J Neurosurg Spine 2007;6(3):280–283. 216. Byrd RL, Rohrbaugh TM, Raney RB, Jr. et al. Transient cortical blindness secondary to vincristine therapy in childhood malignancies. Cancer 1981;47(1):37–40.

Chapter 17 / Neurologic Complications of Chemotherapy

321

217. Haefner MD, Siciliano RD, Widmer LA et al. Reversible posterior leukoencephalopathy syndrome after treatment of diffuse large B-cell lymphoma. Onkologie 2007;30(3):138–140. 218. Ozyurek H, Oguz G. Ozen, S et al. Reversible posterior leukoencephalopathy syndrome: report of three cases. J Child Neurol 2005;20(12):990–993. 219. Riehl JL, Brown WJ. Acute cerebellar syndrome secondary to 5-fluorouracil therapy. Neurology 1964;14:961–967. 220. Boileau G, Piro AJ, Lahiri SR et al. Cerebellar ataxia during 5-fluorouracil (NSC-19893) therapy. Cancer Chemother Rep 1971;55(5):595–598. 221. Greenwald ES. Letter: Organic mental changes with fluorouracil therapy. JAMA 1976;235(3):248–249. 222. Lynch HT, Droszcz CP, Albano WA et al. “Organic brain syndrome” secondary to 5-fluorouracil toxicity. Dis Colon Rectum 1981;24(2):130–131. 223. Liaw CC, Wang HM, Wang CH et al. Risk of transient hyperammonemic encephalopathy in cancer patients who received continuous infusion of 5-fluorouracil with the complication of dehydration and infection. Anticancer Drugs 1999;10(3):275–281. 224. Adams JW, Bofenkamp TM, Kobrin J et al. Recurrent acute toxic optic neuropathy secondary to 5-FU. Cancer Treat Rep 1984;68(3):565–566. 225. Bixenman WW, Nicholls JV, Warwick OH. Oculomotor disturbances associated with 5-fluorouracil chemotherapy. Am J Ophthalmol 1977;83(6):789–793. 226. Gradishar W, Vokes E, Schilsky R et al. Vascular events in patients receiving high-dose infusional 5-fluorouracil–based chemotherapy: the University of Chicago experience. Med Pediatr Oncol 1991;19(1):8–15. 227. Brashear A, Siemers E. Focal dystonia after chemotherapy: a case series. J Neurooncol 1997;34(2):163–167. 228. Pirzada NA, Ali II, Dafer, RM. Fluorouracil-induced neurotoxicity. Ann Pharmacother 2000;34(1):35–38. 229. Takimoto CH, Lu ZH, Zhang R et al. Severe neurotoxicity following 5-fluorouracil-based chemotherapy in a patient with dihydropyrimidine dehydrogenase deficiency. Clin Cancer Res 1996;2(3):477–481. 230. Ohara S, Hayashi R, Hata S et al. Leukoencephalopathy induced by chemotherapy with tegafur, a 5-fluorouracil derivative. Acta Neuropathol (Berl) 1998;96(5):527–531. 231. Hook CC, Kimmel DW, Kvols LK et al. Multifocal inflammatory leukoencephalopathy with 5-fluorouracil and levamisole. Ann Neurol 1992;31(3):262–267. 232. Kimmel DW, Schutt AJ. Multifocal leukoencephalopathy: occurrence during 5-fluorouracil and levamisole therapy and resolution after discontinuation of chemotherapy. Mayo Clin Proc 1993;68(4):363–365. 233. Savarese DM, Gordon J, Smith TW et al. Cerebral demyelination syndrome in a patient treated with 5-fluorouracil and levamisole: the use of thallium SPECT imaging to assist in noninvasive diagnosis: a case report. Cancer 1996;77(2):387–394. 234. Israel ZH, Lossos A, Barak V et al. Multifocal demyelinative leukoencephalopathy associated with 5-fluorouracil and levamisole. Acta Oncol 2000;39(1):117–120. 235. Chen TC, Hinton DR, Leichman L et al. Multifocal inflammatory leukoencephalopathy associated with levamisole and 5-fluorouracil: case report. Neurosurgery 1994;35(6):1138–1142; discussion 42–43. 236. Niemann B, Rochlitz C, Herrmann R et al. Toxic encephalopathy induced by capecitabine. Oncology 2004;66(4):331–335. 237. Videnovic A, Semenov I, Chua-Adajar R et al. Capecitabine-induced multifocal leukoencephalopathy: a report of five cases. Neurology 2005;65(11):1792–1794; discussion 685. 238. Haskell CM, Canellos GP, Leventhal BG et al. l-asparaginase: therapeutic and toxic effects in patients with neoplastic disease. N Engl J Med 1969;281(19):1028–1034. 239. Haskell CM, Canellos GP, Leventhal BG et al. H. l-asparaginase toxicity. Cancer Res 1969;29(4):974–975. 240. Leonard JV, Kay JD. Acute encephalopathy and hyperammonaemia complicating treatment of acute lymphoblastic leukaemia with asparaginase. Lancet 1986;1(8473):162–163. 241. Moure JM, Whitecar JP, Jr., Bodey GP. Electroencephalogram changes secondary to asparaginase. Arch Neurol 1970;23(4):365–368. 242. Foreman NK, Mahmoud HH, Rivera GK et al. Recurrent cerebrovascular accident with l-asparaginase rechallenge. Med Pediatr Oncol 1992;20(6):532–534. 243. Ott N, Ramsay NK, Priest JR et al. Sequelae of thrombotic or hemorrhagic complications following l-asparaginase therapy for childhood lymphoblastic leukemia. Am J Pediatr Hematol Oncol 1988;10(3):191–195. 244. Cairo MS, Lazarus K, Gilmore RL et al. Intracranial hemorrhage and focal seizures secondary to use of l-asparaginase during induction therapy of acute lymphocytic leukemia. J Pediatr 1980;97(5):829–833. 245. Priest JR, Ramsay NK, Steinherz PG et al. A syndrome of thrombosis and hemorrhage complicating l-asparaginase therapy for childhood acute lymphoblastic leukemia. J Pediatr 1982;100(6):984–989. 246. Schick RM, Jolesz F, Barnes PD et al. MR diagnosis of dural venous sinus thrombosis complicating l-asparaginase therapy. Comput Med Imaging Graph 1989;13(4):319–327. 247. Lee AY, Levine MN. The thrombophilic state induced by therapeutic agents in the cancer patient. Semin Thromb Hemost 1999;25(2):137–145. 248. Hamdan MY, Frenkel EP, Bick R. l-asparaginase-provoked seizures as singular expression of central nervous toxicity. Clin Appl Thromb Hemost 2000;6(4):234–238. 249. Paleologos N. Complications of chemotherapy. In: Biller J (ed). Iatrogenic Neurology. Boston: Butterworth-Heinemann; 1998:439–460. 250. Cheson BD, Vena DA, Foss FM et al. Neurotoxicity of purine analogs: a review. J Clin Oncol 1994;12(10):2216–2228. 251. Ignoffo RJ, Viele CS, Damon LE et al. Cancer Chemotherapy Pocket Guide. Philadelphia: Lippincott-Raven; 1998. 252. Schachter S, Freeman R. Transient ischemic attack and adriamycin cardiomyopathy. Neurology 1982;32(12):1380–1381. 253. Arico M, Nespoli L, Porta F et al. Severe acute encephalopathy following inadvertent intrathecal doxorubicin administration. Med Pediatr Oncol 1990;18(3):261–263.

322

Part VI / Complications of Cancer Therapy

254. Neuwelt EA, Pagel M, Barnett P et al. Pharmacology and toxicity of intracarotid adriamycin administration following osmotic blood–brain barrier modification. Cancer Res 1981;41(11 Pt 1):4466–4470. 255. Barbui T, Rambaldi A, Parenzan L et al. Neurological symptoms and coma associated with doxorubicin administration during chronic cyclosporin therapy. Lancet 1992;339(8806):1421. 256. Hall C, Dougherty WJ, Lebish IJ et al. Warning against use of intrathecal mitoxantrone. Lancet 1989;1(8640):734. 257. Lakhani AK, Zuiable AG, Pollard CM et al. Paraplegia after intrathecal mitozantrone. Lancet 1986;2(8520):1393. 258. Wilhelm M, O’Brien S, Rios MB et al. Phase I study of arabinosyl-5-azacytidine (fazarabine) in adult acute leukemia and chronic myelogenous leukemia in blastic phase. Leuk Lymphoma 1999;34(5–6):511–518. 259. Doll DC, Yarbro JW. Vascular toxicity associated with antineoplastic agents. Semin Oncol 1992;19(5):580–596. 260. Vogelzang NJ. Vascular and other complications of chemotherapy for testicular cancer. Worl J Urol 1984;2:32–37. 261. Vassal G, Deroussent A, Hartmann O et al. Dose-dependent neurotoxicity of high-dose busulfan in children: a clinical and pharmacological study. Cancer Res 1990;50(19):6203–6207. 262. Walker RW, Rosenblum MK, Kempin SJ et al. Carboplatin-associated thrombotic microangiopathic hemolytic anemia. Cancer 1989;64(5):1017–1020. 263. O’Brien ME, Tonge K, Blake P et al. Blindness associated with high-dose carboplatin. Lancet 1992;339(8792):558. 264. Stewart DJ, Belanger JM, Grahovac Z et al. Phase I study of intracarotid administration of carboplatin. Neurosurgery 1992;30(4): 512–516; discussion 6–7. 265. Quasthoff S, Hartung HP. Chemotherapy-induced peripheral neuropathy. J Neurol 2002;249(1):9–17. 266. Alberts DS. Clinical pharmacology of carboplatin. Semin Oncol 1990;17(4 Suppl 7):6–8. 267. Vandenberg SA, Kulig K, Spoerke DG et al. Chlorambucil overdose: accidental ingestion of an antineoplastic drug. J Emerg Med 1988;6(6):495–498. 268. Wyllie AR, Bayliff CD, Kovacs MJ. Myoclonus due to chlorambucil in two adults with lymphoma. Ann Pharmacother 1997;31(2): 171–174. 269. Blumenreich MS, Woodcock TM., Sherrill EJ et al. A phase I trial of chlorambucil administered in short pulses in patients with advanced malignancies. Cancer Invest 1988;6(4):371–375. 270. Salloum E, Khan KK, Cooper DL. Chlorambucil-induced seizures. Cancer 1997;79(5):1009–1013. 271. Burns LJ. Ocular toxicities of chemotherapy. Semin Oncol 1992;19(5):492–500. 272. Chabner BA, Longo DL. Cancer Chemotherapy and Biotherapy: Principles and Practice. 2nd ed. Philadelphia: Lippincott-Raven; 1996. 273. Phillips PC, Than TT, Cork LC et al. Intrathecal 4-hydroperoxycyclophosphamide: neurotoxicity, cerebrospinal fluid pharmacokinetics, and antitumor activity in a rabbit model of VX2 leptomeningeal carcinomatosis. Cancer Res 1992;52(22):6168–6174. 274. Paterson AH, McPherson, TA. A possible neurologic complication of DTIC. Cancer Treat Rep 1977;61(1):105–106. 275. Lubiniecki GM, Berlin JA, Weinstein RB et al. Thromboembolic events with estramustine phosphate-based chemotherapy in patients with hormone-refractory prostate carcinoma: results of a meta-analysis. Cancer 2004;101(12):2755–2759. 276. Leff RS, Thompson JM, Daly MB et al. Acute neurologic dysfunction after high-dose etoposide therapy for malignant glioma. Cancer 1988;62(1):32–35. 277. Cohen RB, Abdallah JM, Gray JR et al. Reversible neurologic toxicity in patients treated with standard-dose fludarabine phosphate for mycosis fungoides and chronic lymphocytic leukemia. Ann Intern Med 1993;118(2):114–116. 278. Spriggs DR, Stopa E, Mayer RJ et al. Fludarabine phosphate (NSC 312878) infusions for the treatment of acute leukemia: phase I and neuropathological study. Cancer Res 1986;46(11):5953–5958. 279. Chun HG, Leyland-Jones BR, Caryk SM et al. Central nervous system toxicity of fludarabine phosphate. Cancer Treat Rep 1986;70(10):1225–1228. 280. Warrell RP, Jr., Berman E. Phase I and II study of fludarabine phosphate in leukemia: therapeutic efficacy with delayed central nervous system toxicity. J Clin Oncol 1986;4(1):74–79. 281. Mielke S, Potthoff K, Feuerhake F et al. Fatal leukoencephalopathy after reduced-intensity allogeneic stem cell transplantation. Onkologie 2007;30(1–2):49–52. 282. Rodriguez L, Ribera JM, Batlle M et al. Progressive multifocal leukoencephalopathy shortly after the diagnosis of follicular lymphoma in a patient treated with fludarabine. Haematologica 2002;87(7):ECR26. 283. Kiewe P, Seyfert S, Korper S et al. Progressive multifocal leukoencephalopathy with detection of JC virus in a patient with chronic lymphocytic leukemia parallel to onset of fludarabine therapy. Leuk Lymphoma 2003;44(10):1815–1818. 284. Saumoy M, Castells G, Escoda L et al. Progressive multifocal leukoencephalopathy in chronic lymphocytic leukemia after treatment with fludarabine. Leuk Lymphoma 2002;43(2):433–436. 285. Vidarsson B, Mosher F, Salamat MS et al. T. Progressive multifocal leukoencephalopathy after fludarabine therapy for low-grade lymphoproliferative disease. Am J Hematol 2002;70(1):51–54. 286. Cid J, Revilla M, Cervera A et al. Progressive multifocal leukoencephalopathy following oral fludarabine treatment of chronic lymphocytic leukemia. Ann Hematol 2000;79(7):392–395. 287. Gonzalez H, Bolgert F, Camporo P et al. Progressive multifocal leukoencephalitis (PML) in three patients treated with standard-dose fludarabine (FAMP). Hematol Cell Ther 1999;41(4):183–186. 288. Airoldi M, Cattel L, Passera R et al. Gemcitabine and oxaliplatin in patients with metastatic breast cancer resistant to or pretreated with both anthracyclines and taxanes: clinical and pharmacokinetic data. Am J Clin Oncol 2006;29(5):490–494. 289. Harder J, Riecken B, Kummer O et al. Outpatient chemotherapy with gemcitabine and oxaliplatin in patients with biliary tract cancer. Br J Cancer 2006;95(7):848–852. 290. Liu HM, Hsieh WJ, Yang CC et al. Leukoencephalopathy induced by levamisole alone for the treatment of recurrent aphthous ulcers. Neurology 2006;67(6):1065–1067.

Chapter 17 / Neurologic Complications of Chemotherapy

323

291. Wu VC, Huang JW, Lien HC et al. Levamisole-induced multifocal inflammatory leukoencephalopathy: clinical characteristics, outcome, and impact of treatment in 31 patients. Medicine (Baltimore) 2006;85(4):203–213. 292. Lin CH, Jeng JS, Hsieh ST. et al. Acute disseminated encephalomyelitis: a follow-up study in Taiwan. J Neurol Neurosurg Psychiatry 2007;78(2):162–167. 293. Sullivan KM, Storb R, Shulman HM et al. Immediate and delayed neurotoxicity after mechlorethamine preparation for bone marrow transplantation. Ann Intern Med 1982;97(2):182–189. 294. Shapiro WR, Young DF. Neurological complications of antineoplastic therapy. Acta Neurol Scand Suppl 1984;100:125–132. 295. Shapiro WR, Green SB. Reevaluating the efficacy of intra-arterial BCNU. J Neurosurg 1987;66(2):313–315. 296. Shapiro WR, Green SB, Burger PC et al. A randomized comparison of intra-arterial versus intravenous BCNU, with or without intravenous 5-fluorouracil, for newly diagnosed patients with malignant glioma. J Neurosurg 1992;76(5):772–781. 297. Bashir R, Hochberg FH, Linggood RM et al. Pre-irradiation internal carotid artery BCNU in treatment of glioblastoma multiforme. J Neurosurg 1988;68(6):917–919. 298. Rosenblum MK, Delattre JY, Walker RW et al. Fatal necrotizing encephalopathy complicating treatment of malignant gliomas with intra-arterial BCNU and irradiation: a pathological study. J Neurooncol 1989;7(3):269–281. 299. Kleinschmidt-DeMasters BK, Geier JM. Pathology of high-dose intraarterial BCNU. Surg Neurol 1989;31(6):435–443. 300. Moore-Maxwell CA, Datto MB, Hulette CM. Chemotherapy-induced toxic leukoencephalopathy causes a wide range of symptoms: a series of four autopsies. Mod Pathol 2004;17(2):241–247. 301. Fung LK, Ewend MG, Sills A et al. Pharmacokinetics of interstitial delivery of carmustine, 4-hydroperoxycyclophosphamide, and paclitaxel from a biodegradable polymer implant in the monkey brain. Cancer Res 1998;58(4):672–684. 302. Fleming AB, Saltzman WM. Pharmacokinetics of the carmustine implant. Clin Pharmacokinet 2002;41(6):403–419. 303. Duffner PK. The long-term effects of chemotherapy on the central nervous system. J Biol 2006;5(7):21. 304. Weiss HD, Walker MD, Wiernik PH. Neurotoxicity of commonly used antineoplastic agents (first of two parts). N Engl J Med 1974;291(2):75–81. 305. Postma TJ, van Groeningen CJ, Witjes RJ et al. Neurotoxicity of combination chemotherapy with procarbazine, CCNU, and vincristine (PCV) for recurrent glioma. J Neurooncol 1998;38(1):69–75. 306. LoRusso P, Foster BJ, Poplin E et al. Phase I clinical trial of pyrazoloacridine NSC366140 (PD115934). Clin Cancer Res 1995;1(12):1487–1493. 307. Rowinsky EK, Noe DA, Grochow LB et al. Phase I and pharmacologic studies of pyrazoloacridine, a novel DNA intercalating agent, on single-dosing and multiple-dosing schedules. J Clin Oncol 1995;13(8):1975–1984. 308. Bailey J, Pluda JM, Foli A et al. Phase I/II study of intermittent all-trans-retinoic acid, alone and in combination with interferon alfa-2a, in patients with epidemic Kaposi’s sarcoma. J Clin Oncol 1995;13(8):1966–1974. 309. Selleri C, Pane F, Notaro R et al. All-trans-retinoic acid (ATRA) responsive skin relapses of acute promyelocytic leukaemia followed by ATRA-induced pseudotumour cerebri. Br J Haematol 1996;92(4):937–940. 310. Smith MA, Adamson PC, Balis FM et al. Phase I and pharmacokinetic evaluation of all-trans-retinoic acid in pediatric patients with cancer. J Clin Oncol 1992;10(11):1666–1673. 311. Yamaji S, Kanamori H, Mishima A et al. All-trans retinoic acid-induced multiple mononeuropathies. Am J Hematol 1999;60(4):311. 312. Bigby M, Stern RS. Adverse reactions to isotretinoin: a report from the Adverse Drug Reaction Reporting System. J Am Acad Dermatol 1988;18(3):543–552. 313. Maden M, Holder N. Retinoic acid and development of the central nervous system. Bioessays 1992;14(7):431–438. 314. Durston AJ, Timmermans JP, Hage WJ et al. Retinoic acid causes an anteroposterior transformation in the developing central nervous system. Nature 1989;340(6229):140–144. 315. Yamamoto M, Ullman D, Drager UC et al. Postnatal effects of retinoic acid on cerebellar development. Neurotoxicol Teratol 1999;21(2):141–146. 316. Chaudhry V, Eisenberger MA, Sinibaldi VJ et al. A prospective study of suramin-induced peripheral neuropathy. Brain 1996;119 (Pt 6):2039–2052. 317. La Rocca RV, Meer J, Gilliatt RW et al. Suramin-induced polyneuropathy. Neurology 1990;40(6):954–960. 318. Kaur M, Reed E, Sartor O et al. Suramin’s development: what did we learn? Invest New Drugs 2002;20(2):209–219. 319. Hussain M, Fisher EI, Petrylak DP et al. Androgen deprivation and four courses of fixed-schedule suramin treatment in patients with newly diagnosed metastatic prostate cancer: a Southwest Oncology Group Study. J Clin Oncol 2000;18(5):1043–1049. 320. Yung WK, Prados MD, Yaya-Tur R et al. Multicenter phase II trial of temozolomide in patients with anaplastic astrocytoma or anaplastic oligoastrocytoma at first relapse. Temodal Brain Tumor Group. J Clin Oncol 1999;17(9):2762–2771. 321. Chamberlain MC, Tsao-Wei DD, Groshen S. Temozolomide for treatment-resistant recurrent meningioma. Neurology 2004;62(7): 1210–1212. 322. Tirelli U, Carbone A, Crivellari D et al. A phase II trial of teniposide (VM 26) in advanced non-Hodgkin’s lymphoma, with emphasis on the treatment of elderly patients. Cancer 1984;54(3):393–396. 323. Gutin PH, Levi JA, Wiernik PH et al. Treatment of malignant meningeal disease with intrathecal thioTEPA: a phase II study. Cancer Treat Rep 1977;61(5):885–887. 324. Valteau-Couanet D, Fillipini B, Benhamou E et al. High-dose busulfan and thiotepa followed by autologous stem cell transplantation (ASCT) in previously irradiated medulloblastoma patients: high toxicity and lack of efficacy. Bone Marrow Transplant 2005;36(11):939–945. 325. Nemoto T, Rosner D, Patel JK et al. Aminoglutethimide in patients with metastatic breast cancer. Cancer 1989;63(9):1673–1675. 326. Wiseman LR, Adkins JC. Anastrozole: a review of its use in the management of post-menopausal women with advanced breast cancer. Drugs Aging 1998;13(4):321–332.

324

Part VI / Complications of Cancer Therapy

327. Vecht CJ, Verbiest HBC. Use of glucocorticoids in neuro-oncology. In: Dekker M, (ed.). Neurological Complications of Cancer. New York: Wiley; 1995:199–218. 328. Dropcho EJ, Soong SJ. Steroid-induced weakness in patients with primary brain tumors. Neurology 1991;41(8):1235–1239. 329. Rosener M, Martin E, Zipp F et al. Neurologic side-effects of pharmacologic corticoid therapy. Nervenarzt 1996;67(12):983–986. 330. Stiefel FC, Breitbart WS, Holland JC. Corticosteroids in cancer: neuropsychiatric complications. Cancer Invest 1989;7(5):479–491. 331. Lewis DA, Smith RE. Steroid-induced psychiatric syndromes: a report of 14 cases and a review of the literature. J Affect Disord 1983;5(4):319–332. 332. Baethge BA, Lidsky MD. Intractable hiccups associated with high-dose intravenous methylprednisolone therapy. Ann Intern Med 1986;104(1):58–59. 333. Haddad SF, Hitchon PW, Godersky JC. Idiopathic and glucocorticoid-induced spinal epidural lipomatosis. J Neurosurg 1991;74(1): 38–42. 334. Ernst G, Gericke A, Berg P. Central pain and complex motoric symptoms after gosarelin therapy of prostate cancer. Sci World J 2004;4:969–973. 335. Akaboshi S, Takeshita K. A case of atypical absence seizures induced by leuprolide acetate. Pediatr Neurol 2000;23(3):266–268. 336. Lanser JB, van Seters AP, Moolenaar AJ et al. Neuropsychologic and neurologic side effects of mitotane and reversibility of symptoms. J Clin Oncol 1992;10(9):1504. 337. Kaiser-Kupfer MI, Lippman ME. Tamoxifen retinopathy. Cancer Treat Rep 1978;62(3):315–320. 338. Nayfield SG, Gorin MB. Tamoxifen-associated eye disease: a review. J Clin Oncol 1996;14(3):1018–1026. 339. Pavlidis NA, Petris C, Briassoulis E et al. Clear evidence that long-term, low-dose tamoxifen treatment can induce ocular toxicity: a prospective study of 63 patients. Cancer 1992;69(12):2961–2964. 340. Ashford AR, Donev I, Tiwari RP et al. Reversible ocular toxicity related to tamoxifen therapy. Cancer 1988;61(1):33–35. 341. Ron IG, Inbar MJ, Barak Y et al. Organic delusional syndrome associated with tamoxifen treatment. Cancer 1992;69(6):1415–1417. 342. Pluss JL, DiBella NJ. Reversible central nervous system dysfunction due to tamoxifen in a patient with breast cancer. Ann Intern Med 1984;101(5):652. 343. Chang SM, Barker FG, 2nd, Huhn SL et al. High-dose oral tamoxifen and subcutaneous interferon alpha-2a for recurrent glioma. J Neurooncol 1998;37(2):169–176. 344. Parry BR. Radiation recall induced by tamoxifen. Lancet 1992;340(8810):49. 345. Tang P, Roldan G, Brasher PM et al. A phase II study of carboplatin and chronic high-dose tamoxifen in patients with recurrent malignant glioma. J Neurooncol 2006;78(3):311–316. 346. Saphner T, Tormey DC, Gray R. Venous and arterial thrombosis in patients who received adjuvant therapy for breast cancer. J Clin Oncol 1991;9(2):286–294. 347. Gianni L, Panzini I, Li S et al. Ocular toxicity during adjuvant chemoendocrine therapy for early breast cancer: results from International Breast Cancer Study Group trials. Cancer 2006;106(3):505–513. 348. Caraceni A, Gangeri L, Martini C et al. Neurotoxicity of interferon-alpha in melanoma therapy: results from a randomized controlled trial. Cancer 1998;83(3):482–489. 349. Meyers CA, Obbens EA, Scheibel RS et al. Neurotoxicity of intraventricularly administered alpha-interferon for leptomeningeal disease. Cancer 1991;68(1):88–92. 350. Meyers CA, Scheibel RS, Forman AD. Persistent neurotoxicity of systemically administered interferon-alpha. Neurology 1991;41(5):672–676. 351. Rohatiner AZ, Prior PF, Burton AC et al. Central nervous system toxicity of interferon. Br J Cancer 1983;47(3):419–422. 352. Ulbricht D, Metz RJ, Ries F et al Alpha-interferon encephalopathy. Neurology 2003;61(9):1301. 353. Spiegel RJ. The alpha interferons: clinical overview. Semin Oncol 1987;14(2 Suppl 2):1–12. 354. Smedley H, Katrak M, Sikora K et al. Neurological effects of recombinant human interferon. Br Med J (Clin Res Ed) 1983;286(6361):262–264. 355. Amodio P, De Toni EN, Cavalletto L et al. Mood, cognition, and EEG changes during interferon alpha (alpha-IFN) treatment for chronic hepatitis C. J Affect Disord 2005;84(1):93–98. 356. Suter CC, Westmoreland BF, Sharbrough FW et al. Electroencephalographic abnormalities in interferon encephalopathy: a preliminary report. Mayo Clin Proc 1984;59(12):847–850. 357. Nishihori T, Abdo-Matkiwsky M, Fleishman SB et al. Severe action tremor related to interferon-alpha 2b therapy for malignant melanoma. Am J Clin Oncol 2005;28(5):526. 358. Esmaeli B, Koller C, Papadopoulos N et al. Interferon-induced retinopathy in asymptomatic cancer patients. Ophthalmology 2001;108(5):858–860. 359. Hejny C, Sternberg P, Lawson DH et al. Retinopathy associated with high-dose interferon alfa-2b therapy. Am J Ophthalmol 2001;131(6):782–787. 360. Vesikari T, Nuutila A, Cantell K. Neurologic sequelae following interferon therapy of juvenile laryngeal papilloma. Acta Paediatr Scand 1988;77(4):619–622. 361. Hensley ML, Peterson B, Silver RT et al. Risk factors for severe neuropsychiatric toxicity in patients receiving interferon alfa2b and low-dose cytarabine for chronic myelogenous leukemia: analysis of Cancer and Leukemia Group B 9013. J Clin Oncol 2000;18(6):1301–1308. 362. Wichers MC, Koek GH, Robaeys G et al. IDO and interferon-alpha-induced depressive symptoms: a shift in hypothesis from tryptophan depletion to neurotoxicity. Mol Psychiatry 2005;10(6):538–544. 363. Delattre JY, Vega F, Chen Q. Neurologic complications of immunotherapy. In: Dekker M, (ed.). Neurologic Complications of Cancer. New York: Wiley; 1995:267–293.

Chapter 17 / Neurologic Complications of Chemotherapy

325

364. Denicoff KD, Durkin TM, Lotze MT et al. The neuroendocrine effects of interleukin-2 treatment. J Clin Endocrinol Metab 1989;69(2):402–410. 365. Denicoff KD, Rubinow DR, Papa MZ et al. The neuropsychiatric effects of treatment with interleukin-2 and lymphokine-activated killer cells. Ann Intern Med 1987;107(3):293–300. 366. Siegel JP, Puri RK. Interleukin-2 toxicity. J Clin Oncol 1991;9(4):694–704. 367. Bernard JT, Ameriso S, Kempf RA et al. Transient focal neurologic deficits complicating interleukin-2 therapy. Neurology 1990;40(1):154–155. 368. Karp BI, Yang JC, Khorsand M et al. Multiple cerebral lesions complicating therapy with interleukin-2. Neurology 1996;47(2):417–424. 369. Somers SS, Reynolds JV, Guillou PJ. Multifocal neurotoxicity during interleukin-2 therapy for malignant melanoma. Clin Oncol (R Coll Radiol) 1992;4(2):135–136. 370. Vecht CJ, Keohane C, Menon RS et al. Acute fatal leukoencephalopathy after interleukin-2 therapy. N Engl J Med 1990;323(16): 1146–1147. 371. Barba D, Saris SC, Holder C. et al. Intratumoral LAK cell and interleukin-2 therapy of human gliomas. J Neurosurg 1989;70(2): 175–182. 372. Meyers CA, Yung WK. Delayed neurotoxicity of intraventricular interleukin-2: a case report. J Neuro-oncol 1993;15(3):265–267. 373. Hotton KM, Khorsand M, Hank JA et al. A phase Ib/II trial of granulocyte-macrophage-colony stimulating factor and interleukin-2 for renal cell carcinoma patients with pulmonary metastases: a case of fatal central nervous system thrombosis. Cancer 2000;88(8): 1892–1901. 374. Tulpule A, Joshi B, DeGuzman N et al. Interleukin-4 in the treatment of AIDS-related Kaposi’s sarcoma. Ann Oncol 1997;8(1):79–83. 375. Leniger T, Kastrup O, Diener HC. Reversible posterior leukencephalopathy syndrome induced by granulocyte stimulating factor filgrastim. J Neurol Neurosurg Psychiatry 2000;69(2):280–281. 376. Peterson DC, Inwards DJ, Younge BR. Oprelvekin-associated bilateral optic disk edema. Am J Ophthalmol 2005;139(2):367–368. 377. Grothey A. Recognizing and managing toxicities of molecular targeted therapies for colorectal cancer. Oncology (Williston Park) 2006;20(14 Suppl 10):21–28. 378. Motl S. Bevacizumab in combination chemotherapy for colorecal and other cancers. Am J Health Syst Pharm 2005;62(10):1021–1032. 379. Zhu X, Wu S, Dahut WL et al. Risks of proteinuria and hypertension with bevacizumab, an antibody against vascular endothelial growth factor: systematic review and meta-analysis. Am J Kidney Dis 2007;49(2):186–193. 380. Shih T, Lindley, C. Bevacizumab: an angiogenesis inhibitor for the treatment of solid malignancies. Clin Ther 2006;28(11):1779–1802. 381. Vredenburgh JJ, Desjardins A, Herndon JE, 2nd et al. Phase II trial of bevacizumab and irinotecan in recurrent malignant glioma. Clin Cancer Res 2007;13(4):1253–1259. 382. Herbst RS. Toxicities of antiangiogenic therapy in non-small cell lung cancer. Clin Lung Cancer 2006;8 Suppl 1:S23–S30. 383. Ozcan C, Wong SJ, Hari, P. Reversible posterior leukoencephalopathy syndrome and bevacizumab. N Engl J Med 2006;354(9): 980–982; discussion 982. 384. Glusker P, Recht L, Lane B. Reversible posterior leukoencephalopathy syndrome and bevacizumab. N Engl J Med 2006;354(9): 980–982; discussion 982. 385. Meyer CH, Mennel S, Horle S et al. Visual hallucinations after intravitreal injection of bevacizumab in vascular age-related macular degeneration. Am J Ophthalmol 2007;143(1):169–170. 386. Shah CP, Hsu J, Garg SJ et al. Retinal pigment epithelial tear after intravitreal bevacizumab injection. Am J Ophthalmol 2006;142(6):1070–1072. 387. Spandau UH, Jonas JB. Retinal pigment epithelium tear after intravitreal bevacizumab for exudative age-related macular degeneration. Am J Ophthalmol 2006;142(6):1068–1070. 388. Vanhoefer U, Tewes M, Rojo F et al. Phase I study of the humanized antiepidermal growth factor receptor monoclonal antibody EMD72000 in patients with advanced solid tumors that express the epidermal growth factor receptor. J Clin Oncol 2004;22(1):175–184. 389. Wiseman GA, Leigh B, Erwin WD et al. Radiation dosimetry results for Zevalin radioimmunotherapy of rituximab-refractory nonHodgkin lymphoma. Cancer 2002;94(4 Suppl):1349–1357. 390. Kaminski MS, Estes J, Zasadny KR et al. Radioimmunotherapy with iodine 131 I tositumomab for relapsed or refractory B-cell nonHodgkin lymphoma: updated results and long-term follow-up of the University of Michigan experience. Blood 2000;96(4):1259–1266. 391. Vose JM, Wahl RL, Saleh M et al. Multicenter phase II study of iodine-131 tositumomab for chemotherapy-relapsed/refractory low-grade and transformed low-grade B-cell non-Hodgkin’s lymphomas. J Clin Oncol 2000;18(6):1316–1323. 392. Foran JM, Rohatiner AZ, Cunningham D et al. European phase II study of rituximab (chimeric anti-CD20 monoclonal antibody) for patients with newly diagnosed mantle–cell lymphoma and previously treated mantle-cell lymphoma, immunocytoma, and small B-cell lymphocytic lymphoma. J Clin Oncol 2000;18(2):317–324. 393. Maloney DG, Grillo-Lopez AJ, Bodkin DJ et al. IDEC-C2B8: results of a phase I multiple-dose trial in patients with relapsed non-Hodgkin’s lymphoma. J Clin Oncol 1997;15(10):3266–3274. 394. Maloney DG, Press OW. Newer treatments for non-Hodgkin’s lymphoma: monoclonal antibodies. Oncology (Williston Park) 1998;12(10 Suppl 8):63–76. 395. Steurer M, Clausen J, Gotwald T et al. Progressive multifocal leukoencephalopathy after allogeneic stem cell transplantation and posttransplantation rituximab. Transplantation 2003;76(2):435–436. 396. Matteucci P, Magni M, Di Nicola M et al. Leukoencephalopathy and papovavirus infection after treatment with chemotherapy and anti-CD20 monoclonal antibody. Blood 2002;100(3):1104–1105. 397. Goldberg SL, Pecora AL, Alter RS et al. Unusual viral infections (progressive multifocal leukoencephalopathy and cytomegalovirus disease) after high–dose chemotherapy with autologous blood stem cell rescue and peritransplantation rituximab. Blood 2002;99(4):1486–1488.

326

Part VI / Complications of Cancer Therapy

398. Shak S. Overview of the trastuzumab (Herceptin) anti-HER2 monoclonal antibody clinical program in HER2-overexpressing metastatic breast cancer. Herceptin Multinational Investigator Study Group. Semin Oncol 1999;26(4 Suppl 12):71–77. 399. Cobleigh MA, Vogel CL, Tripathy D et al. Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2–overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 1999;17(9):2639–2648. 400. Baselga J, Carbonell X, Castaneda-Soto NJ et al. Phase II study of efficacy, safety, and pharmacokinetics of trastuzumab monotherapy administered on a 3-weekly schedule. J Clin Oncol 2005;23(10):2162–2171. 401. Richardson, PG, Briemberg H, Jagannath S et al. Frequency, characteristics, and reversibility of peripheral neuropathy during treatment of advanced multiple myeloma with bortezomib. J Clin Oncol 2006;24(19):3113–3120. 402. Herbst RS, Fukuoka M, Baselga, J. Gefitinib: a novel targeted approach to treating cancer. Nat Rev Cancer 2004;4(12):956–965. 403. Baselga J, Rischin D, Ranson M et al. Phase I safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types. J Clin Oncol 2002;20(21):4292–4302. 404. Chau I, Cunningham D, Hickish T et al. Gefitinib and irinotecan in patients with fluoropyrimidine-refractory, irinotecan-naive advanced colorectal cancer: a phase I–II study. Ann Oncol 2007;18(4):730–737. 405. Druker BJ, Sawyers CL, Kantarjian H et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Engl J Med 2001;344(14):1038–1042. 406. Druker BJ, Talpaz M, Resta DJ et al. Efficacy and safety of a specific inhibitor of the BCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001;344(14):1031–1037. 407. Joensuu H, Roberts PJ, Sarlomo-Rikala M. et al. Effect of the tyrosine kinase inhibitor STI571 in a patient with a metastatic gastrointestinal stromal tumor. N Engl J Med 2001;344(14):1052–1056. 408. Kusumi E, Arakawa A, Kami M etc. Visual disturbance due to retinal edema as a complication of imatinib. Leukemia 2004;18(6): 1138–1139. 409. Fraunfelder FW, Solomon J, Druker, BJ et al. Ocular side-effects associated with imatinib mesylate (Gleevec). J Ocul Pharmacol Ther 2003;19(4):371–375. 410. Cohen MH, Williams G, Johnson JR et al. Approval summary for imatinib mesylate capsules in the treatment of chronic myelogenous leukemia. Clin Cancer Res 2002;8(5):935–942. 411. Hotte SJ, Winquist EW, Lamont E et al. Imatinib mesylate in patients with adenoid cystic cancers of the salivary glands expressing c-kit: a Princess Margaret Hospital phase II consortium study. J Clin Oncol 2005;23(3):585–590. 412. Pollack IF, Jakacki RI, Blaney SM et al. Phase I trial of imatinib in children with newly diagnosed brainstem and recurrent malignant gliomas: a Pediatric Brain Tumor Consortium report. Neuro-oncol 2007;9(2):145–160. 413. Song KW., Rifkind J, Al-Beirouti B et al. Subdural hematomas during CML therapy with imatinib mesylate. Leuk Lymphoma 2004;45(8):1633–1636. 414. Escudier B, Eisen T, Stadle, WM. et al. Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med 2007;356(2):125–34. 415. Kane RC, Farrell AT, Saber H et al. Sorafenib for the treatment of advanced renal cell carcinoma. Clin Cancer Res 2006;12(24): 7271–7278. 416. Govindarajan R, Adusumilli J, Baxter DL et al. Reversible posterior leukoencephalopathy syndrome induced by RAF kinase inhibitor BAY 43–9006. J Clin Oncol 2006;24(28):e48. 417. Perez-Ruixo JJ, Chen W, Zhang S et al. Exposure–toxicity relationships for tipifarnib in cancer patients. Br J Clin Pharmacol 2007;64(2):219–232. 418. Karp JE, Lancet JE, Kaufmann SH, et al. Clinical and biologic activity of the farnesyltransferase inhibitor R115777 in adults with refractory and relapsed acute leukemias: a phase 1 clinical-laboratory correlative trial. Blood 2001;97(11):3361–3369. 419. Rao S, Cunningham D, de Gramont, A. et al. Phase III double-blind placebo-controlled study of farnesyl transferase inhibitor R115777 in patients with refractory advanced colorectal cancer. J Clin Oncol 2004;22(19):3950–3957. 420. DiPaola RS and Schuchter, L. Neurologic protection by amifostine. Semin Oncol 1999;26(2 Suppl 7):82–88.

18

Neurological Complications of Hematopoietic Stem Cell Transplantation Eudocia Quant,

MD

and Patrick Y. Wen,

MD

CONTENTS Introduction Toxic-Metabolic Disorders Cerebrovascular Disorders Infections Myelopathy Neuromuscular Disorders Neurologic Manifestations Associated with GVHD Conclusion References

Summary Complications following hematopoietic stem cell transplantation have long been recognized. The causes are numerous, including chemoradiotoxicity, medication toxicity, metabolic abnormalities, organ failure, graft versus host disease, infection, pancytopenia, and platelet dysfunction. This chapter summarizes the disorders affecting the nervous system associated with hematopoietic stem cell transplantation. As the number of transplants performed annually increases, potential neurologic complications are being seen with increasing frequency. Key Words: stem cell transplantation, bone marrow, hematopoietic, graft versus host disease, infection

1. INTRODUCTION Hematopoietic stem cell transplantation (HSCT) is the transfer of hematopoietic stem cells obtained from the bone marrow, peripheral blood, or umbilical cord blood. A vast majority of these transplants are for hematologic and lymphoid cancers, but there are many other diseases commonly treated with HSCT (Table 1). Hematopoietic stem cells are able to sustain the entire hematologic and lymphoid systems including red blood cells, granulocytes, lymphocytes, platelets, and macrophages (e.g., Kupffer cells in the liver, pulmonary alveolar macrophages, osetoclasts, Langerhans cells in the skin, microglial cells in the brain). Internationally, there are 50,000 HSCTs performed each year with an expected growth rate of 10–15% per year (1). Transplants can be categorized according to the relationship between the donor and the recipient. An allogeneic donor is not immunologically identical but is matched using major histocompatability antigens (HLA). HLA matching is important because recipient T-cells will react against mismatched HLA, resulting in graft rejection. Donor T-cells also recognize mismatched recipient antigens, resulting in graft versus host disease (GVHD) or graft versus tumor effect (GVT). Because most transplants are HLA-matched, GVHD or GVT are the result of mismatched minor histocompatibility antigens. Much of the antitumor effect from HSCT depends on this GVT From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

327

328

Part VI / Complications of Cancer Therapy

Table 1 Indications for Hematopoietic Stem Cell Transplantation Autologous transplantation Cancers Multiple myeloma Non-Hodgkin’s lymphoma Hodgkin’s disease Acute myeloid leukemia Neuroblastoma Ovarian cancer Germ cell tumors Other diseases Autoimmune disorders Amyloidosis Allogeneic transplantation Cancers Acute myeloid leukemia Acute lymphoblastic leukemia Chronic myeloid leukemia Myelodysplastic syndromes Myeloproliferative disorders Non-Hodgkin’s lymphoma Hodgkin’s disease Chronic lymphocytic leukemia Multiple myeloma Juvenile chronic myeloid leukemia Other diseases Aplastic anemia Paroxysmal nocturnal hemoglobinuria Fanconi’s anemia Blackfan–Diamond anemia Thalassemia major Sickle cell anemia Severe combined immunodeficiency Wiskott–Aldrich syndrome Inborn errors of metabolism Neurologic diseases (mostly experimental) Chronic inflammatory demyelinating polyneuropathy Multiple sclerosis Leukodystrophies including adrenoleukodystrophy, globoid cell leukodystrophy, Krabbe disease, metachromatic leukodystrophy POEMS syndrome Primary brain tumors including CNS lymphoma, high-grade astrocytomas, medulloblastoma, primitive neuroectodermal tumors, germ cell tumors Modified from Copelan EA. Hematopoietic stem cell transplantation. N Engl J Med 2006;354:1813–1826; with permission. Copyright © 2006 Massachusetts Medical Society. All rights reserved.

response. This is supported by several findings: (i) relapse rates are lowest in patients with GVHD and highest in those without GVHD (2), and (ii) complete remissions following relapse can be achieved by administering viable lymphocytes from the original donor (3). A syngeneic donor is immunologically identical to a recipient (i.e., twin sibling). There is no risk of developing GVHD and no risk for contamination of transplanted stem cells by tumor cells.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

329

In autologous transplantation, the donor is also the recipient. There is no risk of GVHD or graft rejection. However, there is also no GVT effect. The patient receiving an autologous transplant may also relapse as a result of contamination with tumor cells. Multiple procedures have been developed to remove tumor cells, including antibodies against tumor antigens, in vitro incubation with chemotherapy, long-term culture of marrow, and positive selection of stem cells using CD34. Transplants can also be classified according to the source of hematopoietic stem cells. Bone marrow contains a high concentration of hematopoietic stem cells. Aspirates are usually obtained from posterior or anterior iliac crests. Peripheral blood contains hematopoietic stem cells, normally at very low concentrations. This concentration increases after administration of granulocyte-colony stimulating factor (G-CSF) or granulocyte-macrophage colony-stimulating factor (GM-CSF) or during recovery from intensive chemotherapy, at which time cells are obtained via leukopheresis. Peripheral blood stem cells result in more rapid hematopoietic recovery than autologous marrow. The third source for hematopoietic stem cells is umbilical cord blood. Engraftment is slower, leading to a longer period during which patients are susceptible to infections. However, patients are less likely to develop GVHD without losing the GVT effect. Because the quantity obtained from cord blood is small, cord blood transplants are usually more suitable for pediatric populations. Prior to transplantation, the patient receives a preparative or conditioning regimen that kills cancer cells and suppresses the immune system to prevent rejection of donor cells. This stage is known as conditioning. Medications are chosen depending on the underlying disease. The most common regimens may include a combination of busulfan, cyclophosphamide, melphalan, thiotepa, carmustine, etoposide, or total body irradiation. Nonmyeloablative allogeneic HSCT is a newer technique that uses lower doses of chemotherapy and radiation, relaying on the GVT effect to eradicate the disease. Hematopoietic stem cells are then introduced intravenously. The cells home to the bone marrow via integrins. Engraftment is first evident with recovery of peripheral counts. The time to engraftment depends on the source of the stem cells, whether growth factors are used following transplant, and which GVHD prophylaxis is administered. Table 2 Timeline of Common Neurologic Complications Complication Stem cell harvest

Conditioning Infusion Prior to engraftment and marrow reconstitution

Chronic complications (after bone marrow reconstitution)

Intracranial hypotension due to entry into subarachnoid space during bone marrow aspiration Worsening neurologic manifestations of underlying autoimmune syndrome, possibly related to G-CSF Chemoradiation toxicity (see Table 3) Encephalopathy due to DMSO Ischemic stroke, possibly related to DMSO or debris across a PFO Transient global amnesia Cerebrovascular accidents related to Aspergillus or infectious emboli CNS infections including Aspergillus and CMV Coagulopathies resulting in subdural hematoma Drug toxicities including tacrolimus and cyclosporine for GVHD causing PRES as well as antimicrobials causing seizures Idiopathic hyperammonemia Metabolic abnormalities Neuromuscular complications including steroid myopathy, pressure-related peroneal nerve palsies, Guillain-Barre syndrome Systemic organ failure CNS infections including toxoplasmosis, herpes viruses, and nocardia CNS manifestations of chronic GVHD including neuromuscular complications and CNS angiitis Drug toxicities including tacrolimus and cyclosporine for GVHD

Abbreviations: CNS, central nervous system; DMSO., dimethyl sulfoxide; PRES, Posterior reversible encephalopathy

330

Part VI / Complications of Cancer Therapy

Complications following HSCT have long been recognized. In general, these complications are due to chemoradiotoxicity, medication toxicity, metabolic abnormalities, organ failure, GVHD, infection, pancytopenia or platelet dysfunction. Acute GVHD develops within the first 100 days, and chronic GVHD persists beyond 100 days. GVHD may be prevented by early administration of immunosuppressants following transplant. A common regimen is prednisone with cyclosporine or tacrolimus. However, even with prophylaxis, clinically significant acute GVHD occurs in approximately 30% of patients with matched siblings as their donor and 60% of patients with unrelated donors (4). Chronic GVHD occurs in 20–70% patients surviving more than 100 days after allogeneic transplant (5). Risk factors for chronic GVHD include older age, patients receiving mismatched or unrelated stem cells, use of hematopoeitic stem cells obtained from peripheral blood as opposed to bone marrow, and those with a prior episode of acute GVHD (5,6). The medications used for the prevention and treatment of GVHD are another important source of morbidity. Neurological complications adversely affect survival in HSCT patients (7–9); 10–40% of patients undergoing HSCT will develop a clinically significant neurological complication (10). The literature remains limited, however, to clinical and autopsy series. In most clinical case studies, encephalopathy, central nervous system (CNS) infections, and cerebrovascular disorders are most common. In autopsy studies, cerebrovascular complications and CNS infections were the most common findings (8,11). The discrepancy between autopsy and clinical series is due to the inclusion of encephalopathies in clinical studies, which often do not result in death. This chapter summarizes the complications associated with HSCT affecting the central and peripheral nervous systems according to etiology or symptom. Table 2 provides a timeline of common neurological complications occurring at different stages before, during and after transplant.

2. TOXIC-METABOLIC DISORDERS Patients undergoing HSCT are exposed to a wide variety of toxic-metabolic insults, which may result in neurological manifestations include encephalopathy, headache, seizures, and even focal neurological deficits. While HSCT patients are susceptible to more common toxic-metabolic disorders, including acid-base disorders and electrolyte imbalances, this section will focus mainly on less common disorders specific to this population.

2.1. Encephalopathy Encephalopathy may be the most common neurological complication encountered in HSCT patients. It represents a diffuse process altering brain function and/or structure. Causes include infections, metabolic dysfunction, increased intracranial pressure, seizures, toxic agents, poor nutrition, and decreased oxygen to the brain. In a retrospective study of 116 adult HSCT patients, a depressed level of consciousness was a principal reason for admission to intensive care units and conferred a poor prognosis (12). Similarly, in a study of pediatric HSCT patients, encephalopathy was associated with a poor prognosis (13). The signs and symptoms of encephalopathy differ depending on the etiology and severity, but all patients have an altered mental status. Basic work-up should include evaluation for infections, electrolyte abnormalities, and drug levels. Patients with a persistent unexplained encephalopathy or with focal neurological deficits should undergo brain imaging. Brain MRI with and without contrast is recommended. Lumbar puncture may be indicated to look for infectious, inflammatory, or neoplastic processes. If subclinical seizures are suspected, an electroencephalogram may be indicated. 2.1.1. Chemotherapy-Induced Encephalopathy Chemotherapy is well known to cause encephalopathy in a dose-dependent relationship. Even lower doses, as used in nonmyeloablative transplants, can cause CNS dysfunction. The onset of signs or symptoms may be acute or delayed. Table 3 provides a list of chemotherapies commonly used in HSCT associated with encephalopathy. A general discussion of chemotherapy-induced CNS complications can be found in Chapter 17 of this volume. Acute encephalopathy can present within days of receiving chemotherapy. Pyrimidine analogues 5-FU (fluorouracil) and cytarabine (cytosine arabinoside, Ara-C, 1-B-D arabinofuranosylcytosine) may produce a dosedependent acute encephalopathy, which resolves over several weeks. High-dose cytarabine (HIDAC) therapy given intravenously, especially if infused within 1 hr, may cause a diffuse generalized encephalopathy characterized by somnolence, confusion, disorientation, memory loss, and psychosis.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

331

Table 3 Neurological Syndromes Associated with Commonly Used Chemotherapeutic Agents Agent

Syndromes

Busulfan

• Seizures are common, but readily prevented by phenytoin prophylaxis.

Carboplatin

• Microangiopathic hemolytic anemia with secondary cerebral ischemia and focal/diffuse neurologic dysfunction • Capillary leak syndrome • Peripheral neuropathy • Delayed onset encephalopathy (25–47 days after treatment) with lesions in basis pontis, corpus callosum, spinal cord, cerebral hemispheres. • Peripheral neuropathy • Impaired cognition • Posterior reversible encephalopathy syndrome

Carmustine (BCNU) Cyclophosphamide Cyclosporine Cytarabine (cytosine arabinoside, Ara-C, Cytosar)

Etoposide (VP-16)

FK506 (Tacrolimus) Fludarabine

Ifosfamide Melphalan Methotrexate

Thiotepa

CNS effects are not common with standard doses of cytarabine, but CNS toxicity may be associated with high-dose therapy. This includes: • Acute encephalopathy • Cerebellar syndrome occurring 3–8 days after initiation of treatment; patients with renal insufficiency have increased risk of cytarabine neurotoxicity. • Sudden onset severe encephalopathy 1–2 weeks after therapy; reversible with dexamethasone • Cerebral edema with capillary leak syndrome • Acute dystonia • Neuropathy • Posterior reversible encephalopathy syndrome • Reversible encephalopathies after conventional therapy • High doses can cause delayed, severe encephalopathy • Delayed encephalopathy with white matter changes on MRI and elevations in CSF protein • Early onset severe encephalopathy and seizures • High dose melphalan (200 mg/m2 ) may cause severe encephalopathy and generalized tonic-clonic seizures within weeks of transplant • Severe additive neurotoxicity with cranial irradiation • Reversible encephalopathy within 1 week after treatment onset • Short latency encephalopathy in patients given extremely high-dose thiotepa

Common Uses in HSCT Preparative regimen in allogeneic or autologous transplants for acute or chronic leukemia and other nonmalignant disorders. Most commonly used as part of allogeneic transplant in CML. Preparative regimen in autologous transplants

Preparative regimens in autologous transplants and less frequently allogeneic transplants Preparative regimen for autologous and allogeneic transplants. Post-transplant GVHD prophylaxis and treatment Preparative regimens in allogeneic transplants

Preparative regimen for autologous transplants

Post-transplant GVHD prophylaxis and treatment Reduced intensity conditioning regimen

Salvage therapy with autologous transplants Preparative regimen in peripheral blood stem cell transplants for multiple myeloma Post-transplant GVHD prophylaxis

Preparative regimen in allogeneic transplants

332

Part VI / Complications of Cancer Therapy

HIDAC more commonly causes a cerebellar syndrome occurring 3–8 days after initiation of treatment. The typical features of this syndrome include dysarthria, dysdiadochokinesia, dysmetria, nystagmus, and ataxia. Approximately 70% of patients with this cerebellar syndrome will recover, provided that the administration of cytarabine is promptly discontinued. A few will have persistent cerebellar damage due to permanent loss of Purkinje cells in the cerebellar hemispheres and vermis (14). Mechlorethamine is an alkylating agent that may result in both acute and chronic encephalopathy in a strong dose relationship (0.5–2.0 mg/kg). The acute encephalopathy often clears, but sometimes persists to become a chronic condition with personality changes, dementia, and confusion. CT scans may reveal ventricular dilatation but no intraparenchymal abnormalities (15). High-dose melphalan (200 mg/m2 ), commonly used in PBSCT for multiple myeloma, is reported to cause severe encephalopathy and generalized tonic–clonic seizures within weeks of transplant (16). Ifosfamide, used in high-dose chemotherapy regimens, causes a transient, subacute encephalopathy which can be prevented and treated with methylene blue (17). This was seen in a small number of patients (8–9%), but EEG abnormalities without clinical symptoms are noted more frequently (18). The clinical spectrum includes confusion, hallucinations, mutism, and, rarely, profound coma. Extrapyramidal symptoms with opisthotonus, choreoathetosis, and myoclonus may also occur (19). Predisposing factors to ifofsfamide-related encephalopathy include low serum albumin level (< 3.5 g/dl), elevated serum creatinine level, and the presence of pelvic neoplastic disease (18,20). Metabolites of ifosfamide such as chloracetaldehyde are thought to be responsible for this syndrome. Delayed encephalopathy, occurring months after undergoing HSCT, has also been attributed to chemotherapy. Purine analogues such as fludarabine, cladribine, and pentostatin produce a delayed encephalopathy with white matter changes on MRI and elevations in CSF protein (21). Calcineurin inhibitors, cyclosporine and tacrolimus, are known to cause reversible encephalopathy with MRI findings; this is discussed further below. The combination of methotrexate and cranial irradiation produces white matter damage in 1.7% of patients (22). 2.1.2. Antimicrobial-Induced Encephalopathy Antimicrobial agents listed in Table 4 have been reported in association with encephalopathy with or without seizures, but usually without abnormalities on neuroimaging studies (23–25). Antiviral agents such as acyclovir, ganciclovir, and foscarnet are now established agents in the preemptive management of cytomegalovirus-positive HSCT patients, particularly in cases of allogeneic marrow recipients with GVHD. The main side effects of these drugs are neutropenia (thus creating a new risk for other infections), thrombocytopenia, and nephrotoxicity. Particularly in combination with renal dysfunction, reports of encephalopathy with seizures, tremors, headache, and vertigo have been associated with the use of these agents. Exceptional cases of coma, usually reversible, have resulted from both oral and intravenous use of acyclovir (27–29). 2.1.3. Encephalopathy Related to Organ Failure Encephalopathy may arise in the setting of liver, lung, or kidney dysfunction. Veno-occlusive disease (VOD) of the liver occurs in up to 54% of HSCT recipients, is strongly associated with multi-organ failure, and is the cause of death in almost 15% of cases (29). Hemolytic–uremic syndrome (HUS) and intravascular hemolysis with renal insufficiency (due to acute radiation nephritis) are well-known complications in this patient population, occasionally progressing to coma and death (30,31). CNS dysfunction after HSCT may be an early manifestation of a systemic disorder known as multiple organ dysfunction syndrome (MODS) (9). Although it is unclear what leads to MODS, it may arise from systemic inflammatory response syndrome (SIRS), a complex interaction of cytokines, coagulation proteins, complement cascade, lymphocytes, phagocytes, platelets, vascular endothelial cells, and other unrecognized factors (32). The preparative regimen may be the inciting stimulus. CNS, pulmonary, or hepatic dysfunction develops 2 weeks later. Patients who ultimately develop MODS have had more manifestations of SIRS, for longer periods of time, than those who do not develop organ dysfunction. Patients presenting with either pulmonary or CNS dysfunction were up to five times more likely to subsequently develop other organ dysfunction and up to eighteen times more likely to die than patients without pulmonary of CNS dysfunction (32). MODS is more commonly found in patients who have undergone an allogeneic transplant and in patients who have received a total-body irradiation-based preparative regimen.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

333

Table 4 Drugs Used in HSCT That Cause Seizures ± Encephalopathy Antineoplastic agents Cytarabine (Ara-C) Busulfan Methotrexate BCNU Mechlorethamine Ifosfamide Cisplatin Immunosuppressive agents Cyclosporine Tacrolimus (FK-506) Muromonab-CD3 (OKT3) Antimicrobial agents Aminoglycosides Penicillins Cephalosporins Imipenem Vancomycin Isoniazid Metronidazole

Antiviral agents Acyclovir Ganciclovir Foscarnet Antifungal agents Amphotericin B Miscellaneous Lidocaine Narcotics Theophylline Tricyclic antidepressants Antipsychotics Aqueous iodinated contrast agents

2.1.4. Immunosuppressive Neurotoxicity and Posterior Reversible Encephalopathy Syndrome Neurotoxicity from immunosuppressive agents commonly used to prevent and treat GVHD, including cyclosporine, tacrolimus (FK506), and muromonab-3 (OKT3), may cause early neurological manifestations usually without abnormalities on brain MRI. The early symptoms occurring within 2 weeks after transplantation include postural tremor (usually mild), delusions, visual hallucinations, and disorientation to time and place. Additionally, cyclosporine can cause nonsensical speech. Tacrolimus use may result in perioral paresthesias, hyperesthesias of the hands, and “restless legs.” All three agents can cause headache (33). Because the aforementioned symptoms and signs are now well known among transplant physicians, discontinuation or adjustment of the dose of a given immunosuppressive drug usually results in improvement. Unfortunately, serum levels are not a sensitive diagnostic marker of severe tacrolimus-induced neurotoxicity so diagnosis depends on clinical symptoms (34). Immunosuppressive neurotoxicity occurs in most patients during intravenous loading of tacrolimus or cyclosporine, and is much less common and severe after the post-transplant phase has subsided and the patient is stable. A more severe manifestation of cyclosporine or tacrolimus neurotoxicity is posterior reversible encephalopathy syndrome (PRES) (35,36). These medications are effective agents for prevention of GVHD after allogeneic stem cell transplants. The overall incidence of tacrolimus-induced PRES is 1.6% (34). Most cases occur in the early post-transplantation period. Patients may present with headache, seizures, visual changes, and mental status changes. The pathogenesis regarding PRES is poorly understood. Cyclosporine and tacrolimus may have direct toxic effects on vascular endothelium, resulting in endothelial dysfunction, vasospasm, activation of the coagulation cascade, and microthrombosis (37). Other hypotheses regarding medication-related PRES include secondary hypertension and hypomagnesemia, but tacrolimus or cyclosporine-induced PRES may arise even in the absence of hypertension or hypomagnesemia. Additionally, this syndrome may occur even while drug levels are in therapeutic range (38). MRI is the most sensitive diagnostic test for PRES (see Fig. 1). Imaging demonstrates vasogenic cerebral edema, particularly in the posterior circulation, demonstrated as T2 and FLAIR hyperintensities on MRI. These changes are similar to those found in hypertensive encephalopathy. PRES predominantly involves the white matter, but gray matter may also be affected. PRES is often reversible by decreasing the dosage or withholding the drug

334

Part VI / Complications of Cancer Therapy

Fig. 1. PRES due to tacrolimus. MR imaging in a patient with tacrolimus neurotoxicity. Note diffuse white matter hyperintensities (A,B) with no specific predilection for the posterior hemispheres. The diffusion-weighted MR (C,D) was normal (further mapping showed increased ADC levels consistent with vasogenic edema).

for a few days. In one series, most patients had no recurrence of symptoms with continuation of treatment (34). Although signs and symptoms typically resolve, there are reported cases of irreversible PRES. Other medications used during HSCT may cause PRES, although less frequently. Vincristine, given as induction therapy, caused reversible encephalopathy with seizures (39). Bevacizumab-induced PRES is also described in the literature, with one case series including a BMT patient (37). However, this patient was also on cyclosporine. High-dose cytarabine, cisplatin, and granylocyte-stimulating factor result in a reversible leukoencephalopathy, but is not described thus far specifically in the HSCT population. PRES demonstrated radiographically has occurred following infusion of DMSO-cryopreserved autologous stem cells (40).

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

335

A syndrome involving irreversible leukoencephalopathy of unknown etiology has also been described in the literature. In these cases, the symptoms did not resolve after stopping cyclosporine. It is unclear if this represents nonreversible PRES or an entirely different entity. 2.1.5. Wernicke’s Encephalopathy Conditioning therapy and GVHD may cause side effects such as nausea, vomiting, mucositis, and diarrhea, necessitating total parental nutrition (TPN) to meet caloric and nutrient demands. These patients are at risk for thiamine (vitamin B1) deficiency because TPN does not contain thiamine (41,42). Wernicke’s encephalopathy is a metabolic disorder caused by thiamine deficiency, classically characterized by the triad of ataxia, ophthalmoplegia, and altered mental status. Disease onset may be gradual, evolving over several days. Autopsy studies suggest that 6% of patients will develop Wernicke’s encephalopathy within the first month following BMT. Raspberry tongue may be an important early clinical finding (42). Rapid onset encephalopathy follows as well as severe metabolic acidosis, confusion, papilledema, nystagmus, blindness, coma, and possibly death. Symptoms were first noted within 2–6 weeks post-transplant in patients receiving less than 100 mg of thiamine per day (42). Brain MRI may reveal contrast enhancement of the mamillary bodies (43). Neuropathological findings include petechial hemorrhages in the periventricular gray matter, medulla oblongata, and thalamus, but surprisingly none in the mamillary bodies (42). In one case report, thiamine treatment resulted in dramatic improvement but neurological deficits did not completely resolve (41). Patients on long-term TPN and glucose-containing intravenous fluids require larger amounts of thiamine to metabolize their carbohydrate intake. Therefore, any patients receiving long-term TPN should be placed on thiamine replacement (100 mg IV daily) to prevent Wernicke’s encephalopathy.

2.2. Central Pontine and Extrapontine Myelinolysis Central pontine myelinolysis (CPM) and extrapontine myelinolysis (EPM) are well recognized in orthotopic liver transplant patients and have been reported in patients after HSCT. CPM is noninflammatory demyelination within the basis pontis, most commonly due to rapid correction of severe, prolonged hyponatremia. The clinical features include confusion, ophthalmoplegia, pseudobulbar palsy, and spastic quadriplegia. EPM is similar to CPM in pathology and etiology but also affects extrapontine white matter. Clinical manifestations in EPM may include mutism, parkinsonism, dystonia, and catatonia. In HSCT patients, cyclosporine may play a role (44,45). Brain MRI findings are diagnostic with T2 and FLAIR hyperintensities in the areas where demyelination occurs (see Fig. 2). Pathogenesis may be related to osmotic disturbances. Prognosis for recovery is generally poor.

2.3. Idiopathic Hyperammonemia HSCT patients may develop a rare but fatal complication known as idiopathic hyperammonemia (IHA). It occurs during and shortly after high-dose chemotherapy when the patient is neutropenic (46,47). IHA presents as a progressive encephalopathy characterized by acute onset of lethargy, confusion, disorientation, and agitation, and may be accompanied by seizures and cerebellar dysfunction. Rapid deterioration to coma and death frequently follow, sometimes within 24 hrs of the first symptoms (48). The diagnosis is based on elevated serum ammonia levels (> 70 μmol/L), normal liver function tests, normal plasma amino acid levels (which rule out enzymatic deficiencies of the urea cycle) and neurologic deterioration without any other obvious etiology. Usually these patients have a profound leukopenia and respiratory alkalosis. Neuroimaging studies of the brain may be normal (see Fig. 3), and EEG shows nonspecific abnormalities consistent with a metabolic encephalopathy (diffuse slow background activity). The etiology of IHA remains unclear. Autopsy studies show diffuse cerebral edema due to astrocytic swelling. There are clinical similarities to fulminant hepatic failure and inherited defects of urea synthesis, but without the typical abnormal laboratory findings. IHA is also clinically similar to Reye’s syndrome, but no ultrastructural abnormalities of the mitochondria have been found in IHA as seen in Reye’s syndrome (48). Heterozygosity for ornithine carbamoyltransferase deficiency has to be excluded. Finally, hyperammonemia may be caused by asparaginase (49) and valproic acid. Absolute levels of plasma ammonia do not correlate well with the severity of the clinical findings. There is no data in IHA patients regarding glutamine levels in the CSF, which correlates with mental status changes in hepatic encephalopathy; glutamine is the primary metabolic

336

Part VI / Complications of Cancer Therapy

Fig. 2. Central pontine myelinolysis. Axial MRI scan (FLAIR A; T1-weighted with gadolinium, B) of a 48-year-old female 9 months after match-related T-cell depleted bone marrow transplant for CLL. An episode of headaches, fever, nausea, and vomiting preceded mental status changes, somnolence, and coma. Hyperintense lesions in the tegmentum pontis and dorsal portion of medulla oblongata were noted at that time. All metabolic and infectious studies in serum and CSF were unremarkable. Tonic–clonic seizures developed and were treated with phenytoin. Spontaneous clinical and radiographic recovery were seen after one week (C and D).

product of ammonia in the brain. Davies et al. speculate that the IHA following chemotherapy may be due to the combination of increased catabolism and high exogenous nitrogen load such as GI bleeding (50). Another hypothesis postulates that IHA is due to an acquired deficiency of glutamine synthetase, resulting in an inability to convert ammonia to glutamine (51). Treatment should consist of a reduction of the exogenous nitrogen load (i.e., discontinue parenteral nutrition, treat gastrointestinal bleeding) as well as hemodialysis and ammonia-trapping therapy to increase nitrogen excretion. Ammonia-trapping treatment consists of infusion of sodium benzoate or sodium phenylacetate with the resultant compounds hippurate or phenylacetylglutamine, respectively, being excreted in the urine without further metabolism. Although a reversible stage of coma in IHA has been postulated and some HSCT patients survive this complication, most patients progress into an irreversible coma and die. Therefore, early recognition of IHA may prevent this dismal prognosis.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

337

Fig. 3. Idiopathic hyperammonemia. Nonspecific, multiple, small, high-intensity lesions (FLAIR images; (A), nonenhancing (T1-weighted, with gadolinium) (B) compatible with ischemic small vessel disease on a MRI scan in a 51-year-old female 5 months after allogeneic bone marrow transplant for recurrent NHL. She also had a 4-year history of hypertension and had developed chronic GVHD. Rapidly progressive coma (< 48 hrs) set in, with respiratory alkalosis. Serum ammonia level was 451 μmol/L, liver enzymes were only mildly elevated, liver biopsy was normal. Treatment with lactulose resulted in decreased ammonia level (61 μmol/L) and improved neurological status. Within one week, however, her ammonia level was again up to 376 μmol/L. Generalized seizures developed and she expired in spite of treatment with phenylacetate/benzoate.

2.4. Seizures The incidence of seizures (generalized, partial, or status epilepticus) in HSCT patients varies between 3% and 29% (52,53). The underlying primary disease may play a role, with the highest percentage of seizures seen in patients treated for sickle cell anemia (53). Potential causes of seizures in HSCT patients are numerous. Head imaging may be normal or abnormal depending on the underlying etiology. CNS infections, infarctions, hemorrhages, and leukoencephalopathies often produce imaging findings, whereas electrolyte disturbances, acidbase abnormalities, and drugs are typically associated with normal head imaging. The most commonly implicated drugs are listed in Table 4. Several of these agents are part of the conditioning regimen given prior to HSCT, whereas others are used to treat the underlying malignancy or complications of HSCT. High-dose busulfan causes seizures in approximately 10% of patients undergoing HSCT with a busulfanbased conditioning regimen. Generalized tonic–clonic seizures may be related to a direct toxic effect of busulfan on the cerebral cortex due to its rapid entry into the cerebrospinal fluid (54). Prophylaxis with phenytoin commencing 2 days prior to busulfan therefore may be reasonable (55). Methotrexate used for GVHD prophylaxis is another cause of seizures in the HSCT population. Neuroimaging studies are usually normal, but the more serious long-term and frequently fatal, chronic leukoencephalopathy resulting from combined intrathecal methotrexate and whole-brain radiation therapy will yield abnormal imaging results (56). HIDAC therapy may also cause seizures in addition to encephalopathy. These seizures usually generalized tonic–clonic, but complex partial seizures may also occur. Seizures related to cyclosporine or tacrolimus usually occur with high drug levels, but can also occur with normal or even low levels. Seizures may be due to direct neurotoxic effects or related to PRES. Patients with thalessemia receiving busulfan and cyclophosphamide as a preparative regimen are particularly at risk for cyclosporine-induced seizures. Prophylactic anticonvulsant medication, such as phenytoin or levetiracetam, should be considered in all patients receiving busulfan/cyclophosphamide as conditioning therapy, even though this cannot always prevent the occurrence of seizures (57). Antibiotics such as cefepime and imipenem are also associated with seizures and nonconvulsive status epilepticus. Particularly pertinent to HSCT is a late-onset seizure as a first manifestation of recurrent primary disease in the brain in patients who have undergone HSCT for leukemia or lymphoma (58). Even more uncommon, but disproportionately overrepresented in HSCT patients, are de novo neoplasms of the CNS (lymphomas and gliomas) as a late complication of HSCT (59–61). There is a seven-fold increase of malignant tumors of the CNS

338

Part VI / Complications of Cancer Therapy

in long-term survivors of HSCT, particularly in patients with leukemia transplanted at a young age (62). Children who received cranial or craniospinal radiation therapy as prophylaxis as part of their transplant procedure are also at risk (63).

2.5. Toxic Neuropathies Chemotherapy is a common cause of peripheral neuropathy in HSCT patients. Toxic neuropathies are discussed in Chapter 17.

3. CEREBROVASCULAR DISORDERS Cerebrovascular accidents occur in approximately 3% of HSCT patients (64). They develop a median of 28 days after HSCT and are associated with poor outcomes; hospital mortality after CVA was 69.4% in one series (64). Risk factors for cerebrovascular accidents include thrombocytopenia, platelet dysfunction, underlying disease, coagulopathies, infections, metabolic disorders, leukopenia, intrathecal chemotherapy, irradiation, lumbar punctures, mechanical ventilation, transplant type, and nonbacterial thrombotic endocarditis. Management following CVA is similar to that in the nontransplant population with two notable exceptions. Given a relatively higher frequency of cerebrovascular accidents associated with infections, a thorough search for an underlying infection should be performed. Thrombocytopenia should also be corrected in patients with hemorrhagic lesions. Chapter 14 is devoted to cerebrovascular complications in cancer. The following sections summarize cerebrovascular complications specific to the HSCT population.

3.1. Intracranial Hemorrhages Intracranial hemorrhage (ICH) includes intraparenchymal hemorrhages (IPH), subarachnoid hemorrhages (SAH), and subdural hemorrhages (SDH). Mortality rate from ICH is high. The prevalence in HSCT patients varies between autopsy series but may be as high as 32.2% (65). Subdural hemorrhages are strongly associated with intrathecal methotrexate conditioning, post-lumbar puncture headache, prolonged thrombocytopenia, underlying malignancy, and coagulopathy. The incidence in the HSCT population is approximately 2% (66). Computed tomography is the usual diagnostic modality of choice, but initial scans may be negative (66). In one series, definitive diagnosis could only be made by MRI in 25% of cases (67). SDH should not be confused with subdural hygroma, which refers to a subdural collection of cerebrospinal fluid as opposed to blood. SDH may be managed conservatively with platelet transfusion and correction of coagulopathy (66). The need for surgical intervention depends on the size of the SDH and whether or not there is mass effect and significant neurologic compromise. Subarachnoid hemorrhages are rare in the HSCT population and are typically located in sulci as opposed to basal cisterns. Mycotic aneurysms should be suspected in appropriate clinical circumstances. SAH most frequently occur in the setting of relapsing leukemia; many of these patients concomitantly display epistaxis and pulmonary hemorrhage. Outcome in this population is determined by control of blast crisis and rarely by the SAH. Intraparenchymal hemorrhages may be due to infestation with Aspergillus species or cerebral venous sinus thrombosis (68,69). Lobar hematomas have been linked to cyclosporine toxicity due to direct damage to endothelial cells (70). Intracerebral hemorrhage is often a fatal event in a moribund relapsed end stage patient.

3.2. Ischemic Strokes Ischemic lesions occur less frequently than hemorrhagic lesions in HSCT patients and present with focal neurological deficits. Traditional risk factors including hypertension and atrial fibrillation are important considerations. Other less common etiologies seen in the HSCT population include hypercoagulable states, paradoxical emboli during infusions, medication toxicity, infectious vasculitis, endocarditis (either nonthrombotic or infectious), GVHD vasculitis, or post-transplantation thrombotic microangiopathy. Protein C deficiency as well as deficiencies of factors XII and VII, antithrombin 3, and the presence of anticardiolipin antibodies may contribute to a hypercoagulable state in HSCT patients (71–74). Additional factors are endothelial cell injury and microangiopathy possibly due to cyclosporine and methylprednisolone (75,76).

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

339

Stem cell infusions have been associated with ischemic strokes, possibly due to direct toxicity from a component of the preserving solution, 10% dimethyl sulfoxide (DMSO) (77), or due to paradoxical embolism of debris across a patent foramen ovale (79). Aspergillus has a predilection for blood vessels, resulting in arterial occlusion by hyphal elements and thus ischemic strokes as well as secondary hemorrhagic conversion. Aspergillus is further discussed in Section 4 below. A well-known predisposing mechanism to embolic infarction is nonbacterial thrombotic endocarditis (NBTE) (79). One autopsy series found an increased prevalence of NBTE in BMT patients (80). Factors predisposing to NBTE may include preparatory regimens that damage cardiac endothelium such as high-dose cyclophosphamide and radiation, fibrin, and other debris infused during transplant and GVHD. Antemortem diagnosis of NBTE is difficult because vegetations are often too small to be detected by echocardiography (80). Detection of disseminated intravascular coagulopathy should increase the index of suspicion for NBTE in patients with arterial occlusions. Thrombotic microangiopathy syndromes (TMS) such as hemolytic uremic syndrome (HUS) or thrombotic thrombocytopenic purpura (TTP) may result in fluctuating neurological deficits (81,82). Management of TMS should include discontinuation of calcineurin inhibitors, which may be one cause of this syndrome. Response to plasma exchange in HSCT-related TMS is generally poorer than TMS unrelated to HSCT, with less than 50% response rates and mortality rates greater than 80% in patients treated with plasma exchange (82). Primary angiitis is considered unlikely after HSCT, but CNS angiitis has been reported in the literature. These cases may be a consequence of long-standing GVHD (83–85). Symptoms may include acute to subacute onset confusion, seizures, focal neurological deficits, or neuropsychological impairment. MRI reveals multifocal or confluent white matter changes (83), but these findings are sometimes delayed by several weeks, lagging behind clinical symptoms (84). Diagnosis may require brain biopsy because cerebral angiograms do not always demonstrate classic manifestations. Response to cyclophosphamide and corticosteroids is good clinically and radiographically.

4. INFECTIONS It has been reported that 5–8% of HSCT patients with infectious complications develop an infection involving the CNS. Even though it is less common than other systemic infections, CNS infections can be a highly fatal event, particularly fungal and Toxoplasma gondii infections (86). The risk of infection in the post-transplantation period is based on the status of immune system recovery. Prior to engraftment, patients are neutropenic and their mucosal barriers are disrupted by medications. The most common pathogens encountered systemically are bacteria, Candida and herpes simplex virus (HSV). If the period of neutropenia is prolonged, the risk of Aspergillus increases. In the early post-engraftment period, patients have deficient cellular immunity caused by acute GVHD and immunosuppressant medications. In this setting, fungal, CMV, and gram-positive bacterial infections are seen. In the late post-engraftment period (typically within 3 months, but can last up to one year), patients are susceptible to encapsulated bacteria and HSV. Table 5 demonstrates compares MRI characteristics of focal CNS infections. These are discussed in more detail below.

4.1. Bacterial Infections Autopsy studies include cases of bacterial abscesses due to Staphylococcus epidermidis and Staphylococcus aureus (86). All patients had diffuse cerebritis with none dying from their CNS infections. Patients initially presented with diffuse alveolar infiltrate. The etiologic agents were all previously identified via positive blood cultures. Nocardiosis is a rare opportunistic infection occurring in 0.3% of patients undergoing allogeneic BMT (87). It is a branching gram-positive rod and usually establishes infection through inhalation of aerosolized organism. The lungs are most commonly affected, but hematogenous spread may occur to skin, bone, and CNS. The CNS manifestations include parenchymal abscess, meningitis, cerebritis, or meningocerebritis, presenting as fever with headaches, altered mentation, seizures, and focal deficits. Yet, despite its special tropism for neural tissues, cerebral nocardiosis in HSCT patients has only been described in a few case reports (88–92). Most patients received

340

Part VI / Complications of Cancer Therapy

Table 5 MRI Characteristics of Focal CNS Infections in HSCT Patients Infections

Preferential Location

Nocardiosis

Any site

Aspergillosis

Subcortical cerebral/cerebellar hemispheres; basal ganglia Basal ganglia; cortical medullary junction of cerebral/cerebellar hemispheres Medial temporal; multiple lobes

Toxoplasmosis

Herpes virus infections (HSV; HHV-6) Progressive multifocal leukoencephalopathy

Cerebral subcortical white matter at any site, deep white matter later

Number of Lesions

T1

T2/FLAIR

Enhancement

Multiple > Solitary Multiple

↓

↑

+

↓

↑

+

Multiple

↓

↑

+

Solitary (uni/bilateral) or multiple Multiple

↓

↑

+

↓

↑

+ or −

allogeneic transplants and were on immunosuppressants for chronic GVHD. Predisposing factors for developing nocardiosis included active acute or chronic GVHD, neutropenia, and a lack of trimethoprim-sulfamethoxazole (TMP-SMX) prophylaxis (87). Patients most often presented with pulmonary symptoms. Presentations of CNS disease included confusion, fever, headaches and gait instability. Other patients had no neurologic symptoms. Nocardia asteroides was found in sputum cultures, bronchoalveolar lavage, lung biopsy, and brain abscess aspirate. One case grew Nocardia nova instead. Results from imaging studies (CT or MRI) were variable; some patients had multiple, enhancing lesions while others had no findings. Nocardia asteroides is usually sensitive to TMP-SMX, amikacin, imipenem, and minocycline. Aggressive treatment with TMP-SMX and a second agent is necessary to avoid treatment failure or progression (91). Most cases also required surgical drainage of the cerebral abscess. Mycobacterium tuberculosis (MTB) is a slow-growing acid-fast bacillus that also establishes infection through inhalation of aerosolized organisms. Tuberculosis is less frequent in the HSCT population than other immunocompromised populations for unclear reasons. Incidences of MTB range from 0.5% to 3.0%, mostly in T-cell depleted allogeneic graft recipients and those with severe chronic GVHD (93). The typical CNS manifestation of MTB is subacute meningitis with headaches, fever, altered consciousness, sixth or third cranial nerve palsies, and papilledema. CSF abnormalities initially include polymorphic pleocytosis but later evolve into classic lymphocytic pleocytosis, hypoglycorrhacia. and elevated protein. CNS involvement of MTB in HSCT patients is rarely described in the literature (93–96). Two cases involve patients with tuberculous meningitis. CT scan demonstrated either no abnormalities (95) or only ventricular dilatation (96). The latter patient developed a focal lesion on a CT scan 48 hrs later. Another case involves a patient who developed a temporal lobe tuberculoma 3 months after diagnosis of pulmonary tuberculosis despite tuberculostatic therapy (93). Three of five reported patients with CNS tuberculosis died. Treatment of CNS tuberculosis involves multiple drug therapy including isoniazid, rifampin, and pyrazinamide. Resistance requires the addition of other drugs such as ethambutol or streptomycin. Lyme disease is due to infection with Borrelia burgdorferi, a spirochete that is transmitted by deer tick. A case of Lyme meningoradiculits and myositis was reported following allogeneic hematopoietic stem cell transplantation (97). Diagnosis may be made by serologic and CSF testing for B. burgdorferi, antibodies and PCR. Treatment of Lyme neuroborreliosis requires intravenous ceftriaxone.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

341

4.2. Fungal Infections Aspergillus fumigatus is the most common cause of fungal CNS infections in HSCT patients. In 25–50% of patients with invasive aspergillosis, the CNS is involved at the time of diagnosis of the systemic infection. Infection usually occurs following engraftment. Most patients with CNS involvement already have evidence of pulmonary disease and abnormal findings on head imaging (either single or multiple lesions). The usual route of infection is inhalation of excessive Aspergillus spores in contaminated air with resultant invasive disease in the lungs and paranasal sinuses. Pulmonary symptoms and signs include dyspnea with cough, pleurisy, or hemoptysis. Hematogenous dissemination occurs because of blood vessel invasion. After the brain parenchyma is invaded, this same pathophysiological mechanism will result in the development of arterial occlusion by hyphal elements resulting in infarction and frequently secondary hemorrhagic conversion. These lesions may be localized in the subcortical areas of the cerebral hemispheres, the cerebellum (98), or the basal ganglia (99). Rarely, patients may develop fungal vasculitis or mycotic aneurysms with resultant SAH. Involvement of the meninges in the inflammatory process is distinctly uncommon. Clinical and laboratory diagnosis of aspergillosis is difficult. Presenting symptoms are nonspecific and may include hemiparesis, unilateral cranial nerve palsies, intention tremor, seizures, headaches, or dysmetria (98,99). Fever and nuchal rigidity may be absent. There is a relative paucity of CSF abnormalities. Pleocytosis (usually a mix of polymorphonuclear and mononuclear cells) is usually less than 100/mm3 ; CSF protein content is only mildly elevated and glucose level is normal or mildly decreased. CSF cultures for Aspergillus are rarely positive (99). Serologic testing in immunocompromised patients yields inconclusive results. MRI is the imaging modality of choice (see Fig. 4). Two different patterns are described: (i) nonenhancing lesions located in the basal ganglia and thalami representing small infarctions of the lenticulostriate and thalamoperforator arteries, and (ii) large cerebral artery infarctions with early intravascular and meningeal enhancement (99). Most cases do not demonstrate contrast enhancement but ring or nodular enhancement has been described in patients who survived (100). Because the corpus callosum is not commonly involved in pyogenic infection or thromboembolic infarctions, callosal lesions are suggestive of aspergillosis (100). Early diagnosis is important because mortality is almost 100% in most series with only a few case reports of HSCT patients surviving CNS aspergillosis (55,98,99,101,102). Survival is usually only 2 days to 3 weeks after onset of neurological symptoms (100). Diagnosis of invasive aspergillosis in a patient with characteristic brain MRI imaging may be established by biopsy of a lung lesion. Occasionally biopsy of a cerebral lesion may be necessary to document fungus (101,102). Treatment of CNS aspergillosis consists of amphotericin B, usually at 1.0–1.5 mg/kg/day intravenously. Lipid formulations may be used instead as initial therapy (at > 5mg/kg/day) because of increased efficacy and significantly less nephrotoxicity. A second drug such as rifampin (600–900 mg/day in divided doses) is frequently added in CNS aspergillosis patients given their dismal prognosis. Duration of treatment is unknown but should probably be continued for 2–3 months after the MRI scan has normalized. Prevention of aspergillosis is very important and includes clear air supply by high-efficiency particulate air filters on the hospital ward; prevention of CMV infection (which seems to predispose to invasive aspergillosis) and preemptive therapy with amphotericin B after colonization of airways with Aspergillus species has been discovered. Candida is a common systemic infection in HSCT patients but rarely leads to CNS involvement. The risk of candidal infection is closely linked to neutropenia. In granulocytopenic HSCT patients, candidiasis is often disseminated, involving the liver, spleen, kidney, heart, gastrointestinal tract, lungs, skin, and brain (103). Mortality rates with disseminated Candida may be as high as 90% and is almost always fatal when brain parenchyma is involved (86). Patients with CNS Candida tend to be asymptomatic, neutropenic, and fungemic with normal CNS imaging studies. One retrospective series of 58 HSCT patients identified 19 patients with Candida abscesses (15 Candida albicans, 2 Candida tropicalis, 2 unknown species) (104). Only one patient survived, but ultimately died from congestive heart failure. Of these 19 patients, 12 had positive blood cultures. Another study by Maschke and collaborators reported one patient with Candida encephalitis occurring 24 days after BMT (105). Brain MRI demonstrated multiple lesions in the basal ganglia and cerebellum, which were hypointense on T1-weighted sequences, intermediate signal on T2-weighted images, and ring-enhancing after gadolinium administration. The

342

Part VI / Complications of Cancer Therapy

Fig. 4. Aspergillus. Axial MRI, T2-weighted (A and B) and T1-weighted post-gadolinium (C and D) of Aspergillus flavus lesions in the brain of a 31-year-old male 10 months after allogeneic bone marrow transplantation for AML. He was admitted one month prior to death with respiratory insufficiency, and developed generalized tonic–clonic seizures two weeks later. Autopsy confirmed the presence of A. flavus from a right temporal lesion. Note the very faint enhancement of the left basal ganglia and left frontal lesions.

patient died after 19 days from respiratory failure. All allogeneic recipients and select autologous recipients should receive fluconazole prophylaxis (400 mg per day) during neutropenia to prevent invasive disease (106). Other fungal infections are uncommon in the HSCT population, including Coccidiodes immitis, Crypotoccus neoformans, and Histoplasma capsulatum (107). Saprophytes of the Zygomycetes group, most notably Rhizopus, Absidia, and Mucor, may occasionally produce brain abscess in HSCT patients, usually under the predisposing circumstances of neutropenia, ketoacidosis due to hyperglycemia, and corticosteroid use. These species are responsible for a rhinocerebral zygomycosis with or without concomitant pulmonary infection. Invasion directly from the involved nasal sinuses is usually seen. Prognosis is as grim as with the previously discussed fungal CNS infections, in spite of treatment with conventional or liposomal amphotericin B and surgical debridement of the nasal sinuses (107,108). Finally, a unique and again fatal case of Microsicinereus brain abscess in a 28-year-old female recipient of an allogeneic HSCT has been reported (109).

4.3. Parasitic Infections Toxoplasmosis is caused by Toxoplasma gondii, an intracellular protozoan parasite, and most often affects patients who are immunocompromised or pregnant. Transmission may occur transplacentally, via ingestion of raw or undercooked meat containing T. gondii cysts, or by exposure to oocytes from cat feces. It is seen in 0.3%

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

343

of HSCT patients. However, this incidence increases to 2% in patients who had positive toxoplasma serology prior to transplant (110,111). Most cases are seen after allogeneic transplant but there are reported cases following autologous transplant. Clinical disease usually results from reactivation of latent disease during immunosuppression, particularly with concurrent GVHD. Rarely, it occurs as a primary infection acquired from the allograft into the seronegative recipient (110). Following a mononucleosis-like prodromal stage, toxoplasmosis disseminates to the lungs, liver, bone marrow, and brain. Encephalitis is the most common CNS presentation. Clinical symptoms and signs include headaches, low-grade fever, lethargy, focal seizures without or with secondary generalization, and focal neurological deficits depending on the location of the lesions. Symptom onset usually occurs between the second and sixth month following transplant, but may occur as early as 9 days. The majority presents within 3 months (111–115). Definitive proof of CNS toxoplasmosis relies upon histologic demonstration of tachyzoites of T. gondii from brain biopsy (112,115,116). Serological tests measuring IgG antibodies against T. gondii in blood are of limited value given the prevalence of latent infection. Increased IgM levels may indicate recent activation of infection, but false-positive and false-negative cases have been reported (110,112,117,118). Routine CSF parameters such as cell count and protein level are either normal or mildly elevated due to the immunosuppressed state of the patient. PCR assay in CSF may be a useful diagnostic tool in the early detection of T. gondii (116,119). However, even PCR assay in CSF may be negative, at least early in the infection (111). Multiple lesions in the basal ganglia and at the cortico-medullary junction of the cerebral and cerebellar hemispheres are usually present, with low/isointense signal on T1-weighted and isointense/high signal on T2-weighted MR images representing central coagulation necrosis (see Fig. 5) (115). Enhancement after gadolinium administration may or may not be present, depending on the ability of the patient’s immune system to muster a meaningful inflammatory response (115). Leptomeningeal enhancement (120) and hemorrhage (121) are rare. Differential diagnosis based on MRI characteristics may include other opportunistic infections such as aspergillosis, mucormycosis, as well as progressive multifocal leukoencephalopathy and post-transplant lymphoproliferative disorders/primary CNS lymphoma. Additional techniques such as positron emission tomography (PET), single-photon emission CT (SPECT), or MR spectroscopy (MRS) may be useful for differentiation between these possibilities. The Centers for Disease Control (CDC) recommends prophylaxis for seropositive allogeneic recipients with active GVHD or a prior history of toxoplasmic chorioretinitis (106). Trimethoprim-sulfamethoxazole 80 mg/400 mg once per day or 160 mg/800 mg three times per week should be started after engraftment and be administered for as long as the patient remains on immunosuppressive therapy (which is generally for 6 months following HSCT). It should be noted that this prophylactic regimen may not be sufficient for a seronegative recipient with a seropositive donor who then also develops severe GVHD. Treatment should consist of pyrimethamine (50–100 mg/day) and a sulfonamide (2–4 g/day). Intolerance to sulfonamides (allergic reactions, gastrointestinal symptoms) is not uncommon, at which time drugs such as clindamycin, atovaquone, azithromycin, or clarithromycin will need to be given. Prognosis of patients with toxoplasmic encephalitis is poor, with only 1 in 10 patients surviving. Patients who receive adequate therapy and those who develop late infection (> 63 days after HSCT) are less likely to die from toxoplasmosis (122). Acanthamoeba meningoencephalitis has been reported following HSCT (123,124). Acanthamoeba species are found in soil, dust, fresh water sources, sewage, heating, ventilatory, and air conditioning units. Infection is introduced via the skin or nostrils. It disseminates hematogenously to the lungs and the CNS, resulting in granulomatous amebic encephalitis. Onset is insidious. Patients may present with fever, respiratory symptoms, seizures, focal neurological deficits, and rapid mental status deterioration. CSF studies demonstrate increased white blood cells, elevated protein levels, and decreased glucose levels. Imaging studies are usually nonspecific although MRI may reveal a ring-enhancing lesion (124). Diagnosis is confirmed via brain biopsy or autopsy. No effective treatment exists. The mortality rate is nearly 100%.

344

Part VI / Complications of Cancer Therapy

Fig. 5. Toxoplasmosis. Multiple lesions in both cerebral and cerebellar hemispheres (A and B) in a 46-year-old female 8 months after allogeneic bone marrow transplantation for chronic myelogenous leukemia. She presented with dizziness, ataxia, nystagmus, and diplopia. Mild right upper motor neuron facial palsy and right upper extremity weakness were found. CSF analysis including toxoplasmosis IgG and IgM were normal; however, serum toxoplasmosis IgG, IgM, and IgG quantitative index were elevated. After prolonged treatment (> 4 months) with pyrimethamine, sulfadiazine, and leucovorin, there has been significant clinical improvement but only incomplete neuroradiological response (C and D).

4.4. Viral Infections HSCT patients are susceptible to viruses, but CNS involvement is relatively rare. Herpes viruses tend to present in a temporal sequence post-transplant with HSV appearing in the first week, Epstein–Barr virus (EBV) between 2 and 3 weeks, cytomegalovirus (CMV) at 7 weeks, and varicella zoster virus (VZV) at 5 months. Most patients are placed on prophylactic acyclovir following HSCT to prevent herpes infections. Rare cases of herpes simplex encephalitis have been reported. VZV encephalitis usually follows a cutaneous herpes zoster infection. Whereas a 4% incidence of VZV encephalitis in HSCT patients with active zoster or varicella infection was still seen in the 1970s (125), this complication has become uncommon due to acyclovir prophylaxis. Various patterns of dermatomal involvement (cranial, cervical, and lumbosacral) may result in concomitant meningoradiculitis (55). CMV encephalitis is rarely seen. EBV may lead to a post-transplant lymphoproliferative disorder in the CNS, presenting as focal lesions or lymphoblastic meningeosis. One herpes virus with particular predilection for the CNS is human herpesvirus-6 (HHV-6). More than 90% of the general population is seropositive for HHV-6 (126). The virus is generally acquired during childhood, presenting as exanthema subitum and febrile illness. After the initial infection, this lymphotropic and neurotropic virus persists indefinitely and may flare up in a spectrum of diseases in the immunocompromised patient. HHV-6

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

345

infections in the HSCT population generally represent reactivation and have been documented in 38–60% of patients following HSCT (127). In the CNS, HHV-6 reactivation causes encephalitis. Variant B is most commonly involved, but variant A may be more virulent. Most cases of HHV-6 encephalitis occur within 12 weeks of transplantation (128). The most common clinical manifestations are mental changes, seizures, memory disturbance, headaches, and speech disturbance. Focal neurologic symptoms and fever are less common. Brain MRI demonstrates T2 hyperintensity involving one or both hippocampi with variable involvement of adjacent medial temporal lobe structures of the limbic system including amygdalae and parahippocampal gyri (see Fig. 6) (129,130). These findings are similar to other infectious etiologies of limbic encephalitis including HSV, VZV, and neurosyphilis. HHV-6 encephalitis is diagnosed by CSF PCR. It should be noted that HHV-6 may be seen in normal brain tissues. On the other hand, HHV-6 is rarely found in CSF from asymptomatic patients (131). CSF pleocytosis and elevated CSF protein may be seen. HHV-6 is potentially a treatable virus. Overall mortality in transplants patients with HHV-6 encephalitis is 58% (128). Antiviral prophylaxis with acyclovir, ganciclovir, or valacyclovir frequently does not prevent HHV-6 encephalitis, but cures have been documented with ganciclovir or foscarnet. One case report describes successful treatment of HHV-6 encephalitis with donor lymphocyte infusion (132). Progressive multifocal leukoencephalopathy (PML) is another rare complication following HSCT. PML is a demyelinating disorder caused by infection of glial tissue but predominantly oligodendroglial cells. JC virus, a neurotropic polyomavirus of the Papovaviridae family, has been identified as the infectious agent. JC virus is endemic in humans, with seroconversion up to 60% in young adults and up to 80% in the elderly (133). Primary infection with JC virus occurs during childhood and is often asymptomatic. Clinical symptomatology occurs following reactivation and is nonspecific (personality changes, confusion, or dementia, as well as more localizing signs such as aphasia, apraxia, hemiparesis, cerebellar dysfunction, or visual processing abnormalities). MR scan is the imaging modality of choice, usually demonstrating multifocal hyperintensities on T2-weighted and FLAIR images. Changes are mostly located in the white matter but may extend into the gray matter. Diagnosis is based on demonstration of JC virus in CSF by means of PCR assay, in situ hybridization, or immunohistochemistry. However, PCR is negative in approximately 25% of PML cases (134). Treatment is

Fig. 6. HHV-6 encephalitis. Axial MRI scan (FLAIR, A; Diffusion, B) of a 61-year-old male 22 days after myeloablative doublecord transplant for myeloproliferative disorder demonstrating T2 hyperintense lesions in the bilateral temporal lobes that extended to the orbital frontal cortex and subtle restricted diffusion. He developed mental status changes (disorientation, agitation, paranoia, and visual hallucinations) on post-transplant day 16. Lumbar puncture yielded four white blood cells, predominantly lymphocytes, one red blood cell, and elevated protein. HHV-6 PCR was positive from the CSF. An EEG showed right-sided periodic lateralizing epileptiform discharges over a poorly developed background. His mental status and renal function declined rapidly over the next several days. Despite broad spectrum antimicrobials including foscarnet, he became hemodynamically unstable and expired.

346

Part VI / Complications of Cancer Therapy

empiric, and successes have been reported with intrathecal cytarabine, subcutaneous interferon, subcutaneous interleukin-2 or combinations of these. However, prognosis remains very poor with a median survival of less than 6 months. Long-term survivors (more than one year) have been noted in HSCT patients (134,135). In these cases, a relatively prominent inflammatory response in the biopsied brain tissue was found. Spontaneous remission of PML lesions may occur.

5. MYELOPATHY The spinal cord is rarely the predominantly involved portion of the CNS in HSCT patients. Only a few cases have been reported (136–139). In two patients, radiation therapy appeared to be the cause although standard doses and fractionation were given (137,138). In three other patients, the presence of GVHD and the response to corticosteroids or plasmapheresis suggested an immune-mediated mechanism. One patient developed transient optic neuritis. MRI in these patients has demonstrated abnormalities, frequently contrast enhancing, at various levels of the spinal cord.

6. NEUROMUSCULAR DISORDERS Neuromuscular complications occur less frequently than CNS complications in the HSCT population. Peripheral nerves can be damaged by catheter placement, immobilization, cachexia, medications, toxins, or herpes zoster infection. Additionally, HSCT patients may develop sepsis-associated critical illness polyneuropathy (140,141). Toxic neuropathies are described in Chapter 17. Several disorders usually associated with immune dysfunction have been described in HSCT patients including chronic inflammatory demyelinating polyneuropathy (CIDP), acute inflammatory demyelinating polyneuropathy (AIDP), myasthenia gravis (MG), and inflammatory myopathy. These disorders typically occur in the setting of chronic GVHD, but the nature of the relationship remains unclear. Some hypothesize that these neuromuscular disorders are direct manifestations of GVHD, a result of donor T-cell response against neuromuscular antigens. Another hypothesis is that these disorders are simply the outcome of a generally dysregulated immune system following HSCT. A third hypothesis is that they occur as a result of drug toxicity. AIDP or Guillain–Barré syndrome (GBS) typically occurs following viral infections or immunizations as well as in the setting of dysimmune states such as malignancies and immunotherapy. The prevalence following allogeneic transplant is estimated at 1% (105). HSCT patients may develop AIDP as early as immediately posttransplant or even as late as one year after transplant. Clinical manifestations include rapidly progressive limb weakness, hyporeflexia/areflexia, sensory deficits, and occasionally autonomic dysfunction. Muscle cramps and prickling sensations may be present. Progression does not seem to involve the respiratory muscles. Pure autonomic neuropathy is reported as well following allogenic BMT (142). A change in immunosuppression, usually a reduction in the dose, may precede the ictus. Diagnosis is made on the basis of clinical presentation, CSF studies demonstrating albuminocytological dissociation and/or electrophysiological studies. In HSCT patients, symptoms resolve after resolution of GVHD or after treatment with glucocorticosteroids and azathioprine, both abandoned therapies for typical GBS and CIDP (143,144). Other case reports indicate plasma exchange as an effective therapy (145,146). In HSCT patients, polymyositis (PM) is associated with chronic GHVD (147–151). Only rarely has it been reported in association with acute GVHD (152). Clinical presentation is similar to idiopathic PM with subacute onset of proximal muscle weakness, elevated muscle enzymes, typical pathology on muscle biopsy, and typical electromyographic patterns. One unusual case presented as profound weakness involving the neck extensors as well as respiratory failure requiring mechanical ventilation (149). Onset has been reported from 2 to 69 months after HSCT (148). The majority of cases respond to immunosuppressive treatment such as steroids and cyclosporine (150). Clinical remission may occur with treatment within 2–7 months (151). The duration of clinical remission varies, but recurrences may occur on steroid taper (150). Myasthenia gravis in the HSCT population is also considered a rare manifestation of chronic GVHD (153). MG is a neuromuscular disorder in which antibodies are directed against the acetylcholine receptor, resulting in fatiguable weakness. Reported cases of MG developed 22–60 months following transplant (148). Diagnosis is based on clinical picture, typical electromyographic pattern, positive response to cholinesterase inhibitors, and/or the presence of acetylcholine receptor antibodies. Myasthenia gravis has not been associated with thymoma

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

347

in HSCT patients, and acetylcholine receptor antibodies may be negative despite characteristic increased jitter on EMG (154). Traditional MG treatment including corticosteroids, azathioprine, cyclosporine, plasmapheresis, intravenous immunoglobulin, and pyridostigmine are effective in the HSCT population (155,156). Rituximab was successful in one refractory patient with myasthenia gravis (157).

7. NEUROLOGIC MANIFESTATIONS ASSOCIATED WITH GVHD GVHD results from donor T-cells attacking mismatched recipient antigens and usually occurs following allogeneic HSCT (rarely in syngeneic or autologous HSCT). The most commonly involved tissues are skin, liver, mouth, and eyes. However, GVHD is also a significant source of neurological complications, both indirectly through medications used in the treatment or prevention of GVHD and directly through immunologic effects on the nervous system. Treatment of GVHD must suppress the dysregulated immune cells and cytokines, which result in damage to various tissues and organs. Patients undergoing HSCT usually receive immunoprophylaxis with agents such as tacrolimus or cyclosporine. Following successful transplantation, these medications can sometimes be weaned off. However, patients who develop chronic GVHD will require longer-term treatment with immunosuppressive medications. Common treatments include corticosteroids, mycophenolate mofetil, sirolimus, rituximab, extracorporeal phototherapy, thalidomide, pulse cyclophosphamide, and hydroxychloroquine. These treatments weaken the immune system, making patients more susceptible to infection. Additionally, each medication has its own set of neurotoxic side effects or may aggravate neutropenia. Independent of treatment, GVHD is associated with significant immunodeficiency. Patients with chronic GVHD may have hypogammaglobulinemia, reduction of IgG subclasses or deficiencies of secretory IgA (158) and may be functionally asplenic (63). Such patients are particularly susceptible to bacterial infections. Some neurological complications may be directly related to GVHD. As discussed above, GVHD can result in polymyositis, myasthenia gravis, and CIDP. Whereas GVHD seems an established cause of peripheral nervous system disorders, it is unclear if GVHD affects the CNS. In an infant with severe combined immune deficiency (SCID) who received a maternal haploidentical transplant at 3 weeks of age, acute and chronic GVHD developed. This patient died after multiple complications. At autopsy, the brain stem and right hippocampus showed focal lymphohistiocytic aggregates similar to those found in the cardiac conduction system and diaphragm. The authors raise the alternative explanation of a very early stage of a lymphoproliferative process (e.g., Epstein–Barr virus driven B-cell lymphoma), rather than chronic GVHD (159). Similarly, a case report of seizures in a patient with chronic GVHD following allogeneic HSCT for chronic myelogenous leukemia documents hyperintense lesions on T2-weighted MR images involving the posterior white matter around the lateral ventricles; no tissue diagnosis is provided (160). More recently, two patients who presented with seizures, encephalopathy, and focal deficits and had abnormal T2 lesions on MRI, were found after pathology, to have a perivascular lymphocytic infiltrate composed predominantly of T-lymphocytes from the donor, suggesting the possibility of CNS GVHD (161). Extremely rare are cases of immune-mediated myelopathy after allogeneic HSCT (136,139). In these instances, a relationship with GVHD has been postulated. More recent literature suggests an association between CNS angiitis and GVHD (see section 3). Nonetheless, the relationship between GVHD and the brain requires further study.

8. CONCLUSION Complications following hematopoietic stem cell transplantation have long been recognized. The causes are numerous, including chemoradiotoxicity, medication toxicity, metabolic abnormalities, organ failure, graft versus host disease, infection, pancytopenia, and platelet dysfunction. As the number of transplants performed annually increases, potential neurologic complications are being seen with increasing frequency.

REFERENCES 1. Goldman JM, Horowitz MM. The international bone marrow transplant registry. Int J Hematol 2002;76:393–397. 2. Horowitz MM, Gale RP, Sondel PM et al. Graft-versus-leukemia reactions after bone marrow transplantation. Blood 1990;75:555–562.

348

Part VI / Complications of Cancer Therapy

3. Peggs KS, Thomson K, Hart DP et al. Dose-escalated donor lymphocyte infusions following reduced intensity transplantation: toxicity, chimerism, and disease responses. Blood 2004;103:1548–1556. 4. Applebaum FR. Hematopoietic cell transplantation. In: Kasper DL et al. (eds.). Harrison’s Principles of Internal Medicine, 16th ed., vol. 1. New York: McGraw-Hill, 2005:668–673. 5. Bhushan V, Collins RH. Chronic graft-versus-host disease. JAMA 2003;290:2599–2603. 6. Culter C, Giri S, Jeyapalan S et al. Acute and chronic graft-versus-host disease after allogeneic peripheral-blood stem cell and bone marrow transplantation: a meta-analysis. J Clin Oncol. 2001;19:3685–3691. 7. Antonini G, Ceschin V, Morino S et al. Early neurologic complications following allogeneic bone marrow transplant for leukemia. Neurology 1998;50:1441–1445. 8. Bleggi-Torres LF, de Medeiros BC, Werner B et al. Neuropathological findings after bone marrow transplantation: an autopsy study of 180 cases. Bone Marrow Transplant 2000;25:301–307. 9. Gordon B, Lyden E, Lynch J et al. Central nervous system dysfunction as the first manifestation of multiple organ dysfunction syndrome in stem cell transplant patients. Bone Marrow Transplant 2000;25:79–83. 10. Rosenfeld MS, Pruitt A. Neurologic complications of bone marrow, stem cell, and organ transplantation in patients with cancer. Semin Oncol 2006;33:352–361. 11. Mohrmann RL, Mah V, Vinters HV. Neuropathologic findings after bone marrow transplantation. Hum Pathol 1990;21:630–639. 12. Jackson SR, Tweeddale MG, Barnett MJ et al. Admission of bone marrow transplant recipients to the intensive care unit: outcome, survival and prognostic factors. Bone Marrow Transplant 1998;21:697–704. 13. Woodard P, Helton K, McDaniel H et al. Encephalopathy in pediatric patients after allogeneic hematopoietic stem cell transplantation is associated with a poor prognosis. BMT 2004;33:1151–1157. 14. Baker WJ, Royer GL, Weiss RB. Cytarabine and neurologic toxicity. J Clin Oncol 1991;9:679–693 15. Sullivan KM, Storb R, Shulman HM et al. Immediate and delayed neurotoxicity after mechlorethamine preparation for bone marrow transplantation. Ann Intern Med 1982;97:182–189. 16. Schuh A, Dandridge J, Haydon P et al. Encephalopathy complicating high-dose melphalan. Bone Marrow Transplant 1999;24: 1141–1143. 17. Tahsildar HI, REmler BF, Creger RJ et al. Delayed, transient encephalopathy after marrow transplantation: case reports and MRI fidings in four patients. J Neurooncol 1996;247:241–250. 18. Meanwell CA, Blake AE, Kelly KA et al. Prediction of ifosfamide/mesna associated encephalopathy. Eur J Cancer Clin Oncol 1986;22:815–819. 19. Anderson NR, Tandon DS. Ifosfamide extrapyramidal neurotoxicity. Cancer 1991;68:72–75. 20. Curtin JP, Koonings PP, Gutierrez M et al. Ifosfamide-induced neurotoxicity. Gynecol Oncol 1991;42:193–196. 21. Cheson BD, Vena DA, Foss FM et al. Neurotoxicity of purine analogues: a review. J Clin Oncol 1994;112:2216–2228. 22. Atkinson K, Clink H, LaMer S et al. Encephalopathy following bone marrow transplantation. Eur J Cancer 1977;13:623–625. 23. Snavely SR, Hodges GR. The neurotoxicity of antibacterial agents. Ann Intern Med 1984;101:92–104. 24. Eng RHK, Munsif AN, Yangco BG et al. Seizure propensity with imipenem. Arch Intern Med 1989;149:1881–1883. 25. Frytak S, Moertel CG, Childs DS. Neurologic toxicity associated with high-dose metronidazole therapy. Ann Intern Med 1978;88: 361–362. 26. Wade JC, Meyers JD. Neurologic symptoms associated with parenteral acyclovir treatment after marrow transplantation. Ann Intern Med 1983;98:921–925 27. Swan SK, Bennett WM. Oral acyclovir and neurotoxicity. Ann Intern Med 1989;111:188. 28. Cohen SMZ, Minkove JA, Zebley JW III et al. Severe but reversible neurotoxicity from acyclovir. Ann Intern Med 1984; 100:920. 29. McDonald GB, Hinds MS, Fisher LD et al. Veno-occlusive disease of the liver and multiorgan failure after bone marrow transplantation: a cohort study of 355 patients. Ann Intern Med 1993;118:255–267. 30. Rabinowe SN, Soiffer RJ, Tarell NJ et al. Hemolytic-uremic syndrome following bone marrow transplantation in adults for hematologic malignancies. Blood 1991;77:1837–1844. 31. Guinan EC, Tarbell NJ, Niemeyer CM et al. Intravascular hemolysis and renal insufficiency after bone marrow transplantation. Blood 1988; 72:451–455. 32. Haire WD. Multiple organ dysfunction syndrome in hematopoietic stem cell transplantation. Crit Care Med 2002;30[Suppl]:S257–S262. 33. Shah AK. Cyclosporine A neurotoxicity among bone marrow transplant recipients. Clin Neuropharmacol 1999;22:67–73. 34. Wong R, Beguelin GZ, de Lima M et al. Tacrolimus-associated posterior reversible encephalopathy syndrome after allogeneic haematopoietic stem cell transplantation. Br J Haematol 2003;122:128–134. 35. Steg RE, Kessinger A, Wszolek ZK. Cortical blindness and seizures in a patient receiving FK506 after bone marrow transplantation. Bone Marrow Transplant 1999;23:959–962. 36. Memon M, deMagalhaes-Silverman M, Bloom EJ et al. Reversible cyclosporine-induced cortical blindness in allogeneic bone marrow transplant recipients. Bone Marrow Transplant 1995;15:283–286. 37. Hinchey J, Chaves C, Appignani B et al. A reversible posterior leukoencephalopathy syndrome. N Engl J Med 1996;334:494–500. 38. Stott VL, Hurrell MA, Anderson TJ. Reversible posterior leukoencephalopathy syndrome: a misnomer reviewed. Internal Med J 2005;35:83–90. 39. Hurwitz RL, Mahoney DH, Armstrong DL et al. Reversible encephalopathy and seizure as a result of conventional vincristine administration. Med Pediatr Oncol 1988;16:216–219. 40. Higman MA, Port JD, Beauchamp NJ et al. Reversible leukoencephalpathy associated with re-infusion of DMSO preserved stem cells. BMT 2000:26;797–800. 41. Majolino I, Caponetto A, Scime R et al. Wernicke-like encephalopathy after autologous bone marrow transplantation. Haematologica 1990;75:282–284.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

349

42. Bleggi-Torres LF, de Medeiros BC, Ogasawara VSA et al. Iatrogenic Wernicke’s encephalopathy in allogeneic bone marrow transplantation: a study of eight cases. Bone Marrow Transplant 1997;20:391–395. 43. Baek JH, Sohn SK, Dim DH et al. Wernicke’s encephalopathy after allogeneic stem cell transplantation. Bone Marrow Transplant 2005;35:829–830. 44. Boqué C, Petit J, Aguilera C et al. Central and extrapontine myelinolysis following allogeneic peripheral haematopoietic progenitor cell transplantation. Bone Marrow Transplant 2003;31:61–64. 45. Fraser C, Charnas L, Orchard P. Central pontine myelinolysis following bone marrow transplantation complicated by severe hepatic veno-occlusive disease. Bone Marrow Transplant 2005;36:733–734. 46. Mitchell RB, Wagner JE, Karp JE et al. Syndrome of idiopathic hyperammonemia after high-dose chemotherapy: review of nine cases. Amer J Med 1988;85:662–667. 47. Sharp RA, Lang CC. Hyperammonaemic encephalopathy in chronic myelomonocytic leukemia. Lancet 1987;1:805. 48. Tse N, Cederbaum S, Glaspy JA. Hyperammonemia following allogeneic bone marrow transplantation. Amer J Hematology 1991;38:140–141. 49. Leonard JV, Kay JDS. Acute encephalopathy and hyperammonaemia complicating treatment of acute lymphoblastic leukemia with asparaginase. Lancet 1986; i:162–163 (letter). 50. Davies SM, Szabo E, Wagner JE et al. Idiopathic hyperammonemia: a frequently lethal complication of bone marrow transplantation. Bone Marrow Transplant 1996;17:1119–1125. 51. Berry GT, Bridges ND, Nathanson KL et al. Successful use of alternate waste nitrogen agents and hemodialysis in a patient with hyperammonemic coma after heart–lung transplantation. Arch Neurol 1999;56:481–484. 52. Snider S, Bashir R, Bierman P. Neurologic complications after high-dose chemotherapy and autologous bone marrow transplantation for Hodgkin’s disease. Neurology 1994;44:681–684. 53. Walters MC, Patience M, Leisenring W et al. Bone marrow transplantation for sickle cell disease. N Engl J Med 1996;335:369–376. 54. De la Camara R, Tomas JF, Figuera A et al. High-dose busulfan and seizures. Bone Marrow Transplant 1991;7:363–364. 55. Openshaw H, Slatkin NE. Differential diagnosis of neurological complications in bone marrow transplantation. The Neurologist 1995;1:191–206. 56. Ebner F, Ranner G, Slavc I et al. MR findings in methotrexate-induced CNS abnormalities. AJNR 1989;10:959–964. 57. Zakrzewski JL. Cyclosporin A–associated status epilepticus related to hematopoeitic stem cell transplantation for thalassemia. Pediatr Hematol Oncol 2003;20:481–486. 58. Wszolek ZK, Steg RE, Armitage JO. Complex partial status epilepticus after bone marrow transplantation for non-Hodgkin’s lymphoma. Bone Marrow Transplant 1997;19:637–638. 59. Penn I. De novo malignant lesions of the central nervous system. In: Wijdicks EFM (ed.). Blue Books of Practical Neurology: Neurologic Complications in Organ Transplant Recipients, Volume 21. Boston: Butterworth-Heinemann, 1999: 217–227. 60. Kolb HJ, Bender-Gotze CH. Late complications after allogeneic bone marrow transplantation for leukemia. Bone Marrow Transplant 1990;6:61–72. 61. Witherspoon RP, Fisher LD, Schoch G et al. Secondary cancers after bone marrow transplantation for leukemia or aplastic anemia. N Engl J Med 1989;321:784–789. 62. Socie G, Curtis RE, Deeg HJ et al. New malignant diseases after allogeneic marrow transplantation for childhood acute leukemia. J Clin Oncol 2000;18:348–357. 63. Kulkarni S, Powles R, Treleaven J et al. Chronic graft-versus-host disease is associated with long-term risk for pneumococcal infections in recipients of bone marrow transplants. Blood 2000;95:3683–3686. 64. Coplin WM, Cochran MS, Levine SR et al. Stroke after bone marrow transplantation: frequency, aetiology, and outcome. Brain 2001;124:1043–1051. 65. Bleggi-Torres, LF, Werner B, Gasparetto EL. Intracranial hemorrhage following bone marrow transplantation: an autopsy study of 58 patients. Bone Marrow Transplant 2002;29:29–32. 66. Colosimo M, McCarthy N, Jayasinghe R et al. Diagnosis and management of subdural haematoma complicating bone marrow transplantation. Bone Marrow Transplant 2000;25:549–552. 67. Kannan K, Koh LP, Linn YC. Subdural hematoma in two hematopoietic stem cell transplant patients with post-dural puncture headache and initially normal CT brain scan. Ann Hematol 2002;81:540–542. 68. Harvey CJ, Peniket AJ, Miszkiel K et al. MR angiographic diagnosis of cerebral venous signs thrombosis following allogeneic bone marrow transplantation. Bone Marrow Transplant 2000;25:791–795. 69. Bertz H, Laubenberger J, Steinfurth G et al. Sinus venous thrombosis: an unusual cause for neurologic symptoms after bone marrow transplantation under immunosuppression. Transplantation 1998;66:241–244. 70. Mori A, Tanaka J, Kobayashi S et al. Fatal cerebral hemorrhage associated with cyclosporin-A/FK506-related encephalopathy after allogeneic bone marrow transplantation. Ann Hematol 2000;79:588–592. 71. Gordon B, Haire W, Kessinger A et al. High frequency of antithrombin-3 and protein C deficiency following autologous bone marrow transplantation for lymphoma. Bone Marrow Transplant 1991;8:497–502. 72. Gordon BG, Saving Kl, McCallister JA et al. Cerebral infarction associated with protein C deficiency following allogeneic bone marrow transplantation. Bone Marrow Transplant 1991;8:323–325. 73. Kaufman PA, Jones RB, Greenberg CS et al. Autologous bone marrow transplantation and factor XII, factor VII, and protein C deficiencies. Cancer 1990;66:515–521. 74. Qasim W, Gerritsen B, Veys P. Anticardiolipin antibodies and thromboembolism after BMT. Bone Marrow Transplant 1998;21: 845–847. 75. Richard S, Seigneur M, Blann A et al. Vascular endothelial lesion in patients undergoing bone marrow transplantation. Bone Marrow Transplant 1996;18:955–959.

350

Part VI / Complications of Cancer Therapy

76. Kalhs P, Brugger S, Schwarzinger I et al. Microangiopathy following allogeneic marrow transplantation. Transplantation 1995;60: 949–957. 77. Windrum P, Morris TCM. Severe neurotoxicity because of dimethyl sulphoxide following peripheral blood stem cell transplantation. Bone Marrow Transplant 2003;31:315. 78. Darabi K, Brown JR, Kao GS. Paradoxical embolism after peripheral blood stem cell infusion. Bone Marrow Transplant 2005;36: 561–562. 79. Jerman MR, Fick RB, Jr. Nonbacterial thrombotic endocarditis associated with bone marrow transplanation. Chest 1986;90:919–922. 80. Patchell RA, White CL, 3rd, Clark AW et al. Nonbacterial thrombotic endocarditis in bone marrow transplant patients. Cancer 1985;55:631–635. 81. Tschuchnigg M, Bradstock KF, Koutts J et al. A case of thrombotic thrombocytopenic purpura following allogeneic bone marrow transplantation. Bone Marrow Transplant 1990;5:61–63. 82. Ho VT, Cutler C, Carter S et al. Blood and marrow transplant clinical trials network toxicity committee consensus summary: thrombotic microangiopathy after hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11:571–575. 83. Padovan CS, Bise K, Hahn J et al. Angiitis of the central nervous system after allogeneic bone marrow transplantation? Stroke 1999;30:1651–1656. 84. Ma M, Barnes G, Pulliam J et al. CNS angiitis in graft vs. host disease. Neurology 2002;59:1994–1997. 85. Campbell JN, Morris PP. Cerebral vasculitis in graft-versus-host disease. Am J Neuroradiol 2005;26:654–656. 86. De Medeiros BC, de Medeiros CR, Werner B et al. Central nervous system infections following bone marrow transplantation: an autopsy report of 27 cases. J Hematother Stem Cell Res 2000;9:535–540. 87. Van Burik JA, Hackman RC, Nadeem SQ et al. Nocardiosis after bone marrow transplantation: a retrospective study. Clin Infect Dis 1997;24:1154–1160. 88. Chouciño C, Goodman SA, Greer JP et al. Nocardial infections in bone marrow transplant recipients. Clin Infect Dis 1996;23: 1012–1019. 89. Machado CM, Macedo MC, Castelli JB et al. Clinical features and successful recovery from disseminated nocardiosis after BMT. Bone Marrow Transplant. 1997;17:81–82. 90. Elliott MA, Tefferi A, Marshall WF et al. Disseminated nocardiosis after allogeneic bone marrow transplantation. Bone Marrow Transplant. 1997;20:425–426. 91. Carradice D, Szer J. Cerebral nocardiosis after allogeneic bone marrow transplantation. Intern Med J. 2004;34:698–699. 92. Laurence AD, Peggs KS. Cerebral and pulmonary nocardia in a bone marrow transplant patient. Br J Haematol. 2005;129:711. 93. Campos A, Vaz CP, Campilho F et al. Central nervous system tuberculosis following allogeneic stem cell transplantation. Bone Marrow Transplant. 2000;25:567–569. 94. Graus F, Saiz A, Sierra J et al. Neurologic complications of autologous and allogeneic bone marrow transplantation inpatients with leukemia. Neurology 1996;46:1004–1009. 95. Martino R, Martinez C, Brunet S et al. Tuberculosis in bone marrow transplant recipients: report of two cases and review of the literature. Bone Marrow Transplant. 1996;18:809–812. 96. Aljurf M, Gyger M, Alrajhi A et al. Mycobacterium tuberculosis infection in allogeneic bone marrow transplantation patients. Bone Marrow Transplant. 1999;24:551–554. 97. Rodriguez M, Chou S, Fisher DC et al. Lyme meningoradiculitis and myositis after allogeneic hematopoietic stem cell transplantation. Clin Infect Dis. 2005;41:112–114. 98. Walsh TJ, Hier DB, Caplan LR. Aspergillosis of the central nervous system: clinicopathological analysis of 17 patients. Ann Neurol 1985;18:574–582. 99. Miaux Y, Ribaud P, Williams M et al. MR of cerebral aspergillosis in patients who have had bone marrow transplantation. Am J Neuroradiol 1995;16:555–562. 100. Guermazi A, Gluckman E, Tabti B et al. Invasive central nervous system aspergillosis in bone marrow transplantation recipients: an overview. Eur Radiol 2003;13:377–388. 101. Khoury H, Adkins D, Miller G et al. Resolution of invasive central nervous system aspergillosis in a transplant recipient. Bone Marrow Transplant 1997;20:179–180. 102. Baslar Z, Soysal T, Hanci M et al. Successfully treated invasive central nervous system aspergillosis in an allogeneic stem cell transplant recipient. Bone Marrow Transplant 1998;22:404–405. 103. LaRocco MT, Burgert SJ. Infection in the bone marrow transplant recipient and role of the microbiology laboratory in clinical transplantation. Clin Microbiol Rev 1997;10:277–297. 104. Hagensee ME, Bauwens JE, Kjos B et al. Brain abscess following marrow transplantation: experience at the Fred Hutchinson Cancer Research Center, 1984–1992. Clin Infect Dis 1994;19:402–408. 105. Maschke M, Dietrich U, Prumbaum M et al. Opportunistic CNS infection after bone marrow transplantation. Bone Marrow Transplant 1999;23:1167–1176. 106. Centers for Disease Control and Prevention. Guidelines for preventing opportunistic infections among hematopoietic stem cell transplant recipients: recommendations of CDC, the Infectious Disease Society of America, and the American Society of Blood and Marrow Transplantation. MMWR 2000;49:23,26,107. 107. Tolkoff-Rubin NE, Hoving GK, Rubin RH. Central nervous system infections. Chapter 10 in: Wijdicks EFM (ed.). Neurologic Complications in Organ Transplant Recipients. Butterworth-Heinemann, 1999. 108. Nenoff P, Kellermann S, Schober R et al. Rhinocerebral zygomycosis following bone marrow transplantation in chronic myelogenous leukaemia: report of a case and review of the literature. Mycoses 1998;41:365–372. 109. Baddley JW, Moser SA, Sutton DA et al. Microascus cinereus (Anamorph scopulariopsis) brain abscess in a bone marrow transplant recipient. J Clin Microbiol 2000;38:395–397.

Chapter 18 / Neurological Complications of Hematopoietic Stem Cell Transplantation

351

110. Chandrasekar PH, Momin F. Disseminated toxoplasmosis in marrow recipients: a report of three cases and a review of the literature. Bone Marrow Transplant 1997;19:685–689. 111. Brinkman K, Debast S, Sauerwein R et al. Toxoplasma retinitis/encephalitis 9 months after allogeneic bone marrow transplantation. Bone Marrow Transplant 1998;21:635–636. 112. Derouin F, Devergie A, Auber P et al. Toxoplasmosis in bone marrow-transplant recipients: report of seven cases and review. Clin Infect Dis 1992;15:267–270. 113. Bretagne S, Costa JM, Kuentz M et al. Late toxoplasmosis evidenced by PCR in a marrow transplant recipient. Bone Marrow Tranplant 1995;15:809–811. 114. Tefferi A, O’Neill BP, Inwards DJ. Late-onset cerebral toxoplasmosis after allogeneic bone marrow transplantation. Bone Marrow Transplant 1998;21:1285–1286. 115. Dietrich U, Maschke M, Dörfler A et al. MRI of intracranial toxoplasmosis after bone marrow transplantation. Neuroradiology 2000;42:14–18. 116. Khoury H, Adkins D, Brown R et al. Successful treatment of cerebral toxoplasmosis in a marrow transplant recipient: contribution of a PCR test in diagnosis and early detection. Bone Marrow Transplant 1999;23:409–411. 117. Löwenberg B, van Gijn J, Prins E et al. Fatal cerebral toxoplasmosis in a bone marrow transplant recipient with leukemia. Transplantation 1983;35:30–34. 118. Slavin MA, Meyers JD, Remington JS et al. Toxoplasma gondii infection in marrow transplant recipients: a 20-year experience. Bone Marrow Transplant 1994;13:549–557. 119. Roemer E, Blau IW, Basra N et al. Toxoplasmosis, a severe complication in allogeneic hematopoietic stem cell transplantation: successful treatment strategies during a 5-year single-center experience. Clin Infect Dis 2001;32:1–8. 120. Seong DC, Przepiorka D, Bruner JM et al. Leptomeningeal toxoplasmosis after allogeneic marrow transplantation. Am J Clin Oncol 1993;16:105–108. 121. Bleggi-Torres LF, de Medeiros BC, Werner B et al. Unusual presentation of cerebral toxoplasmosis after BMT. Bone Marrow Transplant 1999;23:855–856. 122. Martino R, Maertens J, Bretagne S et al. Toxoplasmosis after hematopoietic stem cell transplantation. Clin Infect Dis 2000;31: 1158–1194. 123. Anderlini P, Przepiorka D, Luna M et al. Acanthamoeba meningoencephalitis after bone marrow transplantation. Bone Marrow Transplant 1994;14:459–461. 124. Feingold JM, Abraham J, Bilgrami S et al. Acanthamoeba meningoencephalitis following autologous peripheral stem cell transplantation. Bone Marrow Transplant 1998;22:297–300. 125. Atkinson K, Meyers JD, Storb R et al. Varicella zoster virus infection after marrow transplantation for aplastic anemia or leukemia. Transplantation. 1980;29(1):47–50. 126. Kimberlin DW, Whitley RJ. Human herpesvirus-6: neurologic implications of a newly described viral pathogen. J Neurovirol 1998;4:474–485. 127. Bosi A, Zazzi M, Amantini A et al. Fatal herpesvirus-6 encephalitis after unrelated bone marrow transplant. Bone Marrow Transplant 1998;22:285–288. 128. Singh N. Paterson D. Encephalitis caused by human herpesvirus-6 in transplant recipients: relevance of a novel neurotropic virus. Transplantation 2000;69:2474–2479. 129. Wainwright MS, Martin PL, Morse RP et al. Human herpesvirus-6 limbic encephalitis after stem cell transplantation. Ann Neurol 2001;50:612–619. 130. Gorniak RJT, Young GS, Wiese DE et al. MR imaging of human herpesvirus-6-associated encephalitis in 4 patients with anterograde amnesia after allogeneic hematopoietic stem cell transplantation. Am J Neuroradiol 2006;27:87–891. 131. Wang FZ, Linde A, Hägglund H et al. Human herpesvirus-6 DNA in cerebrospinal fluid specimens from allogeneic bone marrow transplant patients: does it have clinical significance? Clin Infect Dis 1999;28:562–568. 132. Yoshihara S, Kato R, Inoue T et al. Successful treatment of life-threatening human herpesvirus-6 encephalitis with donor lymphocyte infusion in a patient who had undergone human leukocyte antigen–haploidentical nonmyeloablative stem cell transplantation. Transplantation 2004;77:835–838. 133. Demeter LM. JC, BK, and other polyomavirus; progressive multifocal leukoencephalopathy. In: Mandell GL, Bennett JE, Dolin R (eds.). Principles and Practice of Infectious Diseases, 5th ed. Churchill Livingstone: London, 2000:1645–1651. 134. Re D, Bamborschke S, Feiden W et al. Progressive multifocal leukoencephalopathy after autologous bone marrow transplantation and alpha-interferon immunotherapy. Bone Marrow Transplant 1999;23:295–298. 135. Przepiorka D, Jaeckle KA, Birdwell RR et al. Successful treatment of progressive multifocal leukoencephalopathy with low-dose interleukin-2. Bone Marrow Transplant 1997;20:983–987. 136. Openshaw H, Slatkin NE, Parker PM et al. Immune-mediated myelopathy after allogeneic marrow transplantation. Bone Marrow Transplant 1995;15:633–636. 137. Chao MWT, Wirth A, Ryan G et al. Radiation myelopathy following transplantation and radiotherapy for non-Hodgkin’s lymphoma. Int J Radiation Oncol Biol Phys 1998;5:1057–1061. 138. Schwartz DL, Schechter GP, Seltzer S et al. Radiation myelitis following allogeneic stem cell transplantation and consolidation radiotherapy for non-Hodgkin’s lymphoma. Bone Marrow Transplant 2000;26:1355–1359. 139. Richard S, Fruchtman S, Scigliano E et al. An immunological syndrome featuring transverse myelitis, Evans syndrome, and pulmonary infiltrates after unrelated bone marrow transplant in a patient with severe aplastic anemia. Bone Marrow Transplant 2000;26: 1225–1228. 140. Bolton CF, Young GB. Critical illness polyneuropathy: current treatment options in neurology 2000;2:489–498. 141. Zifko UA. Long-term outcome of critical illness polyneuropathy. Muscle Nerve 2000;999:S49–S52.

352

Part VI / Complications of Cancer Therapy

142. Roskrow MA, Kelsey SM, McCarthy M et al. Selective autonomic neuropathy as a novel complication of BMT. Bone Marrow Transplant 1992;10:469–470. 143. Greenspan A, Deeg HG, Cottler-Fox M et al. Incapacitating peripheral neuropathy as a manifestation of chronic graft-versus–host disease. Bone Marrow Transplant 1990;5:349–352. 144. Amato AA, Barohn RJ, Sahenk Z et al. Polyneuropathy complicating bone marrow and solid organ transplantation. Neurology 1993;43:1513–1518. 145. Griggs JJ, Commichau CS, Rapoport AP et al. Chronic inflammatory demyelinating polyneuropathy in non-Hodgkin’s lymphoma. Am J Hematol 1997;54:332–334. 146. Wen PY, Aleya EP, Simon D et al. Guillain–Barré syndrome following allogeneic bone marrow transplantation. Neurology 1997;49:1711–1714. 147. Parker P, Chao NJ, Ben–Ezra J et al. Polymyositis is a manifestation of chronic graft-versus-host disease. Medicine 1996;75:279–285. 148. Tse S, Saunders EF, Silverman E et al. Myasthenia gravis and polymyositis as manifestations of chronic graft-versus-host disease. Bone Marrow Transplant 1999;23:397–399. 149. Leano AM, Miller K, White AC. Chronic graft-versus-host disease related polymyositis as a cause of respiratory failure following allogeneic bone marrow transplant. Bone Marrow Transplant 2000;26:1117–1120. 150. Couriel DR, Geguelin GZ, Giralt S et al. Chronic graft-versus-host disease manifesting as polymyositis: an uncommon presentation. Bone Marrow Transplant 2002;30:543–546. 151. Stevens AM, Sullivan KM, Nelson JL. Polymyositis as a manifestation of chronic graft-versus-host disease. Rheumatology 2003;42: 34–39. 152. Lin PC, Hsiao LT, Chen PM. Acute polymyositis after donor lymphocyte infusion. Eur J Haematol 2005;74:166–168. 153. Bolger GB, Sullivan KM, Spence AM et al. Myasthenia gravis after allogeneic bone marrow transplantation: relationship to chronic graft-versus-host disease. Neurology 1986;36:1087–1091. 154. Baron F, Sadzot B, Wang F et al. Myasthenia gravis without chronic GVHD after allogeneic bone marrow transplantation. Bone Marrow Transplant 1998;22:197–200. 155. Mackey JR, Desai S, Larratt L et al. Myasthenia gravis in association with allogeneic bone marrow transplantation: clinical observations, therapeutic implications and review of literature. Bone Marrow Transplant 1997;19:939–942. 156. Dowell JE, Moots PL, Stein RS. Myasthenia gravis after allogeneic bone marrow transplantation for lymphoblastic lymphoma. Bone Marrow Transplant 1999;24:1359–1361. 157. Zaja F, Russo D, Fuga G et al. Rituximab for myasthenia gravis developing after bone marrow transplant. Neurology 2000;55: 1062–1063. 158. Maury S, Mary JY, Rabian C et al. Prolonged immune deficiency following allogeneic stem cell transplantation: risk factors and complications in adult patients. Br J Haematol 2001;115:630–641. 159. Rouah E, Gruber R, Shearer W et al. Graft-versus-host disease in the central nervous system: a real entity? Am J Clin Pathol 1988;89:543–546. 160. Azuno Y, Yaga K, Kaneko T et al. Chronic graft-versus-host disease and seizure. Blood 1998;91:2626–2628. 161. Kamble RT, Chang C-C, Sanchez S et al. Central nervous system graft-versus-host disease: report of two cases and review of the literature. Bone Marrow Transplant 2007;39:49–52.

19

Central Nervous System Infections in Cancer Patients Amy A. Pruitt,

MD

CONTENTS Introduction Approach to the Cancer Patient with a Possible CNS Infection High-Risk Patient Groups Clinical Manifestations and Management of Common CNS Infections Conclusion References

Summary Despite the development of effective prophylactic regimens and better antimicrobials for active infection, central nervous system infections in cancer patients continue to be a source of significant morbidity and mortality. The combination of more intense immunosuppression and longer survival has changed the spectrum of infections in several vulnerable populations. Patients at highest risk for CNS infection include hematopoietic cell transplant recipients and neurosurgical patients, although the intensity of chemotherapeutic regimens has resulted in an increased infection rate in patients with hematologic malignancies who do not receive transplants. The clinical presentation and radiographic appearance of infections in cancer patients may differ from those occurring in other populations, including HIV/AIDS patients. This chapter presents a systematic approach to the diagnosis of CNS infections in the most commonly affected patient groups based on patient risk factors, neuroanatomic site of disease, and laboratory tests. It then offers detailed clinical descriptions and management recommendations for the most common infections. Key Words: CNS infections, hematopoietic stem cell transplantation, chemotherapy, immunosuppression

1. INTRODUCTION In the past decade numerous new cancer therapies have been introduced. In addition to radiation therapy, cytotoxic chemotherapy and surgery, the therapeutic armamentarium now includes biologic response modifiers such as tyrosine kinase inhibitors, monoclonal antibodies, and numerous versions of hematopoietic stem cell transplantation. The refinement of hematopoietic growth factors has allowed more aggressive therapy with preservation of bone marrow function. With these improved therapies, cancer patients now survive longer. Despite these advances, however, the acute complications of intensive therapies and the risks of chronic immunosuppression have led to an increased incidence of central nervous system (CNS) complications, including infections. The severity of these infections produces significant challenges for physicians responsible for such patients’ care. The presentation and course of these infections may be different from those in patients without cancer, and new syndromes related to drug combinations or the degree of immune suppression surface regularly with earlier and more aggressive infections reflecting the disease and/or therapy-induced immune impairment. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

353

354

Part VI / Complications of Cancer Therapy

The precise spectrum of infections varies with geography and local medical practices, conditions to which each consultant must be sensitive. The clinician must also consider noninfectious disorders that can mimic CNS infections. These include the adverse effects of drug treatments, vascular lesions, radiation effects, and tumor recurrence. These complexities are significant, but it is now possible to approach each patient with a diagnostic strategy that clarifies differential diagnostic possibilities and produces a thorough, but efficient work-up. The major clinical presentations of CNS infections can be grouped into the meningoencephalitic syndromes and those due to focal brain lesions. Similarly, the range of pathogens can be narrowed by considering the type(s) of immune deficits pertinent to the particular patient. The two groups of patients who most frequently develop CNS infections are those undergoing procedures for primary brain tumors and hematopoietic stem cell transplant recipients. This chapter presents a diagnostic approach to the problem of suspected CNS infections. After a summary of the epidemiology of CNS infections in this patient population, an outline of disease-specific risk factors, clinical syndromes, and neuroimaging and cerebrospinal fluid (CSF) diagnostic testing is surveyed. The major groups of cancer patients at risk for CNS infections are then surveyed. Several recently recognized infectious syndromes and treatment complications and their time courses with respect to treatment regimens, including reversible posterior leukoencephalopathy (RPLS), immune reconstitution inflammatory syndrome (IRIS), and Epstein–Barr virus (EBV) reactivation with post-transplantation lymphoproliferative disorder (PTLD) will be covered. The chapter concludes with two discussions: (i) general recommendations for steroid use in acute infections and choice of antiepileptic drugs, and (ii) clinical and pathophysiologic information about organism-specific clinical syndromes and their treatment.

2. APPROACH TO THE CANCER PATIENT WITH A POSSIBLE CNS INFECTION 2.1. Clinical Challenges The diagnosis of CNS infection in cancer patients in a timely fashion that will ensure meaningful survival without neurological sequelae poses considerable challenges for many reasons: (i) The list of potential pathogens includes many organisms of low pathogenicity in the immunocompetent host. At the same time, the list of noninfectious causes of fever in cancer patients, particularly transplant recipients, is long and includes antibiotic and other drug toxicity, engraftment syndrome, graft-versus-host disease (GVHD), venous thromboembolism, CNS vasculitis, nonbacterial thrombotic endocarditis, RPLS, IRIS, and PTLD. (ii) Infection with more than one agent or sequential infection is common, and prior antibiotic use may lower the culture yield, while coexisting conditions such as metabolic encephalopathy may complicate the picture. (iii) Because of the host’s diminished inflammatory response or because of concurrent corticosteroid treatment, clinical clues such as headache, meningismus, and fever may be absent or mild. Conversely, as the host’s immune system reconstitutes after effective treatment, the clinical manifestations may mimic those of recurrent infection. (iv) Neuroimaging of CNS infections may be nonspecific and may mimic treatment-related abnormalities such as radiation necrosis or drug-induced leukoencephalopathy, making consideration of invasive brain biopsy an unfortunately frequent clinical problem. Knowledge of when to use more specialized testing such as magnetic resonance spectroscopy (MRS) or positron emission tomography (PET) accelerates diagnosis and minimizes invasive interventions. (v) Perhaps the major clinical challenge is keeping up-to-date with evolving cancer treatment regimens and with the neurological effects of new drugs and drug combinations. Major changes in infection risks, patterns, and syndromes that have occurred in the past 10 years in non-HIV-infected patients include the following epidemiologic and toxicologic observations from the author’s experience: (a) With the growing use in hematologic malignancies of potent immunosuppressive purine analogues such as fludarabine, pentostatin, and cladribine as well as anti-T and anti-B cell antibodies such as rituximab and alemtuzumab in dose-intense regimens that produce long-lasting cellular immune deficits, the number of nontransplant patients at risk nearly equals those receiving allogeneic stem cell transplants (1). (b) Nonmyeloablative allogeneic transplant patients (“mini-transplant”) have been shown to be at risk for serious infections as well as those receiving hematopoietic stem cell transplantations (HCT). Autologous HCT does not require immunosuppression after engraftment, but CD34+ selection has been used in such grafts to reduce tumor recurrence and has been associated with a different spectrum of CNS infection, particularly cerebral

Chapter 19 / Central Nervous System Infections in Cancer Patients

(c)

(d)

(e)

(f)

355

toxoplasmosis (2). Similarly, newer conditioning regimens that involve the use of imatinib, rituximab, and alemtuzumab change the spectrum and timing of opportunistic infections with fungi and viruses. The enhanced use of antimicrobial regimens has been accompanied by the selection of resistant organisms. Resistant bacteria [particularly enterococci and methicillin-resistant Staphylococcus aureus (MRSA) acquired both in the hospital and, increasingly, in the general community where over 60% of acquired infections are MRSA] assume a larger role and can be associated with lethal necrotizing fasciitis (3). Opportunistic fungi have become the most frequent and lethal pathogens in the past 25 years as nosocomial fungal infection disease rates have doubled (1). There is an increased incidence of fungal and other opportunistic diseases among patients who are not in an end-stage of their underlying disease, and the timing with respect to organ transplantation has changed so that, for example, Aspergillus infections occur later than they did 10 yeas ago and patients with chronic myelocytic leukemia now experience an increased incidence of dermatomal localized infections with varicella zoster virus (VZV) after treatment with imatinib (4). Nearly 200 million intravascular devices (IVDs) are sold in the United States each year, and the most common complication of vascular access devices is bloodstream infection (5). The spectrum of infections caused by IVDs ranges from local colonization to bacteremia or candidemia with septic shock. The increasing use of anti-infective coated catheters may modify the sensitivity of bloodstream cultures and require alternate strategies to confirm septicemia. There is a slightly lower incidence of meningitis due to Listeria monocytogenes as a result of trimethoprim/sulfamethoxazole prophylaxis for Toxoplasma gondii and Pneumocystis jirovecii. However, the increasing use of alemtuzumab in hematologic malignancies with resulting long-term T-cell depletion has caused reactivation of cytomegalovirus (CMV), which in turn increases the risk of Listeria. Similarly, prophylaxis with acyclovir in HCT patients has reduced the incidence of herpes viruses and CMV. However, progressive multifocal leukoencephalopathy (PML) is encountered in a broader group of patients, including those who have received rituximab (6) Reactivation of viruses such as EBV has led to a higher incidence of PTLD, the diagnosis and management of which with rituximab remain controversial (7).

2.2. Clinical Approach to the Diagnosis of Potential CNS Infection: Four Steps Accurate diagnosis requires four steps of data acquisition. 2.2.1. Epidemiologic Clues Series from specialized cancer hospitals and referral centers suggest that most CNS infections occur in a relatively small subset of cancer patients. Patients with HCT are a particularly high-risk group, while patients with leukemia or lymphoma represent more than a quarter of those with CNS infections. Another 16% of patients with CNS infections have primary CNS tumors (8). The responsible pathogens can be predicted in part by the type of immune defect and the degree and duration of immunosuppression. 2.2.1.1. Barrier Disruption. Shunts, monitoring devices, ventricular reservoirs, cranial surgery, central lines or ports, gastrointestinal surgery, urinary catheters and loss of cutaneous or mucosal integrity predispose to skin or gastrointestinal derived organisms including bacteria (S. aureus, S. epidermidis, Enterobacter, Escherichia coli, Klebsiella, Streptococcus bovis, Propionibacter acnes, Acinetobacter) and fungi (Aspergillus fumigatus and Candida albicans). In particular, S. bovis and Listeria meningitis are associated with gastrointestinal procedures and neoplasms. Another gastrointestinal pathogen associated with CNS infection is Strongyloides stercoralis, a nematode that typically colonizes the gut without symptoms; in immunosuppressed patients the larvae may circulate carrying enteric pathogens, causing a gram-negative bacillary meningitis. 2.2.1.2. Neutropenia. Bone marrow infiltration by leukemia, lymphoma, or solid tumors and drug- or radiationinduced marrow failure predispose to bacteria (S. aureus, S. phenomia, Pseudomonas aeruginosa, E. coli), fungi (A. fumigatus, C. albicans, Mucoraceae ) and viruses (CMV, HSV, human herpes viruses 6 and 7) and blood transfusion–associated viral infections such as adenovirus and West Nile virus. 2.2.1.3. B-lymphocyte/Immunoglobulin. Predisposing disorders include chronic lymphocytic leukemia (CLL), multiple myeloma, splenectomy, IgA deficiency, and Waldenstrom’s macroglobulinemia. Bacterial infections with S. pneumoniae, Haemophilus influenzae, Klebsiella pneumoniae, and P. aeruginosa and viruses such as measles and entroviruses are common pathogens.

356

Part VI / Complications of Cancer Therapy

2.2.1.4. T-lymphocyte/Macrophage. Perhaps the numerically largest group, patients in this group include those with HIV/AIDS, lymphoreticular neoplasms, organ transplantation, chronic corticosteroid use, and immunosuppressive therapy with cyclosporine, mycophenolate, tacrolimus, or alemtuzumab. Relevant viruses include HIV, CMV, HSV, VZV, HHV-6, HHV-7, EBV (PTLD), and JC virus (PML). Fungi include Cryptococcus neoformans, A. fumigatus, C. albicans, Mucoraceae, Pseudoallescheria boydii, and the parasites T. gondii and Strongyloides stercoralis. Bacteria include Listeria monocytogenes, N. asteroides, and Mycobacterium tuberculosis. 2.2.2. Clinical Syndromes CNS infections present with two broad categories of symptoms: (i) meningitis/meningocencephalitis with a variable combination of headache, meningismus, and altered sensorium, and (ii) focal signs with a primarily brain parenchymal process. The later group includes CNS mass lesions such as brain abscess, leukoencephalopathy, stroke-like vascular distribution lesions, and more clinically and radiographically restricted processes such as limbic encephalitis and brain stem syndromes. Unfortunately, depending on the degree of immunosuppression, patients may have atypical presentations that defy conventional clinical maxims. For example, disseminated CNS Nocardiosis can present as a neutrophilic meningitis with multiple nonspecific enhancing lesions instead of the usually isolated mass lesion described in earlier reports (9,10). Table 1 summarizes the most common infections by CNS neuroanatomic presenting site and magnetic resonance imaging (MRI) features. 2.2.3. Noninfectious Mimes The third step in diagnostic evaluation is to consider the possibility that a noninfectious process is mimicking an infectious one. Table 2 complements Table 1 by outlining evaluations of lymphocyte-predominant meningitis for both infectious and noninfectious conditions. Among the latter are many antineoplastic and antimicrobial agents. An ever-expanding list of antineoplastic agents and procedures raises the potential for novel adverse effects. Amphotericin B has been associated with a parkinsonian-like state that is reversible (11). Cefepime and cefixime have been associated with encephalopathy and nonconvulsive status epilepticus with or without renal dysfunction (12,13). Ifosfamide, busulfan, etoposide, and thiotepa all cause encephalopathy with somnolence, confusion, and sometimes seizures. 2.2.4. Laboratory Tests Based on the first three sets of information, the clinician can design a cost-effective group of laboratory tests. These may include serologic tests, lumbar puncture, CT, MRI, and, at times, brain or meningeal biopsy. CT and MRI have significant limitations in cancer patients who are taking corticosteroids that reduce contrast enhancement. Diffuse meningeal enhancement, on the other hand, can be seen during infections but also after blood–brain barrier disruption due to neoplastic, chemical, or ictal processes. Neither CT nor MRI can yet reliably distinguish tumor from abscess or radiation treatment effect, but the use of multiple MRI sequences including diffusion-weighted imaging (DWI), apparent diffusion coefficient (ADC) maps, and MR spectroscopy have improved the ability to distinguish between tumor, infection, and radiation necrosis. For example, using specific ADC cutoffs, it has been possible to differentiate toxoplasmosis from lymphoma with excellent reliability (14). Magnetic resonance angiography (MRA) can be helpful in the evaluation of arteritis associated with VZV, mycobacteria, and mucormycosis. Transfemoral arteriography remains the definitive test to exclude infectious aneurysm in patients with known or suspected endocarditis. Lumbar puncture is a major diagnostic procedure. In a cancer patient with known solid tumor, a screening CT or MRI is recommended prior to spinal tap if there is suspicion of metastatic disease or other mass lesion. Platelet counts of less than 50,000 should be fortified with platelet transfusions prior to the lumbar puncture. The interpretation of the resulting cerebrospinal fluid analysis (CSF) will depend on the extent of the patient’s immunosuppression and ability to mount an inflammatory response. All patients should have cell count and differential, glucose, and protein concentrations, routine bacterial cultures, and in appropriate situations cytology. A predominantly polymorphonuclear leukocytic (PMN) pleocytosis and CSF cell count of greater than 200 is suggestive of bacterial meningitis, though this range of WBC and differential can be seen in West Nile meningoencephalitis as well as with some fungi (15–18).

Paraneoplastic syndromes: Anti-Hu, Ma1, Ma2, Voltage-gated potassium channel antibodies

Repeated seizures

Nonbacterial thrombotic endocarditis

CNS vasculitis (graft-vs.-host disease, granulomatous arteritis) Chemotherapy DMSO preservative stem cell infusion

Hashimoto’s encephalopathy

Radiation-related arteriopathy

Cytomegalovirus*

Limbic Encephalitis Herpes simplex types 1 and 2 Human herpes virus

*Variable manifestations: diffuse encephalitis, mass lesions, myelitis, polyradiculitis.

Central pontine and extra-pontine myelinolysis Amphotericin (mainly frontal and post XRT) Rituximab Valproate Acyclovir

Reversible posterior leukoencephalopathy syndrome (RPLS) methotrexate, cyclosporine, cisplatin, l-asparaginase, tacrolimus, DMSO-treated stem cells, metronidazole, ifosfamide, cytosine arabinoside, gemcitabine Acute disseminated (toxic) leukoencephalopathy

Noninfectious conditions Immune reconstitution inflammatory syndrome (IRIS)

Emboli due to endocarditis

Progressive multifocal leukoencephalopathy (PML) Aspergillus

VZV

Stroke(s)

Infections Varicella zoster virus (VZV)

Leukoencephalopathy

Radiation necrosis

Immune reconstitution inflammatory syndrome (IRIS) Secondary tumor: Lymphoma Astrocytic tumor Metastases

Toxoplasma gondii EBV-virus associated CNS lymphoma

Bacteria (S. aureus/bacterioides P. acnes) Nocardia asteroides

Aspergillus fumigatus

Mass Lesion(s)

Graft-vs.-host disease

Central pontine myelinolysis

Wernicke’s encephalopathy

Varicella zoster virus PML

Listeria monocytogenes Cryptococcus neoformans

Brainstem

Table 1 Differential Diagnosis of CNS Infection by Predominant Focal Clinical Syndrome and MRI Appearance: Infections and Processes that Mimic Infections

358

Part VI / Complications of Cancer Therapy

Table 2 Major Causes of Aseptic Meningitis Syndrome in Immunocompromised Hosts Infectious Etiologies Viruses

Noninfectious Etiologies

Enteroviruses Herpes simplex types 1,2,6 Varicella-zoster Epstein–Barr HIV West Nile

Diagnostic Test (on CSF, unless indicated) polymerase chain reaction (PCR) PCR* PCR, virus-specific antibody PCR Blood PCR, virus-specific antibody*

Bacteria Partially-treated bacterial meningitis

Culture, low CSF glucose

Endocarditis Parameningeal infection Ventriculitis post neurosurgical procedure M. pneumoniae M. tuberculosis B. burgdoferi T. pallidum Fungi Cryptococcus neoformans Histoplasma capsulatum** Coccidioides immitis**

TEE, blood cultures MRI/CT of suspected region(s) Shunt/reservoir tap & removal Chext X-ray, serology PCR Blood: ELISA, Western blot VDRL/RPR blood and CSF MHATP, FTA-ABS

Adverse drug reactions NSAIDS Cox2 inhibitors OKT3 Valacyclovir, azathioprine, isoniazid IVIG Intrathecal chemotherapy Methotrexate, Ara-c, cephalosporins ADEM CNS vasculitis Arachnoiditis Prolonged status epilepticus PTLD

India ink, cryptococcal antigen Serum: CSF antibody Serum: CSF antibody

∗ For viruses such as HHV-6, pcr does not establish active infection and causation for clinical syndrome; acute and convalescent sera for virus-specific IgG to viruses may be obtained to confirm pathogenicity of specific infectious agent ∗∗ If patient lives in or has been in appropriate geographic regions. Abbreviations: ADEM, acute disseminated encephalomyelitis; Cox2, Cyclooxygenase-2; ELISA, enzyme-linked immunosorbent assay; FTA/ABS, Fluorescent Treponemal Antibody/Absorption; HIV, human immunodeficiency virus; IVIG, intravenous immunoglobulin; MHATP, microhemagglutinin-treponema pallidum; NSAIDS, nonsteroidal anti-inflammatory agents; OKT3, muromonab-CD3; PCR, polymerase chain reaction; PTLD, post-transplantation lymphoproliferative disorder; RPR, rapid plasma reagin; TEE, transesophagaeal echocardiogram; VDRL, Venereal Disease Research Laboratory

CSF glucose less than 50% of concurrent blood glucose supports a bacterial, fungal, or neoplastic process. CSF pleocytosis with lymphocytic predominance is the most common finding in cancer patients and includes a broad spectrum of possible etiologies summarized in Table 2 along with guidelines for polymerase chain reaction (PCR) and other studies that detect pathogens. Brain biopsy remains the definitive procedure in a few situations in which the etiology of a mass lesion remains unclear after MRI, MRS, and CSF studies, or if a lesion thought to be toxoplasmosis fails to respond to 2–3 weeks of appropriate antimicrobial therapy.

