Thyroid Cancer
Thvroid Cancer r/
A Comprehensive Guide to Clinical Management
Edited by
Leonard Wartofsky MD, MPH, MACP Washington Hospital Center, Washington,DC
Foreword by
Ernest L. Mazzafem, MD, MACP Ohio StateUniversityMedical Center, Columbus,OH
Humana Press
%
Totowa, New Jersey
0 2000 Humana Press Inc. 999 Riverview Drive, Suite 208 Totowa, New Jersey 075 12
All rights resewed. No part of this book may bereproduced, storedin a retrieval system,or transmitted in any formor by any means,electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permissi from the Publisher. All authored papers, comments, opinions, conclusions,or recommendations are those of the author(& and do not necessarily reflect the views of the publisher.
This publication is printedon acid-free paper. @ ANSI 239.48-1984 (American Standards Institute) Permanence of Paper for Printed Library Materials. Cover illustration: “Histopathologic sectionof papillary thyroid carcinoma showing typical nuclear crowding, with variation in size and shape of cells, some of which have emptyor pale nuclei,” courtesyof James Oertel,MD. Cover designby Patricia F. Cleary.
For additional copies, pricingfor bulk purchases, and/or information aboutother Humana titles, contact Humana at the above address or at any of the following numbers: Tel.: 973-256-1699; Fax: 973-256-8341; E-mail:
[email protected]; Website: http://humanapress.com
Photocopy Authorization Policy: Authorization to photocopy items for internal orpersonal use,or the internalor personal useofspecific clients, is grante by Humana Press Inc., provided that the base fee US of S 10.00 per copy, plusUS $00.25 per page, is paiddirectly to the Copyright Clearance Center at 222 Rosewood Drive, Danvers,MA 01923. For those organizations that have bee granted a photocopy license from the CCC, aseparate system ofpayment has beenarrangedand is acceptableto Humana Press Inc. The feecode for usersof the Transactional Reporting Service is:[0-89603-429-1/00 S10.00 + $00.251. Printed in the United States of America. 10 9 8 7 6 5 4 3 2 I Library of Congress Cataloging-in-PublicationData
cm.
Thyroid cancer :a comprehensive guide to clinical management I edited by Leonard Wartofsky. P. Includes bibliographical references and index. ISBN 0-89603-429-1 (alk. paper) 1. Thyroid gland-Cancer-Treatment. 1. Wartofsky, L. [DNLM: 1. Thyroid Neoplasms. WK 270 T5494 I9991 RC280.T6T527 1999 61 6.99’4444~21
DNLMlDLC for Library of Congress
99-2 1826
CIP
Dedicated to the memoryo f a great thyroidologist, teacher, and friend, Dr. Sidney H. Ingbar, and to all of our patients who struggle under the cloud of an uncertain future with their malignancies
Foreword
As the 20th century draws to a close, it seems like a propitious to look timeback upon the advances we have made in understanding thyroid carcinoma, since our knowledge today will certainly serve to light the path of discovery in the next century. Gazing at world through a small looking glass focused on thyroid carcinoma seems an appropr way to begin thinking about the clinical management of this group of diseases. Wha the important things that we have learned in recent years that form the basis of ou clinical knowledge? How can we best use that information in the care of our patients? Dr. Leonard Wartofsky's new and sharply focused text, Thyroid Cancer, promises to answer this hypothetical set of questions in a succinct and clinically relevant way. It sometimes seems that thyroid carcinoma is a neglected orphan among human cancers, which is at the root of some important issues. Thyroid carcinomas comprise a diverse group of malignancies ranging from indolent microscopic papillary carcinomas that pose no threat to survival to anaplastic carcinomas that are the most vicious carcinomas afflicting humans. Yet, because of its low incidence, there have been no prospective randomized clinical trials of the treatment of thyroid carcinoma. Furthermore, none are likely to be done, given the prolonged survival and relatively low mo tality rates associated with the majority of these cancers. Nonetheless, patients often suffer greatly from this disease: many have serious recurrences and some die from relentlessly progressive and untreatable cancer. This is a disease that knows no boundaries, striking young and old alike. Unfortunately, management paradigmsfrom derive retrospective studies, and few new drugs have been added our to therapeutic armamentarium. One would thus anticipate a deep void in our understanding of these tumors. Despite these shortcomings, the 20th century has seen major advances in our understanding of their etiology, pathophysiology, and management. The good newsis that the advances have been rapidly translated into improved outcomes for many patients with thyroid carcinoma. For example, data from the National Cancer Institute shows that, although the incidence of thyroid carcinoma has increased significantly-almost 2 8 Y e i n c e the early1970s in the United States, cancer-specific mortality rates during this same period have dropped significantly-by almost 2 1%. In my viewthis results from the earlier diagnosis of the cancer, which allows the full impact of effective therapy, and which I believe has dramatically altered the clinical course of these tumors. One of the dazzling success stories in medicine in the of lastthis halfcenturyis that with medullary thyroid carcinoma, a truly orphan tumor afflicting relatively few p First identified in1959 as a discrete entity, this tumor was identified before calcitonin was known to exist and before the mystery of the multiple endocrine neoplasia syndromes had been completely unraveled. The pieces of the puzzle fell together at ligh ning speed over a few decades. The Ret proto-oncogene mutations recently identified vii
viii
Foreword
in this tumor will serve as the portal to our eventual complete understanding of its biology and are already the keystone to its diagnosis in members of afflicted Now children with this genetic defect can be identified with molecular testing well before medullary thyroid carcinoma becomes clinically manifest or is identifiable by any other test, resulting in thyroidectomy that cures the disease. What a wondrous group of developments to pass on to the patients and physicians of the next cent This work serves as a model for the scientific investigation of malignant tumors. We are also acquiring a clearer view of the molecular biology of well-diff Ret rearrangements found in papillary c papillary and fo1licular”thyroid carcinomas. cinomas of humans have been shown to produce the tumor in transgenic mice, of papillary carcinoma. Study o underscoring the central role Ret of in the pathogenesis familial papillary thyroidcarcinoma-now recognized to occur in a small but impo subset of patients in whom it may be transmitted as an autosomal dominant trait-ndoubtedly will provide important new information. These and other exciting di such as the identification of the sodium-iodide symporter in laboratory animals an mans, portend more basic discoveries that will generate currently unimaginable is tic and therapeutic tools. The latest example of this success in the laboratory recombinant h m a n TSH, which was recently introduced into clinical practice and already is dramatically improving and simplifying the care of patients with differen thyroid carcinoma. 50 years we have learned much about the important etiologic r During the past ionizing radiation in thyroid carcinoma. Introduced at the turn of the 20th century b Roentgen, external radiation soon became routine practice in the United States benign clinical conditions ranging from “status thymicolymphaticus” to acne. It to is extremely over 50 years, however, to understand that the thyroid gland of children sensitive to the carcinogenic effects of ionizing radiation and that this therapy itsel caused papillary thyroid carcinoma, often decades after the exposure. Studies of th Japanese survivors of the atomic bombings of Nagasaki and Hiroshima first documented thyroid carcinoma as a consequence of radioactive fallout. Nonetheless, the notion was long held that internal radiation of the thyroid from ingested radioactiv iodine was not a thyroid carcinogen. The outbreak of papillary thyroid carcinoma among children exposed to radioactive iodine fallout from the nuclear reactor in Chernobyl, however, abruptly closed the door on this notion. This accident deadly exclamation mark after the statement that small doses of radioiodine carcinogenic to the thyroid glands of infants and children, and sparked renewed c cerns about the aboveground nuclear weapons testing program in Nevada 1950 betw and 1960, during which radioactive iodine fallout rained down on nearly the entire continental United States. The National Cancer Institute estimates that a substantial excess of thyroid carcinomas has probably occurred and perhaps will continu as a result of this exposure.How clinicians will deal with this information, includin what tests should be done, is under discussion, but national screening studies are likely to be done. We also have learned much about the pathology of thyroid carcinoma during the 20th century. The early observations about the prognostic implications of tumor s and invasion through the thyroid capsule are now well accepted. In addition, patho
Forword
ix
gists now recognize a number of histologic variants of papillary and follicular carc that have important implications that must be carefilly factored into the assessment of a tumor’s prognosis. Other important advances in our understanding of thyroid patholog have occurred in this decade. What was once considered small-cell anaplastic thyroid carcinoma is now recognizedas thyroid lymphom-own to be a rare complication of Hashimoto’s disease-which nevertheless seems to be occurring with increasing frequency. While we were busy discovering thyroid lymphoma, the incidence of anaplasti as a result of early diagnosis and thyroid carcinoma has been quietly declining, probably tumorsthat often serve as its forerunner. All of us breath treatment of well-differentiated a quiet sigh of relief at this improvement. Now it is well appreciated that a tumor’s prognosis cannot be fully assessed until its final histology has been carefilly studied, of its sometimes both histologically and immunochemically, to uncover the dark secrets origin. This has therapeutic implications. I think much of our success in reducing mortality from thyroid carcinoma stems from early diagnosis. Thirty years ago the main diagnostic identifi tests toa malignant nodule were thyroid hormone suppression and radionuclide imaging. Now the standard of care in a clinically euthyroid patient is to perform a fine-needle aspiration biopsy of the nodule before any other tests are done. Though it remains a less than perfect test, the study of fine-needle cytology has prevented unnecessary surgery in many patients while increasing the yield of carcinoma among those undergoing thyroidectomy. I think fine-needle aspiration of nodules has saved more lives than is generally acknowledged-by preventing long periods of thyroid hormone suppression-while malignant nodules sometimes became wildly metastatic. There is evidence that long delays in therapy significantly increase cancer-specific mortality rates of papillary and follicular thyroid carcinoma. The key to fine-needle aspiration diagnosis is to understand the diagnostic details of the cytology report and to act accordingly. Much of the current debate on thyroid carcinoma has revolved around the extent of initial therapy, both surgical and medical, that is necessary for patients with differe ated thyroid carcinoma. Almost everyone believes that some differentiated thyroid carcinomas require minimal therapy, whereas others require more aggressive manage ment. The problem lies in defining aggressive tumors. Several staging systems and prognostic scoring systems have been devised to discriminate between low-risk patients who are anticipated to have a good outcome with minimal therapy and higher risk patients who require aggressive therapy to avoid morbidity or mortality from thyroid carcinoma. However, most of the prognostic systems do not identify the variants of papillary and follicular carcinoma that have remarkably different behaviors. Most prognostic scoring systems have been created with multiple regression analysis to f predictive combinations of factors, but almost none include therapy in the analysis. Moreover, almost all of them have considered cancer mortality as the endpoint of therapy, ignoring tumor recurrence or disease-free survival. This becomes problem40 typically have low cancer mortality atic in defining risk because patients under age rates, but experience high rates of tumor recurrence. Because most recurrences are in the neck and are easily treated, some clinicians regard them as trivialproblemMut my patients find this notion incomprehensible. Most patientsare devastated by a re-
x
Foreword
currence of tumor, regardless of its site. The greatest utility of prognostic scoring systems lies in epidemiological studies and as tools to stratify patients for prospecti therapy trials, but they are least useful in determining treatment for individua In the past few decades most have come to believe that near-total or total thyroidectomy is the optimal treatment for thyroid carcinomas, even for patients at relatively low risk of mortality from their carcinomas. The main reason do total not tothyroidectomy is that it is associated with higher complication rates than those of lobectomy for a long However, there is now evidence documenting what most of us have known time: surgeons with the most experience have the lowest complication rates, regardless of the extent of the thyroidectomy. Given the low frequency of th a compelling argument can be made to refer patients to centers with highly surgeons for their initial management. Followup of differentiated thyroid carcinomas and medullary thyroid carcinom is greatly facilitated by sensitive serum tests-thyroglobulin and calcitonin-and the use I believe that what we term recurrence of tum is of a variety of scanning techniques. actually persistent disease that previously fell below the radar of our older, le tive detection tools. With newer sensitive tests including, for example, thyroglobu measured by messenger RNA, we now have the opportunity to identify and treat roid cancers at an earlier and more responsive stage. Perhaps the most vivid exam of this is the identification of diffuse pulmonary metastases among patients wit serum thyroglobulin levels and negative diagnostic imaging studies, which are only seen on posttherapeutic whole-body scans done after large therapeutic doses ofl3II. Whether this enhances survival continues to be debated, but Ithink there are compelling reasons to suggest that it does improve outcome. Thus many relatively new observations and management tools that have lar developed in the last half of this century are being brought to the bedside to su tially enhance our ability to improve the outcome of most patients with thyroid noma. Many challenges remain, however. More effective therapy is urgently needed for patients with widely metastatic disease thatis unresponsive to current therapies. We need to understand more about the molecular predictors of recurrence and deat from thyroid cancer. Nonetheless, our present state of knowledge provides cli wide variety of diagnostic and therapeutic modalities to effectively manage thi in Cancer will give the practi of cancers. I believe the knowledge contained Thyroid ing clinician the necessary information to provide patients the latest and best tic and therapeutic techniques. Ernest L. Mazzaferri, MD, MACP
Preface
Publication of a new medical text should require some justification. We believe that no text has been published or currently exists that serves the needs of Thyroid practicing physicians for clinically relevant information about thyroid cancer. Cancer: A Comprehensive Guide toClinical Management is intended to fill that need and serve as an extensive and inclusive reference source or handbook to clinicians managing patients with thyroid cancer. The various sections have been written by highly knowledgeable experts, and the editor has functioned to organize and systematize the materials into a readable and practical whole. But, equally important to the authors’ expertise is the fact that they are clinicians who are aware of the questions being asked by practicing physicians, whether in pediatrics, family practice, internal medicine, endocrinology, endocrine surgery, radiation therapy,or nuclear medicine. Indeed, no one author could have the requisite clinical experience to write knowledgeA word of warningis appropriate, however. Because ably on all of the topics covered. of the absence of large, well-controlled clinical trials on the management of thyroid cancer, many aspects of the standard practices or recommendations of the authors will be controversial and not universally accepted. In fact, there are differences of opinion expressed within the various sections of the book, and the editor has not endeavor to suppress controversy or achieve unanimity. Rather, an attempt has been made present evidence-based recommendations where possible, and to indicate clearly wher evidence is lacking and management might be deemed controversial. Thyroid Cancer has a somewhat unusual format in that the same author has written each comparable subsection within each major tumor section. For example, rather t one chapter covering the pathology of all the thyroid cancers, there are separate p ogy chapters within the broad headings of papillary cancer, follicular cancer, medullary cancer, etc., and separate chapters within each of these sections dealing with the clinical presentation, surgical approach, nuclear medicine studies, chemotherapy, etc. Thus, the arrangement should allow the reader to look up the specific cancer or in his her patient and find everything they need to know in one convenient place in an economically concise format. This organization was not always practicaltheand original plan breaks down somewhat with, for example, the surgical management of papillary and follicular thyroid carcinomas, which is similar except for the usual need for a more complete thyroidectomyfor the latter. The minimal differences did not justifL a separate chapter in our view, and in this case the surgery section is described for differ ated thyroid carcinoma. In either case, the emphasis on practical is clinical management such that the volume should represent both an authoritative reference source and a practical handbook for management. xi
xii
Prefkce
It is clear that the pace of accumulating information about thyroid cance Thyroid Cancer is an up-to-date sum been both rapid and accelerating in recent years; known mary of the current state of our knowledge regarding thyroid nodules, and and clinically significant variant of thyroid malignancy. Each section starts with th clinical presentation and then proceeds through diagnostic evaluation, histopatho medical or surgical approaches to therapy, and then to natural history and prognos After the initial sections on thyroid nodules, there is a scholarly review of the patho genesis and epidemiology of thyroid cancer. In the latter context, we have learned much about the risks of radiation exposure from1986 the Chernobyl accident, and th relevance of these data to other lesser radiation exposures is discussed. Also in the p decade, remarkable advances have been made in our ability to diagnose and mana thyroid cancer. Isotopic scanning modalities other than radioiodine have been dev oped, and the recent availability of recombinant human thyrotropin is transfig our management, particularly of low risk patients. Advances in our understanding of immunology of thyroid cancer and the genetic alterations leading to malignancy tinue unabated and the current state of our knowledge in these areas is also tho reviewed. Precision in diagnosis of cancer has been facilitated by fine needle aspira tion cytology, andin cancer followupby development of more specific and sensitiv assays for serum thyroglobulin, and by sensitive imaging techniques including ult sound, CT, MRI, and the newer scanning modalities. Much has been learned about the appropriate surgical procedures that permi an optimal prognosis without unnecessary disfigurement. Recent appreciationof the phenomenon of radionuclide “stunning” has led to development of protocols t mize therapeutic efficacy of radioiodine therapy. Separate sections of the volum with the special aspects of thyroid cancer in children, and the thorny problem of management of the patient with suspected residual thyroid cancer on the basis of me surable thyroglobulin levels but who has radioiodine scan surveys are negative that and fail to identify a source of the thyroglobulin, i.e., the cancer. Though the major port of the text deals with all aspects of the most common thyroid carcinomas, the wel differentiated papillary and follicular types, there are extensive treatments as well of the primary thyroidal cancers that do not arise from follicular epithelium, e.g., lary and anaplastic cancer, and lymphoma. Separate sections also deal with the m rare and unusual malignant thyroid tumors, both in regard to their pathology a clinical presentation and management. Those desiring a greater in-depth discu provided a current and exhaustive bibliography in each section. I thank the many authors for their outstanding contributions, and the ef the skilled executive staff of Humana Publishers, including Thomas Lanigan, Paul professional Dolgert, Craig Adams, and James Geronimo who produced the in so book and timely a manner. We bring this volume to its readership with great enth with the sincere hope that it will prove to be of substantial utility to physic them to provide the greatest benefit to their patients with thyroid cancer.
Leonard Wartofsky, MD
Contents ~~
~
Foreword by ErnestL. Mazzaferri, MD, MACP .............................................................
vii
Preface ...........................................................................................................................
xi
Contributors .................................................................................................................
xix
Part I The Thyroid Nodule The Thyroid Nodule: Pathogenesis, Evaluation, and Risk of Malignancy Leonard Wartofsky
...............................................................................................
3
Nonisotopic Imaging of the Neck in Patients with Thyroid Nodules or Cancer Manfred Blum....................................................................................................... 9 The Thyroid Nodule:Fine Needle Aspiration Biopsy Yolanda C. Oertel...............................................................................................
35
The Thyroid Nodule:Medical Management Leonard Wartofs............................................................................................... 39 Thyroid Nodules and Cancer Risk: Surgical Management Orlo H. Clark...........................................................................................,..,.......
49
Part I1 Thyroid Cancer: General Considerations Molecular Pathogenesis of Thyroid Cancer James Figge
57
Epidemiology of Thyroid Cancer James Figge
77
........................................................................................................ ........................................................................................................
Radiation and Thyroid Cancer James Figge, Timothy Jennings, and Gregory Gerasimov
...............
Classification of Thyroid Malignancies James Oerteland Yolanda Oertel.
................................................................... xiii
85
117
xiv
Contents
10 Thyroid Cancerin Children and Adolescents Merrily Poth....................,..................
...............................................................
121
11 Immunologic Aspects of Thyroid Follicular Neoplasms
James R. Baker, Jr...........................................................................................
129
General Considerations4 12 Radioiodine Therapy of Thyroid Cancer:
.............................................................
13
14
15
Gerald Johnston and Diane Sweeney
I47
Radioiodine Therapy of Thyroid Cancer: General Considerations-11 Side Eflects of Radioiodine Therapyfor Thyroid Cancer Diane Sweeneyand Gerald Johnston
.............................................................
155
Recombinant Human Thyrotropin Matthew D. Ringel
............................................................................................
163
Chemotherapy for Thyroid Cancer Lawrence S. Lessin and My0 Min
I79
...................................................................
Part 111 Differentiated Tumors of the Thyroid Gland: A. Papillary Carcinoma 16
17
18
19
Papillary Carcinoma: Clinical Aspects Leonard Wallfofs
............................................................................................
185
Papillary Carcinoma: Cytology and Pathology James Oerteland Yolanda Oertel................
193
Surgical Approach to Papillary Carcinoma Orlo H. Clark....................................................................................................
209
Differentiated Thyroid Carcinoma:Radioiodine TherappI Gerald Johnston and Diane Sweeney
213
....................................................
.............................................................
20
Chemotherapy of Differentiated (Papillary or Follicular) Thyroid Carcinoma Lawrence S. Lessin and My0 Min...........,....................................................... 221
21
Management of Papillary Thyroid Carcinoma: External Radiation Therapy Robert L. White 225
.................................................................................................
xu
Contents 22 23
Papillary Thyroid Cancer:Follow-Up Henry B.Burch
................................................................................................
Radioiodine Treatment of Thyroid Cancer-11: and Diagnostic I3'I Uptake Diane Sweeneyand Gerald Johnston
229
Maximizing Therapeutic
.............................................................
239
24
An Approach to the Management of Patients with Scan Negative, Thyroglobulin Positive, Differentiated Thyroid Cancer: Alternative Imaging Procedures Leonard Wartofs.............................................................................................. 251
25
Papillary Thyroid Cancer:Prognosis Henry B. Burch
263
Papillary Cancer:Special Aspects in Children Merrily Poth
267
26
................................................................................................
......................................................................................................
Part IV Differentiated lbmors of the Thyroid Gland: B. Follicular Carcinoma 27
Follicular Thyroid Carcinoma:Clinical Aspects
Leonard Wartofs.............................................................................................. 279 28 29
Pathology of Follicular Cancer James Oerteland Yolanda Oertel....................................................................
289
Surgical Management of Follicular Cancer Orlo H. Clark....................................................................................................
297
30
Follicular Carcinoma of the Thyroid: External Radiation Therapy Robert L. Whiteand Leonard Wartofs........................................................... 301
31
Follicular Thyroid Cancer:Follow-Up Henry B. Burch
307
Follicular Thyroid Cancer:Prognosis Henry B. Burch
311
32
................................................................................................
................................................................................................
Contents
mi 33
Follicular Thyroid Cancer:Special Aspects in Children and Adolescents Merrily Poth
......................................................................................................
315
Part V Undifferentiated Cancers: A. Anaplastic Carcinoma 34 Anaplastic Carcinoma: Clinical Aspects
Steven I. Sherman
............................................................................................
35 Anaplastic Carcinoma: Pathology James Oerteland Yolanda Oertel.................................................................... 36 37
319 327
Anaplastic Carcinoma Management:Surgery Orlo H. Clark................... .................................................................................
333
Chemotherapy of Anaplastic Thyroid Cancer Lawrence S. Lessin and My0 Min
337
...................................................................
38
Management of Anaplastic Carcinoma:External Radiation Therapy Robert L. Whiteand Leonard Wartofs. .......................................................... 341
39
Anaplastic Carcinoma: Prognosis Steven I. Sherman
............................................................................................
345
Part VI Undifferentiated Cancers: B. Lymphoma 40
ThyroidLymphoma Steven I. Sherman
............................................................................................
41 ThyroidLymphoma: Pathology James Oerteland Yolanda Oertel....................................................................
351
359
Part VI1 Undifferentiated Cancers: C. Medullary Carcinoma 42 43
MedullaryThyroidCarcinoma Douglas W.Ball
365
MedullaryThyroidCancer: Pathology James Oerteland Yolanda Oertel...........................,........................................
383
................................................................................................
Contents 44
xvii
Medullary Carcinoma of the Thyroid: Nuclear Medicine Imaging and Treatment Diane Sweeney and Gerald Johnston
389
Management of Medullary Carcinoma of the Thyroid: Surgery Orlo H. Clark
399
Medullary Carcinoma Management: External Radiation Therapy Robert L. White and Leonard W a ~ o f s
401
Medullary Carcinoma of the Thyroid: Chemotherapy Lawrence S. Lessin and My0 Min
405
.............................................................
45
46
47
....................................................................................................
............................................................
...................................................................
Part VI11 Miscellaneous and Unusual Cancersof the Thyroid 48
49
Pathology of Miscellaneous and Unusual Cancersof the Thyroid James Oertel and Yolanda Oertel.
...................................................................
411
Clinical Aspects of Miscellaneous and Unusual Types of Thyroid Cancers Matthew D. Ringel, Kenneth D. Burman, and Barry M. Shmookler 421
...........
Part IX Future Directions 50
Thyroid Cancer: DNA Ploidy, Tumor Markers, and Cancer-Causing Genes Michael McDermott 455
51
New Approaches to Chemotherapy for Thyroid Cancer Lawrence S. Lessin and My0 Min
491
Advances in Radiotherapy for Thyroid Cancer Robert L. White
.................................................................................................
495
Index ..........................................................................................................................
.49 7
52
.........................................................................................
...................................................................
Contributors
JAMESR. BAKER, JR., MD Associate Professor, Internal Medicine and Pathology, and Chiefl Division of Allergy, University of Michigan Medical Center, Ann Arbor, M DOUGLAS W. BALL,MD Assistant Professor of Medicine and Oncology, Johns Hopkins Oncology Center, Johns Hopkins University School of Medicine, Baltimore, MD MANFRED BLUM, MD Professor of Clinical Medicine and Radiology and Director of Nuclear Endocrine Laboratory, New York UniversityMedical Center, New York, NY HENRY B. BURCH, MD Assistant Chiefl Endocrinology Division, Department of Medicine, Walter Reed Army Medical Center, and Assistant Professor of Medicine, Uniformed Services University of the Health Sciences, Bethesda, MD KENNETH D. BURMAN, MD Director, Division of Endocrinology, Washington Hospital Center, Clinical Professor of Medicine, Georgetown University School of Medici and George Washington University School of Medicine and Health Sciences, Washington, DC, and Professor of Medicine, Uniformed Services University of Health Sciences, Bethesda, MD ORLO H. CLARK, MD Professor of Surgery, University of California, San Francisco, School of Medicine and Chief of Surgery, Mt. Zion Hospital, San Francisco, CA JAMESFIGGE, MD Associate Professor of Medicine and Director, Thyroid Cancer NY Program, State Universityof New York at Albany, St. Peter’s Hospital, Albany, GREGORY GERASIMOV, MD Director, Department of Endocrinology, Russian Endocrine Research Center, Moscow, Russia TIMOTHY JENNINGS,MD Associate Professor of Pathology,Albany Medical Center, Albany, NY GERALD JOHNSTON,MD Director, Division of Nuclear Medicine, Washington Hospital Center, Washington,DC LAWRENCE S. LESSIN, MD Director, Washington Cancer Institute, Washington Hospital Center, and Clinical Professor of Medicine, George Washington University School of Medicine and HealthSciences, Washington, DC ERNEST L. MAZZAFERRI, MD Professor of Medicine and Chairman, Department of Medicine, Ohio State UniversityMedical Center, Columbus, OH MICHAEL T. MCDERMOTT, MD Associate Professor of Medicine, Division of Endocrinology, Department of Medicine, University of Colorado Medical Center, Denver, CO MYOMIN,MD Division of Hematology/Oncology, WashingtonCancer Institute, WashingtonHospital Center, Washington, DC xix
JAMESOERTEL, MD Chairman Emeritus, Department of Endocrine Pathology, Armed Forces Institute of Pathology, ~ashington,DC YOLANDA C. OERTEL, MD * Director E~eritus, ~ytopathologyDivision, George ~ashingtonUniversity School of Medicine and Health Sciences and Department of Pathology, WashingtonHospital Center, Washington,DC MERRILY POTH,MD * Professor, Department of Pediatrics, Un~ormedServices Universi~of the Health Sciences, ~ethesda,MD WITHEW RINGEL,MD * Assistant Professor of Medicine, U n ~ o ~Services ed and George r ~ ~ h i n Universities, ~ o n and Co-Director,Laboratory of M o l e ~ l aEndocrinolo~, Medlantic Research I ~ t i ~ tDepar~ent e, of Medicine, W~hington Hospital Center, ~ ~ h i n g t oDC n, STEVEN I. SHERMAN * Section of Endocrine Neoplasia and Hormonal Disorders, M.D. Anderson Cancer Center, and Assistant Professor of Medicine, ~ n i v e r sof i~ Texa+Houston Medical School, Houston, TX BARRY M, SHMOOKLER, MD Department of Pathology, ~ashington Hospital Center and Medl~nticResearch Institute, ~ashington,DC DIANESWEENEY, MD * Associate Director, Division of Nuclear ~ e d i c i n e ~ashington , Hos~italCenter, ~ashington,DC LEONARD WARTOFSKY ,MD ~ h a i ~ aDepartment n, of ~edicine,~ashington~ospital ed Universi~ Center, and Professor of Medicine and Physiolo~,U n ~ o r ~Services of Health Sciences, and Clinical Professor of Medicine, ~eorgetown,Howard, and George ~ ~ h i n g t oUniversity n Schools of Medicine, ~ashington,DC R O ~ E RLARRY T WHITE,MD * Director of Education and Research, Depart~entof ~adiationOncology, Washington Cancer Institute, ~ a s h i n gHospitul t~~ Center, Washin~ton,DC 0
0
9
I The Thyroid Nodule
1 The Thyroid Nodule Pathogenesis, Evaluation, and Risk of Malignancy Leonard Wartofsky INTRODUCTION
The clinical management of nodular thyroid disease remains an active topic of (1-10), although a consensus on guidelines for the diagnosis discussion and controversy and management of thyroid nodules has been reached by an authoritative body (11). Palpable nodules of the thyroid are frequently encountered in clinical practice, and their evaluation requires the physician to be familiar with a growing of diagnostic number tools in order to identify those nodules representing cases of carcinoma requiring surgical intervention. More cost effective and precise diagnosis arguablymay be best performed by an endocrinologist (12). PREVALENCE
Solitary nodulesof the thyroid gland are presentin about 6.4% of women and 1.5% of men (3,4,6). The prevalence is low in children (about 1.5%), and increases linearly with age. Many single palpable nodules thought to be solitary are actually in a mu as manyas50%of lar thyroid gland. Autopsy studies indicate thyroid nodules in (1,3,5). consecutive necropsies, although many may be small and clinically inapparent High-resolution ultrasound has identified nodules 1340% in of patients being evaluated for nonthyroid problems (13-15). Thus, a discrepancy exists between the true prevalen of thyroid nodules and that apparent by physical examination. Generally, nodules must approach 1 cm diameter to be recognized on palpation. The prevalence of nonpalpable nodules incidentally detected by ultrasound (“incidentalomas”) is 3040% in autopsy 4%. studies and 19-67% in clinical studies(16) with an average risk for malignancy of PATHOGENESIS
While the cause of thyroid nodules is not known, associations with iodine deficiency and indirect evidence of thyrotropin (TSH) effect suggest possible relationships. Cold nodules occur about 2.5 times more frequently in areas of low naturally occurring of thyroid iodine. In rats, iodine deficiency enhances TSH secretion and the development nodules, someof which are malignant(17). The relationship to TSH is unclear, although
From: Thymid Cancer: A Comprehensive Guideto Clinical Management Edited by: L. Wartofsky 0 Humna Press Inc., T o t m , NJ
3
4
Wartofsky
the response of benign nodular thyroid enlargement to thyroxine (18), as well as the improved prognosis of patients with papillary thyroid cancer treated with thyroxine (19), suggest a role of TSH in human neoplasia. Radiation exposure can cause thyroid neoplasia, with a linear relationship between radiation doses up to 1800 cGy and the incidence of thyroid nodules and cancer. The increased risk of clinically significant thyroid cancer associated with prior radiotherapy to the head and neck given for thymic enlargement, tonsillitis, acne, and adenitis, is around 3% (20-22). Radiation exposure as a child is more likely to produce thyroid neoplasia than similar exposure at a later age, possibly related to greater cellular mi activity at the earlier age of insult. Among individuals in the United States receiving head and neck irradiation in childhood, palpable nodules are found in 1629% and carcinoma in one-third of these nodules (23,24). Most nodules tend to occur within 10-20 yr of exposure, but the risk may exist for over 35 yr. The irradiated thyroid gland often presents with multiple nodules and at surgery the lesion of initial concern may prove to be benign, although one or more carcinomas will be found elsewherein the gland. Thus, those nodules associated with a radiation history do not demonstrate a reduction of cancer risk when the thyroid contains multiple nodules. The dramatic increase in thyroid nodules and thyroid cancer occurring in Belarus after the 1986 Chernobyl nuclear disasteris discussed in detailin Chapter 8 by Figge and associates. Higher doses of irradiation, suchas those used for Hodgkin’s disease (>2000 rads), and I3’I therapy do not appear to be related to subsequent development of thyroid carcinoma. In both cases, the high-dose exposure with attendant cell destruction, f and hypothyroidism may serve to attenuate any carcinogenic effect.