3. HIGH-RISK PATIENT GROUPS As emphasized at the start of this chapter, while hematopoietic stem cell transplant and other tissue recipients as well as neurosurgical patients are the groups at greatest risk for CNS infections, other patients receiving intensive immunosuppression for all types of cancers, particularly hematologic malignancies in the absence of transplantation, also pose challenges for the clinician attempting to diagnose a clinical problem that may be of infectious origin.

3.1. Hematopoietic Stem Cell Transplant and Solid Organ Transplant Recipients Hematopoietic cell transplantation (HCT) has become standard treatment for numerous oncologic disorders. Hematopoietic progenitor cells can be obtained from a closely matched sibling or unrelated donor (allogeneic

Chapter 19 / Central Nervous System Infections in Cancer Patients

359

transplantation) or from the patient prior to administration of chemotherapy (autologous transplantation). The cells may come directly from the bone marrow or may be collected from peripheral blood after stimulation by hematopoietic growth factors, the latter being the most commonly used technique at present. HCT refers to both sources of cells. In allogeneic transplantation, peripheral blood stem cells, which contain more T-cells than marrow, increase the incidence and duration of graft-versuss-host disease (GVHD). The cellsurface maker CD34 is used to estimate the peripheral blood stem cells, which are mobilized from the marrow using granulocyte colony-stimulating factor. Patients with allogeneic HCT need chronic immunosuppression to prevent graft rejection and GVHD. The frequency of CNS infections depends not only on the type of transplant received but also on the underlying disease, with leukemia patients having the highest rate of neurological complications (16). Allogeneic transplant recipients have a higher incidence of infectious as well as noninfectious complications. HCT transplant recipients are vulnerable to a wide variety of organisms presenting as meningoencephalitis and to a more restricted number of pathogens that produce mass lesions. The “Big Ten” organisms that produce mass lesions include four fungi— Aspergillus, Zygomyces, Cryptococcus neoformans, and C. albicans; two bacteria— N. asteroides, M. tuberculosis; three viruses—JC, VZV, EBV; and the parasite T. gondii. Table 2 summarizes the multiple causes of meningeal syndromes. In a recent series, CNS infections occurred in 4.2% of patients marked by onset within the first 4 months in 87% of patients and a high mortality rate of 47% (16). In this series from France, cerebral toxoplasmosis, fungal infections, and viral encephalitis predominated. The authors emphasized the increasing recognition that autologous transplant recipients are also at risk for infection in that two CD34+ selected autologous HCT recipients developed cerebral toxoplasmosis. The series is noteworthy in its emphasis on local clinical patterns of disease. Seroprevalence for toxoplasmosis is higher in Europe than in the United States, where invasive fungal infections outnumber cerebral toxoplasmosis in the first month after transplantation. Denier and colleagues further note that CD34+ selection in autologous peripheral blood stem cell transplantation results in delayed lymphoid reconstitution of both T- and B-cells as compared with recipients of non-CD34+ selected HCT transplants. This procedure also has been associated with an increased risk of CMV infections (17,18). Appropriate prophylactic recommendations for toxoplasmosis and CMV for this group of patients are in the process of development. Several additional excellent surveys of neurological complications of HCT have appeared in recent years, varying in their definition of a neurological complication, but largely confirming the clinically useful temporal division of the types of infectious complications likely to occur based on the interval from institution of HCT (19–24). The following discussion emphasizes both infectious and infection-mimicking conditions at various times from the transplant (Table 3). 3.1.1. Conditioning and Infusion Harvesting of stem cells from bone marrow or peripheral blood is quite safe. Patients with underlying autoimmune disorders such as scleroderma, chronic inflammatory demyelinating polyneuropathy, or multiple sclerosis may have an exacerbation of their disease while receiving recombinant human granulocyte colonystimulating factor prior to apheresis (20). Some patients will experience headache after diagnostic lumbar puncture and have MRIs that show diffuse dural enhancement. This finding should not be confused with meningeal inflammation and is a sign of low intracranial pressure. Several complications of stem cell infusion have been ascribed to the preservative dimethyl sulfoxide (DMSO). Transient global amnesia, seizures, and stroke syndromes have been reported. Several patients have been described with stem cell infusion-related cerebral infarction and also cardiac enzyme elevations consistent with myocardial ischemia (21). 3.1.2. Bone Marrow Depletion (0–30 days) The period of neutropenia before engraftment is a period of risk for infection from hospital-acquired organisms, intravenous line-related sepsis, and from infections acquired from the host tissue. Candida fungemia occurs during this period as a complication of line sepsis and prolonged hospitalization. It can be difficult to diagnosis because the picture is often one of a nonspecific meningoencephalitis although at post-mortem examination multiple microabscesses can be found.

360

Part VI / Complications of Cancer Therapy

Table 3 Time Course of Neurological Complications of Hematopoietic Cell Transplantation Time from Transplant

Infectious Conditions

Conditioning and infusion

Noninfectious Conditions Drug-related encephalopathy (busulfan, etoposide, ifosfamide, methotrexate, ara-C), rh-G-CSF-induced MS exacerbation*, DMSO-related stroke, RPLS, TGA, seizures, intracranial hypotension post-LP

< 1 month, neutropenic period

CMV HHV-6 Aspergillus Toxoplasmosis Infections acquired from donor tissue: LCMV, West Nile, rabies, claustridium, adenovirus, coxsackie virus (B4) Candida (from IV lines)

Engraftment syndrome Delirum due to organ failure Seizures: cefepime, imipenem RPLS: cyclosporine,tacrolimus Encephalopathy/parkinsonism: Amphotericin B, valproic acid SDH due to coagulopathy or LP Intraparenchymal brain hemorrhages**

1–6 months

HHV-6 HSV Toxoplasmosis VZV Aspergillus

ADEM Osmotic demyelination syndrome Brain atrophy (Ref. (138))

> 6 months, chronic conditions

VZV CMV reactivation PML EBV-associated PTLD+

GVHD (polymyositis, myasthenia, GBS or CIDP) Secondary malignancy, including brain tumors)# (Refs. (139,140)) Disease relapse

*rh-G-CSF = recombinant human granulocyte colony-stimulating factor (may exacerbate multiple sclerosis or other autoimmune disorders). ** Seen primarily in AML. + Risk related to intensity of T-cell depletion. # Almost all patients had received cranial or craniospinal radiation: tumors included astrocytoma/PNET (8.8 yr interval from treatment) and meningioma (20 yr from treatment). Abbreviations: ADEM, acute disseminated encephalomyelitis; CIDP, chronic inflammatory demyelinating polyneuropathy; DMSO, dimethyl sulfoxide; GBS, Guillain–Barré syndrome; LCMV, lymphocytic choriomeningitis virus; LP, lumbar puncture; PML, progressive multifocal leukoencephalopathy; PNET, primitive neuroectodermal tumor; RPLS, reversible posterior leukoencephalopathy syndrome; SDH, subdural hematoma; TGA, transient global amnesia.

The evolving epidemiology of tissue-acquired infections deserves some detailed discussion. In the past three years, transmission of organisms from host tissue has been reported for several virulent organisms: West Nile virus from an organ donor to four transplant recipients, and rabies from another donor whose disease was not recognized before the organ donation to another four transplant recipients (22,23). The transmission of West Nile virus by transplantation is of particular epidemiologic importance. The usual route of transmission is through mosquito bites, but blood and organ transplantation transmission yields particularly virulent disease in immunocompromised hosts. Clinical manifestations range from nonfocal meningoencephalitis to focal, flaccid paralysis. The presence of a febrile encephalopathic illness with poliomyelitis-like anterior horn cell syndrome with flaccid paralysis should focus on West Nile virus. A parkinsonian syndrome or other movement disorders develop in a minority of patients during or after the acute encephalitis (24,25). MRI shows progressive involvement of deep gray matter structures (globus pallidus, thalamus, substantia nigra, pons. and dentate nuclei), but the MRI early in infection often is normal, thus differing significantly from MRIs of the herpes viruses. Antibodies can be detected in blood 3–5 days into the clinical illness, and one of the striking CSF

Chapter 19 / Central Nervous System Infections in Cancer Patients

361

characteristics is a neutrophilic pleocytosis with a normal glucose. No drugs are effective for West Nile virus. Transmission of lymphocytic choriomeningitis virus by organ transplantation has been reported in eight patients all of whom had mental status changes, while several also had seizures. Seven of the eight reported patients died, underscoring the unusually virulent clinical syndrome in immunosuppressed patients and the need for careful donor screening (26). Bacterial meningitis with S. aureus and gram-negative hospital-acquired organisms is a threat at this time after the transplant, and antimicrobial coverage should reflect local hospital resistance patterns. Metabolic and drug-related encephalopathies are frequent in this period as are complications related to coagulopathies such as subdural hematoma (see Table 3). Patients receiving peripheral stem cell HCT may experience fever, rash, and headache that mimic CNS infection just as the absolute neutrophil count exceeds 500 mg/m3 . This engraftment syndrome is though to be due to upregulation of cytokines by neutrophils after administration of colony-stimulating factors (27). It is at this 2–4 week interval after transplantation time that infection with HHV-6 presents—a feature that distinguishes HHV-6 from other herpes virus infections that usually occur later. This rather specific syndrome of limbic encephalitis with confusion, short-term memory problems, sleep disturbance, and sometimes seizures has a specific MRI appearance with hyperintensity in the hippocampi on T2-weighted and fluid-attenuated inversion recovery (FLAIR) MRI (28–30) (Fig. 1). Cerebrospinal fluid may be normal. Most HHV-6 infections are considered to be reactivation through transmission of the virus from donor with the allograft has been documented (31). HHV-6, predominantly type B, has been documented in 38% of HCT recipients and in at least that number of solid organ transplant recipients (32). Risk factors include a conditioning regimen with OKT3 monoclonal antibodies or antithymocyte globulin, and seroconversion may be more frequent in patients who received immunosuppressive regimens containing sirolimus and IL-12 receptor antibodies as induction (33). Memory difficulties dominate the clinical picture and raise a differential diagnosis that includes paraneoplastic limbic encephalitis, voltage-gated potassium channel autoimmune limbic encephalitis, herpes simplex virus encephalitis, status epilepticus, Wernicke’s encephalopathy, and immunosuppressive treatment-related toxicity (34–36). There is some evidence that the occurrence of hypersensitivity reactions to drugs such as phenytoin or carbamazepine increases the risk of HHV-6 reactivation (37,38). The highest risk group of HCT patients is the group receiving grafts from unrelated donors or patients with severe GVHD. Increased levels of HHV-6 DNA are associated with increased requirement for platelet transfusions due to delayed platelet engraftment. The prognosis

Fig. 1. MRI FLAIR image shows hyperintensity in hippocampi in a hematopoietic cell transplant patient who developed fever, confusion, and memory loss around the time of engraftment. HHV-6 polymerase chain reaction was positive in CSF.

362

Part VI / Complications of Cancer Therapy

is not clear, because many patients have received antiviral therapy only rather late in their course, but mortality appears to be between 25% and 40% (39). It is important to recognize HHV-6 early because it can be treated with ganciclovir or foscarnet. Foscarnet may be preferable to ganciclovir in heavily bone marrow–suppressed patients, though nephrotoxicity is a concern with both antivirals. HHV-6 is also an immunosuppressive virus that may increase the risk of hepatitis C disease and super-infections with other opportunistic pathogens including invasive fungal infections (40). The indirect effects of HHV-6 appear to be more common in solid organ transplant recipients, while the encephalitic complications dominate the picture in HCT patients. Human herpes virus-7 also has been isolated from HCT recipients and has been reported to produce transverse myelitis and optic neuritis in addition to meningitis (41,42). 3.1.3. One to Six Months Post-Transplantation The period from one to approximately six months post-transplantation is the time when the risk of CNS infections is greatest. As neutrophil counts rise, the risk of bacterial infection decreases, but patients with allogeneic transplants remain at risk for several groups of infections including opportunistic fungi and parasites, herpes viruses and CMV. The prophylactic use of ganciclovir during the first 3 post-transplant months has reduced the incidence of herpes virus encephalitis and CMV infections, which are the most common infections occurring during this interval. Because of their importance and management issues Aspergillus, PML, and VZV are discussed in detail in a separate section on therapeutics at the end of this chapter. Other pathogens are increasingly recognized in the HCT population. Scedosporium apiospermum, a form of Pseudoallescheria boydii, is a mold that is widespread in the environment. It produces localized infection of bones and joint and disseminated infections including the CNS. These fungi now account for 25% of non-Aspergillus mold infections in organ transplant recipients, and nearly 30% of these infections affect the CNS. As is the case with Aspergillus, these infections tend to occur later than they did in earlier cohorts of patients, possibly due to antifungal prophylaxis at the onset of transplantation. Receipt of voriconazole treatment was associated with better survival compared with Amphotericin B (43). 3.1.4. More Than Six Months Post-Transplantation Patients requiring continued high-dose immunosuppression remain at risk for CNS infections during this later period. Confusion with disease recurrence becomes part of the difficult differential diagnosis. Patients who develop chronic GVHD require increased immunosuppression. While the CNS is rarely involved in GVHD, its presence should be viewed as a marker for the degree of immunosuppression and a barometer for infection risk. Autopsy series of patients with chronic GVHD reveal lymphocytic infiltration consistent with possible CNS viral infections (44,45). T. gondii remains an important pathogen with an MRI picture of multiple mass lesions, predominantly in the basal ganglia with minimal enhancement (Fig. 2) (46). The use of trimethoprimsulfamethoxazole partially reduces the risk. In areas of endemic fungal infection with Coccidioides immitis, organ transplant recipients are at risk for severe infection. Most of the cases occur more than six months after transplantation and those who receive targeted antifungal prophylaxis appear to be protected (47). Most cases of C. meningitis occur during this period. Patients on high-dose maintenance immunosuppression remain at risk for bacterial infections such as Listeria and Nocardia. VZV reactivation can occur at any point during the transplant process, but tends to be most common later several months. The rate is reported to been as high as 40% for allogeneic HCT and 25% for autologous HCT. Progressive multifocal leukoencephalopathy due to JC virus, a papovavirus-related DNA virus, becomes a risk at this stage from transplantation. The changing spectrum of conditioning regimens and chronic immunosuppressive agents has led to several newly recognized diseases or risk patterns and to an expansion of agents associated with syndromes or medication complications occurring up to several years after transplantion, about which neurologic consultants should be aware. Six special clinical situations occur predictably in HCT patients due both to the types of diseases treated and to the transplantation treatment: 3.1.4.1. Post-Transplantation Lymphoproliferative Disorder (PTLD). PTLD involves polyclonal-B cell infiltration of multiple organ systems, usually the abdomen, thorax, and allograft. In 15–35% of these patients, the CNS is involved and is the only site of abnormality in 85% of these patients or at least 10% of all PTLD

Chapter 19 / Central Nervous System Infections in Cancer Patients

363

Fig. 2. MRI FLAIR image with multiple deep bilateral lesions in a patient receiving an allogeneic stem cell transplant. Serology was positive for Toxoplasma gondii.

patients (48). PTLD may occur many months to years following transplantation. More than 90% of these lesions are EBV-associated lymphocyte proliferations of B-cell origin. Risk factors for EBV reactivation and subsequent EBV-LPD (Epstein–Barr virus lymphoproliferative disorder) include the use of unrelated or mismatched family donors, T-cell depletion by antithymocyte globulin, and nonmyeloablative stem cell transplants. The introduction of real-time polymerase chain reaction (PCR) to monitor EBV reactivation in asymptomatic patients is a potent tool to predict the risk of developing EBV-LPD. Both prophylactic and therapeutic strategies have been reported to prevent or treat EBV-LPD by B-cell depletion of the graft of restoration of EBV-specific T-cell immunity by infusion of donor lymphocyte (6). Preemptive treatment with B-cell depletion by the CD20 monoclonal antibody rituximab has been used. This agent is used in a variety of settings and continues to shape the pattern of late transplantation CNS infections. 3.1.4.2. Use of Rituximab. Rituximab has been used as a purging agent to eliminate tumor cells before and during stem cell collection and has a role in the treatment of minimal residual disease after transplantation as well for treatment of EBV-associated LPD described above. While it does not affect T-cells and thus has been considered to be a safe drug with respect to infection, its increasing use as consolidation therapy (four once-weekly rituximab infusions 2 months after high-dose chemotherapy) leads to B-cell depletion that persists for 2–6 months. Delayed onset neutropenia has been reported between 1 and 5 months after infusion (49). Reactivation of hepatitis B virus has been reported, though the prophylactic use of lamivudine allows administration of rituximab to patients with hepatitis B and C virus (50). Cases of PML and CMV have been reported in patients undergoing autologous HCT after the introduction of peritransplantation rituximab into treatment protocols (55). An additional case of PML associated with BK papovavirus infection, which is ten-fold less common than JC papovavirus, has been reported (51). While the contributory role of rituximab remains tentative, these cases emphasize the need for accurate surveillance in patients receiving peritransplantation rituximab. A recent advisory from the American College of Rheumatology for physicians considering rituximab for systemic lupus erythematosus and other rhematological conditions underscores the change in labeling for rituximab to include information about patient with nonHodgkin’s lymphoma who developed serious viral infections with CMV, HSV, VZV, and PML after treatment with rituximab. As of December 2006, 23 cases of PML had been reported following rituximab therapy in patients with NHL and other hematologic malignancies (52). The safety profile of rituximab is of most concern when it is considered

364

Part VI / Complications of Cancer Therapy

Fig. 3. Patient with follicular lymphoma diagnosed only 3 months prior to these MRI FLAIR images which show multiple areas of predominantly white matter abnormality that do not enhance on gadolinium sequences (not shown). Patient had received three courses of R-CHOP chemotherapy. Though she did not receive further immunosuppression, lesions progressed and were rapidly fatal.

as a choice of treatment of patients with more indolent lymphoma (53). While rituximab may improve the immediate outcome for patients with non-Hodgkin’s lymphoma, it has been associated with early PML infection when added to the CHOP chemotherapy regimen (cyclophosphamide, doxorubicin, vincristine, and prednisone) (54). At the author’s institution we have observed a rapid progression of PML after only three cycles of R-CHOP for a patient with a low-grade lymphoma (Fig. 3). 3.1.4.3. Use of Alemtuzumab. Alemtuzumab is a monoclonal antibody directed against the CD52 glycoprotein expressed on both B- and T-lymphocytes, monocytes, and natural killer cells that produces profound sustained lymphocyte depletion. The drug is approved for the treatment of B-cell lymphocytic leukemia but increasingly is used in stem cell and organ transplantation. Rates of opportunistic infection appear significant with CMV reactivation a particular issue. In one series 10% of patients developed opportunistic infection including CMV, BK virus, PTLD, HHV-5, and Nocardia (55). Alemtuzumab is also associated with an increased risk of fungal infections (56). Patients receiving alemtuzumab for rejection therapy were at higher risk for infection than those receiving the drug for induction therapy. Median time to infection after alemtuzumab was 84 days and CD4 count within one month of infection, when available, was 10 cells/mm3 . The author has cared for one patient who did not receive a transplant but was given alemtuzumab as salvage therapy for CLL. CMV reactivated and she then presented with a virulent and ultimately fatal meningitis thought likely to be Listeria on the basis of her MRIs, though cultures obtained after broad spectrum antibiotic treatment were negative (Fig. 4).

Fig. 4. Patient with CLL received alemtuzumab as salvage therapy. Cytomegalovirus reactivation was documented and treated with ganciclovir, but patient remained encephalopathic with seizures and with CSF pleocytosis and leptomeningeal enhancement (arrows) on gadolinium-enhanced MRI. CSF cultures remained negative, but the suspicion was of Listeria monocytogenes meningitis.

Chapter 19 / Central Nervous System Infections in Cancer Patients

365

3.1.4.4. Immune Reconstitution Inflammatory Syndrome (IRIS). IRIS denotes a series of infection reactivations recognized in the HIV/AIDS population with the advent of effective antiretroviral therapy. The primary CNS infections in which the syndrome has been recognized are cryptococcal meningitis, VZV, and PML (57–61). A progressive dementing syndrome with brain biopsy showing infiltration of CD8 T lymphocytes also has been described (62). As the HIV patient’s immune system becomes more effective, recrudescence of symptoms related to a prior CNS infection may occur or a new infection may be recognized. The host’s inflammatory response may be exuberant and may lead to new enhancement and/or edema in areas of prior CNS involvement (63). Such reactions also can occur in HCT recipients and should not be confused with ongoing infection as the treatment is directed not just at the organism but also at suppressing the inflammatory response. The inflammatory response can be primarily a meningoencephalitis with raised intracranial pressure or a recurrent mass lesion that can mimic abscess or tumor such as lymphoma (64). Guillain–Barré syndrome also has been described (65). As the immune system of heavily suppressed HCT patients reconstitutes, similar syndromes will likely be recognized. Distinguishing IRIS from worsening of the initial infection is important yet not always possible and the temporal spectrum of reactivation may be different in the AIDS population when compared to patients with HCT. The role of corticosteroids in blunting the host’s inflammatory response remains uncertain, though the author has found steroids useful in several patients with markedly raised intracranial pressure due to IRIS, and the risk of reactivating infection such as herpes zoster while treating brain edema of IRIS-associated cryptococcal meningitis is underscored by the above reports. Unfortunately, current immune modulating therapies such as steroids cannot selectively blunt the inflammatory response against normal tissue while maintaining the cytotoxic immune response against the underlying infection. 3.1.4.5. Reversible Posterior Leukoencephalopathy Syndrome (RPLS). The term reversible posterior leukoencephalopathy syndrome describes headaches, seizures, cortical visual disturbances, and delirium associated with transient cerebral lesions on FLAIR MRI. RPLS occurs in association with hypertensive encephalopathy, renal failure, eclampsia, intra-arterial angiography, and in response to a growing list of immunosuppressive and chemotherapeutic agents including the RAF kinase inhibitor sorafenib (66) and angiogenesis modifiers such as the vascular endothelial growth factor inhibitor bevacizumab used in various malignancies (67). Though the lesions may be predominant posteriorly, this is not always the case. A wider spectrum of MRI and clinical abnormalities is now recognized. Isolated cerebellar involvement has been reported (68) as has been a brainstem variant (69), and variable amounts of enhancement can be seen (Fig. 5). The differential diagnosis includes acute disseminated encephalomyelitis, arteritis, septic emboli, CNS lymphoma, and, in the appropriate settings, radiation therapy or metastatic tumor-related changes. Hypertension and elevated levels of cyclosporine or tacrolimus are the most common precipitating factors in the HCT population. 3.1.4.6. Graft-versus-Host Disease (GVHD). HCT recipients are more heavily immunosuppressed than their counterparts with solid organ transplants because of their underlying diseases and their conditioning regimens.

Fig. 5. Patient with breast cancer and widespread systemic relapse received one dose of gemcitabine and presented with confusion and visual symptoms. MRI (FLAIR) shows multiple areas of largely white matter abnormalities some of which enhanced consistent with reversible posterior leukoencephalopathy syndrome. Lesions partly resolved over the next several weeks, but patient continued to be confused.

366

Part VI / Complications of Cancer Therapy

Chronic immunosuppression to prevent GVHD keeps them susceptible to viral, parasitic, and fungal infections, though the use of prophylactic agents has reduced the incidence (or delayed the onset) of CMV. Episodes of acute GVHD require increase in immunosuppression and are associated with CMV and Aspergillus infection. Chronic GVHD has characteristics of multisystem-affecting autoimmune disorders, commonly involving the skin, bowel, and liver. Most of the manifestations of GVHD affecting the nervous system are peripheral, specifically polymyositis, peripheral neuropathy consistent with chronic inflammatory demyelinating polyneuropathy, and myasthenia gravis. This is not thought to be due to passive transfer of donor antibodies (70). The CNS rarely is involved, but Solaro et al. reported a recurring cerebellar syndrome associated with progressive motor weakness paralleling activity of GVHD (50). Reports of CNS angitis, some with angiographic confirmation of arteritic changes, suggest that the complication can occur up to several years after the HCT and may present with a stroke-like or hemorrhagic onset (71). Treatment with tacrolimus was reported by the authors of the cited report.

3.2. Neurosurgical Patients Patients having neurosurgical procedures are a second group at high risk for CNS infections. Patients with brain tumors, including both primary CNS malignancies and systemic metastases, who have had neurosurgical procedures account for 25% of CNS infections among cancer patients. Risk factors include barrier disruption, often with poor wound healing after multiple procedures and radiation therapy. Patients chronically on corticosteroids also have deficits in T-cell immunity. Pediatric patients have a similar set of risk factors and in a series of cancer patients with bacterial or fungal meningitis nearly two-thirds had recent neurosurgery, a CNS monitoring device or reservoir, or a CSF leak. S. aureus and S. pneumoniae were the most common microbiologic isolates (72). A particularly high incidence of S. bovis meningitis among neurosurgical patients also has been reported. Ommaya reservoirs, which are entered frequently for fluid sampling and chemotherapy, become infected with skin organisms in up to 15% of patients. Tumor cavity implantable carmustine (BCNU wafers, Gliadel® ) is approved for high-grade astrocytic tumors at initial diagnosis and for recurrence following radiation therapy. The robust cerebritis and vasogenic edema associated with these wafers requires large doses of corticosteroids at a time. High infection rates up to 28% (9/32, four of whom had cerebral abscesses) have been reported (73). MRI appearance has been described as “Swiss cheese enhancement.” Other patients have a persistent ring-enhancing lesion. Organisms are skin pathogens, although some of the cavities and biopsied areas of enhancement are sterile cerebritis (Fig. 6).

Fig. 6. Patient with glioblastoma received Gliadel wafers at relapse. MRI with gadolinium shows cerebral edema on FLAIR (A), “Swiss-cheese” enhancement (B) consistent with sterile cerebritis.

Chapter 19 / Central Nervous System Infections in Cancer Patients

367

Table 4 Treatment of Common CNS Infections in Cancer Patients Organism Bacteria Staphylococcus aureus Streptococcus pneumoniae Penicillin intermediate resistance Penicillin resistant Listeria monocytogenes Gram negatives (except Pseudomonas) Pseudomonas aeruginosa Nocardia asteroids Viruses Herpes simplex (encephalitis) Varicella zoster (encephalitis) (dermatomal) HHV-6, types A and B Cytomegalovirus Epstein-Barr virus (PTLD) Enteroviruses Fungi Cryptococcus neoformans

Aspergillus species**

Mucoraceae Candida species*** Histoplasma capsulatum Parasites Toxoplasma gondii

Antibiotic Regimen (Intravenous Route except as noted) Methicillin-sensitive: Nafcillin 2g q4h PLUS Cefotaxime 2 g q6h Methicillin-resistant: Vancomycin 500 mg q6h +/– intraventricular, Vancomycin 20 mg/d MIC* < 0.1–1μg/mL, Cefepime 2 g q12h OR Ceftriaxone 2 g q12h OR Cefotaxime 2 g q4h MIC > 1 μg/mL one of above cephalosporins PLUS Vancomycin 500 mg q6h +/– intraventricular Vancomycin 20 mg/d Ampicillin 2–3 g q4h PLUS Gentamicin 2 mg/kg q8h Ceftriaxone PLUS Gentamicin 1.5 mg/kg q8h, granulocyte transfusions for neutropenia (Ref. (85)) Ceftazidime 2 g q8h OR Cefepime 2 g q12h OR Meropenem 2 g q8h Sulfadiazine 8–12 g/d Acylovir 10–12 mg/kg q 8h§ (? chronic po Valacyclovir)¶ Acyclovir 10–12 mg /kg q 8 h Valacyclovir 1000 mg po twice daily for 10 days OR Famciclovir 500 mg po 3 times daily for 10 days OR Acyclovir 200 mg po 5 times daily for 10 days Foscarnet 60 mg/kg q8h Ganciclovir 5 mg/kg q12h PLUS foscarnet Acyclovir 10 mg/kg q8h Pleconaril 200 mg po 3 times daily for 7 days Amphotericin B 0.7 mg/kg per day followed by Fluconazole 400–800 mg/d po OR AmBisome 4 mg/kg per day OR ABLC† 5 mg/kg per day for 2 weeks PLUS Flucytosine 150 mg/kg/d po for 6 weeks Amphotericin B 0.8–1.25 mg/kg per day OR AmBisome OR ABLC PLUS (for all three above drugs) Itracanazole 600–800 mg/day po X 4d OR ABLC 5 mg/kg per day OR AmBisome 5 mg/kg per day PLUS surgical debridement Amphotericin 0.7 mg/kg per day PLUS Flucytosine 25 mg/kg 4 times daily Amphotericin 0.7–1.0 mg/kg per day PLUS Itraconazole 400 mg/d Sulfadiazine 1.5–2 g po 4 times daily PLUS Pyrimethamine 100–200 mg po load then 75–100 mg po qd PLUS Folinic acid 10–50 mg po qd

* MIC minimal inhibitory concentration. ** Aspergillus species treatment with caspofungin and voraconazole under study (Munoz) *** New class of antifungal agents anidulafungin under study (Vazquez). † ABLC amphotericin B lipid complex. § Adjustment of dose required in renal failure/dialysis patients. ¶ Chronic po prophylaxis currently under study.

368

Part VI / Complications of Cancer Therapy

Cranial irradiation and corticosteroid therapy, both commonly employed for brain tumor treatment, pose risks for CNS infections. Radiation therapy can exacerbate wound healing problems and has been reported to reactivate herpes simplex virus, causing encephalitis (74). As a group, brain tumor patients are exposed chronically to some of the highest doses of corticosteroids and are therefore at risk for Pneumocysitis jirovecii (carinii) pneumonia. They should receive trimethoprim/sulfamethoxazole prophylaxis three times weekly. Prophylaxis with the above regimen also decreases the risk of Listeria, toxoplasmosis, nocardiosis, and urinary tract infections. VZV reactivation is common. No study has yet investigated the efficacy of offering VZV vaccination to patients under age 60 about to undertake steroid courses, but vaccination for everyone over age 60 is now recommended and may reduce the frequency of symptomatic VZV reactivation (75). Reactivation of VZV in the scalp dermatomes puts patients at risk for poor healing and for post-herpetic neuralgia. They should be considered for intravenous acyclovir therapy followed by valacyclovir (Table 4). Reactivation of other viruses such as hepatitis B and C complicates the neurological course in many patients and can cause clinicians erroneously to attribute altered liver function tests (LFTs) to chemotherapy toxicity. Hepatitis B reactivation, especially in patients with lymphoma or breast cancer, is well described (76). Rituximab has been associated with early or delayed reactivation of hepatitis B alone or in combination with R-CHOP therapy. An example of hepatitis B reactivation during glioblastoma treatment with temozolomide with therapeutic response to lamivudine illustrates the increasing scope of the problem (77). Given the widespread seroprevalence of hepatitis B, screening prior to temozolomide chemotherapy is suggested by the authors who advocate prophylactic lamivudine during and for at least 2 months after temozolomide treatment. Chemotherapy patients with elevated LFTs should have viral hepatitis included in the differential diagnosis along with medication-induced hepatotoxicity.

4. CLINICAL MANIFESTATIONS AND MANAGEMENT OF COMMON CNS INFECTIONS This section outlines the varied manifestations and management of common CNS infections in cancer patients These diseases are highlighted because of the evolving spectrum with respect to timing from cancer treatment or because of new management recommendations. Table 4 summarizes specific antimicrobial treatment recommendations.

4.1. General Medical Management Issues Regardless of the specific CNS infection, two medical management issues—corticosteroid supplementation and seizure management—require special mention. Every cancer patient with suspected CNS infection is at risk for adrenal insufficiency. Many such patients have been treated with large doses of corticosteroids in the recent past and when stressed with an acute infection will have insufficient reserves (78). Adrenal insufficiency presents as hypotension unresponsive to volume repletion and requires urgent intravenous hydrocortisone (79). Another medical complication to which patients with meningoencephalitis and, more frequently, infectious mass lesions are prone to is seizures. Treatment of seizures raises complex management issues. Most seizures in the transplant population are isolated events due to toxic drug reaction metabolic or electrolyte abnormalities, or to RPLS, or that do not require long-term antiseizure medicine (80). While there are no definitive guidelines, some clinicians treat with antiepileptic drugs (AEDs) for at least one month after a recognized seizure. Cerebrovascular disease in the form of bland infarction, intracerebral hemorrhage, venous sinus thrombosis, or abscess may cause longer-term risks. Seizure-producing drugs likely to be encountered in the transplant population include: tacrolimus, cyclosporine, muromonab-CD3 (OKT3), busulfan, quinolone antibiotics, beta lactams, penicillin, cephalosporins, and bupropion. If it is deemed necessary to treat seizures, the choice of AEDs should consider interactions between the seizure treatment and HCT regimen. Commonly used AEDs such as phenytoin and fosphenytoin may reduce immunsuppressant blood levels. Drugs such as phenytoin, carbamazepine, and valproate are heavily protein-bound, and in transplant patients whose serum protein levels may be low, free-level measurement is indicated. Hypersensitivity reactions to phenytoin, carbamazepine, oxcarbazepine, and lamotrigine present with fever and rash, and, at times, a Stevens–Johnson syndrome. Pancytopenia may be present. Pancreatitis and hepatic dysfunction with valproate along with an encephalopathy and occasional parkisonian-like appearance have been reported.

Chapter 19 / Central Nervous System Infections in Cancer Patients

369

Enzyme-inducing drugs reduce cyclosporine, tacrolimus, and corticosteroid levels. Newer agents such as levetiracetam, recently available in parenteral form, have the advantage of rapid onset of activity, oral preparations, and absence of significant protein binding or hepatic enzyme induction. With levetiracetam, which is emerging as the drug of choice in the HCT population, no drug–drug interactions are noteworthy, though about 5% of patients may experience adverse neuropsychiatric effects that can look like behavior associated more commonly with corticosteroid psychosis. Renal failure may reduce clearance of levetiracetam. Hemodialysis removes small nonprotein bound drugs such as levetiracetam and gabapentin. Drug levels must be adjusted for renal failure and supplementation after dialysis is necessary. The manufacturer’s recommendations should be followed.

4.2. Bacterial Meningitis Potentially the most immediately lethal of cancer patient CNS infections, bacterial meningitis, is caused by many of the same organisms that afflict the normal host such as S. pneumoniae, but also, particularly when the infection is nosocomially acquired, by gram-negative organisms, by S. aureus and S. bovis in neurosurgical patients (81) and by L. monocytogenes. This last organism does not offer the characteristic rhombencephalitis seen in normal hosts with a brainstem encephalitic picture (82). Instead, cancer patients are more likely to have a nonspecific menigoencephalitis picture which in recent years occasionally has been confused with West Nile encephalitis or even with leptomeningeal metastases (83,84). While Table 4 provides general guidelines, successful treatment of bacterial meningitis depends on hospitalspecific guidelines for the use of antimicrobial agents in neutropenic cancer patients. It is recommended that adult patients with bacterial meningitis receive dexamethasone 10 mg intravenously 15 min before the first dose of antibiotics. In the original Dutch studies, this dose of dexamethasone was continued for four days. Dosage and duration of dexamethasone treatment remain under investigation, but the principle that steroid treatment before antibiotics reduces mortality and neurological sequelae has achieved consensus in adults and children with Hemofluous influenzae meningitis or pneumococal meningitis (85–87). Although there has been no prospective randomized clinical trial, adjunctive dexamethasone for 2 days only may be optimal and likely will reduce the risk of steroid complications. Empiric treatment for meningoencephalitis in a HCT recipient should include a combination of a third-generation cephalosporin (cefotaxime or ceftriaxone) or a fourth-generation cephalosporin (cefepime) plus vancomycin plus ampicillin plus acyclovir for communityacquired infections. Infections acquired nosocomially require ceftazidime or cefepime for P. aeruginosa coverage. Amphotericin B is not usually introduced as part of the empiric regimen. Vancomycin should never be given alone with dexamethasone in the treatment of bacterial meningitis because reduction in inflammation may reduce access to the CSF and delay sterilization. Dexamethasone is not currently recommended for the treatment of gram-negative bacillary meningitis.

4.3. Fungal Infections 4.3.1. Aspergillus Aspergillus has become the most common fungal infection in HCT and solid organ transplant recipients and increasingly is a problem in other cancer patients on prolonged antibiotics (88–90). Of the many Aspergillus species, A. fumigatus and A. flavus account for most cases of invasive aspergillosis. The portal of entry is usually the lung, but infection also can be acquired though skin barrier breakdown and the ear or cornea. Most patients have pulmonary disease or sinusitis. Neutropenia is a major risk factor. Cerebral involvement with the fungus occurs in up to 20% of invasive aspergillosis cases and Aspergillus accounts for 50% of brain abscess in bone marrow transplant recipients (91). A particularly serious clinical feature is its tendency to cause vascular thrombosis with stroke and hemorrhage due to septic aneurysm or vasculitis (92). Multiple brain abscesses as well as paraplegia from spinal cord involvement have been described as well (93). Despite the advent of diagnostic methods that speed pathogen identification and the introduction of new antifungal agents with better CNS penetration, the prognosis for invasive aspergillosis remains poor. While solitary or multiple cerebral abscesses are the most common clinical presentations, infectious aneurysms, myelitis and carotid artery invasion have been reported and extensively reviewed (94). CSF cultures are almost always negative, and definitive diagnosis requires a biopsy.

370

Part VI / Complications of Cancer Therapy

CT and MRI have been helpful in raising suspicion of Aspergillus in the appropriate clinical setting of a neutropenic patient with abnormal chest X-ray. There can be multiple embolic lesions with ring or solid enhancement consistent with abscess or granuloma, usually in less severely immunocompromised patients or, in particularly vulnerable hosts, rapid progression of stroke-like syndrome (Fig. 7). A promising test for early diagnosis of fungal infection is ELISA assay for galactomannan, a cell-wall component of Aspergillus that has shown sensitivity of 90% and specificity of 98%, the abnormality being detected up to a week before the appearance of signs of infection on chest radiography. Another antigenic test is 1,3--d-glucan, a cell wall component. These tests are usually done with blood but can be performed on CSF as well (95). Although amphotericin B remains the recommendation of first line treatment for CNS aspergillosis, the efficacy of voriconazole has been documented in several studies and may be superior to Amphotericin B (96). Echinocandins are a new class of drugs represented by caspofungin and micafungin that have a more favorable adverse event profile than the traditional antifungal agents, particularly for patients with renal failure. Although few patients with CNS aspergillosis were included in the study, caspofungin monotherapy was recently shown to have a response rate of 45% in refractory invasive aspergillosis (97). Combination therapy with voriconazole plus caspofungin as initial therapy for invasive asperillosis in solid organ transplant recipients demonstrated improved results, particularly in patients with renal failure, and combination therapy increasingly is considered (98,99). Anidulafungin, a novel echinocandin with excellent in vitro activity against Aspergillus and many Candida species, is effective in treatment of esophageal candidaiasis, including azole-refractory disease. Unlike caspofungin, its pharmacokinetics are not affected by concomitant treatment with cytochrome P450 modifiers and it does not alter cyclosporine levels (100). This agent may be used safely with liposomal Amphotericin B; further trials for combination efficacy in invasive aspergillosis are necessary.

Fig. 7. Multiple ring-enhancing lesions consistent with aspergillosis (A, B) in heavily immunosuppressed patient with non-Hodgkin’s lymphoma in relapse. Rapid progression of posterior fossa lesions led to brain death examination within days (C, D).

Chapter 19 / Central Nervous System Infections in Cancer Patients

371

Clearly, with the poor prognosis associated with invasive aspergillosis, determination of optimal prophylactic regimens becomes essential. Fluconzaole effectively prevents Candida infection, but Aspergillus species and other molds are unaffected. A new azole drug, posaconazole, has activity against a wide range of yeasts (including Candida species) and molds (including Aspergillus and Zygomycetes). Two controlled trials recently evaluated prophylactic use of posaconazole. The first study evaluated patients undergoing chemotherapy for acute leukemia or myelodysplastic syndromes. Patients receiving posaconazole had significantly lower rates of invasive fungal disease than those receiving fluconazole or itraconazole, though posaconazole had a higher incidence of adverse events (101). In the second study posaconazole was compared to fluconazole, and had a favorable profile for prophylaxis in HCT recipients receiving immunosuppressive therapy for GVHD (102). An accompanying editorial emphasizes that the situation remains confusing. Posaconazole appears to be the drug of choice for prophylaxis, whereas voriconazole is the preferred treatment for proven or probable asperillosis, while caspofungin and liposomal amphotericin B are options for empiric therapy in patients with persistent fever and neutropenia (103,104). A potential drug interaction in allogeneic HCT recipients is the rise in sirolimus levels when voriconzaole is co-administered. A 90% reduction in sirolimus dose is required in the presence of voriconazole (105). 4.3.2. Candida Species Central venous catheters, antibiotics, corticosteroids, diabetes mellitus, prolonged neutropenia, and HCT are all risk factors for Candida species infection. C. albicans, once the predominant species, now represents only about 45% of isolates and infections due to Candida glabrata, parapsilosis, and tropicalis, which are generally less susceptible to traditional antifungal agents, are an increasing problem (106). Antemortem diagnosis is difficult in immunocompromised patients. CSF may be negative, though a neutrophilic meningitis may be present. Mannan is a surface antigen on the cell wall of C. albicans and can be detected in the serum and CSF of patients with candidal meningitis (107). Voriconazole has good CSF penetration and has been described as effective against invasive candidaisis (108). It can be considered an alternative to amphotericin B and flucytosine (109). Drugs with acceptable CSF penetration include 5-flucytosine, fluconzaole, and voriconazole (least). Amphotericin B and itraconazole, the lipid formulations of amphotericin B (AmBisome), ABLC (amphotericin B lipid complex), and ABCD (amphotericin B colloidal dispersion), caspofungin, and posaconazole all have poor CNS penetration. 4.3.3. Cryptococcus Cryptococcus neoformans is one of the most common causes of meningitis in patients with defects of cellmediated immunity of greater than 6 weeks duration, a scenario common to patients with HIV or those on chronic corticosteroid therapy. Among patients with malignancies, those with lymphoma, CLL, and AML are at highest risk. The disease may present either as focal mass lesions or as a more diffuse meningoencephalitis, characteristically with greatly elevated intracranial pressure and variable inflammatory response correlated with the host’s degree of immunosuppression. One of the major risks early in the course of cryptococcal meningitis is visual loss, either from persistently elevated intracranial pressure or from direct invasion of the optic nerves (110). Treatment recommendations are summarized in Table 4 and involve triple drug therapy. Morbidity remains high and obstructive hydrocephalus requiring ventriculostomy or shunting is a frequent complication in severely immunocompromised patients. Detection of cryptococcal polysaccharide antigen allows prompt diagnosis. The suggested optimal treatment for HIV-negative immunosuppressed patients is amphotericin B. Liposomal amphotericin B appears to be equally effective and better tolerated and in in vitro studies posaconazole appears to have good activity. The combination of posaconazole and amphotericin B needs to be evaluated in cryptococcal meningitis. The echinocandins have poor activity against Cryptococcus. 4.3.4. Zygomycetes The order Mucorales has become a more prominent pathogen in heavily immunosuppressed patients, perhaps because of the excessive use of some antifungal agents (111). Vascular invasion and tissue necrosis are the hallmarks and prognosis is poor. Risk factors for mucormycosis include acute leukemia or lymphoma HCT,

372

Part VI / Complications of Cancer Therapy

prolonged neutropenia, diabetes mellitus, and renal failure. Controlling hyperglycemia, surgical debridement, reduction of immunosuppression, and the transfusion of colony-stimulating factors or granulocytes may be helpful in controlling infection. Fluconazole and 5-flucytosine are ineffective and caspofungin and voriconazole are also unhelpful (112).

4.4. Viral Infections Herpes viruses are the most common CNS pathogens in cancer patients. Several of these, including HHV-6, HHV-7, and EBV have been discussed earlier in this chapter. 4.4.1. Varicella Zoster Virus As many as 15% of patients with leukemia and lymphoma, and those with HCT will develop symptomatic VZV infection (113–115). Such patients are at risk for dissemination of initial dermatomal infection. VZV reactivation can occur at any time post-transplantation, but the particular time of risk is with reconstitution of the immune system following bone marrow engraftment (116). Most VZV-related fatalities occur from disseminated disease and the use of intravenous acyclovir for localized VZV is highly effective at halting disease progression (117). Intravenous acyclovir is indicated in all patients with allogeneic HCTs and HCT with moderate or severe acute or chronic GVHD. The option of treating less severely immunocompromised patients with oral antiviral agents is not yet well established, but it appears that famciclovir and valacyclovir treatment in the outpatient setting may be appropriate. Brivudin is not recommended for immunocompromised patients because of potentially fatal interaction with 5-fluororuracil and other 5-fluropyrimidines used in chemotherapy (117). Post-herpetic neuralgia may occur as much as two to three times more frequently in cancer patients than in the general population and early use of antiviral agents also helps reduce postherpetic pain. The American Academy of Neurology Practice Parameter suggests that gabapentin, pregabalin, tricyclic antidepressants, and topical lidocaine patches are effective and can be used in the treatment of postherpetic neuralgia. Antiepileptic drugs such as carbamazepine are not recommended (118). There is no evidence that systemic steroids are helpful and they may be harmful in the heavily immunocompromised population. Quan and colleagues have reported improvement in post-herpetic neuralgia after treatment with intravenous acyclovir followed by oral valacyclovir, and this promising strategy deserves further investigation (119). Less commonly seen peripheral syndromes of VZV are focal segmental weakness, peripheral facial palsy, and hearing loss from herpes zoster oticus which has been associated with segmental pontine encephalitis (120). Herpes zoster may be seen without a rash and may present with a variety of unusual syndromes so that early suspicion and confirmation of diagnosis with PCR is essential for prompt treatment (121). Spinal dissemination can take the form of acute myelitis (122,123). Spinal cord involvement as seen in Fig. 8 can occur weeks after the dermatomal involvement, sometimes just after prophylaxis is discontinued and suggests that long-term oral valacyclovir may be important. Dissemination of VZV to the brain takes the form of either an acute, necrotizing encephalitis (124) or a discrete cerebrovascular disorder following trigeminal involvement with VZV arteritis, multiple strokes, or carotid occlusion (125). Acute retinal necrosis may develop early or be delayed by several months. Patients with ophthalmic zoster should be referred early for expert ophthalmologic care. 4.4.2. PML Nearly 50 years ago, E.P. Richardson and colleagues described two patients with CLL and one with Hodgkin’s disease who had multiple demyelinating lesions of the brain. A clue to the viral etiology came from inclusion bodies in oligodendrocytes and ultimately papovavirus particles were seen and the virus was isolated. In the past 25 years since the advent of HIV, PML has been recognized as a major opportunistic infection. In the HIV era and after introduction of highly active antiretroviral therapy (HAART), 80% of PML patients have AIDS, 13% have hematological malignancies, and 5% are transplant recipients (126). An enlarging group of patients with inflammatory conditions including rheumatologic disease and multiple sclerosis treated with immune-modifying agents is emerging. Among HIV-negative cases at the Mayo Clinic, 55% have hematological malignancies and

Chapter 19 / Central Nervous System Infections in Cancer Patients

373

Fig. 8. Patient received HCT for ALL. He had been taking valacyclovir after a thoracic dermatomal herpes zoster outbreak. Two weeks after stopping the antiviral agent, he developed paraparesis and bladder dysfunction. MRI shows multisegmental involvement of the cord with enhancement (A: T2-weighted, B: T1 post-gadolinium) consistent with varicella zoster myelitis. PCR was positive in CSF, which showed pleocytosis of 85 lymphocytes. Patient did not improve with acyclovir intravenous therapy.

7% are transplant recipients. Among HIV-negative patients with lymphoproliferative disorders, those treated with purine analogues such as fludarabine may be at greatest risk (127). The typical clinical presentation is of subacute focal deficits (weakness, numbness, ataxia, hemianopsia). A version of PML isolated to the brainstem has been described (128). In HIV patients HAART therapy has improved survival. However, in HCT and other cancer patients, survival is dependent on the ability to reduce immunosuppression while controlling the underlying malignancy and preserving the transplant. Diagnosis is suspected on MRI on which there are often asymmetric lesions, hyperintense on T2 and FLAIR, with minimal enhancement (Fig. 9). An exception to the situation of minimal enhancement is when PML is new or worsening shortly after initiation of HAART with a rise in CD4+ counts. (See section 3.1.4.4) Demyelination of the CNS is not the only disease caused by JC virus. A cerebellar syndrome with ataxia and cerebellar atrophy with focal cell loss in the internal granule cell cerebellar layer but without classic PML demyelinating lesions has been described. This novel syndrome has been called JCV granule cell neuronopathy (129).

Fig. 9. Progressive multifocal leukoencephalopathy in a patient with leukemia after HCT. Diagnosis is suspected on MRI on which there are often asymmetric lesions with minimal enhancement (A) and hyperintense on T2 and FLAIR (B).

374

Part VI / Complications of Cancer Therapy

Brain biopsy has been the definitive diagnostic procedure, though until recently CSF JC virus PCR had a sensitivity of 72–92% and a specificity of nearly 100% (130). However, since the advent of HAART, partial recovery of the immune system has been associated with decreased viral replication and clearance of JCV from CSF with reduction in PCR sensitivity to about 58%. There is no specific treatment for PML. In HIV patients HAART is the best option, and in HIV-negative patients reduction of immunosuppressives such as corticosteroids or calcineurin inhibitors is preferred. Numerous drugs have been tried, but cytosine arabinoside appears not to be effective either intravenously or intrathecally. Interestingly, clinical improvement was noted in one natalizumab-treated multiple sclerosis patient with PML at a time of massive IRIS inflammatory reaction 3 months after discontinuation of natalizumab and concomitant with treatment with cytarabine (131–133). Cidofovir, a medicine with excellent activity against CMV, was active against polyoma virus in animal models, but was of no additional benefit in HIV patients treated with HAART (134). On the other hand, successful cidofovir therapy in a patient with angioimmunoblastic T-cell lymphoma has been reported (135). Topetecan and interferon -2B are ineffective. However, with increasing information about how JC virus penetrates glial cells via the cellular 5-hydroxytryptamine-2a (serotonin) receptor, blockade of this receptor with agents such as mirtazapine has been tried (136). While optimal therapy has yet to be elucidated, it is clear that the presentation of PML is more varied both in terms of brain distribution and in terms of enhancement characteristics than original described. Koralnik has suggested that the name PML is a misnomer since the disease is neither invariably progressive nor always multifocal nor even exclusively in white matter (130).

5. CONCLUSION The diagnosis and management of CNS infections remain ever-changing and persistently challenging areas of clinical care. Despite efforts to stratify patients by epidemiologic risk factors, clinical syndromes and appropriate diagnostic studies, diagnostic and therapeutic uncertainties are common and morbidity and mortality from CNS infections continue to be high in the cancer population. The clinician should remain aware of emerging infections, transfusion safety issues (137), changing microbial susceptibilities, synergistic infections, and evolving novel cancer therapies that will continue to impact the nervous system in new ways (138–140). The effective neurologic diagnostician will keep all these parameters in mind while remembering that two or more infections or infectious mimes may coexist.

REFERENCES 1. Blijlevens NM, Donnelly JP, DePauw BE. Microbiologic consequences of new approaches to managing hematologic malignancies. Rev Clin Exp Hematol 2005;9:E2. 2. Denier C, Bourhis, J-H, Lacroix C et al. Spectrum and prognosis of neurologic complications after hematopoietc transplantation. Neurology 2006;67:1990–1997. 3. King MD, Humphrey BJ, Wang YF et al. Emergence of community-acquired methicillin-resistant Staphylococcus aureus USA 300 clone as the predominant cause of skin and soft-tissue infections. Ann Intern Med 2006;144:309–317. 4. Mattiuzzi GN, Cortes JE, Talpaz M et al Development of varicella-zoster virus infection in patients with chronic myelogenous leukemia treated with imatinib mesylate. Clin Cancer Res 2003;9:976–980. 5. Safdar N, Fine JP, Maki DG. Meta-analysis: Methods for diagnosing intravascular device-related bloodstream infection. Ann Intern Med 2005;142:451–466. 6. Goldberg SLS, Pecora AL, Aler RS et al. Unusual viral infections (progressive multifocal leukoencephalopathy and cytomegalovirus disease) after high-dose chemotherapy with autologous blood stem cell rescue and peritransplantation rituximab. Blood 2002;99: 1486–1488. 7. Annels NE, Kalpoe JS, Brodius RGM et al. Management of Epstein–Barr virus (EBV) reactivation after allogeneic stem cell transplantation by simultaneous analysis of EBV DNA load and EBV-specific T cell reconstitution. Clin Infect Dis 2006;42:1743–1748. 8. Pruitt AA. Nervous system infections in patients with cancer. Neurol Clin N Amer 2003;21:193–219. 9. Zaatreh M, Alabulkarim W. Images in clinical medicine: disseminated central nervous system nocardiosis. N Engl J Med 2006;354:2802. 10. Choucino C, Goodman SA, Greer JP et al. Nocardial infections in bone marrow transplant recipients. Clin Infect Dis 1996;23: 1012–1019. 11. Antonini G, Morino S, Fiorelli M et al. Reversal of encephalopathy during treatment with amphotericin B. J Neurol Sci 1996;144: 212–213. 12. Capparelli FJ, Diaz MF, Hlavnika A et al. Cefepime and cefixime-induced encephalopathy in a patient with normal renal function. Neurology 2005;65:1840.

Chapter 19 / Central Nervous System Infections in Cancer Patients

375

13. Fernandez-Torre JL, Martinez-Martinez M, Gonzalez-Rato J et al. Cephalosporin-induced nonconvulsive status epilepticus: clinical and electroencephalographic features. Epilepsia 2005;46:1550–1552. 14. Camacho DL, Smith JK, Castillo M. Differentiation of toxoplasmosis and lymphoma in AIDS patients by using apparent diffusion coefficients. Am J Neuroradiol 2003;24:633–637. 15. Davis LE, DeBiasi R, Goade DE et al. West Nile virus neuroinvasive disease. Ann Neurol 2006;60:286–300. 16. Saiz A, Graus F. Neurological complications of hematopoietic cell transplantation. Semin Neurol 2004;24:427–434. 17. Holmberg LA, Boeckh M, Hooper H et al. Increased incidence of cytomegalovirus disease after autologous CD34-selected peripheral blood stem cell transplantation. Blood 1999;94:4029–4035. 18. Crippa F, Holmberg L, Carter RA et al. Infectious complications after autologous CD34-selected peripheral blood stem cell transplantation. Biol Blood Marrow Transplant 2002;8:281–289. 19. Rosenfeld MR, Pruitt A. Neurologic complications of bone marrow, stem cell, and organ transplantation in patients with cancer. Semin Oncol 2006;33:352–361. 20. Openshaw H, Stuve O, Antel JP et al. Multiple sclerosis flares associated with recombinant granulocyte colony-stimulating factor. Neurology 2000;54:2147–2150. 21. Chen-Plotkin AS, Vossel KA, Samuels MA et al. Encephalopathy, stroke, and myocardial infarction with DMSO use in stem cell transplantation. Neurology 2007;68:859–861. 22. Iwamoto M, Jernigan DB, Gusash A et al. Transmission of West Nile virus from an organ donor to four transplant recipients. N Engl J Med 2003;348:2196–2203. 23. Srinivasan A, Burton EC, Kuehnert MJ et al. Transmission of rabies virus from an organ donor to four transplant recipients. N Engl J Med 2005;352:1103–1111. 24. Hollander H, Schaefer PW, Hedley-Whyte TE. Case records of the Massachusetts General Hospital. Case 22–2005: An 81-year-old man with cough, fever, and altered mental status. N Engl M Med 2005;353:287–295. 25. Solomon T, Fisher AF, Beasley DW et al. Natural and nosocomial infection in a patient with West Nile encephalitis and extrapyramidal movement disorders. Clin Infect Dis 2003;E140–E145. 26. Fischer SA, Graham MB, Kuehnert MJ et al. Transmission of lymphocytic choriomeningitis virus by organ transplantation. N Engl J Med 2006;354:2235–2249. 27. Spitzer TR. Engraftment syndrome following hematopoietic stem cell transplantation. Bone Marrow Transplant 2001;27:893–898. 28. Singh N, Paterson DL. Encephalitis caused by human herpes virus-6 in transplant recipients: relevance of a novel neurotropic virus. Transplantation 2000;69:2474–2479. 29. Wainwright MS, Martin PL, Morse RP et al. Human herpes virus-6 limbic encephalitis after stem cell transplantation. Ann Neurol 2001;50:612–619. 30. MacLean HJ, Douen AG. Severe amnesia associated with human herpes virus-6 encephalitis after bone marrow transplantation. Transplantation 2002;73:1086–1089. 31. Lau YL, Peiris M, Chan GC et al. Primary human herpes virus-6 infection transmitted from donor to recipient through bone marrow infusion. Bone Marrow Transplant 1998;21:1063–1066. 32. Ogata M, Kikuchi H, Satou T et al. Human herpes virus-6 DNA in plasma after allogeneic stem cell transplantation: incidence and clinical significance. J Infect Dis 2006;193:68–79. 33. Deborska D, Durlik M, Sadowska A et al. Human herpes virus-6 in renal transplant recipients: potential risk factors for the development of human herpes virus-6 seroconversion. Transplant Proc 2003;35:2199–2201 34. Thieben M, Lennon V, Boeve B et al. Potentially reversible autoimmune limbic encephalitis with neuronal potassium channel antibody Neurology 2004;62:1177–1182. 35. Vincent A, Buckley C, Schott JM et al. Potassium channel antibody-associated encephalopathy: a potentially immunotherapy-responsive form of limbic encephalitis. Brain 2004;127:701–712. 36. Gultekin SH, Rosenfeld MR Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain 2000;123:1481–1494. 37. Hashimoto K, Yasukawa M, Tohyama M. Human herpes virus-6 and drug allergy. Curr Opin Allergy Clin Immunol 2003;3:255–260. 38. Fujino Y, Nakajima M, Inoue H et al. Human herpes virus-6 encephalitis associated with hypersensitivity syndrome. Ann Neurol 2002;51:771–774. 39. Zerr DM. Human herpes virus-6 and central nervous system disease in hematopoietic cell transplantation. J Cli Virol 2006;37 Suppl 1:S52–S56. 40. Zerr DM, Corey L, Kim HW et al. Clinical outcomes of human herpes virus-6 reactivation after hematopoietic cell transplantation. Clin Infect Dis 2005;40:932–940. 41. Yoshikawa T, Yoshida J, Hamaguchi M et al. Human herpes virus-7–associated meningitis and optic neuritis in a patient after allogeneic stem cell transplantation. J Med Virol 2003;70:440–443. 42. Ward KN, White RP, Mackinnon S et al. Human herpes virus-7 infection of the CNS with acute myelitis in an adult bone marrow recipient. Bone Marrow Transplant 2002;30:983–985. 43. Husain S, Munoz P, Forrest G et al. Infections due to Scedosporium apiospermum and Scedosporium prolificans in transplant recipients: clinical characteristics and impact of antifungal agent therapy on outcome. Clin Infect Dis 2005; 40:89–99. 44. Ma M, Barnes G, Oulliam J et al. CNS angiitis in graft-vs.-host disease. Neurology 2002;59:1994–1997. 45. Solaro C, Murialdo A, Giunti D et al. Central and peripheral nervous system complications following allogeneic bone marrow transplantation. Eur J Neruol 2001;8:77–80. 46. Mueller-Mang C, Mang TG, Kalhs P et al. Imaging characteristics of toxoplasmosis encephalitis after bone marrow transplantation: report of two cases and review of the literature. Neuroradiology 2006;48:84–89.

376

Part VI / Complications of Cancer Therapy

47. Braddy CM, Heilman RL, Blair JE. Coccidioidomycosis after renal transplantation in an endemic area. Am J Transplant 2006;6: 340–345. 48. Neill TA, Lineberry K, Nabors LB. Incidence of post-transplant lymphoproliferative disorder isolated to the central nervous system in renal transplant patients. Neurology 2004;62(suppl 5)A479. 49. Voog E, Morschhauser F, Solal-Celigny P. Neutropenia in patients treated with rituximab. N Engl J Med 2003;348:2691–2694. 50. Kimby E Tolerability and safety of rituximab (MabThera). Cancer Treat Rev 2005;31:456–473. 51. Matteucci P, Magni M, Di Nicola M et al. Leukoencephalopathy and papovavirus infection after treatment with chemotherapy and anti-CD20 monoclonal antibody. Blood 2002;100:1104–1105. 52. American College of Rheumatology Hotline. Rituximab and Progressive Multifocal Leukoencephalopathy. January 2, 2007. 53. Coiffier B, LePage E, Briere J et al. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 2002;346:235–242. 54. Kaplan LD, Lee JY, Ambinder RF et al. Rituximab does not improve clinical outcome in a randomized phase 3 trial of CHOP with or without rituximab in patients with HIV-associated non-Hodgkin lymphoma: AIDS Malignancies Consortium Trial 010. Blood 2005;106:1538–1543. 55. Peleg AY, Husain S, Kwak EJ et al. Opportunistic infections in 547 organ transplant recipients receiving alemtuzumab, a humanized monoclonal CD–52 antibody. Clin Infec Dis 2007;44:204–212. 56. Nath DS, Kandaswamy R, Gruessner R et al. Fungal infections in transplant recipients receiving alemtuzumab. Transplant Proc 2005;37:934–936. 57. Shelburne SA III, Hamill RJ. The immune reconstitution inflammatory syndrome. AIDS Rev 2003;5:67–79. 58. King MD, Perlino CA, Cinnamon J et al. Paradoxical recurrent meningitis following therapy of cryptococcal meningitis: an immune reconstitution syndrome after initiation of highly active antiretroviral therapy. Int J STD AIDS 2002;13:724–726. 59. Domingo P, Torres OH, Ris J et al. Herpes zoster as an immune reconstitution disease after initiation of combination antiretroviral therapy in patients with human immunodeficiency virus type-1 infection. Am J Med 2001;110:605–609. 60. Gilden DH, Kleinschmidt-DeMasters BK, LaGuardia JJ et al. Neurologic complications of the reactivation of varicella-zoster virus. N Engl J Med 2000;342:635–645. 61. Jenny-Avital ER, Abadi M. Immune reconstitution cryptococcosis after initiation of successful highly active antiretroviral therapy. Clin Infect Dis 2002;35:e128–e133. 62. Venkataramana A, Pardo CA, McArthur JC et al. Immune reconstitution inflammatory syndrome in the CNS of HIV-infected patients. Neurology 2006;67:383–388. 63. Buckanovich RJ, Liu G Stricker C et al Nonmyeloablative allogeneic stem cell transplantation for refractory Hodgkin’s lymphoma complicated by interleukin-2 responsive progressive multifocal leukoencephalopathy. Ann Hematol 2002;81:410–413. 64. Powles T, Thirlwell C, Nelson M et al. Immune reconstitution inflammatory syndrome mimicking relapse of AIDS related lymphoma in patients with HIV-1 infection. Leuk Lymphoma 2003;44:1417–1419. 65. Piliero PJ, Fish DG, Preston S et al. Guillain–Barré syndrome associated with immune reconstitution. Clin Infect Dis 2003;36: e111–e114. 66. Govindarajan R, Adusumilli J, Baxter DL. Reversible posterior leukoencephalopathy syndrome induced by RAF kinase inhibitor BAY 43–9006. J Clin Oncol 2006;24:e48. 67. Allen JA, Adlakha A, Bergethon PR. Reversible posterior leukoencephalopathy syndrome after bevacizumab/FOLFIRI regimen for metastatic colon cancer. Arch Neurol 2006;63:1475–1478. 68. Soysal DD, Caliskan M, Aydin K et al. Isolated cerebellar involvement in a case of posterior reversible leukoencephalopathy. Clin Radiol 2006;61:983–986. 69. Kitaguchi H, Tomimoto H, Miki Y et al. A brainstem variant of reversible posterior leukoencephalopathy syndrome. Neuroradiology 2005;47:652–656. 70. Tse S, Saunders EF, Silverman E et al. Myasthenia gravis and polymyositis as manifestations of chronic graft-vs.-host disease. Bone Marrow Transplant 1999;23:397–399. 71. Campbell JN, Morris PP. Cerebral vasculitis in graft-vs.-host disease: a case report. Am J Neuroradiol 2005;26:654–656. 72. Sommers LM, Hawkins DS. Meningitis in pediatric cancer patients: a review of forty cases from a single institution. Pediatr Infect Dis J 1999;18:902–907. 73. McGovern PC, Lautenbach E, Brennan PJ et al. Risk factors for postcraniotomy surgical site infection after 1,3-bis(2-chloroethyl)-1nitrosourea (Gliadel) wafer placement. Clin Infect Dis 2003;36:759–765. 74. Riel-Romero RM, Baumann RJ. Herpes simplex encephalitis and radiotherapy. Pediatr Neurol 2003;29:69–71. 75. Kimberlin DW, Whitley RJ. Varicella-zoster vaccine for the prevention of herpes zoster. N Engl J Med 2007;356:1338–1343. 76. Law JK Ho JK Hoskins PJ et al. Fatal reactivation of hepatitis B post-chemotherapy for lymphoma in a hepatitis B surface antigennegative, hepatitis B core antibody-positive patient: potential implications for future prophylaxis recommendations. Leuk Lymphoma 2005;46:1085–1089. 77. Chheda MG, Drappatz J, Greenberger NJ et al. Hepatitis B reactivation during glioblastoma treatment with temozolomide: a cautionary note. Neurology 2007;68:955–956. 78. Krasner A. Glucocorticoid-induced adrenal insufficiency. JAMA 1999;282:671–676. 79. Coursin D, Wood K. Corticosteroid supplementation for adrenal insufficiency. JAMA 2002;287:236–240. 80. Chabolla DR, Wszolek ZK. Pharmacologic management of seizures in organ transplant. Neurology 2006;67(Suppl 4):S34–S38. 81. Cohen LF, Dunbar SA. Streptococcus bovis infection of the central nervous system: report of two cases and review. Clin Infect Dis 1997;25:819–823. 82. Bartt R. Listeria and atypical presentations of Listeria in the central nervous system. Semin Neurol 200;20:361–373. 83. Cunha BA, Filozov A, Reme P. Listeria monocytogenes encephalitis mimicking West Nile encephalitis. Heart Lung 2004;33:61–64.

Chapter 19 / Central Nervous System Infections in Cancer Patients

377

84. Mileshkin L, Michael M. CNS listeriosis confused with leptomeningeal carcinomatosis in a patient with a malignant insulinoma. Am J Clin Oncol 2002;25:576–579. 85. Van de Beek D, de Gans J, McIntyre P et al. Steroids in adults with acute bacterial meningitis: a systematic review. Lancet Infect Dis 2004;4:139–143. 86. De Gans J, Van de Beek D. Dexamethasone in adults with bacterial meningitis. N Engl J Med 2002;347:1549–1556. 87. Chaudhuri A. Adjunctive dexamethasone treatment in acute bacterial meningitis. Lancet Neurol 2004;3:54–62. 88. Gavalda J, Len O, San Juan R et al. Risk factors for invasive aspergillosis in solid-organ transplant recipients: a case-control study. Clin Infect Dis 2005;41:52–59. 89. Guermazi A, Gluckman E, Tabti B et al. Invasive central nervous system aspergillosis in bone marrow transplantation recipients: an overview. Eur Radiol 2003;13:377–388. 90. Pfeiffer CD, Fine JP, Safdar N et al. Diagnosis of invasive aspergillosis using a galactomannan assay: a meta-analysis. Clin Infect Dis 2006;42:1417–1427. 91. Denning DW. Invasive aspergillosis. Clin Infect Dis 1998;26:781–805. 92. Mathisen GE, Johnson JP. Brain abscess. Clin Infect Dis 1997;25:763–781. 93. Karakousis PC, Magill SS, Gupta A. Paraplegia caused by invasive spinal aspergillosis. Neurology 2007;68:158. 94. Singh N, Paterson D. Aspergillus infections in transplant recipients. Clin Microbiol Rev 2005;18:44–69. 95. Kami M, Ogawa S, Kanda Y et al Early diagnosis of central nervous system aspergillosis using polymerase chain reaction, latex agglutination test and enzyme-linked immunosorbent assay. Br J Haematol 1999;106:536–537. 96. Herbrecht R, Denning DW, Patterson TG et al. Voriconazole versus amphotericin B for primary therapy of invasive aspergillosis. N Engl J Med 2002;347:408–415. 97. Kartsonis NA, Saah AJ, Joy Lipka C et al. Salvage therapy with caspofungin for invasive aspergillosis: results form the caspofungin compassionate use study. J Infection 2005;50:196–205. 98. Singh N, Limaye AP, Forrest G et al. Combination of voriconazole and caspofungin as primary therapy for invasive aspergillosis in solid organ transplant recipients: a prospective, multicenter, observational study. Transplantation 2006;81:320–326. 99. Munoz P, Singh N, Bouza E. Treatment of solid organ transplant patients with invasive fungal infections: should a combination of antifungal drugs be used? Curr Opin Infec Dis 2006;19:365–370. 100. Vazquez JA, Sobel JD. Anidulafungin: a novel echinocandin. Clin Infect Dis 2006;43:215–222. 101. Cornely OA, Maertens J, Winston DJ et al. Posaconazole vs. fluconazole or itraconazole prophylaxis in patients with neutropenia. N Engl J Med 2007;356:348–359. 102. Ullmann AJ, Lipton JH, Vesole DH et al. Posaconazole or fluconazole for prophylaxis in severe graft-versus-host disease. N Engl J Med 2007;356:335–347. 103. De Pauw BE, Donnelly JP. Prophylaxis and aspergillosis: has the principle been proven? N Engl J Med 2007;356:409–411. 104. Walsh TJ , Teppler H, Donowitz GR et al. Caspofungin versus liposomal amphotericn B for empirical antifungal therapy in patients with persistent fever and neutropenia. N Engl J Med 2004;351:1391–1402. 105. Marty FM, Lowry CM, Cutler CS et al. Voriconazole and sirolimus coadministration after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2006;12:552–559. 106. Hajjeh RA, Sofair AN, Harrison LH et al. Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program. J Clin Microbiol 2004;42:1519–1527. 107. Verduyn-Lunel FM, Voss A, Kuijper EJ. Detection of the Candida antigen mannan in cerebrospinal fluid specimens from patients suspected of having Candida meningitis. J Clin Microbiol 2004;42:867–870. 108. Perfect JR, Marr KA, Walsh TJ et al. Voriconazole treatment for less-common, emerging, or refractory fungal infections. Clin Infect Dis 2003;36:1112–1131. 109. Mattiuzzi G, Giles FJ. Management of intracranial fungal infections in patients with haematological malignancies. Br J Haematol 2005;131:287–300. 110. Rex JH, Larsen RA, Dismukes WE et al. Catastrophic visual loss due to Cryptococcus neoformans meningitis. Medicine (Baltimore) 1993;72:207–224. 111. Kauffman CA. Zygomycosis: reemergence of an old pathogen. Clin Infect Dis 2004;39:588–590. 112. Greenberg RN, Scott LJ, Vaughn HH et al. Zygomycosis (mucormycosis): emerging clinical importance and new treatments. Curr Opin Infect Dis 2004;17:517–525. 113. Antonini G, Ceschin V, Morino S et al. Early neurologic complications following allogeneic bone marrow transplant for leukemia: a prospective study. Neurology 1998;50:1441–1445. 114. Maschke M, Dietrich U, Prumbaum M et al. Opportunistic CNS infection after bone marrow transplantation. Bone Marrow Transplant 1999;23:1167–1176. 115. Pruitt AA. Central nervous system infections in cancer patients. Semin Neurol 2004;24:435–452. 116. Rosenfeld MR, Pruitt AA. Neurologic complications of bone marrow, stemcell and organ transplantation in patients with cancer. Sem Oncol 2006;33:352–361. 117. Dworkin RH, Johnson RW, Breuer J et al Recommendations for the management of herpes zoster. Clin Infect Dis 2007;44:S1–S26 118. Dubinsky RM, Kabbani AH, El-Chami Z, et al. Practice parameter: treatment of post-herpetic neuralgia: an evidence-based report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2004;63:959–965. 119. Quan D, Hammack BN, Kittelson J et al. Improvement of post-herpetic neuralgia after treatment with intravenous acyclovir followed by oral valacyclovir. Arch Neurol 2006;63:940–942. 120. Mizock BA, Bartt R, Agbemazdo B. Herpes zoster oticus with pontine lesion: segmental brain-stem encephalitis. Clin Infect Dis 2000;30:229–230.

378

Part VI / Complications of Cancer Therapy

121. Fox RJ, Galetta SL, Mahalingam R et al. Acute, chronic, and recurrent varicella zoster virus neuropathy without zoster rash. Neurology 2001;57:351–354. 122. De Silva SM, Mark AS, Gilden DH et al. Zoster myelitis: improvement with antiviral therapy in two cases. Neurology 1996;47:929–931 . 123. Gilden DH, Beinlich BR, Rubinstien EM et al Varicella-zoster virus myelitis: an expanding spectrum. Neurology 1994;44:1818–1823 124. Weaver S, Rosenblum MK, DeAngelis LM. Herpes varicella-zoster encephalitis in immunocompromised patients. Neurology 1999;52:193–195. 125. Gilden DH, Lipton H, Wolf J et al. Two patients with unusual forms of varicella-zoster virus vasculopathy. N Engl J Med 2002;347:1500–1503. 126. Koralnik IJ, Schellingerhout D, Frosch MP. Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Case 14–2004. A 66-year-old man with progressive neurologic deficits. N Engl J Med 2004;350:1882–1893. 127. Garcia-Suarez J, de Miguel D, Krsnik I et al. Changes in the natural history of progressive multifocal leukoencephalopathy in HIV-negative lymphoproliferative disorders: impact of novel therapies. Am J Hematol 2005;80:271–281. 128. Kastrup O, Maschke M, Diener HC et al. Progressive multifocal leukoencephalopathy limited to the brain stem. Neuroradiology 2002;44:227–229. 129. Koralnik IJ, Wuthrich C, Dang X et al. JC virus granule cell neuronopathy: a novel clinical syndrome distinct from progressive multifocal leukoencephalopathy. Ann Neurol 2005;57:576–580. 130. Koralnik IJ. Progressive multifocal leukoencephalopathy revisited: has the disease outgrown its name? Ann Neurol 2006;60:162–173. 131. Berger JR, Koralnik IJ. Progressive multifocal leukoencephalopathy and natalizumab: unforessen consequences. N Engl J Med 2005;353:414–416. 132. Kleinschmidt-DeMasters BK,Tyler KL. Progressive multifocal leukoencephalopathy complicating treatment with natalizumab and interferon-beta-1a for multiple sclerosis. N Engl J Med 2005;353:369–374. 133. Langer-Gould A, Atlas SW, Green AJ et al. Progressive multifocal leukoencephalopathy in a patient treated with natalizumab. N Engl J Med 2005;353:375–381. 134. Marra CM, Rajicic N, Barker DE et al. A pilot study of cidofovir for progressive multifocal leukoencephalopathy in AIDS. AIDS 2002;16:1791–1797. 135. Viallard JF, Lazaro E, Lafon ME et al. Successful cidofovir therapy of progressive multifocal leukoencephalopathy preceding angioimmunoblastic T-cell lymphoma. Leuk Lymphoma 2005;46:1659–1662. 136. Vulliemoz S, Lurati-Ruiz F, Borruat FX et al. Favourable outcome of progressive multifocal leukoencephalopathy in two patients with dermatomyositis. J Neurol Neurosurg Psychiatry 2006;77:1079–1082. 137. Dodd RY. Emerging infections, transfusion safety, and epidemiology. N Engl J Med 2003;349:1205–1206. 138. Chen JT, Collins DL, Atkins HL et al Brain atrophy after immunoablation and stem cell transplantation in multiple sclerosis. Neurology 2006;66:1935–1937. 139. Curtis RE, Rowlings PA, Deeg HJ et al. Solid cancers after bone marrow transplantation. N Engl J Med 1997;336:897–904. 140. Hijiya N, Hudson MM, Lensing S et al. Cumulative incidence of secondary neoplasms as a first event after childhood acute lymphoblastic leukemia. JAMA 2007;297:1207–1215.

VII

Neurologic Complications of Specific Malignancies

20

Neurological Complications of Primary Brain Tumors Tracy T. Batchelor,

MD,

and Thomas N. Byrne,

MD

CONTENTS Clinical Manifestations of Primary Brain Tumors Specific Neurological Complications Conclusion References

Summary Neurological complications of primary brain tumors may occur as either direct or indirect effects of the tumor; may herald the initial diagnosis or recurrence of the tumor; or may develop as a consequence of therapy directed against the tumor. In this chapter we will begin with a discussion of the clinical manifestations of primary brain tumors. Subsequently, the pathogenesis of cerebral edema and seizures along with their therapies and the potential complications of these therapies are presented. Finally, psychiatric disorders, hydrocephalus, and leptomeningeal metastases as complications of primary brain tumors are discussed. Key Words: primary brain tumor, cerebral edema, seizure, hydrocephalus, leptomeningeal

1. CLINICAL MANIFESTATIONS OF PRIMARY BRAIN TUMORS Primary brain tumors typically present with progressive focal and/or diffuse clinical manifestations. If the lesion is in an eloquent area of brain, the initial clinical manifestation may be dysfunction attributable to that brain locus such as paresis, aphasia, or loss of vision. Alternatively, patients may present with large space-occupying lesions arising in other locations such as the frontal or temporal lobes causing diffuse cerebral symptoms. Common examples of diffuse dysfunction include cognitive or behavioral disturbance, headache, and gait disorder without focal symptoms. The headache of primary brain tumors arises from compression of innervated large intracranial blood vessels and meninges. While the classical brain tumor headache is more severe after a period of recumbency (e.g., headache upon awakening in the morning), when intracranial venous blood is increased leading to worsening of intracerebral pressure, the most common headache type to herald a primary brain tumor has insidious onset and is progressive. The location of the primary brain tumor determines the focal neurological manifestations that occur. Frontal lobe lesions may result in inattention, depression, and lack of motivation. Some patients may be considered to have a psychiatric disease because there are few, if any, focal neurological signs to alert the clinician to a neurological etiology. In some such cases, imaging may reveal an extensive, infiltrative glioma crossing the corpus callosum without involving motor pathways. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

381

382

Part VII / Neurologic Complications of Specific Malignancies

Tumors in the temporal lobes often present with neuropsychiatric manifestations such as memory impairment, hypergraphia, mood disturbance, and déjà vu. Dominant temporal lobe lesions may cause aphasia. A homonymous superior quadrantanopia may be seen on visual field testing because the visual pathways may be interrupted. Parietal lobe tumors may exhibit contralateral motor and sensory disturbances and homonymous hemianopia. Dominant parietal lobe tumors may cause dysphasia. Nondominant lesions may cause geographic agnosia, dressing apraxia, and, rarely, denial of the contralateral side and anosognosia. Occipital lobe tumors may cause a contralateral homonymous hemianopia. The corpus callosum is often eventually involved by invasion of the tumor along white matter tracts which leads to invasion of the contralateral occipital lobe. This may lead to an inability to read and an inability to name objects presented in the nondominant visual field, or cortical blindness. Primary neoplasms of the posterior fossa are more often seen in childhood and include medulloblastoma, cerebellar astrocytoma, and brainstem glioma. Brainstem gliomas often cause cranial neuropathies with symptoms such as diplopia, facial weakness, and dysarthria. Cerebellar astrocytomas are usually slow growing and may present with dizziness, nausea, headache, and ataxia. If the tumor is in the cerebellar hemisphere, ipsilateral appendicular ataxia is frequent, and if it is in the vermis, then truncal and gait ataxia are common. There are some clinical brain tumor syndromes that bear special mention. Midline tumors resulting in ventricular obstruction may cause the “3M” syndrome, namely, maximal disability, minimal signs, and midline lesion. Common clinical manifestations include diffuse headache, cognitive or behavioral changes, and gait/truncal ataxia without lateralizing findings. Tumors of the fourth ventricle are common causes of the “3M” syndrome. Another manifestation of a posterior fossa mass causing obstructive hydrocephalus is projectile vomiting, which is more common in children than adults. Posterior fossa medulloblastomas, ependymomas, and cerebellar astrocytomas may all present in this manner. A rare presentation of primary brain tumors is nausea without neurological signs and can be seen with tumors of the insula or floor of the fourth ventricle irritating the area postrema. Patients with such tumors often have extensive gastrointestinal evaluations including endoscopy only to have a glioma of the insula or ependymoma/subependymoma of the fourth ventricle found after some years. Resection of the neoplasm often aborts the gastrointestinal complaints Seizures are a common manifestation of gliomas and are discussed below. A seizure in a middle-aged person without a history of head trauma or toxic-metabolic cause should raise the suspicion of a possible underlying primary brain tumor and necessitates brain imaging (1).

2. SPECIFIC NEUROLOGICAL COMPLICATIONS 2.1. Brain Edema Brain edema is defined as an increase in brain volume due to an increase in brain water and sodium content (2,3), and may lead to focal or generalized brain dysfunction (4). There are three general categories of brain edema as outlined in Table 1. Brain edema that occurs as a consequence of a brain tumor is vasogenic in origin and results from the increased permeability of the brain tumor vasculature. Vasogenic edema has the characteristics of a plasma exudate with an increased content of sodium and protein in the extracellular fluid space (2). Vasogenic edema associated with tumors occurs primarily within the tumor and spreads secondarily into the surrounding brain (Fig. 1) (5–7). 2.1.1. Blood–Brain Barrier and Edemagenesis The pathogenesis of vasogenic brain edema involves disruption of the blood–brain barrier. This physiological barrier consisting of specialized endothelium of capillaries surrounded by pericytes normally impedes the entry of most water-soluble but not lipid-soluble agents (8). The anatomic features essential for the normal function of the blood–brain barrier are highlighted in Fig. 2 and Color Plate 8 (9). Disruption of the blood–brain barrier by a brain tumor results in the increased entry of water-soluble substances and macromolecules such as plasma protein into the tumor and surrounding brain. Because there is no lymphatic system in the CNS, these substances are not easily eliminated and are driven into surrounding brain tissue by increased hydrostatic pressure within the tumor, resulting in extracellular brain edema. The edema tends to extend along white matter tracts rather than in the more closely packed gray matter (10), and the water content can be measured by diffusion tensor magnetic resonance imaging (11).

Chapter 20 / Neurological Complications of Primary Brain Tumors

383

Table 1 Categories of Brain Edema Vasogenic Pathogenesis

Cytotoxic

Increased capillary permeability Chiefly white matter

Cellular swelling, glial, neuronal, endothelial Gray and white matter

Plasma filtrate, including plasma proteins Increased Increased

Increased intracellular water and sodium Decreased Normal

Beneficial in brain tumor and abscess

Not effective (possibly Reye’s syndrome)

Osmotherapy

Acutely reduces volume of normal brain tissue only Improves compliance

Acetazolamide Furosemide

? Effect ? Effect

Acutely reduces brain volume in hypo-osmolality Improves compliance No direct effect No direct effect

Location of edema Edema fluid composition Extracellular fluid volume Capillary permeability to large molecules (inulin) Therapeutic effects Steroids

Hydrocephalic Increased brain fluid due to block of CSF absorption Chiefly periventricular white matter CSF Increased Normal

Uncertain effectiveness (possibly pseudotumor and meningitis) Rarely useful Improves compliance Minor usefulness Minor usefulness

CSF = cerebrospinal fluid. Adapted from Fishman RA: Cerebrospinal Fluid in Diseases of the Nervous System, 2nd ed. Philadelphia: W.B. Saunders, 1992; 122.

The mechanisms of blood–brain barrier disruption leading to brain edema are incompletely understood, and multiple factors are involved. When brain tumors achieve a size greater than 1–2 mm, new blood vessels are formed that lack a physiological blood–brain barrier. Tumor endothelium reveals an increased number of intercellular junctions, discontinuous tight junctions, membranous fenestrations, noncontiguous basement membranes, increased vessel diameter, reduced pericyte coverage, and active micropinocytosis. The net result of these abnormalities is an increased permeability to macromolecules, ions, and proteins, with formation of vasogenic edema. All of these microvascular abnormalities are most common in the central portion of the tumor and least common at the interface between brain and tumor (12–15). The neoplastic blood vessels of glioblastoma, the most common malignant glioma, express both VEGF and PDGF receptors (16). Expression of these molecules is associated with the development of vasogenic brain edema. Arachidonic acid and its leukotriene metabolites also promote vasogenic edema (17–20). Finally, macrophage infiltration in and around the tumor are capable of elaborating a variety of secretory factors that are associated with increased vascular permeability (3). One study demonstrated that the amount of tumor-associated edema visible on CT scans correlated with the extent of macrophage infiltration observed on pathological study (21). The relative contribution of each of these mechanisms to edemagenesis is unknown. Blood–brain barrier disruption and the subsequent development of vasogenic edema eventually lead to increased intracranial pressure (ICP). Because the increased intracranial pressure is distributed unevenly within the intracranial compartment it may result in brain herniation and compromise of local blood supply (22). These events contribute to the neurological symptoms and signs observed in patients with brain tumors. 2.1.2. Steroid Therapy Glucocorticoids are the mainstay of treatment for vasogenic brain edema. Kofman and colleagues (23) were the first to demonstrate the responsiveness of brain tumor edema to glucocorticoids. Subsequently, Galicich and French (24) introduced dexamethasone therapy as the standard treatment for tumor-associated edema. Glucocorticoids

384

Part VII / Neurologic Complications of Specific Malignancies

Fig. 1. The top panel demonstrates a right frontal nonenhancing glioma with increased T2 (left) and FLAIR (right) signal abnormality. It can be difficult distinguishing vasogenic edema from infiltrative tumor in a case such as this. The bottom panel demonstrates a contrast-enhancing left frontal metastasis (left) with an extensive amount of surrounding vasogenic edema shown as bright signal on the FLAIR sequence (right).

reduce the filtration of plasma-derived fluid across tumor capillaries (25) and reduce the movement of albumin through the extracellular space by solvent drag (26). While the mechanism of action of glucocorticoids to reduce vasogenic brain edema is incompletely understood, it is commonly thought that their beneficial effects is a direct action on endothelial cell function that restores normal vascular permeability (27,28). One possible mechanism involves the ability of glucocorticoids to inhibit the release of arachidonic acid (10,27). Finally, there is indirect evidence that dexamethasone causes cerebral vasoconstriction (29). All of these mechanisms likely contribute to the ability of steroids to stabilize the blood–brain barrier and lead to resolution of vasogenic brain edema. Using diffusion tensor MR imaging, Sinha and colleagues have shown that administration of corticosteroids decreases peritumoral extracellular water content in edematous brain but does not affect the water content of contralateral normal brain (30). The optimal dose of steroids for vasogenic brain edema has not been established. Dexamethasone is commonly used in clinical practice and is probably the best glucocorticoid for the treatment of vasogenic cerebral edema associated with primary brain tumors. The advantages of dexamethasone include the absence of mineralocorticoid effect, which makes salt retention and systemic edema less likely. Furthermore, there is evidence that of all the steroids, dexamethasone is less likely to be associated with infection and cognitive impairment (31). Alternatively, fluorinated steroids such as dexamethasone are more likely to cause myopathy (32). Because steroids cause adverse side effects and induce metabolism of other drugs, the benefits of steroids in patients with asymptomatic brain edema are generally outweighed by these side effects. The patient should be maintained on the lowest dose that controls neurological symptoms in an effort to avoid the development

Chapter 20 / Neurological Complications of Primary Brain Tumors

385

Fig. 2. Normal blood–brain barrier demonstrating tight junctions between endothelial cells; a normal basement membrane and adjacent astrocyte foot processes. (Adapted with permission.) (see Color Plate 8).

of adverse effects. The drug is absorbed from the gastrointestinal tract, but first-pass hepatic metabolism may decrease effectiveness, especially in patients also receiving phenytoin (33). Although many clinicians prescribe dexamethasone four times daily, its half-life permits twice-daily dosing (34). The usual starting schedule is a 10-mg oral dose followed by 4 mg four times daily or 8 mg twice daily, which is the bioequivalent of 20 times the rate of endogenous cortisol production. There is some evidence that doses lower than 16 mg daily may be equally effective with the same degree of clinical improvement after one week of treatment (34). A rapid clinical response to steroid administration indicates that the symptoms may be primarily due to the tumor-associated edema rather than to actual tumor mass (35). Symptoms and signs of generalized brain dysfunction such as headache and lethargy respond more rapidly and dramatically to corticosteroids than focal neurological signs. If the standard dose fails to achieve a clinical response in 48 hrs, then the dose can be increased. Up to 100 mg of dexamethasone over 24 hrs may be required in some patients (36,37). In patients with primary brain tumors, symptomatic improvement usually begins within hours of administration of corticosteroids. Positron emission tomography (PET) scans of humans with brain tumors demonstrate an effect on the blood–brain barrier as soon as 6 hrs after the intravenous bolus (38). Maximal clinical improvement usually occurs within 36–72 hrs (39). It has been shown that the first change is a decrease in plateau waves followed by a gradual decline of increased intracranial pressure over a period of 48–72 hrs. Improvement on CT and MR imaging studies may lag behind clinical improvement, although scans may show decreased contrast enhancement, suggesting partial restoration of the blood-brain barrier (40–42). In order to avoid the deleterious effects of steroids, the patient should be treated with the smallest effective dose for the shortest time possible (43). While the approach needs to be tailored to the individual needs of the patient, some general guidelines can be offered. An attempt to taper the corticosteroids should be made during

386

Part VII / Neurologic Complications of Specific Malignancies

or after more definitive treatment such as surgery, radiation therapy, or chemotherapy. If possible, the taper should generally start 3–4 days after surgery or during the second week of radiation therapy. Ordinarily, patients suffering from significant mass effect should receive standard dose steroids for 48–72 hrs prior to initiating brain irradiation to reduce intracranial pressure and minimize acute radiation toxicity. For patients on a standard dose of 16 mg of dexamethasone per day, decreasing the dose by 2–4 mg each week may be feasible. If symptoms of steroid withdrawal or increased brain edema occur, the patient may benefit from increasing the dose to the immediately preceding level for 4–8 days before starting the taper again. A more rapid taper may be used safely to minimize steroid toxicity, decreasing the dose from 16 mg/day for 4 days to 8 mg/day for 4 days followed by 4 mg/day until completion of radiation therapy (43). In patients on steroids for many months who fail the usual taper schedule, or in patients with a large amount of residual tumor, the drug is tapered more slowly (e.g., 1–2 mg/week) to the lowest dose possible. Patients on large doses of steroids (e.g, 100 mg of dexamethasone per day) who have stabilized and are receiving definitive treatment may tolerate halving the dose every 4–5 days depending on the clinical condition. 2.1.2.1. Adverse Steroid Effects. There are several well-known adverse effects of corticosteroids, some of which bear discussion. These include potentially serious side effects such as myopathy cognitive impairment, gastrointestinal dysfunction, opportunistic infection and osteoporosis (10). Many patients treated with conventional doses of steroid develop myopathy. In one study, among 15 cancer patients being treated with dexamethasone, 9 developed some form of myopathy within 15 days. The cumulative dose was high, ranging from 186 to 1846 mg, and the myopathy correlated with the total dose rather than the duration or daily dose of dexamethasone. The mechanism of steroid-induced myopathy is uncertain but may involve inhibition of protein synthesis (44). Muscle biopsy shows atrophy of type II fibers, the fibers characterized by high glycolytic and low oxidative capacity (45). Serum muscle enzymes are not elevated. As with other forms of myopathy the proximal muscle groups are affected most severely. One of the most common complaints is an inability to arise from a chair. Treatment includes reduction or discontinuation of steroids, if possible. Because fluorinated steroids (dexamethasone, triamcinolone) are associated with more type IIb fiber atrophy than nonfluorinated steroids (prednisolone, methylprednisolone) avoidance of the former may lower the risk of steroid myopathy, although this has not been studied systematically. Preclinical experimental observations have suggested that alternate-day dosing of methylprednisolone reduces the severity of myopathy compared to continuous daily dosing of the same drug (46). Exercise, physiotherapy, and a high-protein diet during steroid treatment may attenuate the disorder (47). One study reported a decreased frequency of myopathy in tumor patients treated with both steroids and phenytoin which may have been due to increased catabolism of the steroid after induction of the hepatic microsomal system by phenytoin (48,49). Neuropsychiatric complications may develop in as many as 5% of patients receiving exogenous steroids. Anxiety, mood, and sleep disorders are the most common manifestations. Occasionally, psychotic reactions occur (50,51). Most patients with psychiatric complications make a full recovery with reduction of the dose, but symptomatic treatment may be necessary. If necessary, neuroleptic administration is usually effective for psychotic symptoms (50). One report suggests that lithium prophylaxis lessens the likelihood of a psychotic reaction to steroids, although it is not routinely recommended (52). Steroid withdrawal can also lead to depressive symptoms (53). Corticosteroids may also cause acute memory impairment, possibly through inhibition of blood flow to the medial temporal lobe (54). Healthy subjects were injected with stress doses of cortisone (25 mg) and underwent declarative memory testing and blood flow measurements using cranial PET studies. The cortisone induced a significant reduction in blood flow to the right posterior medial temporal lobe, a region associated with successful verbal retrieval. The cortisone also significantly impaired word recall in these subjects. The authors concluded that the impaired recall could be due to the impaired blood flow. Although upper gastrointestinal bleeding is infrequent in patients taking steroids with no previous history of such bleeding, ulceration and perforation remain feared complications of corticosteroids (55,56). The incidence is much higher with simultaneous use of nonsteroidal anti-inflammatory agents, anticoagulants, or in patients with a prior history of upper gastrointestinal bleeding, although the overall risk in one study was less than 1% for such patients if treated with steroids for less than 1 month (57).

Chapter 20 / Neurological Complications of Primary Brain Tumors

387

Nielsen et al. found a relative risk of hospitalizations due to gastrointestinal bleeding of 4.9 in a study of nearly 46,000 patients using steroids. However, the relative risk fell to 2.9 among patients using steroids alone, without the use of other drugs known to cause gastrointestinal bleeding (e.g., aspirin) (58). The efficacy of gastrointestinal prophylaxis with antacids, H2 -blockers, proton pump inhibitors, or other anti-ulcer agents in these patients remains controversial (59). Bowel perforation is a serious complication in steroid-treated patients. It usually occurs in patients treated with high doses of steroids who have been constipated as a result of medication, immobility, or neurological dysfunction. The perforation usually affects the sigmoid colon and may not be accompanied by the usual abdominal symptoms and signs due to the masking effects of steroids or co-morbid neurological disease (60). Plain radiographs usually are diagnostic, and surgical repair remains the definitive treatment. Attention and treatment of constipation, including adequate bulk in the diet, hydration, stool softeners, and laxatives as necessary may prevent this complication. Steroids are immunosuppressive drugs. In one study, 24% of primary brain tumor patients receiving concurrent steroids and radiation experienced a reduction in their CD4+ cell count to < 200 cells/mm3 (61). Opportunistic infections secondary to immunosuppression from steroids include Candida mucositis and esophagitis as well as Pneumocystis pneumonia (62). The rate of Pneumocystis pneumonia in patients with brain tumors is increasing, and studies have demonstrated incidence rates of 1–6% for this group of patients. Most of these patients were also receiving steroids for prolonged periods, and infection was most likely to occur during the steroid taper (63,64). Trimethoprim-sulfamethoxazole given as one double-strength tablet daily or one double-strength tablet for 3 days each week during steroid administration and for one additional month afterward is often used as prophylaxis for Pneumocystis pneumonia and should be considered in patients with brain tumors likely to require prolonged steroid treatment (62,63). Osteoporosis is a common complication of prolonged steroid use. While most patients with brain tumors do not live long enough for this to lead to fractures, it is important to recognize that steroid-induced osteoporosis can be prevented and treatments instituted, as appropriate, to the individual patient (65,66). The American College of Rheumatology has recommended that supplementation with calcium and vitamin D, should be offered to all patients receiving glucocorticoids, to restore normal calcium balance (67). There are several studies that demonstrate the efficacy of biphosphonates, such as alendronate or risedronate, in the prevention of osteoporosis in patients taking chronic steroids (68,69). Another bone complication of steroid use is avascular necrosis of the hip or other bones, which may develop following prolonged use of steroids or may occur after only a few weeks of therapy (70). Other possible side effects include hyperglycemia, which occurred in 19% of neuro-oncology patients receiving steroids in one survey (43). The majority of such patients required insulin or modification of the steroid dose to control hyperglycemia in another study. Transient anogenital burning or tingling may occur with rapid intravenous administration of dexamethasone and can be distressing if the patient is not warned of such a possibility (71). Hiccups, nocturia, and diminished sense of smell and taste have also been reported as complications of steroids (72). Another potential hazard of steroid use is the development of a withdrawal syndrome during the steroid taper. Steroid pseudorheumatism is the most common withdrawal syndrome and is heralded by the onset of diffuse arthralgias and myalgias mimicking rheumatoid arthritis. These symptoms may be debilitating, but there are usually few physical findings. Either reintroduction of low-dose steroids followed by a slower taper or treatment with aspirin or other nonsteroidal anti-inflammatory agents may result in improvement (10). Amatruda et al. (73) described a steroid withdrawal syndrome which may include lethargy, headache, dizziness, anorexia, and nausea. These symptoms may confuse the clinician by suggesting worsening of the underlying brain tumor. Because dexamethasone results in the potent induction of specific cytochrome P450 (CYP450) isozymes (CYP3A4; CYP2C8; CYP2C9), there is the potential for significant drug interactions with other agents metabolized by this system. Importantly, dexamethasone has been reported to enhance the metabolism several drugs including phenytoin (74). 2.1.3. Anti-VEGF Therapy Treatment of glioblastoma patients with drugs that inhibit VEGF-mediated pathways led to the observation that these drugs reduce vascular permeability and vasogenic brain edema (16). There is preclinical and clinical evidence that the use of anti-VEGF therapy (cediranib, bevacizumab) transiently “normalizes” the tumor vasculature and

388

Part VII / Neurologic Complications of Specific Malignancies

reduces vascular permeability (75–77). In one study the effect on permeability was observed within 24 hrs after a single dose of the drug. The anti-edema effects of cediranib lasted for up to 6 months and were associated with a steroid-sparing effect in this population of glioblastoma patients. These observations require further assessment but raise the possibility that one or more of these drugs may demonstrate clinical utility as a treatment of vasogenic brain edema. 2.1.4. Emergency Therapy of Brain Edema and Increased Intracranial Pressure 2.1.4.1. Hyperventilation. Immediate treatment of brain edema and increased intracranial pressure (ICP) may occasionally be necessary to prevent death or cerebral herniation. The methods available for such treatment are outlined in Table 2. Hyperventilation is the most rapidly effective technique available for decreasing ICP. Hyperventilation decreases the partial pressure of carbon dioxide (pCO2 ), which causes cerebral vasoconstriction in undamaged areas of the brain and a consequent decrease in cerebral blood volume and ICP. Intracranial pressure decreases within 30 sec of lowering pCO2 and remains low for 15–20 min but usually returns to the original level by 1 hr (78). Usually, the patient is intubated and is ventilated to decrease the pCO2 to 25–30 mm Hg. The patient must be monitored carefully because mechanical ventilation may occasionally increase ICP, and patients with brain lesions are especially susceptible to this effect. 2.1.4.2. Osmotherapy. The exact mechanism by which hyperosmolar agents lower ICP remains a matter of dispute (79,80). At least part of the explanation is the ability of these agents to create an osmotic gradient between the blood and that part of the brain with an intact blood–brain barrier driving the movement of water from the extracellular space to the site of the higher osmolarity in the blood. The agent most commonly used for osmotherapy is mannitol, which is usually given as 20–25% solution in a 0.5–2.0 g/kg intravenous bolus over 10–20 min. Mannitol is effective within minutes, and the effect is sustained for several hours (81). If there is clinical worsening after initial improvement, smaller intravenous boluses of mannitol may be administered, but repeated doses of mannitol may cause a rebound increase in ICP, especially in patients with vasogenic brain edema. There is also some evidence to suggest that the combination of mannitol and a loop diuretic such as furosemide produces a more significant and sustained decline in ICP, although furosemide has no role in the chronic management of tumor-related edema (82). 2.1.4.3. High-Dose Steroid Therapy. In patients with cerebral herniation, plateau waves, or signs of increased ICP, an intravenous bolus of dexamethasone (e.g., 40–100 mg) followed by doses of 40–100 mg/day may be effective in reversing brain herniation (4). The addition of furosemide (40–120 mg intravenously) to the steroid dose may be better than steroids alone (83). Other available methods of lowering ICP such as barbiturate anesthesia and hypothermia are reviewed in detail elsewhere (84). With the above-mentioned emergency therapies, most patients herniating from the effects of a brain tumor stabilize and improve.

2.2. Hydrocephalus Patients presenting with posterior fossa or pineal region tumors often develop obstructive/noncommunicating hydrocephalus that must be alleviated prior to excision of the neoplasm. In addition, patients with leptomeningeal Table 2 Emergency Treatment of Cerebral Herniation Therapy Hyperventilation (minutes) Osmotherapy (hours) Corticosteroids (days)

Dosage or Procedure

Onset (Duration) of Action

Lower pCO2 to 25–30 mm Hg Mannitol, 0.5–2.0 g/kg (IV) over 15 min followed by 25-g booster doses (IV) as needed Dexamethasone 100 mg IV push followed by 40–100 mg every 24 hrs depending on symptoms

Seconds Minutes Hours

From Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis, 1995, p. 51. Used with permission.

Chapter 20 / Neurological Complications of Primary Brain Tumors

389

metastases may develop nonobstructive/communicating hydrocephalus secondary to diffuse involvement of the meningeal covering of the brain and impairment of CSF resorption. There is controversy over the need for permanent versus temporary ventricular drainage in such patients (85). In some cases temporary external ventricular drainage suffices, in others a ventriculoperitoneal shunt or endoscopic third ventriculostomy is needed. Among the reasons for concern over the use of a ventriculoperitoneal shunt is infection. A recent multicenter study of shunts in children inserted for a variety of causes cited a 10% frequency of infections (86). The diagnosis of shunt infection or infection in the post-operative tumor bed can be challenging. The characteristic MRI finding of pyogenic abscess is markedly hyperintense signal on DWI (87,88). However, reoperation with collection of fluid and tissue for gram stain and culture are often necessary to definitively establish the diagnosis and to determine organism drug sensitivities. Hydrocephalus may also develop in patients who develop neurotoxicity after receiving treatment including brain irradiation. These patients usually develop some clinical manifestations of normal pressure hydrocephalus. Although anecdotal responses to ventriculoperitoneal shunts occur this form of communicating hydrocephalus is usually more refractory to treatment (89,90).

2.3. Seizures 2.3.1. Epidemiology and Pathogenesis Seizures are common in patients with primary brain tumors. The frequency of epilepsy varies with the tumor type. Investigators have reported epilepsy in > 80% of patients with low-grade gliomas (91), 40% of meningioma patients (92), and 20% of patients with primary CNS lymphoma (93). In a series of patients receiving chemotherapy for supratentorial tumors (nearly all gliomas), Hildebrand et al. (94) reported that 78% of 234 patients had epilepsy. In 86% of patients with epilepsy, seizures were an early manifestation of the disease and often the presenting manifestation. In only 14% of subjects did the epilepsy begin with malignant transformation of the glioma. Seizures were clinically characterized as focal, focal with secondary generalization, and generalized. Focal seizures alone were more common late in the course of the disease. The cause of this high rate of seizures in brain tumor patients may be related to the presence of neoplastic astrocytes. While normal astrocytes take up extracellular glutamate, glioma cells have been reported to release excitotoxic levels of glutamate (95,96), which has been hypothesized as a mechanism for tissue invasion (97). Additionally, astrocytes have been implicated in the genesis of seizures through the release of glutamate (98). Furthermore, Takano et al. (97) report that the anticonvulsants valproate, phenytoin, and gabapentin decreased the calciummediated glutamate release by astrocytes and decreased experimental seizures as well. Accordingly, glutamate may mediate both tissue invasion of gliomas as well as epileptogenesis. Whether glutamate antagonists may have an antineoplastic effect as demonstrated in animal models by Takano et al. remains to be demonstrated (97). Because seizures are associated with increased cerebral blood flow, they may significantly increase ICP and potentially lead to a herniation syndrome. Furthermore, seizures in patients with primary brain tumors are more likely to result in a Todd’s paralysis (4). Status epilepticus is rare in patients with brain tumors but can occur and has an associated mortality of 6–35% (99). 2.3.2. Symptomatic Treatment Patients with brain tumors causing seizures have a high risk of recurrence of seizures; thus, treatment of these patients with antiepileptic drugs (AEDs) is warranted. The selection of a particular AED requires consideration of individual patient and drug factors as well as the other types of therapy the patient is receiving. Many of the commonly used AEDs (e.g., phenytoin, carbamazepine, oxcarbamazepine, and phenobarbital) induce the following cytochrome P450 (CYP450) enzymes: CYP3A4, CYP2C8, and CYP2C9. Thus, there is a potential for significant drug interactions in brain tumor patients who are receiving anti-neoplastic drugs metabolized by the same enzymes. In fact, reduction in the plasma levels of chemotherapeutic drugs that can be clinically significant has been observed (Table 3) (100–104). Many of the newer agents (gabapentin, lamotrigine, levetiracetam, tiagibine, topiramate, zonisamide) do not induce the CYP450 system, so these drugs are attractive options for these patients. However, there have been no randomized studies demonstrating either the superiority or the equivalency of the newer agents over the traditional ones in controlling seizures in the brain tumor patient population. When available, blood levels of AEDs should

390

Part VII / Neurologic Complications of Specific Malignancies

Table 3 Influence of Enzyme-Inducing Anti-Seizure Drugs on the Total Body Clearance of Intravenously Administered Chemotherapeutic Agents in Cancer Patients Dose(mg/m2 ) Anticancer agent Etoposide Irinotecan Paclitaxel Teniposide Topotecan Vincristine∗∗

Infusion time (h) 6.0 1.5 3.0 4.0 0.5 0.25

Total Body Clearance∗ (l/h/m2 )

–EIASD

+EIASD

–EIASD

+EIASD

Difference in Clearance (%)

320–500 112–125 240 200 2.0 2.0

320–500 411 240 200 2.0 2.0

0.80 18.8 4.76 0.78 20.8 34.1

1.42 29.7 9.75 1.92 30.6 55.5

+76.9 +58.0 +104.8 +146.2 +47.1 +62.6

∗

Mean or median values. Dose and clearance values are not normalized to body surface area. Adapted from Mrugala MM, Batchelor TT, Supko JG. Delivering anticancer drugs to brain tumors. In: Chabner BA, Longo DL, eds. Cancer Chemotherapy and Biotherapy, 4th ed. Philadelphia: Lippincott-Raven 2005: 484–501 ∗∗

be monitored, as many patients with brain tumors are also receiving other medications such as dexamethasone which may cause AED levels to fluctuate (105). Treatment of gliomas with antineoplastic therapy can reduce the prevalence of seizures. In the EORTC trial (106) of early versus delayed radiation therapy in the management of low-grade gliomas, van den Bent et al. found that at one year following diagnosis, 25% of patients who had received early radiotherapy had seizures, whereas 41% of those who had not received early radiotherapy had seizures (p = 0.03); the prevalence of seizures was the same at diagnosis. Additionally, there has been a report of a patient with an oligodendroglioma-induced epilepsy that was refractory to 14 anticonvulsants but responded to temozolomide (107). 2.3.3. Prophylactic Treatment and Adverse Effects Although many patients with primary brain tumors have seizures, as noted above, the question often arises as to whether primary brain tumor patients without seizures should be placed on prophylactic AEDs. This is a critical question because several of the AEDs such as phenytoin and carbamazepine are CYP450 enzyme-inducing agents that affect the metabolism of other drugs (Table 3). Alternatively, valproic acid may inhibit the metabolism of specific chemotherapeutic agents leading to greater toxicity (108). Although the literature and practice parameters cited below argue against the use of prophylactic AEDs, in practice, these drugs are commonly used. In a survey of physicians in one state, 53% of the neurologists and 81% of the neurosurgeons prescribed prophylactic AEDs to brain tumor patients (109). However, a meta-analysis of four randomized trials revealed no evidence of reduction in the frequency of first seizures in patients receiving prophylactic anticonvulsants (110). Other well-recognized reasons not to prescribe prophylactic AEDs are the potential adverse effects of these drugs in this population. Some of these adverse effects are unique or far more common and serious in primary brain tumor patients. Thus, brain tumor patients experience a higher frequency of anticonvulsant complications (20–40% of patients) compared to the general epilepsy population taking these drugs (110). In six studies reporting anticonvulsant side effects, 24% (5–38%) of brain tumor patients experienced adverse effects severe enough to warrant discontinuation of the drug. Rash has been reported to occur in over 20% of glioma patients receiving phenytoin and carbamazepine (111). Other anticonvulsant adverse effects include nausea or vomiting (5%); encephalopathy (5%); myelosuppression (3%); ataxia; and increased liver enzymes or gum pain (5%) (110,112). A side effect worthy of special mention in patients with brain tumors who are also receiving cranial radiation while on a decreasing dose of steroids is the Stevens–Johnson syndrome,which has been reported with phenytoin and, less frequently, carbamazepine. Finally, Klein et al. (113) reported that AEDs were associated with six times the risk of reduced psychomotor speed and attention/executive dysfunction in patients who had undergone focal radiotherapy.

Chapter 20 / Neurological Complications of Primary Brain Tumors

391

Finally, the American Academy of Neurology (AAN) has issued a practice parameter recommending that prophylactic anticonvulsants should not be routinely used in patients with newly diagnosed brain tumors (110). In addition, the AAN has also issued a guideline that it is appropriate to taper and discontinue anticonvulsants after the first post-operative week in brain tumor patients who have not had a seizure, who are medically stable and who are experiencing anticonvulsant-related side effects (110). A meta-analysis of the value of prophylactic AEDs in patients with gliomas, meningiomas, and metastases determined that there was no prophylactic benefit of these agents (114).

2.4. Psychiatric Disorders Depression is a major problem in primary brain tumor patients. In a longitudinal study of 598 malignant glioma patients, depression was diagnosed using DSM-IV criteria in 15% of the study subjects in the early post-operative period. However, 93% of these patients reported symptoms of depression during this period. At follow-up intervals of 3 months and 6 months, physicians reported depression in 22% of patients while 90% of patients still reported symptoms of depression (112). These observations suggest that depression is underdiagnosed and, hence, undertreated, in the brain tumor patient population. Of significant concern is that in this same study, subjects who were depressed had a shorter survival compared to those patients who did not report symptoms of depression. Even when physicians diagnose depression in brain tumor patients there is a reluctance to treat. Only 29.7% of newly diagnosed and 60% of followup brain tumor patients who were diagnosed as depressed by their physicians were prescribed antidepressant medications. Remarkably, only 15% of brain tumor patients who reported depressive symptoms were prescribed antidepressant medications by their health care providers. This study underscores the importance of a thorough psychiatric history in all brain tumor patients; early referral to a psychiatrist or to a support group as appropriate and the use of antidepressant medications when indicated. Attentional deficits and fatigue are common among brain tumor patients. Mulhern et al. investigated the benefit of methylphenidate in 83 children surviving acute lymphoblastic leukemia and malignant brain tumors who had a higher incidence of attention and learning problems in school than their peers (115). In this placebo-controlled trial, the investigators found that methylphenidate improved cognitive and social functioning among these patients as judged by parents and teachers. Similarly, Meyers and colleagues studied the effect of methylphenidate in patients with malignant glioma who were developing progressive neurobehavioral deficits over the course of their illness caused both by the effects of the disease and the effects of radiation and chemotherapy (116). These investigators observed significant improvements in cognitive function at the 10-mg twice-daily dose. Functional improvements included improved gait, increased stamina, and motivation to perform activities. Adverse effects were minimal. Importantly, there was no increase in seizure frequency and the majority of patients on glucocorticoid therapy were able to decrease their dose.

2.5. Leptomeningeal Metastases Patients with primary brain tumors may develop leptomeningeal dissemination or metastasis of their tumor and suffer serious morbidity and shorter survival as a consequence. Patients with leptomeningeal metastases typically present with multifocal neurological symptoms and signs affecting different parts of the neuraxis. Various combinations of cranial or spinal pain, encephalopathy, cranial nerve dysfunction, motor and sensory abnormalities, and autonomic failure are typical. The risk of leptomeningeal metastasis is a function of the type of primary brain tumor. Medulloblastoma, the most common primary brain tumor of childhood, and primitive neuroectodermal tumor (PNET) are associated with leptomeningeal dissemination (LMD) in approximately 30% of cases at the time of initial diagnosis (117). While the diagnosis can be suspected based on the clinical history and a neurological examination suggestive of dysfunction at multiple levels along the neuraxis, it is usually confirmed by contrastenhanced MRI and/or cytology. There has been considerable debate over the relative sensitivity of CSF cytology versus MRI to confirm the diagnosis. Fouladi et al. studied 106 consecutive patients with PNETs or medulloblastomas who had both CSF cytology and gadolinium-enhanced MRI of the spine performed as part of their initial staging evaluation. The

392

Part VII / Neurologic Complications of Specific Malignancies

two diagnostic methods were performed within 48 hrs of each other, 2–3 weeks post-surgery. These investigators observed that if CSF cytology had been the only diagnostic procedure used about 14% of patients with LMD would have been missed. If only MRI of the spine had been used, about 18% of patients with LMD would have been missed. Because the presence of LMD alters therapy the authors concluded that both MRI and CSF cytology should be performed in staging of such patients. Gajjar et al. also examined the question whether ventricular shunt taps were as sensitive as lumbar taps in the diagnosis of LMD in pediatric brain tumor patients. In this series, primary CNS tumors included medulloblastoma, astrocytoma, ependymoma, germinoma, atypical teratoid rhabdoid tumor, choroid plexus carcinoma, and pineoblastoma (118). The investigators observed that of 90 paired shunt and lumbar CSF samples malignant cells were detected at a significantly higher rate in lumbar CSF than in shunt CSF (p = .0018). They concluded that lumbar CSF should remain the specimen of choice for the routine cytological detection of malignant cells in the CSF of children. Another primary brain tumor that commonly causes leptomeningeal seeding is primary CNS lymphoma. Primary CNS lymphoma has been reported to show leptomeningeal seeding in 20–30% of patients at diagnosis (119,120). The diagnosis is usually confirmed by MRI with contrast or CSF cytology, flow cytometry and/or gene rearrangement studies (121,122). Patients with malignant gliomas typically do not have diagnostic CSF analysis as part of a routine evaluation. However, Wagner et al. reported that 46/270 (17%) of patients with high-grade gliomas and diffuse brainstem gliomas developed LMD (123). In a series of children with diffuse pontine glioma, Gururangen et al. found 18/96 (6%) of patients developed leptomeningeal and/or subependymal metastases (124). Leptomeningeal metastasis in patients with medulloblastoma is usually treated with the addition of chemotherapy. Individuals presenting with meningeal metastases are classified as high-risk compared to those with disease confined to the posterior fossa. Standard radiotherapy alone in this high-risk group yields a 5-year event-free survival of less than 40% (125,126). The addition of pre- or post-RT chemotherapy with a combination of agents such as vincristine, cyclophosphamide, and nitrosoureas improves survival with reduced neurological toxicity (127). Leptomeningeal lymphoma in the setting of primary CNS lymphoma is usually responsive to high-dose methotrexate as is intraparenchymal primary CNS lymphoma (120). More recently, rituximab has been reported to be effective for leptomeningeal lymphoma although this remains an experimental therapy (128–131).

3. CONCLUSION The neurological complications of primary brain tumors cause considerable morbidity and yet are often responsive to treatment. For example, vasogenic edema is responsive to corticosteroids and, recently, has been shown to be responsive to anti-VEGF agents. The appropriate selection of patients requiring antiepileptic agents and the selection of agents that do not interfere with the metabolism of chemotherapeutics is important in the neuro-oncology patient population. Psychiatric disorders and fatigue are also major complications of brain tumors, and treatment for these conditions may significantly improve the quality of life for patients. Finally, leptomeningeal metastases from primary brain tumors such as medulloblastoma and primary CNS lymphoma are not uncommon and can be treated with chemotherapy, new biological agents, and/or radiotherapy.

REFERENCES 1. Practice parameter: neuroimaging in the emergency patient presenting with seizure: summary statement. Quality Standards Subcommittee of the American Academy of Neurology in cooperation with American College of Emergency Physicians, American Association of Neurological Surgeons, and American Society of Neuroradiology. Neurology 1996;47(1):288–291. 2. Klatzo I. Presidential address: neuropathological aspects of brain edema. J Neuropathol Exp Neurol 1967;26(1):1–14. 3. Del Maestro RF, Megyesi JF, Farrell CL. Mechanisms of tumor-associated edema: a review. Can J Neurol Sci 1990;17(2):177–183. 4. Posner JB. Neurological Complications of Cancer. Philadelphia: F.A. Davis; 1995. 5. Yamada K, Ushio Y, Hayakawa T et al. Effects of methylprednisolone on peritumoral brain edema: a quantitative autoradiographic study. J Neurosurg 1983;59(4):612–619. 6. Reichman HR, Farrell CL, Del Maestro RF. Effects of steroids and nonsteroid anti-inflammatory agents on vascular permeability in a rat glioma model. J Neurosurg 1986;65(2):233–237.

Chapter 20 / Neurological Complications of Primary Brain Tumors

393

7. Shivers RR, Edmonds CL, Del Maestro RF. Microvascular permeability in induced astrocytomas and peritumor neuropil of rat brain: a high-voltage electron microscope–protein tracer study. Acta Neuropathol (Berl) 1984;64(3):192–202. 8. Rosenblum WI. Aspects of endothelial malfunction and function in cerebral microvessels. Lab Invest 1986;55(3):252–268. 9. Francis K, Van Beek J, Canova C et al. Innate immunity and brain inflammation: the key role of complement. Expert Rev Mol Med 2003;2003:1–19. 10. Weissman DE. Glucocorticoid treatment for brain metastases and epidural spinal cord compression: a review. J Clin Oncol 1988;6(3):543–551. 11. Lu S, Ahn D, Johnson G et al. Peritumoral diffusion tensor imaging of high-grade gliomas and metastatic brain tumors. AJNR Am J Neuroradiol 2003;24(5):937–941. 12. Criscuolo GR, Merrill MJ, Oldfield EH. Further characterization of malignant glioma–derived vascular permeability factor. J Neurosurg 1988;69(2):254–262. 13. Zhang RD, Price JE, Fujimaki T et al. Differential permeability of the blood–brain barrier in experimental brain metastases produced by human neoplasms implanted into nude mice. Am J Pathol 1992;141(5):1115–1124. 14. Silbergeld DL, Ali-Osman F. Isolation and characterization of microvessels from normal brain and brain tumors. J Neurooncol 1991;11(1):49–55. 15. Long DM. Capillary ultrastructure in human metastatic brain tumors. J Neurosurg 1979;51(1):53–58. 16. Batchelor TT, Sorensen AG, di Tomaso E et al. AZD2171, a pan-VEGF receptor tyrosine kinase inhibitor, normalizes tumor vasculature and alleviates edema in glioblastoma patients. Cancer Cell 2007;11(1):83–95. 17. Ohnishi T, Posner JB, Shapiro WR. Vasogenic brain edema induced by arachidonic acid: role of extracellular arachidonic acid in blood–brain barrier dysfunction. Neurosurgery 1992;30(4):545–551. 18. Baba T, Chio CC, Black KL. The effect of 5-lipoxygenase inhibition on blood–brain barrier permeability in experimental brain tumors. J Neurosurg 1992;77(3):403–406. 19. Chio CC, Baba T, Black KL. Selective blood–tumor barrier disruption by leukotrienes. J Neurosurg 1992;77(3):407–410. 20. Black KL, Hoff JT, McGillicuddy JE et al. Increased leukotriene C4 and vasogenic edema surrounding brain tumors in humans. Ann Neurol 1986;19(6):592–595. 21. Shinonaga M, Chang CC, Suzuki N et al. Immunohistological evaluation of macrophage infiltrates in brain tumors: correlation with peritumoral edema. J Neurosurg 1988;68(2):259–265. 22. Weaver DD, Winn HR, Jane JA. Differential intracranial pressure in patients with unilateral mass lesions. J Neurosurg 1982;56(5): 660–665. 23. Kofman S, Garvin JS, Nagamani D 3rd. Treatment of cerebral metastases from breast carcinoma with prednisolone. J Am Med Assoc 1957;163(16):1473–1476. 24. Galicich JH, French LA, Melby JC. Use of dexamethasone in treatment of cerebral edema associated with brain tumors. J Lancet 1961;81:46–53. 25. Nakagawa H, Groothuis DR, Owens ES et al. Dexamethasone effects on 125I-albumin distribution in experimental RG-2 gliomas and adjacent brain. J Cereb Blood Flow Metab 1987;7(6):687–701. 26. Heiss JD, Papavassiliou E, Merrill MJ et al. Mechanism of dexamethasone suppression of brain tumor–associated vascular permeability in rats: involvement of the glucocorticoid receptor and vascular permeability factor. J Clin Invest 1996;98(6):1400–1408. 27. Chan PH, Fishman RA. The role of arachidonic acid in vasogenic brain edema. Fed Proc 1984;43(2):210–213. 28. Hedley-Whyte ET, Hsu DW. Effect of dexamethasone on blood–brain barrier in the normal mouse. Ann Neurol 1986;19(4):373–377. 29. Leenders KL, Beaney RP, Brooks DJ et al. Dexamethasone treatment of brain tumor patients: effects on regional cerebral blood flow, blood volume, and oxygen utilization. Neurology 1985;35(11):1610–1606. 30. Sinha S, Bastin ME, Wardlaw JM et al. Effects of dexamethasone on peritumoural oedematous brain: a DT-MRI study. J Neurol Neurosurg Psychiatry 2004;75(11):1632–1635. 31. Peters WP, Holland JF, Senn H et al. Corticosteroid administration and localized leukocyte mobilization in man. N Engl J Med 1972;286(7):342–345. 32. van Balkom RH, van der Heijden HF, van Herwaarden CL et al. Corticosteroid-induced myopathy of the respiratory muscles. N J Med 1994;45(3):114–122. 33. Lackner TE. Interaction of dexamethasone with phenytoin. Pharmacotherapy 1991;11(4):344–347. 34. Vecht CJ, Hovestadt A, Verbiest HB et al. Dose–effect relationship of dexamethasone on Karnofsky performance in metastatic brain tumors: a randomized study of doses of 4, 8, and 16 mg per day. Neurology 1994;44(4):675–680. 35. Ruderman NB, Hall TC. Use of glucocorticoids in the palliative treatment of metastatic brain tumors. Cancer 1965;18:298–306. 36. Lieberman A, LeBrun Y, Glass P et al. Use of high-dose corticosteroids in patients with inoperable brain tumours. J Neurol Neurosurg Psychiatry 1977;40(7):678–682. 37. Renaudin J, Fewer D, Wilson CB et al. Dose dependency of decadron in patients with partially excised brain tumors. J Neurosurg 1973;39(3):302–305. 38. Jarden JO, Dhawan V, Moeller JR et al. The time course of steroid action on blood-to-brain and blood-to-tumor transport of 82Rb: a positron emission tomographic study. Ann Neurol 1989;25(3):239–245. 39. Alberti E, Hartmann A, Schutz HJ et al. The effect of large doses of dexamethasone on the cerebrospinal fluid pressure in patients with supratentorial tumors. J Neurol 1978;217(3):173–181. 40. Andersen C, Astrup J, Gyldensted C. Quantitation of peritumoural oedema and the effect of steroids using NMR-relaxation time imaging and blood–brain barrier analysis. Acta Neurochir Suppl (Wien) 1994;60:413–415. 41. Crocker EF, Zimmerman RA, Phelps ME et al. The effect of steroids on the extravascular distribution of radiographic contrast material and technetium pertechnetate in brain tumors as determined by computed tomography. Radiology 1976;119(2):471–474.

394

Part VII / Neurologic Complications of Specific Malignancies

42. Yeung WT, Lee TY, Del Maestro RF et al. Effect of steroids on iopamidol blood–brain transfer constant and plasma volume in brain tumors measured with X-ray computed tomography. J Neurooncol 1994;18(1):53–60. 43. Weissman DE, Dufer D, Vogel V et al. Corticosteroid toxicity in neuro-oncology patients. J Neurooncol 1987;5(2):125–128. 44. Owczarek J, Jasinska M, Orszulak-Michalak D. Drug-induced myopathies: an overview of the possible mechanisms. Pharmacol Rep 2005;57(1):23–34. 45. Sieb JP, Gillessen T. Iatrogenic and toxic myopathies. Muscle Nerve 2003;27(2):142–156. 46. van Balkom RH, van der Heijden HF, van Moerkerk HT et al. Effects of different treatment regimens of methylprednisolone on rat diaphragm contractility, immunohistochemistry, and biochemistry. Eur Respir J 1996;9(6):1217–1223. 47. Bowyer SL, LaMothe MP, Hollister JR. Steroid myopathy: incidence and detection in a population with asthma. J Allergy Clin Immunol 1985;76(2 Pt 1):234–242. 48. Dropcho EJ, Soong SJ. Steroid-induced weakness in patients with primary brain tumors. Neurology 1991;41(8):1235–1239. 49. Chalk JB, Ridgeway K, Brophy T et al. Phenytoin impairs the bioavailability of dexamethasone in neurological and neurosurgical patients. J Neurol Neurosurg Psychiatry 1984;47(10):1087–1090. 50. Lewis DA, Smith RE. Steroid-induced psychiatric syndromes: a report of 14 cases and a review of the literature. J Affect Disord 1983;5(4):319–332. 51. Wolkowitz OM, Reus VI, Weingartner H et al. Cognitive effects of corticosteroids. Am J Psychiatry 1990;147(10):1297–1303. 52. Falk WE, Mahnke MW, Poskanzer DC. Lithium prophylaxis of corticotropin-induced psychosis. JAMA 1979;241(10):1011–1012. 53. Patten SB, Neutel CI. Corticosteroid-induced adverse psychiatric effects: incidence, diagnosis, and management. Drug Saf 2000;22(2):111–122. 54. de Quervain DJ, Henke K, Aerni A et al. Glucocorticoid-induced impairment of declarative memory retrieval is associated with reduced blood flow in the medial temporal lobe. Eur J Neurosci 2003;17(6):1296–1302. 55. Messer J, Reitman D, Sacks HS et al. Association of adrenocorticosteroid therapy and peptic-ulcer disease. N Engl J Med 1983;309(1):21–24. 56. Conn HO, Blitzer BL. Nonassociation of adrenocorticosteroid therapy and peptic ulcer. N Engl J Med 1976;294(9):473–479. 57. Carson JL, Strom BL, Schinnar R et al. The low risk of upper gastrointestinal bleeding in patients dispensed corticosteroids. Am J Med 1991;91(3):223–228. 58. Nielsen GL, Sorensen HT, Mellemkjoer L et al. Risk of hospitalization resulting from upper gastrointestinal bleeding among patients taking corticosteroids: a register-based cohort study. Am J Med 2001;111(7):541–545. 59. Tryba M. Side effects of stress bleeding prophylaxis. Am J Med 1989;86(6A):85–93. 60. Fadul CE, Lemann W, Thaler HT et al. Perforation of the gastrointestinal tract in patients receiving steroids for neurologic disease. Neurology 1988;38(3):348–352. 61. Hughes MA, Parisi M, Grossman S et al. Primary brain tumors treated with steroids and radiotherapy: low CD4 counts and risk of infection. Int J Radiat Oncol Biol Phys 2005;62(5):1423–1426. 62. Thomas CF, Jr., Limper AH. Pneumocystis pneumonia. N Engl J Med 2004;350(24):2487–2498. 63. Henson JW, Jalaj JK, Walker RW et al. Pneumocystis carinii pneumonia in patients with primary brain tumors. Arch Neurol 1991;48(4):406–409. 64. Schiff D. Pneumocystis pneumonia in brain tumor patients: risk factors and clinical features. J Neurooncol 1996;27(3):235–240. 65. Sambrook PN. How to prevent steroid-induced osteoporosis. Ann Rheum Dis 2005;64(2):176–178. 66. Dougherty JA. Risedronate for the prevention and treatment of corticosteroid-induced osteoporosis. Ann Pharmacother 2002;36(3): 512–516. 67. Recommendations for the prevention and treatment of glucocorticoid-induced osteoporosis: 2001 update. American College of Rheumatology Ad Hoc Committee on Glucocorticoid-Induced Osteoporosis. Arthritis Rheum 2001;44(7):1496–1503. 68. Saag KG, Emkey R, Schnitzer TJ et al. Alendronate for the prevention and treatment of glucocorticoid-induced osteoporosis: glucocorticoid-induced Osteoporosis Intervention Study Group. N Engl J Med 1998;339(5):292–299. 69. Wallach S, Cohen S, Reid DM et al. Effects of risedronate treatment on bone density and vertebral fracture in patients on corticosteroid therapy. Calcif Tissue Int 2000;67(4):277–285. 70. Assouline-Dayan Y, Chang C, Greenspan A et al. Pathogenesis and natural history of osteonecrosis. Semin Arthritis Rheum 2002;32(2):94–124. 71. Baharav E, Harpaz D, Mittelman M et al. Dexamethasone-induced perineal irritation. N Engl J Med 1986;314(8):515–516. 72. Baethge BA, Lidsky MD. Intractable hiccups associated with high-dose intravenous methylprednisolone therapy. Ann Intern Med 1986;104(1):58–59. 73. Amatruda TT, Jr., Hurst MM, D’Esopo ND. Certain endocrine and metabolic facets of the steroid withdrawal syndrome. J Clin Endocrinol Metab 1965;25(9):1207–1217. 74. Lawson LA, Blouin RA, Smith RB et al. Phenytoin–dexamethasone interaction: a previously unreported observation. Surg Neurol 1981;16(1):23–24. 75. Willett CG, Boucher Y, di Tomaso E et al. Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 2004;10(2):145–147. 76. Jain RK. Tumor angiogenesis and accessibility: role of vascular endothelial growth factor. Semin Oncol 2002;29(6 Suppl 16):3–9. 77. Jain RK. Angiogenesis and lymphangiogenesis in tumors: insights from intravital microscopy. Cold Spring Harb Symp Quant Biol 2002;67:239–248. 78. Ropper AH, King RB. Intracranial pressure monitoring in comatose patients with cerebral hemorrhage. Arch Neurol 1984;41(7): 725–728. 79. Donato T, Shapira Y, Artru A et al. Effect of mannitol on cerebrospinal fluid dynamics and brain tissue edema. Anesth Analg 1994;78(1):58–66.

Chapter 20 / Neurological Complications of Primary Brain Tumors

395

80. Hartwell RC, Sutton LN. Mannitol, intracranial pressure, and vasogenic edema. Neurosurgery 1993;32(3):444–450; discussion 50. 81. Ravussin P, Abou-Madi M, Archer D et al. Changes in CSF pressure after mannitol in patients with and without elevated CSF pressure. J Neurosurg 1988;69(6):869–876. 82. Pollay M, Fullenwider C, Roberts PA et al. Effect of mannitol and furosemide on blood–brain osmotic gradient and intracranial pressure. J Neurosurg 1983;59(6):945–950. 83. Rottenberg DA, Hurwitz BJ, Posner JB. The effect of oral glycerol on intraventricular pressure in man. Neurology 1977;27(7):600–608. 84. Ropper AH. Neurological and Neurosurgical Intensive Care, 3rd ed. New York: Raven; 1993. 85. Fritsch MJ, Doerner L, Kienke S et al. Hydrocephalus in children with posterior fossa tumors: role of endoscopic third ventriculostomy. J Neurosurg 2005;103(1 Suppl):40–42. 86. Kestle JR, Garton HJ, Whitehead WE et al. Management of shunt infections: a multicenter pilot study. J Neurosurg 2006;105(3 Suppl):177–181. 87. Kastrup O, Wanke I, Maschke M. Neuroimaging of infections. NeuroRx 2005;2(2):324–332. 88. Bink A, Gaa J, Franz K et al. Importance of diffusion-weighted imaging in the diagnosis of cystic brain tumors and intracerebral abscesses. Zentralbl Neurochir 2005;66(3):119–125. 89. DeAngelis LM, Delattre JY, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology 1989;39(6): 789–796. 90. Thiessen B, DeAngelis LM. Hydrocephalus in radiation leukoencephalopathy: results of ventriculoperitoneal shunting. Arch Neurol 1998;55(5):705–710. 91. Vertosick FT, Jr., Selker RG, Arena VC. Survival of patients with well-differentiated astrocytomas diagnosed in the era of computed tomography. Neurosurgery 1991;28(4):496–501. 92. Lieu AS, Howng SL. Intracranial meningiomas and epilepsy: incidence, prognosis and influencing factors. Epilepsy Res 2000;38(1): 45–52. 93. Hochberg FH, Miller DC. Primary central nervous system lymphoma. J Neurosurg 1988;68(6):835–853. 94. Hildebrand J, Lecaille C, Perennes J et al. Epileptic seizures during follow-up of patients treated for primary brain tumors. Neurology 2005;65(2):212–215. 95. Ye ZC, Sontheimer H. Glioma cells release excitotoxic concentrations of glutamate. Cancer Res 1999;59(17):4383–4391. 96. Behrens PF, Langemann H, Strohschein R et al. Extracellular glutamate and other metabolites in and around RG2 rat glioma: an intracerebral microdialysis study. J Neuro-oncol 2000;47(1):11–22. 97. Takano T, Lin JH, Arcuino G et al. Glutamate release promotes growth of malignant gliomas. Nat Med 2001;7(9):1010–1015. 98. Tian GF, Azmi H, Takano T et al. An astrocytic basis of epilepsy. Nat Med 2005;11(9):973–981. 99. Engel J. Seizures and Epilepsy. Philadelphia: F.A. Davis; 1989. 100. Rodman JH, Murry DJ, Madden T et al. Altered etoposide pharmacokinetics and time to engraftment in pediatric patients undergoing autologous bone marrow transplantation. J Clin Oncol 1994;12(11):2390–2397. 101. Gilbert MR, Supko JG, Batchelor T et al. Phase I clinical and pharmacokinetic study of irinotecan in adults with recurrent malignant glioma. Clin Cancer Res 2003;9(8):2940–2949. 102. Reardon DA, Friedman HS, Powell JB, Jr. et al. Irinotecan: promising activity in the treatment of malignant glioma. Oncology (Williston Park) 2003;17(5 Suppl 5):9–14. 103. Zamboni WC, Gajjar AJ, Heideman RL et al. Phenytoin alters the disposition of topotecan and N-desmethyl topotecan in a patient with medulloblastoma. Clin Cancer Res 1998;4(3):783–789. 104. Villikka K, Kivisto KT, Maenpaa H et al. Cytochrome P450-inducing antiepileptics increase the clearance of vincristine in patients with brain tumors. Clin Pharmacol Ther 1999;66(6):589–593. 105. Vecht CJ, Wagner GL, Wilms EB. Interactions between antiepileptic and chemotherapeutic drugs. Lancet Neurol 2003;2(7):404–409. 106. van den Bent MJ, Afra D, de Witte O et al. Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial. Lancet 2005;366(9490):985–990. 107. Ngo L, Nei M, Glass J. Temozolomide treatment of refractory epilepsy in a patient with an oligodendroglioma. Epilepsia 2006;47(7):1237–1238. 108. Vecht CJ, Wagner GL, Wilms EB. Treating seizures in patients with brain tumors: drug interactions between antiepileptic and chemotherapeutic agents. Semin Oncol 2003;30(6 Suppl 19):49–52. 109. Glantz MJ, Cole BF, Friedberg MH et al. A randomized, blinded, placebo-controlled trial of divalproex sodium prophylaxis in adults with newly diagnosed brain tumors. Neurology 1996;46(4):985–991. 110. Glantz MJ, Cole BF, Forsyth PA et al. Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology 2000;54(10):1886–1893. 111. Mamon HJ, Wen PY, Burns AC et al. Allergic skin reactions to anticonvulsant medications in patients receiving cranial radiation therapy. Epilepsia 1999;40(3):341–344. 112. Chang SM, Parney IF, Huang W et al. Patterns of care for adults with newly diagnosed malignant glioma. JAMA 2005;293(5):557–564. 113. Klein M, Heimans JJ, Aaronson NK et al. Effect of radiotherapy and other treatment-related factors on mid-term to long-term cognitive sequelae in low-grade gliomas: a comparative study. Lancet 2002;360(9343):1361–1368. 114. Sirven JI, Wingerchuk DM, Drazkowski JF et al. Seizure prophylaxis in patients with brain tumors: a meta-analysis. Mayo Clin Proc 2004;79(12):1489–1494. 115. Mulhern RK, Khan RB, Kaplan S et al. Short-term efficacy of methylphenidate: a randomized, double-blind, placebo-controlled trial among survivors of childhood cancer. J Clin Oncol 2004;22(23):4795”–4803. 116. Meyers CA, Weitzner MA, Valentine AD et al. Methylphenidate therapy improves cognition, mood, and function of brain tumor patients. J Clin Oncol 1998;16(7):2522–2527.

396

Part VII / Neurologic Complications of Specific Malignancies

117. Fouladi M, Gajjar A, Boyett JM et al. Comparison of CSF cytology and spinal magnetic resonance imaging in the detection of leptomeningeal disease in pediatric medulloblastoma or primitive neuroectodermal tumor. J Clin Oncol 1999;17(10):3234–3237. 118. Gajjar A, Fouladi M, Walter AW, et al. Comparison of lumbar and shunt cerebrospinal fluid specimens for cytologic detection of leptomeningeal disease in pediatric patients with brain tumors. J Clin Oncol 1999;17(6):1825–1828. 119. Abrey LE, Yahalom J, DeAngelis LM. Treatment for primary CNS lymphoma: the next step. J Clin Oncol 2000;18(17):3144–3150. 120. Khan RB, Shi W, Thaler HT et al. Is intrathecal methotrexate necessary in the treatment of primary CNS lymphoma? J Neuro-oncol 2002;58(2):175–178. 121. Ekstein D, Ben-Yehuda D, Slyusarevsky E et al. CSF analysis of IgH gene rearrangement in CNS lymphoma: relationship to the disease course. J Neurol Sci 2006;247(1):39–46. 122. Schinstine M, Filie AC, Wilson W et al. Detection of malignant hematopoietic cells in cerebral spinal fluid previously diagnosed as atypical or suspicious. Cancer 2006;108(3):157–162. 123. Wagner S, Benesch M, Berthold F et al. Secondary dissemination in children with high-grade malignant gliomas and diffuse intrinsic pontine gliomas. Br J Cancer 2006;95(8):991–997. 124. Gururangan S, McLaughlin CA, Brashears J et al. Incidence and patterns of neuraxis metastases in children with diffuse pontine glioma. J Neurooncol 2006;77(2):207–212. 125. Evans AE, Jenkin RD, Sposto R et al. The treatment of medulloblastoma: results of a prospective randomized trial of radiation therapy with and without CCNU, vincristine, and prednisone. J Neurosurg 1990;72(4):572–582. 126. Fouladi M, Blaney SM, Poussaint TY et al. Phase II study of oxaliplatin in children with recurrent or refractory medulloblastoma, supratentorial primitive neuroectodermal tumors, and atypical teratoid rhabdoid tumors: a pediatric brain tumor consortium study. Cancer 2006;107(9):2291–2297. 127. Grill J, Dufour C, Kalifa C. High-dose chemotherapy in children with newly diagnosed medulloblastoma. Lancet Oncol 2006;7(10): 787–789. 128. Schulz H, Pels H, Schmidt-Wolf I et al. Intraventricular treatment of relapsed central nervous system lymphoma with the anti-CD20 antibody rituximab. Haematologica 2004;89(6):753–754. 129. Pels H, Schulz H, Schlegel U et al. Treatment of CNS lymphoma with the anti-CD20 antibody rituximab: experience with two cases and review of the literature. Onkologie 2003;26(4):351–354. 130. Pels H, Schulz H, Manzke O et al. Intraventricular and intravenous treatment of a patient with refractory primary CNS lymphoma using rituximab. J Neuro-oncol 2002;59(3):213–216. 131. Rubenstein JL, Combs D, Rosenberg J et al. Rituximab therapy for CNS lymphomas: targeting the leptomeningeal compartment. Blood 2003;101(2):466–468.

21

Neurologic Complications of Lung Cancer Suriya A. Jeyapalan, MD, MA and Anand Mahadevan, MD,

FRCS

CONTENTS Introduction Direct Neurological Complications of Lung Cancer Leptomeningeal Metastases Epidural Spinal Cord Compression Intramedullary Spinal Cord Metastases Skull Base Metastases Plexus and Peripheral Nerve Metastases Indirect Complications of Lung Cancer Conclusion References

Summary Lung cancer frequently causes neurological complications from direct and indirect effects. Brain metastases occur in 41% of patients with non-small cell lung cancer and 35% with small cell lung cancer at autopsy. Presenting symptoms can be quite protean. MRI continues to be the gold standard of detecting metastases. Biopsy or resection should be considered for patients with a single lesion, but is not necessary in the case of multiple metastases and a known history of cancer. Removal of single lesions impacts favorably on survival and allows for the rapid tapering of glucocorticoids. Radiation therapy standards are evolving from wide use of whole-brain radiotherapy to consideration of more focused radiosurgery in selected patients to avoid long-term neurotoxicity. The role of chemotherapy in the treatment of brain metastases is still evolving as results of ongoing randomized and controlled trials using newer blood–brain barrier penetrating agents. Key Words: lung cancer, brain metastasis, whole-brain radiation, small cell lung cancer

1. INTRODUCTION 1.1. Epidemiology Lung cancer is the most frequent form of human cancer and is the leading worldwide cause of cancer-related mortality. In the United States, it is the second leading cause of cancer in both men and women and is the leading cause of cancer-related mortality. In 2007, there were 213,380 new cases of lung cancer, comprising 15% of cancer diagnoses. The incidence is higher in men than women, but while the incidence rate of lung cancer among men has dropped by 2.5% since 1973, the incidence has increased among women by 123%. Concomitantly, mortality rates have been increasing in women more than men. The lifetime risk of developing lung cancer is 8.3% among female smokers and 14.6% among male smokers. Among ethnic groups, blacks have the highest From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

397

398

Part VII / Neurologic Complications of Specific Malignancies

incidence rates (117 per 100,000) while Native Americans have the lowest (14 per 100,000). Lung cancer rarely occurs before the age of 40, but thereafter a sharp increase in rates occurs. The median age at diagnosis is 65–70 years. Fifteen percent of patients are diagnosed initially with localized lung cancer; 23% have regional spread of disease; the majority (48%) have distant metastases. The 5-year survival rate has changed little over time, but there has been some improvement in 2-year survival rates since 1975, from 34.3% to 41%. The 5-year survival for all patients is 15%. Lung cancer is unique among cancers in that its etiology has been attributed to a single strong risk factor— smoking. The increasing incidence among women is thought to be due to the lag of 20 years in the prevalence of smoking among women compared to men.

1.2. Tumor Classification Lung cancer is a generic term used to describe cancer arising from the trachea, bronchi, bronchioles, and alveoli. According to the World Health Organization (WHO), 20% of lung cancer is small cell cancer (SCLC, or oat cell cancer) and 80% is non-small cell lung cancer (NSCLC). NSCLC is subdivided into three groups: (i) adenocarcinoma (40% of all lung cancer), (ii) squamous cell or epidermoid carcinoma (30%), and (iii) large cell undifferentiated carcinoma (10%). These three subgroups are often combined under the heading of NSCLC because they frequently coexist in a single tumor, they are hard to distinguish from one another when poorly differentiated, and because their overall prognosis and treatment is the same (Table 1) (1,2). The revised international staging classification of lung cancer (1997) divides NSCLC into four stages (Table 2) (2–4). This chapter primarily focuses on neurological complications arising from stage IV lung cancer.

1.3. General Oncologic Management 1.3.1. Small Cell Lung Cancer (SCLC) Two-thirds of patients with SCLC have distant metastatic disease at presentation. Patients with stage I or II disease are candidates for surgical resection, but surgery can be also considered as part of combination therapy. The standard approach to these patients is multi-agent chemotherapy, which has an initial response rate of 75–95% (2). Complete responses (CR) are seen in 50–60% of patients with limited disease. These responses, however, are usually short-lived, resulting in overall median survival of less that one year. Among patients with limited stage disease, 20–30% survive two years. In patients with limited stage disease, adjuvant radiotherapy is also beneficial (2). Combination therapy has been shown to moderately improve median survival compared to patients who receive only chemotherapy. Metaanalysis of clinical trials has confirmed this benefit. Timing of radiotherapy may be important, as early therapy may destroy drug-resistant clones that might otherwise survive. Prophylactic cranial irradiation (PCI) will be discussed later in this chapter. 1.3.2. Non-Small Cell Lung Cancer (NSCLC) Surgery for stage I NSCLC results in a 5-year survival of 40–67% (2). In this group, many of the “recurrences” turn out to be second primary cancers. Irradiation is used in patients who do not undergo surgery. Radiotherapy alone has a 25% cure rate in stage I NSCLC. Table 1 WHO Classification of Lung Cancer Type of Cancer Small cell lung cancer (SCLC) Non-small cell lung cancer (NSCLC) Adenocarcinoma (40%) Squamous cell carcinoma or epidermoid carcinoma (30%) Large cell undifferentiated carcinoma (10%)

Incidence 20% 80%

Chapter 21 / Neurologic Complications of Lung Cancer

399

Table 2 Staging of NSCLC Stage 0 IA IB IIA IIB IIIA

IIIB

IV

TNM Subset** Carcinoma in situ T1N0M0 T2N0M0 T1N1M0 T2N1M0 T3N0M0 T3N1M0 T1N2M0 T2N2M0 T3N2M0 T4N0M0 T4N1M0 T4N2M0 T1N3M0 T2N3M0 T3N3M0 T4N3M0 Any T, any N, M1

**TNM: T = Tumor size/invasion, N = nodal involvement, M = distant metastases. T1 = tumor size = 3cm, invasion confined to the lobar bronchus T2 = any of the following: tumor size > 3 cm, invasion of the visceral pleura, involvement of the mainstem bronchus, ≥ 2 cm distant to the carina. Associated with obstructive pneumonitis or atelectasis extending to the hilum, without involvement of the entire lung. N0 = no nodal involvement N1 = no nodal involvement outside of the pleural envelope.

Surgical resection was considered the definitive therapy in stage II NSCLC (2,6). However, there is a small survival benefit with adjuvant platinum-based chemotherapy based on pooled analysis of randomized clinical trials (5). In patients with stage II NSCLC the 5-year survival is 25–55% (2,6). Stage IIIa cancer is treated with a combination of chemotherapy, surgery, and radiation, while stage IIIb is treated with chemotherapy and radiation alone (2,6). A recent phase III comparison of sequential versus concurrent chemoradiation for unresected stage III NSCLC suggested improved median survival with the concurrent treatment strategy (7). In stage IIIb NSCLC, there is often recurrence of disease locally, with median survivals of 10 months or less and a 5-year survival of 5–10%. Stage IV cancer is present in 40–50% of patients at initial diagnosis of NSCLC. Both randomized trials and meta-analysis support the use of palliative chemotherapy in patients with a good performance status. Compared with supportive care alone, chemotherapy improves quality of life and can increase both median and one-year survival.

1.4. Patterns of Spread to the CNS 1.4.1. Small Cell Lung Cancer A retrospective study of the autopsy findings of 537 patients with SCLC found a 35% incidence of brain metastases and an 8% incidence of leptomeningeal metastases (LM) (8). Brain metastases are detected clinically in 12% of patients prior to starting treatment. Patients with a complete response to therapy had a higher rate of brain metastases (60% vs. 40%) and LM (24% vs. 10%) than patients with a partial response.

400

Part VII / Neurologic Complications of Specific Malignancies

1.4.2. Non-Small Cell Lung Cancer Autopsy studies in NSCLC show a 41% incidence of brain metastases and a 9% incidence of LM (9). In patients with LM there is usually coexistence of parenchymal brain metastases. Overall, the metastatic pattern seen in NSCLC does not differ from that of SCLC (8). In a study of 292 patients with NSCLC, brain relapse was shown to be affected predominantly by good performance status at the time of diagnosis; it predicted both improved survival and later relapse (10). Another study of 422 patients with stage IIIa/IIIb NSCLC found that younger age at presentation and nonsquamous pathology were increased risk factors for early relapse (11). Overall, 60% of the patients had a relapse, with 26% having brain involvement (20% had brain as the only site of relapse). About 20% of these occurred in the first four months following treatment and a quarter occurred after 4–6 months of treatment. This indicated that brain metastases did occur early in the course of stage III NSCLC. There are currently trials investigating the use of PCI in this subgroup of patients.

2. DIRECT NEUROLOGICAL COMPLICATIONS OF LUNG CANCER Most studies dealing with neurological complications of systemic cancer include patients with a wide variety of primary tumor types, making it somewhat difficult to draw conclusions about primary-specific complications. However, because lung cancer is a common cause of neurologic complications, it generally comprises a majority of cases in these studies. In this chapter we have attempted to provide as much lung cancer–specific information as possible, but the reader should be aware of the general nature of much of this discussion (Table 3).

2.1. Parenchymal Brain Metastases 2.1.1. Incidence Brain metastases are a common finding in patients with lung cancer. They are present in 10% of patients with SCLC at the time of diagnosis and increase to 20% during therapy and 35% at time of autopsy (12–14). At two years post-diagnosis the cumulative risk of brain metastasis is 47% for patients with limited disease and 69% for those with extensive disease (14). Patients with NSCLC have a 20% incidence of brain metastases, which increases to 40% at time of autopsy (13). Brain metastases are the initial manifestations of disease in as many as 10% of patients with lung cancer. Some experts advocate the routine use of brain CT or MR scan to detect asymptomatic metastases prior to planned curative thoracic resections. Because MR scanning reveals brain metastases in only 3% of such studies, the role of neuroimaging in this setting is arguable.

2.2. Manifestations 2.2.1. Symptoms Presenting symptoms of cerebral metastases can be generalized or focal. Headache is one of the most common symptoms, occurring in up to 50% of patients (15,16). Patients complain of a mild, constant, dull headache, which usually responds well to analgesics. The classic “brain tumor headache,” which is worse in the morning or with Valsalva maneuver, is less common (15). Other symptoms of increased intracranial pressure include impairment of consciousness, nausea, and vomiting and blurring of vision secondary to papilledema (15,17). Table 3 Neurologic Complications of Lung Cancer Metastatic Complications Parenchymal brain metastases Leptomeningeal metastases Epidural metastases Skull base metastases Intramedullary spinal cord metastases Plexopathy

Paraneoplastic Syndromes Encephalomyelitis Cerebellar syndrome Sensory neuronopathy Autonomic neuropathy LEMS Polymyositis/Dermatomyositis

Chapter 21 / Neurologic Complications of Lung Cancer

401

Cognitive disturbances occur in 20–40% of patients and can include such general symptoms as depression, personality change, and memory loss, as well as more focal symptoms such as aphasia, alexia, acalculia, agnosia, and apraxia (15,16,18). Generalized cognitive disturbances are more common in patients with multiple metastases or with increased intracranial pressure. These symptoms may masquerade as psychiatric disease and can lead to delay in diagnosis. Focal weakness is a common finding (16,18). Because tumor-associated edema may also result in brain dysfunction, the pattern of weakness may not accurately reflect the precise location of a cerebral metastasis. Focal or generalized seizures are the presenting complaint in about 20% of patients (16,18) and occur in 30–40% of patients at some point during the course of the disease (15). Ataxia as a presenting symptom is common with cerebellar or brainstem metastases, although gait ataxia may also result from large frontal lobe metastases or hydrocephalus (15,16). 2.2.2. Signs Findings on the neurological examination are typically in excess of the presenting complaints. Cerebellar metastases are an exception, as patients often complain of being more unsteady than actually detected on examination (15). Mental status testing is abnormal in up to 75% of patient with cerebral metastases (15). When present, focal findings will typically identify the neuroanatomic location of metastases or tumor-associated edema.

2.3. Radiologic Findings 2.3.1. CT Cranial CT scanning is a valuable imaging modality in the diagnostic evaluation of cerebral metastases, although it is less sensitive than MRI in the detection of small or infratentorial lesions. Cerebral metastases are usually hypodense on pre-contrast scans, unless the lesions are hemorrhagic or of extremely high cell density (15,19). Metastases become hyperdense after administration of intravenous contrast. 2.3.2. MRI Gadolinium-enhanced cranial MRI is the most sensitive test available for the detection of cerebral metastases (15,19). An MRI should be considered in patients with lung cancer who have the symptoms or signs noted above or are at high risk for brain metastases. Metastases tend to be isointense or mildly hypointense on T1 pre-contrast images when compared to gray matter. T2 images show increased signal in the tumor and surrounding gray and white matter. Almost all cerebral metastases enhance after administration of intravenous gadolinium. Technical advances such as magnetization transfer contrast, MR diffusion, and MR spectroscopy are being investigated. There is some evidence that signal intensity on DWI may reflect the pathology of the metastases, with neuroendocrine tumors being hyperintense while well-differentiated adenocarcinomas are hypointense (20). It has been shown that dedicated PET imaging of the brain does not detect brain metastases from NSCLC (21).

2.4. Diagnosis 2.4.1. Surgical In the setting of known metastatic cancer, biopsy of a lesion with a typical radiographic appearance is usually not indicated. Brain biopsy should be considered in patients without a known cancer, with a remote history of lung cancer, or with indolent, limited stage disease. In one study from the CT era, 11% of the patients with known cancer and a solitary intracranial lesion had diagnoses other than metastases at brain biopsy (22). 2.4.2. Other Tests Biochemical markers such as calcitonin, bombesin, and 2-microglobulin have been examined as possible screening tests in CSF for parenchymal metastases in patients with small cell lung cancer (23,24). Although the specificity of these tests was high, sensitivity was only about 50%, thereby limiting their usefulness as screening measures. Recently S100, an astrocytic protein released into the CSF when there is breakdown of the blood–brain barrier, has been tested in patients with newly diagnosed lung carcinoma (25). It is interesting that none of the

402

Part VII / Neurologic Complications of Specific Malignancies

patients with a normal S100 level had evidence of brain metastases, that is, the negative predictive value of this finding is 1.00. The caveat was that patients with radiological evidence of microvascular changes could also have elevation of their S100 level. However, the authors argued that if one excluded these patients, the positive predictive value of the test rose to 0.875 from 0.471, for the whole series.

2.5. Management 2.5.1. Glucocorticoids Glucocorticoids are effective in the management of cerebral edema associated with metastatic lesions (26). Symptomatic response to glucocorticoids is an important prognosticator of response to further therapy and survival (27). Dexamethasone is the most commonly used glucocorticoid because of its potent effect on cerebral edema and its low mineralocorticoid activity (26). Dexamethasone is often started with a loading dose of 10 mg followed by 4–6 mg qid, although doses of 2 mg bid suffice in many patients. In patients who fail to respond, the dose can be increased up to 40 mg/day. As the patient responds to other treatment modalities, glucocorticoids are slowly tapered off to minimize side effects such as steroid myopathy, diabetes, and immunosuppression. Trimethoprim/sulfamethoxazole prophylaxis against Pneumocystis jirovecii pneumonia should be considered in patients on dexamethasone for longer than two months (28). 2.5.2. Antiepileptic Drugs (AEDs) Antiepileptic agents should be administered to patients with intracranial metastases who have seizures (26). Levetiracetam is a commonly used AED with few side effects; other advantages are twice-a-day dosage and the possibility of intravenous administration. Its main advantage in the treatment of cancer is that it is metabolized exclusively by the kidney and therefore does not interact with most chemotherapies, which are metabolized by the liver. Phenytoin, carbamazepine, valproic acid, and other AEDs are also effective. In randomized trials AED prophylaxis have shown no benefit (26). Therefore, AEDs should be withheld from patients with brain metastases until a first seizure. 2.5.3. Surgery The impact of surgical resection of brain metastases has been difficult to assess due to bias in the selection of patients. In one study, median survival after surgery for brain metastases from lung cancer was 11.6 months (29). Survival rates were 24% at one year and 8% at two years. Performance status after surgery improved in 36% of patients, remained the same in 53%, and became worse in 11%. Favorable prognostic factors included stable systemic disease, good initial performance scores, histologic diagnosis of adenocarcinoma, solitary brain metastases, and receipt of adjuvant therapy. 2.5.4. Role of Surgery in Single and Multiple Metastases Single brain metastases occur in one-quarter to one-third of patients, and about half of metastases are resectable (30,31). Unlike the situation in infiltrative gliomas, metastases are well demarcated from the surrounding brain tissue, making complete resection possible. Resection of a single brain lesion followed by whole-brain radiation (WBRT) appear to improve survival and quality of life and delay recurrence compared with WBRT alone (22,32). In these studies, there was no survival benefit in patients with active extracranial disease. Patients with resection of a single lesion may be able to discontinue their glucocorticoids more rapidly, thereby lowering the incidence of side effects. Patients with single metastases from primary tumors that are highly sensitive to radiotherapy (e.g., SCLC) may be less likely to benefit from resection. A retrospective cohort study of 231 patients with NSCLC who underwent surgical resection of their brain metastases reported a median survival of 11 months (33). Significant positive prognostic factors were female sex (p < 0.02), single metastases (11 vs. 8.5 month median survival, p < 0.02), high KPS, complete resection of primary tumor, and age < 60 years. About one-third of the cohort died of neurological complications while another third died of a combination of systemic and neurological causes. These latter statistics are similar to those of all patients with CNS metastases, suggesting that surgery did not reduce the long-term risk from neurological complications.

Chapter 21 / Neurologic Complications of Lung Cancer

403

It has been suggested that surgical removal of two or three lesions improves survival (14 vs. 6 months) and that the quality of life is similar to that in patients undergoing excision of a single lesion (34). However, another study has found no benefit of surgery for multiple metastases, with median survival of 5 months in this group compared to 12 months for the patients with single intracranial metastases (35). These results do not permit definite recommendations regarding the role of surgery in patients with multiple metastases (26). Currently, therefore, accessible lesions that are symptomatic, large, or life-threatening are most often considered for surgery. 2.5.5. Recurrent Brain Metastases A retrospective study of 214 patients has shown that re-resection of recurrent brain metastases in patients with NSCLC may prolong survival (36). The median survival of patients who underwent a second operation was 15 months from time of the first operation compared to 10 months for those who did not undergo an additional surgery (p < 0.001). Positive predictive factors were female sex, histology of the primary tumor (adenocarcinoma associated with a better survival), disease stage, and extent of resection of the primary tumor. Almost half of the patients were re-irradiated after surgery; the median survival of those re-irradiated was 24 months compared to 14.4 months for those who were not re-irradiated (p = 0.48). A third operation was performed in some patients. The median period until a third recurrence was 4 months. Median survival after this third surgery was 10.5 months. These survival statistics strongly suggest that patients were highly selected for those with limited, indolent disease and that in such patients an aggressive approach to management of CNS disease can produce remarkably long survival.

2.6. Radiation Radiation therapy (RT) is a standard treatment for patients with brain metastases from lung cancer (26). Randomized trials have shown a benefit in survival and quality of life with the addition of RT, even after complete excision of a single metastasis (22,32,37). RT reduces local recurrence rates and may effectively eradicate small metastases thereby reducing distant brain relapse rates (37). The radiotherapeutic options in the management of brain metastasis from lung cancer include whole-brain radiation therapy (WBRT), stereotactic radiosurgery (SRS), or a combination of both (WBRT + SRS). 2.6.1. Whole-Brain Radiation Whole-brain radiation is the standard form of treatment for brain metastases, employing external beam, fractionated megavoltage radiation, the most common regimen being 3000 cGy in 10 fractions over 2 weeks (26,38). Altered fractionation treatments and addition of radiosensitizers offer no benefit over the conventional schedule (38). Administration over shorter periods of time may actually lead to poorer survival rates (15,26). Median survival after WBRT ranges from 2 to 8 months with one-year survival of 5–10% (15,26,39,40) depending on their age, performance status, control of their primary site, and presence of other metastasis. In patients with a solitary metastasis treated with WBRT who have their primary site under control, the most likely cause of death is recurrence of the brain metastasis (15,26,39). Long-term survivors (patients who lived more than a year from treatment) were examined in a recent retrospective analysis of patients with brain metastases treated with WBRT + SRS vs SRS alone. The addition of WBRT to SRS specifically increased local control in the subgroup of patients who had metastases from lung cancer. Other positive factors were that of single metastases, peripheral dose ≥16 Gy, and tumor volume ≤ 2cc. As in other studies, the addition of WBRT did not improve survival (41). Overall, about 60% of patients respond to radiotherapy with improvement or stabilization in their symptoms (42). Median survival is about 4.5 months in these patients compared to 1.5 months in nonresponders. Treatment has been shown to be more effective when started early (26). Response to radiation is often delayed, with about 50–60% of the patients having a partial response (i.e., at least 50% shrinkage in the tumor) at 6 weeks after the completion of WBRT (15). The incidence of brain metastasis in SCLC is greater than 50%. WBRT is generally given to these patients due to multiple brain metastasis and increased radiosensitivity as survival is usually only 4.5 months after its administration. 2.6.2. Sterotactic Radiosurgery Stereotactic radiosurgery (SRS) has gained increased importance in the management of brain metastases from lung cancer. Its role is more important in NSCLC than in SCLC given the latter histology’s typical radiosensitivity

404

Part VII / Neurologic Complications of Specific Malignancies

and multiplicity of the metastases. Part of the rationale for the use of SRS has been the belief that metastases are well-circumscribed and do not infiltrate into the brain. Therefore, they present as a well-defined target for focal radiation to be given, as local control can be easily achieved. This may not be absolutely correct in the case of both NSCLC and SCLC. In a recently published histologic analysis of brain metastases, SCLC had the deepest penetration, and lung cancer, overall, had the greatest degree of infiltration (43). This has implications in treatment of lung metastases with a slightly larger volume and using WBRT alone in the setting of multiple metastases and in addition to SRS for oligometastasis. SRS has been utilized by itself or as an adjuvant boost to WBRT for oligometastatic disease and as a salvage therapy for recurrence following WBRT or resection. NSCLC has been estimated to be the cause of 37–56% of all brain metastases, and in all the trials conducted using SRS for brain metastases lung cancer is the most common primary site. Patients who have only single or solitary lesions synchronously or metachronously with their lung cancer comprise an important group of these patients. Despite this, the role of SRS versus surgery in the treatment of single or solitary metastases from NSCLC has not been clearly defined. There have been two randomized trials comparing SRS alone versus SRS + WBRT and neither showed a survival benefit for either approach (44,45). Local control was poor in the Japanese trial without the addition of WBRT. Two other trials compared the addition SRS to WBRT versus WBRT alone (46,47). Once again there was no survival benefit. However, the well-conducted RTOG 9508 (47) did show benefit to the addition of SRS boost in maintaining performance status, decreased steroid usage and even survival advantage in solitary brain metastasis with squamous histology in good performance patients. As a single treatment of SRS results in a local control rate of >80%, there is debate whether it can be used alone or in addition to WBRT, particularly in good performance NSCLC patients with oligometastasis (48). SRS is established as a treatment modality for recurrence after WBRT, and the RTOG 9005 trial established the dose levels for tumors < 4 cm and demonstrated good control. A recent retrospective review looked at the role of SRS in the treatment of recurrent brain metastases after the administration of WBRT (49). The authors achieved an 81% local control rate and only 4% of patients died solely of progressive intracranial disease. The majority of patients died of a mixture of extracranial and intracranial disease while 41% died of extracranial progression. Median survival was 4.5 months after the administration of SRS. Significant prognostic factors were tumor volume, time between initial diagnosis of SCLC and development of brain metastases, and pre-operative KPS. 2.6.3. Prophylactic Cranial Irradiation (PCI) Though initially responsive to chemotherapy, SCLC progresses with distant metastases occurring early in the course of the disease (13). With combined treatment producing thoracic CR, the risk of thoracic recurrence decreases and brain becomes an important region of tumor failure. It has been presumed that the brain is a pharmacologic sanctuary in which microscopic tumor is protected against typical systemic chemotherapy for SCLC which does not penetrate the intact blood–brain barrier. This has led to numerous trials designed to test whether PCI would decrease the incidence of brain relapse and improve survival in patients who achieved systemic CR. Typically, 24–36 Gy WBRT has been administered in 2–2.5 Gy fractions. These studies commonly demonstrated that PCI decreased risk of brain metastasis and typically showed an insignificant trend towards survival benefit (50,51). A recently published meta-analysis of these studies indicated that PCI reduced the risk of subsequent brain metastasis in half and significantly increased 3-year survival from 15.3% to 20.7% (p = 0.01) (52). The majority of the patients in this meta-analysis had limited disease with the median age being 59 years and over 90% were WHO class I. A retrospective analysis has investigated the role of PCI in patients with a poor performance status (53). They found that brain metastases occurred about 20% less frequently than what would be expected in published rates of similar patients that were not treated with PCI. An ongoing EORTC randomized, controlled trial investigating treatment of SCLC patients with extensive disease should help further define the role of PCI in these patients. Controversy still remains over whether the benefits of PCI outweigh its toxicities. Short-term side effects of PCI are usually benign and consist mostly of headache, loss of hair, transient lack of appetite, decreased

Chapter 21 / Neurologic Complications of Lung Cancer

405

hearing, decreased taste, and fatigue (13). Potential long-term side effects of PCI include leukoencephalopathy and dementia that can occur about a year after radiation is completed. Milder forms of neuropsychological impairment are a long-term complication of cranial radiation recently recognized in adults (54). Two large prospective randomized trials of PCI did not document increased neuropsychological deficits among PCI recipients (50,51). Others have argued that the small numbers of long-term survivors in these trials precluded accurate assessment of the risk of leukoencephalopathy (55), and that since PCI benefited only about one-quarter of its recipients it should not be considered standard (56). There have been four randomized, prospective trials that have investigated the role of PCI in NSCLC. The first two were done in the early 1980s by the RTOG (84-03) and by Umsawasdi et al. The other two trials, the VALG (Veteran’s Administration Lung Group) and the SWOG, were done in the 1990s. Unfortunately, they varied widely in the types of patients that were treated, the doses of radiation that were given, and the outcome measures that were used. No meta-analysis of the trials has been done, as their overall numbers were still small (a total of 791 evaluable patients from all four trials) and the radiation doses varied widely (57). Nevertheless, although it did appear that there was no difference in median survival between those who were treated with PCI and those who were not, the group that was treated with PCI developed fewer brain metastases. Larger trials with more patients need to be done before a definitive statement regarding the role of PCI in NSCLC can be made. RTOG 0214 is an ongoing phase III RCT that plans to enroll 1000 patients who have had definitive treatment of their locally advanced NSCLC to evaluate the value of PCI vs. observation. Detailed QOL and toxicity data will be obtained and the large number of patients may be able to enough to determine whether there is a survival benefit from the addition of PCI.

2.7. Chemotherapy The use of chemotherapy in patients with brain metastases from any type of cancer has been difficult for a variety of reasons (58). Logically, it would make sense to use treatments that have efficacy against the systemic cancer. NSCLC, however, is generally chemoresistant. SCLC is very chemosensitive, but the existence of the blood–brain barrier, which prevents penetration of large molecular weight compounds, makes it difficult to achieve adequate concentrations of any drug into the brain parenchyma. Researchers argue that the presence of the tumor will make this barrier “leaky,” thus theoretically allowing more drug to penetrate the brain; however, these same patients are often treated with dexamethasone to control this same edema, which does partially restore the barrier. Furthermore, after the cancer has had time to metastasize to the brain, it has usually become chemoresistant. Most of the literature supporting the use of chemotherapy in lung cancer metastases is from nonrandomized trials. Patients usually have widespread systemic disease at the time of diagnosis of the brain metastases, resulting in a reduction of their life expectancy irrespective of their intracranial process. Additionally, response may be hard to determine as there is often damage from radiation, that is, “radiation necrosis” that can mimic treatment regrowth. Moreover, corticosteroids profoundly modulate contrast enhancement and may not be used uniformly in all patients on a trial. 2.7.1. Small Cell Lung Cancer The chemosensitivity of systemic SCLC has led to an interest in treatment of brain metastases with chemotherapy. Frequently these patients have had prior WBRT and have multiple metastases as well as systemic relapse, making radiosurgery or surgery unattractive. Postmus et al has carefully reviewed the results with various chemotherapy regimens (59). In general, multidrug regimens have led to radiographic responses in about half of patients, with less impressive results from single-agent trials. He concluded that chemotherapy’s efficacy is roughly comparable to that of WBRT. The utility of chemotherapy alone in SCLC patients with brain metastases who had not received prior WBRT was recently evaluated (12). Patients in this phase III trial were randomized to teniposide alone vs. teniposide plus WBRT. The rationale was that WBRT alone was inappropriate because almost all these patients had or would shortly have systemic relapse. The radiographic response rate (57%; 30% CR, and 27% PR) was significantly higher in the combined-therapy group than in the chemotherapy group (22%; 8% CR and 13% PR).

406

Part VII / Neurologic Complications of Specific Malignancies

Patients receiving teniposide alone had a significantly higher chance of failing in the brain, and median time to brain progression was longer in the combined modality group (11 vs. 7 weeks). The median survival in the two groups was identical: 3.2 vs. 3.5 months. The disappointing results in both arms of this trial may be partly attributable to some patients having prior exposure to etoposide (similar to teniposide), and perhaps newer agents or combinations such as carboplatin/paclitaxel might improve on these results. The authors argue that because of the extremely poor prognosis of this subgroup of patients, it is questionable whether WBRT should be given to patients with progressive disease outside the brain that does not respond to second-line chemotherapy, because most patients will die within a few weeks. 2.7.2. Non-Small Cell Lung Cancer The role of chemotherapy in brain metastases from NSCLC is not clearly defined. Various single-agent and multidrug regimens have been tested in small series of patients with brain metastases from NSCLC. Although occasional responses have been reported, generally response rates, time to progression, and survival have been disappointing (59). Development of more active regimens for systemic metastases from NSCLC will be necessary to improve treatment of brain metastases with this therapeutic modality. There have been numerous case reports that describe the response of NSCLC brain metastases to gefitinib, an EGFR tyrosine kinase inhibitor (60–62). A review of a series of case reports showed a pooled response rate of 18% in 62 patients (63). A prospective trial of 41 consecutive patients had a disease control rate (partial response + stable disease) of 27% (64). The response to the agent was seen early in the course of treatment (64). Favorable prognostic factors were patients with adenocarcinoma and prior treatment with whole-brain radiation (64). In all of these papers, gefitinib was well tolerated. A possible reason for the dramatic response may be the high incidence of EGFR mutations in patients with brain metastases from NSCLC, which may occur prior to the spread of the cancer to the brain (65). A phase II trial was started in Europe to treat 100 chemo-naïve NSCLC patients with stage IV disease (± brain metastases) with temozolomide, but was closed early after no response was seen in 25 patients (12 had brain metastases). The authors concluded that single-agent use of temozolomide had no efficacy in patients with stage IV NSCLC (66). Since the publication of the first edition, there have been two more studies that have investigated chemotherapy as the initial treatment of brain metastases from NSCLC. One was a randomized-controlled trial which looked at concurrent versus delayed WBRT on 171 patients with NSCLC using cisplatin and vinorelbine (67). The RRs, PFS, and 6-month survival were similar in both groups. Two-thirds of the group not treated initially with WBRT ended up requiring it. This suggested to the authors that initial chemotherapy with delayed WBRT was as good as combining the two modalities at the time of diagnosis. The other trial looked at the use of cisplatin, ifosfamide, and irinotecan in 30 patients with brain metastases from NSCLC (68). It reported a median survival of 12.8 months and a 50% partial response rate. 2.7.3. Approach to Progressive or Recurrent Disease Radiosurgery may be administered as the focal therapy for symptomatic, progressive lesions 3 cm or less in diameter. Retreatment with standard WBRT may also be considered. In a retrospective study of 86 patients who were re-irradiated for recurrent brain metastases, two-thirds had partial resolution or complete resolution of their neurologic symptoms, whereas one-third had no benefit or had progression after reirradiation. Most of the patients had no significant short-term side effects from retreatment. Favorable prognostic factors were solitary brain metastases without extracranial metastases and a reirradiation dose greater than 20 Gy (69). In a retrospective study, patients with NSLC who were reirradiated after repeat resection for recurrent metastases showed a trend towards increased survival (36). Chemotherapy may also be considered in patients with SCLC whose cancer is progressing in the brain but have stable systemic disease. Temozolomide, topotecan, or platinum can be administered in the salvage setting. A multicenter phase II trial using topotecan after WBRT in 30 SCLC patients with symptomatic brain metastases has also been done (70). A systemic RR of 29% was reported with a 33% RR of the brain metastases and a median time to progression of 3.3 months. The authors concluded that topotecan merited further investigation because it showed some promising results and was well tolerated, despite being used in heavily pretreated patients. There have been some older trials that investigated role of a combination chemotherapy in SCLC brain metastases, but there have been no randomized trials as of yet (58).

Chapter 21 / Neurologic Complications of Lung Cancer

407

2.8. Prognosis (27,29,39) In two large retrospective studies, median survival of lung cancer patients after diagnosis of metastases to the brain was three months with one year survival of about 10% (27,71). Performance status is the major determinant of survival in patients with brain metastases from cancer. Histologic subtype of lung cancer does not appear to be correlated with survival, although some studies have suggested that adenocarcinomas have a more favorable prognosis than other types (29,30,71). Patients older than 70 years had poorer survival rates than younger patients, even after correction for treatment differences (27,29). Progressive systemic disease from the lung cancer is a negative prognostic factor. In this study the presence of multiple brain metastases did not have the impact on survival that has been seen in some other studies (p < 0.0001) (71). Sen et al. suggested that response to glucocorticoids was a positive prognostic factor (30). Patients who responded to steroids showed a median survival of 4.3 months compared to only 1.6 months in nonresponders. The authors concluded that a “good” prognostic group included patients with inactive systemic tumor, high performance status, and response to steroids.

3. LEPTOMENINGEAL METASTASES 3.1. Epidemiology by Tumor Type Leptomeningeal metastases (LM) complicate 1–15% of cancer cases (72,73). Autopsy studies in lung cancer have found LM in about 10% of cases (8,9,14,74,75). The incidence of LM has been increasing in solid tumors as the overall survival for these patients has risen (72,73). In patients with SCLC who have survived for three years the incidence of LM increased to 25% (74). Almost half of patients with SCLC who relapse in the CNS do so in the meninges, which is the sole site of relapse in about 25% (15). Risk factors associated with the development of LM disease in SCLC are the presence of metastases in other parts of the neuraxis, extensive systemic disease, no response or partial response to therapy for the primary cancer, and male sex (74). A recent prospective review of 458 patients with SCLC has confirmed the findings of prior studies (76). The prevalence of LM, at time of diagnosis, was about 2% and the two-year cumulative incidence was 10%. The authors admit that these numbers may be underestimated as only symptomatic patients underwent lumbar puncture or MRI screening for leptomeningeal disease. Additionally, only 14% of patients underwent autopsy so a true prevalence rate was not established. Median survival after diagnosis of LM was still poor, only 7 weeks. However, they do note that better survival was seen in patients with spinal LM as opposed to cranial LM. They attributed this difference to the finding that cranial LM was associated more with brain metastases than with spinal LM; and the additive adverse effect of brain metastases decreased the survival of these patients. They also corroborated the finding that posterior fossa brain metastasis was more highly associated with LM than supratentorial brain metastasis. Their hypothesis was that this was secondary to direct extension of malignant cells into the CSF.

3.2. Manifestations In general, findings on neurologic examination tend to exceed the patient’s symptoms (73,77) (Table 4). The multifocal nature of leptomeningeal dissemination means there may be signs and symptoms at multiple different levels of the neuraxis. Combinations of mental status changes, cranial nerve, and spinal nerve root signs are suggestive of the diagnosis in cancer patients (73,77). The neurological signs and symptoms of LM patients with SCLC are similar to those found in other malignancies (74). About half of the patients will have signs and symptoms involving different levels of the nervous system, with almost all developing further CNS metastases during the course of their disease (74).

3.3. Symptoms 3.3.1. Parenchymal Headache occurs in approximately 25% of patients with LM (15,77). Episodic headache accompanied by nausea and vomiting may indicate the presence of plateau waves from increased intracranial pressure (73).

408

Part VII / Neurologic Complications of Specific Malignancies

Table 4 Presenting Symptoms and Signs in Patients with Leptomeningeal Metastases*

Headache Nausea/vomiting Seizure Confusion Paresthesia Weakness Cranial neuropathy Meningismus Cauda equina syndrome Myelopathy Back pain or radiculopathy Ataxia ∗

Patients with All Primary Types (n = 40) (77)

Patients with NSCLC (n = 32) (72)

19 – 5 2 – 2 16 3 2 2 1 3

11 – – – – 9 2 3 3 2 5

Adapted from ref 119.

Bifrontal cerebral dysfunction from hydrocephalus or parenchymal invasion may result in cognitive changes and gait apraxia (15). Cognitive dysfunction occurs in 25–33% of patients with LM, while seizures are a presenting symptom in only 3–12% of cases. 3.3.2. Cranial Nerves Cranial nerve symptoms include loss of visual acuity, diplopia, dysarthria, dysphagia, and hoarseness (73,77). Visual changes may occur in patients with cancer involving the optic chiasm or tract (15,77). Sudden hearing loss and vertigo may be seen with involvement of the VIIIth cranial nerve (15,77). Cranial nerve involvement is seen in almost 20% of SCLC patients with LM disease (74). 3.3.3. Spinal Cord/Nerve Root Spinal cord and nerve root symptoms occur in more than 50% of patients with LM and can be divided into radicular and leptomeningeal patterns of involvement (15,77). A higher incidence is seen in patients with SCLC, where two-thirds have spinal cord and nerve root symptoms including radicular pain and sensory loss, weakness (lower extremities more often than upper extremities), and autonomic failure with sexual dysfunction and sphincter disturbances (71). Leptomeningeal infiltration results in neck or back pain with nuchal rigidity.

3.4. Signs 3.4.1. Parenchymal Signs of parenchymal brain dysfunction, such as aphasia, hemiparesis, and hemisensory loss are uncommon, and when present suggest coexisting cerebral metastases or significant invasion of tumor cells into brain along Virchow–Robin spaces (15). Mental status abnormalities are discovered in approximately 50% of patients. In SCLC patients with LM disease, 60% were found to have either limb weakness or mental status changes. None of the patients had seizures (74). Isolated cortical signs or symptoms occur in almost 30% of patients (74). 3.4.2. Cranial Nerves Cranial nerve signs discovered during examination are usually mild and involve paresis of extraocular muscles, decreased facial sensation, facial paresis, and hearing loss (15,78). 3.4.3. Spinal Cord/Nerve Roots When tumor cells infiltrate the parenchyma of the spinal cord, the presentation may be predominantly of upper motor neuron type (weakness, spasticity, hyperreflexia, and Babinski signs). Lower motor neuron findings

Chapter 21 / Neurologic Complications of Lung Cancer

409

(weakness, hypotonia, areflexia, and fasciculations) are from spinal nerve root infiltration. The lower extremities are more commonly involved than the upper extremities, reflecting the lengthy course of the nerves of the cauda equina. Isolated spinal symptoms can occur in almost 30% of patients with SCLC (74).

3.5. Radiologic Findings MRI is more sensitive than CT in detecting LM (73). Both have a high false negative rate, and the radiographic findings are often nonspecific (79,80). The best use of MRI is the detection of bulky disease. It may also be of benefit in the detection of focal leptomeningeal disease, where CSF cytology may have a higher false-negative incidence (81). Contrast enhancement may be seen diffusely or as multiple subarachnoid nodules, especially along the cauda equina. Double-dose gadolinium may increase the sensitivity of MRI (79,80,82). The presence of communicating hydrocephalus is a nonspecific but suggestive finding. When considering patient eligibility for intrathecal chemotherapy, the demonstration of CSF flow patency or the reversibility of its obstruction should be taken into account (83,84). Radionuclide cisternography, which examines CSF flow, can be used to exclude the presence of subarachnoid blocks due to LM disease (83,84).

3.6. Diagnosis Dural puncture with CSF analysis is the definitive diagnostic test for LM. Cerebral metastases with mass effect and obstructive hydrocephalus are relative contraindications to dural puncture (85). All patients being screened for LM should first undergo brain CT or MR scans to exclude such conditions. An adequate volume of CSF (5–10 ml) should be obtained and delivered promptly to the laboratory to maximize diagnostic yield on cytopathology. Opening pressure is elevated, due to impaired CSF absorption, in approximately 60% of SCLC patients with LM. The cell count reveals a mild pleocytosis with a lymphocytic predominance in most cases (15,74). Subarachnoid hemorrhage is uncommon in LM. Protein is typically increased because of blood–brain barrier disruption. Hypoglycorrhachia is present in 68% of SCLC patients with LM (74). CSF cytology is the definitive diagnostic test for LM, but only 50% of patients with LM will have a positive result on the first lumbar puncture (15,73). With serial dural punctures, malignant cells can be identified in up to 90% of patients. In patients with a negative first LP and a high degree of clinical suspicion, at least two additional lumbar punctures should be performed over several days to increase diagnostic yield (15,74). A presumptive diagnosis of leptomeningeal metastases may occasionally be made in the absence of positive cytology in patients with decreased glucose, increased protein, lymphocytic pleocytosis, and negative microbiological studies. In such circumstances, neuroimaging may be useful in supporting the presumptive diagnosis. There is some evidence that cervical level punctures are more sensitive than lumbar level punctures in the diagnosis of LM in breast cancer patients (73,77). 3.6.1. Other Tests Monoclonal antibodies to specific tumor proteins can increase the detection of malignant cells by about 10% (15,73,78). Biochemical markers are generally not useful for patients with lung tumors (73,77,78). Creatine kinase BB may be useful as a screening test for leptomeningeal disease in patients with SCLC. In one series it was found to have a sensitivity of 88% and a specificity of 100% (23). In addition, CSF gastrin-releasing peptide was able to establish the diagnosis of LM spread from SCLC in a patient who had repeatedly negative CSF (86). Flow cytometry, though still under investigation, is occasionally helpful when a routine cytology is negative (15,77,78). Abnormal results include the detection of aneuploid cells and the presence of CEA on cell surfaces.

3.7. Management Treatment of LM should begin as soon as possible after the diagnosis is established. Because the disease involves the subarachnoid space, it can extend from the cerebral convexities to the lumbar cistern, making treatment of the entire neuraxis of concern. In general, focal radiation is employed for treatment of bulky, nodular disease and sites of rapidly progressive symptoms (usually of cranial nerve or cauda equina origin). Intrathecal chemotherapy is used to treat the entire subarachnoid space (74,77,87).

410

Part VII / Neurologic Complications of Specific Malignancies

3.7.1. Radiation Focal external beam RT is an effective means of alleviating LM symptoms (15,77). Irradiation of the entire neuraxis is usually ill-advised due to the amount of bone marrow (40% of the body’s total) affected by this approach. RT should be employed for sites of focal symptoms and sites of bulky disease seen on MRI (15,77). Focal radiation to the skull base can be employed in place of WBRT in patients with cranial nerve symptoms. WBRT is indicated for patients with hydrocephalus, seizures, or other signs of brain parenchymal involvement (15). PCI, which prevents CNS metastases, is not effective in preventing LM in SCLC (15). 3.7.2. Intrathecal Chemotherapy LM is usually treated with administration of chemotherapy directly into the CSF. This allows a high concentration of drug to be delivered to the site of disease (15,77,87). Treatment with systemically administered chemotherapy is usually ineffective as most antineoplastic agents penetrate the blood CSF barrier poorly. Intrathecal chemotherapy may be administered by lumbar puncture (LP) or via an Ommaya reservoir that connects to the lateral ventricle by a cannula. Ommaya reservoirs are generally preferred because of ease of administration of the drug and better distribution throughout the entire CSF system (77). Complications of intraventricular catheter insertion occur in about 5–10% of cases and mainly involve infection or subdural hematoma (77,88). The most commonly used intrathecal chemotherapy agents are methotrexate (MTX), thiotepa, and cytosine arabinoside (ara-C). A liposomal form of ara-C has recently become available. Agents such as mafosphamide (a derivative of cyclophosphamide), monoclonal antibodies, interferon, and interleukin-2 are under investigation. Single agent therapy is usually employed because thus far combination chemotherapy is not more effective and is also associated with a higher incidence of side effects (15,73,77,87). MTX is usually administered at a dose of 12 mg twice a week until the CSF cytology becomes negative or patients experience symptomatic improvement, at which point the frequency is gradually decreased to once every 2 to 4 weeks. Treatment is continued for at least 3 to 6 months in patients who respond (15,73). Oral leucovorin should be administered to patients who develop mucositis or are receiving concurrent systemic chemotherapy. Ara-C is given at a dose of 50 mg twice a week on the same schedule as MTX. Liposomal ara-C has the advantage that it is administered every 2 weeks; however, it has a much higher rate of chemical meningitis than does standard formulation ara-C. It is difficult to know when to recommend intrathecal chemotherapy in an individual patient. Treatment is most often considered in patients with SCLC, in patients with good performance and relatively indolent systemic disease, and in patients without bulky LM on neuroimaging. A retrospective study found that a combination therapy of intrathecal methotrexate and irradiation to symptomatic regions of the neuraxis cleared the CSF of malignant cells in 50% of SCLC patients. Half of these patients had complete or nearly complete resolution of signs or symptoms of their disease. This contrasts with patients treated only with irradiation who experience neither clearance of malignant cells in their CSF nor symptomatic benefit. Four out of five patients treated only with intrathecal MTX cleared their CSF of malignant cells, but only one had symptomatic relief of signs and symptoms (74). 3.7.3. Ventricular Peritoneal Shunting Ventricular peritoneal shunting (VPS) is used for the treatment of symptomatic hydrocephalus. Improvement of symptoms of intracranial hypertension can be seen in 77% of patients, especially with the symptoms of encephalopathy (88). The existence of the shunt is problematic if intrathecal chemotherapy is planned, as the drug will drain out of the ventricle into the abdominal cavity. However, the risk that a VPS may introduce cancer cells into the peritoneal cavity has proven negligible in clinical practice (15,88). The presence of hydrocephalus also raises concern regarding the distribution of intrathecal chemotherapy. Therefore, the combination of VPS and WBRT is commonly employed. 3.7.4. Biologic Therapy Gefitinib is the first oral epidermal growth factor inhibitor that came on the marker. It has been used in the treatment of NSCLC. In patients with NSCLC it was not shown to significantly prolong overall survival, though one publication attributed this to dosing at one-third of its MTD (89). Interestingly, one paper showed that the

Chapter 21 / Neurologic Complications of Lung Cancer

411

CNS (both BM and LM) were a frequent source of failure in patients who had a systemic response of NSCLC to gefitinib (90). A recent case report followed a patient with NSCLC who was treated with geftinib and developed LM (91). Interestingly, increase of the gefitinib dosing caused radiographic and clinical response, but the patient eventually died as his systemic disease developed resistance to the agent. Another case report has also confirmed the response of LM disease from NSCLC to gefitinib, although at a lower dose (92).

3.8. Prognosis Average survival of patients with untreated LM is one-and-half to two months after initial diagnosis (15,73,88). LM from SCLC is usually more responsive than other histologies. Seventy-five percent of these patients will experience symptomatic improvement with a concomitant improvement in their CSF results. Approximately 25% of all patients will have some neurological improvement (73). Nonetheless, the overall median survival after treatment for LM disease in SCLC is only six weeks (74).

4. EPIDURAL SPINAL CORD COMPRESSION 4.1. Incidence The incidence of clinically diagnosed epidural spinal cord compression ESCC in patients with systemic cancer is approximately 5% (93,94). There are approximately 18,000 cases of symptomatic ESCC in the United States each year (93). Danish investigators reported a 5–15% incidence of ESCC in patients with all types of lung cancer. The incidence of ESCC in SCLC is reported to be 3–8% (14,95). ESCC is the initial presentation of cancer in up to 20% of patients with this syndrome (93,94). Patients with lung cancer, hematological malignancies, and cancer of unknown primary are especially likely to present with symptoms and signs of ESCC (96). Patients with SCLC have a tendency to develop ESCC early after diagnosis of their primary cancer (87% within the first three months), whereas a similar percentage is achieved after 30 months in patients with NSCLC. In about 60% of patients with ESCC, vertebral metastases occur in the thoracic spine, 25% in the lumbar spine, and the remaining 15% in the cervical spine (94,97,98). Virtually identical figures pertain to ESCC specifically from lung cancer (95). A large retrospective study found that 29% of patients with ESCC from lung cancer have multiple, synchronous epidural lesions (95).

4.2. Manifestations: Signs and Symptoms 4.2.1. Pain Back pain is the initial symptom in the majority of lung cancer patients with ESCC, occurring in 77% of cases. Pain was either localized to the spine (37%) or radicular in nature (40%) (94,95,99). Pain occurs when vertebral metastases invade the pain-sensitive periosteum, dural or paravetebral soft tissues. Pain can also be caused by compression of the spinal nerve roots from tumor or bone fragments from pathologic fractures. Secretion of prostaglandins by the tumor can promote tumor invasion and cause increased pain sensitivity. Pain may occur 1–2 months before the onset of other symptoms and signs (94). Treatment is most effective when pain is the only symptom. Pain localized to the affected vertebral body is usually steady, aching, and midline. It may be exacerbated by Valsalva maneuver or movement in the case of spinal instability and is usually more severe at night and in the supine position. In contrast, pain from a herniated intervertebral disk or compression fracture is usually alleviated in the supine position. Localized tenderness to palpation is useful but not precise in identifying the affected spinal level (15). Radicular pain is caused by epidural extension or vertebral collapse with compression of the nerve roots within the spinal canal or as they exit through the intervertebral foramen. Cervical root compression may produce pain or paresthesias in one or both arms. Thoracic compression may cause a tight band around the chest or abdomen, while patients with lumbar nerve root compression have radiation of pain down one or both legs (15). Neck movement and straight leg raising may exacerbate radicular pain. Percussion of the spine helps to localize the involved vertebral body.

412

Part VII / Neurologic Complications of Specific Malignancies

4.2.2. Weakness Weakness is the next most common symptom of ESCC (93,94). At presentation, about two-thirds of lung cancer patients had profound motor symptoms and were unable to walk (95). Upper motor neuron weakness occurs with compression of the corticospinal tracts and is usually associated with other signs of myelopathy including hyperreflexia and Babinski signs. Myelopathy usually develops weeks after onset of pain and may progress rapidly within a few days of onset (78). Initially there is symmetrical, proximal leg weakness, producing difficulty climbing stairs and arising from the seated position. Acute weakness may occur on occasion with sudden onset paraplegia, flaccid muscle tone, and areflexia in a “spinal shock” pattern of cord injury. This presentation is often secondary to acute hemorrhage in the epidural tumor, or to collapse of an involved vertebral body. Recovery of ambulation is unlikely to occur in patients presenting in this manner. Lower motor neuron weakness (LMN) occurs when the cauda equina is compressed. Symptoms include decreased muscle tone and absent reflexes (15). Weakness from compression of the cauda equina is usually patchy in distribution and more distal than proximal. 4.2.3. Sensory Loss Sensory loss usually occurs simultaneously with weakness. Symptoms may include decreased sensation or paresthesias in the limbs or trunk below the level of cord compression (94). Two-thirds of lung cancer patients had a sensory level below the area of (ESCC), with 10% retaining normal sensory function (95). However, it should be noted that the sensory level cannot be used to accurately predict the site of ESCC. Cauda equina lesions may result in painful paresthesias of the feet and lower legs. With cord compression, the earliest sensory changes are proprioceptive and vibratory loss (94). Lhermitte’s sign is an electrical sensation, which, upon neck flexion, radiates from the cervical region down the spine and into the legs. In cancer patients this sign may be due to a cervical epidural metastasis, prior irradiation to the cervical spine, or a side effect of certain chemotherapeutic drugs. 4.2.4. Autonomic Dysfunction Bowel or bladder dysfunction, sweating abnormalities, and impotence are rare as isolated or initial manifestations of ESCC (15,94). However, patients with ESCC affecting the tip of the conus medullaris (often secondary to a T11-T12 vertebral metastasis) may have bowel or bladder changes before weakness sets in. In lung cancer patients, 59% had severe bladder dysfunction (usually urinary retention) requiring catheterization; another 12% had milder symptoms of urgency (95). Sphincter disturbances can produce retention or incontinence, the former being more common. Patients may not be aware of severe urinary retention because of sensory loss. Urinary retention may result in frequent episodes of small volume voiding as bladder compliance is exceeded. Urinary retention is usually seen in more rapidly evolving cases of ESCC and is associated with marked sensory and motor losses in the legs (15). Examination of anal tone and assessment of bladder function (post-void residual) can be useful to assess sphincter function when ESCC is suspected. 4.2.5. Ataxia Spinocerebellar signs, including gait ataxia, are often obscured by weakness and sensory loss, and can persist after weakness has improved with treatment. Cerebellar signs suggest that the spinocerebellar tracts of the spinal cord may be involved (15). On rare occasions, ataxia may be the only sign in ESCC, and a diagnostic delay may occur while the cause of cerebellar dysfunction is sought.

4.3. Radiologic Findings and Diagnosis 4.3.1. MRI MRI is the most sensitive and specific test for spinal metastases (15). Because MRI is sensitive to changes in the bone marrow, this technique is able to detect the presence of metastatic disease within the vertebral bodies. The addition of gadolinium is not crucial for the identification of epidural metastases, and may mask bony involvement as T1 hypointense metastases may become isointense after contrast administration (100,101).

Chapter 21 / Neurologic Complications of Lung Cancer

413

Because synchronous epidural metastases occur at multiple levels in approximately 20%–30% of cases, the entire vertebral column should be imaged (94,100). 4.3.2. Plain X-Rays Plain X-rays may be useful for identifying vertebral body metastases (15). Abnormalities include erosion of a pedicle and collapse of the vertebral body. Plain X-rays are suggestive of epidural extension in a vertebral body collapse of more than 50% or pedicle erosion (15,94). 4.3.3. CT CT scans are more sensitive than plain X-rays or bone scans for identifying vertebral metastases and, like MRI, can image paravertebral disease. 4.3.4. Myelography MRI has replaced myelography as the definitive radiographic test for epidural metastases (15,101). However, if the patient is unable to tolerate MRI because of claustrophobia or severe pain, or if there is a strong clinical suspicion of cord compression despite a negative MRI of poor quality, CT-myelography can be useful. The entire spinal canal should be imaged.

4.4. Management 4.4.1. Glucocorticoids Glucocorticoids can improve the symptoms of ESCC, especially pain. In addition, patients treated with highdose dexamethasone plus RT have a better functional outcome than patients treated with RT alone (15,102). 4.4.2. Surgery Surgery should be considered for diagnostic purposes in patients without known cancer, in patients with ESCC in the setting of limited or indolent systemic disease, in cases of RT failure, or in cases of spinal instability (15). A recent phase III randomized trial in single metastases to the spine showed a clear advantage of surgery followed by radiation compared to radiation alone (103). Important caveats were that most of the patients in the trial came from a single institution and there may have been a selection bias benefiting treatment with surgery (high percentage of patients with radiation-resistant tumors and instability of the spine). Vertebral body resection allows for complete removal of tumors. Bone graft or synthetic cement replaces the resected vertebrae. Stabilization of the spinal column is necessary and requires intact vertebral bodies above and below the lesion. Laminectomy can be employed for patients with compression of the spinal cord by metastases to posterior spinal elements (15). 4.4.3. Radiation RT should be implemented soon after diagnosis is made in patients who do not receive surgery. The radiation port is centered on the site of ESCC and includes two vertebral bodies above and below. 4.4.4. Approach to Recurrent Disease Retrospective series of patients with ESCC put the incidence of recurrence between 13% and 20%, with half of the recurrences occurring at a different level than the original site (104–106). Half of the patients who survived two years after diagnosis and nearly all of the patients at three years had recurrence of their disease. Reirradiation of recurrent ESCC may preserve ambulation with low risk of radiation injury to those patients with limited duration of expected survival (106).

4.5. Prognosis The best results from treatment are for pain control, with two-thirds of patients with ESCC from all cancers having durable improvement of pain (98). This figure may be even higher for patients with SCLC treated with radiotherapy. One series reported significant pain relief in eight of nine patients, reflecting the radiosensitive

414

Part VII / Neurologic Complications of Specific Malignancies

nature of this tumor. The majority of patients with ESCC from all cancers who are ambulatory at the initiation of treatment remain so a year later (15,97,98). In lung cancer patients, the number of ambulatory patients improved from 41% to 52% after treatment. If the patient was paraplegic at the beginning of treatment, the chances of walking again are poor (95). However, rare patients have regained ambulation several weeks to months after RT despite severe weakness at initiation of treatment. Patients with SCLC again may be particularly likely to improve, with three of nine initially nonambulatory patients regaining the ability to walk following radiotherapy. Patients with ESCC from all cancers who are paraparetic at start of therapy have a 25–70% chance of being ambulatory after treatment (15,97,98). Patients who do not regain ambulation have shorter survivals. There were no significant improvements in bladder function in lung cancer patients after treatment of ESCC (95). Prognostic factors for survival and function include tumor histology (lung cancer is an unfavorable histology), good initial performance status, complete surgical removal of the metastasis, and cervical location of the metastasis (97,98,104). Patients with multiple epidural metastases had a shorter median survival than patients with a single, spinal metastasis (104). Patients with SCLC showed no difference in outcome regardless of mode of treatment. However, the NSCLC patients did significantly better when treated with the combination therapy of laminectomy and then RT. Only 39% of these patients improved with RT alone, 47% with laminectomy alone, while 82% improved with combined therapy (p = 0.03) (95). Survival also improved in the combined therapy group (median of 3.5 months, with range of 0–132 months) compared with the RT group (median of 1 month, range of 0–59 months) or the laminectomy group (median of 1.5 months, range of 0–32 months) (95). Although there is no significant difference in survival between different lung cancer histologies, the trend shows survival for squamous cell carcinoma to be the worst (median of 1 month, range of 0–32 months) and SCLC being the best (median of 2.5 months, range of 0–132 months) (95). Overall, only 9% of lung cancer patients survived more than one year (95).

5. INTRAMEDULLARY SPINAL CORD METASTASES Lung cancer accounts for about 50% of intramedullary spinal cord metastases (ISCMs), and SCLC accounts for about 60% of ISCM associated with lung cancer (107). In a large unselected series, ISCM developed in 3/203 SCLC patients (1.5%). In another series of 50 patients treated with a specific chemoradiation protocol, ISCM developed in 6/50 (12%). ISCM and leptomeningeal carcinomatosis frequently co-exist in SCLC patients (between 17% and 54% of cases) (107,108). Clinical features and diagnosis of ISCM in lung cancer patients do not differ from these issues in other malignancies and are discussed in Chapter 11. ISCM in patients with SCLC is particularly responsive to radiotherapy and may be undetectable at autopsy in irradiated patients (109). As with ISCM from other malignancies, early diagnosis and rapid institution of radiotherapy offer the best hopes of maintaining or improving neurologic function.

6. SKULL BASE METASTASES Skull base metastases can invade the bones surrounding the middle cranial fossa, as well as the parasellar regions, orbits, occipital condyle, and jugular foramen. Lung, breast, and prostate cancer are the primary tumors that most commonly produce metastases to the base of the skull (16). Clinical syndromes and management are fully disucssed in Chapter 10.

7. PLEXUS AND PERIPHERAL NERVE METASTASES 7.1. Brachial Plexus 7.1.1. Manifestations 7.1.1.1. Symptoms. Pancoast tumor is a term used to describe tumors that originate in the apex of a lung and compress or invade the brachial plexus (15). As the tumor grows into the plexus from below, the C8 and T1 fibers of the cord (which become the ulnar nerve) are usually first affected.

Chapter 21 / Neurologic Complications of Lung Cancer

415

Pain, the most common initial symptom, rapidly progresses from a dull, throbbing feeling in the back or lateral part of the shoulder to involve the medial portions of the upper arm, elbow and forearm. Involvement of the C8 and T1 fibers also results in numbness and tingling in the fourth and fifth fingers, although these symptoms are usually less noticeable by the patient. Weakness in the intrinsic muscles of the hand makes it difficult for the patient to grasp small items. As the tumor spreads to involve the rest of the plexus, involvement of the medial and radial nerve occur, resulting in weakness in the flexors and extensors of the hand and wrist and in the extensors of the elbow. Involvement of muscles innervated by the upper plexus, such as the biceps and brachioradialis, is a late-stage development. Medial progression of the tumor to involve the sympathetic trunk may cause a partial Horner’s with ptosis and anhydrosis. 7.1.1.2. Signs. The neurological examination may be normal at an early stage despite sensations of numbness and tingling. Visual and tactile examination may show smoothing of the well-defined clavicular boundaries with supraclavicular or axillary adenopathy. Percussion of these areas may reproduce the sensory symptoms in the arm. Extension of the brachial plexus with full range of motion of the arm can also reproduce these symptoms. 7.1.2. Radiologic Findings Apical chest X-ray views can be obtained for examination of the superior part of the lung. Imaging of the brachial plexus with either CT or MRI generally demonstrates tumor. Extension of the tumor into the spinal canal via the neural foramina can also be seen. 7.1.3. Diagnosis Diagnosis is usually made by typical radiographic findings. EMG can document involvement of the plexus but is rarely indicated. 7.1.4. Prognosis Debilitating pain and weakness are the most serious complications of this disorder. 7.1.5. Treatment 7.1.5.1. Radiation. Radiation can provide effective pain relief and stabilize or improve motor function. Full recovery can occur if the syndrome is diagnosed early. However, restoration of neurologic function is less likely if the symptoms are advanced. Chemotherapy may result in alleviation of symptoms and signs. 7.1.5.2. Analgesics. Treatment for pain usually includes the use of analgesic agents and opiods. Newer agents such as gabapentin, which are specific for neuropathic pain, should also be employed. 7.1.5.3. Surgery. Failure to control pain may result in consideration of rhizotomy or cordotomy. Alternatively, in the setting of a reflex sympathetic disorder syndrome, blocking the stellate ganglion may produce some benefit.

7.2. Recurrent Laryngeal Nerve 7.2.1. Manifestations The recurrent laryngeal nerve, a distal branch of the vagal nerve, innervates the muscles of the larynx. Its compression by metastatic deposits of lung cancer causes a weak cough and hoarseness secondary to vocal cord paralysis. Dysphagia can occur, with aspiration of liquids. 7.2.1.1. Treatment. Treatment of lung cancer by radiation and chemotherapy can occasionally improve the symptoms. Aspiration may be reduced by laryngoplasty of the paralyzed vocal cord.

7.3. Atypical Facial Pain Patients with lung cancer may describe constant aching unilateral facial pain often located around the ear. The pain is invariably ipsilateral to the thoracic tumor and is likely attributable to intrathoracic vagus nerve compression (Fig. 1). Local management with surgery or radiotherapy usually alleviates the symptom.

416

Part VII / Neurologic Complications of Specific Malignancies

Fig. 1. This 51-year-old man presented with a fall and was found to have NSCLC with a left frontal brain metastasis. He was treated with SRS and systemic chemotherapy and was able to return to work. One year later he returned with right facial and head pain with slight difficulty swallowing. Though brain MRI showed no new lesions, a chest CT showed substantial growth of a right mediastinal mass over the preceding 3 months (B: 3 months after A). He received chest RT with some amelioration of pain.

8. INDIRECT COMPLICATIONS OF LUNG CANCER 8.1. Paraneoplastic Neurologic Syndromes The neurologic paraneoplastic syndromes constitute a group of rare disorders that may affect the nervous system at all levels, and not uncommonly involve multiple sites within a single individual (see Chapter 15 for overview of paraneoplastic syndromes) (15). SCLC appears to have the highest incidence among all cancers of associated paraneoplastic neurologic syndromes; approximately 3% of SCLC patients have paraneoplastic disorders associated with diagnostic antibodies (14,15,110). In contrast, clinically significant paraneoplastic neurologic syndromes probably occur in fewer than 1% of all cancer patients (15,111). 8.1.1. Manifestations 8.1.1.1. Encephalomyelitis. Paraneoplastic limbic encephalomyelitis (PLE) is characterized by memory loss, confusion, personality changes, and hallucinations (112). Less commonly, involvement of the brainstem results in cranial nerve symptoms (such as deafness, vertigo, and diplopia), weakness, central respiratory failure, or involvement of the autonomic system. MRI abnormalities consist of T2-weighted changes in the medial temporal lobes and brainstem that do not enhance. Neuropathological findings show multifocal inflammatory infiltrates (110). In a recent review of 50 patients with PLE (112), lung cancer was the most commonly associated malignancy, making up 50% of the cases. In 60% of patients the neurological symptoms preceded the diagnosis of cancer by a median interval of 3.5 months. One-half of MRIs showed changes in the limbic system. Sixty percent had positive antineuronal antibodies, and 18 of these patients had anti-Hu. In the patients with anti-Hu antibodies, 94% had SCLC. Thirty-eight percent of the anti-Hu patients showed clinical improvement over time, compared with 64% of the patients without anti-Hu antibodies. Treatment directed against the primary tumor was more effective than immunosuppressive therapy. 8.1.1.2. Paraneoplastic Cerebellar Degeneration (PCD). Symptoms of PCD begin as mild truncal ataxia, evolving over the course of several weeks to months to include the limbs and trunk. Dysarthria, nystagmus, vertigo, diplopia, and oscillopsia are common symptoms. After a period of subacute progression, the disease usually stabilizes, leaving the patient severely disabled. Signs are bilateral, though one side can be more affected than the other (15,110). Although signs and symptoms are primarily confined to the cerebellar system, other areas of the nervous system may be affected, producing altered mental status, extrapyramidal signs, hearing loss, hyperreflexia, and peripheral neuropathy (15). The most common association with PCD is lung cancer (15). In patients with SCLC, this syndrome may exist in combination with a widespread encephlomyeloneuritis (113). Detection of anti-Hu or anti-Yo antibodies in

Chapter 21 / Neurologic Complications of Lung Cancer

417

serum or CSF can be useful to confirm the clinical suspicion and to help the search for a primary tumor. Anti-Hu is associated with SCLC whereas anti-Yo is associated with breast and gynecologic cancers in women (113). Over time, neuroimaging shows cerebellar atrophy. At autopsy, there is specific loss of Purkinje cells (110,113). 8.1.1.3. Subacute Sensory Neuronopathy (SSN). This syndrome occurs in patients with SCLC or with autoimmune diseases such as Sjogren’s syndrome. Serum from affected patients usually contains the anti-Hu antibody (114–116). Anti-Hu antibodies have a sensitivity of 82% and specificity of 99% in confirming the diagnosis of SSN (116). The syndrome starts with burning dysesthesias involving the legs and progressing over days and weeks to involve the arms and the face. A severe sensory ataxia results that mimics cerebellar dysfunction. Both small and large neurons are affected, which helps to differentiate SSN from the large fiber neuropathy due to cisplatin. Muscle stretch reflexes are lost, whereas normal muscle strength is maintained (15,110). CSF findings include an inflammatory pleocytosis and high titer of anti-Hu antibody. EMG/NCS studies show normal motor nerve findings but small or absent sensory potentials. Pathologically, there is inflammation of the dorsal root ganglion with loss of neuronal cell bodies (15,110). Evidence suggests that the neuronal damage is mediated by cellular immunity with the anti-Hu antibodies serving as a marker for the condition (116). 8.1.1.4. Autonomic Neuropathy. Autonomic neuropathy is a rare entity that can exist by itself or in combination with other sensory neuropathies. It is primarily associated with small cell lung cancer, usually as part of the anti-Hu syndrome. Typical symptoms include orthostatic hypotension, gastroparesis, neurogenic bladder and papillary asymmetries. Symptoms are usually progressive, but may stabilize with successful treatment of the primary cancer (15,114). 8.1.1.5. Lambert–Eaton Myasthenic Syndrome (LEMS). LEMS is a paraneoplastic disorder of the neuromuscular junction that occurs in patients with SCLC or autoimmune conditions. Patients complain of increasing weakness and fatigability in the proximal musculature. Weakness can involve respiratory muscles, but unlike myasthenia gravis, the oculobulbar muscles are usually spared. Another difference compared to myasthenia gravis is that strength increases with initial effort before the weakness returns. These patients have other cholinergic disturbances such as impotence, constipation and a dry mouth. Freeze-fracture electron micrographs show loss of voltage-gated calcium channels in the presynaptic active zone (15,110). Antibodies to voltage-gated calcium channels at the presynaptic terminal are pathogenic. Electromyography is used to diagnose LEMS, with a decrement of the compound muscle action potential (CMAP) at slow rates of stimulation but an increment at faster rates. This is in contrast to normal musculature which has a similar size CMAP at slow or fast rates or to myasthenia gravis, which has a decremental response to both (15). 8.1.1.6. Polymyositis (PM)/Dermatomyositis (DM). Both of these disorders are inflammatory myopathies of autoimmune origin, and are only rarely associated with an underlying malignancy. DM is more likely than PM to be associated with cancer. Lung cancer is a frequent cause of the paraneoplastic form of this disease. Laboratory findings include autoantibodies (e.g., anti-Jo) and increased serum creatine kinase level. EMG confirms the presence of a myopathy. The conditions respond to immunosuppressive therapy. Dermatologic and muscle symptoms improve in some patients when the underlying primary cancer is treated (15). 8.1.2. Management 8.1.2.1. Treatment of Primary Cancer. Treatment directed at the primary lung tumor stabilizes or partially improves paraneoplastic syndromes in approximately one-quarter of patients. These improvements have been observed between 3 weeks and 3 months after surgery (117). 8.1.2.2. Immunosuppressive Therapies. Because of the apparent role of autoimmunity in many of the paraneoplastic conditions, there has been an attempt to employ immunosuppressive therapies. In general, these approaches have yielded little in the way of benefit, but may be considered in patients for whom other options, such as removal of the primary tumor, are not effective or possible. Treatments have included administration of glucocorticoids, intravenous immunoglobulin, and plasmapheresis. In an analysis of cases published in the literature, 22% of patients with a variety of paraneoplastic neurologic disorders appeared to respond to plasmapheresis. A recent case series of nine anti-Hu and anti-Yo patients

418

Part VII / Neurologic Complications of Specific Malignancies

showed improvement in four of them after treatment with rituximab (118). Patients with LEMS have responded to glucocorticoids and plasmapheresis. Dermatomyositis responds to glucocorticoids (16). 8.1.2.3. Manipulation of Neuromuscular Transmission. Pyridostigmine has some activity in patients with LEMS. Acetylcholine release can be enhanced by blocking potassium channels involved in terminating the action potential, prolonging the time available for calcium entry into the cell, and enhancing acetylcholine release. 4-aminopyridine improves muscle strength, but is associated with an unacceptable lowering of the seizure threshold. 3,4-diaminopyridine (3,4-DAP) has been shown to increase strength and is associated with a much lower incidence of seizures. The medication is well-tolerated and is very effective, but is at present only available as an investigational agent. Efforts are being made to have it commercially available.

9. CONCLUSION The treatment of neurological complications from lung cancer is rapidly evolving. Radiation therapy is being further refined and radiosurgery more frequently used in the treatment of brain metastases. Chemotherapy, long avoided because of the blood–brain barrier, is now being considered following the discovery of new drugs that are able to penetrate the CNS. Randomized, controlled trials are being designed and conducted to further define and evaluate the role of these newer methods in the treatment and control of metastases that spread to the nervous system. As improved systemic treatments allow patients to live longer with their systemic disease, we see the relapse of the disease process in the nervous system. Continued advances in radiation and chemotherapy will soon combat the cancer in its final refuge and increase the survival of these patients.

REFERENCES 1. 2. 3. 4. 5.

6. 7.

8. 9. 10. 11. 12.

13. 14. 15. 16. 17. 18. 19. 20. 21.

Smith RA, and Glynn TJ. Epidemiology of lung cancer. Radiol Clin North Am 2000;38(3):453–470. Hoffman PC, Mauer AM, Vokes EV. Lung cancer. Lancet 2000;355:479–485. Deslauriers J, Gregoire J. Surgical therapy of early non-small cell lung cancer. Chest 2000;117(4, supplement):104S–109S. Bunn PA, Kelly K. New combinations in the treatment of lung cancer. Chest 2000;117:138S–1343S. Pignon JP, Tribodet H, Scagliotti GV, et al. Lung Adjuvant Cisplatin Evaluation (LACE): A pooled analysis of five randomized clinical trials including 4,584 patients (abstract). J Clin Oncol 2006;24:366s. (Abstract available online at www.asco.org/portal/site/ ASCO/menuitem.34d60f5624ba07fd506fe310ee7a01d/?vgnextoid). Johnson DJ. Evolution of cisplatin-based chemotherapy in non-small cell lung cancer. Chest 2000;117(4):133S–137S. Curran WJ, Scott C, Langer C et al. Phase III comparison of sequential vs concurrent chemoradiation for patients (pts) with unresected stage III non–small cell lung cancer (NSCLC): Initial Report of Radiation Therapy Oncology Group (RTOG) 9410. Proc ASCO 2000;19:484a. Elliot JA, Osterlind K, Hirsch FR et al. Metastatic pattern in small-cell lung cancer. J Clin Oncol 1987;5:246–254. Stenbygaard LE, Sorenson JB, Olsen JE. Metastatic pattern at autopsy in non-resectable adenocarcinoma of the lung. Acta Oncologica 1997;36(3):301–306. Tang SG, Tseng CK, Tsay PK et al. Predictors for patterns of brain relapse and overall survival in patients with non-small cell lung cancer. J Neuro-oncol 2005;73(2):153–161. Gaspar LE, Chansky K, Albain KS et al. Time from treatment to subsequent diagnosis of brain metastases in stage III non-small-cell lung cancer: a retrospective review by the Southwest Oncology Group. J Clin Oncol 2005;23(13):2955–2961. Postmus P, Haaxma-Reiche H, Smit E et al. Treatment of brain metastases of small-cell lung cancer: comparing teniposide and teniposide with whole-brain radiotherapy: a phase III study of the European Organization for the Research and Treatment of Cancer Lung Cancer Cooperative Group. J Clin Oncol 2000;18:3400–3408. Glantz MJ, Choy H, and Yee L. Prophylactic cranial irradiation in small cell lung cancer: rationale, results, and recommendations. Semin Oncol 1997;24(4):477–483. van Oosterhout AGM, van de Pol M, ten Velde GPM et al. neurologic disorders in 203 consecutive patients with small cell lung cancer. Cancer 1996;77(8):1434–1441. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. Das A, Hochberg FH. Clinical presentation of intracranial metastases. Neurosurgery Clin North Am 1996;7:377–391. Ampil FL, Nanda A, Willis BK et al. Metastatic disease in the cerebellum: the LSU experience in 1981–1993. Am J Clin Oncol 1996;19:509–511. Nussbaum ES, Djalilian HR, Cjp KH. Brain metastases: histology, multiplicity, surgery, and survival. Cancer 1996;78:1781–1788. Sumida M, Uozomi T, Kiya K et al. Surface anatomy scanning (SAS) in intracranial tumors: comparison with surgical findings. Neuroradiology 1995;37:94–98. Hayashida Y, Hirai T, Morishita S et al. Diffusion-weighted imaging of metastatic brain tumors: comparison with histologic type and tumor cellularity. AJNR Am J Neuroradiol 2006;27(7):1419–1425. Posther KE, McCall LM, Harpole DH, Jr., et al. Yield of brain 18F-FDG PET in evaluating patients with potentially operable non-small cell lung cancer. J Nucl Med 2006;47(10):1607–1611.

Chapter 21 / Neurologic Complications of Lung Cancer

419

22. Patchell R, Tibbs PA, Walsh JW et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322:494–500. 23. Pederson AG, Bach FW, Nissen M et al. Creatine kinase BB and B-2-microglobulin as markers of CNS metastases in patients with small cell lung cancer. J Clin Oncol 1985;3(10):1364–1372. 24. Pederson AG, Becker KL, Bach F et al. Cerbrospinal fluid bombesin and calcition in patients with central nervous system metastases from small cell lung cancer. J Clin Oncol 1986;4(11):1620–1627. 25. Vogelbaum MA, Masaryk T, Mazzone P et al. S100beta as a predictor of brain metastases: brain versus cerebrovascular damage. Cancer 2005;104(4):817–824. 26. Wen PY, Loeffler JS. Management of brain metastases. Oncology 1999;13(7):941–961. 27. Lagerwaard FJ, Levendag PC, Nowak PJCM et al. Identification of prognostic factors in patients with brain metastases: a review of 1292 patients. Int J Rad Onc Biol Phys 1999;43(4):795–803. 28. Henson JW, Jalaj JK, Walker RW et al. Pneumosystis carinii pneumonia in patients with primary brain tumors. Arch Neurol 1991;48:406–409. 29. Nakagawa H, Miyawaki Y, Fujita T et al. Surgical treatment of brain metastases of lung cancer: retrospective analysis of 89 cases. J Neurol Neurosurg Psychiatry 1994;57:950–956. 30. Shahidi H, Kvale PA. Long-term survival following surgical treatment of solitary brain metastasis in non-small cell lung cancer. Chest 1996;109(1):271–276. 31. Delattre J-Y, Krol G, Thaler HT et al. Distribution of brain metastases. Arch Neurol 1988;45:741–744. 32. Noordijk EM, Vecht CJ, Haaxma-Reiche HH et al. The choice of treatment of single brain metastases should be based on extracranial tumor activity and age. Int J Rad Onc Biol Phys 1994;29(4):711–717. 33. Wronski M, Arbit E, Burt M et al. Survival after surgical treatment of brain metastases from lung cancer: a follow-up study of 231 patients treated between 1976 and 1991. J Neurosurgery 1995;86:605–616. 34. Bindal RK, Sawaya R, Leavens ME. Surgical treatment of multiple brain metastases. J Neurosurg 1993;79:210–216. 35. Hazuka MB, Burleson W, Stroud DN. Mulitple brain metastases are associates with poor survival in patients treated with surgery and radiotherapy. J Clin Oncol 1993;11:369–373. 36. Arbit E, Wronski M, Burt M et al. The treatment of patients with recurrent brain metastases. Cancer 1995;76(5):765–773. 37. Patchell RA, Tibbs PA, Regine WF et al. Postoperative radiotherapy in the treatment of single metastases to the brain. JAMA 1998;280(17):1485–1489. 38. Tsao MN, Lloyd N, Wong R et al. Whole brain radiotherapy for the treatment of multiple brain metastases. Cochrane Database Syst Rev 2006;3:CD003869. 39. Sen S, Ayse S, Detingoz R et al. Prognostic factors in lung cancer with brain metastasis. Radiother Oncol 1998;46:33–38. 40. Gaspar L, Scott C, Rotman M et al. Recursive partitioning analysis (RPA) of prognositc factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiation Oncol Biol Phys 1997;37(4):745–751. 41. Varlotto JM, Flickinger JC, Niranjan A et al. Analysis of tumor control and toxicity in patients who have survived at least one year after radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys 2003;57(2):452–464. 42. Postmus E, Haaxma-Reiche H, Gregor A et al. Brain-only metastases of small cell lung cancer; efficacy of whole brain radiotherapy: an EORTC phase II study. Radiother Oncol 1998;46:29–32. 43. Baumert BG, Rutten I, Dehing-Oberije C et al. A pathology-based substrate for target definition in radiosurgery of brain metastases. Int J Radiat Oncol Biol Phys 2006;66(1):187–194. 44. Chougule PB, Burton-Williams M, Saris S et al. Randomized treatment of brain metastases with gamma knife radiosurgery, whole brain radiotherapy or both [abstract]. Int J Radiat Oncol Biol Phys 2000;48:114. 45. Aoyama H, Shirato H, Tago M et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs. stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295(21):2483–2491. 46. Kondziolka D, Patel A, Lunsford LD et al. Stereotactic radiosurgery plus whole brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys 1999;45(2):427–434. 47. Andrews DW, Scott CB, Sperduto PW et al. Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363(9422):1665–1672. 48. Mehta MP, Tsao MN, Whelan TJ et al. The American Society for Therapeutic Radiology and Oncology (ASTRO) evidence-based review of the role of radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys 2005;63(1):37–46. 49. Sheehan J, Kondziolka D, Flickinger J et al. Radiosurgery for patients with recurrent small cell lung carcinoma metastatic to the brain: outcomes and prognostic factors. J Neurosurg 2005;102 Suppl:247–254. 50. Arriagada R, Le Chevalier T, Borie F et al. Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. J Nat Cancer Inst 1995;87:83–90. 51. Gregor A, Cull A, Stephens RJ et al. Prophylactic cranial irradiation is indicated following complete response to induction therapy in small cell lung cancer: results of a multicentre randomised trial. United Kingdom Coordinating Committee for Cancer Research (UKCCCR) and the European Organization for Research and Treatment of Cancer (EORTC). Eur J Cancer 1997;33:1752–1758. 52. Auperin A, Arrigado R, Pignon J-P et al. Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. N Engl J Med 1999;34:476–484. 53. Henry AM, Snee MP. Low-dose prophylactic cranial irradiation in patients with poor prognosis small-cell lung cancer. Clin Oncol (R Coll Radiol) 2006;18(2):129–132. 54. Roman D, Sperduto P. Neuropsychologic effects of cranial radiation: current knowledge and future directions. Int J Rad Oncol Biol Phys 1995;31:983–998. 55. Fonseca R, O’Neill BP, Foote RL et al. Cerebral toxicity in patients treated for small cell carcinoma of the lung. Mayo Clinic Proc 1999;74:461–465.

420

Part VII / Neurologic Complications of Specific Malignancies

56. Bunn PAJ, Kelly K. Prophylactic cranial irradiation for patients with small-cell lung cancer. J Nat Cancer Inst 1995;87:161–162. 57. Lester JF, MacBeth FR, Coles B. Prophylactic cranial irradiation for preventing brain metastases in patients undergoing radical treatment for non-small-cell lung cancer: a Cochrane Review. Int J Radiat Oncol Biol Phys 2005;63(3):690–694. 58. Peereboom DM. Chemotherapy in brain metastases. Neurosurgery 2005;57(5 Suppl):S54–65; discussion S1–4. 59. Postmus PE, Smit EF. Chemotherapy for brain metastases of lung cancer: a review. Ann Oncol 1999;10:753–759. 60. Roggero E, Busi G, Palumbo A et al. Gefitinib (‘Iressa,’ ZD1839) is active against brain metastases in a 77-year-old patient. J Neurooncol 2005;71(3):277–280. 61. Ishida A, Kanoh K, Nishisaka T et al. Gefitinib as a first line of therapy in non-small cell lung cancer with brain metastases. Intern Med 2004;43(8):718–720. 62. Nishi N, Kawai S, Yonezawa T et al. Effect of gefitinib on brain metastases from non-small cell lung cancer. Neurol Med Chir (Tokyo) 2006;46(10):504–507. 63. Katz A, Zalewski P. Quality-of-life benefits and evidence of antitumour activity for patients with brain metastases treated with gefitinib. Br J Cancer 2003;89 Suppl 2:S15–18. 64. Ceresoli GL, Cappuzzo F, Gregorc V et al. Gefitinib in patients with brain metastases from non-small-cell lung cancer: a prospective trial. Ann Oncol 2004;15(7):1042–1047. 65. Matsumoto S, Takahashi K, Iwakawa R et al. Frequent EGFR mutations in brain metastases of lung adenocarcinoma. Int J Cancer 2006;119(6):1491–1494. 66. Dziadziuszko R, Ardizzoni A, Postmus PE et al. Temozolomide in patients with advanced non-small cell lung cancer with and without brain metastases: a phase II study of the EORTC Lung Cancer Group (08965). Eur J Cancer 2003;39(9):1271–1276. 67. Robinet G, Thomas P, Breton JL et al. Results of a phase III study of early versus delayed whole brain radiotherapy with concurrent cisplatin and vinorelbine combination in inoperable brain metastasis of non-small-cell lung cancer: Groupe Francais de PneumoCancerologie (GFPC) Protocol 95–1. Ann Oncol 2001;12(1):59–67. 68. Fujita A, Fukuoka S, Takabatake H et al. Combination chemotherapy of cisplatin, ifosfamide, and irinotecan with rhG-CSF support in patients with brain metastases from non-small cell lung cancer. Oncology 2000;59(4):291–295. 69. Wong WW, Schild SE, Sawyer TE et al. Analysis of outcome in patients reirradiated for brain meastases. Int J Radiation Oncology Biol Phys 1996;34(3):585–590. 70. Korfel A, Oehm C, von Pawel J et al. Response to topotecan of symptomatic brain metastases of small-cell lung cancer also after whole-brain irradiation. a multicentre phase II study. Eur J Cancer 2002;38(13):1724–1729. 71. Ryan GGF, Ball DK, Smith JG. Treatment of brain metastases from primary lung cancer. Int J Radiation Oncology Biol Phys 1995;31(2):273–278. 72. Chamberlain MC, Kormanik P. Carcinoma meningitis secondary to non-small cell lung cancer: combined modality therapy. Arch Neurol 1998;55:506–512. 73. Zachariah B, Zachariah SB, Varghese R et al. Carcinomatous meningitis: clinical manifestation and mangement. Intl J Clin Pharmacol Theraeutics 1995;33(1):7–12. 74. Rosen ST, Aisner J, Makuch RW et al. Carcinomatous leptomeningitis in small cell lung cancer. Medicine 1982;61(1):45–53. 75. Balducci L, Little DD, Khansur T et al. Carcinomatous meningits in small cell lung cancer. Am J Medical Sciences 1984;287(1):31–33. 76. Seute T, Leffers P, ten Velde GP et al. Leptomeningeal metastases from small cell lung carcinoma. Cancer 2005;104(8):1700–1705. 77. Chamberlain MC. Leptomeningeal metastases: a review of evaluation and treatment. J Neuro-oncol 1998;37:271–284. 78. Boogerd W. Central nervous system metastasis in breast cancer. Radiother Oncol 1996;40:5–22. 79. Schaefer PW, Budzik RF, Gonzalez RG et al. Imaging of cerebral metastases. Neurosurg Clin N Am 1996;7(3):393–423. 80. Kallmes DF, Gray L, Glass JP. High-dose gadolinium-enhanced MRI for diagnosis of meningeal metastases. Neuroradiology 1998;40(1):23–26. 81. Gomori JM, Heching N, Siegal T. Leptomeningeal metastases: evaluation by gadolinium-enhanced spinal magnetic resonance imaging. J Neurooncol 1998;36(1):55–60. 82. Fukui MB, Meltzer CC, Kanal E et al. MR imaging of the meninges. Part II. Neoplastic disease. Radiology 1996;201(3):605–612. 83. Chamberlain MC. Spinal 111Indium-DTPA CSF flow studies in leptomeningeal metastasis. J Neuro-oncol 1995;25(2):135–141. 84. Chamberlain MC, Kormanik PA. Prognostic significance of 111indium-DTPA CSF flow studies in leptomeningeal metastases. Neurology 1996;46(6):1674–1677. 85. Aboulafia DM, Taylor LP, Crane RD et al. Carcinomatous meningitis complicating cervical cancer: a clinicopathologic study and literature review. Gynecol Oncol 1996;60(2):313–318. 86. Castro MP, McDonald TJ, Qualman SJ et al. Cerebrospinal fluid gastrin releasing peptide in the diagnosis of leptomeningeal metastases from small cell carcinoma. Cancer 2001;91(11):2122–2126. 87. Chamberlain MC. New approaches to and current treatment of leptomeningeal metastases. Cur Opin Neurol 1994;7:492–500. 88. Omuro AM, Lallana EC, Bilsky MH et al. Ventriculoperitoneal shunt in patients with leptomeningeal metastasis. Neurology 2005;64(9):1625–1627. 89. Comis RL. The current situation: erlotinib (Tarceva) and gefitinib (Iressa) in non-small cell lung cancer. Oncologist 2005;10(7):467– 470. 90. Omuro AM, Kris MG, Miller VA et al. High incidence of disease recurrence in the brain and leptomeninges in patients with non-small cell lung carcinoma after response to gefitinib. Cancer 2005;103(11):2344–2348. 91. Jackman DM, Holmes AJ, Lindeman N et al. Response and resistance in a non-small-cell lung cancer patient with an epidermal growth factor receptor mutation and leptomeningeal metastases treated with high-dose gefitinib. J Clin Oncol 2006;24(27):4517–4520. 92. Sakai M, Ishikawa S, Ito H et al. Carcinomatous meningitis from non-small-cell lung cancer responding to gefitinib. Int J Clin Oncol 2006;11(3):243–245.

Chapter 21 / Neurologic Complications of Lung Cancer

421

93. Milross CG, Davies MA, Fisher R et al. The efficacy of treatment for malignant epidural spinal cord compression. Australasian Radiol 1997;41:137–142. 94. Schiff D, Batchelor T, Wen PY. Neurologic emergencies in cancer patients. Neurol Clin 1998;16(2):449–483. 95. Bach F, Agerlin N, Sorensen JB et al. Metastatic spinal cord compression secondary to lung cancer. J Clin Oncol 1992;10(11): 1781–1787. 96. Posner J. Side effects of radiation therapy. In: Posner J (ed.). Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. 97. Bauer HCF. posterior decompression and stabilization for spinal metastases. J Bone Joint Surg Int 1997;79(4):514–522. 98. Klekamp J, Samii H. Surgical results for spinal metastases. Acta Neurochir (Wien) 1998;140:957–967. 99. Dethy S, Piccart MJ, Paesmans M et al. History of brain and epidural metastases from breast cancer in relation with the disease evolution outside the central nervous system. Eur Neurol 1995;35(1):38–42. 100. Bradley WG. Use of gadolinium chelates in MR imaging of the spine. J Magn Reson Imaging 1997;7(1):38–46. 101. Mehta RC, Marks MP, Hinks RS et al. MR evaluation of vertebral metastases: T1-weighted, short-inversion-time inversion recovery, fast spin-echo, and inversion-recovery fast spin-echo sequences. AJNR Am J Neuroradiol 1995;16(2):281–288. 102. Twycross R. The risks and benefits of corticosteroids in advanced cancer. Drug Safety 1994;11(3):163–78. 103. Patchell RA, Tibbs PA, Regine WF et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet 2005;366(9486):643–648. 104. Schiff D, O’Neill BP, Wang C-H et al. Neuroimaging and treatment implications of patients with multiple epidural spinal metastases. Cancer 1998;83(8):1593–1601. 105. van der Sande JJ, Boogerd W, Kappelle AC. Recurrent spinal epidural metastases: a prospective sutdy with a complete follow up. J Neurol Neurosurg Psychiatry 1999;66:623–627. 106. Schiff D, Shaw EG, Cascino TL. Outcome after spinal reirradiation for malignant epidural spinal cord compression. Neurology 1995;37(5):583–589. 107. Schiff D, O’Neill BP. Intramedullary spinal cord metastases: clinical features and treatment outcome. Neurology 1996;47:906–912. 108. Weissman DE, Grossman SA. Simultaneous leptomeningeal and intramedullary spinal metastases in small cell lung carcinoma. Med Pediatr Oncol 1986;14:54–56. 109. Holoye P, Libnoch J, Cox J et al. Spinal cord metastasis in small cell carcinoma of the lung. Int J Radiat Oncol Biol Phys 1984;10:349–356. 110. Newsom-Davis J. Paraneoplastic neurological disorders. J Roy Col Phys London 1999;33(3):225–227. 111. Posner JB, Dalmau JO. Paraneoplastic syndromes of the nervous system. Clin Chem Lab Med 2000;38(2):117–122. 112. Gutelkin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumor association in 50 patients. Brain 2000;123:1481–1494. 113. Greenlee JE. Cytotoxic T cells in paraneoplastic cerebellar degeneration. Ann Neurol 2000;47:4–5. 114. Dalmau J, Furneaux HM, Gralla RJ et al. Detection of the anti-Hu antibody in the serum of patients with small cell lung cancer: a quantitative western blot analysis. Ann Neurol 1990;27:544–552. 115. Dalmau J, Furneaux HM, Cordon-Cardo C et al. The expression of the Hu (paraneoplastic encephalomyelitis/sensory neuronopathy) antigen in human normal and tumor tissues. Amer J Pathol 1992;141:881–886. 116. Grisold W, Drlicek M. Paraneoplastic neuropathy. Cur Opin Neurol 1999;12:617–625. 117. Das A, Hochberg FH, McNelis S. A review of the therapy of paraneoplastic neurologic syndromes. J Neuro-Oncol 1999;41:181–194. 118. Shams’ili S, de Beukelaar J, Gratama JW et al. An uncontrolled trial of rituximab for antibody-associated paraneoplastic neurological syndromes. J Neurol 2006;253(1):16–20. 119. Jeyapalan S, Batchelor TT. Diagnositic evaluation of neurologic metastases. Cancer Invest 2000;18(4):391–394.

22

Neurologic Complications of Breast Cancer Willem Boogerd,

MD, PHD

CONTENTS Introduction Metastatic Complications Nonmetastatic Complications Neurologic Complications of Treatment Conclusion References

Summary The metastatic and nonmetastatic neurologic complications that occur in patients with breast cancer are reviewed. Breast cancer is the most common primary tumor in brain metastases, leptomeningeal metastases, and spinal epidural metastases in women. The etiology, diagnosis, and treatment of these and other metastases of the central and peripheral nervous systems are discussed in this chapter. Nonmetastatic neurologic complications are common, and include metabolic disorders, cerebrovascular complications, infections, and paraneoplastic disorders, as well as neurologic complications of treatment. Key Words: breast cancer, metastases, leptomeningeal, epidural

1. INTRODUCTION Breast cancer arises from the epithelial cells that line the ducts and lobules of the mammary gland. Its incidence has increased gradually (1–2% a year for the last three decades), but recently the incidence has leveled off and mortality has even decreased slightly. At present, about 1 out of every 10 females in the United States and Western Europe will develop breast cancer. The risk of breast cancer rises steadily from about the age of 25, with some slowing of increase after menopause. Obviously, hormones are important promoters of the disease. The length of menstrual life and the fraction until first pregnancy are established risk factors of breast cancer. Other factors include exposure to radiotherapy (RT) before the age of 30, and to some extent long-term postmenopausal estrogen replacement. A family history of breast cancer is the most important risk factor (1). About 5–10% of all breast cancers occur in high-risk families, including families with the Li-Fraumeni syndrome (germline p53 mutation) and mutation in the tumor suppressor gene BRCA 1 or BRCA 2 (lifetime risk of breast cancer up to 85%). In sporadic breast cancer p53 mutation is found in 40% of patients and overexpression of the oncogene erb B2 (HER-2/neu) in 15–20% of the patients. At diagnosis about 50% of patients with breast cancer have clinically local disease, 40% regional disease, and 10% distant metastases (2). The estimated 5-year survival rates for local, regional, and distant metastatic disease at diagnosis are approximately 85%, 55%, and 10%, respectively (2). Almost half of patients with From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

423

424

Part VII / Neurologic Complications of Specific Malignancies

breast cancer will eventually die from their disease. Recently, mortality of breast cancer has somewhat declined, presumably because of screening women over the age of 50 years and the extended use of adjuvant systemic therapy (3). One-quarter of women who would otherwise die of metastatic disease will remain disease-free with appropriate adjuvant systemic chemotherapy or hormone therapy. Adjuvant chemotherapy with concurrent trastuzumab (a monoclonal antibody directed against HER-2/neu) in HER-2 positive patients further reduces the risk of death by one-third (4). Tumor relapse, and overall survival are dependent on prognostic factors, including age, menopausal status, tumor size and tumor grade, lymph node status, lymphovascular invasion, and hormone receptor status. High expression of HER-2 and mutated p53 also portend a bad prognosis. Half of patients who will ultimately relapse develop metastatic disease more than 5 years after the diagnosis of the primary tumor. The first site of relapse is soft tissue (skin, lymph nodes), bone, or viscera (lung, liver, brain), each accounting for a third of the relapses. The median survival from diagnosis of relapse is about 2–3 years; patients with visceral metastases have the worst prognosis. Patients with bone metastases as the sole site of relapse often have hormone receptor-positive tumors and a relatively indolent course of their disease. Visceral metastases more often develop in relatively young patients with hormone receptor-negative tumor. The response rate to hormonal therapy of tumors with positive estrogen and progesterone receptors is 70%, and still 30–50% to second line hormonal therapy. Patients with hormone receptor-negative tumors, patients with life threatening visceral metastases, and patients refractory to hormonal therapy should receive chemotherapy, if necessary combined with local treatment (RT, surgery) for disabling local disease. Response rates for the combination regimen cyclophosphamide, methotrexate and 5-fluorouracil (5-FU) (CMF) are 30–50%; response rates with anthracycline-based regimens such as CAF or FEC slightly higher with generally acceptable toxicity (5). Patients may respond to CMF for metastatic disease even after previous adjuvant CMF. Tumors with high expression of HER-2 are more aggressive and resistant to chemotherapy, but show a higher probability of response to anthracyclines (6). The taxanes have significant activity in metastatic breast cancer with little cross-resistance with anthracyclines and relatively high activity in visceral disease. Capecitabine is an orally active 5-FU derivate that shows activity as monotherapy even in heavily pretreated patients refractory to 5-FU and taxanes (7). Trastuzumab as single agent in HER-2 overexpressing metastatic breast cancer with progressive disease after chemotherapy has modest activity with a reported response rate of 15% (8). The addition of trastuzumab to standard chemotherapy (anthracyclines or taxoids) in patients with metastatic breast cancer overexpressing HER-2 is associated with a higher response rate, longer duration of response, and longer survival (9). Several phase II studies have reported a high response rate to high dose chemotherapy in combination with autologous stem cell transplantation in patients with metastatic breast cancer. Survival benefit, however, has not been demonstrated in the published randomized studies (10,11). Overall, despite major improvements in treatment of advanced breast cancer, the median survival of patients with metastatic disease at 2–3 years has changed little over the last decades. However, the expansion of therapeutic options and increased knowledge of pathogenesis and prognostic factors of metastatic disease enable the physician to refine the management of the individual patient, often leading to long-term palliation and meaningful quality of life. This also holds true for central nervous system (CNS) metastasis. CNS metastasis is one of the most dreaded complications in metastatic disease. Longer survival of patients with metastatic disease and increased use of adjuvant systemic therapy may lead to a higher incidence of CNS metastasis. The clinical diagnosis of CNS metastasis is made in about 15% of patients with breast cancer. Autopsy studies report an incidence of CNS metastases in more than 30% of patients dying of breast cancer, of whom about 10% have metastatic disease exclusively in the CNS (2,12). Between 5% and 10% of patients with breast cancer will die primarily of CNS failure (13). Similar to CNS involvement in other solid tumors, CNS metastasis in breast cancer is usually associated with widespread systemic disease. Systemic disease in breast cancer is, however, often more amenable to treatment than in other solid tumors. In addition, CNS involvement may be the only site of clinical disease in a small but clinically important proportion of patients with breast cancer (12). This explains why in breast cancer the outcome of patients with CNS metastases often depends primarily on the treatment response of the neurological disease.

Chapter 22 / Neurologic Complications of Breast Cancer

425

2. METASTATIC COMPLICATIONS 2.1. Brain Metastasis Approximately 20% of breast cancer patients have brain metastases (BM) at autopsy, whereas the clinical diagnosis of BM is made before death in 5–10% of patients with breast cancer (14,15). BM is associated with ductal carcinoma, a short cancer history (6 months shorter than patients dying without BM), young age (on average 5 years younger than patients dying without BM), estrogen receptor-negative tumor, and extensive metastatic disease including increased serum lactate dehydrogenase (LDH), liver metastases, and in particular lung metastases (16–18). HER-2 overexpression is associated with more aggressive disease, but its significance as independent risk factor for BM is controversial (17,19–21). In patients with advanced metastatic breast cancer 13.5% had occult BM on screening CT or brain MRI (19). In about 20% of patients with BM the brain is the first site of relapse (22,23). An autopsy study reported that in 13% of patients with CNS involvement no other metastases were found (12). In contrast to lung cancer, BM only rarely is the first clinical manifestation of breast cancer. An increased incidence of the brain as first site of relapse has been associated with use of adjuvant chemotherapy (23). The median disease-free interval in these patients seems the same as in patients without adjuvant treatment who develop systemic metastases, suggesting that chemotherapy administered when the blood–brain barrier is intact does not prevent the development of cerebral micrometastases, whereas it suppresses effectively the development of extracranial metastatic disease. It was suggested that BM occur more frequently in patients treated with trastuzumab, presumably because of lack of penetration through the blood–brain barrier. However, the increased incidence of BM in those patients might be related to the aggressive biology of HER-2 overexpressing tumor, rather than to the use of trastuzumab (17,24). About 40% of BM in breast cancer are single lesions. Symptoms and signs of BM are related to the site of the lesion(s) and accompanying edema (Table 1). Half of patients complain of headache that may be associated with vomiting and drowsiness. One-third present with focal signs and 20% with seizures (25). The diagnosis of BM is confirmed with CT or preferably MRI using intravenous contrast. The lesion(s) may be indistinguishable from primary brain tumors, vascular lesions, infectious processes, radionecrosis, or demyelinating lesions. In general, the correlation between BM and aggressive tumor activity is reflected in the poor outcome with a median survival after treatment of BM of 3–5 months; 10–20% survive more than a year, while half die of their neurological disease (23,24,26). Absence of extracranial metastatic disease and a solitary BM appear the most important prognostic factors. A Karnofsky score of more than 70, and a neurological history of longer than a month prior to the diagnosis of BM are less significant prognostic factors (23,24,26,27). Survival of patients with asymptomatic BM treated with radiotherapy (RT) was less than 5 months and not different from survival of patients with symptomatic BM (19). Thus, it is questionable if detection and treatment of occult BM with RT improves outcome. A remarkably high median survival after diagnosis of BM of more than a year was reported in patients with BM with ongoing treatment with chemotherapy and trastuzumab for HER-2 positive tumor (20,28). Other studies found no relation between survival after diagnosis of BM and ongoing treatment with trastuzumab (17,29). Radiotherapy still is the mainstay of treatment of BM (Table 2). Median survival after RT is about 4 months (22,23,26). RT is usually combined with steroids. Steroids induce neurological improvement in about 75% of patients. There is no correlation between the extent of edema and degree of clinical improvement, nor between response to steroids and survival (30). The optimum dose of steroids is not defined. In patients without imminent herniation a dose of 4 mg per day is as effective as 16 mg, while causing less side effects (31). The RT fractionation scheme and total dose have no significant influence on the rate and duration of response (30,32,33), although in patients with favorable prognostic factors some increase of survival has been suggested with a RT scheme with low fraction size (≤ 3 Gy). Boost RT and the use of radiosensitizers do not result in better outcome, even in the absence of systemic disease (32). A randomized study comparing RT with or without the radiosensitizer Efaproxiral in patients with BM from solid tumors suggested some increase in response rate, but not in survival in the subgroup of patients with breast cancer (34). A confirmatory trial has been initiated in patients with BM from breast cancer. Clinical response following steroids and RT is seen in about 80% of the

426

Part VII / Neurologic Complications of Specific Malignancies

Table 1 Neurologic Symptoms Due to Metastatic Breast Cancer and Its Treatment Symptom

Cause

Headache

Metastasis: BM, LM, dural, skull, skull base: • orbit: periorbital pain (proptosis, diplopia) • parasellar: frontal headache (+ diplopia, sensory loss CN V1 ) • petrous bone: dull pain cheek, forehead • jugular foramen: pain in pharynx • occipital condyle: occipital pain (+ ipsilateral CN XII) Treatment related: • intrathecal chemotherapy • RT (edema, encephalopathy) Post-lumbar puncture Vascular: sinus thrombosis, subdural hematoma/hygroma

Seizures

Metastasis: BM, LM, dural Treatment related: • intrathecal MTX • RT (encephalopathy, radiation necrosis) Vascular: sinus thrombosis, subdural hematoma/hygroma Metabolic (hypercalcemia)

Encephalopathy

Metastasis: BM, LM, dural Metabolic (hypercalcemia, liver metastasis) Drug-induced (opioids, corticosteroids, anticonvulsants) Treatment related: • intrathecal chemotherapy • RT (encephalopathy)

Visual disturbances • decreased visual acuity, blurred vision

• metamorphopsia • diplopia

• oscillopsia • cortical blindness

Metastasis: choroid, optic nerve, orbit, LM, retro-orbital/pituitary Treatment related: • intrathecal MTX, tamoxifen • RT (eye, optic nerve) • corticosteroids Metastastis: choroid Metastasis: ocular, orbit, retro-orbital/parasellar, LM, brain stem Increased intracranial pressure (metastasis, edema) Paraneoplastic: PCD (anti-Yo) Anticonvulsants, opioids Metastasis: brain stem Paraneoplastic: PCD (anti-Yo), OMS (anti-Ri) Metastasis: LM, BM, bilateral occipital Vascular: sinus thrombosis

Arm/leg weakness

Metastasis: cranial dural, BM, LM, ISCM, SEM, plexus RT-induced (ischemic) plexopathy

Tetra/paraparesis

Metastasis: multiple BM, ISCM, SEM Treatment related: • MTX • Ara-C myelopathy • RT myelopathy Vascular: sinus thrombosis Corticosteroids: myopathy (proximal), epidural lipomatosis Hypercalcemia (proximal) Polymyositis (proximal)

Chapter 22 / Neurologic Complications of Breast Cancer

427

Lhermitte’s sign

Metastasis: SEM RT myelopathy (transient) Docetaxel

Sensory disturbances

Metastasis: ISCM, LM, SEM Taxane-induced polyneuropathy RT myelopathy Paraneoplastic sensory neuropathy

BM: brain metastasis; SEM: spinal epidural metastasis; ISCM: intramedullary spinal cord metastasis; RT: radiotherapy; LM: leptomeningeal metastasis; MTX: methotrexate; OMS: opsoclonus myoclonus syndrome.

Table 2 Treatment of Metastases from Breast Cancer Metastatic Site Brain

Leptomeningeal

Intramedullary spinal cord Spinal epidural

Skull base/orbit

Treatment Modality

Neurological Response Rate

Median Survival(Months)

Corticosteroids WBRT (+ steroids) Radiosurgery + WBRT Surgery + WBRT

75% 75% 90%

1 month 3 months 75% 1-year

1–2 4

> 90%

75% 1-year

9–18

Chemotherapy

75%

7 months

6

Intraventricular chemotherapy + RT symptomatic site Systemic chemotherapy + RT symptomatic site RT

50%

1–2 months

3–4

50%

3–4

Sporadically

1

RT (+/− laminectomy)

± 80%

Systemic chemotherapy Surgery + RT (vertebrectomy + stabilization) RT

55% > 90%

50% Brachial plexus

Median Duration

Systemic therapy RT Systemic therapy

50–75% < 50%

± 12 months

9

7 > 12 months

12–24

Comment

Only lesions < 3 cm Most die of systemic disease In nonrandomized studies objective response 55% Great difference in treatment results between various studies No neurotoxicity

> 90% remain ambulant 30% become ambulant Small series selected pts Best method to regain walking ability

428

Part VII / Neurologic Complications of Specific Malignancies

patients with a median duration of response of 3 months (30). At time of death half the patients treated with RT are neurologically improved or stable (22,23,30). Radiosurgery is used with increasing frequency in patients with BM. Radiosurgery of BM provides good local tumor control, low morbidity and mortality, low rate of steroid dependence, and high rate of functional independence, comparable with results of surgical resection (35,36). Radionecrosis is a rare complication that predominantly occurs following whole brain RT (WBRT) and in lesions larger than 3 cm. Whole-brain RT with radiosurgery showed a survival benefit compared to WBRT alone in patients with a single BM from solid tumors (predominantly lung cancer). In that randomized trial no survival benefit was demonstrated in patients with 2 or 3 BM, possibly because of the remarkably high 1-year response rate of 71% in the WBRT group (versus 81% in the WBRT plus radiosurgery group) (37). After radiosurgery without WBRT the risk of cerebral recurrence distant from the irradiated lesion is increased, but these recurrences can be treated successfully again with radiosurgery (36,38). In a randomized trial in patients with one to four BM from solid tumors neurologic functional preservation, cause of death, and survival following radiosurgery plus WBRT did not differ from radiosurgery with salvage WBRT as needed (38). Patients with brain stem metastases without acute neurologic symptoms may achieve effective palliation after radiosurgery (39). Small lesions (< 3 cm) are treated with radiosurgery probably as effective as with surgical resection. Patients with larger but operable lesions without active systemic disease should be treated with surgical resection. Post-operative WBRT significantly prevents relapse at the site of the original metastases, as well as at other sites in the brain, but overall survival does not differ from WBRT as salvage treatment (40). Post-operative local brachytherapy using the GliaSite RT System leads to a local control rate and survival time similar to those achieved with WBRT, but is associated with a high rate of local radiation necrosis (41). Median survival of patients treated with surgery and post-operative WBRT is about 9–18 months (22,23); most of these patients die of systemic disease without evidence of cerebral relapse. In a retrospective analysis of patients with a solitary BM as first site of relapse, operated patients had a clearly better outcome (median survival 23 months) than patients treated with RT or chemotherapy (median survival 9 months) (42). Outcome of patients undergoing resection of multiple, accessible BM appears the same as in patients with a single operated BM (43,55). Because it is recognized that the blood–brain barrier is impaired or absent in large areas of BM because metastatic tumor vasculature lacks the structure of normal brain capillaries, evidence has accumulated that systemic chemotherapy can be an effective treatment in BM from breast cancer (Figure 1). Phase II studies demonstrated

Fig. 1. Brain metastasis; complete remission following systemic chemotherapy. Pretreatment CT-scan (A) shows contrast enhancing lesion (arrow) with surrounding edema. Complete remission after CMF chemotherapy (B) persisted until death due to systemic disease after 19 months.

Chapter 22 / Neurologic Complications of Breast Cancer

429

Fig. 2. Leptomeningeal metastasis; MRI FLAIR image without contrast. Hyperintensity of arachnoid and pia (white, small arrow) signifies leptomeningeal seeding. Uninvolved arachnoid is hypointense (black, large arrow).

an objective response of 50% or more and an overall median survival of about 6 months after primary treatment with standard systemic chemotherapy, comparing favorably to the treatment results of WBRT (44,45). Responses of extracranial disease paralleled the cerebral response in those patients. Chemotherapy including cisplatin and etoposide followed by consolidation WBRT in a number of patients resulted in an objective response rate of 38%, stabilization in 12%, and overall median survival of 7 months (31). The benefit of consolidation WBRT following chemotherapy-induced response of BM instead of reserving RT for local cerebral relapse has not been established. Regression of BM after capecitabine was reported in a heavily pretreated patient (47). Temozolomide as a single agent is ineffective in BM from breast cancer (48). After capecitabine combined with temozolomide for recurrent as well as newly diagnosed BM an objective response was observed in 18%, and stable disease in 50% of the patients with a median time to progression of 12 weeks (49). High-dose intravenous methotrexate for BM as well as leptomeningeal metastasis was effective (response or stable disease) in about half of the treated patients (50). Response of BM to hormonal therapy has been reported occasionally (51,52), although it is assumed that hormonal treatment usually will act too slowly to prevent serious neurological deterioration in patients with symptomatic BM. Several studies reported a remarkably high survival in patients with ongoing treatment with trastuzumab after diagnosis of BM (17,20). The role of trastuzumab and of concurrent chemotherapy in the relatively favorable outcome is unclear. Recurrence of BM in stable or absent extracranial disease is not uncommon in breast cancer patients. A second course of RT or second line chemotherapy after previous systemic treatment of BM will result only occasionally in a meaningful response. However, systemic chemotherapy after previous RT or surgery of BM may induce stabilization or response in more than half of patients (45,53). In selected cases radiosurgery or re-resection affords local tumor control for 6–12 months in about 75% of cases (36,42,54,55). In conclusion, if the variety of therapeutic options including surgery, RT, radiosurgery and systemic therapy is appropriately put into practice, long-term and meaningful palliation may be possible, with survival extending 1 year or more. 2.1.1. Pituitary Metastasis Pituitary metastasis seems to be a relatively frequent finding at autopsy. It was found in 9% of patients dying with breast cancer (56). It is associated more frequently with bone metastasis than with lung metastasis (12). The pituitary gland may be invaded by extension from a bone metastasis in the sella, by hematogenous spread, or from leptomeningeal metastasis (LM). The vast majority of pituitary metastases are asymptomatic (57). Symptoms include headache, anterior hypopituitarism, visual field loss, and rapidly progressive oculomotor weakness in

430

Part VII / Neurologic Complications of Specific Malignancies

cases of cavernous sinus invasion. Diabetes insipidus develops relatively frequently (58). Treatment consists of RT, but systemic chemotherapy may also be effective (59). 2.1.2. Intramedullary Spinal Cord Metastasis Intramedullary spinal cord metastasis (ISCM) is an infrequent complication of breast cancer that will be diagnosed presumably more often by the use of MR scanning. Enhancement of ISCM at T1-weighted images is a reliable finding reflecting disruption of the blood spinal cord barrier. T2-weighted images are particularly sensitive in detecting ISCM, which usually develops in patients with widespread metastatic disease, including parenchymal BM. Most of ISCM results from hematogenous dissemination and occurs in the gray matter because of its greater arterial perfusion. Sometimes ISCM is caused by direct infiltration of tumor cells from LM along the perivascular space into the cord. There are no clinical features characteristic for ISCM by which it can be distinguished with certainty from an extradural compressive lesion. Pain is the initial symptom in the majority of the reported cases; it usually begins as back pain that often soon becomes radicular. The clinical picture of ISCM usually is that of a rapidly progressive myelopathy, initially often asymmetric, leading to a complete loss of cord function in the course of days or a few weeks. Treatment consists of RT with corticosteroids. Treatment results are poor. Reported median survival of patients with symptomatic ISCM is ≤ 3 months. In our experience, early diagnosis and prompt start of treatment may lead to symptomatic improvement lasting a few months. 2.1.3. Leptomeningeal Metastasis Among solid tumors breast cancer is the primary tumor most frequently associated with leptomeningeal metastasis (LM). About 2–5% of patients with metastatic breast cancer will experience LM, usually late in the course of their disease (60). It is relatively more common in lobular carcinoma (61). The relationship to aggressive tumor activity including a short recurrence-free interval is a matter of controversy (60–62). In breast cancer LM is associated with bone metastases. Spread of tumor cells from vertebral metastases perivascularly along the radicular veins appears to be the major route of entrance of tumor cells into the subarachnoid space (63). Direct extension from a subependymal or cortical metastasis, from a paravertebral mass, or by direct seeding from a metastasis in the choroid plexus are occasional causes of LM, which is usually multifocal. It often forms macroscopic tumor masses, sometimes with a local inflammatory reaction. This may cause obliteration of the subarachnoid space with CSF compartmentalization and hydrocephalus. There is a predilection for tumor sedimentation at the base of the brain and in the cauda equina. Tumor cells may infiltrate the nerve or form a cuff surrounding nerve roots. Tumor cells in the leptomeninges may extend into the perivascular spaces, penetrating the brain or spinal cord parenchyma. Concomitant cortical tumor is more likely a late effect of LM than the source of LM. The most common symptoms and signs of LM include headache, confusion, cranial nerve involvement (most commonly cranial nerves III, VI, VII, VIII) and spinal root dysfunction (particularly at the lumbosacral level). The differential diagnosis includes metabolic encephalopathy, infectious meningoencephalitis, paraneoplastic disease, brain metastasis, epidural metastasis, intervertebral disc disease and peripheral neuropathy. The diagnosis of LM is established by demonstration of malignant cells in the CSF. In LM from breast cancer cytology is falsely negative in about 10% upon the first lumbar puncture (60,61). Cisternal CSF cytology may be more sensitive than lumbar CSF in patients with cerebral symptoms (64). Flow cytometry or immunocytochemistry are not superior to cytology in detecting tumor cells in CSF of patients with LM from breast cancer (65,66). Interphase cytogenetic studies of CSF cells correlated better than cytology with the course of disease (67), but relevance in diagnosing or monitoring LM is unclear. CSF chemical composition (protein, glucose, LDH), the cell count, or CSF pressure almost always shows abnormalities in LM (60,68). A variety of specific and non-specific tumor markers has been detected in the CSF, but their clinical relevance is uncertain and CSF cytology is clearly more sensitive and specific in establishing the diagnosis (69). However, vascular endothelial growth factor (VEGF) may be a useful marker for both the diagnosis and prognosis (70,71). Myelography and brain CT show abnormalities suggestive of LM in about 25% of patients. These abnormal findings include meningeal or ependymal enhancement, hydrocephalus and nodular filling defects. About 20%

Chapter 22 / Neurologic Complications of Breast Cancer

431

show concomitant but clinically unsuspected parenchymal BM. Contrast-enhanced brain MRI is abnormal in 70% and spinal MRI in about 30% of patients with LM, although it should be noted that any condition with irritation of the meninges can produce meningeal enhancement. This includes local tumor infiltration, but also infectious meningitis and even post-lumbar puncture CSF hypotension syndrome (72). Occasionally MRI reveals meningeal infiltration concurrent with repeatedly negative CSF cytology. Thus, CSF cytology and MRI are complementary in the diagnosis of LM (73,74). In patients with a normal neurological examination, imaging studies have a low yield of detecting abnormalities. In line with the experience with LM from hematologic malignancies, intraventricular administration of chemotherapy in combination with RT to the symptomatic area has become the recommended treatment of LM from breast cancer. Methotrexate (MTX) is the drug of choice and, given as single agent, less toxic and as effective as multiple agent treatment (75). Other agents for intraventricular use are cytarabine (Ara-C) and thiotepa. The benefit of intraventricular thiotepa is questionable (76): TEPA, an active metabolite of thiotepa that crosses the blood–brain barrier, is formed when thiotepa is given intravenously, whereas it is not formed when thiotepa is given intraventricularly. The alternative of frequent intralumbar administration of MTX is inconvenient for the patient and does not reliably achieve therapeutic ventricular drug concentrations (77). Liposomal depot Ara-C given once per 2 weeks results in continuously cytotoxic CSF concentration, even in the ventricles after intralumbar administration. Intraventricular administration via an Ommaya reservoir is associated with infection, misplacement, or occlusion of the catheter in 10–30% of the patients (60,78,79). The response rate following intraventricular treatment in combination with clinically involved field RT was more than 50% in several early studies with a median survival of about 6 months as compared to rapid clinical deterioration and a median survival of 4–6 weeks for historical controls (68,80,81). However, poor responses and a short median survival of only one or a few months despite similar, or even more intensive therapy were described in more recent studies (60,75,82–84). Obviously, selection of patients played a role in those noncontrolled studies. In a randomized study comparing the efficacy of intraventricular MTX and thiotepa in combination with involved field RT, not a single patient in either treatment arm with a fixed neurological deficit showed neurological improvement and median survival was 15 weeks (82). In all studies at least one-third of the patients die within a few weeks despite intensive treatment. Based on prognostic factors these patients should preferably be excluded from extensive therapy. Negative prognostic factors are age over 55 years, low performance status, progressive visceral metastatic disease, encephalopathy, cranial nerve involvement and a decreased CSF glucose concentration (60,61,85). Normal CSF protein has been associated with a better prognosis, but a markedly increased CSF protein (> 1.0 g/l) does not necessarily imply a short survival (60,75). A markedly increased protein may be related to spinal involvement and a relatively better prognosis (60,62,79). In addition to uncertainty about the actual efficacy of intraventricular treatment, there are no data concerning the effect of dose of MTX, nor from duration of treatment. No differences in efficacy were seen in schedules that included 5, 10, or 20 mg of MTX per injection. A 12 hourly administration of 1 mg was as effective but less toxic than 10 mg MTX twice a week (86). In most studies intraventricular treatment remains intensive as long as CSF cytology is tumor positive, and continues throughout the patient’s life. Larger studies show that continuation of intraventricular therapy beyond 6 weeks does not improve survival, but increases the risk of neurotoxicity (60,79). It should be noted that treatment of LM is purely palliative. A lasting complete remission of LM from solid tumors is hardly ever obtained. The clinical status after the first 6 weeks of treatment is a better predictor of outcome than CSF cytology (60). Patients may have tumor-positive CSF cytology for many months, while they are in a stable clinical condition, even without prolonged intraventricular treatment. Lack of efficacy of intraventricular MTX has been ascribed to compartmentalization and flow disturbances of the subarachnoid space (87,88). However, the reported better outcome in patients with corrected CSF flow after RT does not necessarily imply a direct relationship between flow disturbances and outcome; more bulky meningeal tumor or resistance of tumor to RT may play a role in this matter. In addition, cytotoxic levels of MTX can be found in CSF compartments with a partial block (89). Methotrexate (2x/week) was compared with depot Ara-C (1x/2 weeks) in a randomized trial. Two of the 12 breast cancer patients treated with intraventricular depot Ara-C had a response (defined as negative CSF cytology, and stable or improved neurological examination), versus none of the 10 patients treated with intraventricular

432

Part VII / Neurologic Complications of Specific Malignancies

MTX (90). Other large prospective studies showed similar results of intrathecal depot Ara-C in breast cancer patients, with response rates of 21–24%, duration of response of less than 2 months, and a median survival of about 3 months (91,92). No difference in response was observed between intraventricular and intralumbar administration of depot Ara-C. Patients with CSF flow disturbances were excluded in those studies. Despite anecdotal reports of success, other drugs, investigational intrathecal therapies such as monoclonal antibodies conjugated with toxins or radioisotopes, and intraventricular immunotherapy have not yet been demonstrated to be effective in LM from breast cancer (96–98). Intrathecal injections of trastuzumab were well tolerated, but efficacy without concomitant chemotherapy is not clear (99). Radiotherapy to the symptomatic area often results in stabilization and sometimes in improvement of neurologic deficit, but probably does not influence survival (60). Some authorities report a better outcome with the administration of RT to asymptomatic sites of macroscopic meningeal infiltration (79). Radiotherapy to the site of CSF block as demonstrated on flow studies may restore CSF flow (87,88). Radiotherapy to the complete neuraxis is not feasible in most patients because of effects on bone marrow, previous RT to parts of the neuraxis, and the requirement for a protracted treatment course in a disease with short median survival. Acute and subacute neurologic side effects of intrathecal chemotherapy, including aseptic menigitis, myelopathy, and encephalopathy, are usually mild and transient. On occasion, however, they are progressive and fatal. Aseptic meningitis of any grade was observed in 60% of patients treated with intrathecal MTX or depot Ara-C without oral dexamethasone prophylaxis, and in about 15% of the patients with dexamethasone (90). Slow but sustained absorption of MTX from the CSF into the plasma may lead to systemic side effects including myelosuppression and mucositis; the use of leucovorin protects against these complications. Late neurotoxicity consisting of leukoencephalopathy was reported only occasionally in early studies, but more recent studies report that this serious complication with progressive ataxia and dementia develops in as many as half of long-term survivors (60,79). High peak levels of MTX, a high cumulative dose of MTX, and cranial RT are associated with an increased risk of leukoencephalopathy (93,94). In this respect flow disturbances in the subarachnoid space may play a role (87). Intrathecal Ara-C in combination with RT to the spinal cord is associated with serious myelopathy (95). Several lines of evidence suggest potential utility of systemic chemotherapy in LM from breast cancer. Studies of experimental LM have demonstrated that LM are well vascularized with highly permeable blood vessels (100,101). Contrast enhancement of LM on neuroimaging supports this observation. Penetration of chemotherapy from the CSF space into these tumors is only a few cell layers. Systemically administered drugs better penetrate these tumor deposits than drugs dissolved in CSF, and efficacy will not be affected by CSF flow obstruction. Patients with LM treated with intraventricular chemotherapy seem to have a better neurological response and a longer survival when they also receive systemic chemotherapy (60,79,87). A few nonrandomized studies showed that patients treated with standard systemic chemotherapy and involved field RT (Figure 3) have the same response, median survival, and proportion of long-term survivors as patients treated with intraventricular chemotherapy combined with RT, without the serious neurotoxicity associated with intraventricular treatment (60,102). High-dose intravenous MTX crosses the blood–brain barrier, and was tolerated well in patients with LM, but its efficacy in patients with LM from breast cancer is not clear (103,104). One small randomized study was performed in patients with LM from breast cancer, comparing intraventricular MTX, involved field RT, and appropriate systemic therapy with appropriate systemic therapy, involved field RT, but without intrathecal chemotherapy (105). Neurological response or stabilization (59% vs. 67%), time to neurologic progression (23 weeks vs. 24 weeks), median survival (18 weeks vs. 30 weeks), and cause of death were not different. However, neurotoxic complications of treatment occurred significantly more often in the intraventricular MTX group (47% vs. 6%). Thus, adding intrathecal chemotherapy to systemic treatment and involved field RT will in general not improve neurological outcome or survival, but often will affect quality of life by neurologic complications. A larger randomized study should be performed to confirm these results. From the data of the randomized study it remains possible that a subgroup of patients might benefit from a combination of intrathecal treatment and systemic therapy. Favorable responses of LM have also been observed after oral capecitabine (106,107). In selected patients even hormonal therapy can induce a long-term response and survival in LM from breast cancer (108). However,

Chapter 22 / Neurologic Complications of Breast Cancer

433

Fig. 3. Leptomeningeal metastasis; response to local RT and systemic chemotherapy. The pretreatment, post-contrast, T1-weighted MRI (A) shows that the lumbar subarachnoid space is filled with tumor with thickening of the nerve roots. After local RT and systemic CMF chemotherapy near-complete disappearance of intradural tumor (B). The patient survived 48 months with only slight residual neurological impairment without intrathecal treatment.

most of the patients who present with LM have previously received systemic therapy, which reduces the chance of a successful systemic treatment of LM. Intrathecal chemotherapy may still be indicated in heavily pretreated patients for whom systemic therapy is not a feasible option. 2.1.4. Cranial Dural Metastasis Cranial dural metastasis includes metastases that involve either or both the epidural or subdural spaces (Figure 4). Epidural cranial metastasis usually occurs through direct extension from calvarial metastatic lesions,

Fig. 4. Calvarial metastasis with epidural and subdural tumor. The postcontrast T1-weighted MRI shows contrast enhancement of the calvarial lesion and epidural as well as subdural tumor extension. Contrast-enhanced dura: thick arrow. Tabula externa: thin arrow. Tabula interna: small arrow.

434

Part VII / Neurologic Complications of Specific Malignancies

or occasionally hematogenously through the external carotid artery or the vertebral veins. Cranial epidural tumor often appears to spread over a much larger area than the adjacent osseous metastasis from which it originates. It may invade the cranial dura causing a diffuse thickening of the dura, whereas it only rarely transgresses the dura to give rise to a subdural tumor. Solitary subdural metastasis may also occur without adjacent epidural tumor. It may develop in the falx cerebri where it should be differentiated from a meningioma. Subdural tumor may invade the adjacent leptomeninges to produce LM. Overall, dural and subdural metastasis constitute only a minor clinical problem because of rare symptomatic occurrence (15). At autopsy, however, cranial dural metastasis can be found in 15% of patients who die of breast cancer (14). Symptoms of cranial dural metastasis include contralateral weakness, signs of intracranial hypertension, and cranial nerve involvement (Figure 5). Abrupt onset of focal neurological deficit and seizures suggest direct parenchymal invasion through perivascular spaces, hemorrhage, or venous sinus occlusion. MRI is the most accurate diagnostic method. Its advantages over CT scanning include multiplanar capability and identification of sinus occlusion. Symptomatic cranial dural or epidural metastasis is treated usually with RT, but in our experience systemic therapy as single treatment can also be successful. Occasionally surgical resection followed by RT is the treatment of choice for a large dural metastasis. As cranial dural metastasis is associated with bone metastasis, survival after treatment is usually better than in patients with parenchymal brain metastasis. 2.1.5. Spinal Epidural Metastasis Spinal epidural metastasis (SEM) in breast cancer is considered of utmost clinical importance because of its high incidence and risk of spinal cord or cauda equina compression. In breast cancer, SEM is usually caused by extension of a vertebral metastasis through the bony cortex into the extradural space. Vertebral bone marrow apparently is a suitable environment for tumor growth: 60% of patients with breast cancer develop vertebral metastases, predominantly in the posterior part of the vertebral bodies. The majority of SEM is found at the lower thoracic and lumbar level. Less than 10% develop in the cervical spine. SEM extends over one vertebra in half of the patients, over two vertebrae in a quarter, and over three or more vertebrae in another quarter (109). Multiple SEM (MSEM) occur in at least 20% of patients suspected of SEM (109,110). These MSEM are usually asymptomatic, but of course may influence the further course of disease.

Fig. 5. Parasellar epidural metastasis (small arrow) extending through the foramen ovale (large arrow) into the lateral pterygoid muscle in a patient with pain and swelling of the right cheek.

Chapter 22 / Neurologic Complications of Breast Cancer

435

Asymptomatic SEM rarely occur in the cervical region, so it is recommended that at least the thoracic and lumbosacral spinal canal are imaged in patients suspected of SEM. Neurological symptoms of SEM are the result of direct compression of spinal roots and cord by tumor or bone fragments, and of changes in blood flow and vascular permeability in the spinal cord. Back pain at the site of the vertebral metastasis is nearly always the initial symptom (111). Typically, a few months later radicular signs develop due to epidural extension or vertebral collapse. The pain associated with SEM often increases with recumbency. Signs of myelopathy due to spinal cord compression develop an average of one month after the first signs of radiculopathy, but may also develop rapidly or even may be the first manifestation of SEM. The risk of rapid development of paraplegia is great once signs of myelopathy are present. Physicians and patients should be familiar with early symptoms of myelopathy in order to institute adequate treatment promptly. The pretreatment ambulatory status is the critical prognostic factor for neurological outcome and thus for the quality of life. In contrast to SEM from other primaries (110), the degree of spinal block and presence of MSEM are not prognostic for outcome (109). In acute spinal cord injury, as occurs with spinal instability, the damage develops within hours in the center of the cord, with hemorrhage in the gray matter followed by decreased blood flow and necrosis. In chronic cord compression the damage is mainly in the lateral columns with relatively few changes in the gray matter. Thus, the rate of progression of myelopathy can be important for neurological outcome: occasionally functional recovery may occur after prolonged development of paraplegia, even if complete paralysis exists for more than 24 hrs (112). Plain films of the vertebral column are the first diagnostic procedure in patients suspected of SEM. A vertebral collapse of 50% or more carries a 50% risk of SEM (113). However, more than 30–70% bony destruction is required before a plain spinal X-ray becomes abnormal. Consequently, normal plain films do not exclude SEM. Plain films did not accurately localize the site of epidural lesion in a third of patients with breast cancer. In addition, clinical data were not localizing in another third of SEM (109). Thus, in patients suspected of SEM the entire spinal canal, or at least the thoracic and lumbar area should be imaged, preferably with MRI. If MRI is contraindicated, myelography or CT scanning with intrathecal or bolus intravenous contrast are alternative techniques (114). Myelography has the added advantage of CSF examination: in 9% of breast cancer patients who underwent myelography for SEM, CSF disclosed LM (109). The aim of treating SEM is to relieve pain, to maintain or restore neurological function and spinal stability, and to achieve local tumor control. Whether asymptomatic SEM should be treated is uncertain (110,115). Compared to BM from breast cancer, SEM occurs in older patients and is related to bone metastases, implying some longer median survival. Median survival of SEM from breast cancer is 6 to 9 months, and the 1-year survival 40% (109,116). Unsurprisingly, nonambulatory patients have a significantly worse prognosis (109). Radiotherapy in combination with corticosteroids is the standard treatment of SEM. Corticosteroids are an important adjuvant treatment in the early stage (117,118). Most patients experience dramatic relief of pain at the first day of steroid treatment, and neurological outcome is significantly better than without steroids (118). Experimental studies showed a dosedependent effect of dexamethasone on the reduction of spinal cord edema (119), so high-dose dexamethasone of 100 mg per day is recommended in patients with severe or rapidly progressive neurologic deficit. Usually, after an initial bolus of 10 mg dexamethasone, 4 mg four times a day is given during the first 3–7 days and then tapered off over 1–2 weeks. The risk of major side effects is small when it can be discontinued within 3 weeks, even after doses of 100 mg during the first 3 days (117,118). More than 90% of ambulatory patients treated with RT for SEM from breast cancer remain ambulatory, whereas one-third of paraparetic patients regain walking ability. Restoration of sphincter dysfunction is achieved in one-third of the patients. Generally RT ports are two vertebrae above and below SEM. Smaller RT fields were associated with increased risk of local recurrence (120). Overall, recurrent SEM occurs in about 20% of patients, expectedly more often in long survivors: recurrent SEM occurs in one-third of patients surviving 6 months, and in one-half of patients surviving two years (121). About one-half of recurrent SEM develops at the same level as the previous SEM. The optimal radiation dose and schedule for SEM remain undetermined. Influence of total RT dose or fraction size on posttreatment neurological function has not been demonstrated. In a noncontrolled study in-field recurrence was observed more often after 1 × 8 Gy and 5 × 4 Gy, than after 10 × 3 Gy, 15 × 2.5 Gy or 20 × 2 Gy (122). So,

436

Part VII / Neurologic Complications of Specific Malignancies

patients with a poor prognosis of < 6 months survival could be treated with 1 × 8 Gy, and patients with a better prognosis with 10×3 Gy. Recurrent SEM is treated with RT, surgery or systemic therapy. Reirradiation carries only a small risk of radiation myelopathy, whereas half or more of the patients remain ambulatory after re-irradiation at one year follow-up (123). A third recurrence occurs in half of patients experiencing a second recurrence, most of whom will remain ambulatory after repeated and timely treatment, which confirms the importance of continuous neurological observation (121). Because MSEM occur as frequently in recurrent SEM as in initial SEM, imaging of as much of the spinal canal as possible is required also in recurrent SEM. Surgery is reserved for instability of the spine, bony compression of the spinal cord, serious clinical deterioration while on RT, for recurrent SEM in patients who have received the maximum tolerated dose of RT, and according to recent data in selected patients with a satisfactory medical status. Surgery should be performed at the site of the offending tumor. Because at least 85% of SEM invade the spinal canal anteriorly of the cord, an anterior approach including tumor and vertebral body resection and stabilization instrumentation will be the usual procedure. For this technique nonmetastatic adjacent vertebrae are required. Nonrandomized studies using this technique show prolonged clinical improvement in the vast majority of patients with a high proportion of nonambulatory patients regaining ability to walk (124–126). A randomized study compared immediate direct circumferential decompression of the spinal cord followed by RT (10 × 3 Gy) with the same RT regimen without surgery in patients with SEM restricted to a single area (127). Thirteen of the 101 included patients had breast cancer. In the surgery group the post-treatment ambulatory status was significantly better: 94% of the patients in the surgery group continued to walk, compared to 74% in the RT group; 62% in the surgery group regained the ability to walk, compared to 19% in the RT group. The 30-day mortality rates were 6% (surgery group) and 14% (RT group). Ten patients in the RT group crossed over to receive surgery: three regained the ability to walk, and four patients had surgical complications. This randomized trial shows that decompressive surgery as primary treatment should be considered more frequently in patients with satisfactory medical condition. Systemic therapy can be given as an adjuvantto RT or surgery. As sole treatment it is generally considered inadequate because it will take too long to induce tumor reduction. We have used chemotherapy and hormonal therapy as primary treatment in a few selected patients with SEM, predominantly in recurrent SEM after RT, with rate and duration of response being comparable to RT results (109,128). Thus, early recognition of (recurrent) SEM may permit successful treatment with systemic therapy. 2.1.6. Skull and Skull Base Metastasis Metastasis to the skull usually is asymptomatic if it is confined to the pain-insensitive bone marrow. It may become palpable by expanding the tabula interna and externa. Pain ensues if the tumor breaks through the bony cortex. Rarely, skull metastases may cause increased intracranial pressure or local infarction by compression of draining veins or sinus. Contrast-enhanced MRI is superior to CT with bone window setting for detecting subtle intradiploic metastases. It is also the best technique for evaluation of the patency of the dural sinuses that are adjacent to calvarial lesions. CT is a sensitive method to identify lesions confined to the bone, particularly when the metastasis is osteoblastic, whereas MRI is sensitive in identifying tumor growth along the nerve sheaths and into the extradural space. Occasionally skull base metastasis is visible on technetium SPECT scan when both MRI and CT are negative (129). Skull base metastasis is more likely than LM to cause cranial nerve involvement in patients without macroscopic intracranial lesions (130). Cranial nerves V and VII are involved most frequently. Sporadically isolated cranial nerve involvement occurs without LM or adjacent bone metastasis. It is presumably caused by hematogenous spread and is predominantly found in cranial nerves II and V (25). The numb chin syndrome, which is associated in particular with breast cancer and characterized by numbness restricted to the distribution of the mental nerve, is usually caused by bone metastasis in the mandible or at the base of the skull and less frequently by LM (131). Improvement of cranial nerve involvement from skull base metastasis can be expected in at least half of the patients following local RT or systemic therapy. Improvement is more likely to occur in recently developed cranial nerve dysfunction.

Chapter 22 / Neurologic Complications of Breast Cancer

437

2.1.7. Orbital and Ocular Metastasis Breast cancer may metastasize both to bone and soft tissue of the orbit. Orbital metastases develop an average of 5 years after initial cancer diagnosis. Presenting symptoms are proptosis, ptosis, pain, and diplopia. Orbital RT is the treatment of choice, but systemic treatment may also be successful. Orbital metastases are associated with bone metastases, and as a consequence the median survival of patients with orbital metastasis is relatively long (approximately 1.5 years). Ocular metastasis is probably associated with a less favorable prognosis than orbital metastasis (132). The majority of choroidal metastases remain asymptomatic. Symptoms of choroidal metastasis include decreased visual acuity, metamorphopsia, diplopia, and less frequently pain, and photophobia. RT is the treatment of choice. Except in unusual cases of anteriorly located lesions, RT to the lens can be avoided. Approximately 70% of patients improve, although vision may deteriorate for a few weeks following RT because of subretinal fluid and retinal detachment. Response of choroidal metastasis following systemic therapy has been documented; if the patient can be carefully observed by an ophthalmologist and shows no progression, such a treatment may be an option. 2.1.8. Tumor-Induced Plexopathy 2.1.8.1. Brachial Plexopathy. Despite proximity to the draining lymph nodes of the breast, tumor involvement of the brachial plexus is relatively uncommon. Lymph node metastases most frequently develop in the axillary region, so the lower trunk (root level C8 and T1) is usually involved. Involvement of the upper plexus (root level C5 and C6) due to supraclavicular lymph node metastasis is less common. The clinical picture of tumor-induced and radiation-induced plexopathy may be similar, but usually the two conditions can be differentiated after a careful history and neurological examination. The most prominent clinical difference is the frequency and intensity of pain as initial symptom. In tumorinduced plexopathy pain is almost always the first and dominant symptom (25). Initially it usually is located in and around the shoulder, irrespective of the site of plexus involvement. After a few weeks or months pain and paraesthesias radiate into the arm and hand, and motor signs develop corresponding to the site of plexus involvement. Medial extension of tumor along the lower trunk may induce a Horner’s syndrome and subsequently may progress into the epidural space at the level of C7–T1 and T1–T2. Physical examination often reveals a palpable mass in the axilla or in the supraclavicular fossa. Compression of the subclavian vein may cause an increased venous pattern around the shoulder. Lymphedema of the arm sometimes occurs when tumor infiltrates a previously irradiated brachial plexus. Imaging of the brachial plexus including the paraspinal and epidural space with CT or MR scan is required to determine the cause and the extent of the lesion. CT scan should preferably include both the symptomatic and asymptomatic plexus for comparison. CT may reveal soft tissue density changes in the brachial plexus and paraspinal extension of the tumor. Limitations of CT compared to MRI are the ability to image a single plane, artifacts from bone, and inaccuracy in distinguishing vascular structures from neural tissue. In addition, MRI is superior in imaging the epidural space. Tumor is best demonstrated on T1-weighted images. Early enhancement after bolus intravenous contrast helps differentiate tumor from radiation fibrosis (133). Loss of fat planes and local increased signal on T2-weighted images are not helpful in differentiating tumor from radiation injury. Overall, MRI diagnoses the nature of the plexus lesion correctly in more than 80% (134). In a selected group of 16 patients with tumor plexopathy MRI was nondiagnostic in 2 and CT in 6 patients. Theoretically, PET scanning may be a useful technique to differentiate between local tumor and radiation plexopathy, but at present PET scanning appears less accurate than MRI in defining the nature of the lesion. Particularly in diffusely infiltrating tumor, imaging techniques may be inconclusive. In these instances, electrophysiologic studies, a close follow-up, or eventually surgical exploration may be required. There are no electrophysiological features pathognomonic for tumor-induced plexopathy. Motor or sensory nerve conduction abnormalities are found in the majority of patients but do not differentiate between tumor-induced and radiation-induced plexopathy. Similarly, needle examination may reveal fibrillation potentials and fasciculations in both conditions. Myokymic discharges (pseudomyotonia) occur in about half of the patients with radiation-induced plexopathy and only very

438

Part VII / Neurologic Complications of Specific Malignancies

rarely in tumor-induced plexopathy, being the only electrophysiologic finding possibly useful in distinguishing the two conditions (135,136). Occasionally the clinical picture, imaging studies, and electrophysiologic studies cannot define the nature of the plexus lesion. The differential diagnosis also includes idiopathic neuralgic amyotrophy and ischemic brachial plexopathy. Ischemic brachial plexopathy results from a radiation-induced thrombotic occlusion of the subclavian artery (137). The clinical picture of acute paresis and paresthesias with absence of peripheral arterial pulsations and signs of ischemia is usually readily distinguishable from the other causes of plexopathy. Idiopathic plexus amyotrophy is characterized by acute and severe pain over the shoulder. After 1–2 weeks paresis develops, particularly in the muscles of the shoulder girdle, and pain subsides spontaneously. Paresis recovers gradually and usually completely within a few weeks or months. Dependent on the severity of pain and neurological deficit, the actual tumor state and the therapeutic options in case of local tumor, close clinical follow-up or occasionally surgical exploration may be considered. However, a negative exploration does not reliably exclude tumor. Moreover, debulking of tumor infiltrating the brachial plexus is unlikely to improve neurologic symptoms and function. Therefore, surgery should be considered only rarely and predominantly to differentiate between metastatic and primary or radiation-induced tumor. The latter is a consideration in patients at least several years out from their initial radiotherapy; the 10-year actuarial risk of radiation-induced sarcoma is estimated to be 0.8%. Treatment of tumor-induced plexopathy consists of pain management and antineoplastic therapy. In previously unirradiated patients RT is the treatment of choice. Local tumor control and pain relief is achieved in about 50– 75% of the patients. When radiation is not an option, systemic chemotherapy or hormonal therapy afford the only chance for tumor control, but this will be achieved in a minority of the patients. Prompt and adequate treatment of pain is essential to also to prevent development of a chronic pain syndrome and pain-induced dysfunction including frozen shoulder and dystrophy of the arm. In our experience corticosteroids are of some benefit in neoplastic brachial plexopathy. 2.1.8.2. Lumbosacral Plexopathy. Tumor involvement of the lumbosacral plexus in patients with breast cancer is caused by extension from bone metastases, particularly of the sacral bone. It typically starts with asymmetrical low back pain, radiating into the buttock and the posterior side of one leg. Pain often increases with recumbency. Bilateral radiating pain with incontinence indicates epidural tumor extension into the sacral spinal canal. On neurological examination, absence of radicular signs and normal or only slightly abnormal motor, sensory, and bladder function may help distinguish pure sacral plexopathy from leptomeningeal and epidural metastasis. CT or MR scan of the sacral area confirms the diagnosis, with MRI superior in demonstrating epidural extension of tumor. Local RT is the treatment of choice.

3. NONMETASTATIC COMPLICATIONS 3.1. Paraneoplastic Neurologic Disorders Paraneoplastic neurological disorders (PND) are very rare in patients with breast cancer. These disorders presumably arise when the primary tumor expresses antigens that are normally found only in neural tissue (onconeural antigens). The pathogenesis of PND is attributed to the immune response elicited by the tumor antigens which is also directed to the parts of the nervous system that share the same antigen (25,138). This explains why the tumor status associated with PND is usually limited; often the neurological signs precede the diagnosis of the primary tumor. Antibodies against discrete onconeural antigens are found in serum and cerebrospinal fluid in some patients with PND, but with the exception of the antibodies in Lambert–Eaton myasthenic syndrome in patients with small cell lung cancer, these antibodies do not appear to be pathogenic. In breast cancer patients a few distinct clinico-pathological paraneoplastic syndromes are recognized. Paraneoplastic cerebellar degeneration (PCD) is the most common syndrome, with an estimated incidence of one in 3000 patients. The cerebellar disease is usually subacute in onset and stabilizes after a couple of weeks. Dysarthria and truncal ataxia are prominent. Most patients are unable to walk, write, or read because of diplopia or oscillopsia. After they are stabilized, the neurologic deficits do not change even with successful treatment of the primary tumor.

Chapter 22 / Neurologic Complications of Breast Cancer

439

Histopathology shows loss of Purkinje cells with diffuse lymphocytic infiltration. Specific antibodies against cytoplasmic Purkinje cell antigens (anti-Yo) are found in serum and CSF of a substantial proportion of patients with PCD, but therapy directed against antibody activity does not affect the course of the disease (139). Cytotoxic T-lymphocytes reactive to the onconeural antigen were demonstrated in blood and CSF of patients with PCD, indicating that a T-cell–mediated autoimmune response probably is responsible for the Purkinje cell degeneration. Tacrolimus, the effect of which is directed against activated T-cells, may be effective provided that it is given in the very early stage of disease (140). The differential diagnosis of PCD includes posterior fossa metastasis, Wernicke disease, infectious or postviral encephalitis, multiple sclerosis, and spinocerebellar atrophy. Additionally, the paraneoplastic opsoclonus– myoclonus syndrome (OMS) may somewhat resemble PCD. Saccadic eye movements and opsoclonus are often associated with truncal ataxia, dysarthria, and vertigo. Symptoms of OMS may fluctuate spontaneously or react to cancer treatment, and sometimes respond to treatment with steroids or benzodiazepines, suggesting a functional rather than structural neural damage (141,142). MRI is normal, and CSF is normal or may show some pleocytosis. OMS in breast cancer patients is associated with antibodies (anti-Ri) that react with virtually all neuronal nuclei of CNS. Cerebellar and brain stem dysfunction without opsoclonus do not rule out the possibility of anti-Ri associated paraneoplastic disorder (141). Stiff-man syndrome is a very rare paraneoplastic disorder described in a few patients with breast cancer. It is characterized by axial rigidity with painful spasms that responds to intravenous benzodiazepines and disappears during sleep. Anti-amphiphysin antibodies identified in the serum of these patients react with synapses of CNS neurons (138). Paraneoplastic sensory neuropathy (PSN) predominantly occurs in patients with small cell lung cancer, but sometimes is associated with breast cancer. Histopathology shows degeneration of the dorsal root ganglion neurons with perivascular inflammatory infiltrates. Symptoms are painful paresthesias and numbness in arms or legs of asymmetric distribution. Cranial nerves may be involved, leading to facial numbness, loss of taste, or deafness. Patients may become wheelchair-bound because of sensory ataxia, although the course of disease in breast cancer often appears less severe than in patients with small cell lung cancer. Anti-Hu antibodies may be found in patients with PSN, both in serum and in CSF. Electrophysiologic studies show decreased or absent sensory potentials and markedly decreased sensory conduction velocity, while motor nerve conduction is usually normal. The differential diagnosis includes chemotherapy-induced sensory neuropathy (taxanes), pyridoxine intoxication and immune mediated sensory neuropathy, particularly Sjögren’s syndrome. Patients with Sjögren’s syndrome do not harbor anti-Hu antibodies unless they also have cancer and PSN. Polymyositis (PM) and dermatomyositis (DM) are inflammatory, probably immune-mediated muscle diseases, that may occur more frequently in patients with breast cancer (25,143). Patients usually present with proximal muscular weakness. Muscular tenderness is somewhat less common. Symptoms often precede identification of tumor. There is no absolutely diagnostic laboratory test. ESR and creatinine kinase are usually elevated but may be normal. EMG changes include spontaneous fibrillation and polyphasic activity. Treatment consists of steroids and other immunosuppressants. Successful treatment of breast cancer may lead also to improvement in myositis.

3.2. Cerebrovascular Complications Patients with breast cancer appear to harbor some increased risk of cerebrovascular complications (144). Several factors have been implicated as possible etiologic mechanism. These include a tumor-related hypercoagulable state, nonbacterial thrombotic endocarditis, consumptive coagulopathy, and tumor-related thrombocytopenia. There is also a link between breast cancer and the metabolic syndrome with an increased risk of stroke (145). In addition, chemotherapy may be associated with an increased risk of thromboembolic complications (146). Protein C and protein S deficiency have been found during CMF chemotherapy (147). Other possible risk factors for stroke during chemotherapy are hypovolemia due to vomiting and a hypercoagulable state because of release of procoagulant tissue factors derived from tumor cells, or because of injury of endothelial cells. Deficiency of protein C and antithrombin III has been described in patients receiving tamoxifen. On the other hand, use of tamoxifen is associated with a decrease of LDL cholesterol. A clear increase in risk of stroke was not observed in patients receiving tamoxifen (148). Overall, some increase of stroke may occur in patients with breast cancer. A slight increase was observed during the first year after breast cancer diagnosis, possibly caused

440

Part VII / Neurologic Complications of Specific Malignancies

by tumor-related coagulation disorders (149). There are no laboratory tests specific for the hypercoagulable state in these patients. Treatment of these thrombotic events is the same as in patients without cancer. A decreased risk of stroke was observed in 10-year survivors of breast cancer (150). Notably, radiation fields including the carotid arteries were not associated with a higher risk of stroke (150,151). However, in long-survivors hormone therapy was associated with some increased risk of stroke (150). Intracranial hemorrhages are rare in patients with breast cancer, despite the high incidence of BM and occurrence of tumor or treatment-related thrombocytopenia. Hemorrhages into cerebral metastases occur in < 1% of patients with BM. Bleeding is usually from veins or small arterioles and often superficial. The prognosis for immediate survival and functional recovery is usually better than in hypertensive cerebral hemorrhage. In general, management and outcome of hemorrhagic brain metastases is the same as in non-hemorrhagic brain metastases. Subdural hematoma is usually caused by bleeding into a dural metastasis or by rupture of small vessels due to venous obstruction caused by dural tumor, with or without cancer-induced coagulopathy (144). Occasionally, insertion of an Ommaya reservoir for the treatment of LM is complicated by a subdural hematoma or hygroma. Subdural hematoma may be asymptomatic and an incidental finding on CT or MRI brain or at autopsy. Asymptomatic or small subdural hematomas or hygromas may be treated conservatively. Progressive lesions will require surgical evacuation and, in case of (sub)dural tumor, RT to prevent recurrence. Sinus thrombosis is a rare complication in patients with breast cancer, occurring when a metastatic tumor compresses the sinus. Treatment-induced protein C or protein S deficiency is apparently not associated with an increased risk of sinus thrombosis.

3.3. Infection Infections of the nervous system are uncommon in patients with breast cancer. Patients who are immunocompromised from chemotherapy or chronic corticosteroid use are accordingly susceptible to the usual microorganisms. About 5–10% of the patients treated for leptomeningeal metastasis with intraventricular chemotherapy develop infectious meningitis (60,78). Staphylococcus epidermidis usually is the infecting organism. With appropriate intravenous and intraventricular antibiotics the Ommaya reservoir can be preserved in the majority of cases. Infectious meningitis may run a fulminant and fatal course, particularly in patients receiving corticosteroids.

3.4. Metabolic Disorders Metabolic CNS dysfunction is a common complication in patients with cancer. In a survey of neurological complications in patients referred to the Memorial Sloan-Kettering Cancer Center, more than 10% of patients had an admitting diagnosis of metabolic encephalopathy. In 61% of these patients metabolic or drug-related encephalopathy was the cause of mental disturbance (152). The most common causes are use of opioids, sepsis, and electrolyte imbalance. Often the clinical picture results from multiple systemic factors. Both overdose and withdrawal of drugs can cause a confusional state or delirium. An adrenal crisis with decreased consciousness may be the result of withdrawal after prolonged use of steroids. Hypercalcemia probably is the most important cause of metabolic encephalopathy in breast cancer patients, usually associated with extensive bone metastasis. It can also cause proximal weakness and diminished reflexes.

4. NEUROLOGIC COMPLICATIONS OF TREATMENT 4.1. Neurologic Complications of Cytostatic Drugs 4.1.1. Central Nervous System Toxicity Systemic chemotherapy regimens as employed in patients with breast cancer are not associated with specific acute complications of the CNS. Mild long-term CNS toxicity, consisting of some cognitive impairment, has been observed in breast cancer patients treated adjuvantly with high dose chemotherapy (153), and also after standard CMF (154). Severe chemotherapy-induced CNS toxicity is predominantly observed after intrathecal (IT) chemotherapy. Use of IT MTX can be complicated by acute, subacute, and late neurotoxicity (Table 3) (155). Acute aseptic meningitis may develop within a few hours or days after IT MTX. Symptoms are usually mild and self-limiting

Chapter 22 / Neurologic Complications of Breast Cancer

441

Table 3 Neurologic Complications of Intraventricular Methotrexate Related to Ommaya reservoir

Related to Methotrexate +/– RT

Bacterial meningitis Intracranial hemorrhage Subdural hematoma/hygroma Focal leukoencephalopathy due to CSF leakage along the reservoir Aseptic meningitis (transient) Seizures (very rare) Acute encephalomyelopathy (rare) Subacute encephalopathy (usually mild and transient) Optic neuropathy/radiculopathy Myelopathy Late leukoencephalopathy (progressive, 50% of long survivors)

if treatment is discontinued. Rarely, and sometimes concurrent with MTX meningitis the patient may develop transient or progressive encephalopathy or myelopathy. Rapidly fatal cases have been considered idiosyncratic, but in patients with a subacute course axonal swelling and demyelination were found, similar to the findings in late leukoencephalopathy. Spinal roots and cranial nerves may be involved (156). Mild and transient subacute encephalopathy seems related to development of late and progressive leukoencephalopathy (94). Leukoencephalopathy is the most important late neurologic complication of IT MTX. It usually occurs after combined treatment of cranial RT and IT MTX, but it may also develop after IT MTX alone (93,94). The first symptoms of subcortical dementia and ataxia usually are noted between 4 and 6 months after IT MTX. Recovery is observed only very rarely. The beneficial effect of leucovorin is uncertain. The risk of leukoencephalopathy is related to the dose of whole-brain RT, the cumulative dose of MTX, and the presence of CSF flow disturbances. Histopathology shows foci of demyelination and coagulation necrosis, axonal swelling and relative absence of inflammatory reaction. The pathogenesis of these changes is still unknown. Damage to endothelial cells, oligodendrocytes and microglia, but also primarily to the neuron itself have been postulated. Myelopathy has also been observed following IT Ara-C and IT Thiotepa. The combination of spinal RT and previous IT MTX administration increases the risk of Ara-C myelopathy, which is clinically and histopathologically indistinguishable from MTX myelopathy (95). The combination of IT Ara-C and cranial RT may be complicated by optic neuropathy (157). Hormonal agents are very rarely associated with neurotoxicity. Optic neuropathy, retinopathy, and reversible encephalopathy have been reported following Tamoxifen (155,158). 4.1.2. Peripheral Neuropathy The taxanes (paclitaxel and docetaxel) are the cytotoxic drugs principally responsible for peripheral neuropathy in breast cancer patients. These agents interact with microtubule activity and so affect axoplasmic transport. Paclitaxel induces a predominantly sensory, symmetrically distributed distal polyneuropathy. Paresthesias, numbness, and sometimes pain in feet and hands are early symptoms. Myalgia and arthralgia are transient disabling symptoms, especially with high doses. Neurotoxicity is uncommon at doses below 200 mg/m² per course until a cumulative dose of about 1500 mg/m² is administered. In diabetics, neurotoxicity may be dose limiting at 175 mg/m² (159). Motor signs are mild; some weakness may develop at doses of 250 mg/m². Optic nerve toxicity causing transient scotomas has been reported at doses between 175 and 225 mg/m² (160). Paclitaxel neuropathy is at least partly reversible. Agents to prevent this toxicity have not been successful. Neuropathy due to docetaxel is usually mild, with acral paresthesias and numbness, and occasionally Lhermitte’s sign (161). Symptoms may become seriously disabling with cumulative doses of more than 600 mg/m². Predominantly proximal motor weakness has also been reported (162). Taxane-induced neuropathy may deteriorate the first weeks after cessation of treatment.

442

Part VII / Neurologic Complications of Specific Malignancies

4.2. Neurologic Complications of Radiotherapy Serious neurologic complications of radiotherapy (RT) are uncommon in patients with breast cancer. Occasionally whole-brain RT (WBRT) is followed within 2 months by somnolence, anorexia, and irritability. This early delayed reaction is attributed to damage of the oligodendrocytes (163). CT or MRI may show some increased subcortical hypodensity. Usually, this syndrome recovers completely within a few weeks or months. Late reactions are caused by vascular injury in combination with demyelination and are usually irreversible. They may develop with conventional RT fractionation schemes. Symptoms range from some neuropsychological impairment to severe dementia. Pathological changes include atrophy, hydrocephalus, leukoencephalopathy and focal necrosis, not necessarily at the site of the irradiated BM and not always restricted to the white matter. These complications, occurring a median of 1 year after WBRT, are rare in our experience, although one study estimated their occurrence in almost 20% of long survivors treated with conventional RT (164). Transient paresthesias and Lhermitte’s sign may develop a few months after RT that included the spinal cord. These are attributed to RT-induced demyelination of the posterior columns. Late RT myelopathy is irreversible and progresses to a complete paraplegia in one-half of patients. Vascular injury is the pathogenic factor, initially affecting the posterior columns. Because paraplegia is such a debilitating complication, the threshhold of tolerance is usually set at a low dose level of 50 Gy in 2.0 Gy fractions or 33 Gy in 3.0 Gy fractions. Clinical practice shows that re-irradiation of the spinal column for recurrent EM, with a cumulative dose well above this threshold is only rarely complicated by radiation myelopathy (123). Moreover, the average interval of 5–6 months to onset of RT myelopathy after re-irradiation should be weighed against the expected survival of the patient with recurrent EM. Brachial plexopathy is a well-known complication of radiation therapy in breast cancer patients. Two distinct entities are identified: an early, transient, and usually mild plexopathy, and the more common delayed and often progressive plexopathy. Early and transient plexopathy has been reported in only two series of patients (165,166). Symptoms usually are mild and consist of paresthesias, and mild shoulder and axilla pain. Weakness occurred in about one-half of patients. Symptoms develop a few months after radiation therapy and usually resolve completely within 1 year. The incidence of this mild and reversible plexopathy was 1% and 1.4% in the two series. No definite relationship was found with RT dose. Concurrent chemotherapy may be a contributing factor (166). Occasionally such early and reversible plexopathy has a more serious and prolonged course. The pathogenesis of this apparently distinct syndrome of transient plexopathy is unclear. Classic radiation-induced brachial plexopathy is caused by perineural fibrosis. Epineural vessel wall thickening occurs that may eventually result in vascular occlusion. Surgical dissection contributes to the extent of fibrosis and microvascular damage. The development of plexus fibrosis and its course are related to the total dose of RT and fraction size. Stoll in 1966 reported an incidence of plexus injury in 73% of patients who had received a total dose of 56.7 Gy in large fractions of 4.7 Gy (167). Fraction size appears to be the crucial factor. In a randomized study with 45 Gy in 15 fractions versus 54 Gy in 30 fractions the incidence of radiation-induced brachial plexus injury was 5.9% and 1.0%, respectively (168). Obviously, overlap of radiation fields will increase the risk of plexus injury. In a retrospective analysis adjuvant chemotherapy and younger age (premenopausal) portended an increased risk of plexopathy (169), but others found no significant impact of chemotherapy or hormone therapy (170). Clinical symptoms usually occur 1-4 years from RT, but may vary from a few months to > 15 years. Paresthesias are the presenting symptom in the vast majority of patients. Pain, usually described as dull and mildly disabling, is an initial symptom in one-quarter of the patients. Gradually weakness develops. The predominant level of clinical involvement is the upper trunk or the entire plexus in most of the patients (123). It has been suggested that the clavicle may spare the lower plexus to some extent to radiation injury. Horner’s syndrome is extremely rare in radiation plexopathy. Arm lymphedema is found in a substantial proportion of patients, but it probably does not influence the further course of disease. The natural history is highly variable; weakness may stabilize or increase to complete paralysis. Occurrence of severe pain is uncommon and should lead to further investigation for recurrent tumor. In addition to the history and neurological examination CT or MRI scan should be performed to differentiate between tumor-induced and radiation-induced plexopathy. Myokymic discharges on EMG are highly suggestive of radiation-induced plexopathy (135,136). There is no treatment that can reverse the neurologic damage. Hyperbaric oxygen has no significant effect (171). Surgery

Chapter 22 / Neurologic Complications of Breast Cancer

443

may be associated with worsening of fibrosis and vascular damage. The effect of anticoagulants or antiplatelet drugs has not been investigated in controlled studies. Treatment consists of measurements to manage subluxation of the shoulder and lymphedema of the arm and physical therapy to try to improve or maintain residual strength.

4.3. Neurologic Complications of Surgery Axillary lymph node dissection may be complicated by a chronic deafferentiation pain due to a lesion of the intercostobrachial nerve (172). Pain usually develops days or weeks after surgery with disturbed sensation of the lateral chest wall, in the axilla, and over the inner area of the upper arm. Shoulder movement increases the pain, and frozen shoulder can develop. Tricyclic antidepressants, carbamazepine, or topical lidocaine or capsaicin are the recommended treatments but usually are only moderately effective. Neuroablative techniques are not effective in deafferentiation pain, and may lead to increase of pain distributed over a larger skin area.

5. CONCLUSION Central nervous system metastasis usually develops in patients with widespread metastatic disease, but in a substantial subgroup it is the only site of active disease. In most of these patients neurologic disease determines the outcome. Making the right diagnosis in breast cancer patients with neurological complaints can be a real challenge. Clinically, but also radiologically, it can be difficult to differentiate between metastatic lesions at the different sites of the central and peripheral nervous system, and nonmetastatic neurologic disorders or treatment-related neurologic complications of the central or peripheral nervous system. Neurotoxicity will become increasingly important, since patients are treated more intensively and survival increases. Treatment of brain metastasis is no longer restricted to whole-brain radiotherapy and surgery, but now also includes stereotactic radiotherapy, systemic chemotherapy, and a combination or sequence of various treatment modalities. In patients with leptomeningeal metastasis it has been demonstrated that adding intraventricular chemotherapy to systemic therapy and involved field radiotherapy does generally not improve neurological outcome or survival and may adversely impact quality of life from its side effects. Early diagnosis of spinal epidural metastasis remains crucial for functional outcome. In patients with spinal epidural metastasis restricted to a single area decompressive surgery should be considered as primary treatment.

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13. 14. 15.

Armstrong K, Eisen A, Weber B. Assessing the risk of breast cancer. N Engl J Med 2000; 342:564–571. Lee Y-TN. Patterns of metastasis and natural courses of breast carcinoma. Cancer Metast Rev 1985; 4:153–172. Kerlikowske K, Grady D, Rubin SM et al. Efficacy of screening mammography: a meta-analysis. JAMA 1995; 273:149–154. Romond EH, Perez EA, Bryant J et al. Trastuzumab plus adjuvant chemotherapy for operable Her2-positive breast cancer. N Eng J Med 2005; 73:1673–1678. Hortobagyi GN. Treatment of breast cancer. N Engl J Med 1998; 339:974–984. Paik S, Bryant J, Park C et al. ErbB-2 and response to doxorubicin in patients with axillary lymph node-positive, hormone receptornegative breast cancer. J Natl Cancer Inst 1998; 90:1361–1370. Pierga JY, Fumoleau P, Brewer Y et al. Efficacy and safety of single-agent capecitabine in pretreated metastatic breast cancer patient from the French compassionate use program. Breast Cancer Res Treat 2004; 88:117–129. Cobleigh MA, Vogel CL, Tripathy D et al. Multinational study of the efficacy and safety of humanized anti-Her2 monoclonal antibody in women who have HER2-overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 1999; 17:2639–2648. Slamon DJ, Leyland-Jones B, Shak S et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that over expresses HER2. N Engl J Med 2001; 344: 783–792. Rodenhuis S, Richel DJ, van der Wall E et al. Randomised trial of high-dose chemotherapy and haemopoietic progenitor-cell support in operable breast cancer with extensive axillary lymph-node involvement. Lancet 1998 352:515–521. Stadtmauer EA, O’Neill A, Goldstein LJ et al. Conventional-dose chemotherapy compared with high-dose chemotherapy plus autologous hematopoietic stem-cell transplantation for metastatic breast cancer. N Engl J Med 2000; 342:1069–1076. Viadana E, Bross ID, Pickren JW. An autopsy study of some routes of dissemination of cancer of the breast. Br J Cancer 1973; 27:336–340. Hagemeister FB, Buzdar AU, Lunda MA et al. Causes of death in breast cancer: a clinicopathologic study. Cancer 1980; 46:162–167. Tsukada Y, Fouad A, Pickren J et al. Central nervous system metastasis from breast carcinoma: autopsy study. Cancer 1983: 52:2349–2354. Kamby C. The pattern of metastases in human breast cancer: methodological aspects and influence of prognostic factors. Cancer Treatm Rev 1990; 17:37–61.