DIFFERENTIAL DIAGNOSIS
As indicatedin Table 1, the differential diagnosis of apparent thyroid nodules cove (Id). Most (2740%) true intrathyroidal nodules will awiderangeofpathology represent colloid adenomas or simple follicular adenomas (2640%). About 5% of to thyroid nodules are classified as “hot” on the basis of a relative increased ability trap iodide. Most of these hot nodules are autonomously functioning, and more than one-half in patients over 60 will cause hyperthyroidism. Twenty percent of nodules (25,26),compared with greater than 3 cm diameter are associated with hyperthyroidism 2% of smaller lesions. Although most toxic autonomous nodules secrete both thyroxin (T4) and triiodothyronine(T3), elevations of T3 or T4alone occasionally may be seen. Moreover, even when T3 and T4levels are “normal,” low serum TSH by a sensitive (TRH) stimulation assay or a blunted response of TSH to thyrotropin-releasing hormone is common, suggesting supraphysiologic iodothyronine production. Cancers are found in 10-14% of patients presenting with palpable thyroid nodules (Id). In the United States, papillary carcinomas account for about 70% of all thyroid cancer, with follicular being the next most common (20-25%), and anaplastic and medullary thyroid carcinomas each comprising 3-5%. The thyroid gland has a rich blood supply, and a thyroid nodule occasionally may represent a secondary or neoplasm, including malignant melanoma, and renal cell, breast, and bronchogenic of new thyroid cancer cases appears to be carcinomas. The frequency of diagnosis increasing with about 11,300 cases and lo00 thyroid-related deaths occurring yearly
Pathogenesis, Evaluation, and Risk of Malignancy
5
Table 1 Differential Diagnosisof Apparent Thyroid Nodules Benign Thyroid Neoplasms Other Thyroid Abnormalities Thyroiditis adenoma Follicular Thyroid Colloid Hemiagenetic Simple Fetal sarcoidosis) (e.g., Granulomatous Embryonal disease Htirthle cell Nonthyroidal Lesions Papillary adenoma Lymphadenopathy Teratoma Aneurysm Lipoma Thyroglossal duct cyst C-cell adenoma Parathyroid cyst Dermoid cyst Parathyroid adenoma Malignant Thyroid Neoplasms Laryngocele hygroma Cystic carcinoma Papillary Follicular carcinoma Medullary thyroid carcinoma Anaplastic carcinoma Metastatic carcinoma Sarcoma Lymphoma
in theUNted States in1989 (271, whereas the American Cancer Society projects 18,800 new cases in 1999 with perhaps 1500 deaths. Autopsy studies have revealed occult thyroid cancer in 6% of autopsies in North American series (28). There is general agreement that these small, occult, and mostly papillary cancers are of little or no clinical significance, and their increased prevalence does not correlate with an increase in the death rate from thyroid carcinoma(29). DIAGNOSTIC EVALUATION (Fig. 1) Sincethevastmajorityofthyroidnodulemorbidity is relatedtothoselesions representing carcinoma, the evaluation is focused on identification of those nodules that may be malignant. HISTORY AND PHYSICAL EXAMINATION The single most important historical risk factor for canceris exposure to radiation. It is important to determine the age at time of exposure, exact region of the body of radiation to the thyroid. Although irradiated, and, if possible, the type and dose women are more prone to thyroid nodules and cancer than men, the probability of cancer is higher among men with nodules. The incidence of thyroid cancer increases with age, but a higher percentage of nodulesin patients less than 20 years of age will be malignant. Thyroid lymphoma should be considered in patients with rapid thyroid enlargement and a previous diagnosis of Hashimoto’s thyroiditis, especially in those as adominant“cold”nodule,and womenoverage 50. Suchlesionsmaypresent
6
Wartofsky Table 2 Physical and Historical Factors Increasing Risk of Carcinoma in a Thyroid Nodule
ysical Cervical
History Radiation history Family of MEN wthDocumentedgrowth Rapid cord Vocal Hoarseness Pain Homer’s Dysphagia Respiratory obstructive symptoms Growth on thyroxine medication
Firmness Fixation
there is often coincident diabetes mellitus. A family history of pheochromocytoma, hypercalcemia, mucosal abnormalities, or medullary thyroid carcinoma raises suspi of the latter diagnosis as part of a multiple endocrine neoplasia (MEN) syndrome. While family history of benign goiter may be reassuring, the rare Pendred’s syndrom of familial goiter and deaf mutism is associated with a higher cancer risk(1,3). Most thyroid nodulesare discovered incidentally in asymptomatic patients. As noted in Table 2, a number of symptoms or physical findings are felt to be more common in malignant than benign nodules, although as few 5-10’3as 1 of patients with malignancy present with symptoms. Patients with advanced disease may present with lymp all thy, growth of hard nodules, thyroid pain and tenderness, and vocal cord paralysis, of which point to the likelihood of malignancy.
REFERENCES 1. AshcraftMW,VanHerleAJ. Managementof thyroid nodules. I. History and physical examina
tion, blood tests, X-ray tests, and ultrasonography. Head Neck Surg 1981; 3:216-230. 2.Ashcraft MW, VanHerle AJ. Management of thyroid nodules. It. Scanning techniques, thyroid suppressive therapy, and fine needle aspiration. Head Neck Surg 1981; 3:297-322. 3. Mazzaferri EL. Management of a solitary thyroid nodule. N Engl JMed 1993; 328:553-559. 4. Burch HB. Evaluation and management of the solid thyroid nodule. Endocrin Metab Clin N Amer 1995; 24:663-710. 5. Sheppard MC, Franklyn JA. Management of the single thyroid nodule. Clin Endocrinol 1992; 37:398-401. of a solitary thyroid nodule. J Clin 6. Ridgway EC. Clinical review 30: clinician’s evaluation Endocrinol Metab 1992; 74:231-235. 7. Molitch ME, Beck JR, Lheisman M, Gottlieb JE, Pauker SG. The cold thyroid nodule: an analysis of diagnostic and therapeutic options. Endocr Rev 1984; 5:185-199. 8. Hermus A R , Huysmans DA. Treatment of benign nodular thyroid disease. N Engl J Med 1998; 338:1438-1447. 9. Gharib H, Mazzafeni EL. Thyroxine suppressive therapy in patients with nodular thyroid disease. Ann Intern Med 1998; 128:386-94. of benign thyroid nodules: have we defined a benefit? An 10. Ridgway EC. Medical treatment Intern Med 1998; 128:403-405. 11. Singer PA, Cooper DA, Daniels GH, Ladenson F’W, Greenspan FS, Levy EG, et al. Treatment
Pathogenesis, Evaluation,
and Risk of Malignancy
7
guidelines for patients with thyroid nodules and well differentiated thyroid cancer. Arch Int Med 1996; 156:2165-2172. 12. Ortiz R, Hupart KH, DeFesi CR, Surks MI. Effect of early referral to an endocrinologist on efficiency and cost of evaluation and development of treatment plan in patients with thyroid nodules. J Clin Endocrinol Metab 1998; 83:3803-3807. 13. Carroll BA. Asymptomatic thyroid nodules: incidental sonographic detection. Am J Roentgenol 1982:138:499-501. 14. Horlocker 'IT, Hay JE, James EM. Prevalence of incidental nodular thyroid disease detected Gaitan E (Eds.) during high resolution parathyroid ultrasonography. In Medeiros-Net0 G, Frontiers in Thyroidology, Vol. 1, Plenum Press, New York, 1986, pp. 1309-1312. 15. Brander A, Viikinkoski P, Nickels J, Kivisaari L. Thyroid gland: ultrasound screening in middle aged women with no previous thyroid disease. Radiology 1989; 173507-510. 16. Tan GH, Gharib H. Thyroid Incidentalomas: Management approaches to nonpalpable nodules discovered incidentallyon thyroid imaging. Ann Intern Med 1997; 126:226-231. 17. BorrowGN. The thyroid: nodules and neoplasia. In Felig P, Baxter JD, Broadus AE, Frohman LA (Eds.) Endocrinology and Metabolism, McGraw-Hill, New York, 1987, pp. 473-507. 18. Greer MA, Astwood EB. Treatment of simple goiter with thyroid. J Clin Endocrinol Metab 1953;13:1312-1331. RL, Oertel E, KemmererWT, Page CP.Papillary thyroid carcinoma: 19. Mazzafeni EL, Young the impact of therapy in 576 patients. Medicine 1977; 56:171-196. 20. Same D, Schneider AB. External radiation and thyroid neoplasia. Endocrin Metab Clin North h e r 1996; 25:181-196. 21. Griffin JE. Management of thyroid nodules. Presented at Southwestern Internal Medicine Conference. Am J Med Sci 1988; 296:336-347. 22. Ron E, Kleineman RE, Boice JD Jr, LiVolsi VA, Flannery JT, Fraumeni JF Jr. A populationbased case-control studyof thyroid cancer. J Natl Cancer Inst 1987; 79:l-12. 23. Favus MJ, Schneider AB, Stachura ME, Arnold JE, Ryo W, Pinsky SM, et al. Thyroid cancer occurringas a late consequence of head-and-neck irradiation. N Engl J Med1976; 294~1019-1025. of 24. DeGroot LJ, Reilly M, Pinnameneni K, Refetoff S. Retrospective and prospective study radiation-induced thyroid disease,Am J Med 1983; 749524362. 25. Hamburger JI. Evolution of toxicity in solitary nontoxic autonomously functioning thyroid nodules. J Clin Endocrinol Metab 1980; 50:1089-1093. 26. Hamburger JI.The autonomously functioning thyroid nodule: Goetsch's disease. Endocrine Rev 1987; 8 : 4 3 9 4 7 . 27. Silverberg BS, Lubera JA. Cancer statistics 1989. Cancer 1989; 39:3-7. 28. Sampson RJ, Woolner LB, Bahn RC. Occult thyroid carcinoma in Olmsted County, Minne1974; sota. Prevalence at autopsy compared with that in Hiroshima and Nagasaki. Cancer 34:2070-2076. 29. Sampson RJ. Prevalence and significance of occult thyroid cancer. Radiation-Associated Thyroid Carcinoma, DeGroot LJ, Frohman LA, Kaplan EL, Refetoff S (Eds.) Grune & Stratton, New York, 1977, pp. 137-143.
8
2 Nonisotopic Imaging of the Neck in Patients with Thyroid Nodules or Cancer Manfred Blum
This chapter discusses the clinical relevanceof nonisotopic imaging of the neck in patients in whom thyroid canceris suspected or those with a history of thyroid cancer. twohas aspects. Initially, The traditional approach to the diagnosis of thyroid cancer the physician must determine if a thyroid nodule is benign or malignant. Then, after a thyroid cancer has been removed surgically, residual or recurrent cancer must be detected early, accurately, and safely. However, recent societal economic constraints in clinical practice may be on medicine and the greater acceptability of uncertainty perceived to suggest a lower diagnostic standard ( I ) . This should not be the case. While uncertainty is and has always been inescapable, it is minimized by integrating knowledge about the disease process, clinical skills and judgment. Economy can be enhanced by theoptimaluseofcurrenttechnology,which,forthyroidcancer,isthefocusof this presentation. Although thyroid nodules are very common, fewer than 5% are malignant. Considerable time, money, and health-care resources are spent to identify this small number of 6 7 % of all people who have patients with cancer and to spare the approximately benign thyroid nodules from needless operations. From the perspective of economists and those who fund health care for a large population of people, cost is particularly important with thyroid cancer because it is very uncommon,is generally slow-growing, usually cured by the initial surgical procedure. Furthermore rarely results in death, isand recurrence of this malignancy is also uncommon and rarely results in death. Yet, to the patient and family who are insured or who are members of a managed care syste is not a great concern and who seek medical attention this for problem, cost containment at the time of illness. The clinician is challenged to meet the expectations of both the patient and the payer and to uncover those few patients who need additional treatmen CHOOSE THE OPTIMAL AND ESSENTIAL IMAGING PROCEDURE AND CORRELATIVE DATA WITH THE CLINICAL SITUATION The diagnosis of thyroid nodules is beyond the scope of this discussion but does bear on the issue because the same principles apply to cancer(2,3).
Fmm: Thymid C a n c e r : A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 8 Humana Press Inc., T o t m , NJ
9
10
Blum
The medical history and clinical examination alert the clinician to the problem and direct the selection of testing. In the past, it was perceived that a solitary nodule had the highest risk cancer. Now there appears to be controversy among experts about the cancer risk of a solitary nodule in an otherwise norm~-feelingthyroid gland as opposed to a nodule in a goiter. The perception of the controversy clouds the issue and confuses the clinician. ~ u l t i p l ebenign thyroid nodules are very ~ommon,increasingly so with age, particularly in women. Modern technology, especially ultrasonography, has revealed that approximately 30-50% of people have small nodules, that have been called incidentalomas, and are not palpable or discoverable by scintiscanning (4). Therefore many patients with an apparently s o l i t ~ nodule y really have multiple nodules. The chance that any one nodule in a clinical goiter or in a thyroid gland that has subclinical nodules is cancerous is exceedingly small unless there are clinical characteristics that suggest a higher risk. The clinical implication is that a nodule may not demand immediate diagnostic or therapeutic attention because of the low virulence of most thyroid tumors, but it cannot be ignored because it could be malignant. Factors that are associated with an increased likelihood that a specific nodule is a cancer have been identified. The nodules that are the most suspicious of thyroid cancer, either before any surgery or when there is a new nodule after a prior thyroid cancer had been removed, occur when there is a history of therapeutic i~adiationto the head or neck, hard consistency, lymphadenopathy or evidence of invasion. Also suspicious is enlargement of a nodule to make it a dominant mass in a goiter. A nodule may be associated with a 10% risk of cancer if it does not accumulate radioiodine as efficiently as the rest of the thyroid gland, less than 1% if it accumulates the isotope more avidly, and 30% if there is a history,of radiograp~ctherapy. Cancer is several times more likely in a male than a female, and much more on in a child than an adult. Furthermore, because of the multifocal tendency o id cancer, there is a high risk of cancer when a nodule emerges in the con~alateralthyroid tissue after surgical lobectomy for cancer, especially when thyrotropin (TSH) is suppressed. The risk of recurrent cancer is exceedingly high when there is persistent painless regional lymphadenopathy after a thyroidectomy for cancer. Generally, clinical methods such as visual inspection and palpation together with the history provide adequate clues about the thyroid region and may be sufficient for diagnosis and a management decision. Only when these data are not adequate, and for specific indications, is it necessary to proceed to imagin~procedures. Then, the optimal test must be chosen to achieve the diagnosis safely, expeditiously, and economically. A working knowledge of the imaging methods and an understand anatomy of the thyroid gland are required so that clinician may presc that is best designed to answer the patient’s problem. There are pitfalls in testing. A precise questio must be relayed to the radiologist so that the images may be interpreted in context. It must be appreciated that imaging data alone may mislead unless integrated with the rest of the clinical situation. It is important when interpreting images to be aware of the principles of statistics and the limitations inherent in the method. The beauty and apparent detail of the images of the thyroid gland and its surroundings are simultaneouslya promise for enhanced diagnosis and a hazard for misinformation; overinterpretation may lead to the wrong diagnosis. Synthesizing the images from the different t e c ~ i q u ewill s provide appropriate anatomic
Imaging Nonisotopic Neck
of the
11
information and avoid traps and controversies that occur all too frequently when the diverse data and images are interpreted independently and without meticulous attention to the clinical problem. Although the most cost-effective diagnostic tool for a nodule in the thyroid region of the neck is fine needle aspiration biopsy and cytologic analysis, regional imaging can also provide useful ancillary information. Radionuclide scanning identifies tissue that is iodine-avid, whichis most informative after thyroidectomy for cancer to demonstrate that a nodule is thyroid tissue. Otherwise, its role in assessing a patient with a thyroid nodule is controversial (see Chapters 1 and 4). Ultrasonography may be used todepicttheregionalanatomyaccurately,safely,andeconomically.Amongother capabilities, it canidentifythesolidcomponent of acomplexnodule,providean anatomic guide for fine needle aspiration, document comparative size of nodules in patients who are under observation (especially when they are taking TSH-suppressive therapy), detect a small nodule in patients who were exposed to therapeutic irradiation of the head or neck, and reveal even nonpalpable recurrent thyroid cancer after surgery. The more costly sectional imaging methods, computed tomography (CT) and magnetic resonance (MR), play no role in management of the patient with an average thyroid nodule. CT and MR images may be important to answer a specific clinical question, for example 1. To evaluate patients with an invasive nodule 2. To depict the gross anatomywhenthere are crypticsymptoms,confusingfindingson palpation,conflictingresultsfromotherimagingtests,oralteredanatomyafterother regional operations 3. To identify metastatic malignancy, especially in regions that are blind to sonography
ULTRASONOGRAPHY (SONOGRAPHY) Ultrasonography plays an important role in the diagnostic evaluation of the thyroid gland and the surrounding tissues because its of safety, low cost, clear images displayed in real time, remarkable resolving power, and wide availability(5-8). Sonography has gained the primary role to depict the thyroid region in the patient with thyroid cancer. Figure 1 shows a postoperative thyroid bed that is free of tumor. Figure 2 shows a local recurrence of cancer after surgery even before the lesion is palpable. Figure 3 shows ultrasound imaging of regional nonpalpable adenopathy.
Principles and Method Gray Scale Ultrasonography
Ultrasonography involves the intermittent generation of a pulse of sound energy and the reception of the reflected echoes to produce an image of the tissues that have been traversed by the sound. Current technology produces high-resolution thyroid images million MHz. These frequenby employing sound frequencies between 5 million 10 and cies are well above the range audible by the human ear. The sound waves penetrate tissues and a portion of the energy is reflected at tissue interfaces up to a 5-cm depth using typical equipment. The superficial location in the neck of the thyroid gland or regional metastases is well within this limitation. Current clinical equipment provides high resolution of structures as small as 2 mm. By contrast, all of the other imaging
12
Blum
Fig. 1. Use of sonography in routine follow-up; no lesion. Sonogram from 53-year-old male who had a total thyroidectomy 15 years previously for a follicular thyroid cancer. There is no evidence of thyroid tissue or tumor in the left thyroid bed (or right, not shown). T, trachea; SM, sternocleidomastoid muscle;C, left common carotid artery; J,jugular vein.
methods are considerably less sensitive. Linear array transducers are preferred to sec transducers because they minimize distortion, produce superior images from the very superficial portions, and facilitate correlation with images derived by other techniqu Some factors that omit the usefulness of sonography include attenuation of the highfrequency sound waves in deeper tissues, which may be an issue with very larg distortion by air-filled structures such as the trachea, blockade of the ultrasound by calcified deposits, and inaccessibility of substernal tumor. The process of producing an optimal ultrasound image requires an operator whois familiarwiththeanatomyandthesuspectedpathology.Ultrasoundscanning is a subjective art, and skill improves with experience.A diligent search for the answer to the question that has been posed by the clinical problem is required. The average technician cannot optimally perform the procedure and then submit the films for later interpretation. Rather, a well-trained and experienced sonography technician, active participation by the radiologist or sonologists, and input by the clinician enhance the usefulness of the procedure. Indeed, many clinicians perform their own sonograms at the time of the patient’s visit. Images are best obtained with the patient lying supine with the neck maximally extended, consistent with comfort. Patience and attention to positioning, body habitu and factors such as arthritis will be rewarded with improved images. Anatomic landmarks, the thyroid gland, and abnormalities must be carefully palpated and the noted. Especially in patients with thyroid cancer, the entire region must be examined in the midline completely in both the transverse and longitudinal planes, beginning and extending laterally to encompass nodal regions. Scanning must be done from the sternal notch to the chin. The entire length of the carotid sheath must be explored to identify enlarged lymph nodes. The esophagus can be differentiated from adenopathy
Nonisotopic Imaging of the Neck
13
A
I
Fig. 2. Use of sonography to detect a very small nonpalpable local recurrence of thyroid cancer. MRI confirms the abnormality. Sonogrm of the left side of the neck from a 30-yearold female who hada thyroidectomy fora 1-cm papillary thyroid cancer in the left lobe 4 years previously. At the time of an annual reevaluation there was no palpable mass in the neck. TSH (A) Two films from a transverse sonogram. was suppressed and thyroglobulin was not detectable. &j?: Doppler-augmented examination highlighting the left common carotid artery as a bright circle. Medial to thatis a hypoechoic mass. Right: Non-Doppler study. The noduleis 4.4 mm I3'I, andfineneedleaspirationbiopsywasnot wide.Sincethenoduledidnotaccumulate toto obtain more anatomic information. (B) Coronal diagnostic, an MRI of the neck was donetry view from the MRI examination. A nodule is seen to be left of the trachea. (C)Transverse TZweighted image shows that the mass is brighter than muscle, suggesting cancer, and also shows adenopathy. Thyroid malignancy in the nodule and the lymph nodes was confirmed at surgery. T, trachea;+ +, mass; L, lymph node; C, left common carotid artery; J, jugular vein.
14
Blum
A
Fig. 3.Use of sonography in routine follow-up; nonpalpable adenopathy MRI found. confirms the abnormality.A 48-year-old muscular male who had a total thyroidectomy 5 years previously for a papillary thyroid cancer. An examination was physiological, third generation TSH was 0.04 pIU/ml,and thyroglobulin less than 0.5 ng/dl. while he was taking200 pg of L-thyroxine daily. A sonogram of the neck was done as part of routine reevaluation. (A) One film of the sonogram demonstrating a rounded pathological lymph node with loss of the fatty h i l u m . The rectangle demarcates a Doppler examination. The bright spots indicate blood vessels. was unwilling to accept the need for treatment until anMRI of the neck also demonstrated a as a bright mass on the =-weighted image, lesion. (B) One MRI film. The lesion was confirmed TSH 'and consistentwithcancer.Subsequently,thesuppressivetherapywasdiscontinued, I3lI. L, thyroglobulin levels rose, a whole-body I3'I scan was done, and he was treated with pathological lymph node.
by having the patient sip water, which is made possible because images are obtained in real time.
Color Flow Doppler Imaging Color flow doppler imaging of the thyroid gland adds dynamicflow information to a static gray scale image. Color-encoded signals are superimposed on real-time gray scale images to indicate both the direction (phase shift) and the velocity (frequency (9), which shift) of blood flow and are useful in depicting the degree of vascularity
Nonisotopic Imaging of the Neck
15
B
Fig. 3. (Continued)
may be diagnostically useful when evaluating tumors but more often to identify blood vessels (Fig. 2).
Correlation of the Ultrasonic Properties of Tissues with Pathology This discussion is derived from experience with the current commercially available high-resolution ultrasound equipment using a7.5-10 MHz transducer and a reviewof the literature. It seems safe to anticipate that enhanced images and,it is hoped, better correlation with pathology will occur with the extremely high-resolution devices who 1995 RSNAmeeting.Experimentalscannershavea prototypeswereshownatthe mechanized drive, video spatial depiction of the signal from the ultrasound probe that is quickly digitalized, and computer reconstructed to produce three-dimensional imag There is a growing dilemma in clinical management about the meaning of nodules that have been detected with ultrasound. Recent technological advances have permitte the detection of nodules in the millimeter range, which is an enormous advance but, at the same time, is the source of problems (5). Sonography can detect nonpalpable micronodules, “incidentalomas,” which are of indeterminate significance but are usually benign and not in need of therapy. Some of these lesions may indicate occult thyroid cancer, whose incidence in patients varies from a few percent in the United States to perhaps as much as 20% in other parts of the world. Data suggest that these lesions are of no clinical consequence in most patients and their discovery during echography mayoccasionneedlessconcernandtherapy.Yet,somesmallnodulesareclinical
16
Blum
carcinomas, a few of which metastasize, and a very rare one may cause death (10). Therefore, finding an incidentaloma cannot be dismissed. As discussed below, when of cancer, such as a history of irradiation, there is an incidentaloma and an increased risk ultrasound-guided aspiration biopsy may be diagnostically useful. Correlation of the ultrasonic properties of tissues and histopathology is poor. Sonography merely depicts the anatomy. Benign and malignant lesions are not differentiated. Nevertheless certain associations with diseasesbecan made using state-of-the-art equip ment. However, these associations reflect statistical probabilities that have been draw from relatively small numbers of patients and offer limited insight into a single person’s problem. (11-16). Productive Early correlations were made in patients with subacute thyroiditis analyses of ultrasound pattern of thyroid tumors occurred later, except that the findin of an ultrasonically unique area in a patient with a goiter or thyroiditis was employe (17). to alert the clinician to a second pathological entity, including cancer or lymphom Analysis of the criteria that have been suggested for identifying thyroid canc thyroid nodules have yielded conflicting results (17J8). These criteria are not useful either preoperatively when cancer is suspected or postoperatively in the residual thyroi tissue after cancer has been removed elsewhere in the gland. Sonographic characteristic of nodules that have been examined for correlation with malignancy include t of the echos, sharpness of a boundary, the presence of a “halo,” calcifications, and internal structure (9,19). Although many thyroid cancers are less echogenic than the 2 and 4), manybenignnodulesarealso surroundingnormalthyroidtissue(Figs. of benign disease. hypoechoic. Noris the findingof hyperechogenicity a reliable index a distinct border surround There is little, if any diagnostic value in knowing that isthere ing a nodule. Nevertheless, selected grossly invasive nodules may be shown to exhibit a poorly defined margin in the invaded area. In addition to simple margination of a nodule, some nodulesare partly or completely surrounded by a sonolucent halo. Som echographers have suggested that a thin halo may be more likely associated wi disease and a thick, partial halo may be more often associated with malignant disea (20). It seems approOthers have not been able to confirm either of these conclusions priate to consider the halo as an interface between two different types of thyroid tiss that in some cases represents a capsule, in others compressed or atrophic thyro and in still others local inflammation or edema. Color flow Doppler imaging has that in many cases the halo is vascular and may represent capsular vessels. The inter structure of a nodule has not been a useful indicator of malignancy. Cancers may be sonographically solid tissue or complex, having undergone cystic or hemorrhagic degeneration or both, which is a common occurrence in large nodules. Similarly, cys degeneration may be seen in cancerous regional lymph nodes. Calcification of benign and malignant nodules is also common and not predictive of histopathology. However, some information about the nature of a nodule can be derived from the nature of the calcification in a thyroid nodule or in a pathological lymph node. A peripheral rim or eggshell-like calcification is indicative of chronicity, which favors the benign nature of a nodule. Coarse, scattered calcifications may be seen in benign or malignant nodules that have undergone hemorrhage. Large calcifications can be useful as an index of malignancy in patients who are suspected of having medullary
Nonisotopic NeckImaging of the
17
Fig. 4. Use of sonography to detect a nonpalpable thyroid primary when adenopathy is palpated. A 37.-year-old obese female who had a palpable, firm 2-cm lymph node in the right jugular chain of nc,des. The thyroid region was not abnormal on palpation. Sonograms of the neck in the longitudinal plane are shown. (Top) A 1-cm hypoechoic mass in the upper posterior (Bottom) the lymph portion of the right thyroid lobe. Minute bright spots are calcifications. node that was palpated. It is 1.8 cm in diameter, rounded, and also has calcifications. Surgery revealed papillary thyroid cancer with psammoma bodies and adenopathy.T, thyroid lobe; L, lymph node; arrow, calcification (psammoma body).
Blum
Fig. 5. Use of sonography in routine follow-up; nonpalpable, nonsuspicious lymph node wa revealed; no tumor was found. Two films from a sonogram of the right side of the neck the same patient depicted in Figure2. She had tumor on the left side of the neck but no pathology on the right side. (Left) Transverse plane (Right) Sagittal plane. There is a 10.8 x 2.0 x 4.4 mm lymph node. Note the narrow central slitlike hilum and thin elliptical shape that suggest benignity. At surgery, benign nodes but no malignancy were foundon the right side. The lymph node is marked by the + + and x x symbols.
carcinoma because of an increased concentration of calcitonin or in the clinical setting (MEN). Punctate of one of the syndromes of familial multiple endocrine neoplasia calcifications tend to correlate with the histopathological demonstrationof psammoma bodies in papillary carcinoma of the thyroid (Fig. 4). It is not cost-effective to routinely perform a sonogram on patients with a thyroid nodule before thyroid surgery. However, sonography can be useful in selected patients to answer specific questions or to detect gross evidenceof invasion or encasement of regional structures when there is a large nodule. Ultrasonography may offer diagnostic insights for patients who have lymphadeno thy and who are suspected of having thyroid cancer or who have had a thyroidectom for cancer. This examination is valuable to diagnose the natureof palpable nodes and also to detect and characterize impalpable lymphadenopathy. Benign nodes are 5). Sutton They tend to be elliptical, and have a narrow central hilum (Fig. and colleagues (21) reported that size, shapeand internal architecture of lymph nodes did not reliably (22), using segregate benign and malignant lesions. However, Vassallo and associates rotation of the scanning plane to i d e n w the largest diameter of a node, confirmed observations that benign lymph nodes tend to be oval in shape, and malignant ones are usually round. However, the consensus is that there are no significant differences in size between benign and malignant nodes. Solbaiti and associates (23) evaluating 291 lymph nodes in 143 patients before surgical dissection of the neck because of thyroid cancer, reported other ultrasonic characteristics of lymph nodes that correlate, albeit imperfectly, with histology. They demonstrated that ultrasound can show echogenic hilum and a peripheral hypoechogenic cortex. The combination of nar of the hilum and widening of the cortex was seenin 90% of malignant nodes, but in only 54% of benign nodes. Finding an elliptical or wide hilum was a useful criterion for identifying low likelihood of malignancy. This characteristic was confirmed by Vassallo’s group (22), who observed it in 15 of 26 (58%) of benign reactive nodes, but in only 5 of 68 (8%) of malignant nodes. By contrast, the inability to identify a
ImagingNonisotopic Neck
of the
19
hilum with high-resolution equipment correlated with malignancy, having been obse in only 2 of 26 (8%) if reactive nodes but in 36 of 68 (44%) of nodes that were occupied by malignancy. Finding a “slitlike” narrow hilum was a less useful criterion 33 of for identifying a cancerous node, having been seen in 68 (49%)of benign nodules and 9 of 26 (35%) of benign nodules. Thus, sonographic features of a cervical lymph node that should be regarded with suspicion for malignancy include a plump round shape, narrowing of the hilum associated with wideningof the cortex, and inability to identify a hilum (Fig. 4).
Sonography in the Patient with Thyroid Cancer Finding Small Lesions in the Thyroid Cancer Patient with High-Resolution Ultrasonography Sonography has become the most frequently used imaging procedure in the manage ment of thyroid cancer. Its value was recognized even before the advent of current high resolution equipment (7). More recently, in a study of 1 0 0patients with thyroid carcinoma, Simeone and colleagues (19) reported that sonography is actually the preferred method for evaluating postoperative thyroid tissue after partial thyroidectomy. Periodic sonography may detect recurrent carcinoma in the thyroid bed after surgery, the contralateral lobe, or in lymph nodes even before it has grown sufficiently large to be palpable (Fig. 2). Sonography is highly efficient in detecting a thyroid mass when a patient has cervical adenopathy due to thyroid cancer but the primary lesion is not palpable even if the scintiscanis normal (Fig.4). Furthermore sonography can demonstrate much smaller lesions than any of the other imaging methods. The procedure can be done during replacement or suppressive therapy, avoiding the inconvenience and risks of hypothyroidism that would be needed for scintillation scanning, and at a much lower expense than CT or MRT (see below). After partial removal of the gland has revealed cancer, the finding of a nodule by sonograminresidualthyroidtissue,evenifthereisnopalpablenodule,maybe consideredastrongindicationindecidingwhetheracompletionthyroidectomy is necessary. The role of ultrasound guided fine needle aspiration biopsy inthis circumstance is controversial. Cytological identification of malignant cells would be an indicationforadditionalsurgerybutsamplingerrorandlimitationsthatareinherentin cytologic examinationof thyroid nodules require caution in interpreting the significance of a “negative” aspiration. Ultrasound-Assisted Fine Needle Aspiration Biopsy in the Patient with Thyroid Cancer Most clinically detected and palpable thyroid nodules may be punctured directly; ultrasound guidance is not needed, makes the procedure more complicated, and adds to the expense. Direct ultrasound guidance for percutaneousisbiopsy generally reserved for certain conditidns(24-26): 1. 2. 3. 4. 5. 6.
Unusually deep nodules, particularly in the obese, muscular, or large-framed patient Very small nodules Nonpalpablenodules Ultrasonically detected incidentalomas that are associated with cancer risk factors Some complex degenerated nodules Nonpalpableadenopathy.
20
Blum
The most simple and frequently used method for ultrasound-assisted aspiration biopsy for palpable nodules is to locate the lesion on the film or screen, taking note of the solid and cystic components of the nodule in relation to palpable or visual land marks, The operator then punctures free-hand in the proper location preferably to sample this approach is not possibleor has not been successful, tissue. For small nodules, where real-time ultrasound may be employed to actually observe the insertion of a needle The free-hand free-hand or using a special needle guide attached to the transducer. technique offers flexibility. The transducer is placed at some distance from the point of needle insertion. The needle pathis observed on the screen in more than one plane, while the operator maneuvers the needle to puncture the target. Transducers that are fitted with a needle guide are preferred by some physicians, but require considerable practice and need special attention to sterile technique. Some ultrasound units have computer-generated grids which map the path of the needle and estimate the distance to the target. Ultrasound biopsy probes for intraoperative use are also available bu widely employed. Cytology on minute nodules can be obtained. The questions are: How often is it accurate with nonpalpable nodules that are smaller than 1 cm in size? How sure can one be that the end of the needle is within the nodule at the time of sampling? What is the clinical impact of the information? These questions cannot currently be answe critically. However, it would appear that the success rate, accuracy, and reliability are significantly higher when the procedure is done by highly experienced investigators than by sonographers and clinicians who only occasionally do ultrasound-guide The main point is that the demonstration that a malignancy is present is persuasive, but the absence of cytological evidence of malignancy should be interpreted with caution. Investigation is attempting to define the clinical role of ultrasound guided aspira The success rateis better for large nodules than for small ones. Boland and coworke (26) showed 91% sensitivity (102 out of 112 masses). There were 7 nondiagnostic punctures, 4 in complex nodules, and 3 in nodules that were less than 1 cm (0.8-0.9 cm in diameter). Since only 29 of the 112 nodules were less than 1 cm, the failure 5% among larger lesions. It is noteworrate was 10%among subcentimeter nodules and thy that the failures with large nodules occurred when there was evidence of cystic or hemorrhagic degeneration. Sonography is useful to aspirate modules that have been In the same study, there were 71 patients in whom imaged but have not been palpated. a nodule was not palpated. The technique may help when prior nonguided biopsy attempts were unsuccessful, as was the case in 16 in this series. Percutaneous biopsy of pathological lymph nodes guided by ultrasound has also been studied. Takashima reported correlation with surgical findingsin 62 patients who 0.8 cmwitha hadimpalpablecervicallymphnodeswhosemeandiameterwas sensitivity of 95%, specificity 94%, and accuracy of 94% (27). Lee and colleagues (28) reported on36 cervical lymph node biopsies in 29 patients with cervical lym nopathy and suspected recurrent differentiated thyroid cancer. They reported a 91% sensitivity and 100% specificity. The combined diagnostic sensitivity and .specificity of tissue marker analysis (thyroglobulin or calcitonin) and cytopathological examin was 100% (28).
Imaging Nonisotopic Neck
of the
21
Applications of Color Doppler Imaging in the Thyroid Cancer Patient The value of color Doppler imaging of the thyroid in the diagnosis of cancer is under current investigation. To date, the clinical usefulness of color Doppler is best exemplified by the detectionof diffuse hyperemia in the thyroid gland of patients with Graves’ disease (29). However, an increased vascularity on color Doppler has been reported in some focal lesions, especially autonomous “hot” nodules, whose risk of cancer is small (30). Another area of diagnostic study is the sonographic halo around a nodule which was discussed above. Criteria for altered vascularity in the halo, the significance of interruption of the halo, and correlation with pathological findings have not been well defined. Investigation with color flow Doppler in cervical lymph nodes has shown vascular patterns that are significantly different in benign and malignant lymphadenopathy with a reported sensitivity of 93%, a specificity of 86%, and an accuracy of 89% (31). The enhanced sensitivity of Doppler technology, percutaneous biopsy of adenopathy and the possible useof sonographic contrast agents that are now in experimental stages suggest the potential for insights and specific characterization of tumors, lymph nodes, and goiters. SECTIONAL IMAGING: COMPUTED TOMOGRAPHY AND MAGNETIC RESONANCE IMAGING Computed tomography (CT) and magnetic resonance imaging (MRI) are sectional are occasionally employed in the evaluation of thyroid cancer. imaging techniques which The utility of both modalities reflects their ability to accurately define the regional anatomy of the infrahyoid neck and superior mediastinum (18,32-34).
PrincipZes and Methods . Computed Tomography CT images are generated by computer-assisted analysisof the attenuation of x-rays that have been transmitted through the patient. For certain purposes, the images may be examined directly because the thyroid gland is somewhat more radiopaque than the rest of the soft tissue structures of the neck due to its high iodine content. More often, the contrast needs to be enhanced by the intravenous administration of iodinated material to make the studies clinically useful. Unfortunately, however, the iodine in the dye may be counterproductive to further diagnostic tests and treatment of thyroid disorders. This is a major limitation to using CT for thyroid diagnosis. Therefore, in most centers, CT has assumed a role in the diagnosis and management of thyroid problems that is complementary to MRI (32,35,36).
Magnetic Resonance M R images are created by computer-assisted analysis of electromagnetic waves interacting with a patient lying within a magnetic field. These interactions occur b thehydrogenatomsinthepatient’sbodyandradiowaves of aspecificfrequency transmitted through the patient during the examination.By varying the magnetic field withinwhichthepatientlies,differentpropertiesofthehydrogenatomscan be
22
Blum
selectively emphasized. The two properties most commonly encountered in MR imaging of the thyroid are termedT1 and T2. Since the hydrogen atomsof various tissues have specific T1 and T2 properties, a computer-assisted analysis of differences between ‘‘Tl weighted” and “T2 weighted” images is used to identify the thyroid gland, skeletal muscle, blood vessels, or regional lymph nodes. Tissue characterization may be further enhanced on MR images by the intravenous administration of noniodinated contrast agents, such as gadolinium-DTPA or suppressing the signal derived from fat [short tau inversion recovery (STIR)] (36). MRI has problems. The magnetic field interferes with pacemakers. The test may cause discomfort, principally for a claustrophobic patient. It is time-consuming, considerable noise is inherenttothetechnique,and,generally,theentirebodymustbe inserted into a large cylinder. However, recent advances employing special surface electromagnets over the neck and “open” systems may enhance the usefulness ofMRI to thyroid diagnosis but provide poorer images. The equipment is very costly and in great demand for other types of examinations. Studies arein progress to correlate images with the biochemistry and histopatholo of tissue (37,38). Similarities in the proton response of thyroid tissuein the neck and (39) and differencesin these characteristicsin malignant chest have been demonstrated and benign thyroid tissues have been studied in vitro (40). Currently, MRI may identify a thyroid mass as malignant with high probability. Normal thyroid tissue tends to be slightly more intense than muscle on a T1-weighted imageandtumorfrequentlyappearsevenmoreintense,orbrighter,thannormal thyroid (18).
Comparing Computed Tomography and Magnetic Resonance Imaging No study directly comparing the utility of CT and MRI in thyroid disease has yet been published. Since MRI does not require ionizing radiation or iodinated contrast agents to produce its images, it is often assumed that MRJ is “better” than CT scans for this purpose. However, CT scans provide better spatial resolution than MRI. MRI offers several advantages in the assessment of suspected thyroid pathology. This Direct sagittal, coronal, and axial images may be obtained with the patient supine. multiplanar capability facilitates neoplasm localization (41) (Figs. 2 and 3). MR is superior to CT in the differentiation of postoperative scar from recurrent tumor (42). MR images are not degraded by the shoulder artifact commonly found on CT (43). CT offers its own advantages in the assessment of thyroid disease but it should be considered a complementary examination to MRI. CT is more sensitive than MFU in detecting small metastases (S1.5 cm in diameter) to cervical or mediastinal lymph nodes (44). CT is currently more reliable than MR in the detection of small nodules, especially pulmonary (45,46).The total examination time for CTis shorter than MRI, an important considerationin unstable or claustrophobic patients. Patients with cardi (47).Access pacemakers or other biomechanical devices can only be assessed with CT to CT examinations is greater than to MRI because of the larger number of scanners available. Finally, the cost of a CT examination is considerably lower than MRI. Generally, isotope studies, when needed, should be done before contrast CT to contamination with iodine (Fig. 6). This issue is of special importance in the patient with thyroid cancer who may need an I3’Iwhole-body scan or therapy with I3lI. The
Nonisotopic Imaging of the Neck
23
Fig. 6. Use of CT to demonstrate adenopathy.A 43-year-old woman with a remote history of thyroid cancer that was treated surgically in another country was found to have a mass in the right side of the neck. Fine needle aspiration biopsy of the mass was not successful. CT after intravenous iodinated contrast medium wasatdone another medical facility, demonstrating a large pathological lymph node deep to the sternocleidomastoid muscle. '''I whole-body scan"II, ning had to be delayed untilthe iodide from the dye was excreted. The node accumulated confirming thyroid cancer, Therefore, she was treatedI3lI. with L, lymph node;SM,sternocleidomastoid muscle.
iodide from the contrast agent will block the uptakeof radioiodine by normal thyroid tissue or cancer even if TSH is very high and delay or preclude testing and therapy. In patients who are taking suppressive therapy with thyroid hormone, and in whom a contrast CT is required, it is best to continue the medication for several days after the contrast CT until iodide from the dyeis excreted, to minimize the effects of excessive iodine on the thyroid gland. Alteration of thyroid functionby contrast agent iodide is also a serious issue. If the patient has not had a thyroidectomy, the excessive iodide may cause either hyperthyroidism or hypothyroidism, depending on the underlying thyroid condition. It is imperative for the radiologist and the clinician to discuss these aspects prior to the performanceof a contrast CT and to consider a noncontrast study,
24
BIum
which, although not optimal, may be adequate to answer the clinical question. Iodina dye is not used for MRI examinations, a distinct advantage.
Usefulness of Sectional Imaging in Clinical Management Sectional imaging techniques should not be used to search for thyroid pathology or to evaluate the usual thyroid nodule or goiter because they are too expansive and insufficientlyspecifictobeusefulintheinitialdiagnosis (18). Sectionalmethods become necessary only when the results of other methods are inadequate. Nevertheless thyroid lesions may be noted incidentally on CT or MRI examinations of the cervical spine or other regional structures, leading to further investigation. A preoperative sectional imaging examination is useful for a thyroid nodule for specific situations. When the clinical examination demonstratesmass a that is fixed to surrounding tissue or when extrathyroid masses are palpated, the study may provide a surgical guideor may demonstrate that total excision is precluded, thereby allowing an appropriate plan for a palliative procedure or the need for a specialized surgeon. MRI or CT is useful when there is an unusually large mass that obstructs the thoracic inlet and impinges on other structures or extends substernally. Important information may beobtainedabouttrachealcompression or invasionevenwhenconventional Evaluating substernalor retrotraradiographs fail to demonstrate such evidence 7). (Fig. cheal extension may lead to involving a chest surgeon although in most cases a cervica approach is adequate. The results of CT altered surgical planning in 5 of 19 patients who had intrathoracic extension of a thyroid tumor and in 3 of 19 with laryngeal or esophageal invasion (48). Sectional imaging is uniquely valuable to detect recurrent thyroid cancer that is located in the mediastinum (Fig. 8). Sectional imaging procedures have been especially useful in the assessment and management of patients after thyroid cancer surgery. The major uses in these patients are to detect the following: 1. Recurrent thyroid cancer 2. Cervical or mediastinal lymphadenopathy 3. Regional metastases 4. New masses that have been palpated
5. Evaluating cryptic findings on palpation, sonography, or scintiscan
The h4R characteristics of recurrent thyroid carcinoma may allow differentiation of thyroid tumor and scarring in the normal thyroid bed, providing the study is delayed (49). In one study, CT until postoperative edema, infection, or bleeding have resolved correlated with tumor invasion of the carotid artery (7/7)internal jugular vein (9/10), larynx (5/6), trachea (8/10), esophagus (4/5), mediastinum (5/5), and regional lymph nodes (14/16) (SO). Tumor, scar, tissue deformity, displaced normal structures, and cryptic findings may be elucidated. FREQUENCY AND CIRCUMSTANCES FOR PERFORMING NONISOTOPIC IMAGING STUDIESPERSONAL PERSPECTIW How often and under what circumstances is it necessary to perform non isotopic imaging studies when there is a history of thyroid cancer? Should any of the tests be
25
Nonisotopic Imaging of the Neck
4
I
Fig. 7. Use of MRI to demonstrate tracheal invasion. MRI of the neck from a 64-year-old man who had a right modified neck dissection and thyroidectomy for a tall-cell papillary thyroid cancer. He had recurrent cancer. This film shows invasion of the trachea on the right side. The abnormal signal of the mass replaces the signal from the trachea. Compare the appearance of the right and left sides of the trachea. The arrows point to the tracheal cartilage. M, mass.
used routinely?Is it enough to examine the patients clinically? How do these examinations relate to isotope scanning and other studies such as thyroglobulin assays? The answer to each of these questions is judgmental. Certain issues have not been and may never be evaluated critically. Some of our current insightsare tentative and intuitive. Theywillevolvewithexperience,theoutcome of investigation,andtechnologic advances. I shall express my views. For most patients whose tumor is not progressing rapidly, clinical evaluation with history taking, palpation and assaying thyroglobulin may be done at yearly intervals or when there are new complaints or findings. This frequency of testing is probably adequate to detect recurrence of cancer in a timely fashion. It is unknown if there is an additional benefit from performing imaging studies to detect subclinical lesions. or complete thyroidectomy for cancer, sonography However, I believe that, after partial is the most cost-effective, sensitive, and accurate method is available that for identifying persistent or recurrent cancer and is superior to palpation. Therefore, .I advise sono-” graphic examination of the neck yearly for the first several years and then every five years orso. I do not adviseCT or M R routinely. Rather,I employ these tests selectively
26
to demonstrate substernal adenopathy. MRI of the upper mediastinurn from a 44-year-old woman who had a thyroidectomy for papillary thyroid cancer. A pathological right substernal lymph node is ~emonstrated.Note the brightness of the signal. The arrow points to the pathologi~allymph node. S , sternum.
to elucidate distorted anatomy, conflicting medical opinions, or other test results that require objective analysis. It is unclear how long to continue yearly follow-up. roid cancer has been reported and reevaluation at unspecified very late recu~enceo rther localize anato~callya lesion that takes up I3lI. radioactive iodine are too small to be detected by es, it is useful to assess the actual volume of the mass or its extent and involvement of regional s ~ c ~ r eFor s .these purposes, sonography is usually adequate but sectional imaging, avoiding iodinated contrast medium9may indications and fre~uencyof isotope scanning is discusse~else~here I shall offer an opinion for testing in selected circumstances, as outlined in Tables 1, 2, 3, and 4.
Nonisotopic imaging of the neck plays an important role in the ~ ~ a g e m eofn patients t with thyroid cancer. These imaging tests are expensive and for cost-effectiveness, they e employed selectively to answer specific questions that are posed by the clinical problem. ~onographyhas become the most fre~uentlyused imaging procedure in the thyroid cancer patient to depict the regional ~ a t o m yaccurately, safely, and e c o n o ~ cally. The major capability is to reveal nonpal recurrent th~roidcancer, including a d e n o ~ a t ~after y 9 surgery. ~ l ~ a s o u also nd p an a n a t o ~ guide c for fine needle as pi ratio^ biopsy and ocuments comparative size of nodules and lymph nodes. or CT becomes necessary only when palpation and ultrasonography are inadequate. They can i~entifymetastatic ~alignancyin regions that are blind to sonography,
Table 1 Role of Nonisotopic Imaging in Clinical Management when Thyroid Cancer is Suspected but Undiagnosed ~~
~
~
~
_
_
_
_
lu
Is “solitary” thyroid nodule malignant? Dominant nodule in a diffuse or nodular goiter or goiters of Hashimoto’s or Graves’ disease
u
History of exposure to radiation therapy during youth but no palpable nodule History of surgery for benign disease (adenoma) that may have malignant potential Worrisome chemical marker: elevated thyroglobulin or calcitonin
~~~~~~~
~
Role of Sonogram
Clinical Circumstance
Minimal if any, may show lymphadenopathy May disclose region with unique appearing ultrasound texture that is suspicious, may have “halo” around nodule, may demonstrate psammoma bodies May show one or more nodules (some say a source of confusion, others say useful) May show new nodule in contralateral lobe, may show lymphadenopathy May show nodule that has not been palpated, may show lymphadenopathy
MRI, magnetic resonance imaging; CT, computed tomography.
Role of MRI
Role of CT
None None
None None
None
None
None
None
None
None
Table 2 Role of Nonisotopic Imaging in Detecting Persistent or Recurrent Thyroid Cancer Shortly After the Initial Surgery for Cancer Clinical Circumstance After lobectomy or nodule removal
After near total thyroidectomy h,
O0
After complete thyroidectomy
After removal of thyroid cancer in a lymph node or discovery of distant thyroid metastasis After surgery for inoperable cancer (residual tumor)
~~~
Role of MRI
Role of Sonogram
Role of CT
May show nodule in contralateral lobe that has not been palpated, may show lymphadenopathy May show unsuspected residual tissue, may show lymphadenopathy May show unsuspected residual tissue, may show lymphadenopathy May disclose primary thyroid lesion
None
None
None
None
None
None
Rarely adds information to sonogram
Rarely adds information to sonogram
Documents baseline anatomy objectively
If sonogram is cryptic documents baseline anatomy objectively
If sonogram is cryptic and MFU is not available or cannot be used (pacemaker) documents baseline anatomy objectively
~
MRI, magnetic resonance imaging; CT, computed tomography.
7 70 c
Y
0
C
9 5
Y
0
C
Y
4
30
32
Blum
and further define the anatomy when there are cryptic symptoms, confusing findings, conflicting results, or altered anatomy after regional operations.MRI offers the advantage of not requiring iodinated contrast agents.
REFERENCES
1. Logan RL, Scott PJ. Uncertainty in clinical practice: implications for quality and costs of health care. Lancet 1996; 347:595-598. 2. Van Herle A J , Rich P, Ljung BME, Ashcraft MW, Solomon DH, Keeler EB. The thyroid nodule. Ann Intern Med 1982; 96:221. 3. Molitch ME, Beck JR, Dreisman M, Gottlieb JE, Pauker G. The cold thyroid nodule: an analysis of the diagnostic and therapeutic options. Endocr Rev 1984; 5:185. 4. Brander A, Viikinkoski P, Nickels J, Kivisaari L. Thyroid ultrasound screening in a rand adult population. Radiology 1991; 181:683-687. 5. Blum M, Goldman AB, Herskovic A, Hernberg J. Clinical applications of thyroid echogr phy. N Engl J Med 1972; 287:1164-1169. 6. Leopold GR. Ultrasonography of superficially located structures. Radiol Clin North Am 1980; 18:161. 7. Butch RJ, Simeone JF, Mueller PR. Thyroid and parathyroid ultrasonography. Radiol Cli North A m 1985; 23:57. 8. Blum M. Ultrasonography and computed tomography of the thyroid gland. In Ingbar SH, Braveman LE, editors. Werner’s the thyroid, 5th ed. New York Lippincott, 1986:576-591. 9. James EM, CharboneauJ W , Hay ID. The thyroid. In Rumack CM, Wilson SR, Charbone J W , editors. Diagnostic ultrasound Vol 1. St. Louis,MO: Mosby Year Book, 1991507-523 10.Foley WD. Physical Principles and Instrumentation. Chapter 1. In Color Doppler flow imaging, Foley WD, editor. Boston: Andover Medical Publishers, Inc., 1991; p. 3. 11. Boehm T M , Rothouse L, Wartofsky L. Occult follicular carcinomaof the thyroid with a solitary slowly growing metastasis.JAMA 1976; 2352420. 12. Blum M, PassalaqueAM, Sackler J, Pudiowski R. Thyroid echography of subacute thyro tis. Radiology 1977; 124795-799. 13. Espinassse P. L’echographie thyroidienne dans les thyroidities lymphocytaires chroniques autoimmunes. J Radiol 1983; 64537-544. 14. Hayashi N, Tamaki N, Konishi J, Yonekura Y , Senda M, Kasagi K, et al. Sonography of Hashimoto’s thyroiditis. J Clin Ultrasound 1986; 14:123-126. 15. Jayaran G, Marwaha RK, Gupta RK,S h m a SK. Cytomorphologicaspects of thyroiditis: a study of 51 cases with functional, immunologic and ultrasonographic data. Acta Cytol 1987; 31~687-693. 16. Gutenkust R, Hafermann W, ManskyT, Scriba PC. Ultrasonography related to clinical and laboratory findingsinlymphocytic thyroiditis. ActaEndocrinol (Copenh) 121:129-135. 1989; 17. Marcocci C, Vitti P, Cetani F, Catalan0 F, Concetti R, Pinchera A. Thyroid ultrasonogr helps to identify patients with diffuse lymphocytic thyroiditis who are prone to develop hypothyroidism. J Clin Endocrinol Metabol 1991; 72:209-213. 18. Blum M. Evaluation of thyroid function: sonography, computed tomography and magne resonance imaging. In Becker KL, editors. Principles and practice of endocrinology and metabolism. Philadelphia: Lippincott, 1990; 289-293. 19. SimeoneJF, Daniels GH, Hall DA, McCarthy K, Kopans DB, Muelleral. PR, Sonography et in the follow-up of 100 patients with thyroid carcinoma. AJR 1987; 148:45-49. 20. Proper RA, Skolnick ML, Weinstein BJ, Decker A. The nonspecificity of the “halo” sign. J Clin Ultrasound 1980; 8:129. J W , James EM, Grant CS, Hay ID. US-guided biopsy 21. Sutton RT, Reading CC, Charboneau of patients with thyroid cancer. Radiology of neck masses in preoperative management 1988;168:769-772.
Imaging Nonisotopic Neck
of the
33
22. Vassallo P, Wernecke K, Roos N, Peters PE. Differentiation of benign from malignant superficial lymphadenopathy: the role of high-resolution US. Radiology 1992; 183:215-220 23. Solbiati L, Arsizio B, Rizzatto G, Bellotti E, Montali G, Cioffi V, et al. High esolution sonography of cervical lymph nodes in head and neck cancer: criteria for differentiation of reactive versus malignant nodes. [abstract]. Radiology 1988; 169:113. 24. Rizzatto G, Solbiati L, Croce F, LE. Derci Aspiration biopsyof superficial lesions: ultrasonic guidance with a linear-array probe.AJR 1987; 148:623-625. 25.MatalonTAS,SilverB.USguidanceofinterventionalprocedures.Radiology1990; 174:43-47. 26. Boland GW, Lee MJ, Mueller PR, Mayo-Smith W, Dawson SL, Simeone JF. Efficacy of sonographically guided biopsy of thyroid masses and cervical lymph nodes. AJR 1993; 161L:1053-1056. Oi H, Okamoto S. Nonpalpable 27. Takashima S, Yoshida J, Kishimoto H, Matsushita M, lymph nodes of theneck assessment with US and US-guided fine-needle aspiration biopsy. [abstract]. Radiology 1995; 197(suppl):270. 28. Lee MJ, Ross DS, Mueller PR, Daniels GH, Dawson SL,Simeone JF. Fine-needle biopsy of cervical lymph nodes in patients with thyroid cancer: a prospective comparison of cystopathologic and tissue marker analysis. Radiology 1993; 187:851-854. 29. Ralls PW, Mayekawa DS, Lee K, Colletti PM, Radin DR, Boswell W D , et al. Color-flow Doppler sonography in Graves’ disease: “thyroid inferno.” Am J Roentgen01 1988; 150 781-784. H, WolfKT. Appearance of thyroid diseases 30. Fobbe F, Finke R, Reichenstein E, Schleusener using colour-coded duplex sonography. Eur J Radiol 1989; 9:29-31. 3 1. Tschammler A,Ott G, Schang T, Hoehmann D. Seelbach-Goebel B, Michel C. Vascular patterns in reactive and malignant lymph nodes. [Abstract]. Radiology 1995; 197(suppl):270. DF. Computerized axial 32. Blum M, Reede DL, SeltzerTF, Burroughs VJ, Greene LW, Roses tomography in the diagnosis and management of thyroid and parathyroid disorders.Am J Med Sci 1984; 287:34-39. 33. Bahist B, Ellis K, Gold RP. Computed tomography of intrathoracic goiters. AJR 1983; 140:455-460. 34. Higgins CB, Auffermann W.MR imaging of thyroid and parathyroid glands: a review of current status. AJR 1988; 151:1095-1106. 35. Blum M. Practical application of modem technology in thyroid evaluation. In Van Middlesworth L (Ed). The thyroid gland: practical clinical treatise. Chicago: Year Book Medical Publishers; 1986:47. 36. Blum M, Braverman LE, Holliday RA, McDougall IR, Simkin PH, Spencer CA, YeeJM. The thyroid diagnosis. In Wagner HN, Szabo Z, Buchanan J W , editors. Principles of nuclear medicine, 2nd ed. Philadelphia: WB Saunders 1995595-621. 37. Charkes ND, Mauer AH, Siege1 JA, Radecki PD, Malmud LS. MR imaging in thyroid disorders: correlation of signal intensity with Graves’ disease activity. Radiology 1987; 164:491. JC.imaging of the thyroid comparison 38. Mountz JM, Glazer GM, Dmuchowski C, SissonMR with scintigraphy in the normal and diseased gland. J Comput Assist Tomogr 1987; 11: 39. Sandler MP, Putton JA, Sccks GA, Shaff MI, Kudkaoni MV, Partain C. Evaluation of intrathoracic goiter with 1-123 scintigraphy and nuclear magnetic resonance imaging. J Nu Med 1984; 25874-876. MRI 40. Tennvall J, Biorklund A, Moller T, Clsson M, Persson B, Akerman M. Studies of relaxation times in malignant and normal tissues of the human thyroid gland. Prog Nucl Med 1984; 8:142-148. 41. Mancuso AA, Dillon WP. The neck. Radiol Clin North Am 1989; 27:407-434. 42. Glazer HS, Niemeyer JH, Balfe DM, Hayden RE, Emani B, Devineni VR, et al. Neck neoplasms: MR imaging. Part II.Posttreatment evaluation. Radiology 1986; 160:349-354.
34
Blum
43. Freeman M, Toriumi DM, Mafee MF. Diagnostic imaging techniques in thyroid cancer. Am J Surg 1988; 155:215-223. 44. Yousem DM, Som PM, Hackney DB, Schwaibold F, Hendrix RA. Central nodal necrosis M R imaging versus CT. and extracapsular neoplastic spread in cervical lymph nodes: Radiology 1992;182:753-759. 45. Webb WR, Sostman HD. MR imaging of thoracic disease: clinical uses. Radiology 1992; 182:621-630. 46. Davis SD.CT evaluationof pulmonary metastases in patients with extrathoracic malig Radiology 1991;18O:l-12. 47. Shellock FG.MR imaging of metallic implants and materials: a compilation of the literature. AJR 1988; 15~811-814. 48. Auffernann W, Clark OH, Thurner S, Galante M, Higgins CB. Recurrent thyroid carcinom characteristic on M R images. Radiology 1988; 168:753-757. 49. Takashima S, Morimoto S, Ikezoe J, Takai S, Kobayashi T, Koyama H, Nishiyama K, Kozuka T. CTEvaluation of anaplasticthyroidcarcinoma. Am J Neuroradiol 1990; 11~361-367. 50. Cooper JC, Nakielny R, Talbot CH. The useof computered tomography in the evaluation of large multinodular goiters.Ann R Coll Surgeons Engl1991; 73:32-35. 51. Blum M, Perlman S. Reducing suppressive therapy in patients with a history of thyroid cancer. [Letter to the Editor.] Ann Intern Med 1995; 123907-809.
3 The Thyroid Nodule Fine Needle Aspiration Biopsy Yolanda C. Oertel Fine needle aspiration (FNA) is a valuable procedure in assessing the nature of a thyroidal mass. Details of the technique as we practice it and the equipment required are described in references I and 2. Of particular importance are the use of a syringe holder and needles with clear plastic hubs (3). It is essential to use little or no suction when aspirating the lesions (4). EQUIPMENT REQUIRED 1. Syringe holder or handle: This is an indispensable item. We prefer the Cameco Syringe Pistol (Precision Dynamics Corporation,3031Thornton Avenue, Burbank,CA 91504). A 7 Sherman lessexpensive-plastic-handle is the Aspir Gun (The Everest Company, Street, Linden, NJ 07036). 2. Plastic disposable syringes, 10-cc and, rarely, 20-cc. 3. Disposable needles:22, 23, and 25 gauge; 1 and 1.5 inches long; withclear plastic hubs. 4. Plain glass slides, preferably with one frosted end. 5. Hemacytometer cover glass. This is a thick pieceof glass, narrower than the width of the regular glass microslide, which we use to smear the aspirated material the on glass slide (Fisher Scientific, Pittsburgh, PA 15219-4785; 1-800-766-7000). 6. Appropriate staining solutions. We prefer the Diff-Quik@ stain, a hematological stain similar to the May-GrUnwald Gierpsa stain.
PERFORMANCE OF THE ASPIRATION Ask the patient ifhe or she knows what you are going to do. When you explain the procedure, ask the patient to hold an ice cube (for mild anesthesia) on the area to be aspirated. Then proceed in the following way: 1. Place the mass between your index and middle fingers in a position suitable for needling.
2. Clean the skin with a cotton swab soaked in ethyl alcohol. Dry the skin with a gauze sponge to avoid the stinging sensation caused by residual
alcohol when inserting the needle.
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wmtofsky 0 Humnnn Press Inc., Totawn, N]
35
36
Oertel
The cytotechnologist gives the aspiration device to the pathologist and then hoMs the patient’s h a n d . (Patients will often mentionhow helpful and reassuring itis for someone to hold their hands.) 3. Ask the patient to swallow (if necessary, water can be provided through a bent straw). After the patient has swallowed, hold down the lesion between your left index and middle fingers. 4. Introduce the needle through the skin, making sure that the syringe is in “the resting position” (plunger at the 0-cc mark). 5. Advance the needle perpendicularly into the lesion. 6. Once the needle has entered the nodule, move it back and forth in thesame plane (do not vary the angle of the needle)without applying anysuction. Do this twice, andif nothing appears in the clear plastic needle hub, then apply suction very gradually and gentlyby pulling the plungerof the syringe. 7. While jabbing the needle in the lesion, keep talking to the patient: “You are doing fine “Everything is all right.” “It looks pretty good.” “You are helping me beautifully.” Such As you apply suction, gradually jab the needle again, moving expressions calm the patient. it with gentle movements back and forth in the lesion, maintaining the suction. Usually when the plunger of the syringe is at the 2- or 3-cc mark, you will see material in the clear plastic needle hub. However, if the lesion is firm and no material appears in the needle hub, keep applying suction until you reach the 10-cc mark. Again, move the need back and forth with gentle jabbing movements. By now you should have hemorrhagic material in the needle hub. Release the plungerto stop the suction. Shift your fingers f the “trigger” to hold the outside of the handle. 8. Withdraw the needle. 9. Ask the patient to apply flm pressure at the site of the aspiration using the same piece of gauze that you used to dry the skin. 10. While the patient is still applying pressure, move very quickly to prepare the smears. T cytotechnologist assists inthis procedure. Because the aspirate tends to clot promptly, w cannot overemphasize that speedy preparation of the smears is of extreme importance. Once the smears have been prepared and the technologist has started drymg and staining one slide, the pathologist may help the patientto sit up and then continue to apply press at the site of aspiration, until the smear is ready for microscopic examination. Then tell the patientit is herturn to apply pressure while we examine the slide under the micr The patient remains seated on the examining table until after we have examined the and decided which size needle to use for the next aspiration. These simple precautions sitting up between aspirates (to improve the venous drainage) and applying steady p at the puncture site-prevent the formation of hematomas. However, if the patient feels dizzy and does not want to sit up, she should be allowed to remain lying down. Also, if the patient cannot tolerate pressure on the neck (e.g., the lesion is over the trachea and pressure produces coughing), apply an ice cube to the site. 11. Once we are satisfied that we have sampled the lesion thoroughly, we tell the patient th there are no restrictions, that she can go back to work or to her routine activities. Very seldom do we put a small adhesive bandageon the area that has been aspirated.
NOTES: While applying suction and moving the needlein the lesion, look at the needle hub
a. If fluid appears, continue applying suction until filling up the syringe or until no more fluid is aspirated. b. If blood appears in the needle hub, stop applying suction immediately, whether th is at the l-cc mark or the 4-cc mark. c. If no blood appears, continue applying suction up to the 10-cc mark on the syringe.
Fine
37
Observe your patient’s face. If you see grimacing or any signs of discomfort, tell her that you are almost done, that it will take a little longer, to bear with you, etc. If you notice that the patient is about to swallow, release the plunger immediately and pull out of the lesion. Sometimes a patient willstart swallowing and surprise you; then just move along, do not offer resistance, release the plunger and pull outof the lesion as quickly as possible. Do not forget to release the plunger. It is a common mistake for beginners to withdraw the needle from the nodule while still applying suction. This will cause all the aspirated material to flow into the syringe. To recover it, you will have to rinse the syringe and prepare a filter specimen. TECHNICAL HINTS FOR FINE NEEDLE ASPIRATION Most publications mention the use of fill suction once the needle penetrates into the lesion. We do not advocate this procedure. After inserting the needle in the lesion, the needle should be moved up and down (if the patient is lying down), or back and forth (if the patient is sitting upright), with gentle jabbing movements, without applying any suction. There should not be any lateral movements. If bright red blood is seen immediately in the needle hub, the procedure should be stopped, and the needle withdrawn at once. One smear should be checked under the microscope and, depending on the microscopic findings, one could change the needle gauge or decide to apply more suction in the next aspirate. Although the techniqueis not complicated,we have seen such a high rate of failure this is apparently among internists, endocrinologists, and surgeons that we believe that more difficult than it seems. Probably these physicians can palpate thyroid nodules better than the pathologist,so we do notthink the problem liesin localizing the lesion. Usually they apply too much suction too soon. WestainthesmearswithDiff-Quikm,andthecytologicdiagnosticcriteriawe describe are based onthis staining method. Other pathologists prefer the Papanicolaou stain or hematoxylinandeosin;someofthediagnosticcriteriaaredifferentwith these stains. REFERENCES 1. OertelYC.Fine-needleaspiration of thethyroid. In Moore WT, Eastman RC, editors. Diagnostic endocrinology, 2nd ed. St. Louis: Mosby-Year Book, 1996:211-228. 2. Oertel YC. Fine-needle aspiration and the diagnosis of thyroidcancer.EndocrinolMetab Clin North Am 1996; 2569-91. 3. Oertel YC. Fine-needle aspiration: a personal view. Lab Med 1982; 13:343-347. 4. Oertel YC. A pathologist’s comments on diagnosisof thyroid nodules by fine needle aspiration. J Clin Endocrinol Metab 1995; 80:1467-1468.