444

Part VII / Neurologic Complications of Specific Malignancies

16. Ryberg M, Nielsen D, Osterlind K et al. Predictors of central nervous system metastasis in patients with metastatic breast cancer: a competing risk analysis of 579 patients treated with epirubicin-based chemotherapy. Breast Cancer Res Treat 2005; 91:217–225. 17. Lai R, Dang CT, Malkin MG et al. The risk of central nervous system metastases after trastuzumab therapy in patients with breast carcinoma. Cancer 2004; 101:810–816. 18. Pestalozzi BC, Zahrieh D, Price KN et al. Identifying breast cancer patients at risk for central nervous system (CNS) metastases in trials of the International Breast Cancer Study Group (IBCSG). Ann Oncol 2006; 17:935–944. 19. Miller KD, Weathers T, Haney LG et al. Occult central nervous system involvement in patients with metastatic breast cancer: prevalence, predictive factors and impact on overall survival. Ann Oncol 2003; 14:1072–1077. 20. Bendell JC, Domchek SM, Burstein HJ et al. Central nervous system metastases in women who receive trastuzumab-based therapy for metastatic breast carcinoma. Cancer 2003; 97:2972–2977. 21. Tham YL, Sexton K, Kramer R et al. Primary breast cancer phenotypes associates with propensity for central nervous system metastases. Cancer 2006; 107:696–704. 22. Distefano A, Yap Y, Hortobagyi GN et al. The natural history of breast cancer patients with brain metastases. Cancer 1979: 44:1913–1918. 23. Boogerd W, Vos VW, Hart AA et al. Brain metastases in breast cancer: natural history, prognostic factors, and outcome. J Neuro-oncol 1993; 15: 165–174. 24. Lower EE, Drosick DR, Blau R et al. Increased rate of brain metastasis with trastuzumab therapy not associated with impaired survival. Clin Breast Cancer 2003; 4:114–119. 25. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. 26. Kamby C, Soerensen PS. Characteristics of patients with short and long survivals after detection of intracranial metastases from breast cancer. J Neuro-oncol 1988; 6:37–45. 27. Diener-West M, Dobbins TW, Phillips TL et al. Identification of an optimal subgroup for treatment evaluation of patients with brain metastases using RTOG study 7916. Int J Radiat Oncol Biol Phys 1989: 16:669–673. 28. Stemmer HJ, Kahlert S, Siekiera W et al. Characteristics of patients with brain metastases receiving trastuzumab for HER2 overexpressing metastatic breast cancer. Breast 2006; 15:219–225. 29. Clayton AJ, Danson S, Jolly S et al. Imcidence of cerebral metastases in patients treated with trastuzumab for metastatic breast cancer. Br J Cancer 2004; 91: 639–643. 30. Borgelt N, Gelber R, Kramer S et al. The palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1980; 6:1–19. 31. Vecht CHJ, Hovestadt A, Verbiest HBC et al. Dose–effect relation of dexamethasone on brain tumor edema: a randomized study with 16 mg, 8 mg, and 4 mg per day. Neurology 1994; 44: 675–680. 32. Coia LR. The role of radiation therapy in the treatment of brain metastases. Int J Radiat Oncol Biol Phys 1992; 23:229–238. 33. Priestman TJ, Dunn J, Brada M et al. Final results of the Royal College of Radiologists’ trial comparing two different radiotherapy schedules in the treatment of cerebral metastases. Clin Oncol 1996; 8:308–315. 34. Suh JH, Stea B, Nabid A et al. Phase III study of efaproxiral as an adjunct to whole-brain radiation therapy for brain metastases. J Clin Oncol 2006; 24:106–114. 35. Auchter RM, Lamond JP, Alexander E et al. A multi-institutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1996; 35:27–35. 36. Muacevic A, Kreth FW, Horstmann GA et al. Surgery and radiotherapy compared with gamma knife radiosurgery in the treatment of solitary cerebral metastases of small diameter. J Neurosurg 1999; 91:35–43. 37. Andrews DW, Scott CB, Sperduto PW et al. Whole-brain radiation therapy with or without stereotactic radiosurgery boost for patients with 1–3 brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004; 363:1665–1672. 38. Aoyama H, Shirato H, Tago M et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs. stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006; 295:2483–2491. 39. Huang C-F, Kondziolka D, Flickinger JC et al. Stereotactic radiosurgery for brainstem metastases. J Neurosurg 1999; 91:563–568. 40. Patchell RA, Tibbs PA, Regine WF et al. Post-operative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998; 280:1485–1489. 41. Rogers LR, Rock JP, Sills AK et al. Results of a phase II trial of the GliaSite radiation therapy system for the treatment of newly diagnosed, resected single brain metastases. J. Neurosurg 2006; 105:375–384. 42. Boogerd W, Hart AAM, Tjahja IS. Treatment and outcome of brain metastasis as first site of distant metastasis from breast cancer. J Neuro-oncol 1997; 35:161–167. 43. Bindal RK, Sawaya R, Leavens ME et al. Surgical treatment of multiple brain metastases. J Neurosurg 1993; 79:210–116. 44. Rosner D, Takuma N, Lane WW. Chemotherapy induces regression of brain metastases in breast carcinoma. Cancer 1986; 58:832–839. 45. Boogerd W, Dalesio O, Bais EM et al. Response of brain metastases from breast cancer to systemic chemotherapy. Cancer 1992; 69:972–980. 46. Franciosi V, Cocconi G, Michiara M et al. Front-line chemotherapy with cisplatin and etoposide for patients with brain metastases from breast carcinoma, non-small cell lung carcinoma, or malignant melanoma. Cancer 1999; 85:1599–1605. 47. Wang MLH, Yung WAK, Royce ME et al. Capecitabine for 5-fluorouracil-resistant brain metastases from breast cancer. Am J Clin Oncol 2001; 24:421–424. 48. Abrey LE, Olson JD, Raizer JJ et al. A phase II trial of temozolomide for patients with recurrent or progressive brain metastases. J Neuro-oncol 2001; 53:259–265. 49. Rivera E, Meyers C, Grovers M et al. Phase I study of capecitabine in combination with temozolomide in the treatment of patients with brain metastases from breast carcinoma. Cancer 2006; 107:1348–1354.

Chapter 22 / Neurologic Complications of Breast Cancer

445

50. Lassman AB, Abrey LE, Shah GG et al. Systemic high-dose intravenous methotrexate for central nervous system metastases. J Neurooncol 2006: 78:255–260. 51. Carey RW, Davis JM, Zervas NT. Tamoxifen-induced regression of cerebral metastases in breast carcinoma. Cancer Treatm Rep 1981; 65:793–795. 52. Van der Gaast A, Alexieva-Figusch J, Vecht C et al. Complete remission of a brain metastases to third-line hormonal treatment with megestrol acetate. Am J Clin Oncol 1990; 13:507–509. 53. Kaba SE, Kyritsis P, Hess K et al. TPDC-FuHu chemotherapy for the treatment of recurrent metastatic brain tumors. J Clin Oncol 1997; 15:1063–1070. 54. Ledermann G, Wronski M, Fine M. Fractionated radiosurgery for brain metastases in 43 patients with breast carcinoma. Breast Cancer Res and Treat 2001; 65:145–154. 55. Wrónski M, Arbit E, McCormick B. Surgical treatment of 70 patients with brain metastases from breast carcinoma. Cancer 1997; 80:1746–1754. 56. Abrams HL, Spiro R, Goldstein N. Metastases in carcinoma: analysis of 1000 autopsied cases. Cancer 1950; 3:74–85. 57. Fassett DR, Couldwell WT. Metastases to the pituitary gland. Neurosurg Focus 2004;16: E8. 58. Sioytos P, Yen V, Arbit E. Pituitary gland metastases. Ann Surg Oncol 1996; 3(1):94–99. 59. Yap HY, Tashima CK, Blumenschein GR et al. Diabetes insipidus and breast cancer. Arch Intern Med 1979; 139:1009–1011. 60. Boogerd W, Hart AAM, van der Sande JJ et al. Meningeal carcinomatosis in breast cancer: prognostic factors and influence of treatment. Cancer 1991; 67:1685–1695. 61. Jayson C, Howell A, Harris M et al. Carcinomatous meningitis in patients with breast cancer. Cancer 1994; 74:3135–3141. 62. Clamon G, Doebbeling B. Meningeal carcinomatosis from breast cancer: spinal cord vs. brain involvement. Breast Cancer Res Treatm 1987; 9:213–217. 63. Kokkoris CHP. Leptomeningeal carcinomatosis: how does cancer reach the pia-arachnoid? Cancer 1983; 51:154–160. 64. Murray JJ, Greco FA, Wolff SN et al. Neoplastic meningitis: marked variations of cerebrospinal fluid composition in the absence of extradural block. Am J Med 1983; 75:289–294. 65. Boogerd W, Vroom ThM, van Heerde P et al. CSF cytology versus immunocytochemistry in meningeal carcinomatosis. J Neurol Neurosurg Psychiatry 1988; 51:142–145. 66. Cibas ES, Malkin MG, Posner JB et al. Detection of DNA abnormalities by flow cytometry in cells from cerebrospinal fluid. Am J Clin Pathol 1987; 88:570–577. 67. Van Oostenbrugge RJ, Hopman AHN, Arends JW et al. Treatment of leptomeningeal metastases evaluated by interphase cytogenetics. J Clin Oncol 2000; 18:2053–2058. 68. Wasserstrom WR, Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal carcinomatosis from solid tumors: experience with 90 patients. Cancer 1982; 49:759–772. 69. Bach F, Bjerregaards B, Sölétormos G et al. Diagnostic value of cerebrospinal fluid cytology in comparison with tumor marker activity in central nervous system metastases secondary to breast cancer. Cancer 1993; 72:2376–2382. 70. Stockhammer G, Poewe W, Burgstaller S et al. Vascular endothelial growth factor in CSF. Neurology 2000; 54:1670–1675. 71. Herrlinger U, Wiendl H, Renninger M et al. Vascular endothelial growth factor (VEGF) in leptomeningeal metastasis: diagnostic and prognostic value. Br J Cancer 2004; 91:219–224. 72. Pannullo SC, Reich JB, Krol G et al. MRI changes in intracranial hypotension. Neurology 1993; 43:919–926. 73. Freilich RJ, Krol G, DeAngelis LM. Neuroimaging and cerebrospinal fluid cytology in the diagnosis of leptomeningeal metastasis. Ann Neurol 1995; 38:51–57. 74. Straathof CSM, de Bruin HG, Dippel DWJ et al. The diagnostic accuracy of magnetic resonance imaging and cerebrospinal fluid cytology in leptomeningeal metastasis. J Neurol 1999; 246:810–814. 75. Hitchins RN, Bell DR, Woods RL et al. A prospective trial of single agent versus combination chemotherapy in meningeal carcinomatosis. J Clin Oncol 1987; 5:1655–1662. 76. Blaney SM, Balis FM, Poplack DG. Pharmacologic approaches to the treatment of meningeal malignancy. Oncology 1991; 5:107–116. 77. Shapiro WR, Young DF, Metha BM. Methotrexate: distribution in the cerebrospinal fluid after intravenous, intraventricular and lumbar injections. N Engl J Med 1975; 293:161–166. 78. Lishner M, Perrin RG, Feld R et al. Complications associated with Ommaya reservoirs in patients with cancer: the Princess Margaret Hospital experience and a review of the literature. Arch Intern Med 1990; 150:173–176. 79. Siegal T, Lossos A, Pfeffer MR. Leptomeningeal metastases: analysis of 31 patients with sustained off-therapy response following combined-modality therapy. Neurology 1994; 44:1463–1469. 80. Shapiro WR, Posner JB, Ushio Y et al. Treatment of meningeal neoplasms. Cancer Treatm Rep 1977; 61:733–743. 81. Ongerboer de Visser BW, Somers R, Nooyen WH et al. Intraventricular methotrexate therapy of leptomeningeal metastases from breast carcinoma. Neurology 1983; 33:1565–1572. 82. Grossman SA, Finkelstein DM, Ruckdeschel JC et al. Randomized prospective comparison of intraventricular methotrexate and thiotepa in patients with previously untreated neoplastic meningitis. J Clin Oncol 1993; 11:561–569. 83. Grant R, Naylor B, Greenberg HS et al. Clinical outcome in aggressively treated meningeal carcinomatosis. Arch Neurol 1994; 51:457–461. 84. Orlando L, Curigliano G, Colleoni M et al. Intrathecal chemotherapy in carcinomatous meningitis from breast cancer. Anticancer Res 2002; 22:3057–3059. 85. Chamberlain MC, Tsao-Wei D, Groshen S. Neoplastic meningitis related encephalopathy: prognostic significance. Neurology 2004; 63:2159–2161. 86. Bleyer WA, Poplack DG, Simon RM et al. “Concentration x time” methotrexate via a subcutaneous reservoir: a less toxic regimen for intraventricular chemotherapy of central nervous system neoplasms. Blood 1978: 51:835–842.

446

Part VII / Neurologic Complications of Specific Malignancies

87. Glantz MJ, Hall WA, Cole BF et al. Diagnosis, management, and survival of patients with leptomeningeal cancer based on cerebrospinal fluid-flow status. Cancer 1995; 75:2919–2931. 88. Chamberlain MC, Kormanik PA. Prognostic significance of 111 Indium-DTPA CSF flow studies in leptomeningeal metastases. Neurology 1996; 46:1674–1677. 89. Mason WP, Yeh SD, DeAngelis LM. 111 Indium-diethylenetriamine pentaacetic acid cerebrospinal fluid flow studies predict distribution of intrathecally administered chemotherapy and outcome in patients with leptomenigeal metastases. Neurology 1998; 50:438–444. 90. Glantz MJ, Jaeckle KA, Chamberlain MC et al. A randomized controlled trial comparing intrathecal sustained-release cytarabine (DepoCyt) to intrathecal methotrexate in patients with neoplastic meningitis from solid tumors. Clin Cancer Res 1999; 5:3394–3402. 91. Jaeckle KA, Phuphanich S, Bent MJ et al. Intrathecal treatment of neoplastic meningitis due to breast cancer with a slow-release formulation of cytarabine. Br J Cancer 2001; 84:157–163. 92. Jaeckle KA, Batchelor T, O’Day SJ et al. An open label trial of sustained-release cytarabine (DepoCyt) for the intrathecal treatment of solid tumor neoplastic meningitis. J Neuro-oncol 2002; 57:231–239. 93. Bleyer WA. Neurologic sequelae of methotrexate and ionizing radiation: a new classification. Cancer Treatm Rep 1981; 65: 89–98. 94. Boogerd W, van der Sande JJ, Moffie D. Acute fever and delayed leukoencephalopathy following low dose intraventricular methotrexate. J Neurol Neurosurg Psychiatry 1988; 51:1277–1283. 95. Watterson J, Toogood I, Nieder M et al. Excessive spinal cord toxicity from intensive central nervous system-directed therapies. Cancer 1994; 74:3034–3041. 96. Nakagawa H, Yamada M, Maeda N et al. Clinical trial of intrathecal administration of 5-fluoro-2’-deoxyuridine for treatment of meningeal dissemination of malignant tumors. J Neuro-oncol 1999; 45:175–183. 97. Benjamin JC, Moss T, Moseley RP et al. Cerebral distribution of immunoconjugate after treatment for neoplastic meningitis using an intrathecal radiolabeled monoclonal antibody. Neurosurg 1989; 25: 253–258. 98. Moseley RP, Benjamin JC, Ashpole RD et al. Carcinomatous meningitis: antibody-guided therapy with I-131 HMFG1. J Neurol Neurosurg 1991; 54:260–265. 99. Platini C, Long J, Walter S. Meningeal carcinomatosis from breast cancer treated with intrathecal trastuzumab. Lancet Oncol 2006; 7:778–780. 100. Siegal T, Sandbank U, Gabrizon A et al. Alteration of blood–brain–CSF barrier in experimental meningeal carcinomatosis: a morphologic and adriamycin-penetration study. J Neuro-oncol 1987; 4:233–242. 101. Ushio Y, Shunizuk K, Aragaki Y et al. Alteration of blood–CSF barrier by tumor invasion into the meninges. J Neurosurg 1981; 55:445–449. 102. Bokstein F, Lossos A, Siegal T. Leptomeningeal metastases from solid tumors. Cancer 1998; 82:1756–1763. 103. Glantz MJ, Cole BF, Recht L et al. High dose intravenous methotrexate for patients with nonleukemic leptomeningeal cancer: is intrathecal chemotherapy necessary? J Clin Oncol 1998; 16:1561–1567. 104. Tetef ML, Margolin KA, Doroshow JH et al. Pharmacokinetics and toxicity of high-dose intravenous methotrexate in the treatment of leptomingeal carcinomatosis. Cancer Chemother Pharmacol 2000; 46:19–26. 105. Boogerd W, Bent MJ van den, Koehler PJ et al. The relevance of intraventricular chemotherapy for leptomeningeal metastasis in breast cancer: a randomized study. Eur J Cancer 2004; 40:2726–2733. 106. Giglio P, Tremont-Lukats IW, Groves MD. Response of neoplastic meneingitis from tumors to oral capecitabine. J Neuro-oncol 2003; 65:167–172. 107. Rogers LR, Remer SE, Tejwani S. Durable response of breast cancer leptomeningeal metastasis to capecitabine monotherapy. Neurooncol 2004; 6(1):63–64. 108. Boogerd W, Dorresteijn LDA, van der Sande JJ et al. Response of leptomeningeal metastases from breast cancer to hormonal therapy. Neurology 2000; 55:117–119. 109. Boogerd W, van der Sande JJ, Kröger R. Early diagnosis and treatment of spinal epidural metastasis in breast cancer: a prospective study. J Neurol Neurosurg Psychiatry 1992; 55:1188–1193. 110. Schiff D, O’Neill BP, Wang C-H et al. Neuroimaging and treatment implications of patients with multiple epidural spinal metastases. Cancer 1998; 83:1593–1601. 111. Boogerd W, van der Sande JJ. Diagnosis and treatment of spinal cord compression in malignant disease. Cancer Treatm Rev 1993; 19: 129–150. 112. Helweg-Larsen S, Rasmussen B, Sörensen PS. Recovery of gait after radiotherapy in paralytic patients with metastatic epidural spinal cord compression. Neurology 1990; 40: 1234–1236. 113. Graus F, Krol G, Foley K. Early diagnosis of spinal epidural metastasis: correlation with clinical and radiological findings. Proc ASCO 1985; 4:269. 114. Boogerd W, Kröger R. Intravenous contrast in spinal computed tomography to identify epidural metastases. Clin Neurol Neurosurg 1991; 93:195–199. 115. Helweg-Larsen S, Hansen SW, Sorensen PS. Second occurrence of symptomatic metastatic spinal cord compression and findings of multiple spinal epidural metastases. Int J Radiat Oncol Biol Phys 1995; 33:595–598. 116. Maranzano E, Latini P. Effectiveness of radiation therapy without surgery in metastatic spinal cord compression: final results from a prospective trial. Int J Radiat Oncol Biol Phys 1995; 32:959–967. 117. Greenberg HS, Kim JH, Posner JB. Epidral spinal cord compression from metastatic tumor: results with a new treatment protocol. Ann Neurol 1980; 8:361–366. 118. Sorensen PS, Helweg-Larsen S, Mouridsen H et al. Effect of high-dose dexamethasone in carcinomatous metastatic spinal cord compression treated with radiotherapy: a randomised trial. Eur J Cancer 1994; 30:22–27. 119. Delattre JY, Arbit E, Thaler HT et al. A dose–response study of dexamethasone in a model of spinal cord compression caused by epidral tumor. J Neurosurg 1989; 70:920–925.

Chapter 22 / Neurologic Complications of Breast Cancer

447

120. Kaminskyi HJ, Diwan VG, Ruff RL. Second occurrence of spinal epidural metastases. Neurology 1991; 41:744–746. 121. Van der Sande JJ, Boogerd W, Kröger R et al. Recurrent spinal epidural metastases: a prospective study with a complete follow-up. J Neurol Neurosurg Psychiatry 1996; 66:623–627. 122. Rades D, Stalpers LJA, Veninga T et al. Evaluation of five radiation schedules and prognostic factors for metastatic spinal cord compression. J Clin Oncol 2005; 23:3366–3375. 123. Schiff D, Shaw EG, Casino TL. Outcome after spinal reirradiation for malignant epidural spinal cord compression. Ann Neurol 1995; 37:583–589. 124. Siegal T, Siegal T. Surgical decompression of anterior and posterior malignant epidural tumors compressing the spinal cord: a prospective study. Neurosurg 1985; 17:424–432. 125. Fidler MW. Anterior decompression and stabilization of metastatic spinal fractures. J Bone Joint Surg 1986; 68:83–90. 126. Sundaresan N, Digiacinto GV, Hughes JEO et al. Treatment of neoplastic spinal cord compression: results of a prospective study. Neurosurg 1991; 29:645–650. 127. Patchell RA, Tibbs PA, Regine WD et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomized trial. Lancet 2005; 366:643–648. 128. Boogerd W, van der Sande JJ, Kröger R et al. Effective systemic therapy for spinal epidural metastasis from breast carcinoma. Eur J Cancer Clin Oncol 1989; 25:149–153. 129. Jansen BP, Pillay M, de Bruin HG et al. 99m Tc SPECT in the diagnosis of skull base metastasis. Neurology 1997; 48:1326–1330. 130. Hall SM, Buzdar U, Blumenschein R. Cranial nerve palsies in metastatic breast cancer due to osseous metastasis without intracranial involvement. Cancer 1983: 52:180–184. 131. Bruyn RPM, Boogerd W. The numb chin. Clin Neurol Neurosurg 1991; 93:187–193. 132. McCormick BA, Abramson DH. Ocular metastases. In: Harris JR (ed.). Diseases of the Breast. Philadelphia: Lippincott;2000:889–891. 133. Dao TH, Rahmouni A, Campana F et al. Tumor recurrence versus fibrosis in the irradiated breast: differentiation with dynamic gadolinium-enhanced MR imaging. Radiology 1993; 1187:751. 134. Thyagarajan D, Cascino T, Harms G. Magnetic resonance imaging in brachial plexopathy of cancer. Neurology 1995; 45:421–427. 135. Lederman RJ, Wilbourn AJ. Brachial plexopathy: recurrent cancer or radiation? Neurology 1984; 34:1331–1335. 136. Harper CM, Thomas JE, Cascino TL et al. Distinction between neoplastic and radiation-induced brachial plexopathy, with emphasis on the role of EMG. Neurology 1989; 39:502–506. 137. Gerard JM, Franck N, Moussa Z et al. Acute ischemic brachial plexus neuropathy following radiation therapy. Neurology 1989; 39:450–451. 138. Dalmau JO, Posner JB. Paraneoplastic syndromes. Arch Neurol 1999; 56:405–408. 139. Greenlee JE. Cytotoxic T cells in paraneoplastic cerebellar degeneration. Ann Neurol 2000; 47:4–5. 140. Albert ML, Austin LM, Darnell RB. Detection and treatment of activated T cells in the cerebrospinal fluid of patients with paraneoplastic cerebellar degeneration. Ann Neurol 2000; 47:9–17. 141. Escudero D, Barnadas A, Codina M et al. Anti-Ri-associated paraneoplastic neurologic disorder without opsoclonus in a patient with breast cancer. Neurology 1993; 43:1605–1606. 142. Dropcho EJ, Kline LB, Riser J. Antineuronal (anti-Ri) antibodies in a patient with steroid-responsive opsoclonus–myoclonus. Neurology 1993; 43:207–211. 143. Buchbinder R, Forbes A, Hall S et al. Incidence of malignant disease in biopsy-proven inflammatory myopathy. Ann Intern Med 2001; 134:1087–1095. 144. Rogers LR. Cerebrovascular complications in cancer patients. Oncology 1994; 8:23–30. 145. Stoll BA. Western nutrition and the insulin resistance syndrome: a link to breast cancer. Eur. J Clin Nutr 1999; 53(2):83–87. 146. Wall JG, Weiss RB, Norton L et al. Arterial thrombosis associated with adjuvant chemotherapy for breast carcinoma: a Cancer and Leukemia Group B study. Am J Med 1989; 87:501–504. 147. Rogers JS, Murgo AJ, Fontana JA et al. Chemotherapy for breast cancer decreases plasma protein C and protein S. J Clin Oncol 1988; 6:276–281. 148. Saphner T, Tormey DC, Gray R. Venous and arterial thrombosis in patients who received adjuvant therapy for breast cancer. J Clin Oncol 1991; 9:286–294. 149. Nilsson G, Holmberg L, Garmo H et al. Increased incidence of stroke in women with breast cancer. Eur J Cancer 2005; 41:423–429. 150. Hooning MJ, Dorresteijn LD, Aleman BM et al. Decreased risk of stroke among 10-year survivors of breast cancer. J Clin Oncol 2006; 24:5388–5394. 151. Woodward WA, Giordano SH, Duan Z et al. Supraclavicular radiation for breast cancer does not increase the 10-year risk of stroke. Cancer 2006; 106:2556–2562. 152. Clouston PD, De Angelis LM, Posner JB. The spectrum of neurological disease in patients with systemic cancer. Ann Neurol 1992; 31:268–273. 153. Van Dam FSAM, Schagen SB et al. Impairment of cognitive function in women receiving adjuvant treatment for high-risk breast cancer: high-dose versus standard-dose chemotherapy. J Natl Cancer Inst 1998; 90:210–218. 154. Schagen SB, van Dam FSAM, Muller MJ et al. Cognitive deficits after post-operative adjuvant chemotherapy for breast carcinoma. Cancer 1999; 85:640–650. 155. Boogerd W. Neurological complications of chemotherapy. In: Vinken PJ, Bruyn GW (eds.). Handbook of Clinical Neurology. Elsevier;1995:527–546. 156. Boogerd W, Moffie D, Smets LA. Early blindness and coma during intrathecal chemotherapy for meningeal carcinomatosis. Cancer 1990; 65:452–457. 157. Fishman ML, Bean SC, Gogan DG. Optic atrophy following prophylactic chemotherapy and cranial radiation for acute lymphocytic leukemia. Am J Ophthalmol 1976; 82:571–576.

448

Part VII / Neurologic Complications of Specific Malignancies

158. Pavlidis NA, Petris C, Briassoulis E et al. Clear evidence that long-term, low-dose tamoxifen treatment can induce ocular toxicity: a prospective study of 63 patients. Cancer 1992; 69:2961–2964. 159. Postma TJ, Vermorken JB, Liefting AJM et al. Paclitaxel-induced neuropathy. Ann Oncol 1995; 6:489–494. 160. Capri G, Munzone E, Tarenzi E et al. Optic nerve disturbances: a new form of paclitaxel neurotoxicity. J Natl Cancer Inst 1994; 86:1099–1101. 161. Hilkens PHE, Verweij J, Stoter G et al. Peripheral neurotoxicity induced by docetaxel. Neurology 1996; 46:104–108. 162. Freilich RJ, Balmaceda C, Seidman AD et al. Motor neuropathy due to docetaxel and paclitaxel. Neurology 1996; 47:115–118. 163. Van Daal WAJ, van der Kogel AJ. Side effects of radiation on the nervous system. In: Twijnstra A, Keyser A, Ongerboer de Visser BW (eds.). Neuro-oncology. Elsevier;1993:357–365. 164. DeAngelis LM, Delattre J-Y, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology 1989; 39: 789–796. 165. Salner AL, Botnick LE, Herzog AG et al. Reversible brachial plexopathy following primary radiation therapy for breast cancer. Cancer Treat Rep 1981; 65:797–802. 166. Pierce SM, Recht A, Lingos TI et al. Long-term radiation complications following conservative surgery and radiation therapy in patients with early stage breast cancer. Int J Radiation Oncol Biol Phys 1992; 23:915–923. 167. Stoll BA, Andrews JT. Radiation-induced peripheral neuropathy. Br Med J 1966; 1:834–837. 168. Powell S, Cooke J, Parsons C. Radiation-induced brachial plexus injury: follow-up of two different fractionation schedules. Radiother Oncol 1990; 18:213–220. 169. Olsen NK, Pfeiffer P, Johannsen L et al. Radiation-induced brachial plexopathy: neurological follow-up in 161 recurrence-free breast cancer patients. Int J Rad Oncol Biol Phys 1993; 26:43–49. 170. Bajrovic A, Rades D, Fehlauer F et al. Is there a life-long risk of brachial plexopathy after radiotherapy of supravlavicular lymph nodes in breast cancer patients? Radiother Oncol 2004; 71:297–301. 171. Pitchard J, Anand P, Broome J et al. Double-blind randomized phase II study of hyperbaric oxygen in patients with radiation-induced brachial plexopathy. Radiother Oncol 2001; 58:279–286. 172. Vecht CHJ, van de Brand HJ, Wajer OJM. Post-axillary dissection pain in breast cancer due to a lesion of the intercostobrachial nerve. Pain 1989; 38:171–176.

23

Neurologic Complications of Female Reproductive Tract Cancer Lauren E. Abrey,

MD

CONTENTS Introduction Ovarian Cancer Other Gynecologic Cancers Neurologic Complications of Tumor Neurologic Complications of Treatment Conclusion References

Summary Neurologic problems associated with gynecologic malignancy are uncommon. However, the full spectrum of neurologic complications is more common than generally appreciated. This chapter reviews the common malignancies of the female reproductive tract and the specific neurologic problems encountered as a result of these tumors or their treatment. Key Words: paraneoplastic syndrome, gynecologic cancer

1. INTRODUCTION Neurologic problems in patients with gynecologic malignancy may be the result of treatment, metastatic disease, paraneoplastic syndromes, or coincidental neurologic disease (Table 1). Gynecologic cancers are uncommon causes of brain metastases and neurologic paraneoplastic syndromes are rare, but the full spectrum of neurologic complications in these patients is more common than generally appreciated. Metabolic encephalopathies and secondary neurologic complications are common in patients with cancer and are the most common reason for neurologic consultation. In particular, patients with pelvic tumors are at high risk to develop obstructive renal failure with uremia, mental status changes and seizures. This chapter will review the common malignancies of the female reproductive tract and the specific neurologic problems encountered as a result of these tumors or their treatment.

2. OVARIAN CANCER Ovarian cancer is the most deadly of the female reproductive tract malignancies with only 39% of all women surviving for 5 years. Approximately 24,000 new cases are diagnosed and 13,600 deaths attributed to ovarian cancer annually (1). Oral contraceptives may reduce the risk of ovarian cancer, but infertility treatments have been associated with an increased risk. Family history is the most significant risk factor, and there are several familial ovarian cancer syndromes that include associations with breast, endometrial and colorectal cancer. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

449

450

Part VII / Neurologic Complications of Specific Malignancies

Table 1 Incidence of Neurologic Complications in Gynecologic Malignancy Incidence Tumor-related Brain metastases Spine metastases and cord compression Leptomeningeal carcinomatosis Lumbosacral plexopathy Paraneoplastic disorders Treatment-related Surgical complications Peripheral nerve injury Radiotherapy complications Lumbosacral plexopathy Chemotherapy complications Peripheral neuropathy Ifosfamide encephalopathy

0–20% < 1% < 1% 2% < 1%

1% 0.2% up to 60% up to 30%

The disease is often clinically silent in its early stages. Increasing abdominal girth with or without abdominal pain or bloating is the most common initial symptom, usually a manifestation of advanced disease. As many as 85% of tumors are malignant at diagnosis and may spread via local extension, lymphatic invasion, peritoneal implants, and hematogenous metastasis. Tumor markers such as lactate dehydrogenase (LDH), human chorionic gonadotropin (HCG), and alphafetoprotein (AFP) are useful in distinguishing germ cell tumors from ovarian carcinoma. Serum CA 125 is a valuable marker of disease activity in serous ovarian carcinoma. Ovarian cancer is treated with a combination of surgery and adjuvant therapy. The role of surgery is two-fold. Firstly, methodical surgical procedure is critical to accurately stage patients; secondly, optimal debulking and cytoreduction are therapeutic. Pelvic or abdominal radiotherapy is important adjuvant therapy for patients with stage II and III disease. Chemotherapy—most commonly regimens incorporating either paclitaxel or a platinumbased chemotherapy—is used for patients with advanced-stage disease and may be used preoperatively to decrease tumor size and allow optimal debulking. Ovarian germ cell tumors are a relatively uncommon subset of ovarian cancer seen most often in young women. A combination of chemotherapy and conservative surgery may be used in select patients in an effort to preserve fertility.

3. OTHER GYNECOLOGIC CANCERS 3.1. Endometrial Cancer Endometrial cancer is the most common cancer of the female reproductive tract and ranks as the fourth most common cancer of women behind breast, lung, and colorectal carcinoma. In 1994, there were 31,000 new cases of endometrial cancer and 5,900 deaths (1). The median age at diagnosis is 58 years and most patients are postmenopausal; however, 25% of cases occur in premenopausal women. Risk factors include obesity, diabetes, nulliparity, unopposed estrogen exposure, pelvic irradiation, and late menopause (2). Unopposed endogenous or exogenous estrogen appears to play the most important role in pathogenesis. The vast majority of patients present with post-menopausal bleeding, and the older the patient is, the greater the likelihood that cancer is the cause of the bleeding. Locally, endometrial cancer metastasizes to the vagina, peritoneum, and inguinal lymph nodes, while hematogenous spread results in metastases to the lungs, liver, and bones. Elevated levels of CA 125 at diagnosis are highly correlated with advanced, metastatic disease. Total abdominal hysterectomy with bilateral salpingo-oophorectomy (TAH/BSO) is the standard treatment; the removal of the ovaries is important as these often contain micrometastatic disease. Furthermore, the ovaries are a source of estrogen production that may stimulate growth of residual tumor. Adjuvant radiotherapy is recommended for all patients with advanced-stage disease or any known risk factor for more aggressive tumor. Advanced

Chapter 23 / Neurologic Complications of Female Reproductive Tract Cancer

451

endometrial cancer should be treated with a multimodality approach including surgery, radiation, chemotherapy, and/or progestin therapy. The most common chemotherapeutic agents used are cisplatin and doxorubicin.

3.2. Cervical Cancer Cervical cancer is diagnosed in approximately 15,000 women annually and accounts for 4,600 cancer-related deaths (1). Risk factors include lower socioeconomic status, history of multiple sexual partners, intercourse at a young age, and a large number of pregnancies (3). It is rare in sexually inactive or nulliparous women. Human papillomavirus and several oncogenes, including c-myc and ras, have been strongly implicated in the pathogenesis of cervical cancer (4–6). Cervical cancer spreads by local extension and lymphatic invasion; tumor infiltrates locally to involve the upper vagina, parametria, bladder, and rectum. The typical clinical presentation is a complaint of vaginal bleeding, including heavy menses, intermenstrual bleeding, and post-coital bleeding. Hydronephrosis and uremia are late or end-stage symptoms. Distant metastases are relatively uncommon. The treatment of cervical cancer is based on the clinical stage of the disease. Surgical resection—radical hysterectomy—is recommended for locally confined disease and has an associated 5-year survival of 90%. Locally invasive tumors (stage IIB–IV) are treated routinely with post-operative radiotherapy to a total dose of 4500–5000 cGy with 5-year survivals ranging between 45% and 65%. Neoadjuvant chemotherapy may be useful to consolidate local disease to allow a safe radical hysterectomy and to decrease the frequency of regional lymph node metastases.

3.3. Other Gynecologic Malignancy Other primary gynecologic malignancies are rare and include fallopian tube cancer, vulvar malignancies, and vaginal carcinomas. These are primarily tumors that spread by local extension. Surgery and radiotherapy are the primary treatment modalities. Choriocarcinoma, a gestational trophoblastic tumor, is an aggressive germ cell tumor of particular neurologic interest because it frequently metastasizes to the brain. The prognosis for patients with this tumor has improved dramatically with the use of multiagent chemotherapy; however, up to 50% of deaths from choriocarcinoma are related to intracranial metastases (7).

4. NEUROLOGIC COMPLICATIONS OF TUMOR 4.1. Brain Metastases Brain metastases are relatively uncommon in patients with cancer of the female reproductive tract. However, an autopsy series from Memorial Sloan–Kettering Cancer Center found that 7% of patients with tumors of the female genital tract had intracranial metastasis; therefore, an estimated 1,764 patients with female genital tract tumors and intracranial tumor in the United States died in 1994 (8,9). While brain metastases can develop in any lobe of the brain, these pelvic tumors may have a predilection to metastasize to the posterior fossa (10). There has been concern that the incidence of brain metastases secondary to ovarian cancer may be increasing as a result of prolonged survival from improved treatment of systemic disease. Eradication of systemic tumor is increasingly successful, but the brain can act as a sanctuary site for tumor cells which then lead to central nervous system metastases. Longitudinal studies of ovarian cancer estimate that brain metastases occur in 0.29–4% of patients (11–16). Metastases may be single or multiple and can occur at any time during the course of the disease (17). A recent study suggests that ovarian cancer patients with CNS and non-CNS metastases have similar disease characteristics and outcome (18). Median survival following a diagnosis of brain metastases ranges from 1 to 6 months; however, aggressive multimodality therapy incorporating surgery, radiotherapy, and/or chemotherapy has been demonstrated to result in an improved overall survival (19,20). Combined treatment using stereotactic radiosurgery plus whole-brain radiotherapy has been reported to have a 69% radiographic response rate and a 2-year survival rate of 60% (21). Response to chemotherapy alone has been clearly documented (Fig. 1) and may be an important treatment alternative for patients previously treated with cranial radiation or for those patients in whom radiation may confer an unacceptable risk of neurologic morbidity.

452

Part VII / Neurologic Complications of Specific Malignancies

Fig. 1. Gadolinium-enhanced MRI scan of a 50-year-old ovarian cancer patient with a brain metastasis treated with ifosfamide. Panel A is pretreatment and panel B is after four cycles of chemotherapy. This patient was felt to be a poor candidate for radiotherapy as she also had a diagnosis of multiple sclerosis.

Fewer than 1% of patients with endometrial or cervical cancer develop symptomatic brain metastases, although autopsy series report rates as high as 10% (22,23). Metastases usually develop in the setting of widely disseminated disease although they are rarely the presenting symptom of the underlying tumor. Therefore, the diagnosis of brain metastases in a patient with cervical or endometrial cancer should prompt systemic restaging. Whole-brain radiation therapy is the standard of care; however, patients with single lesions may be candidates for surgical resection or stereotactic radiosurgery. Choriocarcinoma is the gynecologic malignancy most likely to metastasize to the brain and comprises up to 35% of all brain metastases ascribed to gynecologic malignancies (22). These metastases are particularly prone to hemorrhage and may be the presenting feature of the primary malignancy. The presence of pulmonary metastases, histologic invasion into surrounding tissue, and an increase in urine HCG levels during treatment are associated with an increased risk of brain metastases (7). Gadolinium-enhanced MRI of the brain should be performed in every patient with choriocarcinoma to evaluate the extent of disease. Unlike other brain metastases, these are often exquisitely sensitive to chemotherapy and whole brain radiation, allowing for the possibility of prolonged remission or even cure in a subset of patients. Therefore, these patients should have an aggressive multimodality approach to treatment including neurosurgical decompression in those patients with evidence of increased intracranial pressure. The use of corticosteroids to treat increased intracranial pressure may be relatively contraindicated in these patients as there are reports of steroids stimulating tumor growth (7,24).

4.2. Spine Metastases and Cord Compression A review of more than 23 series of patients suggests that gynecologic malignancies account for 3% or less of all epidural metastases causing spinal cord and cauda equina compression (25). However, pelvic tumors may metastasize directly to the lumbar vertebral bodies via Batson’s venous plexus without having evidence of other distant metastases, and autopsy series suggest that as many as 20% of women with tumors of the genitourinary tract have metastatic disease in the vertebral column (26). In addition, back pain is a common symptom of pelvic tumors and is the most common presenting symptom of spinal cord compression. Therefore, patients with back pain merit careful neurologic evaluation and consideration of epidural tumor. Diagnostic neuroimaging should be obtained in those patients with an abnormal neurologic evaluation or symptoms suggestive of spinal tumor

Chapter 23 / Neurologic Complications of Female Reproductive Tract Cancer

453

(increased pain when lying flat, urinary retention, radiculopathy, etc.). Prompt treatment with corticosteroids and radiation therapy is critical to preserve neurologic function in patients with spinal cord compression. In a series of neurologic consultations performed on patients with ovarian cancer there were only four with a vertebral body metastasis, all of whom presented with pain (17); none had evidence of spinal cord compression. A report of seven ovarian cancer patients with neurologic complications included four with epidural spinal cord compression (27). Three patients presented with leg weakness and sphincter dysfunction with compression at the level of the conus medullaris; the fourth had cervical cord compression with myelopathy. Two of the four patients had significant improvement following radiation therapy and one attained a remission lasting at least 44 months. A single patient with ovarian cancer and an intramedullary spinal cord metastasis has been reported (28). This patient had a cervical cord lesion presenting with progressive paraparesis and left arm weakness following treatment for brain metastases. She improved with focal radiotherapy and died 6 months later from progressive systemic tumor. Intramedullary spinal cord metastasis and epidural spinal cord compression are rare complications of cervical cancer with only two reported cases in a review of 2261 patients (23). Robinson et al., however, reported five patients with cervical cancer and spinal cord compression, two of whom were not known to have cancer. A thorough history revealed a complaint of abnormal vaginal bleeding in both patients (29). This series highlights the importance of considering pelvic malignancy in the patient whose presenting tumor symptom is spinal cord compression.

4.3. Leptomeningeal Carcinomatosis Meningeal carcinomatosis complicating malignant gynecologic neoplasms is extremely rare. The incidence of leptomeningeal metastases, however, appears to be increasing as more effective systemic treatments are developed. Diagnosis is usually established by the demonstration of malignant cells in the cerebrospinal fluid or unequivocal leptomeningeal tumor visualized on cranial or spinal MRI scan. For patients with ovarian cancer, CA 125 can be assayed in the CSF and may be useful in the diagnosis and monitoring of leptomeningeal tumor. Treatment options include whole-brain radiotherapy, radiation to symptomatic sites of disease, intrathecal chemotherapy, and systemic chemotherapy. In spite of these treatments prognosis is generally poor, with patients rarely surviving more than a few months (30,31).

4.4. Lumbosacral Plexopathy Lumbosacral plexopathy was the most common neurologic complication in a review of 2261 cervical cancer patients at the University of Kentucky, affecting 50 patients (2%) (23). All patients with plexopathy had a diagnosis of squamous cell cancer. Patients developed plexopathy on average 20 months after the diagnosis of cervical cancer. Pain was the most common presenting symptom. CT or MR scan demonstrated local compression by a retroperitoneal mass (tumor or lymphadenopathy) in the majority of patients. Half had extension of tumor into adjacent lumbar or sacral vertebrae and most had ipsilateral hydronephrosis. Survival was less than 6 months in all patients with plexopathy. Radiation therapy was helpful in controlling pain in approximately one-third of patients treated; however, no patient had improved neurologic function as a result of radiation. A review of 85 patients with malignant lumbosacral plexopathy included twelve (14%) with tumors of the female reproductive tract (six cervical cancer, three ovarian cancer, and three uterine cancer) (32). In this series pain preceded any clinical neurologic signs by weeks to months. Patients with genitourinary tumors were more likely to develop a panplexopathy with pain primarily localized in the lumbosacral region and variable radicular or referred pain. Sensory loss and paresthesias were typically seen in the anterior thigh and foot with weakness of knee flexion, ankle dorsiflexion, and inversion. Nearly half of patients studied also had evidence of epidural spinal tumor. Treatment provided symptomatic relief to only about one-third of patients, and the median survival was 5.5 months, emphasizing the need for aggressive pain and palliative care management in these patients.

4.5. Paraneoplastic Disorders Paraneoplastic cerebellar degeneration is the most common and best characterized of the neurologic paraneoplastic disorders. Patients with gynecologic malignancy develop a particular type of paraneoplastic cerebellar

454

Part VII / Neurologic Complications of Specific Malignancies

degeneration characterized by the anti-Yo antibody. In a report of 55 patients with anti-Yo–associated paraneoplastic cerebellar degeneration, 26 had a diagnosis of ovarian cancer, and eight had other gynecologic malignancies including endometrial, fallopian tube, and mesovarium carcinoma (33). Another patient developed typical paraneoplastic cerebellar degeneration with anti-Yo antibodies and an elevated CA 125 that decreased to normal levels following TAH/BSO despite the absence of tumor on detailed pathologic examination. In addition, a number of patients with gynecologic malignancy and paraneoplastic cerebellar degeneration have been described with other autoantibodies directed against the cytoplasm of Purkinje cells or other neurons. Most patients present with an acute to subacute onset of neurologic symptoms preceding the diagnosis of malignancy. The typical patient develops a pancerebellar syndrome characterized by axial and appendicular ataxia, dysarthria, and nystagmus. The nystagmus often has a downbeat component. These symptoms stabilize within weeks to months leaving the patient with severe neurologic disability. Most patients are unable to read or watch television because of oscillopsia, and many are wheelchair-bound as a result of gait ataxia. Treatment of the underlying malignancy often results in oncologic cure but the patient does not improve neurologically. Specific therapies aimed at treating the neurologic disease have been unsuccessful although there are a few individual reports of transient clinical improvement following plasmapheresis (34), high-dose corticosteroids (33) or other immunosuppressive agents. Although the neurologic disability often stabilizes, many patients die as a result of complications of immobility. Persistent anti-Yo antibody titers can be detected years after diagnosis and treatment in patients in remission from their cancer and with stable neurologic disability. The pathogenesis of paraneoplastic cerebellar degeneration is unknown. The presence of high titer antibody suggests an autoimmune etiology, but there is no direct evidence for either a humoral or cell-mediated immune reaction. Attempts to use paraneoplastic antibodies to create an animal model have been unsuccessful to date. The anti-Yo antibody interacts with the cdr2 antigen expressed in the cytoplasm of Purkinje cells. This onconeural antigen is also expressed by many breast and ovarian cancers including tumors in neurologically normal patients (35). There is evidence that patients with paraneoplastic cerebellar degeneration have expanded populations of cdr2 specific cytotoxic T-lymphocytes in their blood and cerebrospinal fluid, leading to the hypothesis that these lymphocytes may play an important role in the etiology of paraneoplastic cerebellar degeneration as well as mediating antitumor immunity (36). Ongoing clinical investigations will try to diminish this T-cell population in patients with active paraneoplastic cerebellar degeneration in an effort to decrease Purkinje cell destruction and minimize neurologic dysfunction (37). Recently, a new syndrome of paraneoplastic encephalitis associated with ovarian teratoma has been described (38). Patients present with acute psychiatric symptoms, seizures, and central hypoventilation, often requiring ventilatory support. Serum and CSF antibodies to hippocampal NMDA receptor subunits have been identified as pathogenic. Recognition of this disorder is critical as the majority of patients recover following tumor resection with or without the use of immunosuppressants. Other neurologic paraneoplastic disorders have been reported in patients with gynecologic malignancy including limbic encephalitis, retinal degeneration, and opsoclonus associated with the anti-Ri antibody (39). There is also a single case report of a patient with a mixed mullerian tumor of the uterus developing myasthenia gravis with a positive Tensilon test and elevated acetylcholine receptor antibodies (40). Interestingly, her tumor contained a mesenchymal element with striated muscle differentiation.

4.6. Cerebrovascular Disease Cerebrovascular disease was more common than anticipated in a series of ovarian cancer patients, accounting for 12% of neurologic consults (19). Large retrospective series indicate that 15% of all cancer patients have cerebrovascular lesions at autopsy; only half are ever symptomatic and the pathogenesis of cerebrovascular disease is usually cancer-related. Ovarian cancer patients suffering a cerebrovascular event were older than the average patient and most had risk factors for stroke that were not related to their cancer. However, all of the patients had active advanced stage ovarian cancer and in more than one-third the etiology of the cerebrovascular event was related to the tumor or treatment. Volume depletion, electrolyte abnormalities, and alteration in baseline blood pressure are all potential side effects of abdominal surgery and chemotherapy, and may increase the risk of stroke in patients with

Chapter 23 / Neurologic Complications of Female Reproductive Tract Cancer

455

preexisting cerebrovascular risk factors. Hypercoagulable states and nonbacterial thrombotic endocarditis are complications of disseminated cancer, and patients with the mucinous form of ovarian cancer are particularly prone to develop a coagulopathy. Hemorrhagic stroke or cerebral sinus thrombosis may occur following chemotherapy. Choriocarcinoma may result in formation of a neoplastic aneurysm via tumor embolization to the brain and present with subarachnoid or intracerebral hemorrhage. Neoplastic aneurysm should be considered in the differential diagnosis of subarachnoid hemorrhage in women of childbearing age (41). Neonates with intracerebral hemorrhage secondary to metastatic choriocarcinoma originating in the placenta have also been reported. The treatment should include combined chemotherapy and radiation in conjunction with definitive treatment of the aneurysm. The role of neurosurgery must be decided on a case-by-case basis (42,43). Some patients may benefit from neurosurgical decompression of a large hematoma with improvement in neurologic function as well as their ability to tolerate chemotherapy or cranial radiotherapy. The benefit of resecting or clipping an unruptured neoplastic aneurysm is unclear although some authors report improved survival (44).

5. NEUROLOGIC COMPLICATIONS OF TREATMENT 5.1. Surgical Complications 5.1.1. Peripheral Nerve Injury Patients undergoing hysterectomy or radical pelvic surgery for tumor debulking are at risk for peripheral nerve injury (45,46). There are no good prospective studies that assess the risk of neuropathy, but the available reports in the literature suggest that the risk is quite low. Femoral neuropathy is the most common injury reported, but damage to any of the individual nerves or lumbosacral plexus may occur. The most likely etiology is compression either as a result of self-retaining retractors or the positioning of the patient in stirrups. Excessive flexion or external rotation of the hip may result in stretch injury to the femoral nerve. Surgical ligation or severe vascular injury is rare. Co-morbidities such as diabetes mellitus or uremia play an important role in patient susceptibility and prognosis. Most patients have a gradual recovery of neurologic function over weeks to months. Delayed neuropathies, particularly of the ilioinguinal or iliohypogastric nerve, can also be the result of surgery if the nerve becomes entrapped secondary to postoperative scar tissue or adhesions. 5.1.2. Other An unusual case of a vaginal CSF leak was seen in a 42-year-old woman with endometrial cancer following pelvic exenteration. In the post-operative period the patient complained of clear colorless fluid leaking from her vagina after assuming an upright posture. She subsequently developed a postural headache consistent with intracranial hypotension. A nuclear medicine study confirmed leakage of CSF into her pelvis and onto a sanitary napkin. A myelogram demonstrated a fistula of her S3-4 nerve root. The patient was managed conservatively with bed rest and placement of a lumbar drain and had complete resolution of her symptoms at the time of discharge. This complication has not previously been reported following gynecologic surgery but is a known complication of thoracic procedures (47).

5.2. Radiotherapy Complications 5.2.1. Plexopathy Radiation-induced lumbosacral plexopathy is a risk for patients treated with local radiotherapy to the pelvic organs (48). It is distinguished from metastatic plexopathy by the absence of tumor on imaging studies, the presence of bilateral signs and symptoms, predominant motor dysfunction, and the long survival of afflicted patients. Radiation induced plexopathy may develop from months to years after treatment with cases reported up to 14 years. Patients typically develop a flaccid painless weakness of the legs without bowel or bladder dysfunction; sensory symptoms are present in roughly one-half of patients. MRI or CT scanning should be used to exclude recurrent tumor. Myokymic discharges seen on electromyogram differentiate radiation-induced plexopathy from a compressive etiology. No effective treatment exists, but patients may benefit from physical therapy.

456

Part VII / Neurologic Complications of Specific Malignancies

5.3. Chemotherapy Complications 5.3.1. Peripheral Neuropathy Cis-platinum and paclitaxel alone or in combination are commonly used in the treatment of gynecologic malignancy and account for the majority of peripheral neuropathies seen in these patients. Cis-platinum typically produces a large fiber neuropathy with loss of vibratory and position sense, while paclitaxel affects all sensory fibers and may also cause a proximal motor neuropathy. Either concomitant or sequential administration of cis-platinum and paclitaxel enhances neurotoxicity and can result in debilitating peripheral neuropathy (49,50). The only available intervention for chemotherapy-induced peripheral neuropathy is to discontinue the chemotherapy. This may be difficult with cisplatin, which may cause a progressive neuropathy after discontinuation of the agent. Maximal nerve damage may have occurred without the patient or physician realizing it during the course of treatment. However, if debilitating neuropathy develops during therapy, stopping or reducing the dose of the offending agent is often necessary. This is a difficult decision if the drug is having a beneficial antineoplastic effect. The neurologist must eliminate other potential causes of peripheral neuropathy so that effective chemotherapy is not stopped unnecessarily. Several potential neuroprotective and therapeutic agents (Org 2766, nerve growth factors, and amifostine) are in different stages of preclinical and clinical evaluation (51–53).

5.4. Ifosfamide Encephalopathy Ifosfamide is an alkylating agent used in the treatment of both cervical and ovarian carcinoma. Up to 30% of patients treated with ifosfamide will develop an encephalopathy. The risk of encephalopathy is increased by oral administration or rapid intravenous infusion; therefore, ifosfamide is administered over 5 days. In addition, patients with renal dysfunction, hypocalcemia, or those receiving sedatives are at increased risk of encephalopathy. The clinical picture ranges from mild somnolence to an agitated delirium to deep coma; cerebellar dysfunction, hallucinations, and seizures are common. The encephalopathy is usually reversible within 3–4 days, but prolonged symptoms and death have been reported. Methylene blue has been reported as an effective treatment in shortening the duration of symptoms and may be used prophylactically in patients who need to be retreated with ifosfamide after developing an encephalopathy (54).

5.5. Other Intra-arterial chemoembolization has been used in the treatment of predominantly unilateral cervical cancer in an attempt to minimize the sequelae of systemic chemotherapy administration. In one study utilizing a combination of cisplatin and collagen injected into the internal iliac artery, three patients developed significant neurologic toxicity (55). All three patients had evidence of an acute femoral neuropathy thought to be ischemic in origin; in two of the three the sciatic nerve was also affected. In each patient the onset of symptoms began abruptly within 12 hrs of the chemoembolization procedure. Neurologic recovery occurred over a period of several months.

6. CONCLUSION Neurologic complications of female reproductive tract tumors are more common and more diverse than usually recognized. In particular, patients with pelvic tumors are at high risk of local neurologic complications, including compression of the lumbosacral plexus or peripheral nerves, metastasis to the spinal column, and neurologic sequelae of obstructive hydronephrosis. As systemic treatment of gynecologic malignancy improves, it is likely that neurologists will encounter increasing numbers of patients with metastasis to the central nervous system. Studies of neurologic paraneoplastic syndromes may provide important clues to tumor immunology and its therapeutic potential.

REFERENCES 1. Boring, CC, Squires TS, Tong T et al. Cancer statistics. CA Cancer J Clin 1994;44:7–26. 2. Park, RC, Grigsby PW, Muss, HB et al. Corpus: epithelial tumors. In: Hoskins WJ, Perez CA, Young RC (eds.). Principles and Practice of Gynecologic Oncology. Philadelphia: J.B. Lipincott; 1992:664. 3. Morrow, CP, Townsend DE. Tumors of the cervix. In: Morrow CP, Townsend DE (eds.). Synopsis of Gynecologic Oncology. 3rd ed. New York: John Wiley; 1987:107.

Chapter 23 / Neurologic Complications of Female Reproductive Tract Cancer

457

4. Syrjanen, K, Mantyjarvi R, Vayrynen M et al. Evolution of human papillomavirus infections in the uterine cervix during a long-term prospective follow-up. Appl Pathol 1987;5:121–135. 5. Riou GF, Bourhis J, Le MG. The c-myc proto-oncogene in invasive carcinomas of the uterine cervix: clinical relevance of overexpression in early stages of the cancer. Anticancer Res 1990;10:1225–1231. 6. Hayashi Y, Hachisuga T, Iwasaka T et al. Expression of ras oncogene product and egf receptor in cervical squamous cell carcinomas and its relationship to lymph node involvement. Gynecol Oncol 1991;40:147–151. 7. Ishizuka T, Tomoda Y, Kaseki S et al. Intracranial metastasis of choriocarcinoma. Cancer 1983;52:1896–1903. 8. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. 9. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol 1978;19:575–587. 10. Delattre JY, Krol G, Thaler HT et al. Distribution of brain metastases. Arch Neurol 1988;45:741–744. 11. Stein M, Steiner M, Klein B et al. Involvement of the central nervous system by ovarian carcinoma. Cancer 1986;58:2066–2068. 12. Larson D, Copeland L, Moser R et al. Central nervous system metastases in epithelial ovarian cancer. Obstet Gynecol 1986;68:746–750. 13. Dauplat J, Nieberg RK, Hacker NF. Central nervous system metastases in epithelial ovarian cancer. Cancer 1987;60:2559–2562. 14. Deutsch M, Beck D, Manor D et al. Metastatic brain tumor following second look operation in ovarian carcinoma. Gynecol Oncol 1987;27:116–120. 15. Ross WM, Carmichael VA, Shelley WE. Advanced carcinoma of the ovary with central nervous system relapse. Gynecol Oncol 1988;30:398–406. 16. Ziegler J, Gliedman P, Fass D et al. Brain metastases from ovarian cancer. J Neuro-oncol 1987;5:211–215. 17. Pectasides D, Aravantinos G, Fountzilas G et al. Brain metastases from epithelial ovarian cancer. The Hellenic Cooperative Oncology Group (HeCOG) Experience and review of the literature. Anticancer Res. 2005;25:3553–3558. 18. Kastritis E, Efstathiou E, Gika D et al. Brain metastases as isolated site of relapse in patients with epithelial ovarian cancer previously treated with platinum and paclitaxel-based chemotherapy. Int J Gynecol Cancer 2006;16:994–999. 19. Abrey LE, Dalmau JO. Neurologic complications of ovarian carcinoma. Cancer 1999;85:127–133. 20. Rodriguez GC, Soper JT, Berchuck A et al. Improved palliation of cerebral metastases in epithelial ovarian cancer using a combined modality approach including radiation therapy, chemotherapy, and surgery. J Clin Oncol 1992;10:1553–1560. 21. Corn BW, Greven KM, Randall ME et al. The efficacy of cranial irradiation in ovarian cancer metastatic to the brain: analysis of 32 cases. Obstet Gynecol 1995;86:955–959. 22. Kottke-Marchant K, Estes ML, Nunez C. Early brain metastases in endometrial carcinoma. Gynecol Oncol 1991;41:67–73. 23. Saphner T, Gallion HH, Van Nagell JR et al. Neurologic complications of cervical cancer. Cancer 1989;64:1147–1151. 24. Tomoda Y, Fuma M, Saiki N et al. Immunologic studies in patients with trophoblastic neoplasia. Am J Obstet Gynecol 1976;126: 661–667. 25. Kapp DS, LiVolsi VA, Kohorn EI. Cauda equina compression secondary to metastatic carcinoma of the uterine corpus: preservation of neurologic function and long-term survival following surgical decompression and radiation therapy. Gynecol Oncol 1985;20:209–218. 26. Wong DA, Fornasier VL, MacNab I. Spinal metastases: the obvious, the occult, and the impostors. Spine 1990;15(1):1–4. 27. Hoffman JS, Pena YM. Central nervous system lesions and advanced ovarian cancer. Gynecol Oncol 1988;30:87–97. 28. Thomas AW, Simon SR, Evans C. Intramedullary spinal cord metastases from epithelial ovarian carcinoma. Gynecol Oncol 1992;44:195–197. 29. Robinson WR, Muderspach LI. Spinal cord compression in metastatic cervical cancer. Gynecol Oncol 1993;48:269–271. 30. Khalil AM, Yamout BI, Tabbal SD et al. Case report and review of literature: leptomeningeal relapse in epithelial ovarian cancer. Gynecol Oncol 1994;54:227–231. 31. Aboulafia DM, Taylor LP, Crane RD et al. Carcinomatous meningitis complicating cervical cancer: a clinicopathologic study and literature review. Gynecol Oncol 1996;60:313–318. 32. Jaeckle KA, Young DF, Foley KM. The natural history of lumbosacral plexopathy in cancer. Neurology 1985;35:8–15. 33. Peterson K, Rosenblum MK, Kotanides H et al. Paraneoplastic cerebellar degeneration. I. a clinical analysis of 55 anti-Yo antibodypositive patients. Neurology 1992;42:1931–1937. 34. Cocconi G, Ceci G, Juvarra G et al. Successful treatment of subacute cerebellar degeneration in ovarian carcinoma with plasmapheresis. Cancer 1985;56:2318–2320. 35. Darnell JC, Albert ML, Darnell RB. Cdr2, a target antigen of naturally occurring human tumor immunity, is widely expressed in gynecological tumors. Cancer Res 2000;60(8):2136–2139. 36. Albert ML, Darnell JC, Bender A et al. Tumor-specific killer cells in paraneoplastic cerebellar degeneration. Nat Med 1998;4(11): 1321–1324. 37. Albert ML, Austin LM, Darnell RB. Detection and treatment of activated T cells in the cerebrospinal fluid of patients with paraneoplastic cerebellar degeneration. Ann Neurol 2000;47(1):9–17. 38. Dalmau J, Tuzun E, Wu H-Y et al. Paraenoplastic anti-NMDA receptor encephalitis associated with ovarian teratoma. Ann Neurol 2007;61:25–36. 39. Ashour AA, Verschraegen CF, Kudelka AP et al. Paraneoplastic syndromes of gynecologic neoplasms. J Clin Oncol 1997;15: 1272–1282. 40. Ariad S, Geffen DB, Yanai-Inbar I et al. Myasthenia gravis associated with malignant mixed mullerian tumor of the uterus. Gynecol Oncol 1997;64:510–515. 41. Kalafut M, Vinuela F, Saver JL et al. Multiple cerebral pseudoaneurysms and hemorrhages: the expanding spectrum of metastatic cerebral choriocarcinoma. J Neuroimaging 1998;8(1):44–47. 42. Fujiwara T, Mino S, Nagao S et al. Metastatic choriocarcinoma with neoplastic aneurysms cured by aneurysm resection and chemotherapy: case report. J Neurosurg 1992;76(1):148–151.

458

Part VII / Neurologic Complications of Specific Malignancies

43. Giannakopoulos G, Nair S, Snider C et al. Implications for the pathogenesis of aneurysm formation: metastatic choriocarcinoma with spontaneous splenic rupture: case report and a review. Surg Neurol 1992;38(3):236–240. 44. Chandra SA, Gilbert EF, Viseskul C et al. Neonatal intracranial choriocarcinoma. Arch Pathol Lab Med 1990;114(10):1079–1082. 45. Hoffman MS, Roberts WS, Cavanagh D. Neuropathies associated with radical pelvic surgery for gynecologic cancer. Gynecol Oncol 1988;31(3):462–466. 46. Alsever JD. Lumbosacral plexopathy after gynecologic surgery: case report and review of the literature. Am J Obstet Gynecol 1996;174:1769–1778. 47. Assietti R, Kibble MB, Bakay RA. Iatrogenic cerebrospinal fluid fistula to the pleural cavity: case report and literature review. Neurosurgery 1993;33(6):1104–1108. 48. Georgiou A, Grigsby PW, Perez CA. Radiation-induced lumbosacral plexopathy in gynecologic tumors: clinical findings and dosimetric analysis. Int J Rad Oncol Biol Phys 1993;26(479):482. 49. Cavaletti G, Bogliun G, Marzorati L et al. Peripheral neurotoxicity of Taxol in patients previously treated with cisplatin. Cancer 1995;75:1141–1150. 50. Berger T, Malayeri R, Doppelbauer A et al. Neurological monitoring of neurotoxicity induced by paclitaxel/cisplatin chemotherapy. Eur J Cancer 1997;33:1393–1399. 51. Apfel SC, Arezzo JC, Lipson L et al. Nerve growth factor prevents experimental cisplatin neuropathy. Ann Neurol 1992;31:76–80. 52. Mollman JE, Glover DJ, Hogan WM et al. Cisplatin neuropathy risk factors, prognosis, and protection by WR-2721. Cancer 1988;61:2192–2195. 53. van der Hoop RG, Vecht CJ, van der Berg MEL et al. Prevention of cisplatin neurotoxicity with an acth(4–9) analogue in patients with ovarian cancer. New Engl J Med 1990;322:89–94. 54. Pelgrims J, de Vos F, van den Brande J et al. Methylene blue in the treatment and prevention of ifosfamide-induced encephalopathy: report of 12 cases and a review of the literature. Br J Cancer 2000;82(2):291–294. 55. Quinn SF, Frau DM, Saff GN et al. A. Neurologic complications of pelvic intra-arterial chemoembolization performed with collagen material and cisplatin. Radiology 1988;167:55–57.

24

Neurologic Complications of Genitourinary Cancer David E Traul,

MD, PHD,

and David Schiff,

MD

CONTENTS Introduction Prostate Cancer Testicular Cancer Bladder Cancer Renal Carcinoma Conclusion References

Summary It is not uncommon to attribute dysfunction of both the central and the peripheral nervous systems to genitourinary cancers or their treatment. The association of these malignancies and their management strategies with neurologic morbidity is often specific to the primary cancer. Unfortunately, neurologic complications are usually associated with advanced stages of genitourinary cancers. As novel approaches in treatment regimens prolong the survival rates in patients with these cancers, a higher incidence of neurologic complications can be expected. The management options of these complications should not only take into account the primary cancer, but also the age, health, and expectations of the individual patient. Key Words: genitourinary cancer, prostate, renal cancer, testicular cancer, kidney

1. INTRODUCTION Nearly 25% of the estimated new cases of cancer in 2007 involved primary malignancies of the genitourinary system (i.e., prostate, kidneys, bladder and testicles) (1). Malignancies specific to the female genitourinary system are discussed in Chapter 23. Neurologic complications arising from genitourinary malignancies may be associated with complications of the central nervous system (metastases, paraneoplastic syndromes), the peripheral nervous system (mechanical forces, paraneoplastic syndromes), or both. Additionally, treatments for genitourinary malignancies may produce neurologic morbidity. The propensity of genitourinary malignancies and their management strategies to produce neurologic complications is often specific to the primary malignancy. For example, prostate cancer is most commonly associated with bony metastases of the spine while testicular cancer and renal cell carcinoma are associated with metastases to brain parenchyma. This chapter will provide an overview of the neurologic complications associated with each of these genitourinary malignancies, emphasizing the syndromes most commonly encountered. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

459

460

Part VII / Neurologic Complications of Specific Malignancies

2. PROSTATE CANCER Prostate cancer is the most diagnosed cancer in men, and accounted for one-third of all estimated new cases of cancer in 2006. Approximately 235,000 new cases of prostate cancer in the United States were diagnosed in 2006, causing an estimated 27,000 deaths (1). Risks for developing prostate cancer include both genetic susceptibility and environmental factors (2). The incidence rate of prostate cancer has risen over the past decade, most likely due to improved screening methods such as prostate-specific antigen (PSA) testing. In patients deemed at risk for prostate cancer through screening methods, clinical diagnosis centers on transrectal ultrasound and biopsy. The best prognostic indicator for prostate cancer is the stage of disease based on results of PSA measurement, lymph node dissection, and skeletal assessment by radionucleotide bone scan. Prostate cancer diagnosed at local or regional stages was expected to account for over 90% of the new cases of prostate cancer in 2007. Although prostate cancer may spread by direct extension and through the lymphatic system, hematogenous spread is responsible for most neurologic morbidity. Autopsy studies evaluating hematogenous spread by prostate cancer indicate high metastasis rate to bone (90%), lung (46%), and liver (25%) (3). Treatment of prostate cancer depends on the stage of disease in the context of patient age and health. Management options for localized prostate cancer include conservative observation, external beam radiotherapy, and brachytherapy. In addition, localized prostate cancer may be treated by radical prostatectomy; however, associated morbidity such as impotence and urinary incontinence may be more common than with other options (4). Locally advanced prostate cancer (tumor extending through the prostate capsule, diagnosed with radical prostatectomy) has proven responsive to adjuvant radiotherapy (5). Recent trials of neoadjuvant hormonal therapy in patients with locally advanced prostate cancer have also appeared promising (6,7). Metastatic or lymph node–positive prostate cancer is treated with androgen deprivation by pharmacologic methods or by surgical methods such as bilateral orchiectomy. Pharmacologic agents include analogues of luteinizing hormone releasing hormone (LHRH) such as leuprolide, goserelin, and buserelin. Analogues of LHRH may initially increase testosterone levels, and should be given concomitantly with antiandrogens such as flutamide in patients with suspected metastatic epidural spinal cord compression. Hormone-refractory disease may be managed with docetaxol-based chemotherapy that, in recent trials, appears to improve survival rates (8,9). Additionally, palliative radiotherapy provides symptomatic relief of hormone-resistant prostate cancer. Due to the tendency of prostate cancer to metastasize to bone, most associated neurologic morbidity develops from spinal or skull metastases. Paraneoplastic syndromes associated with prostate cancer have also been described. Parenchymal metastases from prostate cancer are relatively rare, but may produce severe neurologic morbidity.

2.1. Spinal Disease Osseous spinal metastases from the hematogenous spread of prostate cancer are a common cause of neurologic morbidity. Bubendorf performed autopsies on 1589 patients with known prostate cancer and found that 556 (35%) had hematogenous metastases (3). Evaluation of the metastatic pattern of these patients indicated that 501 of the 556 patients had osseous metastases. Anatomically, the most affected spinal level was the lumbar region (90%), followed by the thoracic (66%), and then the cervical region (38%). Epidural spinal cord compression (ESCC) is a common cause of neurologic morbidity due to spinal metastases in patients with prostate cancer. In men, prostate cancer is the second most common cause of metastatic ESCC next to lung cancer (10). Between 2% and 10% of patients with prostate cancer will develop ESCC (11–13), and spinal cord compression may be the presenting symptom in 12–20% of these patients (11,12,14). Risk factors for metastatic ESCC in patients with prostate cancer include advanced disease and poor differentiation of primary tumor (12). The most common site of metastases causing ESCC is the thoracic spine, followed by the lumbar spine (10,12,13,15,16). With the improvement of screening methods such as PSA testing, the increase in diagnoses of local or regional prostate cancer combined with earlier and more effective treatment is likely to lower the incidence of metastatic ESCC. It is important, however, to have a high index of suspicion because metastatic ESCC is treatable when diagnosed early and associated with a poor outcome once neurologic dysfunction is present. A prospective study by Bayley that evaluated 68 patients with metastatic prostate cancer with spinal MRI found that 32% of these patients had clinically occult epidural metastases (17). Most of these patients had undergone

Chapter 24 / Neurologic Complications of Genitourinary Cancer

461

prior hormonal therapy. Multivariate analysis found that patients with > 20 osseous metastases had a risk of occult compression that increased from 32% to 44% as the duration of continuous hormonal therapy prior to study increased from 0 to 24 months. Neither back pain nor the use of narcotic analgesics was a reliable predictor of ESCC. Additionally, if the initial MRI was negative for ESCC, the actuarial risk of developing clinical spinal cord compression was 3% at 1 year and 14% at 2 years. Bayley’s study suggests that spinal MRI screening for occult metastatic ESCC in high-risk patients is appropriate. The pathophysiology, symptoms, signs and work-up of suspected ESCC in prostate cancer do not differ from ESCC in other malignancies (discussed in Chapter 11). Treatment of ESCC from prostate cancer is directed at palliation of pain and prevention of further neurologic morbidity. Most patients receive corticosteroids in the acute setting for rapid pain control, followed by more individualized definitive treatment. In patients with ESCC who are naïve to hormonal therapy, androgen deprivation given at the time of diagnosis may be effective at relieving pain and improving neurological deficits. In a study of the treatment outcomes in 69 prostate cancer patients with ESCC undergoing radiation therapy with or without surgical decompression, Huddart found that a subset of 17 hormone-naive patients receiving adjuvant androgen deprivation had a median survival time of 42 months compared to 7.6 months of previously hormonerefractory patients (18). Additionally, on multivariate analysis, no prior hormonal treatment was predictive of better functional outcome. Similar improvements in survival time and neurologic function after adjuvant androgen deprivation in hormone-naïve patients with ESCC were seen by Tazi and Flynn (10,15). Stand-alone androgen deprivation may also improve pain control and functional outcome in hormone-naïve patients unfit for other treatment methods (19). Radiation therapy continues to be the treatment of choice for patients with metastatic ESCC from prostate cancer. Prognostic factors of radiation therapy correlate with pre-treatment ambulatory status and post-treatment documentation of response by MRI or myelogram (20). Most patients receive 30–40 Gy given in 10–20 fractions over 2–4 weeks. Reporting on 26 prostate cancer patients treated with steroids, androgen deprivation therapy, and radiation therapy for metastatic ESCC, Smith found that 12/12 patients ambulatory at start of treatment responded to radiation therapy (14). In patients paraparetic at diagnosis, 10/12 (83%) became ambulatory after treatment, with 8 of these patients remaining ambulatory till death or last follow-up. Only 1 of 5 paraplegic patients responded to therapy. The median survival after treatment was 19 months in patients without paraplegia compared to 4 months in patients with paraplegia. Flynn reported that ambulation following radiation therapy was recorded in all patients who were ambulatory at the start of treatment, in 52% of paraparetic patients, and in only 38% of paraplegic patients (15). More recently, a retrospective analysis demonstrated radiation therapy alone was effective in all patients with focal neurologic deficits, in two-thirds of patients with paraparesis, and in one-fourth of patients with paraplegia at the start of treatment (10). Despite these results, radiation therapy should also be considered in all patients with metastatic ESCC from prostate cancer regardless of neurologic morbidity. A prospective study of 49 patients with metastatic ESCC due to hormone-resistant prostate cancer found that radiation therapy may offer improvement in activities of daily living (21). Surgery combined with radiotherapy appears to be superior to radiotherapy alone in patients who present with paraparesis or paraplegia from metastatic ESCC due to prostate cancer. Tazi found that 8 out of 9 patients treated with laminectomy combined with radiotherapy were ambulatory after treatment compared to 7 out of 12 patients treated with radiotherapy alone (10). Flynn reported that 75% of paraplegic patients treated with combined surgery and radiotherapy were ambulatory post-treatment compared with 33% of paraplegic patients treated with radiotherapy alone (15). Surgery alone for metastatic ESCC from prostate cancer is considered less often than combined modalities. Surgery as the sole treatment for metastatic ESCC was efficacious in one study reporting that 79% of patients treated with laminectomy and decompression demonstrated improvement in presenting complaints and that 63% of immobile patients became ambulatory after surgery (16). Surgery is also advisable in patients with an expected lifespan of at least 6 months who deteriorate during radiation, who have had previous radiation to the involved site, or who have a potentially correctable unstable spine (15). The role of bisphosphonates for treatment of bone pain and prevention of metastatic ESCC due prostate cancer is unclear. Randomized trials of two bisphosphonates—pamidronate and clodronate—failed to show a significant improvement in pain control for patients with hormone-resistant prostate cancer (22,23). A third-generation

462

Part VII / Neurologic Complications of Specific Malignancies

bisphosphonate, zolendronic acid, did significantly reduce skeletal-related events such as pathologic fractures and spinal compression in a randomized trial (24). However, zolendroic acid did not improve quality of life for patients with hormone-resistant prostate cancer. Overall, the prognosis for patients with metastatic ESCC from prostate cancer is poor, with reported median survival times ranging from 3.5 to 18 months (10,14–16,18,21). The poor prognosis derives from the fact the most patients with ESCC have hormone-resistant metastatic prostate cancer. Patients naïve to hormone therapy, ambulatory at time of diagnosis, with a single site of spinal compression, and under 65 years of age appear to have a significantly better prognosis (10,15,18).

2.2. Skull Metastases Metastases to the skull are relatively common in patients with prostate cancer. Long evaluated 508 patients with CT scan to identify unusual patterns of metastatic disease and found skull/orbit metastases in 11 patients (2%) (25). All 11 patients had evidence of metastatic disease elsewhere in the body. In an autopsy study, Bubendorf reported the incidence of prostate metastases to the skull to be 8% (3). Together, these studies indicate that skull metastases are associated with more advanced prostate cancer. Prostate metastases reach the skull through arterial hematogenous spread or via Batson’s venous plexus. Calvarial metastases are common and usually asymptomatic, although they may occasionally cause pain or present as a palpable mass. Less commonly, calvarial metastases compress the underlying brain and produce neurological symptoms. Raizer reported a patient with headaches and papilledema who was found to have calvarial metastases from prostate cancer compressing the sagittal or lateral sinus (26). Occasionally, the sinuses become occluded and venous infarction of the brain results in headache, seizures, and focal neurological deficits. Sakurai described a patient with metastatic prostate cancer and venous sinus thrombosis who subsequently developed dural arteriovenous malformations (27). Asymptomatic calvarial metastases may not require any treatment. Symptomatic lesions usually respond to radiotherapy; however, surgery may occasionally be required to remove a large lesion, especially if the lesion is compressing the underlying brain. Prostate metastases to the base of the skull commonly produce cranial nerve dysfunction by neural compression or infiltration (Fig. 1). McDermott described 15 patients with prostate metastases to the skull of whom 12 had cranial mononeuropathy (28). Similar findings by Seymore suggest that cranial mononeuropathy, in particular cranial nerve VII, is a common sign of skull base metastases (29). Additionally, occipital condyle syndrome

Fig. 1. A 73-year-old man with a remote history of prostate cancer and a recently rising PSA presented with right facial numbness and horizontal diplopia and eventually headache over a one-month course. The examination revealed right CN Vth and VIth nerval palsies. Contrast-enhanced CT (A) showed a large enhancing, erosive mass involving the basisphenoid, basiocciput, sphenoid sinus, bilateral clinoids, both cavernous sinuses, and the medial right temporal lobe. Post-contrast T1-weighted MRI (B) confirmed this finding and revealed perineural spread along trigmeninal branches. Biopsy confirmed metastatic prostate cancer, and the patient was treated with fractionated radiotherapy with good clinical response.

Chapter 24 / Neurologic Complications of Genitourinary Cancer

463

consisting of unilateral occipital pain with ipsilateral XII cranial nerve dysfunction, and optic neuropathy producing blindness have been described (30,31). Several syndromes of multiple cranial neuropathies associated with prostatic skull base metastases have been reported. Pershant described a patient with Collet–Siccard syndrome, palsy of the lower four cranial nerves, in a patient with calvarial prostate metastases (32). Additionally, Sicenica described a patient with Villaret syndrome, palsy of the lower four cranial nerve with ipsilateral Horner’s syndrome (33). Due to their long path through the mandible, the inferior alveolar and mental branches of the mandibular nerve are susceptible to osseous metastases to the jaw. Compression of these branches produce numbness over the ipsilateral chin referred to as “numb chin syndrome,” which has been described in a patient with metastatic prostate cancer (34). Radiation therapy is the mainstay of treatment for symptomatic skull base metastases from prostate cancer. Complete or partial response rates greater than 90% have been reported with radiation therapy in several small series (28,29). The presence of skull base metastases heralds advanced prostate cancer; therefore, survival times after treatment are typically less than 1 year (28,29).

2.3. Dural Metastases Dural metastases are a common cause of neurological complications associated with prostate cancer. In an analysis of 131 patients with craniospinal metastases from prostate cancer, 28 (21%) patients had dural metastases (35). Additionally, in a review of 33 surgical and 27 autopsy cases for dural metastases, KleinschmidtDeMasters found that 8 patients (7 autopsy, 1 surgical) had prostate cancer as the primary malignancy (36). Mode of prostate cancer metastasis to the dura appeared to be via direct extension from the adjacent skull or by hematogenous spread. The higher incidence in the autopsy cases suggests that dural metastases are a complication of advanced prostate cancer. Dural metastases may act as a mass lesion, compressing the underlying brain, producing seizures, headaches, and focal neurological deficits. Radiographically, a solitary dural-based metastasis may be mistaken for a more benign lesion. There are several case reports of a dural metastasis from prostate cancer mimicking a meningioma (37,38). Additionally, dural-based metastasis from prostate cancer may be diagnosed preoperatively as subdural hematoma (39,40). Rarely, a dural metastasis may cause symptoms by exuding fluid into the subdural space, thereby producing a true subdural hematoma or effusion (41). The diagnosis of multiple dural metastases associated with metastatic prostate cancer is usually straightforward. Treatment usually consists of corticosteroids and external beam radiotherapy. Decompression surgery may be warranted in selected cases, but is generally not feasible in patients with multiple dural-based metastases. For single isolated dural metastases or tumor that has recurred after external beam radiotherapy, stereotactic radiosurgery may have a role.

2.4. Brain Metastases Parenchymal brain metastases are an extremely rare cause of neurological complications in patients with prostate cancer. Among 16,280 cases of prostate cancer at MD Anderson Cancer Center, 103 patients (0.63%) had parenchymal metastases identified by either neuroimaging or autopsy (35). Of these patients, 86% had single metastases, and the most frequent site of the lesion was supratentorial (76%). Most patients complained of nonfocal symptoms such as headache or cognitive changes that were attributable to intracranial hypertension or diffuse cortical dysfunction. Although adenocarcinoma is the most frequent histologic subtype of prostate cancer, the incidence of brain parenchymal metastases from prostatic small cell cancer (SCC) is overrepresented in many studies (35,42). Reasons for the proclivity of prostatic SCC to metastasize to the brain are unknown. Occasionally, prostate metastases to brain parenchyma may be the first manifestation of primary cancer (43). More often, patients have undergone previous prostate cancer treatment and have known metastatic disease elsewhere, especially the lung (42). Treatment is similar to other brain parenchymal metastases and involves acute corticosteroids, surgery when feasible for symptomatic single lesions, whole-brain external radiotherapy (WBRT), and stereotactic radiosurgery. Overall mean survival time after diagnosis of brain metastases ranges between 4 and 9.2 months (35,42).

464

Part VII / Neurologic Complications of Specific Malignancies

2.5. Leptomeningeal Disease Leptomeningeal carcinomatosis is exceedingly uncommon in prostate cancer. Tremont-Lukats reported leptomeningeal disease in 5 out of 131 (4%) prostate cancer patients with known craniospinal metastases (35). Several case reports identify the usefulness of MRI and elevated CSF prostate-specific antigen (PSA) in the evaluation of leptomeningeal carcinomatosis from prostate cancer (44–46). The response to treatment is generally poor, although one reported patient responded to androgen blockade (47).

2.6. Peripheral Nerve Compression Prostate cancer tends to metastasize to the spine and produce neurologic morbidity in the lower extremities by compressing the cauda equina or the lumbosacral nerve roots. Rarely, prostate cancer may spread along prostatic nerves and infiltrate the lumbosacral plexus, producing radicular pain, weakness and paresthesias in the legs (48). Additionally, a case report describes an osteolytic solitary radial head metastasis presenting with wrist drop in a patient with prostate cancer (49). The wrist drop responded to treatment of the primary cancer with hormonal therapy. Patients with advanced cancer, including prostate cancer, are at risk for developing peroneal neuropathies, characterized by foot drop and numbness over the anterolateral aspect of the shin and the dorsum of foot (50). This condition results from compression of the common peroneal nerve at the level of the fibular head. Predisposing factors include weight loss, prolonged bed rest, leg crossing, and chemotherapy. In general, the prognosis is good and the neuropathy improves in the majority of patients.

2.7. Stroke Stroke is an infrequent cause of neurological morbidity in patients with prostate cancer. The most common etiology of stroke in patients with prostate cancer is controversial. In an autopsy study of patients with cancer, Graus concluded that the majority of asymptomatic ischemic events were due to traditional risk factors such as atherosclerosis rather than being attributable to the neoplasm (51). However, the majority of symptomatic ischemic events were due to pathophysiologic abnormalities related to the neoplasm, including direct effects of the tumor, coagulation disorders, and infections. Recently, a retrospective review of cancer patients diagnosed at Memorial Sloan-Kettering Cancer Center suggested that embolic events related to hypercoagulability were more common than atherosclerosis as the etiology of ischemia in cancer patients (52). The mechanism by which tumor-related hypercoagulable states cause ischemic events in cancer patients is unknown. Hypercoagulability in cancer patients may be associated with nonbacterial thrombotic endocarditis (NBTE), characterized by the formation of platelet-fibrin vegetations on heart valves. These vegetations can embolize to cerebral vessels of any size, typically producing focal deficits or less often encephalopathy without focal features. Transesophageal echocardiography (TEE) demonstrating valvular vegetations in the setting of negative blood cultures is an effective method for diagnosing NBTE. In many patients, these vegetations are often small and undetectable. Some patients experience concomitant venous thromboembolic disease. Treatment of NBTE is often directed at the underlying tumor, although heparin may be of some benefit. Surgical options may also be considered (53). Chronic disseminated intravascular coagulation (DIC) may co-exist with NBTE or can occur as a separate entity. DIC most commonly produces an encephalopathy without focal findings. The diagnosis of chronic DIC may be difficult, as hematologic parameters such as prothrombin time, partial thromboplastin time, and platelet count may be normal. Elevated fibrin degradation products and d-dimer may suggest DIC; however, these markers of intravascular clotting may be abnormal in the presence of an underlying malignancy or from the stroke itself. Treatment is directed at the underlying condition, and may include supportive care with blood products, antithrombin, heparin, and other agents. Hemorrhagic strokes are rarely associated with prostate cancer. Inamasu reported a patient with prostate cancer diagnosed with cerebellar hemorrhage secondary to cranial metastases causing obstructive hydrocephalus (54). Rarely, dural-based metastases from prostate cancer will cause venous sinus thrombosis and hemorrhagic infarction (41). Tumor hemorrhage into a prostatic parenchymal brain metastasis is exceptional.

Chapter 24 / Neurologic Complications of Genitourinary Cancer

465

2.8. Paraneoplastic Syndromes Neurologic complications due to paraneoplastic syndromes are not uncommonly associated with prostate cancer. One of the most commonly encountered paraneoplastic syndromes associated with prostate cancer is defined by the presence of the anti-Hu antibody, also known as anti-neuronal nuclear antibody type I (ANNA-I). Although anti-Hu antibody is more frequently found in small cell lung carcinoma, patients with prostate cancer may also harbor anti-Hu (51,55,56). This antibody binds to neuronal nuclei in both the central and peripheral nervous system. The most common manifestation of anti-Hu paraneoplastic syndrome is sensory neuropathy; however, dysautonomia, cerebellar syndromes, limbic or brainstem encephalitis may also occur. Treatment of the underlying cancer may initially improve symptoms of the paraneoplastic syndrome (56), but overall the outcome is poor as patients typically have advanced disease. Baloh reported two patients with prostatic cancer who developed a suspected paraneoplastic syndrome consisting of loss of voluntary horizontal eye movements, severe persistent muscle spasms of the face, jaw, and pharynx, and mild gait unsteadiness (57). No antineuronal antibodies were detected, and both patients had evidence of chronic inflammation in the pons and medulla. Although typically affiliated with lung and breast cancer, dermatomyositis and polymyositis are also associated with a slightly increased risk of prostate cancer (58). There are several case reports of both dermatomyositis and polymyositis associated with prostate cancer (59–61). Prostate cancer is a rare cause of paraneoplastic cerebellar degeneration. Matschke reported a patient with adenocarcinoma of the prostate exhibiting signs of subacute cerebellar degeneration (62). Interestingly, work-up found anti-Yo antibodies in both the serum and CSF of the patient. Expression of anti-Yo antigen was also detected in the tumor tissue. Anti-Yo antibodies are directed toward antigens expressed in the cytoplasm of Purkinje cells, and are usually associated with ovarian or breast cancers. Finally, Lambert-Eaton myasthenic syndrome (LEMS) has been associated with prostate cancer (63,64).

2.9. Complications of Treatment Impotence is a frequent consequence of radical retropubic prostatectomy in the treatment of prostate cancer. Surgical injury to the neurovascular bundle crossing along the posterolateral aspect of the prostate disrupts the autonomic supply to the corpus cavernosa. The advent of nerve-sparing laparoscopic radical prostatectomy may offer an advantage over the open surgical technique; however, the apparent benefit of this newer technique remains controversial (65). Recent techniques involving nerve grafting to improve sexual function after radical prostatectomy appear promising for selected patients (66,67). Impotence may also be seen with radiation therapy and brachytherapy. In a study of 441 patients with nonmetastatic prostate cancer, Beard reported that 62% of patients experienced impotence after treatment with small field external beam radiotherapy (68). Rarely, radiation therapy for spinal cord compression may result in radiation myelopathy. Brachytherapy, while associated with a lower risk of impotency than radical prostatectomy, has been shown to produce sexual dysfunction in a significant number of patients (65). Obturator nerve injury may occur during radical prostatectomy with lymphadenectomy (69). Reapproximation of the nerve usually results in minimal neurological morbidity. Less commonly, the lateral femoral cutaneous nerve or genitofemoral nerve is injured in extensive resections, producing paresthesias in the superolateral thigh or groin and scrotum.

3. TESTICULAR CANCER Testicular cancers represent nearly 1% of the estimated new cancer cases in the United States with over 8000 cases being diagnosed each uear. Testicular cancer is the leading cause of cancer among males in the age range 15–34 years of age. Germ cell tumors account for > 95% of all testicular cancers; non–germ cell tumors such as lymphoma and Leydig cell tumors are exceptional and occur more frequently in the elderly population. Germ cell tumors occasionally arise in extragonadal locations, including the sacrum, retroperineum, and mediastinum. Primary intracranial germ cell tumors are rare and typically located in the hypothalamus or pineal gland. Germ cell tumors are subtyped into pure seminomas and nonseminomas. Seminomas are the most common histologic subtype accounting for nearly 50% of all germ cell tumors. Nonseminomas include pathological cell lines such

466

Part VII / Neurologic Complications of Specific Malignancies

as embryonic cell carcinoma, choriocarcinoma, teratoma, yolk sac tumors, or a mixture of these cell lines with seminoma. Germ cell tumors metastasize primarily through the lymphatic system; however, hematogenous spread may occur early in the disease course with choriocarcinoma. Retroperitoneal lymph nodes are the most common route of metastatic spread with inguinal nodes less frequently involved. Tumor staging and differentiation between seminomatous and nonseminomatous subtypes are important considerations in the management and prognosis of germ cell tumors. Tumor staging typically involves ultrasonogram of the testicles and CT scan of the chest, abdomen and pelvis. Serum tumor markers may also be useful in the diagnosis and management of germ cell tumors. For instance, serologic elevation of alpha fetoprotein is not seen with pure seminomas. While human chorionic gonadotropin (HCG) is secreted by both seminomas and nonseminomas, high serum HCG levels may suggest aggressive choriocarcinoma. Germ cell neoplasms may also produce lactate dehydrogenase, a relatively nonspecific finding. Seminomatous tumors are highly radiosensitive, and management of low-stage seminomas involves orchiectomy combined with low-dose adjuvant retroperitoneal or inguinal radiotherapy. In metastatic seminomas, orchiectomy is combined with bleomycin, etoposide, and cisplatin-based (BEP) chemotherapy. The prognosis for seminomatous testicular cancer is excellent with a 99% cure rate for low-stage seminomas (70), and an overall 5-year survival rate > 90%. Nonseminomatous tumors are less radiosensitive, and initial orchiectomy is combined with: (i) surveillance and treatment at relapse for low risk patients; (ii) retroperitoneal lymph node resection with or without adjuvant chemotherapy; or (iii) adjuvant chemotherapy. The overall survival rate for early-stage nonseminomas is greater than 95%. With metastatic nonseminomas, overall 10-year survival rates approach 85% (70). Brain metastases, while relatively uncommon, account for substantial neurologic morbidity associated with germ cell tumors. Additionally, paraneoplastic syndromes and complications of surgery may also result in neurologic complications. With the high cure rates associated with some testicular cancers, long-term effects of radiation therapy and chemotherapy may also contribute to important but less serious neurologic deficits.

3.1. Brain Metastases Brain metastases are an infrequent neurologic complication of testicular cancers. With the advent of novel chemotherapy regimens, the incidence of brain metastases from germ cell tumors has sharply declined to 1–3% (71–75). Concomitant lung metastases are seen in nearly 80% of patients with brain metastases from germ cell tumors (71,74–78), indicating that brain metastases from germ cell tumors arise in the setting of widespread disease. Nonseminomatous tumors (especially choriocarcinoma), can spread hematogenously, and have a higher association with brain metastases then seminomatous tumors (74,76). The brain may also serve as a site for relapse of germ cell tumors in patients who previously responded to chemotherapy. Commonly, extracerebral relapse occurs concomitantly to cerebral relapse; however, the brain may serve as a sanctuary site for germ cell tumors (76–79). These relapses may be asymptomatic, but are detectable with brain MRI and CT and may produce elevations of serum tumor markers (77). Typically, treatment for brain metastases from germ cell tumors involves cisplatin-based chemotherapy with or without surgical intervention and radiotherapy. Patients presenting with brain metastases at time of diagnosis of metastatic germ cell tumors have a more favorable prognosis than patients in whom cerebral metastases are diagnosed after initiation of treatment with chemotherapy. Additionally, non–treatment related factors predicting a more favorable outcome include single brain metastases and the absence of bone or liver metastases (76). For nonseminomatous tumors, histologic subtype does not appear to affect survival after treatment (71); however, one study found that cerebral metastases containing choriocarcinomatous elements were associated with a less favorable outcome (80). When cerebral metastases are detected at diagnosis, 5-year survival rates range from 45% to 53% with cisplatinbased chemotherapy (77,80). Surgical intervention has also been shown to produce a more favorable outcome in these patients; however, limitations of most studies include selection bias of patients with single or oligometastatic brain lesions (76,77,80). WBRT has been shown in one study not to affect survival outcome in patients with brain metastases detected at time of cancer diagnosis (77). Brain metastases that develop after initial chemotherapy have a less favorable prognosis, and long-term survival rates in these patients range between 12% and 20% (77,80). The 5-year survival rate for patients with tumor recurrence isolated to the brain approaches 40% (77). Due to the possibility of increased

Chapter 24 / Neurologic Complications of Genitourinary Cancer

467

chemo-resistance of brain metastases developing after induction chemotherapy, WBRT and neurosurgery should be considered in this subset of patients (73,77,80).

3.2. Paraneoplastic Syndromes Testicular germ cell neoplasms occasionally give rise to paraneoplastic syndromes. Specifically, testicular cancer is over-represented in patients diagnosed with paraneoplastic limbic or brainstem encephalitis (81). These patients usually present with psychiatric symptoms, hypothalamic dysfunction, memory loss, seizures, or brainstem dysfunction prior to diagnosis of cancer. Paraneoplastic limbic or brainstem encephalitis may occur with both seminomatous and nonseminomatous tumors (82,83). Brain MRI displays T2 hyperintensities in the mesial temporal lobes and the brainstem that may enhance with gadolinium. In testicular cancer, paraneoplastic limbic encephalitis may be associated with antineuronal antibody, anti-Ta, detectable in the serum and CSF (81,82). Anti-Ta recognizes the Ma2 protein normally expressed in neurons and sometimes in testicular tumor tissue (81,82). Not all patients with testicular cancer displaying symptoms of paraneoplastic limbic encephalitis will have detectable levels of anti-Ta antibodies (82,84). Clinically, a combination of symptoms involving the disruption of limbic, diencephalic, and upper brain stem structures may distinguish anti-Ma2 antibody-associated encephalitis from classical paraneoplastic limbic or brainstem encephalitis (85). Anti-Ma2 antibodies in patients with testicular cancer may also be associated with cerebellar degeneration (82) and motor neuron disease (86). A similar protein, Ma1, is expressed in normal brain and testicular tissue. Although usually expressed in patients with non-germ cell tumors, anti-Ma1 antibodies may also be present in patients with testicular cancer (85). Paraneoplastic syndromes associated with anti-Ma1 antibodies have a greater incidence of cerebellar dysfunction and are associated with a poor prognosis (87). The functions of Ma1 and Ma2 protein are unknown. Additionally, it is uncertain if the antibodies are directly involved with the pathogenesis of the paraneoplastic neurologic syndrome. Brain biopsy in one testicular cancer patient with anti-Ma2 antibodies revealed CD8+ lymphocytes clustered around neurons, suggesting cytotoxic T-cells explain syndrome pathogenesis (82). Treatment with immunomodulatory agents such as intravenous immunoglobin and corticosteroids may improve neurological symptoms (85,86).

3.3. Leptomeningeal Disease Leptomeningeal disease associated with testicular cancer is rare but has been reported with both seminomatous (88) and nonseminomatous (89) tumors. Due to lack of reported cases, data on treatment outcomes and prognosis associated with leptomeningeal involvement in testicular cancer are scarce. A single patient with a nonseminomatous tumor treated with ablative chemotherapy combined with autologous peripheral stem cell re-infusion was reported to have a 1-year remission (89). Dural metastasis of seminoma and choriocarcinoma producing subdural hematoma has also been described (90).

3.4. Spinal Disease Spinal cord compression due to metastases is rarely associated with testicular cancer. In a series of 297 patients with metastatic germ cell cancer, Hitchins reported that 11 patients had bone involvement and only two of these developed epidural spinal cord compression (91). In another retrospective study, symptomatic spinal cord compression was seen in 2.6% of patients diagnosed with metastatic germ cell tumors (92). In both studies, lumbar spine was the most frequent site involved. Pain is the most common presenting symptom and may be present for months prior to diagnosis. Sensorimotor deficits, when present, tend to develop later in the disease course, but may also be the presenting sign and progress rapidly (93–95). Surgical intervention is typically considered for rapidly progressive neurological deficits. The chemosensitivity of germ cell tumors makes cisplatin-based chemotherapy with or without corticosteroids another effective treatment option (92,96). Patients with large areas of osseous destruction undergoing chemotherapy should be closely monitored for the possibility of vertebral collapse related to rapid tumor lysis (91).

468

Part VII / Neurologic Complications of Specific Malignancies

3.5. Peripheral Nerve Compression Involvement of the peripheral nerve system is a rare direct complication of testicular cancer. In a series of 85 cancer patients diagnosed with lumbosacral plexopathy, three patients developed this neurologic complication due to testicular cancer metastases (97). Recurrent laryngeal nerve paralysis presenting with hoarseness has been associated with mediastinal seminoma, and successfully treated with chemotherapy (98). Brachial plexus neuropathy complicating testicular cancer, resolving after resection and radiation therapy, has also been reported (99).

3.6. Complications of Treatment Adjuvant radiation therapy to para-aortic lymph nodes and distal spine in conjunction with orchiectomy has been infrequently associated with a late-developing progressive lower motor neuron syndrome. Bowen reported a series of 6 patients presenting with predominantly motor weakness up to 25 years following radiation therapy (> 40 Gy) for testicular cancer (100). Neuropathology in one of these patients displayed a radiation-induced vasculopathy of proximal spinal roots with preservation of motor neuron cell bodies suggesting a radiculopathy rather than a motor neuronopathy (100). Additionally, prolonged but reversible motor deficits have been reported after lower-dose (36–40 Gy) radiation therapy for testicular seminoma (101) Current radiation therapy strategies for testicular cancer employ target doses < 30 Gy, which may circumvent the lower motor neuron syndrome; however, transient sensory deficits following doses of 26–30 Gy have been reported (101), Another uncommon complication of radiation therapy for testicular cancer is the late development of malignant peripheral nerve sheath tumors (MPNST). At least four patients undergoing adjuvant para-aortic radiotherapy for testicular cancer have been diagnosed with MPNST presenting as radicular pain up to ten years following initial treatment (102–104). Chemotherapy for testicular cancers is cisplatin-based and may also include vinblastine: two drugs known to produce neurotoxic effects. The most widely reported neurotoxic effect associated with chemotherapy regimens is peripheral neuropathy. Paresthesias immediately during or following treatment are reported by 45–83% of patients receiving cisplatin-based chemotherapy (105–110). These sensory symptoms may persist for decades in some patients (105,107–109,111). A correlation between higher cumulative doses of cisplatin (> 400 mg/m2 ) and the persistence of sensorypredominant neuropathy has been suggested (105), but is not confirmed in all studies (106). Ototoxicity is another commonly reported side effect of cisplatin-based chemotherapy for testicular cancers. Ototoxic symptoms, usually tinnitus, are reported by 30–50% of patients (106,109). Objective testing with audiography shows changes in the high-frequency region of the audiogram in 60% of patients that may be asymptomatic (109). Ototoxic symptoms can be persistent in >25% of patients and appear correlated to serum concentrations of cisplatin during chemotherapy (109). Primary or post-chemotherapy retroperitoneal lymph node dissection (RPLND) is often utilized for accurate staging and treatment of nonseminomatous germ cell tumors. Although historically associated with sympathetic nerve disruption and consequent ejaculatory dysfunction, modern advancements including nerve sparing techniques have decreased the incidence of this complication (112). Femoral neuropathy presenting with sensorimotor deficits and hypoactive reflexes after RPLND for testicular cancer has also been reported (113). Spinal cord ischemia following RPLND is a rare but serious neurological complication associated with testicular cancer. Leibovitch reported this surgical complication in 4 of 712 post-chemotherapy patients (two of whom had also undergone radiation therapy) but in none of 735 prechemotherapy surgeries (114). Asymmetric paraparesis, sensory deficits, and neurogenic bladder were the most reported symptoms in patients with spinal ischemia.

4. BLADDER CANCER Bladder cancer represents nearly 4% of all estimated new cancer cases in the United States. Bladder cancer is the third most diagnosed cancer in men and the ninth most diagnosed cancer in women (reflecting that nearly 75% of bladder cancer is diagnosed in men). An estimated 61,000 new cases and 13,000 deaths occurred in 2006 from bladder cancer (1). Transitional cell carcinoma is the most common histological subtype in the

Chapter 24 / Neurologic Complications of Genitourinary Cancer

469

United States, followed by squamous cell carcinoma, adenocarcinoma, rhabdomyosarcoma, and undifferentiated carcinoma. Metastatic disease associated with bladder cancer typically involves liver, lung, lymph nodes, and bone. Mechanism of spread is through both lymphatic and hematogenous dissemination. Treatment options and the management of bladder cancer depend on tumor staging that categorizes tumors as superficial (80%) or invasive (20%) based on involvement of the muscularis propria. Superficial tumors may be managed with intravesical instillation of chemotherapeutic agents such as thiotepa, doxorubicin, and mitomycin which may decrease tumor recurrence and slow disease progression. Superficial tumors also appear sensitive to intravesical immunologic agents such as bacillus Calmette–Guerin (BCG). Invasive tumors are often managed with neoadjuvant or adjuvant chemotherapy with radical cystectomy; however, the role of chemotherapy is still under study. The most commonly utilized chemotherapeutic regimen involves methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC), which may improve remission-free survival in some patient populations (115–118). Radiotherapy may also be a palliative measure for highly invasive tumors. Neurologic complications from bladder cancer metastases are uncommon, but may affect brain, spinal cord, and cauda equina. Nonmetastatic neurologic complications such as paraneoplastic disease, stroke, and peripheral neuropathy are also rare.

4.1. Brain Metastases Traditionally, brain metastases from bladder cancer are rare with a reported incidence of 1–7% (119–121). The mean interval between diagnosis of bladder cancer and the identification of brain metastases ranges from 8 to 23 months, and multiple lesions are more common than solitary lesions (116,121,122). Interestingly, the advent of novel treatment options may increase the incidence of brain metastases in bladder cancer patients. Incidence rates of 16% have been reported in patients treated with MVAC, suggesting the increased frequency of this complication may be due to improved survival rates or poor penetration of MVAC across the blood–brain barrier (116,117). Brain metastases from bladder cancer are commonly seen in conjunction with extracerebral metastases, especially in the lung. Rarely, isolated brain metastases arise (117,120,121,123–125). Treatment for brain metastases from bladder cancer is not yet optimized. Typically, surgical resection, when feasible, is recommended. The benefit of adjuvant radiation therapy is uncertain. In a series of 19 patients with brain metastases from bladder cancer, Rosenstein demonstrated a longer survival time in patients undergoing surgical resection combined with WBRT compared to surgery alone (126). The benefit of combined treatment modalities was not observed in another study (116). Even with treatment, the mean survival time after the diagnosis of brain metastases is 1–6 months (116,119,121). Patients with isolated brain lesions survive only slightly longer than patients with multiple lesions. Chemotherapy may serve an adjuvant role to radiation therapy when directed at palliative efforts for patients with brain metastases from bladder cancer (127).

4.2. Leptomeningeal Disease Leptomeningeal carcinomatosis is an uncommon but reported cause of neurological deficits associated with bladder cancer. Most bladder cancer patients diagnosed with leptomeningeal carcinomatosis have previously undergone chemotherapy for systemic metastases (128–131). Again, this suggests that the incidence of leptomeningeal carcinomatosis cancer may increase with longer survival times from chemotherapy or be instigated by poor blood–brain barrier penetration of chemotherapeutic agents. Leptomeningeal carcinomatosis as the initial manifestation of bladder cancer has also been reported (132).

4.3. Paraneoplastic Syndromes Paraneoplastic syndromes associated with bladder cancer are rare. Prestigiacomo reported a paraneoplastic syndrome associated with the antineuronal antibody anti-Ri presenting as progressive opsoclonus, ataxia, and memory deficits in a patient with transitional cell carcinoma (133). Anti-Ri antibodies were detectable in the CSF and serum, and immunohistochemical analysis confirmed the target antigen was expressed in tumor tissue. Although anti-Ri antibodies are believed to target neuronal RNA splicing proteins expressed predominately in the brainstem and cerebellum, autopsy in this patient demonstrated widespread inflammatory changes to the basal ganglia, limbic structures, frontal cortex, subcortical white matter, and spinal cord.

470

Part VII / Neurologic Complications of Specific Malignancies

Another antineuronal antibody, anti-Yo, has been associated with a patient with bladder cancer who presented with diplopia, headache and vertigo (134). Anti-Yo antibodies, which recognize target antigen on Purkinje cells, were found in the patient’s CSF and serum. Histological analysis demonstrated expression of the anti-Yo antigen in tumor tissue. Paraneoplastic brainstem encephalitis was suspected in one patient with transitional cell carcinoma who presented with visual changes, glossal spasm, and dysphagia that resolved after resection of the tumor, experienced neurologic relapse with tumor recurrence, and subsequently remitted with further chemotherapy (135). Paraneoplastic dermatomyositis and dermatomyositis with inclusion body myositis have also been associated with bladder cancer (136–138). Paraneoplastic polymyositis heralding a diagnosis of transitional cell carcinoma was reported in a patient who presented with dysphagia, progressive symmetrical weakness of proximal muscles, and normal deep tendon reflexes after extensive serologic workup was performed (139). Finally, a paraneoplastic leukemoid reaction was suspected in a patient with transitional cell carcinoma after a negative workup for infection (140).

4.4. Spinal Disease Despite the predilection of bladder cancer to spread to bone, spinal cord compression is uncommon. Liskow reported a series of 685 patients with bladder cancer observed for over a decade in which only a single patient was diagnosed with epidural spinal cord compression after developing progressive paraplegia (13). More recently, in a retrospective study of 359 patients with bladder cancer, six patients (1.7%) were diagnosed with epidural spinal cord compression (119). Four of these patients were found to have developed spinal cord compression after hematogenous spread of the tumor to vertebral column with secondary extension into the epidural space. In two patients, tumor expansion through spinal bone and into the epidural space caused the compression.

4.5. Peripheral Nerve Compression Peripheral nerve disease is infrequently associated with bladder cancer. Direct extension of the tumor or metastases to regional lymph nodes may result in lumbosacral plexopathy. In Jaeckle’s series of 85 cancer patients evaluated for lumbosacral plexopathy, three patients had documented bladder cancer (97). These patients predominantly reported local or radicular pain and paresthesias and demonstrated sensorimotor deficits, hyporeflexia, and positive straight leg raising tests. Obturator mononeuropathy has been described in six patients with new or recurrent bladder cancer (141). These patients presented with leg pain, leg edema, and were found to have sensorimotor deficits on clinical exam. Treatment with radiation therapy or chemotherapy provided symptomatic relief for most patients.

4.6. Complications of Treatment Neurotoxicity from cisplatin-based chemotherapy, notably neuropathy and ototoxicity, is not uncommon in bladder cancer patients and in one study was reported by 58% of patients treated with MVAC (115). Radical cystectomy for aggressive bladder cancer can be associated with neurologic urinary and sexual dysfunction; however, nerve-sparing techniques may decrease the incidence of this complication in selected patients (142). Additionally, post-operative femoral neuropathy, thought secondary to mechanical compression during RPLND, has been reported in patients with bladder cancer (113). Patients typically complained of numbness and paresthesias on the anteromedial aspect of the thigh, and displayed weakness in hip flexion, adduction, and external rotation. Ligation of both internal iliac arteries to control bleeding during radical cystectomy has been associated with ischemia of the cauda equina manifesting as distal lower extremity weakness and numbness (143). The posterior trunks of these vessels help supply the cauda equina via the iliolumbar and lateral sacral arteries.

5. RENAL CARCINOMA Renal cell carcinoma (RCC) accounts for nearly 3% of the estimated cancers cases in the United States with over 38,000 cases diagnoses annually. Over twice as many males than females are diagnoses each year. An estimated 12,900 deaths from renal carcinoma occurred in 2007. The majority of RCC are clear cell adenocarcinomas. With modern use of ultrasonography and MRI, about 75% of patients present with localized disease. About half of these patients will develop metastatic disease. RCC metastasizes via the lymphatic system to local, regional, and

Chapter 24 / Neurologic Complications of Genitourinary Cancer

471

mediastinal lymph nodes. Hematogenous metastasis to distal sites primarily involves the lung, bone, liver, and brain. Treatment for localized RCC centers around radical nephrectomy, and 5-year survival rate may exceed 80% when the tumor is confined to the kidney or found incidentally (144). Surgical excision, when feasible, is recommended for locally invasive renal cell carcinoma, but the benefit of adjuvant radiotherapy is unproven (145). For metastatic RCC, radical nephrectomy is generally recommended. Removal of the primary tumor may result in a better response to adjuvant therapy and palliate symptoms of locally advanced disease. Occasionally, nephrectomy may also result in spontaneous remission of metastatic disease (146). Overall, the 2-year survival rate of patients with metastatic RCC is 10–20%. Various adjuvant therapies have been investigated. Cytokine-based adjuvant therapy with interferon alpha or interleukin-2 has shown modest survival benefit in certain patient populations, however patients with cerebral metastases are typically excluded from trials as the brain is considered a nonresponder site (147). Recently, agents directed at inhibiting tumor angiogenesis, such as sorafenib, sunitinib, temsirolimus, and the monoclonal antibody bevacizumab, have shown encouraging results in slowing the progression of RCC (148–151). Hormonal therapy for metastatic RCC, including tamoxifen and corticosteroids, has not shown a significant survival benefit when compared to cytokine-based therapy (152). Although adjuvant gemcitabine and 5-FU-based chemotherapy has been suggested to offer some survival benefit (153), RCC is generally considered one of the most resistant solid tumors to classic cytotoxic agents. Rarely, brain and pulmonary metastases from RCC display spontaneous complete or partial regression (154,155). Brain and spinal metastases are a common cause of neurologic morbidity in RCC. Although less common, paraneoplastic syndromes associated with RCC may also cause significant neurologic complications.

5.1. Brain Metastases Brain metastases from RCC are not uncommon (Fig. 2). In a review of 114 patients with renal carcinoma followed over 5 years, diagnosis of brain metastases occurred in 5.2% of patients after 1 year and in 9.8% of patients after 5 years (156). An earlier autopsy study reported an 11% incidence of brain metastases from RCC (157). In 10–25% of patients with metastatic RCC, brain metastases are detected at the time of diagnosis of primary cancer (156,158,159). Otherwise, the mean interval from the diagnosis of RCC to the diagnosis of brain metastases is less than 2 years (156,158,160). Rarely, brain metastases will be diagnosed years after treatment for primary cancer (161,162). At the time of diagnosis of brain metastases, the majority of patients have metastases to other organs, most commonly the lungs (156,158). Most patients will have neurologic signs or symptoms including headache, motor deficits,

Fig. 2. Multiple intracranial metastases from renal cell carcinoma. These gadolinium-enhanced MR images demonstrate three parenchymal and one calvarial metastasis. The left frontal parenchymal metastasis demonstrated pre-contrast T1 shortening and extracellular methemoglobin on other sequences indicative of hemorrhage.

472

Part VII / Neurologic Complications of Specific Malignancies

ataxia, mental status changes, or seizures (158,160,163). Intratumoral hemorrhage, occurring spontaneously or in conjunction with therapy, is common in RCC brain metastases (160,164). The median survival of patients with RCC after diagnosis of brain metastases is 4–13 months (158,160,163,165). Besides parenchymal brain lesions, metastases to choroid plexus and calvarium are rarely associated with RCC. Surgery is often considered in RCC patients with single brain metastases or with multiple metastases that are accessible through craniotomy. In Badalament’s series of 22 patients diagnosed with < 3 accessible brain metastases from renal cell carcinoma, 2 died within 30 days of surgery and the median survival of 20 patients was 21 months (166). The majority of patients were previously treated with WBRT or received post-operative radiation therapy. Eight of these patients had CNS relapse, 3 locally, and 5 distantly. Patients diagnosed with brain metastases more than 1 year after treatment for RCC had a median survival time of 30 months compared to 9 months for those diagnosed with brain metastases within a year of RCC diagnosis. Wronski reported a median survival rate of 12.6 months following surgical resection of brain metastases in 50 patients with RCC (160). Twenty-two of these patients had post-operative WBRT, which did not provide any survival benefit. Brain metastases from RCC are relatively radioresistant. In a study of palliative radiotherapy for brain metastases in 49 RCC patients, Maor reported that only 30% of patients had a response to WBRT (167). Median survival time for patients who responded to WBRT was 17 months. Wronski reviewed the outcome of 119 patients treated with WBRT for brain metastases from RCC and reported a median survival time of 4.4 months (158). In a multivariate analysis of possible prognostic factors, only single brain metastases, lack of distant metastases at the time of diagnosis, and tumor diameter of < 2 cm were significant. When feasible, WBRT combined with surgical resection appears superior to WBRT alone as treatment for RCC brain metastases (163,168,169). Chemotherapy is presently considered ineffective in the treatment for brain metastases from RCC. Chemotherapeutic agents able to cross the blood–brain barrier, such as gemcitabine, may potentially serve as adjuvant radio-sensitizing agents (170). The role of immunotherapy is also not clearly defined. Unlike systemic metastases, brain metastases from RCC are generally considered nonresponsive to immunotherapy due to blood–brain barrier limitations. Recent advances in radiosurgery offer another treatment option for patients with brain metastases from RCC. The attraction of radiosurgery over surgical resection include its non-invasiveness, the ability to treat multiple lesions in a single setting, the potential to treat lesions not accessible by craniotomy, and the ability to be performed repeatedly for tumor recurrence. Single-fraction high-dose radiation circumvents the relative resistance of RCC brain metastases to fractionated radiotherapy. Additionally, radiosurgery is associated with low morbidity and mortality rates. Radiosurgery provides local tumor control in 83–96% of brain metastases from RCC (164, 171–174). Median survival time following radiosurgery ranges between 6 and 9.5 months (164,172,175). WBRT in conjunction with radiosurgery has been reported to increase overall survival time in RCC patients with brain metastases (175); however, this is not confirmed in all studies (171,174). Fractionated stereotactic radiotherapy—the use of stereotactic techniques to deliver repeated doses of highly focal radiation to the target tissue—has been reported to produce a high rate of local tumor control similar to that of radiosurgery (176). The advantages of fractionated stereotactic radiotherapy over radiosurgery are unclear.

5.2. Leptomeningeal Disease Leptomeningeal carcinomatosis associated with RCC is extremely rare. A single case study of an untreated RCC patient diagnosed with leptomeningeal carcinomatosis after presenting with progressive bilateral leg weakness and gait instability has been reported (177).

5.3. Spinal Cord Disease Spinal metastases from RCC are a common cause of neurologic morbidity. In Saitoh’s autopsy study of distant metastases of renal cell carcinoma, 37% of cases had metastases to the bone with thoracic spine, lumbar spine, and ribs having the highest frequencies of lesions (157). Symptomatic patients typically complain of local or radicular pain and sensorimotor deficits. Surgical techniques and palliative radiotherapy are the basis of treatment for patients with significant morbidity as osseous metastases from RCC are resistant to chemotherapy. Due to

Chapter 24 / Neurologic Complications of Genitourinary Cancer

473

the high vascularity of RCC metastases, spinal angiography and tumor embolization is commonly utilized in non-emergent cases of spinal metastases as an effective technique to reduce intra-operative blood loss (178). Jackson evaluated the effectiveness of spinal surgery in 79 patients at MD Anderson Cancer Center with metastatic RCC (179). After surgery, 88% of patients reported an improvement in pain. Functional improvement was seen in 36 of 55 patients with neurologic deficits, with 14 regaining ambulation. Preoperative embolization was performed in 47 patients and was associated with major morbidity in 8.5% of cases. Preoperative radiotherapy was performed in 40 patients, and pre- or post-operative chemotherapy/immunotherapy was performed in 7 patients. Median survival following surgical intervention was 12.3 months. Reoperation for tumor recurrence was performed in 16% of patients. Similar outcomes were reported by Sundaresan in a retrospective analysis of 30 patients with spinal metastases from RCC who underwent spinal surgery (180). Twenty-seven of the 30 patients (90%) had post-operative neurologic improvement. Pre-operative embolization was performed in 17 patients. Seventeen patients had failed previous radiotherapy. Blood loss was a major surgical complication seen in 11 patients; 9 of these patients did not undergo presurgical embolization. The median survival time from spinal surgery was 16 months. In a earlier study, Durr reported a 49% 1-year survival rate following resection of osseous metastases (one-third of which arose from the spine) (181). Multivariate analysis identified solitary osseous metastasis and latency between diagnosis of RCC and detection of osseous metastasis as good prognostic indicators. Gael reported the surgical outcomes in a small series of 11 RCC patients with either progressive paraparesis (4 patients) or immobility (8 patients) from spinal metastases (182). At time of discharge, all patients were mobile with little reported pain. Mean survival time after surgery was 10 months for patients with multiple spinal metastases. Patients with solitary metastases had a longer postoperative survival time and appreciated long-term mobility. Surgery for tumor recurrence was performed in 3 patients. The role of spinal angiography and tumor embolization as primary therapy in spinal metastases is controversial, but has been associated with neurologic improvement or stabilization in a small number of patients (183,184). Although spinal metastases from RCC are generally considered radioresistant, palliative radiotherapy for osseous metastases should be considered. Reddy examined the effectiveness of palliative radiation for bone metastases from RCC and noted that 59% of patients reported some pain relief at 2 weeks, but only 12% had complete pain relief at 8 weeks (185). Dibiase also reported favorable results with radiotherapy to palliate focally symptomatic metastases from renal cell carcinoma including bone and spinal lesions (186). The use of stereotactic radiosurgery in the treatment of spinal metastases from RCC has recently been investigated. In Gerszten’s study of 48 patients with symptomatic RCC spinal metastases, 28 of 34 patients treated with a mean of 20 Gy for pain reported an improvement in symptoms (187). Additionally, 7 of 8 patients treated for radiographic tumor progression had tumor control after stereotactic radiosurgery. Overall, 6 patients underwent subsequent surgical intervention for progression of neurologic deficits. RCC occasionally metastasizes to the intramedullary compartment of the spinal cord (Fig. 3). Management of this complication does not differ from the general guidelines outlining treatment of intramedullary spinal cord metastases discussed in Chapter 11.

5.4. Paraneoplastic Syndromes Paraneoplastic neurologic complications of RCC are relatively rare. Paraneoplastic motor neuron disease presenting with symptoms mimicking amyotrophic lateral sclerosis in one patient and with hypercapnic respiratory failure in another patient have been reported (188,189). Although the association between RCC and motor neuron disease in these patients is controversial, both patients improved after nephrectomy. Paraneoplastic polymyositis presenting as weakness has also been associated with RCC (190–192). One reported patient developed myasthenia gravis that heralded RCC diagnosis and improved after nephrectomy (193). More recently, Hagel described a paraneoplastic frontal lobe disorder and ataxia in a patient with RCC (194). An extensive workup for antineuronal antibodies in this patient was negative. The pre-nephrectomy immunoglobulin fraction of the patient was demonstrated to bind to neurons of normal brain tissue, whereas the post-nephrectomy fraction did not. This finding may suggest an antibody-mediated paraneoplastic syndrome.

474

Part VII / Neurologic Complications of Specific Malignancies

Fig. 3. Intramedullary spinal cord metastasis from renal cell carcinoma. A 2-cm enhancing mass is seen within the spinal cord at the T5-6 level on post-contrast axial (A) and sagittal (B) images. Increased signal on the T2-weighted images (C) is seen in the central portion of the cord extending superiorly and inferiorly from the mass consistent with a tumor-associated syrinx. This patient developed paraplegia three weeks after diagnosis of a small frontal lobe metastasis.

5.5. Complications of Treatment Neurologic complications of immunotherapy for treatment of metastatic renal cell carcinoma are covered in Chapter 17. In the recent phase III trial of the multikinase inhibitor sorafenib, 59 of 451 (13%) patients reported sensory neuropathy associated with the study drug (148).

6. CONCLUSION Dysfunction of both the central and the peripheral nervous systems is not uncommonly attributable to genitourinary cancers or their treatment. The association of these malignancies and their management strategies with neurologic morbidity is often specific to the primary cancer. Unfortunately, neurologic complications are usually associated with advanced stages of genitourinary cancers. Because novel approaches in treatment regimens prolong the survival rates in patients with these cancers, a higher incidence of neurologic complications can be expected. The management options of these complications should not only take into account the primary cancer, but also the age, health, and expectations of the individual patient.

Chapter 24 / Neurologic Complications of Genitourinary Cancer

475

REFERENCES 1. Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA Cancer J Clin. 2007 Jan–Feb;57(1):43–66. 2. Nelson WG, De Marzo AM, Isaacs WB. Prostate cancer. N Engl J Med. 2003 Jul 24;349(4):366–381. 3. Bubendorf L, Schopfer A, Wagner U et al. Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol. 2000 May;31(5):578–583. 4. Zincke H, Oesterling JE, Blute ML et al. Long-term (15 years) results after radical prostatectomy for clinically localized (stage T2c or lower) prostate cancer. J Urol. 1994 Nov;152(5 Pt 2):1850–1857. 5. Thompson IM, Jr., Tangen CM, Paradelo J et al. Adjuvant radiotherapy for pathologically advanced prostate cancer: a randomized clinical trial. JAMA. 2006 Nov 15;296(19):2329–2335. 6. See WA, Wirth MP, McLeod DG et al. Bicalutamide as immediate therapy either alone or as adjuvant to standard care of patients with localized or locally advanced prostate cancer: first analysis of the early prostate cancer program. J Urol. 2002 Aug;168(2):429–435. 7. Bolla M, Collette L, Blank L et al. Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet. 2002 Jul 13;360(9327):103–106. 8. Tannock IF, de Wit R, Berry WR et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med. 2004 Oct 7;351(15):1502–1512. 9. Petrylak DP, Tangen CM, Hussain MH et al. Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med. 2004 Oct 7;351(15):1513–1520. 10. Tazi H, Manunta A, Rodriguez A et al. Spinal cord compression in metastatic prostate cancer. Eur Urol. 2003 Nov;44(5):527–532. 11. Rosenthal MA, Rosen D, Raghavan D et al. Spinal cord compression in prostate cancer: a 10-year experience. Br J Urol. 1992 May;69(5):530–533. 12. Kuban DA, el–Mahdi AM, Sigfred SV, Schellhammer PF, Babb TJ. Characteristics of spinal cord compression in adenocarcinoma of prostate. Urology. 1986 Nov;28(5):364–9. 13. Liskow A, Chang CH, DeSanctis P et al. Epidural cord compression in association with genitourinary neoplasms. Cancer. 1986 Aug 15;58(4):949–954. 14. Smith EM, Hampel N, Ruff RL et al. Spinal cord compression secondary to prostate carcinoma: treatment and prognosis. J Urol. 1993 Feb;149(2):330–333. 15. Flynn DF, Shipley WU. Management of spinal cord compression secondary to metastatic prostatic carcinoma. Urol Clin North Am. 1991 Feb;18(1):145–152. 16. Shoskes DA, Perrin RG. The role of surgical management for symptomatic spinal cord compression in patients with metastatic prostate cancer. J Urol. 1989 Aug;142(2 Pt 1):337–339. 17. Bayley A, Milosevic M, Blend R et al. A prospective study of factors predicting clinically occult spinal cord compression in patients with metastatic prostate carcinoma. Cancer. 2001 Jul 15;92(2):303–310. 18. Huddart RA, Rajan B, Law M et al. Spinal cord compression in prostate cancer: treatment outcome and prognostic factors. Radiother Oncol. 1997 Sep;44(3):229–236. 19. Susuki K, Matsumoto S, Kitagawa N et al. Epidural compression of the cauda equina caused by vertebral osteoblastic metastasis of prostatic carcinoma: resolution by hormonal therapy. J Neurol Neurosurg Psychiatry. 2000 Apr;68(4):514–515. 20. Zelefsky MJ, Scher HI, Krol G et al. Spinal epidural tumor in patients with prostate cancer: clinical and radiographic predictors of response to radiation therapy. Cancer. 1992 Nov 1;70(9):2319–2325. 21. Aass N, Fossa SD. Pre- and post-treatment daily life function in patients with hormone-resistant prostate carcinoma treated with radiotherapy for spinal cord compression. Radiother Oncol. 2005 Mar;74(3):259–265. 22. Ernst DS, Tannock IF, Winquist EW et al. Randomized, double-blind, controlled trial of mitoxantrone/prednisone and clodronate versus mitoxantrone/prednisone and placebo in patients with hormone-refractory prostate cancer and pain. J Clin Oncol. 2003 Sep 1;21(17):3335–3342. 23. Small EJ, Smith MR, Seaman JJ et al. Combined analysis of two multicenter, randomized, placebo-controlled studies of pamidronate disodium for the palliation of bone pain in men with metastatic prostate cancer. J Clin Oncol. 2003 Dec 1;21(23):4277–4284. 24. Saad F, Gleason DM, Murray R et al. A randomized, placebo-controlled trial of zoledronic acid in patients with hormone-refractory metastatic prostate carcinoma. J Nat Cancer Inst. 2002 Oct 2;94(19):1458–1468. 25. Long MA, Husband JE. Features of unusual metastases from prostate cancer. Br J Radiol. 1999 Oct;72(862):933–941. 26. Raizer JJ, DeAngelis LM. Cerebral sinus thrombosis diagnosed by MRI and MR venography in cancer patients. Neurol. 2000 Mar 28;54(6):1222–1226. 27. Sakurai N, Koike Y, Hashizume Y et al. Dural arteriovenous malformation and sinus thromboses in a patient with prostate cancer: an autopsy case. Intern Med (Tokyo Jpn). 1992 Aug;31(8):1032–1037. 28. McDermott RS, Anderson PR, Greenberg RE et al. Cranial nerve deficits in patients with metastatic prostate carcinoma: clinical features and treatment outcomes. Cancer. 2004 Oct 1;101(7):1639–1643. 29. Seymore CH, Peeples WJ. Cranial nerve involvement with carcinoma of prostate. Urol. 1988 Mar;31(3):211–213. 30. Capobianco DJ, Brazis PW, Rubino FA et al. Occipital condyle syndrome. Headache. 2002 Feb;42(2):142–146. 31. Kattah JC, Chrousos GC, Roberts J et al. Metastatic prostate cancer to the optic canal. Ophthalmol. 1993 Nov;100(11):1711–1715. 32. Prashant R, Franks A. Collet–Sicard syndrome: a report and review. Lancet Oncol. 2003 Jun;4(6):376–377. 33. Sicenica T, Venkata Balaji G, Klein A et al. Villaret’s syndrome in a man with prostate carcinoma. Am J Med. 2000 Apr 15;108(6): 516–517. 34. Halachmi S, Madeb R, Madjar S et al. Numb chin syndrome as the presenting symptom of metastatic prostate carcinoma. Urol. 2000 Feb;55(2):286.

476

Part VII / Neurologic Complications of Specific Malignancies

35. Tremont-Lukats IW, Bobustuc G, Lagos GK et al. Brain metastasis from prostate carcinoma: The M.D. Anderson Cancer Center experience. Cancer. 2003 Jul 15;98(2):363–368. 36. Kleinschmidt-DeMasters BK. Dural metastases. A retrospective surgical and autopsy series. Arch Path Lab Med. 2001 Jul;125(7): 880–887. 37. Lyons MK, Drazkowski JF, Wong WW et al. Metastatic prostate carcinoma mimicking meningioma: case report and review of the literature. Neurologist. 2006 Jan;12(1):48–52. 38. Lippman SM, Buzaid AC, Iacono RP et al. Cranial metastases from prostate cancer–simulating meningioma: report of two cases and review of the literature. Neurosurgery. 1986 Nov;19(5):820–823. 39. Tomlin JM, Alleyne CH. Transdural metastasis from adenocarcinoma of the prostate mimicking subdural hematoma: case report. Surg Neurol. 2002 Nov;58(5):329–331; discussion 331. 40. Scarrow AM, Rajendran PR, Marion D. Metastatic prostate adenocarcinoma of the dura mater. Br J Neurosurg. 2000 Oct;14(5): 473–474. 41. Barolat-Romana G, Maiman D, Dernbach P et al. Prostate carcinoma presenting as intracranial hemorrhage: case report. J Neurosurg. 1984 Feb;60(2):414–416. 42. McCutcheon IE, Eng DY, Logothetis CJ. Brain metastasis from prostate carcinoma: antemortem recognition and outcome after treatment. Cancer. 1999 Dec 1;86(11):2301–2311. 43. Sutton MA, Watkins HL, Green LK et al. Intracranial metastases as the first manifestation of prostate cancer. Urol. 1996 Nov;48(5): 789–793. 44. Cone LA, Koochek K, Henager HA et al. Leptomeningeal carcinomatosis in a patient with metastatic prostate cancer: case report and literature review. Surg Neurol. 2006 Apr;65(4):372–375, discussion 5–6. 45. Honda M, Miyagawa I. Prostatic meningeal carcinomatosis with low serum level of prostate-specific antigen. Urol. 2005 Dec;66(6):1320. 46. Schaller B, Merlo A, Kirsch E et al. Prostate-specific antigen in the cerebrospinal fluid leads to diagnosis of solitary cauda equina metastasis: a unique case report and review of the literature. Br J Cancer. 1998 Jun;77(12):2386–2389. 47. Mencel PJ, DeAngelis LM, Motzer RJ. Hormonal ablation as effective therapy for carcinomatous meningitis from prostatic carcinoma. Cancer. 1994 Apr 1;73(7):1892–1894. 48. Ladha SS, Spinner RJ, Suarez GA et al. Neoplastic lumbosacral radiculoplexopathy in prostate cancer by direct perineural spread: an unusual entity. Muscle Nerve. 2006 Nov;34(5):659–665. 49. Ansari MS, Nabi G, Aron M. Solitary radial head metastasis with wrist drop: a rare presentation of metastatic prostate cancer. Urol Int. 2003;70(1):77–79. 50. Rubin DI, Kimmel DW, Cascino TL. Outcome of peroneal neuropathies in patients with systemic malignant disease. Cancer. 1998 Oct 15;83(8):1602–1606. 51. Graus F, Rogers LR, Posner JB. Cerebrovascular complications in patients with cancer. Medicine. 1985 Jan;64(1):16–35. 52. Cestari DM, Weine DM, Panageas KS et al. Stroke in patients with cancer: incidence and etiology. Neurol. 2004 Jun 8;62(11): 2025–2030. 53. Rabinstein AA, Giovanelli C, Romano JG et al. Surgical treatment of nonbacterial thrombotic endocarditis presenting with stroke. J Neurol. 2005 Mar;252(3):352–355. 54. Inamasu J, Nakamura Y, Saito R et al. Cerebellar hemorrhage secondary to cranial metastasis of prostate cancer: case report. Neurol Med. 2004 Feb;44(2):82–85. 55. Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and oncologic profiles of patients seropositive for type 1 antineuronal nuclear autoantibodies. Neurol. 1998 Mar;50(3):652–657. 56. Baird AD, Cornford PA, Helliwell T et al. Small cell prostate cancer with anti-Hu positive peripheral neuropathy. J Urol. 2002 Jul;168(1):192. 57. Baloh RW, DeRossett SE, Cloughesy TF et al. Novel brainstem syndrome associated with prostate carcinoma. Neurology. 1993 Dec;43(12):2591–2596. 58. Hill CL, Zhang Y, Sigurgeirsson B et al. Frequency of specific cancer types in dermatomyositis and polymyositis: a population-based study. Lancet. 2001 Jan 13;357(9250):96–100. 59. Subramonian K, Sundaram SK, MacDonald Hull SP. Carcinoma of the prostate associated with dermatomyositis. BJU Int. 2000 Aug;86(3):401–402. 60. Masuda H, Urushibara M, Kihara K. Successful treatment of dermatomyositis associated with adenocarcinoma of the prostate after radical prostatectomy. J Urol. 2003 Mar;169(3):1084. 61. Park Y, Oster MW, Olarte MR. Prostatic cancer with an unusual presentation: polymyositis and mediastinal adenopathy. Cancer. 1981 Sep 1;48(5):1262–1264. 62. Matschke J, Kromminga A, Erbersdobler A et al. Paraneoplastic cerebellar degeneration and anti-Yo antibodies in a man with prostatic adenocarcinoma. J Neurol Neurosurg Psychiatry. 2006 Dec 22. 63. Delahunt B, Abernethy DA, Johnson CA et al. Prostate carcinoma and the Lambert–Eaton myasthenic syndrome. J Urol. 2003 Jan;169(1):278–279. 64. Tetu B, Ro JY, Ayala AG et al. Small cell carcinoma of prostate associated with myasthenic (Eaton–Lambert) syndrome. Urol. 1989 Feb;33(2):148–152. 65. Talcott JA, Manola J, Clark JA et al. Time course and predictors of symptoms after primary prostate cancer therapy. J Clin Oncol. 2003 Nov 1;21(21):3979–3986. 66. Nelson BA, Chang SS, Cookson MS et al. Morbidity and efficacy of genitofemoral nerve grafts with radical retropubic prostatectomy. Urol. 2006 Apr;67(4):789–792.

Chapter 24 / Neurologic Complications of Genitourinary Cancer

477

67. Secin FP, Koppie TM, Scardino PT et al. Bilateral cavernous nerve interposition grafting during radical retropubic prostatectomy: memorial Sloan-Kettering Cancer Center experience. J Urol. 2007 Feb;177(2):664–668. 68. Beard CJ, Lamb C, Buswell L et al. Radiation-associated morbidity in patients undergoing small-field external beam irradiation for prostate cancer. Int J Radiat Oncol Biol Phys. 1998 May 1;41(2):257–262. 69. Spaliviero M, Steinberg AP, Kaouk JH et al. Laparoscopic injury and repair of obturator nerve during radical prostatectomy. Urol. 2004 Nov;64(5):1030. 70. Horwich A, Shipley J, Huddart R. Testicular germ-cell cancer. Lancet. 2006 Mar 4;367(9512):754–765. 71. Spears WT, Morphis JG, 2nd, Lester SG et al. Brain metastases and testicular tumors: long-term survival. Int J Radiat Oncol Biol Phys. 1992;22(1):17–22. 72. International Germ Cell Consensus Classification: a prognostic factor-based staging system for metastatic germ cell cancers. International Germ Cell Cancer Collaborative Group. J Clin Oncol. 1997 Feb;15(2):594–603. 73. Lutterbach J, Spetzger U, Bartelt S et al. Malignant germ cell tumors metastatic to the brain: a model for a curable neoplasm? The Freiburg experience and a review of the literature. J Neuro-oncol. 2002 Jun;58(2):147–156. 74. Spunt SL, Walsh MF, Krasin MJ et al. Brain metastases of malignant germ cell tumors in children and adolescents. Cancer. 2004 Aug 1;101(3):620–626. 75. Mahalati K, Bilen CY, Ozen H et al. The management of brain metastasis in nonseminomatous germ cell tumours. BJU Int. 1999 Mar;83(4):457–461. 76. Bokemeyer C, Nowak P, Haupt A et al. Treatment of brain metastases in patients with testicular cancer. J Clin Oncol. 1997 Apr;15(4):1449–1454. 77. Fossa SD, Bokemeyer C, Gerl A et al. Treatment outcome of patients with brain metastases from malignant germ cell tumors. Cancer. 1999 Feb 15;85(4):988–997. 78. Raina V, Singh SP, Kamble N et al. Brain metastasis as the site of relapse in germ cell tumor of testis. Cancer. 1993 Oct 1;72(7): 2182–2185. 79. Crabb SJ, McKendrick JJ, Mead GM. Brain as sanctuary site of relapse in germ cell cancer patients previously treated with chemotherapy. Clin Incol (R Coll Radiol). 2002 Aug;14(4):287–293. 80. Salvati M, Piccirilli M, Raco A et al. Brain metastasis from nonseminomatous germ cell tumors of the testis: indications for aggressive treatment. Neurosurg Rev. 2006 Apr;29(2):130–137. 81. Gultekin SH, Rosenfeld MR, Voltz R et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings and tumour association in 50 patients. Brain. 2000 Jul;123 (Pt 7):1481–1494. 82. Voltz R, Gultekin SH, Rosenfeld MR et al. A serologic marker of paraneoplastic limbic and brain-stem encephalitis in patients with testicular cancer. N Engl J Med. 1999 Jun 10;340(23):1788–1795. 83. Wingerchuk DM, Noseworthy JH, Kimmel DW. Paraneoplastic encephalomyelitis and seminoma: importance of testicular ultrasonography. Neurol. 1998 Nov;51(5):1504–1507. 84. Almeras C, Soussi N, Molko N et al. Paraneoplastic limbic encephalitis, a complication of the testicular cancer. Urol. 2001 Jul;58(1):105. 85. Dalmau J, Graus F, Villarejo A et al. Clinical analysis of anti-Ma2-associated encephalitis. Brain. 2004 Aug;127(Pt 8):1831–1844. 86. Waragai M, Chiba A, Uchibori A et al. Anti-Ma2 associated paraneoplastic neurological syndrome presenting as encephalitis and progressive muscular atrophy. J Neurol Neurosurg Psychiatry. 2006 Jan;77(1):111–113. 87. Dalmau J, Gultekin SH, Voltz R et al. Ma1, a novel neuron- and testis-specific protein, is recognized by the serum of patients with paraneoplastic neurological disorders. Brain. 1999 Jan;122 (Pt 1):27–39. 88. Mackey JR, Venner P. Seminoma with isolated central nervous system relapse, and salvage with craniospinal irradiation. Urol. 1998 Jun;51(6):1043–1045. 89. Denissen NH, van Spronsen DJ, Smilde TJ et al. Leptomeningeal carcinomatosis in relapsed nonseminoma testis: a 1-year complete remission with high-dose chemotherapy. Anticancer Drugs. 2005 Sep;16(8):897–899. 90. Rouah E, Goodman JC, Harper RL. Acute subdural hematoma and metastatic seminoma. Neurol. 1986 Mar;36(3):418–420. 91. Hitchins RN, Philip PA, Wignall B et al. Bone disease in testicular and extragonadal germ cell tumours. Br J Cancer. 1988 Dec;58(6):793–796. 92. Gale J, Mead GM, Simmonds PD. Management of spinal cord and cauda equina compression secondary to epidural metastatic disease in adults with malignant germ cell tumours. Clin Oncol (R Coll Radiol). 2002 Dec;14(6):481–490. 93. Arnold PM, Morgan CJ, Morantz RA et al. Metastatic testicular cancer presenting as spinal cord compression: report of two cases. Surg Neurol. 2000 Jul;54(1):27–33. 94. Sawin PD, VanGilder JC. Spinal cord compression from metastatic Leydig’s cell tumor of the testis: case report. Neurosurgery. 1996 Feb;38(2):407–411. 95. Lee JK, Kim SH, Kim JH et al. Metastatic spinal cord compression of testicular yolk sac tumor. Childs Nerv Syst. 2002 Apr; 18(3–4):171–174. 96. Cooper K, Bajorin D, Shapiro W et al. Decompression of epidural metastases from germ cell tumors with chemotherapy. J Neuro-oncol. 1990 Jun;8(3):275–280. 97. Jaeckle KA, Young DF, Foley KM. The natural history of lumbosacral plexopathy in cancer. Neurol. 1985 Jan;35(1):8–15. 98. Horvath L, Bayfield M, Clifford A et al. Unusual presentations of germ cell tumors. Case 1: recurrent laryngeal nerve palsyin mediastinal seminoma. J Clin Oncol. 2001 Feb 1;19(3):909–911. 99. Hans S, Lindner DW, Webster JD. Brachial plexus neuropathy from metastatic testicular seminoma: prolonged survival after surgery and radiation therapy. Urol. 1985 Apr;25(4):398–400. 100. Bowen J, Gregory R, Squier M et al. The post-irradiation lower motor neuron syndrome neuronopathy or radiculopathy? Brain. 1996 Oct;119 (Pt 5):1429–1439.

478

Part VII / Neurologic Complications of Specific Malignancies

101. Brydoy M, Storstein A, Dahl O. Transient neurological adverse effects following low-dose radiation therapy for early-stage testicular seminoma. Radiother Oncol. 2006 Dec 23. 102. Kourea HP, Bilsky MH, Leung DH et al. Subdiaphragmatic and intrathoracic paraspinal malignant peripheral nerve sheath tumors: a clinicopathologic study of 25 patients and 26 tumors. Cancer. 1998 Jun 1;82(11):2191–2203. 103. West DA, Parra RO, Manepalli A et al. Development of a malignant peripheral nerve sheath tumor following treatment for testicular seminoma. Urol. 1997 Aug;50(2):292–294. 104. Amin A, Saifuddin A, Flanagan A et al. Radiotherapy-induced malignant peripheral nerve sheath tumor of the cauda equina. Spine. 2004 Nov 1;29(21):E506–E509. 105. von Schlippe M, Fowler CJ, Harland SJ. Cisplatin neurotoxicity in the treatment of metastatic germ cell tumour: time course and prognosis. Br J Cancer. 2001 Sep 14;85(6):823–826. 106. Strumberg D, Brugge S, Korn MW et al. Evaluation of long-term toxicity in patients after cisplatin-based chemotherapy for nonseminomatous testicular cancer. Ann Oncol. 2002 Feb;13(2):229–236. 107. Petersen PM, Hansen SW. The course of long-term toxicity in patients treated with cisplatin-based chemotherapy for nonseminomatous germ-cell cancer. Ann Oncol. 1999 Dec;10(12):1475–1483. 108. Hansen SW, Helweg-Larsen S, Trojaborg W. Long-term neurotoxicity in patients treated with cisplatin, vinblastine, and bleomycin for metastatic germ cell cancer. J Clin Oncol. 1989 Oct;7(10):1457–1461. 109. Fossa SD, de Wit R, Roberts JT et al. Quality of life in good prognosis patients with metastatic germ cell cancer: a prospective study of the European Organization for Research and Treatment of Cancer Genitourinary Group/Medical Research Council Testicular Cancer Study Group (30941/TE20). J Clin Oncol. 2003 Mar 15;21(6):1107–1118. 110. Fossa SD, Lehne G, Heimdal K et al. Clinical and biochemical long-term toxicity after post-operative cisplatin-based chemotherapy in patients with low-stage testicular cancer. Oncol. 1995 Jul–Aug;52(4):300–305. 111. Bokemeyer C, Berger CC, Kuczyk MA et al. Evaluation of long-term toxicity after chemotherapy for testicular cancer. J Clin Oncol. 1996 Nov;14(11):2923–2932. 112. Donohue JP, Foster RS, Rowland RG et al. Nerve-sparing retroperitoneal lymphadenectomy with preservation of ejaculation. J Urol. 1990 Aug;144(2 Pt 1):287–291; discussion 91–92. 113. Hall MC, Koch MO, Smith JA, Jr. Femoral neuropathy complicating urologic abdominopelvic procedures. Uroly. 1995 Jan;45(1): 146–149. 114. Leibovitch I, Nash PA, Little JS, Jr. et al. Spinal cord ischemia after post-chemotherapy retroperitoneal lymph node dissection for nonseminomatous germ cell cancer. J Urol. 1996 Mar;155(3):947–951. 115. Petrioli R, Frediani B, Manganelli A et al. Comparison between a cisplatin-containing regimen and a carboplatin-containing regimen for recurrent or metastatic bladder cancer patients: a randomized phase II study. Cancer. 1996 Jan 15;77(2):344–351. 116. Dhote R, Beuzeboc P, Thiounn N et al. High incidence of brain metastases in patients treated with an M-VAC regimen for advanced bladder cancer. Eur Urol. 1998;33(4):392–395. 117. Sternberg CN, Yagoda A, Scher HI et al. Methotrexate, vinblastine, doxorubicin, and cisplatin for advanced transitional cell carcinoma of the urothelium: efficacy and patterns of response and relapse. Cancer. 1989 Dec 15;64(12):2448–2458. 118. Sternberg CN, Yagoda A, Scher HI et al. M-VAC (methotrexate, vinblastine, doxorubicin, and cisplatin) for advanced transitional cell carcinoma of the urothelium. J Urol. 1988 Mar;139(3):461–469. 119. Anderson TS, Regine WF, Kryscio R et ak. Neurologic complications of bladder carcinoma: a review of 359 cases. Cancer. 2003 May 1;97(9):2267–2272. 120. Clatterbuck RE, Sampath P, Olivi A. Transitional cell carcinoma presenting as a solitary brain lesion: a case report and review of the world literature. J Neuro–oncol. 1998 Aug;39(1):91–94. 121. Mahmoud-Ahmed AS, Suh JH, Kupelian PA et al. Brain metastases from bladder carcinoma: presentation, treatment and survival. J Urol. 2002 Jun;167(6):2419–2422. 122. Anderson RS, el-Mahdi AM, Kuban DA et al. Brain metastases from transitional cell carcinoma of urinary bladder. Urol. 1992 Jan;39(1):17–20. 123. Steinfeld AD, Zelefsky M. Brain metastases from carcinoma of bladder. Urol. 1987 Apr;29(4):375–376. 124. Wakisaka S, Miyahara S, Nonaka A et al. Brain metastasis from transitional cell carcinoma of the bladder: case report. Neurol Med. 1990 Mar;30(3):188–190. 125. Davies BJ, Bastacky S, Chung SY. Large cerebellar lesion as original manifestation of transitional cell carcinoma of the bladder. Urol. 2003 Oct;62(4):749. 126. Rosenstein M, Wallner K, Scher H et al. Treatment of brain metastases from bladder cancer. J Urol. 1993 Mar;149(3):480–483. 127. Protzel C, Zimmermann U, Asse E et al. Gemcitabine and radiotherapy in the treatment of brain metastases from transitional cell carcinoma of the bladder: a case report. J Neuro-oncol. 2002 Apr;57(2):141–145. 128. Mandell S, Wernz J, Morales P et al. Carcinomatous meningitis from transitional cell carcinoma of bladder. Urol. 1985 May;25(5): 520–521. 129. Eng C, Cunningham D, Quade BJ et al. Meningeal carcinomatosis from transitional cell carcinoma of the bladder. Cancer. 1993 Jul 15;72(2):553–557. 130. Hust MH, Pfitzer P. Cerebrospinal fluid and metastasis of transitional cell carcinoma of the bladder. Acta Cytol. 1982 Mar–Apr;26(2):217–223. 131. Bloch JL, Nieh PT, Walzak MP. Brain metastases from transitional cell carcinoma. J Urol. 1987 Jan;137(1):97–99. 132. Bruna J, Rojas-Marcos I, Martinez-Yelamos S et al. Meningeal carcinomatosis as the first manifestation of a transitional cell carcinoma of the bladder. J Neuro-oncol. 2003 May;63(1):63–67. 133. Prestigiacomo CJ, Balmaceda C, Dalmau J. Anti-Ri-associated paraneoplastic opsoclonus–ataxia syndrome in a man with transitional cell carcinoma. Cancer. 2001 Apr 15;91(8):1423–1428.

Chapter 24 / Neurologic Complications of Genitourinary Cancer

479

134. Greenlee JE, Dalmau J, Lyons T et al. Association of anti-Yo (type I) antibody with paraneoplastic cerebellar degeneration in the setting of transitional cell carcinoma of the bladder: detection of Yo antigen in tumor tissue and fall in antibody titers following tumor removal. Ann Neurol. 1999 Jun;45(6):805–809. 135. Lowe BA, Mershon C, Mangalik A. Paraneoplastic neurological syndrome in transitional cell carcinoma of the bladder. J Urol. 1992 Feb;147(2):462–464. 136. Mallon E, Osborne G, Dinneen M et al. Dermatomyositis in association with transitional cell carcinoma of the bladder. Clin Exp Dermatol. 1999 Mar;24(2):94–96. 137. Talanin NY, Bushore D, Rasberry R et al. Dermatomyositis with the features of inclusion body myositis associated with carcinoma of the bladder. Br J Dermatol. 1999 Nov;141(5):926–930. 138. Garcia-Donoso C, Sanchez-Munoz A, Lopez-Medrano F. Dermatomyositis and transitional cell carcinoma of the bladder: a rare paraneoplastic syndrome associated with tumor recurrence. Eur J Intern Med. 2003 Oct;14(6):397–398. 139. Bouropoulos C, Kanellakopoulou KD, Zarakovitis IE et al. Paraneoplastic polymyositis associated with transitional cell carcinoma of the bladder. J Urol. 1997 Mar;157(3):950–951. 140. Miller JI, Sarver RG, Drach GW. Leukemoid reaction: a rare paraneoplastic syndrome associated with advanced bladder carcinoma. Urology. 1994 Sep;44(3):444–446. 141. Rogers LR, Borkowski GP, Albers JW et al. Obturator mononeuropathy caused by pelvic cancer: six cases. Neurol. 1993 Aug;43(8):1489–1492. 142. Lane BR, Finelli A, Moinzadeh A et al. Nerve-sparing laparoscopic radical cystectomy: technique and initial outcomes. Urol. 2006 Oct;68(4):778–783. 143. Kaisary AV, Smith P. Spinal cord ischemia after ligation of both internal iliac arteries during radical cystoprostatectomy. Urol. 1985 Apr;25(4):395–397. 144. Sokoloff MH, deKernion JB, Figlin RA et al. Current management of renal cell carcinoma. CA: Cancer J Clin. 1996 Sep–Oct;46(5): 284–302. 145. Rabinovitch RA, Zelefsky MJ, Gaynor JJ et al. Patterns of failure following surgical resection of renal cell carcinoma: implications for adjuvant local and systemic therapy. J Clin Oncol. 1994 Jan;12(1):206–212. 146. Lokich J. Spontaneous regression of metastatic renal cancer: case report and literature review. Am J Clin Oncol. 1997 Aug;20(4): 416–418. 147. De Mulder PH, van Herpen CM, Mulders PA. Current treatment of renal cell carcinoma. Ann Oncol. 2004;15 Suppl 4:iv319–328. 148. Escudier B, Eisen T, Stadler WM et al. Sorafenib in advanced clear-cell renal-cell carcinoma. N Engl J Med. 2007 Jan 11;356(2): 125–134. 149. Motzer RJ, Hutson TE, Tomczak P et al. Sunitinib versus interferon-alfa in metastatic renal-cell carcinoma. N Engl J Med. 2007 Jan 11;356(2):115–124. 150. Yang JC, Haworth L, Sherry RM et al. A randomized trial of bevacizumab, an antivascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med. 2003 Jul 31;349(5):427–434. 151. Atkins MB, Hidalgo M, Stadler WM et al. Randomized phase II study of multiple dose levels of CCI–779, a novel mammalian target of rapamycin kinase inhibitor, in patients with advanced refractory renal cell carcinoma. J Clin Oncol. 2004 Mar 1;22(5):909–918. 152. Interferon-alpha and survival in metastatic renal carcinoma: early results of a randomised controlled trial. Medical Research Council Renal Cancer Collaborators. Lancet. 1999 Jan 2;353(9146):14–17. 153. Stadler WM, Huo D, George C et al. Prognostic factors for survival with gemcitabine plus 5-fluorouracil based regimens for metastatic renal cancer. J Urol. 2003 Oct;170(4 Pt 1):1141–1145. 154. Gleave ME, Elhilali M, Fradet Y et al. Interferon gamma-1b compared with placebo in metastatic renal-cell carcinoma. Canadian Urologic Oncology Group. N Engl J Med. 1998 Apr 30;338(18):1265–1271. 155. Guthbjartsson T, Gislason T. Spontaneous regression of brain metastasis secondary to renal cell carcinoma. Scand J Urol Nephrol. 1995 Jun;29(2):215–217. 156. Schouten LJ, Rutten J, Huveneers HA, Twijnstra A. Incidence of brain metastases in a cohort of patients with carcinoma of the breast, colon, kidney, and lung and melanoma. Cancer. 2002 May 15;94(10):2698–2705. 157. Saitoh H. Distant metastasis of renal adenocarcinoma. Cancer. 1981 Sep 15;48(6):1487–1491. 158. Wronski M, Maor MH, Davis BJ et al. External radiation of brain metastases from renal carcinoma: a retrospective study of 119 patients from the M.D. Anderson Cancer Center. Int J Radiat Oncol Biol Phys. 1997 Mar 1;37(4):753–759. 159. Radley MG, McDonald JV, Pilcher WH et al. Late solitary cerebral metastases from renal cell carcinoma: report of two cases. Surg Neurol. 1993 Mar;39(3):230–234. 160. Wronski M, Arbit E, Russo P et al. Surgical resection of brain metastases from renal cell carcinoma in 50 patients. Urol. 1996 Feb;47(2):187–193. 161. Sadatomo T, Yuki K, Migita K et al. Solitary brain metastasis from renal cell carcinoma 15 years after nephrectomy: case report. Neurol Med. 2005 Aug;45(8):423–427. 162. Roser F, Rosahl SK, Samii M. Single cerebral metastasis 3 and 19 years after primary renal cell carcinoma: case report and review of the literature. J Neurol Neurosurg Psychiatry. 2002 Feb;72(2):257–258. 163. Culine S, Bekradda M, Kramar A et al. Prognostic factors for survival in patients with brain metastases from renal cell carcinoma. Cancer. 1998 Dec 15;83(12):2548–2553. 164. Sheehan JP, Sun MH, Kondziolka D et al. Radiosurgery in patients with renal cell carcinoma metastasis to the brain: long-term outcomes and prognostic factors influencing survival and local tumor control. J Neurosurg. 2003 Feb;98(2):342–349. 165. Citterio G, Bertuzzi A, Tresoldi M et al. Prognostic factors for survival in metastatic renal cell carcinoma: retrospective analysis from 109 consecutive patients. Eur Urol. 1997;31(3):286–291.