4 The Thyroid Nodule Medical Management Leonard Wartofsky LABORATORY TESTS
Thyroid function tests are usually of little value in the evaluation of thyroid nodules, with the exception of possible toxic adenomas. Thyroglobulin levels may be elevated in patients with thyroid malignancy and are very useful as a tumorin marker the routine follow up of patients operated upon for thyroid cancer (1,2), but preoperative blood levels do not differentiate from those associated with benign adenomasor thyroiditis. Serum antithyroglobulin and antimicrosomal (or antithyroid peroxidase) antibodies are also of very limited value (I). Special diagnostic studies are available for the detection of medullary thyroid cancer (MTC), which may present as a dominant cold nodule per se or as part of a multiple endocrine neoplasia (MEN) syndrome (3,4). MEN 2A is characterized by MTC with pheochromocytoma and, in somecases, hyperparathyroidMEN 2A differs from sporadicMTC in being often preceded ism. The familial MTC of by C cell hyperpasia leading to multifocal tumors. MEN 2B includes MTC, pheochromocytoma,andseveralphenotypicabnormalitiesincludingmucosalneuromata.Ithas been shown that theRET protooncogene is the gene responsible forMEN 2A and 2B, and mutations in differing codons and exons ofRET have been identified in sporadic MTC as well. It is now possible to routinely identifyRET in material obtained by fine known for needle aspirationof a thyroid nodule. Differentiation between the mutations sporadic vs familial MTC provides information that helps decide whether or not preop ative screening for pheochromocytoma is necessary. Traditional management approaches have held that basal plasma calcitonin or CEA (carcinoembryonic antigen), and calcium RET and/or pentagastrin stimulation tests or assessment for theprotooncogene intended to identify cases of MTC are not cost effective in the initial or routine evaluation of the nodular thyroid. However, some recent studies have indicated otherwise,Le., that calcitonin measurements may detect unsuspected medullary carcinoma (5,6). FINE NEEDLE ASPIRATION [See Chapters 3 and 171
The single best preoperative method to identify a malignancy is to obtain cells from the nodule for cytopathologic examination by a fine needle aspiration (FNA) technique (7-11). FNA with a 22-25 gauge needle provides the highest rate of successful sampling From: Thyroid Cancer: A Comprehensive Guide to c l i n i c a l Management Edited by: L. Wartofsky 0 Humana Press Inc., Totma, NJ
39
40
Warto$ky
Table 1 Potential Utility of Ultrasonography of Thyroid Nodules Differentiation of solid vs cystic consistency Detection of multinodularity . Detection of occultthyroid malignancy in cases of metastatic cervical lymphadenopathy from unknown primary Monitoring nodule size, including response to suppressive therapy in Determination of solidvs hemorrhagic expansionin thyroid lesions showing rapid increase size Guidance for needle biopsy in difficult selected cases (?) Monitoring irradiated thyroids (?) Monitoring for local recurrence of thyroid carcinoma
and the lowest rate of complications while yielding diagnostic precision is equal that or superior to other methods (12-16). Both the collection technique and the availability of the a skilled cytopathologist are critical to the adequate collection and interpretation specimens. Consequently, results are best when both the operator and pathologist considerable continuing experience,in which case false negatives should average only 1%and false positives 2% or less. The results of FNA of a thyroid nodule be will generally categorized as “benign,” “suspicious,” “malignant,” or “inadequate for diagnosis,” a 3,9, and 17 in this volume devoted to cytology the distinctions are discussed in Chapters FNA has been shown to and pathology by Oertel and Oertel. Use of ultrasound-guided be effective forthe detection of malignancy in nonpalpable lesions(17,18) (Table 1). Some studies have sought to use cytometric DNA analysis to improve the predictiv value of FNA. While it has not been found to be entirely successful in separating benign from malignant disease,it does correlate with outcome and survival in patients with proven malignancy (19,20). PCR amplification of TSH receptor or thyroglobulin FNA of cervical lymph nodes has facilitated earlier transcripts on material obtained by diagnosis of metastatic malignancy (21). FNA carries no significant risk, andno cases of tumor seeding have been reported. Over the past decade, the use of FNA has had a clear salutary effect on the economics of nodule management by reducingthe required frequency of surgical thyroidectomy by approximately 50% while doubling the yield of cancer in those patients operated upon (19,20). Routine re-aspiration in follow-up of nodules found to be benign initially may be of only limited value (22). THYROID SCANNING The majorityof thyroid adenomas and carcinomas have defectsin iodide accumulation and/or organification, which can be demonstrated by reduced trapping of radionuclide, leading to the designation of “cold nodule.” On radionuclide scanning, about 5% of nodules will be found to be “hot” (hyperfunctioning), 10% “warm” (normal functioning), and 84% “cold” (nonfunctioning).
The Hyperfunctioning (“Hot”) Thyroid Nodule Hyperfunctioning thyroid adenomas result from genetic alterations in eitherGsa the protein or in the TSH receptor (22). Hot nodules rarely represent malignancy, warm
Medical Management
41
nodules carry an intermediate risk of around 5%, and cold nodules, while having the highest risk (13,14) of malignancy, still represent benign pathology in more than 80% of cases. Therefore, radioisotopic scans are of low specificity despite their high sensiti ityfornodulesover 1 cmdiameter (24). Scanningisusuallydonewith [1231]- or [Tclpertechnetate. Despite some limitations, the qualitiesof low radiation dose, low cost, short scanning interval, and reliabilityof hypofunctioning thyroids on scan have led to continued use of Tc at many centers. 1231also delivers low radiation and is the In addition to functional information, scans may reveal preferred iodine scanning agent. evidence of multinodularity in up to one-third of clinically palpable solitary lesions, a finding that is associated with a decreased risk of malignancy. A thyroid scan is also useful in identifying hyperfunctioning nodules in patients with symptoms of hyperthyroidism, suppressed TSH levels, or biopsy results suggestive of follicular neoplasm. A recent voguein Europe, which is yet to gain wide popularity in the United States, this approach is the percutaneous injection of 95% ethanol into thyroid nodules. Initially, was applied to hyperfunctioning nodules (25-28), and, more recently, to both benign thyroid cysts after aspiration and benign cold nodules (29-32). The procedure can beverypainfultothepatientandhasbeenassociatedwithtransientincreases in serum thyroglobulin and the thyroid hormones with self-limited thyrotoxicosis. Fever, local pain and hematomata, and vocal cord paralysis are also possible in inexperienced hands. Although hyperfunctioning nodules are rarely the seat of carcinoma, one advantage of surgical excision (usually lobectomy)is the acquisition of definitive histopathologic or ethanol injection. Nevertheless, diagnosis, whichis lacking with radioiodine therapy surgery is infrequently recommended for hyperfunctioning nodules because of its own associated risks, the low incidence of cancer, and the efficacy of radioiodine. One exception may be those very large (Acm diameter) nodules for which the required dose of radioiodine is so great as to itself provide the contralateral lobe with a risk of radiation-induced neoplasia. Efficacious radioiodine treatment of hyperfunctioning nodules appears to correlate best with the ratio of the dose to the nodule area (33). Hemorrhagic necrosis may occur during the natural history of a hyperfunctioning adenoma. The presentation of pain in a nodular goiter may suggest subacute thyroiditis. With infarction of the hyperfunctioning nodule, the subsequent loss of function leads of function in the previously suppressed to returnof TSH levels to normal and resumption extranodular thyroid tissue. The previously hyperfunctioning nodule may then appear “cold”onscintiscanning,whichtakentogetherwiththehistory ofpaincould be misinterpreted to represent a carcinoma.
Other Scanning Modalities Fluorescent thyroid scanning offers special advantages in childhood and pregnancy 100% due to minimal radiation exposure. The procedure has been said to be nearly sensitive but only64% specific when cold areas are taken as positive results(24). The procedure employs xlAm, which has the ability to excite thyroidal iodine causing release of X-rays that quantitatively correlate with iodine content of the imaged tissue. Unfortunately, the required equipment is not widely available and accumulated data remain too limited to recommend standard use of this technique.
42
Wartofsky
A variety of other scanning techniques, including * O I T h , '%elenomethionine, 'j7Ga, and I3'Ce, have been investigated, none of which has proved to be a reliable indicator of malignancy(12). [13'I]MIBG(meta-iodobenzylguanidine)has been used successfully to image medullary carcinoma of the thyroid (34).
THYROID HORMONE SUPPRESSION Thyroid hormone has been used for many years to reduce the size of thyroid thought to be dependent upon TSH stimulation. As a diagnostic test, the assumption is that benign lesions will show preferential reduction in size. Typically, patients are given a 3-to 6-mo trial of L-thyroxine at a dose titrated to result in TSH suppression to below the lower limit of normal of a sensitive TSH assay. Growth of a nodule or lack of reduction in size during therapy raises suspicion of malignancy. Complete responses probably occur in less than 10% of cases, while 50% reduction in size has been reported in an average of30% of cases (13). Trials of Z-thyroxine suppression of 6 mo of treatment colloid nodules showed no change in size compared to placebo after in one study (35), but more than 50% reduction of nodule volume in 56% of patients in another (36). The issue remains controversial (37) with some relatively well-controll studies that have failed to observe significant shrinkage of nodules on thyroxine therap (38,39), while other workers continue to suggest that suppressive therapy may shrink thyroid nodules (40-42), or prevent the appearance of new nodules (43). A careful reviewof all of the relevant studies on the efficacy of L-thyroxine suppression of thyroid nodules concluded that treatment is beneficial in only10-20% of lesions proven benign by aspiration cytology(44). Similar responses to thyroxine therapy w (45). And in spite ofthe large numberof published noted in a recent small metaanalysis reports onthis issue, critical analysis suggests that additional carefully controlled of large numbers of patients are still required to clarify the management of thyroid nodules with suppression therapy (46).Adverse effects of suppressive doses of thy (47). include altered myocardial contractility, increased heart rate, and atrial arrhythm Co-administration of p-blocking drugs may ameliorate the hyperadrenergic sympto in patients sensitive to suppressive dosage (48). The increasing concern about possible ofthyroidhormonehasbeen risk of osteopenia after long-term suppressive doses allayed somewhat by careful analyses of the data (49). Use of prudent suppression (50). Moreover, doses of thyroxine has been shown to not contribute to osteopenia supplementation of thyroxine with estrogen replacement in postmenopausal women (51). It appears that failure of may totally obviate any potential risk of osteopenia nodule reduction to suppression minimally increases the probability of cancer while successful suppression reduces the probability by about 25% (52). A trial of thyroid hormone suppression alone is neither sensitive nor specific, but may haveutility as an adjunct to other modalities of evaluation. In addition, suppre therapy may be of benefit in preventing development of additional nodules (43). One study, looking at recurrence rates for thyroid nodules after partial surgical th for benign diseasein patients with a previous history of radiation, showed that trea with thyroid hormone postoperatively decreased the risk of benign recurrence from 36% to 8.4% (53).
Medical
Management 43
Dosage
Some patients being given thyroxine may have concomitant autoimmune thyroid disease and/or other hyperfunctioning nonsuppressible nodules in addition to the cold nodule being treated; these patients may require less L-thyroxine because of the presence of functioning tissue which complements the exogenously administered hormone.An approximate suppressive dosage is slightly in excess 1.7 of pgkg body weight per day (54), a dose that usually results in a serum T4 at or somewhat above the upper limit of the normal range. The dose is incremented by 0.025 pglday every 5-6 wk with TSH monitoring until a suppressed TSHis observed. Nodules are assessed for change in size by physical examination every 6 wk for the first 6 mo (or less frequently by ultrasound if required). The follow-up intervals may be more prolonged when signifi decreases in size are observed, extending eventually to annual follow-up. Because of its long half-life, L-thyroxine is administered as a single daily dose. While the serum T3may be superior to the serum T4 as an indicator of the metabolic state in the patien receivingL-thyroxine,theoptimaldose is best determined by clinical criteria and elevated serum TSH indicates measurement of serum TSH by an ultrasensitive An assay. that treatment is insufficient, and an elevated serum T3 that it is excessive. There are a number of clinical circumstances in which a previously stable dosage of L-thyroxine may need to be either increased or decreased. Increases of 25-50 pg daily may be required during pregnancy, andthis need will be detected with frequent monitoring of serum TSH during pregnancy. Progressive increases in dose may be as further requiredduringlong-termfollow-up ofHashimoto’sthyroiditispatients atrophy of their glands occurs, or in Graves’ patients who are postthyroidectomy or postradioiodine therapy.A number of drugs may interfere with gastrointestinal absorption of thyroxine or enhance its metabolic clearance. Similarly, decreases in dosage may be required with spontaneous disappearance of TSH receptor blocking antibodies in Hashimoto’s disease, or with increases in stimulatory TSH-R antibodies causing reactivation of Graves’ disease,or with progressive emergence of autonomy and hypermay be prompted by exposure function ina uninodular or multinodular goiter. The latter to a high iodine source. FNA biopsy should be repeated immediately when a nodule is found to be enlarging on suppressive therapy, and surgical exploration should be deemed inevitable unless cystic fluid or hemorrhage with benign cytologyis obtained. FNA should also be repeated when there is failure to obtain significant reduction in nodule size after 6-12 mo of suppressive therapy.
SUMMARY: APPROACH TO THE THYROID NODULE In summary, the majority of thyroid nodules will be follicular adenomas, which are benign tumors that may occur by itself or in multiples and may mimic normal thyroid function, trapping iodide and producing thyroid hormones. On radionuclide scan, they maybenonfunctional(“cold”),normallyfunctional(“warm”), or hyperfunctioning (“hot”). Hot nodules are almost always benign. Hot nodules with hyperthyroidism are treated with radioiodine or by surgery, whereas patients with euthyroid hot nodules may befollowedwithouttherapy,advisedtoavoidiodineexcess,andmonitored
44
Wartofiky
periodically with thyroid function tests. Other common benign nodules include coll adenomas or cysts. Concern should be raised that a thyroid nodule may be malignant in the setting of a history of irradiation to the neck in childhood, of associated cervical lymphad or of recent or rapid nodule enlargement. Routine laboratory tests are of little value in distinguishing between benign and malignant nodules. Fine needle aspiration for c ogy is the initial procedure of choice (7,55). Ultrasound for sizing and detection of cystic components, and scintiscansto confirm functional state maybe useful. Cysts or FNA mayreducetheirsizewith colloidadenomasdemonstratedtobebenignon levothyroxine therapy by approximately 50%. L-T~treatment is contraindicated for autonomous hyperfunctioning adenomata. The clinical challenge in thyroid nodule management is formulation of the most accurate but cost-effective diagnostic protocol. Evaluation may be most efficiently and cost-effectively performed by a consultant endocrinologist(56). Decision analysis has suggested thatFNA, the most accurate single evaluation parameter, provides a minim 1 (p. 8) advantage in quality-adjusted life expectancy over thyroid suppression. Figure suggests an algorithm that may be useful in a practice whereFNA is frequently utilized with experienced cytopathology support. Use of radionuclide scans as the initial step 5-10% of scans will obviate the need for may result in increased cost, since only aspiration, whereas 60-80% ofFNAs will eliminate scan requirements. Because aspi are only rarely simple cysts, sonogr tion will identify predominantly cystic lesions that phy has limited utility initially, but may be of value to follow results of suppressive therapy. Thyroxine-suppressive therapyis usually achieved ata dose of approx 1.7pg/ kg. Patientsmay be started on lower doses on the basis of age and potential underlying cardiovascular disease, with patient reevaluations within wk 6to monitor for symptom of hyperthyroidism and to evaluate serum TSH levels. From the initial dose, the dose TSH suppression or near suppression is then incremented by 0.025 mg every 6 wk until is documented, depending upon the target TSH range desired, unless the patient has symptoms that require lowering the dose of thyroxine. Nodules should be assessed fo change in size every 2 mo for the first 6 mo. If the nodule significantly decreases in size, intervals of follow-up may be gradually prolonged, with eventual yearly followup. FNA may be repeated when nodules fail to respond to thyroxine suppression after 6-12 mo, and sooner for any nodule that seems to be enlarging. Ninety-five percent of repeat FNA’s confirm the original diagnosis (57). Patients with a history of irradiation present a special situation. Historically these patients have been immediately referred to surgery because of their higher cancer Some clinicians now advocate FNA in the management of these patients as well, although sufficient evidencefor reliability of benign resultsis still lacking, owing to th (58). Follow-up evaluations frequent coexistence of both benign and malignant nodules employing ultrasound may be more sensitive for detection of nodules than physical examination or scanning (59). When surgical therapyis recommended, an ipsilateral lobectomy and isthmusectom is the most common approach in single nodules where the preoperative diagnosis is uncertain (see Chapter5). Frequently, frozen section histologic evaluationis inconclusive or unreliable and final diagnosis will require careful examination of permanent sections. Many papillary carcinomas have multicentric growth, with tumor foci in the
Medical Management
45
contralateral lobe in 30-82% of cases (60).Accordingly, when the ultimate diagnosis of is carcinoma, it is customary to complete a near-total thyroidectomy within 1 wk the first surgery. Studies have shown no differences in survival or recurrence rates between total and near-total thyroidectomy, and greater morbidity associated with total thyroidectomy (see Chapter 18). Inspection of regional lymph nodes with excision of suspicious nodes should also be performed in all cases of thyroid cancer. Near-total thyroidectomy is the initial procedure of choice in patients having thyroid nodules and a history of thyroid irradiation, due to the high incidence (54-75%) of bilateral disease (61).
REFERENCES
1. Spencer CA, Wang C-C. Thyroglobulin measurement: techniques, clinical benefits, and pitfalls. Endocrin Metab Clin N Amer 1995; 24:841-864. 2. Torrens JI, Burch HB. Serum thyroglobulin measurement: utility in clinical practice. Endocrinologist 1996; 6: 125-144. 3. Marsh DJ, Learoyd DL, Robinson BG. Medullary thyroid carcinoma: recent advances and management update. Thyroid 1995; 5:407-424. 4. HeshmatiHM, Gharib H, van Heerden JA, Sizemore GW. Advances and controversies in th diagnosis and management of medullary thyroid carcinoma. Amer J Med 1997; 103:60-69. 5. Vierhapper H, Raber W, Bieglmayer C, Kaserer K, Weinhausl A, Niederle B. Routine measurement of plasma calcitonin in nodular thyroid disease. J Clin Endocrinol Metab 1997; 8211589-1593. G, etal.Routine 6.PaciniF,FontanelliM,Fugazzola L, EliseiR,RomeiC,DiCoscio measurement of serum calcitonin in nodular thyroid diseases allows the preoperative dia sis of unsuspected sporadic medullary thyroid carcinoma. J Clin Endocrinol Metab 1994; 78:826-829. 7. Oertel YC. Fine needle aspiration and the diagnosis of thyroid cancer. Endocrin Metab Clin N Amer 1996; 25:69-92. 8. Gharib H, Goellner J R . Fine needle aspiration biopsyof the thyroid an appraisal. Annals Int Med 1993; 118~282-289. 9. Gharib H, Goellner J R , Johnson DA. Fine-needle aspiration cytologyof the thyroid. A 12year experience with 11,000 biopsies. Clin LabMed 1993; 13:699-709. 10. Caraway N P , Sneige N, Samaan NA. Diagnostic pitfalls in thyroid fine-needle aspiration: a review of 394 cases. Diag Cytopathol 1993; 9:345-350. 11. Gharib H, GoellnerJ R , Zinsmeister AR, Grant CS, VanHeerden JA. Fine neede aspiration biopsy in the thyroid. Ann Intern Med 1984; 101:25-28. 12. Ashcraft MW, VanHerle AJ. Management of thyroid nodules. II. Scanning techniques, thyroid suppressive therapy, and fine needle aspiration. Head Neck Surg 1981; 3:297-322. 13. MazzaferriEL. Management ofa solitary thyroid nodule. N Engl J Med 1993; 328553-559. 14. Ridgway EC. Clinical review 30: clinician’s evaluationof a solitary thyroid nodule. J Clin Endocrinol Metab 1992; 74:231-235. of thyroid neoplasms. Endocrine Pathol15. OertelYC. Fine-needle aspiration in the evaluation ogy 1997; 8~215-224. 16.Oertel YC, OertelJE.Diagnosisofbenignthyroidlesions:fine-needleaspirationand histopathologic correlation.Ann Diagnost Path 1998; 2:250-263. 17. Leenhardt L, Hejblum G, Franc B, DuPasquier-Fediaevsky L, Delbot T, LeGuillouzic D, et al. Indications and limits of ultrasound-guided cytology in the management of nonpalpable thyroid nodules. J Clin Endocrinol Metab 1999; 84:24-28. 18. Hatada T,Okada K, Ishii H, Ichii S , Utsunomiya J. Evaluation of ultrasound-guided fineneedle aspiration biopsy for thyroid nodules. AmerJ Surg 1998; 175:133-136.
46
Wartofiky
19. Singer PA. Evaluation and management of the thyroid nodule, Otolaryngol Clin N Am 1996; 29577-592. 20. Backdahl M, Wallin G, Lowhagen T, Auer G, Granberg, P. Fine-needle biopsy cytology
of patients with thyroid and DNA analysis: their place in the evaluation and treatment neoplasms. Surg Clin North Am 1987; 67:197-211. 21. Arturi F,Russo D, Giuffrida D, Ippolito A, Perrotti N, Vignen R, Filetti S. Early diagnosis by genetic analysisof differentiated thyroid cancer metastases in small lymph nodes. J Endocrinol Metab 1997; 82:1638-1640. 22. Erdogan MF, Kame1 N, Aras D, Akdogan A, Baskal N, Erdogan G.Value of re-aspirations in benign nodular thyroid disease. Thyroid 1998; 8:1087-1090. 23. RUSSOD, Arturi F, Suarez HG, Schlumberger M, DuVillard J-A, Crocetti U, Filetti S. Thyrotropin receptor gene alterations in thyroid hyperfunctioning adenomas. End&J Clin no1 Metab 1996; 81:1548-1551. 2 4 . VanHerle AJ, Rich P, Ljung BE, Ashcraft MW, Solomon DH, Keller EB. The thyroid nodule. Ann Intern Med 1982; 96:221-232. 25. Monzani F,Lippi F, Goletti 0, Del GuerraP, Caraccio N, Lippolis PV, et Percutaneous aspirationandethanolsclerotherapy for thyroid cysts. J Clin Endocrinol Metab 1994;
al.
78:800-802. 26. Papini E, Pacella C M , Verde G. Percutaneous ethanol injection CpEI): what is its role in the treatment of benign thyroid nodules? Thyroid 1995; 5:147-150. 27. Lippi F,Ferrari C, Manetti L, Rag0 T, Santini F, Monzani F, et al. Treatment of solitary of anresults Italian multicenautonomous thyroid nodules by percutaneous ethanol injection: ter study. J Clin Endocrinol Metab 1996; 81:3261-3264. 28. Mincheva L, Simeonov S , Troev D, Mitkov M, Pavlova M, niev D, Botushanov N. Percutaneousethanolsclerotherapy of autonomousthyroidnodules:preliminaryresults.Folia Medica 1997;39:49-54. 29. Caraccio N, Goletti 0, Lippolis PV, Casolaro A, Cavina E, Miccoli P, Monzani F. Is
percutaneous ethanol injection a useful alternative for the treatment of the cold benign thyroid nodule? Five years experience. Thyroid 1997; 7:699-704. 30. Goletti 0,Monzani F, Lenziardi M, Lippolis PV, DeNegri F, Carraccio N, et al. Cold thyroid nodules: a new application of percutaneous ethanol injection treatment. J Clin Ultrasound 1994; 22:175-178. 31. Zingrillo M, Collura D, Ghiggi MR, Nirchio V, Trischitta V. Treatment of large cold benign thyroid nodules not eligible for surgery with percutaneous ethanol injection. J Cli Endocrinol Metab 1998; 8333905-3907. 32. Bennedbrek FN, Nielsen LK, Hegedus L. Effect of percutaneous ethanol injection therapy Versus suppressive doses of L-thyroxine on benign solitary solid cold thyroid nodules: a randomized trial. 3 Clin Endocrinol Metab 1998;83:830-835. . 33. Estour B,Millot L, Vergely N, Clavier A, Dhondt0, Caillot A, et al. Efficacy of low doses of radioiodine in the treatment of autonomous thyroid nodules: importance of dosdarea ratio. Thyroid 1997; 7:357-361. 34. Asari AN,Siege1ME, DeQuattroV. Imaging of medullary thyroid carcinoma and hyperfun tioning adrenal medulla using iodine-131 metaiodobenzylguanidine. J NuclMed 1986;
27:1858-1860. for solitary 35. Gharib H, James EM, Charboneau JW.Suppressive therapy with levothyroxine nodules. N Engl J Med 1987; 317:70-75. 36. Celani MF, Mariani M, Mariani G. On the usefulnessof levothyroxine suppressive therapy in the medical treatment of benign solitary, solid, or predominantly solid thyroid nodules Acta Endocrinol 1990; 123:603-608. 37. Cooper DS. Thyroxine suppression therapy for benign nodular disease. J Clin Endocrinol Metab 1995; 80:331-334.
Medical
Management 47
38. Reverter JL, Lucas A, Salinas I, Audi L, Fox M, Sanmarti A. Suppressive therapy with levothyroxine for solitary thyroid nodules. Clin Endocrinol1992; 36:25-28. of benign solitary thyroid 39. Cheung PSY,Lee JMH, Boey JH. Thyroxine suppressive therapy nodules: a prospective randomized study. World J Surg 1989; 13:818-822. 40. Papini E, Bacci V, Panunzi C, Pacella CM, Fabbrini R, Bizzarri G, et al. A prospective randomized trial of levothyroxine suppressive therapy for solitary thyroid nodules. Clin Endocrinol 1993; 38507-513. 41. LaRosa GL, Lupo L, Giuffrida D, Gull0 D, Vigneri R, Belfiore A. Levothyroxine and potassium iodide are both effective in treating benign solitary solid cold nodules of the thyroid. Ann Int Med 1995; 122:l-8. 42. Lima N, Knobel M, Cavaliere H, Sztejnsznajd C, Tomimori E, Medeiros-Net0 G. Levothyroxine suppressive therapy is partially effective in treating patients with benign, solid thyr nodules and multinodular goiters. Thyroid1997; 7:691-97. et Long-term changes 43. Papini E, Petrucci L, Guglielmi R, PanunziC, Rinaldi R, Bacci V, al. in nodular goiter: A 5-year prospective randomizedoftrial levothyroxine suppressive therapy for benign cold thyroid nodules. J Clin Endocrinol Metab1998; 83:780-783. 44. Gharib H, Mazzaferri EL. Thyroxine suppressive therapy in patients with nodular thyroid disease. Ann Int Med 1998; 128:386-394. 45. Zelmanovitz F, Genro S, Gross JL. Suppressive therapy with levothyroxine for solitary thyroid nodules: a double-blind controlled clinical study and cumulative meta-analyses. J Clin Endocrinol Metab 1998; 83:3881-3885. 46. Ridgway EC. Medical treatment of benign thyroid nodules: have we defined a benefit? Ann Int Med 1998; 128:403-405. 47. Biondi B, Fazio S, Carella C, Amato G, Cittadini A, Lupoli G, et al. Cardiac effects of long term thyrotropin-suppressive therapy with levothyroxine. J Clin Endocrinol Metab 1993; 77~334-338. 48. BiondiB,Fazio S, CarellaC,SabatiniD,AmatoG,CittadiniA,etal.Control of adrenergic overactivity by P-blockade improves the quality of life in patients receiving long term suppressive therapy with levothyroxine. J Clin Endocrinol1994; Metab 78:10281033. 49. Wartofsky L.Does replacement L-thyroxine therapy cause osteoporosis? Advances Internal Med 1993; 4157-175. 50. Marcocci C, Golia F, Bruno-Bossio G, Vignali E, Pinchera A. Carefully monitored levothyroxine suppressive therapy is not associated with bone loss in premenopausal women. J Clin Endocrinol Metab 1994; 78:818-823. 51. Schneider DL, Bmett-Connor EL, Morton DJ. Thyroid hormone use and bone mineral density in elderly women: effects of estrogen. JAMA 1994; 271:1245-1249. 52. Molitch ME, Beck JR, Dreisman M, Gottlieb JE, Pauker SG. The cold thyroid nodule: an analysis of diagnostic and therapeutic options. Endocr Rev1984; 5185-199. 53. Fogelfeld L, Wiviott MBT, Shore-Freedman E, Blend M, Bekerman C, Pinsky S, et al. Recurrence of thyroid nodules after surgical removal in patients irradiated in childhood for benign conditions. N EnglJ Med 1989; 320:835-840. 54. Hennessey J V , Evaul JE, Tseng YL, Burman KD, Wartofsky L. L-thyroxine dosage: a reevaluation of therapy with contemporary preparations.Ann Intern Med 1986; 105:ll-16. 55. Wartofsky L,Oertel Y. Fine needle aspiration biopsy of thyroid nodules. In Nuclear Medicine Atlas, Van Nostrand D (Ed.) J. B. Lippincott, Philadelphia, 1987, pp. 193-200. 56. Ortiz R, Hupart KH, DeFesi CR, Surks MI. Effect of early referral to an endocrinologist on efficiency and cost of evaluation and development of treatment plan in patients with thyroid nodules. J Clin Endocrinol Metab 1998; 83:3803-3807. 57. Hamburger JI, Hamburger SW. Fine needle biopsy of thyroid nodules: avoiding the pitfalls, NY State J Med 1986; 86:241-249.
48
Wartojkky
58. Rosen IB, Palmer JA, Bain J, Strawbridge H, Walfish PG. Efficacy of needle biopsy in postradiation thyroid disease. Surgery 1983; 94:1002-1007. 59. Schneider A B , Bekerman C, Leland J, Rosengarten J, Hyun H, Collins B, et al. Thyroid nodules in the follow-up of irradiated individuals: comparisonof thyroid ultrasound with scanning and palpation. J Clin Endocrinol Metab 1997; 82:4020-4027. 60. Lennquist S . The thyroid nodule: diagnosis and surgical treatment. Surg Clin North Am 1987; 67:213-232. 61. Norton JA, DoppmanJL, Jensen RT. Cancer ofthe endocrine system. In Cancer: Principles and Practiceof Oncology, DevitaVT Jr, Helman L, Rosenberg SA(Eds.) J. B. Lippincott, Philadelphia, 1989, pp. 1269-1287.
5 Thyroid Nodules and Cancer Risk Surgical Management Orlo H. Clark
Patients with thyroid nodules are selected for thyroidectomy if they are candidates for cancer or have proven cancer, are experiencing related symptoms, or have cosmetic abnormalities. Since about4% of the United States population has thyroid nodules, as determined by the Framingham studies, and yet only 40 patients per million have clinical thyroid cancer, a selective approach must be used to determine who will be from thyroidectomy ( I ) . Factors that increase the risk that a thyroid nodule may be cancer are listed in Table 1. Patients with a family history of thyroid cancer are much more likely to develop thyroid cancer. Thus, about30% of patients with medullary thyroid cancer (MTC) have familial rather than sporadic disease (2). This familial disease occurs in four forms: 1. Familial medullary thyroid cancer without other endocrinopathies. 2. Familial medullary cancer with MEiN2A (MTC, hyperparathyroidism, and pheochromocy-
tomas). Some patients also have cutaneous lichen amyloidosis with a pruritic plaquelike skin rash over the scapular region and or concurrent Hirschsprung’s disease. m N 2 B (pheochromocytomas, marfanoid habitus, 3. Familial medullary thyroid cancer with mucosal neuromas, and ganglioneuromatosis). 4. Familial MTC as well as papillary thyroid cancer (3-6). Patients with familial MTC can now be diagnosed by a blood test for RET germ line point mutations (3-5).
RET somatic mutations are present in about 50% of sporadic medullary thyroid cancers, and specific mutations appear to correlate with tumor behavior (7,s). About 5% of patients with papillary and Hiirthle cell cancer have familial thyroid (9,IO). Patients cancer, but follicular thyroid cancer does not appear to be familial with autosomal dominant disorders causing disseminated gastrointestinal polyposis or Gardner syndrome (large and small bowel tumors, desmoid tumors, lipomas, and epidermoid cysts), and Cowden syndrome (multiple hemartomas, breast cancer, colon cancer, and nodular goiter) have an increased risk of thyroid cancer ( I O J I ) . Thyroid MENl and with familial cancer also appears to be more common in patients with hyperparathyroidism without MENl (I2,13). Loss of genetic material at or close to the MENl centromeric region on chromosome 1 1 has been documented in some sporadic
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, NJ
49
isk of Thyroid Cancer 1. Family history of thyroid cancer A. me dull^ thyroid cancer or MEN2 €3. Familial n o n m e d u l l ~thyroid cancer 1. Familial nonmedullary thyroid cancer 2. Cowden syndrome 3. Familial polyposis (Gardner’s) 4. Mul~pleendocrine neoplasia type I or familial ~ype~arathyroidism 2. Exposure to low or moderate dose ~ e r a p e ~ tradiation ic A. External radiation €3. Nuclear fallout 3. Hard thyroid nodule 4. New thyroid nodule in young person (under 20 years) or older person (over 60 years) 5. Thyroid nodule with adjacent lymphadenopathy 6. History of hoarseness with vocal cord paralysis.
thyroid cancers of follicular cell origin (14). Patients with other thyroid pathology also appear to be predisposed to develop thyroid cancer (15,16). E x p o s ~to e low or moderate doses of therapeutic radiation also ~amaticallyincreases the risk of thyroid cancer (17). There does not ear to be a ~ e s h o l ddose, since exposure to as little as 6 cCy of radiation appears to increase the risk of thyroid cancer sixfold (18). An almost linear increase in cancer frequency occurs as the dose of radiation increases from 6 to 2000 cGy, Higher doses of radiation, such as 5000 to 6000 cCy, can cause hypothyroidism, but thyroid cancer does not appear to increase appreciably probably because the thyroid cells are destroyed (17). ~ y r o i dcancer is more likely to occur in younger children after radiation exposure, which su~gestsa her edit^ predisposition to the increased frequency of thyroid cancer after exposure to low-dose therapeutic radiation or radiation fallout (19,20). Thus, after e x p o s ~ eto low-dose therapeutic radiation, or after exposure to radiation from the ~ h e ~ o bnuclear yl accident, thyroid cancer developed in several members of some families, but, in other neighbo~ngfamilies, tumors did not develop (19,2U). Thyroid nodules that develop in persons under 20 or over 60 years of age are more likely to be cancer, as are nodules that are associated with vocal cord par hoarseness, rapidly growing nodules, hard, solitary nodules, fixed nodules associated with palpable ipsilateral lymphadenopathy. In my experience, nodules that ache or are minimally painful are more likely to be me dull^ thyroid cancer. Regardless of the risk of cancer, fine needle aspiration (FNA) for cytological e x a ~ n a tion helps determine the histological nature of a nodule. FNA requires an experienced cytologistto interpret the aspirate. One can usually d e t e ~ i nby e cytologic^ e x ~ n a t i o n whether a thyroid nodule is benign (95% reliability), malignant (99% relia~ilit~), or suspicious (about 20% cancer), or whether the biopsy specimen is inadequate and warrants a repeat biopsy (21,22). When cytologic examina~onsuggests me thyroid cancer, the specimen should be stained for amyloid and for calcitonin
Surgical
Management 51
(carcinoembryonic antigen). The patient should also have a blood test for calcitonin and a urine test for VMA (vanillylmandelic acid), catecholamines, and metanephrine, to rule out a coexistent pheochromocytoma. FNA is often unreliable in patients with a thyroid nodule who have been exposed to low-dose therapeutic radiation because of (23). Around 40% of these patients the multifocal natureof the tumors in these patients will have thyroid cancer somewherein the thyroid gland, and the dominant noduleis the cancer in these patients only 60% of the time. Thus, there is a high false-negative rate (23). As a consequence, we recommend total or near-total thyroidectomy for these patients rather than needle biopsy. Patients with follicular neoplasms, follicular varian of papillary thyroid cancer, and Hiirthle cell neoplasms usually have thyroid nodules that are interpreted as suspicious by cytological examination. When these nodules are “cold” by radioiodine scanning, about 20% will be malignant, so that I recommend (22). When nodules are suspicious for papillary thyroid cancer, removal in most patients repeat FWA biopsy may help clarify the cytologic diagnosis, and ultimately about 50% of these tumors are found to be malignant (24).
OPERATIVE APPROACH Considerable controversy continues relating to the most appropriate treatment of patients with thyroid cancer. Since there are no prospective studies comparing various surgical or postsurgical therapies, this debate will probably continue. Most surgeons and endocrinologists agree that the minimal thyroid operation that should be done for a thyroid nodule that might be malignant is an ipsilateral total thyroid lobectomy and isthmusectomy. The reason for this recommendation is that, if further surgeryis needed, one does not have to operate in an area of scar tissue, and thus there should be no appreciable increased risk of complications such as hypoparathyroidism or recurrent laryngeal nerve injury (24). It is also more difficult to remove all of the remaining thyroid gland after a partial thyroidectomy because the remaining thyroid tissue is often adherent to the surrounding structures when less than a thyroid lobectomy has been performed. One reason for the controversy concerning the extent of thyroidectomy required is that most patients with papillary thyroid cancer have an excellent prognosis. Thus, patients with occult( 4 cm) papillary thyroid cancers without nodal involvement have 0.2% mortality rate(26,27). It is, therefore, difficult a 6%to 8% recurrence rate but only to improve on these numbers. When lymph node metastases are present or when there is angioinvasion within the occult papillary thyroid cancer, the recurrence rate and death rates are higher (26,27). We recommend total or near-total thyroidectomy for 1 cm. We realize that virtually all patients with papillary thyroid cancer larger than the mortality rate of patients considered to be at low risk byTNM, the AGES, M S , or MACIS classifications is less than 5%, and about 75% of all patients with thyroid (28). We believe, however, that, ifwecan cancerwouldbeclassified at low risk decrease this mortality rate further,it is advisable, unless the improvement in survival rate is countered by a high complication rate. It is also important to mention that the AGES, AMES, and MACIS classifications are postoperative classifications. For example, local invasion, tumor differentiation, resectability, and even distant metastases,
52
Clark
often are not recognized until after the operationor after a postoperative 1311 scan and serum thyroglobulin determination. Grant and colleagues (29) also reported that recurrent cancer is less common after bilateral thyroid operations in both low- and highrisk patients, as determined by the AGES classification. DeGroot and coworkers(30) and Mazzafem and Jhiang (31) also reported fewer recurrences and improved survival in patients after total or near-total thyroidectomy. The major reasons we recommend total thyroidectomy is that one can then make use of serum thyroglobulin levels and radioactive scanning to determine if all tumor has been removed or if residual tumor needs to be ablated with radioiodine (32-34). One might ask why not wait to see if patients will develop recurrent disease because most patients do not develop recurrent tumor. The problem withthis approach is that once the recurrent tumor becomes clinically evident or is evident on a chest radiograp the chance of curative therapy with I3'I decreases from about 70% to about 7% in (33,34). Other reasons to perform a total thyroidectom tumors that take up radioiodine are as follows: 1. it removes multifocal or bilateral disease that occurs in 30% to 87% of patients. 2. it lowers the recurrence rate. 3. it probably improves survival, as one-third to half of the patients who develop recurrent thyroid cancer die of thyroid cancer (32-34).
The easiest time to do a total thyroidectomy is also at the initial operation. Near-total thyroidectomy, leaving less than 1 g of thyroid tissue, rather than total thyroidectomy, should be done when the surgeon is concerned about the viability of the parathyroid glands or the recurrent laryngeal nerve during the initial dissection o the sideof the tumor. Leaving a small remnant of normal thyroid tissue on the contral eral side to the tumor that can subsequently be ablated with I3II is preferable in this situation. Numerous retrospective studies report that totalor near-total thyroidectomy followed by 13'1 ablation and TSH suppressive therapy resultsin the fewest recurrences and the best survival (30-32). Before thyroidectomy and during the initial thyroid operation, the surgeon should carefilly look for, and palpate for, enlarged lymph nod adjacent to the thyroid tumor and medial or lateral to the carotid sheath. All nodes in the central neck should be removed, and patients with palpable nodes in the lateral neck benefit from an ipsilateral modified radical neck dissection. For most patients today the histology of the thyroid tumoris known preoperatively a fine needle biopsy performed before the opera because of information gained from procedure.Needle biopsy is quite accurate for papillary, medullary, and anaplastic thyroid cancers, but cannot differentiate between a follicular or Htirthle cell adenom and a follicular or HUrthle cell carcinoma. After needle biopsy, one can usually plan the definitive operation and discuss what will be done with the patient preoperative At operationin patients with follicular or Hlirthle cell neoplasms by cytological tion, the surgeon should look for lymph nodes, and, if present, removefor them frozen section examination. About half the patients with follicular neoplasms confirmed by cytological examinations who are found to have thyroid cancer have a follicular of papillary thyroid cancer; nodal involvement is quite common in these patients, whereas only about 10% of patients with follicular cancer have nodal involvement (36).Frozen-section examination is unfortunately not very effective in differentiating
Surgical
Management 53
between benign or malignant follicular or Hitrthle cell neoplasms. However, those 4 cm, or occur in older patients, follicular and Hiirthle cell neoplasms that are larger than are more likely to be malignant. In patients with follicular or H W e cell neoplasms, I usually perform a thyroid lobectomy, since most patients will have benign disease. I also discuss the various options with the patient before surgery, including that in about 10% ofcasesasecondoperation(completiontotalthyroidectomy)maybe necessary if cancer is diagnosed only by permanent histological examination. Patients who havesolitary follicular adenomas usually do not require thyroxine postoperatively because recurrent follicular adenomas are rare.
REFERENCES 1. Vander JB, Gaston EA, Dawber TR. The significance of nontoxic thyroid nodules: final report of a 15-year study of the incidence of thyroid malignancy. Ann Intern Med 1968; 69537-540. 2. Wohllk N, Cote GJ, Evans DB, Goepfert TH, Ordonez NG, Gage1 RF. Application of genetic screening information to the management of medullary thyroid carcinoma and multiple endocrine neoplasia type 2. Endocrinol Metab Clin North Am, 1996; 251-25. 3. Donis-Keller H, Dou S , Chi D, Carlson KM, Toshima K, Lairmore TC, Howe JR, Moley JF, Goodfellow P, Wells SA Jr. Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet 1993; 2:851-856. 4. Mulligan LM,Kwok JB, Healey CS, Elsdon MJ, EngC,Gardner E, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature 1993; 363:458460. 5. Carlson KM, Dou S , Chi D, Scavarda N, Toshima K, Jackson C E , Wells SA Jr, Goodfellow PJ, Donis-Keller H. Single missense mutation in the tyrosine kinase catalytic domain of the RET protooncogeneis associated with multiple endocrine neoplasia type 2B. Proc Natl Acad Sci USA, 1994; 91:1579-1583. 6. Sabanci U, Al-Sobhi S, Galante M, Siperstein A E , Duh Q-Y, Clark OH. Simultaneous occurrence of medullary and papillary carcinoma in the same patient. American Thyroid Association, San Diego, November 1996. TM, Clark OH, Falko JM, 7. Jhiang SM, Fithian L, Weghorst CM, Mazzaferri EL, O’Dorisio George JM: RETmutationscreeninginpatientswithfamilialandsporadicmedullary thyroid carcinoma disease and a discovery of mutation in a sporadic disease. Thyroid a novel 1996; 115-121. 8. GrossmanRF, Tu SH, DQY, Siperstein A E , Novosolov F, Clark OH. Familial nonmedullary thyroid cancer: an emerging entity that warrants aggressive treatment. Arch Surg 1995; 130:892-897; discussion 898-899. 9. Ozaki 0, Ito K, Kobayashi K, Suzuki A, Manabe Y, Hosoda Y. Familial occurrence of differentiated, nonmedullary thyroid carcinoma. WorldJ Surg 1988; 12565-571. 10. Camiel MR. Mule JE, Alexander LL, Benninghoff DL. Association of thyroid carcinoma with Gardner’s syndrome in siblings. N Engl J Med 1968; 278:1056-1058. 11. Weary PE, GorlinFU, Gentry WC Jr, ComerJE, Greer KE. Multiple hamartoma syndrome (Cowden’s disease). Arch Dermatol 1972; 106:682-690. 12. Huang SM, DuhQY, Shaver J, SipersteinAE, Kraimps J, Clark OH. Familial hyperparathyroidism without multiple endocrine neoplasia. World J Surg 1997; 21:22-29. 13. Lips CJ, Vasen HF, Lamers CB. Multiple endocrine neoplasia syndromes. Crit Rev OncoY Hematol 1984; 2:117-184. 14. Matsuo K, Tang SH, Fagin JA. Allelotype of human thyroid tumors: loss of chromosome llq13 sequences in follicular neoplasms. Mol Endocrinol 1991;5:1873-1879.
54
Clark
15. Ron E, KeinnermanRA, Boice JD Jr, LiVolsi VA, Flannery JT, Fraumeni JF Jr. A population-based case-control studyof thyroid cancer. J Natl Cancer Inst1987; 79:l-12. 16. D’Avanzo B, La Vecchia C, Franceschi S, Negri E, Talamini R. History of thyroid diseases and subsequent thyroid cancer risk. Cancer Epidemiol Biomarkers 1995; Prevent 4:193-199. 17. Greenspan FS. Radiation exposure and thyroid cancer.JAMA 1977; 237:2089-2091. 18. ModanB,RonE,WernerA.Thyroidcancerfollowing scalpirradiation.Radiology 1977; 123~741-744. 19. Balter M.Chernobyl’s thyroid cancer toll [news]. Science 1995; 2701758-1759. 20. Schachner SH, Riley TR, Old J W , Taft DA, Hamwi GJ. Familial hyperparathyroidism, Arch Intern Med 1966; 117:417. 21. Lowhagen T, Granberg PO, Lundell G, Skinnari P, Sundblad R, Willems JS. Aspiration biopsy cytology (ABC) in nodues of the thyroid gland suspected to be malignant. Surg Clin North Am 1979; 593-18. 22. Gharib H, Goellner JR, Zinsmeister A R , Grant CS,Van HeerdenJA. Fine-needle aspiration biopsyofthe thyroid theproblemofsuspiciouscytologicfindings.AnnInternMed 1984; 101:25-28. 23. Rosen IB, Azadian A, Walfish PG, SalemS, Lansdown E, BedardYC. Ultrasound-guided of thyroid disease. Am J Surg 1993; fine-needle aspiration biopsy in the management 166~346-349. 24. Zieger MA, Chen H, Clark D,- Udelsman R, Westra WH. Papillary thyroid cancer: can operative management be solely based on fine needle aspiration? Paper presented at the American Collegeof Surgeons, 82nd Annual Clinical Congress, San Francisco, Californi October 7, 1996. 25. Levin KE, Clark AH, Duh QY, Demeure M, SipersteinA E , Clark OH. Reoperative thyroid surgery. [Comments]. Surgery 1992; 11 1:604-609. SM. Differentiated thyroid cancer long-term impact of initial the 26. Mazzafem EL, Jhiang Trans Am Clin Climatol Assoc 1994; 106:151-168; discussion 168-170. 27. Mazzafem EL. Papillarythyroidcarcinoma:factorsinfluencingprognosisandcurrent 1988 Jun; 15(3):x]. SeminOncol therapy[publishederratumappearsinSeminOncol 1987; 14315-332. 28. Hay ID.Papillary thyriod carcinoma. Endocrinol Metab Clin North Am 1990; 19545476. 29. Grant CS, Hay ID, Gough IR, Bergstralh El, Goellner JR, McConahey M. Local recurrence in papillary thyroid carcinoma: is extent of surgical resection important? Surgery 1988; 104~954-962. 30. DeGroot LJ, Kaplan EL, McCormick M, StrausFH.Natural history, treatment, and course of papillary thyroid carcinoma. J Clin Endocrinol Metab1990; 71:414424. 31. Mazzafeni EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer [see comments] [published erratum appears inAm J Med 1995 Feb; 98(2):215]. Am J Med 1994; 97:418-428. 32. Clark OH. Total thyroidectomy: the treatment of choice for patients with differentiated thyroid cancer. Ann Surg 1982; 196:361-370. 33. Clark OH, Levin K, Zeng QH, Greenspan FS, Siperstein A. Thyroid cancer: the case for total thyroidectomy. Eur J Cancer Clin Oncol 1988; 24:305-313. C, Gardet P, Travagli JP, Fragu P, 34. Schlumberger M, Tubiana M, De Vathaire F, Hill Lumbroso J, Caillou B, Parmentier C. Long-term resultsof treatment of 283 patients with lung and bone metastases from differentiated thyroid carcinoma. J Clin Endocrinol Metab 1986; 63:960-967. 35. Casara D, Rubello D, Saladini G, Masarotto G, Favero A, Girelli ME, Busnardo B. Differen features of pulmonary metastases in differentiated thyroid cancer: natural history and variate statistical analysisof prognostic variables. J Nucl Med 1993; 34:1626-1631. 36. Emerick GT, Duh QY, Siperstein AE, Burrow GN, Clark OH. Diagnosis, treatment, and outcome of follicular thyroid carcinoma. [Comments]. Cancer1993; 72:3287-3295.
I1 Thyroid Cancer General Considerations
6 Molecular Pathogenesis of Thyroid Cancer James Figge CANCERS OF THE THYROID FOLLICULAR EPITHELIUM
Signal Transduction Pathways The thyroid follicular epithelial cell (thyrocyte) responds to myriad growth-stimulating substances, including hormones, growth factors, cytokines, and other mitogens (1-13) as exemplified in Table 1. Thyrocyte responses to these factors are mediated by three distinct signal transduction pathways (Figs. 1-3). Each pathway features a cell surface receptor linked to a specific cytoplasmic signal transduction cascade: 1. Tyrosine kinase receptor/ras/mitogen-activated protein kinase pathway (Fig. 1) 2. Thyrotropin (TSH)receptodadenylate cyclase/protein kinaseA pathway (Fig. 2) 3. Receptor/phospholipase C/protein kinase C pathway (Fig. 3)
These pathways transmit mitogenic signals from the cell surface through the cytoplasm into the nucleus. Activation of some pathways will increase the concentration of cytoplasmic second messengers such as cyclic adenosine monophosphate (CAMP) or calcium. The tyrosine kinase receptor pathway activates a series of protein pho lation events that are involved in signal transduction. All of the pathways eventually activate nuclear transcription factors, stimulate new protein synthesis and interact with the cell-cycle machinery of the nucleus. The initial signals generated by each cascade in the cytoplasm are distinct; however, there is considerable (but never complete) convergence of the distal branches of the pathways, particularly within the cell nucleu Two distinct outcomes occuras a result of pathway activationin thyrocytes. The TSH receptodadenylate cyclase/protein kinase A pathway (Fig. 2) stimulates proliferation two pathways (Figs. 1 and 3) stimulate and maintains thyrocyte differentiation; the other two outcomes result from proliferation but promote thyrocyte dedifferentiation. These differential regulatory effects of the pathways on the synthesis of proteins that are involved in maintaining the normal thyrocyte phenotype (e.g., thyroglobulin, thyroid peroxidase, the TSH receptor, proteins involved in iodine trapping, and cell adhesion molecules such as E-cadherin).
The Tyrosine Kinase ReceptorlraslMAP Kinase Pathway Thetyrosinekinase receptor/ras/mitogen-activated proteinkinase (MAP kinase) pathway is depicted in Figure 1 (14). Activation of this pathway stimulates thyrocyte From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, NI
57
58
Figge Table 1 Selected Growth-Stimulating Factors for Thyroid Follicular Cells Reference
Hormones Thyrotropin (TSH) Human Chorionic gonadotropin (hCG)
2 3
Growth factors
Insulinlike growth factor4 (IGF-I) Epidermal growth factor (EGF) Tumor growth factor-alpha (TGF-a)
4 5 6
Cytokines Prostaglandin El(PGE,) Prostaglandin E, (PGE, Prostacyclin Iz (PGIJ Interleukin 1 (IL-1) Other mitogenic factors CAMP increase Agents that Bradykinin Adenosine Acetylcholine Thyroid stimulating immunoglobulin
2
10 11 (T.SI)
13
proliferation and lossof differentiation. The epidermal growth factor receptor (EGF-R is a classic example of a tyrosine kinase receptor that transmits mitogenic signals through this pathway. Ligands for EGF-R are epidermal growth factor itself, as well as tumor growth factor-alpha (TGF-a) (Table 1).TGF-a is known to be an autocrine growth stimulator of the thyroid follicular cell in certain proliferative conditions (15). Upon ligand binding, EGF receptors dimerize and activate their tyrosine kinase f resulting in receptor autophosphorylation. The mitogenic signalis then relayed via an adapter molecule (Grb2) and a guanine nucleotide exchange factor (mSOS) to a ras protein. The ras proteins are a family of 21-kDa guanine nucleotide binding proteins anchored to the plasma membrane. Three types are known in humans: H-ras, K-ras, two forms, an inactive guanosine ahd N-ras. Each type of ras protein can exist in diphosphate (GDP)-bound form andan activated guanosine triphosphate (GTP)-bound form. The mSOS protein facilitates the binding of GTP to ras, resulting in its activ (16). Inactivation of ras occurs by hydrolysis of GTP, a reaction that is catalyzed by the intrinsic GTPase activityof the ras protein. Mutations in the GTP binding domain of ras (encoded by codons 12 and 13) allow GTP to bind but lock the protein in its rus gene inactivate active GTP-bound state. Likewise, mutations in codon 61 of the the intrinsic GTpase function and result in permanent activationof the protein(17,18). Either type of mutation will result in the continuous unregulated activation of the as rus that downstream signal transduction pathway. Growth-stimulatory genes such
Pathogenesis Molecular
59
Cancer of Thyroid
I Tyrosine Kinase Receptor (EGF-R)
active)
GrbP MAPKK
-
c fos
Proliferation Effects:
Differentiation Genomic Instability
0
0 0
Fig. 1. Tyrosine kinase receptor/rm/mitogen-activatedprotein kinase pathway.
can be activated by genetic alterations (e.g., single base substitutions, gene amplifi or chromosomal rearrangements) are known as protooncogenes. The activated forms of these genes are called oncogenes. Activating mutations in ras genes are frequently found in human thyroid cancers (see the subsection ras Proteins below). Furthermore, transgenic mice harboring an activated mutant ras gene under control ofa thyroglobulin promoter develop papillary carcinomas, thus demonstrating the role of ras mutations in thyroid oncogenesis (19). Following activation, ras can in turn activate a cascade of kinases that are involved in transmitting the proliferative signal to the cell nucleus. The first of these kinases is another protooncogene product known as Raf-1. The signal is then transmitted to the next kinase in the cascade, MAP kinase kinase (MAPKK or MEK). This kinase, in turn, activatesMAPkinase(MAPK),whichcanphosphorylatealargenumberof rsk, which can phosphoryregulatory proteins in the nucleus. MAP kinase also activates late and activate the nuclear transcription factors, c-fos and (20)c-jun (see the subsection Nuclear Transcription Factors below). Two linesof evidence demonstrate that expression of an activated mutant ras protein can induce genomic instability. The first evidence was obtained using the bacterial
60
Figge
~~
-
TSH R
n
c AMP
PK-A
Effects:
Proliferation Differentiation
Fig. 2. TSH receptorladenylate cyclaselprotein kinaseA pathway.
I
IP3
DAG
Ca2+
PK-C
4
J.
4
CALMODULIN CAM KINASE
Effects:
Proliferation Differentiation
Fig. 3. Receptorlphospholipase Uprotein kinase C pathway.
0 0
Pathogenesis Molecular
Cancer of Thyroid
61
regulatory elements of the E. coli lac operon, which have previously been shown to stringently regulate the expression of a target gene integrated into the context of a mammalian chromosome (21,22). In the basal state, expression of the target gene is nearly fully repressed; however, on addition of the lactose analogue, isopropyl-B-Dthiogalactoside (IPTG), transcriptionof the target geneis induced (21). Stambrook and this system to regulate the expression of an activated mutant colleagues (23,24) adopted rus gene in a cell line, NIH 3T3. It should be noted that these cells are known to harbor p53 mutations and are, therefore, susceptible to genomic damage. Upon treating the cells with IPTG, a transformed phenotype was induced (23). Furthermore, within the time frame needed for one cell cycle, there was a marked increase in the number of gross chromosomal aberrations noted on karyotype analysis, indicating that expression of a mutant ras protein can rapidly induce genomic instability (24). In the second experimental paradigm, the normal ras allele in a rat fibrobast line was replaced byan activated mutant ras gene using the technique of homologous recombination (25). Expression of the mutant ras gene under control of its natural promoter increased the rate of spontaneous transformation in these cells, and the m ras gene was amplified in the majority of transformed cells. These data indicate that expression of a mutant ras protein at normal levels is not sufficient to directly transform cells, yet it is sufficient to induce gene amplification events that eventually result in overexpression of the mutant allele.
TSH ReceptorlAdenylate CyclaselProtein Kinase A Pathway The TSH receptor/adenylate cyclase/protein kinase A pathway is depicted in Figure
2 (26). As would be expected, activation of this pathway stimulates thyrocyte prolifera-
tion and maintenance of the differentiated phenotype. Upon TSH binding, the TSH is a receptor changes its conformation and activates the stimulatory Gs protein. Gs heterotrimeric complex containing an active component (alpha subunit, Gsa), which is encoded by the gsp gene. In the inactive basal state,Gsa is bound to GDP, whereas two compoupon activation, Gsa exchanges GTPfor GDP, dissociates from the other nents of the complex, and activates adenylate cyclase, thereby generating CAMP. The increased level of the second messenger, CAMP, then activates protein kinase A (PKA), which mediates downstream effects of CAMP. Predictably, activating mutations of both TSH-R and gsp genes are commonly found in toxic (hyperfunctioning) adenomas (27). These adenomas are driven by the constitutive activation of the adenylate cyclase pathway and maintain differentiated function (e.g., thyroid hormone secretion). Surprisingly, mutations ingsp are also found commonly in thyroid cancers (see the subsection G Proteins below); therefore, gsp is also classified as a protooncogene.
ReceptorlPhospholipase CIProtein Kinase C Pathway The phospholipase Uprotein kinase C (PK-C) pathway (28) (Fig. 3) is also active in thyrocytes and stimulates proliferation andloss of differentiation. The pathway can be activated by a number of mitogens, including bradykinin, adenosine, and acetylcholine (Table l), aswell as TSH. Activation ofan appropriate receptor can activate another G protein known as Gq. The activated GTP-bound Gqa subunit can activate phospholipase C (PL-C) which, in turn,converts phosphatidylinositol 4,5-biphosphate to inositol 1,4,5-triphosphate (Ip3) and diacylglycerol (DAG). IP3 increases the level
62
Figge
of intracellular calcium which, upon binding to calmodulin, activates a set of kinases. PK-C is activated by DAG.
Nuclear Pathways That Control Genomic Integrity The nuclear protein, p53,is a tumor suppressor gene product that functions in cell DNA (29). Tumor suppressor gene product cycle control and in the repair of damaged normally function to restrict cell proliferation. Therefore, genetic alterations that ina vate tumor suppressor genes will promote unregulated thyrocyte proliferation. p53, (30) andinducesexpressionof whenactivated,functionsasatranscriptionfactor p21c'p'mF' and other target gene products that arrest cells at a checkpoint in the late G1 phase of the cell cycle (31). In the presence of DNA damage, this p53-mediated pathway allows DNA repair processes to occur before the onset ofDNA synthesis (S phase) and mitosis, thus reducing the likelihood that mutations will be passed to cells (32). Under some conditions that are not completely understood, p53 can also trigger apoptosis (programmed cell death) (33). Cells lacking normal p53 function may develop genomic instability (34) because they lack the normal checkpoint control in late G1, a situation that contributes to the pathogenesis or progression of malignancy. Such cells may accumulate mutations that affect the functioningof other critical growth-controlling genes. For example, germl transmission of a mutant p53 allele in the Li-Fraumeni syndrome results in a marked increase in cancer risk (35). p53 mutations are the most frequently observed genetic defects in human cancers (36). There are five known mechanisms that can inactivate p53 function (31):
1. Missense mutations that usually disrupt the specific DNA binding activity of the p53 protein 2. Nonsense or splice site mutations that result in expression of a truncated p53 protein 3. Gene deletion, eliminating expression of the protein from the affected allele 4. Complex formation with a viral oncogene product, suchas SV40 T antigen, which inactivates the p53 protein 5. An increased level of the MDM2 protein, which binds to p53 protein and inactivates its transcriptional activation function
Inactivating mutations of p53 are frequently observed in human anaplastic thyroid carcinomas (see the subsection p53 below). The human MDM2 gene was initially found to be amplified in sarcomas (37), and subsequently it was shown that MDM2 protein can form a stable complex with p53 protein (38). p53 protein can be rendered nonfunctional by complex formation with MDM2protein (31,38,39). Specifically,MDM2binds to theacidictranscriptional activation domain in the amino terminal portion of p53 (39), thereby preventing p53 frominteractingwiththetranscriptionalmachineryinthecellnucleus.Therefore, overexpression of MDM2 protein inhibits the ability of p53 to activate the transcri of its target genes such as ~ 2 1 ~(38). ~ It' is~ also ~ of ' interest to note that the p53 gene product can stimulate the transcriptionof the MDMZ gene (40). Therefore, there is an autoregulatory feedback loop that involves both p53 and MDM2 (41).
Specific Molecular Aberrations in Human Thyroid Cancers A variety of molecular aberrations have been described in human thyroid cancers involving:
Molecular Pathogenesis of Thyroid Cancer
63
1. DNA methylation 2. Growth factors
3. Cell surface receptors 4. Signal transduction proteins
5. Nuclear transcription factors 6. Nuclear proteins that control genomic integrity 7. Cell surface adhesion molecules
In some cases the aberrations represent primary etiological events, such as the act of an oncogene by point mutation, gene amplificationor chromosomal rearrangement, or the inactivation of a tumor suppressor gene by mutation or deletion. According to the “multiple hit” theory of carcinogenesis, thyroid cancer is thought to arise from a series of genetic alterations that result in the sequential activation of oncogenes and the inactivation of tumor suppressor genes. In some cases a genetic alteration can be inherited, but usually these lesions are acquired. Thyrocytes harboring an activated oncogene oran inactivated tumor suppressor gene may derive a growth advantage over normal cells, leading to the clonal expansion of a population of abnormal cells. Some of these abnormal cells may acquire additional mutations, resulting in further growth advantage or greater susceptibility to genetic damage. Certain genetic lesions (e.g., ras activation, p53 inactivation) appear to render rapidly dividing cells more susceptible so that hyperplastic thyroid tissue harboring such mutations to further genetic damage, would be subject to developing mutations in other critical growth-controlling genes. The “multiple hit” theory therefore predicts a process of clonal evolution of thyroid neoplasms. The clonality of both benign and malignant thyroid neoplasms has been observed experimentally (42-44).
DNA Methylation Methylation of cytosine in eukaryotic DNA occurs in specific patterns controlled by enzymatic processes. Aberrant DNA methylation patterns occur frequently in ben (45), suggesting that this is an early event in the evolution and malignant thyroid tumors of thyroid neoplasia. Growth Factors
Some papillary cancers produce TGF-a. As previously noted, this can bind to EGF receptors and may function as a growth-stimulating factor in an autocrine positive feedbackloop (46). Insulinlikegrowthfactor I (IGF-I)mightalsofunctionasan autocrine factor in thyroid neoplasms (47,48).
Cell Surjace Receptors c-erbB and c-erbB2/neu Thec-erbBprotooncogeneencodestheEGFreceptor.Acloselyrelatedgene, c-erbB2heu encodes a tyrosine kinase receptor with an unknown ligand. Some pa two-at to threefold higher levels carcinomas express mRNA transcripts from these genes than normal thyroid tissue (49). This is a secondary rather than primary abnormality. PDGF Receptor The receptor for platelet-derived growth factor (PDGF) is overexpressed in a certain anaplastic cancer cell line (50).
64
Figge Tyrosine Kinase Domain
TM
C-ret
I
ret / PTC1
ret / PTC2
I
ret / PTC3
ELE 1
Fig. 4. Patterns of ret oncogene rearrangement found in papillary carcinomas.
ret and trk The ret and trk protooncogenes are activated as primary genetic events in papilla carcinomas (52-64). Both encode tyrosine kinase receptors. In the case of the trk protooncogene, the encoded receptor is for nerve growth factor(NGF). The ligand for the ret protooncogene product is glial-cell-line-derived neurotrophic factor (GDNF). Bothreceptorsareexpressedincells of neuroectodermalorigin,andaresilentin normal thyrocytes. Activation to oncogene status occurs as a result of a chromosomal rearrangement in papillary carcinoma cells. The effectof the rearrangement is to link the tyrosine kinase domainof the receptor to an unrelated protein segment, forming of an active promoter in thyroid follic hybrid protein thatis expressed under control cells. This results in the expression of a constitutively activated tyrosine kinase d in the cancer cell. The activated tyrosine kinase domain is oncogenic, and can induce thyroid papillary carcinomas in transgenic mice (63u,63b). There are severalknown ret rearrangements; three types are shown in Figure 4. The arm of human chromosome10 (lOql1.2). ret protooncogene (c-ret) resides on the long The retPTC1 rearrangement features a fusion between the tyrosine kinase domainof c-ret and theH4 locus, also on the long arm of chromosome 10 (1Oq21). This indicates that an intrachromosomal rearrangement (called an inversion) has resulted in the between the two genes. The promoter region and5’ end of the chimeric sequence are derived from the H4 locus. This places the expression of the tyrosine kinase domain under the control of a heterologous promoter (H4). The ret/PTC3 rearrangement also arises from an inversion and features a fusion between the tyrosine kinase domain o c-ret and the ELEl gene (also located on the long arm of chromosome 10). Finally,
Pathogenesis Molecular
Cancer of Thyroid
65
the ret/PTC2 rearrangement features a fusion between the tyrosine kinase domain of c-ret and theRI-a gene from chromosome 17.This rearrangement is believed to result from a reciprocal translocation between chromosomes 10 and 17. Rearrangements of ret are highly specific for papillary thyroid cancer (55,60) and are rarely found in tumorsof other tissues orin other types of thyroid neoplasms. Ret is activated in about 10% to 30% of papillary cancers in different series, but the rate is lower in some geographic areas(62). Ret rearrangements may be induced as a result ofradiationexposure (59), andretactivationhasbeenseeninChernobyl-related papillary cancers in children (56-58) (see Chapter 8). TSH Receptor
Activating mutations of the TSH receptor have been described in toxic (hypefinctioning) adenomas (27). Itis well known that these adenomas rarely exhibit malignant behavior. This is in keeping with the idea that activation of the adenylate cyclase pathway (Fig. 2) maintains the differentiated thyrocyte phenotype. However, the first report of a TSH receptor mutation occurring in three differentiated thyroid carcinomas was recently presented (65). The mutation was discovered in the third intracellular loop of the receptor, in a region critical for signal transduction. This mutation would be expected to cause constitutive activation of the CAMP pathway. As predicted, the affected cancers had increased basal levels of CAMP.