480

Part VII / Neurologic Complications of Specific Malignancies

166. Badalament RA, Gluck RW, Wong GY et al. Surgical treatment of brain metastases from renal cell carcinoma. Urol. 1990 Aug;36(2):112–117. 167. Maor MH, Frias AE, Oswald MJ. Palliative radiotherapy for brain metastases in renal carcinoma. Cancer. 1988 Nov 1;62(9):1912–1917. 168. Patchell RA, Tibbs PA, Walsh JW et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med. 1990 Feb 22;322(8):494–500. 169. Vecht CJ, Haaxma-Reiche H, Noordijk EM et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol. 1993 Jun;33(6):583–590. 170. Maraveyas A, Sgouros J, Upadhyay S et al. Gemcitabine twice weekly as a radiosensitiser for the treatment of brain metastases in patients with carcinoma: a phase I study. Br J Cancer. 2005 Mar 14;92(5):815–819. 171. Schoggl A, Kitz K, Ertl A et al. Gamma-knife radiosurgery for brain metastases of renal cell carcinoma: results in 23 patients. Acta Neurochir. 1998;140(6):549–555. 172. Shuto T, Inomori S, Fujino H et al. Gamma knife surgery for metastatic brain tumors from renal cell carcinoma. J Neurosurg. 2006 Oct;105(4):555–560. 173. Payne BR, Prasad D, Szeifert G et al. Gamma surgery for intracranial metastases from renal cell carcinoma. J Neurosurg. 2000 May;92(5):760–765. 174. Mori Y, Kondziolka D, Flickinger JC et al. Stereotactic radiosurgery for brain metastasis from renal cell carcinoma. Cancer. 1998 Jul 15;83(2):344–353. 175. Hernandez L, Zamorano L, Sloan A et al. Gamma knife radiosurgery for renal cell carcinoma brain metastases. J Neurosurg. 2002 Dec;97(5 Suppl):489–493. 176. Ikushima H, Tokuuye K, Sumi M et al. Fractionated stereotactic radiotherapy of brain metastases from renal cell carcinoma. Int J Radiat Oncol Biol Phys. 2000 Dec 1;48(5):1389–1393. 177. Crino PB, Sater RA, Sperling M et al. Renal cell carcinomatous meningitis: pathologic and immunohistochemical features. Neurol. 1995 Jan;45(1):189–191. 178. Manke C, Bretschneider T, Lenhart M et al. Spinal metastases from renal cell carcinoma: effect of preoperative particle embolization on intraoperative blood loss. AJNR. 2001 May;22(5):997–1003. 179. Jackson RJ, Loh SC, Gokaslan ZL. Metastatic renal cell carcinoma of the spine: surgical treatment and results. J Neurosurg. 2001 Jan;94(1 Suppl):18–24. 180. Sundaresan N, Choi IS, Hughes JE et al. Treatment of spinal metastases from kidney cancer by presurgical embolization and resection. J Neurosurg. 1990 Oct;73(4):548–554. 181. Durr HR, Maier M, Pfahler M et al. Surgical treatment of osseous metastases in patients with renal cell carcinoma. Clin Orthopaedic Related Res. 1999 Oct(367):283–290. 182. Giehl JP, Kluba T. Metastatic spine disease in renal cell carcinoma: indication and results of surgery. Anticancer Research. 1999 Mar–Apr;19(2C):1619–1623. 183. Kuether TA, Nesbit GM, Barnwell SL. Embolization as treatment for spinal cord compression from renal cell carcinoma: case report. Neurosurgery. 1996 Dec;39(6):1260–1262; discussion 2–3. 184. O’Reilly GV, Kleefield J, Klein LA et al. Embolization of solitary spinal metastases from renal cell carcinoma: alternative therapy for spinal cord or nerve root compression. Surg Neurol. 1989 Apr;31(4):268–271. 185. Reddy S, Hendrickson FR, Hoeksema J et al. The role of radiation therapy in the palliation of metastatic genitourinary tract carcinomas: a study of the Radiation Therapy Oncology Group. Cancer. 1983 Jul 1;52(1):25–29. 186. DiBiase SJ, Valicenti RK, Schultz D et al. Palliative irradiation for focally symptomatic metastatic renal cell carcinoma: support for dose escalation based on a biological model. J Urol. 1997 Sep;158(3 Pt 1):746–749. 187. Gerszten PC, Burton SA, Ozhasoglu C et al. Stereotactic radiosurgery for spinal metastases from renal cell carcinoma. J Neurosurg Spine. 2005 Oct;3(4):288–295. 188. Forman D, Rae-Grant AD, Matchett SC et al. A reversible cause of hypercapnic respiratory failure: lower motor neuronopathy associated with renal cell carcinoma. Chest. 1999 Mar;115(3):899–901. 189. Evans BK, Fagan C, Arnold T et al. Paraneoplastic motor neuron disease and renal cell carcinoma: improvement after nephrectomy. Neurology. 1990 Jun;40(6):960–962. 190. Klausner AP, Ost MC, Waterhouse RL, Jr. et al. Occult renal cell carcinoma in a patient with polymyositis. Urol. 2002 May;59(5):773. 191. Solon AA, Gilbert CS, Meyer C. Myopathy as a paraneoplastic manifestation of renal cell carcinoma. Am J Med. 1994 Nov;97(5): 491–492. 192. Wurzer H, Brandstatter G, Harnoncourt K et al. Paraneoplastic polymyositis associated with a renal carcinoma. J Int Med. 1993 Nov;234(5):521–524. 193. Torgerson EL, Khalili R, Dobkin BH et al. Myasthenia gravis as a paraneoplastic syndrome associated with renal cell carcinoma. J Urol. 1999 Jul;162(1):154. 194. Hagel C, Stavrou D, Hansen HC. Paraneoplastic frontal lobe disorder and ataxia in renal cell carcinoma. Neuropathol Appl Neurobiol. 2005 Feb;31(1):97–99.

25

Neurologic Complications of Gastrointestinal Cancer Jeffrey Raizer,

MD,

and Jeffrey Cilley,

MD

CONTENTS Overview of Gastrointestinal Malignancies Esophageal Cancer Gastric Cancer Colorectal Cancer Hepatocellular Carcinoma (HCC) Gallbladder and Bile Duct (Cholangiocarcinoma) Carcinomas Pancreatic Cancer Peripheral Nervous System Complications Metabolic Abnormalities Chemotherapy-Related Neurologic Complications Conclusion References

Summary Neurologic complications are a relatively rare complication of gastrointestinal malignancies. In most cases, the complications that result are from direct spread of the individual tumor types to the affected areas of the central or peripheral nervous system. In this chapter, we review the relevant neurologic complications that can be seen in this group of malignancies. Key Words: gastrointestinal, colorectal cancer, esophageal cancer, gastrointestinal stromal tumor, hepatocellular carcinoma

1. OVERVIEW OF GASTROINTESTINAL MALIGNANCIES Malignancies of the gastrointestinal (GI) tract are one of the most common tumors presenting in people in the United States. There were an estimated 271,250 new cases diagnosed in 2007 and 134,710 deaths (1). The most commonly occurring GI malignancy is colorectal cancer, which comprises 10% of all new cancer diagnoses and is the third most common malignancy in men and women. Of all GI malignancies, the greatest increase in survival has been in patients with colorectal cancer. This is due to the combined use of modern surgical techniques and effective adjuvant chemotherapy, and in the case of rectal cancer the neoadjuvant use of radiation. Generally, staging takes into account the depth of invasion of the primary tumor into the wall of the viscus, the number and location of involved lymph nodes, and whether or not there are metastatic sites of disease. Treatment strategies usually favor surgical resection when possible followed by adjuvant chemotherapy or combined chemoradiation therapy to prevent local or distant recurrences. Neoadjuvant strategies using chemotherapy or chemoradiation therapy before surgery have also been looked at especially for tumors of the pancreas, esophagus, and rectum. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

481

482

Part VII / Neurologic Complications of Specific Malignancies

The approach to squamous cell anal cancer is different and consists principally of combined chemoradiotherapy. Treatment principles for gastrointestinal stromal tumors and lymphoma use targeted therapies like imatinib or rituximab. Advanced or metastatic disease is considered incurable and treated mainly with palliative chemotherapy. The median survival with metastatic disease is at best one year; however, the treatment of metastatic colon cancer has had significant advances using multiple agents in sequence, thereby increasing survivals by approximately 10% over the past 20 years (1). Systemic treatments are beyond the scope of this chapter, but readers are referred to the National Comprehensive Cancer Network practice guideline for each of the malignancies covered in this chapter. Neurologic complications from GI malignancies may be direct [i.e., brain metastases (BM)] or indirect (i.e., complications related to chemotherapy). They are varied but the incidence is far less than with some other solid tumors such as breast or lung cancer. For example, gastrointestinal metastases to the central nervous system (CNS) account for approximately 4–6% of solid tumor metastatic brain lesions (2). However, with improved therapies and survivorship, the incidence of metastatic CNS disease from GI malignancies may increase due to poor CNS penetration of most of the agents used for these tumors. As a rule, CNS disease is usually associated with more extensive systemic disease, especially liver and lung metastases. Intracranial metastases from gastrointestinal malignancies do not present distinctly from other solid tumors. CNS invasion is thought to be hematogenous via the arterial circulation or Batson’s plexus, but rarely is direct extension from skull or dural metastases. We review some of the direct and indirect neurologic complications of GI malignancies. These are rare complications but need to be considered by physicians treating such patients. Symptoms from BM are based on the location of the tumor and are attributable to multiple mechanisms, including direct displacement or irritation of brain tissue, vasogenic edema, and disruption of nearby venous and arterial blood flow. The most common presenting symptoms are focal weakness, impaired cognition, and headache with or without nausea and vomiting. Imaging with contrast MRI or CT scans will identify lesions and associated edema; MRI is the modality of choice for optimal resolution of the nervous system, especially the posterior fossa. Acute medical treatment should include corticosteroid therapy such as dexamethasone dosed every 6 to 8 hrs and anti-seizure medications if the patient presents or develops generalized or partial seizures. More definitive therapy is the same as for other solid tumor metastases to the brain. Patients with a solitary lesion benefit from surgical resection followed by whole-brain radiation therapy (WBRT) (3–5); patients with > 4 BM should receive WBRT. There is some debate about the best approach to patients who have 2–3 brain metastases. Chemotherapy may be attempted for patients who fail surgery and radiation, but data on effective therapies using this approach are scarce. Metastatic lesions to the spine, leptomeninges and dura, are rare in gastrointestinal disease and again not treated differently from other solid tumor metastases. Symptoms at presentation may sometimes be different than symptoms from brain lesions. For example, patients with epidural spinal cord compression present with focal back pain and may have weakness, sensory changes or autonomic dysfunction such as urinary retention. Management of these lesions from GI malignancies is again not unique. In the acute setting, corticosteroids (high or low dose depending on clinical picture) should be used when spinal cord is compromised and/or there is neurologic dysfunction or significant back pain. Treatment is radiation therapy in most cases, but surgery may be of benefit as many of the GI malignancies are radioresistant. Surgery may also be used for cases with mechanical instability, bone fragments in the spinal canal, or radiation failures. A recent trial suggested that for some tumor types, surgery followed by radiation therapy was better than radiation alone (6). Treatment of leptomeningeal metastases (LM) is limited to either intrathecal (IT) or systemically administered agents. The available IT agents have limited to no activity against GI malignancies and systemically administered agents have either poor CNS penetration or limited to no activity, hence the poor outcomes for these patients. Radiation should be used to treat symptomatic sites for palliation. Dural lesions can be treated with surgery and/or radiation.

2. ESOPHAGEAL CANCER The estimated number of new cases in 2007 was 15,560 with 13,940 deaths. Most patients present with dysphagia for solids and later liquids. Weight loss is the second most common symptom and occurs in more than 50% of people with esophageal carcinoma.

Chapter 25 / Neurologic Complications of Gastrointestinal Cancer

483

Diagnosis can be made when a mass lesion is seen on imaging or esophagoscopy; the lesion is then biopsied or surgically removed. The typical histologies are squamous cell followed by adenocarcinoma. Some risk factors include smoking, heavy alcohol use, Barrett’s esophagus from gastro-esophageal reflux disease and achalasia. Until the 1970s, squamous cell carcinoma was the most common type of esophageal cancer (90–95%); but since then, the incidence of adenocarcinoma of the distal esophagus and gastroesophageal junction has progressively increased. Currently, it accounts for more than 50% of all new cases of esophageal cancer. The 5-year survival rate for patients with esophageal cancer is 16–32%; left untreated it is on the order of months. Brain Metastases. Weinberg et al. (7) found that 27 out of 1588 (1.7%) of patients with esophageal cancer had BM; three-quarters presented with neurologic symptoms. In an older study, 11 of 343 (3.6%) patients with esophageal cancer who underwent esophagectomies had BM (8). The only risk factor for BM was increased tumor size in association with local invasion and lymph node metastases. Median survival was short at 3.8 months; there was trend for worse outcome in patients with concomitant liver disease and increased age and lower performance status. Ogawa et al. (9) reported on 36 patients with BM over a 14-year period. The median survival was 3.9 months with a 1-year survival of 14%; interestingly, many of these patients were without lung metastases. On multivariate analyses, performance status and type of treatment significantly impacted outcome. Khuntia et al. (10) found a median survival of 3.6 months in a retrospective review of patients with BM; those with higher KPS and aggressive treatment had a significant increase in survival on multivariate analysis. Patients treated aggressively had 1-year survival of 36% compared to 6% for WBRT alone. Leptomeningeal Metastases. In the few cases reported, patients presented with bilateral deafness (11), progressive headache and loss of consciousness (12) and progressive neurologic decline (13). One of the largest series reported on seven cases, all of whom were male. Specific symptoms were not listed but headache, visual symptoms, vertigo, nausea, and vomiting were observed in 33–52% of patients. Median survival was 8 weeks but ranged from 0 to 28 weeks (14). Paraneoplastic Syndromes. One patient with numbness and fever was found to have vasculitis on muscle biopsy; symptoms resolved after esophagectomy (15). Two patients with paraneoplastic cerebellar degeneration (PCD) due to anti-Yo antibody have been reported (16,17). A patient with small cell lung cancer developed a sensorimotor neuropathy three years prior to diagnosis of esophageal cancer (18).

3. GASTRIC CANCER Gastric cancer affects approximately 21,260 patients a year in the United States with 11,210 deaths per year. Rates are higher in certain Asian countries, Chile, and Iceland. Associated causes for gastric cancer include Helicobacter pylori infections; nitrites; diets low in fruits and vegetables and high in salted, smoked, or preserved foods; family history; cigarette smoking; and gastric adenomatous polyps or familial adenomatous polyposis. Early disease has no associated symptoms; however, some patients with incidental complaints are diagnosed with early gastric cancer. Most symptoms of gastric cancer reflect advanced disease where patients may complain of indigestion, nausea or vomiting, dysphagia, postprandial fullness, loss of appetite, and weight loss. Late complications include pathologic peritoneal and pleural effusions; obstruction of the gastric outlet, gastroesophageal junction, or small bowel; and bleeding in from esophageal varices and jaundice. Adenocarcinoma of the stomach comprises 90–95% of all gastric malignancies. Others include lymphomas, leiomyosarcomas, carcinoids, adenoacanthomas, and squamous cell carcinomas. Diagnosis is made by imaging or endoscopy and then biopsy or resection. Early-stage disease accounts for only 10–20% of all cases diagnosed in the United States. Five-year survival is > 90% if detected early but for most patients it is 15–20% because diagnosis occurs when there is not limited stage disease. Brain Metastases. Two reports exist where a patient’s first presentation was neurologic. One case was a patient who presented with headache and dizziness from a cerebellar metastasis, and the other was a patient who presented with seizures and alteration in consciousness (19,20). Another report describes a patient with milliary metastases from small cell gastric cancer, a neuroendocrine tumor with a predilection for the CNS (21). York et al. (22) reviewed 3300 patients with gastric cancer over a 40-year period and found 24 cases with BM. Nineteen of these patients presented with neurologic symptoms, eight before their primary cancer diagnosis.

484

Part VII / Neurologic Complications of Specific Malignancies

The most common symptoms were weakness, headache, gait abnormalities, nausea, and vomiting. Average time to neurologic symptoms from diagnosis was 9 months. Approximately equal numbers of patients had single or multiple metastases with most patients having involvement of the supratentorial compartment. Median survival was only 2.4 months. Leptomeningeal Metastases. In several cases LM was the presenting manifestation of malignancy (19,23–25). Lisenko et al. (26) report on eight cases with an average time to presentation of 12 months after diagnosis of gastric cancer. Presenting symptoms were headache, mental status changes, seizures, and visual problems. Unique features included paucity and low volume of liver metastases, “atypical” sites of metastatic disease, and diffuse nature of dissemination (i.e., skin and meninges). Giglio et al. (14) report on eight cases, the most common symptoms listed in the esophageal cancer section. Median survival ranged from 2 to 38 weeks. In general, these patients had advanced systemic disease. Lee et al. (27) reported on 19 patients with a median survival of 4 weeks (range 3–39) and they note a possible benefit to IT treatment. Four cases of LM from gastric linitis adenocarcinoma (a variant of gastric carcinoma) have been reported out of 80 cases reviewed in a 5-year period. Symptoms were cranial nerve palsies, headache, radiculopathy, ataxia, and dysarthria; three of the four patients had some clinical improvement with treatment with IT MTX (28). Paraneoplastic Syndromes. Among cases reported, two patients with adenocarcinoma had PCD associated with anti-Yo antibody; in one patient titers dropped after resection of the tumor (29,30). Another patient with paraneoplastic cerebellar degeneration due to anti-Ri antibody had a mixed tumor of neuroendocrine (reactive part of tumor) and adenocarcinoma (31). Other cases include a sensorimotor neuropathy and encephalopathy with an antibody to alpha-enolase (32), peripheral neuropathy with arteritis of the sciatic nerve (33), and the opsoclonus–myoclonus syndrome (34).

4. COLORECTAL CANCER There were an estimated 153,760 cases of colorectal cancer diagnosed in 2007 with 52,100 deaths, making colorectal cancer the third most common cancer in both men and women. The incidence and mortality from colorectal cancer has decreased in the last two decades, most likely due to increased screening and improved treatments. Groups that have a high incidence of colorectal cancer include those with hereditary conditions, such as familial polyposis (autosomal dominant), HNPCC or Lynch syndrome variants I and II, and Turcot’s syndrome (autosomal recessive) where patients can have multiple polyps that can degenerate into carcinomas and also have brain tumors (gliomas, ependymomas, medulloblastomas). Genetic syndromes account for 10–15% of colorectal cancers. More common risk factors include a first-degree family history of colorectal cancer or adenomas and a personal history of ovarian, endometrial, or breast cancer and those with a history of ulcerative colitis or Crohn’s disease (i.e., 10 years of ulcerative colitis carries an increased risk of about 1% annually). These high-risk groups account for 23% of all colorectal cancers. Other risk factors include a diet high in red meats, high fat, and low fiber; less related are alcohol intake and obesity. Surgery is the primary therapy for colon and rectal cancers. Stage II/III rectal disease should be treated with chemoradiation. Advanced stage colon cancer requires 5-fluorouracil (FU)-based chemotherapy. The 5-year survival rate is about 60% but drops to less than 10% when metastatic disease is present. Brain Metastases. In a review of patients with BM from Memorial Sloan-Kettering Cancer Center, 14 out of 210 had GI malignancies—8 colon, 4 esophageal, 1 pancreatic, and 1 gastric (35). Another older review found a BM incidence of 0.3–6.5% in patients with colon cancer, often with associated lung or liver disease (36). Cascino et al. (37) evaluated 40 patients with colon cancer who had BM, a 4% incidence of all colon cancer cases seen; almost all patients had advanced disease with poor outcome. Median time to presentation from colon cancer diagnosis was 24 months with a median survival of 9 weeks. Sundermyer et al. (38) reviewed records over a 20-year period and found a 3% incidence of BM out of 1020 patients reviewed. Patients with lung metastases were more likely to develop BM than those without it. A large study of 73 patients with colorectal cancer found that a median interval to develop BM was 27 months (39). Median survival for these patients post craniotomy was 8.3 months, but patients with cerebellar metastases died approximately 4 months earlier (39). Cerebellar lesions occurred in a third of the patients, but this figure may be as high as 55% (40)—much higher than some other solid tumors.

Chapter 25 / Neurologic Complications of Gastrointestinal Cancer

485

Alden et al. (40) identified 19 patients over a 14-year period with metastatic colorectal cancer to the brain. Primary tumor location was distributed throughout the colon; 58% had disseminated disease when first diagnosed. All patients were symptomatic. Lesions were solitary in 63%, unilateral in 89%, and hemispheric in 53%. The median survival was 2.8 months with no patients alive at one year. Schouten et al. (41) reviewed the records of 720 patients with colorectal cancer, of whom only 10 developed BM, during a 10-year period. The one-year incidence of BM was 0.6% but doubling to 1.2% at 5 years. Ko et al. (42) report on 53 cases of BM over a 26-year period. Brain metastases developed a median of 36 months after initial colon surgery. They were more commonly seen in rectal cancer and often occurred concurrently with lung metastases. Forty of these patients received active intervention in terms of surgery, chemotherapy, or radiotherapy, with surgical intervention achieving a significantly increased mean survival time compared with chemotherapy or radiotherapy or both. Hammoud et al. (43) reported on 100 patients diagnosed between 1980 and 1994 with metastatic brain tumors secondary to colorectal adenocarcinoma. The most common primary sites were the sigmoid colon and rectum (65%). Brain metastases with concomitant liver and/or lung metastases were seen more frequently than BM alone. The median interval between the diagnosis of primary cancer and the diagnosis of BM was 26 months. The median survival time after the diagnosis of BM was 1 month for patients who received only steroids, 3 months for those who received radiotherapy, and 9 months for those who underwent surgery. The extent of noncerebral systemic disease was not correlated with survival, but early onset of brain metastasis was significantly associated with poor prognosis. As a rule, most patients with BM also have systemic metastatic disease (44). Spinal Cord Compression (SCC). In one review from Memorial Sloan-Kettering Cancer Center, 9 of 235 patients had SCC from GI malignancies. The low incidence may in part be due the lack of “bone affinity” by these tumors (45). Brown et al. (46) examined the outcome of 34 patients with colorectal cancer who had SCC and who received radiation therapy (RT). The most common symptoms were back pain (97%), weakness and sensory complaints, and radiculopathy. Patients were treated with a median dose of 3000 cGy (1800–4750 cGy). The median overall survival for the entire cohort was 4.1 months, but was significantly better if the primary cancer was rectal (median 7.9 months). Patients who received a total dose of more than 3000 cGy had a better survival (7 months) than those who received 3000 cGy or less (3.1 months). Unlike other primary tumors in which approximately 70% of lesions are located in the thoracic spine, the location of epidural metastasis in this cohort was most frequently in the lumbar spine (55% of lesions). Almost 90% of patients had other sites of metastatic disease (47). Leptomeningeal Metastases. This complication has an incidence of 1%. Giglio et al. (14) report on five cases, 80% of whom had other sites of disease. Median survival was 5 weeks, ranging from 0 to 14 weeks, after diagnosis of LM. Other cases are also reported (48,49). Paraneoplastic Syndromes. One reported patient had limbic encephalitis and PCD with high titers of an antiHu–like antibody; the patient’s encephalitis improved after removal of the colonic polyp but not the PCD (50). One patient had retinal anti-bipolar antibodies and was ultimately found to have adenocarcinoma of the colon, after resection of the tumor no antibodies could be detected and the electroretinography response normalized (51). Two patients with sensorimotor neuropathy have been reported; one was found to have rectal cancer, after resection and chemotherapy, the neuropathy resolved (52,53).

5. HEPATOCELLULAR CARCINOMA (HCC) The incidence of hepatic or intrahepatic cancer is about 19,000 cases per year with about 17,000 deaths. There is a higher incidence in males and also in Asia compared to the United States. Both hepatitis B and C infection can lead to HCC. Hepatocellular carcinoma is associated with cirrhosis in 50–80% of patients; 5% of cirrhotic patients eventually develop HCC. Hepatitis B and C infection appear to be the most significant causes of HCC worldwide; this risk may be increased when patients also consume large amounts of alcohol. Aflatoxin has also been implicated as an etiologic factor for primary liver cancer in parts of the world where this mycotoxin occurs in high levels in ingested food. The primary symptoms of HCC are those of a hepatic mass. Among patients with underlying cirrhotic disease, a progressive increase in alpha-fetoprotein (AFP) and/or in alkaline phosphatase or a rapid deterioration of hepatic

486

Part VII / Neurologic Complications of Specific Malignancies

function may be the only clue to the presence of the neoplasm. Infrequently, patients with this disease have polycythemia, hypoglycemia, hypercalcemia, or dysfibrinogenemia. Surgery for local disease yields a 5-year survival of 60–70% but only a minority of patients will fall into this category; when there is local or disseminated spread, outcomes are poor irrespective of treatment. Brain Metastases. In a review of 403 patients with HCC at the University of Pittsburgh Medical Center, 0.74% had BM (54). Kim et al. (55) report a 0.2% incidence of BM out of 3100 Korean patients with HCC. Four of these patients had a stroke-like presentation; median time to presentation was 15.3 months. A study of 45 patients from Taiwan found a median time to presentation of 10.5 months after initial diagnosis. In these patients, 18 presented with intracranial hemorrhage with signs and symptoms dependent on location of tumor(s). Survival was short, less than 4 weeks without treatment but 4 or more with treatment. A radiographic study from Japan reviewed 16 cases; 8 had an acute onset of neurologic symptoms (56). Over 80% had associated pulmonary metastases and median survival from BM was 6 weeks. Fourteen of the patients had hemorrhagic BM which was correlated to lesion size. Most BM are supratentorial; they may be single or multiple and hemorrhagic as noted above (55–57). Often, other sites of disease are present but the brain may be the initial site of disease (55,57). Skull metastases have also been reported (55,57,58). Paraneoplastic Syndromes. Several cases of a demyelinating neuropathy have been reported (59–61). Patients with cancer-associated retinopathy and polymyositis that resolved after surgical removal of the tumor have been reported (62,63).

6. GALLBLADDER AND BILE DUCT (CHOLANGIOCARCINOMA) CARCINOMAS These malignancies are rarer than other GI tumors. There are about 9,250 cases per year with 3,250 deaths. The most common symptoms caused by gallbladder cancer are jaundice, pain, and fever. When cancer is discovered in the mucosa of the gallbladder at pathologic examination, it is curable in more than 80% of cases; however, when there is penetration of the muscularis and serosa it is curable in fewer than 5% of patients. Cancer arising in the extrahepatic bile duct is uncommon, curable by surgery in fewer than 10% of all cases. Bile duct cancer may occur more frequently in patients with a history of primary sclerosing cholangitis, chronic ulcerative colitis, choledochal cysts, or infections with the fluke Clonorchis sinensis. The most common symptoms caused by bile duct cancer are jaundice, pain, fever, and pruritus. Brain Metastases. This is a rare complication with a single report for each tumor, one with hydrocephalus due to a cerebellar metastasis (64,65). Leptomeningeal Metastasis. Although this complication is in general rarer than BM, there are more cases of LM reported for these tumor types. One patient with gallbladder cancer presented with psychosis (66), others presented with headaches and cranial neuropathies (67,68) and meningitic picture (69). One patient with cholangiocarcinoma and LM is reported (70). Paraneoplastic Syndromes. One case of Guillain–Barré syndrome that may have been paraneoplastic has been reported as well as one case of opsoclonus in a patient with gallbladder cancer (71,72).

7. PANCREATIC CANCER Pancreatic cancer has an incidence of 37,000 case per year and nearly the same number of deaths. It is second to colorectal cancer in incidence. High fat diet is a risk factor for pancreatic cancer, whereas diets high in fiber, fresh fruits, and vegetables have an inverse relationship with developing pancreatic cancer. Diagnosis is often late in the course of the illness and is associated with weight loss, abdominal pain, jaundice, and pruritis. Depression and venous thrombosis can herald the disease. Psychiatric symptoms are higher in patients with pancreatic cancers than other GI malignancies (73). Approximately 80% of patients present with locally advanced or metastatic disease, leaving 20% eligible for surgery, which has the best chance for cure. Median survival is 12–18 months for resectable patients but 6 months if unresectable. Treatment usually consists of radiation and chemotherapy. Brain Metastases. These are rare and limited to a few case reports involving the brain or skull, in one case the BM was found on autopsy (58,74,75). Park et al. (76) reviewed 1229 case of pancreatic cancer, of whom only

Chapter 25 / Neurologic Complications of Gastrointestinal Cancer

487

four patients had BM. In two patients this was the initial manifestation and other two presented 4 and 5 months after diagnosis; median survival was less than 3 months. All patients had other sites of metastatic disease. Leptomeningeal Metastases. Giglio et al. (14) report on one patient with this complication and note only seven other reported cases. Most patients had other sites of metastases with survival ranging from weeks up to 1 year. Paraneoplastic Syndromes. One patient with encephalomyelitis was found to have anti-GAD antibodies and another with small cell pancreatic cancer who presented with PCD and later polyneuropathy had anti-Hu antibodies (77,78).

8. PERIPHERAL NERVOUS SYSTEM COMPLICATIONS A rare complication of colorectal cancer is lumbosacral plexopathy. This may be a direct effect of the tumor or a secondary complication of radiation therapy. Direct compression from tumor causes back and leg pain followed by sensory changes and weakness. In a review of 85 patients with lumbosacral plexopathy, 17 had colon and 2 gastric cancer (79). Symptoms differed slightly between colon and rectal cancers. Colon cancer produced radicular pain down the posterior aspect of the leg from lower plexus compression, while rectal cancer was associated with perineal sensory changes from coccygeal plexus involvement. Patients did poorly with a median survival of 5 months from diagnosis of plexopathy. Radiation-induced lumbosacral plexopathy is characterized by painless weakness that progresses and ultimately stabilizes with a fixed deficit. Myokymia may be seen on EMG and provide a clue that neurologic dysfunction is a complication of radiation.

9. METABOLIC ABNORMALITIES Volume depletion from vomiting or diarrhea results in the secretion of antidiuretic hormone (ADH), which when coupled with free water intake may lead to hypo-osmotic hyponatremia. Low serum sodium levels can manifest symptomatically as lethargy, confusion, seizures, or even coma. Continued volume depletion can further lead to deceased renal perfusion with hypokalemia and azotemia. Severe and persistent emesis can lead to hypokalemic metabolic alkalosis. Hypo- and hyperkalemia can present as muscle weakness while uremia can result in mental status changes. McKittrick–Wheelock syndrome is the constellation of dehydration, hyponatremia, hypokalemia, and azotemia that is directly associated with malignancies in the rectum, most commonly a villous adenoma, although rectal adenocarcinoma has also been implicated (80). Treatment of volume depletion and electrolyte disorders is supportive and often can be managed with isotonic fluids. Electrolyte replacement should be done with care; it is recommended that the correction rate of serum sodium not exceed 10–12 mEq/L per day in order to prevent osmotic demyelination. Treatment for the underlying cause of the electrolyte imbalance may require surgical intervention to relieve a small bowel obstruction or resect the rectal tumor as the case of McKittrick–Wheelock syndrome. The administration of octreotide or sandostatin LAR can be very helpful in reducing the diarrhea associated with carcinoid syndrome and VIPomas. Some of the physiologic changes that occur with gastric cancer are related to surgery (81). One complication is a “gastric dumping” syndrome, where there is a delay in the transportation of food into the small intestine due to loss of a functional pylorus. A second complication is iron and B12 deficiency, the latter due to the loss of intrinsic factor which can cause pernicious anemia, peripheral neuropathy, and subacute combined degeneration of the spinal cord. This has a delayed onset and patients have loss of proprioception and vibration, ataxia, and loss of deep tendon reflexes. Treatment is with parenteral vitamin B12 replacement.

10. CHEMOTHERAPY-RELATED NEUROLOGIC COMPLICATIONS There is a lot of overlap between the types of chemotherapy used to treat gastrointestinal malignancies and other cancers; the agents most commonly used are covered.

488

Part VII / Neurologic Complications of Specific Malignancies

10.1. 5-Fluorouracil (5-FU) Intravenous fluorouracil (5-FU) can rarely be associated with acute and chronic neurotoxicities. The acute toxicities have two clinical presentations: the acute cerebellar syndrome characterized by ataxia, confusion, drowsiness, disorientation, euphoria, headache, nystagmus and visual disturbances or an encephalopathy with the notable biochemical changes: hyperammonemia and lactic acidosis (41). These toxicities usually develop shortly after therapy and persist for 48–72 hrs after therapy has stopped (Fig. 1). Dihydropyrimidine dehydrogenase (DPD), which is necessary in clearing 5-FU, is deficient in 2.4% of cancer patients; its absence has been linked to an increase in neurotoxicity (82). In early studies combining levamisole and 5-FU in the treatment of metastatic colorectal cancer, some patients developed a subacute multifocal leukoencephalopathy manifested as cognitive abnormalities, disturbances of consciousness, dysarthria, focal extremity weakness, and gait and limb ataxia 3–5 months post-therapy (83, 84). Brain MRI revealed multifocal enhancing white matter lesions, which were both supra and infratentorial. Discontinuation of levamisole generally results in improvement. Other reported side effects include ophthalmoplegia, optic neuropathy, encephalopathy, focal dystonias, and parkinsonism (47).

10.2. Bevacizumab Bevacizumab is a monoclonal antibody against VEGF. It has been associated with reversible posterior leukoencephalopathy syndrome (RPLS), which may present with varied neurologic symptoms including headaches, seizures, lethargy, confusion, blindness, or other visual disturbances. Hypertension may precede the symptoms but is not necessary for diagnosis. Magnetic resonance imaging is used to confirm the diagnosis based on characteristic findings. The incidence is less than 0.1% (85). Bevacizumab increases the risk of arterial thromboembolic events including stroke, transient ischemic attacks and myocardial infarctions. Although less common than venous thrombotic disease in general, the morbidity associated with arterial events can be quite significant. The bevacizumab study AVF2107g, reported 13 (3.3%) events compared to 5 (1.3%) when treated with and without bevacizumab (86). Similarly, AVF2192g reported an absolute doubling of the rate of arterial thrombotic

Fig. 1. Leukoencephalopathy in a woman with colon adenocarcinoma treated with 5-FU and leucovorin. She presented with ataxia and confusion. She was found later to have profound DPD deficiency with no detectable enzyme activity. (A) T2-weighted axial MRI shows hyperintensity in periventricular white matter. (B) Coronal MRI shows similar changes, as well as involvement of the deeper white matter and corticospinal tracts sparing U-fibers. (Figures courtesy of Robert Diasio, M.D., Director, Mayo Clinic Comprehensive Cancer Center, Rochester MN).

Chapter 25 / Neurologic Complications of Gastrointestinal Cancer

489

events when bevacizumab was used (10% vs. 4.8%). Of note, the study population had median age of 70 years (87). In practice, clinicians must exercise caution in prescribing bevacizumab for patients with risk factors for or with known vascular disease. Intratumoral bleeding is another side effect that may occur in tumors; for intracranial tumors this could be fatal.

10.3. Oxaliplatin Oxaliplatin is a third-generation platinum compound that causes acute and chronic peripheral neuropathies. The acute neurotoxicity may occur during, shortly after, or 1–2 days post-infusion of the drug and is associated with paresthesias, hypesthesias, and dysesthesias. These usually begin in the hands or feet, but may occur around the mouth or in the throat as well. Acute side effects occur at a dose of about 130 mg/m2 (88). Patients may also have a sense of dyspnea or dysphagia without bronchospasm, wheezing, stridor, or laryngospasm. Patients have described an unusual sensation in the tongue, jaw spasms, eye pain, and muscle spasms or cramps, which are sometimes described as stiffness in the hands or feet or an inability to release grip. Cold temperature may exacerbate symptoms and patients are educated to avoid cold drinks, wear gloves when handling refrigerated items, and avoid inhaling cold air. Symptoms usually last only a few days post-therapy (89). One suggestion for the mechanism by which oxaliplatin causes an acute neurotoxicity has been coined “channelopathy.” Oxaliplatin has been shown to be associated with the prolonged opening of sodium-gated channels on peripheral nerves that leads to hyperexcitability (90–92). Whether this is a direct effect or not is unclear but may be related to the sequestration of calcium by the oxaliplatin–oxalate metabolite. There is little published data on how to prevent and treat the acute neurotoxicity associated with oxaliplatin. Using a lower dose or increasing the infusion time has been thought to lessen the occurrence of these symptoms (93). Administering calcium and magnesium salts like calcium gluconate and magnesium chloride has reportedly decreased the occurrence of pharyngolaryngeal dysesthesia (1.6% vs. 26%) (88,94). Amifostine has also been studied as a preventative measure in reducing acute oxaliplatin-induced neuropathy. Patients receiving oxaliplatin, 5-FU, and leucovorin (a common first- or second-line therapy for metastatic colon cancer) in addition to amifostine reported less cold-induced sensitivity. However, there are significant toxicities associated with the administration of intravenous amifostine including hypotension, nausea, and vomiting that may limit its practical use; therefore, a subcutaneous preparation is recommended (95). Symptoms associated with a more prolonged administration of oxaliplatin (total doses of ≥540–850 mg/m2 ) include paresthesia, hypesthesia, dysesthesia, and changes in propioception that do not resolve between cycles. Proprioceptive dysfunction may present with difficulties in fine motor coordination required for writing, holding objects, picking up coins, and buttoning shirts. The chronic neuropathy is cumulative with a reported incidence of grade 3 toxicity occurring 10% of the time after nine cycles and roughly 50% of the time at 12–14 cycles (based on oxaliplatin doses of 85 mg/m2 infused over 2 hrs every 14 days) (96,97). Lhermitte’s phenomenon, an electric sensation experienced with neck flexion, has also been reported as a manifestation of chronic oxaliplatin-induced neuropathy (98). Other central neuropathic symptoms such as urinary retention have also been reported less commonly. Symptoms usually last months with most resolving completely or to grade 1–2 toxicity within 12 months (99). Rare symptoms include optic neuritis and visual field deficits (100). Preventive strategies such as those outlined above (e.g., longer infusion time, administration of calcium or magnesium salts) have some reported efficacy in helping to prevent or at least minimize the chronic neuropathic effects of oxaliplatin. Gabapentin has also been shown to reduce the acute neuropathic toxicity but also prevents the chronic form as well. In one very small study, 10 patients were treated with 100 mg daily of gabapentin and increased to 300 mg daily if symptoms persisted for 3 days. Patients who continued on this medication for all cycles (up to 14) had no reports of neuropathy (101).

10.4. Capecitabine Capecitabine is a prodrug metabolized to 5-FU by thymidine phosphorylase. Neurologic toxicity is rare and limited to one case of peripheral neuropathy and several cases of encephalopathy. The latter is different from that seen with 5-FU/levamisole, as this is a reversible process with diffusion-restricted changes that do not enhance on brain MRI. This process starts earlier than that of 5-FU/levamisole (102,103). The erythpalmar dysesthia may mimic symptoms of neuropathy or give a sense that an underlying neuropathy is worsening.

490

Part VII / Neurologic Complications of Specific Malignancies

10.5. Gemcitabine Gemcitabine is a deoxycytidine analogue with minimal CNS effects. About 1% of patients complain of mild paresthesias and rare autonomic neuropathy is reported (104). It may increase neurotoxicity when given after WBRT (105).

10.6. Irinotecan Irinotecan is a topoisomerase inhibitor with no major neurologic toxicity except for visual changes, dizziness and drowsiness.

11. CONCLUSION Neurologic complications from gastrointestinal malignancies are relatively rare compared to other solid tumors such as breast or lung cancer. Nonetheless, they do occur and need to be part of the differential diagnosis in the evaluation of patients with these specific malignancies. Most commonly, these malignancies will cause direct effects, such as brain metastases, but indirect effects of these tumors and chemotherapy can occur.

REFERENCES 1. Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA Cancer J Clin 2007; 57(1):43–66. 2. Newton HB. Neurologic complications of systemic cancer. Am Fam Physician 1999; 59(4):878–886. 3. Patchell RA, Tibbs PA, Regine WF et al. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 1998; 280(17):1485–1489. 4. Patchell RA, Tibbs PA, Walsh JW et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990; 322(8):494–500. 5. Vecht CJ, Haaxma-Reiche H, Noordijk EM et al. Treatment of single-brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993; 33(6):583–590. 6. Patchell RA, Tibbs PA, Regine WF et al. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Lancet 2005; 366(9486):643–648. 7. Weinberg JS, Lang FF, Sawaya R. Surgical management of brain metastases. Curr Oncol Rep 2001; 3(6):476–483. 8. Gabrielsen TO, Eldevik OP, Orringer MB et al. Esophageal carcinoma metastatic to the brain: clinical value and cost-effectiveness of routine enhanced head CT before esophagectomy. AJNR Am J Neuroradiol 1995; 16(9):1915–1921. 9. Ogawa K, Toita T, Sueyama H et al. Brain metastases from esophageal carcinoma: natural history, prognostic factors, and outcome. Cancer 2002; 94(3):759–764. 10. Khuntia D, Sajja R, Chidel MA et al. Factors associated with improved survival in patients with brain metastases from esophageal cancer: a retrospective review. Technol Cancer Res Treat 2003; 2(3):267–272. 11. Wagemakers M, Verhagen W, Borne B et al. Bilateral profound hearing loss due to meningeal carcinomatosis. J Clin Neurosci 2005; 12(3):315–318. 12. Abdo AA, Coderre S, Bridges RJ. Leptomeningeal carcinomatosis secondary to gastroesophageal adenocarcinoma: a case report and review of the literature. Can J Gastroenterol 2002; 16(11):807–811. 13. Giglio P, Tremont-Lukats IW, Groves MD. Response of neoplastic meningitis from solid tumors to oral capecitabine. J Neurooncol 2003; 65(2):167–172. 14. Giglio P, Weinberg JS, Forman AD et al. Neoplastic meningitis in patients with adenocarcinoma of the gastrointestinal tract. Cancer 2005; 103(11):2355–2362. 15. Mita T, Nakanishi Y, Ochiai A et al. Paraneoplastic vasculitis associated with esophageal carcinoma. Pathol Int 1999; 49(7):643–647. 16. Xia K, Saltzman JR, Carr-Locke DL. Anti-Yo antibody-mediated paraneoplastic cerebellar degeneration in a man with esophageal adenocarcinoma. Med Gen Med 2003; 5(3):18. 17. Sutton IJ, Fursdon Davis CJ, Esiri MM et al. Anti-Yo antibodies and cerebellar degeneration in a man with adenocarcinoma of the esophagus. Ann Neurol 2001; 49(2):253–257. 18. Shimoda T, Koizumi W, Tanabe S et al. Small cell carcinoma of the esophagus associated with a paraneoplastic neurological syndrome: a case report documenting a complete response. Jpn J Clin Oncol 2006; 36(2):109–112. 19. Bredin C, Terris B, Sogni P et al [Carcinomatous meningitis as a rare presentation of gastric cancer]. Presse Med 2005; 34(7):509–510. 20. Nomura T, Yoshikawa T, Kato H et al. Early gastric cancer manifested as brain metastasis: report of a case. Surg Today 1997; 27(4):334–336. 21. Bugalho P, Chorao M, Fontoura P. Miliary brain metastases from primary gastric small cell carcinoma: illustrating the seed and soil hypothesis. J Neurooncol 2005; 73(1):53–56. 22. York JE, Stringer J, Ajani JA et al. Gastric cancer and metastasis to the brain. Ann Surg Oncol 1999; 6(8):771–776. 23. Braeuninger S, Mawrin C, Malfertheiner P et al. Gastric adenocarcinoma with leptomeningeal carcinomatosis as the presenting manifestation: an autopsy case report. Eur J Gastroenterol Hepatol 2005; 17(5):577–579. 24. Deeb LS, Yamout BI, Shamseddine AI et al. Meningeal carcinomatosis as the presenting manifestation of gastric adenocarcinoma. Am J Gastroenterol 1997; 92(2):329–331.

Chapter 25 / Neurologic Complications of Gastrointestinal Cancer

491

25. Grove A. Meningeal carcinomatosis from a clinically undiagnosed early gastric cancer. Pathol Res Pract 1991; 187(2–3):341–345. 26. Lisenko Y, Kumar AJ, Yao J et al. Leptomeningeal carcinomatosis originating from gastric cancer: report of eight cases and review of the literature. Am J Clin Oncol 2003; 26(2):165–170. 27. Lee JL, Kang YK, Kim TW et al. Leptomeningeal carcinomatosis in gastric cancer. J Neurooncol 2004; 66(1–2):167–174. 28. Delaunoit T, Boige V, Belloc J et al. Gastric linitis adenocarcinoma and carcinomatous meningitis: an infrequent but aggressive association: report of four cases. Ann Oncol 2001; 12(6):869–871. 29. Meglic B, Graus F, Grad A. Anti-Yo-associated paraneoplastic cerebellar degeneration in a man with gastric adenocarcinoma. J Neurol Sci 2001; 185(2):135–138. 30. Goto A, Kusumi M, Wakutani Y et al. [Anti-Yo antibody associated paraneoplastic cerebellar degeneration with gastric adenocarcinoma in a male patient: a case report]. Rinsho Shinkeigaku 2006; 46(2):144–147. 31. Kikuchi H, Yamada T, Okayama A et al. Anti-Ri–associated paraneoplastic cerebellar degeneration without opsoclonus in a patient with a neuroendocrine carcinoma of the stomach. Fukuoka Igaku Zasshi 2000; 91(4):104–109. 32. Tojo K, Tokuda T, Yazaki M et al. Paraneoplastic sensorimotor neuropathy and encephalopathy associated with anti-alpha-enolase antibody in a case of gastric adenocarcinoma. Eur Neurol 2004; 51(4):231–233. 33. Naka T, Yorifuji S, Fujimura H et al. [A case of paraneoplastic neuropathy with necrotizing arteritis localized in the peripheral nervous system]. Rinsho Shinkeigaku 1991; 31(4):427–432. 34. Bataller L, Graus F, Saiz A et al. Clinical outcome in adult onset idiopathic or paraneoplastic opsoclonus–myoclonus. Brain 2001; 124(Pt 2):437–443. 35. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol 1978; 19:579–592. 36. Rovirosa A, Bodi R, Vicente P et al. [Cerebral metastases in adenocarcinoma of the colon]. Rev Esp Enferm Dig 1991; 79(4):281–283. 37. Cascino TL, Leavengood JM, Kemeny N et al. Brain metastases from colon cancer. J Neurooncol 1983; 1(3):203–209. 38. Sundermeyer ML, Meropol NJ, Rogatko A et al. Changing patterns of bone and brain metastases in patients with colorectal cancer. Clin Colorectal Cancer 2005; 5(2):108–113. 39. Wronski M, Arbit E. Resection of brain metastases from colorectal carcinoma in 73 patients. Cancer 1999; 85(8):1677–1685. 40. Alden TD, Gianino JW, Saclarides TJ. Brain metastases from colorectal cancer. Dis Colon Rectum 1996; 39(5):541–545. 41. Schouten LJ, Rutten J, Huveneers HA et al. Incidence of brain metastases in a cohort of patients with carcinoma of the breast, colon, kidney, and lung and melanoma. Cancer 2002; 94(10):2698–2705. 42. Ko FC, Liu JM, Chen WS et al. Risk and patterns of brain metastases in colorectal cancer: 27-year experience. Dis Colon Rectum 1999; 42(11):1467–1471. 43. Hammoud MA, McCutcheon IE, Elsouki R et al. Colorectal carcinoma and brain metastasis: distribution, treatment, and survival. Ann Surg Oncol 1996; 3(5):453–463. 44. Amichetti M, Lay G, Dessi M et al. Results of whole-brain radiation therapy in patients with brain metastases from colorectal carcinoma. Tumori 2005; 91(2):163–167. 45. Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 1978; 3(1):40–51. 46. Brown PD, Stafford SL, Schild SE et al. Metastatic spinal cord compression in patients with colorectal cancer. J Neurooncol 1999; 44(2):175–180. 47. Posner JB. Neurologic Complications of Cancer. Philadelphia: F.A. Davis; 1995. 48. Kato H, Emura S, Takashima T et al. Gadolinium-enhanced magnetic resonance imaging of meningeal carcinomatosis in colon cancer. Tohoku J Exp Med 1995; 176(2):121–126. 49. Fisher MA, Weiss RB. Carcinomatous meningitis in gastrointestinal malignancies. South Med J 1979; 72(8):930–932. 50. Tsukamoto T, Mochizuki R, Mochizuki H et al. Paraneoplastic cerebellar degeneration and limbic encephalitis in a patient with adenocarcinoma of the colon. J Neurol Neurosurg Psychiatry 1993; 56(6):713–716. 51. Jacobson DM, Adamus G. Retinal anti-bipolar cell antibodies in a patient with paraneoplastic retinopathy and colon carcinoma. Am J Ophthalmol 2001; 131(6):806–808. 52. Pantalone D, Muscas GC, Tings T et al. Peripheral paraneoplastic neuropathy, an uncommon clinical onset of sigmoid cance: case report and review of the literature. Tumori 2002; 88(4):347–349. 53. Sharma BC, Ghoshal UC, Saraswat VA. Carcinoma colon presenting as paraneoplastic sensorimotor neuropathy. J Assoc Physicians India 1998; 46(2):239. 54. Katyal S, Oliver JH, III, Peterson MS et al. Extrahepatic metastases of hepatocellular carcinoma. Radiology 2000; 216(3):698–703. 55. Kim M, Na DL, Park SH et al. Nervous system involvement by metastatic hepatocellular carcinoma. J Neurooncol 1998; 36(1):85–90. 56. Murakami K, Nawano S, Moriyama N et al. Intracranial metastases of hepatocellular carcinoma: CT and MRI. Neuroradiology 1996; 38 Suppl 1:S31–S35. 57. Chang L, Chen YL, Kao MC. Intracranial metastasis of hepatocellular carcinoma: review of 45 cases. Surg Neurol 2004; 62(2):172–177. 58. Kuratsu J, Murakami M, Uemura S et al. Brain and skull metastases of hepatic or pancreatic cancer: report of six cases. Neurol Med Chir (Tokyo) 1990; 30(7):476–482. 59. Walcher J, Witter T, Rupprecht HD. Hepatocellular carcinoma presenting with paraneoplastic demyelinating polyneuropathy and PR3-antineutrophil cytoplasmic antibody. J Clin Gastroenterol 2002; 35(4):364–365. 60. Sugai F, Abe K, Fujimoto T et al. Chronic inflammatory demyelinating polyneuropathy accompanied by hepatocellular carcinoma. Intern Med 1997; 36(1):53–55. 61. Arguedas MR, McGuire BM. Hepatocellular carcinoma presenting with chronic inflammatory demyelinating polyradiculoneuropathy. Dig Dis Sci 2000; 45(12):2369–2373. 62. Chang PY, Yang CH, Yang CM. Cancer-associated retinopathy in a patient with hepatocellular carcinoma: case report and literature review. Retina 2005; 25(8):1093–1096.

492

Part VII / Neurologic Complications of Specific Malignancies

63. Hasegawa K, Uesugi H, Kubota K et al. Polymyositis as a paraneoplastic manifestation of hepatocellular carcinoma. Hepatogastroenterology 2000; 47(35):1425–1427. 64. Gudesblatt MS, Sencer W, Sacher M et al. Cholangiocarcinoma presenting as a cerebellar metastasis: case report and review of the literature. J Comput Tomogr 1984; 8(3):191–195. 65. Takano S, Yoshii Y, Owada T et al. Central nervous system metastasis from gallbladder carcinoma: case report. Neurol Med Chir (Tokyo) 1991; 31(12):782–786. 66. Higes-Pascual F, Beroiz-Groh P, Bravo-Guillen AI et al. [Leptomeningeal carcinomatosis as presenting symptom of a gallbladder carcinoma]. Rev Neurol 2000; 30(9):841–844. 67. Gaumann A, Marx J, Bohl J et al. Leptomeningeal carcinomatosis and cranial nerve palsy as presenting symptoms of a clinically inapparent gallbladder carcinoma. Pathol Res Pract 1999; 195(7):495–499. 68. Tans RJ, Koudstaal J, Koehler PJ. Meningeal carcinomatosis as presenting symptom of a gallbladder carcinoma. Clin Neurol Neurosurg 1993; 95(3):253–256. 69. Miyagui T, Luchemback L, Teixeira GH et al. Meningeal carcinomatosis as the initial manifestation of a gallbladder adenocarcinoma associated with a Krukenberg tumor. Rev Hosp Clin Fac Med Sao Paulo 2003; 58(3):169–172. 70. Huffman JL, Yeatman TJ, Smith JB. Leptomeningeal carcinomatosis: a sequela of cholangiocarcinoma. Am Surg 1997; 63(4):310–313. 71. Corcia P, de Toffel B, Hommet C et al. Paraneoplastic opsoclonus associated with cancer of the gallbladder. J Neurol Neurosurg Psychiatry 1997; 62(3):293. 72. Phan TG, Hersch M, Zagami AS. Guillain–Barré syndrome and adenocarcinoma of the gallbladder: a paraneoplastic phenomenon? Muscle Nerve 1999; 22(1):141–142. 73. Holland JC, Korzun AH, Tross S et al. Comparative psychological disturbance in patients with pancreatic and gastric cancer. Am J Psychiatry 1986; 143(8):982–986. 74. El Kamar FG, Jindal K, Grossbard ML et al. Pancreatic carcinoma with brain metastases: case report and literature review. Dig Liver Dis 2004; 36(5):355–360. 75. Yamada K, Miura M, Miyayama H et al. Brain metastases from asymptomatic adenocarcinoma of the pancreas: an autopsy case report. Surg Neurol 2002; 58(5):332–336. 76. Park KS, Kim M, Park SH et al. Nervous system involvement by pancreatic cancer. J Neurooncol 2003; 63(3):313–316. 77. Hernandez-Echebarria L, Saiz A, Ares A et al. Paraneoplastic encephalomyelitis associated with pancreatic tumor and anti-GAD antibodies. Neurology 2006; 66(3):450–451. 78. Salmeron-Ato P, Medrano V, Morales-Ortiz A et al. [Paraneoplastic cerebellar degeneration as initial presentation of a pancreatic small cell carcinoma]. Rev Neurol 2002; 35(12):1112–1115. 79. Jaeckle KA, Young DF, Foley KM. The natural history of lumbosacral plexopathy in cancer. Neurology 1985; 35(1):8–15. 80. Lepur D, Klinar I, Mise B et al. McKittrick–Wheelock syndrome: a rare cause of diarrhoea. Eur J Gastroenterol Hepatol 2006; 18(5):557–559. 81. Williams JA, Hall GS, Thompson AG et al. Neurological disease after partial gastrectomy. Br Med J 1969; 3(5664):210–212. 82. Pirzada NA, Ali II, Dafer RM. Fluorouracil-induced neurotoxicity. Ann Pharmacother 2000; 34(1):35–38. 83. Chen TC, Hinton DR, Leichman L et al. Multifocal inflammatory leukoencephalopathy associated with levamisole and 5-fluorouracil: case report. Neurosurgery 1994; 35(6):1138–1142. 84. Hook CC, Kimmel DW, Kvols LK et al. Multifocal inflammatory leukoencephalopathy with 5-fluorouracil and levamisole. Ann Neurol 1992; 31(3):262–267. 85. Glusker P, Recht L, Lane B. Reversible posterior leukoencephalopathy syndrome and bevacizumab. N Engl J Med 2006; 354(9): 980–982. 86. Novotny WF, Holmgren E, Nelson B et al. Bevacizumab (a monoclonal antibody to vascular endothelial growth factor) does not increase the incidence of venous thromboembolism when added to first-line chemotherapy to treat metastatic colorectal cancer. J Clin Oncol ASCO Annual Meeting Proceedings 22[14S]. 2004. 87. Kabbinavar FF, Schulz J, McCleod M et al. Addition of bevacizumab to bolus fluorouracil and leucovorin in first-line metastatic colorectal cancer: results of a randomized phase II trial. J Clin Oncol 2005; 23(16):3697–3705. 88. Gamelin E, Gamelin L, Bossi L et al. Clinical aspects and molecular basis of oxaliplatin neurotoxicity: current management and development of preventive measures. Semin Oncol 2002; 29(5 Suppl 15):21–33. 89. Cersosimo RJ. Oxaliplatin-associated neuropathy: a review. Ann Pharmacother 2005; 39(1):128–135. 90. Grolleau F, Gamelin L, Boisdron-Celle M et al. A possible explanation for a neurotoxic effect of the anticancer agent oxaliplatin on neuronal voltage–gated sodium channels. J Neurophysiol 2001; 85(5):2293–2297. 91. Lehky TJ, Leonard GD, Wilson RH et al. Oxaliplatin-induced neurotoxicity: acute hyperexcitability and chronic neuropathy. Muscle Nerve 2004; 29(3):387–392. 92. Adelsberger H, Quasthoff S, Grosskreutz J et al. The chemotherapeutic oxaliplatin alters voltage-gated Na(+) channel kinetics on rat sensory neurons. Eur J Pharmacol 2000; 406(1):25–32. 93. Screnci D, McKeage MJ. Platinum neurotoxicity: clinical profiles, experimental models and neuroprotective approaches. J Inorg Biochem 1999; 77(1–2):105–110. 94. Gamelin L, Boisdron-Celle M, Delva R et al. Prevention of oxaliplatin-related neurotoxicity by calcium and magnesium infusions: a retrospective study of 161 patients receiving oxaliplatin combined with 5-Fluorouracil and leucovorin for advanced colorectal cancer. Clin Cancer Res 2004; 10(12 Pt 1):4055–4061. 95. Koukourakis MI, Simopoulos C, Minopoulos G et al. Amifostine before chemotherapy: improved tolerance profile of the subcutaneous over the intravenous route. Clin Cancer Res 2003; 9(9):3288–3293. 96. de GA, Figer A, Seymour M et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18(16):2938–2947.

Chapter 25 / Neurologic Complications of Gastrointestinal Cancer

493

97. Souglakos J, Mavroudis D, Kakolyris S et al. Triplet combination with irinotecan plus oxaliplatin plus continuous-infusion fluorouracil and leucovorin as first-line treatment in metastatic colorectal cancer: a multicenter phase II trial. J Clin Oncol 2002; 20(11):2651–2657. 98. Taieb S, Trillet-Lenoir V, Rambaud L et al. Lhermitte sign and urinary retention: atypical presentation of oxaliplatin neurotoxicity in four patients. Cancer 2002; 94(9):2434–2440. 99. Andre T, Boni C, Mounedji-Boudiaf L et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med 2004; 350(23):2343–2351. 100. Sul JK, Deangelis LM. Neurologic complications of cancer chemotherapy. Semin Oncol 2006; 33(3):324–332. 101. Mariani G, Garrone O, Granetto C et al. Oxaliplatin-induced neuropathy: could gabapentin be the answer? Proc Am Soc Clin Oncol 19. 2000. 102. Videnovic A, Semenov I, Chua-Adajar R et al. Capecitabine-induced multifocal leukoencephalopathy: a report of five cases. Neurology 2005; 65(11):1792–1794. 103. Saif MW, Wood TE, McGee PJ et al. Peripheral neuropathy associated with capecitabine. Anticancer Drugs 2004; 15(8):767–771. 104. Dormann AJ, Grunewald T, Wigginghaus B et al. Gemcitabine-associated autonomic neuropathy. Lancet 1998; 351(9103):644. 105. Jeter MD, Janne PA, Brooks S et al. Gemcitabine-induced radiation recall. Int J Radiat Oncol Biol Phys 2002; 53(2):394–400.

26

Neurologic Complications of Sarcoma Santosh Kesari,

MD, PHD,

and Lara J. Kunschner,

MD

CONTENTS Introduction Chondrosarcoma Malignant Fibrous Histiocytoma Hemangiopericytoma Rhabdomyosarcoma Leiomyosarcoma Malignant Peripheral Nerve Sheath Tumor Osteogenic Sarcomas Ewing’s Sarcoma Gliosarcoma Gastrointestinal Stromal Tumors Conclusion References

Summary Sarcomas are a heterogeneous group of tumors that rarely involve the nervous system. Neurologic effects of sarcoma are more often due to tumors outside of the central nervous system. However, as long-term survival rates in childhood sarcoma improves, reports of late neurologic complications have increased. With recent advances in treating local sarcomas with targeted molecular therapies, the incidence of late neurologic complications, such as brain metastases, will probably continue to increase. Key Words: sarcoma, gastrointestinal stromal tumor, gliosarcoma, brain metastases

1. INTRODUCTION Tumors that arise from mesenchymal tissue rarely occur within the central nervous system. Sarcomas develop from a wide variety of tissues types: fat, smooth or striated muscle, vascular tissue, and peripheral nerve. The sarcomas are an extremely heterogeneous group of rather rare tumors that comprise less than 1% of adult malignancies, and approximately 15% of pediatric malignancies. Identification of mutations and translocations associated with these tumors has illuminated aberrant signaling pathways that cause these diseases, determine their behavior, and allow for the development of therapeutic drug targets. Estimates of incidence suggest that perhaps 0.1% of intracranial tumors are sarcomas (1). A recent survey of 1100 children at a single institution with extracranial solid tumors showed that only 16 developed brain metastases (2). Some tumors previously designated as sarcomas have been renamed due to advances in molecular genetics and profiling; for example, reticular sarcoma is now recognized as malignant lymphoma, and cerebellar arachnoidal sarcoma is recognized as desmoplastic medulloblastoma. From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

495

496

Part VII / Neurologic Complications of Specific Malignancies

Table 1 Malignant Mesenchymal, Nonmeningothelial Tumors of the CNS and Relative Incidence Tissue of Origin Fibrohistiocytic and fibrous tissue Adipose tissue Muscle tissue Blood vessels Cartilage and bone Pluripotential mesenchyme Uncertain origin

Unclassified

Tumor

Age at Diagnosis

Fibrosarcoma MFH Liposarcoma Leimyosarcoma Rhabdomyosarcoma Angiosarcoma Hemangiopericytoma Chondrosarcoma Osteosarcoma Ectomesenchymoma Ewing’s sarcoma Rhabdoid tumor Meningeal sarcomatosis Sarcoma NOS

Adult 50–60 yr Adult Child Child Young adult Adult Adult Adolescent Adolescent Child to young adult Young child Infant or child

Incidence ++ +++ + + +++ +++ + + ++ + + ++ ++ ++

According to recent criteria proposed by WHO, several tumors comprise the class of malignant mesenchymal, nonmeningothelial tumors of the central nervous system (3). Table 1 outlines the malignant tumors in this category, the relative incidence with respect to other members of the category, and typical age at presentation. Paulus et al. examined the histopathologic diagnosis of 25,000 intracranial tumors and found 19 sarcomas: 6 malignant fibrous histiocytoma, 3 leiomyosarcoma, 2 each of rhabdomyosarcoma and angiosarcoma, and 1 each of fibrosarcoma, mesenchymal chondrosarcoma, differentiatied chondrosarcoma, fibromyxoid sarcoma, malignant ectomesenchymoma, and Ewing’s sarcoma (1). This list highlights the relative scarcity of any one tumor type within the class. Neurologic effects of sarcoma are more often due to tumors outside of the central nervous system. A paraspinal mass may occur due to any tumor in the family, although certain tumors predominate in patients of different ages. In early childhood, neuroblastoma is more common than sarcoma, but causes radicular or spinal symptoms in only 4% of patients. A young child found to have a spinal extradural lesion in the latter half of the first decade is more likely to have a sarcoma than a neuroblastoma (4). The incidence of spinal cord compression in childhood sarcoma is 12–15% (5,6). As long-term survival rates in childhood sarcoma improves, reports of neurologic complications have increased (7). A review of common hisotological subtypes of sarcomas that cause neurologic complications will follow.

2. CHONDROSARCOMA Chondrosarcomas are rare tumors consisting of malignant chondrocytes with variable mesenchymal components such as atypical cartilage and vascular tissue (Fig. 1). Infrequently they occur in the head and neck, where it is felt they arise from cartilaginous remnants in the petro-clival, spheno-occipital, and fronto-parietal synchondroses. When involving the CNS, they usually arise from an extradural and have imaging characteristics suggesting a meningioma; rare tumors are intradural. These tumors often arise in locations typical for meningioma, largely from the skull base (8). The distribution of intracranial chondrosarcoma in the largest series published to date, including 177 cases, listed location as petrosal bone (37%), occipital bone/clivus (23%), sphenoid (20%), frontal/ethmoid/parietal (14%) and any intradural location (6%) (9). When these tumors are midline, the differential diagnosis also includes chordoma, which can be differentiated using immunohistochemical stains for cytokeratin and epithelial membrane antigen that will be positive in chordoma but not chondrosarcoma. Rare orbital cases (17 to date) with intracranial extension have been reported (10). Several case reports and a small series of 21 cases reveal that the tumor usually is attached to the dura, though very rarely may be found intraparenchymally in the frontoparietal region (11–13). On standard CT and MRI imaging, the tumor appears as a contrast enhancing, lobulated mass. Imaging often reveals bony destruction and

Chapter 26 / Neurologic Complications of Sarcoma

497

Fig. 1. Intracranial chondrosarcoma. (A) Histology at high power showing hyaline cartilage surrounding isolated polymorphic small cells. (B) intracranial chondrosarcoma at lower power demonstrating typical small cell component. (Figure courtesy of Henry Brown, M.D., Allegheny General Hospital, Pittsburgh, Pennsylvania.)

variable degrees of intratumoral calcification; however, angiography typically reveals the tumor to be relatively hypovascular. Rare chondrosarcomas are extensively vascular, with the appearance of an arteriovascular malformation on imaging (9). Chondrosarcoma has been reported in a wide range of patients, ages 5–51 years, with no true peak in incidence. A mean age of 37 and no predisposition for either sex was seen in the series reported by Korta et al. (9). Intracranial chondrosarcoma can occur in the setting of Maffuci’s syndrome (multiple enchondromas and subcutaneous hemangiomas) and Ollier’s disease (multiple skeletal enchondromas) (14,15). A patient with Goldenhar’s syndrome of multiple facial, vertebral and jaw anomalies also developed an intracranial mesenchymal chondrosarcoma, initially misdiagnosed as meningioma (16). Typical presenting clinical features reflect the common localization of the tumor. The most common reported symptoms are oculomotor dysfunction and diplopia (51%), headache (31%), and auditory/vestibular dysfunction such as decreased hearing, dizziness, and tinnitus (21%) (9). Clinical features in patients with chordoma and chondrosarcoma overlap considerably. Both entities produce diplopia or visual impairment as the initial symptom in approximately 50% of patients, but multiple cranial neuropathies are more common in chondrosarcoma while most patients with chordoma have normal neurologic examinations at presentation (17). Recently, a patient with resection of tibial myxoid chondrosarcoma developed systemic embolization with cerebral ischemia due to lupus anticoagulant (18). These tumors vary in malignant potential, although the majority appear to have slow growth. Survival appears to correlate with the degree of differentiation of the tumor, as well as the degree of initial tumor resection (13). Treatment includes surgical resection, either complete resection alone or followed by focal radiotherapy if resection is incomplete. Regrowth after surgery is common, reported by Korta et al. in 53% of patients treated with surgery alone (9). Treatment with surgical resection alone resulted in 1- and 3-year overall survival of 81% and 45%, respectively, in spinal meningeal chondrosarcoma (19). Fractionated focal radiotherapy has been used with prolonged overall survival and time to progression in low-grade chondrosarcoma (20). Estimated 5-year survival of 83–94%, and local control at 5 years of 78–91% reported after radiation therapy suggest a standard role for this modality in the adjuvant setting (9). Chemotherapy has been used in both the adjuvant setting for higher-grade chondrosarcomas and at recurrence. Too few tumors have been reported to determine the role of chemotherapy is this tumor.

3. MALIGNANT FIBROUS HISTIOCYTOMA Malignant fibrous histiocytoma (MFH) is the most common adult soft tissue sarcoma, commonly occurring in the lower extremities, less often in the upper extremities or retroperitoneum. MFH is a malignant tumor characterized by spindled, pleomorphic giant cells in a storiform background. This tumor is seen mostly in middle-age adults, typically 50–70 years of age, with a slight male predominance. It is extremely rare (< 3% of cases) to identify a MFH arising in the CNS. MFH has been reported both as an intracranial primary tumor and as cerebral metastasis from systemic primary tumors. Two cases of primary leptomeningeal MFH have been reported (21). Several case reports in middle-age adults of primary intracranial

498

Part VII / Neurologic Complications of Specific Malignancies

MFH suggest a possible etiology of previous radiation or trauma, but true etiologic correlation is difficult to assess (22). A recent case of a low-grade variant of primary CNS histiocytic sarcoma has been described (23). Neurologic involvement due to MFH is more often seen as an indirect effect by peripheral nerve compression secondary to tumor. Retroperitoneal MFH typically presents with constant, moderately severe, back pain. Compression of peripheral nerves of the lumbar or sacral plexus may occur within the retroperitoneum, and in the extremity, an isolated peripheral nerve compression may occur. Several cases of paraneoplastic syndromes due to MFH have been recently reported including opsoclonus–myoclonus syndrome due to occult MFH of the retroperitoneum detected at autopsy (24) and brainstem encephalitis due to atypical antineuronal antibodies (25). Treatment for both intracranial and systemic MFH involves surgical resection. Prognosis is usually quite poor, with rapid local recurrence and very rare survival 2 years after diagnosis (22).

4. HEMANGIOPERICYTOMA Hemangiopericytoma (HPC) is a rare vascular tumor that may arise anywhere in the body, and rarely within the CNS where it is usually in close approximation to the leptomeninges. Formerly, intracranial HPC was classified as angioblastic meningioma. The 1993 WHO classification reclassified the tumor as a distinct entity and subsequently as a group by itself, as it is now recognized to arise from pericapillary mesenchymal cells (3). They are very vascular and highly cellular tumors and can present as intracerebral hemorrhages (26). The majority of HPC occur in adults, with a mean age at diagnosis of 40–50 years old. Children account for only 10% of intracranial HPC. The tumor usually is found attached to the dura without infiltration into the brain or spine parenchyma, although it does have a tendency to metastasize outside of the central nervous system (27). The primary location typically reflects the usual distribution of meningiomas; supratentorially over the convexities, along the petrous ridge, along the tentorium, and less often in the posterior fossa or spinal canal. HPC shares many imaging characteristics of the more common meningioma and therefore is commonly not recognized prior to surgical resection. Features that may differentiate HPC from meningioma are that it often is multilobulated with a narrow dural tail, may show bony erosion but not hyperostosis, and lacks intratumoral calcification (28). Figure 2 demonstrates nonetheless that the imaging features may closely mimic meningioma or metastatic tumors. In children, however, meningiomas are quite uncommon and several tumors may mimic meningioma on standard imaging. A recent series by Demirtas et al. reported that in a series of apparent childhood meningioma on

Fig. 2. Intracranial hemangiopericytoma in a 48-year-old woman. (A) T2-weighted FLAIR image. (B) T1 gadolinium-contrasted image demonstrates a large, well-circumscribed, homogeneously enhancing mass with relatively modest surrounding edema and mass effect. The tumor extends to the dural surface as a small tail only at the inferior–lateral edge of the mass.

Chapter 26 / Neurologic Complications of Sarcoma

499

MRI, 7 of 18 (38.8%) tumors showed anaplastic features, including two HPC, one mesenchymal chondrosarcoma, and one pleomorphic sarcoma (29). Papillary meningiomas with HPC-like solid areas were seen frequently (15.3%). Angiography typically shows small corkscrew vessels in a densely stained tumor. Treatment of HPC consists of complete surgical extirpation, if possible, followed by focal radiation therapy. Surgery is sometimes preceded by embolization of the tumor to partially devascularize these very vascular tumors, because HPC has been known to bleed extensively perioperatively. Despite this, the usual course of adult HPC is one of frequent local recurrence, estimated at 27% at 5 years and 67% at 10 years (30). Radiosurgery has been attempted in small series to enhance local control with modest apparent success. Using Gamma Knife, Payne et al. reported that 4 of 9 tumors shrank an average of 22 months after treatment (31). Distant metastasis throughout the CNS and to multiple systemic sites, including bone, pancreas, liver, heart, kidney, and skin, has been reported (32,33). The role of chemotherapy for recurrent and metastatic HPC has not been clarified; however, one series reported improved survival at 24 months, 90% vs. 60%, in patients treated with adjuvant radiation or chemotherapy versus surgery alone (27). An infantile form of HPC is recognized with a quite benign clinical course despite a histopathological appearance that in an adult would be associated with a malignant course (34). The infantile form typically presents within days of birth as a large mass within the cerebral hemispheres. With aggressive surgical removal the tumor usually follows a very indolent clinical course (35).

5. RHABDOMYOSARCOMA Rhabdomyosarcoma (RMS) is an embryonal sarcoma derived from primitive mesenchymal cells throughout the body that shows evidence for muscle differentiation within the mass. It is the most common soft tissue sarcoma in children. Two histological types are seen—embryonal and alveolar rhabdomyosarcoma. Only embryonal RMS has been reported in the central nervous system. RMS is a childhood tumor with a median age of 10–10.5 years. Usually seen sporadically, it can be associated with neurofibromatosis type 1, the Li–Fraumeni syndrome, and the Beckwith–Wiedemann syndrome. Most childhood rhabdomyosarcomas occur in the head and neck region, typically in the orbit, paranasal sinuses, pterygopalatine fossa, the infratemporal fossa, middle ear, and the parotid gland. Skull base rhabdomyosarcoma in this region is usually an invasive tumor, often extending intracranially and producing a neoplastic meningitis (36). Infiltration of the leptomeninges has been estimated to occur in one-third of patients (37). Primary intracranial rhabdomyosarcoma is very rare, with most cases occurring in children as well (38). Fewer than 50 cases of primary intracranial rhabdomyosarcoma and fewer than 10 cases of primary meningeal rhabdomyosarcoma have been reported to date (39,40). Brain metastases from systemic RMS are uncommon, usually only seen concurrent with or following developing after lung metastasis. Patients with RMS may have a very short duration of symptoms prior to diagnosis. Presenting neurologic symptoms largely reflect the location of the mass and frequently include headache, visual disturbance due to orbital tumor, hearing loss, and rarely unilateral facial paralysis due to middle ear tumor. Occasionally, RMS will present with a pseudotumor cerebri-like syndrome due to tumor in or compressing the jugular vein from extrapharyngeal areas or tumor in the heart causing venous congestion. Rarely, an intracranial mass will present with seizure or acute focal neurologic symptoms. This presentation usually signals hemorrhage into a brain metastasis. Rhabdomyosarcoma is one of the metastatic tumors prone to bleeding. Treatment for head and neck RMS has been developed according to International Rhabdomyosarcoma Society protocols that take into consideration several risk factors identified for development of leptomeningeal involvement with tumor. One or more of the following factors increases this risk: skull base erosion, cranial nerve palsy, and intracranial extension. Current recommendations are that concomitant combination chemotherapy and focal radiation therapy be administered if one or more of the risk factors is present (36). Radiation is delivered to a 2-cm margin around the gross tumor to a dose of approximately 50.4 Gy. Combination chemotherapy includes ifosfamide or melphalan, followed by vincristine, adriamycin, and cyclophosphamide. If no risk factors are present the radiation therapy is held until after completion of chemotherapy. Multimodality therapy for head and neck RMS has resulted in excellent long-term survival in most patients. Wharam reported 5-year progression-free survival of 71% for parameningeal RMS (36). Intracranial extension

500

Part VII / Neurologic Complications of Specific Malignancies

and meningeal involvement, however, portend a shorter survival than those cases without the above noted risk factors. Late effects of treatment are relatively common in patients treated for RMS. These effects include frequent abnormal facial growth, neuroendocrine abnormalities, hearing loss, visual disturbances, and cognitive loss (especially in children who receive whole-brain radiation for meningeal involvement) (41). In rare instances, distant cerebral metastases occur from systemic RMS (2% of cases) with ominous implications. The median survival, only 2.7 months in one large series, has been extremely short despite aggressive therapy (42). Primary intracranial RMS has also had a poor prognosis, and most reported cases had survivals of only months after diagnosis.

6. LEIOMYOSARCOMA Leiomyosarcoma is a previously extremely rare smooth muscle–derived tumor that is becoming somewhat more frequent as a tumor related to acquired immunodeficiency syndrome (AIDS). The most common soft tissue site is the retroperitoneum, although the tumor is commonly found in the gastrointestinal tract and uterus. Any age group can be affected. Long clinical latency of this tumor is one hallmark. Tumors are often large and unresectable at diagnosis. Examples exist of long-standing radicular-type extremity pain due to pelvic leiomyosarcoma (43). In children with AIDS the tumor is usually seen in the chest or abdomen; however, several case reports of intracranial leiomyosarcoma exist in AIDS patients. The majority of these have involved young adult AIDS patients found to have leiomyosarcoma in the sellar or suprasellar region, though cases have been reported in the pontine cistern and the spinal canal (44,45). The AIDS-related tumors have been found to be EBV-positive by in situ hybridization, suggesting that reactivation of the virus may have a role in the development of these tumors (44,46). Interestingly, a single case of a 14-year-old with common variable immunodeficiency syndrome with a temporal lobe leimyosarcoma has been reported, in which in situ hybridization revealed extensive EBV infection in the tumor (47). Fewer than 5 cases of non-AIDS related intracranial leiomyosarcoma have been reported, all in children less than 10 years old with uniformly poor outcome (48). Non-AIDS related cases of leiomyosarcoma more commonly impact the nervous system by extrinsic compression of peripheral nerves by a systemic tumor. Pelvic tumors, for example, may produce lower back pain and lumbosacral plexus dysfunction with unilateral leg weakness or numbness. A case of epithelioid leiomyosarcoma of the spine was reported in a patient who presented with progressive spastic quadriparesis (49). Treatment of intracranial leiomyosarcoma has been largely surgical resection alone, although focal radiation therapy has been administered in a few cases. The benefit of radiation therapy remains undefined due to small numbers of cases treated. Systemic tumors, likewise, are surgically resected as primary treatment. Recurrence occurs in 40–60% of cases, and repeat resection is often the treatment. The role of radiation and chemotherapy remains unclear.

7. MALIGNANT PERIPHERAL NERVE SHEATH TUMOR Malignant peripheral nerve sheath tumor (MPNST) makes up 5–10% of soft tissue sarcomas. MPNST can be divided into four groups: (i) sporadic MPNST, (ii) MPNST related to neurofibromatosis type 1 (NF-1), (iii) MPNST resulting from prior ionizing radiation exposure, and (iv) MPNST arising within ganglioneuroma or carcinoid tumor. MPNSTs are a wide group of tumors with neural differentiation that most commonly present as a spindle cell neoplasm. A retrospective institutional review of MPNST of the buttock and extremity over 35 years revealed that 53% of patients had NF-1 (50). In general, risk of death due to NF-1 is largely related to the development of malignant neoplasms, the most common of which is MPNST. There is a 7–10% lifetime chance of malignant peripheral nerve sheath tumor in NF-1 patients (51). A large population-based Finnish study identified an 8% risk of NF-1 related malignancy, the most common of which was malignant peripheral nerve sheath tumors leading directly to the patient’s death in most cases (52). A retrospective series from St. Jude’s Hospital of 28 patients with 29 MPNSTs found a 5-year overall survival of 39% (53). In most cases MPNSTs are very aggressive tumors. MPNST often appears histologically as a spindle cell neoplasm with tightly packed cells, marked pleomorphism, numerous mitotic figures, and geographic necrosis. Variants exist with glandular, epitheliod and rhabdomyoloid

Chapter 26 / Neurologic Complications of Sarcoma

501

Fig. 3. (A) Neurofibroma. Ovoid to spindle-shaped, curved cells in a matrix of collagen fibers showing very rare mitoses. (B) Malignant nerve sheath tumor. Spindle-type cells showing marked nuclear pleomorphism and frequent mitotic figures in a disorganized background. (Figure courtesy of Henry Brown, M.D., Allegheny General Hospital, Pittsburgh, Pennsylvania.)

differentiation. Most MPNST arise within benign neurofibromas; 81% did so in cases of NF-1 associated and 41% of sporadic cases in a series from the Mayo Clinic. Plexiform neurofibromas appear to have a greater predisposition to malignant transformation than do standard neurofibromas. Histopathologic differences between neurofibroma are highlighted in Fig. 3. Progression from benign to a malignant nerve sheath tumor is poorly understood. Malignant transformation in NF-1 may be related to genetic changes leading to gains of 17q or loss of 13q which can be identified in multiple MPNSTs in NF-1 but not in sporadic MPNST (54). Allelic loss of both the short and long arm of 17 has been reported for MPNST, as well as loss of heterozygosity of 17p in a patient with concurrent gliosarcoma (55). Distinguishing a benign from a malignant PNST on imaging studies can be difficult. Both tumors can cause neural foramen widening due to a dumbbell-shaped tumor extending through the foramen. The “target sign” on T2-weighted MR images—round lesions with a central hypointensity and a hyperintense rim—is helpful in distinguishing a neurofibroma. MPNSTs very rarely display the target sign, whereas neurofibroma usually will (56). Clinical presentation of MPNST varies depending on location of the mass. Tumors of the cranial nerves, spinal nerve roots, and peripheral nerves and plexuses have been reported. Spinal nerve sheath tumors along the dorsal nerve roots are commonly found in NF-1 and NF-2. One of the more common presentations of MPNST is extradural spinal cord compression due to MPNST either arising within the spinal canal or extending into the canal through a neural foramen (Fig. 4). MPNSTs arising from peripheral nerves result less frequently in neurologic signs and symptoms, despite their ability to grow to quite large size along medium to large nerves. The common presentation is of an enlarging mass in an extremity with or without neurologic symptoms. Metastatc MPNST can also present as an intracerebral hemorrhage (57). MPNST is a very aggressive tumor. Low-grade tumors occur in only 10–15% cases. A recent review estimated the overall 5-year survival in the range 34–52% (58). The most significant prognostic factor identified in MPNST associated with NF-1 was extent of initial resection. Patients with a gross total resection had a 65% 5-year overall survival; those with subtotal resection had poorer survival; no patient was alive greater than 25 months (53). Distant metastasis, 18% in one series, have been reported to multiple systemic sites, including lung and brain (59–61). Outcome has been extremely poor after development of metastasic disease from MPNST. Treatment includes surgical resection followed by focal radiation therapy; however, rapid regrowth is not uncommon.

8. OSTEOGENIC SARCOMAS Osteosarcoma is the most common primary malignancy of bone. This tumor has typical sarcomatous features, but also has direct bone formation within the tumor. The usual sites of occurrence are distal femur, proximal tibia, humerus, and pelvis. Primary osteosarcoma of the vertebral column is rare. Fewer than 100 cases, mostly in the thoracic and lumbar spine, have been reported to date (62). Osteosarcoma arising from the spine can present as acute spinal cord compression due to local kyphosis and vertebral body collapse. Brain metastases are uncommon, usually only seen in concert with lung metastasis. Isolated brain metastasis without lung involvement

502

Part VII / Neurologic Complications of Specific Malignancies

Fig. 4. MPNST arising within the spinal canal, compressing the lower thoracic spinal cord (A; arrows). A fourth ventricular mass was noted to be a metastasis from the lesion in this patient without evidence for neurofibromatosis type-1 (B; arrow).

has only been convincingly shown in a case of a boy with a patent foramen ovale that permitted bypass of the lung vasculature by metastasis (63). Osteosarcomas of the skull are rare. Risk factors include Paget’s disease and prior radiotherapy. Prior radiotherapy is also a risk factor for development of fibrosarcoma of the skull (64). In contrast to sarcomas of long bones, which frequently metastasize, post-radiation sarcomas of the skull rarely do so. Local progression results in headache and the development of neurologic symptoms related to compression of underlying brain. Treatment includes aggressive resection and fractionated radiotherapy, but such tumors are usually fatal within a few years.

9. EWING’S SARCOMA Ewing’s sarcoma of the spine is an extremely rare condition with a typical clinical triad of local back pain, neurologic deficit, and a palpable mass (65). Back pain not relieved by conservative measures, especially with symptoms suggesting sciatica, a cauda equina syndroma, or a conus medullaris syndrome has been reported in multiple teenagers as presenting symptoms of Ewing’s sarcoma (66). Rarely, spinal cord compression has been the presenting symptom. Ewing’s sarcoma may also be complicated by cerebral metastasis both in children and in adults (42,67). Ewing sarcoma rarely involves the cerebrum with only a few reported cases—most recently of a 7-year-old girl with an anterior cranial fossa mass (68).

Chapter 26 / Neurologic Complications of Sarcoma

503

Fig. 5. Gliosarcoma histopathological appearance. (A) Biphasic tumor with clear demarcation between glial and mesenchymal cells. (B) Higher-powered view of sarcomatous tissue showing spindle cells with pleomorphic nuclei. (Figure courtesy of Henry Brown, M.D., Allegheny General Hospital, Pittsburgh, Pennsylvania.)

10. GLIOSARCOMA Gliosarcoma is a variant of glioblastoma that has clear biphasic areas of glial and mesenchymal differentiation within the tumor. Approximately 2% of glioblastoma meet WHO criteria for gliosarcoma (69). Histologically, the sarcomatous portion of the tumor is demarcated from the glial portion and may show differentiation along cartilaginous, bony, or smooth muscle lines (Fig. 5). The gross appearance may be firmer and more discrete than usual for glioblastoma. The first report of a liposarcomatous differentiation of gliosarcoma has been reported (70) as well as a possible gliosarcoma arising in a oligodendroglial tumor (“oligosarcoma”) (71). Gliosarcoma tends to mimic the clinical characteristics of glioblastoma, with regards to presentation, imaging, and response to treatment. There does not seem to be a survival difference between glioblastoma and gliosarcoma (72).

11. GASTROINTESTINAL STROMAL TUMORS Gastrointestinal stromal tumors (GIST) are the most common mesenchymal neoplasm of the gastrointestinal tract and are highly resistant to conventional chemotherapy and radiotherapy (73). GIST usually metastasize to the lungs and liver while intracranial metastasis is extremely rare with several case reports in the literature of parenchymal metastases (74,75). Such tumors usually have activating mutations in either c-kit (75–80%) or PDGFRA (5–10%) tyrosine kinases and may occur in higher frequency in patients with neurofibromatosis-1 (76). Targeting these activated kinases with small-molecule kinase inhibitor (imatinib mesylate) has been effective in recurrent disease. However, resistance to imatinib is a growing problem that may result in a increase in neurologic complications such as CNS metastases (74,75,77) (Fig. 6). This group of tumors serves as a model to understand

Fig. 6. A 45-year-old man with a history of recurrent GIST was started on imatinib mesylate therapy at 400 mg daily. Doses were escalated to 800 mg day due to progressive intrabdominal masses that stabilized the lesions. Two months later he presented with intractable vomiting. A cranial MRI showed subtle enhancing lesions throughout the brain (arrows). Patient died a few months later of neurologic decline.

504

Part VII / Neurologic Complications of Specific Malignancies

the molecular mechanisms of CNS recurrence in light of systemic control of disease with these new class of targeted agents.

12. CONCLUSION Sarcomas are a heterogeneous group of tumors that rarely involve the nervous system. Neurologic effects of sarcoma are more often due to tumors outside of the central nervous system. However, as long-term survival rates in childhood sarcoma improves, reports of late neurologic complications have increased. Identification of mutations and translocations associated with these tumors has illuminated aberrant signaling pathways that cause these diseases, and allowed for development of targeted molecular therapies, There is early evidence of systemic efficacy with these agents but physicians must be alert to the late neurologic complications, such as brain metastases, that will also likely continue to be more apparent.

REFERENCES 1. Paulus W, Slowik F, Jellinger K. Primary intracranial sarcomas: histopathological features of 19 cases. Histopathology 1991;18(5): 395–402. 2. Kebudi R, Ayan I, Gorgun O et al. Brain metastasis in pediatric extracranial solid tumors: survey and literature review. J Neurooncol 2005;71(1):43–48. 3. Paulus W, Scheithauer BW. Mesenchymal, nonmeningothelial tumors. In: Kleihues P, Cavenee WK (eds). Tumors of the Nervous System. Lyon: IARC Press; 2000:185–92. 4. Di Lorenzo N, Giuffre R, Fortuna A. Primary spinal neoplasms in childhood: analysis of 1234 published cases (including 56 personal cases) by pathology, sex, age and site: differences from the situation in adults. Neurochirurgia (Stuttg) 1982;25(5):153–64. 5. Geller TJ, Kotagal S. Myeloradicular features as the initial presentation of sarcomas of childhood. Pediatr Neurol 1996;14(4):297–300. 6. Lewis DW, Packer RJ, Raney B et al. Incidence, presentation, and outcome of spinal cord disease in children with systemic cancer. Pediatrics 1986;78(3):438–443. 7. Miller RW, McKay FW. Decline in US childhood cancer mortality: 1950 through 1980. JAMA 1984;251(12):1567–1570. 8. Brackmann DE, Teufert KB. Chondrosarcoma of the skull base: long-term follow-up. Otol Neurotol 2006;27(7):981–991. 9. Korten AG, ter Berg HJ, Spincemaille GH et al. Intracranial chondrosarcoma: review of the literature and report of 15 cases. J Neurol Neurosurg Psychiatry 1998;65(1):88–92. 10. Khouja N, Ben Amor S, Jemel H et al. Mesenchymal extraskeletal chondrosarcoma of the orbit: report of a case and review of the literature. Surg Neurol 1999;52(1):50–53. 11. Harsh GRt, Wilson CB. Central nervous system mesenchymal chondrosarcoma: case report. J Neurosurg 1984;61(2):375–381. 12. Hassounah M, Al-Mefty O, Akhtar M et al. Primary cranial and intracranial chondrosarcoma: a survey. Acta Neurochir (Wien) 1985;78(3–4):123–132. 13. Parker JR, Zarabi MC, Parker JC, Jr. Intracerebral mesenchymal chondrosarcoma. Ann Clin Lab Sci 1989;19(6):401–407. 14. Balcer LJ, Galetta SL, Cornblath WT et al. Neuro-ophthalmologic manifestations of Maffucci’s syndrome and Ollier’s disease. J Neuroophthalmol 1999;19(1):62–66. 15. Ramina R, Coelho Neto M, Meneses MS et al. Maffucci’s syndrome associated with a cranial base chondrosarcoma: case report and literature review. Neurosurgery 1997;41(1):269–272. 16. Ostlere SJ, McDonald B, Athanasou NA. Mesenchymal chondrosarcoma associated with Goldenhar’s syndrome. Arch Orthop Trauma Surg 1999;119(5–6):347–348. 17. Volpe NJ, Liebsch NJ, Munzenrider JE et al. Neuro-ophthalmologic findings in chordoma and chondrosarcoma of the skull base. Am J Ophthalmol 1993;115(1):97–104. 18. Chen WH, Liu JS. Chondrosarcoma, lupus anticoagulant, and cerebral ischaemia. J Clin Neurosci 2005;12(3):305–307. 19. Forbes RB, Eljamel MS. Meningeal chondrosarcomas: a review of 31 patients. Br J Neurosurg 1998;12(5):461–464. 20. Austin-Seymour M, Munzenrider J, Goitein M et al. Fractionated proton radiation therapy of chordoma and low-grade chondrosarcoma of the base of the skull. J Neurosurg 1989;70(1):13–17. 21. Akimoto J, Takeda Y, Hasue M et al. Primary meningeal malignant fibrous histiocytoma with cerebrospinal dissemination and pulmonary metastasis. Acta Neurochir (Wien) 1998;140(11):1191–1196. 22. Amendola BE, Amendola MA, McClatchey KD. Radiation-induced malignant fibrous histiocytoma: a report of five cases including two occurring post whole-brain irradiation. Cancer Invest 1985;3(6):507–513. 23. Cao M, Eshoa C, Schultz C et al. Primary central nervous system histiocytic sarcoma with relapse to mediastinum: a case report and review of the literature. Arch Pathol Lab Med 2007;131(2):301–305. 24. Zamecnik J, Cerny R, Bartos A et al. Paraneoplastic opsoclonus–myoclonus syndrome associated with malignant fibrous histiocytoma: neuropathological findings. Cesk Patol 2004;40(2):63–67. 25. Corato M, Marinou-Aktipi K, Nano R et al. Paraneoplastic brainstem encephalitis in a patient with malignant fibrous histiocytoma and atypical antineuronal antibodies. J Neurol 2004;251(11):1415–1417. 26. Kaen A, Arrese I, Lagares A et al. Haemangiopericytoma presenting with acute intracerebral haemorrhage. Acta Neurochir (Wien) 2007;149(4):415–418.

Chapter 26 / Neurologic Complications of Sarcoma

505

27. Spatola C, Privitera G. Recurrent intracranial hemangiopericytoma with extracranial and unusual multiple metastases: case report and review of the literature. Tumori 2004;90(2):265–268. 28. Chiechi MV, Smirniotopoulos JG, Mena, H. Intracranial hemangiopericytomas: MR and CT features. AJNR Am J Neuroradiol 1996;17(7):1365–1371. 29. Demirtas E, Ersahin Y, Yilmaz F et al. Intracranial meningeal tumours in childhood: a clinicopathologic study including MIB-1 immunohistochemistry. Pathol Res Pract 2000;196(3):151–158. 30. Jaaskelainen J, Servo A, Haltia M et al. Intracranial hemangiopericytoma: radiology, surgery, radiotherapy, and outcome in 21 patients. Surg Neurol 1985;23(3):227–236. 31. Payne BR, Prasad D, Steiner M et al. Gamma surgery for hemangiopericytomas. Acta Neurochir (Wien) 2000;142(5):527–536; discussion 36–37. 32. Jeong YI, Chang SE, Lee MW et al. Case of cutaneous metastasis from intracranial hemangiopericytoma. Int J Dermatol 2005;44(10):870–872. 33. Hammontree LN, Whitehead K, Markert JM. Bilateral metastatic renal hemangiopericytoma ten years after primary intracranial lesion. Int Braz J Urol 2006;32(3):306–307. 34. Bailey PV, Weber TR, Tracy TF, Jr., et al. Congenital hemangiopericytoma: an unusual vascular neoplasm of infancy. Surgery 1993;114(5):936–941. 35. Herzog CE, Leeds NE, Bruner JM et al. Intracranial hemangiopericytomas in children. Pediatr Neurosurg 1995;22(5):274–279. 36. Wharam MD, Jr. Rhabdomyosarcoma of parameningeal sites. Semin Radiat Oncol 1997;7(3):212–216. 37. Gerson JM, Jaffe N, Donaldson MH et al. Meningeal seeding from rhabdomyosarcoma of the head and neck with base of the skull invasion: recognition of the clinical evolution and suggestions for management. Med Pediatr Oncol 1978;5(1):137–144. 38. Hawkins C, Muller P, Bilbao JM. 44-year-old man with a bleeding intracerebral tumor. Brain Pathol 1999;9(4):741–742. 39. Dropcho EJ, Allen JC. Primary intracranial rhabdomyosarcoma: case report and review of the literature. J Neuro-oncol 1987;5(2): 139–150. 40. Xu F, De Las Casas LE, Dobbs, LJ, Jr. Primary meningeal rhabdomyosarcoma in a child with hypomelanosis of Ito. Arch Pathol Lab Med 2000;124(5):762–765. 41. Paulino AC, Simon JH, Zhen W et al. Long-term effects in children treated with radiotherapy for head and neck rhabdomyosarcoma. Int J Radiat Oncol Biol Phys 2000;48(5):1489–1495. 42. Parasuraman S, Langston J, Rao BN et al. Brain metastases in pediatric Ewing sarcoma and rhabdomyosarcoma: the St. Jude Children’s Research Hospital experience. J Pediatr Hematol Oncol 1999;21(5):370–377. 43. Benyahya E, Etaouil N, Janani S et al. Sciatica as the first manifestation of a leiomyosarcoma of the buttock. Rev Rhum Engl Ed 1997;64(2):135–137. 44. Brown HG, Burger PC, Olivi A et al. R. Intracranial leiomyosarcoma in a patient with AIDS. Neuroradiology 1999;41(1):35–39. 45. Morgello S, Kotsianti A, Gumprecht JP et al. Epstein–Barr virus–associated dural leiomyosarcoma in a man infected with human immunodeficiency virus: case report. J Neurosurg 1997;86(5):883–887. 46. Kleinschmidt-DeMasters BK, Mierau GW, Sze CI et al. Unusual dural- and skull-based mesenchymal neoplasms: a report of four cases. Hum Pathol 1998;29(3):240–245. 47. Mierau GW, Greffe BS, Weeks DA. Primary leiomyosarcoma of brain in an adolescent with common variable immunodeficiency syndrome. Ultrastruct Pathol 1997;21(3):301–305. 48. Lee TT, Page LK. Primary cerebral leiomyosarcoma. Clin Neurol Neurosurg 1997;99(3):210–212. 49. Marshman LA, Pollock JR, King A et al. Primary extradural epithelioid leiomyosarcoma of the cervical spine: case report and literature review. Neurosurgery 2005;57(2):E372. 50. Hruban RH, Shiu MH, Senie RT et al. Malignant peripheral nerve sheath tumors of the buttock and lower extremity: a study of 43 cases. Cancer 1990;66(6):1253–1265. 51. Tonsgard JH. Clinical manifestations and management of neurofibromatosis type-1. Semin Pediatr Neurol 2006;13(1):2–7. 52. Poyhonen M, Niemela S, Herva R. Risk of malignancy and death in neurofibromatosis. Arch Pathol Lab Med 1997;121(2):139–143. 53. deCou JM, Rao BN, Parham DM et al. Malignant peripheral nerve sheath tumors: the St. Jude Children’s Research Hospital experience. Ann Surg Oncol 1995;2(6):524–529. 54. Lothe RA, Karhu R, Mandahl N et al. Gain of 17q24-qter detected by comparative genomic hybridization in malignant tumors from patients with von Recklinghausen’s neurofibromatosis. Cancer Res 1996;56(20):4778–4781. 55. Glover TW, Stein CK, Legius E et al. Molecular and cytogenetic analysis of tumors in von Recklinghausen neurofibromatosis. Genes Chromosomes Cancer 1991;3(1):62–70. 56. Bhargava R, Parham DM, Lasater OE et al. MR imaging differentiation of benign and malignant peripheral nerve sheath tumors: use of the target sign. Pediatr Radiol 1997;27(2):124–129. 57. Park SK, Yi HJ, Paik SS, et al. Metastasizing malignant peripheral nerve sheath tumor initially presenting as intracerebral hemorrhage: case report and review of the literature. Surg Neurol 2007;68(1):79–84. 58. Woodruff JM. Pathology of tumors of the peripheral nerve sheath in type-1 neurofibromatosis. Am J Med Genet 1999;89(1):23–30. 59. Fenzi F, Moretto G, Zamboni G et al. Brain metastases from post-radiation malignant peripheral nerve sheath tumour. Ital J Neurol Sci 1995;16(7):495–498. 60. Oishi H, Ishii K, Bandou K et al. Malignant schwannoma metastasizing to the parenchyma of the brain: case report. Neurol Med Chir (Tokyo) 2000;40(2):116–119. 61. Vege DS, Chinoy RF, Ganesh B et al. Malignant peripheral nerve sheath tumors of the head and neck: a clinicopathological study. J Surg Oncol 1994;55(2):100–103. 62. Korovessis P, Repanti M, Stamatakis M. Primary osteosarcoma of the L2 lamina presenting as “silent” paraplegia: case report and review of the literature. Eur Spine J 1995;4(6):375–378.

506

Part VII / Neurologic Complications of Specific Malignancies

63. Menassa L, Haddad S, Aoun N et al. Isolated brain metastases of osteosarcoma in a patient presenting with a patent foramen ovale. Eur Radiol 1997;7(3):365–367. 64. Dodick DW, Mokri B, Shaw EG et al. Sarcomas of calvarial bones: rare remote effect of radiation therapy for brain tumors. Neurology 1994;44(5):908–912. 65. Sharafuddin MJ, Haddad FS, Hitchon PW et al. Treatment options in primary Ewing’s sarcoma of the spine: report of seven cases and review of the literature. Neurosurgery 1992;30(4):610–618; discussion 8–9. 66. Paul FA. Ewing’s sarcoma as an etiology for persistent back pain in a 17-year-old girl after trauma to the back. J Am Osteopath Assoc 1995;95(1):58–61. 67. Olivi A, Donehower RC, Mann RB et al. Solitary, isolated metastasis from Ewing’s sarcoma to the brain: case report. Surg Neurol 1991;35(3):239–243. 68. Kazmi SA, Perry A, Pressey JG et al. Primary Ewing’s sarcoma of the brain: a case report and literature review. Diagn Mol Pathol 2007;16(2):108–111. 69. Ohgaki H, Biernat W, Reis R et al. Gliosarcoma. In: Kleihues P, Cavenee WK, (eds.). Tumors of the Nervous System. Lyon: IARC Press; 2000:42–44. 70. Vlodavsky E, Konstantinesku M, Soustiel JF. Gliosarcoma with liposarcomatous differentiation: the new member of the lipid-containing brain tumors family. Arch Pathol Lab Med 2006;130(3):381–384. 71. Rodriguez FJ, Scheithauer BW, Jenkins R et al. Gliosarcoma arising in oligodendroglial tumors (“oligosarcoma”): a clinicopathologic study. Am J Surg Pathol 2007;31(3):351–362. 72. Galanis E, Buckner JC, Dinapoli RP et al. Clinical outcome of gliosarcoma compared with glioblastoma multiforme: North Central Cancer Treatment Group results. J Neurosurg 1998;89(3):425–430. 73. Rubin BP, Heinrich MC, Corless CL. Gastrointestinal stromal tumour. Lancet 2007;369(9574):1731–1741. 74. Hughes B, Yip D, Goldstein D et al. Cerebral relapse of metastatic gastrointestinal stromal tumor during treatment with imatinib mesylate: case report. BMC Cancer 2004;4:74. 75. Kaku S, Tanaka T, Ohtuka T et al. Perisacral gastrointestinal stromal tumor with intracranial metastasis: case report. Neurol Med Chir (Tokyo) 2006;46(5):254–257. 76. Miettinen M, Fetsch JF, Sobin LH et al. Gastrointestinal stromal tumors in patients with neurofibromatosis 1: a clinicopathologic and molecular genetic study of 45 cases. Am J Surg Pathol 2006;30(1):90–96. 77. Gerin F, Baloglu O, Morgan JA et al. Central nervous system metastases from imatinib mesylate resistant gastrointestinal stromal tumor. J Neurooncol 2007;82(2):227–228.

27

Neurologic Complications of Head and Neck Cancer Katherine B. Peters,

MD, PHD,

and David Schiff,

MD

CONTENTS Overview of Cancer Direct Complications Indirect Complications Treatment-Related Neurologic Complications Specific to Cancer Conclusions References

Summary Cancers of the oral cavity, nasal cavity, salivary glands, pharynx, and larynx comprise head and neck cancers. These cancers are present not only in the United States but also around the world. Risk factors such as smoking, alcohol, and human papillomavirus are clearly linked to the prevalence of these cancers. Because of the obvious close anatomical proximity of structures involved in head and neck tumors, neurologic complications—in particular involving the cranial nerves and brain—can occur. The most common neurologic complication is cranial nerve dysfunction, but in rare instances the brain, spine, leptomeninges, vascular structures, and immune system through paraneoplastic processes can be affected. Neurologic dysfunction can occur not only secondary to direct involvement of head and neck malignancies but also from the treatment of the disease, particular from radiation and surgery. Key Words: head and neck cancer, cranial nerves, nasopharynx, oral cavity, larynx, salivary gland, perineural invasion

1. OVERVIEW OF CANCER Head and neck cancers represent a heterogenous group of malignancies involving the oral cavity, nasal cavity, salivary glands, pharynx, and larynx. In 2007, head and neck cancers contributed approximately 45,000 new cases of cancer and 11,300 deaths associated with cancer in the United States (1). This problem not only extends to developed countries such as the United States and Europe, but also affects developing countries where environmental risk factors are prevalent (2–6). External risk factors such as smoking (5,7–9), tobacco use (i.e., chewing tobacco) (5,7,10,11), alcohol (5,7–9), dietary deficiencies (5,7–10,12–15), poor dental hygiene (5,7,16), Epstein–Barr virus (5,17–22), and human papillomavirus (HPV) (2,5,23–27) are associated with an increased prevalence of head and neck cancer. In addition to external risk factors, genetic susceptibility is also present in the development of head and neck malignancies (28–30). The prognosis of patients with head and neck cancer depends on the site and stage of the malignancy (3,31). The standard treatment for head and neck malignancies depends uon the location and nature of the tumor. The mainstay of therapy for head and neck malignancies has traditionally been surgery and radiotherapy (32,33). From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

507

508

Part VII / Neurologic Complications of Specific Malignancies

Radical neck dissection and larygectomy are used to treat head and neck cancers, and these methods have been refined not only to excise tumors more effectively but also to preserve structures such as the larynx and hypopharynx (34–42). In the last two decades, adjuvant chemotherapy with surgery and/or radiotherapy has improved survival in patients with advanced head and neck cancer (32,33,43–45). Cisplatin, 5-fluorouracil, and taxanes such as docetaxel are the agents of choice as adjuvant therapy for advanced head and neck cancer (46–55). Cetuximab is a monoclonal antibody to the epidermal growth factor receptor and up-regulation of epidermal growth factor has been shown to occur in head and neck malignancies (56). Recently, cetuximab has been approved as adjuvant therapy with radiation for the treatment of squamous cell head and neck cancer that cannot be surgically resected (57). The anatomical categorization of head and neck cancer includes cancers of the oral cavity, nasal cavity, salivary glands, pharynx, and larynx. Most of the malignancies arise from the oral cavity, followed by the larynx, pharynx, and other locations. While other malignancies such as lymphomas, skin cancers of the face, melanoma, thyroid malignancies, brain tumors, and nerve sheath tumors can inhabit the anatomical areas mentioned above, this chapter will follow the head and neck cancers as defined by the International Classification of Diseases (ICD-10). For this chapter, the term “oral cancers” will encompass tumors of the lip, tongue, gum, palate, and floor of the mouth. Tumors of the pharynx will include the structures of the nasopharynx, hypopharynx, and oropharynx. Other structures discussed will be the larynx and salivary glands. In light of the obvious anatomical closeness of head and neck malignancies to multiple neurologic structures (in particular brain and cranial nerves), neurologic complications in these malignancies are common, but varying phenomena. Aiken has previously reviewed the topic of neurologic complications of head and neck cancers and focused primarily on the different types of malignancies (58). In this chapter, the discussion is organized into direct complications (i.e., anatomical), indirect complications (i.e., cerebrovascular, metabolic, paraneoplastic, etc.), and treatment-associated complications.

2. DIRECT COMPLICATIONS 2.1. Cranial Nerves Cranial nerve (CN) palsies and dysfunctions can arise from head and neck cancers because of the anatomical proximity. Because the cranial nerves control the motor, sensory, and special sensory functions of the head and neck, it is easy to ascertain that cranial nerve dysfunction can be a consequence of head and neck malignancies. The usual presentation can range from an isolated cranial nerve palsy to multiple cranial nerve palsies at the time of presentation. Neurologic recovery from cranial nerve palsies due to head and neck cancers depends primarily on the duration of the palsies before treatment (59). Cancers involving the oral cavity, pharynx, larynx, and salivary glands can lead to cranial nerve pathology. Tumors of the oral cavity arise from the tongue, gums, lips, floor of the mouth, and palate. Location dictates which cranial nerves are affected. Squamous cell cancers arising from the lip can lead to cranial nerve dysfunction. Cancer cells can invade the nerve through perineural invasion (60–65), which can present microscopically (identified on biopsy) or can be identified clinically (on exam or on radiological testing). “Perineural spread” refers to dissemination of tumor cells via the nerve itself without spread via lymphatics or vascular system. Perineural spread of squamous cell carcinomas of the head and neck is associated with a poorer prognosis (66). The most commonly affected nerve is the inferior alveolar nerve, a branch of the mandibular nerve, which is itself the third branch (V3) of the trigeminal nerve (CN V) (67–70). Patients can present with pain and paresthesias in the V3 distribution, which includes the lower lip, the lower teeth and gums, the floor of the mouth, the anterior two-thirds of the tongue, the chin and jaw, and parts of the external ear. The sensation of “ants” or “bugs” crawling underneath the skin was a common early complaint in patients who were later identified as having perineural spread of tumors (71). Other cranial nerves can be involved, including CN II, CN III, CN IV, CN VII, and CN VIII (71,72). After nerves undergo perineural invasion and neuropathies develop, the neuropathies rarely improve with treatment. A study by Garcia-Serra and colleagues revealed that only 7% patients with cranial neuropathies due to perineural spread improved with treatment (71). The most neurologic complication associated with nasopharyngeal cancer is cranial nerve abnormalities (73). Upper CN palsies (II–VIII) and lower CN palsies (IX–XII) can occur as a complication of nasopharyngeal cancer

Chapter 27 / Neurologic Complications of Head and Neck Cancer

509

(59,74–88). For nasopharyngeal tumors, studies by Li et al. found that cranial nerve palsies involving CN V, CN VI, and CN XII accounted for a majority of the cases (89). Other palsies included CN II, CN III, CN IV, CN VII, CN IX, and CN XI. The facial nerve travels through the parotid gland on its way to supplying the muscles of facial expression. Therefore, facial nerve dysfunction can be an obvious manifestation of tumors of the parotid gland, in particular adenoid cystic carcinoma (90,91).

2.2. Brain Metastatic disease associated with head and neck cancers is rare, as local involvement usually predominates the progression of these tumors (92). Brain and intracranial metastatic disease from head and neck cancers has been reported from oral squamous cell cancers (93), adenoid cystic carcinoma of the parotid gland (94–97), nasopharyngeal carcinoma (98–101), and laryngeal carcinoma (92). Through perineural spread, carcinoma of the lip has been shown to spread to the middle cranial fossa, in particular spread along the mandibular branch of CN V to gain access to the CNS (61).

2.3. Spinal Cord/Dura In rare instances, vertebral metastases can be seen in patients with oral squamous cell carcinoma (102) and pleomorphic adenoma of the parotid gland (103). Nasopharyngeal carcinoma has been documented to metastasize to the intramedullary cervical cord (104), intradural space of the spine (98), and extradural space of the spine (105,106).

2.4. Leptomeninges Carcinomatosis meningitis/leptomeningeal dissemination can occur in patients with oral squamous cell carcinoma (65), facial squamous cell carcinoma (107), nasopharyngeal cancer (108,109), squamous cell carcinoma of the larynx (110), tonsillar carcinoma (111), and uvular carcinoma (112).

2.5. Bony Invasion The invasive nature of certain head and neck malignancies can compromise the bones of the face and skull base. This type of spread can lead to communication of the cerebrospinal fluid and even the brain parenchyma with the outside environment. In rare cases, tumors such as nasopharyngeal carcinoma can erode bone thereby leading to fistula (113,114). These fistulae can be a nidus for infection, in particular meningitis (111,112). Organisms responsible for meningitis due to fistula developed from head and neck tumor invasion have included Pseudomonas aeruginosa (114) and Mycobacterium tuberculosis (115).

3. INDIRECT COMPLICATIONS In rare instances, head and neck tumors have been implicated in neurologic paraneoplastic syndromes. Pillay and Estes described a case of a woman who developed acute necrotizing myopathy that was associated with comorbid squamous cell carcinoma of the tongue (116). Another report discusses a case of anti-Hu associated encephalomyelitis in a patient with laryngeal carcinoma (117). Pharyngeal and tonsillar carcinoma has been reported in associated with paraneoplastic sensory neuropathy (118). Dermatomyositis has been reported in patients with nasopharyngeal carcinoma (119–126), hypopharyngeal carcinoma (122), and salivary pleomorphic adenoma (127). Polymyositis has also been seen in patients with nasopharyngeal carcinoma (128). Tumors of the head and neck can infiltrate not only local nerves but also local vascular structures. The internal carotid artery, common carotid artery, and jugular vein can be infiltrated by head and neck cancers, which can lead to strokes, transient ischemic attacks, and increased tumor spread (129,130). Sometimes involvement of these vessels, in particular the internal carotid artery, can be identified on physical exam by palpitation of a nonmovable mass adherent to the vessel, but imaging techniques such as computed tomography (CT), magnetic resonance imaging (MRI), and duplex Doppler ultrasound are now being used to detect early invasion or involvement of the vascular structures (131,132).

510

Part VII / Neurologic Complications of Specific Malignancies

Fig. 1. This 63-year-old woman with a long-standing history of medullary thyroid carcinoma with metastases to cervical lymph nodes presented with stereotypical presyncopal episodes triggered by turning her head. Axial post-contrast MRI demonstrates a 8.5 x 10 mm mass in contiguity with the carotid bulb (short arrow). Holter monitor was unremarkable. A trial of carbamazepine dramatically reduced the frequency of her spells. (Courtesy of Dawit Aregawi, M.D., University of Virginia).

Invasion of the vascular structures of the neck is a rare occurrence and has been estimated to occur about 5–10% of the time in patients with known cervical metastases (132). This invasion usually does not extend to the lumen of the vessel; the adventia and external elastic lamina of the vessel were involved 37.5–42% of the time in patients who underwent carotid artery resection (133,134). Invasion of the carotid sinus by head and neck cancers can be a rare cause of syncope (135–138). Syncope as a manifestation of head and neck malignancy has also been attributed to involvement of CN IX and CN X (Fig. 1) (139–143).

4. TREATMENT-RELATED NEUROLOGIC COMPLICATIONS SPECIFIC TO CANCER 4.1. Radiation-Induced Neurologic Complications Use of radiation in the treatment of head and neck cancers can cause local damage to structures such as the spinal cord, brain, and cranial nerves (144). Even structures such as the vessels that supply central nervous system and bony structures that protect the central nervous system can be affected in the long-term by radiation (144). Patients who have received radiotherapy and surgery for oral and pharyngeal cancer have reported changes in focal sensation after these therapies (145). This decrease in sensation could also be associated with decreased oral ability to recognize shapes and objects (146,147). Dysphagia and dysarthria may occur in patients with head and neck cancer who have received radiotherapy and surgery (148–155). Dysphagia, difficulty and pain with swallowing, is present in 42–87% of patients who receive radiotherapy and chemotherapy for head and neck (152–154). This has very important implications, as dysphagia can lead to severe weight loss in patients with head and neck cancer (152). Dysarthria, difficulty speaking due to weakness or dysfunction of the muscles responsible for speech, also develops in patients who receive treatment for head and neck cancer and usually develops before dysphagia (155). Radiation has been implicated in damage to cranial nerves during treatment of nasopharyngeal cancer (156–161). Palsies involving the hypoglossal nerve, vagus nerve, recurrent laryngeal nerve, and accessory nerve are the most common neuropathies associated with patients who received radiotherapy for the treatment of nasopharyngeal cancer (156–161). Among the previously mentioned nerve palsies, hypoglossal nerve palsy, manifested as weakness of the muscles of the tongue, is the most common cranial neuropathy seen in patients with nasopharyngeal cancer after radiotherapy (156).

Chapter 27 / Neurologic Complications of Head and Neck Cancer

511

Delayed injury to the brain parenchyma from radiation received during treatment for head and neck malignancies may also ensue (144). Recently, Kishimoto and colleagues investigated the types of brain injury that can be seen after irradiation for head and neck cancers (162). In these 40 patients, the initial injury to the brain parenchyma was almost always in the field of irradiation (162). Lesions included edema, white matter changes on T2 or FLAIR imaging, and development of cystic lesions (162). The peak for the development of brain injury was approximately 15–24 months (162). Clinical presentation of these changes included symptoms consistent with mass lesions, headaches, vertigo, and convulsions (162). Woo and colleagues described 11 cases of memory disturbance, complex partial seizures, and hypopituitarism in patients with nasopharyngeal carcinoma who received radiotherapy (163). These symptoms were attributed to temporal lobe and hypothalamic-pituitary axis dysfunction after radiotherapy (163). Delayed damage to the temporal lobe can be seen in patients who received radiotherapy for the treatment of nasopharyngeal cancer. Case reports involving the late effects of radiotherapy in patients with nasopharyngeal cancer have documented the following pathological conditions: cognitive dysfunction (164,165), Kluver–Bucy syndrome (166), and functional experiential hallucinosis (167) Because the spinal cord, in particular the cervical spine cord, is in close proximity to the radiation field for head and neck cancers, there has been documented spinal cord pathology after radiotherapy. Radiation-induced myelopathy can occur as a late finding after radiotherapy for nasopharyngeal cancer (168,169) and laryngeal carcinoma (170). Radiation-induced myelopathy can present as an isolated motor or sensory dysfunction and/or mixed motor sensory dysfunction along with bowel and bladder dysfunction (168,169). MRI imaging shows changes in the spinal cord such as increased signal of T2-weighted images and focal contrast enhancement on post gadolinium T1-weighted images (Fig. 2) (169,171). Other causes of myelopathy such as vitamin deficiencies, inflammatory myelitis such as transverse myelitis, and epidural spinal cord compression should be ruled out with imaging, serum testing, and cerebrospinal fluid analysis. Treatment of radiation-induced myelopathy is very difficult and attempts have been made to use surgery, corticosteroids, hyperbaric oxygen, and anticoagulation with heparin, enoxaprin, and/or warfarin (172–174). One

Fig. 2. This 48-year-old man was treated for tonsillar squamous cell carcinoma and cervical adenopathy with chemoradiotherapy including cisplatin, docetaxel, and radiation to a total dose of 70 Gy to the primary site and 54 Gy to cervical nodes. A subsequent radical neck dissection revealed no tumor. One year after diagnosis the patient presented with a gradually progressive asymmetric quadriparesis and sensory dysfunction. The MR scan demonstrates a relatively homogenous enhancing intramedullary lesion at the cervicomedullary junction (arrow). A PET scan showed no hypermetabolism. His neurologic condition stabilized over the ensuing year (perhaps coincidentally) with hyperbaric oxygen, vitamin E, and pentoxyphylline.

512

Part VII / Neurologic Complications of Specific Malignancies

case report has shown the development of syringomyelia in a patient with advanced oropharyngeal cancer after radiotherapy and chemotherapy (175). The bones of the cervical spine, face, and skull can be damaged by radiotherapy for head and neck cancers (176). Osteonecrosis or osteoradionecrosis are used to describe damage to bones caused by radiotherapy. The development of osteonecrosis depends on the radiation dose; patients who received greater than 6500 rads have a higher chance of suffering from osteonecrosis (176). Clinical manifestations of osteonecrosis vary on which bony structures are affected. Imaging including CT and MRI can show bony destruction and exposure along with development of sequestra within or surrounding necrotic bone (177). Treatment of osteoradionecrosis presents a similar challenge to the treatment of radiation-induced myelopathy. Current therapies include hyperbaric oxygen, surgery, corticosteroids, and antibiotics if infection occurs secondarily (177,178). Osteonecrosis of the cervical spine can lead to cervical cord compression and secondary osteomyelitis (179–181). Cerebrospinal fluid leakage from the ears (otorrhea) and cerebrospinal fluid accumulations or effusions in the middle ear can occur as a consequence of radiation induced osteonecrosis of the temporal bone (182,183). This has been documented in patients who received radiation for nasopharyngeal cancer and adenoid cystic cancer of the submandibular gland (183–186). Patients with otorrhea can present with unilateral hearing loss, in particular conductive hearing loss, along with leakage of cerebrospinal fluid from the ear canal and are at an increased risk for meningitis (187). Because of this open communication to the brain tissue, these patients are at an increased risk for abscesses in the temporal lobe or cerebellum (183). If osteoradionecrosis develops at the skull base, there is a serious but uncommon risk of the development of pneumocephalus: a collection of intracranial air that can be located in the epidural space, subdural space, ventricles, or in the parenchyma (188–190). The symptoms of tension pneumocephalus include headache, cerebrospinal fluid leakage from the nose (rhinorrhea), blurring of vision, vomiting, and progressive decline in consciousness (188–190). This has been documented in patients who received high-dose radiotherapy for nasopharyngeal cancer (188–191). Treatment of pneumocephalus includes emergent decompression with intraventricular catheter (if there is increased intracranial pressure), surgical repair of defect, and antibiotics (188–190). If the radiation field includes underlying brain, especially in lesions near the temporal bone, there can be long-term damage to the brain parenchyma. Wang and colleagues described a case of development of cystic brain lesions in the temporal lobe that developed 10 years following radiation therapy for cancer involving the left ear canal (192). Other temporal lobe pathology arising in patients that received radiotherapy for nasopharyngeal cancer includes abscesses (193,194), hemorrhages (195), and necrosis (196–200). As with neural tissue, vascular structures are also prone to damage from local radiotherapy and surgery. Both symptomatic and asymptomatic carotid stenosis can occur after local radiotherapy to the neck (201–217). Cheng and colleagues evaluated the development of carotid artery stenosis in patients who received radiotherapy for head and neck cancers in comparison to control patients (211). In this study, carotid arteries from control patients had adjusted freedom from progression rate at 3 years of 87% and carotid arteries from patients who receive radiotherapy had 65% for irradiated arteries (211). Brown and colleagues recently conducted a historical prospective cohort study to evaluate the incidence of carotid artery stenosis after radiotherapy for head and neck cancers (218). In this study, 44 patients were identified and followed after receiving unilateral irradiation for head and neck cancers (218). Using carotid Doppler ultrasound, they showed increased incidence of carotid stenosis of the irradiated side of the neck compared to the unirradiated side (218). Management of carotid stenosis in these patients should be evaluated on a case-by-case basis, but recent advances in angioplasty and stenting are promising (219,220). Carotid stenosis does not necessarily have implications for mortality because most patients with carotid stenosis after irradiation for head and neck malignancies eventually succumb to their malignancy, and not stroke (221). Following surgery and radiation for tumors of the larynx, carotid damage, such as carotid blowout, hemorrhage, aneurysm, and pseudoaneurysm, can occur (222–225). Iguchi and colleagues described a case of a 61-year-old man who developed carotid artery pseudoaneurysm after development of a pharyngocutaneous fistula after recurrent laryngeal carcinoma (223). The fistula provided a nidus for infection and 3 months post-operatively, imaging showed a pseudoaneurysm of the left carotid artery bifurcation (223).

Chapter 27 / Neurologic Complications of Head and Neck Cancer

513

4.2. Surgical Complications Radical neck dissection is commonly employed in the treatment of head and neck cancers. The main role of radical neck dissection is the exploration and removal of the lymphatic vessels that drain the neck, thereby providing improved staging and control of head and neck malignancies. A recent retrospective analysis by Prim and colleagues documented the neurologic complications after radical neck dissection (226). These complications included paralysis of CN XI, CN VII (in particular the marginal branch), CN XII, and Horner’s syndrome (226). These cranial neuropathies can occur separately as an isolated accessory nerve palsy, isolated hypoglossal nerve palsy, or isolated facial nerve palsy but they can also occur in combination. Paralysis of CN XI was the most common complication associated with radical neck dissection with incidence as high as 62% (226). Patients who undergo surgery for parotid tumors can experience either transient or permanent facial nerve palsy (227–231). The presence of facial nerve dysfunction and its duration are associated with extent of the parotidectomy and inclusion of radical neck dissection (231). Patients who underwent more extensive surgeries and/or radical neck dissection are more likely to have facial nerve dysfunction, particularly involving the marginal mandibular branch of the nerve (231). With the advent of more conservative resection such as superficial partial parotidectomies, patients are less likely to develop facial nerve dysfunction (228). Patient who have undergone radical neck dissection also have a higher risk for the development of carotid pathology including carotid stenosis (218), carotid blowout, carotid hemorrhage, carotid aneurysm, and carotid pseudoaneurysm (222–225). In rare cases when unilateral or bilateral internal jugular vein ligation is required during radical neck dissection, increased intracranial pressure can develop (232,233). Therefore, in radical neck dissection, care should be taken to monitor for signs of increased intracranial pressure postoperatively if the internal jugular vein is disrupted (232,233). Pseudotumor cerebri, which is characterized by headaches, papilledema, progressive visual loss due to increased intracranial pressure, can also be a rare complication of unilateral and bilateral radical neck dissection (234,235). Pseudotumor cerebri is felt to occur when the venous outflow from the brain is disturbed (234,235).

4.3. Chemotherapy-Related Complications Cisplatin and 5-fluorouracil are the most commonly used chemotherapeutic agents to treat head and neck cancers. Cisplatin routinely produces a dose-dependent large fiber sensory polyneuropathy and ototoxicity (236, 237). Paclitaxel, one of the taxanes utilized in treatment of head and neck cancer, can cause painful peripheral sensory neuropathy (238,239). Docetaxel can also cause a peripheral sensory neuropathy but it occurs less commonly than paclitaxel-associated neuropathy (239). For both cisplatin- and taxane-induced neuropathies, the distribution of the neuropathy localizes primarily to the lower extremities, and symptoms may worsen after withdrawal of the offending agent (236). There is considerable research to develop agents to prevent the development of chemotherapy induced neuropathy and these have included vitamin E, amifostine, growth factors, glutathione, Org 2766, and acetyl-l-carnitine (236,240–251). Recent Cochrane review shows that there is insufficient evidence to support the use of any of the above agents in the prevention of peripheral neuropathy (243). High-dose 5-fluorouracil therapy has been linked to development of leukoencephalopathy with changes on MRI which can present as seizures, encephalopathy, or cerebellar dysfunction (252–256). Case reports have also shown that cisplatin therapy can cause not only peripheral nervous system complications but also central nervous system complications (257). These central nervous system complications include seizures and encephalopathy with changes on MRI consistent with posterior reversible leukoencephalopathy (257). Patients usually recover completely from leukoencephalopathy due to 5-fluorouracil and cisplatin after supportive therapy and withdrawal of offending agent dysfunction (252–257). On rare occasions, patients do not recover after removal of the agent and complications can progress to status epilepticus, cerebellar dysfunction, and even death (252–257). To date, cetuximab, a monoclonal antibody to epidermal growth factor receptor utilized in head and neck cancer, has not been associated with neurologic complications (258).

514

Part VII / Neurologic Complications of Specific Malignancies

5. CONCLUSIONS Neurologic complications of malignancies are becoming increasingly common, and patients with head and neck cancer are at risk of developing neurologic complications. Structures such as the brain, cranial vasculature, and cranial nerves are at particular risk of being involved because of their anatomical closeness. Several areas of the neuraxis can be affected by head and neck cancers, but the cranial nerves are most vulnerable to injury. Cranial nerve dysfunction can arise from direct invasion by the cancer, damage by radiotherapy, and damage from surgery. The most insidious cause of cranial nerve dysfunction is perineural spread of tumors as this type can be delayed in detection until the cancer is widespread. Moreover, prognosis and outcomes after treatment are poor in patients with perineural spread of head and neck malignancies. Other structures such as the brain, spinal cord, bones of the spine and skull, dural layer, leptomeninges, and vascular structures can be affected by head and neck malignancies. Neurologic dysfunction can stem from direct extension of tumors into these areas. However, development of damage to these structures can be a consequence of therapy including chemotherapy, radiotherapy, and surgery. For patients who have received radiotherapy for head and neck cancer, aggressive monitoring for conditions such as internal carotid stenosis and osteonecrosis of the cervical spine is increasingly important. With development of new radiological techniques and new therapies for malignancy, our ability to monitor and treat neurologic complications will probably improve. In fact, the best tool to monitor and treat these side effects is forethought about the mechanisms and development of these complications.

REFERENCES 1. Jemal A, Siegel R, Ward E et al. Cancer statistics, 2007. CA: Cancer J Clin. 2007 Jan–Feb;57(1):43–66. 2. Spitz MR. Epidemiology and risk factors for head and neck cancer. Semin Oncol. 1994 Jun;21(3):281–288. 3. Sankaranarayanan R, Masuyer E, Swaminathan R et al. Head and neck cancer: a global perspective on epidemiology and prognosis. Anticancer Res. 1998 Nov–Dec;18(6B):4779–4786. 4. Argiris A, Eng C. Epidemiology, staging, and screening of head and neck cancer. Cancer Treat Res. 2003;114:15–60. 5. Sturgis EM, Wei Q, Spitz MR. Descriptive epidemiology and risk factors for head and neck cancer. Semin Oncol. 2004 Dec;31(6): 726–733. 6. Dobrossy L. Epidemiology of head and neck cancer: magnitude of the problem. Cancer Metastasis Rev. 2005 Jan;24(1):9–17. 7. Graham S, Dayal H, Rohrer T et al. Dentition, diet, tobacco, and alcohol in the epidemiology of oral cancer. J Nat Cancer Inst. 1977 Dec;59(6):1611–1608. 8. Rao DN, Ganesh B, Rao RS et al. Risk assessment of tobacco, alcohol, and diet in oral cancer: a case-control study. Int J Cancer. 1994 Aug 15;58(4):469–473. 9. Rao DN, Desai PB. Risk assessment of tobacco, alcohol, and diet in cancers of base tongue and oral tongue: a case control study. Indian J Cancer. 1998 Jun;35(2):65–72. 10. Kaugars GE, Brandt RB, Chan W et al. Evaluation of risk factors in smokeless tobacco-associated oral lesions. Oral Surg Oral Med Oral Pathol. 1991 Sep;72(3):326–331. 11. Winn DM, Blot WJ, Shy CM et al. Snuff dipping and oral cancer among women in the southern United States. N Engl J Med. 1981 Mar 26;304(13):745–749. 12. Winn DM, Ziegler RG, Pickle LW et al. Diet in the etiology of oral and pharyngeal cancer among women from the southern United States. Cancer Res. 1984 Mar;44(3):1216–1222. 13. Notani PN, Jayant K. Role of diet in upper aerodigestive tract cancers. Nutrition Cancer. 1987;10(1–2):103–113. 14. La Vecchia C, Negri E, D’Avanzo B et al. Dietary indicators of oral and pharyngeal cancer. Int J Epidemiol. 1991 Mar;20(1):39–44. 15. Levi F, Pasche C, La Vecchia C et al. Food groups and risk of oral and pharyngeal cancer. Int J Cancer. 1998 Aug 31;77(5):705–709. 16. Maier H, Zoller J, Herrmann A et al. Dental status and oral hygiene in patients with head and neck cancer. Otolaryngol Head Neck Surg. 1993 Jun;108(6):655–661. 17. Chien YC, Chen JY, Liu MY et al. Serologic markers of Epstein–Barr virus infection and nasopharyngeal carcinoma in Taiwanese men. N Engl J Med. 2001 Dec 27;345(26):1877–1882. 18. Zeng Y, Zhang LG, Wu YC et al. Prospective studies on nasopharyngeal carcinoma in Epstein–Barr virus IgA/VCA antibody-positive persons in Wuzhou City, China. Int J Cancer. 1985 Nov 15;36(5):545–547. 19. Naegele RF, Champion J, Murphy S et al. Nasopharyngeal carcinoma in American Children: Epstein–Barr virus-specific antibody titers and prognosis. Int J Cancer. 1982 Feb 15;29(2):209–212. 20. Feng P, Chan SH, Soo MY et al. Antibody response to Epstein–Barr virus Rta protein in patients with nasopharyngeal carcinoma: a new serologic parameter for diagnosis. Cancer. 2001 Oct 1;92(7):1872–1880. 21. Tam JS, Murray HG. Nasopharyngeal carcinoma and Epstein–Barr virus–associated serologic markers. Ear Nose Throat J. 1990 Apr;69(4):261–267. 22. Levine PH, Connelly RR, Milman G et al. Epstein–Barr virus serology in the control of nasopharyngeal carcinoma. Cancer Detect Prevent. 1988;12(1–6):357–362.

Chapter 27 / Neurologic Complications of Head and Neck Cancer

515

23. Fukushima K, Ogura H, Watanabe S et al. Human papillomavirus type 16 DNA detected by the polymerase chain reaction in noncancer tissues of the head and neck. Eur Arch Otorhinolaryngol. 1994;251(2):109–112. 24. Gillison ML, Lowy DR. A causal role for human papillomavirus in head and neck cancer. Lancet. 2004 May 8;363(9420):1488–1489. 25. Szentirmay Z, Polus K, Tamas L et al. Human papillomavirus in head and neck cancer: molecular biology and clinicopathological correlations. Cancer Metastasis Rev. 2005 Jan;24(1):19–34. 26. Hobbs CG, Sterne JA, Bailey M et al. Human papillomavirus and head and neck cancer: a systematic review and meta-analysis. Clin Otolaryngol. 2006 Aug;31(4):259–266. 27. Tran N, Rose BR, O’Brien CJ. Role of human papillomavirus in the etiology of head and neck cancer. Head Neck. 2006 Jul 5. 28. Copper MP, Jovanovic A, Nauta JJ et al. Role of genetic factors in the etiology of squamous cell carcinoma of the head and neck. Arch Otolaryngol Head Neck Surg. 1995 Feb;121(2):157–160. 29. Foulkes WD, Brunet JS, Sieh W et al. Familial risks of squamous cell carcinoma of the head and neck: retrospective case-control study. BMJ (Clinical research ed.) 1996 Sep 21;313(7059):716–721. 30. Sturgis EM, Wei Q. Genetic susceptibility: molecular epidemiology of head and neck cancer. Curr Opin Oncol. 2002 May;14(3): 310–317. 31. Carvalho AL, Nishimoto IN, Califano JA et al. Trends in incidence and prognosis for head and neck cancer in the United States: a site-specific analysis of the SEER database. Int J Cancer. 2005 May 1;114(5):806–816. 32. Posner MR. Paradigm shift in the treatment of head and neck cancer: the role of neoadjuvant chemotherapy. The Oncologist. 2005;10 Suppl 3:11–19. 33. Forastiere A, Koch W, Trotti A et al. Head and neck cancer. N Engl J Med. 2001 Dec 27;345(26):1890–1900. 34. Weinstein GS. Surgical approach to organ preservation in the treatment of cancer of the larynx. Oncology (Williston Park). 2001 Jun;15(6):785–796; discussion 98–803. 35. Chepeha DB, Hoff PT, Taylor RJ et al. Selective neck dissection for the treatment of neck metastasis from squamous cell carcinoma of the head and neck. The Laryngoscope. 2002 Mar;112(3):434–438. 36. Taylor RJ, Chepeha JC, Teknos TN et al. Development and validation of the neck dissection impairment index: a quality of life measure. Arch Otolaryngol Head Neck Surg. 2002 Jan;128(1):44–49. 37. Teknos TN, Hogikyan ND, Wolf GT. Conservation laryngeal surgery for malignant tumors of the larynx and pyriform sinus. Hematol Oncol Clin N Am. 2001 Apr;15(2):261–276. 38. Ferlito A, Shaha AR, Lefebvre JL et al. Organ and voice preservation in advanced laryngeal cancer. Acta Otolaryngol. 2002 Jun;122(4):438–442. 39. Lefebvre JL. What is the role of primary surgery in the treatment of laryngeal and hypopharyngeal cancer? Hayes Martin Lecture. Arch Otolaryngol Head Neck Surg. 2000 Mar;126(3):285–288. 40. Lefebvre JL. Laryngeal preservation in head and neck cancer: multidisciplinary approach. Lancet Oncol. 2006 Sep;7(9):747–755. 41. Lefebvre JL, Coche-Dequeant B, Degardin M et al. Treatment of laryngeal cancer: the permanent challenge. Expert Rev Anticancer Ther. 2004 Oct;4(5):913–920. 42. Pfister DG, Laurie SA, Weinstein GS et al. American Society of Clinical Oncology clinical practice guideline for the use of larynxpreservation strategies in the treatment of laryngeal cancer. J Clin Oncol. 2006 Aug 1;24(22):3693–3704. 43. Posner MR. Adjuvant post-operative chemoradiotherapy in head and neck cancer: a standard of care? The Oncologist. 2005 Mar;10(3):174–175. 44. Posner MR, Haddad RI, Wirth L et al. Induction chemotherapy in locally advanced squamous cell cancer of the head and neck: evolution of the sequential treatment approach. Semin Oncol. 2004 Dec;31(6):778–785. 45. Merlano M, Vitale V, Rosso R et al. Treatment of advanced squamous-cell carcinoma of the head and neck with alternating chemotherapy and radiotherapy. N Engl J Med. 1992 Oct 15;327(16):1115–21. 46. Clark JR, Busse PM, Norris CM, Jr. et al. Induction chemotherapy with cisplatin, fluorouracil, and high-dose leucovorin for squamous cell carcinoma of the head and neck: long-term results. J Clin Oncol. 1997 Sep;15(9):3100–3110. 47. Colevas AD, Busse PM, Norris CM et al. Induction chemotherapy with docetaxel, cisplatin, fluorouracil, and leucovorin for squamous cell carcinoma of the head and neck: a phase I/II trial. J Clin Oncol. 1998 Apr;16(4):1331–1339. 48. Colevas AD, Norris CM, Tishler RB et al. Phase II trial of docetaxel, cisplatin, fluorouracil, and leucovorin as induction for squamous cell carcinoma of the head and neck. J Clin Oncol. 1999 Nov;17(11):3503–3511. 49. Colevas AD, Norris CM, Tishler RB et al. Phase I/II trial of outpatient docetaxel, cisplatin, 5-fluorouracil, leucovorin (opTPFL) as induction for squamous cell carcinoma of the head and neck (SCCHN). Am J Clinical Oncol. 2002 Apr;25(2):153–159. 50. Ervin TJ, Clark JR, Weichselbaum RR et al. An analysis of induction and adjuvant chemotherapy in the multidisciplinary treatment of squamous-cell carcinoma of the head and neck. J Clin Oncol. 1987 Jan;5(1):10–20. 51. Haddad R, Tishler RB, Norris CM et al. Docetaxel, cisplatin, 5-fluorouracil (TPF)-based induction chemotherapy for head and neck cancer and the case for sequential, combined-modality treatment. The Oncologist. 2003;8(1):35–44. 52. Haddad RI, Wirth L, Posner MR. Integration of chemotherapy in the curative treatment of locally advanced head and neck cancer. Expert Rev Anticancer Ther. 2003 Jun;3(3):331–338. 53. Posner MR, Colevas AD. Induction chemotherapy in the management of squamous cell cancer of the head and neck. Cancer J Sci Am. 1997 Mar–Apr;3(2):73–75. 54. Tishler RB, Busse PM, Norris CM, Jr. et al. An initial experience using concurrent paclitaxel and radiation in the treatment of head and neck malignancies. Int J Radiat Oncol Biol Phys. 1999 Mar 15;43(5):1001–1008. 55. Tishler RB, Norris CM, Jr., Colevas AD et al. A Phase I/II trial of concurrent docetaxel and radiation after induction chemotherapy in patients with poor prognosis squamous cell carcinoma of the head and neck. Cancer. 2002 Oct 1;95(7):1472–1481. 56. Harding J, Burtness B. Cetuximab: an epidermal growth factor receptor chemeric human-murine monoclonal antibody. Drugs Today (Barc). 2005 Feb;41(2):107–127.

516

Part VII / Neurologic Complications of Specific Malignancies

57. Cetuximab approved by FDA for treatment of head and neck squamous cell cancer. Cancer Biol Therapy. 2006 Apr;5(4):340–342. 58. Aiken RD. Neurologic complications of head and neck cancers. Semin Oncol. 2006 Jun;33(3):348–51. 59. Turgut M, Erturk O, Saygi S et al. Importance of cranial nerve involvement in nasopharyngeal carcinoma: a clinical study comprising 124 cases with special reference to clinical presentation and prognosis. Neurosurg Rev. 1998;21(4):243–248. 60. Mickalites CJ, Rappaport I. Perineural invasion by squamous-cell carcinoma of the lower lip: review of the literature and report of a case. Oral Surg Oral Med Oral Pathol. 1978 Jul;46(1):74–78. 61. Anderson C, Krutchkoff D, Ludwig M. Carcinoma of the lower lip with perineural extension to the middle cranial fossa. Oral Surg Oral Med Oral Pathol. 1990 May;69(5):614–618. 62. Bagatin M, Orihovac Z, Mohammed AM. Perineural invasion by carcinoma of the lower lip. J Craniomaxillofac Surg. 1995 Jun;23(3):155–159. 63. Pyle MA, Zak J, Bath M et al. Perineural spread of squamous cell carcinoma of the lip: the importance of follow-up and collaboration. Spec Care Dentist. 1999 May–Jun;19(3):118–122. 64. Bhatnagar AK, Heron DE, Schaitkin B. Perineural invasion of squamous cell carcinoma of the lip with occult involvement of the infra-orbital nerve detected by PET-CT and treated with MRI-based IMRT: a case report. Technol Cancer Res Treat. 2005 Jun;4(3):251–253. 65. Sullivan LM, Smee R. Leptomeningeal carcinomatosis from perineural invasion of a lip squamous cell carcinoma. Australasian Radiol. 2006 Jun;50(3):262–266. 66. Soo KC, Carter RL, O’Brien CJ et al. Prognostic implications of perineural spread in squamous carcinomas of the head and neck. The Laryngoscope. 1986 Oct;96(10):1145–1148. 67. Schmidseder R, Dick H. Spread of epidermoid carcinoma of the lip along the inferior alveolar nerve. Oral Surg Oral Med Oral Pathol. 1977 Apr;43(4):517–520. 68. Califano L, Maremonti P, Zupi A et al. Spread of squamous cell carcinoma of the lower lip along the inferior alveolar nerve: a case report. J Oral Maxillofac Surg. 1995 Sep;53(9):1108–1110. 69. Fujita M, Hirokawa Y, Naito K et al. Recurrent lower gingival squamous cell carcinoma spreading along the pathway of the inferior alveolar nerve. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 1995 Sep;80(3):369–375. 70. Zupi A, Mangone GM, Piombino P et al. Perineural invasion of the lower alveolar nerve by oral cancer: a follow-up study of 12 cases. J Craniomaxillofac Surg. 1998 Oct;26(5):318–321. 71. Garcia-Serra A, Hinerman RW, Mendenhall WM et al. Carcinoma of the skin with perineural invasion. Head Neck. 2003 Dec;25(12):1027–1033. 72. Banerjee TK, Gottschalk PG. Unusual manifestations of multiple cranial nerve palsies and mandibular metastasis in a patient with squamous cell carcinoma of the lip. Cancer. 1984 Jan 15;53(2):346–348. 73. Turgman J, Braham J, Modan B et al. Neurological complications in patients with malignant tumors of the nasopharynx. Eur Neurol. 1978;17(3):149–154. 74. Ilhan O, Sener EC, Ozyar E. Outcome of abducens nerve paralysis in patients with nasopharyngeal carcinoma. Eur J Ophthalmol. 2002 Jan–Feb;12(1):55–59. 75. Low WK. Facial palsy from metastatic nasopharyngeal carcinoma at various sites: three reports. Ear Nose Throat J. 2002 Feb;81(2): 99–101. 76. Chong VF, Fan YF. Hypoglossal nerve palsy in nasopharyngeal carcinoma. Eur Radiol. 1998;8(6):939–945. 77. Luo CB, Teng MM, Chen SS et al. Orbital invasion in nasopharyngeal carcinoma: evaluation with computed tomography and magnetic resonance imaging. Zhonghua yi xue za zhi (Chinese Medical Journal); Free China ed. 1998 Jul;61(7):382–388. 78. Chong VF, Fan YF. Pterygopalatine fossa and maxillary nerve infiltration in nasopharyngeal carcinoma. Head Neck. 1997 Mar;19(2):121–125. 79. Chong VF, Fan YF. Maxillary nerve involvement in nasopharyngeal carcinoma. AJR. 1996 Nov;167(5):1309–1312. 80. Chong VF, Fan YF. Jugular foramen involvement in nasopharyngeal carcinoma. J Laryngol Otol . 1996 Oct;110(10):987–990. 81. Su CY, Lui CC. Perineural invasion of the trigeminal nerve in patients with nasopharyngeal carcinoma: imaging and clinical correlations. Cancer. 1996 Nov 15;78(10):2063–2069. 82. Chong VF. Trigeminal neuralgia in nasopharyngeal carcinoma. J Laryngol Otol . 1996 Apr;110(4):394–396. 83. Ozyar E, Atahan IL, Akyol FH et al. Cranial nerve involvement in nasopharyngeal carcinoma: its prognostic role and response to radiotherapy. Radiat Med. 1994 Mar–Apr;12(2):65–68. 84. Tao ZD. Oculomotor neuropathy syndrome: a diagnostic challenge in nasopharyngeal carcinoma. Zhonghua yi xue za zhi (Chinese Medical Journal). 1992 Jul;105(7):567–571. 85. Schifter M, Barrett AP. Multiple cranial nerve involvement leading to diagnosis of nasopharyngeal carcinoma: case report. J Oral Maxillofac Surg. 1992 Apr;50(4):400–402. 86. Prasad U, Doraisamy S. Optic nerve involvement in nasopharyngeal carcinoma. Eur J Surg Oncol. 1991 Oct;17(5):536–540. 87. Sham JS, Cheung YK, Choy D et al. Cranial nerve involvement and base of the skull erosion in nasopharyngeal carcinoma. Cancer. 1991 Jul 15;68(2):422–426. 88. Leung SF, Tsao SY, Teo P et al. Cranial nerve involvement by nasopharyngeal carcinoma: response to treatment and clinical significance. Clin Oncol (R Coll Radiol). 1990 May;2(3):138–141. 89. Li JC, Mayr NA, Yuh WT et al. Cranial nerve involvement in nasopharyngeal carcinoma: response to radiotherapy and its clinical impact. Ann Otol Rhinol Laryngol. 2006 May;115(5):340–345. 90. Brach JS, VanSwearingen JM. Not all facial paralysis is Bell’s palsy: a case report. Arch Phys Med Rehabil. 1999 Jul;80(7):857–859. 91. Terhaard C, Lubsen H, Tan B et al. Facial nerve function in carcinoma of the parotid gland. Eur J Cancer. 2006 Nov;42(16):2744–2750. 92. de Bree R, Mehta DM, Snow GB et al. Intracranial metastases in patients with squamous cell carcinoma of the head and neck. Otolaryngol Head Neck Surg. 2001 Feb;124(2):217–221.

Chapter 27 / Neurologic Complications of Head and Neck Cancer

517

93. Hardee PS, Hutchison IL. Intracranial metastases from oral squamous cell carcinoma. Br J Oral Maxillofac Surg. 2001 Aug;39(4): 282–285. 94. Kazumoto K, Hayase N, Kurosumi M et al. Multiple brain metastases from adenoid cystic carcinoma of the parotid gland: case report and review of the literature. Surg Neurol. 1998 Nov;50(5):475–479. 95. Gross SW, Friedman AP. Cerebral metastases from a mixed tumor of the parotid gland. Neurology. 1955 Jun;5(6):435–437. 96. Hammoud MA, Hassenbusch SJ, Fuller GN et al. Multiple brain metastases: a rare manifestation of adenoid cystic carcinoma of the parotid gland. J Neuro-oncol. 1996 Jan;27(1):61–64. 97. Sheedy SP, Welker KM, DeLone DR et al. CNS metastases of carcinoma ex pleomorphic adenoma of the parotid gland. AJNR. 2006 Aug;27(7):1483–1485. 98. Ngan RK, Yiu HH, Cheng HK et al. Central nervous system metastasis from nasopharyngeal carcinoma: a report of two patients and a review of the literature. Cancer. 2002 Jan 15;94(2):398–405. 99. Liaw CC, Ho YS, Koon-Kwan NG et al. Nasopharyngeal carcinoma with brain metastasis: a case report. J Neuro-oncol. 1994;22(3):227–230. 100. Young YH, Lin CY, Lou PJ et al. Intracranial relapse of nasopharyngeal carcinoma manifested as sudden deafness. Otol Neurotol. 2001 May;22(3):392–396. 101. Low WK, Fong KW, Chong VF. Cerebellopontine angle involvement by nasopharyngeal carcinoma. Am J Otol . 2000 Nov;21(6): 871–876. 102. Carlson ER, Ord RA. Vertebral metastases from oral squamous cell carcinoma. J Oral Maxillofac Surg. 2002 Aug;60(8):858–862. 103. Hay MA, Witterick IJ, Mock D. Recurrent pleomorphic adenoma of the parotid gland with cervical metastasis. J Otolaryngol. 2001 Dec;30(6):361–365. 104. Morariu MA, Serban M. Intramedullary cervical cord metastasis from a nasopharynx carcinoma. Eur Neurol. 1974;11(5):317–322. 105. Bagatzounis A, Erakleous E, Michaelides I. Epidural metastasis in nasopharyngeal carcinoma. Strahlenther Onkol. 2003 Feb;179(2):123–128. 106. Elango S, Kareem BA, Chandrasekaran S et al. Nasopharyngeal carcinoma with spinal secondaries. J Laryngol Otol . 1991 Sep;105(9):772–773. 107. Zhu JJ, Padillo O, Duff J et al. Cavernous sinus and leptomeningeal metastases arising from a squamous cell carcinoma of the face: case report. Neurosurgery. 2004 Feb;54(2):492–498; discussion 8–9. 108. Lee O, Cromwell LD, Weider DJ. Carcinomatous meningitis arising from primary nasopharyngeal carcinoma. Am J Otolaryngol. 2005 May–Jun;26(3):193–197. 109. Wang CJ, Wang CY. Nasopharyngeal carcinoma with leptomeningeal dissemination: case report. Chang Gung Medical Journal. 2000 Feb;23(2):118–122. 110. Thompson SR, Veness MJ, Morgan GJ et al. Leptomeningeal carcinomatosis from squamous cell carcinoma of the supraglottic larynx. Australasian Radiol. 2003 Sep;47(3):325–330. 111. Biswal BM, Goyal M, Prasad RR et al. Leptomeningeal carcinomatosis from carcinoma of the palatine tonsil. Australasian Radiol. 1998 Feb;42(1):66–68. 112. Garcia C, Rubio G, Molina F et al. Gadolinium-DTPA MRI in the diagnosis of a patient with leptomeningeal metastasis produced by uvular carcinoma. Neuroradiology. 1991;33(3):282–283. 113. Sculier JP, Klastersky J. Rhinorrhea and Pseudomonas meningitis associated with a rhinopharyngeal tumor. Arch Otolaryngol. 1982 Jan;108(1):36–37. 114. Tang LM, Chen ST, Ng SH. Bacterial meningitis in patients with nasopharyngeal carcinoma. QJM. 1996 Jan;89(1):71–76. 115. Tang LM, Chen ST, Chang HS. Tuberculous meningitis in patients with nasopharyngeal carcinoma. Scand J Infect Dis. 1996;28(2): 195–196. 116. Pillay PK, Estes ML. Acute necrotising myopathy in association with carcinoma of the tongue. Ann Acad Med Singapore. 1993 May;22(3 Suppl):516–517. 117. Baijens LW, Manni JJ. Paraneoplastic syndromes in patients with primary malignancies of the head and neck: four cases and a review of the literature. Eur Arch Otorhinolaryngol. 2006 Jan;263(1):32–36. 118. Pericot I, Rio J, Jaen A et al. [Paraneoplastic sensory neuropathy due to pharynx and tonsil squamous cell carcinoma]. Neurologia (Barcelona, Spain). 2001 May;16(5):237–239. 119. Fung WK, Chan HL, Lam WM. Amyopathic dermatomyositis in Hong Kong: association with nasopharyngeal carcinoma. Int J Dermatol. 1998 Sep;37(9):659–663. 120. Boussen H, Mebazaa A, Gritli S et al. [Dermatomyositis and nasopharyngeal carcinoma: 3 cases]. Ann Dermatol Venereol. 2000 Apr;127(4):389–392. 121. Boussen H, Mebazaa A, Nasr C et al. Dermatomyositis and nasopharyngeal carcinoma: report of 8 cases. Arch Dermatol. 2006 Jan;142(1):112–113. 122. Liu W, Gao Z, Ni D et al. [Dermatomyositis/polymyositis associated with head and neck malignancy]. Lin Chuang Er Bi Yan Hou Ke Za Zhi (Journal of Clinical Otorhinolaryngology). 2005 Apr;19(8):340–341, 344. 123. Abraham Z, Rosner I, Rozenbaum M et al. Dermatomyositis and nasopharyngeal carcinoma. J Dermatol . 1998 Aug;25(8):539–543. 124. Hu WH. [Nasopharyngeal carcinoma (NPC) with dermatomyositis: an analysis of 30 cases]. Zhonghua zhong liu za zhi [Chinese Journal of Oncology]. 1986 Mar;8(2):133–135. 125. Hu WJ, Chen DL, Min HQ. Study of 45 cases of nasopharyngeal carcinoma with dermatomyositis. Am J Clin Oncol. 1996 Feb;19(1): 35–38. 126. Teo P, Tai TH, Choy D. Nasopharyngeal carcinoma with dermatomyositis. Int J Radiat Oncol Biol Phys. 1989 Feb;16(2):471–474. 127. Holden CA, Davis RW, MacDonald DM. Dermatomyositis and salivary pleomorphic adenoma. J R Soc Med. 1983 Sep;76(9):787–788.

518

Part VII / Neurologic Complications of Specific Malignancies

128. Lee KH, Ong B, Lim TK et al. Polymyositis associated with symptomless nasopharyngeal carcinoma. Singapore Med J. 1994 Jun;35(3):323–324. 129. Yu Q, Wang P, Shi H et al. Carotid artery and jugular vein invasion of oral-maxillofacial and neck malignant tumors: diagnostic value of computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2003 Sep;96(3):368–372. 130. Yoo GH, Hocwald E, Korkmaz H et al. Assessment of carotid artery invasion in patients with head and neck cancer. The Laryngoscope. 2000 Mar;110(3 Pt 1):386–390. 131. Freeman SB, Hamaker RC, Borrowdale RB et al. Management of neck metastasis with carotid artery involvement. The Laryngoscope. 2004 Jan;114(1):20–24. 132. Nemeth Z, Domotor G, Talos M et al. Resection and replacement of the carotid artery in metastatic head and neck cancer: literature review and case report. Int J Oral Maxillofac Surg. 2003 Dec;32(6):645–650. 133. Huvos AG, Leaming RH, Moore OS. Clinicopathologic study of the resected carotid artery. Analysis of sixty-four cases. Am J Surg. 1973 Oct;126(4):570–574. 134. McCready RA, Miller SK, Hamaker RC et al. What is the role of carotid arterial resection in the management of advanced cervical cancer? J Vasc Surg. 1989 Sep;10(3):274–280. 135. Bauer CA, Redleaf MI, Gartlan MG et al. Carotid sinus syncope in head and neck cancer. The Laryngoscope. 1994 Apr;104(4):497–503. 136. Muntz HR, Smith PG. Carotid sinus hypersensitivity: a cause of syncope in patients with tumors of the head and neck. The Laryngoscope. 1983 Oct;93(10):1290–1293. 137. Tulchinsky M, Krasnow SH. Carotid sinus syndrome associated with an occult primary nasopharyngeal carcinoma. Arch Intern Med. 1988 May;148(5):1217–1219. 138. Tang Y, Wang JM, Huang CH. Syncope in nasopharyngeal carcinoma: report of three cases and review of the literature. Changgeng Yi Xue Za Zhi/Changgeng Ji Nian Yi Yuan (Chang Gung Medical Journal/Chang Gung Memorial Hospital). 1993 Mar;16(1):59–65. 139. Macdonald DR, Strong E, Nielsen S et al. Syncope from head and neck cancer. J Neuro-oncol. 1983;1(3):257–267. 140. Papay FA, Roberts JK, Wegryn TL et al. Evaluation of syncope from head and neck cancer. The Laryngoscope. 1989 Apr;99(4): 382–388. 141. Lin RH, Teng MM, Wang SJ et al. Syncope as the presenting symptom of nasopharyngeal carcinoma. Clin Neurol Neurosurg. 1994 May;96(2):152–155. 142. Worth PF, Stevens JC, Lasri F et al. Syncope associated with pain as the presenting feature of neck malignancy: failure of cardiac pacemaker to prevent attacks in two cases. J Neurol Neurosurg Psychiatry. 2005 Sep;76(9):1301–1303. 143. Chen-Scarabelli C, Kaza AR, Scarabelli T. Syncope due to nasopharyngeal carcinoma. Lancet Oncol. 2005 May;6(5):347–349. 144. Becker M, Schroth G, Zbaren P et al. Long-term changes induced by high-dose irradiation of the head and neck region: imaging findings. Radiographics. 1997 Jan–Feb;17(1):5–26. 145. Bodin I, Jaghagen EL, Isberg A. Intraoral sensation before and after radiotherapy and surgery for oral and pharyngeal cancer. Head Neck. 2004 Nov;26(11):923–929. 146. Bodin I, Lind MG, Henningsson G, Isberg A. Deterioration of intraoral hole size identification after treatment of oral and pharyngeal cancer. Acta Otolaryngol. 1999;119(5):609–616. 147. Bodin I, Lind M, Henningsson G et al. Deterioration of intraoral recognition of shapes after treatment of oral and pharyngeal cancer. Otolaryngol Head Neck Surg. 2000 Apr;122(4):584–589. 148. Baker BM, Fraser AM, Baker CD. Long-term postoperative dysphagia in oral/pharyngeal surgery patients: subjects’ perceptions vs. videofluoroscopic observations. Dysphagia. 1991;6(1):11–16. 149. Mady K, Sader R, Hoole PH et al. Speech evaluation and swallowing ability after intra-oral cancer. Clin Linguist Phon. 2003 Jun–Aug;17(4–5):411–420. 150. Nicoletti G, Soutar DS, Jackson MS et al. Chewing and swallowing after surgical treatment for oral cancer: functional evaluation in 196 selected cases. Plas Reconstruct Surg. 2004 Aug;114(2):329–338. 151. Nicoletti G, Soutar DS, Jackson MS et al. Objective assessment of speech after surgical treatment for oral cancer: experience from 196 selected cases. Plas Reconstruct Surg. 2004 Jan;113(1):114–125. 152. Nguyen NP, North D, Smith HJ et al. Safety and effectiveness of prophylactic gastrostomy tubes for head and neck cancer patients undergoing chemoradiation. Surg Oncol. 2007;15(4):199–203. 153. Nguyen NP, Frank C, Moltz CC et al. Impact of dysphagia on quality of life after treatment of head-and-neck cancer. Int J Radiat Oncol Biol Phys. 2005 Mar 1;61(3):772–778. 154. Nguyen NP, Moltz CC, Frank C et al. Dysphagia following chemoradiation for locally advanced head and neck cancer. Ann Oncol. 2004 Mar;15(3):383–388. 155. Radford K, Woods H, Lowe D et al. A UK multicentre pilot study of speech and swallowing outcomes following head and neck cancer. Clin Otolaryngol Allied Sci. 2004 Aug;29(4):376–381. 156. Lin YS, Jen YM, Lin JC. Radiation-related cranial nerve palsy in patients with nasopharyngeal carcinoma. Cancer. 2002 Jul 15;95(2):404–409. 157. Shapiro BE, Rordorf G, Schwamm L et al. Delayed radiation-induced bulbar palsy. Neurology. 1996 Jun;46(6):1604–1606. 158. Chaudhry MR, Akhtar S. Bilateral vocal cord paralysis following radiation therapy for nasopharyngeal carcinoma. ORL; J Oto-RhinoLaryngol Related Specialties. 1995 Jan–Feb;57(1):48–49. 159. Stern Y, Marshak G, Shpitzer T et al. Vocal cord palsy: possible late complication of radiotherapy for head and neck cancer. Ann Otol Rhinol Laryngol. 1995 Apr;104(4 Pt 1):294–296. 160. Takimoto T, Saito Y, Suzuki M et al. Radiation-induced cranial nerve palsy: hypoglossal nerve and vocal cord palsies. J Laryngol Otol. 1991 Jan;105(1):44–45. 161. Kang MY, Holland JM, Stevens KR, Jr. Cranial neuropathy following curative chemotherapy and radiotherapy for carcinoma of the nasopharynx. J Laryngol Otol. 2000 Apr;114(4):308–310.

Chapter 27 / Neurologic Complications of Head and Neck Cancer

519

162. Kishimoto R, Mizoe JE, Komatsu S et al. MR imaging of brain injury induced by carbon ion radiotherapy for head and neck tumors. Magn Reson Med Sci. 2005 Dec 31;4(4):159–164. 163. Woo E, Lam K, Yu YL et al. Temporal lobe and hypothalamic-pituitary dysfunctions after radiotherapy for nasopharyngeal carcinoma: a distinct clinical syndrome. J Neurol Neurosurg Psychiatry. 1988 Oct;51(10):1302–1307. 164. Cheung M, Chan AS, Law SC et al. Cognitive function of patients with nasopharyngeal carcinoma with and without temporal lobe radionecrosis. Arch Neurol. 2000 Sep;57(9):1347–1352. 165. Cheung MC, Chan AS, Law SC et al. Impact of radionecrosis on cognitive dysfunction in patients after radiotherapy for nasopharyngeal carcinoma. Cancer. 2003 Apr 15;97(8):2019–2026. 166. Lam LC, Chiu HF. Kluver–Bucy syndrome in a patient with nasopharyngeal carcinoma: a late complication of radiation brain injury. J Geriatr Psychiatry Neurol. 1997 Jul;10(3):111–113. 167. Lam LC, Leung SF, Chow LY. Functional experiential hallucinosis after radiotherapy for nasopharyngeal carcinoma. J Neurol Neurosurg Psychiatry. 1998 Feb;64(2):259–261. 168. Ballantyne AJ. Late sequelae of radiation therapy in cancer of the head and neck with particular reference to the nasopharynx. Am J Surg. 1975 Oct;130(4):433–436. 169. Wang PY, Shen WC. Magnetic resonance imaging in two patients with radiation myelopathy. Journal of the Formosan Medical Association. 1991 Jun;90(6):583–585. 170. Moskal J, Kluczewska E, Moskal W et al. [Post-radiation cervical myelopathy after radiotherapy of laryngeal carcinoma]. Neurol Neurochir Pol. 1997 Nov–Dec;31(6):1245–1251. 171. Wang PY, Shen WC, Jan JS. Serial MRI changes in radiation myelopathy. Neuroradiology. 1995 Jul;37(5):374–377. 172. Shaw PJ, Bates D. Conservative treatment of delayed cerebral radiation necrosis. J Neurol Neurosurg Psychiatry. 1984 Dec;47(12):1338–1341. 173. Liu CY, Yim BT, Wozniak AJ. Anticoagulation therapy for radiation-induced myelopathy. Ann Pharmacotherapy. 2001 Feb;35(2): 188–191. 174. Chuba PJ, Aronin P, Bhambhani K et al. Hyperbaric oxygen therapy for radiation-induced brain injury in children. Cancer. 1997 Nov 15;80(10):2005–2012. 175. Ostler PJ, Patel N, Grant HR et al. Syringomyelia after chemotherapy and radiotherapy for advanced oropharyngeal carcinoma: cause or coincidence? Clin Oncol (R Coll Radiol). 1998;10(5):324–326. 176. Morrish RB, Jr., Chan E, Silverman S, Jr. et al. Osteonecrosis in patients irradiated for head and neck carcinoma. Cancer. 1981 Apr 15;47(8):1980–1983. 177. Huang XM, Zheng YQ, Zhang XM et al. Diagnosis and management of skull base osteoradionecrosis after radiotherapy for nasopharyngeal carcinoma. The Laryngoscope. 2006 Sep;116(9):1626–1631. 178. London SD, Park SS, Gampper TJ et al. Hyperbaric oxygen for the management of radionecrosis of bone and cartilage. The Laryngoscope. 1998 Sep;108(9):1291–1296. 179. Donovan DJ, Huynh TV, Purdom EB et al. Osteoradionecrosis of the cervical spine resulting from radiotherapy for primary head and neck malignancies: operative and nonoperative management: case report. J Neurosurg. 2005 Aug;3(2):159–164. 180. Lim AA, Karakla DW, Watkins DV. Osteoradionecrosis of the cervical vertebrae and occipital bone: a case report and brief review of the literature. Am J Otolaryngol. 1999 Nov–Dec;20(6):408–411. 181. Mut M, Schiff D, Miller B et al. Osteoradionecrosis mimicking metastatic epidural spinal cord compression. Neurology. 2005 Jan 25;64(2):396–397. 182. Sikand A, Longridge N. CSF otorrhea complicating osteoradionecrosis of the temporal bone. J Otolaryngol. 1991 Jun;20(3):209–211. 183. Guida RA, Finn DG, Buchalter IH et al. Radiation injury to the temporal bone. Am J Otol. 1990 Jan;11(1):6–11. 184. Lim BY, Pang KP, Low WK et al. CSF otorrhea complicating temporal bone osteoradionecrosis in a patient with nasopharyngeal carcinoma. Ear Nose Throat J. 2005 Jan;84(1):39–40. 185. Ou Y, Zheng Y, Chen S et al. [Diagnosis and treatment on osteoradionecrosis of temporal bone in cases with nasopharyngeal carcinoma after radiotherapy]. Lin Chuang Er Bi Yan Hou Ke Za Zhi (Journal of Clinical Otorhinolaryngology). 2006 Oct;20(19):865–867. 186. Tsang WS, Ku PK, Andrew van Hasselt C. Osteoradionecrosis of the temporal bone in nasopharyngeal carcinoma after radiotherapy: a case report. Ear Nose Throat J. 2000 Feb;79(2):94–95. 187. Brown NE, Grundfast KM, Jabre A et al. Diagnosis and management of spontaneous cerebrospinal fluid–middle ear effusion and otorrhea. The Laryngoscope. 2004 May;114(5):800–805. 188. Ng WF, Fung KH, Sham JS. Tension pneumocephalus: a rare complication of radiotherapy in nasopharyngeal carcinoma. Pathology. 1995 Apr;27(2):204–208. 189. Wang HC, Hwang JC, Peng JP et al. Tension pneumocephalus: a rare complication of radiotherapy: a case report. J Emerg Med . 2006 Nov;31(4):387–389. 190. Wu CT, Lee ST. Delayed spontaneous tension pneumocephalus caused by radionecrosis of the skull base. Br J Neurosurg. 1999 Apr;13(2):214–216. 191. Tai CF, Kuo WR, Juan KH et al. Pneumocephalus in a case of nasopharyngeal carcinoma. Kaohsiung Journal of Medical Sciences . 1996 Nov;12(11):646–649. 192. Wang PC, Tu TY, Liu KD. Cystic brain necrosis and temporal bone osteoradionecrosis after radiotherapy and surgery in a patient of ear carcinoma. J Chin Med Assoc. 2004 Sep;67(9):487–491. 193. Cheng KM, Chan CM, Fu YT et al. Brain abscess formation in radiation necrosis of the temporal lobe following radiation therapy for nasopharyngeal carcinoma. Acta Neurochir. 2000;142(4):435–440; discussion 40–41. 194. Wong WC, Cheng PW, Chan FL et al. Improved diagnosis of a temporal lobe abscess in a post-irradiated nasopharyngeal carcinoma patient using diffusion-weighted magnetic resonance imaging. Clin Radiol. 2002 Nov;57(11):1040–1043.

520

Part VII / Neurologic Complications of Specific Malignancies

195. Cheng KM, Chan CM, Fu YT et al. Acute hemorrhage in late radiation necrosis of the temporal lobe: report of five cases and review of the literature. J Neuro-oncol. 2001 Jan;51(2):143–150. 196. Chong VF, Fan YF, Chan LL. Temporal lobe necrosis in nasopharyngeal carcinoma: pictorial essay. Australasian Radiol. 1997 Nov;41(4):392–397. 197. Chong VF, Rumpel H, Aw YS et al. Temporal lobe necrosis following radiation therapy for nasopharyngeal carcinoma: 1H MR spectroscopic findings. Int J Radiat Oncol Biol Phys. 1999 Oct 1;45(3):699–705. 198. Lee AW, Ng SH, Ho JH et al. Clinical diagnosis of late temporal lobe necrosis following radiation therapy for nasopharyngeal carcinoma. Cancer. 1988 Apr 15;61(8):1535–1542. 199. Rajendra T, Lee KS, Leo KW et al. Previously treated nasopharyngeal carcinoma with cystic lesions in the temporal lobe. Singapore Med J. 2004 Dec;45(12):590–593. 200. Tsui EY, Chan JH, Ramsey RG et al. Late temporal lobe necrosis in patients with nasopharyngeal carcinoma: evaluation with combined multisection diffusion weighted and perfusion weighted MR imaging. Eur J Radiol. 2001 Sep;39(3):133–138. 201. Halak M, Fajer S, Ben-Meir H et al. Neck irradiation: a risk factor for occlusive carotid artery disease. Eur J Vasc Endovasc Surg. 2002 Apr;23(4):299–302. 202. Cheng SW, Ting AC, Lam LK et al. Carotid stenosis after radiotherapy for nasopharyngeal carcinoma. J Emerg Med. 2000 Apr;126(4):517–21. 203. Lam WW, Ho SS, Leung SF et al. Cerebral blood flow measurement by color velocity imaging in radiation-induced carotid stenosis. J Ultrasound Med. 2003 Oct;22(10):1055–1060. 204. Lam WW, Leung SF, So NM et al. Incidence of carotid stenosis in nasopharyngeal carcinoma patients after radiotherapy. Cancer. 2001 Nov 1;92(9):2357–2363. 205. Lam WW, Liu KH, Leung SF et al. Sonographic characterisation of radiation-induced carotid artery stenosis. Cerebrovasc Dis. 2002;13(3):168–173. 206. Lam WW, Yuen HY, Wong KS et al. Clinically underdetected asymptomatic and symptomatic carotid stenosis as a late complication of radiotherapy in Chinese nasopharyngeal carcinoma patients. Head Neck. 2001 Sep;23(9):780–784. 207. Dorresteijn LD, Kappelle AC, Scholz NM et al. Increased carotid wall thickening after radiotherapy on the neck. Eur J Cancer. 2005 May;41(7):1026–1030. 208. Abayomi OK. Neck irradiation, carotid injury and its consequences. Oral Oncol. 2004 Oct;40(9):872–878. 209. Call GK, Bray PF, Smoker WR et al. Carotid thrombosis following neck irradiation. Int J Radiat Oncol Biol Phys. 1990 Mar;18(3): 635–640. 210. Cazaban S, Maiza D, Coffin O et al. Surgical treatment of recurrent carotid artery stenosis and carotid artery stenosis after neck irradiation: evaluation of operative risk. Ann Vasc Surg. 2003 Jul;17(4):393–400. 211. Cheng SW, Ting AC, Ho P et al. Accelerated progression of carotid stenosis in patients with previous external neck irradiation. J Vasc Surg. 2004 Feb;39(2):409–415. 212. Cheng SW, Wu LL, Ting AC et al. Irradiation-induced extracranial carotid stenosis in patients with head and neck malignancies. Am J Surg. 1999 Oct;178(4):323–328. 213. Friedell ML, Joseph BP, Cohen MJ et al. Surgery for carotid artery stenosis following neck irradiation. Ann Vasc Surg. 2001 Jan;15(1):13–18. 214. Martin JD, Buckley AR, Graeb D et al. Carotid artery stenosis in asymptomatic patients who have received unilateral head-and-neck irradiation. Int J Radiat Oncol Biol Phys. 2005 Nov 15;63(4):1197–1205. 215. Minion DJ, Lynch TG, Baxter BT et al. Pseudoaneurysm of the external carotid artery following radical neck dissection and irradiation: a case report and review of the literature. Cardiovasc Surg. 1994 Oct;2(5):607–611. 216. Reed R, Sadiq S. Acute carotid artery thrombosis after neck irradiation. J Ultrasound Med. 1994 Aug;13(8):641–644. 217. So NM, Lam WW, Chook P et al. Carotid intima-media thickness in patients with head and neck irradiation for the treatment of nasopharyngeal carcinoma. Clin Radiol. 2002 Jul;57(7):600–603. 218. Brown PD, Foote RL, McLaughlin MP et al. A historical prospective cohort study of carotid artery stenosis after radiotherapy for head and neck malignancies. Int J Radiat Oncol Biol Phys. 2005 Dec 1;63(5):1361–1367. 219. Harrod-Kim P, Kadkhodayan Y, Derdeyn CP et al. Outcomes of carotid angioplasty and stenting for radiation-associated stenosis. AJNR. 2005 Aug;26(7):1781–1788. 220. Ecker RD, Brown RD, Jr., Nichols DA et al. Cost of treating high-risk symptomatic carotid artery stenosis: stent insertion and angioplasty compared with endarterectomy. J Neurosurg. 2004 Dec;101(6):904–907. 221. Marcel M, Leys D, Mounier-Vehier F et al. Clinical outcome in patients with high-grade internal carotid artery stenosis after irradiation. Neurology. 2005 Sep 27;65(6):959–961. 222. Sarkar S, Mehta SA, Tiwari J et al. Complications following surgery for cancer of the larynx and pyriform fossa. J Surg Oncol. 1990 Apr;43(4):245–249. 223. Iguchi H, Takayama M, Kusuki M et al. Carotid artery pseudoaneurysm as a rare sequela of surgery for laryngeal cancer. Acta Otolaryngol. 2006 May;126(5):557–560. 224. Lam HC, Abdullah VJ, Wormald PJ et al. Internal carotid artery hemorrhage after irradiation and osteoradionecrosis of the skull base. Otolaryngol Head Neck Surg. 2001 Nov;125(5):522–527. 225. Lau WY, Chow CK. Radiation-induced petrous internal carotid artery aneurysm. Ann Otol Rhinol Laryngol. 2005 Dec;114(12): 939–940. 226. Prim MP, De Diego JI, Verdaguer JM et al. neurologic complications following functional neck dissection. Eur Arch Otorhinolaryngol. 2006 May;263(5):473–476. 227. Nichols RD, Stine PH, Bartschi LR. Facial nerve function in 100 consecutive parotidectomies. The Laryngoscope. 1979 Dec;89(12):1930–1934.

Chapter 27 / Neurologic Complications of Head and Neck Cancer

521

228. Witt RL. Facial nerve function after partial superficial parotidectomy: an 11-year review (1987–1997). Otolaryngol Head Neck Surg. 1999 Sep;121(3):210–213. 229. Witt RL. Facial nerve monitoring in parotid surgery: the standard of care? Otolaryngol Head Neck Surg. 1998 Nov;119(5):468–470. 230. Watanabe Y, Ishikawa M, Shojaku H et al. Facial nerve palsy as a complication of parotid gland surgery and its prevention. Acta Otolaryngol Suppl. 1993;504:137–139. 231. Bron LP, O’Brien CJ. Facial nerve function after parotidectomy. Arch Otolaryngol Head Neck Surg. 1997 Oct;123(10):1091–1096. 232. Weiss KL, Wax MK, Haydon RC, 3rd et al. Intracranial pressure changes during bilateral radical neck dissections. Head Neck. 1993 Nov–Dec;15(6):546–552. 233. Lydiatt DD, Ogren FP, Lydiatt WM et al. Increased intracranial pressure as a complication of unilateral radical neck dissection in a patient with congenital absence of the transverse sinus. Head Neck. 1991 Jul–Aug;13(4):359–362. 234. Lam BL, Schatz NJ, Glaser JS et al. Pseudotumor cerebri from cranial venous obstruction. Ophthalmol. 1992 May;99(5):706–712. 235. Doepp F, Schreiber SJ, Benndorf G et al. Venous drainage patterns in a case of pseudotumor cerebri following unilateral radical neck dissection. Acta Otolaryngol. 2003 Oct;123(8):994–997. 236. Hildebrand J. Neurological complications of cancer chemotherapy. Curr Opin Oncol. 2006 Jul;18(4):321–324. 237. Tuxen MK, Hansen SW. Neurotoxicity secondary to antineoplastic drugs. Cancer Treat Rev. 1994 Apr;20(2):191–214. 238. Siau C, Xiao W, Bennett GJ. Paclitaxel- and vincristine-evoked painful peripheral neuropathies: loss of epidermal innervation and activation of Langerhans cells. Exp Neurol. 2006 Oct;201(2):507–514. 239. Lee JJ, Swain SM. Peripheral neuropathy induced by microtubule-stabilizing agents. J Clin Oncol. 2006 Apr 1;24(10):1633–1642. 240. Argyriou AA, Chroni E, Koutras A et al. Vitamin E for prophylaxis against chemotherapy-induced neuropathy: a randomized controlled trial. Neurology. 2005 Jan 11;64(1):26–31. 241. Argyriou AA, Chroni E, Koutras A et al. Preventing paclitaxel-induced peripheral neuropathy: a phase II trial of vitamin E supplementation. J Pain Symptom Manage. 2006 Sep;32(3):237–44. 242. Argyriou AA, Chroni E, Koutras A et al. A randomized controlled trial evaluating the efficacy and safety of vitamin E supplementation for protection against cisplatin-induced peripheral neuropathy: final results. Support Care Cancer. 2006 Nov;14(11):1134–1140. 243. Albers J, Chaudhry V, Cavaletti G et al. Interventions for preventing neuropathy caused by cisplatin and related compounds. Cochrane Database System Rev (online). 2007(1):CD005228. 244. Tredici G, Cavaletti G, Petruccioli MG et al. Low–dose glutathione administration in the prevention of cisplatin-induced peripheral neuropathy in rats. Neurotoxicology. 1994 Fall;15(3):701–704. 245. Roberts JA, Jenison EL, Kim K et al. A randomized, multicenter, double-blind, placebo-controlled, dose-finding study of ORG 2766 in the prevention or delay of cisplatin-induced neuropathies in women with ovarian cancer. Gynecol Oncol. 1997 Nov;67(2):172–177. 246. Kanat O, Evrensel T, Baran I et al. Protective effect of amifostine against toxicity of paclitaxel and carboplatin in non-small cell lung cancer: a single center randomized study. Med Oncol. 2003;20(3):237–245. 247. Pace A, Savarese A, Picardo M, Maresca V et al. Neuroprotective effect of vitamin E supplementation in patients treated with cisplatin chemotherapy. J Clin Oncol. 2003 Mar 1;21(5):927–931. 248. Hovestadt A, van der Burg ME, Verbiest HB et al. The course of neuropathy after cessation of cisplatin treatment, combined with Org 2766 or placebo. J Neurol. 1992 Mar;239(3):143–6. 249. van Gerven JM, Hovestadt A, Moll JW et al. The effects of an ACTH (4–9) analogue on development of cisplatin neuropathy in testicular cancer: a randomized trial. J Neurol. 1994 Jun;241(7):432–435. 250. Planting AS, Catimel G, de Mulder PH et al. Randomized study of a short course of weekly cisplatin with or without amifostine in advanced head and neck cancer. EORTC Head and Neck Cooperative Group. Ann Oncol. 1999 Jun;10(6):693–700. 251. van der Hoop RG, Vecht CJ, van der Burg ME et al. Prevention of cisplatin neurotoxicity with an ACTH(4–9) analogue in patients with ovarian cancer. N Engl J Med. 1990 Jan 11;322(2):89–94. 252. Elkiran ET, Altundag K, Beyazit Y et al. Fluorouracil-induced neurotoxicity presenting with generalized tonic–clonic seizure. Ann Pharmacotherapy. 2004 Dec;38(12):2171. 253. Ki SS, Jeong JM, Kim SH et al. A case of neurotoxicity following 5-fluorouracil–based chemotherapy. Korean J Internal Med. 2002 Mar;17(1):73–77. 254. Pirzada NA, Ali, II, Dafer RM. Fluorouracil-induced neurotoxicity. Ann Pharmacotherapy. 2000 Jan;34(1):35–38. 255. Sucker C, Lambers C, Stockschlader M et al. [Neurotoxicity of 5-fluorouracil]. Deutsche medizinische Wochenschrift (1946). 2002 Sep 27;127(39):2011–2014. 256. Langer CJ, Hageboutros A, Kloth DD et al. Acute encephalopathy attributed to 5-FU. Pharmacotherapy. 1996 Mar–Apr;16(2):311–313. 257. Steeghs N, de Jongh FE, Sillevis Smitt PA et al. Cisplatin-induced encephalopathy and seizures. Anti-Cancer Drugs. 2003 Jul;14(6):443–446. 258. Klastersky J. Adverse effects of the humanized antibodies used as cancer therapeutics. Curr Opin Oncol. 2006 Jul;18(4):316–20.

28

Neurologic Complications of Melanoma Denise M. Damek,

MD

CONTENTS Malignant Melanoma CNS Metastases Parenchymal Brain Metastases Leptomeningeal Metastases Spinal Metastases Intramedullary Spinal Cord Metastases Plexus/Peripheral Nerve Metastases Paraneoplastic Disorders Neurologic Complications Related to Therapy Conclusion References

Summary Central nervous system (CNS) metastases manifest clinically in 10–40% of patients with melanoma, and in up to two-thirds of patients at autopsy. While the prognosis of CNS metastases from most cancers is contingent upon control of underlying systemic disease, the vast majority of patients with CNS metastases from melanoma suffer a neurologic death. Preliminary study results suggest that the addition of chemotherapy drugs with good CNS penetration to treatment regimens at the initial management of metastatic melanoma may decrease the incidence of CNS metastases. While palliative therapy is reasonable in patients with concurrent widely disseminated disease, more aggressive multimodality treatment approaches are warranted in patients with good performance status and absent or limited extracranial disease. Key Words: melanoma, brain metastases, leptomeningeal metastasis

1. MALIGNANT MELANOMA 1.1. Melanoma Incidence and Age Groups Malignant melanoma of the skin comprises roughly 10% of all skin cancers, but is responsible for more than 75% of skin-cancer–related deaths. The incidence of melanoma has steadily increased over the last several decades throughout the world at a rate of 5–7% per year (1–3). In the United States, the estimated lifetime risk of developing melanoma has increased from 1 in 105 to 1 in 59 (4–6). Currently melanoma is the sixth most common cancer among both men and women in the United States (7). The American Cancer Institute estimated that in the year 2007, over 48,000 Americans would be diagnosed with in situ melanoma, approximately 60,000 would be diagnosed with malignant melanoma, and over 8,000 would succumb to this disease (7). Although malignant melanoma may affect adults of all age groups, the median age at diagnosis is 59 years (8). However, melanoma is distinctly uncommon in the pediatric population. Fewer than 3% of patients with this malignancy are younger than 20 years of age (8–11). The occurrence of melanoma in prepubertal children is From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

523

524

Part VII / Neurologic Complications of Specific Malignancies

particularly rare, representing between 0.4% and 1% of cases (12–14). Surveillance Epidemiology and End Results (SEER) data from 2000–2004 found that approximately 1.0% of malignant melanoma patients were diagnosed under age 20; 8% between 20 and 34; 14% between 35 and 44; 19% between 45 and 54; 19% between 55 and 64; 18% between 65 and 74; 16% between 75 and 84; and 5% 85+ years of age (8).

1.2. Risk Factors Risk factors associated with the development of malignant melanoma include personal characteristics, ultraviolet light exposure history, and genetic factors. Personal characteristics include pigmentary traits of light colored skin, blue eyes, red hair, and lack of tanning ability; increased number of pigmented skin lesions, either freckles or common acquired nevi; and history of dysplastic nevi or large congenital nevi. Of these personal characteristics, the strongest risk factor is a large number of dysplastic nevi in combination with fair skin. Notably, a prior history of melanoma increases one’s risk of developing cutaneous melanoma by 11–24% (15). History of excessive sun exposure, blistering burns, especially in childhood, and the use of tanning beds correlates with malignant melanoma risk. The risks associated with ultraviolet light exposure differ with age. In children, blistering burns and high levels of sun exposure irrespective of sunburn, correlate best with the development of cutaneous melanoma (16). In adults, however, intermittent high sun exposure to unacclimatized skin is a particular risk factor (17,18). At least 10% of all melanoma cases are familial (19). While some melanomas are derived from an autosomal dominant inheritable syndrome of atypical moles, i.e., familial atypical multiple mole melanoma (FAMMM) (20,21); the pattern of inheritance in most malignant melanoma appears to be multifactoral or polygenic. Germline mutations in three highly penetrant gene products—p16, alternate reading frame (ARF), and cyclin-dependent kinase 4 (CDK4)—are associated with familial melanoma syndromes. Inherited mutations in the melanocortin-1 receptor have been linked to red-headed individuals, photosensitivity, and an increased risk of cutaneous melanoma. Inherited mutations in the CDKN2A and CDK4 genes have been identified in some families with hereditary melanoma (22), and confer a 60–90% lifetime risk of melanoma (23). Cutaneous malignant melanomas have been associated with neurofibromas (24,25) and meningiomas (26), implicating their common neural crest cell origin. Reports of family kindreds with a high incidence of malignant melanoma postulate genetic linkage of malignant melanoma to essential myoclonus (27) and Charcot–Marie– Tooth neuropathy (28). In rare instances, melanoma is associated with other genetic diseases, such as xeroderma pigmentosum and neurocutaneous melanosis (25) and Li–Fraumeni syndrome. Additional risk factors for the development of melanoma include ionizing radiation exposure and immunosuppressant therapy. The etiologic role of immunosuppressants in the development of malignant melanoma is supported by a 2–8 fold increase in melanoma incidence post-transplantation compared with nontransplant patient populations (29–31). Work place exposure to benzene and/or benzene-containing solvents, or vinyl chloride has been linked to the development of melanoma, particularly in non–sun exposed areas, but causality has not been established or corroborated in large studies (32,33). While it is well recognized that patients with Parkinson’s disease have twice the risk of melanoma as that of the general population, significant controversy exists over whether this risk is inherent to the disease itself or associated with levodopa treatment. Levodopa, dopamine agonists, and other pharmacologic agents for the treatment of Parkinson’s disease carry United States Food and Drug Administration (FDA) warning or precaution labels highlighting a potential causal relationship of these drugs with increased melanoma risk. However, many reviews have disputed a causal relationship (34–37), and a recent large epidemiological study concluded that prevalence rates of malignant melanoma increased before the diagnosis of Parkinson’s disease (38). 1.2.1. Early Detection and Prevention The detection of cutaneous melanoma is facilitated by melanocyte pigment production in over 95% of primary lesions. The majority of cutaneous melanomas evolve through a predictable sequence of growth beginning with a visually recognizable but biologically indolent phase of radial growth. The classic features of superficially spreading melanoma are summarized by the “ABCD(E)” rule, an acronym for Asymmetry, Border irregularity, Color variability, Diameter greater than 6 mm, and Elevation or Evolution.

Chapter 28 / Neurologic Complications of Melanoma

525

Melanoma confined to the epidermal skin layer, or associated with superficial invasion of the papillary dermis (melanoma in situ), has no capacity to metastasize. It is at this phase of growth that melanoma is surgically curable with 5-year survival rates approaching 100%. After a melanoma lesion enters its vertical growth phase through the dermal skin layers, a significant decline in 5-year survival rates are seen. Characteristic changes in pigmented lesions on visual inspection of the skin, and the potential for cure with early detection underscores the importance of surveillance in at-risk populations. On the other hand, screening of the general population at large is of uncertain benefit. It is logistically challenging, not cost-effective, and has yet to demonstrate a reduction in mortality from malignant melanoma. In the 1970s, Australia was the first of many countries to institute primary prevention campaigns to promote sunscreen use and limited sun exposure. To date, a direct correlation between preventative measures and a reduction in melanoma has not yet been realized.

1.3. Staging and Treatment Overview When a diagnosis of malignant melanoma has been established, several prognostic factors have been identified to help guide treatment. The current staging system for melanoma was approved by the American Joint Committee on Cancer in 2001 (39,40) (Table 1). In this system, the T stage, or tumor classification, is largely determined by melanoma thickness and the presence of microscopic tumor ulceration pathologically. Tumor thickness is graded from 0 to 4, and ulceration is denoted by the letter b. An increased risk of local recurrence, regional metastases, distant metastases, and decreased melanoma-specific survival is seen with increasing melanoma thickness. In addition, the presence of ulceration signifies a locally advanced lesion and correlates with a higher risk for metastases. Ulceration upgrades the tumor stage by one substage. In patients with melanomas 1.0 mm or less in thickness, the likelihood of detecting nodal disease is approximately 1% and nodal failure is rare (41–43). A limited staging evaluation and treatment with surgical resection alone is indicated. However, if pathological examination identifies negative prognostic factors such as ulceration or involvement of the lower dermis or subcutis (Clark level IV or V), despite thickness less than 1.0 mm, there is an increase in concurrent nodal disease. A sentinel lymph node biopsy would be considered in this situation (39,40,42). Furthermore, in patients with melanomas 4.0 mm or greater in thickness, the likelihood of nodal metastases is 36% (41,42). More extensive staging evaluation and sentinel lymph node biopsy are warranted. The N stage, or nodal classification, is based on the number of metastatic lymph nodes and the delineation of micrometastatic disease based on the sentinel lymph node biopsy technique or elective nodal dissection versus macroscopic nodal metastases. The N category is subdivided 0 to 3 based on the number of lymph nodes involved or the spread of melanoma in skin towards a lymph node area. Microscopic disease is designated by the letter a, macroscopic disease by the letter b, and skin spread towards a lymph node area by the letter c. Nodal status has been established as a powerful predictor of recurrence and melanoma-related survival. In the clinical staging system, lymph node involvement is based on the presence or absence of clinically palpable lymph nodes or radiographic evidence of lymphadenopathy. In the pathologic staging system, lymph node status is determined by a combination of clinical, radiographic, and pathological assessments. Sentinel lymph node biopsy, in which the first draining node from a primary melanoma lesion is sampled, has essentially replaced elective node dissection in patients with stage II disease to determine the presence or absence of occult nodal metastases. In general, the sentinel lymph node is identifiable in more than 95% of cases with less than 5% false negative results (44). Nodal metastases are found in 20–25% of patients with stage II disease, and are present in over 60% of patients with stage III disease. The presence of nodal metastases correlates with incidence of systemic metastases. Patients with stage II disease have less than a 25% incidence of systemic metastases, whereas patients with stage III disease have more than a 70% incidence of occult distant metastases at the time of initial presentation. Patients with positive nodal disease require further therapy with regional or completion lymph node dissection and adjuvant or neoadjuvant chemotherapy. Immediate lymphadenectomy in patients with subclinical sentinel node metastases is associated with improved melanoma-specific 5-year survival rate compared to delayed lymphadenectomy for clinically detectable nodal disease (39,40,45–47).

IV

III

II

I

0

Stage

Any thickness +/– Ulceration

Any thickness +/– Ulceration

IIIB

IIIC

Any thickness +/– Ulceration

Any thickness No ulceration

Between 1.01 and 2 mm thick with ulceration (T2b, N0, M0) Between 2.01 and 4 mm without ulceration (T3a, N0, M0) Between 2.01 and 4 mm thick with ulceration (T3b, N0, M0) Thicker than 4 mm without ulceration (T4a, N0, M0) Thicker than 4 mm thick with ulceration (T4b, N0, M0)

Less than 1 mm thick and involves only upper dermis (T1a, N0, M0) Less than 1 mm thick and involves lower dermis or subcutis or is ulcerated (T1b, N0, M0) Between 1.01 and 2 mm thick without ulceration (T2a, N0, M0)

Melanoma in situ, i.e., tumor involvement limited to epidermis (Tis, N0, M0)

Tumor

IIIA

IIC

IIB

IIA

IB

IA

Substage

Involves 1-3 lymph nodes, without nodal enlargement, and microscopic disease only (T1a–4a, N1a or N2a, M0) If ulceration is present, microscopic involvement of 1–3 lymph nodes without nodal enlargement (T1b-4b, N1a, or N2a, M0) If ulceration is absent, involvement of 1–3 lymph nodes with enlargement of nodes (T1a–T4a, N1b or N2b, M0) +/– ulceration, no lymph node involvement but spread to skin lymphatic channels around original tumor (T1a/b–4a/b, N2c M0, ) If ulceration is present, involvement of 1–3 lymph nodes with nodal enlargement (T1b-4b, N1b or N2b, M0) +/– ulceration and involvement of greater than 4 lymph nodes, clumping of lymph nodes, or spread to local skin lymphatic channels and lymph nodes (Any T, N3, M0) Absent or present lymph node involvement clinically or pathologically

—

—

27–52%

—

18%

50–68%

—

Present (Any T, Any N, M1)

Limited data

56%

68%

78%

92%

99%

Almost 100%

5-Year Survival

—

—

—

—

—

—

—

—

—

Metastases

—

—

Nodes

Table 1 American Joint Committee on Cancer (AJCC) TNM System with Survival Rates by Stage

14%

22–37%

44–60%

Limited data

48%

59%

66%

86%

97%

Almost 100%

10-Year Survival

Chapter 28 / Neurologic Complications of Melanoma

527

The M stage, or distant metastases classification, is further refined by site(s) of distant metastases and the presence of elevated serum lactic dehydrogenase (LDH). Survival differences between the M categories are small. Patients with distant cutaneous, subcutaneous, or distant lymph node metastases have a better prognosis than other sites of metastases with a one-year survival rate of 59% (39,40). Lung metastases portend intermediate prognosis with one-year survival rate of 57%, while other visceral metastastases have a relatively worse prognosis with one-year survival rate of 41% (39,40). Elevation of serum LDH above the upper limits of normal on at least two occasions more than 24 hours apart at the time of staging is associated with poor prognosis irrespective of site of distant metastases with one year survival rate of 41% (39,40). Stage IV disease has a uniformly poor prognosis, with a median survival of 7.5–7.9 months (48,49). Aggressive treatment approaches often include high-dose bolus interleukin-2 or immunotherapy combined with chemotherapy (i.e., biochemotherapy). Additional therapeutic approaches to metastatic melanoma include supportive care only, surgical resection with or without adjuvant radiation, limb perfusion, chemotherapy, or experimental drug trials. According to SEER data, using historical staging over the time period of 1996–2002, 80% of patients had localized disease at presentation, 12% regional disease, and 4% distant metastases, and 4% of patients had unknown staging information (7). The corresponding 5-year relative survival rates were 99 % for localized disease; 65% for regional disease; 15% for distant metastatic disease (7). Since the publication of the revised AJCC staging classification for melanoma (39,40) several molecular factors with potential clinical and prognostic significance have been reported. Nuclear receptor coactivator-3 (NCOA3) gene overexpression may predict sentinel lymph node metastasis and increased tumor burden, with correspondingly decreased relapse free survival and disease-specific survival (50). The identification of circulating malignant cells via mRNA expression of tyrosinase and MelanA/MART1 in peripheral blood may have prognostic value (51). In addition, the use of radiographic staging with computed tomography and magnetic resonance imaging may be of prognostic value in patients with thick and/or ulcerated primary melanoma lesions (52).

1.4. Patterns of Distant Metastases Melanoma can metastasize to virtually any body organ. At the initial diagnosis of stage IV disease, 18–36% of patients have disease within the lung, 14–20% within the liver, and 12–20% within the brain. Distant sites of metastases found at autopsy in decreasing order of frequency are lung, liver, bowel, brain, kidneys, and bone (53). Brain involvement usually manifests late in the disease course, occurring at disease relapse or progression. At malignant melanoma relapse, brain metastases are the first site of distant metastasis in 12–20% of patients (53).

1.5. Metastatic Melanoma of Unknown Primary Site In large series of metastatic malignant melanoma, patients with an unknown site of primary melanoma constitute 2–6% of patients with metastatic disease; figures as high as 15%, however, have been reported (54–60). About two-thirds of patients with occult primary melanoma have regional lymph node metastases. The remaining onethird have distant metastases involving skin, subcutaneous tissue, lung, or brain (61). Patients presenting with “primary” cerebral melanoma invariably develop multiple systemic metastases (62,63). Outcomes are similar between these patients and those patients with known primary sites when matched for prognostic factors (61).

2. CNS METASTASES 2.1. Incidence As treatment advances increase survival times in some types of cancer, the frequency of cerebral metastases is escalating (64–66). While malignant melanoma is the fourth most common cause of central nervous system metastases after lung and breast carcinomas and unknown primary tumors, it has the highest propensity to metastasize to the brain (67–70). Clinical data from large databases report an incidence of CNS metastases in patients with malignant melanoma of 6–10% (63,71–73). However, if the data from smaller series is incorporated, the incidence ranges from 6% to 43% (68,74–78). Autopsy series report brain metastases in 12–74% of patients succumbing to metastatic melanoma (75,77,79–85).

528

Part VII / Neurologic Complications of Specific Malignancies

2.2. Co-existent Extracranial Disease Brain metastasis in malignant melanoma generally occurs in patients with concurrent extracranial disease. However, the brain is the only site of distant metastases in 20–60% of malignant melanoma patients with brain metastasis (86–90). In one case series, isolated brain metastases occurred in less than 5% of cases (91). The presence or absence of extracranial disease appears related to the multiplicity of brain metastases. While the majority of patients with metastatic brain disease have multiple lesions, 20–60% of patients will have single metastatic deposits (63,91–99). Some 40–45% of patients with single brain metastases have co-existing systemic disease compared with 80% of patients with multiple brain metastases (88,99).

2.3. Risk Factors for Developing CNS Metastases in Malignant Melanoma Several factors have been associated with the development of central nervous system disease in malignant melanoma. Primary melanomas involving the head and neck regions are the most likely to spread to the central nervous system (73,84,92,96,100,101). Brain metastases are more common in males than females (73,87,90,95, 96,100,102), perhaps because primary melanoma lesions in males more often develop in the head, neck, and trunk regions. The presence of lymph node or visceral metastases at the time of diagnosis is associated with a higher risk of developing brain metastasis (63). Nearly half of patients with multiple visceral metastases have concomitant brain involvement or will subsequently develop CNS metastasis. Specific features of primary melanoma lesions that are associated with the development of brain metastases include acral lentiginous or nodular histological types (63), deeply invasive primary lesions or primary lesions located on mucosal surfaces (63,73,92), and primary lesions that are ulcerated or have large superficial diameters (63,73). Overall, these factors identify a population at risk for the widespread metastatic dissemination of disease rather than specifically delineating a population at risk for brain metastases (63).

2.4. Interval Between Initial Diagnosis and Brain Metastases Central nervous system metastases rarely herald the diagnosis of malignant melanoma. Instead, they typically occur later in the disease course and often represent the first metastatic focus (97,103,104). The interval between the diagnosis of a primary skin lesion and the first appearance of CNS involvement varies according to the initial site of the primary lesion, occurring earlier in those patients with head and neck primary tumors. The first appearance of neurological signs or symptoms of CNS involvement typically manifests 2–4 years after diagnosis (63,73,99,100,105).

3. PARENCHYMAL BRAIN METASTASES 3.1. General Patients with brain metastases typically have advanced, widespread, and uncontrolled systemic cancer. Not surprisingly, more than 70% of patients with brain metastases from grouped histologies eventually succumb to their systemic disease. In contrast, brain metastases from malignant melanoma directly contribute to patient death in up to 95% of patients (63). In addition, compared with patients with other forms of brain metastasis, patients with melanoma tend to be younger (< 50 years), often have excellent performance scale scores, and have a longer disease-free survival between the initial diagnosis and development of metastatic disease (62). Because patients with brain metastases from melanoma usually die from neurological sequelae, most clinical research trials exploring treatments for metastatic melanoma exclude patients with brain metastases. It is notable that rare case reports describe long-term survivals of patients with melanomatous brain metastases, ranging between 3 and 18 years (87,106–110). Brain metastases carry the worst prognosis of all distant metastases in patients with melanoma (90,111–113). Central nervous system metastases occur in up to 75% of all patients who die from melanoma and contribute to morbidity and mortality in one-half of all patients with disseminated disease (61,63,81,84,85,100). Of patients with brain metastases from malignant melanoma, those with metastatic disease limited to the brain have the best outcomes (63,90,99,100,103,114,115).

Chapter 28 / Neurologic Complications of Melanoma

529

The presence of widely disseminated disease portends a poor prognosis for patients with brain metastases (63,88). However, the complete surgical resection of metastases at other visceral sites may improve survival (90,111,116–120). Patients with brain metastases and concomitant single organ extracranial disease have survival rates comparable to patients with brain involvement alone. Exceptions are coexistent lung or liver metastases, which result in a significantly worse prognosis compared to other sites of single organ involvement (63,90,100). General negative prognostic indicators derived from a recursive partitioning analysis of prognostic factors in patients with brain metastases from grouped histologies include poor performance status, the presence of systemic tumor activity, and age over 70 years (121) (Table 2). Specific favorable prognostic indications derived from a recursive partitioning analysis in a large review of patients with brain metastases from melanoma include solitary brain metastasis, patients without lung or visceral metastases, and patients whose brain metastases heralded the diagnosis of melanoma (122). In addition, there are some data suggesting that patients who do not improve neurologically with steroid therapy have poorer response rates to radiotherapy and decreased overall survival rates (121).

3.2. Clinical Presentation The majority of brain metastases are symptomatic during the lifetime of the patient. The signs and symptoms of brain metastases vary according to tumor growth rate and localization within the brain. Tumors can produce symptoms directly via destruction of neurons or by the compression of surrounding brain structures. Indirect tumor effects include seizures related to the focal irritation of neurons, mass effect related to vasogenic edema surrounding the tumor, or hydrocephalus produced by obstruction of ventricular outflow. Brain metastases from melanoma are associated with seizures in nearly half of patients, which is considerably higher than the 25% incidence seen with brain metastases from other primaries (91). Other common presenting symptoms include headache, focal weakness, cognitive impairment, and behavioral disturbances. Symptoms inconsistent with tumor location on imaging studies should raise suspicion for leptomeningeal spread or multifocal tumor.

3.3. Radiographic Findings Similar to other primary tumors, brain metastases from malignant melanoma have a predilection for localization to the gray–white matter junction and are distributed roughly in proportion to brain volume and blood flow distribution. As such, 80–85% of metastatic lesions involve the cerebrum, 10–15% the cerebellum, and 3–5% the brainstem (95,102,123–125). A large autopsy study of patients with cerebral metastases from all histologies found a single mestastasis in 40%, two or three lesions in 27%, and four or more lesions in 33% (126). Single metastases are present in one-fourth to one-third of patients on magnetic resonance imaging (MRI) studies (127). Compared to other primary tumors, melanoma has the highest tendency to produce multiple lesions (76,82,109,128).

Table 2 Radiation Therapy Oncology Group (RTOG) Recursive Partitioning Analysis (RPA) Class and Survival Primary Lesion Status Class: I II III

Controlled Uncontrolled Controlled Controlled Any

Presence of Extracranial Metastases No Yes or no Yes or no Yes Any

Age

Karnofsky Performance Scale Score

Median Survival for All Histologies*

Median Survival for Melanoma**

< 65 years Any age ≥ 65 years < 65 years Any

≥ 70 ≥ 70 ≥ 70 ≥ 70 < 70

7.1 months 4.2 months

10.5 months 5.9 months

2.3 months

1.8 months

* Gaspar L, Scott C, Rotman M et al. Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys 1997;37:745–751. ** Buchsbaum JC, Suh JH, Lee S-Y et al. Survival by Radiation Therapy Oncology Group recursive partitioning analysis class and treatment modality in patients with brain metastases from malignant melanoma: a retrospective study. Cancer 2002;94:2265–2272.

530

Part VII / Neurologic Complications of Specific Malignancies

The superiority of gadolinium-enhanced MRI over contrast-enhanced computerized tomography (CT) in the diagnosis of brain metastases is unquestioned. MRI is particularly sensitive in the detection of smaller metastatic deposits and infratentorial lesions (129–131). Melanoma metastases are often heterogeneous in appearance, demonstrating contrast enhancement as well as necrosis and hemorrhage within individual lesions. These tumors characteristically show increased signal intensity on noncontrast T1-weighted imaging sequences as a consequence of methemoglobin and melanin. Intraoperatively, the hyperpigmented lesions of melanoma metastases grossly appear well demarcated from the surrounding brain tissue. However, microscopically these tumors can display infiltrative tendencies (132). Microscopic invasion may play an importance role in recurrence after gross total resection of single metastasis lesions.

3.4. Intratumoral Hemorrhage Acute bleeding into a metastasis represents a small proportion of intracerebral hemorrhage. 0.6–1% of hematomas in neurosurgical series (133,134) and 5.5% in one autopsy series (135) are attributable to intratumoral hemorrhage. Despite this, intratumoral hemorrhage can be a significant factor in the management of some types of brain metastases. Up to 20% of adult patients with brain metastases may have a catastrophic event secondary to intracranial bleeding (125,136). Neoplastic hemorrhage is most common in patients with malignant melanoma and germ cell tumors (134,137–143). Clinical series report intratumoral hemorrhage in 25–50% of patients with brain metastases from malignant melanoma (62,68,91,105,125,136,139,144). The majority of patients with intratumoral hemorrhage present with subacute progressive symptoms typical of nonhemorrhagic brain metastases (139). In some patients, the clinical presentation may mimic an evolving stroke with abrupt onset of headache followed by focal neurologic signs and progressive obtundation. Others exhibit acute worsening of pre-existing neurological symptoms (139,143,145,146). In one series, a significant proportion of patients were asymptomatic (137). Medical conditions often associated with intracerebral hemorrhage, such as hypertension or thrombocytopenia, do not appear to contribute to intratumoral hemorrhage in malignant melanoma brain metastases (144). Up to 80% of brain metastases from melanoma have evidence of macroscopic bleeding (62,139). When multiple metastases are present, simultaneous bleeding into most or all metastases is more common than an isolated hemorrhage (139). Histopathologic evidence of prior hemorrhage is found in up to 25% of presumptive nonhemorrhagic metastases (62). Hemorrhage may be confined to the metastatic brain tumor itself, occur in the area immediately adjacent to the tumor, or exist as an intracerebral hematoma intermixed with scattered tumor fragments. Both peripherally located and deep tumors can rupture to the surface of the brain or intraventricularly (143). The mechanism of intratumoral hemorrhage is multifactorial, involving loss of vessel integrity associated with inter-and intratumoral necrosis and neovascularization (144). Lesion size appears unrelated to bleeding tendency. Hemorrhagic metastatic tumors range in size from microscopic tumor deposits to 8 cm in diameter (143). Imaging characteristics suggestive of intratumoral hemorrhage include the presence of additional noncontiguous enhancing lesions, hemorrhage occurring outside of the basal ganglia region (139,147), and enhancement adjacent to the blood clot (139,145). Brain MRI may help differentiate neoplastic from non-neoplastic hemorrhages. Neoplastic lesions do not display the usual pattern of hematoma resolution on serial imaging studies, and often have a heterogeneous intensity pattern on spin-echo sequences (148). 3.4.1. Subdural Hematoma Metastatic subdural hematomas are associated with neoplastic infiltration of the dura at the level of the hematoma (198), and are associated with tumors, such as melanoma, that have a proclivity to spread peripherally. One case series reports subdural hematoma in approximately 50% of patients with cerebral melanoma (149), but this is not typical. Tumor contact with the leptomeninges promotes the development of an arterial supply from meningeal vessels (146,150), and neovascularization at this level is postulated to be an etiological factor. Of note, thrombocytopenia is rarely associated with subdural hematomas of nonleukemic origin (139). The majority of metastatic subdural hematomas are discovered as incidental imaging findings. Nonspecific symptoms of acute confusion and lethargy are present in one-fourth of patients. Less than 10% of patients have focal neurologic signs or symptoms (139).

Chapter 28 / Neurologic Complications of Melanoma

531

3.5. Treatment Overview and Prognosis Treatment options for patients with cerebral metastases depend largely on the number and the size of metastatic deposits as well as the extent of extracranial disease. Most patients survive 2–5 months after diagnosis of CNS metastases from malignant melanoma (70,151,152); however, isolated case reports note patients with solitary brain metastases surviving greater than 5 years (153–155). Without treatment, the median survival of patients with brain metastases from malignant melanoma is 3 weeks to 2 months (95,96,156,157). Whole-brain radiation therapy (WBRT) may alleviate neurological symptoms and improve quality of life; however, it has a limited impact on life expectancy, improving median survival rates to between 2 and 6 months (87,91,93,100,103,158–164). Surgical resection of solitary or symptomatic brain metastases increases median survival to between 5 and 22 months (78,86,87,91,96,100,111,113,156,161,165–169). Similar, if not better, results are seen with radiosurgery (88,91,99,115,121,157,161,170–175). Whole-brain irradiation following surgical resection or radiosurgery is of uncertain benefit. Some reports suggest a delay in time to neurologic disease progression and an overall trend towards improved regional disease control and survival (88,136). The role of chemotherapy, immunotherapy, and tumor vaccines in the treatment of brain metastases from melanoma is growing, but remains less well defined. 3.5.1. Whole-Brain Radiation Therapy for Brain Metastases For decades, whole-brain radiation therapy has been considered the standard of care for melanoma patients with brain metastases. It is commonly used as the only mode of therapy in patients with multiple cranial metastases, or as adjunct therapy following surgical resection or stereotactic radiosurgery. Most of the data for the use of WBRT in melanoma patients is derived from retrospective and nonrandomized trials. While WBRT essentially doubles the median survival compared to supportive care, the overall survival benefit is modest. Reported median survival is in the range of 2–6 months (87,91,95,100,103,151,158–164). As such, palliation of neurological symptoms is the primary treatment goal of whole-brain irradiation. Roughly 50% of treated individuals are able to successfully taper off of steroids following WBRT, implicating at least short-term symptomatic benefit (151). While whole-brain radiation therapy addresses both gross tumor as well as microscopic tumor deposits within seemingly normal brain tissue, the large treatment volume limits the total radiation dose, potentially compromising local tumor control. The minimum radiation dose required to achieve antitumor effect is 30 Gy (176). Relative contraindications to radiation therapy include prior whole-brain irradiation, active intracranial infection, and severe collagen vascular or cerebrovascular disease (177). Despite the relative radioresistance of melanomatous metastases, a subset of treated patients will achieve tumor stabilization or manifest a reduction in tumor volume following WBRT (151). In radioresistant brain metastases, WBRT increases local control rates and decreases the distant failure rate (178,179). However, high rates of local recurrence are reported. Over one-half of patients demonstrate local failure, which ultimately contributes to patient death (162,180,181). Acute neurological side effects occurring during radiation therapy typically result from increased perilesional vasogenic edema. Corresponding symptoms may include headaches, nausea, lethargy, increased intracranial pressure, seizures, and worsening of focal neurological deficits. Early delayed side effects result from damage to myelin and manifest as a self-limited fatigue syndrome 1–4 months following completion of radiation. A significant proportion of surviving patients will develop late effects of radiation. Symptoms result from radiation damage to normal brain tissue and usually occur 6–12 months after whole-brain irradiation. Approximately 11% of 1-year survivors and up to 50% of 2-year survivors develop dementia (182,183). Other potential late effects include radiation necrosis, cerebral atrophy, and leukoencephalopathy (182–184). 3.5.2. Surgery for Brain Metastases The goals of surgical resection are improvement or resolution of neurological deficits and increased survival. Surgical management of brain metastases is generally limited to those patients with a solitary brain lesion, well-controlled or limited systemic disease, and good performance status. However, patients with multiple brain metastases in whom all of the lesions can be surgically excised may realize a prognosis similar to that of patients

532

Part VII / Neurologic Complications of Specific Malignancies

with a solitary resectable lesion (185,186). In addition, surgical resection is often considered if one of multiple brain metastases is life-threatening or symptomatic, or when a pathologic diagnosis is required. Improvement of neurological symptoms or deficits is reported in 30–100% of patients undergoing surgical resection (62,90,99,115,156). However, surgical resection is not without a small but tangible risk of morbidity and mortality. Increased postoperative neurologic deficits are seen in less than 5% of patients (86,183,185,187). Other operative complications, including infections and hematomas, occur in 8-9% of all craniotomies for brain metastases (86,183,185,187). In addition, an estimated 10% of surgically treated will develop thromboembolic complications post-surgery (188). Less than 3% of patients die during surgery or in the immediate post-operative period (183). Gross total resection is essential to fully realize the benefits of surgical intervention (90,99,115,123,156). Some 31–48% of surgically treated patients will develop recurrence in the brain (62,183,185,189,190). For these patients, second resections may provide symptomatic relief (63,90) and can result in increased survival time (191). In addition, post-operative whole-brain irradiation may provide a survival advantage. Several nonrandomized studies, including some specifically evaluating melanoma brain metastases, have found increased survival and/or improved local disease control in those patients treated with post-operative wholebrain radiotherapy (88,99,115,169,192). However, the survival benefit was limited and did not reach statistical significance in any of the reports. One study evaluating patients with limited or no systemic disease at the time of craniotomy reported a median survival of almost 20 months when post-operative radiation doses ranged between 4000 and 5000 cGy (88). This result was not confirmed in other series. No prospective randomized trials have specifically evaluated surgical resection for brain metastases from melanoma. Three such trials, however, have evaluated the role of surgery as an adjunct to whole-brain radiotherapy for patients with a single brain metastasis from grouped histologies (187,193,194). In two of these trials, patients treated with surgical resection followed by whole-brain irradiation lived longer, had fewer local recurrences within the CNS, and had fewer neurological symptoms referable to brain metastases compared to patients receiving radiation therapy alone (187,193). However, both of these trials studied a select patient population with controlled or limited extracranial disease. Notably, subgroup analysis in one of these trials found that patients with active systemic disease had a poorer prognosis (187). Moreover, in a similar trial evaluating a more broad patient population which included patients with advanced systemic disease and poorer performance status, the combination of surgical resection and brain irradiation showed no survival benefit over radiation alone, but did decrease the distant failure rate and incidence of neurological death (194). Overall, based on available data, strong consideration of whole brain irradiation is warranted following the surgical resection of single or multiple brain metastases from melanoma. 3.5.3. Radiosurgery for Brain Metastases Radiosurgery is an external irradiation technique that provides a single high dose of radiation to a small target volume. Three types of radiation technologies are utilized—high-energy x-rays produced by linear accelerators, gamma rays produced by the Gamma Knife, and charged particles such as protons produced by cyclotrons. The hallmark of these radiation techniques is the rapid fall-off of the radiation dose at the target edge, minimizing dose delivery to the surrounding normal brain. Target lesions for radiosurgery should be generally less than 4 cm in maximal dimension and at least several millimeters removed from critical structures such as the optic chiasm. Brain metastases are particularly amenable to radiosurgical treatment because they are generally small, spherical lesions that exhibit minimal invasion of surrounding tissue. Many patients with brain metastases are not candidates for surgical resection because of lesion location, multiplicity of tumors, or confounding medical issues. Radiosurgery is an appealing treatment option under these conditions. When used as initial therapy, radiosurgery is either provided by itself or as a boost with whole-brain irradiation. The primary goal of radiosurgery provided as monotherapy is to achieve control of intracranial disease without the neurocognitive side effects of whole-brain irradiation. The intent of radiosurgery provided as a boost is to improve local disease control. In addition, radiosurgery has a role in the treatment of recurrent brain metastases following whole-brain irradiation or surgical resection. Three randomized trials examining the use of whole-brain radiotherapy with or without radiosurgery boost in selected patients with brain metastases report local control rates at one year from 82% to 92% (195–197).

Chapter 28 / Neurologic Complications of Melanoma

533

Retrospective studies of patients with metastatic brain tumors treated with radiosurgery report crude local control rates of 73–98% with median follow-up of 5–26 months (157,174,198–201), and 1-year actuarial freedom from progression rates ranging from 67% to 80% (172,201–204). There is no apparent correlation between local control and number of lesions treated or type of histology (202). Radiosurgery may provide better tumor control in relatively radioresistant tumor types, such as melanoma or renal cell carcinoma, compared to other tumor histologies (172,173). However, in one report, univariant analysis found no differences between melanoma and adenocarcinoma, and multivariate subset analysis for differences in radiosurgical parameters and other prognostic factors showed significantly worse freedom from progression for melanomas (203). Improved local control does not necessarily equate to prolonged survival. Statistically insignificant trends toward improved survival are reported in most radiosurgical series. While lesion multiplicity is not related to local control rates, it may be linked to survival. Patients with multiple lesions may fare worse than patients with single lesions (177). Patients with one to two brain metastases treated with radiosurgery survive 6–16 months (173,174,178,205–216), while those with three or more brain metastases generally survive only 3–6 months (206,216). However, one retrospective review of patients receiving SRS with four or more intracranial metastases reported median overall survival of 8 months (217). In this series, the total treatment volume, rather than the number of metastases, was the most important predictor of survival (217). In another report, however, total intracerebral tumor volume was associated with significantly longer survival only for patients with 1–3 cerebral metastases, and not for patients with ≥4 metastases (216). Radiographic appearance, in particular ring-enhancing and heterogeneous patterns of lesion enhancement at the time of radiosurgery correlate negatively with both radiographic response and freedom from progression (203,218,219). Brain metastases, with the exception of small cell carcinoma and melanoma, generally have a well-defined border. Not surprisingly, histopathologic tumor infiltration beyond the lesion border of ≥1 mm was found in a significant percentage of small cell lung cancer and melanoma brain metastases in one study (220). Currently there is no consensus among radiation oncologists regarding the clinical target volume for the stereotactic radiosurgery treatment of brain metastases. Some plan treatment based on only the area of contrast enhancement while others add a margin of 1–2 mm to include any invisible infiltrative growth. One study comparing these two approaches found a significant improvement in local control if a 1 mm treatment margin was added to the radiographically visible tumor border (221). The impact of microscopic tumor infiltration and inconsistent radiosurgical target volumes on the outcome of clinical trials evaluating stereotactic radiosurgery of brain metastases remains speculative at best. Complications of radiosurgery occur in < 10% of patients reported on in the published literature (222). Most studies do not report significant side effects from the radiosurgical treatment of brain metastases, likely attributable to the relatively short survival period following treatment. When reported, early radiation reactions are the most common. Brain edema can result in transient worsening of pre-existing neurologic symptoms, increased intracranial pressure, and breakthrough localization related seizures. These symptoms generally resolve with conservative management including corticosteroid therapy and anticonvulsant medications. The incidence of acute effects is not known, but small retrospective studies suggest that 3–8% of patients experience acute side effects (177). Radiation necrosis is estimated to occur in 5–10% of patients with brain metastases, but occurrence in up to 18% of treated patients has been reported (198,206,211,223). However, because patients may succumb to their systemic disease before any radiation-induced toxicities are noted, the actual incidence may be higher. In one study, the development of symptomatic radiation necrosis requiring reoperation for increasing mass effect and steroid dependency occurred in 6% of patients (173). Misinterpretation of radiation necrosis as progressive tumor on imaging studies may further confound the issue. Risk factors for developing radiation necrosis include tumor volumes greater than 3 cm3 and prior whole-brain radiotherapy to doses greater than 40 Gy (198). Three randomized clinical trials have examined the benefit of the addition of a radiosurgical boost to WBRT in the treatment of patients with up to four brain metastases ≤4 cm in diameter and a Karnofsky performance scale score ≥70 (195–197). All three studies report improved local brain control in patients treated with whole-brain irradiation and radiosurgery boost compared to WBRT alone; however, no statistically significant improvement

534

Part VII / Neurologic Complications of Specific Malignancies

in overall survival was seen. Similar results were reported in several retrospective reviews assessing the provision of whole brain irradiation with radiosurgery boost in patients with brain metastases from melanoma (136,207,208, 210,215,224). In addition, the Andrews et al. study also evaluated a subset of patients with single brain metastases that were not eligible for surgery either because the lesion was unresectable or the patients themselves were not operative candidates (196). In this patient group with solitary metastasis, radiosurgery boost with WBRT did provide a survival advantage over WBRT alone (196). Although quality of life was not formally assessed using validated instruments, a statistically significant decrease in steroid requirement and improvement in performance status in patients in the radiosurgery treatment arm was noted (196). Two randomized trials have evaluated radiosurgery alone as initial therapy, comparing radiosurgery alone to whole-brain radiotherapy with radiosurgery boost in patients with 1–4 brain metastases (197,225). These trials are currently reported only in abstract form, and available data is incomplete. Improved local and intracranial disease control was seen with the addition of whole-brain irradiation to radiosurgery, although this did not result in a survival advantage. Similar results are reported in additional nonrandomized prospective or retrospective studies (206,226–233). Patients with parenchyaml brain melanoma appear to respond to radiosurgery as well as patients with relatively more radiosensitive tumors. In patients with melanoma brain metastases treated with stereotactic radiosurgery, the overall median survival time ranged from 5.7 to 16.2 months (178,207–210,212–215,234,235). A prospective multi-institutional phase II trial evaluated the provision of radiosurgery alone for patients with radioresistant tumors (renal cell carcinoma, melanoma, and sarcoma) with one to three brain metastases (236). A high intracranial failure rate was seen, but overall survival (8.2 months) was comparable to that of whole-brain irradiation with a radiosurgery boost. Similar findings are reported in retrospective series assessing up-front radiosurgery specifically in melanoma brain metastases (201,209,235,237). While most clinical trials reporting on radiosurgery emphasize local disease control rates, it should be noted that radiosurgery may induce objective radiographic responses even in tumors considered relatively radioresistant such as melanoma. One series evaluating radiosurgery in melanoma brain metastases reported that 61% of lesions treated with radiosurgery regressed by more than 50% of pretreatment volume and 13% resolved completely (235). Currently based on data from these studies, many propose initial treatment of brain metastases with either surgical resection or radiosurgery, deferring whole-brain radiotherapy for salvage therapy at brain metastases recurrence. However, the effect of whole-brain radiotherapy on local tumor control in the setting of recurrent metastases following radiosurgery remains uncertain (172). Another argument in favor of reserving whole-brain radiotherapy for salvage therapy is the potential to avoid the neurocognitive and neuropsychiatric toxicities of whole-brain irradiation. While surrogate markers of quality of life, such as performance status and ability to taper steroids, have been evaluated in some patient cohorts treated solely with radiosurgery, validated quality of life outcomes have not been reported (196,225). One prospective phase II study formally addressed the quality of life of 20 patients undergoing radiosurgery treatment for cerebral metastases using the Spitzer Quality of Life Scale (238). Notably, 75% of patients in this cohort had previously received whole-brain radiotherapy. In this small series, quality of life scores decreased in patients with intracranial and extracranial tumor relapse and progression, but increased or remained stable in patients without evidence of tumor progression. Further study is needed. Only a few small retrospective studies have evaluated radiosurgery as salvage therapy for patients with disease progression following WBRT. In selected patients, radiosurgery can achieve a tumor response with 1-year survivals ranging from 26% to 40% (239). However, these results may be in part due to selection bias. Overall, radiosurgery remains a viable option for metachronous brain lesions after initial treatment in patients with controlled systemic disease and good performance status. 3.5.4. Radiosurgery Versus Surgical Resection Both surgical resection and radiosurgery offer a rapid means of achieving local tumor control. This is of particular advantage for patients being considered for study protocols which exclude patients with uncontrolled central nervous system disease. Radiosurgery, however, offers reduced morbidity and decreased health care costs

Chapter 28 / Neurologic Complications of Melanoma

535

(240). Local tumor control rates for radiosurgery equals or surpasses those reported for surgical resection with or without fractionated radiotherapy (163,172,187,239,241,242). In patients managed at presentation, the combination of whole-brain irradiation with either radiosurgery or surgical resection improves local tumor and regional disease control but not survival. In addition, combination therapy delays the time to neurologic relapse within distant sites of the brain and the leptomeninges (88). A direct comparison of radiosurgery and surgical resection for the treatment of solitary brain metastases is not likely as a randomized trial designed to address this issue was closed early due to poor patient accrual (202). 3.5.5. Chemotherapy With the exception of germ cell tumors and possibly breast carcinoma and small cell lung cancer, chemotherapy for the treatment of brain metastases has not yet demonstrated a survival advantage (243–248). Therefore, its role in the treatment of melanoma brain metastases is largely palliative. Successful chemotherapeutic agents must not only demonstrate activity against melanoma, but also must adequately penetrate the blood–brain barrier. Throughout the world, dacarbazine (DTIC) is a standard chemotherapy agent for the systemic treatment of advanced stage melanoma. Combination chemotherapy regimens containing DTIC are frequently used; however, studies have not shown increased response rates or improved survival with the combination regimens compared with DTIC monotherapy. Responses to chemotherapy in metastatic melanoma appear to be disease site dependent. Patients with skin, subcutaneous tissue, and lymph node involvement respond most frequently to DTIC with responses in 25–35% of cases, compared to 15% in patients with lung metastases, and 5–10% in those patients with bone, liver, and brain metastases (249). The median duration of treatment response is 3–6 months. The lack of efficacy against brain metastases is not surprising given the poor CNS penetration of dacarbazine. Other single agents with activity in metastatic melanoma include the platinum compounds, nitrosureas, and tubular toxins. All show similar activity, with systemic response rates of 10–15% (250). Scant reports metastatic brain melanoma responses to these agents exist (251–255). However, when administered to patients with coexisting brain metastases, these regimens, in particular high-dose therapy with bone marrow rescue, were often associated with an increased risk of hemorrhage within the brain metastases directly contributing to mortality rates as high as 25% (254). Biochemotherapy regimens, or the combination of multiagent cytotoxic chemotherapy and biologic response modifiers such as interleukin 2 (IL-2) and interferon-alpha (INF), are increasingly provided for metastatic melanoma. In clinical studies, response rates in systemic disease of greater than 50% have been achieved, with duration of responses from 5 to greater than 12 months (256). Similar to dacarbazine, IL-2 based immunotherapy regimens have not demonstrated efficacy in the CNS (256,257). However, depending upon the chemotherapy agents utilized, there have been reports of CNS response to biochemotherapy. A report of combination chemotherapy (carmustine, cisplatin, dacarbazine) with interleukin-2 and interferonalpha observed partial radiographic responses in 7 (47%) of 15 patients with melanoma brain metastases. The median time to disease progression and median survival in this subgroup was 6 and 6.5 months, respectively (258). A second study evaluated cisplatin and IL-2 based regimens of concurrent biochemotherapy in selected patients with active or pretreated brain metastases (259). No difference was found in response rates, toxicity, time to progression, or overall survival comparing patients without initial brain metastases to those with brain metastases. Notably, 5 of 26 patients with pretreatment CNS involvement were alive 24 months after treatment. Multivariant analysis found decreased survival in patients with untreated brain metastases compared to those receiving surgery or radiosurgery for their brain metastases prior to biochemotherapy. This, in conjunction with a small number of long-term survivors, has led the authors to recommend therapy of CNS metastases prior to the initiation of systemic biochemotherapy in those patients with pretreatment CNS involvement. Temozolomide, a prodrug of monomethyl 5-triazeno imidazole carboxamide (MTIC), which is the active metabolite of DTIC, has demonstrated activity against melanoma. In phase I and II trials, patients with advanced metastatic melanoma achieved response rates of 17–21% with single-agent temozolomide (260,261). A large randomized phase III study comparing temozolomide with dacarbazine in patients with metastatic melanoma found similar response rates between the two groups and a statistically insignificant favorable trend toward improved survival in the temozolomide treatment group (262). However, the response rate did not exceed 15%.

536

Part VII / Neurologic Complications of Specific Malignancies

Overall, these studies demonstrates that temozolomide is at least equal in efficacy to DTIC for metastatic melanoma. Patients with CNS disease were excluded from these initial studies. However, unlike DTIC, temozolomide does penetrate the blood–brain barrier. This has formed the basis of the hypothesis that replacing DTIC with temozolomide in biochemotherapy regimens may reduce the incidence of CNS relapse and increase the number of long-term survivors. Because most active drugs for stage IV melanoma do not adequately penetrate the blood–brain barrier, the CNS is a sanctuary site for malignant melanoma. This is of significance because 50% of patients with stage IV melanoma that have responded to initial therapy will relapse in the brain as the first and only site of relapse (263,264). To address whether or not temozolomide reduces the risk of CNS relapse compared to DTIC-based therapy, the incidence of CNS relapse in patients with advanced melanoma who had previously responded to either temozolomide or DTIC was examined in a retrospective case-controlled study of 41 patients (265). Fewer CNS relapses were seen in patients who were treated with temozolomide; however, small patient numbers preclude specific conclusions. Further exploration of the potential for temozolomide plus combination immunotherapy to prevent brain metastases in metastatic melanoma was a major objective of another clinical study (266). Kersten et al reported that none of the patients who achieved a response or maintained stable disease following temozolomide and IL-2 based biochemotherapy developed brain metastases at a median follow-up of 10 months. Other phase II trials have observed similar findings, reporting a 0–11% CNS failure rate in patients with advanced melanoma responding to temozolomide-containing regimens (266–270). This suggests that therapeutic agents with activity against melanoma and reasonable penetration of the blood–brain barrier may prevent or delay the time to CNS relapse. Single agent chemotherapy can induce durable, complete radiographic resolution of melanoma brain metastases in a very limited number of patients. Three reports each describe a patient with a complete radiographic response of multiple brain metastases from melanoma to temozolomide monotherapy (271–273). Additional reported cases of complete regression of brain metastases from melanoma include two patients treated with temozolomide chemotherapy and whole brain irradiation (274,275), and another patient who was treated with temozolomide chemotherapy and radiosurgery (276). Furthermore, there are reports of patients achieving a partial radiographic response to chemotherapy, either temozolomide alone or in combination with docetaxel or thalidomide, which realized long-term remissions in the brain of over one year (277,278). Fotemustine, a newer phosphalanine-modified nitrosurea, has demonstrated activity in brain metastases from melanoma. Systemic disease response rates of fotemustine are similar to those obtained with other nitrosourea compounds. Notably, phase II studies evaluating fotemustine in patients with melanoma brain metastases, report objective response rates ranging from 8% to 25% (279–282). The response rate was more modest in a large phase III study comparing fotemustine to dacarbazine in patients with metastatic melanoma (283). This study reported a 6% response rate in patients with brain metastases after treatment with fotemustine, and no responses after treatment with dacarbazine (283). Unfortunately fotemustine has a less than favorable toxicity profile, and myelotoxicity has limited its combination with other chemotherapy drugs or whole-brain radiation therapy (279). Of note, fostemustine is currently not approved by the Food and Drug administration for use in the United States. Phase II trials suggest that temozolomide, either as a single agent or in combination with other chemotherapeutic drugs, has modest activity in the initial treatment of melanoma brain metastases producing an overall 5–10% objective response rate with no significant survival advantage (269,272,277,284–289). Agarwala et al. has reported the largest clinical experience of any single chemotherapy agent in patients with melanoma brain metastases, enrolling 151 patients in a phase II trial evaluating temozolomide using the standard 5-day dosing schedule (272). In this trial, patients were stratified into two groups based upon whether or not they had received prior chemotherapy for systemic disease. Among 117 chemotherapy-naive patients, 7% achieved an objective response including one complete responder, and 29% had stable disease. A single partial response was seen in the 34 patients who had been previously treated with chemotherapy, and 18% of these patients had stable disease. Considering all 151 patients, a 6% response rate was realized with a median overall survival of 3.2 months. Not surprising, chemotherapy-naïve patients faired somewhat better with an overall median survival of 3.5 compared to 2.2 for those patients who had received prior chemotherapy. This trend was also noted in another study using the same stratification schema (284). Utilizing a dose-intensified temzolomide regimen, 40% of 21 melanoma patients with brain metastases achieved a partial response or stable disease in one study (290).

Chapter 28 / Neurologic Complications of Melanoma

537

This result, however, was not duplicated in another study utilizing the same intensified temozolomide dosing schedule (284). Combining temozolomide with docetaxol (277), thalidomide (278,285,286), carboplatin (287), lomustine (288), or immunotherapy (289) did not improve upon objective response rates or impact overall survival (277,278,285,286). However, more toxicity was usually seen. This, in particular, is the case with the combination of temozolomide and thalidomide, which was associated with an unacceptable incidence of toxicity, both thromboembolic events and intratumoral hemorrhage (278,285,286). A few small phase II studies have addressed the efficacy of chemotherapy combined with whole-brain irradiation for the treatment of brain metastases from melanoma (275,291,292). While objective responses were reported in individual patients receiving chemotherapy, either fotemustine or temozolomide, in addition to whole-brain irradiation, no clear survival advantage was seen. However, in a prospective, randomized trial of WBRT plus fostemustine compared to WBRT alone for the treatment of brain metastases from melanoma, a significant increase was seen in time to disease progression in those patients receiving fotemustine (293). One retrospective study has evaluated the feasibility of radiotherapy as adjunct treatment to temozolomide for the treatment of melanoma brain metastases. In this report, all 35 patients had unresectable melanoma brain metastases and received temozolomide on the standard 5-day dosing schedule. Twelve patients with 1–3 brain metastases additionally were treated with radiosurgery, and 10 patients with disseminated brain metastases and a satisfactory performance status additionally received whole-brain irradiation. The frequency of objective responses was very low with one complete response (3%) and two partial responses (6%). An additional 9/35 patients (26%) had stable disease. Resolution of four of five brain metastases (partial response) after three courses of temozolomide therapy occurred in a patient treated with TMZ alone. Complete remission was seen after treatment with radiosurgery and temozolomide in another patient. Median survival from the start of therapy was 8 months. Limited data are available to address the role of salvage chemotherapy for patients with recurrent brain metastases. Three patients with recurrent melanoma brain metastases after treatment with whole-brain irradiation were treated with temozolomide as part of a phase II trial in grouped histologies (294). No responses were seen in this very limited patient group. Similar to surgical resection and radiosurgery, the best results of chemotherapy are achieved in previously untreated patients with a single brain metastasis and limited systemic disease. Further studies are needed to define the role of chemotherapy in the management of brain metastases from malignant melanoma.

4. LEPTOMENINGEAL METASTASES Up to 19% of cancer patients with neurologic signs and symptoms will have evidence of meningeal involvement at autopsy (295). Melanoma is the third most common solid tumor to metastasize to the leptomeninges after breast and lung carcinoma (295–298). It accounts for 6–18% of all leptomeningeal carcinomatosis in large series (295,297–299). Alternatively, approximately 23% of patients with melanoma will develop leptomeningeal metastases (300). One-half of patients with meningeal carcinomatosis have concomitant brain relapse (88). Clinical manifestations of leptomeningeal carcinomatosis result from obstruction of normal CSF flow, infiltration of nerves or occlusion of pial blood vessels within the subarachnoid space, irritation or invasion of underlying brain parenchyma, and alteration of CNS metabolism. The initial presentation is headache in up to one-half of patients and cranial nerve dysfunction in one third of patients (298). Mental status changes, seizures, back or radicular pain, incontinence, lower motor neuron weakness, and sensory abnormalities are also common symptoms. In one clinical series, the most common neurological examination finding was the asymmetric loss of deep tendon reflexes in 70% of patients, followed by cranial nerve abnormalities in 55%, cauda equina syndrome in 33%, and cognitive abnormalities in 31% (298) Neurologic dysfunction at multiple levels of the neuraxis is found in most patients (278–298,301,302). Diagnostic confirmation of leptomeningeal carcinomatosis can be difficult. Definitive diagnosis relies on the finding of malignant cells in cerebrospinal fluid, tumor nodules on nerve roots at myelography, or convincing evidence of leptomeningeal tumor with magnetic resonance imaging. Malignant cells are found in the CSF at the

538

Part VII / Neurologic Complications of Specific Malignancies

first examination in 45–77% of cases with the diagnostic yield increasing up to 94% if multiple CSF samples are taken at different times or from differing locations (298,301,303,304). The diagnostic utility of biochemical markers in cerebrospinal fluid has yet to be realized in this disease. Elevated CSF levels of lactate dehydrogenase (LDH) and LDH isoenzyme-5 are often present in patients with leptomeningeal involvement from melanoma (305–310); however, these markers lack specificity and can be seen in nonmalignant conditions such as stroke, bacterial meningitis, and head trauma (309). Radiographic abnormalities are found in approximately 50% of patients with spinal symptoms from leptomeningeal carcinomatosis (298,310). Findings include meningeal enhancement along the cortical convexities or basilar cisterns, enlarged enhancing nerve roots, or hydrocephalus in the absence of ventricular obstruction. While these findings are suggestive of leptomeningeal carcinomatosis, they are nonspecific and of limited diagnostic utility. In contrast, multiple nodular enhancing subarachnoid masses on CT myelography or MRI are highly suggestive if not diagnostic of leptomeningeal carcinomatosis. Most imaging manifestations are better visualized with gadolinium-enhanced MRI than contrast-enhanced CT (311,312). Imaging also provides additional information that may impact diagnostic evaluation and/or treatment recommendations. It serves to exclude concurrent intraparenchymal or epidural metastases, identify areas of bulky disease, and demonstrate potential contraindications for lumbar puncture such as obstructive hydrocephalus or spinal block. In the appropriate clinical setting, imaging findings and relatively nonspecific spinal fluid abnormalities, such as an elevated protein or pleocytosis, are sufficient for a clinical diagnosis of leptomeningeal carcinomatosis. Treatment of leptomeningeal carcinomatosis must address widespread tumor within the cerebrospinal fluid space, bulky tumor deposits, and tumor spread within nerve root sleeves and Virchow–Robin spaces. Craniospinal irradiation adequately targets all sites of disease, but this treatment modality is limited by the relative radioresistance of melanoma as well as side-effects inherent to this approach such as myelosuppression. Intrathecal chemotherapy does not adequately penetrate into bulky subarachnoid tumor masses, and its distribution within the cerebrospinal fluid may be impeded by alterations in CSF bulk flow (313). Radionuclide CSF flow scans demonstrate abnormalities in as many as 70% of patients with solid tumors (314,315). Local irradiation to areas of symptomatic disease and abnormal CSF flow in combination with intrathecal chemotherapy provides treatment encompassing all disease sites, and is the mainstay of current therapy. Antineoplastic agents available for intrathecal administration, mainly methotrexate, thiotepa, and cytosine arabinoside (Ara-C), have not demonstrated significant activity against melanoma. Overall, the response to therapy in leptomeningeal seeding from malignant melanoma is poor. Most evidence suggests that untreated, clinically evident leptomeningeal carcinomatosis produces precipitous neurological decline and death from neurologic disease with a median survival of 4-6 weeks (296–298). Treatment is palliative, and rarely results in effective local control (316). Diffuse brain symptoms, such as altered mental status, respond to therapy more often than fixed neurological deficits. Even with aggressive treatment, between 35% and 76% of patients die as a direct consequence of their neoplastic meningitis rather than of their systemic disease (298,303,304).

5. SPINAL METASTASES The spine is a common site of skeletal metastases in melanoma (317). However, spinal cord compression due to metastatic malignant melanoma occurs infrequently, accounting for approximately 5% of cases of spinal cord compression (318). Patients with epidural spinal cord compression present with pain in 83–96%, sensory loss in 51–90%, weakness in 76–94%, and autonomic dysfunction in 57–69% (319). Approximately one-third of patients with symptomatic epidural spinal cord compression have multiple sites of involvement radiographically (320–322). Spinal cord compression due to epidural metastases has a poor prognosis. As such, the primary goal of treatment is reduction of pain symptoms and functional improvement or stabilization of neurological deficits. Non-weightbearing individuals at presentation are unlikely to regain ambulatory capacity (322). General treatment options for epidural spinal cord metastases include radiotherapy, radiosurgery, radionuclide therapy, surgical decompression and/or stabilization, systemic chemotherapy, biphosphonates, embolization, and bracing (323,324). Small retrospective case series reviewing symptomatic spinal metastases from melanoma report neurological improvement in 20–71% of patients treated with radiation therapy alone (168,318,324,325). Similar results are

Chapter 28 / Neurologic Complications of Melanoma

539

noted with surgical intervention (160,326–330). One report of stereotactic radiosurgery reported pain relief in 96% of treated patients (324). There are no randomized studies addressing whether or not surgical decompression provides better or more durable neurological responses compared to radiation therapy alone. A higher complication rate of surgery may exist in patients with melanoma compared to other tumor histologies, presumably due to the vascularity of melanoma metastases (331). The median survival for patients with symptomatic spinal metastases undergoing surgical decompressive laminectomy is approximately 1.5 months (326–330,332). Treatment recommendations must take into consideration the risk and recovery period of an extensive operative procedure in light of the uncertainty of clinical improvement and overall poor prognosis. Surgery may be justified in selected patients with good general health and no evidence of visceral or other distant metastases. The benefit of post-operative radiation therapy is unclear. In one report, radiation therapy alone was as effective as the combination of surgery and radiotherapy in radioresistant tumors. The provision of post-operative radiation therapy did not improve overall response in another series (160). Conventional radiotherapy delivers full-dose radiation to a treatment field that encompasses both the vertebral body and the adjacent spinal cord. The maximally tolerated radiation dose of the spinal cord is well below the optimal therapeutic dose for the vast majority of tumors, significantly limiting the efficacy of conventional radiation (333–335). Radiosurgery overcomes this limitation through the delivery of a very conformal high-dose fraction of radiation to the tumor bed sparing the adjacent tissues. Radiosurgery has long been utilized in the treatment of brain metastases; however, only recently have technological advances made stereotactic radiosurgical treatment of the spine feasible. Numerous studies have demonstrated feasibility, efficacy, and tolerability of spinal radiosurgery at doses compared to those used for the treatment of brain metastases (324,335–342). Clinical or radiographic evidence of acute or subacute spinal cord damage has not been reported (324). Compared to external beam irradiation, spinal radiosurgey provides a larger total radiobiological equivalent dose, limits spinal column irradiation and subsequent bone marrow suppression, and allows completion of treatment in a single day. However, there is limited data on the comparative efficacy of external beam irradiation, stereotactic radiosurgery, and surgery. Unlike surgery, radiation therapy inadequately addresses pain caused by mechanical instability. Vertobroplasty and kyphoplasty offer a minimally invasive approach for the stabilization of pathologic vertebral body compression fractures. Vertobroplasty consists of the percutaneous injection of an acrylic bone cement into the fractured vertebral body to stabilize the remaining bone structure. Kyphoplasty takes the process one step further by attempting to restore vertebral body height. In this procedure a specialized inflatable balloon is inserted through a cannula into the collapsed vertebral body. When inflated, the balloon creates a cavity into which the bone cement is injected allowing for possible re-establishment of vertebral body height. A small study of kyphoplasty in conjunction with radiosurgery for patients with symptomatic, unstable pathological compression fractures, reported long-term pain relief in 92% of 26 patients (343). No acute or subacute radiation-induced spinal cord damage was seen, even in a subgroup of seven patients previously treated with external beam irradiation (343). This treatment paradigm shows promise of providing the maximal benefits of surgery with a minimally invasive and time-wise approach.

6. INTRAMEDULLARY SPINAL CORD METASTASES Intramedullary spinal cord metastases are much less common than metastatic epidural spinal cord compression. In all cancer types, they account for less than 4% of symptomatic metastases affecting the spinal cord, and less than 9% of metastases to the central nervous system (344,345). Metastases from melanoma comprise approximately 9% of intramedullary spinal cord tumors (344,346,347). While intramedullary metastases may occur as isolated central nervous system metastases, they are most often accompanied by disease at other levels of the neuraxis. Some 50–60% of patients will have a history of prior brain metastases or simultaneous brain metastases (319,348,349), while 15% to 44% have coexistent leptomeningeal disease (319). One-third of patients have multiple spinal cord lesions (347). Prognosis is universally poor with average survival 4–6 months from diagnosis. The presenting symptoms of intramedullary spinal cord metastases are similar to those of epidural spinal metastases—pain, sensory loss, weakness, and autonomic dysfunction. However, the presence of asymmetric spinal cord dysfunction in the pattern of Brown–Sequard syndrome or pseudo-Brown–Sequard syndrome, implicates

540

Part VII / Neurologic Complications of Specific Malignancies

intramedullary disease (319). Spinal magnetic resonance imaging readily distinguishes the two entities and is the diagnostic tool of choice. There are limited data on the treatment of intramedullary spinal cord metastases of melanoma. Symptomatic intramedullary spinal cord metastases are most often treated with spinal irradiation. Radiation therapy with concurrent corticosteroids can stabilize or alleviate neurological dysfunction. As is the case with epidural spinal metastases, the loss of ambulatory capacity at presentation portends poor prognosis for neurological recovery. Other treatment options include aggressive surgical resection often followed by post-surgical radiation therapy or chemotherapy.

7. PLEXUS/PERIPHERAL NERVE METASTASES Cutaneous melanoma itself rarely metastases to nerve root/peripheral nerve. However, a variant of melanoma, desmoplastic neurotropic melanoma, has a unique proclivity to extend along small peripheral nerves. This entity is synonymous with spindle-cell malignant melanoma with neurotropism, desmoplastic melanoma, and neurotropic melanoma. Desmoplastic neurotropic melanoma usually occurs in the head and neck region, but it can be located in the extremities (350,351). Typically patients present in their seventh decade and have precursor lentigo maligna (352). The primary differential diagnosis is malignant peripheral nerve sheath tumors (MPNSTs), which can simulate desmoplastic neurotropic melanoma clinically and histologically (353–357). Both types of tumors may express S-100 protein, and neural proteins may stain both Schwann cells and melanocytes. While melanoma-associated antigen HMB-45 is commonly found in melanoma, it can be rarely expressed in MPNSTs (353). In the setting of immunohistochemical ambiguity, the pattern of disease involvement may help differentiate between these two entities. In particular, the presence of lymph node metastases implicates malignant melanoma (358). The mechanism of neurotropism in malignant melanoma is poorly understood. The nerve growth factor (NGF)/NGF-receptor system likely plays a role (359). In vitro studies have demonstrated that the binding of NGF to its receptor can stimulate melanoma cell proliferation (360). Nerve growth factor is produced in peripheral nerves by Schwann cells. NGF receptors are rarely expressed in conventional epithelioid melanomas, but are highly expressed in neurotropic melanoma.

8. PARANEOPLASTIC DISORDERS 8.1. Melanoma-Associated Retinopathy Melanoma-associated retinopathy (MAR) is one of three distinct paraneoplastic syndromes presenting as visual symptoms: cancer-associated retinopathy (CAR), melanoma-associated retinopathy, and paraneoplastic optic neuropathy. MAR was first described in 1988 (361), and as the name suggests, is thought to be exclusively associated with melanoma. However, two reports of a MAR-like syndrome in patients with colon cancer suggest the possibly of syndromic overlap with CAR (362,363). MAR is characterized by night blindness (nyctalopia), electroretinogram (ERG) findings of rod dysfunction with relatively normal cone function, and the demonstration of circulating IgG autoantibodies to retinal bipolar cells. Clinically, patients report abrupt onset of painless visual symptoms consisting of night blindness, shimmering or flickering light phenomena (photopsias), and peripheral visual field loss. Symptoms typically involve both eyes, either simultaneously or sequentially, and manifest abruptly over days. MAR appears to be a unimodal disease with very few reported patients demonstrating a pattern of slowly progressive deficits (364–366). A male sex predominance exists (367). Unlike cancer-associated retinopathy, in which visual symptoms often herald the diagnosis of cancer, the visual symptoms of MAR occur after the diagnosis of malignant melanoma, either heralding or following the development of metastatic disease (367–375). The average latency from melanoma diagnosis to onset of visual symptoms in MAR is 3.6 years (376). MAR rarely precedes the diagnosis of melanoma (377). The spectrum of visual field abnormalities in MAR include general constriction, peripheral depression, or midperipheral scotoma (376). Visual acuity, color vision, and central visual fields are normal or mildly impaired at presentation (376). Initial fundoscopic examination findings are frequently normal; however, vascular attenuation,

Chapter 28 / Neurologic Complications of Melanoma

541

thinning, and mottling of the retinal pigment epithelium, vitreous cells, and optic disc pallor can be seen (378). After some time, attenuation of the arterioles and retinal pigment epithelium changes are characteristic findings (378). Histopathologic examination of eyes with MAR demonstrates a marked reduction of bipolar neurons in the inner nuclear layer and evidence of trans-synaptic ganglion cell atrophy (379). Diagnostic studies reveal an electroretinopathy pattern mimicking congenital night blindness. A markedly reduced or absent dark-adapted b-wave response is characteristically seen indicating rod dysfunction with relatively normal cone function. However, the hallmark of MAR is the demonstration of circulating immunoglobulin G (IgG) autoantibodies that cross-react with retinal epitopes. Antibodies within the serum of patients with MAR most commonly react with retinal cells in the bipolar cell layer, but reaction with retinal cells in the outer plexiform and nerve fiber layers, as well as the phototransduction protein transducin, have been demonstrated (373,378,380,381). The corresponding antigen has not yet been identified. In fact, multiple retinal antigens and/or the heterogeneity in autoantibody specificity may be present. Testing for anti-bipolar cell antibodies is currently available commercially and at several research laboratories. However, interpretation of testing results may be difficult. Not all patients with MAR possess anti-bipolar cell antibodies. Furthermore, patients with melanoma without ocular symptoms may have anti-retinal antibodies (382). In addition, anti-bipolar antibodies have been reported in a patient with CAR and colon cancer (373). Relatively few of the reported cases of MAR describe treatment or its outcome. Overall, MAR symptoms appear relatively refractory to immunosuppressive therapies (369,373). However, anecdotal reports describe visual improvement of varying degrees with corticosteroid therapy (368,373,383,384), plasmapheresis in combination with corticosteroids (378), intravenous immunoglobulin (IVIG) provided either alone or in combination with cytoreductive surgery (378,385), and cytoreduction via surgery or radiation therapy (378,386). Treatment strategies aggressively managing the primary tumor in addition to immunosuppressive therapy have shown the most promise. Current research focuses on strategies to block antibody-mediated apoptosis and evaluation of the activation of recoverin-specific antitumor cytotoxic T-lymphocytes.

8.2. Anti-Hu–Related Encephalomyelitis A high titer of anti-Hu antibody, also known as type 1 antineuronal nuclear autoantibody or ANNA-1, generally implies the presence of small cell lung cancer (SCLC). However, it is rarely associated with other tumors including melanoma (387). Malignant melanoma can express a Hu antigen (387), and isolated cases of anti-Hu paraneoplastic disorders have been reported in malignant melanoma patients (388). In a group of ANNA-1 positive patients, one series found a 13% frequency of an unrelated primary malignancy coexisting with small cell lung cancer (388). This in conjunction with the overwhelming association of anti-Hu paraneoplastic disorders with small cell lung cancer underscores the importance of vigilance for SCLC in melanoma patients with anti-Hu antibodies.

8.3. Chronic Inflammatory Demyelinating Polyneuropathy Isolated reports have proposed an association between malignant melanoma and chronic inflammatory demyelinating polyneuropathy (389,390). This concept is supported by the rare observation of an inflammatory demyelinating polyneuropathy in patients receiving melanoma immunotherapy, in particular following the administration of monoclonal anti-GD2 antibodies or intradermal vaccinia melanoma cell lysates for metastatic melanoma (391,392). Both peripheral nerve Schwann cells and melanocytes are derived from the neural crest and may share common antigenic components.

9. NEUROLOGIC COMPLICATIONS RELATED TO THERAPY 9.1. Alpha-Interferon Therapy Biochemotherapy, which is becoming the mainstay of treatment for metastatic melanoma, is not without side effects. Neurotoxicity is the major dose-limiting factor for alpha-interferon therapy. Although symptoms can occur at any time throughout treatment, 90% of side effects occur within 3 months of starting treatment, 60% within one month, 40% within 2 weeks, and 20% within one week (393). Systemically administered alpha-interferon can produce subacute, reversible symptoms of impaired concentration, memory loss, cognitive slowing, reduction in goal directed behavior, and frontal lobe dysfunction.

542

Part VII / Neurologic Complications of Specific Malignancies

Incoordination and gait disturbance can also occur (394–401). Later on during therapy dysphoria, helplessness, and anhedonia may manifest (402). Suicidal ideation and severe dysphoria can occur at the beginning of treatment or at the time of dose escalation (402). For the majority of patients, discontinuation of IFN therapy results in remission of side effects in 2–3 weeks (403), although persistent symptoms have been reported up to 3 years after treatment (404). Long-term IFN therapy is associated with prominent dose-limiting fatigue (403). In patients with leptomeningeal carcinomatosis, intraventricular administration of alpha interferon at a cumulative dose from 14 x 106 IU to 54 x 106 IU was associated with a subacute reversible progressive wakeful vegetative state resembling catatonia. Prior brain irradiation may potentiate neurotoxicity in this patient population (405).

10. CONCLUSION Several characteristics distinguish melanoma brain metastases from brain metastases of other tumors. Brain metastases from melanoma have a strong tendency for both macroscopic and microscopic intratumoral hemorrhage. Central nervous system metastases from malignant melanoma carry the worst prognosis of all sites of distant metastases in this disease. While the majority of patients with brain metastases from grouped histologies die from uncontrolled systemic disease, patients with melanoma brain metastases usually succumb to a neurologic death. For this reason, patients with central nervous system disease are usually excluded from clinical trials for metastatic melanoma. Preliminary study results suggest that improved disease control at the initial management of metastatic melanoma may decrease the incidence of central nervous system disease. While palliative therapy is reasonable in patients with concurrent widely disseminated disease, more aggressive multiple modality treatment approaches are warranted in those patients with good performance status and absent or limited extracranial disease. Management decisions must consider both quality and quantity of life.

REFERENCES 1. Swerdlow AJ. International trends in cutaneous melanoma. Ann N Y Acad Sci 1990;60:235–251. 2. La Vecchia C, Lucchini F, Negri E et al. Recent declines in worldwide mortality from cutaneous melanoma in youth and middle age. Int J Cancer 1999;81:62–66. 3. Wingo PA, Tong T, Bolden S. Cancer statistics, 1995. CA Cancer J Clin 1995;45:8–30. 4. Khayat D, Coeffic D, Antoine E-C: Overview of medical treatments of metastatic malignant melanoma. ASCO Educational Book, Spring 414–427, 2000. 5. Greenlee RT, Murray T, Bolden S et al. Cancer statistics, 2000. CA Cancer J Clin 2000;50:7–33. 6. Cancer Stat Fact Sheets. Melanoma of the Skin. National Cancer Institute. (Accessed October 15, 2006, at http://seer.cancer.gov/statfacts/html/melan/html.) 7. Jemal A, Siegel R, Ward E et al. CA Cancer J Clin 2007;57:43–66. 8. SEER Cancer Statistics Review 1975–2004. Melanoma of the skin section. National Cancer Institute (Accessed October 15, 2006 at http://seer.cancer.gov/csr/1975_2004/results_merged/sect_16_melanoma.pdf) 9. Pratt CB, Palmer MK, Thatcher N et al. Malignant melanoma in children and adolescents. Cancer 1981;47:392–397. 10. Davidoff AM, Cirrincone C, Seigler HF. Malignant melanoma in children. Ann Surg Oncol 1994;1:278–282. 11. Tate PS, Ronan SG, Feucht KA et al. Melanoma in childhood and adolescence: clinical and pathological features of 48 cases. J Pediatr Surg 1993;28:217–222. 12. Mehregan AH, Mehregan DA. Malignant melanoma in childhood. Cancer 1993;71:4096–4103. 13. Barnhill RL, Flotte TJ, Fleischli M et al. Cutaneous melanoma and atypical Spitz tumours in childhood. Cancer 1995;76:1833–1845. 14. Ceballos PI, Ruiz-Maldonado R, Mihm MC. Melanoma in children. N Engl J Med 1995;332:656–662. 15. Ferrone CR, Ben Porat L, Panageas KS et al. Clinicopatholoical features of and risk factors for multiple primary melanomas. JAMA 2005;294:1647–1654. 16. Siskind V, Aiten J, Green A, Martin N. Sun exposure and interaction with family history in risk of melanoma, Queensland, Australia. Int J Cancer 2002;97:90–95. 17. Elwood JM. Melanoma and sun exposure: contrasts between intermittent and chronic exposure. World J Surg 1992;16:157–165. 18. Berwick M, Armstrong BK, Ben-Porat L et al. Sun exposure and mortality from melanoma. J Natl Cancer Inst 2005;97:195–199. 19. Wallace DC, Beardmore GL, Exton LA. Familial malignant melanoma. Ann Surg 1973;177:15–20. 20. Reimer RR, Clark WC, Greene MH et al. Precursor lesions in familial melanoma: a genetic preneoplastic syndrome. JAMA 1978;239:744–746. 21. Anderson DE, Smith JL, McBride CM. Hereditary aspects of malignant melanoma. JAMA 1967;200:741–746. 22. Tsao H, Sober A. Acqired precursor lesions and markers of increased risk for cutaneous melanoma. In: Balch C, Houghton A, Sober Aet al. (eds.). Cutaneous Melanoma. 4th ed. St. Louis: Quality Medical Publishing, 2003:121–134. 23. Bishop DT, Demenais F, Goldstein AM et al. Geographical variation in the pentrance of CDKN2A mutations for melanoma. J Natl Cancer Inst 2002;94:894–903.

Chapter 28 / Neurologic Complications of Melanoma

543

24. Duve S, Rakoski J. Cutaneous melanoma in a patient with neurofibromatosis: a case report and review of the literature. Br J Dermatol 1994;131:290–294. 25. Greene MH, Fraumeni JF, Jr. The hereditary variant of cutaneous malignant melanoma. In: Clark W, Goldman L, Mastrangelo M (eds.). Human Malignant Melanoma. New York: Grune and Stratton, 1979:139–166. 26. Azizi E, Friedman J, Pavlotsky F et al. Familial cutaneous malignant melanoma and tumors of the nervous system. Cancer 1995;76:1571–1578. 27. Nutt JG, Bird TD. Essential myoclonus in a kindred with familial malignant melanoma. Arch Neurol 1984;41:189–191. 28. Greene MH, Mead GD, Reimer RR et al. Malignant melanoma and Charcot–Marie–Tooth disease. Am J Med Genet 1980;5:69–71. 29. Le Mire L, Hollowood K, Gray D et al. Melanomas in renal transplant recipients. Br J Dermatol 2006;154:472–477. 30. Bouwes Bavinck JN, Hardie DR, Green A et al. The risk of skin cancer in renal transplant recipients in Queensland, Australia: a follow-up study. Transplantation 1996;61:715–721. 31. Leveque L, Dalac S, Dompmartin A et al. Mélanome chez le transplante. Ann Dermatol Venereol 2000;127 :160–165. 32. Mehlman MA. Causal relationship from exposure to chemicals in oil refining and chemical industries and malignant melanoma. Ann NY Acad Sci 2006;1076:822–828. 33. Langard S, Rosenberg J, Andersen A et al. Incidence of cancer among workers exposed to vinyl chloride in polyvinyl chloride manufacture. Occup Environ Med 2000;57:65–68. 34. Zanetti R, Loria D, Rossa S. Melanoma, Parkinson’s disease, and levodopa: causal or spurious link? A review of the literature. Melanoma Res 2006;16:201–206. 35. Fiala KH, Whetteckey J, Manyam BV. Malignant melanoma and levodopa in Parkinson’s disease: causality or coincidence? Parkinsonism Relat Dis 2003;9:321–327. 36. Siple JF, Schneider DC, Wanlass WA et al. Levodopa therapy and the risk of malignant melanoma. Ann Pharmacother 2000;34: 382–385. 37. Pfützner W, Przybilla B. Malignant melanoma and levodopa: is there a relationship? Two new cases and a review of the literature. J Am Acad Dermatol 1998;37:332–336. 38. Olsen JH, Friis S, Frederiksen K. Malignant melanoma and other types of cancer preceding Parkinson disease. Epidemiology 2006;17:582–587. 39. Balch CM, Buzaid AC, Soong SJ et al. Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol 2001;19:3635–3648. 40. Balch CM, Soong SJ, Gershenwald JE et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 2001;19:3622–3634. 41. Lens MB, Dawes M, Goodacre T et al. Excision margins in the treatment of primary cutaneous melanoma: a systematic review of randomized controlled trials comparing narrow vs wide excision. Arch Surg 2002;137:1101–1105. 42. Bleicher RJ, Essner R, Foshag LJ et al. Role of sentinel lymphadenectomy in thin invasive cutaneous melanomas. J Clin Oncol 2003;21:1326–1331. 43. Kuhn JA, McCarty TM. Malignant melanoma and the sentinel lymph node biopsy. Cancer Invest 1999;17:39–46. 44. Reintgen DS, Jakub JW, Pendas S et al. The staging of malignant melanoma and the Florida melanoma trial. Ann Surg Oncol 2004;11:186S–191S. 45. Morton DL, Thompson JF, Cochran AJ et al. Sentinel-node biopsy or nodal observation in melanoma. N Engl J Med 2006;355: 1307–1317. 46. Morton DL, Wanek L, Nizze JA et al. Improved long-term survival after lymphadenectomy of melanoma metastatic to regional nodes: analysis of prognostic factors in 1314 patients from the John Wayne Cancer Clinic. Ann Surg 1991;214:491–499. 47. Cascinelli N, Morabito A, Santinami M et al. Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: a randomised trial. Lancet 1998;351:793–796. 48. Barth A, Wanek LA, Morton DL. Prognostic factors in 1521 melanoma patients with distant metastases. J Am Coll Surg 1995;181: 193–201. 49. Tomsu K, Van Eschen KB, Lee MA. Meta-analysis of median survival of patients with stage IV melanoma. Proc Am Soc Clin Onc 1997;16:1784. 50. Rangel J, Torabian S, Shaikh L et al. Prognostic significance of nuclear receptor coactivator-3 overexpression in primary cutaneous melanoma. J Clin Oncol 2006;24:4565–4569. 51. Palmiere G, Satriano SMR, Budroni M et al. Serial detection of circulating tumour cells by reverse transcriptase-polymerase chain reaction assays is a marker for poor outcome in patients with malignant melanoma. BMC Cancer 2006;6:266–273. 52. Aloia TA, Gershenwald JE, Andtbacka RH et al. Utility of computed tomography and magnetic resonance imaging staging before completion lymphadenectomy in patients with sentinel lymph node-positive melanoma. J Clin Oncol 2006;24:2858–2865. 53. Balch CM, Reintgen DS, Kirkwood JM et al. Cutaneous melanoma. In: DeVita VT, Jr, Hellman S, Rosenberg SA (eds.) Cancer: Principles & Practice of Oncology. Vol 2. 5th Ed. Philadelphia: Lippincott-Raven, 1997:1947–1994. 54. Chang P, Knapper WH. Metastatic melanoma of unknown primary. Cancer 1982;49:1106–1111. 55. Das Gupta T, Bowden L, Berg JW. Malignant melanoma of unknown primary origin. Surg Gynec Obstet 1963;117:341–345. 56. Baab GH, McBride CM. Malignant melanoma: the patient with an unknown site of primary origin. Arch Surg 1975;110:896–900. 57. Milton GW, Shaw HM, McCarthy WH. Occult primary malignant melanoma: factors influencing survival. Br J Surg 1977;64:805–808. 58. Guiliano AE, Moseley HS, Morton DL. Clinical aspects of unknown primary melanoma. Ann Surg 1980;191:98–104. 59. Reintgen DS, McCarty KS, Woodard B et al. Metastatic malignant melanoma with an unknown primary. Surg Gynecol Obstet 1983;156:335–340. 60. Huffman TA, Sterin WK. Ten-year survival with multiple metastatic malignant melanoma: primary site unknown. Arch Surg 1973;106:234–235.

544

Part VII / Neurologic Complications of Specific Malignancies

61. Balch CM, Houghton AN. Diagnosis of metastatic melanoma at distant sites. In: Balch CM, Houghton AN, Milton GW (eds.) Cutaneous Melanoma. Philadelphia: J.B. Lippincott, 1992:439–467. 62. Wronski M, Arbit E. Surgical treatment of brain metastases from melanoma: a retrospective study of 91 patients. J Neurosurg 2000;93:9–18. 63. Sampson JH, Carter JH, Friedman AH et al. Demographics, prognosis, and therapy in 702 patients with brain metastases from malignant melanoma. J Neurosurg 1998;88:11–20. 64. Patchell RA. Brain metastases. Neurol Clin 1991;9:817–824. 65. Hansen HH. Should initial treatment of small cell carcinoma include systemic chemotherapy and brain irradiation? Cancer Chemother Rep Suppl 1973:4:239–241. 66. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. In: Schoenberg BS (ed.). Neurological Epidemiology: Principles and Clinical Applications. Advances in Neurology, Vol 19. New York: Raven Press. 1978:579–592. 67. Grant R, Whittle IR, Collie DA et al. Referral pattern and management of patients with malignant brain tumours in southeast Scotland. Health Bull 1996;54:204–211. 68. Nussbaum ES, Djalilian HR, Cho KH et al. Brain metastases: histology, multiplicity, surgery, and survival. Cancer 1996;78: 1781–1788. 69. Patchell RA. Metastatic brain tumors. Neurol Clin 1995;13:915–925. 70. Sampson JH, Carter JH Jr, Friedman AH et al. Demographics, prognosis, and therapy in 702 patients with brain metastases from malignant melanoma. J Neurosurg 1998;88:11–20. 71. Moser RP. Surgery. In: Twijnstra A, Keyser A, Ongerboer de Visser BW (eds.) Neuro-oncology: Primary Tumors and Neurological Complications of Cancer. Amsterdam: Elsevier, 1993:208. 72. Moon D, Maafs E, Peterson-Schaefer K et al. A review of 567 cases of brain metastases from malignant melanoma. Melanoma Res 1993;3:40 (abstract). 73. Fife KM, Colman MH, Stevens GN et al. Determinants of outcome in melanoma patients with cerebral metastases. J Clin Oncol 2004;22:1293–1300. 74. Baker AB. Metastatic tumors of the nervous system. Arch Pathol 1942;34:495–537. 75. McNeer G, Das Gupta T. Problem of recurrence in the management of melanoma. CA Cancer J Clin 1965;15:270–274. 76. Stehlin JS, Jr, Hills WJ, Rufine C. Disseminated melanoma biologic behavior and treatment. Arch Surg 1967;94:495–501. 77. Amer MH, Al-Sarraf M, Vaitkevicius VK. Clinical presentation, natural history and prognostic factors in advanced malignant melanoma. Surg Gynecol Obstet 1979;149:687–692. 78. Atkinson L. Melanoma of the central nervous system. Aust N Z J Surg 1978;48:14–16. 79. Beresford HR. Melanoma of the central nervous system: treatment with corticosteroids and radiation. Neurology 1969;19:59–65. 80. Chason JL, Walker FB, Landers JW. Metastatic carcinoma in the central nervous system and dorsal root ganglia: a prospective autopsy study. Cancer 1963;16:781–787. 81. Budman DR, Camacho E, Wittes RE. The current causes of death in patients with malignant melanoma. Eur J Cancer 1978;14: 327–330. 82. Das Gupta T, Brasfield R. Metastatic melanoma: a clinico-pathological study. Cancer 1964;17:1323–39. 83. De la Monte SM, Moore GW, Hutchins GM. Patterned distribution of metastases for malignant melanoma in humans. Cancer Res 1983;43:3427–3433. 84. Einhorn LH, Burgess MA, Vallejos C et al. Prognostic correlations and response to treatment in advanced metastatic malignant melanoma. Cancer Res 1974;34:1995–2004. 85. Patel JK, Didolkar MS, Pickren JW et al. Metastatic pattern of malignant melanoma: a study of 216 autopsy cases. Am J Surg 1978;135:807–810. 86. Brega K, Robinson WA, Winston K et al. Surgical treatment of brain metastases in malignant melanoma. Cancer 1990;66:2105–110. 87. Fell DA, Leavens ME, McBride CM. Surgical versus nonsurgical management of metastatic melanoma of the brain. Neurosurgery 1980;7:238–242. 88. Hagen NA, Cirrincione C, Thaler HT et al. The role of radiation therapy following resection of single brain metastasis from melanoma. Neurology 1990;40:158–160. 89. Neuss M, Konig A, Herrmann HD. Intracranial melanoma metastases: surgical treatment and follow-up of 18 patients. Anticancer Res 1987;7:445–446. 90. Overett TK, Shiu MH. Surgical treatment of distant metastatic melanoma: indications and results. Cancer 1985;56:1222–1230. 91. Byrne TN, Cascino TL, Posner JB. Brain metastases from melanoma. J Neuro-oncol 1983;1:313–317. 92. Bullard DE, Cox EB, Seigler HF. Central nervous system metastases in malignant melanoma. Neurosurgery 1981;8:26–30. 93. Ginaldi S, Wallace S, Shalen P et al. Cranial computed tomography of malignant melanoma (stage IVA melanoma). AJR 1981;136: 145–149. 94. Katz HR. The results of different fractionation schemes in the palliative irradiation of metastatic melanoma. Int J Radiat Oncology Biol Phys 1981;7:907–911. 95. Madajewicz S, Karokousis C, West CR et al. Malignant melanoma brain metastases: review of Roswell Park Memorial Institute experience. Cancer 1984;53:2550–2552. 96. Mendez IM, Del Maestro RF. Cerebral metastases from malignant melanoma. Can J Neurol Sci 1988;15:119–123. 97. Pennington DG, Milton GW. Cerebral metastasis from melanoma. Aust N Z J Surg 1975;45:405–409. 98. Posner JB, Chernik NL. Intracranial metastases from systemic cancer. Adv Neurol 1978;19:579–592. 99. Stevens G, Firth I, Coates A. Cerebral metastases from malignant melanoma. Radiother Oncol 1992;23:185–191. 100. Amer MH, Al-Sarraf M, Baker LH et al. Malignant melanoma and central nervous system metastases: incidence, diagnosis, treatment, and survival. Cancer 1978;42:660–668.

Chapter 28 / Neurologic Complications of Melanoma

545

101. Shah JP, Huvos AG, Strong EW. Mucosal melanomas of the head and neck. Am J Surg 1977;134:531–535. 102. Takakura K, Sano K, Hojo S et al. Metastatic Tumors of the Central Nervous System. New York: Igaku-Shoin, Ltd., 1982. 103. Gottlieb JA, Frei E, III, Luce JK. An evaluation of the management of patients with cerebral metastases from malignant melanoma. Cancer 1972;29:701–705. 104. Hayward RD. Malignant melanoma and the central nervous system: a guide for classification based on the clinical findings. J Neurol Neurosurg Psychiatry 1976;39:526–530. 105. Retsas S, Gershuny AR. Central nervous system involvement in malignant melanoma. Cancer 1988;61:1926–1934. 106. Fernandez E, Maira G, Puca A et al. Multiple intracranial metastases of malignant melanoma with long-term survival. J Neurosurg 1984;60:621–624. 107. Bauman ML, Price TR. Intracranial metastatic malignant melanoma: long-term survival following subtotal resection. South J Medicine 1972;65:344–346. 108. McCann KWP, Weir BKA, Elridge AR. Long-term survival after removal of metastatic malignant melanoma of the brain: report of two cases. J Neurosurg 1968:28:483–487. 109. Lang EF, Slater J. Metastatic brain tumors: results of surgical and nonsurgical treatment. Surg Clin N Am 1964;44:865–872. 110. McNeel DP, Leavens ME. Long-term survival with recurrent metastatic intracranial melanoma: case report. J Neurosurg 1968;29: 91–93. 111. Feun LG, Gutterman J, Burgess MA et al. The natural history of respectable metastatic malanoma (stage IVa melanoma). Cancer 1982;50:1656–1663. 112. Hena MA, Emrich LJ, Nambisan RN et al. Effect of surgical treatment on stage IV melanoma. Am J Surg 1987;153:270–275. 113. Wornom IL, III, Smith JW, Soong SJ et al. Surgery as palliative treatment for distant metastases of melanoma. Ann Surg 1986;204: 181–185. 114. Akslen LA, Hove LM, Hartveit F. Metastatic distribution in malignant melanoma: a 30-year autopsy study. Invas Metast 1987;7: 253–263. 115. Choi KN, Withers HR, Rotman M. Metastatic melanoma of the brain: rapid treatment or large dose fractions. Cancer 1985;56:10–15. 116. Branum GD, Seigler HF. Role of surgical intervention in the management of intestinal metastases from malignant melanoma. Am J Surg 1991;162:428–431. 117. Harpole DH, Jr, Johnson CM, Wolfe WG et al. Analysis of 945 cases of pulmonary metastatic melanoma. J Thorac Cardiovasc Surg 1992;103:743–748; discussion 748–750. 118. Lejeune FJ, Lienard D, Sales F et al. Surgical management of distant melanoma metastases. Semin Surg Oncol 1992;8:381–391. 119. Roses DF, Karp NS, Oratz R et al. Survival with regional and distant metastases from cutaneous malignant melanoma. Surg Gynecol Obstet 1991;172:262–268. 120. Wong JH, Euhus DM, Morton DL. Surgical resection for metastatic melanoma to the lung. Arch Surg 1988;123:1091–1095. 121. Lagerwaard FJ, Levendag PC, Nowak PJCM et al. Identification of prognostic factors in patients with brain metastases: a review of 1292 patients. Int J Radiat Oncol Biol Phys 1999;43:795–803. 122. Buchsbaum JC, Suh JH, Lee SY et al. Survival by Radiation Therapy Oncology Group recursive partitioning analysis class and treatment modality in patients with brain metastases from malignant melanoma: a retrospective study. Cancer 2002;94:2265–2272. 123. Haar F, Patterson RH. Surgery for metastatic intracranial neoplasms. Cancer 1972;30:1241–1245. 124. Delattre JY, Krol G, Thaler HT et al. Distribution of brain metastases. Arch Neurol 1988;45:741–744. 125. Enzmann DR, Kramer R, Norman D et al. Malignant melanoma metastatic to the central nervous system. Radiology 1978;127: 177–180. 126. Pickren J, Lopez G, Tsukada Y et al. Brain metastases: an autopsy study. Cancer Treat Symptoms 1983;2:295–313. 127. Yuh W, Engelken J, Mulhonem M et al. Experience with high-dose gadolinium in the evaluation of brain metastases. Am J Neuroradiol 1992;13:335–345. 128. Beresford HR. Melanoma of the nervous system: treatment with corticosteroids and radiation. Neurology 1969;19:59–65. 129. Kuhn MJ, Hammer GM, Swenson LC et al. MRI evaluation of “solitary” brain metastases with triple-dose gadoteridol: comparison with contrast-enhanced CT and conventional-dose gadopentetate dimeglumine MRI studies in the same patient. Comput Med Imaging Graph 1994;18:391–399. 130. Nomoto Y, Miyamoto T, Yamaguchi Y. Brain metastasis of small cell lung carcinoma: comparison of Gd-DTPA enhanced magnetic resonance imaging and enhanced computerized tomography. Jpn J Clin Oncol 1994;24:258–262. 131. Sze G, Milano E, Johnson C et al. Detection of brain metastases: comparison of contrast-enhanced MR with unenhanced MR and enhanced CT. Am J Neuroradiol 1990;11:785–791. 132. Kaye A. Malignant brain tumors. In: Little J, Awad I (eds.). Reoperative Neurosurgery. Baltimore: Williams & Wilkins, 1992:49–76. 133. Louria DB, Stiff DP, Bennett B. Disseminated moniliasis in the adult. Medicine 1962;41:307. 134. Scott M. Spontaneous intracerebral hematoma caused by cerebral neoplasms: report of eight verified cases. J Neurosurg 1975;42: 338–342 135. McCormik WF, Rosenfield DB. Massive brain hemorrhage: a review of 144 cases and examination of their causes. Stroke 1973;4: 946–954. 136. Somaza S, Kondziolka D, Lunsford JD et al. Stereotactic radiosurgery for cerebral metastatic melanoma. J Neurosurg 1993;79: 661–666. 137. Schold SC, Vurgin D, Golbey RB et al. Central nervous system metastases from germ cell carcinoma of testis. Semin Oncol 1979;6:102–108. 138. Dagi TF, Maccabe JJ. Metastatic trophoblastic disease presenting as a subarachnoid hermorrhage: report of two cases and review of the literature. Surg Neurol 1980;14:1975–1984. 139. Graus F, Rogers LR, Posner JB. Cerebrovascular complications in patients with cancer. Medicine 1985;64:16–35.

546

Part VII / Neurologic Complications of Specific Malignancies

140. Globus JH, Sapirstein M. Massive hemorrhage into brain tumors. JAMA 1942;120:348–352. 141. Madow L, Alpers BJ. Cerebral vascular complications of metastatic carcinoma. J Neuropathol Exp Neurol 1952;11:137–148. 142. Strang RR, Ljungdahl TI. Carcinoma of the lung with a cerebral metastasis presenting as subarachnoid hemorrhage. Med J Aust 1962;1:90–91. 143. Mandybur TI. Intracranial hemorrhage caused by metastatic tumors. Neurology 1977;27:650–655. 144. Kondziolka D, Bernstein M, Resch L et al. Significance of hemorrhage into brain tumors: clinicopathological study. J Neurosurg 1987;67:852–857. 145. Little JR, Dial B, Belanger G et al. Brain hemorrhage from intracranial tumor. Stroke 1979;10:283–288. 146. Palmer FJ, Poulgrain AP. Metastatic melanoma simulating subdural hematoma. J Neurosurg 1978;49:301–302. 147. Bitoh S, Hasegawa H, Ohtsuki H et al. Cerebral neoplasms initially presenting with massive intracerebral hemorrhage. Surg Neurol 1984;22:57–62. 148. Altas SW, Grossman RI, Gomori JM et al. Hemorrhagic intracranial malignant neoplasms: spin-echo MR imaging. Radiology 1987;164:71–77. 149. Madonick MJ, Savitsky N. Subarachnoid hemorrhage in melanoma of the brain. Arch Neurol Psychiatry 1951:65:628–636. 150. Wolpert SM, Zimmer A, Schechter MM et al: The neuroradiology of melanomas of the central nervous system. Am J Roentgenol Radium Ther Nucl Med 1967;101:178–187. 151. Ellerhorst J, Strom E, Nardone E et al. Whole brain irradiation for patients with metastatic melanoma: a review of 87 cases. Int J Radiat Oncol Biol Phys 2001;49:93–97. 152. Hillner BE, Kirkwood JM, Agarwala SS. Burden of illness associated with metastatic melanoma: an audit of 100 consecutive referral center cases. Cancer 2001;91:1814–1821. 153. Hafstrom L, Jonsson P-E, Stromblad L-G. Intracranial metastases of malignant melanoma treated by surgery. Cancer 1980;46: 2088–2090. 154. McCann WP, Weir BKA, Elvidge AR. Long-term survival after removal of metastatic malignant melanoma of the brain: report of two cases. J Neurosurg 1968; 29:483–487. 155. Karakousis CP, Velez A, Driscoll DL et al. Metastasectomy in malignant melanoma. Surgery 1994;115:295–302. 156. Saha S, Meyer M, Krementz ET et al. Prognostic evaluation of intracranial metastasis in malignant melanoma. Ann Surg Oncol 1994;1:38–44. 157. Mehta MP, Rozental JM, Levin AB et al. Defining the role of radiosurgery in the management of brain metastases. Int J Radiat Oncol Biol Phys 1992;24:619–625. 158. Cooper JS, Carella R. Radiotherapy of intracerebral metastatic malignant melanoma. Radiology 1980;134:735–738. 159. Stridsklev IC, Hagen S, Klepp O. Radiation therapy for brain metastases from malignant melanoma. Acta Radiol Oncol 1984;23: 231–235. 160. Carella RJ, Gelber R, Hendrikson F et al. Value of radiation therapy in the management of patients with cerebral metastases from malignant melanoma. Cancer 1980;45:679–683. 161. Katz HR. The relative effectiveness of radiation therapy, corticosteroids, and surgery in the management of melanoma metastatic to the central nervous system. Int J Radiat Oncology Biol Phys 1981;7:897–906. 162. Borgelt B, Gelber R, Kramer S et al. The palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1980;6:1–9. 163. Mandell L, Hilaris B, Sullivan M et al. The treatment of single brain metastasis from non–oat cell lung carcinoma: surgery and radiation versus radiation alone. Cancer 1986;58:641–649. 164. Patchell RA, Cirrincione C, Thaler HT et al. Single brain metastases: Surgery plus radiation or radiation alone. Neurology 1986;36: 447–453. 165. Bremer AM, West CR, Didolkar MS. An evaluation of the surgical management of melanoma of the brain. J Surg Oncol 1978;10: 211–219. 166. Hafstrom L, Jansson PE, Stromblad LG. Intracranial metastases of malignant melanoma treated by surgery. Cancer 1980;46: 2088–2090. 167. Hayward RD. Secondary malignant melanoma of the brain. Clin Oncol 1976;2:227–232. 168. Oredsson S, Ingvarf C, Stromblad LG, et al. Palliative surgery for brain metastases of malignant melanoma. Eur J Surg Oncol 1990;16:451–456. 169. Rate WR, Solin LJ, Turrisi AT. Palliative radiotherapy for metastatic malignant melanoma: brain metastases, bone metastases, and spinal cord compression. Int J Radiat Oncol Biol Phy 1988;15:859–864. 170. Pladdet I, Boven E, Nauta J et al. Palliative care for brain metastases of solid tumour types. Neth J Med 1989;34:10–21. 171. Zimm S, Wampler GL, Stablein D et al. Intracerebral metastases in solid-tumor patients: natural history and results of treatment. Cancer 1981;48:384–395. 172. Flickinger JC, Kondziolka D, Lunsford LD et al. A multi-institutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Radiation Oncology Biol Phy 1994;28:797–802 173. Alexander E, III, Moriarty TM, Davis RB et al. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Natl Cancer Inst 1995;87;34–40. 174. Engenhart R, Kimmig BN, Hover KH et al. Long-term follow-up for brain metastases treated by percutaneous stereotactic single high-dose irradiation. Cancer 1993;71:1353–1361. 175. Shu H-KG, Sneed PK, Shiau C-Y et al.. Factors influencing survival after gamma knife radiosurgery for brain metastases: implications for development of treatment guidelines in patients with single and multiple lesions. Cancer J Sci Am 1996:2:335. 176. Isokangas OP, Muhonen T, Kajanti M et al. Radiation therapy of intracranial malignant melanoma. Radiother Oncol 1996;38:139–144.

Chapter 28 / Neurologic Complications of Melanoma

547

177. Seung SK, Shu HG, McDermott MW et al. Stereotactic radiosurgery for malignant melanoma to the brain. Surg Clin North Am 1996;76:1399–1411. 178. Brown PD, Brown CA, Polloc BE et al. Stereotactic radiosurgery for patients with “radioresistant” brain metastases. Neurosurgery 2002;51:656–667. 179. Yu C, Chen JC, Apuzzo ML et al. Metastatic melanoma to the brain: prognostic factors after gamma knife radiosurgery. Int J Radiat Oncol Biol Phys 2002;52:1277–1287. 180. Borgelt B, Gelber R, Larson M et al. Ultra-rapid high dose irradiation schedules for the palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1981;7:1633–1638. 181. Kurtz JM, Gelber R, Brady LW et al. The palliation of brain metastases in a favorable patient population: a randomized clinical trial by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 1981:7:891–895. 182. DeAngelis L, Delattre J, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology 1989;39:789–796. 183. Sundaresan N, Galicich J. Surgical treatment of brain metastases: clinical and computerized tomography evaluation of the results of treatment. Cancer 1985;55:1382–1388. 184. Lee Y, Naubert C, Glass P. Treatment related white matter changes in cancer patients. Cancer 1986;57:1473–1482. 185. Bindal RK, Sawaya R, Leavens ME et al. Surgical treatment of multiple brain metastases. J Neurosurg 1993;79:210–216. 186. Ewend MG, Carey LA, Brem H. Treatment of melanoma metastases in the brain. Semin Surg Oncol 1996;12:429–435. 187. Patchell RA, Tibbs PA, Walsh JW et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med 1990;322:494–500. 188. Constantini S, Lornowski R, Pomeranz S et al. Thromboembolic phenomena in neurosurgical patients operated upon for primary and metastatic brain tumours. Acta Neurochir (Wein) 1991;109:93–97. 189. Burt M, Wronski M, Arbit E et al. Resection of brain metastases from non-small cell lung carcinoma: results of therapy. Memorial Sloan–Kettering Cancer Center Thoracic Surgical Staff. J Thorac Cardiovasc Surg 1992;103:399–410; discussion 410–411. 190. Sawaya R, Bindal RK. Metastatic brain tumors. In: Kaye AH, Laws ER, Jr. (eds.). Brain Tumors: An Encyclopedic Approach. Edinburgh: Churchill Livingstone, 1995:923–946. 191. Bindal RK, Sawaya R, Leavens ME et al. Reoperation for recurrent metastatic brain tumors. J Neurosurg 1995;83:600–604. 192. Skibber JM, Soong SJ, Austin L et al. Cranial irradiation after surgical excision of brain metastases in melanoma patients. Ann Surg Oncol 1996;3:118–123. 193. Vecht CJ, Haaxma-Reiche H, Noordijk EM et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol 1993:33:583–590. 194. Mintz AH, Kestle J, Rathbone MP et al. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer 1996;78:1470–1476. 195. Kondziolka D, Patel A, Lunsford LD et al. Stereotactic radiosurgery plus whole brain radiotherapy versus radiotherapy alone for patients with multiple brain metastases. Int J Radiat Oncol Biol Phys 1999;45:427–434. 196. Andrews DW, Scott CB, Sperduto PW et al. Whole-brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 2004;363:1665–1672. 197. Chougule PB, Burton-Williams M, Saris S et al. Randomized treatment of brain metastases with gamma knife radiosurgery, whole brain radiotherapy or both. Int J Radiat Oncol Biol Phys 2000:48:114. 198. Adler JR, Cox RS, Kaplan I et al. Stereotactic radiosurgical treatment of brain metastases. J Neurosurg 1992;76:444–449. 199. Kihlstrom L, Karlsson B, Lindquist C. Gamma knife surgery for cerebral metastases: implications for survival based on 16 years experience. Stereotact Funct Neurosurg 1993;1:45–50. 200. Young RF, Vermeulen SS, Posewitz AE et al. Gamma knife radiosurgery for treatment of multiple brain metastases. In: Second Congress of the International Stereotactic Radiosurgery Society, Boston, June 14–17, 1995:64–65. 201. Gaudy-Marqueste C, Regis J-M, Muracciole X et al. Gamma-knife radiosurgery in the management of melanoma patients with brain metastases: a series of 106 patients without whole-brain radiotherapy. Int J Radiat oncol Biol Phys 2006;65:809–816. 202. Loeffler JS, Barker FG, Chapman PH. Role of radiosurgery in the management of central nervous system metastases. Cancer Chemother Pharmacol 1999;43 (Suppl):S11–S14. 203. Shiau C, Sneed PK, Shu HG et al. Radiosurgery for brain metastases: relationship of dose and pattern of enhancement to local control. Int J Radiation Oncology Biol Phys 1997;37:375–383. 204. Auchter R, Lamond J, Alexander E et al. A multi-institutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis. Int J Radiat Oncol Biol Phys 1996;35:27–35. 205. Flickinger J, Kondziolka D, Lunsford L et al. A multi-institutional experience with stereotactic radiosurgery for solitary brain metastasis. Int J Radiat oncol Biol Phys 1994;28:797–802. 206. Joseph J, Adler JR, Cox RS et al. Linear accelerator-based stereotaxic radiosurgery for brain metastases: the influence of number of lesions on survival. J Clin Oncol 1996;14:1085–1092. 207. Gieger M, Wu J, Ling MN et al. Response of intracranial melanoma metastases to stereotactic radiosurgery. Radiat Oncol Invest 1997;5:72–80. 208. Mori Y, Kondziolka D, Flickinger JC et al. Stereotactic radiosurgery for cerebral metastatic melanoma: factors affecting local disease control and survival. Int J Radiat Oncol Biol Phys 1998;42:581–589. 209. Grob JJ, Regis J, Laurans R et al. Radiosurgery without whole-brain radiotherapy in melanoma brain metastases. Eur J Cancer 1998;34:1187–1192. 210. Seung SK, Sneed PK, McDermott MW et al. Gamma knife radiosurgery for malignant melanoma brain metastases. Cancer J Sci Am 1998;4:103–109. 211. Chang SD, Lee E, Sakamoto GT et al. Stereotactic radiosurgery in patients with multiple brain metastases. Neurosurg Focus 2000;9:1–5.

548

Part VII / Neurologic Complications of Specific Malignancies

212. Mingione V, Oliveria M, Prasad D et al. Gamma surgery for melanoma metastases in the brain. J Neurosurg 2002;96:544–551. 213. Fife KM, Colman MH, Stevens GN et al. Determinants of outcome in melanoma patients with cerebral metastases. J Clin Oncol 2004;22:1293–1300. 214. Chang El, Selek U, Hassesnbusch SJ, III et al. Outcome variation among “radioresistanat” brain metastases treated with stereotactic radiosurgery. Neurosurgery 2005;56:936–945. 215. Rao G, Klimo P, Thompson CJ et al. Stereotactic radiosurgery as therapy for melanoma, renal carcinoma, and sarcoma brain metastases: impact of added surgical resection and whole brain radiotherapy. Int J Radiat Oncol Biol Phys 2006:66:S20–S25. 216. DiLuna ML, King JT, Knisely JPS et al. Prognostic factors for survival after stereotactic radiosurgery vary with the number of cerebral metastases. Cancer 2007;107:135–145. 217. Bhatnagar AK, Flickinger JC, Kondziolka D et al. Stereotactic radiosurgery for four or more intracranial metastases. Int J Radiat Oncol Biol Phys 2006;64:898–903. 218. Peterson AM, Meltzer CC, Evanson EJ et al. MR imaging response of brain metastases after gamma knife stereotactic radiosurgery. Radiology 1999;211:807–814. 219. Goodman K, Sneed PK, McDermott M et al. Relationship between pattern of enhancement and local control of brain metastases after radiosurgery. Int J Radiat Oncol Biol Phys 2001;50:139–146. 220. Baumert BG, Rutten I, Dehing-Oberije C et al. A pathology-based substrate for target definition in radiosurgery of brain metastases. Int J Radiat Oncol Biol Phys 2006;66:187–194. 221. Noel G, Simon JM, Valery CA et al. Radiosurgery for brain metastasis: impact of CTV on local control. Radiother Oncol 2003;68: 15–21. 222. Loeffler JS, Flickinger JC, Shrieve DC. Radiosurgery for the treatment of intracranial lesions. In: Devita VT, Hellman S, Rosenberg SA, (eds.) Important Advances in Oncology. Philadelphia: W.B. Lippincott, 1995:141–156. 223. Lutterbach J, Cyron D, Henne K et al. Radiosurgery followed by planned observation in patients with one to three brain metastases. Neurosurgery 2003;52:1066–1074. 224. Somaza S, Kondziolka D, Lunsford LD et al. Stereotactic radiosurgery for cerebral metastatic melanoma. J Neurosurg 1993;79: 661–666. 225. Aoyama H Shirato H, Nakagawa K et al. Interim report of the JROSG99–1 multi-institutional randomized trial, comparing SRS alone vs. WBI+SRS for 1–4 brain metastases. J Clin Oncol 2004;22:108 (suppl; abstr 1506). 226. Shehata MK, Young B, Reid B et al. Stereotactic radiosurgery of 468 brain metastases ≤2 cm: implications for SRS dose and whole-brain radiation therapy. Int J Radiat Oncol Biol Phys 2004;59:87–93. 227. Chang EL, Hasenbusch SJ, 3rd , Shiu AS et al. The role of tumor size in the radiosurgical mangagement of patients with ambiguous brain metastases. Neurosurgery 2003;53:272–280. 228. Varlotto JM, Flickinger JC, Niranjan A et al. Analysis of tumor control and toxicity in patients who have survived at least one year after radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys 2003;57:452–464. 229. Chidel MA, Suh JH, Reddy CA et al. Application of recursive partitioning analysis and evaluation of the use of whole-brain radiation among patients treated with stereotactic radiosurgery for newly diagnosed brain metastases. Int J Radiat Oncol Biol Phys 2000;47:993–999. 230. Prizkall A, Debus J, Lohr F et al. Radiosurgery alone of in combination with whole-brain radiotherapy for brain metastases. J Clin Oncol 1998;16:3563–3569. 231. Sneed PK, Suh JH, Goetsch SJ et al. A multi-institutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases. Int J Radiat Oncol Biol Phys 2002;53:519–526. 232. Sansur CA, Chin LS, Ames JW et al. Gamma knife radiosurgery for the treatment of brain metastases. Stereo Funct Neurosurg 2000;74:37–51. 233. Simonova G, Liscak R, Novotny J, Jr. et al. Solitary brain metastases treated with the Leskell gamma knife: prognostic factors for patients. Radiother Oncol 2000;57:519–526. 234. Friehs GM, Legat J, Zheng Z et al. Outcomes in patients treated with gamma knife radiosurgery for brain metastases from malignant melanoma. Neurosurg Focus 1998;4:e1. 235. Christodoulou C, Bafaloukos D, Linardou H et al. Temozolomide (TMZ) combined with cisplatin (CDDP) in patients with brain metastases from solid tumors: a Hellenic Cooperative Oncology Group (HeCOG) phase II study. J Neurooncol 2005;71:61–65. 236. Manon R, O’Neill A, Knisely J et al. Phase II trial of radiosurgery for one to three newly diagnosed brain metastases from renal cell carcinoma, melanoma, and sarcoma: an Eastern Cooperative Oncology Group Study (E6397). J Clin Oncol 2005:23:8870–8876. 237. Lavine SD, Petrovich Z, Cohen-Gadol AA et al. Gamma knife radiosurgery for metastatic melanoma: an analysis of survival, outcome, and complications. Neurosurgery 1999;44:59–64. 238. DiBiase SJ, Chin LS, Ma L. Influence of gamma knife radiosurgery on the quality of life in patients with brain metastases. Am J Clin Oncol 2002;25:131–134. 239. Mehta MP, Tsao MN, Whelan TJ et al. The American Society for Therapeutic Radiology and Oncology (ASTRO) evidence-based review of the role of radiosurgery for brain metastases. Int J Radiat Oncol Biol Phys 2005;63:37–46. 240. Rutiglian MJ, Lunsford LD, Kondziolka D et al. The cost-effectiveness of stereotactic radiosurgery versus surgical resection in the treatment of solitary metastatic brain tumors. Neurosurgery 1995;37:445–453. 241. Smalley SR, Laws ER, O’Tallon JR et al. Resection for solitary brain metastases. J Neurosurg 1992;77:531–540. 242. O’Neill B, Itirria N, Link M et al. A comparison of surgical resection and stereotactic radiosurgery in the treatment of solitary brain metastases. Int J Radiat Oncol Biol Phys 2003; 55:1169–1176. 243. Greig NH. Chemotherapy of brain metastases: current status. Cancer Treat Rev 1984;11:157–186. 244. Rosner D, Nemoto T, Pickren J et al. Management of brain metastases from breast cancer by combination chemotherapy. J Neurooncol 1983;1:131–137.

Chapter 28 / Neurologic Complications of Melanoma

549

245. Cascino T, Byrne M, Deck H et al. Intra-arterial BCNU in the treatment of metastatic brain tumors. J Neuro-oncol 1983;1:211–218. 246. Lange O, Scheef W, Haase K. Palliative radio-chemotherapy with ifosfamide and BCNU for breast cancer patients with cerebral metastases: a 5-year experience. Cancer Chemother Pharmacol 1990;26(suppl):78–80. 247. Ushio Y, Arita N, Hayakawa T et al. Chemotherapy of brain metastases from lung carcinoma: a controlled randomized study. Neurosurgery 1991;28:201–205. 248. Twelves C, Shouhami R, Harper P et al. The response of cerebral metastases in small cell lung cancer to systemic chemotherapy. Br J Cancer 1990;61:147–150. 249. Comis RL. DTIC in malignant melanoma: a perspective. Cancer Treat Rep 1976;60:165–176. 250. Lee SM, Betticher DC, Thatcher N. Melanoma: chemotherapy. Br Med Bull 1995;51:609–630. 251. Kolaric K, Roth A, Jelicic I et al. Phase II clinical trial of cis-dichlorodiammine platinum (cis DDP) in metastatic brain tumors. J Cancer Res Clin Oncol 1982;104:287–293. 252. Legha S, Ring S, Plager C et al. Evaluation of a triple-drug regimen contain cisplatin (C), vinblastine (V), and DTIC (D) in patients (pts) with metastatic melanoma. Proc Am Soc Clin Oncol 1988;7:250 (abstract). 253. Herzig RH, Wolff SN, Fay JW et al. Treatment of advanced melanoma with high-dose chemotherapy anad autologous marrow transplantation. Proceedings of the Third International Autologous Bone Marrow Transplantation Symposium, Houston, TX 1987: 531–536. 254. Lazarus HM, Herzig RH, Wolff SN et al. Treatment of metastatic malignant melanoma with intensive melphalan and autologous bone marrow transplantation. Cancer Treat Rep 1985;69:473–477. 255. Thomas MR, Robinson WA, Glode LM et al. Treatment of advanced malignant melanoma with high-dose chemotherapy and autologous bone marrow transplantation: preliminary results—phase I study. Am J Clin Oncol 1982;5:611–622. 256. Legha SS. Durable complete responses in metastatic melanoma treated with interleukin-2 in combination with interferon-alpha and chemotherapy. Semin Oncol 1997;24(1 suppl 4):S39–S43. 257. Atkins MB, Lotze MT, Dutcher JP et al. High-dose recombinant interleukin-2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol 1999;17:2105–2116. 258. Richards JM, Gale D, Mehta N et al. Combination of chemotherapy with interleukin-2 and interferon-alpha for the treatment of metastatic melanoma. J Clin Oncol 1999;17:651–657. 259. Boasberg PD, O’Day SJ, Kristedja TS et al. Biochemotherapy for metastatic melanoma with limited central nervous system involvement. Oncology 2003; 64:328–335. 260. Newlands ES, Blackledge GR, Slack JA et al. Phase I trial of temozolomide (CCRG 81045: M&B 39831: NSC 362856). Br J Cancer 1992;65:287–291. 261. Bleehen NM, Newlands ES, Lee SM et al. Cancer Research Campaign Phase II trial of temozolomide in metastatic melanoma. J Clin Oncol 1995;13:910–913. 262. Middleton MR, Grob JJ, Aronson N et al. Randomized phase III study of temozolomide versus dacarbazine in the treatment of patients with advanced metastatic malignant melanoma. J Clin Onc 2000;18:158–166. 263. Agarwala SS, Kirkwood JM. Temozolomide, a novel alkylating agent with activity in the central nervous system, may improve the treatment of advanced metastatic melanoma. Oncologist 2000;5:144–151. 264. Dreiling L, Hoffman S, Robinson WA. Melanoma: epidemiology, pathogenesis, and new modes of treatment. Adv Intern Med 1996;41:553–604. 265. Paul MJ, Summers Y, Calvert AH et al. Effect of temozolomide on central nervous system relapse in patients with advanced melanoma. Melanoma Res 2002;12:175–178. 266. Kersten MJ, Boogerd W, Batchelor D et al. Combined immunotherapy with GM-CSF, IF-2, and IFN allows dose escalation of temozolomide and prevention of lymphocytopenia and brain metastases in metastatic malignant melanoma. 36th Annual Meeting of the American Society of Clinical Oncology, May 20–23, 2000. Abstract 2244. 267. Atkins MB, Bollob JA, Sosman JA et al. A phase II pilot trial of concurrent biochemotherapy with cisplatin, vinblastine, temozolomide, interleukin-2, and IFN-alpha 2B in patients with metastatic melanoma. Clin Cancer Res 2002;8:3075–3081. 268. Summers Y, Middleton MR, Calvert H et al. Effect of temozolomide (TMZ) on central nervous system relapse in patients with advanced melanoma. 35th Annual Meeting of the American Society of Clinical Oncology, May 15–18, 1999. Abstract 2048. 269. Bafaloukos D, Gogas H, Gerogoulias V et al. Temozolomide in combination with docetaxel in patients with advanced melanoma: a phase II study of the Hellenic Co–operative Oncology Group. J Clin Oncol 2002;20:420–425. 270. Ready N, Aronson F, Wanebo H et al. A low rate of central nervous system progression in a phase II trial of outpatient chemobiologic therapy with cisplatin, temozolomide, interleukin-2, and interferon alfa 2-B for metastatic malignant melanoma. Am J Clin Oncol 2005;28:479–484. 271. Biasco G, Pantaleo MA, Casadei S. Treatment of brain metastases of malignant melanoma with temozolomide. N Engl J Med 2001;345:621–622. 272. Agarwala SS, Kirkwod JM, Gore M et al. Temozolomide for the treatment of brain metastases associated with metastatic melanoma: a phase II study. J Clin Oncol 2004;22:2101–2107. 273. Dvorak J, Melichar B, Zizka J et al. Complete response of multiple melanoma brain metastases after treatment with temozolomide. Onkologie 2004;27:171–174. 274. Conill C, Pig S, Toscas I, Castel T. Complete response of cerebral metastatic melanoma after a combined treatment of radiotherapy and temozolomide. Med Clin (Barc) 2002;119:758–759 275. Margolin K, Atkins MB, Thompson JA et al. Temozolomide and whole-brain irradiation in melanoma metastatic to the brain: a phase II trial of the Cytokine Working Group. J Cancer Res Clin Oncol 2002;128:214–218. 276. Hoffman M, Kiecker F, Wurm R et al. Temozolomide with or without radiotherapy in melanoma with unresectable brain metastases. J Neuro-oncol 2006;76:59–64.

550

Part VII / Neurologic Complications of Specific Malignancies

277. Bafaloukos D, Tsoutsos D, Fountzilas G et al. The effect of temozolomide-based chemotherapy in patients with cerebral metastases from melanoma. Melanoma Res 2004;14:289–294. 278. Hwu WJ, Raizer J, Panageas KS, Lis E. Treatment of metastatic melanoma in the brain with temozolomide and thalidomide. Lancet Oncol 2001;2:634–635. 279. Jacquillat C, Khayat D, Banzet P et al. Final report of the French multicenter phase II study of nitrosourea fotemustine in 153 evaluable patients with disseminated malignant melanoma including patients with cerebral metastases. Cancer 1990;66:1873–1878. 280. Kleeberg UR, Engel E, Israels P et al. Palliative therapy of melanoma patients with fotemustine: inverse relationship between tumour load and treatment effectiveness. A multicenter phase II trial of the EORTC. Melanoma Cooperative Group (MCG). Melanoma Res 1995;5:195–200. 281. Calabresi F, Aapro M, Becquart D et al. Multicenter phase II trial of the single agent fotemustine in patients with advanced malignant melanoma. Ann Oncol 1991;2:377–378. 282. Falkson CI, Falkson G, Falkson HC. Phase II trial of fotemustine patients with metastatic malignant melanoma. Invest New Drugs 1994;12:251–254. 283. Avril MF, Aamdal S, Grob JJ et al. Fotemustine compared with dacarbazine in patients with disseminated malignant melanoma: a phase III study. J Clin Oncol 2004;22:1118–1125. 284. Schadendorf D, Hauschild A, Ugurel S et al. Dose-intensified biweekly temozolomide in patients with asymptomatic brain metastases from malignant melanoma: a phase II DeCOG/ADO study. Ann Oncol 2006;17:1592–1597. 285. Hwu W-J, Lis E, Menell JH et al. Temozolomide plus thalidomide in patients with brain metatases from melanoma: a phase II study. Cancer 2005;103:2590–2597. 286. Krown S, Niedzwiecki D, Hwu W-J et al. Phase II study of temozolomide and thalidomide in patients with metastatic melanoma in the brain. Cancer 2006;107:1883–1890. 287. Strauss SJ Marples M, Napier MP et al. A phase I (tumor site–specific) study of carboplatin and temozolomide in patients with advanced melanoma. Br J Cancer 2003;89:1901–1905. 288. Larkin JMG, Hughes SA, Beirne DA et al. A phase I/II study of lomustine and temozolomide in patients with cerebral metastases from malignant melanoma. Br J Cancer 2007;96:44–48. 289. Boogerd W, de Gast GC, Dalesio O. Temozolomide in advanced malignant melanoma with small brain metastases: can we withhold cranial irradiation? Cancer 2007;109:306–312. 290. Siena S, Landonio G, Baietta E et al. Multicenter phase II study of temozolomide therapy for brain metastasis in patients with malignant melanoma, breast cancer, and non-small cell lung cancer. Proc Am Soc Clin Oncol 2003;22:[Abstract 407]. 291. Brocker EB, Bohndorf W, Kampgen E et al. Fotemustine given simultaneously with total brain irradiation in multiple brain metastases for malignant melanoma: report on a poilot study. Melanoma Res 1996;6:399–401. 292. Ulrich J, Gademann G, Gollnick H. Mangement of cerebral metastases from malignant melanoma: results of a combined simultaneous treatment with fotemustine and irradiation. J Neuro-oncol 1999;43:173–178. 293. Mornex F, Thomas L, Mohr P et al. Randomised phase III trial of fotemustine versus fotemustine plus whole brain irradiation in cerebral metastases of melanoma. Cancer Radiother 2003;7:1–8. 294. Abrey LE, Olson JD, Raizer JJ et al. A phase II trial of temozolomide for patients with recurrent or progressive brain metastses. J Neuro-oncol 2001;53:259–265. 295. Glass JP, Melamed M, Chernik NL et al. Malignant cells in cerebrospinal fluid (CSF): the meaning of a positive CSF cytology. Neurology 1979;29:1369–1375. 296. Olson ME, Chernik NL, Posner JB. Infiltration of the leptomeninges by systemic cancer” a clinical and pathologic study. Arch Neurol 1974;30:122– 137. 297. Little JR, Dale AJ, Okazaki H. Meningeal carcinomatosis: clinical manifestations. Arch Neurol 1974;30:138–143. 298. Wasserstrom WR, Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49:759–772. 299. Ehya H, Hajdu SI, Melamed MR. Cytopathology of nonlymphoreticular neoplasms metastatic to the central nervous system. Acta Cytol 1981;25:599–610. 300. Amer MH, Al-Sarraf M, Baker L et al. Malignant melanoma and central nervous system metastases: incidence, diagnosis, treatment and survival. Cancer 1978;42:660–668. 301. Theodore WH, Gendelman S. Meningeal carcinomatosis. Arch Neurol 1981;38:696–699. 302. Twijnstra A, zan Zanten AP, Hart AA et al. Serial lumbar and ventricle cerebrospinal fluid lactate dehydrogenase activities in patients with leptomeningeal metastases from solid and haematological tumours. J Neurol Neurosurg Psychiatry 1987;50:313–320. 303. Boogerd W, Hart AAM, Van der Sande JJ et al. Meningeal carcinomatosis in breast cancer. Cancer 1991;67:1685–1695. 304. Balm M, Hammack J. Leptomeningeal carcinomatosis: presenting features and prognostic factors. Arch Neurol 1996;53:626–632. 305. Wasserstrom WR, Schwartz MK, Fleisher M et al. Cerebrospinal fluid biochemical markers in central nervous system tumors: a review. Ann Clin Lab Sci 1981;11:239–251. 306. Twijnstra A, Ongerboer de Visser BW, van Zanten AP. Diagnosis of leptomeningeal metastasis. Clin Neurol Neurosurg 1987;89: 79–85. 307. Fleisher M, Wasserstrom WR, Schold SC et al. Lactic dehydrogenase isoenzymes in the cerebrospinal fluid of patients with systemic cancer. Cancer 1981;47:2654–2659. 308. Van Zanten AP, Twijnstra A, Ongerboer de Visser BW et al. Tumour markers in the cerebrospinal fluid of patients with central nervous system metastases from extracranial malignancies. Clin Chim Acta 1988;175:157–166. 309. Grossman SA, Krabak MJ. Leptomeningeal carcinomatosis. Cancer Treat Rev 1999;25:103–119. 310. Krol G, Sze G, Malkin M et al. MR of cranial and spinal meningeal carcinomatosis: comparison with CT and myelography. AJR 1988:151;583–588.