Signal Transduction Proteins ras Proteins As shown in Table 2, ras proteins are activated in a variety of thyroid cancers (52,53,66-86). Single base substitution mutations in codons 12, 13, or 61 have been described in all three rus oncogenes. In addition, activationof ras by gene amplification has also been described in thyroid cancer (72). As previously described, these mutatio are expected to constitutively activate MAP the kinase cascade (Fig. 1)and will stimulate proliferation, loss of the differentiated phenotype, and genomic instability. When data are pooled from multiple studies (Table 2), it can be seen that rus point mutations (50%)and follicular cancers (29%) than in papillary occur more frequently in anaplastic carcinomas (12%), although there are wide variations from one study to the next; for 0% to 62% in papillary carcinomas. Multiple example, the prevalence varies from factors may account for the variations between different studies, including small sa sizes, differences in iodine intake, different levels of serum TSH, environmental exporus mutations. Furthermore, sure to radiation, and differences in genetic predisposition to the majority of studies relied on DNA hybridization techniques to detect single base in only a few samples. changes and the results were confirmed with DNA sequencing This type of indirect technique may yield false-negative and false-positive results. Only one study employed direct DNA sequencing without prior screening to detect mutatio (85). Thus,methodologicaldifferencesmayalsoaccountforsomevariability between studies. It is of interest to note that the prevalencerusofmutations is about equal in follicular adenomas (28%) and follicular carcinomas (29%), suggesting thatrus mutations occur rus mutation may early in the pathogenesis of follicular cancers. For example, the
Figge
66 Table 2 Point Mutations in ras Genes in Thyroid Neoplasms
Prevalence of Point Mutations in All Three R a s Genes Reference
Jxmoine (66,68) and Wright (70) Fusco (52) Bongarzone (53) Namba (73) Suarez (75)
Karga (76)
shi (77)
Iodine-sufficient area Iodine-deficient area
Hara (81) Manenti (83) Hone (85) Pooled results
Papillary Carcinoma
Follicular Adenoma
Follicular Carcinoma
Anaplastic Carcinoma
3/17 (18%) 0120
8/24 (33%) nt
8115 (53%) nt
(60%)
nt
nt
nt
6/24 (25%) 6/13 (46%) 019
013 (0%)
nt
(0%)
2/16 (13%) 3/14 (21%) 8/13 (62%) 1/15 (7%) 0110 (0%)
0112
(0%)
13191 (14%) 013 1
(0%)
2/12 (17%) 11/13 (85%) nt
(0%)
1/22 (5%)
311261 (12%)
34/121 (28%)
6/10 nt
1/1 (100%) 2/14 (14%) 1/10 (10%) 316 (50%) nt
111 (100%) nt
5/21 (24%) nt
115 (20%) nt
20170
8/16 (50%)
(29%)
nt nt nt
nt, not tested
occur before or during the follicular adenoma stage, which might, in certain cases, progress to the follicular carcinoma stage. It is not clear whether M S lesions occur early in the pathogenesis of papillary cancers. Several studies suggest that mutations in N-ras codons are more commonly found in aggressive follicular and papillary (76,79,81). Further studies using larger sample sizes and rigorous methodology with confirmation of results byDNA sequencing are needed to better understand the corre tion between rus mutations and clinical tumor behavior. G proteins The oncogene gsp, which encodes theGsa subunit, is commonly mutated in hyperfunctioning adenomas (27). Activating mutations in gsp have also been described in thyroid cancers(79,87). This suggests that CAMP is a relevant growth signal that, und certain circumstances, can contribute to oncogenesis. It is likely that gsp mustbe activated in concert with another oncogene in order to promote cancer formation.
Pathogenesis Molecular
of Thyroid Cancer
67
Nuclear Transcription Factors Nuclear transcription factors such as c-fos and c-myc are overexpressed at the level of mRNA in some thyroid cancers(88), but this appears to be a secondary consequence of increased proliferation. No structural rearrangements have been identified in these genes in thyroid cancers. Nuclear Proteins That Control Genomic Integrity
P53 Recent data have established that inactivation of the p53 tumor suppressor protein by gene mutation is frequently implicatedin the pathogenesis of anaplastic (undifferentiated) thyroid carcinoma (89-95). However, most authors have not found p53 gene mutations in well-differentiated papillary thyroid carcinomas (89-93,9596). Zou and colleagues (94) reported finding p53 gene mutations in 7 well-differentiated papillary carcinomas and 3 papillary carcinomas with evidence of regional dedifferentiation (solid foci) in a groupof 40 papillary carcinomas. Three of the reported mutations did not result in an amino acid change. Alterations involving p53 have also been detected in some Chernobyl-related papillary carcinomas (see Chapter8). Fagin and associates (93) found a p53 mutation in 1 of 11 follicular carcinomas, and Zou and colleagues (94) identified a mutation in 1 of 4 follicular carcinomas. Some anaplastic carcinomas are thought to arise from preexisting foci of differe thyroid cancer. Thereis strong evidence that mutational inactivation of p53 is involved in the transition from differentiated to undifferentiated (anaplastic) thyroid cancer. Donghi and associates(92) studied an anaplastic carcinoma that contained a differentiA p53 mutation was present in cells ated region as well as the undifferentiated portion. from the undifferentiated area and also from a lymph node metastasis, but not from cells derived from the more differentiated region.A similar observation was reported by It0 and coworkers (90). These observations suggest that p53 mutations arise relatively late in the evolutionof anaplastic cancer. MDM2 Zou and colleagues (97) found twofold overexpression of MDM2 mRNA in 19% (3/16) of papillary carcinomas and threefold overexpression in a single follicular carcinoma that was studied. Jennings and colleagues (98) observed nuclear MDM2 protein accumulation in 33% (8/24) of papillary carcinomas. These observations suggest that p53 might be inactivated in a subset of well-differentiated thyroid carcinomas due to overexpression of MDM2 protein.
Cell Surface Adhesion Molecules
CD44 CD44 is a polymorphic family of integral membrane proteoglycans and glycoproteins implicated in diverse processes such as cell-cell adhesion, cell-matrix adhesion, cell migration and tumor metastasis (99,100). CD44 is a major receptor for hyaluronate (101). The heterogeneity of CD44 results from posttranslational modifications as well as “alternative mRNA splicing” of up to 10 variant exons that encode parts of the extracellular domain. The process of alternative mRNA splicing allows different combinations of the variant exons to be incorporated into CD44 mRNA transcripts, resulting
68
Figge
in the generation of multiple different CD44 protein isoforms. In rodent models, some CD44 isoforms can confer metastatic behavior to tumor cells (102,103). Recent data demonstrate that variant CD44 molecules are expressed widely throughout the huma body on epithelial cells in a tissue-specific pattern (104,105), suggesting that the process of alternative splicing is normally tightly regulated. Significant levels of CD44 protein (106). Ermak are expressed on the plasma membranes of papillary thyroid cancer cells is deregulated andcoworkers (107,108) reported that alternative splicing of CD44 in a variety of thyroid lesions (goiters, adenomas, papillary carcinomas). Papillary carcinomas exhibit specific patterns of aberrant CD44 mRNA splicing (108). These aberrations are postulated to affect the function of CD44 protein molecules on the c surface and might, at least in part, regulate papillary thyroid cancer growth patterns and metastatic potential. E-Cadherin E-cadherin is a calcium-dependent cell adhesion molecule required for normal ep lial function and postulated to play a role in tumor invasion. Data from several model suggest that E-cadherin is a suppressor of tumor spreading and invasion (109,110). E-cadherin mRNA levels and protein immunoreactivity are equally high in normal thyroid tissue and benign thyroid disorders, but are both markedly reduced in anap mRNA levels and immuthyroid carcinomas(111). In papillary carcinomas, E-cadherin noreactivity are variable, ranging from normal to markedly reduced. The E-cadherin mRNA levels in follicular carcinomas are high but immunoreactivity varies considerably. A good correlation was found between the level of E-cadherin and steady-state TSH receptor mRNA, suggesting that E-cadherin is a marker of differentation in thyro malignancies (111). The loss of E-cadherin expression in anaplastic carcinomas may, in part, explain the aggressive behavior of these cancers at the molecular level. Other Genetic Loci Several additional genes that might be important in thyroid carcinogenesis ha mapped (112,113) to chromosome llq13, and chromosome 3p. These loci may be of particular importance in the pathogenesis of follicular neoplasms. Genetic Syndromes Associated with Thyroid Cancer The prevalenceof thyroid canceris increased in certain genetic syndromes (114-11 7) such as Gardner syndrome, adenomatous polyposis coli, and Cowden’s disease. There are also a few families with a clustering of papillary cancers. MEDULLARY THYROID CARCINOMA
Medullary thyroid carcinomas arise from the C cells of the thyroid and may be sporadic (80%) or familial (20%). The genetic predisposition to develop a familial medullary carcinoma is conferred by a point mutationin the germline DNA encoding the ret oncogene (118). Figure 5 summarizes the current data regarding the status of inherited ret mutations in three inherited medullary carcinoma syndromes: MEN2A, MEN2B, and familial non-MEN medullary carcinoma(FMTC). These mutations serve to constitutively activate the tyrosine kinase function of the ret gene product and predispose to development of neoplasia. Lesions conferring susceptibility to MEN2A map to exons 10 and 11,encoding part of a cysteine-rich region in the extracellular
69
Molecular Pathogenesis of Thyroid Cancer Sporadic
FMTC
I
MEN 2A 609.611
Extracellular Domain
R I
MEN 28
H
Intracellular Domain
Fig. 5. Map of inherited ret mutations predisposing to medullary carcinoma in MEN 2A, MEN 2B, and FMTC. Some sporadic medullary carcinomas harborret mutations in the tumor DNA but not the germline DNA.
domain of the receptor.MEN2B maps most commonlyto codon918 in exon 16, which codes for part of the tyrosine kinase domain. Some sporadicalso tumors have mutations (in the tumor DNA but not germline DNA) that map to the tyrosine kinase domain.
REFERENCES
1.LewinskiA,PawlikowskiM,CardinaliDP.Thyroidgrowth-stimulatingandgrowthinhibiting factors. Biol Signals 1993;2:313-351. J, Dumont JE.Mitogenic effectsof thyrotropin and adenosine 2. Roger P, Taton M, Van Sande 3’,5’-monophosphate in differentiated normal human thyroid cells in vitro. J Clin Endocrino1 Metab 1988; 66:1158-1165. 3. Hershman JM, Lee HY, Sugawara M, Mirell C J , Pang X P , Yanagisawa M, Pekary AE. Human chorionic gonadotropin stimulates iodide uptake, adenylate cyclase, and deoxyrib nucleic acid synthesis in cultured rat thyroid cells. J Clin Endocrinol Metab 1988; 67:74-79. AC, Ingbar SH.Insulin-like growth factor-I stimulates 4. Tramontano D, Cushing GW, Moses the growthof rat thyroid cells in culture and synergizes the stimulation of DNA synthesis induced by TSH and Graves’-IgG. Endocrinology 1986; 119:940-942. 5. Roger PP, Dumont JE. Epidermal growth factor controls the proliferation and the expres 1982; 1U.209-212. of differentiation in canine thyroid cells in primary culture.Lett FEBS 6. Grubeck-Loebenstein B, Buchan G, Sadeghi R, Kissonerghis M, Londei M, Turner M, Pirich K, Roka R, NiederleB, KassalH, Waldhausl W, FeldmanM. Transforming growth factor beta regulates thyroid growth: role in the pathogenesis of nontoxic goiter. J Clin Invest 1989; 83:764-770. 7. Lupulescu A. Goiter formation following prostaglandin administration in Amrats. J Path01 1976; 85~21-35. 8. Pawlikowski M, Kunert-Radek J, Lewinski A. Effect of prostaglandins on the mitotic activity of rat thyroid in organ culture. Endokrynol Pol 1982; 33:129-134.
70
Figge
9. Mine M, Tramontano D, Chin WW, Ingbar SH. Interleukin-1 stimulates thyroidgrowth cell and increases the concentrationof the c-myc protooncogene mRNAin thyroid follicular cells in culture. Endocrinology 1987; 120:1212-1214. 10. Raspe E,Laurent E, Andry G, Dumont JE. ATP, bradykinin, TRH and TSH activate the Caz+-phosphatidylinositolcascade of human thyrocytes in primay culture. Mol Cell Endocrinol 1991; 81:175-183. 11. Sho K, Okajima F, Majid MA, Kondo Y.Reciprocal modulation of thyrotropin actions by P1-purinergicagonistsin FRTL-5 thyroidcells.JBiolChem 1991; 266:12,18012,184. 12. RaspeE,LaurentE,CorvilainB,Verjans B, ErneuxC,Dumont JE. Control of the intracellular Ca2+ concentration and inositol phosphate accumulation in dog thyrocyte cascade activaprimary culture: evidence for different kinetics of Ca2+-phosphatidylinositol 1991; tion and for involvement in the regulation of H202production. J Cell Physiol 146242-250. 13. ValenteWA,VittiP,RotellaCM,Vaughan MM, Aloj SM,GrollmanEF,AmbesiImpiombato FS, Kohn LD. Antibodies that promote thyroid growth: a distinct populati of thyroid-stimulating autoantibodies. N Engl JMed 1983; 309:1028-1034. 14. Cantly LC, Auger K R , Carpenter C, Duckworth B, Graziani A, Kapeller R, Soltoff S. Oncogenes and signal transduction. Cell 1991; 64:281-302. 15. Lemoine N R , Hughes CM, Gullick WJ, Brown CL, Wynford-Thomas D. Abnormalities of the EGF receptor system in human thyroid neoplasia. Int J Cancer 1991; 49558-561. 16. Egan SE, Giddings BW, BrooksMW, Buday L, SizelandAM, Weinberg RA.Association is implicated in tyrosine kinase signal transductio of Sos Ras exchange protein with Grb2 and transformation. Nature 1993; 363:45-51. 17. Bourne HR, Sanders DA, McCormick F. The GTPase superfamily: a conserved switch for diverse cell functions. Nature 1990; 348:125-132. 18. Bourne HR, Sanders DA, McCormick F. The GTPase superfamily: conserved structure and molecular mechanism. Nature 1991; 349:117-127. 19. Rochefort P, Caillou B, Michiels FM, Ledent C, Talbot M, Schlumberger M, Lavelle F, Monier R, Feunteun J. Thyroid pathologies in transgenic mice expressing a hu ras gene driven by a thyroglobulin promoter. Oncogene1996; 12:11 1-1 18. 20. Roberts TM. A signal chain of events. Nature 1992; 360534-535. TM, Livingston DM. Stringent regulation of stably 21. Figge J, Wright C, Collins CJ, Roberts integrated chloramphenicol acetyl transferase genes by E. coli lac repressor in monkey cells. Cell 1988; 52:713-722. 22. Brown M,Figge J, Hansen U, Wright C, Jeang K, Khoury G, Livingston DM, Roberts TM. lac repressor can regulate expression from a hybrid SV40 early promoter conta a lac operator in animal cells. Cell 1987; 49:603-612. 23. LiuHS,Scrable H, Villaret DB, Lieberman MA, StambrookPJ.Controlof Ha-rusmediated mammalian cell transformation by Escherichia coli regulatory elements. Cancer Res 1992; 52:983-989. 24. Denko NC, GiacciaAJ, Stringer J R , Stambrook PJ. The human Hu-ras oncogene induces Proc Natl Acad Sci USA genomic instability in murine fibroblasts within one cell cycle. 1994; 915124-5128. 25. Finney RE, Bishop JM. Predisposition to neoplastic transformation caused by gene re ment of H-rad. Science 1993; 260:1524-1527. of cell prolifera26. Dumont JE, Jauniaux JC, Roger PP. The cyclic AMP-mediated stimulation tion. Trends Biochem Sci 1989; 1467-71. 27. Russo D,Arturi F, Wicker R, Chazenbalk GD, Schlumberger M, DuViard JA, CaillouB, Monier R, Rapoport B, Filetti S, Suarez HG. Genetic alterations in thyroid hyperfunctio adenomas. J Clin Endocrinol Metab 1995; 80:1347-1351.
Pathogenesis Molecular
Cancer of Thyroid
71
28.BerridgeMJ,Irvine RF. Inositolphosophatesandcellsignalling.Nature1989;341: 197-205. 29. Harris CC, Hollstein M. Clinical Implications of the p53 tumor-suppressor gene. N Engl J Med 1993; 329:1318-1327. 30. Fields S, Jang SK. Presence of a potent transcription activating sequence in the p53 protein Science 1990; 249:1046-1049. 31. Vogelstein B, Kinzler KW. p53 function and dysfunction. Cell 1992; 705234526. in a cell cycle checkpoint may be responsible for the genomic instability 32. Hartwell L. Defects of cancer cells. Cell 1992; 71543-546. 33. Clarke A R , Purdie CA, Harrison DJ, Moms RG,BirdCC,Hooper ML, Wyllie AH. Thymocyteapoptosisinducedbyp53-dependentandindependentpathways.Nature 1993; 362~849-852. White A,Sprouse J, LivanosE,JacksT, Tlsty TD. Alteredcell 34.LivingstoneLR, cyclearrestandgeneamplificationpotentialaccompanyloss ofwild-typep53. Cell 1992; 70~923-935. 35. Srivastava S, Zou ZQ, Pirollo K, Blattner W, Chang EH. Germ-line transmission of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome. Nature 1990; 348~747-749. 36. Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in human cancers. Science 1991; 253:49-53. 37. Oliner JD, Kinzler KW, Meltzer PS,George DL, VogelsteinB. Amplification of a gene encoding a p53-associated protein in human sarcomas. Nature 1992; 358:80-83. AJ. The mdm-2 oncogene product 38. Momand J, Zambetti GP, Olson DC, George DL, Levine forms a complex with the p53 protein and inhibits p53-mediated transactivation. Cell 1992; 69:1237-1245. 39. Oliner J, Pietenpol J, Thiagalingam S, Gyuris J, Kinzler K, Vogelstein B. Oncoprotein MDM2concealstheactivationdomainoftumorsuppressorp53.Nature1993;362: 857-860. 40. Bar& Y, Juven T, Haffner R, Oren M. mdm2 expression is induced by wild type p53 activity. EMBO J 1993; 12:461-468. 41. Wu X, Bayle JH, Olson D, Levine AJ. The p53-mdm2 autoregulatory feedback loop. Genes Dev 1993; 7:1126-1132. 42. Namba H, Matsuo K, Fagin JA. Clonal composition of benign and malignant human thyroid tumors. J Clin Invest 1990; 86:120-125. 43. Thomas GA, Williams D, Williams ED. The clonal origin of thyroid nodules and ade Am J Path01 1989; 134:141-147. 44. Gerber H, Burgi U, Peter HJ. Etiology and pathogenesis of thyroid nodules. Exp Clin Endocrinol 1993; 101:97-101. 45. Matsuo K, Tang SH,Zeki K, Gutman R 4 , Fagin JA. Aberrant DNA methylation in human thyroid tumors, J Clin Endocrinol Metab 1993; 77:991-995. 46. Aasland R, Akslen LA, Varhaug JE, Lillehaug JR. Co-expression of the genes encoding transforming growth factor-alpha and its receptor in papillary carcinomas of the thyroid. Int J Cancer 1990; 46:382-387. 47. Williams DW, Williams ED, Wynford-Thomas D. Evidence for autocrine production of IGF-1 in human thyroid adenomas. Mol Cell Endocrinol 1989; 61:139-143. 48. Tode B, Serio M, Rotella CM, Galli G, Franceschelli F, Tanini A, Toccafondi R. Insulinlike growth factor I: autocrine secretion by human thyroid follicular cells in primary culture. J Clin Endocrinol Metab 1989; 69:639-647. 49. Aasland R, Lillehaug JR, Male R, Josendal 0, Varhaug JE, Kleppe K. Expression of oncogenes in thyroid tumors: coexpression of c-erbB2heu and c-erbB. Br J Cancer 1988; 57~358-363.
72
Figge
50. Heldin N E , Gustavsson B, Claesson-Welsh L, H m a c h e r A, MarkJ, Heldi CH, Westermark B. Aberrant expression of receptors for platelet-derivedgrowth factor inan anaplastic thyroid carcinoma cell line.Proc Natl Acad Sci USA 1988; 85:9302-9306. 51. Di Renzo MF, Narsimhan RP, Oliver0 M, Bretti S, Giordano S, Medico E, Gaglia P, Zara P, Comoglio PM. Expression of the met/HGF receptor in normal and neoplastic human tissues. Oncogene 199 ;16: 1997-2003. 52. FUSCO A, Grieco M, Santoro M, Berlingieri MT, Pilotti S, Pierotti MA, Della Porta G, Vecchio G. A new oncogene in human thyroid papillary carcinomas and their lymphnodal metastases. Nature 1987; 328:170-172. 53. Bongarzone I, Pierotti MA, Monzini N, Mondellini P, Manenti G, Donghi R, Pilotti S , Grieco M, Santoro M, Fusco A, Vecchio G , Della PortaG. High frequencyof activation of tyrosine kinase oncogenes in human papillary thyroid carcinoma. Oncogene 1989; 4:1457-1462. 54. Grieco M, Santoro M, Berlingieri MT, Melillo RM, Donghi R, Bonganone I, Pierotti MA, Della Porta G, FuscoA, Vecchio G. PTC is a novel rearranged form of the ret protooncogene and is frequently detected in vivo in human thyroid papillary carcinomas. Cell 1990; 60557-563. 55. Santoro M, Carlomagno F, Hay ID, Herrmann MA, Grieco M, Melillo R, pierotti MA, Bongarzone I, Della Porta G, Berger N, Peix JL, Paulin C, Fabien N, Vecchio G, Jenki RB, Fusco A. Ret oncogene activation in human thyroid neoplasms is restricted to the papillary cancer subtype.J Clin Invest 1992; 89:1517-1522. 56. It0 T, Seyama T, Iwamoto KS, Mizuno T, Tronko N D , Komissarenko IV, Cherstovoy ED, Satow Y, Takeichi N, Dohi K, Akiyama M. Activated RET oncogene in thyroid cancers of childrenfromareascontaminated by Chernobylaccident.Lancet1994; 344259. HM.High prevalenceof RETrearrange57. Klugbauer S, Lengfelder E, Demidchik EP, Rabes ment in thyroid tumors of children from Belarus after the Chemobyl reactor accident. Oncogene 1995; 112459-2467. 58. Fugazzola L, Pilotti S , Pinchera A, Vorontsova T V , Mondellini P, Bongarzone I, Greco A, Astakhova L, Butti MG, Demidchik EP, Pacini F, PierottiMA. Oncogenic rearrangements of the RET proto-oncogene in papillary thyroid carcinomas from children exposed to the Chernobyl nuclear accident. Cancer Res 1995; 555617-5620. 59. It0 T, Seyama T, Iwamoto KS, Hayashi T, Mizuno T, Tsuyama N, Dohi K, Nakamura N, Akiyama M. In vitro irradiationis able to causeRET oncogene rearrangement. Cancer Res 1993; 53:2940-2943. 60. Santoro M, Sabino N, Ishizaka Y, Ushijima T, Carlomagno F, Cerrato A, Grieco M, Battaglia C, Martelli ML, Paulin C, Fabien N, Sugimura T, Fusco A, Nagao M. Involv of RET oncogene in human tumours: specificity RET of activation to thyroid tumours. B J Cancer 1993; 68:460-464. Miozzo M, Fusco A, Grieco M, Santoro M, 61. Donghi R, Sozzi G, Pierotti MA, Biunno I, Vecchio G, Spurr NK, Della Porta G. The oncogene associated with human papillary thyroid carcinoma (PTC) is assignedto chromosome 10 q l l - q l 2 in the same regionas multiple endocrine neoplasia type 2a (MEN2A). Oncogene 1989; 4:521-523. 62. Zou M,Shi Y, Farid NR.Low rateof ret proto-oncogene activation (PTC/retm) in papillary thyroid carcinomas from Saudi Arabia. Cancer 1994; 73:176-180. 63. Jhiang SM, Caruso DR, Gilmore E, Ishizaka Y, Tahira T, NagaoM, Chiu IM, Mazzafem EL. Detection of the PTC/retTPC oncogenein human thyroid cancers. Oncogene 1992; 7~1331-1337. 63a. Jhiang SM, S a g e JE, Tong Q, Parker-ThomburgJ, Capen CC, ChoJY, Xing S, Ledent C. Targeted expression of the retlPTCI oncogene induces papillary thyroid carcinomas. Endocrinology 1996; 137:375-378.
Pathogenesis Molecular
Cancer of Thyroid
73
L, Vecchio G, Fusco A. Develop63b. Santoro MyGhiappetta G, Cerrato A, Salvatore D, Zhang of the RET/ ment of thyroid papillary carcinomas secondary to tissue-specific expression PTC1 oncogene in transgenic mice. Oncogene 1996; 121821-1826. 6 4 . Greco A, Pierotti MA, Bongarzone I, Pagliardini S, Lanzi CyDella Porta G. TRK-T1 is a novel oncogene formed by the fusionof TPR andTRK genes in human papillary thyroid carcinomas. Oncogene 1992; 7:237-242. 65. Russo D, Arturi F, SchlumbergerMyCaillou ByMonier R, Filetti S, Suarez HG. Activating mutations of the TSH receptor in differentiated thyroid carcinomas. Oncogene 1995; 11~1907-1911. C J , Hughes D, Padua Thurston v, williams 66. Lemoine N R , Mayall ES, Wyllie FS, Farr ED, Wynford-Thomas D. Activated ras oncogenesin human thyroid cancers. Cancer Res 1988; 48:4459-4463. CyMonier 67. Suarez HG, DuVillard JA, Caillou B, Schlumberger M, Tubiana M, Parmentier R. Detection of activated ras oncogenes in human thyroid carcinomas. Oncogene 1988; 2:403-406. My Stringer By Wynford68. Lemoine N R , Mayall ES, Wyllie FS, Williams ED, Goyns Thomas D. High frequency of rus oncogene activation in all stages of human thyroid tumorigenesis. Oncogene 1989; 4:159-164. JM, Parker MH, Seid JM, Hearn PRY Wynford-Thomas D, Ingemans69. Stringer BM, Rowson son S , Woodhouse N, Goyns “I. Detection of the H-rus oncogene in human thyroid anaplastic carcinomas. Experientia 1989; 45:372-376. 70, Wright PA, Lemoine N R , Mayall ES, Wyllie FS, Hughes D, Williams ED, WynfordThomas D. Papillary and follicular thyroid carcinomas show a different pattern of ras oncogene mutation. Br J Cancer 1989; 60576-577. 71. Dockhorn-DworniczakB, Caspari S, Schroder S, Bocker W, DworniczakB. Demonstration of activated oncogenes of the rus family in human thyroid tumors using the polymerase chain reaction. [In German].Verhandl Dtsch Ges Path01 1990; 74415418. 72. Namba H, Gutman Matsuo K, Alvarez A, Fagin JA. H-rus protooncogene mutations in human thyroid neoplasms. J Clin Endocrinol Metab 1990; 71:223-229. 73. Namba H, Rubin SA, Fagin JA. Point mutations of ras oncogenes are an early event in thyroid tumorigenesis. Mol Endocrinol 1990; 4: 1474-1479. RF, Westbrook CA, Straus FH, Kaplan EL.N-ras 61 oncogene 74. Schark C, Fulton N, Jacoby mutations in H W e cell tumors. Surgery 1990; 108:994-999. 75. SuarezHG, duVillard JAYSeverinoM, Caillou B, Schlumberger M, Tubiana M, Parmentier C, Monier R. Presence of mutations in all three ras genes in human thyroid tumors. Oncogene 1990; 5565-570. 76. Karga H, Lee JK, Vickery AL, Thor A, Gaz RD, Jameson JL. ras oncogene mutations in benign and malignant thyroid neoplasms. J Clin Endocrinol Metab 1991; 73:832-836. NR. Highrates of 77. Shi Y,ZouM, SchmidtH,JuhaszF,StenskyV,RobbD,Farid rus codon 61 mutation in thyroid tumors in an iodide-deficient area. Cancer Res 1991; 51~2690-2693. 78. Wright PA, Williams ED, Lemoine N R , Wynford-Thomas D. Radiation-associated and “spontaneous” human thyroid carcinomas show a different pattern of rus oncogene mutation. Oncogene 1991; 6:471473. 79. Goretzki PE, Lyons J, Stacy-PhippsS , Rosenau W, Demeure M, Clark OH, McCormick F, RoherHD, Bourne HR. Mutational activationof rus and gsp oncogenes in differentiated thyroid cancer and their biological implications. World J Surg 1992; 16576-581. 80. Yoshimoto K, Iwahana H, Fukuda A, Sano T, Katsuragi K, Kinoshita MySaito S, Itakura M. ras mutations in endocrine tumors: mutation detection by polymerase chain reactionsingle strand conformation polymorphism. JpnJ Cancer Res 1992; 83:1057-1062. N, Yashiro T, It0 K, DeGrootLJ, Kaplan EL. N-ras mutation: an independent 81. Hara H,.Fulton R
R
A
Y
A
Y
74
Figge
prognosticfactorforaggressiveness ofpapillarythyroidcarcinoma.Surgery1994; 116:lOlO-1016. 82. Kaihara M, Taniyama M, Tadatomo J, Tobe T, Tomita M, It0 K, Ban Y, Katagiri T. specific PCR amplification for N-ras mutations in neoplastic thyroid diseases. EndocrJ 1994; 41:301-308. 83. Manenti G, Pilotti S, Re FC, Della Porta G, Pierotti MA. Selective activation of ras Eur JCancer1994; oncogenesinfollicularandundifferentiatedthyroidcarcinomas. 30Az987-993. 84. Challeton C, BounacerDu A,Villard JA, Caillou B, De Vathaire F, Monier R, Schlumberger M, S U ~HG. Z Pattern of ras and gsp oncogene mutations in radiation-associated human thyroid tumors. Oncogene 1995; 11:601-603. 85. Hone H, Yokogoshi Y, Tsuyuguchi M, Saito S. Point mutations of ras and Gsa subunit genes in thyroid tumors. Jpn J Cancer Res 1995; 86737-742. 86. Oyama T, Suzuki T, Hara F,Iino Y, Ishida T, Sakamoto A, Nakajima T. N-ras mutatio of thyroid tumor with special reference to the follicular type. Path01 Int 1995; 4545-5 87. Suarez HG, du Villard JA, Caillou B, Schlumberger M, Parmentier C, Monier R. gsp mutations in human thyroid tumors. Oncogene 1991; 6:677-679. 88. Terrier P, Sheng ZM, Schlumberger M, Tubiana M, Caillou B, Travagli JP, Fragu P, Parmentier C, Riou G. Structure and expression of c-myc and c-fos proto-oncogenes in thyroid carcinomas. Br J Cancer 1988; 57:43-47. 89. It0 T, Seyama T, Mizuno T, TsuyamaN, Hayashi T, Hayashi Y, Dohi K, Nakamura N, Akiyama M. Unique association of p53 mutations with undifferentiated but not with differentiated carcinomasof the thyroid gland. Cancer Res 1992; 52:1369-1371. 90. It0 T, SeyamaT, Mizuno T, Tsuyama N, Hayashi Y, Dohi K, Nakamura N, Akiyama M. Genetic alterations in thyroid tumor progression: association with p53 gene mutations. Jpn J Cancer Res 1993; 84526-531. E, Karakawa K, Fujita S, Miya A, Mori T, 91. Nakamura T, Yana I, Kobayashi T, Shin Nishisho I, Takai S. p53 gene mutations associated with anaplastic transformation of human thyroid carcinomas. JpnJ Cancer Res 1992; 83:1293-1298. 92. Donghi R, Longoni A, Pilotti S, Michieli P, Della Porta G, Pierotti MA.Gene p53 mutation are restricted to poorly differentiated and undifferentiated carcinomas of the thyroid gland J Clin Invest 1993; 91:1753-1760. 93. Fagin JA, MatsuoK, Karmakar A, Chen DL, Tang SH, KoefflerHP. High prevalence of mutations of the p53 gene in poorly differentiated human thyroid carcinomas. J Cli 1993; 91~179-184. 94.Zou M, Shi Y, Farid NR. P53 mutations in all stages of thyroid carcinomas. J Clin Endocrinol Metab 1993; 77:1054-1058. 95. Dobashi Y, Sugimura H, Sakamoto A, Mernyei M, Mori M, Oyama T, Machinami R. Stepwise participation of p53 gene mutation during dedifferentiation of human thyroid carcinomas. Diagn Mol Pathol 1994; 3:9-14. 96. Gerasimov G, Bronstein M, Troshina K, Alexandrova G, Dedov I, Jennings T, Kallakury BVS, Izquierdo R, Boguniewicz A, Figge H, Robinson L, Breese K, Ross JS, Figge J. Nuclear p53 immunoreactivityin papillary thyroid cancersis associated with two established indicators of poor prognosis. Exp Mol Path01 1995; 6252-62. 97. Zou M, Shi Y, Al-Sedairy S, Hussain SS, Farid NR. The expressionof the "2gene, a p53 binding protein, in thyroid carcinogenesis. Cancer 1995; 76:314-318. 98. JenningsT, Bratslavsky G, Gerasimov G, TroshinaK, Bronstein M, DedovI, Alexandrova G, Figge J. Nuclear accumulation of MDM2 protein in well-differentiated papill carcinomas. Exp Mol Path01 1995; 62:199-206. 99. Gunthert U. CD44: a multitude of isoforms with diverse functions. Curr Top Microbiol Immunol 1993; 184:47-63.