Chapter 28 / Neurologic Complications of Melanoma

551

311. Chamberlain MC, Sandy AD, Press GA. Leptomeningeal metastasis: a comparison of gadolinium-enhanced MR and contrast-enhanced CT of the brain. Neurology 1990;40:85–89. 312. Sze G, Soletsky S, Bronen E et al. MR imaging of the cranial meninges with emphasis on contrast enhancement and meningeal carcinomatosis. AJNR 1989;10:965–997. 313. Glantz MJ, Hall WA, Cole BF et al. Diagnosis, management of, and survival of patients with leptomeningeal cancer based on cerebrospinal fluid-flow status. Cancer 1995;75:2919–2931. 314. Grossman SA,Trump DL, Chen DCP et al. Cerebrospinal fluid flow abnormalities in patients with neoplastic meningitis: an evaluation using 111 Indium-DPTA ventriculography. Am J Med 1982;73:641–647. 315. Chamberlain MC, Corey-Bloom J. Leptomeningeal metastases: 111 Indium-DPTA CSF flow studies. Neurology 1991;41:1765–1769. 316. Mitchell MS. Relapse in the central nervous system in melanoma patients successfully treated with biomodulators. J Clin Oncol 1989;7:1701–1709. 317. Gokaslan Z, Aladag M, Ellerhorst J. Melanoma metastatic to the spine: a review of 133 cases. Melanoma Res 2000;10:78–80. 318. Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 1978;3:40–51. 319. Schiff D, Oneill BP. Intramedullary spinal cord metastases: clinical features and treatment outcome. Neurology 1996;47:906–912. 320. Helwig-Larsen S, Hansen SW, Sorensen PS. Second occurrence of symptomatic metastatic spinal cord compression and findings of multiple spinal epidural meatastases. Int J Radiat Oncol Biol Phys 1995;33:595–598. 321. Schiff D, O’Neill BP, Suman VJ. Spinal epidural metastasis as the initial manifestation of malignancy: clinical features and diagnostic approach. Neurology 1997;49:452–456. 322. Chamberlain MC, Kormanik PA. Epidural spinal cord compression: a single institution’s retrospective experience. Neuro-oncology 1999;1:120–123. 323. Schachar NS. An update on the nonoperative treatment of patients with metastatic bone disease. Clin Orthop Relat Res 2001;382: 75–81. 324. Gerszten PC, Burton SA, Quinn AE et al. Radiosurgery for the treatment of spinal melanoma metastases. Stereotact Funct Neurosurg 2005;83:213–221. 325. Strauss A, Dritschilo A, Nathanson L et al. Radiation therapy of malignant melanomas: an evaluation of clinically used fractionation schemes. Cancer 1981;47:1262–1266. 326. Donaldson WF, III, Peppelmann WC, Yaw KM. Symptomatic malignant melanoma to the spine. J Spinal Disord 1993;6:360–363. 327. Helweg-Larson S. Clinical outcome in metastatic spinal cord compression: a prospective study of 153 patients. Acta Neurol Scand 1996;94:269–275. 328. Kocialkowski A, Webb JK. Metastatic spinal tumors: survival after surgery. Eur Spine J 1992;1:43–48. 329. Speigel DA, Sampson JH, Richardson WJ et al. Metastatic melanoma to the spine. Spine 1995;20:2141–2146. 330. Sorensen S, Borgesen SE, Rohde K et al. Metastatic epidural spinal cord compression. Cancer 1990;65:1502–1508. 331. Gokaslan ZL, Aladaq MA, Ellerhorst JA. Melanoma metastatic to the spine: a review of 133 cases. Melanoma Res 2000;10:78–80. 332. Weigel B, Maghsudi M, Neumann C et al. Surgical management of symptomatic spinal metastases. Spine 1999;24:2240–2246. 333. Faul CM, Flickinger JC. The use of radiation in the management of spinal metastases. J Neuro-oncol 1995;23:149–161. 334. Loblaw DA, Laperriere NJ. Emergency treatment of malignant extradural spinal cord compression: an evidence-based guideline. J Clin Oncol 1998; 16:1613–1624. 335. Ryu S, Chang S, Kim D et al. Image–guided hypofractionated stereotactic radiosurgery to spinal lesions. Neurosurgery 2001:49: 838–846. 336. Ryu S, Yin FF, Rock J et al. Image-guided and intensity-modulated radiosurgery for patients with spinal metastasis. Cancer 2003;97:2013–2018. 337. Milker-Zabel S, Zabel A, Thilmann C et al. Clinical results of retreatment of vertebral bone metastases by stereotactic conformal radiotherapy and intensity–modulated radiotherapy. Int J Radiat Oncol Biol Phys 2003;55:162–167. 338. Benzil DL, Saboori M, Mogilner AY et al. Safety and efficacy of stereotactic radiosurgery for tumors of the spine. J Neurosurg 2004;101:413–418. 339. Bilsky MH, Yamada Y, Yenice KM et al. Intensity-modulated stereotactic radiotherapy of paraspinal tumors: a preliminary report. Neurosurgery 2004;54:823–830. 340. Chang El, Shiu AS, Lii MF et al. Phase I clinical evaluation of near-simultaneous computed tomographic image-guided stereotactic body radiotherapy for spinal metastases. Int J Radiat Oncol Biol Phys 2004;59:1288–1294. 341. De Salles AA, Pedroso AG, Medin P et al. Spinal lesions treated with Novalis shaped beam intensity-modulated radiosurgery and stereotactic radiotherapy. J Neurosurg 2004;101(suppl 3):435–440. 342. Degen JW, Gagnon GJ, Voyadzis JM et al. CyberKnife stereotactic radiosurgerical treatment of spinal tumors for pain control and quality of life. J Neurosurg Spine 2005;2:540–549. 343. Gerszten PC, Germanwala A, Burton SA et al. Combination kyphoplasty and spinal radiosurgery: a new treatment paradigm for pathological fractures. J Neurosurg Spine 2005;3:296–301. 344. Connolly ES, Winfree CJ, McCormick PC et al. Intramedullary spinal cord metastasis: report of three cases and review of the literature. Surg Neurol 1996;46:329–338. 345. Costigan D, Winkleman MD. Intramedullary spinal cord metastasis: a clinicopathological study of 13 cases. J Neurosurg 1985;62: 227–233. 346. Schwechheimer K, Lemminger JM. Intramedullary metastases: report of 4 cases and review of the literature. Clin Neuropathol 1984:4:28–37. 347. Crasto S, Duca S, Davini O et al. MRI diagnosis of intramedullary metastases from extra-CNS tumors. Eur Radiol 1997;7:732–736.

552

Part VII / Neurologic Complications of Specific Malignancies

348. Dunne JW, Harper CG, Pamphlett R. Intramedullary spinal cord metastases: a clinical and pathological study of nine cases. QJM 1986:61:1003–1020. 349. Grem JL, Burgess J, Trump DL. Clinical features and natural history of intramedullary spinal cord metastasis. Cancer 1985;56: 2305–2314. 350. Carlson JA, Dickersin GR, Sober AF et al. Desmoplastic neurotropic melanoma: a clinicopathologic analysis of 28 cases. Cancer 1995;75:478–494. 351. Ogose A, Emura I, Iwabuchi Y et al. Malignant melanoma extending along the ulnar, median, and musculocutaneous nerves: a case report. J Hand Surg 1998;23A:875–878. 352. Hufnagel T, Savino PJ, Zimmerman RA et al. Painful ophthalmoplegia caused by neurotropic malignant melanoma. Can J Ophthalmol 1990;25:38–41. 353. Laskin WB, Weiss SW, Bratthauer GL. Epithelioid variant of malignant peripheral nerve sheath tumor (malignant epithelioid schwannoma). Am J Surg Pathol 1991;15:1136–1145. 354. King R, Busam K, Rosai J. Metastatic malignant melanoma resembling malignant peripheral nerve sheath tumor. Am J Surg Pathol 1999;23:1499–1505. 355. Schadendorf D, Haas N, Ostmeier H et al. Amelanotic malignant melanoma presenting as malignant schwannoma. Br J Dermatol 1993;129:609–14. 356. Sexton M, Maize JC. Malignant melanoma simulating schwannian differentiation. Am J Dermatopathol 1985;7(suppl):171–176. 357. Diaz-Cascajo C, Hoos A. Histopathologic features of malignant peripheral nerve sheath tumors are not restricted to metastatic malignant melanoma and can be found in primary malignant melanoma also. Am J Surg Pathol 2000; 24:1438–1439. 358. Enzinger FM, Weiss SW. Tumors of peripheral nerves. In: Soft Tissue Tumors. 3rd ed. St Louis: Mosby, 1995:889–928. 359. Iwamoto S, Odland PB, Piepkorn M et al. Evidence that the p75 neurotrophin receptor mediates perineural spread of desmoplastic melanoma. J Am Acad Dermatol 1996;35:725–731. 360. Rakowicz-Szulczynska EM, Reddy U, Vorbrodt A et al. Chromatin and cell surface receptors mediate melanoma cell growth response to nerve growth factor. Mol Carcinog 1991;4:388–396. 361. Berson EL, Lessell S. Paraneoplastic night blindness with malignant melanoma. Am J Ophthalmol 1988;106:307–311. 362. Jacobson DM, Adamus G. Retinal antibipolar cell antibodies in a patient with paraneoplastic retinopathy and colon carcinoma. Am J Ophthalmol 2001; 131:806–808. 363. Milam AH, Saari JC, Jacobson SG et al. Autoantibodies against retinal bipolar cells in cutaneous melanoma-associated retinopathy. Invest Ophthalmol Vis Sci 1993; 34:91–100. 364. Potter MJ, Thirkill CE, Dam OM et al. Clinical and immunocytochemical findings in a case of melanoma-associated retinopathy. Ophthalmology 1999;106: 2121–2125. 365. Alexander KR, Fishman GA, Peachey NS et al. “On” response defect in paraneoplastic night blindness with cutaneous malignant melanoma. Invest Ophthalmol Vis Sci 1992;33:477–483. 366. Kiratli H, Thirkill CE, Bilgic S et al. Paraneoplastic retinopathy associated with metastatic cutaneous melanoma of unknown primary site. Eye 1997;11:889–92. 367. Boeck K, Hofmann S, Klopfer M et al. Br J Dermatol 1997;137:457–460. 368. Chan JW. Paraneoplastic retinopathies and optic neuropathies. Surv Ophthalmol 2003;48:12–38. 369. Rush JA. Paraneoplastic retinopathy in malignant melanoma. Am J Ophthalmol 1993;115:390–391. 370. Potter MJ, Thirkill CE, Dam OM et al. Clinical and immunocytochemical findings in a case of melanoma-associated retinopathy. Ophthalmology 1999;106:2121–2125. 371. Berson EL, Lessel S. Paraneoplastic night blindness with malignant melanoma. Am J Ophthalmol 1998;106:307–311. 372. Weinstein JM, Kelman SE, Bresnick GH et al. Paraneoplastic retinopathy associated with antiretinal bipolar cell antibodies in cutaneous malignant melanoma. Ophthalmology 1994:101:1236–1243. 373. Milam AH, Saari JC, Jacobson SG et al. Autoantibodies against retinal bipolar cells in cutaneous melanoma-associated retinopathy. Invest Ophthalmol Vis Sci 1993;34:91–100. 374. Remulla JF, Pineda R, Gaudio AR et al. Cutaneous melanoma-associated retinopathy with retinal periphlebits. Arch Ophthalmol 1995;113:854–855. 375. Gittinger JW, Smith TW. Cutaneous melanoma-associated paraneoplastic retinopathy: histopathologic observations. Am J Ophthalmol 1999;27:612–614. 376. Ling CPW, Pavesio C. Paraneoplastic syndromes associated with visual loss. Curr Opin Ophthalmol 2003;14:426–432. 377. Singh AD, Milam AH, Shields CL et al. Melanoma-associated retinopathy. Am J Ophthalmol 1995;119:369–370. 378. Keltner JL, Thirkill CE, Yip PT. Clinical and immunologic characteristics of melanoma-associated retinopathy syndrome: eleven new cases and a review of 51 previously published cases. J Neuro Ophthalamol 2001;21:173–87. 379. Gittinger JW, Jr, Smith TW. Cutaneous melanoma-associated paraneoplastic retinopathy: histopathologic observations. Am J Ophthalmol 1999;127:612–614. 380. Sotodeh M, Paridaens D, Keunen J et al. Paraneoplastic vitelliform retinopathy associated with cutaneous or uveal melanoma and metastases. Klin Monatsbl Augenheilkd 2005;222:910–914. 381. Potter MJ, Adamus G, Szabo SM et al. Autoantibodies to transducin in a patient with melanoma-associated retinopathy. Am J Ophthalmol 2002;134:128–130. 382. Ladewig G, Reinhold U, Thirkill CE et al. Incidence of antiretinal antibodies in melanoma: screening of 77 serum samples from 51 patients with American Joint Committee on Cancer stage I–IV. Br J Dermat 2005;152:931–938. 383. Kellner U, Bornfeld N, Foerster MH. Severe course of cutaneous melanoma associated retinopathy. Invest Ophthalmol Vis Sci 1994; 35:2117.

Chapter 28 / Neurologic Complications of Melanoma

553

384. Remulla FJ, Pineda R, Gaudio AR et al. Cutaneous melanoma-associated retinopathy with retinal periphlebitis. Arch Ophthalmol 1995; 113:854–855. 385. Vaphiades MS, Brown H, Whitcup SM. Node way out. Surv Ophthalmol 2000; 45:77–83. 386. Milam AH. Clinical aspects: paraneoplastic retinopathy. In: Djamgoz MBA, Archer SN, Vallerga S (eds.). Neurobiology and Clinical Aspects of the Outer Retina. London: Chapman and Hall 1995:461–471. 387. Dalmau J, Furneaux HM, Cordon-Cardo C et al. The expression of the Hu (paraneoplastic encephalomyelitis sensory neuronopathy) antigen in human normal and tumor tissues. Am J Pathol 1992;141:881–886. 388. Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and oncologic profiles of patients seropositive for type I antineuronal nuclear autoantibodies. Neurology 1998;50:652–657. 389. Antonine JC, Mosnier FJ, Lapras J et al. Chronic inflammatory demyelinating polyneuropathy associated with carcinoma. J Neurol Neurosurg Psychiatry 1996;60:188–190. 390. Bird SJ, Brown MJ, Shy ME et al. Chronic inflammatory demyelinating polyneuropathy associated with malignant melanoma. Neurology 1996:46:822–824. 391. Saleh MN, Khazaeli MB, Wheeler RH et al. Phase I trial of the murine monoclonal anti-Gd2 antibody 14F2a in metastatic melanoma. Cancer Res 1992;52:4342–4347. 392. Fuller GN, Spies JM, Pollard JD et al. Demyelinating neuropathies triggered by melanoma immunotherapy. Neurology 1994;44: 2404–2405. 393. Yokoyama A, Kimura Y, Shigemura J. Psychiatric side effects of interferon. J Toxicol Sci 1996;21:93–96. 394. Adams F, Fernandez F, Mavligit G. Interferon-induced organic mental disorders associated with unsuspected pre-existing neurologic abnormalities. J Neurooncol 1988;6:355–3559. 395. Iivanainen M, Laaksonen R, Niemi ML et al. Memory and psychomotor impairment following high-dose interferon treatment in amyotrophic lateral sclerosis. Acta Neurol Scand 1985;72:475–480. 396. Muss HB, Richards F, Homesley HD et al. A phase I trial of recombinant leukocyte alpha 2 interferon in patients with advanced malignancy. Am J Clin Oncol 1985;8:97–107. 397. Farkkila M, Iivanainen M, Roine R et al. Neurotoxic and other side effects of high-dose interferon in amyotrophic lateral sclerosis. Acta Neurol Scand 1984;70:42–46. 398. Muss HB, Kempf RA, Martino S et al. A phase II study of recombinant alpha interferon in patients with recurrent or metastatic breast cancer. J Clin Oncol 1984;2:1012–6. 399. Suter CC, Westmoreland BF, Sharbrough FW et al. Electroencephalographic abnormalities in interferon encephalopathy: a preliminary report. Mayo Clin Proc 1984;59:847–850. 400. Honigsberger L, Fielding JW, Priestman TJ. Neurologic effects of recombinant human interferon (letter). Br Med J 1983;286:719. 401. Rohatiner AZ, Prior PF, Burton AC et al. Central nervous system toxicity of interferon. Br J Cancer 1983:47:419–422. 402. Valentine AD, Meyers CA, Kling MA et al. Mood and cognitive side effects of interferon-a therapy. Semin Oncol 1998(suppl 1);25: 39–47. 403. Bocci V. Central nervous system toxicity of interferons and other cytokines. J Biol Regul Homeos Agents 1988;2:107–118. 404. Meyers CA, Scheibel RS, Forman AD. Persistent neurotoxicity of systemically administered interferon-. Neurology 1991;41: 672–676. 405. Meyers CA, Obbens EAMT, Scheibel RS et al. Neurotoxicity of intraventricularly administered alpha-interferon for leptomeningeal disease. Cancer 1991; 68:88–92.

29

Neurologic Complications of Leukemia Marc C. Chamberlain,

MD

CONTENTS Introduction Leukemic Parenchymal Tumor Intracranial Hemorrhage Encephalopathy Meningitis Epidural Spinal Cord Compression Radiculopathy Peripheral Neuropathy Myopathy Conclusions References

Summary Leukemia affects both the central and peripheral nervous systems. Neurological complications are a consequence of both direct leukemic infiltration, as occurs with leukemic meningitis, and complications of either antileukemic treatment (e.g., thrombocytopenic or DIC-related intracranial hemorrhage, steroid myopathy, vinca alkaloid peripheral neuropathy, posterior reversible encephalopathy syndrome, multifocal necrotizing leukoencephalopathy) or immune compromise (e.g., Herpes zoster shingles or Aspergillus infection). Key Words: leukemia, meningitis, neurologic complications, leptomeningeal metastases

1. INTRODUCTION Leukemia is classified into acute and chronic types, and further separated into subtypes based on whether tumors are comprised of cells that appear mature (chronic leukemia) or immature (acute leukemia) and whether cell lineage is lymphoid or myeloid (1–3). Within each category, distinct leukemias are defined according to a combination of morphology, immunophenotype, and cytogenetic features in addition to clinical syndrome. An estimated 30,800–33,400 new cases of leukemia will be diagnosed in the United States this year. Acute leukemia, a clonal disease of hematopoietic stem cells, account for slightly more than half of all new leukemias in the United States annually. Hematopoietic stem cells may differentiate along lymphoid or myeloid lines. In adults, acute myelogenous leukemia (AML), also called acute nonlymphocytic leukemia, is three times more common than acute lymphocytic leukemia (ALL) and represents 60–70% of all acute leukemia; 11,000–12,000 new cases occur annually in the United States. AML is most common in individuals older than 50 years of age, whereas ALL is more common in children and young adults. Approximately one-third of patients with either ALL or AML achieve long-term survival; however, outcome depends significantly on cytogenetic profile. Chronic lymphocytic leukemia (CLL) is the second most From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

555

556

Part VII / Neurologic Complications of Specific Malignancies

common adult leukemia and affects 8,000–9,000 persons in the United States annually. Like AML, CLL is more common in the elderly. CLL represents a monoclonal disorder with expansion of small lymphocytes of B-cell (95%) or T-cell (5%) lineage. Median survival is 6 years, but depends on staging (as per the Rai staging system) at disease presentation. CLL only rarely progresses to a more malignant phenotype. Chronic myelogenous leukemia (CML) is characterized by excessive clonal proliferation of myeloid cells and affects 4,000–5,000 adults annually in the

Table 1 Neurologic Complications of Leukemia • Direct J Meningeal  Leukemic J Parenchymal  Tumor  Hemorrhage Vascular sludging/stasis due to hyperleukocytosis Thrombocytopenia due to leukemia or treatment J Epidural  Leukemic • Indirect J Meningeal  Infectious Bacterial meningitis Fungal meningitis Chemical meningitis  Headache Low-pressure headache (post-lumbar puncture) Subdural hematoma J Parenchymal  Hemorrhage Treatment-induced sinus thrombosis (l-asparaginase) Treatment-induced thrombocytopenia Moyamoya disease Disseminated intravascular coagulation Fungal-related • Mycotic aneurysm • Vasculitis  Encephalopathy Radiation-related Methotrexate related Toxic-metabolic Organ failure Posterior reversible encephalopathy syndrome (PRES) Multifocal necrotizing leukoencephalopathy J Spinal  Treatment-related myelopathy (intrathecal drugs) J Epidural  Hemorrhage Treatment-induced thrombocytopenia  Steroid-related epidural lipomatosis J Peripheral neuropathy  Treatment-related (vinca alkaloids) J Myopathy  Treatment-related (corticosteroids)  

  

 

    

     



Chapter 29 / Neurologic Complications of Leukemia

557

United States. The disease can be divided into two phases, an initial chronic phase in which cell maturation is normal followed by an acute phase (blast crisis), characterized by maturation arrest. Median survival of CML is 4 years. The neurological manifestations of leukemia are diverse and reflect either direct tumor involvement or indirect complications of immunosuppression or therapy. The following discussion of the neurological complications of leukemia amplifies that outlined in Table 1.

2. LEUKEMIC PARENCHYMAL TUMOR AML may give rise to solid tumors consisting of myeloid leukemic blasts called granulocytic sarcomas or chloromas (4–6). The term chloroma results from the greenish color of these tumors caused by the presence of myeloperoxidase. Chloromas usually have a dural attachment, and although rare, parenchymal tumors have been reported. These tumors are hypercellular and avidly enhance after contrast administration with either cranial MR or CT imaging. Neurologic findings depend on tumor location. Chloromas most often occur in dura (where they may simulate meningiomas) or bone. Bone involvement may result in epidural spinal cord compression, while involvement of the orbit may result in proptosis and a restrictive ophthalmopathy. Although chloromas are very radiosensitive, their presence typically heralds aggressive systemic disease such that disease control is a function of extracranial therapy and response.

3. INTRACRANIAL HEMORRHAGE Hemorrhagic complications are common in patients with acute leukemia (approximately 20%) and constitute the second most common cause of death in such patients (20% of all leukemic deaths result from intracranial hemorrhage) (4,7–15). Intracranial hemorrhage (ICH) is the most common hemorrhagic complication in acute promyelocytic leukemia and is not infrequent in AML and ALL (2–18% of all patients with acute leukemia). ICH may occur at time of diagnosis (early hemorrhage) or subsequent to diagnosis and following initial treatment (late hemorrhage). Disseminated intravascular coagulation (DIC), disseminated aspergillosis or mucormycosis, leukemic cell infiltration, thrombocytopenia or l-asparaginase chemotherapy-related in that order, are the most common etiologies for ICH. Both DIC (especially common in the M3 subtype of AML) and thrombocytopenia typically result in a solitary, often massive ICH, whereas ICH occuring during neutropenia is associated with disseminated fungal infection and is the result of hemorrhagic infarction. Leukemic cell infiltration occurs with marked leukocytosis (defined as >300,000 leukemic cells/μL) and results in multiple intracranial hemorrhages. l-asparaginase may induce hyperfibrinogenemia and result in cortical vein or sinus thrombosis with resulting venous infarction. Fungal-related mycotic aneurysms and ICH would be a consideration in a patient with blood culture positive for fungus. Topographically, the majority of ICH is intraparenchymal with cerebral hemorrhage more common than cerebellar hemorrhage. Subdural hematoma is a relatively infrequent complication except following stem cell transplantation. Aside from symptomatic treatment of ICH, treatment is directed at the underlying cause of hemorrhage (e.g., correction of an underlying coagulopathy, whole-brain irradiation or systemic chemotherapy for hyperleukocytosis and brain leukemic infiltration). In general, ICH in the patient with leukemia portends shortened survival.

4. ENCEPHALOPATHY A variety of etiologies may account for encephalopathy (defined as an alteration in consciousness, neurobehavioral abnormalities, seizures, or focal neurological deficits) in the leukemic patient. Most commonly, toxic/metabolic (narcotic overmedication, hyponatremia, uremia, organ failure) causes are identified; however, consideration of DIC, sinus thrombosis, ICH, radiation and chemotherapy-related (either high-dose methotrexate or cytarabine) complications, and infections (disseminated Candida or Aspergillus) is necessary (4,7–15). Evaluation for DIC is warranted in any patient with leukemia and encephalopathy and should include a coagulopathy blood evaluation. Sinus thrombosis is occasionally due to leukemic infiltration of the superior sagittal sinus but more often occurs with dehydration, sepsis-related coagulopathy or l-asparaginase chemotherapy. Clinical presentation may be as an isolated headache, raised intracranial pressure syndrome (headache, nausea, vomiting, transient visual obscurations, and diplopia), hemiparesis, or encephalopathy. Cranial imaging most often

558

Part VII / Neurologic Complications of Specific Malignancies

demonstrates a venous hemorrhagic stroke. Chemotherapy-related encephalopathy is seen following high-dose methotrexate as either a transient diffuse encephalopathy or occasionally as a posterior reversible leukoencephalopathy defined best by cranial MR with posterior quadrant white matter high signal abnormalities. Highdose cytarabine (ara-C), used in the treatment of AML, causes either a pure cerebellar syndrome or diffuse encephalopathy and is more common in the elderly patient (age >60 years) and in association with renal impairment. Radiation-related encephalopathy occurs in two contexts—as an early-delayed side effect or, more commonly, as a late-delayed radiation complication (4,12,16). In both instances, radiation therapy is given as prophylactic whole-brain radiotherapy in high-risk (for leukemic meningitis) patients. Early-delayed radiation complication, occurring weeks after completion of radiation therapy, is a generalized demyelinating syndrome that presents with hypersomnolence, is benign, and resolves with steroid treatment. Late-delayed radiation complication (occurring years after radiotherapy) has two major forms, mineralizing arteriopathy, and a necrotizing leukoencephalopathy (Fig. 1). The former is reflected as dystrophic calcification in small blood vessels and is commonly seen in the basal ganglia, dentate nuclei, thalami, and subcortical white matter. Necrotizing leukoencephalopathy is an admixture of demyelination, astrogliosis, and necrosis presenting with either a static or progressive encephalopathy. Neither condition is treatable; however, they may be related to both chemotherapy (in particular both systemic and intra-CSF methotrexate) and radiation dose, and therefore is potentially modifiable because acute leukemia induction regimens are tailored to the risk of CNS relapse. Not uncommon in leukemic patients and particularly in relation to high-dose systemic treatment is the occurrence of the posterior reversible leukoencephalopathy syndrome (17–20). The cardinal features of the posterior reversible leukoencephalopathy syndrome (PRES) are both clinical (i.e., hypertension, headaches, seizures, altered mental status, and visual disturbances) and radiological. The radiological findings are characterized by cortical gyriform abnormalities best visualized by FLAIR and T2-weighted MRI and topography of involvement. PRES is usually limited to the white matter of the parietal and occipital lobes; however, anterior and infratentorial involvement have also been described. Most often PRES is associated with malignant hypertension or eclampsia. But PRES is also a well-described complication of high-dose chemotherapy and tacrolimus and related immunophilins (i.e., cyclosporine) administered in conjunction with stem cell transplantation (18,19).

Fig. 1. MRI of a 21-year-old patient with T-ALL with CNS involvement who received systemic chemotherapy (vincristine, doxorubicin, 6-MP) in addition to cranial irradiation and intrathecal methotrexate and cytarabine. One year later the patient developed headaches and was found to have a patchy enhancing lesion on MRI in the left frontal lobe (A) with associated surrounding edema extending into the corpus callosum (B) concerning for tumor. Biopsy of the left frontal lesion showed white matter necrosis, fibrinoid vascular changes, and early mineralization consistent with toxic leukoencephalopathy. (Courtesy of Dr. Santosh Kesari, Dana-Farber/Brigham and Women’s Cancer Center, Boston, Massachusetts.)

Chapter 29 / Neurologic Complications of Leukemia

559

The precipitating event in the development of PRES is acute endothelial cell damage resulting in a microangiopathy, cerebrovascular dysregulation, and vasogenic edema. Another rare cause of encephalopathy in leukemic patients is multifocal necrotizing leukoencephalopathy (MNL) (21). MNL is a rare neurological treatment-related complication infrequently seen in patients treated for acute leukemia or other acquired immunodeficiency syndromes such as AIDS and chronic steroid-requiring diseases. Most commonly the lesions of MNL are found in the pons but may occasionally be extrapontine. Due to the predominantly pontine location MNL, most series are postmortem analyses. Histology reveals scattered infiltration by lipid-laden macrophages, accompanied by focal vacuolation, occasional axonal spheroids, and dystrophic calcification. Reactive astrocytosis is noted in the periphery. The lesions display myelin loss on Luxol fast blue stain. MRI reveals contrast enhancement on T1-weighted images and often scattered calcification best seen by cranial CT.

5. MENINGITIS Meningitis in leukemia may result from leptomeningeal infiltration of tumor (LM), subarachnoid hemorrhage, chemical causes (treatment-related following intra-CSF instillation of chemotherapy), or infectious causes (bacterial or fungal) (4–6,10,11,13–15,22–28). The presence or absence of LM always needs to be ascertained; if LM is diagnosed, prognosis is profoundly affected. Subarachnoid hemorrhage often occurs in the context of ICH, either in isolation or more frequently as more diffuse hemorrhage secondary to DIC. Spinal subarachnoid hemorrhage may occur in the context of DIC and acute promyelocytic (M3) leukemia and presents primarily with back pain that migrates rostrocaudally. Chemical meningitis (typically due to intraCSF cytarabine or methotrexate and most often when given intraventricularly) is temporally related to intra-CSF chemotherapy. Chemical meningitis begins 1–2 days after intra-CSF chemotherapy administration, is transient (typically lasting less than 5 days), and demonstrates no evidence of infection with CSF culture. Like other meningitic syndromes, patients complain of headache, fever, nausea, vomiting, photophobia, and meningismus. Notwithstanding an inflammatory CSF, chemical meningitis rapidly abates and is mitigated by oral steroids. Infectious meningitis occurs in leukemia due to immunosuppression both as a result of the underlying disease and its treatment. Listeria, Candida, and Aspergillus are common infectious etiologies with differing clinical presentations. Listeria presents as a meningitic syndrome, whereas Candida presents with a diffuse encephalopathy and multiple small brain abscesses and Aspergillus presents with progressive hemorrhagic stroke confined to a single vascular territory. Acute leukemia, in particular ALL, has the highest propensity to invade the meninges and result in leukemic meningitis (LM) (12,14,22). This is also true for Burkitt’s lymphoma and lymphoblastic lymphoma (2–3% all adult NHL), two subtypes of what is now considered ALL (6). Although AML infrequently results in LM, an unusual subtype, acute myelomonocytic leukemia (AMML), is at high risk (estimated at 20%) for the development of LM (5,24). Prior to CNS prophylaxis, 70% of autopsied patients with ALL had postmortem evidence of LM. However, with contemporary induction protocols incorporating CNS prophylaxis, only 5–10% of adult patients with acute leukemia develop CNS disease (14,15). Nonetheless, patients who develop CNS recurrence with leukemia have a poor prognosis. Chronic leukemia (CLL and CML), the most common adult leukemia encountered, rarely causes LM (10,11,23). Leukemic meningitis may be seen at diagnosis (3–5% all adult patients with ALL) or at relapse (5–7% of adult patients with ALL and prior CNS prophylaxis) (13,22). Three groups of patients with LM at relapse are recognized; CNS only (53%), bone marrow relapse followed by CNS (24%) and simultaneous CNS and bone marrow relapse (23%). In Surapaneni’s series of 527 consecutive adult patients with ALL, among patients with isolated LM, 88% subsequently relapsed in the bone marrow (22). As a consequence, the presence of LM— regardless of time of occurrence after induction therapy—is a predictor of systemic disease recurrence and poor outcome. Therefore the treatment of adult acute leukemia increasingly utilizes CNS risk stratification and tailors CNS prophylaxis accordingly to prevent CNS relapse (29). Risk for relapse of LM is associated with several prognostic factors in adults including young age, leukocytosis, presence of extramedullary disease, a high leukemia cell proliferation rate (S + GM fraction >14%), an elevated serum LDH level (>600 U/L), mature B-cell immunophenotype (L3), Philadelphia chromosome positivity [t (9; 22)], CD56 expression by leukemia cells, and an elevated serum beta 2-microglobulin level (>4 mg/dl)

560

Part VII / Neurologic Complications of Specific Malignancies

(29–34). Kantarjian utilized three risk factors (elevated serum LDH, elevated serum beta 2-microglobulin, and a high leukemia cell proliferation rate) in adult ALL and determined the risk of CNS relapse. Four groups were identified; in patients with one risk factor the risk of LM at one year exceeded 13% and increased to >20% if two or more risk factors were present. This approach has resulted in CNS disease risk stratification and intensification of CNS prophylaxis in adult ALL in an attempt to mitigate the emergence of LM (29). Similar to data with lymphomatous meningitis, after LM has occurred prognosis is poor with median survival of 6 months (22). In ALL patients with evidence of CSF leukemic blasts at diagnosis (3–5% of all patients with ALL), survival varies according to CSF category. CSF categories are as follows: • • • • •

normal CSF without blast cells (CNS1), normal CSF, that is, no evidence of pleocytosis (< 5 WBC/μL of CSF) and blasts (CNS2), CSF pleocytosis and blasts (CNS3), traumatic LP with blasts (TLP+), and traumatic lumbar puncture without blasts (TLP–) (35).

Patients with CNS1, CNS2, and TLP– have similar overall survival, whereas patients with CNS3 have a markedly worse prognosis and overall survival. Patients with TLP+ have intermediate survival relative to CNS1/2 and CNS3. These data suggest two further groups of patients (CNS3 and TLP+) who may benefit from more aggressive CNS prophylaxis than that administered to CNS1 (and CNS2 and TLP–) patients. The clinical presentation of LM in patients with leukemia is similar to that seen in patients with meningeal seeding from solid tumors (24,27,28,36). However, patients with hematologic malignancies present with a higher frequency of cranial nerve signs (e.g., trigeminal, mental or oculomotor neuropathies) as initial manifestations of neoplastic meningitis (24). LM is varied in its clinical presentation as it affects all levels of the CNS (24,25,27,28,36). In general, three domains of neurologic disturbance are characterized as affected by LM including: (1) the cerebral hemispheres, (2) the cranial nerves, and (3) the spinal cord and roots. The common symptoms of cerebral hemispheric dysfunction are headache and mental status change. Signs found in patients with LM and cerebral hemisphere disturbance encompass mental status changes including confusion and dementia, seizures, and hemiparesis. The single most useful laboratory test in diagnosing LM is an examination of the CSF usually obtained by lumbar puncture (24,25,27,28,36,37). In nearly all patients with LM, the CSF is abnormal regardless of the results of CSF cytology. CSF cytology positive for malignant cells is the standard method in most clinical series by which LM is diagnosed. Numerous biochemical markers have been evaluated, but in general poor sensitivity and specificity limit their use (38–44). In leukemia, monoclonal antibodies against cell surface markers can be used to distinguish between reactive and neoplastic lymphocytes in the CSF (45,46). Furthermore the demonstration by immunohistochemistry of monoclonality of CSF cells is as compelling as positive cytology. Lastly, the finding of CSF lymphocytes all of B-cell lineage is highly suggestive of LM as reactive lymphocytes in CSF are of T-cell lineage. Cytogenetic studies have also been evaluated in an attempt to improve the diagnostic accuracy of leptomeningeal metastases. Flow cytometry and DNA single cell cytometry, techniques that measure the chromosomal contents of cells, and fluorescent in situ hybridization (FISH), which detects numerical and structural genetic aberrations as a sign of malignancy, can give additional diagnostic information, but still have a low sensitivity (47–49). Polymerase chain reaction (PCR) can establish a correct diagnosis when cytology is inconclusive, but the genetic alteration of the neoplasia must be known for it to be amplified with this technique, and this may be helpful with hematological malignancies (46–49). A variety of neuroradiographic methods are available to evaluate patients with suspected LM including cranial computed tomography, brain and spine magnetic resonance imaging (MRI), computerized tomographic (CT) myelography, and radionuclide CSF flow studies (50–55). Despite the superiority of cranial contrast enhanced (CE) MRI or CT in the evaluation of LM, both studies have a high incidence of false negatives (30% by MRI and 58% by CT). Normal studies by either methodology do not exclude a diagnosis of LM in patients with negative CSF cytologies; however, positive MRI or CT may be suggestive or diagnostic of LM (24,25,27,28). MRI and CT are most useful in demonstrating bulky disease, a pattern of disease most responsive to radiotherapy and least responsive to intra-CSF chemotherapy (see below) though less commonly seen with LM.

∗

C × T = concentration × time

Thiotepa

50 mg every 2 weeks (total 8 weeks) 10 mg 2 or 3 times weekly (total 4 weeks)

25–100 mg 2 or 3 times weekly (total 4 weeks)

Cytarabine

DepoCyt

10–15 mg twice weekly (total 4 weeks)

Bolus Regimen

10 mg/day for 3 days weekly (total 4 weeks)

50 mg every 4 weeks (total 24 weeks) 10 mg once weekly (total 4 weeks)

25–100 mg once weekly (total 4 weeks)

25 mg/day for 3 days weekly (total 4 weeks)

—

10–15 mg once weekly (total 4 weeks)

Bolus Regimen

10 mg/day for 3 days every other week (total 4 weeks)

10 mg/day for 3 days once a month

10 mg once a month

25 mg/day for 3 days once a month

25–100 mg once a month

—

2 mg/day for 5 days once a month

10–15 mg once a month

2 mg/day for 5 days every other week (total 4 weeks) 25 mg/day for 3 days every other week (total 4 weeks) — —

C × T Regimen

Bolus Regimen

Maintenance regimen

C × T Regimen

Consolidation Regimen

2 mg/day for 5 days every other week (total 8 weeks)

C × T Regimen∗

Induction Regimens

Methotrexate

Drugs

Table 2 Intra-CSF Chemotherapy for Neoplastic Meningitis

562

Part VII / Neurologic Complications of Specific Malignancies

Radionuclide CSF flow studies (FS), or so-called “radionuclide ventriculography,” provide a safe physiological assessment of the functional anatomy of the CSF spaces (28,55). In prior reports FS demonstrated superiority in detecting interruption of CSF flow in patients with LM when compared to CT-myelography and contrast-enhanced spine MRI. However, FS are informative only with respect to compartmentalization of CSF and provide no information regarding bulky leptomeningeal disease, an aspect of LM best addressed by CT myelography or spine MRI. In addition, CT myelography or spine MRI are clearly superior to FS in detecting epidural spinal cord compression or intraparenchymal spinal cord metastases, two infrequent CNS complications of metastatic systemic cancer requiring emergency radiotherapy. Therefore, patients suspected of LM should undergo: (1) one or two lumbar punctures for CSF cytology and then, if negative, proceed to either a ventricular or lateral cervical CSF analysis; (2) contrast-enhanced cranial imaging (MR preferred to CT); (3) contrast-enhanced spine MR in patients with spinal symptoms; and (4) CSF flow study either by lumbar or ventricular radioisotope administration. Increasingly, the hematologic oncologist has adopted a risk-oriented approach to the prophylaxis of LM, with therapeutic regimens tailored to the risk of the individual patient. This approach is a reflection of the impoverished patient survival following the development of LM (2–6 month median survival) (29). A compilation of studies from adult patients with ALL and CNS prophylaxis treatment regimens suggests the following: regimens without cranial irradiation are effective; high-dose systemic therapy for low-risk disease is sufficient without intrathecal therapy; intrathecal methotrexate or alternating with cytarabine is effective without need for triple intrathecal therapy; intrathecal therapy and high-dose systemic chemotherapy are effective for high-risk disease; and a risk-oriented approach is optimal. How best to define the optimal CNS prophylaxis for the high-risk leukemic patient remains problematic and under investigation. The goal of treatment of LM is palliative and meant to improve or delay progression of neurologic symptoms and signs. The treatment of LM includes craniospinal irradiation, traditional systemic chemotherapy, intrathecal chemotherapy, and high-dose chemotherapy with hematopoietic stem cell rescue (Table 2). Because most CNS disease in ALL and NHL occurs in the setting of advanced or relapsed systemic disease, control of local or systemic disease is critical.

6. EPIDURAL SPINAL CORD COMPRESSION Leukemic epidural spinal cord compression (ESCC) is relatively rare (1% occurrence) with two exceptions: Burkitt’s lymphoma and lymphoblastic lymphoma (incidence 10–18%), both of which are presently considered as part of the ALL spectrum and similarly treated (4,6,15). Unlike solid cancers that initially metastasize to the vertebral body, leukemic ESCC originates in the paravertebral space and extends through the intervertebral foramina with resultant cord compression. As a consequence, bone involvement on neuroimaging is absent. Additionally and unlike solid cancer, there are no issues of spinal instability due to the lack of vertebral body involvement. Otherwise the presentation of leukemic ESCC is similar to that commonly seen with lymphoma and solid cancer beginning with pain (local, referred, or radicular) and evolving to myelopathy. Surgery is rarely contemplated (unless the primary is unknown) as leukemic ESCC is exquisitely radiosensitive and additionally responds to systemic chemotherapy. The latter approach, however, is reserved for patients with pain-only ESCC syndromes. Most importantly, the presence of ESCC in patients with leukemia does not negatively affect survival because treatment most often results in complete tumor eradication. Two other considerations in leukemic patients with ESCC include an epidural hematoma seem most often in the thrombocytopenic patient following a lumbar puncture or in the patient on chronic steroids wherein steroid-induced epidural lipomatosis may occur.

7. RADICULOPATHY Herpes zoster is a common cause of dermatomal vesicular rash in leukemic patients and is most common in CLL where 7% of patients have at least one Herpes zoster infection during the course of their disease (4,5,10,11). Most problematic of acute Herpes zoster is an acute pain syndrome that may evolve into post-herpetic neuralgia, a chronic pain syndrome. Dissemination may occur in up to 20% during which neurological involvement is seen in 50%. Neurological manifestations occurring in the context of disseminated Herpes zoster may include encephalitis, meningitis, and motor neuropathies.

Chapter 29 / Neurologic Complications of Leukemia

563

8. PERIPHERAL NEUROPATHY Neuropathies occur in two contexts in leukemia, either by direct tumor infiltration or as a consequence of chemotherapy (4,12). Optic neuropathy, either unilateral or bilateral, is a common presentation of LM (incidence 20–30%) and warrants emergency radiotherapy to preserve vision. Another common cranial neuropathy involved in the context of LM is the numb chin syndrome wherein leukemic cells preferentially affect the mental nerve, a subdivision of the mandibular branch of the trigeminal nerve. Either of these neuropathies in a leukemic patient is an indicator of LM and warrants LM-directed therapy. The most common peripheral neuropathy that occurs in leukemic patients is a length-dependent axonal sensorimotor neuropathy caused by vinca alkaloids. Initial symptoms are paresthesias of the hands and feet followed by progressive motor dysfunction culminating in foot and wrist drop. Occasionally, symptoms may be transiently worsened by administration of granulocyte- or granulocyte-macrophage colony stimulating factor. Though the neuropathy may resolve after discontinuing the drug, early dose modification based on clinical symptoms and signs mitigate the development of a disabling chronic neuropathy. Cranial neuropathies, though uncommon, may occur and affect oculomotor, trigeminal, facial, or recurrent laryngeal nerves manifested as ptosis, diplopia, jaw pain, facial paresis, and vocal cord paralysis. Lastly, transient autonomic neuropathy is common (20–30% of patients) with vinca alkaloids typically seen as abdominal pain with constipation. Rarely seen in patients treated with high-dose cytarabine is an acute demyelinating neuropathy, which resembles Guillain–Barré in its clinical symptoms and signs, frequently requiring transient respiratory support.

9. MYOPATHY Myopathy is seen in the majority of leukemic patients after several weeks (>3 weeks) of therapy, though in susceptible individuals (i.e., the elderly, deconditioned, or malnourished) it may appear within days of steroid therapy (4,12). The myopathy is proximal, characterized histologically by bland atrophy of type 2 (fast twitch) fibers and preferentially affects the lower extremities; however, over time shoulder weakness is also seen. Therapy entails steroid reduction and if possible discontinuance. Recovery after steroid taper not infrequently requires months before power returns to normal.

10. CONCLUSIONS Leukemia is associated with a myriad of neurological complications (Table 1) that occur both as a direct consequence of leukemia (leukemic meningitis, chloromas) and indirectly due to treatment or immunosuppression. Most relevant with respect to differential diagnosis, however, is leukemic meningitis. LM, with its ability to affect the entire CNS, may present in a pleomorphic manner and mimic a variety of neurologic syndromes. Therefore, in essentially all leukemic patients with CNS dysfunction a CSF examination is necessary. By contrast, peripheral nervous system disorders are nearly always treatment related (steroid myopathy, vinca alkaloid neuropathy) and respond best to discontinuance of the neurotoxic agent.

REFERENCES 1. Xie Y, Davies SM, Xiang Y et al. Trends in leukemia incidence and survival in the United States (1973–1998); Cancer 2004 (97) 9:2229–2235. 2. Hoelzer D, Gokbuget N, Ottmann O et al. Acute lymphoblastic leukemia. Hematology 2002 (1):162–171. 3. Pui C-H, Evans WE. Acute lymphoblastic leukemia. Drug Therapy (339) 9:605–615. 4. Recht L, Mrugala M. Neurologic complications of hematologic neoplasms. Neurol Clin N Am. 2003 (21):87–105. 5. Meyer RJ, Ferreira, PP, Cuttner J et al. Central nervous system involvement at presentation in acute granulocytic leukemia: a prospective cytocentrifuge study. Am J Med 1980; 68:691–694. 6. Teshima T, Akashi K, Shibuka T et al. Central nervous system involvement in adult T-cell leukemia/lymphoma. Cancer 1990; 65:327–332. 7. Kim H, Lee J-H, Choi S-J et al. Analysis of fatal intracranial hemorrhage in 792 acute leukemia patients. Haematologica 2004; 89:622–624. 8. Kawanami T, Kurita K, Yamakawa M et al. Cerebrovascular disease in acute leukemia: a clinicopathological study of 14 patients. Intern Med. 2002; 41(12): 1130–1134.

564

Part VII / Neurologic Complications of Specific Malignancies

9. Sostak P, Padovan CS, Yousry TA et al. Prospective evaluation of neurological complications after allogeneic bone marrow transplantation. Neurology 2003; 60:842–848. 10. Bower JH, Hammack JE, McDonnell SK et al. The neurologic complications of B-cell chronic lymphocytic leukemia. Neurology 1997; 48:407–412. 11. Cramer SC, Glaspy JA, Efird JT et al. Chronic lymphocytic leukemia and the central nervous system: a clinical and pathological study. Neurology 1996; 46:19–25. 12. Plotkin SR, Wen PY. Neurologic complications of cancer therapy. Neurol Clin N Am 2003; 21:279–318. 13. Cortes J. Central nervous system involvement in adult acute lymphocytic leukemia. Hematol Oncol Clin North Am. 2001; 15:145–162. 14. Wolk RW, Masse SR, Conklin R, Freireich EJ. The incidence of central nervous system leukemia in adults with acute leukemia. Cancer 1974; 33:863–871. 15. Stewart DJ, Keating MJ, McCredie KB et al. Natural history of central nervous system acute leukemia in adults. Cancer 1981; 47:184–196. 16. Chen C-Y, Zimmerman RA, Faro S et al. Childhood leukemia: central nervous system abnormalities during and after treatment. AJNR 1996; 17:295–310. 17. Hinchey J, Chaves C, Appignani B et al. A reversible posterior leukoencephalopathy syndrome. N Engl J Med 1996; 334:494. 18. Shin RK, Stern JW, Janss AJ et al. Reversible posterior leukoencephalopathy during the treatment of acute lymphoblastic leukemia. Neurology 2001; 56:388–391. 19. Lavigne CM, Shrier DA, Ketkar M t al. Tacrolimus leukoencephalopathy: a neuropathological confirmation. Neurology 2004; 2: 1132–1133. 20. Lamy C, Oppenheim C, Meder JF et al. Neuroimaging in posterior reversible encephalopathy syndrome. J Neuroimaging 2004; 14:89–96. 21. Anders KH, Becker SP, Holden JK et al. Multifocal necrotizing leukoencephalopathy with pontine predilection in immunosuppressed patients: a clinicopathologic review of 16 cases. Hum Pathol 1993; 24(8):897–904. 22. Surapaneni UR, Cortes JE, Thomas D et al. Central nervous system relapse in adults with acute lymphoblastic leukemia. Cancer 2002; 94 (3):773–780. 23. Cash J, Fehir KM, Pollack MS. Meningeal involvement in early stage chronic lymphocytic leukemia. Cancer 1987; 59:798–800. 24. Dekker AW, Elderson A, Punt K et al. Meningeal involvement in patients with acute nonlymphocytic leukemia. Cancer 1985; 56:2078–2082. 25. Kaplan JG, DeSouza TG, Farkash A et al. Leptomeningeal metastases: comparison of clinical features and laboratory data of solid tumors, lymphomas, and leukemias. J Neuro Oncol 1990; 9:225–229. 26. Chamberlain M. Leptomeningeal metastases. In: Cancer in the Nervous System. Levin V (ed.). New York: Churchill Livingstone, 1996:282–290. 27. Grossman SA, Krabak MJ. Leptomeningeal carcinomatosis. Cancer Treat Rev 1999; 25(2):103–119. 28. Chamberlain MC. Leptomeningeal metastases: a review of evaluation and treatment. J Neuro Oncol 1998; 37:271–284. 29. Kantarjian HM, Walters RS, Smith TL et al. Identification of risk groups for development of central nervous system leukemia in adults with acute lymphocytic leukemia. Blood 1988; 72:1784–1789. 30. Kantarjian HM, Smith T, Estey E et al. Prognostic significance of elevated serum beta 2-microglobulin levels in acute lymphocytic leukemia. Am J Med. 1992; 93:599–604. 31. Ravandi F, Cortes J, Estrov Z et al. CD56 expression predicts occurrence of CNS disease in acute lymphoblastic leukemia. Leuk Res. 2002; 26:643–649. 32. Kantarijian HM, O’Brien S, Smith TL et al. Results of treatment with hyper-CVAD, a dose-intensive regimen, in adult a cute lymphocytic leukemia. J Clin Oncol. 2000; 18:547–561. 33. Mavlight GM, Stuckey SE, Cabanillas FF et al. Diagnosis of leukemia or lymphoma in the central nervous system by beta 2microglobulin determination. N Engl J Med. 1980; 303:718–722. 34. Kersten MJ, Evers LM, Dellemijn PL et al. Elevation of cerebrospinal fluid soluble CD27 levels in patients with meningeal localization of lymphoid malignancies. Blood 1996; 87:1985–1999. 35. Burger B, Zimmermann M, Mann G et al. Diagnostic cerebrospinal fluid examination in children with acute lymphoblastic leukemia: significance of low leukocyte counts with blasts or traumatic lumbar puncture. J Clin Oncol 2003; 21 (2):184–188. 36. Wasserstrom W, Glass J, Posner J. Diagnosis and treatment of leptomeningeal metastases from solid tumors experience with 90 patients. Cancer 1982, 49:759–772. 37. Van Oostenbrugge RJ, Twijnstra A. Presenting features and value of diagnostic procedures in leptomeningeal metastases. Neurology 1999; 53:382–385. 38. Wasserstrom WR, Schwartz MK, Fleisher M et al. Cerebrospinal fluid biochemical markers in central nervous system tumors: a review. Ann Clin Lab Sci 1981; 11(3):239–251. 39. Van Zanten AP, Twijnstra A, Hart AA et al. Cerebrospinal fluid lactate dehydrogenase activities in patients with central nervous system metastases. Clin Chim Acta 1986; 161(3):259–268. 40. Klee GG, Tallman RD, Goellner JR et al. Elevation of carcinoembryonic antigen in cerebrospinal fluid among patients with meningeal carcinomatosis. Mayo Clin Proc 1986; 61(1):9–13. 41. Twijnstra A, van Zanten AP, Hart AA et al. Serial lumbar and ventricle cerebrospinal fluid lactate dehydrogenase activities in patients with leptomeningeal metastases from solid and haematological tumours. J Neurol Neurosurg Psychiatry 1987; 50(3):313–320. 42. Twijnstra A, Ongerboer dV, van Zanten AP et al. Serial lumbar and ventricular cerebrospinal fluid biochemical marker measurements in patients with leptomeningeal metastases from solid and hematological tumors. J Neuro Oncol 1989; 7(1):57–63. 43. Chamberlain MC. Cytologically negative carcinomatous meningitis: usefulness of CSF biochemical markers. Neurology 1998; 50(4):1173–1175.

Chapter 29 / Neurologic Complications of Leukemia

565

44. Garson JA, Coakham HB, Kemshead JT et al. The role of monoclonal antibodies in brain tumour diagnosis and cerebrospinal fluid (CSF) cytology. J Neuro Oncol 1985; 3(2):165–171. 45. Boogerd W, Vroom TM, van Heerde P et al. CSF cytology versus immunocytochemistry in meningeal carcinomatosis. J Neurol Neurosurg Psychiatry 1988; 51(1):142–145. 46. Van Oostenbrugge RJ, Hopman AH, Ramaekers FC et al. In situ hybridization: a possible diagnostic aid in leptomeningeal metastasis. J Neuro Oncol 1998; 38 (2–3):127–133. 47. Cibas ES, Malkin MG, Posner JB et al. Detection of DNA abnormalities by flow cytometry in cells from cerebrospinal fluid. Am J Clin Pathol 1987; 88(5):570–577. 48. Biesterfeld S, Bernhard B, Bamborschke S et al. DNA single cell cytometry in lymphocytic pleocytosis of the cerebrospinal fluid. Acta Neuropathol (Berl) 1993; 86(5):428–432. 49. Van Oostenbrugge RJ, Hopman AH, Arends JW et al. The value of interphase cytogenetics in cytology for the diagnosis of leptomeningeal metastases. Neurology 1998; 51(3):906–908. 50. Chamberlain M, Sandy A, Press G. Leptomeningeal metastasis a comparison of gadolinium-enhanced MR and contrast-enhanced CT of the brain. Neurology 1990, 40:435–438. 51. Gomori JM, Heching N, Siegal T. Leptomeningeal metastases: evaluation by gadolinium-enhanced spinal magnetic resonance imaging. J Neuro Oncol 1998; 36:55–60. 52. Sze G, Soletsky S, Bronen R et al. MR Imaging of the cranial meninges with emphasis on contrast enhancements and meningeal carcinomatosis. AJNR 1989; 10:965–975. 53. Chamberlain MC. Comparative spine imaging in leptomeningeal metastases. Neuro-oncol 1995; 23:233–238. 54. Kramer E, Rafto S, Packer R et al. Comparison of myelography with CT follow-up versus gadolinium MRI for subarachnoid metastatic disease in children. Neurology 1991; 41:46–50. 55. Chamberlain M, Corey-Bloom J. Leptomeningeal metastases indium-DTPA CSF flow studies. Neurology 1991; 41:1765–1769.

30

Neurological Complications of Lymphomas Brian Patrick O’Neill,

MD

CONTENTS Introduction Central Nervous System (Intradural) Peripheral Nervous System (Dural and Extradural) Paraneoplastic Disorders in the Lymphomas Newer Diagnostic Techniques Conclusion References

Summary The neurologic complications of the malignant lymphomas are common, serious, and treatable. Some complications, such as dural involvement by MALT tumor and paraneoplastic ataxia of Hodgkin’s disease, are unique to these malignancies. Compared to more common cancers the malignant lymphomas appear to affect the central and peripheral nervous systems disproportionately. As such they have the potential to significantly impact patients’ quality of life because of effects on special senses, mobility, communication, and cognition. Because of the inextricable intersection with disordered immune system function, an additional challenge for the treating physician is determining whether the neurologic syndrome is due to cancer or to autoimmunity. The interplay between the two contribute to the morbidity and pose challenges to treatment. The potential for improvement validates an intensive review of these complications including new insights into pathogenesis, diagnosis, and treatment. Key Words: lymphoma, Hodgkin’s disease, neoplastic meningitis

1. INTRODUCTION Lymphomas are malignant neoplasms of the hematopoietic system, specifically lymphocytes. They are a diverse group of disorders clinically, histologically, immunophenotypically, and genotypically. They vary in their natural history, treatments, response to therapy, and survival rates. Lymphomas typically arise in lymph nodes or in lymphoid tissue but also may involve extranodal tissues such as the liver and lungs. Of these latter cases, immunologically privileged sites such as the testes, the eyes, and the central nervous system (CNS) appear to be preferentially involved. The most common and most prevalent subtypes in the United States are Hodgkin’s disease (HD) and the non-Hodgkin’s lymphomas (NHL). These will be the focus of this chapter. Most NHLs originate from a monoclonal population of B-lymphocytes. NHLs derived from T cells are less common. Although its tumorigenesis is much less understood, HD is characterized by a unique cell—the Reed–Sternberg cell. Both HD and NHL appear to require some form of altered host immumocompetence for their development. A direct proportion between immunocompetence and risk for development of the malignant lymphomas has been consistently demonstrated. Multiple classification schemes have been proposed for both HD and NHL. The most recent, and most widely used, classification for NHL is the World Health Organization From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

567

568

Part VII / Neurologic Complications of Specific Malignancies

(WHO) Classification, which was developed from the Revised European American Lymphoid Neoplasms (REAL) Classification (1). Similarly, the latest international classification system for HD is the Cotswolds’ modification of the Ann Arbor Classification (2). Treatment varies with the histologic classification, type of presentation, extent of disease, new versus relapsed disease, and, if relapsed, prior treatment. In 1993, the International Non-Hodgkin’s Lymphoma Prognostic Factors Project established a new prognostic classification, the International Prognostic Index, or IPI (3). The IPI was developed for intermediate- and highgrade NHL, has been well-validated, and is widely used. IPI was based on five independent prognostic factors including age, Ann Arbor tumor stage, serum lactate dehydrogenase, performance status, and number of extranodal sites. Patients with a high risk based on this index are less likely to obtain a complete response (CR) to chemotherapy and have a higher relapse rate following a CR. Nevertheless there continues to be marked residual heterogeneity in the outcome of patients with identical prognostic scores. The latter has been attributed to the marked genetic and molecular heterogeneity that underlies disease aggressiveness and tumor progression, and led to evaluation of molecular and genetic markers associated with patients’ survival (4). A similar metric has not yet been developed for HD. Apropos this discussion there is wide variation in the incidence, frequency, and type of neurologic involvement in the lymphomas. In 1980, Cairncross and Posner described the differing neurologic manifestations of hematologic malignancies, in particular the differing manifestations of HD and NHL (5). For example, in the former case, dural involvement is common and leptomeningeal and parenchymal (intradural) disease are rare; the opposite is true for NHL. Reviews since that seminal contribution have maintained the classification and will be maintained herein (6,7). Table 1 summarizes the differences between the neurologic complications of HD and NHL. Five caveats relate to this variability in neurologic involvement. Firstly, the lymphomas occur by clonal expansion of normal lymphocytes. Thus, the malignant lymphocytes may retain some or most properties of normal lymphocytes, at least temporarily (undoubtedly they lose these characteristics as they become progressively more dedifferentiated). This characteristic is best exemplified by the “ghost tumor” response, that is, lymphocyte apoptosis in response to corticosteroids (8,9). Secondly, these patients may have simultaneous or sequential disease involving multiple sites in the central and peripheral nervous systems, and the deficits associated with these disorders may be amplified by the effects of previous therapy and, since these patients are often older, co-morbidity (10). Special testing and procedures may be required to detect them. Thirdly, by inference, lymphoma occurs in hosts who have alterations in their immunocompetence (11). Thus, patients may not only have neurologic disease directly related to lymphomatous involvement; they may also have neurologic disorders that are dysimmune in origin. For example, we have had examples of patients developing corticosteroid-responsive idiopathic thrombocytopenic purpura and Hashimoto’s thyroiditis many years after complete remission of the presenting cancer.

Table 1 Neurologic Complications of HD and NHL Hodgkin’s Disease Intradural • Parenchymal CNS • Neoplastic meningitis • Other Dural and extradural • Dural • Epidural spinal cord compression • Other Paraneoplastic

Exceedingly rare Very rare Eosinophilic meningitis Lymphomatoid granulomatosis Common Common Cerebellar degeneration

Non-Hodgkin’s Lymphoma Rare Common; may be presenting syndrome Intravascular lymphoma Pituitary lymphoma Common Common Neurolymphomatosis Subacute motor neuronopathy

Chapter 30 / Neurological Complications of Lymphomas

569

Fourthly, biologic differences themselves likely contribute to the inter- and intratumoral phenotypic variation discussed two decades ago by Cairncross and Posner. Fifthly, sustained control or cure of neurologic disease will only occur if systemic disease can be successfully treated. Although some of these complications may be the presenting symptoms of systemic lymphoma, they typically are late phenomena and their occurrence correlates with, among other things, high-grade, widespread, and especially extranodal disease, and bone marrow involvement (12,13). Lastly, we do not yet know whether what have been called “metastases” are in fact separate lymphomas that “home” to specific sites in CNS and PNS. Current data suggests that when NHL occurs or recurs in nervous system, they do so as different tumors. This chapter reviews the neurologic complications of Hodgkin’s disease and non-Hodgkin’s lymphomas. Several arbitrary decisions have been made. Firstly, I have chosen to describe those conditions due to actual lymphomatous involvement of the nervous system and those directly related to the disease, such as paraneoplastic disorders. Secondly, based on nearly 20 years of consultation for neurologic problems of patients with these diseases, I have chosen to highlight concepts by examples from my own experience. Thirdly, neurologic complications of treatment such as encephalopathy and neuropathy are purposely underdeveloped. Reviews of these complications are widely available from multiple sources, especially Posner’s excellent monograph, Neurologic Complications of Cancer (14), and elsewhere in this volume. Specific examples are noted when appropriate such as in the differential diagnosis of plexopathies. Lastly, controversy exists over whether primary CNS non-Hodgkin’s lymphoma (PCNSL) is a brain tumor that happens to be a lymphoma or a lymphoma that happens to be in the brain. Conventional thinking treats PCNSL as a unique entity and thus it will not be specifically discussed in this chapter. The vast majority of neurologic complications of lymphomas occur in non-Hodgkin’s lymphomas, and of these, specifically the diffuse large B-cell type (DLC NHL). In one series, 98% of the neurologic complications occurred in DLC NHL (15). Nevertheless these are relatively uncommon overall. Liang and colleagues estimated a 6% (51/833) risk of neurologic complications in 833 NHL cases diagnosed in the modern neuroimaging era (16). In this series, no patient with low-grade lymphoma developed neurologic involvement. However, 6.5% and 16.7% of patients with intermediate- and high-grade lymphomas, respectively, did have involvement. The majority of cases had either epidural spinal cord compression or leptomeningeal lymphoma. In the Liang series a significantly higher incidence of CNS disease was seen in patients with lymphoma involving orbit (43%), testis (40%), peripheral blood (33%), bone (29%), nasal/paranasal sinuses region (23%), and bone marrow (20%). Stage IV disease, an elevated serum LDH, the presence of circulatory lymphoma cells, and the presence of B symptoms were also associated with an increased risk of CNS disease (12,13). By inference these complications should also correlate with IPI, although a large-scale study has not yet been reported. Approximately 10% of all NHL are T-cell lymphoma (TCL). In contrast to B-cell lymphomas, Kaufman and colleagues reviewed the incidence of neurologic complications in TCL and found that the overall rate of neurological complications was 7.9% (17). The frequency of neurological complications in peripheral TCL and cutaneous TCL was 17% and 3%, respectively, with at least half of the neurological complications in both conditions due to direct involvement of the nervous system (leptomeningeal and parenchymal involvement). Unlike the situation in B-cell NHL there were no cases of epidural spinal cord disease. Since the publication of the first edition, additional literature comprised only three case reports and no additional series of neurological complications in T-cell NHL (18–20). In HD precise figures are lacking. One can only conclude that this is partly because of its rarity. In one series of more than 2000 patients with HD, none presented with direct neurologic involvement (21). Neurologic disease, when present, was usually seen with advanced HD (22). In preparation for the first edition of this book, I conducted a review of the Entrez PubMed resource over the entire time period of its existence, January 1966–December 2000. Only 82 citations were retrieved when “Hodgkin’s disease” and “neurologic disease” were used as key words. The vast majority of these citations described indirect complications, that is, paraneoplastic, neurotoxic, and infectious complications. They were from the 1970s and 1980s and arose mostly from the London group of Henson and Currie (23). In preparation for this second edition, I again reviewed Entrez PubMed and identified an additional 115 citations. The spectrum remains the same, but modern imaging has led to improved understanding of the nature and relationships of the neurologic complications in HD (24,25).

570

Part VII / Neurologic Complications of Specific Malignancies

2. CENTRAL NERVOUS SYSTEM (INTRADURAL) There are no reports of primary intradural parenchymal HD. In those reported as “primary,” careful analysis shows that every example had a dural component. Thus, it is reasonable to conclude that these infrequent reports [e.g., (26)] support the hypothesis of centripetal extension of HD. Also, no such cases of “primary HD” have been identified at our institution during the modern neuro-imaging era (since 1975). Unusual forms of parenchymal disease do occur in HD but are not thought to be due to the actual presence of malignant lymphocytes. These include eosinophilic meningitis (27) and granulomatous angiitis (28,29). These are not documented in NHL and this may be explainable by inherent biologic differences as to how these two lymphomas affect the nervous system. Intradural NHL, though not common, does occur. The mechanism of direct spread of NHL intradurally remains a puzzle. In the early part of the twentieth century, intradural NHL was variously described as reticulum cell sarcoma, microglioma, and perithelial sarcoma because they were thought to arise from nonlymphocytic microglial cells, or dedifferentiation of multipotential cells in the perivascular spaces of the CNS (30). It has only been in the past two decades that intradural NHL is viewed as a unique extranodal form. Specifically, by histologic, ultrastructural, and immunophenotypic criteria intradural malignant lymphocytes are identical to malignant lymphocytes elsewhere (31–33). Newer theories focus on how malignant lymphocytes or cells destined to acquire the malignant phenotype can gain entry in and out of the CNS via normal lymphocyte flux and lymphocytic “homing” via blood-borne “metastases” (34–36). A special consideration for prophylactic treatment is relevant here. The CNS has been assumed to be a “sanctuary” site for systemic chemotherapy. As with acute lymphocytic lymphoma, a high incidence of intradural “failure” has been described in NHL at certain sites such as sinus and testicular lymphomas, and Burkitt’s lymphoma (37–39). Furthermore, this failure occurs at a time in the patients’ illness atypical for NHL at large [i.e., early in their course or even at presentation, or at a time later in their illness when they are well (39)]. A large literature has accumulated about the issue of prophylatic treatment of the intradural nervous system, its type of treatment (or treatments), their timing, and indications (40,41). The assumption was that prophylaxis was necessary to prevent CNS relapse and even systemic reseeding. Because many of the reports upon which these recommendations were based preceded the modern neuroimaging era, it is not clear whether patients had been properly staged. Since even with prophylaxis a substantial number of patients will develop CNS disease and prophylactic therapy may be neurotoxic, we have taken a cautious view towards prophylaxis (39). A recent survey appears to bear this out. Of 1693 patients treated in protocols of the German High-Grade Non-Hodgkin’s Lymphoma Study Group (DSHNHL), the incidence of CNS relapse in the patients treated for aggressive lymphomas on DSHNHL protocols from 1990 to 2000 was low (2.2%), although CNS prophylaxis was administered to < 5% of patients (42).

2.1. Brain With the exception of PCNSL, brain parenchymal involvement is unusual in NHL. In an autopsy series that provides the most convincing data, Schaumberg and colleagues at Massachusetts General Hospital found that only 2 of 121 consecutive postmortem examinations of NHL patients harbored parenchymal lymphomatous deposits (43). More recent studies employing modern neuroimaging have confirmed the rarity. In one series a total of 2514 NHL patients who had no CNS involvement at diagnosis were entered prospectively and were treated according to standard protocols. Only 22 cases of parenchymal recurrence/relapse was documented (0.9%). Only those patients with either Burkitt’s or lymphoblastic lymphomas appeared to be at higher risk. We reviewed the Mayo Lymphoma Database for all patients who presented as intradural NHL and were presumed to have PCNSL. Staging demonstrated occult systemic disease in 3.9% of cases, typically present in intra-abdominal or intra-pelvic lymph nodes (44). Because our population is a highly selected one, the actual incidence is probably much lower. When parenchymal lymphoma has been diagnosed it has usually been as a late complication, occurring at a time when there is widespread systemic disease and bone marrow involvement (12,13). A special situation involves Richter’s transformation. Several researchers have described parenchymal lymphoma occurring some years after another hematologic malignancy, including systemic NHL and HD (45,46). In some cases no treatment for the original tumor had been administered, thus excluding an induced second malignancy. In other patients

Chapter 30 / Neurological Complications of Lymphomas

571

the histology of the original lymphoma and the parenchymal disease have been discordant (47), supporting the contention that these are in fact second malignancies rather than recurrence. In another patient, identical immunoglobulin gene rearrangements were identified, providing evidence for the evolution of two morphologically distinct neoplasms from the same clone (48). In fact, the imaging features of these instances of Richter’s transformation are similar to those seen in de novo PCNSL, that is, intense and homogeneous contrast enhancement, less mass effect proportional to the tumor bulk, and contact with a CSF surface (30). Decision making regarding treatment ultimately rests with the responsible clinician and is based on realistic goals and expectations. Treatment may be supportive care, palliative radiotherapy, or salvage attempts with “penetrating” chemotherapy (49), typically high-dose intravenous methotrexate (50) or high-dose intravenous cytarabine (51). In some ways therapy decisions are similar whether the intradural disease is PCNSL or parenchymal NHL as a secondary site of systemic disease. No trial or large-scale review is available to inform treatment decisions. However, considering its positive effect on survival and the potential improvement in neurological symptoms and quality of life, salvage therapy seems to be a valuable treatment. A review by Reni and colleagues described preliminary results from small pilot studies with topotecan, rituximab, temozolomide, the PCV regimen, and HD-chemotherapy supported by autologous or allogeneic peripheral blood stem cells transplantation (52).

2.2. Spinal Cord Cord involvement has been demonstrated far less frequently than brain parenchymal disease. Except for the special circumstance of cord involvement in the context of leptomeningeal lymphoma (see below), the mechanism is unknown. No doubt this is an under-recognized complication of systemic lymphoma. Because cord lymphoma rarely expands the cord, a myelogram would be unlikely to give diagnostic information (53). The more recent widespread use of MR scanning of the spine has contributed to the increased number of reports in the past decade compared to the two decades prior (54,55), as well as displaying unique complications of treatment such as radiation-induced spinal cord glioma subsequent to treatment of HD (56). In patients with an isolated CNS site as the site of late recurrence of systemic disease, we are now much more likely to stage the entire nervous system and in the process have identified occult parenchymal masses. For instance, a patient was observed with progressive difficulty walking at a time when he had no evident residual NHL; a biopsy confirmed this lesion as lymphoma. A biopsy should be considered in those patients who are candidates for aggressive treatment. Enhancing cord masses without significant mass effect occurring in the context of systemic NHL have been assumed to be infectious, inflammatory, or post-irradiation. Because these have often occurred as end-of-life illnesses, an aggressive work-up may not have been considered. Again, treatment decisions rest with the responsible clinician and are based on realistic goals and expectations. Treatment may be supportive care, palliative radiotherapy, or salvage. Only anecdotal reports support the use of one over another.

2.3. Cranial and Spinal Nerves Three specific forms are encountered clinically: (i) encasement by leptomeningeal lymphoma (which is by definition intradural); (ii) centripetal spread intradurally from contiguous extradural sites of lymphoma; and (iii) neurolymphomatosis, a unique intraneural tumor that may be intra- or extradural. Cavernous sinus lymphoma, a not uncommon phenomenon in systemic lymphoma, is discussed in Section 3 on the “Peripheral Nervous System.” 2.3.1. Lymphomatous Meningitis (LM) This entity is common; in some series it occurs in as many as 5% of patients overall with large cell lymphoma (57). Furthermore, 10–30% of all patients are diagnosed at the time of initial presentation of the lymphoma (22). In B-cell lymphomas this entity comprises nearly half of the neurologic complications (15). However, it may be as common in T-cell lymphomas. In the Kaufman series (17), 2.4% of T-cell lymphoma cases overall had leptomeningeal syndromes comprising 36.8% of the neurologic complications (7/19 patients). Nevertheless, as with parenchymal disease, LM is typically seen in patients with insufficient control of systemic disease and high-grade disease.

572

Part VII / Neurologic Complications of Specific Malignancies

Pathogenesis is uncertain. Most reports associate the development of lymphomatous meningitis with direct entry of malignant lymphocytes into the cerebrospinal fluid (CSF) rather than dural invasion. Our experience is that even with aggressive dural lymphoma intradural involvement is uncommon. LM is an especially dreaded complication of systemic lymphoma because its diagnosis may be so elusive and its disability progressive and additive. I have had patients who accrued cranial or spinal nerve involvement daily. Current treatment regimens are rarely curative and may produce additional neurologic toxicity, thus adding to the disease burden. Diagnosis requires a high index of suspicion especially when lymphoma has not yet been discovered. The differential diagnosis is broad and includes dysimmune states, opportunistic infections, and other malignancies. Malignant lymphoma cells tend to cluster (presumably by stasis) at the basilar meninges and the cauda equina. Thus, patients typically present with meningeal and/or radicular pain, and cranial and spinal neuropathies (58). A few unusual syndromes have been described including nonconvulsive status epilepticus presenting as a confusional state (59) and optic neuropathy (60). Sometimes, if cells obstruct CSF pathways a communicating hydrocephalus syndrome may be superimposed. Invasion along the Virchow–Robin spaces may give focal or multifocal CNS parenchymal signs including frank myelopathy due to spinal cord invasion (61). As many as 10% of cases are asymptomatic (40). A careful neurologic examination may reveal disease at several levels even if the patient presents with symptoms referable to only one cranial or spinal nerve [“more signs than symptoms” (14)]. Usually the diagnosis is secured by CSF analysis. CSF cytology may be abnormal in the first lumbar puncture (or ventricular fluid if an extraventicular drain is inserted because of hydrocephalus) but in approximately one-half the cases at least three CSF examinations are necessary (58). CSF analysis may be problematic. The gold standard is the demonstration of malignant lymphocytes in the CSF. However, CSF lymphocytes obtained from either the ventricular fluid or the lumbar fluid may be deceptively normal (62). Sometimes it is necessary to obtain fluid from both sites (63). Demonstration of surface immunoglobulin monoclonality had been assumed to be an acceptable surrogate. However, many researchers have had patients with inflammatory disorders of the CNS (e.g., multiple sclerosis) who have had small numbers of CSF lymphocytes demonstrating monoclonality. Thus, the issue now is whether there is a monoclonality threshold for the diagnosis of lymphomatous meningitis. At the time of this writing this has not been settled. Newer techniques may render this last point moot. Hug and colleagues described the use of a polymerase chain reaction (PCR) approach to characterize the clonally diverse gene encoding the immunoglobulin heavychain (IgH) determining regions (CDR3) of single B-cells. This single-cell PCR analysis may be valuable if cytology, immunocytochemistry, flow cytometry, and automated gene scanning of CSF samples are unable to detect malignant monoclonal proliferation (64). Lastly, some investigators consider “atypical” lymphocytes to be equivalent to malignant lymphocytes (65). Since morphology is dependent on many factors our approach has been to make treatment decisions on the presence of “positive” CSF cytology as defined by frankly malignant lymphocytes and/or a monoclonal population that is at least 20% of the total. In some cases CSF cytology remains negative despite multiple spinal taps. In these cases, a meningeal biopsy is required for diagnosis. Cheng and colleagues reviewed the Mayo experience of the use of such biopsies in 37 patients with chronic meningitis of unknown cause seen during the MRI era (66). A definitive diagnosis was made in 16 of 41 biopsies (39%), but in cases where enhancement was present and the enhancing meninges were biopsied, a diagnosis was obtained in 80% (12 of 15 cases). In some patients who presented with a cauda equina syndrome and had contrast enhancement only in the lumbosacral spine a “mini”-hemilaminectomy was performed to remove an involved rootlet and its coverings. Cancer was diagnosed in over one-half of these cases. Only 2 of 22 biopsies (9%) from nonenhancing regions were diagnostic. The authors emphasized that a well-coordinated approach between clinician, surgeon, radiologist, and pathologist will assure the best results with the lowest morbidity. Treatment, as with all decisions regarding these patients, is predicated on the overall status of the patient and consensus among the responsible clinician, the patient, and the patient’s family. Several treatment algorithms have been published (14,22). In the past, most of these focused on some variation of intrathecal administration (intralumbar administration and/or intraventricular administration) of short-acting water-soluble chemotherapy such as methotrexate or cytarabine, usually with irradiation to the symptomatic area (s). If intralumbar administration was chosen, many advocated the concomitant use of corticosteroids and some advocated alternation of

Chapter 30 / Neurological Complications of Lymphomas

573

agents such as methotrexate and cytarabine (67). Control of systemic disease was a prerequisite for successful treatment (22). Up until the wide acceptance of high-dose intravenous methotrexate, intra-Ommaya treatment was the preferred route of administration if the CSF is to be directly treated. It “goes with the flow” (i.e, administered drug follows normal CSF circulation); it permits frequent administrations of chemotherapy without the need for repeated lumbar punctures; and it allows CSF to be sampled as a means of monitoring therapy. Usually a radionuclide flow study is performed first to be certain that there is free communication between the intracranial and intraspinal compartments (63). If there is a hindrance to flow drug may accumulate in the ventricles increasing the risk of neurotoxicity (69). A slow-release formulation of cytarabine designed to maintain cytotoxic CSF concentrations for more than 14 days is now available for clinical use (70). This agent (DepoCyt) is purported to produce a high response rate and a better quality of life as measured by Karnofsky score relative to that produced by free drug given by intralumbar injection twice a week. However, time to neurologic progression and survival was not significantly different between the two arms. Newer therapies now entering clinical trials include the use of radiolabeled monoclonal antibodies directly instilled into the CSF (71). Some unusual but nevertheless serious complications of Ommaya reservoir placement and use have been described. Firstly, if ventriculomegaly is not present, freehand placement of the ventricular end may be difficult. Obviously, multiple passes through the brain cortex expose the patient to immediate and longer-term complications such as hemorrhage, epilepsy, and intracranial infection. In 1987, Hagen and colleagues described the use of stereotaxy to place ventricular catheters into the lateral ventricle, thus simplifying the insertion (72). Secondly, most surgeons advise waiting several days after placement before introducing drug. However, some patients may literally add to disease burden day-by-day; any time lost before treatment onset can be clinically significant. Thirdly, systemic lymphoma patients are highly vulnerable to infection and bleeding. Thus, as with shunts, the placement of an intraventricular catheter may increase the patients’ risks of serious complications such as seizures, hemorrhage, and infection. Lastly, pericatheter necrosis is a rare and potentially devastating complication (73). This is thought to arise by capillary movement of drug along the outside of the Ommaya tubing, thus exposing the contiguous brain tissue to high concentrations of drug. Early diagnosis is critical because the necrosis may progress after cessation of treatment. Most neurosurgeons do not recommend removal of such catheters for fear of removing brain tissue with it, but if infection is suspected removal will be necessary. Systemic administration of high-dose chemotherapy that penetrates the blood–brain barrier (BBB) is now viewed by many to be the preferred means of treating leptomeningeal lymphoma (74). In the past two years we have not placed an Ommaya reservoir for treatment of neoplastic meningitis. Instead for those patients thought salvageable we have employed high-dose systemic chemotherapy. Obviously this decision is individualized based on the patient’s prior treatment history and the primary tumor. Usually therapy has consisted of high-dose (3.5 g/m2 and higher) methotrexate (MTX) administered intravenously or high-dose (1–3 g/m2 intravenous) ARA-C. Studies compared CSF levels from intralumbar, intra-Ommaya, and intravenous administrations and have found the latter to be equivalent to those directly introduced (75). Because the myelosuppressive effect of MTX can be abrogated by the administration of leucovorin, MTX patients with active systemic disease may be able to continue to receive their systemic chemotherapy, an important factor in overall patient outcome. Although a devastating MTX-induced leukoencephalopathy has been described the risks of this are less if the drug is given intravenously without, or at least before, head irradiation (69,76). In our experience, high-dose intravenous MTX is usually sufficient to treat disease. 2.3.2. Centripetal Spread Because retroperitoneal adenopathy is common in both HD and in systemic NHL involvement of adjacent peripheral nervous system structures should occur. However intradural extension via centripetal spread occurs only with NHL. Again, biologic differences probably account for this observation. Lymphoma may dissect intradurally along nerve fibers from contiguous sites extradurally such as the dorsal root ganglia and spinal roots. Although the mechanism of centripetal spread is only partly understood it probably reflects the ability of malignant lymphocytes to transgress across endothelia (77) or to migrate within myelinated structures (78). This has been well-documented in multiple sclerosis and appears to be mediated by integrins and adhesion molecules (79).

574

Part VII / Neurologic Complications of Specific Malignancies

This may be much more common than recognized clinically. In one autopsy series invasion of peripheral nerves was noted in over one-third of cases (80). 2.3.3. Neurolymphomatosis Primary localization of malignant lymphoma to a peripheral nerve is rare. No series has been presented and fewer than ten case reports are in the literature. A more common situation, but no more clearly understood phenomenon, is neurolymphomatosis (NL). Diaz-Arrastia and colleagues in 1992 (81) described a clinicopathologic syndrome characterized as “a clinical disorder with signs of peripheral neuropathy that is confirmed by histopathologic evidence of lymphomatous infiltration of the nerves as seen by nerve biopsy or at autopsy.” The lymphocytes have the appearance of those of NHL rather than HD; only 1 of 40 patients studied had HD. Their review defined four clinical syndromes: (i) an acute sensorimotor illness similar to Guillain–Barré syndrome; (ii) a subacute progressive neuropathy; (iii) a mononeuropathy syndrome; and (iv) a cauda equina syndrome. Only one patient was asymptomatic. More than one half (52%) had no known lymphoma at presentation. Only nerve biopsy allowed correct diagnosis during life. Treatment when given usually involved systemic chemotherapy. Batchelor and colleagues have reported success with high-dose MTX (82). Our experience is limited but similar. A recent patient is described in Fig. 1. This 45-year-old woman presented with a mononeuropathy syndrome 24 years after treatment and complete remission of stage IIA HD. After a sural nerve biopsy confirmed neurolymphomatosis, she had a complete response to fludarabine. A similar complete response to HD-MTX was described by Ghobrial and colleagues to treat NL as the first and only site of relapse after a complete response of systemic disease following nonpenetrating chemotherapy [CHOP (83)].

2.4. Other Some unusual intradural syndromes have been described. Although rare, they may pose diagnostic and therapeutic challenges. Four such syndromes have been selected for further discussion—eosinophilic meningitis and granulomatous angiitis from HD, and pituitary and intravascular lymphoma from NHL. 2.4.1. Eosinophilic Meningitis Eosinophilic meningitis can occur from parasitic and autoimmune causes. In 1981, Patchell and Perry described CSF eosinophils in an autopsy-proven case of HD (84). A more detailed report in 1988 by Mulligan and colleagues described a 31-year-old man in remission after radiotherapy for HD who developed meningitis characterized by an eosinophilic pleocytosis (27). A neoplastic cause was suspected because of “variant” Reed–Sternberg cells in the CSF. The patient promptly responded to oral dexamethasone and intrathecal methotrexate. Systemic relapse occurred 10 months later. As of this writing it is not resolved whether this represents a true neoplastic meningitis. 2.4.2. Granulomatous Angiitis A paraneoplastic vasculitis has been described in HD. In some, an infectious cause was suspected because of the ultrastructural demonstration of what were felt to be viral particles in brain capillary endothelia. Inwards and

Fig. 1. Nerve biopsy of 45-year-old woman who presented with a mononeuritis multiplex syndrome 24 years after treatment and complete remission of stage IIA HD. Biopsy demonstrates thick epineural lymphoplasmocytic infiltrate (A) that displayed IgG kappa predominance (B).

Chapter 30 / Neurological Complications of Lymphomas

575

colleagues report a 28-year-old man who had a 5-month history of focal and generalized neurologic symptoms culminating in a thoracic myelopathy. Evaluation revealed granulomatous angiitis of the spinal cord in association with occult nodular sclerosing Hodgkin’s disease (28). Other cases reported involved brain parenchyma (85). Successful therapy for Hodgkin’s disease may result in marked improvement of associated granulomatous angiitis, whereas the lack or failure of therapy results in a uniformly fatal outcome. Definitive antemortem diagnosis of granulomatous angiitis requires a biopsy of involved tissue. The cause of granulomatous angiitis, as well as the nature of its association with Hodgkin’s disease, remains unexplained. 2.4.3. Pituitary Lymphoma In 1983, Kimmel et al. described 25 patients with diabetes insipidus (DI) on the basis of systemic cancer (86). This represented 14% of all cases of DI diagnosed at the Mayo Clinic during the period of study. DI was the initial presentation of the cancer in 11 of the 25, and 4 of these 11 were due to systemic lymphoma. Anterior pituitary and visual system involvement was uncommon. Although skull X-rays were usually normal, computed tomography (CT) was abnormal, demonstrating pituitary stalk enlargement, suprasellar masses, or both. These findings have not been updated although there are a small number of corroborating case reports from the modern neuroimaging era that included MR findings.

Fig. 2. Intravascular lymphoma: Multiple punctate and confluent areas of T2 signal intensity in the cerebral white matter (A) with spotty contrast enhancement (B) for unusual processes such as vasculitis, lymphoma, and infection. Biopsy of the lesion showed intravascular lymphomatosis with immunoperoxidase stains which support diagnosis of a large cell lymphoma, B-cell phenotype (C). (see Color Plate 9).

576

Part VII / Neurologic Complications of Specific Malignancies

2.4.4. Intravascular Lymphoma (IVL) Intravascular lymphoma (also known as malignant angioendotheliomatosis or angiotropic lymphoma) is an unusual systemic lymphoma with a predilection for the central nervous system (87). Typically, patients present with multifocal and progressive neurologic involvement in the setting of silent or occult systemic involvement. Stroke syndromes are the most common. A cauda equina syndrome with paraparesis, pain, and incontinence; a mononeuritis multiplex syndrome involving cranial and/or spinal nerves; and a subacute encephalopathy syndrome with confusion and/or delirium syndromes have been reported (88). The unexplained development of one or more of these syndromes, and especially a multi-infarct syndrome, should alert the clinician to the possibility of IVL. “B” symptoms and adenopathy are usually absent. A careful general physical examination may reveal splenomegaly and petechial hemorrhages in the skin that if biopsied may prove diagnostic. Abdominal scanning may not only confirm splenic enlargement, but may also demonstrate adrenal gland enlargement, another nearly pathognomonic feature. It is not clear whether the entity represents actual intravascular proliferation of B-lymphocytes or homing to the luminal surface of endothelia and then blockage of these vessels by malignant lymphocytes (89). Unlike the imaging in “primary” and “secondary” central nervous system lymphoma the imaging is nonspecific. Thus, CT and MR scanning will demonstrate one or more infarcts and angiography may show single or multiple occlusions. Contrast enhancement is either absent or wispy and nondiagnostic (Fig. 2 and Color Plate 9). Often a cerebrovascular biopsy is necessary; if it is, we have found it helpful to obtain a large enough sample to include leptomeninges, pial vessels, and adjacent cortex. Staging procedures usually follow to exclude occult systemic disease (staging tests are usually positive). Treatment when given has usually involved systemic chemotherapy (90).

3. PERIPHERAL NERVOUS SYSTEM (DURAL AND EXTRADURAL) Involvement of the peripheral nervous system by HD and NHL is more intuitively obvious. It usually occurs by direct extension from contiguous sites such as calvarial lymphoma extending from skin through bone to dura, and epidural spinal cord compression extending from nodes in the paraspinal “gutter” to the epidural space. Epidural spinal cord compression may be the initial presentation of systemic HD or NHL (22), although a more likely occurrence with NHL. The diagnosis of either should always lead to a search for occult systemic disease. Listing these following disorders under “peripheral nervous system” may prove to have been as arbitrary and as simplistic as the classification for “central nervous system.” Clearly calvarial lymphoma may transgress the dura and give an intradural component. As already mentioned, neurolymphomatosis seems to encompass a syndrome that has both central and peripheral (that is, intra- and extradural) components. Mechanisms that underlie these complications may be similar for intra- and extradural disorders. For example, malignant lymphocytes seem to be able to transgress the blood–nerve barrier just as they may the BBB (77). Indeed, the particular phenotype may not be simply due to invasion from a contiguous area but may also be biologically preordained. Nevertheless, we will use a traditional classification system for the peripheral nervous system disorders.

3.1. Dural Syndromes Unlike the situation intraspinally the intracranial epidural space is a virtual one. The dura is relatively impenetrable; when it does occur, invasion is an extremely poor prognostic sign. 3.1.1. Calvarial Lymphoma and Other Forms of Dural Lymphoma Herkes and colleagues described two patients who had focal neurological deficits as the initial manifestation of a malignant lymphoma involving the skull (91). Soft tissue masses and variable bone destruction were the predominant computed tomographic findings. Magnetic resonance imaging studies revealed sinus thrombosis in one case and meningeal involvement in the other (Fig. 3). Systemic lymphomas initially appearing in the skull are rare, but these lesions should be considered in patients with a rapidly developing scalp mass and invasion should be suspected in anyone who has focal neurological signs. In another report, Isla and colleagues described the imaging similarity of cranial vault lymphoma to meningioma. They reviewed other reports of patients not known to have lymphoma where the diagnosis was surgically confirmed (92). Because meningiomas are being treated more often with radiosurgery (where the diagnosis is

Chapter 30 / Neurological Complications of Lymphomas

577

Fig. 3. Superior sagittal sinus involvement by dural lymphoma. An 18-year-old woman with life-long history of Still’s disease developed progressive headache and seizures at 1 year. The dural mass enlarged (A) over one year with increased attenuation of the superior sagittal sinus (B; arrow).

based on clinical and imaging characteristics) it becomes even more important to recognize the bony component of calvarial lymphoma to avoid misdiagnosis. In some cases, calvarial lymphoma occurred in well-established patients who were in full remission. We have experience with eight such patients. None of these patients had evidence of neoplastic meningitis, although two had parenchymal swelling and pial enhancement adjacent to the involved dura. Assuming that the dura has not been broached, we assume that these patients have disease outside the BBB and proceed with systemic chemotherapy (93,94). If they are nonresponsive or if chemotherapy is not indicated, involved-field radiotherapy is employed. Tu and colleagues described intracranial marginal zone lymphoma affecting intracranial dura without a calvarial component (IC MZBCL) that mimicked meningioma (95). In this report 15 tumors were studied clinically, pathologically, and genetically. These preferentially affected middle-aged women (female-to-male ratio, 4:1), with 93% presenting as dural-based masses mimicking meningioma (Fig. 4). Like MZBCLs outside of the nervous system, they consisted of CD20+, CD3– small B-lymphocytes with varying degrees of plasmacytic differentiation and predominantly kappa light-chain restriction (78%) (Fig. 5 and Color Plate 10). Systemic disease was either not present or of low burden. Treatment is usually systemic chemotherapy and is effective. Ten patients with 1 to 7.6 years of follow-up after diagnosis showed no evidence of disease after radiation and/or chemotherapy. 3.1.2. Cavernous Sinus Lymphoma A not uncommon presentation of both HD and NHL is lymphomatous involvement of the cavernous sinus dura. Patients may present with pain in the distribution of one of the branches of the trigeminal nerve (typically V1) and/or extraocular muscle palsies and local ophthalmic signs such as chemosis. In patients with no known lymphoma the differential diagnosis is broad but typically includes cavernous sinus meningioma and inflammatory conditions. Although cavernous sinus is one of the skull base syndromes initially described by Greenberg and colleagues (96), most of these syndromes occurred with solid tumors that go to bone such as breast and prostate cancers. In solid tumors, bony disease is usually apparent on skull X-rays, bone scan, or CT scan with bone windows. In lymphoma, bone changes are unusual and a tissue diagnosis is usually required. In our experience an image-guided (usually CT) needle biopsy should be first considered to obtain tissue; when successful this will quickly and safely provide the diagnosis (97). 3.1.3. Dural Sinus Thrombosis (DST) Dural invasion by HD or NHL adjacent to a venous sinus may lead to thrombosis. This may occur by either compression or intraluminal invasion. The likelihood of subsequent thrombosis may be enhanced by such procoagulant factors as dehydration and dysproteinemia. In fact, DST is more common without apparent sinus invasion by cancer and is usually attributable to a hypercoagulable state (although many of the clinical reports did not include postmortem inspection of the sinuses).

578

Part VII / Neurologic Complications of Specific Malignancies

Fig. 4. Intracranial marginal zone lymphoma (“DALT’oma”). (A) Dural-based mass with hyperdense signal on computed tomography scan; (B,C) homogeneous contrast enhancement and a dural tail sign, mimicking meningioma, on the T1-weighted magnetic resonance images (MRIs). (D) Fluid-attenuated inversion recovery (FLAIR) MRI demonstrates prominent peritumoral edema. The hypodense regions in (C) and (D), and hyperdense signals in (A) were large amyloid deposits.

In superior sagittal sinus thrombosis, patients may present with seizures, headache, obtundation, focal and/or multifocal signs, and increased intracranial pressure (98). Imaging with MR is useful, particularly if magnetic resonance venography can be planned. DST is a potentially lethal situation when a major sinus is involved such as the superior sagittal sinus, the cavernous sinus, or the transverse sinus. Controversy exists over whether these patients should receive anticoagulation (99) but some form of anti-lymphoma therapy is usually employed. Again, treatment decisions rest with the responsible clinician and are based on realistic goals and expectations.

3.2. Epidural Syndromes Intraspinally the epidural space is a true space with traversing veins, arteries, and epidural fat. The dura reflects back on itself at the dorsal root ganglion so that this structure is technically extraspinal but intradural. Here too the dura is relatively impenetrable so that invasion when it does occur is an extremely poor prognostic sign.

Chapter 30 / Neurological Complications of Lymphomas

579

Fig. 5. Histologic features of dural marginal zone B-cell lymphomas. (A) Lymphoid infiltrate in the leptomeninges extending into the Virchow–Robin space. (B) Small- to medium-sized neoplastic cells with a distinct monocytoid appearance and frequent plasmacytoid/plasma cells (arrow). (C) Reactive lymphoid follicles with germinal centers in some patients. (D) Scattered areas with increased mitotic figures (arrowheads) were seen in two patients. (see Color Plate 10).

3.2.1. Epidural Spinal Cord Compression (ESCC) Because ESCC can be demonstrated by myelography (and thus does not depend on modern neuroimaging) an extensive literature describes this condition with important reports dating back to the early 1970s (100,101). Nevertheless, MR has clearly contributed to improved knowledge about ESCC. For example, MR can tell us about the tissue components of the mass (e.g., bone fragments, blood clot, tumor tissue), the relationship of the mass with extra- and juxtaspinal structures, and whether there is abnormal signal within adjacent spinal cord parenchyma, which is a harbinger of poor neurologic outcome. Also, Schiff and colleagues demonstrated how MR could accurately define the number of other spinal lesions in patients presenting with ESCC (102). In that study, 32% of patients with ESCC and complete spinal MR imaging had multiple epidural metastases; however, lymphoma (HD and NHL) patients were much more likely than solid tumor patients to have only a single lesion (only 16% had multiple epidural metastases). There are other ways that ESCC from solid tumors differ from lymphomatous ESCC. In ESCC of solid tumor origin (e.g., metastatic adenocarcinoma of the prostate) the classic presentation consists of bone pain followed by monoradicular pain then asymmetric myelopathy or cauda equina syndrome. Symptoms of bony instability may also be present relative to the status of the involved vertebral body. In ESCC of lymphoma this presentation is often modified. Firstly, because malignant lymphocytes may access the spinal epidural space from retroperitoneal lymph nodes, bone pain may be absent. Secondly, the syndromic time line is usually short. The mean duration of symptoms of ESCC of all tumors is approximately 6 weeks from onset to diagnosis. A more abrupt presentation may occur because of vertebral body collapse and pathologic fracture and cord ischemia. In lymphomatous ESCC the course is often very subacute even without vertebral body collapse; we have had patients whose ESCC occurred within 24 hrs; presumably this reflects an ischemic component. Thirdly, the corticosteroid benefit seen in ESCC from solid tumor is usually much more dramatic in lymphomatous ESCC. This is probably because of corticosteroidinduced lymphocyte apoptosis [the so-called “ghost tumor” effect (9)] (103). Because highly malignant lymphoma cells lose many “normal” properties, a lack of corticosteroid response does not rule this out.

580

Part VII / Neurologic Complications of Specific Malignancies

Lymphoma patients may present with ESCC although it is more commonly seen in later stages of the disease (5,22). Schiff and colleagues reviewed Mayo Clinic records of 337 patients with ESCC (104). In that study 20% of these patients presented with ESCC. Although NHL represented 16.4% of these cases overall nearly half (44%) of the lymphoma patients presented with ESCC. We have also had experience with several unusual variations on this theme. In 1992 and 1996, we described a group of patients who presented with ESCC and never developed other sites of disease (105,106). There were no distinguishing features between these patients and the more typical patient presenting with ESCC. These papers, and subsequent papers from our group, made the argument for percutaneous CT-guided needle biopsy. Lastly, even within NHL there is variation. In the previously cited Kaufman paper on neurologic complications of T-cell NHL none had ESCC (17). MR is the preferred modality for evaluation of the patient suspected to have ESCC. MR reveals T2 signal change in cord parenchyma (which has been shown to correlate with a poor outcome); it defines the tissue characteristics of the compressing mass; it can reveal clinically occult other lesions; it can demonstrate complications of therapy such as radiation myelitis; and it can show the response to therapy and pattern(s) of failure. In addition, MR can give information about spine integrity and the extent of paraspinal disease. In patients in whom MR is not feasible (unavailable, patient with pacemaker, claustrophobic patient) CT-myelogram will be the next best test. Unfortunately CT-myelography has not been shown to be comparable to MR in displaying tissue characteristics of the compressing mass, occult other disease, and cord parenchymal change. Apropos lymphoma presenting as ESCC a tissue confirmation is required and it has been our practice, as with cavernous sinus lymphoma, to first consider a CT-guided biopsy (104). Similarly, we have found this to be safe, efficient, and reliable with little or no “down time” so that therapy can commence as soon as the diagnosis is established. Because lymphomatous masses appear to respond as well to chemotherapy as they do to radiotherapy and because there does not seem to be an advantage (and perhaps a disadvantage) to surgical debulking we now avoid neurosurgical procedures if at all possible. Under certain circumstances CT-guided biopsy should also be considered in those patients with known lymphoma who develop a spinal syndrome, and whose imaging demonstrates a compressing mass. These circumstances include long disease-free interval, atypical presentation, and low-grade histology of the original NHL. As an example we have had the experience of a patient with follicular NHL who after a long period of stable disease developed a subacute myelopathy. A CT-guided biopsy showed the compressing mass to be NHL, diffuse large cell type, presumably as a “Richter’s transformation.” We recently employed positron emission tomography (PET) to demonstrate new disease in the cauda equina region 14 months after successful treatment of a more rostral lesion (Fig. 6 and Color Plate 11). As stated previously, treatment is predicated on the overall status of the patient and consensus among the responsible clinician, the patient, and the patient’s family. Traditionally, radiotherapy (RT) has been employed for treating HD and NHL ESCC (107). However, RT has important disadvantages to this group of patients, more so than in ESCC from solid tumors. Firstly, because most cases of ESCC occur within the thoracolumbar spine, RT will impact active bone marrow. Because these patients typically have active disease and are more likely to receive concurrent or subsequent chemotherapy, treatment options could be limited by RT. Furthermore, previous treatment may either limit the amount of RT a person can receive or produce more intensive side effects due to the cumulative myelosuppression. Secondly, even though HD and NHL were less likely than solid tumors to have other occult sites of ESCC, a commitment to employ RT for the symptomatic lesion might imply that the additional sites would also need to receive RT (thus, potential cumulative myelosuppression). Thirdly, RT has its own menu of local and neurologic side effects including radiation myelitis. Avoidance of additional spinal disease burden intuitively seems to be a realistic goal. For these reasons, both Burch and Grossman (108) and Wong and colleagues (109) described a satisfactory and satisfying response of ESCC to chemotherapy alone in HD and NHL, respectively. In the former, two HD patients responded “dramatically” to systemic chemotherapy. In the latter report, seven episodes of ESCC from NHL were analyzed. Five episodes were asymptomatic at presentation; one patient had back pain, leg numbness, and tingling; and one had radicular pain and mild leg weakness. After chemotherapy alone, five of seven episodes showed radiographic resolution of ESCC and improvement of neurologic deficits. One patient received consolidation radiotherapy (2,700 cGy) to the spine after chemotherapy

Chapter 30 / Neurological Complications of Lymphomas

581

Fig. 6. Role of positron emission tomography in diagnosis and management lymphoma. Sixty-two-year-old female with PCNSL involving cauda equina, conus and left retina, status post L2, L3 laminectomy, and cauda equina biopsy. PET is negative in the eyes, but positive in the conus (thin arrow) and cauda equina (thick arrow). (A) 3-D maximum intensity projection (MIP) image, (B) sagittal fused PET-CT image, (C) transaxial images of CT(I), PET(II), and fused PET/CT. (see Color Plate 11).

and had a complete response. One patient had progression of systemic lymphoma and ESCC despite chemotherapy. More recent reports bear out this initial enthusiasm (110). Our experience has been similar and we typically consider chemotherapy as the first choice of treatment for spinal cord compression caused by malignant lymphoma (110). Because both NHL and HD are highly sensitive to both radiation therapy and chemotherapy, surgery is usually only employed to obtain tissue for histologic diagnosis in patients without known lymphoma or when the diagnosis is in doubt and CT-guided biopsy is not feasible. Unlike the case in ESCC of solid tumor origin, vertebral body resection and stabilization (111) is rarely required. In patients who have recurrent ESCC after a course of radiotherapy adequate palliation can be achieved with a course of re-irradiation. Schiff and colleagues reported 54 patients who underwent two or more courses of radiotherapy to the same segment of the spinal column with radiographically documented epidural disease at the time of re-irradiation to determine outcome as measured by the ability to walk and by survival. They concluded that for patients with progressive epidural disease following radiotherapy, re-irradiation frequently preserves ambulation and carries minimal risk of radiation myelopathy during the patient’s lifetime (112).

3.3. Nerve Entrapment Syndromes 3.3.1. Numb Chin Syndrome In 1970, Simpson described mental neuropathy and coined the term “numb chin syndrome” (NCS) (113). Since then, more than 30 reports have described this syndrome in patients with metastatic cancer including lymphoma. The frequency and stereotypy of this phenomenon has led to its recognition as being due to “cancer until proven otherwise.” The syndrome is attributed to entrapment of the mental nerve by tumor as it passes through its foramen in the mandible involved with blastic metastases [breast, prostate (114)] or marrow overgrowth as in lymphoma and leukemia (115). A plain radiograph (panorex views) will usually demonstrate an abnormal foramen. Sometimes a similar syndrome can occur with more proximal disease such as involvement of the alveolar nerve along its interosseous course in the mandible (116). NCS from disease at the level of the trigeminal ganglion and even intradurally have been described (117). Thus, if neither panorex views and mandible views displays the responsible pathology we

582

Part VII / Neurologic Complications of Specific Malignancies

Fig. 7. Abnormal signal changes T12-L1 vertebrae and adjacent epidural space. Needle biopsies were obtained and were positive for large B cell lymphoma.

usually will do an enhanced MR scan and, if there is no contraindication, a lumbar puncture. In some cases a CT scan with bone windows will give additional information. Treatment is usually defined by the status of the responsible malignancy, but often consists of involved field radiotherapy. 3.3.2. Root Entrapment at Cranial and Spinal Nerve Foramens A process similar to the numb chin syndrome may occur at one or several bony egresses. Blastic lesions (the so-called “ivory” vertebra, Fig. 7) and less commonly fractures with encroachment on the foramen can cause a mononeuropathy (or multiple mononeuropathy) syndrome (118). The typical setting is a painful spinal mononeuropathy without cord or cauda signs. A simple yet helpful clue is the “winking owl” sign of a destroyed pedicle ipsilateral to the radiculopathy (Fig. 8 and Color Plate 12). Given suspicion of an epidural lesion, a CT-myelogram and/or MR is performed and is normal. However, careful views of the foramen will usually display the responsible pathology. One MR pitfall is the variability in epidural fat so that the changes at the root exit are not seen. The previously mentioned paper described five metastatic skull base syndromes: orbital, parasellar, middle fossa, jugular foramen, and occipital condyle (96). Frontal headache, diplopia, and first-division trigeminal sensory loss characterized the orbital and parasellar syndromes. Proptosis occurred with the orbital but not the parasellar syndrome. Facial pain or numbness characterized the middle fossa syndrome. The jugular foramen syndrome was characterized by hoarseness and dysphagia, with paralysis of the ninth through eleventh cranial nerves. Unilateral occipital pain and unilateral tongue paralysis characterized the occipital condyle syndrome. In our experience a common pitfall in the evaluation of the patient is the all-too-quick ascription of a cranial nerve palsy to lymphomatous meningitis. Disease at the skull base is an important differential point since by definition it is “outside” the BBB. Although localized RT is usually employed systemic chemotherapy is not unreasonable in selected patients. As was mentioned in Section 3 on “entrapment neuropathies,” careful and individualized imaging is key (119).

Fig. 8. Root entrapment at cranial and spinal nerve foramens. A simple yet helpful clue is the “winking owl” sign (A) of a destroyed pedicle ipsilateral to the radiculopathy as shown on the left in this case. (see Color Plate 12).

Chapter 30 / Neurological Complications of Lymphomas

583

A recurring clinical problem is the failure of standard CT-myelography and MR to adequately cover the lower sacral nerve roots. For example a patient may present with S2 to S5 symptoms and signs. Standard MR imaging typically goes no further than S1 or the S1-2 interspace. Again, discussion with the imager will avoid missing the responsible pathology or having to repeat the study. For the sacrum we have found simple bone X-rays to be invaluable. The bony “struts” that arch over the foramina bilaterally at each level are usually well seen. Although the “winking owl” sign does not apply (because of the different architecture of the sacrum) the same principle holds, that is, the foramina should be displayed symmetrically. Electromyography (EMG) may be useful if sufficient time has elapsed since the onset of the syndrome since proximal denervation changes may not be apparent until a minimum of 14 days after injury (120). Options for treatment include surgery, chemotherapy, and radiotherapy. The same surgical concerns that were discussed under epidural disease apply here.

3.4. Peripheral Nerve Syndromes 3.4.1. Plexopathies As with solid tumors, brachial and crural plexopathies may be of malignant origin in systemic lymphoma (121–123). Unlike solid tumors, however, the diagnosis may be more elusive. In lymphomatous plexopathies imaging is less likely to show a discrete mass and inflammatory plexopathies of dysimmune origin may occur and simulate the symptoms of plexopathies of malignant origin (124). EMG may help differentiate between an intra- and an extradural process, and in the case of the latter, localize the process to the plexus rather than the proximal nerve. Often a biopsy is necessary to establish the diagnosis. It is our practice to obtain a CSF sample and spinal MR in lymphoma patients presenting with plexopathy because as many as one-third of patients may have coincident intradural disease (5). Whether this occurs as a manifestation of neurolymphomatosis or simply multifocal disease is uncertain. Usually the lymphomatosis plexopathies occur in the context of adjacent malignant lymphadenopathy. I recently evaluated a 53-year-old man who had an explosive return of large cell NHL and began experiencing neuropathic pain along the anterior surface of the left thigh followed by leg weakness and quadriceps atrophy. Pelvis MRI demonstrated enormous adenopathy and the involvement of the adjacent plexus. Treatment with systemic chemotherapy resulted in significant improvement in both the adenopathy and the plexopathy. Radiation therapy is often deferred in such patients to avoid marrow suppression that might hinder the use of subsequent chemotherapy. Patients can usually be salvaged with external beam radiotherapy. When adjacent nodes appear normal, and especially when the patient is disease free, alternate diagnoses need to be considered. We had the experience of a 36-year-old man who was disease-free and presented with brachial plexus symptoms over 8 weeks. Imaging and electrodiagnostic testing were inconclusive. A brachial plexus exploration and biopsy revealed inflammatory cells only (125). Treatment with corticosteroids resulted in a complete remission and the patient remains disease-free 15½ years later. Unknown is whether metabolic imaging such as PET will enhance accuracy of diagnosis. Obviously the slow development of plexus symptoms and signs (especially without pain) in the distribution of an irradiated plexus raises the spectrum of radiation-induced plexopathies. Cascino and colleagues have reported on the imaging and electrophysiologic features of these conditions (126,127). Both diagnostic tests may reveal distinctive features; EMG may reveal myokymia and MRI can demonstrate wispy enhancement without a discrete mass or adjacent adenopathy. Although not exact, the window for development of postirradiation plexopathy is sufficiently reliable (127). If plexus symptoms slowly develop many years (typically a hiatus of at least a decade) after irradiation one needs to consider a radiation-induced nerve sheath tumor (128). 3.4.2. Peripheral Nerve Lymphoma As mentioned in the section on neurolymphomatosis, primary localization of malignant lymphoma to a peripheral nerve is rare. No series has been presented and fewer than 10 case reports are in the literature. Interestingly, T-cell lymphoma and the sciatic location predominate in these (129,130). Because some of these reports preceded modern neuroimaging it is possible that these reports represent cases of centripetal spread from an intradural site or cases of neurolymphomatosis. However, a carefully worked-up case by Kanamori and colleagues described a purely intraneural lesion (129). Nevertheless, the rarity of such cases means that such patients have

584

Part VII / Neurologic Complications of Specific Malignancies

disease elsewhere until proven otherwise. Treatment must be individualized. Extrapolation from parenchymal lymphoma and neurolymphomatosis would suggest that, because of the blood–nerve barrier, chemotherapy that “penetrates”—such as high-dose systemic methotrexate—might be considered.

4. PARANEOPLASTIC DISORDERS IN THE LYMPHOMAS Every paraneoplastic disorder can also be seen without known malignancy, but with clinical and paraclinical features of a dysimmune disorder (131). These syndromes presumably occur by some dysimmune process of which certain malignancies are the most common trigger. As with other malignancies, paraneoplastic disorders associated with the lymphomas have been amply described in the literature. Unlike other malignancies, however, the lymphoma-associated disorders are almost exclusively neurologic and are relatively restricted. Comprehensive and current reviews are plentiful especially the recent monograph by Hagler and Lynch (132). However, two of these lymphoma-associated syndromes are sufficiently distinctive to merit special attention—cerebellar degeneration associated with HD and subacute motor neuronopathy associated with NHL.

4.1. Subacute Cerebellar Degeneration Associated with Hodgkin’s Disease In 1951, Brain first described the association of a subacute cerebellar degeneration syndrome and cancer (133). This syndrome was further clarified in the 1980s by the identification of specific antineuronal autoantibodies to Purkinje cells and other neurons (134,135). The syndrome now known as paraneoplastic cerebellar degeneration (PCD) is profoundly disabling, often precedes the initial diagnosis of or the recurrence of a malignancy, and by clinical phenotype and antibody characteristic may predict the type of malignancy (136). Hammack and colleagues described a group of Hodgkin’s disease-associated PCD including 5 patients whose PCD heralded tumor recurrence and 2 patients who had significant spontaneous neurologic improvement (137). The report also described a specific autoantibody, later named anti-Tr. For unknown reasons PCD is not characteristic of NHL, presumably because of inherent biologic differences between HD and NHL.

4.2. Subacute Motor Neuronopathy Associated with Non-Hodgkin’s Lymphoma Schold and colleagues described a group of 10 patients with a subacute lower motor neuron syndrome associated with lymphoma. It may occur prior to the discovery of the lymphoma or after a complete remission (138). The syndrome as initially described was principally characterized by muscle weakness, often asymmetric, and purely lower motor neuron in type. Spontaneous improvement of neurological function occurred in 7 of 10 patients, and improvement was independent of the activity of the underlying neoplasm. In 2 patients postmortem examinations revealed prominent neural degeneration restricted to the anterior horns of the spinal cord and mild posterior demyelination. The mechanism is unknown.

5. NEWER DIAGNOSTIC TECHNIQUES Newer techniques have been developed to help improve the accuracy of detecting subclinical disease and reducing the number of false negative tests. Similar to CSF analysis, PCR of the rearranged immunoglobulin heavy-chain genes (139) may be useful in detecting subclinical systemic disease. In this report identical dominant PCR products were found in bone marrow aspirates, blood samples, and tumor biopsy specimens of 2 patients, indicating that the same tumor cell population is present in the CNS and in extracerebral sites. Follow-up IgH PCR demonstrated a persistent monoclonal amplification in blood in one of these patients 27 months after diagnosis presumably due to occult systemic disease. Another newer technique is the use of nuclear medicine imaging such as PET and gallium scanning. For example, PET appears to be particularly useful to determine systemic disease at staging or restaging after effective salvage therapy (140). There are few articles in the literature in which the utility of FDG PET in diagnosis of unsuspected lymphoma was closely examined. The studies that have been published deal primarily with the differentiation of central nervous system lymphoma from toxoplasmosis or other inflammatory lesions in patients with human immunodeficiency virus infection. In this role, PET has been shown to be accurate for distinguishing between malignant and inflammatory lesions (141). Nuclear medicine imaging pertains to a different domain,

Chapter 30 / Neurological Complications of Lymphomas

585

metabolism, than typically employed anatomic imaging such as CT and MR. The strengths of these latter scans are accuracy for the detection of structural abnormalities and its ability to define structures that are below the resolution of nuclear medicine imaging. However, anatomic imaging is hampered by its reliance on anatomic criteria in order to identify pathologic conditions. By assessing metabolic activity PET is not directly reliant on size to determine the presence or absence of malignancy (142). PET is also accurate for the identification of specific sites of disease and may be of particular value in assessing disorders of the peripheral nervous system. In a large study of lymph node regions evaluated with both FDG PET and CT, discordant interpretations between PET and CT images were almost always resolved in favor of PET (142). Identification of additional sites of disease resulted in an increase in disease stage in more than half of the patients (142). Gallium scintigraphy has many of the advantages of FDG PET. Both techniques are a reflection of tumor physiology rather than anatomy. Increased gallium uptake is seen in many types of lymphoma, as well as in other types of tumors. Like FDG PET, gallium-67 scintigraphy is not directly reliant on node size for determination of disease. However, this advantage is less often realized in gallium scintigraphy due to an imaging resolution of greater than 1 cm for most single photon emission computed tomography (SPECT) systems. Although relatively few studies have been performed to directly compare the accuracy of FDG PET and gallium scintigraphy for evaluation of patients with lymphoma, the data that do exist indicate that PET is the superior imaging modality. In a larger number of patients, including patients with newly diagnosed and recurrent Hodgkin’s and non-Hodgkin’s lymphoma, PET was more accurate for determination of disease stage than was gallium scintigraphy, as confirmed with pathologic or radiographic follow-up (143).

6. CONCLUSION The neurologic complications of the malignant lymphomas are common, serious, and treatable. Some, such as dural involvement by MALT tumor and paraneoplastic ataxia of Hodgkins disease, are unique to these malignancies. Compared to more common cancers the malignant lymphomas appear to affect the central and peripheral nervous systems disproportionately. As such they have the potential to significantly impact patients’ quality-of-life because of effects on special senses, mobility, communication, and cognition. Because of the inextricable intersection with disordered immune system function, an additional challenge for the treating physician is determining whether the neurologic syndrome is due to cancer or to autoimmunity. The interplay between the two contributes to the morbidity and poses a challenge to treatment. The new insights into pathogenesis and early diagnosis of these disorders will allow physicians to improve the care of these patients.

REFERENCES 1. Harris NL, Jaffe ES, Diebold J et al. World Health Organization classification of neoplastic diseases of the hematopoetic and lymphoid tissues: report of the Clinical Advisory Committee meeting: Airlie House, Virginia, November 1997. J Clin Oncol 1999;17:3835–3849. 2. Lister TA, Crowther D, Sutcliff SB et al. Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin’s disease: Cotswolds meeting. J Clin Oncol 1989;7:1630–1636. 3. Shipp MA, Harrington DP, Anderson JR et al A predictive model for aggressive non-Hodgkin’s lymphoma: the international nonHodgkin’s lymphoma prognostic factors project. N Engl J Med 1993;329:987–994. 4. Lossos IS, Morgensztern D. Prognostic biomarkers in diffuse large B-cell lymphoma. J Clin Oncol. 2006;24:995–1007. 5. Cairncross JG, Posner JB. Neurological complications of malignant lymphoma. In: Vinken PJ, Bruyn GW (eds.). Handbook of Clinical Neurology, Amsterdam: North Holland, 1980; 39:27–62. 6. Glass, J, Neurologic complications of lymphoma and leukemia. Semin Oncol. 2006;33(3):342–347 7. Recht LD, Neurologic complications of systemic lymphoma. Neurol Clin. 199;9(4):1001–1015. 8. Gametchu B. Glucocorticoid receptor-like antigen in lymphoma cell membranes: correlation to cell lysis. Science 1987;236:456–461. 9. Vaquero J, Martinez R, Rossi E et al. Primary cerebral lymphoma: the “ghost tumor.” J Neurosurg 1984;60:174–176. 10. Tucha O, Smely C, Preier M et al. Cognitive deficits before treatment among patients with brain tumors. Neurosurg 2000; 47:324–333. 11. Slivnick DJ, Ellis TM, Nawrocki JF et al. The impact of Hodgkin’s disease on the immune system. Semin Oncol 1990;17:673–682. 12. Levitt LJ, Dawson DM, Rosenthal DS et al. CNS Involvement in the non–Hodgkin’s lymphomas. Cancer 1980;45:545–452. 13. MacKintosh FR, Colby TV, Podolsky WJ et al. Central nervous system involvement in non–Hodgkin’s lymphoma: an analysis of 105 cases. Cancer 1982;49:586–595. 14. Posner JP. Neurologic Complications of Cancer. Philadelphia: F.A. Davis, 1995. 15. Hoerni-Simon G, Suchaud JP, Eghbali H et al. Secondary involvement of the central nervous system in malignant non–Hodgkin’s lymphoma: a study of 30 cases in a series of 498 patients. Oncology 1987;44:98–101. 16. Liang R, Chiu E, Loke SL. Secondary central nervous system involvement by non–Hodgkin’s lymphoma: the risk factors. Hematol Oncol 1990;8:141–145.

586

Part VII / Neurologic Complications of Specific Malignancies

17. Kaufman DK, Habermann TM, Kurtin PJ et al. Neurologic complications of peripheral and cutaneous T-cell lymphomas. Ann Neurol 1994;36:625–629. 18. Ho TP, Carrie S, Meikle D et al. T-cell lymphoma presenting as acute mastoiditis with familial palsy. Int J Pediatr Otorhinolaryngol 2004;68:1199–1201. 19. Doran M, Du Plessis DG, Larner AJ. Disseminated enteropathy-type T-cell lymphoma: cauda equina syndrome complicating coeliac disease. Clin Neurol Neurosurg 2005;107:517–520. 20. Luther N, Greenfield JP, Chadburn A et al. Intracranial nasal natural killer/T-cell lymphoma: immunopathologically confirmed case and review of literature. J Neurooncol 2005;75:185–188. 21. Sapozink MD, Kaplan HS. Intracranial Hodgkin’s disease: a report of 12 cases and review of the literature. Cancer 1983;52:1301–1307. 22. van den Bent MJ. Neurological complications of systemic lymphoma. In: Vecht ChJ (ed.). Handbook of Clinical Neurology, Elsevier Science B.V., Amsterdam 1997;69:261–287. 23. Henson RA, Urich H. Cancer and the Nervous System. Oxford: Blackwell Scientific, 1982. 24. Hsia AW, Katz JS, Hancock SL et al. Post-irradiation polyradiculopathy mimics leptomeningeal tumor on MRI. Neurology 2003;60:1694–1696. 25. Heran NS, Yong RL, Heran MS et al. Primary intradural extraarachnoid Hodgkin’s lymphoma of the cervical spine: case report. J Neurosurg Spine 2006;5(1):61–64. 26. Scheithauer BW. Cerebral metastasis in Hodgkin’s disease. Arch Pathol Lab Med 1979;103:284–287. 27. Mulligan MJ, Vasu R, Grossi CE et al. Care report: neoplastic meningitis with eosinophilic pleocytosis in Hodgkin’s disease: a case with cerebellar dysfunction. Am J Med Sci 1988;296:322–326. 28. Inwards DJ, Piepgras DG, O’Neill BP et al. Granulomatous angiitis of the spinal cord associated with Hodgkin’s disease. Cancer 1991;68:1318–1322. 29. Sheehy N, Sheehan K, Brett F et al. Hodgkins disease presenting with granulomatous angiitis of the central nervous system. J Neurol 2003;250:112–113. 30. O’Neill BP, Illig JJ. Primary central nervous system lymphoma. Mayo Clin Proc 1989;64:1005–1020. 31. Helle TL, Britt RH, Colby TV. Primary lymphoma of the central nervous system: clinicopathological study of experience at Stanford. J Neurosurg 1984;60: 94–103. 32. Horvat B, Pena C, Fisher ER. Primary reticulum cell sarcoma (microglioma) of the brain: an electron microscopic study. Arch Pathol 1969;87:609–616. 33. Kumanishi T, Washiyama K, Saito T et al. Primary malignant lymphoma of the brain: an immunohistochemical study of eight cases using a panel of monoclonal and heterologous antibodies. Acta Neuropathol (Berl) 1986;71:190–196. 34. Ersboll J, Schultz HB, Thomsen BLR et al. Meningeal involvement in non–Hodgkin’s lymphoma: symptoms, incidence, risk factors and treatment. Scand J Haematol 1985;35:487–496. 35. Bashir R, Coakham H, Hochberg F. Expression of LFA-1/ICAM-1 in CNS lymphomas: possible mechanisms for lymphoma homing into the brain. J Neuro-oncol 1992;12:103–110. 36. Steffen BJ, Butcher EC, Engelhardt B. Evidence for involvement of ICAM-1 and VCAM-1 in lymphocyte interaction with endothelium in experimental autoimmune encephalomyelitis in the central nervous system in the SJL/J mouse. Amer J Pathol 1994;145:189–201. 37. Ratech H, Burke JS, Blayney DW et al. A clinicopathologic study of malignant lymphomas of the nose, paranasal sinuses, and hard palate, including cases of lethal midline granuloma. Cancer 1989;64:2525–2531. 38. Todeschini G, Tecchio C, Degani D et al. Eighty-one percent event-free survival in advanced Burkitt’s lymphoma/leukemia: no differences in outcome between pediatric and adult patients treated with the same intensive pediatric protocol. An Oncol 1997;8(Suppl 1): 77–81. 39. Fonseca R, Habermann TM, Colgan JP et al. Testicular lymphoma is associated with a high incidence of extranodal relapse. Cancer 2000;88:154–161. 40. Young RC, Howser DM, Anderson T et al. Central nervous system complications of non–Hodgkin’s lymphoma: the potential role for prophylactic therapy. Am J Med 1979;66:435–443. 41. Recht L, Straus DJ, Cirrincione C et al. Central nervous system metastases from non–Hodgkin’s lymphoma: treatment and prophylaxis. Am J Med 1988;84:425–435. 42. Boehme V, Zeynalova S, Kloess M et al. Incidence and risk factors of central nervous system recurrence in aggressive lymphoma: a survey of 1693 patients treated in protocols of the German High-Grade Non–Hodgkin’s Lymphoma Study Group (DSHNHL). Ann Oncol 2007;18:149–157. 43. Schaumburg HH, Plank CR, Adams RD. The reticulum cell sarcoma–microglioma group of brain tumours. Brain 1972;95:199–212. 44. O’Neill BP, Dinapoli RP, Kurtin PJ et al. Occult systemic non–Hodgkin’s lymphoma (NHL) in patients I diagnosed as primary central nervous system lymphoma (PCNSL): How much staging is enough? J Neurooncol 1995;25:67–71. 45. O’Neill BP, Habermann TM, Banks PM et al. Primary central nervous system lymphoma as a variant of Richter’s syndrome in two patients with chronic lymphocytic leukemia. Cancer 1989;64:1296–1300. 46. DeAngelis LM. Primary central nervous system lymphoma as a secondary malignancy. Cancer. 1991;67:1431–1435. 47. Lossos A, Ben–Yahuda D, Bokstein F et al. Does late CNS relapse of systemic non–Hodgkin’s lymphoma (NHL) represent a residual primary disease or a second novel lymphoma? Neurology 2000;54(Suppl 3):39–40. 48. Bayliss KM, Kueck BD, Hanson CA et al. Richter’s syndrome presenting as primary central nervous system lymphoma: transformation of an identical clone. Am J Clin Pathol. 1990;93(1):117–123. 49. Landys KE, Berg GE, Torgeson JS et al. Bulky centroblastic non–Hodgkin’s lymphoma mimicking brain involvement managed with chemotherapy: a case report. Cancer 1995;76:1261–1270. 50. Cher L, Glass J, Harsh GR et al. Therapy of primary CNS lymphoma with methotrexate-based chemotherapy and deferred radiotherapy: preliminary results. Neurology 1996;46:1757–1759.

Chapter 30 / Neurological Complications of Lymphomas

587

51. Frick JC, Hansen RM, Anderson T et al. Successful high-dose intravenous cytarabine treatment of parenchymal brain involvement from malignant lymphoma. Arch Intern Med 1986;146:791–792. 52. Reni M, Ferreri AJ. Therapeutic management of refractory or relapsed primary central nervous system lymphomas. Ann Hematol. 2001;80 Suppl 3:B113–B117 53. Winkelman MD, Adelstein DJ, Karlins NL. Intramedullary spinal cord metastasis: diagnostic and therapeutic considerations. Arch Neurol 1987;44:526–531. 54. Lyding JM, Tseng A, Newman A et al. Intramedullary spinal cord metastasis in Hodgkin’s disease: rapid diagnosis and treatment resulting in neurologic recovery. Cancer 1987;60:1741–1744. 55. Lee DK, Chung CK, Kim HJ et al. Multifocal primary CNS T cell lymphoma of the spinal cord. Clin Neuropathol. 2002;21(4):149–155. 56. Riffaud L, Bernard M, Lesimple T et al. Radiation-induced spinal cord glioma subsequent to treatment of Hodgkin’s disease: case report and review. J Neurooncol. 2006;76(2):207–211. 57. Bashir RM, Bierman PJ, Vose JM et al. Central nervous system involvement in patients with diffuse aggressive non–Hodgkin’s lymphoma. Am J Clin Oncol 1991;14:478–482,. 58. Wasserstrom WR. Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49(4):759–772. 59. Broderick JP, Cascino TL. Nonconvulsive status epilepticus in a patient with leptomeningeal cancer. Mayo Clin Proc 1987;62: 835–837. 60. Behbehani RS, Vacarezza N, Sergott RC et al. Isolated optic nerve lymphoma diagnosed by optic nerve biopsy. Am J Ophthalmol 2005;139(6):1128–1130. 61. Dunne JW, Harper CG, Pamphlett R. Intramedullary spinal cord metastases: a clinical and pathological study of nine cases. Q J Med. 1986;61(235):1003–1020. 62. Schiff D, Feske SK, Wen PY. Deceptively normal ventricular fluid in lymphomatous meningitis. Arch Intern Med 1993;153:389–390. 63. Chamberlain MC, Kormanik PA, Glantz MJ. A comparison between ventricular and lumbar cerebrospinal fluid cytology in adult patients with leptomeningeal metastases. Neuro-oncol 2001;3:42–45. 64. Hug A, Storch-Hagenlocher B, Haas J et al. Single-cell PCR analysis of the immunoglobulin heavy-chain CDR3 region for the diagnosis of leptomeningeal involvement of B-cell malignancies using standard cerebrospinal fluid cytospins. J Neurol Sci 2004;219(1–2):83–88. 65. Balmaceda C, Gaynor JJ, Sun M et al. Leptomeningeal tumor in primary central nervous system lymphoma: recognition, significance, and implications. Ann Neurol 1995;38:202–209. 66. Cheng TM, O’Neill BP, Scheithauer BW et al. Chronic meningitis: the role of meningeal or cortical biopsy. Neurosurg 1994;34: 590–596. 67. Chamberlain MC. Leptomeningeal metastases: a review of evaluation and treatment. J Neuro-oncol 1998;37:271–284. 68. Chamberlain MC, Corey-Bloom J. Leptomeningeal metastases: 111 Indium–DPTA CSF flow studies. Neurology 1991;41:1765–9. 69. Bleyer WA. Neurologic sequelae of methotrexate and ionizing radiation: a new classification. Cancer Treat Rep 1981;65(Suppl 1): 89–98. 70. Glantz MJ, LaFollette S, Jaeckle KA et al. Randomized trial of a slow-release versus a standard formulation of cytarabine for the intrathecal treatment of lymphomatous meningitis. J Clin Oncol 1999;17:3110–3116. 71. Coakham HB, Kemshead JT. Treatment of neoplastic meningitis by targeted radiation using 131 I-radiolabelled monoclonal antibodies: results of responses and long-term follow-up in 40 patients. J Neuro-oncol 1998;38:225–232. 72. Hagen NA, O’Neill BP, Kelly PJ. Computer-assisted stereotactic placement of Ommaya reservoirs for delivery of chemotherapeutic agents in cancer patients. J Neuro-oncology 1987;5:273–276. 73. Obbens EA, Leavens ME, Beal JW et al. Ommaya reservoirs in 387 cancer patients: a 15-year experience. Neurology 1985;35: 1274–1278. 74. Bokstein F, Lossos A, Siegal T. Leptomeningeal metastases from solid tumors: a comparison of two prospective series treated with and without intra-cerebrospinal fluid chemotherapy. Cancer 1998;82:1756–1763. 75. Glantz, MJ, Hall WF, Cole BS et al. Diagnosis, management, and survival of patients with leptomeningeal cancer based on cerebrospinal fluid-flow status. Cancer 1995;75:2919–2931. 76. Allen JC, Rosen G, Mehta BM et al. Leukoencephalopathy following high-dose IV methotrexate chemotherapy with leucovorin rescue. Cancer Treat Rep 1980;64:1261–1273. 77. Azzarelli B, Easterling K, Norton JA. Leukemic cell–encothelial cell interaction in leukemic cell dissemination. Lab Invest 1989;60: 45–64. 78. Washington R, Burton J, Todd RF, 3rd et al. Expression of immunologically relevant endothelial cell activation antigens on isolated central nervous system microvessels from patients with multiple sclerosis. Ann Neurol 1994;35:89–97. 79. Trojano M, Avolio C, Simone IL et al. Soluble intercellular adhesion molecule-1 in serum and cerebrospinal fluid of clinically active relapsing–remitting multiple sclerosis: correlation with Gd–DPTA magnetic resonance imaging enhancement and cerebrospinal fluid findings. Neurology 1996;47:1535–1541. 80. Jellinger K, Radaszkiewez T. Involvement of the central nervous system in malignant lymphoma. Virchows Arch 1976;48:330–332. 81. Diaz-Arrastia R, Younger DS, Hair L et al. Neurolymphomatosis: a clinicopathological syndrome re-emerges. Neurology 1992;42: 1136–1141. 82. Quinones-Hinojosa A, Friedlander RM, Boyer PJ et al. Solitary sciatic nerve lymphoma as an initial manifestation of diffuse neurolymphomatosis: case report and review of the literature. J Neurosurg 2000;92:165–169. 83. Ghobrial IM, Buadi F, Spinner RT et al. High-dose intravenous methotrexate followed by autologous stem cell transplantation as a potentially effective therapy for neurolymphomatosis. Cancer. 2004;100(11):2403–2407. 84. Patchell R, Perry MC. Eosinophilic meningitis in Hodgkin’s disease. Neurology 1981;31:887–888.

588

Part VII / Neurologic Complications of Specific Malignancies

85. Delobel P, Brassat D, Danjoux M. Granulomatous angiitis of the central nervous system revealing Hodgkin’s disease. J Neurol. 2004;251(5):611–612. 86. Kimmel, DW, O’Neill, BP. Systemic cancer presenting as diabetes insipidus (di): clinical and radiographic features of eleven patients with a review of metastatic-induced diabetes insipidus. Cancer 1983;52:2355–2358. 87. Wick MR, Mills SE, Scheithauer BW et al. Reassessment of malignant “angioendotheliomatosis”: evidence in favor of its reclassification as “intravascular lymphomatosis.” Am J Surg Path 1986;10:112–123. 88. Glass J, Hochberg FC, Miller DC. Intravascular lymphomastosis. Cancer 1993;75:2919–2931. 89. Ferry JA, Harris NL, Picker JL. Intravascular lymphomatosis (malignant angioendotheliomatosis): a B-cell neoplasm expressing surface homing receptors. Mod Pathol. 1988;1(6):444–452. 90. DiGiuseppe JA, Nelson WG, Seifter EJ et al. Intravascular lymphomatosis: a clinicopathologic study of 10 cases and assessment of response to chemotherapy. J Clin Oncol 1994;12:2573–2579. 91. Herkes GK, Partington MD, O’Neill BP. Neurological features of cranial vault lymphoma: report of two cases. Neurosurgery 1991;29:898–901. 92. Isla A, Alvarez F, Gutierrez M et al. Primary cranial vault lymphoma mimicking meningioma. Neuroradiology 1996;38:211–213. 93. Mokri B. Spontaneous cerebrospinal fluid leaks: from intracranial hypotension to cerebrospinal fluid hypovolemia—evolution of a concept. Mayo Clin Proc 1999;74:1113–1123. 94. Stewart DJ. A critique of the role of the blood–brain barrier in the chemotherapy of human brain tumors. J Neuro-oncol 1994;20: 121–134 95. Tu PH, Giannini C, Judkins AR et al. Clinicopathologic and genetic profile of intracranial marginal zone lymphoma: a primary low-grade CNS lymphoma that mimics meningioma. J Clin Oncol. 2005;23(24):5718–5727. 96. Greenberg HS, Deck MD, Vikram B et al. Metastasis to the base of the skull: clinical findings in 43 patients. Neurology 1981;31: 530–537. 97. Mut M, Schiff D, Shaffrey ME. Metastasis to nervous system: spinal epidural and intramedullary metastases. J Neuro-oncol. 2005;75(1):43–56. 98. Sigsbee B, Deck MD, Posner JB. Nonmetastatic superior sagittal sinus thrombosis complicating systemic cancer. Neurology 1979;29:139–146. 99. Packer RJ, Rorke LB, Lange BJ et al. Cerebrovascular accidents in children with cancer. Pediatrics 1985;76:194–201. 100. Haddad P, Thaell JF, Keily JM et al. Lymphoma of the spinal epidural space. Cancer 1976;38:1862–1866. 101. Gilbert RW, Kim J-H, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol 1978;3:40–51. 102. Schiff D, O’Neill BP, Wang C-H et al. Neuroimaging and treatment implications of multiple spinal epidural metastases. Cancer 1998;83:1593–1601. 103. Posner JB, Howieson J, Cvitkovic E. “Disappearing” spinal cord compression: oncolytic effect of glucocorticoids (and other chemotherapeutic agents) on epidural metastases. Ann Neurol 1977;2:409–413. 104. Schiff D, O’Neill BP, Suman VJ. Spinal epidural metastasis as the initial manifestation of malignancy: clinical features and diagnostic approach. Neurology 1997;49:452–456. 105. Lyons MK, O’Neill BP, Marsh WR et al. Primary spinal epidural non–Hodgkin’s lymphoma: report of eight CT era cases and review of the literature. Neurosurg 1992;30:675–680. 106. Lyons MK, O’Neill BP, Kurtin PJ et al. Diagnosis and management of primary spinal epidural non–Hodgkin’s lymphoma. Mayo Clin Proc 1996;71:453–457. 107. Greenberg HG, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor; results with a new treatment protocol. Ann Neurol 1980;8:361–366. 108. Burch PA, Grossman SA. Treatment of epidural cord compressions from Hodgkin’s disease with chemotherapy: a report of two cases and a review of the literature. Am J Med 1988;84:555–558. 109. Wong ET, Portlock CS, O’Brien JP et al. Chemosensitive epidural spinal cord disease in non–Hodgkins lymphoma. Neurology 1996;46:1543–1547. 110. Motsubana H, Watanabe K, Sakai H et al. Rapid improvement of paraplegia caused by epidural involvements of Burkitt’s lymphoma with chemotherapy. Spine 2004;29(1):E4–E6. 111. Siegal T, Tiqva P, Siegal T. Vertebral body resection for epidural compression by malignant tumors. J Bone Joint Surg 1985;67: 375–382. 112. Schiff D, Shaw EG, Cascino TL. Outcome after spinal re-irradiation for malignant epidural spinal cord compression. Ann Neurol 1995;37:583–589. 113. Simpson JF. Numb-chin syndrome. Lancet 1970 2:1366. 114. Thompson A, Pearce I, Walton G et al. Numb-chin syndrome: an unusual presentation of metastatic prostate cancer. BJU Internat 2000;85:377–378. 115. Lossos A, Siegal T. Numb-chin syndrome in cancer patients: etiology, response to treatment, and prognostic significance. Neurology 1992;42:1181–1184. 116. Rowe WE, Cohen G, Scopp I. Numb-chin syndrome: mandibular metastasis of a reticulum-cell sarcoma. J Oral Med 1974;29:102–104. 117. Laurencet FM, Anchisi S, Tullen E et al. Mental neuropathy: report of five cases and review of the literature. Crit Rev Oncol-Hematol 2000;34:71–79. 118. Wang AM, Lewis ML, Rumbaugh CL et al. Spinal cord or nerve root compression in patients with malignant disease: CT evaluation. J Comput Assist Tomogr 1984;8:420–428. 119. Ransom DT, Dinapoli RP, Richardson RL. Cranial nerve lesions due to base of the skull metastases in prostate carcinoma. Cancer 1990;65:586–589.

Chapter 30 / Neurological Complications of Lymphomas

589

120. Kimura J. Electrodiagnosis of neuromuscular disorders. In: Bradley WL, Daroff RB, Fenichel GM et al (eds.). Neurology in Clinical Practice. Boston: Butterworth-Heinemann, 1996:477–498. 121. Cascino TL, Kori S, Krol G et al. CT of the brachial plexus in patients with cancer. Neurology 1983;33:1553–1557. 122. Kori S, Foley KM, Posner JB. Brachial plexus lesions in patients with cancer: 100 cases. Neurology 1981;31:45–50. 123. Jaeckle KA, Young DF, Foley KM. The natural history of lumbosacral plexopathy in cancer. Neurology 1985; 35:8–15. 124. Hughes RA, Britton T, Richards M. Effects of lymphoma on the peripheral nervous system. J Royal Soc Med 1994;87:526–530. 125. Lachance DH, O’Neill BP, Harper CM et al. Paraneoplastic brachial plexopathy with Hodgkin’s disease. Mayo Clin Proc 1991;66: 97–101. 126. Kimmel DW, Krecke KN, Cascino TL. Magnetic resonance imaging in cancer-related lumbosacral plexopathy Mayo Clin Proc 1997;72:823–829. 127. Harper CM, Jr, Thomas JE, Cascino TL et al. Distinction between neoplastic and radiation-induced brachial plexopathy. Neurology 1989;39:502–506. 128. Kori SH. Diagnosis and management of brachial plexus lesions in cancer patients. Oncology 1995;9:756–760. 129. Kanamori M, Matsui H, Yudoh K. Solitary T-cell lymphoma of the sciatic nerve: case report. Neurosurgery 1995;36:1203–1205. 130. Krendel DA, Stahl RL, Chan WC. Lymphomatous polyneuropathy: biopsy of clinically involved nerve and successful treatment. Arch Neurol 1991;48:330–332. 131. Lucchinetti CF, Kimmel DW, Lennon VA. Paraneoplastic and oncologic profiles of patients seropositive for type-1 antineuronal nuclear autoantibodies. Neurology 1998;50:652–657. 132. Hagler KT, Lynch JW, Jr. Paraneoplastic manifestations of lymphoma. Clin Lymphoma. 2004;5(1):29–36. 133. Brain WR, Daniel PM, Greenfield JG. Subacute cortical cerebellar degeneration and its relation to cancer. J Neurol Neurosurg Psychiatry 1951;14(2):59–75. 134. Greenlee JE, Brashear HR. Antibodies to cerebellar Purkinje cells in patients with paraneoplastic cerebellar degeneration and ovarian carcinoma. Ann Neurol 1983;14:609–613. 135. Dropcho EJ, Chen YT, Posner JB et al. Cloning of a brain protein identified by auto-antibodes from a patient with paraneoplastic cerebellar degeneration. Proc Natl Acad Sci USA 1987;84:4552–4556. 136. Dropcho EJ. Paraneoplastic disorders. Continuum 1999;5:6–99. 137. Hammack JE, Kotanides H, Rosenblum MK et al. Paraneoplastic cerebellar degeneration. I. Clinical and immunologic findings in 21 patients with Hodgkin’s disease. Neurology 1992;42:1938–1943. 138. Schold SC, Cho ES, Somasundaram M et al. Subacute motor neuronopathy: a remote effect of lymphoma. Ann Neuro 1979;5:271–287. 139. Jahnke K, Hummel M, Korfel A et al. Detection of subclinical systemic disease in primary CNS lymphoma by polymerase chain reaction of the rearranged immunoglobulin heavy-chain genes. J Clin Oncol. 2006;24(29):4754–4757. 140. Rosenfeld SS, Hoffman JM, Coleman RE et al. Studies of primary central nervous system lymphoma with fluorine-18fluorodeoxyglucose positron emission tomography. J Nucl Med. 1992;33(4):532–536. 141. Heald AE, Hoffman JM, Bartlett JA et al. Differentiation of central nervous system lesions in AIDS patients using positron emission tomography (PET). Int J STD AIDS. 1996;7(5):337–346. 142. Rohren EM, Turkington TG, Coleman RE. Clinical applications of PET in oncology. Radiology. 2004;231(2):305–332. 143. Fruchart C, Reman O, LeStang N et al. Prognostic value of early 18 fluorodeoxyglucose positron emission tomography and gallium-67 scintigraphy in aggressive lymphoma: a prospective comparative study. Leuk Lymphoma. 2006;47(12):2547–2557.

31

Neurologic Complications of Plasma Cell Dyscrasias John J. Kelly,

MD

CONTENTS Introduction Laboratory Screening for Plasma Cell Dyscrasias and Monoclonal Proteins Benign Plasma Cell Dyscrasias Multiple Myeloma Primary Systemic Amyloidosis Miscellaneous Syndromes Conclusion References

Summary Plasma cell dyscrasias are an uncommon but important cause of neurologic morbidity. The pathophysiology of these disorders is quite varied and ranges from direct effects of cancer on neurologic tissues to remote effects caused by monoclonal antibodies. This chapter discusses these hematologic disorders and their neurologic consequences including monoclonal gammopathy of undetermined significance, multiple myeloma and its rare variants, osteosclerotic myeloma and POEMS syndrome, and primary systemic amyloidosis. While not all of these disorders are classified as malignancies, they all share derivation from a single clone of plasma cells and thus are best discussed as a group. This review gives the clinician a framework for evaluating and treating these patients. Key Words: plasma cell dyscrasia, monoclonal gammopathy, anti-MAG neuropathy, multiple myeloma, osteosclerotic myeloma, POEMS syndrome, primary systemic amyloidosis

1. INTRODUCTION Plasma cell dyscrasias (PCDs), both malignant and benign, are frequently associated with neurologic diseases (1). These syndromes may be distinctive and, in some cases, are classical paraneoplastic syndrome due to the direct effects of their monoclonal proteins (M-protein) on peripheral nerves (2). This chapter will outline our current knowledge of PCDs and provide a clinical approach to these patients. A PCD is defined as a proliferation of a single clone of plasma cells, either neoplastic or non-neoplastic, and is usually associated with a monoclonal serum or urine protein (Table 1) (1,3). M-proteins consist of a single heavy chain (M, G, or A) and a single light chain (kappa or lambda) (1). Polyclonal gammopathies contain both light chains and generally more than one heavy chain, and are usually a non-neoplastic reaction to inflammatory disease or neoplasia. Occasionally, in a monoclonal gammopathy (MG), only the light chain or heavy chain is secreted in serum (light or heavy chain disease) or the light chain in urine (Bence–Jones protein). Formerly, M-proteins were thought to be biologically inert. Recently, however, these proteins have been found to possess activity directed at specific antigens, probably accounting for most of the remote effects of these disorders (4). From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

591

592

Part VII / Neurologic Complications of Specific Malignancies

Table 1 Classification of Common Plasma Cell Dyscrasias Disorder

Diagnostic Criteria

Monoclonal gammopathy of undetermined significance Osteosclerotic myeloma Multiple myeloma Waldenström’s Primary systemic amyloidosis Gamma heavy chain disease

MP in serum < 3 g/dL and no malignancy or amyloid Solitary or multiple plasmacytomas with osteosclerotic features > 10% abnormal plasma cells in bone marrow or plasmacytoma and MP in serum or urine or osteolytic lesions IgM-MP > 3 g/dl; > 10% lymphs or macroglobulinemia plasma cells in bone marrow Light chain amyloid by histology Monoclonal heavy chain in serum or urine

Abbreviations: MP, monoclonal protein. Adapted from Kelly JJ, Jr, Kyle RA, Latov N. Polyneuropathies Associated with Plasma Cell Dyscrasias. Boston: Martinus-Nijhoff, 1987.

2. LABORATORY SCREENING FOR PLASMA CELL DYSCRASIAS AND MONOCLONAL PROTEINS The M-protein is detected by screening patients with serum protein (cellulose acetate) electrophoresis (SPEP) (1). In cases where a suspicious peak is seen on SPEP and in all cases where a monoclonal gammopathy is suspected, such as idiopathic polyneuropathy or atypical motor neuron disease (MND), serum immunoelectrophoresis (IEP), or immunofixation electrophoresis (IFE) should be performed, even with a normal SPEP (3). IEP and IFE are more sensitive than SPEP for the presence of a small M-protein and allow characterization of a single heavy and light chain, thus verifying the monoclonal nature of the immunoglobulin (3). Of these two, IFE is more sensitive and will occasionally detect M-proteins when IEP and SPEP are negative. A concentrated urine specimen should also be examined because monoclonal light chains may appear in urine when serum is normal, suggesting either a malignant PCD or light chain amyloidosis. However, serum IFE and free light chain assay may eventually replace urine studies (5). After identification and characterization of an M-protein in serum or urine, further hematologic evaluation should be done to classify the PCD (Table 2) (1,6). If a diagnosis of monoclonal gammopathy of undetermined significance (MGUS) is made, M-protein levels should be monitored on a yearly basis because a sudden increase can indicate malignant transformation of a benign plasma cell dyscrasia, which occurs at a rate of approximately 1% per year (7).

Table 2 Hematologic Diagnosis of 28 Patients with PCD and Polyneuropathy Diagnosis Monoclonal gammopathy of undetermined significance Primary systemic amyloidosis Multiple myeloma (includes osteosclerotic myeloma) Waldenström’s macroglobulinemia Gamma heavy chain disease

Number 16 7 3 1 1

Abbreviation: PCD, plasma cell dyscrasia. (Adapted from Kelly JJ, Jr, Kyle RA, O’Brien PC et al. Prevalence of monoclonal protein in peripheral neuropathy. Neurology 1981;31:1480.)

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

593

3. BENIGN PLASMA CELL DYSCRASIAS 3.1. MGUS Monoclonal gammopathy of undetermined significance (MGUS), rather than the older name “benign monoclonal gammopathy,” is now the preferred term for patients with small M-protein spikes without evidence of an underlying malignancy (1,7). These disorders are often associated with peripheral neuropathies (3,6). IgMMGUS accounts for 50% or more neuropathies in most series. These syndromes are not uniform and are best approached by dividing them into IgM-MGUS and non-IgM-MGUS (IgG or IgA) (Tables 3 and 4). 3.1.1. IgM-MGUS Roughly 50% of MGUS neuropathies occur in patients with an IgM M-protein (4,8). Because the percentage of IgM gammopathies in the general population is very low, investigators have long suspected an etiologic link between neuropathies and IgM gammopathies. Subsequent studies have shown that about half of IgMMGUS neuropathies display antinerve antibody activity. Anti-MAG (myelin-associated glycoprotein) associated polyneuropathies are the most common and the prototype of this group. 3.1.1.1. Anti-MAG Neuropathy. This disorder, first described by Latov and colleagues in 1980, accounts for about 25% of the neuropathies associated with MGUS (2). Later studies by this group and others showed that the M-protein was an IgM antibody directed at MAG and other glycosphingolipids in the myelin sheath (9,10). This led to the discovery of other antinerve antibodies (Table 5) (11,12), most of which appear to be of uncertain clinical significance. Table 3 Features of Dysproteinemia Polyneuropathy Syndromes Class MGUS-IgM MGUS-IgG,A Amyloidosis OSM WM

Weakness

Sensory

Autonomic

CSF

MNCV

+ ++ +/++ +++ ++

+++ ++ ++++ ++ ++

— — ++ — —

++ + + +++ ++

D D A or D D D or A

Abbreviations: CSF, cerebrospinal fluid protein concentration; MNCV, motor nerve conduction velocity; MGUS, monoclonal gammopathy of undetermined significance; OSM, osteosclerotic myeloma; WM, Waldenström’s macroglobulinemia; D, segmental demyelination pattern; A, axonal degeneration pattern. (Adapted from Kelly JJ, Jr, Kyle RA, Latov N. Polyneuropathies Associated with Plasma Cell Dyscrasias. Boston: Martinus-Nijhoff, 1987.)

Table 4 Major Electrodiagnostic Features of PN Associated with PCD Type of PN MGUS-IgM MGUS-IgG,A OSM PSA MM

Demyel

Axonal

CTS

Pure Sensory

Other

+++ ++ +++ — +

+ ++ + +++ ++

— — — ++ +

++ + — + +

+ + — +++* ++**

Abbreviations: PN, polyneuropathy; PCD, plasma cell dyscrasia; CTS, carpal tunnel syndrome superimposed on polyneuropathy; MGUS, monoclonal gammopathy of undetermined significance; OSM, osteosclerotic myeloma; PSA, primary systemic amyloidosis; MM, multiple myeloma;*, autonomic involvement; ** , root involvement and polyradiculopathies superimposed on PN. (Adapted from Kelly JJ, Jr. Peripheral neuropathies associated with monoclonal proteins: a clinical review. Muscle Nerve 1985;8:138.)

594

Part VII / Neurologic Complications of Specific Malignancies

Table 5 Antibody Activities of Monoclonal IgM in Peripheral Nerve Disorders Antibody Activity MAG Acidic glycolipids Gangliosides GM1 and GD1b Chondroiton sulfate C Intermediate filaments Neurofilament Sulfatide

Clinical Syndrome

Pathology

Sensory > motor polyn. Polyneuropathy Motor neuron disease Sensory polyneuropathy Polyneuropathy Polyneuropathy Sensory polyn.

SD ? SD, ?AD AD SD AD AD

Abbreviations: MAG, myelin associated glycoprotein; SD, segmental demyelination; AD, axonal degeneration. (Adapted from Steck AJ, Murray N, Dellagi K et al. Peripheral neuropathy associated with monoclonal IgM autoantibody. Ann Neurol 1987;22:764.)