Pathogenesis Molecular
of Thyroid Cancer
75
100. Lesley J, Hyman R, Kincade PW. CD44 andits interaction with extracellular matrix. Adv Immunol 1993; 54:271-335. 101. Aruffo A, StamenkovicI, Melnick M, Underhill CB, Seed B. CD44 is the principal cell surface receptor for hyaluronate. Cell1990; 61:1303-1313. 102. Hofmann M, Rudy W, Zoller M, Tolg C, Ponta H, Herrlich P, Gunthert U. CD44 splice variants confer metastatic behavior in rats: homologous sequences are expressed in human tumor lines. Cancer Res 1991; 515292-5297. 103. Rudy W, Hofmann M, Schwartz-Albiez R, Zoller M, Heider KH, Ponta H, Herrlich P. are derived from Two major CD44 proteins expressedon a metastatic rat tumor cell line different splice variants: each one individually suffices to confer metastatic behavior. Cancer Res 1993; 53:1262-1268. 104. Fox SB, Fawcett J, Jackson DG, Collins I, Gatter KC, Harris AL, Gearing A, Simmons DL. Normal human tissues, in addition to some tumors, express multiple different CD44 isoforms. Cancer Res 1994; 54:4539-4546. 105. Mackay CR, Terpe HJ, Stauder R, Marston WL, Stark H, Gunthert U. Expression and modulation of CD44 variant isoforms in humans. J Cell Biol 1994; 124:71-82. 106. Figge J, del RosarioAD, Gerasimov G, Dedov I, Bronstein M, TroshinaK, Alexandrova G, Kallakury BVS, Bui HX, Bratslavslq G, Ross JS. Preferential expressionof the cell adhesionmoleculeCD44inpapillarythyroidcarcinoma.ExpMolPath01 1994; 61: 203-211. 107. Ermak G, Gerasimov G, Troshina K, Jennings T, Robinson L, Ross JS, Figge J. Dereg alternative splicingof CD44 messenger RNA transcriptsin neoplastic and nonneoplastic lesions of the human thyroid. Cancer Res 1995; 55:4594-4598. 108. Ermak G, Jennings T, Robinson L, Ross JS, Figge J. Restricted patterns of CD44 variant 56:1037-1042. exon expression in human papillary thyroid carcinoma. Cancer 1996; Res 109. Chen W, Obrink B. Cell-cell contacts mediated by E-cadherin (uvomorulin) restrict invasive behavior of L-cells. J Cell Biol 1991; 114:319-327. 110. Vleminckx K, Vakaet L, Mareel M, Fiers W, Van Roy F. Genetic manipulation of Ecadherin expression by epithelial tumor cells reveals an invasion suppressor role. Cell 1991; 66:107-119. 11 1. Brabant G, Hoang-Vu C, Cetin Y,Dralle H, Scheumann G, MolneJ, Hansson G, Jansson S, Ericson LE, NilssonM. E-cadherin: a differentiation marker in thyroid malignancies. Cancer Res 1993; 53:4987-4993. 112. Matsuo K, Tang S-H, Fagin JA. Allelotype of human thyroid tumors:ofloss chromosome llq13 sequences in follicular neoplasms. Mol Endocrinol1991; 5:1873-1879. 113. Herrmann MA, Hay ID, Bartelt DH, Ritland SR, Dahl RJ, Grant CS, JenkinsRB. Cytogenetic and molecular genetic studies of follicular and papillary thyroid cancers. J Clin Invest 1991; 8811596-1604. 114. Plail RO, Bussey HJ, Glazer G, Thompson JP. Adenomatous polyposis: an association with carcinoma of the thyroid. Br J Surg 1987; 74:377-380. 115. Lote K,Andersen K, Nordal E, Brennhovd IO. Familial occurrence of papillary thyroid carcinoma. Cancer 1980; 46:1291-1297. 116. Sogol PB, Sugawara M, Gordon HE, ShellowWV, Hernandez F, Hershman JW.Cowden’s disease:familialgoiterandskin hamartomas-a report of threecases.WestJMed 1983; 139~324-328. 117. Camiel M R , Mule JE, Alexander LL, Beninghoff DL. Association of thyroid carcioma with Gardner’s syndrome in siblings. N Engl J Med 1968; 278:1056-1058. 118. Mulligan LM, Ponder BAJ. Genetic basis of endocrine disease: multiple endocrine neoplasia type 2.J Clin Endocrinol Metab 1995; 80:1989-1995.
7 Epidemiology of Thyroid Cancer James Figge THYROID CANCER INCIDENCE Thyroid cancer is the most common of the endocrine malignancies, accountingfor
1.1%of all new malignant tumors (excluding skin cancer) diagnosed annually in the United States (0.5% of cancers in men, 2.0% in women) (I). Annual incidence rates
vary by geographic area, age, and sex. The age-adjusted annual incidence in the United States is 55 new cases per million (2,3), with a higher incidence in women (80 per million) than men (29 per million) (2,4). Approximately 15,600 new cases of thyroid cancer are now diagnosed annually in the United States, with a female : male ratio of close to 3:l ( I ) . Worldwide, incidence rates are highest in certain geographic areas such as Hawaii(104 per million women and 39 per million men), probably as a result (2,5). Rates in Poland are among the lowest recorded: of local environmental influences 14 per million women, and 4 per million men (6). Thyroid cancer is very rare in children under age 15. The annual incidence inthis population in the United Statesis 2.2 per million girls and 0.9 per million boys (7). The annual incidence of thyroid cancer increases with age, peaking at per million by the 90-100 fifth to eighth decade (2). The incidence of thyroid cancer has increased over a period of several decades in as well as several other countries, particularly among women (2,3,4,8the United States 20). For example, in Connecticut, the annual age-standardized incidence in women has progressively increased from13 per million in 1935-1939, to 36 per million in 19651969, to 45 per million in 1985-1989, reaching 58 per million in 1990-1991. The corresponding figures for men are 2 per million, 18 per million, 21 per million, and 26 per million, respectively (4). The precise reasons for the increase are not clearly understood, butmay be related, at least in part, to the introduction of improved diagno methodology(e.g.,ultrasound,thyroidscans,fineneedleaspirationbiopsy),and improvements in cancer registration($20). In the United States, the increased incidence between 1935and 1975might also be a consequence of therapeutic radiation treatments that had been administered to the head and neck region of children (9,2I) (see Chapter 8). However, increases in thyroid cancer incidence were documented in other countries wherechildhoodradiationtreatmentswerenevercommonlyemployed (I3,15,19); therefore, other factors must also be involved. Exposure to fallout from nuclear weap testing has been suggested as a factor in Europe, but epidemiological data suggest that
From: Thyroid Cancer: A Comprehensive Guide to clinical Management Edited by: L. Wmtofiky 0 Humana Press Inc., Totma, NJ
77
78
Figge
there are still more important factors (14). The incidence of thyroid canceris no longe increasing in certain countries such as Norway and Iceland (15,16,17), but continues to rise in the United States (2).
THYROID CANCER PREVALENCE
Thyroid cancer prevalence rates vary widely by geographic area, patient populatio and method of survey. Autopsy rates ranging from 0.03% to over 2% have be (22-26). Mortensen and colleagues(22) reported on 1000 consecutive routine autop and found a 2.8% prevalence rate of thyroid carcinoma. The high cancer prevalence (22). On routine might be attributed to the meticulous histological evaluation protocol clinical assessment, 61% (17/28) of the cancers originated from thyroid glands that were apparently normal (23). Similar prevalence rates (2.3% to 2.7%) were reported by Bisi and colleagues(24) and Silverberg and Vidone(25). The high prevalence rate reported in the latter two studies might also have been influenced by the high inpatientpopulationsthatwerestudiedand maynotreflect theprevalenceinthe general population. Small foci of papillary thyroid carcinoma, measuring 1 cm in diameter or less, can (27) andoccurfrequentlyinautopsy beclassifiedas“papillarymicrocarcinomas” 28). Most of these papillary microcarcinomas measure b material (reviewed in ref. 4 and 7 mm (29). These can be subdivided into “tiny” (5-10 mm in diameter) and “minute” carcinomas (<5 mm in diameter) (27,30-33). The term “occult” carcinoma has no pathological meaning and should be abandoned in favor of these more precisel (27). Papillary microcarcinomas are usually defined terms as advocated by LiVolsi detected by meticulous sectioning of the thyroid at 2- to 3-mm intervals, with detai microscopic examination of each section. The highest prevalence rate of papillary thyroid microcarcinoma (diameter 21 cm) was reported from Finland(34), with 33.7% of 101 cases harboring this finding. Rates over 20% have been reported from Japan (35,36), whereas the rate of papillary microcarcinoma in Olmsted County, Minnesota is much lower, on the order of 5.1% (37). Minute papillary carcinomas ( 4 mm) are often not detected clinically and are believed to exhibit a relatively benign clinical course. There are, however, occasional reports of distant metastases (e.g., pulmonary metastases) arising from minute papillary carcinomas (38). Thyroidcancerprevalenceratesinthepopulationaresignificantlygreaterthan incidence figures, reflecting the fact that substantial numbers of patients survive sever decades or longer. Data in the Connecticut registry show a prevalence of rate677 cases per million in women and 237 cases per million in men(39). These data refer only clinically apparent disease and therefore are lower than the rates given in many of th autopsy series (22-25).
THYROID CANCER MORTALITY
The annual mortality from thyroid cancer is low, on the order of 3 to 4 deaths pe million individuals per year (2), presumably reflecting the good prognosis for most thyroid cancers. Mortality rates are lowest in individuals under age 50, and increase sharply thereafter (2). About 1200 deaths from thyroid cancer occur annually in the
Epidemiology of Thyroid Cancer
79
United States ( I ) , accounting for 0.2% of all cancer deaths (0.15% of cancer deaths in men, 0.3% in women). While the incidence of thyroid cancer has been increasing over time in both men and women as noted above, mortality has decreased (2). The reduced mortalityis due to earlier diagnosis, improved treatment and a decreasein the incidence of anaplastic carcinoma. For example, relative 5-year survival rates for thyroid cancer have increased in Caucasians from 83% in 1960-1963 to 95% in 1986-1991 (I). DISTRIBUTION OF THYROID CANCER BY HISTOLOGICAL TYPE
The relative proportion of differentiated (follicular and papillary) thyroid cancers in a given geographic area depends upon the dietary iodine intake. Papillary cancers predominate in iodine-sufficient areas. For example, in Iceland, which has ample iodine intake, the proportions were 85% papillary and 15% follicular cancer from 1955 throu 1984 (I7),whereas in Bavaria, Germany, an iodine-deficient area, the proportions were 35% papillary and 65% follicular during 1960 to 1975 (40). The introduction of iodine supplementation in an endemic goiter region results in an increase in the proportion of papillary cancers (41). In the United States, approximately 80% of thyroid cancers are papillary carcinomas (42). Papillary cancer has a peak incidence in the third and fourth decades of life(43) and affects women two to three times more frequently than men. Follicular carcinoma accounts for approximately 5 to 10% of cases in the United States (42) and has a peak fifth decade. The tumor is three times more common in women than men. incidence in the Medullary carcinomas comprise about 5% to 10% of thyroid carcinomas (44). Of (44). The sporadic these, 80% are sporadic and20% are familial, mostly MEN 11-related form presents most commonly in the fifth and sixth decades of life, and affects females 1.5 times morel than males (45). MEN IIa-related medullary carcinomas present in the first and second decades, and MEN IIb-associated medullary cancers present during the first decade of life (44). Familial non-MEN medullary thyroid carcinomas present in the sixth decade and beyond (44). Familial forms of medullary carcinoma occur with equal frequency in females and males. Anaplastic cancers and lymphomas account for the remainder of cases. The incide of anaplastic cancer has recently declined, a factor that has contributed to the decrease (42). The peak incidence of anaplastic in overall thyroid cancer mortality as noted above cancer is in the seventh decade; the female : male ratio is 1.51. Lymphomas account for about 5% of thyroid malignancies, with a mean age of 60 to 65 at the time of presentation (46,47).Females predominate at all ages; in patients under 60 the ratiois 1.5:1, in patients over 60 the ratio ranges from 3 to 8: 1 (46,47). FACTORS ASSOCIATED WTTH THYROID CANCER RISK There are several strong associations between thyroid cancer incidence and certain risk factors: 1. Thyroid cancer incidence increases with age. 2. Thyroid canceris more common in females than males. The female predominance suggests be involved.Somestudiessuggestthatbiologicalchanges thathormonalfactorsmay . occurringduringpregnancymayincreasethe risk ofthyroidcarcinoma (48-50).
80
Figge
3. Several genetic syndromes such as Gardner syndrome and Cowden’s disease are with an increased risk of thyroid cancer and are discussed in Chapter 5. 4. Radiation exposure is the only factor that has been shown unequivocally to cause thyro
cancer and is discussed in detail in Chapter8. 5. There is strong evidence that individuals with Hashimoto’s thyroiditis are at increa for the developmentof thyroid lymphoma (51).
In addition to the aforementioned well-established associations, there are a numbe of postulated risk factors for thyroid carcinoma that remain unproven. These include iodine deficiency and endemic goiter (52), which may result in prolonged stimulation of thyroid tissue by elevated thyroid-stimulating hormone (TSH) levels. Data on this postulated association are inconsistent (50,5242). A major study comparing goiter prevalence and the effect of iodine supplementation with thyroid cancer rates in the United States failed to support the existence of a relationship between endemic goiter and thyroid cancer(62). Graves’ disease has also been postulated to be associated w an increased incidence of thyroid cancer. This hypothesis is of interest because of the TSH-like activity of thyroid-stimulating immunoglobulins (TSI). However, the data remain inconclusive (63-76), with reported cancer rates ranging from 0.06% (66) to as high as 8.7% (68) in glands affected by Graves’ disease. Lower rates were report in older studies (63-66), and several recent studies (70-72) have reported rates in the range of 5.1% to 7.0%. The possibility that other benign diseasesof the thyroid could increase the risk of cancer has also been entertained (50,51,53,57,77-81). These data are difficult to interpret, given the strong possibility of ascertainment bias. Furthermor it is well-established that pathological examinations of thyroid tissue can reveal a h rate of unsuspected microcarcinomas that may be of little clinical significance. Thus, it remains uncertain as to whether patients with preexisting thyroid disease are at increased risk of developing clinically significant thyroid carcinoma.
REFERENCES
1. Parker SL, Tong T, Bolden S , Wingo PA. Cancer statistics, 1996. CA 1996; 46:5-27. 2. Ries LAG, Kosary CL, Hankey BF, Miller BA, Clegg L, Edwards BK (eds). SEER C Statistics Review, 1973-1996, National Cancer Institute. Bethesda,MD, 1999. 3. Verby JE, Woolner LB, Nobrega FT, Kurland LT, McConahey WM. Thyroid cancer in Olmsted County, 1935-1965. J Natl Cancer Inst 1969; 43:813-820. 4. Polednak AP. TrendsincancerincidenceinConnecticut, 1935-1991. Cancer 1994; 74:2863-2872. 5. Goodman MT, Yoshizawa CN, Kolonel LN. Descriptive epidemiology of thyroid cancer in Hawaii. Cancer 1988; 61:1272-1281. 6. Whelan SL, Parkin DM, Masuyer E. Patterns of cancer in five continents. IARC Sci Pub 1990;102~1-159. 7. Parkin DM, Stiller CA, Draper et GJ,al. International incidence of childhood cancer. IAR Sci Publ 1988; 87:1-401. 8. Weiss W. Changing incidence of thyroid cancer. J Natl Cancer Inst 1979; 62:1137-1142. Day NE, Pickle LW, Fraumeni JF Jr. Thyroid cancerin Connecticut 9. Pottern LM, Stone BJ, 1935-1975: an analysis by cell type. Am J Epidemiol 1980; 112:764-774. 10. Carroll RE, Haddon W Jr, Handy V H , Wieben EE. Thyroid cancer: cohort analysis of increasing incidencein New York State, 1941-1962. J Natl Cancer Inst1964; 33:277-283. 11. Waterhouse J, Muir C, Correa P, Powell J. Cancer incidence in five continents, vol. 3. Lyon, France: International Agency for Research on Cancer, 1976.
Epidemiology of Thyroid Cancer
81
12. Waterhouse J, MuirCyShanugaratnamK. Cancer incidence in five continents, vol. 4. Lyon, France: International Agency for Research on Cancer, 1982. 13. Pettersson B, Adami H-0, Wilander E, Coleman MP. Trends in thyroid cancer incidence in Sweden, 1958-1981, by histopathologic type. Int J Cancer 1991; 48:28-33. 14. dos Santos Silva I, Swerdlow AJ. Thyroid cancer epidemiology in England and Wales: time trends and geographical distribution. Br J Cancer 1993; 67:330-340. 15. Akslen LA, Haldorsen T, Thoresen SO, Glattre E. Incidence pattern of thyroid cancer in Norway: influence of birth cohort and time period. Int J Cancer 1993; 53:183-187. 16. Glattre E, Akslen LA, Thoresen S, Haldoren T. Geographic patterns and trends in the incidence of thyroid cancer in Norway1970-1986. Cancer Detect Prev1990; 14625-631. 17. Hrafnkelsson J, Jonasson JG, Sigurdsson G, Sigvaldason H, Tulinius H. Thyroid cancer in Iceland 1955-1984. Acta Endocrinol 1988; 118566-572. 18. Staunton MD, Bourne H. Thyroid cancer in the 1980's: a decade of change. Ann Acad Med (Singapore) 1993; 22:613-616. 19. Levi F, Franceschi S , Te VC, Negri E, La Vecchia C. Descriptive epidemiology of thyroid cancer in the Swiss canton of Vaud. J Cancer Res Clin Oncol 1990; 116:639-647. Schymura MJ, White C. Cancer risk and incidence trends: the 20. Roush G C , Holford Connecticut perspective.New York Hemisphere Publishing,1987. 21. Same D, Schneider AB. External radiation and thyroid neoplasia. Endocrinol Metab Clin North Am 1996; 25~181-195. 22. Mortensen J D , Bennett WA, Woolner LB. Incidenceof carcinoma in thyroid glands removed at 1000 consecutive routine necropsies. Surg Forum1954; 5659-663. 23. Mortensen JD, Woolner LB, Bennett WA. Gross and microscopic findings in clinically normal thyroid glands. J Clin Endocrinol Metab 1955; 15:1270-1280. 24. Bisi H, Fernandes VS, de Camargo RY, Koch L, Abdo AH, de Brito T. The prevalence of unsuspected thyroid pathology in300 sequential autopsies, with special reference to the incidental carcinoma. Cancer 1989; 64:1888-1893. 25. Silverberg SG, VidoneR4. Carcinoma of the thyroid in surgical and postmortem material: analysis of 300 cases at autopsy and literature review.Ann Surg 1966; 1W291-299. 26. VanderLaan WP. The occurrence of carcinoma of the thyroid gland in autopsy material. N Engl J Med 1947; 237:221-222. '27. LiVolsi, VA. Papillary neoplasms of the thyroid. Am J Clin Path01 1992; 97:426-434. 28. Ain KB. Papillary thyroid carcinoma. Endocrinol Metab Clin North Am 1995; 24:711-760. 29. Vickery AL Jr, CarcangiuML, Johannessen J V , Sobrinho-Simoes M. Papillary carcinoma. Semin Diagn Path01 1985; 2:90-100. 30. KasaiN,SakamotoA.Newsubgrouping of smallthyroidcarcinomas.Cancer 1987; 60~1767-1770. 31. Naruse T, Koike A, Kanemitsu T, Kat0 K. Minimal thyroid carcinoma: a report of nine cases discovered by cervical lymph node metastases. Jpn J Surg1984; 14:118-121. S , Akiyama T, Miyazaki I, Michigishi T, Tonami N, Hisada K, Terahata 32. Noguchi My Tanaka S , Matsubara F. Clinicopathological studies of minimal thyroid and ordinary thyroid cance Jpn J Surg 1984; 14110-117. 33. Yamashita H, NakayamaI, Noguchi S, Murakami N, Moriuchi A, Yokoyama S, Mochifllki Y, Noguchi A. Thyroid carcinoma in benign thyroid diseases: an analysis from minute carcinoma. Acta Path01 Jpn 1985; 35:781-788. 34. Harach HR, Franssila KO, Wasenius V-M. Occult papillary carcinoma of the thyroid-a "normal" finding in Finland: a systematic autopsy study. Cancer 1985; 56531-538. 35. Fukunaga FH, Yatani R. Geographic pathology of occult thyroid carcinomas. Cancer1975; 36: 1095-1099. 36. Sampson RJ. Prevalence and significance of occult thyroid cancer. In DeGroot LJ, editor. Radiation-associated thyroid carcinoma.New York Grune 8z Stratton, 1997:137-153. 37. Sampson RJ, Woolner LB, Bahn RC, Kurland LT. Occult thyroid carcinoma in Olmsted T
R
Y
82
Figg
county, Minnesota: prevalence at autopsy compared with that in Hiroshima and Naga Japan. Cancer 1974; 342072-2076. 38. Strate SM, Lee EL, Childers JH. Occult papillary carcinoma of the thyroid with distant metastases. Cancer 1984; 54:1093-1100. 39. Feldman A R , Kessler L, Myers MH, Naughton MD. The prevalence of cancer: estimates based on the Connecticut tumor registry. N Engl JMed 1986; 3151394-1397. 40. Lohrs U, Permanetter W, Spelsberg F, Beitinger M. Investigationof frequency and spread of the different histological types of thyroid cancer in an endemic goiter region. [In Ger Verhandl Dtsch Ges Path01 1977; 61:268-274. J, Saravia Day E, Williams ED. 41. Harach HR, Escalante DA, Onativa A, Lederer Outes Thyroid carcinoma and thyroiditis in an endemic goitre region before and after iodine prophylaxis. Acta Endocrinol 1985; 108:55-60. In Becker KL, (Ed). Principles and practice of endocrinolog 42. Mazzaferri EL. Thyroid cancer. and metabolism. 2nd ed. Philadelphia: Lippincott, 1995354366. 43. McDermott WV Jr, Morgan WS, Hamlin E Jr, Cope 0. Cancer of the thyroid. J Clin Endocrinol Metab 1954; 14:1336. 44. Ledger GA, Khosla S , Lindor NM, Thibodeau SN, Gharib H. Genetic testing in the di II. Ann Intern Med 1995;122: and management of multiple endocrine neoplasia type 118-124. 45. Emmertsen K. Medullary thyroid carcinoma and calcitonin. Dan Med Bull 1985; 32:l-28. 46. Mazzaferri EL, Ortel YC. Primary malignant lymphoma and related lymphoproliferative MA disorders. In Mazzaferri EL, Samaan NA, editors. Endocrine tumors. Cambridge, Blackwell Scientific, 1993:348. 47. AnscombeA M , Wright DH. Primary malignant lymphoma of the thyroid-a tumor associated lymphoid tissue: reviewof seventy six cases. Histopathology 1985; 9:81-97. 48. Kravdal 0,Glattre E, Haldorsen T. Positive correlation between parity and incidence of thyroid cancer: new evidence based on complete Norwegian birth cohorts. Int J Cancer 1991; 49931-836. 49. Glattre E, Kravdal 0. Male and female parity and risk of thyroid cancer. Int J Cancer 1994; 58:616-617. 50. Ron E, KleinermanRA, Boice JD Jr, LiVolsi VA, Flannery JT, Fraumeni JF Jr. A populatio based case-control studyof thyroid cancer. J Natl Cancer Inst1987; 79:l-12. 51. Holm LE, Blomgren H, Lowhagen T. Cancer risks in patients with chronic lymphocytic thyroiditis. N Engl J Med 1985; 312:601-604. 52. Wegelin C, Malignant disease of the thyroid gland and its relation to goiter in men and animals. Cancer Rev 1928; 3:297. 53. Franceschi S , Fassina A, Talamini R, Mazzolini A, Vianello S, Bidoli E, Serraino D 1989 La Vecchia C. Risk factors for thyroid cancer in northern Italy. Int J Epidemiol 18578-584. 54. Franceschi S, Talamini R, Fassina A, Bidoli E. Diet and epithelial cancer of the thyroid gland. Tumori 1990; 76:331-338. 55. Kolonel LN, Hankin JH, Wilkens LR, FukunagaFH, Hinds MW. An epidemiologic stud of thyroid cancer in Hawaii. Cancer Causes Control 1990; 1:223-234. 56. Glattre E, Haldorsen T, Berg JP, Stensvold I, Solvoll K. Norwegian case-control study testing the hypothesis that seafood increases the risk of thyroid cancer. Cancer Causes Control 1993; 4:ll-16. 57. Preston-Martin S, Jin F, Duda MJ, Mack WJ. A case-control study of thyroid cancer in women under age 55 in Shanghai (People’s Republic of China). Cancer Causes Contro 1993; 4 : 4 3 1 4 0 . 58. Hallquist A, Hardell L, Degerman A, Boquist L. Thyroid cancer: reproductive factors, previous diseases, drug intake,family history and diet: a case-control study. Eur J Cance Prev 1994; 3:481-488.
Epidemiology of Thyroid Cancer
83
59. Franceschi S, Levi F, Negri E, Fassina A, LaVecchia C. Diet and thyroid cancer: a pooled analysis of four European case-control studies. Int J Cancer 1991; 48:395-398.
60. Correa P, LlanosG. Morbidity and mortality from cancer in Cali, Columbia. J Natl Cancer Inst 1966; 36~717-745. 61. Franssila K, Saxen E, Teppo L, Bjarnason 0, Tulinius H, NormanT, Ringertz N. Incidence of different morphological types of thyroid cancer in the Nordic countries. Acta Path01 Microbiol Scand A 1981; 89:49-55. BK, Marcus SC. Thyroid cancer and thyrotoxicosis in the United 62. Pendergrast WJ, Milmore States: their relation to endemic goiter.J Chronic Dis 1961; 13:22-38. 63. Beahrs OH, PembertonJJ, Black BM. Nodular goiter and malignant lesions of the thyroid gland. J Clin Endocrinol 1951; 11:1157-1165. 6 4 . Pemberton J, Black BM. The association of carcinoma of the thyroid gland and exophthalmic goiter. Surg Clin North Am 1948; 28:935-952. 65. Olen E,Klinck GH. Hyperthyroidism and thyroid cancer. Arch Path01 1966; 81531-535. 66. Sokal JE. Incidence of malignancy in toxic and non-toxic nodular goiter. J A M A 1954;
154:1321-1325. 67. Carnell NE, Valente WA. Thyroid nodules in Graves’ disease: classification, characterization, and response to treatment. Thyroid1998; 8:647-652. 68. Shapiro SJ, Friedman N B , Perzik SI, Catz B. Incidence of thyroid carcinoma in Graves’ disease. Cancer 1970; 26:1261-1270. 69. Wahl R A , Goretzki P, Meybier H, Nitschke J, Linder M, Roher H-D. Coexistence of hyperthyroidism and thyroid cancer. WorldJ Surgery 1982; 6:385-390. A M , Paloyan E. Thyroid carcinoma in Graves’ disease. 70. Farbota LM, Calandra DB, Lawrence Surgery 1985;98:1149-1153. M, WuT-C, McCormick M, Straus FH, DeGroot LJ, Kaplan EL. Graves’ 71. Behar R, Arganini disease and thyroid cancer. Surgery 1986; 100:1121-1127. 72. Pacini F, Elisei R, Di Coscio GC, Anelli S , Macchia E, Concetti R, Miccoli P, Arganini M, Pinchera A. Thyroid carcinoma in thyrotoxic patients treated by surgery.J Endocrinol Invest 1988; 11:107-112. 73. Ozaki 0,Ito K, Kobayashi K, Toshima K, Iwasaki H, Yashiro T. Thyroid carcinoma in Graves’ disease. World J Surg 1990; 1443740. S, Fiumara A, Ippolito0, Vigneri 74. Belfiore A, GamfaloM R , Giuffiida D, Runello F, Filetti
J Clin R. Increased aggressiveness of thyroidcancerinpatientswithGraves’disease. Endocrinol Metab 1990; 70:830-835. 75. Hales lB, McElduff A, Crummer P, Clifton-Bligh P, Delbridge L, Hoschl R, Poole A, Reeve TS, Wilmshurst E, Wiseman J. Does Graves’ disease or thyrotoxicosis affect the prognosis of thyroid cancer. J Clin Endocrinol Metab1992; 75886-889. 76. Cady B. Papillary carcinoma of the thyroid. Semin Surg Oncol 1991; 7:81-86. 77. Goldman MB, Monson RR, Maloof F. Cancer mortality in women with thyroid disease. Cancer Res 1990; 50:2283-2289. 78. Levi F, Franceschi S , La Vecchia C, Negri E, Gulie C, Duruz G, Scazziga B. Previous thyroid disease and risk of thyroid cancer in Switzerland. EurJ Cancer 1991; 27:85-88. 79. Preston-Martin S , Bernstein L, Pike MC, Maldonado AA, Henderson BE. Thyroid cancer among young women related to prior thyroid disease and pregnancy history. BrJ Cancer 1987; 55~191-195. 80. Wingren G, Hatschek T, Axelson0. Determinants of papillary cancer of the thyroid.Am J Epidemiol 1993; 138:482491. 81. McTiernan A M , Weiss NS, Daling JR. Incidence of thyroid cancer in women in relation
to previous exposure to radiation therapy and history of thyroid disease. J Natl Cancer Inst 1984; 73575-581.
8 Radiation and Thyroid Cancer James Figge, Timothy Jennings, and Gregory Gerasimov Radiation is one of the few accepted risk factors for thyroid cancer. Numerous studies have confirmed that the thyroid gland is one of the most radiation-sensitive human organs and that thyroid cancer is one of the most common radiogenic malignancies. Analysis of these studies is problematic, however, due to difficulties in dose assessment, long-term follow-up of thousands of exposed subjects, definition and confirmation of pathological diagnoses, and differences in exposure modalities. The first part of this chapter briefly outlines the nature and methods of the most significant studies to date, and analyzes the data available to define the characteristics of the risks of radiation to the thyroid on subsequent development of thyroid cancer. in children exposed The second partof the chapter presents an update on thyroid cancer to fallout from the Chernobyl accident. Terminology used throughout the chapter is defined in Table 1.