Clinical Features. Anti-MAG neuropathy has a fairly homogenous clinical presentation (4,13–17). Typically, these patients are older (6th –9th decades) and present with a slowly progressive, sensory dominant neuropathy. Unlike many of the mild and painful benign neuropathies of late life, these neuropathies are relatively painless. Patients instead complain of numbness and paresthesias of the feet and distal legs and gradually increasing unsteadiness due to sensory ataxia. Weakness is less prominent but becomes more evident as the disease progresses. Rare patients may present with predominant weakness, resembling chronic inflammatory demyelinating polyneuropathy. An action tremor of the hands is also prominent in some patients (18). Examination reveals striking discriminative sensory loss, including loss of vibration sense in feet and impaired position sense, accounting for the sensory ataxia usually accompanied by a positive Romberg sign. Cutaneous sensory modalities (pain and temperature) are less severely affected and autonomic dysfunction rarely occurs, helping to separate this disorder from amyloidosis. Motor strength is often impaired distally to a much lesser extent. Reflexes are usually absent in legs and depressed in arms. Nerves can be thickened and firm to palpation. The symptoms are very chronic, often progressing for months or years. Some patients are relatively stable for several years but most progress slowly. In severe cases, patients are unable to walk due mostly to sensory ataxia with varying degrees of weakness. Laboratory Tests. The EMG is very helpful and shows, in all but the earliest cases, the classical findings of a demyelinating polyneuropathy. These include marked slowing of motor conduction velocities, very prolonged distal latencies and areas of conduction block and dispersion on proximal stimulation with secondary axonal degenerative changes (15,19,20). Sensory potentials are absent or attenuated. These findings, suggestive of a demyelinative process, greatly simplify the differential diagnosis. Cerebrospinal fluid shows a high protein concentration with nonspecific features and a normal glucose and cell count. Nerve biopsy is almost pathognomonic, showing IgM deposition on the myelin sheath using immunofluorescent techniques and splitting and separation of the outer layers of compacted myelin with electron microscopy (4,21–23). General laboratory and hematological tests are negative in these patients, except for occasional patients who have Waldenström’s macroglobulinemia (24). The SPEP usually shows a small monoclonal spike. However, a negative SPEP may occur and should not obviate further testing for antinerve antibodies in the appropriate setting. IEP or IFE confirms the presence of an IgM M-protein, most commonly with kappa light chains. Further testing using ELISA and Western blot shows that the IgM antibody reacts with MAG and other sphingoglycolipid epitopes, thus establishing the diagnosis (10,11,16). Pathogenesis. There is now overwhelming evidence that the M-protein causes the neuropathy. Small studies comparing neuropathy patients with IgM M-proteins with and without anti-MAG antibody activity cannot demonstrate a difference in attributes of neuropathy between the two groups (25–27). However, laboratory data are much more convincing. The anti-MAG antibodies are deposited in the outer layers of the myelin sheath causing complement-mediated damage to the myelin sheath (4,21,23).

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

595

Separation of the outer lamellae of myelin occurs, a finding highly characteristic of this neuropathy, which is presumably due to the specificity of these antibodies for the adhesion molecules of the myelin sheath (4,21). In addition, although intraneural injection of serum in rats has not demonstrated pathologic changes comparable to those in humans, systemic injection into higher animals has demonstrated identical changes (28). Moreover, although there are exceptions, clinical improvement is generally associated with a reduction of the M-protein level in serum (14,29). Additionally, the presence of anti-MAG antibodies predicts future development of neuropathy (30). Variability of the clinical course and time of onset in individual patients may be related to differing binding affinities of the anti-MAG proteins (31). Thus, most investigators now accept the premise that anti-MAG neuropathy is an autoimmune disease caused by direct damage of the M-protein, although further work needs to be done to elucidate the specific epitopes affected and the exact mechanism of myelin damage. Treatment. Treatment is problematic in these patients, because the M-protein is difficult to eliminate. Plasmapheresis and intravenous gamma globulin infusions can help modestly, but the marked chronicity of this disorder necessitating frequent and lifelong treatments make this impractical (8,25,32–34). Cytotoxic drugs such as cyclophosphamide and fludarabine are effective, presumably due to the lowering of the M-protein level in serum (14,21,29,35). Some patients, however, respond without lowering of the M-protein level (35). Therefore, the mechanism of action of these drugs is unclear. Toxicity of cytotoxic drugs is a limiting factor, especially in elderly patients. If cytotoxics are used, monthly intravenous therapy is thought to be less toxic than daily oral therapy. Anecdotal reports have suggested that rituximab, an anti-CD20 monoclonal antibody, is helpful with limited toxicity, but its role needs to be determined by controlled trials (20,36–38). A preliminary study found that interferon-alpha seemed to help some patients (39). Careful consideration in each case must be given to whether or not to treat and, if so, how aggressively. Many patients with mild disease should not be treated aggressively unless their disease accelerates (8). 3.1.1.2. IgM Non-MAG Neuropathies. This group accounts for about 25% of the MGUS neuropathies (1,6, 16,40,41). The patients with anti-MAG–negative IgM neuropathies cannot be separated from the IgM anti-MAG cases without serological testing (25–27). Clinical manifestations (25,26) are similar to those described above for anti-MAG neuropathy. Occasionally, these patients have IgM antibody activity directed at other antigens, such as sulfatides (11). The significance of this antibody activity is uncertain because non-MAG patients are less likely to respond to immunosuppressive or cytotoxic treatments and the pathologic data is less compelling. Treatment consists of immunosuppression or cytotoxic drugs. In general, these patients respond less well to these treatments than do the anti-MAG patients (14), but there have been no controlled studies of this group. However, most patients with severe and progressing neuropathies deserve a trial of therapy. More work needs to be done on this group, with careful separation of cases, to determine if there are homogeneous groups with specific pathophysiology and if target antigens on myelin or axons can be identified. 3.1.2. IgG and IgA Neuropathies This group accounts for approximately 50% of MGUS neuropathies (6). In general, these patients are quite heterogeneous. Antinerve antibody activity is infrequently found and is of unclear significance (42,43). 3.1.2.1. MGUS-Chronic Inflammatory Demyelinating Polyneuropathy (CIDP). One distinct group presents with a syndrome resembling chronic inflammatory demyelinating polyneuropathy (44) with a subacute or chronic progressive or relapsing and remitting motor dominant polyradiculoneuropathy. EMG typically shows changes of a demyelinating neuropathy in most. Cerebrospinal fluid protein concentration (CSF) shows the typical albumino-cytologic dissociation. Nerve biopsy is nonspecific with mixed axonal and demyelinating changes with or without inflammatory cell infiltration. One reported patient, however, had an IgA gammopathy with IgA and complementary deposition on the myelin sheath associated with splitting of the myelin sheath lamellae, similar to anti-MAG neuropathy, but without detectable anti-MAG or other glycolipid reactivity (45). Otherwise, serological testing rarely identifies antinerve antibodies in these patients. Therapy is similar to CIDP, requiring long-term immunosuppresants. These patients generally respond well and have a good prognosis if treated early before severe axonal damage occurs. Thus, these patients may have conventional CIDP with the chance association of an M-protein. 3.1.2.2. Sensory or Mild Sensorimotor Neuropathy with IgG or IgA M-Protein. Sensory or mild sensorimotor neuropathy in the setting of a small IgG or IgA M-protein is the most frequent syndrome detected in MGUS

596

Part VII / Neurologic Complications of Specific Malignancies

patients (26,46). This neuropathy is generally fairly mild but symptoms are disturbing (1,46). These patients are usually older and complain of painful dysesthesias with or without autonomic disturbances. Motor manifestations are usually mild. Progression is usually very slow and symptoms are more of a nuisance than a real impairment, similar to the many cases of idiopathic sensory neuropathy that occur in elderly patients. Generally, laboratory tests, with the exception of the protein studies and EMG, are normal in this group. The presence of anemia, an elevated sedimentation rate, proteinuria, hypercalcemia or other laboratory findings should raise the question of amyloidosis or a malignant PCD (1). The SPEP typically shows a small spike and IFE confirms an IgA or G M-protein with a low concentration and no suppression of the gamma globulin fraction. EMG usually shows a mild axonal neuropathy with predominant sensory involvement. Nerve biopsy and CSF exam are not helpful and are generally not indicated. Pain control is the main goal of treatment. These patients require analgesics and other pain control medications especially at night, when the discomfort keeps them awake. If mild non-narcotic analgesics such as nonsteroidal anti-inflammatory drugs are not helpful, then the tricyclic antidepressants such as amitriptyline in a dose of 25–75 mg at bedtime should be tried. For patients in whom this is not helpful or the side-effects are limiting, gabapentin, pregabalin, or duloxetine can often help with generally fewer unpleasant side-effects (37,47). For patients not helped optimally by these treatments, the judicious use of narcotics, in a medically stable and reliable patient, can help a great deal. This syndrome is usually very slowly progressive. Immunosuppression seldom helps and is not indicated in these patients. The pathophysiology is unknown. Some patients have anti-sulfatide antibodies (11,12) or antibodies against other nerve antigens. However, the relevance of these antibodies to the nerve damage is unestablished. 3.1.2.3. Motor Neuron Disease and Monoclonal Gammopathy. There are occasional reports of patients with MND and M-protein. These usually fall into one of three categories. First, patients with amyotrophic lateral sclerosis (ALS) may have a small IgG or IgA M-protein in their blood (MGUS). This is most likely the chance co-occurrence of the M-protein and ALS in an elderly patient, which is not thought to play an etiologic role. Second, patients with an M-protein (usually IgG or A MGUS) may also have multifocal motor neuropathy, which can superficially resemble MND. The diagnosis is clarified by the restricted nature of the findings, lack of upper motor neuron involvement and characteristic EMG changes in MMN. Third, there are rare cases with apparent MND and an IgM gammopathy occasionally with antinerve activity against neural antigens (Table 5). These are likely restricted motor neuropathies or polyradiculopathies and not true motor neuron disorders and generally can be separated by lack of upper motor neuron involvement, careful EMG and clinical studies. Thus, the general consensus at this time is that true MND or ALS is not caused by a M-protein. 3.1.2.4. Others. This group consists of a number of other disorders that occur rarely in association with IgA or G M-proteins. A number of these are discussed below. They include primary systemic amyloidosis (PSA) which, in early stages, may be difficult to separate from sensory polyneuropathies associated with IgG and IgA MGUS. Rapid progression, marked pain and autonomic involvement, other organ dysfunction and abnormal laboratory studies are a clue to that diagnosis. Other rare patients have myeloma, lymphoma or cryoglobulinemia. Occasional reports cite the occurrence of a Guillain–Barré-like syndrome.

4. MULTIPLE MYELOMA Multiple myeloma (MM) is a malignant PCD with high serum and urinary concentrations of M-protein, infiltration of bone marrow by malignant plasma cells, and multiple bony plasmacytomas (3). Most neurologic complications are due to secondary effects of the tumor (hypercalcemia, infections) or to malignant infiltration of vertebrae with secondary compression of spinal cord or nerve roots due to vertebral fractures and collapse. Remote effects are much less common.

4.1. Direct Effects of Myeloma Most neurologic symptoms associated with MM are due to malignant infiltration of the vertebral column or metabolic and toxic manifestations of the malignancy or its treatment. Patients with spinal involvement usually present with segmental spinal pain and symptoms of spinal cord or cauda equina disruption below that level.

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

597

Local pain is due to infiltration of the vertebrae with collapse. Unless the cord or roots are involved, pain is usually localized to the region of the vertebra and does not radiate into the trunk or extremities. There may be tenderness to percussion over the vertebra. The pain is aggravated by spinal movements and is often worse at night when the patient is supine. If there is no clinical evidence of root or cord compression, plain X-ray or CT is often adequate for diagnosis, although MRI may be warranted to judge whether there should be concern for later epidural encroachment and cord compression. Treatment of uncomplicated vertebral involvement consists of adequate pain control, localized radiation therapy, and chemotherapy. The patient needs to be followed carefully for development of cord compression. Root and spinal cord compression cause local pain plus a radicular or cord syndrome or both, depending on the level and extent of compression. If the compression is in the thoracic region, the patient may have girdle pain around the flanks and into the chest or abdomen due to root involvement, which may suggest cardiac or abdominal disease initially. In addition, if the cord is compressed, leg symptoms and signs appear. These may be subtle at first, consisting only of mild weakness, sensory symptoms, or even isolated ataxia. Without rapid diagnosis and treatment, these patients can progress to bilateral leg weakness or paralysis, dense sensory loss to the spinal level of compression, and bladder and bowel paralysis. Patients with lesions in the cervical or lumbar area may have added limb radicular symptoms and signs, which may be the only finding early on. These patients may have root pain, lost reflexes, weakness, and dermatomal sensory loss in affected root distributions. If the cord is compressed in the cervical area, patients develop involvement of legs and arms with upper motor neuron signs in a distribution depending on the level of involvement of the cord. The cord syndrome may be asymmetrical in cases of mild or early compression but is generally more symmetrical in advanced cases. In the lumbar region, because the cord terminates normally at L1, root and cauda equina symptoms and signs predominate. In this setting, there is no evidence of a myelopathic deficit such as Babinski signs, hyperreflexia, and long tract sensory loss. Patients instead display lost reflexes, weakness and atrophy of muscles, radicular pain, and sensory loss in root distributions with bladder and bowel involvement if the conus medullaris is affected or the roots (cauda equina) are affected severely and bilaterally. Diagnosis is suspected by the association of local symptoms (spinal pain, radicular pain, and findings) and/or a spinal cord pattern of involvement of the lower extremities. Diagnosis is urgent in these cases to prevent further worsening. Emergency MRI is warranted with imaging of the suspected area. Because incomplete spinal lesions are often difficult to carefully localize with confidence, a sagittal screening MRI of the entire spine is often helpful as an initial step to detect suspicious areas that allows careful localized imaging. In patients who cannot have MRI, spinal plain films and radionuclide bone scans may be helpful in localizing collapsed vertebrae and eroded pedicles, thus suggesting the level of compression. This can be followed by CT of selected areas to look for an epidural mass. If the findings fit with the clinical picture, treatment can commence without myelography. However, since the lesion seen on CT may not be the proximate cause, careful follow-up must be maintained and any worsening should prompt urgent CT myelography. Treatment generally consists of high-dose corticosteroids, pain control, localized irradiation and chemotherapy. As mentioned above, careful monitoring is necessary to detect deterioration, which may necessitate surgical decompression, although usually this is not necessary if treatment is commenced promptly. Surgery is complicated by the frequent involvement of adjacent vertebrae, which may render the post-operative spine unstable. Outcome usually depends on speed of diagnosis, rapidity of commencing treatment, and the neurologic status before treatment. Severe pretreatment impairment usually predicts a poor result. Neuropathies or plexopathies due to localized deposits in the peripheral nerves are quite uncommon (48). Direct involvement of plasma cell disorders of the intracranial compartment is also rare but well documented. Leptomeningeal myelomatosis can occur in advanced cases of MM (49). Solitary extramedullary plasmacytomas also rarely involve the dura and are successfully managed with surgery and radiotherapy (50). Parenchymal brain plasmacytomas are exceedingly uncommon (51).

4.2. Metabolic, Toxic and Infectious Effects of Myeloma Metabolic, toxic, and infectious disorders can also cause neurologic syndromes in myeloma. These patients can develop encephalopathy, sometimes with seizures, from renal insufficiency, dehydration, hypercalcemia,

598

Part VII / Neurologic Complications of Specific Malignancies

and associated metabolic failure. Light chain deposition can cause a rapidly progressive nephropathy. Anemia, immunosuppression, and secondary infections are common. In patients with IgM myeloma, a hyperviscosity syndrome or malignant infiltration can rarely present as a CNS syndrome. In evaluation of encephalopathic patients with plasma cell dyscrasias, thorough metabolic screening with careful review of medications is indicated. If the patient is febrile or no obvious metabolic or toxic cause is found, evaluation for infection should be carried out including CSF exam for bacteria, fungi and tuberculosis. Brain imaging with CT or MRI and EEG are indicated in all cases where cause is not clear. These patients generally do well when the cause is found and the underlying problem is reversed, but recovery to baseline often takes several days or longer.

4.3. Remote Effects of Myeloma Remote effects consist mostly of peripheral neuropathies of various types. Other remote effects, such as paraneoplastic cerebellar ataxia, are extremely rare. 4.3.1. Typical Lytic MM Polyneuropathies are uncommon (53,54). They occur in only a few percent of MM patients and are diverse in nature, similar to the polyneuropathies associated with other malignancies. The exception is osteosclerotic myeloma (OSM), discussed separately below. Neuropathies associated with typical lytic MM include distal sensorimotor axonopathy, a CIDP-like syndrome, and a sensory neuropathy. In addition, these patients may also develop amyloid polyneuropathy, also known as primary systemic amyloid (PSA or AL) neuropathy or neuropathy due to deposition of light chain derived amyloid in nerve and other tissue. In one series, 20% of MM-associated neuropathies were due to AL (53). Superimposed root involvement in occasional patients may confuse the clinician by mistakenly suggesting a picture of mononeuritis multiplex. The root and cord compressive syndromes should be managed by conventional means as discussed above. The neuropathies should be separately classified by the usual techniques and treated according to type. Electromyography can be very helpful in properly classifying these disorders. Of the four types mentioned above, only the CIDP-like neuropathy is amenable to treatment using conventional immunosuppression. 4.3.2. Osteosclerotic Myeloma (OSM) and Polyneuropathy (and Related Syndromes) OSM is a rare and relatively benign variant of MM (53,55). Fewer than 3% of untreated myeloma patients have sclerotic bony lesions. In addition, while polyneuropathy is rare with typical MM, it occurs in 50% or more of reported cases with OSM. Also, in contrast to typical MM, patients with OSM are usually not systemically ill. They present because of the neuropathy or other remote effects of the malignancy rather than as a direct effect of the malignancy. Anemia, hypercalcemia, and renal insufficiency are uncommon in OSM, the bone marrow is rarely infiltrated with malignant plasma cells, and the serum M-protein concentration is low. Finally, the course of OSM is more indolent, and these patients often have prolonged survivals even without treatment. Thus, the syndrome of OSM and its paraneoplastic accompaniments are quite different from MM. 4.3.2.1. Clinical Features. Unlike MM, the polyneuropathy accompanying OSM is distinctive and homogeneous (55). Deficits are mainly motor and slowly progressive without sudden changes in severity or tempo of progression. Patients present with the onset of weakness, mostly in distal limbs initially, with gradual proximal spread accompanied by reflex loss. Sensory loss is typically less striking and tends to affect the larger sensory fibers disproportionately with greater loss of discriminative than cutaneous sensation. Pain and autonomic dysfunction, with the exception of impotence (actually due to endocrine insufficiency), are very uncommon. Nerves may be palpably thickened. The deficit is usually very symmetrical and the speed of progression is slow, often over months to years. In keeping with the nature of the underlying disorder, general laboratory studies are usually relatively uninformative. The best clue to the diagnosis is the presence of a serum M-protein, which is present in about 75–80% of patients. However, the M-protein may be very small and obscured by the normal serum protein components in the electrophoresis, emphasizing the importance of IEP or IFE in all patients with idiopathic polyneuropathy. The M-protein is characteristically IgG or IgA with a lambda light chain (occasionally kappa), and rarely present in the urine, as opposed to MM and AL.

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

599

4.3.2.2. Laboratory Tests. Neurodiagnostic studies are helpful but nonspecific (55). Nerve biopsy studies disclose a reduced concentration of myelinated fibers with changes of mixed demyelination and axonal degeneration (55). There may be mild foci of mononuclear cells in the epineurium surrounding blood vessels. These changes are nonspecific and occur in a number of neuropathies, including CIDP and diabetic polyneuropathy. The EMG (Table 4) reveals a mixture of axonal and demyelinating features that is also nonspecific but helpful in categorizing the neuropathy into the group with clear-cut demyelinating features (15,55). This feature is helpful because the differential diagnosis of demyelinating neuropathies is very limited. Cerebrospinal fluid typically discloses a normal cell count but a very high protein concentration, generally greater than 100 mg/dL and sometimes as high as several hundred milligram per deciliter. Because all these findings are nonspecific, the diagnosis often hinges on the discovery of the characteristic bony lesions and subsequent bone biopsy. The osteosclerotic lesions may be solitary or multiple (55). They tend to affect the axial skeleton and very proximal long bones but spare the distal long bones and skull (Fig. 1) (56). They may be purely sclerotic, mixed sclerotic, and lytic or lytic with a rim of sclerosis. Radioactive bone scans, although more sensitive than radiographs as a rule in detecting spinal metastases, are not as sensitive as X-rays in detecting typical MM and especially OSM lesions, probably due to the indolent nature and relative paucity of osteoclastic activity in these plasmacytomas (55,57,58). Thus, all patients with unexplained polyneuropathies that fit the clinical profile as described above should be screened with a radiographic skeletal survey, even in the absence of a serum or urine M-protein. On occasion, these lesions are misinterpreted by radiologists who are unfamiliar with their appearance and significance. Three of our patients were believed to have benign osteoslcerotic lesions (fibrous dysplasia in a rib in two and a vertebral hemangioma in one) with negative radionuclide bone scans. We insisted on biopsy because of the clinical picture and the presence of a serum M-protein and plasmacytomas were discovered, leading to diagnosis and effective treatment. Thus, if there is any question of the significance of a bony lesion in a patient with a suggestive clinical picture, the neurologist and the radiologist should review the X-rays and the lesion should be biopsied if doubt remains. In our experience open biopsy has been preferable to needle biopsy. 4.3.2.3. Pathogenesis. The cause of the polyneuropathy is not known, but most theories of pathogenesis have focused on some secretory product of the tumor, most likely the M-protein itself. However, other secretory or autoimmune products are possible including cytokines. One study suggested that cytokines IL-1 beta, TNF-alpha, and IL-6 may play a role in pathogenesis (59). Studies of antinerve antibody activity in the serum of these patients and immunocytochemical studies of nerves have been negative to date, although one study showed deposition of antibody in the endoneurium in 3 of 4 cases (60). The pathogenesis of nerve damage in this disorder and whether or not it is an axonopathy or a primary demyelinating disorder remains unresolved at this time. Recent studies

Fig. 1. Various osteosclerotic lesions from our original series of patients with osteosclerotic myeloma. Plain radiographs showing sclerotic and lucent areas in pelvis and spine.

600

Part VII / Neurologic Complications of Specific Malignancies

have focused on the effects of vascular endothelial growth factor (VEGF) secreted by the tumor, especially in the setting of POEMS syndrome (see below) (61,62). 4.3.2.4. Treatment. The diagnosis of this disorder is of more than academic interest because these patients may be helped by tumoricidal treatment. Patients with solitary lesions do best. Radiation therapy in tumoricidal doses to the lesion or surgical excision eliminates the M-protein from the serum and allows gradual recovery of the neuropathy and other symptoms over the ensuing months. However, these patients should be followed because they often relapse with the development of new lesions months to years later. This is usually heralded by the return of the neuropathy and other symptoms and the reappearance of the serum M-protein. Patients with multiple lesions are more difficult to treat. Radiation therapy is generally not an option due to the risk of toxicity. In some cases, aggressive chemotherapy, with or without local radiation therapy to large lesions, can help these patients (55,63,64), but in general the outcome is less favorable than for solitary lesions. Treatment usually requires large doses of steroids and alkylating agents. Treatments that are usually effective in autoimmune inflammatory neuropathies, such as steroids, azathioprine, plasmapheresis, and IV-Ig, are typically ineffective in these patients. Recent reports suggest that autologous stem cell transplantation may be an effective treatment (65). 4.3.2.5. Systemic Features. This disorder is also of considerable interest because some of these patients develop a multisystem syndrome that goes by a variety of names including the POEMS syndrome (polyneuropathy, organomegaly, endocrinopathy, M-protein, skin changes) (66) or the Crow–Fukase syndrome (67). These patients have, in addition to polyneuropathy, other features (Table 6) suggesting the presence of an underlying endocrinopathy or malignancy (55,66–68). The reason for the endocrinopathy is unclear. Limited data suggests a disturbance of the hypothalamicpituitary axis rather than primary end-organ failure, possibly due to antibody activity against pituitary tissue (55). The organomegaly is usually nonspecific pathologically. Biopsy of affected lymph nodes generally discloses hyperplastic changes, sometimes resembling the pathologic findings in the syndrome of angiofollicular lymph Table 6 Nonneurologic Abnormalites in 16 Patients with OSM and Polyneuropathy Abnormality Gyneomastia Hepatomegaly Splenomegaly Hyperpigmentation Edema Lymphadenopathy Papilledema Digit clubbing White nails Hypertrichosis Atrophic testes Impotence Polycythemia Leucocytosis Thrombocythemia Hypotestosterone Hyperestrogen Hypothyroidism Hyperglycemia

Patients 2 5 2 5 3 2 4 3 2 3 3 4 5 3 12 5 3 2 1

Abbreviation: OSM, osteosclerotic myeloma. (Adapted from Kelly JJ, Jr, Kyle RA, Miles JM et al. Osteosclerotic myeloma and peripheral neuropathy. Neurology (NY) 1983;33:202.)

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

601

node hyperplasia (Castleman’s disease), which is a benign localized or generalized hyperplastic lymph node syndrome of unknown etiology. Of interest, patients with generalized angiofollicular lymph node hyperplasia without bony lesions may also have the manifestations of Crow–Fukase syndrome associated with serum Mproteins or polyclonal gammopathies. Thus, it is likely that the main pathogenetic determinant of these syndromes is the presence of a serum product that is secreted by the tumor and directed against neural and other tissues. Another possibility is the secretion by the tumor of active substances such as VEGF, which has been implicated in the etiology (61,62). The term “POEMS syndrome” for these patients, however, is not entirely accurate and focuses attention on a small number of these patients to the exclusion of others (55,68). For example, of the patients with OSM polyneuropathy, most have features other than neuropathy that are fragments of a multisystemic disorder, but only a few would qualify for the term POEMS. Also, patients without myeloma may develop all the features of the POEMS syndrome. Thus, I prefer the term “Crow–Fukase syndrome” when referring to patients with polyneuropathy and multisystemic disorder, as suggested by Nakanishi and colleagues (67).

5. PRIMARY SYSTEMIC AMYLOIDOSIS PSA or ALL can occur in the setting of MM (53) or Waldenström’s macroglobulinemia (29) although most commonly presenting in the absence of a malignant plasma cell dyscrasia (69–72). This disorder should be considered in patients who develop a neuropathy in the setting of a malignant or benign plasma cell dyscrasia or in any patient who presents with a predominantly painful small fiber neuropathy with attendant autonomic symptoms.

5.1. Clinical Features This syndrome is perhaps the best characterized of the polyneuropathies associated with M-proteins and accounts for up to one-quarter of cases in some series (6). This neuropathy characteristically occurs in older men and is very rare prior to the sixth decade. Most cases are unassociated with an underlying illness but a few are associated with hematologic malignancies such as myeloma and Waldenström’s macroglobulinemia. PSA generally presents as a multisystem disease due to the deposition of fragments of the variable portion of a monoclonal light chain, most often lambda, in tissue (70–72). Patients present with a medical disease with associated (sometimes incidental) polyneuropathy (60%) or severe polyneuropathy with minimal organ involvement (40%) (71). A similar illness can occur in a variety of inherited amyloid polyneuropathies due to an abnormal circulating prealbumin (transthyretin) protein with a single amino acid substitution. Polyneuropathy does not occur in amyloidosis secondary to chronic inflammatory disease or familial CNS amyloidosis. Medical syndromes (Table 7) include the nephrotic syndrome due to amyloid infiltration of the kidneys, cardiac failure due to amyloid cardiomyopathy, chronic diarrhea with wasting due to amyloid infiltration of the gut wall and autonomic neuropathy with prominent orthostatic hypotension (71). General laboratory studies reflect the medical syndromes with proteinuria occurring in a high percentage, elevated erythrocyte sedimentation rate in about half and a mild increase in benign appearing plasma cells in bone marrow in many. Up to 90% have an M-protein in serum or a monoclonal light chain in urine when thoroughly screened with serum and urine IFE. Patients who lack an M-protein may have inherited amyloid neuropathy. Table 7 Medical Syndromes in Amyloid Polyneuropathy Syndrome Orthostatic hypotension Nephrotic syndrome Cardiac failure Malabsorption

Percent Frequency 42% 23% 23% 16%

(Adapted from Kelly, JJ, Jr, Kyle RA, O’Brien PC et al. The natural history of peripheral neuropathy in primary systemic amyloidosis. Ann Neurol 1979;6:1–7.)

602

Part VII / Neurologic Complications of Specific Malignancies

However, most have PSA and are called nonsecretory, assuming the lack of a measurable M-protein in blood or urine. In fact, immunocytologic studies of their tissue disclose amyloid derived from single (monoclonal) light chains. Presumably, the serum light chain concentration is too low in these patients. The light chains are deposited in tissue where they are digested by macrophages with the production of amyloid fibrils, which are insoluble. The polyneuropathy has been well characterized (71,72). Sensory symptoms are typically most prominent and the earliest to appear. Almost all present with numbness of the hands and legs with complaints such as burning, aching, stabbing, and shooting pains. In greater than half of patients, cutaneous sensation (light touch, pain, temperature) is more frequently and severely affected than discriminative sensation (vibration and position sense). Occasional patients (about 20%) present with the typical symptoms of carpal tunnel syndrome due to amyloid infiltration of the flexor retinaculum of the wrist before distal neuropathy symptoms appear. Rare patients present with symptoms of autonomic dysfunction without symptoms of somatic sensory dysfunction. Symptoms and signs of weakness generally follow. These are usually less prominent than the sensory findings although rare patients may present with predominantly motor findings (73). Occasional patients with amyloid infiltrative myopathy present with proximal muscle weakness and patients with malignant plasma cell dyscrasias, such as myeloma, may present with additional compressive radiculopathies which can mimic mononeuropathies or plexopathies. Otherwise the findings tend to be symmetric and predominant distally with gradual proximal spread. Most patients soon complain of autonomic dysfunction with orthostatic lightheadedness and syncope, bowel and bladder disturbances, impotence, and sweating disturbances. Hypoactive pupils and orthostatic blood pressure drop with a fixed heart rate are the most easily detected autonomic signs at the bedside.

5.2. Laboratory Tests Electrophysiologic studies (Table 4) confirm the presence of a distal axonopathy that is maximal in the legs (71). Motor conduction velocities in the “demyelinating” range (< 60% of the mean normal for that nerve) occur rarely and then only in severely affected nerves where the evoked compound muscle action potential is very low in amplitude. Sensory nerve action potentials are usually absent. Often, there is evidence of carpal tunnel syndrome, which should suggest the diagnosis. Needle EMG shows the changes expected of a distal axonopathy, with abundant signs of distal denervation and reinnervation. Cerebrospinal fluid is usually acellular typically with mild elevations of protein levels, in the 50–70 mg/dL range. Diagnosis depends on the discovery of amyloid in tissue. Sural nerve biopsy detects amyloid in almost all cases, although occasionally requiring multiple sections (71). One small study, however, reported that 6 of 10 patients with PSA neuropathy had negative nerve biopsies (74), so it is often advisable to biopsy more than one tissue. Amorphous deposits of amyloid on Congo red or cresyl violet stains typically appear in the perivascular regions of the epineurium or occasionally in the endoneurium. Amyloid is classically defined by its appearance under polarized light where the Congo red –stained deposits emit an apple green birefringence. Electron microscopy can also be used to identify the characteristic beta-pleated fibrils. Immunofluorescent staining for monoclonal light chain fragments is helpful but is technically more demanding and should be limited to experienced labs. Other useful biopsy sites include rectum, fat pad aspiration [reported to be 82% sensitive (75)] and other affected organs (Table 8) (71). Teased fiber studies show predominant axonal degeneration. The mechanism of nerve fiber damage, however, is not always readily apparent in all cases. In some instances, marked axonal degeneration appears with minimal amyloid infiltration, possibly caused by more proximal amyloid deposited at the level of the dorsal root ganglion.

5.3. Pathogenesis There have been many theories of pathogenesis of PSA neuropathy proposed, including vascular and pressure changes by the amyloid deposits. However, direct toxic effects of the amyloid fibrils on nerve fibers and dorsal root ganglion cells seem more likely (71).

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

603

Table 8 Results of Biopsy in Primary Amyloidosis with Neuropathy Site

Number of patients

Rectum Kidney Liver Small intestine Bone marrow Sural nerve Other (skin, gingiva)

Percent Positive

25 4 2 2 21 10 2

88 75 100 100 33 100 100

(Adapted from Kelly JJ, Jr, Kyle RA, O’Brien PC et al. The natural history of peripheral neuropathy in primary systemic amyloidosis. Ann Neurol 1979;6:1–7.)

5.4. Treatment Treatment is problematic. The amyloid fibrils are insoluble once deposited in tissue. Improvement is unlikely, even if amyloid deposition is halted. Thus far, the neuropathy has resisted all attempts to halt its progression with combinations of anti-inflammatory medications including steroids, alkylating agents such as melphalan and cyclophosphamide designed to slow production of the light chains, and even prolonged plasmapheresis aimed at lowering the light chain concentration in serum (71,76,77). However, the nephropathy due to light chain deposition has been shown to be at least partially reversible with a combination of melphalan and prednisone (76,77). Recently, treatment of this disorder with autologous stem cell transplantation has been effective (52,78). Gertz et al. reported a 2-year actuarial survival of 70% in a highly selected group of patients with minimal cardiac or renal involvement (52). Without effective treatment, these patients usually progress inexorably with increasing numbness and pain, autonomic failure, and weakness with added multi-organ failure in many cases. Death usually occurs 2–4 years from the time of diagnosis, and is generally due to major organ failure, cardiac most commonly. Diagnosis is delayed most in patients with relatively pure neuropathies without significant organ failure (median 26 months) (71,79). The disease has a dismal prognosis and 85% are dead within 25 months (71).

6. MISCELLANEOUS SYNDROMES 6.1. Waldenström’s Macroglobulinemia (WM) Separating WM from IgM-MGUS is sometimes difficult and the latter may evolve into WM over time (80). Thus, similar polyneuropathy syndromes occur in both. The most frequent polyneuropathy encountered is likely that associated with anti-MAG antibodies (16). This syndrome has the same features and clinical course as in IgM-MGUS since, despite the presence of a malignant plasma cell dyscrasia, the anti-MAG antibody determines the type of neuropathy. One patient with anti-MAG neuropathy and WM reportedly responded to bone marrow transplantation (80). Other patients may have a CIDP-like picture, a distal axonal neuropathy, typical amyloid polyneuropathy or even the sensory neuronopathy syndrome usually seen with small cell cancer of the lung. Rare patients develop central nervous system symptoms due to hyperviscosity, requiring urgent lowering of viscosity via plasmapheresis. These patients present with encephalopathy with or without seizures. Treatment is based on rapid lowering of IgM levels, hydration, and chemotherapy to lower IgM production. The prognosis is often poor unless the patient is treated promptly before there is significant neurologic deterioration.

6.2. Cryoglobulinemia This disorder is usually divided into three types (41,81). In type 1, the M-protein itself is a cryoglobulin in the setting of a plasma cell dyscrasia. In type 2, the cryoglobulin is a mixture of an M-protein of IgM type with rheumatoid factor activity against polyclonal immunoglobulins, usually occurring in the setting of a lymphoproliferative disorder. Type 3 occurs in the setting of a collagen-vascular or other chronic inflammatory

604

Part VII / Neurologic Complications of Specific Malignancies

disease and the cryoglobulin consists of polyclonal immunoglobulins. The polyneuropathy in all these syndromes is painful, symmetrical or asymmetrical, sensorimotor, and axonal in nature. Purpura occurs in distal limbs in a high percentage of patients, and the neuropathy is generally due to a vasculopathy or vasculitis of skin and vasa nervorum.

6.3. Lymphoma, Leukemia, Cancer These disorders can be associated with M-protein and polyneuropathy (16). In lymphoma with IgM M-protein, the IgM may have anti-MAG activity with the usual clinical and pathological features. Other syndromes without clear antinerve activity in the M-protein fraction may respond to ablation of the malignancy. Still others have an unclear relation to the malignancy and show little response to tumoricidal treatment or to lowering of the M-protein concentration in serum.

7. CONCLUSION The topic of plasma cell dyscrasias and neurologic disease has been a fruitful area for active research over the last decade and more, and an organized approach, as presented in this review, helps with diagnosis. Prompt treatment can reverse many of these syndromes. The neuropathy patients especially are of great importance to recognize because treatment may lead to remission in some cases and careful study of these patients may lead to a better understanding of the pathogenesis of polyneuropathies and possibly MND. This may in turn lead to effective treatment for conditions for which there are now no effective treatments. Therefore, despite their relative infrequency, increased recognition of these neuropathies will continue to be a high priority for both peripheral nerve specialists and for general neurologists.

REFERENCES 1. Kelly JJ, Jr: Peripheral neuropathies associated with monoclonal proteins: a clinical review. Muscle Nerve 1985;8:138–150. 2. Latov N, Sherman WH, Nemni R et al. Plasma cell dyscrasia and peripheral neuropathy with a monoclonal antibody to peripheral nerve myelin. N Engl J Med 1980;303:618–621. 3. Kyle RA. Plasma cell dyscrasias. In: Spitell JA, Jr, (ed.). Clinical Medicine. Philadelphia: Harper & Row, 1981:1–35. 4. Latov NR, Hays AP, Sherman WH. Peripheral neuropathy and anti-MAG antibodies. CRC Crit Rev Neurobiol 1988;3:301–332. 5. Katzman JA, Disperienzi A, Kyle RA et al. Elimination of the need for urine studies in the screening algorithm for monocalon gammopathies by using serum immunfixation and free light chain assays. Mayo Clinic Proc 2006;81(12):1575–1578. 6. Kelly JJ, Jr, Kyle RA, O’Brien PC et al. Prevalence of monoclonal protein in peripheral neuropathy. Neurology 1981;31:1480–1483. 7. Kyle RA. “Benign” monoclonal gammopathy: A misnomer? JAMA 1984;251:1849–1854. 8. Kelly JJ, Jr, Kyle RA, Latov N. Polyneuropathis Associated with Plasma Cell Dyscrasias. Boston: Martinus-Nijhoff, 1987. 9. Latov N, Braun PE, Gross RA et al. Plasma cell dyscrasia and peripheral neuropathy: identification of the myelin antigens that react with human paraproteins. Proc Nat Acad Sci 1981;78:7139–7142. 10. Yu RK, Ariga T. The role of glycosphingolipids in neurological disorders: mechanisms of immune action. Ann NY Acad Sci 1998;19:285–306. 11. Ilyas AA, Cook SD, Dalakas MC et al. Anti-MAG IgM paraproteins from some patients with polyneuropathy associated with IgM paraproteinemia also react with sulfatide. J Neuroimmunol 1992;37:85–92. 12. Steck AJ, Murray N, Dellagi K et al. Peripheral neuropathy associated with monoclonal IgM autoantibody. Ann Neurol 1987;22: 764–767. 13. Chassande B, Leger JM, Younes-Chennoufi AB, et al. Peripheral neuropathy associated with IgM monoclonal gammopathy: correlations between M-protein antibody activity and clinical/electrophysiological features in 40 cases. Muscle Nerve 1998;21:55–62. 14. Kelly JJ, Jr, Adelman LS, Berkman E et al. Polyneuropathies associated with IgM monoclonal gammopathies. Arch Neurol 1988;45;1355–1359. 15. Kelly JJ, Jr. The electrodiagnostic findings in polyneuropathies associated with IgM monoclonal gammopathies. Muscle Nerve. 1990;13:1113–1117. 16. Latov N, Wokke JHJ, Kelly JJ, Jr. Immunological and infectious diseases of the peripheral nerves. Cambridge: Cambridge University Press, 1997. 17. Melmed C, Frail DE, Duncan I et al. Peripheral neuropathy with IgM kappa monoclonal immunoglobulin directed against myelinassociated glycoprotein. Neurology 1983;33:1397–1405. 18. Bain PG, Britton TC, Jenkins IH et al. Tremor associated with benign IgM paraproteinemia. Brain 1996;119:789–799. 19. Kaku DA, England JD, Sumner AJ. Distal accentuation of conduction slowing in polyneuropathy associated with antibodies to myelin-associated glycoprotein and sulphated glucuronyl paragloboside. Brain 1994;117:941–947. 20. Kelly JJ. Neurologic disorders in benign and malignant plasma cell dyscrasias. In: Noseworthy JL, (ed.) Neurologic Therapeutics: Principles and Practice, Vol. 2, 2nd ed., London: Martin Dunitz, 2006:1361–1372.

Chapter 31 / Neurologic Complications of Plasma Cell Dyscrasias

605

21. Ellie E, Vital A, Steck A et al. Neuropathy associated with “benign” anti-myelin-associated glycoprotein IgM gammopathy: clinical, immunological, neurophysiological, and pathological findings and response to treatment in 33 cases. J Neurol 1996;243:34–43. 22. Jacobs JM. Morphological changes at paranodes in IgM paraproteinaemic neuropathy. Microsc Res Tech 1996;34:544–553. 23. Nemni R, Galassi G, Latov N et al. Polyneuropathy in nonmalignant IgM plasma cell dyscrasia: a morphological study. Ann Neurol 1983;14:43–54. 24. Levine T, Pestronk A, Florence J et al. Peripheral neuropathies in Waldenström’s macroglobulinaemia. J Neurol Neurosurg Psychiatry. 2006;77:224–228. 25. Suarez GA, Kelly JJ, Jr. Polyneuropathy associated with monoclonal gammopathy of undetermined significance: further evidence that IgM-MGUS neuropathies are different than IgG-MGUS. Neurology 1993;43:1304–1308. 26. Gosselin S, Kyle R, Dyck P. Neuropathy associated with monoclonal gammopathies of undetermined significance. Ann Neurol 1991;30:54–61. 27. Simovic D, Gorson KC, Ropper AH. Comparison of IgM-MGUS and IgG-MGUS polyneuropathy. Acta Neurol Scand 1998;97: 194–200. 28. Tatum AH. Experimental paraprotein neuropathy, demyelination by passive transfer of human IgM anti-myelin-associated glycoprotein. Ann Neurol 1993:33;502–506. 29. Notermans NC, Lokhorst HM, Franssen H et al. Intermittent cyclophosphamide and prednisone treatment of polyneuropathy associated with monoclonal gammopathy of undetermined significance. Neurology 1996;47:1227–1233. 30. Meucci N, Baldini L, Cappellari A, et al. Anti–myelin–associated antibodies predict the development of neuropathy in asymptomatic patients with IgM monoclonal gammopathy. Ann Neurol 1999;46:119–222. 31. Weiss MD, Dalakas MC, Lauter CJ et al. Variability in the binding of anti-MAG and anti-SGPG antibodies to target antigens in demyelinating neuropathy and IgM paraproteinemias. J Neuroimmunol 1999;95:174–184. 32. Dalakas MC, Quarles RH, Farrer RG et al. A controlled study of intravenous immunoglobulin in demyelinating neuropathy with IgM gammopathy. Ann Neurol 1996;40:792–795. 33. Ernerudh JH, Vrethem M, Andersen O et al. Immunochemical and clinic effects of immunosuppressive treatment in monoclonal IgM neuropathy. J Neurolog Neurosurg Psychiatry 1992;55:930–934. 34. Nobile-Orazio E, Meucci N, Baldini L et al. Long-term prognosis of neuropathy associated with anti-MAG IgM M-proteins and its relationship to immune therapies. Brain 2000;123:710–717. 35. Wilson HC, Lunn MP, Schey S et al. Successful treatment of IgM paraporteinemia with fludarabine. J Neurol Neurosurg Psychiatry 1999;66:575–580. 36. European Federation of Neurological Societies, Peripheral Nerve Society, Hadden RD et al. European Federation of Neurological Societies/Peripheral Nerve Society guideline on management of paraproteinaemic demyelinating neuropathies: report of a joint task force of the European Federation of Neurological Societies and the Peripheral Nerve Society. Eur J Neurol 2006;13:809–818. 37. Sonnett TE, Setter SM, Campbell RK. Pregabalin for the treatment of painful neuropathy. Expert Rev Neurother. 2006;6:1629–1635. 38. Kelly JJ. Chronic peripheral neuropathy responsive to rituximab. Rev Neurol Dis. 2006;3:78–81. 39. Mariette X, Chastang C, Clavelou P et al. A randomized clinical trial comparing interferon-alpha and intravenous immunoglobulin in polyneuropathy associated with monoclonal IgM: the IgM-associated Polyneuropathy Study Group. J Neurol Neurosurg Psychiatry 1997;63:28–34. 40. Osby LE, Noring L, Hast R et al. Benign monoclonal gammopathy and peripheral neuropathy. Br J Haematol 1982;51:531–539. 41. McLeod JG, Walsh JC, Pollard JD. Neuropathies associated with paraproteinemias and dysproteinemias. In: Dyck PJ, Thomas PK, Lambert EH et al. (eds.) Peripheral Neuropathy, 2nd ed. Philadelphia: W.B. Saunders, 1984:1857–1860. 42. DiTroia A, Carpo M, Meucci N et al. Clinical features and antineural reactivity in neuropathy associated with IgG monoclonal gammopathy of undetermined significance. J Neurol Sci 1999;15:64–71. 43. Nobile-Orazio E, Casellato C, Di Troia A. Neuropathies associated with IgG and IgA monoclonal gammopathy. Rev Neurol (Paris) 2002;158:979–987. 44. Simmons Z, Albers JW, Bromberg MB et al. Presentation and initial clinical course in patients with chronic inflammatory demyelinating polyradiculoneuropathy: comparison of patients without and with monoclonal gammopathy. Neurology. 1993;43:2202–2209. 45. Vallat JM, Tabaraud F, Sindou P et al. Myelin widenings and MGUS-IgA: an immunoelectron microscopic study. Ann Neurol 2000;47:808–811. 46. Gorson KC, Ropper AH. Axonal neuropathy associated with monoclonal gammopathy of undetermined significance. J Neurol Neurosurg Psychiatry 1997;63:163–168. 47. Raskin J, Pritchett YL, Wang F et al. A double-blind, randomized multicenter trial comparing duloxetine with placebo in the management of diabetic peripheral neuropathic pain. Pain Med 2005;6:346–356. 48. Denier C, Lozeron P, Adams D et al. Multifocal neuropathy due to plasma cell infiltration of peripheral nerves in multiple myeloma. Neurology 2006;28:917–918. 49. Leifer D, Grabowski T, Simonian N et al. Leptomeningeal myelomatosis presenting with mental status changes and other neurologic findings. Cancer 1992;70:1899–1904. 50. Roddie P, Collie D, Johnson P. Myelomatous involvement of the dura mater: a rare complication of multiple myeloma. J Clin Pathol 2000;53:398–399. 51. Wisniewski T, Sisti M, Inhirami G et al. Intracerebral solitary plasmacytoma. Neurosurgery 1990;27:826–829; discussion 829. 52. Gertz MA, Lacy MQ, Dispenzieri A et al. Stem cell transplantation for the management of primary systemic amyloidosis. Am J Med 2002;113:549–555. 53. Kelly JJ, Jr, Kyle RA, Miles JM et al. The spectrum of peripheral neuropathy in myeloma. Neurology (NY) 1981;31:24–31. 54. Kyle RA. Clinical aspects of multiple myeloma and related disorders including amyloidosis. Pathol Biol 1999;47:148–157. 55. Kelly JJ, Jr, Kyle RA, Miles JM et al. Osteosclerotic myeloma and peripheral neuropathy. Neurology (NY) 1983;33:202–210.

606

Part VII / Neurologic Complications of Specific Malignancies

56. Michel JL, Gaucher-Hugel AS, Reynier C et al. POEMS syndrome: imaging of skeletal manifestations, a study of 8 cases. J Radiol 2003;84:393–397. 57. Lindstrom E, Lindstrom FD. Skeletal scintigraphy with technetium diphosphonate in multiple myeloma: a comparison with skeletal X-ray. Acta Med Scand 1980;208:289–291. 58. Tamir R, Glanz I, Lubin E et al. Comparison of the sensitivity of 99mTc–methyl diphosphonate bone scan with the skeletal X-ray survey in multiple myeloma. Acta Haematol 1983;69:236–242. 59. Gherardi RK, Authier FJ, Belec L. Les cytokines pro-inflammatoires: une cle pathogenique du syndrome POEMS. Rev Neurologique 1996;64:809–812. 60. Adams D, Said G. Ultrastructural characterisation of the M protein in nerve biopsy of patients with POEMS syndrome. J Neurol Neurosurg Psychiatry 1998;64:809–812. 61. Dvorak HF, Brown LF, Detmar M et al. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995;146:1029–1039. 62. Scarlato M, Previtali SC, Carpo M et al. Polyneuropathy in POEMS syndrome: role of angiogenic factors in the pathogenesis. Brain 2005;128:1911–1920. 63. Donofrio PD, Albers JW, Greenberg HS et al. Peripheral neuropathy in osteosclerotic myeloma: clinical and electrodiagnostic improvement with chemotherapy. Muscle Nerve 1984;7:137–141. 64. Kuwabara S, Hattori T, Shimoe Y et al. Long-term melphalan–prednisolone chemotherapy for POEMS syndrome. J Neurol Neurosurg Psychiatry 1997;63:385–387. 65. Dispenzieri A, Moreno-Aspitia A, Suarez GA et al. Peripheral blood stem cell transplantation in 16 patients with POEMS syndrome, and a review of the literature. Blood 2004;104:3400–7. Epub 2004 Jul 27. 66. Bardwick PZ, Zvaifler NJ, Gill GN et al. Plasma cell dyscrasia with polyneuropathy, organomegaly, endocrinopathy, M-protein and skin changes: the POEMS syndrome. Medicine 1980;59:311–322. 67. Nakanishi T, Sobue I, Toyokura Y et al. The Crow–Fukase syndrome: a study of 102 cases in Japan. Neurology 1984;34:712–720. 68. Miralles GD, O’Fallon J, Talley NJ. Plasma cell dyscrasia with polyneuropathy: the spectrum of POEMS syndrome. N Engl J Med 1992;327:1919–1923. 69. DeNavasquez S, Treble HA. A case of primary generalized amyloid disease with involvement of the nerves. Brain 1938;61:116–128. 70. Gertz MA, Lacy MQ, Dispenzieri A. Amyloidosis: recognition, confirmation, prognosis and therapy. Mayo Clin Proc 1999;74:490–494. 71. Kelly JJ, Jr, Kyle RA, O’Brien PC et al. The natural history of peripheral neuropathy in primary systemic amyloidosis. Ann Neurol 1979;6:1–7. 72. Trotter JL, Engel WE, Ignaczak TF. Amyloidosis with plasma cell dyscrasia: an overlooked cause of adult onset sensorimotor polyneuropathy. Arch Neurol 1947;34:209–214. 73. Quattrini A, Nemni R, Sferrazza B et al. Amyloid neuropathy simulating lower motor neuron disease. Neurology 1998;51:600–602. 74. Simmons Z, Blaivas M, Aguilera AJ et al. Low diagnostic yield of sural nerve biopsy in patients with peripheral neuropathy and primary amyloidosis. J Neurol Sci 1993;120:60–63. 75. Masouye I. Diagnostic screening of systemic amyloidosis by abdominal fat aspiration: an analysis of 100 cases. Am J Dermatopathol 1997;19:41–45. 76. Gertz MA, Kyle RZ, Greipp PR. Response rates and survival in primary systemic amyloidosis. Blood 1991;77:257–262. 77. Gertz MA, Kyle RA. Amyloidosis: prognosis and treatment. Semin Arthritis Rheum 1994;24:124–138. 78. Gertz M. Autologous attack on amyloidsis. Mayo Clin Proc 2006;81:874–876. 79. Rajkumar SV, Gertz MA, Kyle RZ. Prognosis of patients with primary systemic amyloidosis who present with dominant neuropathy. Am J Med 1998;104:232–237. 80. Rudnicki SA, Harik SI, Dhodapkar M et al. Nervous system dysfunction in Waldenström’s macroglobulinemia: response to treatment. Neurology 1998;51:1210–1213. 81. Logothetis J, Kennedy WR, Ellington A et al. Cryoglobulinemic neuropathy. Arch Neurol 1968;19:389–397.

32

Neurologic Complications of Pediatric Systemic Cancer Nicole J. Ullrich,

MD, PHD

and Scott L. Pomeroy,

MD, PHD

CONTENTS Introduction Direct Complications of Malignancy in Children Indirect and Treatment-Related Complications in Children Conclusions References

Summary Each year in the United States, an average of one to two children per 10,000 develop cancer. Survival rates for children with cancer have continued to increase quite dramatically over the last several decades. Many cancer patients have symptomatic neurologic complications during the course of their illness, and neurologic problems are a common reason for hospitalization of both adult and pediatric patients with systemic cancer. Therapeutic improvements, longer survival times, and improved imaging modalities have increased both the amount of time at risk and the possibility of detecting such complications. The incidence, timing, etiology and treatment of neurologic sequelae in children with cancer has not been extensively evaluated. The purpose of this review is to outline the major tumor- and treatment-related neurologic sequelae of pediatric cancer. Key Words: cancer, children, late effects, leptomeningeal disease, cognition, neuropathy, seizures, headache

1. INTRODUCTION Cancer is the most common disease-related cause of death for children under the age of 20 years and remains the fourth most common cause of all deaths, after accidents, homicides, and suicide (1). The most frequent cancers in children include leukemia and central nervous system (CNS) tumors, which comprise nearly half of new cases. Brain tumors are the second most common malignancy of childhood and are the most common solid tumors, representing approximately 20% of childhood cancers (Fig. 1 and Color Plate 13) (2). Although there was an overall decline of 40% in cancer mortality between 1975 and 1995 (3), this improved overall survival and increase in cure rates excluded brain tumors, for which the overall prognosis remained quite poor. The apparent increase in brain tumors may have been in part related to improved detection, with wider availability of magnetic resonance imaging (MRI). It has been estimated that over two-thirds of pediatric cancer patients will be long-term survivors of their underlying disease. As survival continues to improve, the detrimental consequences of this treatment on the nervous system, in particular the developing nervous system, are now better appreciated. Exact frequencies and attribution of these effects to the underlying tumor or its treatment are often quite difficult to determine. As with adults, the incidence of neurologic complications from systemic cancer in children has been increasing. This may be partly due to the better overall recognition of these types of issues, but is also likely a reflection From: Current Clinical Oncology: Cancer Neurology in Clinical Practice Edited by: D. Schiff, S. Kesari, and P. Y. Wen © Humana Press, Totowa, NJ

607

608

Part VII / Neurologic Complications of Specific Malignancies Distribution of cancer types: Ages 15–19, all races, both sexes, SEER 1986–1995

NHL Hodgkin's Other Rhabdomyosarcoma Ewings Osteosarcoma Non-RMS Sarcoma AML ALL Melanoma Thyroid CNS Germ cell

Fig. 1. Distribution of cancer types among children 0–19 years old, 1988–2001 (2). (see Color Plate 13).

Table 1 Neurologic Complications of Cancer in Children Direct, Cancer-Related Toxicities Local compression Metastatic disease Leptomeningeal carcinomatosis

Indirect, Non-metastatic neurologic effects Paraneoplastic disorder Seizure Headache/migraine Tic disorder, movement disorder Static encephalopathy

Treatment Related/Iatrogenic Toxicities Headache Seizure Neuropathy Toxic/metabolic encephalopathy Cerebral infarction Drug-specific effects Radiation complications Central nervous system infections Effects of stem cell transplantation

of improved overall survival, particularly in the pediatric population. In the adult population, neurologic effects have been well-documented; it is estimated that as many as two-thirds of patients with cancer will develop some type of neurologic problem during the course of their illness and therapy (4–6). Neurologic complications of childhood solid malignant tumors have rarely been described, although the rate may be as high as one-third of patients (7,8). In general, the focus of prior studies in children has been on metastatic disease (9–11). As is well appreciated, children are not “mini-adults”. Cancer presentation, manifestations, response to therapies, and late effects often differ in children compared to the adult population. Rather than focusing on a particular tumor subtype in children, as has been done for the adult malignancies, this chapter will focus on the unique aspects of pediatric cancer in general. In children as with adults, these complications can be divided into metastatic neurologic effects, nonmetastatic neurologic effects, and the neurologic toxicity of tumor-directed therapies (see Table 1).

2. DIRECT COMPLICATIONS OF MALIGNANCY IN CHILDREN 2.1. Brain Metastases Metastases of somatic cancers to the brain are common in adults with cancer, with approximately 25–35% of adult patients developing intracranial parenchymal lesions (12,13). By contrast, central nervous system metastases are much less common in most extracranial pediatric solid tumors (10,14–17). Brain metastases have been reported

Chapter 32 / Neurologic Complications of Pediatric Systemic Cancer

609

Table 2 Solid Tumors that Lead to Brain Metastases in Children Primary tumor type Osteosarcoma Ewing’s sarcoma family tumors Soft tissue sarcoma Wilms’ tumor Germ cell tumor Neuroblastoma Other

Frequency (range) (10,11,15,17,114) 5–13% 0–17% 2–14% 1–13% 0–50% 0–8% 2–19%

Other = retinoblastoma, hepatoblastoma, melanoma, malignant schwannoma, lung cancer.

mostly as case reports, with an increased predominance in osteosarcoma, Ewing’s sarcoma, soft tissue sarcoma, neuroblastoma, hepatoblastoma, germ cell tumors and retinoblastoma. The overall incidence has been reported in the range of 1.5–2.5% in clinical series and 6–13% in autopsy series (see Table 2) (10,11,15,17,18). Because of the relative rarity of brain metastases in children, information regarding presenting symptoms, pattern of spread, management and further prognosis is quite limited. In the largest series of patterns of brain metastases in pediatric solid tumors, most patients had concurrent pulmonary metastases, suggesting a hematogenous spread of tumor emboli from the pulmonary metastasis (15,16). In neuroblastoma, by contrast, metastases are though to arise from adjacent bone metastases, and in these cases local invasion of the skull or brain parenchyma and spinal metastases were seen, rather than isolated parenchymal lesions (17). Children are thought to have more rapid onset of neurologic manifestations, when compared with adults, perhaps because of a higher proliferative rate of pediatric malignancies (17). Symptoms include increased intracranial pressure, hemiparesis, cranial nerve palsies, mental status changes, and seizures, the latter with an incidence ranging from 8% to 60% (7,15,17). In most cases, surgical extirpation is not attempted. Overall prognosis in children with known brain metastasis is quite poor, independent of underlying pathology or treatment, with a median survival of 1–6 months. Even with control of brain disease, the majority of patients die of other systemic disease (7,10,14–18).

2.2. Leptomeningeal/Spinal Metastatic Disease Leptomeningeal metastases are found most frequently in children with leukemia or primary CNS malignancy (19–21). Prior to the use of prophylactic CNS therapy, nearly 50% of children with leukemia had CNS disease, which was the major cause of treatment failure, even after bone marrow transplantation (20,22). Now, fewer than 10% of patients have CNS leukemia. Patients at highest risk are treated with a combination of intrathecal methotrexate, with or without cytarabine and dexamethasone, and craniospinal irradiation (20). Leptomeningeal lymphoma is more common in those patients with concurrent disease of the bone marrow (21,23–25). Solid tumors can have a leptomeningeal component, either at presentation, as an isolated event, or associated with local disease progression. This is seen most frequently with retinoblastoma (26), neuroblastoma (27–29), rhabdomyosarcoma (30), melanoma (31,32) and Ewing’s sarcoma (33,34) and most commonly with advanced systemic disease. Leptomeningeal disease and spinal cord compression are both thought to be more treatable and have a better prognosis in children than in adults. Leptomeningeal disease can be bulky, nodular, and more focal in nature or more widespread and diffuse involvement. The most frequent mode of metastases is to the epidural or subarachnoid space or by metastatic spread to the cord parenchyma. Presenting features, therefore, reflect the pattern of involvement and may include both nonspecific and focal neurologic symptoms; these symptoms can be the presentation of underlying malignancy in children, either with or without concurrent symptoms of the underlying neoplasm (7,35). Patients may also be asymptomatic, with the diagnosis made by surveillance of CSF cytology and/or neuroimaging studies (19).

610

Part VII / Neurologic Complications of Specific Malignancies

The most common signs and symptoms of leptomeningeal disease are signs of increased intracranial pressure, including headache, nausea, and vomiting. There may also be involvement of the cranial nerves or spinal cord involvement with weakness, sensory loss, pain, ataxia, or paraparesis. The most common nontraumatic cause of paraparesis in children is spinal cord compression from tumor (36). Symptoms of spinal cord compression often include back pain, lower extremity weakness, or paresthesias and urinary retention (7,14,19,37). For leukemia, the most common neurologic symptoms overall include weakness and back pain, even without leptomeningeal infiltration (35). Lymphoma, by contrast, is more likely to show direct meningeal infiltration with spinal cord compression and/or cranial neuropathy (38). Neuroblastoma may present with spinal cord compression due to epidural spinal metastases. Diagnosis is confirmed with sampling of cerebrospinal fluid and imaging studies. A negative cytology does not preclude the diagnosis; in adults, initial CSF was positive in 55–70%, but the yield increases to 80–92% with repeated CSF sampling (23,39). The diagnosis of CSF leukemia requires definable blasts in the CSF, with a stratification based on number of leukocytes and blasts detected. For germ cell tumors, elevated tumor markers such as alpha fetoprotein and human chorionic gonadotropin may support the diagnosis of leptomeningeal metastases, along with simultaneous serum marker studies. Other potential clues include elevated opening pressure, decreased glucose, and increased protein. Imaging studies may be helpful to show bulky disease and to assess CSF flow dynamics. Treatment for spinal cord compression can include a combination of decompressive laminectomy and/or high-dose steroids followed by therapeutic irradiation and intrathecal and/or systemic chemotherapy. In children, laminectomy is avoided as much as possible, particularly in young patients, because of concerns for development of anterior subluxation, kyphosis, or scoliosis; surgery is reserved for cases with rapid neurologic deterioration (40). Children with neuroblastoma and spinal cord compression from local disease are often treated with chemotherapy alone (Fig. 2). The outlook is poor if relapse has occurred after radiation to the neuraxis (19).

Fig. 2. Sagittal post-contrast MRI of a 3-day-old infant who presented with paraparesis and was found to have extensive spinal neuroblastoma and spinal cord compression (arrows show enhancing tumor throughout thoracic and lumbar spine). He was treated with chemotherapy alone. At 15 months, he was crawling and standing with a stander device.

Chapter 32 / Neurologic Complications of Pediatric Systemic Cancer

611

3. INDIRECT AND TREATMENT-RELATED COMPLICATIONS IN CHILDREN 3.1. Paraneoplastic Disorders Paraneoplastic neurologic syndromes are rare complications of malignancy that are seen much more frequently in the adult population. Excluding the more common Lambert–Eaton myasthenic syndrome and myasthenia gravis, paraneoplastic syndromes affect approximately 0.01% of all patients with cancer (41). In children, paraneoplastic syndromes are more commonly associated with neuroblastoma as the opsoclonus– myoclonus syndrome (OMS). OMS is defined by the classical acute onset of rapid and chaotic eye movements, myoclonic jerks, and appendicular and axial ataxia. It can also be associated with behavioral changes and emotional lability. OMS is associated with underlying neuroblastoma in more than one-half of cases and occurs in 2–3% of all children with neuroblastoma (42,43). In at least half of these cases, OMS is the first presenting symptom of the neuroblastoma; the presence of OMS is often thought to portend a more favorable prognosis in terms of favorable histology for the underlying neuroblastoma (44). A thorough search for neuroblastoma is necessary in all patients with OMS. There does not appear to be a universally detectable autoantibody, but an autoimmune pathogenesis has been suspected, particularly with the favorable response to steroids and intravenous immunoglobulin (45). Although the oncologic prognosis of neuroblastoma associated with OMS is better than in non-OMS patients, acute and chronic neurologic effects are commonly observed (42,43,46). Personality changes, developmental regression with loss of speech and language, and motor deficits may remain significant. Many patients will be left with severe learning and language deficits (47). Prompt and strong immunosuppression remain the mainstay of therapy; however, very few children actually receive early therapy (48). Currently the Childrens’ Oncology Group has a phase III study for patients with OMS and neuroblastoma to determine the efficacy and outcomes with immunosuppression. Paraneoplastic limbic encephalitis (PLE) has been reported in 6 patients, all teenage girls, in association with ovarian teratoma (n = 4), Hodgkin’s lymphoma (n = 1), and small cell ovarian adenocarcinoma (n = 1) (49–54). Evaluation in these cases demonstrated mild cerebrospinal fluid pleocytosis, elevated protein, and serum/CSF antineuronal antibodies. In nearly 50% of cases, MR imaging was normal and seizures were frequent. PLE is thus rare in children, seen only in teenage girls, with a predominance of ovarian teratoma as the underlying etiology. Lastly, Guillain–Barré syndrome has been reported in relationship to underlying cancer, and has also been observed after chemotherapy and/or bone marrow transplantation (55,56).

3.2. Chemotherapy-Related Complications in Children Tumor-directed therapy is generally tailored to the underlying tumor type and location in children, as in adults. Chemotherapy-related effects in children are in many cases similar to those experienced and reported in adults. These include seizures, transient ischemic attacks, encephalopathy, ataxia, and myelopathy, as well as movement disorders. The incidence of neurotoxicity from therapy from ALL is thought to range from 5% to 18% (57,58). Moreover, several specific chemotherapeutic agents, which are used with much higher frequency in children based on underlying disease prevalence, are known to have quite specific neurologic effects. These offending agents include methotrexate, cytarabine, vincristine, asparaginase, and corticosteroids; however, because standard therapy often relies on combinations of chemotherapy, it is sometimes difficulty to isolate the offending agent. The following section will review some general complications of chemotherapy in children and will discuss several of these agents and their more common neurotoxic effects (see Table 3 for summary).

3.3. Headaches Headaches in children are a common neurologic complaint and the most common cause for neurologic consultation in children with cancer; the difficulty lies in determining the underlying etiology (Table 4) (59). Migraine is the source of headache in approximately one-third of children with cancer (59). Headaches may also occur in the context of a febrile illness or result from medications. Low-pressure headaches after lumbar puncture, with or without intrathecal chemotherapy, are quite frequent with an incidence that ranges from 8% overall (60) to as high as 50% in the adolescent population (61). For systemic cancers, brain metastases are the most common structural cause of headache, followed by infections such as abscess and meningitis, and then much less frequently

612

Part VII / Neurologic Complications of Specific Malignancies

Table 3 Commonly Used Chemotherapeutic Drugs: Frequency and Type of Neurologic Side Effects Drug Methotrexate Cytarabine Vincristine Asparaginase Corticosteroids Cisplatin/carboplatin Cyclosporine Thalidomide

Common, >10%

Uncommon, <10%

Leukoencephalopathy, especially with RT

Rare, <1%

Myelitis, arachnoiditis

Seizure

Ataxia, arachnoiditis Peripheral/autonomic neuropathy Mental status  Myopathy, tremor, behavioral  Hearing loss Tremor

Seizures Sinus thrombosis Psychosis, seizures, neuropathy Peripheral neuropathy, seizure Seizure, PRES Neuropathy

PRES = posterior reversible leukoencephalopathy syndrome

Table 4 Most Common Reasons for Neurologic Assessment in Children with Systemic Cancer Reason for Neurologic Evaluation Headache Back/neck pain Limb pain/sensory change Extremity weakness Seizure Altered mental status Visual complaints/diplopia

Complaints 34.9% 25% 12% 7.1% 7% 8.4% 6.5%

From Antunes et al. (62).

by intracranial hemorrhage (59). Structural disease should be excluded even in the absence of localizing signs, as 25% of children with systemic cancer and headache have an underlying structural lesion (62).

3.4. Seizures The second most frequent reason for neurologic consultation is seizures, which occur in children with both solid tumors and hematologic malignancies as well as children undergoing bone marrow transplantation. Seizures are more frequent in children than in adults and may result from either tumor- or treatment-related toxicity. Seizure risk may be increased in those patients who are known to have central nervous system involvement. All types of seizures may be observed in children who are treated with chemotherapy. Seizures without evidence for mass lesion are reported in 8–10% of children with leukemia and lymphoma (63,64), although it is estimated that as many as 50% of children with solid tumors who experience a seizure have an underlying structural abnormality (59). A thorough evaluation must be performed to search for intracranial lesions, infection, encephalopathy, and stroke as potential etiologies. The strongest association with seizures and chemotherapy has been reported with methotrexate, which is used to treat many pediatric cancers, including ALL, lymphoma and sarcomas. Estimates of seizure frequency suggest that 7–20% of children with ALL experience seizures at some point during the course of therapy (63,65,66). One study of survivors of ALL showed that 13% of children had experienced a seizure at some point during their treatment, and all but one reported patients experienced seizure after intrathecal methotrexate or subcutaneous asparaginase (64). Other agents known to lower seizure threshold include cisplatin and vincristine, as both of these agents can pass through the blood–brain barrier and can secondarily induce seizures secondary to electrolyte disturbances from hypocalcemia, hypomagnesemia, and hyponatremia.

Chapter 32 / Neurologic Complications of Pediatric Systemic Cancer

613

Seizure prophylaxis must be undertaken with an interdisciplinary approach to determine whether medications are needed and, if so, the specific choice. The choice of anticonvulsant may be challenging, particularly since many of the more traditional anticonvulsants, such as phenytoin, carbamazepine, and phenobarbital, affect metabolism of chemotherapeutic drugs by inducing increased activity of the cytochrome P450 system (67,68). This may result in decreased tumor therapy efficacy, decreased seizure control, and/or unexpected toxicity. Enzyme-inducing or enzyme-inhibiting seizure medications are less preferable. Third-generation anticonvulsants that do not induce the cytochrome P450 system, such as levetiracetam, may ultimately prove to be better suited for patients who require ongoing seizure prophylaxis (69).

3.5. Change in Mental Status Alteration of mental status is a relatively common occurrence in children undergoing chemotherapy and has been reported in as many as 11% of children during the course of treatment (70). Similar to the adult population, the underlying cause of mental status changes in children is toxic–metabolic encephalopathy in two-thirds of the cases and structural causes in the other one-third. Many different agents are known to cause changes level of alertness and/or somnolence in children, including asparaginase, cisplatin, ifosfamide, methotrexate (including intrathecal), cytarabine, cyclosporine, etoposide, and vincristine. Although medications may ultimately be the culprit, it is important to eliminate other potential causes, such as seizure, cerebrovascular event, infection, hemorrhage, or metabolic derangement, so as to address potentially treatable sources.

3.6. Cerebral Infarction The largest autopsy series of stroke in adults with cancer demonstrated that stroke is the second most frequent CNS lesion after metastatic disease. The most common cause of symptomatic cerebral infarction in this population was thromboembolic disease, followed by intravascular coagulation and atherosclerosis (71). In general, this population has more embolic events and less intrinsic vascular disease than most stroke series report from the general population (72). Overall prognosis is poor, regardless of etiology, after the stroke has occurred, with a median survival of 4.5 months (72). Hypercoagulable state in association with advanced malignancy and/or presence of metastatic disease is thought to increase the frequency of embolic events and also contribute to the short survival. Prognosis is influenced by type and stage of malignancy, stroke etiology, and disability following stroke. As in adults, cerebral infarction in childhood cancer patients may be ischemic or hemorrhagic. Stroke manifestations may be age-related. In the younger child, presentation may be more insidious with irritability, altered level of alertness and seizures while older children often report headaches, focal deficits such as visual change, speech impairment and weakness, and seizures (73,74). Strokes have been reported in association with impaired coagulation/disseminated intravascular coagulation, thrombocytopenia, venous sinus thrombosis secondary to l-asparaginase in children with leukemia, thrombotic endocarditis, meningitis, focal deficits related to hemorrhagic metastatic disease, and chemotherapy (75,76). Hemorrhagic strokes presenting as delayed hemorrhagic radiation vasculopathy, separate from tumor recurrence or secondary tumor and unrelated to treatment dose, have been reported after radiation alone or radiation in combination with systemic chemotherapy in children (77). One large study of children with leukemia reported a prevalence of ischemic stroke of 0.47%, all of which were sinovenous thrombosis (Fig. 3) (78). Despite a relatively high frequency of elevated white blood cell counts and decreased platelets in children with cancer, the overall rate of strokes is thought to be quite low as a result of these risk factors. Strokes from chemotherapy itself have been more commonly associated with asparaginase and methotrexate (70,79,80). In the pediatric population, neurologic outcome after sinovenous thrombosis was related to underlying clinical features such as seizures, impaired level of consciousness, coincident intracranial hemorrhage, deep venous location, lack of antithrombotic therapy, and young age, less than six years (81–84).

3.7. Motor Deficits/Neuropathy Weakness may result from spinal cord or peripheral nerve etiologies. Acute spinal cord compression is a neurologic emergency in children as in adults and must be investigated emergently for potentially reversible causes. Acute paralysis has been reported, both alone and in combination with cytarabine, with methotrexate

614

Part VII / Neurologic Complications of Specific Malignancies

Fig. 3. Magnetic resonance venogram in a 4 year old child treated with L-asparaginase for acute leukemia who experienced sagittal sinus thrombosis (arrows) and subsequent venous infarction.

(85,86). More subtle, and often progressive, weakness is common in patients with peripheral neuropathy from vincristine. Children often complain of paresthesias and painful distal paresthesias in the extremities that progress to loss of deep tendon reflexes and foot drop. Cranial nerve deficits may also be observed, particularly in patients who already have some cranial nerve dysfunction. Platinum drugs may also lead to peripheral neuropathy (59). Platinum drugs more commonly lead to high-frequency sensorineural hearing loss; ototoxic effects are related to total dose and often synergistic in patients who have also received cranial irradiation. The underlying mechanism is thought to be related to permanent damage to the mitochondria of the hair cells.

3.8. Drug-Specific Effects Methotrexate is often implicated as a major cause of acute neurologic issues. All of the major risk factors for methotrexate-induced neurotoxicity, including use in higher doses (87), intrathecal administration (88), young age at time of treatment (89,90), and concurrent use of cranial irradiation (90) are all applicable in the pediatric age group. The characteristic radiographic findings in patients with associated neurotoxicity is thought to be reversible leukoencephalopathy, which are hyperintense regions on T2-weighted MRI that are typically located in the periventricular white matter (90,91). These white matter changes often recover spontaneously, but in some cases are irreversible (92). One of the mainstays of therapy for children with acute lymphoblastic leukemia is asparaginase, which functions by depletion of downstream stores of l-asparagine and aspartic acid. Secondary toxicities from asparaginase include alteration of coagulation and hypersensitivity reactions. Reported central nervous system complications include cerebral hemorrhage and cerebral thrombosis (93). Defects in coagulation are thought to result from an imbalance of the pro- and anticoagulating systems, leading to decreases of antithrombin III, protein S, and protein C. Common presenting symptoms of sinus thrombosis, associated with secondary parenchymal hemorrhage, include headache, seizure, and increased intracranial pressure. These symptoms were all evident in this 4-year-old boy with ALL, who experienced a sagittal sinus thrombosis; additional therapy with asparaginase had to be discontinued and fractionated low-molecular weight heparin was initiated (Fig. 3). Vincristine is one of the basic components of therapy for leukemia as well as many CNS tumors, including low-grade glioma. Vincristine is thought to impair motor function by causing a peripheral neuropathy, which manifests as decreased deep tendon reflex responses, decreased peripheral motor abilities, motor clumsiness, and decreased distal sensation (94,95). Although considered reversible, these effects can last months to years after completion of therapy. Signs of motor problems have been reported up to 5 years after therapy for ALL when patients are followed with careful electromyographic studies (96).

3.9. Neurologic Effects of Radiation in Children Radiation therapy remains an important component of cancer treatment. Like many other therapies, radiation has nonspecific cytotoxic effects on the adjacent nervous system when used to treat malignancies located in close proximity to the brain and spinal cord. Although tissue thresholds have been established, individuals may have different radiation tolerances. Moreover, patients may experience different complications related to underlying age, disease status, concomitant therapies, length of survival and radiation features such as dose, size of the field

Chapter 32 / Neurologic Complications of Pediatric Systemic Cancer

615

and fractionation schema (97). Radiation injury may affect each level of the nervous system and may occur at the time of therapy or months or years after completion of therapy. These long-term effects on cognitive function and growth and development emphasize the importance of new clinical trials to examine the efficacy of reduced overall dose of therapy for diseases with high cure rates. This strategy has been used with some success for childhood leukemia to reduce the radiation dose without detrimental effects on long-term outcome. Radiation, used alone or in combination with chemotherapy, is associated with significant adverse neurologic effects, including neurocognitive deficits, attention deficit disorder, endocrine abnormalities, growth retardation, secondary neoplasms, and stroke. The two most common neurologic sequelae in children are cognitive and neuroendocrine damage. The frequency, degree of, and etiology of neurocognitive dysfunction remains incompletely elucidated. Radiotherapy has been implicated as the major cause of damage, but the relationship between radiotherapy and the type of damage caused and the volume and dose of radiotherapy and degree of cognitive damage is unclear. Cognitive deficits are progressive in nature and younger children are more likely to suffer the severest damage; but no patient of any age is free of risk of damage (98,99). In addition to core deficits in attention and concentration (100), declines of intelligence and impairment of working memory and information processing are also seen after cranial irradiation (101,102). Research into the neurocognitive treatment effects has mirrored the progress in the treatment of the medical disease. Because of concerns over neurocognitive sequelae from radiation, there is renewed interest in reduction of total radiation dose, both for solid tumors and for primary brain tumors. For example, with ALL, it is thought that preventive therapy to the CNS with intrathecal chemotherapy and/or irradiation reduces the probability of relapse; currently, however, cranial irradiation is reserved for patients with CNS relapse or high-risk disease. Other potential interventions include using cognitive remediation as an educational strategy to improve preparedness and on task performance (103). In addition, the use of stimulant medications to improve cognitive performance is now being explored (104). Growth hormone impairment is the most common form of neuroendocrinologic dysfunction after cranial irradiation (105,106). Radiation to the brain can lead to dysregulation of the hypothalamic-pituitary axis, typically affecting growth hormonal status. Higher doses of radiation can lead to more widespread effects. Early puberty is now recognized and is often treated with gonadotrophin-releasing hormone analogs in order to maximize final height. Both radiation to the spine and adjuvant chemotherapy may lead to long-term consequences including decreased fertility and premature ovarian failure (107). There is increasing evidence that children with cancer who are long-term survivors are at increased risk for the development of secondary CNS and non-CNS tumors (108–110). The increased incidence of secondary tumors is now becoming more clear as the number of long-term survivors increases; in some cases, the risk is thought to be increased by dose of radiation (109).

3.10. Stem Cell Transplantation-Related Complications Transplant-related complications have been well-characterized in adults, but very little is reported in pediatric series. Hematopoietic stem cell transplantation is now used as the main treatment for a large number of heritable and acquired diseases, including leukemia, lymphoma, histiocytosis, and myelodysplastic syndrome, as well as refractory anemias. Neurologic events are now appreciated as an emerging transplant-related toxicity. In the largest series of 272 consecutive children undergoing allogeneic or autologous hematopoietic stem cell transplant for both hematologic or non-hematologic diseases, 13.6% developed severe neurologic events (111). These events typically occurred within the first year, with a median of 90 days after transplant. The most frequent complications were related to neurotoxicity from cyclosporine A. This was followed by what was thought to be irreversible late effects from irradiation and reversible and irreversible effects from chemotherapy. Other common complications from transplant included CNS infections, metabolic encephalopathies, cerebrovascular events, and peripheral neuropathy (111–113). Risk factors included transplant from allogeneic donor, particularly if unrelated, development of severe graft-versus-host disease, and the use of total body irradiation as a transplant regimen (Table 5). Underlying disease type was not an independent risk factor in the development of post-transplant complications. In another large series, which included children and adults, neurologic complications were present in > 50% of patients; moreover, overall survival was significantly worse in patients with major neurologic issues (112). Of the group with severe neurologic sequelae, mortality ranged from 6% to 30% (111,113).

616

Part VII / Neurologic Complications of Specific Malignancies

Table 5 Risk Factors for Neurologic Complications After Stem-Cell Transplantation (111,112) Total body irradiation Allogeneic donor (particularly if unrelated donor) Acute graft-versus-host disease grade > 2 Drug-related neurotoxicity

4. CONCLUSIONS Therapy for cancer can lead to long-lasting neurologic toxicity, particularly in the setting of the developing nervous system. Many of the complications of cancer and the therapy for cancer in children are often not appreciated until many years after the completion of therapy. The precise mechanism by which neurotoxicity in childhood or young adulthood translates to later functioning is not yet clearly understood, but some neurologic complications ultimately may be long-lasting or permanent. This emphasizes the importance of long-term followup into adulthood for survivors of childhood cancer. Future cooperative group studies should include systematic neurologic assessment to better determine the incidence of neurologic sequelae and therefore provide a means to prevent and/or ameliorate these symptoms.

REFERENCES 1. Hamilton BE, Minino AM, Martin JA et al. Annual summary of vital statistics: 2005. Pediatrics 2007;119:345–360. 2. Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov). SEER*Stat Database: Incidence: Survival Mono, SEER9/RGA/SJM/LA, Nov 2004 Sub (1988–2001), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch, released December 2004. 3. Cancer survivorship: United States, 1971–2001. MMWR Morb Mortal Wkly Rep 2004;53:526–529. 4. Clouston PD, DeAngelis LM, Posner JB. The spectrum of neurologic disease in patients with systemic cancer. Ann Neurol 1992;31: 268–273. 5. Gilbert MR, Grossman SA. Incidence and nature of neurologic problems in patients with solid tumors. Am J Med 1986;81:951–954. 6. Patchell RA, Posner JB. Neurologic complications of systemic cancer. Neurol Clin 1985;3:729–750. 7. Weyl-Ben Arush M, Stein M, Perez-Nachum M et al. Neurologic complications in pediatric solid tumors. Oncology 1995;52:89–92. 8. Tasdemiroglu E, Patchell RA, Kryscio R. Neurologic complications of childhood malignancies. Acta Neurochir (Wien) 1999;141: 1313–1321. 9. Baram TZ, van Tassel P, Jaffe NA. Brain metastases in osteosarcoma: incidence, clinical and neuroradiological findings and management options. J Neurooncol 1988;6:47–52. 10. Graus F, Walker RW, Allen JC. Brain metastases in children. J Pediatr 1983;103:558–561. 11. Vannucci RC, Baten M. Cerebral metastatic disease in childhood. Neurology 1974;24:981–985. 12. Boring CC, Squires TS, Tong T, Montgomery S. Cancer statistics, 1994. CA Cancer J Clin 1994;44:7–26. 13. Subramanian A, Harris A, Piggott K et al. Metastasis to and from the central nervous system: the “relatively protected site.” Lancet Oncol 2002;3:498–507. 14. Bouffet E, Marec-Berard P, Thiesse P et al. Spinal cord compression by secondary epi- and intradural metastases in childhood. Childs Nerv Syst 1997;13:383–387. 15. Bouffet E, Doumi N, Thiesse P et al. Brain metastases in children with solid tumors. Cancer 1997;79:403–410. 16. Deutsch M, Orlando S, Wollman M. Radiotherapy for metastases to the brain in children. Med Pediatr Oncol 2002;39:60–62. 17. Kebudi R, Ayan I, Gorgun O, Agaoglu FY, Vural S, Darendeliler E. Brain metastasis in pediatric extracranial solid tumors: survey and literature review. J Neuro-oncol 2005;71:43–48. 18. Tasdemiroglu E, Patchell RA. Cerebral metastases in childhood malignancies. Acta Neurochir (Wien) 1997;139:182–187. 19. Neville KA, Blaney SM. Leptomeningeal cancer in the pediatric patient. Cancer Treat Res 2005;125:87–106. 20. Bleyer WA, Byrne TN. Leptomeningeal cancer in leukemia and solid tumors. Curr Probl Cancer 1988;12:181–238. 21. Chamberlain MC, Nolan C, Abrey LE. Leukemic and lymphomatous meningitis: incidence, prognosis, and treatment. J Neuro-oncol 2005;75:71–83. 22. Evans AE, Gilbert ES, Zandstra R. The increasing incidence of central nervous system leukemia in children. Children’s Cancer Study Group A. Cancer 1970;26:404–409. 23. Kaplan JG, DeSouza TG, Farkash A et al. Leptomeningeal metastases: comparison of clinical features and laboratory data of solid tumors, lymphomas and leukemias. J Neuro-oncol 1990;9:225–229. 24. Giglio P, Gilbert MR. Neurologic complications of non–Hodgkin’s lymphoma. Curr Oncol Rep 2005;7:61–65. 25. Glass J. Neurologic complications of lymphoma and leukemia. Semin Oncol 2006;33:342–347. 26. Meli FJ, Boccaleri CA, Manzitti J et al. Meningeal dissemination of retinoblastoma: CT findings in eight patients. AJNR Am J Neuroradiol 1990;11:983–986.

Chapter 32 / Neurologic Complications of Pediatric Systemic Cancer

617

27. Blatt J, Fitz C, Mirro J, Jr. Recognition of central nervous system metastases in children with metastatic primary extracranial neuroblastoma. Pediatr Hematol Oncol 1997;14:233–241. 28. Kellie SJ, Hayes FA, Bowman L et al. Primary extracranial neuroblastoma with central nervous system metastases characterization by clinicopathologic findings and neuroimaging. Cancer 1991;68:1999–2006. 29. Shaw PJ, Eden T. Neuroblastoma with intracranial involvement: an ENSG Study. Med Pediatr Oncol 1992;20:149–155. 30. Parasuraman S, Langston J, Rao BN et al. Brain metastases in pediatric Ewing sarcoma and rhabdomyosarcoma: the St. Jude Children’s Research Hospital experience. J Pediatr Hematol Oncol 1999;21:370–377. 31. Rodriguez-Galindo C, Pappo AS, Kaste SC et al. Brain metastases in children with melanoma. Cancer 1997;79:2440–2445. 32. Allcutt D, Michowiz S, Weitzman S et al. Primary leptomeningeal melanoma: an unusually aggressive tumor in childhood. Neurosurgery 1993;32:721–729; discussion 9. 33. Yu L, Craver R, Baliga M et al. Isolated CNS involvement in Ewing’s sarcoma. Med Pediatr Oncol 1990;18:354–358. 34. Trigg ME, Makuch R, Glaubiger D. Actuarial risk of isolated CNS involvement in Ewing’s sarcoma following prophylactic cranial irradiation and intrathecal methotrexate. Int J Radiat Oncol Biol Phys 1985;11:699–702. 35. Aysun S, Topcu M, Gunay M et al. Neurologic features as initial presentations of childhood malignancies. Pediatr Neurol 1994;10: 40–43. 36. Raffel C. Spinal cord compression by epidural tumors in childhood. Neurosurg Clin N Am 1992;3:925–930. 37. Huang LT, Hsiao CC, Weng HH et al. Neurologic complications of pediatric systemic malignancies. J Formos Med Assoc 1996;95: 209–212. 38. Tomita N, Kodama F, Sakai R et al. Predictive factors for central nervous system involvement in non–Hodgkin’s lymphoma: significance of very high serum LDH concentrations. Leuk Lymphoma 2000;38:335–343. 39. Wasserstrom WR, Glass JP, Posner JB. Diagnosis and treatment of leptomeningeal metastases from solid tumors: experience with 90 patients. Cancer 1982;49:759–772. 40. Raffel C, Neave VC, Lavine S et al. Treatment of spinal cord compression by epidural malignancy in childhood. Neurosurgery 1991;28:349–352. 41. Darnell RB, Posner JB. Paraneoplastic syndromes involving the nervous system. N Engl J Med 2003;349:1543–1554. 42. Pranzatelli MR. The neurobiology of the opsoclonus–myoclonus syndrome. Clin Neuropharmacol 1992;15:186–228. 43. Rudnick E, Khakoo Y, Antunes NL et al. Opsoclonus–myoclonus–ataxia syndrome in neuroblastoma: clinical outcome and antineuronal antibodies: a report from the Children’s Cancer Group Study. Med Pediatr Oncol 2001;36:612–622. 44. Altman AJ, Baehner RL. Favorable prognosis for survival in children with coincident opso–myoclonus and neuroblastoma. Cancer 1976;37:846–852. 45. Pranzatelli MR. The immunopharmacology of the opsoclonus–myoclonus syndrome. Clin Neuropharmacol 1996;19:1–47. 46. Russo C, Cohn SL, Petruzzi MJ et al. Long-term neurologic outcome in children with opsoclonus–myoclonus associated with neuroblastoma: a report from the Pediatric Oncology Group. Med Pediatr Oncol 1997;28:284–288. 47. Mitchell WG, Davalos-Gonzalez Y, Brumm VL et al. Opsoclonus–ataxia caused by childhood neuroblastoma: developmental and neurologic sequelae. Pediatrics 2002;109:86–98. 48. Tate ED, Allison TJ, Pranzatelli MR et al. Neuroepidemiologic trends in 105 U.S. cases of pediatric opsoclonus–myoclonus syndrome. J Pediatr Oncol Nurs 2005;22:8–19. 49. Aydiner A, Gurvit H, Baral I. Paraneoplastic limbic encephalitis with immature ovarian teratoma: a case report. J Neurooncol 1998;37:63–66. 50. Lee AC, Ou Y, Lee WK et al. Paraneoplastic limbic encephalitis masquerading as chronic behavioural disturbance in an adolescent girl. Acta Paediatr 2003;92:506–509. 51. Nokura K, Yamamoto H, Okawara Y et al. Reversible limbic encephalitis caused by ovarian teratoma. Acta Neurol Scand 1997;95: 367–373. 52. Okamura H, Oomori N, Uchitomi Y. An acutely confused 15-year-old girl. Lancet 1997;350:488. 53. Taylor RB, Mason W, Kong K et al. Reversible paraneoplastic encephalomyelitis associated with a benign ovarian teratoma. Can J Neurol Sci 1999;26:317–320. 54. Stein-Wexler R, Wootton-Gorges SL, Greco CM et al. Paraneoplastic limbic encephalitis in a teenage girl with an immature ovarian teratoma. Pediatr Radiol 2005;35:694–697. 55. Adams C, August CS, Maguire H et al. Neuromuscular complications of bone marrow transplantation. Pediatr Neurol 1995;12:58–61. 56. Re D, Schwenk A, Hegener P et al. Guillain–Barré syndrome in a patient with non–Hodgkin’s lymphoma. Ann Oncol 2000;11:217–220. 57. Shuper A, Stark B, Kornreich L et al. Methotrexate treatment protocols and the central nervous system: significant cure with significant neurotoxicity. J Child Neurol 2000;15:573–580. 58. Mahoney DH, Jr., Shuster JJ, Nitschke R et al. Acute neurotoxicity in children with B-precursor acute lymphoid leukemia: an association with intermediate-dose intravenous methotrexate and intrathecal triple therapy: a Pediatric Oncology Group study. J Clin Oncol 1998;16:1712–1722. 59. Antunes NL. Acute neurologic complications in children with systemic cancer. J Child Neurol 2000;15:705–716. 60. Ramamoorthy C, Geiduschek JM, Bratton SL et al. Postdural puncture headache in pediatric oncology patients. Clin Pediatr (Phila) 1998;37:247–251. 61. Wee LH, Lam F, Cranston AJ. The incidence of post-dural puncture headache in children. Anaesthesia 1996;51:1164–1166. 62. Antunes NL, De Angelis LM. Neurologic consultations in children with systemic cancer. Pediatr Neurol 1999;20:121–124. 63. Ochs JJ, Bowman WP, Pui CH et al. Seizures in childhood lymphoblastic leukaemia patients. Lancet 1984;2:1422–1424. 64. Maytal J, Grossman R, Yusuf FH et al. Prognosis and treatment of seizures in children with acute lymphoblastic leukemia. Epilepsia 1995;36:831–836.

618

Part VII / Neurologic Complications of Specific Malignancies

65. Mahoney DH, Jr., Shuster JJ, Nitschke R et al. Intensification with intermediate-dose intravenous methotrexate is effective therapy for children with lower-risk B-precursor acute lymphoblastic leukemia: a Pediatric Oncology Group study. J Clin Oncol 2000;18: 1285–1294. 66. Winick NJ, Bowman WP, Kamen BA et al. Unexpected acute neurologic toxicity in the treatment of children with acute lymphoblastic leukemia. J Natl Cancer Inst 1992;84:252–256. 67. Zamboni WC, Gajjar AJ, Heideman RL et al. Phenytoin alters the disposition of topotecan and N-desmethyl topotecan in a patient with medulloblastoma. Clin Cancer Res 1998;4:783–789. 68. Villikka K, Kivisto KT, Maenpaa H et al. Cytochrome P450–inducing antiepileptics increase the clearance of vincristine in patients with brain tumors. Clin Pharmacol Ther 1999;66:589–593. 69. Tibussek D, Distelmaier F, Schonberger S et al. Antiepileptic treatment in paediatric oncology: an interdisciplinary challenge. Klin Padiatr 2006;218:340–349. 70. DiMario FJ, Jr., Packer RJ. Acute mental status changes in children with systemic cancer. Pediatrics 1990;85:353–360. 71. Graus F, Rogers LR, Posner JB. Cerebrovascular complications in patients with cancer. Medicine (Baltimore) 1985;64:16–35. 72. Cestari DM, Weine DM, Panageas KS et al. Stroke in patients with cancer: incidence and etiology. Neurology 2004;62:2025–2030. 73. Lynch JK, Hirtz DG, DeVeber G et al. Report of the National Institute of Neurologic Disorders and Stroke workshop on perinatal and childhood stroke. Pediatrics 2002;109:116–123. 74. deVeber G, Andrew M, Adams C et al. Cerebral sinovenous thrombosis in children. N Engl J Med 2001;345:417–423. 75. Reddingius RE, Patte C, Couanet D et al. Dural sinus thrombosis in children with cancer. Med Pediatr Oncol 1997;29:296–302. 76. Packer RJ, Rorke LB, Lange BJ et al. Cerebrovascular accidents in children with cancer. Pediatrics 1985;76:194–201. 77. Poussaint TY, Siffert J, Barnes PD et al. Hemorrhagic vasculopathy after treatment of central nervous system neoplasia in childhood: diagnosis and follow-up. AJNR 1995;16:693–699. 78. Santoro N, Giordano P, Del Vecchio GC et al. Ischemic stroke in children treated for acute lymphoblastic leukemia: a retrospective study. J Pediatr Hematol Oncol 2005;27:153–157. 79. Fleischhack G, Solymosi L, Reiter A et al. [Imaging methods in diagnosis of cerebrovascular complications with l-asparaginase therapy]. Klin Padiatr 1994;206:334–341. 80. Ott N, Ramsay NK, Priest JR et al. Sequelae of thrombotic or hemorrhagic complications following l-asparaginase therapy for childhood lymphoblastic leukemia. Am J Pediatr Hematol Oncol 1988;10:191–195. 81. von Mering M, Stiefel M, Brockmann K et al. Deep cerebral venous sinus thrombosis often presents with neuropsychologic symptoms. J Clin Neurosci 2003;10:310–312. 82. Buccino G, Scoditti U, Patteri I et al. Neurologic and cognitive long-term outcome in patients with cerebral venous sinus thrombosis. Acta Neurol Scand 2003;107:330–335. 83. de Bruijn SF, de Haan RJ, Stam J. Clinical features and prognostic factors of cerebral venous sinus thrombosis in a prospective series of 59 patients: for The Cerebral Venous Sinus Thrombosis Study Group. J Neurol Neurosurg Psychiatry 2001;70:105–108. 84. Lanthier S, Carmant L, David M et al. Stroke in children: the coexistence of multiple risk factors predicts poor outcome. Neurology 2000;54:371–378. 85. Dunkelman H, Earl HM, Twelves C. Acute reversible neurologic deficit following intrathecal chemotherapy. Cancer Chemother Pharmacol 1991;27:329–330. 86. Watterson J, Toogood I, Nieder M et al. Excessive spinal cord toxicity from intensive central nervous system–directed therapies. Cancer 1994;74:3034–3041. 87. Jaffe N, Takaue Y, Anzai T et al. Transient neurologic disturbances induced by high-dose methotrexate treatment. Cancer 1985;56:1356–1360. 88. Gowan GM, Herrington JD, Simonetta AB. Methotrexate-induced toxic leukoencephalopathy. Pharmacotherapy 2002;22:1183–1187. 89. Chessells JM, Cox TC, Kendall B et al. Neurotoxicity in lymphoblastic leukaemia: comparison of oral and intramuscular methotrexate and two doses of radiation. Arch Dis Child 1990;65:416–422. 90. Matsumoto K, Takahashi S, Sato A et al. Leukoencephalopathy in childhood hematopoietic neoplasm caused by moderate-dose methotrexate and prophylactic cranial radiotherapy: an MR analysis. Int J Radiat Oncol Biol Phys 1995;32:913–918. 91. Asato R, Akiyama Y, Ito M et al. Nuclear magnetic resonance abnormalities of the cerebral white matter in children with acute lymphoblastic leukemia and malignant lymphoma during and after central nervous system prophylactic treatment with intrathecal methotrexate. Cancer 1992;70:1997–2004. 92. Antunes NL, Small TN, George D ey al. Posterior leukoencephalopathy syndrome may not be reversible. Pediatr Neurol 1999;20: 241–243. 93. Kieslich M, Porto L, Lanfermann H et al. Cerebrovascular complications of l-asparaginase in the therapy of acute lymphoblastic leukemia. J Pediatr Hematol Oncol 2003;25:484–487. 94. Casey EB, Jellife AM, Le Quesne PM et al. Vincristine neuropathy: clinical and electrophysiological observations. Brain 1973;96: 69–86. 95. Harila-Saari AH, Huuskonen UE, Tolonen U et al. Motor nervous pathway function is impaired after treatment of childhood acute lymphoblastic leukemia: a study with motor evoked potentials. Med Pediatr Oncol 2001;36:345–351. 96. Lehtinen SS, Huuskonen UE, Harila-Saari AH, et al. Motor nervous system impairment persists in long-term survivors of childhood acute lymphoblastic leukemia. Cancer 2002;94:2466–2473. 97. Cross NE, Glantz MJ. Neurologic complications of radiation therapy. Neurol Clin 2003;21:249–277. 98. Butler RW, Haser JK. Neurocognitive effects of treatment for childhood cancer. Ment Retard Dev Disabil Res Rev 2006;12:184–191. 99. Mulhern RK, Butler RW. Neurocognitive sequelae of childhood cancers and their treatment. Pediatr Rehabil 2004;7:1–14; discussion 5–6. 100. Rodgers J, Horrocks J, Britton PG et al. Attentional ability among survivors of leukaemia. Arch Dis Child 1999;80:318–323.

Chapter 32 / Neurologic Complications of Pediatric Systemic Cancer

619

101. Fletcher JM, Copeland DR. Neurobehavioral effects of central nervous system prophylactic treatment of cancer in children. J Clin Exp Neuropsychol 1988;10:495–537. 102. Schatz J, Kramer JH, Ablin A et al. Processing speed, working memory, and IQ: a developmental model of cognitive deficits following cranial radiation therapy. Neuropsychology 2000;14:189–200. 103. Butler RW. Attentional processes and their remediation in childhood cancer. Med Pediatr Oncol 1998;Suppl 1:75–78. 104. Mulhern RK, Khan RB, Kaplan S et al. Short-term efficacy of methylphenidate: a randomized, double-blind, placebo-controlled trial among survivors of childhood cancer. J Clin Oncol 2004;22:4795–4803. 105. Gleeson HK, Shalet SM. The impact of cancer therapy on the endocrine system in survivors of childhood brain tumours. Endocr Relat Cancer 2004;11:589–602. 106. Cohen LE. Endocrine late effects of cancer treatment. Endocrinol Metab Clin North Am 2005;34:769–789, xi. 107. Brougham MF, Wallace WH. Subfertility in children and young people treated for solid and haematological malignancies. Br J Haematol 2005;131:143–155. 108. Caglar K, Varan A, Akyuz C et al. Second neoplasms in pediatric patients treated for cancer: a center’s 30-year experience. J Pediatr Hematol Oncol 2006;28:374–378. 109. Haddy N, Le Deley MC, Samand A et al. Role of radiotherapy and chemotherapy in the risk of secondary leukaemia after a solid tumour in childhood. Eur J Cancer 2006;42:2757–2764. 110. Lin HM, Teitell MA. Second malignancy after treatment of pediatric Hodgkin disease. J Pediatr Hematol Oncol 2005;27:28–36. 111. Faraci M, Lanino E, Dini G et al. Severe neurologic complications after hematopoietic stem cell transplantation in children. Neurology 2002;59:1895–1904. 112. Antonini G, Ceschin V, Morino S et al. Early neurologic complications following allogeneic bone marrow transplant for leukemia: a prospective study. Neurology 1998;50:1441–1445. 113. Denier C, Bourhis JH, Lacroix C et al. Spectrum and prognosis of neurologic complications after hematopoietic transplantation. Neurology 2006;67:1990–1997. 114. Paulino AC, Nguyen TX, Barker JL, Jr. Brain metastasis in children with sarcoma, neuroblastoma, and Wilms’ tumor. Int J Radiat Oncol Biol Phys 2003;57:177–183.

Index A Acanthamoeba meningoencephalitis, 343 Acoustic nerve dysfunction, 275 Acute encephalopathy, 261–262 Acute lymphocytic leukemia (ALL), 555 Acute myelogenous leukemia (AML), 555 Acute necrotizing myopathy, 250 Adjuvants, 117, 124–125 AEDs. See Antiepileptic drugs (AEDs) AIDP. See Guillain–Barré syndrome (GBS) Alemtuzumab, 364 Aminoglutethimide, 308 Analgesics: pharmacological treatment for cancer pain oral route, 119–121 parenteral route, 121 transdermal route, 122 transmucosal/sublingual route, 122 Anastrozole, 309 Anticonvulsant prophylaxis, 40–41 in patients with brain tumors, 40–41 prescribing prophylactic anticonvulsants, AAN guidelines, 40 prophylactic anticonvulsants in primary and metastatic brain tumors, 41 Anticonvulsant therapy, 34, 40, 41 enzyme-inducing and nonenzyme-inducing, 38 side effects, 37 Antiepileptic drugs (AEDs), 133, 368, 402 Arterial occlusions, 223 Aseptic meningitis syndrome in immunocompromised hosts, causes of, 358 Aspergillus, 369–371 axial MRI, 342 multiple ring-enhancing lesions, 370 Aspergillus fumigatus, 341 Autologous transplantation, 329 B Back pain, 4–5 Benign Plasma Cell Dyscrasias IgM-MGUS, 593 Bevacizumab, 311 Big Ten, 359 Bladder cancer, 468–469 brain metastases, 469 complications of treatment, 470

leptomeningeal disease, 469 paraneoplastic syndromes, 469–470 peripheral nerve compression, 470 spinal disease, 470 Bleeding diathesis/hemorrhage hyperleukocytic syndrome, 225 primary fibrinolysis, 225 thrombocytopenia, 225–226 Blood–Brain barrier (BBB), 53 anatomic features for normal function of, 385 and edemagenesis, 382–383 BM. See Brain metastases (BM) Bone marrow and hematopoietic stem cells, 329 transplantation, 230 Bone pain, 117, 125 Brachial plexopathy, 276 early-delayed, 276 ischemic late-delayed, 278 late-delayed progressive, 276–277 Brachial plexus, metastases to imaging, 208 presentation, 208 treatment, 209 Brain edema, 382 categories, 383 emergency therapy and increased intracranial pressure, 388 Brain metastases (BM), 8–9, 34, 427–429 and breast cancer, 136 cranial dural metastasis, 433–435 epidemiology, 131 imaging features and delayed imaging, 21–22 diffusion-weighted imaging (DWI), 23 double- and triple-dose contrast, 21 evaluation of patients with newly diagnosed brain masses, 24–25 features of patients, in absence of known cancer, 25 higher field MRI, 23 imaging of patients, 25 magnetic resonance spectroscopy (MRS), 23 magnetization transfer (MT), 22 malignant gliomas, 19 melanoma, unique imaging features of, 18 metastases from SCLC, 21 metastatic melanoma, 18 621

622 metastatic ovarian adenocarcinoma, 16 newer contrast agents, 22 nuclear medicine techniques, 24 open ring sign, 21 parenchymal brain hemorrhage, 20 perfusion-weighted imaging (PWI), 23 pitfalls in imaging newly detected brain masses, 17–21 restricted diffusion, 20 strategies to improve detection of metastases by MRI, 21–23 venous infarcts, 22 intramedullary spinal cord metastasis, 430 leptomeningeal metastasis, 430–433 neurologic complications of tumor, 451–452 orbital and ocular metastasis, 437 pathophysiology, 132 pituitary metastasis, 429 presentation and diagnosis, 132 prevention of, 132–133 prophylactic treatment recommendations, 141 RPA class, 141–142, 142 skull and skull base metastasis, 437 spinal epidural metastasis, 435–436 symptomatic, treatment, 133–134 cerebral edema, 134 seizures, 133–134 symptoms of, 8 treatment algorithm for, 141 with corticosteroids/dexamethasone, 8–9 definitive therapy, 134–141 Brain tumor, 34, 40, 50 anticonvulsant prophylaxis in patients with, 40–41 headache, 58 headaches in children with, 60 increased risk of headache in patients with, 59 pathophysiology of headache related to, 59 Breakthrough pain, 124 Breast cancer brachial plexopathy, 437–438 etiology, 423 lumbosacral plexopathy, 437–439 neurologic complications of treatment, 441 nonmetastatic complications, 439–441 peripheral neuropathy, 441–443 metastatic complications, 427–437 neurologic complications of, 423–426 neurologic symptoms and treatment, 424–425 treatment of metastases from, 426 Buprenorphine, 122 C Cachexia-associated neuropathy, 7 cause of, 7 treating/preventing, 7

Index Calvarial metastases, 434 epidemiology, 146 imaging, 147–148 metastatic breast carcinoma, 147 pathophysiology, 146 presentation, 146 treatment, 148–149 Cancer pain adjuvants, 124–125 assessment, 116–117 breakthrough pain, 124 interventional procedures cordotomy and pituitary ablation, 126 spinal route, 125–126 sympathetic blocks, 126–127 pain mechanisms neuropathic pain, 116 nociceptive pain, 115–116 pain syndromes associated with oncologic treatments, 114 treatment high-potency opioid analgesics, 120 low-potency opioid analgesics, 119 non-opioid analgesics, 118 opioid-related adverse effects, 122–123 opioid switching, 123–124 pharmacological treatment, 117–122 tumor-related pain syndromes, 114 Candida species, 371 Centers for Disease Control (CDC), 343 Central nervous system (CNS) dysfunction, 332 infections bacterial infections, 339–340 fungal infections, 341–342 MRI characteristics of focal, 339 tumors of, 496 Central pontine myelinolysis (CPM), 335 axial MRI scan, 336 versus EPM, 335 Cerebellar degeneration, 239 Cerebral edema, 53 Cerebral hemorrhage, 8 Cerebral Herniation, emergency treatment of, 388 Cerebral metastases, 60 Cerebral symptoms and signs, 183 Cerebral venous thrombosis, 8 Cerebrospinal fluid (CSF), 181 Cerebrovascular complications cerebral hemorrhage, 8 cerebral venous thrombosis, 8 ischemic stroke, 8 Cerebrovascular complications of cancer sequelae of cancer diagnostic tests and treatment, 226–230

Index stroke, central nervous system tumor glioblastoma multiforme with intratumoral hemorrhage, 217 intratumoral parenchymal hemorrhage, 216–217 neoplastic infiltration of vessels, 218–220 pituitary apoplexy, 221 subdural hemorrhage, 217–218 tumor embolus, 220–221 stroke, remote effects of tumor: hyperand hypocoagulopathies arterial occlusions, 223 bleeding diathesis/hemorrhage, 225–226 combined hypercoaguability/bleeding diathesis, 224–225 hypercoagulability and thrombosis, 221–222 mucin-positive adenocarcinoma-associated hypercoaguability, 223–224 venous occlusions, 222–223 traditional stroke mechanisms appearing in setting of tumor, 226 Cerebrovascular disease, 217 Cerebrovascular disorders, 338 intracranial hemorrhages, 338 ischemic strokes, 338–339 Cervical cancer, 451 Cervical plexus, metastases to imaging, 207 presentation, 207 treatment, 207 Cetuximab, 312 Chemotherapy, 98–99, 139–141, 287 biologic agents, 310–311 bone marrow transplantation, 230 growth factors, 311 hormonal therapy, 308–310 hypercoagulability and thrombocytopenia, 229 anticoagulation-induced hemorrhage in cancer patient, 230 cardiomyopathy, 230 infection and stroke, 230 other agents, 229–330 lung cancer, 405–407 monoclonal antibodies, 311–312 neurotoxicity causing drugs cisplatin, 288–289 ifosfamide, 294–295 methotrexate, 295 oxaliplatin, 299 taxanes, 299 thalidomide, 299–300 vinca alkaloids, 300 neurotoxicity less frequently causing drugs anthracycline antibiotics, 302–308 neurotoxicity occasionally causing drugs 5-fluorouracil, 300–301

623 l-asparaginase, 301 pentostatin, 301–302 other agents amifostine, 313 denileukin difitox, 314 pamidronate, 314 PT-523, 314 zoledronic acid, 315 small molecule inhibitors bortezomib, 313 gefitinib, 313 imatinib, 313 sorafenib, 313 sunitinib, 313 temsirolimus, 313 tipifarnib, 313 Chemotherapy-related complications in children, 611 cerebral infarction, 613 change in mental status, 613 drug-specific effects, 614 headaches, 611–612 motor deficits/neuropathy, 613–614 neurologic effects of radiation in children, 614–615 seizures, 612–613 stem cell transplantation-related complications, 615–616 risk factors for neurologic complications after stem-cell transplantation, 616 Children, cancer and. See Pediatric systemic cancer Chloroma, 557 Chondrosarcoma, 496–497 intracranial chondrosarcoma, 497 Choriocarcinoma, 452 Chronic lymphocytic leukemia (CLL), 555–556 Chronic myelogenous leukemia (CML), 556–557 Cidofovir, 374 Cisplatin cranial neuropathies, 291–292 cytosine arabinoside, 292–294 for head and neck cancers, 513 other complications, 292 spinal cord involvement (L’hermitte’s sign), 292 Cisplatin, 6 Clinical neuro-oncologic usage of steroids imaging, 50 CNS. See Central nervous system (CNS) CNS infections, 339 bacterial meningitis, 369 in cancer patients, 355–356 challenges in diagnosing, 356–359 clinical syndrome and MRI appearance, 357 data acquisition in diagnosing, 356–357 clinical syndromes, 356 epidemiologic clues, 357 laboratory tests, 356 noninfectious mimes, 356

624 fungal infections Aspergillus, 369–371 Candida species, 371 Cryptococcus, 371 Zygomycetes, 371–372 general medical management issues, 368–369 high-risk patient groups, 358 HSCT and solid organ transplant recipients, 358–366 neurosurgical patients, 366–368 manifestations and management, 368–369 bacterial meningitis, 369 fungal infections, 369 MRI characteristics in HSCT patients, 340 patient with clinical challenges, 356–358 diagnosis: four steps, 355–358 symptoms, 356 treatment of common, in cancer patients, 367 viral infections, 372 PML, 372–374 Varicella Zoster Virus, 372 Cognitive dysfunction fatigue in cancer patients, 104–106 importance of neuropsychological assessment, 93–94 and leukoencephalopathy radiation-induced dementia, 267–268 radiation-induced mild to moderate cognitive impairment, 267 mood and psychiatric disorders in cancer patients, 101–104 neurocognitive functioning in patients with brain tumors, 92–93 neurocognitive side effects of cancer treatment, 94–101 chemotherapy, 98–99 effects of stereotactic radiosurgery or with whole-brain radiation, 97–98 interaction between radiotherapy and chemotherapy, 99–100 radiation therapy, 95–97 role of disease progression, 100 surgery, 94–95 therapies to improve or preserve neurocognition, 100–101 Colony-stimulating factors, growth factors, 311 Colorectal cancer, 484–485 Combined hypercoaguability/bleeding diathesis, 224–225 Computed tomography, 17 Confusion assessment method (CAM), 74 Corticosteroids, 170, 308 and memory impairment, 386 in neuro-oncology cerebral edema, 53–54 drug interaction, 53 imaging and usage of steroids, 50

Index novel anti-edema agents, 54 toxicity and usage of steroids, 50–52 vasogenic edema and dexamethasone, 54 Corticotropin-releasing factor (CRF), 54 Cranial nerve symptoms and signs, 183, 184 Cryoglobulinemia, 604 Cryptococcus, 371 Cryptococcus neoformans, 371 CT, diagnosis of ESCC, 412 Cytotoxic edema, 53 D Danazol, 308 Delayed radiation myelopathy (DRM), 272–273 Delirium, 65 clinical features description, 69–70 epidemiological aspects, 69 other neurologic signs, 70 defined, 66 delirium in elderly, 79 delirium tremens, 79 diagnostic criteria and classifications, 71–72 DSM IIIR criteria, 71 DSM IV criteria, 72 ICD-10 criteria, 72 diagnostic tools and instruments for assessing, 73–74 differential diagnosis, 72–73 etiology and risk factors causes of delirium in cancer patients, 76 clinical situations in which delirium occurs, 78 diagnostic tests, 75 examples of drugs associated with reports of delirium or confusion, 77 exogenous intoxications and alcohol or drug withdrawal, 77 primary CNS diseases, 75–76 risk and precipitating factors, 78 systemic diseases with CNS effects, 76 history and terminology, 66–67 pathophysiology, 67–69 postoperative delirium, 79 prognosis, 84–85 terminal delirium, 79 treatment environmental interventions, 80 etiological interventions, 79–80 pharmacological interventions, 80–84 summary of therapy guidelines, 84 Delirium rating scale (DRS), 74 Depression, 92, 93, 94, 98–99, 101–102, 103–106 Dermatomyositis (DM), 250 Dexamethasone, 48, 54, 62 use of, patient with recurrent glioblastoma, 49

Index Diffusion, 18–19, 20 Diffusion-Weighted Imaging (DWI), 23 Disseminated intravascular coagulation (DIC), 224 DRM. See Delayed radiation myelopathy (DRM) Dropped head syndrome, 276 DSM III-R, 71 DSM IV, 71 Dural Lymphoma, 576–577 Dural metastases epidemiology, 157 imaging, 157–158 MR imaging, 158 pathophysiology, 157 presentation, 157 Dural Sinus Thrombosis (DST), 578–579 DWI. See Diffusion-weighted imaging (DWI) Dysproteinemia polyneuropathy syndromes, features of, 593 E EEG. See Electroencephalogram (EEG) Efaproxiral, 139 Electroencephalogram (EEG), 74 Encephalomyelitis, 27–28 paraneoplastic encephalitis, 28 Encephalopathy, 5, 66, 557–559 associated with brain metastasis, 5–6 cause of, 5 clinical spectrum of, 5 defined, 557 drug-induced encephalopathy, 6 cisplatin, 6 metabolic, 5 necrotizing leukoencephalopathy, 558 risk factor for, 5 Wernicke’s encephalopathy, 5 Encephalopathy, toxic-metabolic disorders, 330 antimicrobial-induced, 332 chemotherapy-induced, 330–332 delayed, 332 immunosuppressive neurotoxicity and posterior reversible, 333–335 related to organ failure, 332 Wernicke’s encephalopathy, 335 Endocrine dysfunction, 270–271 Endometrial cancer, 450–451 Endovascular treatment-associated stroke, 229 Engraftment, 329 Eosinophilic meningitis, 574 Epidural spinal cord compression (ESCC), 4, 26–27 incidence, 410–411 management, 412–413 prognosis, 413 radiologic findings and diagnosis, 412 signs and symptoms, 411–412 spinal metastases, 27

625 Epidural spinal metastases (ESM), 164 Epilepsy, 39 Erythropoietin, growth factors, 311 Esophageal cancer, 482–483 Ewing’s sarcoma, 502 External-beam radiation therapy (EBRT), 172 Extrapontine myelinolysis (EPM), 335 versus CPM, 335 F Facial nerve injury, 275 Fatigue, 91, 92, 93, 101 in cancer patients, 104–106 Female reproductive tract cancer, neurologic complications of, 449 gynecologic cancers, 451 incidence gynecologic malignancy, 450 ovarian cancer, 449–450 treatment, 455–456 and complications, 455, 456 tumor brain metastases, 451–452 cerebrovascular disease, 454–455 leptomeningeal carcinomatosis, 453 lumbosacral plexopathy, 453 paraneoplastic disorders, 454 spine metastases and cord compression, 452–453 Fentanyl, 122 Fluid attenuated inversion recovery (FLAIR), 35 5-fluorouracil for head and neck cancers, 513 Focal brain radionecrosis, 263–265 G Gallbladder and bile duct (cholangiocarcinoma) carcinomas, 486 Gastric cancer, 483–484 Gastrointestinal cancer, neurologic complications of colorectal cancer, 484–485 esophageal cancer, 482–483 gallbladder and bile duct (cholangiocarcinoma) carcinomas, 486 gastric cancer, 483–484 gastrointestinal malignancies, 481–482 hepatocellular carcinoma (HCC), 485–486 pancreatic cancer, 486–487 chemotherapy-related neurologic complications, 487–490 metabolic abnormalities, 487 peripheral nervous system complications, 487 Gastrointestinal malignancies, 481–482 Gastrointestinal stromal tumors, 503–504 recurrent, 503 Gemtuzumab ozogamicin, 312

626 Genitourinary cancer, neurologic complications, 459 prostate cancer, 459–460 brain metastases, 463 complications of treatment, 465 dural metastases, 463 leptomeningeal disease, 464 paraneoplastic syndromes, 465 peripheral nerve compression, 464 skull metastases, 460–462 spinal disease, 460–462 stroke, 464 testicular cancer, 465–466 brain metastases, 465–466 Glioblastoma, 58, 59 Gliosarcoma, 503 histopathological appearance, 503 Glucocorticoids, 402, 412 Goserelin, 309 Graft versus host disease (GVHD), 329, 365–366 complications following HSCT, 330 neurological complications related to, 347 polymyositis (PM), in HSCT patients, 346 Granulocyte-colony stimulating factor (G-CSF) increased concentration of hematopoietic stem cells, 329 Granulocyte-macrophage colony-stimulating factor (GM-CSF) increased concentration of hematopoietic stem cells, 329 Granulomatous angiitis, 575 Growth hormone (GH), 271–272 Guillain–Barré syndrome (GBS), 346 GVHD. See Graft versus host disease (GVHD) H Head and neck cancer bony invasion, 509 brain, 509 cranial nerves, 508–509 incidence, 507 indirect complications, 509–510 leptomeninges, 509 neurologic complications of, 507–508 radical neck dissection as treatment, 513 spinal cord/dura, 509 treatment-related chemotherapy-related complications, 513 radiation-induced, 510–512 surgical complications, 513 Headache brain tumor headache, 58 cause, 59 characteristics in brain tumors, 58 in children with brain tumors, 60 incidence, 58–59 increased risk in patients with brain tumors, 59 management, 62

Index other causes of, 62 pathophysiology, related to brain tumors, 59 in patients with systemic cancer causes of headache in cancer patients, 61 with cerebral metastases, 60 with leptomeningeal metastases, 60 sellar tumors, 61–62 Headache management, 62 Headaches in children with brain tumors, 60 Hemangiopericytoma, 498–499 intracranial, 498 Hematogenous dissemination, 341 Hematologic cancers, treatment. See Hematopoietic stem cell transplantation (HSCT) Hematopoietic stem cell transplantation (HSCT), 329, 358 antitumor effect depending on GVT, 328 Candida, 341 cerebrovascular disorders, 338 intracranial hemorrhages, 338 ischemic strokes, 338–339 chemotherapies used in, associated with encephalopathy, 331 complications following, 330 diseases commonly treated, 328 drugs used in, cause seizures, 332 immune dysfunction disorders, 346 incidence of seizures, 337 indications for, 328 infection involving CNS, 339 infections, 339 bacterial infections, 339–340 fungal infections, 341–342 parasitic infections, 342–344 viral infections, 344–346 MRI characteristics of focal CNS infections, 340 myelopathy, 346 neurologic manifestations associated with GVHD, 347 neurological complications of, 327–330 neuromuscular disorders, 346–347 neutropenia and risk of Aspergillus, 339 peripheral stem cell, complications similar to CNS infections, 361 time course of neurological complications of, 360 toxic-metabolic disorders, 330 central pontine and extrapontine myelinolysis, 335 encephalopathy, 330–335 idiopathic hyperammonemia, 335–336 seizures, 337–338 toxic neuropathies, 338 Hematopoietic stem cells, 329 increased concentration, 329 in peripheral blood, 329 source, 329 Hemolytic uremic syndrome (HUS), 332, 339 Hepatocellular carcinoma (HCC), 485–486

Index HHV-6 encephalitis, 345 High-dose cytarabine (HIDAC) therapy, 330–332 Histocompatability antigens (HLA), 329 and T-cells, 327–328 Hodgkin’s disease (HD), 567 neurologic complications, 568 subacute cerebellar degeneration associated with, 584 Hospital Anxiety and Depression Scale (HAD), 103 HSCT. See Hematopoietic stem cell transplantation (HSCT) HSCT, infections, 339 bacterial infections, 339–340 fungal infections, 341–342 parasitic infections, 342–344 viral infections, 344–346 Hydromorphone, 121 Hypercoagulability, 221–222 and thrombocytopenia anticoagulation-induced hemorrhage in cancer patient, 230 cardiomyopathy, 230 infection and stroke, 230 other agents, 229–230 Hypercoagulability related to surgery, 228 Hyperleukocytic syndrome, 225 I Ibritumomab tiuxetan, 312–313 ICD-10, 71 Idiopathic hyperammonemia (IHA), 335 normal neuroimaging studies and abnormal EEG, 335, 337 treatment, 336 Ifosfamide, 332 Immune Reconstitution Inflammatory Syndrome (IRIS), 365 Infection, in HSCT patients, 339 bacterial infections, 339–340 fungal infections, 341–342 MRI characteristics of focal CNS, 340 parasitic infections, 342–344 viral infections, 344–346 Intensity-modulated radiotherapy (IMRT), 172 Intercellular edema, 53 Interventional procedures of cancer pain cordotomy and pituitary ablation, 126 spinal route, 125–126 sympathetic blocks, 126–127 Intracranial hemorrhage (ICH), 338, 557 prevalence in HSCT patients, 338 Intracranial metastasis (ICMs), 131 Intradural intramedullary spinal cord metastasis (ISCM), 175 site of primary tumor with, 175 Intramedullary spinal cord metastases, 414 Intraneural metastases acute promyelocytic leukemia in remission, 205

627 MRI image of extent of MPNST, 206 neuroleukemia, 205 neurolymphomatosis, 204–205 other PNS tumors, 205–206 Intraparenchymal hemorrhages, 338 Intrathecal chemotherapy, 193 Intratumoral parenchymal hemorrhage, 216–217, 219 Intravascular lymphoma (IVL), 575–576 Intraventricular methotrexate, neurologic complications of, 427 Iodine131 -tositumomab, 313 Ischemic stroke, 8 J Jugular foramen syndrome, 151 K Karnofsky Performance Score (KPS), 5 L Lambert–Eaton myasthenic syndrome (LEMS), 249, 416–417 Leiomyosarcoma, 500 Leptomeningeal lymphoma, 392 Leptomeningeal metastases, treatment of radiotherapy, 192 regional chemotherapy, 193–195 surgery, 196 systemic chemotherapy, 195–196 Leptomeningeal metastases (LM), 25, 60, 391–392, 407 diagnosis, 408–409 differential diagnosis, 190 epidemiology by tumor type, 407 imaging CT, 188 MR, 188–189 spine, 189–191 incidence, 182 laboratory investigations CSF analysis, 185–188 CSF tumor markers, 187 Indium-DTPA CSF flow study, 191 management biologic therapy, 410 intrathecal chemotherapy, 409–410 radiation, 409 ventricular peritoneal shunting, 410 manifestations, 407 MRI FLAIR image, 430 pathology and pathophysiology, 184–185 presenting symptoms and signs in patients with, 417 prognosis, 196–197, 410 radiologic findings, 408

628 signs cranial nerves, 408 parenchymal, 408 spinal cord/nerve roots, 408 staging, 190–191 symptoms and signs cerebral, 183 cranial nerves, 183–184, 408 parenchymal, 407–408 spinal cord and root symptoms and signs, 184 spinal cord/nerve root, 408 treatment limitations, 482 radiotherapy, 192 regional chemotherapy, 193–195 surgery, 196 systemic chemotherapy, 195–196 Lesions producing organisms. See Big Ten Letrozole, 309 Leukemia encephalopathy, 557–559 epidural spinal cord compression, 562 intracranial hemorrhage, 557 leukemic parenchymal tumor, 557 meningitis, 559–562 myopathy, 563 neurologic complications of, 556 peripheral neuropathy, 563 radiculopathy, 562 Leukemic meningitis, 563–564. See also Leptomeningeal metastases (LM) Leukemic parenchymal tumor, 557 Leukoencephalopathy, 265–267 increased risk of, 265 Leuprolide acetate, 309 LM. See Leptomeningeal metastases (LM) Lower cranial nerve involvement, 275–276 Lower motor neuron syndrome, 278–279 Lumbar puncture, 226 Lumbosacral plexopathy, 453 early-delayed, 278 late-delayed, 278 neurologic complications of treatment, 441 nonmetastatic complications cerebrovascular complications, 440 infection, 440 metabolic disorders, 441 paraneoplastic neurologic disorders, 439–440 peripheral neuropathy, 441–442 neurologic complications of radiotherapy, 442–443 neurologic complications of surgery, 443 Lumbosacral plexus, metastases to imaging, 209 presentation, 209 treatment, 209–210

Index Lung cancer AEDs, 402 chemotherapy, 405–407 approach to progressive or recurrent disease, 406 non-small cell lung cancer, 406 small cell lung cancer, 405–406 epidural spinal cord compression incidence, 410–411 management, 412–413 manifestations: signs and symptoms, 411–412 prognosis, 413 radiologic findings and diagnosis, 412 general oncologic management non-small cell lung cancer (NSCLC), 398–399 small-cell lung cancer (SCLC), 398 incidence, 397–398 indirect complications, 400 paraneoplastic neurologic syndromes, 415–417 intramedullary spinal cord metastases, 414 leptomeningeal metastases diagnosis, 408–409 epidemiology by tumor type, 407 management, 409–410 manifestations, 407 prognosis, 410 signs, 408 symptoms, 407–408 management glucocorticoids, 402 recurrent brain metastases, 403 role of surgery in single and multiple metastases, 402–403 surgery, 402 manifestations signs, 401 symptoms, 400–401 neurologic complications of base of skull metastases, 414 direct complications, 400 epidemiology, 397–398 parenchymal brain metastases, 400 patterns of spread to CNS non-small cell lung cancer, 400 small cell lung cancer, 399 plexus and peripheral nerve metastases brachial plexus, 414–415 recurrent laryngeal nerve, 415 prognosis, 407 radiation, 403 chemotherapy, 405–407 Prophylactic Cranial Irradiation (PCI), 404–405 sterotactic radiosurgery, 403–404 whole-brain radiation, 403 radiologic findings, 401 WHO classification of, 398–399

Index Lyme disease, 340 Lymphangiography, 226 Lymphoid cancer, treatment. See Hematopoietic stem cell transplantation (HSCT) Lymphomas central nervous system (intradural), 570 brain, 570–571 cranial and spinal nerves, 571–574 spinal cord, 571 neurological complications of, 567–570 newer diagnostic techniques, 584–585 paraneoplastic disorders in lymphomas, 584 peripheral nervous system (dural and extradural), 576 dural syndromes, 576–578 epidural syndromes, 578–581 nerve entrapment syndromes, 581–583 peripheral nerve syndromes, 583–584 M “3M” syndrome, 382 Magnetic resonance imaging (MRI), 5 Magnetic resonance spectroscopy (MRS), 23 Magnetization transfer (MT), 22 Malignant fibrous histiocytoma, 497–498 Malignant peripheral nerve sheath tumor, 500–501 arising within the spinal canal, 502 malignant nerve sheath tumor, 501 neurofibroma, 501 Malignant peripheral nerve sheath tumors (MPNSTs), 205 Mechlorethamine, 332 Medullary thyroid carcinoma with metastases to cervical lymph, 510 Melanoma, neurologic complications of, 542 CNS metastases co-existent extracranial disease, 528 incidence, 527 interval between initial diagnosis and brain metastases, 528 risk factors for developing CNS metastases in malignant melanoma, 528 risk factors in malignant melanoma, 528 intramedullary spinal cord metastases, 539–540 leptomeningeal metastases, 537–538 malignant melanoma, 523–524 melanoma incidence and age groups, 523–524 metastatic melanoma of unknown primary, 527 patterns of distant metastases, 527 risk factors, 524–525 staging and treatment overview, 525–527 neurologic complications related to therapy alpha-interferon therapy, 541–542 paraneoplastic disorders anti-Hun–related encephalomyelitis, 541 chronic inflammatory demyelinating polyneuropathy, 541

629 melanoma-associated retinopathy, 540–541 parenchymal brain metastases, 528–529 clinical presentation, 529 general, 528–529 intratumoral hemorrhage, 530 radiographic findings, 529–530 radiosurgery for, 532–534 radiosurgery for brain metastases, 532–534 radiosurgery versus surgical resection, 534–537 treatment overview and prognosis, 531 whole-brain radiation therapy for brain metastases, 531–532 plexus/peripheral nerve metastases, 540 spinal metastases, 538–539 Melanoma-associated retinopathy, 540–541 Memorial delirium assessment scale (MDAS), 74 Meningeal metastases, 25–26 leptomeningeal and pial enhancement, 26 Meningeal syndromes, causes, 358 Meningitis, 559–562 CSF categories, 560 Metastatic lesions, 164 Methadone, 121 Methotrexate, 409, 432 high-dose methotrexate neurotoxicity, 297 intrathecal methotrexate toxicity, 295–296 leukoencephalopathy, 297–299 weekly low-dose methotrexate neurotoxicity, 296 Middle fossa syndrome, 151 Mini-Mental State Examination (MMSE), 6 Mitotane, 309 Mood and psychiatric disorders in cancer patients, 101–104 Morphine, 120 Morphine-3-glucuronide (M3G), 120 Morphine-6-glucuronide (M6G), 120 Motor neuron disease (MND), 246–247 muscle atrophy in patient with SCLC and anti-Hu-associated myelitis, 247 MRI. See Magnetic resonance imaging (MRI) MRI, diagnosis of ESCC, 412 Mucin-positive adenocarcinoma-associated hypercoaguability antiphospholipid antibodies, 224 Multiple myeloma, 248, 596 direct effects of myeloma, 597 metabolic, toxic and infectious effects of myeloma, 598 remote effects of myeloma, 598–601 Multiple organ dysfunction syndrome (MODS), 332 Muscles, metastases to imaging, 207 pathophysiology, 206 treatment, 207 Myasthenia gravis, 346 Mycobacterium tuberculosis (MTB), 340 Myelography, 167

630 Myelography, diagnosis of ESCC, 412 Myelopathy, 346 Myopathy, 51–52 N Neoplastic infiltration of vessels arterial infiltration, 218–220 cerebral venous thrombosis, 219 hematologic malignancies, 220 venous infiltration, 218 Neoplastic Meningitis, intra-CSF chemotherapy for, 561 Neurocognition, therapies to improve or preserve, 100–101 Neurofibromatosis type 1 (NF1), 205 Neuroleukemia, 205 Neurologic complications timeline of common, 329 Neurologic diagnoses in cancer, 3–4 Neurologic disease in cancer, prevalence and impact of back pain, 4–5 brain metastases, 8–9 cerebrovascular complications cerebral hemorrhage, 8 cerebral venous thrombosis, 8 ischemic stroke, 8 diagnosis after consultation, 4 encephalopathy associated with brain metastasis, 5–6 drug-induced encephalopathy, 6 metabolic, 5 polyneuropathy cachexia-associated neuropathy, 7 paraneoplastic neuropathy, 7 plexopathy/mononeuropathy, 7 toxic neuropathy, 6–7 quality of life pain control as means to improve quality of life, 9 reasons for consultation to neurology service, 4 Neurolymphomatosis, 204 Neurolytic celiac plexus block (NCPB), 126 Neuromuscular disorders, 346–347 Neuropathic pain, 116 Neuropathy, 248 Neuropsychological assessment, importance of, 93–94 Neuropsychology, 93 Neurotoxicity, 287 causing drugs in chemotherapy, 289–290 azacitidine, 302 bleomycin sulfate, 302 busulfan, 303 capecitabine, 303 carboplatin, 303 chemotherapy, 289–290 chlorambucil, 303 cladribine, 303

Index cyclophosphamide, 303 dacarbazine, 303 doxorubicin, 302 estramustine, 303–304 etoposide, 304 fludarabine, 304 gemcitabine, 304 hexamethylmelamine, 304 hydroxyurea, 304–305 irinotecan, 305 levamisole, 305 mechlorethamine, 306 mitomycin C, 306 nitrosoureas, 306 plicamycin, 306 procarbazine-HCL, 306–307 pyrazolonacridine, 307 retinoic acid, 306 suramin, 307 temozolomide, 307 teniposide, 307 thioguanine, 307 thiotepa, 308 topotecan, 308 N ocardia asteroides, 340 Nocardiosis, 339 Non-Hodgkin’s lymphomas (NHL), 567 neurologic complications, 568 subacute motor neuronopathy associated with, 584 Non-small cell lung cancers (NSCLC), 133 Nonbacterial thrombotic endocarditis (NBTE), 223 Nonsteroidal anti-inflammatory drugs (NSAIDs), 117 Novel anti-edema agents, 54 Nuclear medicine techniques, 24 PET, evaluation of primary brain tumors, 24 Nuclear medicine techniques, 24 Numb chin syndrome, 581–582 O Occipital condyle syndrome, 152 Octreotide, 309 Ocular motor nerve injury, 275 Olfactory nerve injury, 274 Oligodendrocytes, 261 Oncologic disease, imaging neurologic manifestations of epidural spinal cord compression (ESCC), 26–27 imaging features of brain metastasis advanced techniques in brain metastasis imaging, 23–24 approach to evaluation of patients with newly diagnosed brain masses, 24–25 pitfalls in imaging newly detected brain masses, 17–21 strategies to improve detection of metastases by MRI, 21–23

Index meningeal metastases, 25–26 paraneoplastic neurologic syndromes encephalomyelitis, 27–28 paraneoplastic cerebellar degeneration, 28 Oncologic disorders, treatment, 358 Opioids, 113, 117, 118, 119–120, 121–122, 123, 124–125 -related adverse effects, 122–123 switching, 123–124 Oprelvekin, growth factors, 311 Optic neuropathy, 274 Orbital syndrome, 150 Osteogenic sarcomas, 501–502 Osteosclerotic myeloma, 248 Ovarian cancer, 449–550 gadolinium-enhanced MRI scan, 452 Oxycodone, 121 P Pain, 4, 59, 113 neuropathic, 116 nociceptive, 115–116 Palliative care, 70, 78, 82–83 Pancoast tumor, 414 Pancreatic cancer, 486–487 chemotherapy-related neurologic complications, 487–490 5-fluorouracil (5-FU), 488 bevacizumab, 488–489 capecitabine, 489 gemcitabine, 490 irinotecan, 490 oxaliplatin, 489 metabolic abnormalities, 487 peripheral nervous system complications, 487 Panitumumab, 313 Paraneoplastic autonomic dysfunction, 249 Paraneoplastic Cerebellar Degeneration (PCD), 28, 242–244, 416 electrodiagnostic features of PN associated with, 593 hematologic diagnosis of 28 patients with, 592 Paraneoplastic encephalomyelitis anti-CV2/CRMP5, 244 anti-Hu, 244 anti-Ma proteins, 244 anti-NMDAR, 244–245 Paraneoplastic limbic encephalitis, 241–242 Paraneoplastic limbic encephalomyelitis (PLE), 27, 416 Paraneoplastic neurologic disorders, diagnosis of antibodies, 239–240 presentation of symptoms, 239 syndromes similar without cancer association, 239 tumors associated with, 240–241 Paraneoplastic neurologic syndromes, 415–416 encephalomyelitis, 27–28 paraneoplastic cerebellar degeneration, 28 Paraneoplastic neuromyotonia, 249

631 Paraneoplastic neuropathy, 7 Paraneoplastic opsoclonus–myoclonus (POM), 246 Paraneoplastic sensorimotor neuropathy, 247 Paraneoplastic sensory neuronopathy, 245–246 Paraneoplastic stiff-man syndrome, 246 Paraneoplastic syndromes of nervous system acute necrotizing myopathy, 250 autonomic dysfunction, 249 cerebellar degeneration, 242–244 diagnosis of antibodies associated with, 240 paraneoplastic antibodies, 239–240 presentation of symptoms, 239 syndromes similar without cancer association, 239 tumors associated with paraneoplastic disorders, 240–241 encephalomyelitis anti-CV2/CRMP5, 244 anti-Hu, 244 anti-Ma proteins, 244 anti-NMDAR, 244–245 frequency, 238 Lambert–Eaton myasthenic syndrome, 249–250 motor neuron disease, 246–247 neuromyotonia, 249 opsoclonus–myoclonus, 246 pathogenesis, 238–239 polymyositis and dermatomyositis, 250 sensorimotor neuropathy, 247 sensorimotor peripheral neuropathy associated with malignant monoclonal gammopathies multiple myeloma, 248 osteosclerotic myeloma, 248 Waldenström’s macroglobulinemia, 248–249 sensory neuronopathy, 245–246 stiff-man syndrome, 246 vasculitis of nerve and muscle, 248 Paraneoplastic vasculitis of nerve and muscle, 248 Parasellar epidural metastasis, 434 Parasellar syndrome, 151 Parenchymal, 408 brain metastases, 400 Pediatric systemic cancer brain metastases, 608–609 solid tumors leading to, 609 chemotherapeutic drugs and neurologic side effects, 612 chemotherapy-related complications in children, 611–616 distribution of cancer types, 608 leptomeningeal/spinal metastatic disease, 609–610 sagittal post-contrast MRI, 610 neurologic complications of, 607–608 paraneoplastic disorders, 611 Perfusion-Weighted Imaging (PWI), 23 Peripheral nerve lymphoma, 583–584

632 Peripheral Nerve Disorders antibody activities of monoclonal IgM in, 594 Peripheral nerves metastases to acute promyelocytic leukemia, 205 compression of nerves, 204 extent of MPNST, 206 intraneural metastases, 204–206 Peripheral nervous system (PNS), 203 Peripheral nervous system metastases metastases to brachial plexus imaging, 208 presentation, 208 treatment, 209 metastases to cervical plexus imaging, 207 presentation, 207 treatment, 207 metastases to lumbosacral plexus imaging, 209 presentation, 209 treatment, 209–210 metastases to muscles imaging, 207 pathophysiology, 206 treatment, 207 metastases to peripheral nerves, 203–204 compression of nerves, 204 intraneural metastases, 204–206 Pharmacotherapy, 170 Pituitary adenoma, 59, 61 Pituitary apoplexy, 221 Pituitary lymphoma, 575 Pituitary tumor, 57, 61–62 Plasma cell dyscrasias classification of common, 592 clinical syndromes benign plasma cell dyscrasias, 593–596 cryoglobulinemia, 604 laboratory screening, 592 lymphoma, leukemia, cancer, 604 multiple myeloma, 596 direct effects of myeloma, 597 metabolic, toxic and infectious effects of myeloma, 598 remote effects of myeloma, 598–601 neurologic complications of, 591 primary systemic amyloidosis clinical features, 601–602 laboratory tests, 602 pathogenesis, 603 treatment, 603 Waldenström’s Macroglobulinemia (WM), 603 Plexopathies, 583

Index Plexopathy/mononeuropathy, 7 types of, 7 PML, 372–374 Polymyositis (PM), 250 Polymyositis (PM)/Dermatomyositis (DM), 417 Polyneuropathy cachexia-associated neuropathy, 7 paraneoplastic neuropathy, 7 plexopathy/mononeuropathy, 7 toxic neuropathy, 6–7 Positron emission tomography (PET), 24 role of in diagnosis and management lymphoma, 581 scans, 385 Post-Transplantation Lymphoproliferative Disorder (PTLD), 362–363 Posterior reversible encephalopathy syndrome (PRES), 333 MRI diagnostic test and, 334 Primary brain tumors, neurological complications of, 392 clinical manifestations, 381–382 specific neurological complications brain edema, 382–388 hydrocephalus, 388–389 leptomeningeal metastases, 391–392 psychiatric disorders, 391 seizures, 389–391 Primary central nervous system lymphoma (PCNSL), 99 Primary CNS diseases causes of delirium in cancer patients, 76 cerebrovascular diseases, 75 degenerative diseases, 76 epilepsy, 75 infections, 75 intracranial tumor, 75–76 trauma, 75 Primary fibrinolysis, 225 Primary systemic amyloidosis clinical features, 601–602 laboratory tests, 602 pathogenesis, 603 treatment, 603 Progressive multifocal leukoencephalopathy (PML), 345 Prophylactic Cranial Irradiation (PCI), 132, 404–405 effects of, 98 Psychiatric disorders, 391 Psychosis, 66 Q Quality of life (QOL), 91 components of, 91 paincontrol as means to improve quality of life, 9 R Radiation, 413 Radiation enhancers, 139

Index Radiation-induced tumors brain tumors, 268–269 cavernomas, angiomatous malformations, and aneurysms, 270 peripheral nerve sheath tumors, 279 vasculopathy, 269–270 Radiation-induced vasculopathy laser treatment, 227 Radiation therapy, 171–172 Radiation therapy, neurologic complications of consequences of RT on peripheral nervous system brachial plexopathy, 276 dropped head syndrome, 276 lumbosacral plexopathy, 278 early-delayed complications of RT somnolence syndrome, 262 subacute rhombencephalitis, 263 transitory cognitive decline, 262–263 worsening of pre-existing symptoms, or tumor-pseudoprogression, 262 endocrine dysfunction, 270–271 late-delayed complications of RT cognitive dysfunction and leukoencephalopathy, 265–267 focal brain radionecrosis, 263–265 main differential diagnoses of cognitive impairment in cancer patients, 266 lower motor neuron syndrome, 278–279 oligodendrocytes, 261 pathophysiology, 260 radiation and other CNS cell types, 261 radiation-induced brain tumors, 268–269 radiation-induced peripheral nerve sheath tumors, 279 radiation-induced vasculopathy large and medium intra- and extracranial artery injury, 269 radiation-induced cavernomas, angiomatous malformations, and aneurysms, 270 radiation-induced vasculopathy with moyamoya pattern, 269 silent lacunar lesions, 269–270 sequelae of radiation therapy on brain acute encephalopathy, 261–262 sequelae of radiotherapy on cranial nerves acoustic nerve dysfunction, 275 facial nerve injury, 275 lower cranial nerve involvement, 275–276 ocular motor nerve injury, 275 olfactory nerve injury, 274 optic neuropathy, 274 trigeminal nerve dysfunction, 275 sequelae of radiotherapy to spinal cord early-delayed (transient) radiation myelopathy, 271–272

633 late-delayed radiation-induced spinal cord disorders, 272–273 vascular damage, 260–261 Radiation therapy (RT), 259 on brain, sequelae of acute encephalopathy, 261–262 consequences of RT on peripheral nervous system, 276–278 on cranial nerves, 274–276 early-delayed complications of, 262–263 late-delayed complications of, 263–268 to spinal cord, sequelae of early-delayed (transient) radiation myelopathy, 271–272 late-delayed radiation-induced spinal cord disorders, 272 main causes of L’hermitte’s phenomenon in cancer patients, 272 Radiosurgery, 263 Randomized controlled studies (RCTs), 132 summary slide of RCTs comparing WBRT, SRS, and surgery for patients with ICMs, 135 Recurrent disease, approach to, 413 Regional cerebral blood volume (rCBV), 23 Renal carcinoma, 470–471 brain metastases, 471–472 complications of treatment, 474 intramedullary spinal cord metastasis, 471, 474 leptomeningeal disease, 472 paraneoplastic syndromes, 473–474 spinal cord disease, 472–473 Reversible Posterior Leukoencephalopathy Syndrome (RPLS), 365 Rhabdomyosarcoma, 499–500 Rituximab, 313 use of, 363–364 RT. See Radiation therapy (RT) S Sarcoma, neurologic complications, 495–496 chondrosarcoma, 496–497 Ewing’s sarcoma, 502 gastrointestinal stromal tumors, 503–504 gliosarcoma, 503 hemangiopericytoma, 498–499 leiomyosarcoma, 500 malignant fibrous histiocytoma, 497–498 malignant peripheral nerve sheath tumor, 500–501 osteogenic sarcomas, 501–502 rhabdomyosarcoma, 499–500 Seizures, 33, 34, 389–391 differential diagnosis of seizures in patients with cancer, 36 epidemiology and pathogenesis, 389

634 epidemiology of seizures in patients with brain tumors, 34–35 evaluation of seizures in patients with known cancer, 37 frequency of different tumor histologies as cause of first seizures at different ages, 35 frequency of seizures by tumor histology, 34 frequency of tumor as etiology of first seizures, 35 overall frequency of seizures related to tumor growth rate, 34 prophylactic treatment and adverse effects, 390–391 influence of enzyme-inducing anti-seizure drugs, 390 symptomatic treatment, 389–390 Seizures and anti-epileptic drugs in neuro-oncology anticonvulsant prophylaxis in patients with brain tumors, 40–41 physician practices for prescribing prophylactic anticonvulsants before and after publication of AAN guidelines, 40 use of prophylactic anticonvulsants in patients with primary and metastatic brain tumors, 41 clinical features of tumor-associated seizures, 35–36 epidemiology of seizures in patients with brain tumors, 34–35 frequency of different tumor histologies as cause of first seizures at different ages, 35 frequency of seizures by tumor histology, 34 frequency of tumor as etiology of first seizures, 35 overall frequency of seizures related to tumor growth rate, 34 evaluation of seizures in patients with known cancer, 37 differential diagnosis of seizures in patients with cancer, 36 outcome in tumor-associated epilepsy, 39–40 special issues in patients with tumor-associated seizures, 41 treatment of tumor-associated seizures: drug therapy, 37–39 commonly used enzyme-inducing and nonenzyme-inducing anticonvulsants, 38 treatment of tumor-associated seizures: radiation and surgery, 39 Sellar tumors, 61–62 Sensorimotor peripheral neuropathy associated with malignant monoclonal gammopathies multiple myeloma, 248 osteosclerotic myeloma, 248 Waldenström’s macroglobulinemia, 248–249 Sequelae of cancer diagnostic tests and treatment chemotherapy bone marrow transplantation, 230 hypercoagulability and thrombocytopenia, 229 diagnostic procedures bronchoscopy procedure–induced tumor emboli, 228

Index endovascular treatment–associated stroke, 229 hypercoagulability related to surgery, 228 lumbar puncture, 226 lymphangiography, 226 radiation-induced vasculopathy, 227 surgery-associated stroke directly related to surgery, 227–228 Silent lacunar lesions, 269–270 Single-photon emission computed tomography (SPECT), 24 with thallium 201, evaluation of primary tumors, 24 Skull and dural metastases calvarial metastases epidemiology, 146 imaging, 147–148 metastatic breast carcinoma, 147 pathophysiology, 146 presentation, 146 treatment, 148–149 dural metastases epidemiology, 157 imaging, 157–158 pathophysiology, 157 presentation, 157 imaging, 153 Skull base metastases epidemiology, 149 manifestations of metastatic skull base syndromes, 150 nasopharyngeal carcinoma, 156 pathophysiology, 149 presentation, 149–153 jugular foramen syndrome, 151 middle fossa syndrome, 151 occipital condyle syndrome, 152 orbital syndrome, 150 other syndromes, 152–153 parasellar syndrome, 151 sclerosis, 153, 154–155 Small cell lung carcinoma (SCLC), 131 Somnolence syndrome, 262 Spinal cord compression, 3, 4–5 Spinal cord disorders, late-delayed radiation-induced progressive myelopathy, or delayed radiation myelopathy (DRM), 272–273 spinal hematoma, 273 spinal radionecrosis, 273 Spinal metastases bony metastasis, 168 diagnostic work-up, 167–169 epidemiology, 164–165 intradural intramedullary, 175–176 Loblaw’s evaluation of ESM in Ontario, 165 pathogenesis and pathophysiology, 165 presentation, 166

Index prognosis, 169–170 site of primary tumor with epidural, 164 Tokuhashi’s evaluation for prognosis, 170 treatments chemotherapy, 171 pharmacotherapy, 170–171 radiation therapy, 171–172 surgical management, 173–175 uterine cancer, 167 Stem cell transplantation, hematopoietic. See Hematopoietic stem cell transplantation (HSCT) Stereotactic radiosurgery (SRS), 97, 137–139, 403–404 Stereotactic radiosurgery or with whole-brain radiation, effects of, 97–98 Steroid pseudorheumatism, 387 Steroid therapy, 383–387 adverse steroid effects, 386–387 bowel perforation, 387 osteoporosis and prolonged, 387 Steroids, 387 clinical neuro-oncologic usage of efficacy of steroids, 48–50 imaging, 50 Stroke, 8 central nervous system tumor intratumoral parenchymal hemorrhage, 216–217 neoplastic infiltration of vessels, 218–220 pituitary apoplexy, 221 subdural hemorrhage, 217–218 tumor embolus, 220–221 hyper- and hypocoagulopathies arterial occlusions, 223 bleeding diathesis/hemorrhage, 225–226 combined hypercoaguability/bleeding diathesis, 224–225 hypercoagulability and thrombosis, 221–222 mucin-positive adenocarcinoma-associated hypercoaguability, 223–224 venous occlusions, 222–223 Subacute rhombencephalitis, 263 Subacute Sensory Neuronopathy (SSN), 416 Subarachnoid hemorrhages (SAH), 338 Subdural hemorrhages (SDH), 217–218, 338 Surgery, 413 Surgery-associated stroke directly related to surgery, 227–228 bronchoscopy procedure–induced tumor emboli, 228 Surgical management of spine metastasis, 173–175 anterior and posterior surgical intervention, 174 Systemic cancers, 15–16 headache in patients with cerebral metastases, 60 leptomeningeal metastases, 60 Systemic inflammatory response syndrome (SIRS), 332

635 T Tacrolimus (FK-506), 333 Tamoxifen, 309 Therapy of brain metastases chemotherapy, 139–141 radiation enhancers, 139 renal cell carncinoma and single left occipital metastases, 138 stereotactic radiosurgery, 137–139 summary slide of RCTs comparing WBRT, SRS, and surgery for patients with ICMs, 135 surgical resection plus WBRT, 136–137 whole-brain radiation therapy (WBRT), 134–136 Thrombocytopenia, 225–226 Thrombosis, 221–222 Thrombotic microangiopathy syndromes (TMS), 339 Thrombotic thrombocytopenic purpura (TTP), 339 TNF. See Tumor necrosis factor (TNF) Tonsillar squamous cell carcinoma and cervical adenopathy with chemoradiotherapy, 511 Toremifene citrate, 309 Toxic neuropathy, 6–7 Toxicity, 303 Toxicity of corticosteroids complications of corticosteroid therapy, 51 gastrointestinal bleeding, 50–51 infections, 52 mood disturbance, 52 myopathy, 51–52 osteoporosis, 52 other effects, 52 Toxoplasmosis, 342–343, 344 Transitory cognitive decline, 262–263 Transplants autologous, 329 catogorizing according to source of hematopoietic stem cells, 329 relationship between donor and recipient, 329 conditioning before, 329 neurological complications, 329 Trastuzumab, 313 Treatment, cancer pain analgesics: pharmacological treatment, 117–118 high-potency opioid, 120 low-potency opioid, 119 Step 1, 118–119 Step 2, 119 Step 3, 119–122 Treatments for spinal metastasis chemotherapy, 171 pharmacotherapy, 170–171 radiation therapy, 171–127 surgical management, 173–175 Trigeminal nerve dysfunction, 275 Trimethoprim-sulfamethoxazole (TMP-SMX), 340

636

Index

Tumor-associated seizures clinical features of, 35–36 commonly used enzyme-inducing and nonenzyme-inducing anticonvulsants, 38 outcome in, 39–40 special issues in patients with, 41 treatment of drug therapy, 37–39 radiation and surgery, 39 Tumor classification, 398 NSCLC, 400 adenocarcinoma, 398 large cell undifferentiated carcinoma, 398 squamous cell or epidermoid carcinoma, 398 staging of, 399 SCLC, 398, 399 Tumor embolus, 220–221 papillary fibroelastoma of aortic valve, 221 Tumor necrosis factor (TNF), 311 Tumor-pseudoprogression, 262

Vascular damage, 260–261 Vascular endothelial growth factor (VEGF), 54 Vasogenic edema, 382 occurence and spreading, 384 Vasogenic edema, 54 VEGF inhibitors, 54 Veno-occlusive disease (VOD), 332 Venous occlusions, 222–223 Ventricular peritoneal shunting (VPS), 410 Vertebral column metastasis symptoms, 4–5

V Varicella Zoster Virus, 372

Z

W Waldenström’s Macroglobulinemia (WM), 248–249, 603 Whole-Brain Radiation Therapy (WBRT), 97, 134–136 X X-rays, diagnosis of ESCC, 412

Zygomycetes, 371–372