PATHOLOGY Knowledge of the pathology of radiation injury to the thyroidis essential to understanding the data from previous long-term follow-up studies. Thyroid glands exposed to external beam or 13'1 radiation show a variety of histological abnormalities, most often multinodularity, distorting fibrosis, oncocytic change, and chronic inflammation (1-3). At higher (>1.5 Gy) doses, hyperplastic nodules may show cytologic atypia, which requires careful scrutiny to distinguish from malignancy (4). The incidence of benign adenomasin patients having received thyroid irradiation is also greatly increased over nonirradiated individuals, as demonstrated by virtually every study of such populations. Many studies failed to distinguish between benign nodules and carcinoma, and are therefore excluded from this discussion. As early as 1949, Quimby and Werner (5) suggested the possibility of a relationship between radiation and the subsequent development of thyroid carcinoma. Winship and 1948, and their Rosvoll (6) began collecting data on children with thyroid cancer in final report on878 cases worldwide represents the largest to date. They found a history of radiation in 76% of 476 children with available records. Most received radiation for enlarged thymus or tonsils and adenoids, with an average thyroid dose of 0.512 Gy and an average interval to diagnosis of 8.5 years; 72% of the cancers were of
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartojky 0 Humana Press Inc., T o t m , NJ
85
Dejinition,
Figge, Jennings, and Gerasimov
86 Table 1 Definition of Terminology Term
Sievert (Sv)
Becquerel (Bq) RR ERR
EAR
Factors
The gray is the unitof absorbed dose, the amount of energy imparted by ionizing radiation to a unit massof tissue; 1 gray corresponds to 1 joule per kilogram. 1 Gy = 100 rad The sievert is the unitof effective dose. When exposure is to mixed radiation (e.g., alpha and gamma), their contribution is weighted to give an equivalent dose.A further weighting is made to account for different susceptibilitiesof various tissues. 1 Sv = 100 rem The becquerel is the unit of activity, the numberof radioactive transformations taking place per second. 1 curie = 3.7 x 1O’O Bq Relative risk. The risk of developing cancerin a radiation-exposed subject compared to the risk in an unexposed individual. RR = ERR + 1. Excess relative risk is usually specified per Gy. For example, if the ERR is 2.0 per Gy, then the RR would be 3.0 for a 1 Gy exposure, and 5.0 for a 2 Gy exposure. Excess absolute risk, usually expressed per 10,OOO person-years per Gy. Defines the increase in the absoluterisk of developing canceras a result of radiation exposure.
papillary type, and 18% were follicular. Cervical lymph node metastases were prese in74%ofcases,withbilateralneckdiseasein32%.Nearly20%hadpulmonary involvement, generally at presentation. The authors noted a sharpin rise thyroid cancer incidence as of 1945, with the greatest number of cases between 1946 to 1959; they attributed the subsequent decline to the curtailment of the practice of head and neck irradiation in children. A number of additional studies (7,s) have confirmed that the majority of radiationinduced thyroid carcinomas are well-differentiated papillary adenocarcinomas, which more frequently present with extrathyroidal spread and bilateral thyroid lobe involve ment, but with similar recurrence and mortality rates to tumors in nonirradiate The patients are also younger at diagnosis, usually less than 35 years of age, with an average interval to clinical presentation of 25-30 years. The incidence of radiationinduced thyroid carcinoma appeared to increase from 1940 to at least 1970, but since this trendhas discontinuationofwidespreaduse ofx-raytherapyininfancy, decreased (9,10). Clinically occult papillary microcarcinomas are generally not included in analysis ofthesedata,although they are often detected by pathologists examining thyroids removed for larger, benign nodules. Autopsy studies have demonstrated prevalence rates of papillary microcarcinoma (diameter 1 cm or less) of up to 33.7% in general populations, and ethnic andor geographic differences exist (11,12) (see Chapter 7). The prevalence of carcinoma is also dependent on the extent of surgery, the amount of resected thyroid tissue processed for histological assessment, and the absolute n
Radiation
87
of sections examined by the pathologist (13). Care is required in evaluating studies with regard to these issues. Although radiation exposure plays a role in the development of clinical papillary carcinoma, the extentof such risk in a given population cannot always be ascertained, since the number of persons at risk may be unknown. Currently it is estimated that 9% of thyroid cancers may be attributable to radiation (14). Because radiation-induced thyroid carcinomas rarely include the more aggressive anaplastic and medullary types, the fatality rate of radiation-induced thyroid carcinoma is between 3% and 9% (15). A small but significant number of patients with anaplastic thyroid carcinoma have had a history of prior exposure to external irradiation or to 13'I. Suchtherapyfor differentiated thyroid cancer might theoretically induce transformation to an anaplastic carcinoma, but since most cases of anaplastic carcinoma show areasof differentiated tumor, this phenomenon may be an aspect of the natural history of these tumors, and therefore may not be a consequence of radiation (16).
PRIOR STUDIES External Radiation Introduction From 1920to 1960, gamma radiation was commonly used to treat a variety of benign conditions, including a number of head, neck, and upper thoracic sites, which resulted In 1950, D u m andFitzgerald (17) foundthat 9 of 28 inthyroidglandexposure. children with thyroid cancer had received prior irradiation of the thymus as infants. Subsequent reports (18,19) confirmed the risk of thyroid cancer in children exposed to high-dose radiation, and the use of radiation to treat benign disease slowly dimin In addition, the risk of radiation has been analyzed in patients treated for malignant disease, in occupational settings, and in situations of inadvertent exposure. Atomic Bomb Survivors A fixed cohort of nearly 80,000 survivors of the atomic bomb exposures in Hiroshima and Nagasaki, Japan, has been followed since 1958 by the Atomic Bomb Casualty Commission (ABCC) and its successor, the Radiation Effects Research Foundation (RERF). In a comprehensive report (20) on the incidence of and risk estimates for solid tumors diagnosed between 1958 and 1987, the thyroid had one of the highest solid tumor risk estimatesin the Life Span Study cohort, with occult tumors excluded. The mean estimated thyroid dose was0.264 Sv, and a strong linear dose-response was 10 years had an excess relative demonstrated. Persons exposed when younger than age risk (ERR) of 9.46, over three times greater than those in their second decade (see Table 2). Although earlier studies (21) suggested otherwise, this report (20) showed that those individuals over the age 20 of years at the time of the blast had no evidence of an excess of thyroid cancer. Mortality data from the Life Span Study contributed little support for an increased risk of thyroid cancer, since the disease causes so few deaths (22). Cervical Tuberculous Admitis Tisell and colleagues (23) evaluated 444 patients treated with x-rays for cervical tuberculous lymphadenitis between 1913 and 1951in Goteborg, Sweden. The mean
Gerasimov 88
and
Jennings,
Figge,
Table 2 Major Cohort Studies of External Radiation in Childhood Study Re$
No. Thyroid Exposure Exposed
Age at
Dose
Atomic bomb (20) Age
79,972
17.6 yr
Hemangioma
14,351
6 mo
0.26 Gy
1 1,807
5 mo
0.12 Gy
4,296
4.4 yr
0.59 Gy
2,657
5 wk
1.36 Gy
10,834
7.4 yr
0.09 Gy
(25)
Hemangioma
0.264 1,950,567 Sv
(26)
Tonsils and (27) Thymus (33)
Tinea
adenoids
Follow-up
EWGy 95% CZ
PY
1.151s~ (0.5-2.1) 9.461s~ (4.1-19) 406,355 PY 4.92/Gy (1.3-10) 370,517 7.51Gy PY (0.418) 33 yr 3.0lGy (140) 9.0lGy 37.1 yr (4.2-22) 30.2 yr 3OlGy
(34)
EWGy, excess relative risk per gray; PY, person-yem.
age at irradiation was 19 years, with almost 50% of patients between 15 and 24 years of age. The calculated absorbed dose to the thyroid ranged from 0.40 to 50.90 Gy (median: 5.2 Gy; mean: 7.2 Gy); 25 thyroid cancers were found, all but one palpable, with a mean observation time of 43 years. The mean and median latency periods w 40 years to diagnosis. A positive correlation was shown between absorbed dose and No the probability of developing carcinoma, even after doses of more than 20 Gy. significant correlation between age at irradiation and the risk of developing cancer was detected. Cutaneous Hemangioma Furst and coworkers (24) followed up 18,030 patients with skin hemangioma who were treated with external beam radiation between 1920 and 1959 at the Karolinska 1 year of age (median age: 6 Hospital. At the time of therapy 82% were less than months). Treatment methods varied, but the relative risk of thyroid cancer was only slightly increased (1.18) in the group treated with radium-226 or orthovoltage x-rays. In patients receiving contact x-rays or no radiation, no increased risk was noted. No estimation of absorbed thyroid doses was made. A similar analysis (25) of a cohort of 14,351 infants less than 18 months of age (mean: 6 months) irradiated for hemangioma during the period 1920 to 1959 in Stoc holm covered 406,355 person-years at risk, with a mean follow-up of 39 years. The mean absorbed thyroid dose was 0.26 Gy. The Swedish Cancer Registry documente 17 thyroid cancers. Excess cancers began 19 years after radiation and persisted for at least 40 years after radiation therapy (see Table 2). Another study (26) involved 11,807 infants who had been treated with radium-226 5 months. The mean between 1930-1965 in Goteborg, Sweden, at a median age of
Radiation
89
absorbed thyroid dose was 0.12 Gy. Follow-up through the Swedish Cancer Registry yielded 15 thyroid cancers, ERR = 7.5/Gy (see Table 2).
TonsilslAdenoids Extensive data has been reported by Schneider and colleagues (27-30) from long5300 subjects who received external term follow-up studies of a population of more than radiation for a variety of benign head and neck abnormalities, principally for enlarged tonsils and adenoids, during the years 1939 to 1962. In analyzing one cohort of 4296 of these individuals with an average age at first exposureof 4.4 years and an average thyroid dose of 0.59 Gy, they found that ERR = 3/Gy for thyroid cancer (see Table 2). With a mean follow-up of 33 years, the majority of cases occurred in the interval between 20 and 40 years after radiation therapy, peaking at 25 to 29 years, with a significant decline in risk with increasing age at exposure (27). Additional data from this source includes information on the effect of screening, as well as characteristics of the secondary thyroid cancers. The authors documented recurrent malignancy in 13.5% of the 296 patients with thyroid cancer, nearly all within 10 years after primary tumor resection. Significant risk factors for recurrence were identified as the size of the primary lesion, number of lobes involved, histological type, vessel invasion, and lymph node metastasis (28). Longer term follow-up of 118 cases occurring before intensive screening showed recurrences23.7% in of this total;39% of cancers occurring in children recurred versus 15.6% in adults. This established an inverse relationship between the frequency of recurrence and the patient's age at surgery (also between the frequency of recurrence and the latency period between radiation and surgery). Age (29). One additional at radiation and treatment dose were not related to recurrences aspect explored in this group was the possibility of a radiation sensitivity within the population at risk. Patients with secondary salivary gland and/or neural tumors of the head and neck region had a significantly increased frequency of thyroid cancer comp to patients with neither of these tumors, suggesting that additional factors such as radiation sensitivity may account for this increased risk (30).
Acne Paloyan and Lawrence(31) found that20 of 224 patients referredfor thyroidectomy for solitary nodules had received antecedent radiation for treatment of acne vulgaris. Of the 20 patients, 12had thyroid cancer,9 to 41years after radiation therapy. Complete records were unavailable, and no statistical analysis was reported. To date no com sive survey of such patients has been performed.
Thymus Analysis of the effect of radiotherapy for thymic enlargementin infancy on subsequent neoplastic disease was initiated by Hempelmann and his colleagues in Rochester, New York, in the 1950s. They established that the risk of cancer was proportional to the thyroid dose, and raised concern that persons of Jewish ancestry might be at greater risk (32). Extended (average 37 years) follow-up (33) of a cohort of 2657 of these exposed infants and 4833 of their siblings via mail surveys through 1986 confirmed a linear dose-response relationship, with an ERR = of 9/Gy (see Table 2). The median 5 weeks, and 95% were under34 weeks of age. Estimated age at radiation therapy was thyroid doses ranged from 0.03 to >10 Gy, with a mean of 1.36 and median of only
90
Gerasimov Figge, Jennings, and
0.3 Gy. None of the dose fractionation variables examined (dose per fraction, numb
of fractions, and interval between fractions) was significant in modifying risk. Tinea Capitis A major long-term study (34) of 10,834 persons who received x-ray therapy for tinea capitis between1948 and 1960 in Israel compared the effects with a like numbe of nonirradiated individuals and 5392 nonirradiated siblings. All irradiated subjects were under 16 years old at the time of treatment, with a mean age of 7.4 years. The mean thyroid dose was 0.093 Gy, with the dose highly inversely correlated with age at exposure, dueto the proximity of the thyroid to the x-ray fields in smaller children The Israel Cancer Registry documented 98 thyroid cancers, with a mean interval of 17.1 years from radiation therapy to diagnosis. A much higher excess risk of thyroid cancer was found than in other studies (Table 2), possibly related to underestimation of thyroid doses as a result of patient movement.An increased risk in Jewish person may also have been a factor. A similar study of 2215 children irradiated for tinea capitis inNew York found no thyroid cancers via a mailed questionnaire after an average 20.5 year follow-up (35). However, there were less than 300 females and the expected number of thyroid canc 7.9 years, with an would have been only 2.9 (36). The mean age at treatment was estimated thyroid dose of 0.06 Gy (35). Previous Malignancy Tucker and colleagues (37) reported on the experience of the Late Effects Study Group, which followed a roster of 9170 patients surviving any type of malignancy in childhood for over 2 years. The period of risk extended to death, last follow-up, or th date of developing any formof second malignancy, whether thyroid or not. The me age at initial tumor diagnosis was 7 years, with 45% of all patients less than 5 years old. The duration of follow-up beginning 2 years after initial diagnosis was 2 to 48 years(mean: 5.5 years),withanaggregatefollow-upof 50,609 person-years.The radiation dose to the thyroid ranged from 0 to 76 Gy (mean: 12.5; median: 3.6 Gy). 23 secondary thyroid cancers through their 13 centers, yielding The authors documented a 53-fold increased risk over matched controls, with a significantly increased relativ risk among those with early age at initial cancer diagnosis. All of the thyroid cancer patients had received at least 1 Gy to the thyroid. A study of 1787 patients treated for Hodgkin’s disease (38) at Stanford University between 1961and 1989 included 1677patients who had thyroid radiation, most rece 44 Gy. The mean age at time of treatment was 28 years (range: 2-82 years). After an average follow-up of 9.9 years, they found six thyroid cancers, 9 to 19 (median: 13) years after therapy began, for a relative riskof 15.6 times expected. The age of these six patients ranged from 5 to 32 years at the time of exposure. A number of additional studies ( 3 9 4 3 ) have assessed the risk of radiation therapy for malignancies on the subsequent occurrence of thyroid and other second cancers. These include investigationsof large populationsof women treatedfor uterine cervica of secondary thyroid cancer and of males with testicular malignancy; no increased risk cancerhasbeendemonstrated,althoughonesuchstudyinwomenfoundaslight insignificant excess (RR = 1.1) (39). In this study, as well as the others, the thyroid
Radiation
91
gland was outside the field of direct radiation, with an estimated average thyroid dose of 0.15 Gy.
Occupational Exposure Occupational exposure to low-dose radiation has been analyzed in a number of large studies of workers in the nuclear industry. In a report on mortality among radiation workers in the United Kingdom, thyroid cancer was the only malignancy for which ERR = 1.051s~was low (44). the standardized mortality ratio was raised, although the A similar study of employees of the U.K. Atomic Energy Authority demonstrated a slight, but not significant, increase in mortality from thyroid cancer (45). Additional mortality studies in the United States (46-48) failed to demonstrate excess thyroid To date, no analysis of cancer incidence cancer deaths among nuclear materials workers. has been performed in such a cohort. A study of cancer incidence among medical diagnostic x-ray workers in China (49) found 7 thyroid cancers in 27,011 individuals employed between 1950 and 1980, with nearly 700,000 person-years of observation. Thyroid cancers were increased among workers employed for 10or more years and among those who began such work before A study ofmorethan143,000 1960. No dosimetrymeasurementswereobtained. (50) from 1926 to members of the American Registry of Radiologic Technologists 220ofself-reported 1982, who were evaluated through questionnaires, revealed a total thyroid cancers versus an expected number of about 100 cases. These data are preliminary, however, and do not include confirmation of the diagnoses, nor are thyroid dose estimates yet available. Mortality analysis by review of death certificates of British radiologists who died (51). Similar between 1897 and 1976 failed to demonstrate an excess of thyroid cancers mortality data from North American radiologists demonstrated no excess deaths from thyroid cancer compared to other specialty physicians for the period 1920-1969 (52).
Populations Living Near Nuclear Facilities A number of studies from the United States and the United Kingdom (53,54) have investigated the mortality from cancer among persons residing near nuclear power plants.Althoughsuchanalysesareproblematicowingtorelocationofpotentially exposed persons, case ascertainment in different areas, information on individual radia has tion exposures, and a varietyof social issues, no increase in thyroid cancer deaths yet been reported. The Chernobyl accident is considered separately below.
Prenatal Exposure The cancer risk of prenatal irradiation has been analyzed in atomic bomb surv from diagnostic imaging. Although some studies have found evidence of an increased incidence of childhood cancer following prenatal abdominal x-ray exposure, thyroid cancer rates have never been shown to be increased. Many of these reports have been based on mortality data (55,56), which would not be expected to show an increase for thyroid cancer, but several have utilized incidence data as well(57-59). These studies are confounded by a number of factors, however, including difficulties in dose estimation, and maternal issues that may affect the risk of subsequent malignancy, such as prenatal care, maternal age, sibship position, and prior miscarriage. Follow-up via dea
Gerasimov 92
and
Jennings,
Figge,
recordsandtumorregistriesof1630ofthe2802individualssurviving in utero exposure from the Japanese atomic bombs disclosed only one thyroid cancer through 1984 (60).
Internal Irradiation Introduction Human exposure to13'1has been analyzed in patients treated for hyperthyroidism at smaller doses ( 4 mCi) for diagnostic thyroid scans. Radioactive fallout contain l3II and short-lived radioiodines has also resulted in human thyroid irradiation. It is believed that I3'Iis considerably less effective in producing thyroid abnormalities than x-ra although oneof the best controlled animal studies suggests that the carcinogenic e are similar (61).Shorter lived radioisotopes of iodine are more destructive due to the greater penetration of their beta rays and faster dose rate, but their ability to produc thyroid cancer relative to x-rays is uncertain. Diagnostic 1311 In amulticentercohort (62) study of 35,074patients, 50 thyroidcancerswere observed through the Swedish Cancer Registry versus an expected number of 39.4 in the general population. This incidence was insignificantly greater than expected andmayhavebeeninfluencedbytheprevalenceofunderlyingthyroiddiseasein this selected population. The mean age at first I3'I examination was about 44 years, Cy. The mean followwith a mean total dose of 52 pCi and absorbed dose of 0.5 upwas20years,with527,056person-yearsatrisk,excludingthefirst 5 years after examination. Therapeutic 1311 Severalstudiesoftheeffectof 1311 therapyforhyperthyroidismonsubsequent (63), from the Cooperative malignancy have been performed. The largest of these Thyrotoxicosis Therapy Follow-up Study, evaluated 35,593 patients, including 23,02 treated with I3II. An elevated risk of thyroid cancer mortality following 13'1treatment was documented (63). In a group of 4,557 patients who received1311therapy for hyperthyroidism between 1951 and 1975 in Sweden, Holm (64) found no increased risk of thyroid cancer at doses estimated at 60-100 Gy. The mean age at treatment in this study was 56 years, with an average follow-up time of only 9.5 years. An excess of thyroid cancer was found by Hoffman and associates (65) in a study of 1005 women treated with I3'I at the Mayo Clinic, butthis excess was not statistically significant. In the study with the longest follow-up period (mean of 15 years for theof 85% recipient patients surviving) Holm and coworkers (66)found no increased risk of thyroid cancer in 3000 subjects treated for hyperthyroidismor cardiac disease, based on Swedish Cancer Registry Additional mortality studies(67-68) in women receivingl3*Itherapy for hyperthyroid ism have shown no excess thyroid cancer deaths. At the Cleveland Clinic (69),87 children and adolescents less than18 years of age when they received 1311 treatment for hyperthyroidism were evaluated. The mean 13'1 dose was 9.75 mCi. No thyroid cancers were detected in these patients or their off with a mean follow-up period of 12.3 years. Although I3lI in therapeutic doses may
Radiation
93
effect substantial cell killing and thereby mitigate any tumorigenic impact on the t long term follow-up of exposed populations is needed to establish the effectof I3'I on subsequent thyroid cancer risk. Fallout SOUTHWESTERN UNITED STATES PeoplelivinginNevadaand Utah nearthenucleartestsitewereexposedto radioactive fallout in the 1950s. At least 87 of the atmospheric tests between 1951 and 1958 resulted in offsite contamination (70). Thyroid dose estimates range from 0.46 (71) to 25 Gy or more (70), with added uncertainty regarding the amount of consumption of contaminated milk. It is not known whether short-lived isotopes of iodinewereinvolved.Thyroidexamination 12 to 18 yearslaterof 5179 children from the area of greatest exposure failed to disclose any increase in abnormalities in one study (71). However, an interview survey of a 1951cohort of 4125 Mormons in this area disclosed an excess incidence of thyroid cancers throughout the period 1958 to 1980 (70). This report was challenged by a subsequent mortality study of this region, which found no excess thyroid cancer deaths (72). Due to the lack of accurate dose information in this setting, no definite conclusions regarding the risk of fallout exposure are possible. MARSHALL ISLANDS In 1954, after detonation of a 15-megaton nuclear device at Bikini, an unanticipated windshift resulted in exposure to fallout of at least 300 people on at least three of the atolls of the Marshall Islands (73-75). Late effects of this exposure have been (1311, 1321, 1331, and 1351), predominantly thyroid abnormalities from absorbed radioiodines as wellaspenetratingwhole-bodygammaradiation.Significantuncertaintyexists regarding thyroid doses, with rough estimatesof an average of 3.12 Gy in all exposed children. Although a significant increase in nodular thyroid disease and hypothyroidism has been demonstrated throughout the northern atolls, few cancers have been documented, with an estimated risk only1.9 times greater than unexposed Marshallese. The risk of thyroid cancer was lower in children under 10 years old at irradiation than in older populations, suggesting that dose estimates might be too low and that significant cell-killing occurred in the younger group, reflected by their higher of hypothyincidence roidism. No thyroid cancer has been detected in the 10 individuals exposed in utero, although 2 of these developed benign nodules. OTHBR INCIDENCES
Wiklund and colleagues(76) studied a cohortof 2034 reindeer-breeding Lapps who had ingested large amounts of radioactive fallout products from nuclear weapons tests For the period in the USSR. Exposure was through the lichen-reindeer-man food chain. 1961-1984, no excess in thyroid cancer incidence was detected through the Swedish Cancer Registry. I j 1 1 Risk in Children and Adolescents Data in the literature regarding1311 exposures in individuals under age20 are sparse (77). Exposed populations were small and only small numbers of thyroid cancer cases (23 cases) have been reported. Because of these factors,and the fact that some subjects
94
Gerasimov Figge, Jennings, and
were being investigated for thyroid diseases and others were administered I3*Idoses in the cell-sterilization range, there is insufficient scientific information to draw conclu about the risk posed by I3'I in children and adolescents. ANALYSIS OF RISK ASSESSMENT
Introduction The association between radiation exposure and subsequent thyroid cancer has conclusively demonstrated in epidemiological studies of children receiving head and neck irradiation and in survivors of the atomic bomb exposures in Japan. These Gy) is highly associated have demonstrated that radiation to the thyroid at high(>1 doses with the subsequent development of cancer; the effect at lower doses is difficult to assess. Previous studies to assess doses of less 0.10 than Gy have produced no conclusive evidence of significant risk, but the requisite sample of greater thanlOO,O00 exposed individuals and a similar control population have not been identified and analyzed.
Modifiing Factors Type and Duration of Exposure External radiationis roughly fourto five timesas effective in causing thyroid canc as is I3'I for each unit of absorbed dose (78), with other isotopes of iodine probably having an effect between that of l3lI and external radiation. Fractionation appears to (79). However, provide about a30% reduction in the tumorigenic effect on the thyroid x-ray technicians may be at increased risk over the general population (49,501.
Age at Irradiation The thyroid is more radiosensitive in children than in adolescents, and similarly (37) found that individuals more so in adolescents thanin adults. Tucker and colleagues treated at an early age appeared to have a higher relative risk of thyroid cancer and also after lower doses of radiation, suggesting some increased sensitivity to radiatio of thyroid cancer following irradia Shore (78) estimated that the geometric mean ERR 10%that in children.In the atomic bomb survivors, thyro tion in adulthood was about cancer in children had one of the highest ERR estimates among solid malignancies, while there was virtually no ERR for thyroid cancer in adults (20). Large studies of women treated with radiation therapy for cervical cancer (39,40) are among the few in adults that have demonstrated an excess risk for thyroid cancer, but the confidenc intervals were very wide in each study.
Sex The absolute risk in females is two to four times that in males, but the E W G y is about the same in both sexes.Most thyroid cancers occurred in females in a study by Lundell and associates(25), but due to their higher background incidence rate, the s (34) and Shore and specific relative risk estimates were similar. Ron and associates coworkers (80) reported a greater excess numberof cancers among females compared to males, but no significant difference in the relative risk estimates. According to the report of the BEIR V Committee (81), females are about three timesas susceptible to radiogenic and nonradiogenic thyroid cancer as males.
Radiation
95
Race The risk appears to be greater in individuals of Jewish ancestry. Thyroid cancer risks varied among different Jewish subgroups in the Israeli Tinea study, with those born in Israel having one-third the risk of those born in the Middle East or North Africa. Since the fathers of those born in Israel were themselves born in the Middle East or North Africa, environmental rather than genetic issues seem to be(81). operative
Iodine Deficiency Iodine deficiency is a possible promoting factor, since decreased thyroid hormone results in increased stimulation of thyroid epithelium by TSH. However, at least two human studies indicate the opposite effect, with thyroid cancer associated with a high of iodine dietary iodine intake (78). To date, there are no reports on the influence deficiency on the risk of radiation-induced thyroid cancer.
Parity The observation that thyroid cancer among the exposed Marshall Island population occurred exclusively in -multiparous women suggested that parity might increase the risk of radiation-induced thyroid cancer. Shore and colleagues (33) demonstrated that older age at first childbirth significantly increased the risk of radiation-induced thyroid cancer in patients irradiated for thymic enlargement in infancy. A similar effect was found with older age at menarche. Other studies have demonstrated that a history of miscarriage increased this risk, especially for younger women (78).
Latency Period The interval between initial exposure to radiation and detection of thyroid cancer 5 to 50 yearsafterirradiation, varieswidelyamonghumanclinicalstudies,from reflecting in large part the follow-up interval of the study. The latency period may.also increase with the age of the individual at irradiation.
Efect of Screening Basedon an intensivescreeningprogrambegunin1974inChicago,Ronand of secondary thyroid cancer were colleagues (82) reported that adjusted incidence rates seven times greater during the screening period (1974-1979) than before.
Temporal Pattern The temporal pattern of risk remains uncertain due to the limited long-term followup data available. Schneider and colleagues (27) estimated that the increased risk of radiation-induced thyroid cancer probably lasts throughout life. Similarly Thompson and colleagues (20) found no evidence for a decrease in risk with time after exposure. Ron and coworkers (34) reported a continued increase in risk over their entire study periodofup to 38 years. Shore and collaborators (33) reported that the risk ratio declined over time, but remained highly elevated at least 45 years after irradiation. Excess risk began 5 years after exposure. Shore and coworkers(33) found that ERR decreased during the entire study period, but that there was no significant change over time in excess absolute risk (EAR). Conversely, Ron and coworkers (34) demonstrated no significant change in ERR, but 30 years). a continuing increasein EAR over the whole study period (mean follow-up:
96
Gerasimov Figge, Jennings, and
These somewhat contradictory results highlight the need periods to clarify the temporal pattern.
for even longer follow-up
Dose-Response Relationship A strong dose-response relationship between radiation and incidence of thyroid cancer has been documented in Japanese atomic bomb survivors (20) and in studies (79) of of children and adolescents (25,27,34,80). The results of a pooled analysis seven major studies over a wide range of doses demonstrated an ERR of 7.7 per Gy (95% confidence limits: 2.1-28.7). For persons exposed to radiation before age 15 years, linearity best described the dose-response relationship, even down to 0.10 Gy. Although risk estimates are generally those of the. linear no-threshold model, at very (15). Regardless of high doses these estimates might not be valid due to cell killing possible threshold effects at high doses due to cell killing, the need is greatest for an understanding of carcinogenic effects of low-dose radiation.
THE CHERNOBYL ACCIDENT AND THYROID CANCER
Circumstance of the Accident, Radioactivity Release The Chernobyl accident was, without question, the worst technological disaster in 26,1986, at 1:23 A M , two explosions the history of nuclear power generation. On April occurred (due to steam and hydrogen) in reactor number of four the Chernobyl nuclear power station, ejecting large amounts of radioactive material into the atmosphere. Subsequently, the graphite within the reactor ignited and fuel elements in the core of the reactor melted, resulting in the release of volatile radioactive products over a 10day period. The immediate cause of the accident was operator error, but the reactor design (which lacked a concrete containment vessel) has been implicated in the ser (83) indicated consequences of the accident. Initial estimates from officials in Moscow that approximately 4% of the total activity of the core escaped into the atmosphere, resulting in the release of some 50 million Ci (2 x lo'*Bq). However, other researchers concluded that the release was much greater(84,85). After 18 months of study at the reactor site, Sich (86,87) estimated that the total release was actually in the range of 120 to 150 million Ci. Over80 different isotopes were released (88); the most abundant volatile isotopes were those of iodine (I3II, 1321,L331,and 1351),tellurium (L32Te), and cesium (~"CS and 137Cs). Some of the radioactive isotopes released during the accide naturally decay to isotopes of iodine, for example, I3*Te has a 3-day half-life and de to 1321.
Geographic Distribution of Volatile Radioactive Isotopes The distribution of volatile radioactive isotopes to different geographic regions w (89-92). Theinitialplumeof governedbytheprevailingmeteorologicconditions volatileisotopesdriftedovernorthernUkraineandtheGomeloblast(region)of southern Belarus (map, Fig. 1). Contaminated air masses then moved west and then northwest, sweeping across the Brest and Grodno oblasts of Belarus, and resulting in the deposition of isotopes in Sweden on April 27. The wind direction changed to the northeast and to the east on April 29, and a large cloud of radioactivity drifted over southern Belarus and the southwestern corner of the Russian Federation. A substantial
Radiation and Thyroid Cancer
97
areas with contamination higher than 5 wries per square kilometer
0
Fig. 1. Mapshowingthedistribution of '"CS in Belarus,Ukraine,andRussia. Reference 93, courtesy of the World Health Organization.)
(From
deposit of radioactivity in the Gomel and Mogilev oblasts of Belarus and the Bryansk oblast of Russia resulted from rainfall during April 28-30, which washed fallout from the cloud onto the ground. Another substantial deposit about 500km from Chernobyl was formed when the same cloud drifted over theKaluga-fila-Ore1 oblasts of Russia. Rains during April28-30 washed fallout to the ground in these regions. Winds changed to the southand then shifted to the southwest during the last few days of the accident, contaminating the Balkans and Alps. The World Health Organization estimated that 4.9 million people lived in areas where ground surface contamination exceeded 1 CY k m 2 (93).About 2.3 million children lived in areas that were significantly contaminated at the time of the accident (94).
Cesium-137 Release Approximately 2 million Ci (8 x 10l6 Bq) of13'Cs was released, causing widespread soil contamination (89,95).The distribution of 137Cs,which has a half-life of approximately 30 years, has been carefully mapped (89), and was deposited in the following manner: Belarus 33.5%, Russia 24%, Ukraine 20%, Sweden 4.4%, Finland 4.3%. The areas receiving the highest 137Cs contamination are shown in Figure1. Radioiodine Release The heaviest initial exposure to the population resulted from isotopes of iodine. According to recent studies, the release of 1311 (half-life: 8.05 days) was on the order 50 million Ci (approximately 1.7X lo1*Bq), representing about 50-60% of 40 million to
98
Figge, Jennings, and Gerasimov
of the core inventory (95-98). By comparison, theThree Mile Island accident released only 15 to 20 Ci of 1311in the United Statesin 1979. During the first month following which the Chernobyl accident, the major source of internal radiation exposure1311, was was acquired by inhalation as well as ingestion of contaminated food. Deposits of I3II onpasturelands and gardens in the rural agricultural areas surrounding the reactor introduced this radioisotope into the food chain. Ingestion of contaminated milk was the most important source of internal 1311 exposure in children (90). Consumption of contaminated leafy vegetables was a secondary source of internal I3IIexposure. as 1321(half-life: 2.3 hours) and 1331(half-life: Short-lived isotopes of iodine such 21 hours) were also released from Chernobyl4. Very few direct measurements of radioiodines were made in the initial days following the explosion. Therefore, data which were important primarily in the first days following the regarding 13*1 and 1331, 28,1986, inWarsaw,Poland, accident,arescarce.MeasurementsmadeonApril revealed that 28% of the radioactivity in the air was due to short-lived iodine isotop (99).Thus, populations living near the reactor were exposed 1321and to 1331 via inhalation for at least 1or 2 days. Following the accident a limited numberof measurements of the ground deposition density of l3II were conducted in Belarus by the Belarus Institute of Nuclear Physics (Minsk) (100). A map(Fig. 2) ofthe I3'I deposition in Belarus (96) showssome obvious differences in the distribution 1311 of compared with the pattern of 137Cs grou contamination (Fig. 1). In particular, the Gomel and Mogilev oblasts were both hea contaminated with 137Cs, with relatively less contamination in the Brest oblast. By contrast, the 1311contamination was highest in the Gomel oblast, with lower but sig cant levels of deposition in both the Mogilev and Brest oblasts. The contamination in this area the Brest oblast arose from the initial plume of radioactivity that passed over during the first day of the accident.
Reconstruction of Thyroid Doses
Ideally, to support careful epidemiological studies, one would like to have accurate thyroid dose reconstructions that separate out the contribution 1) of external radiation, 2) internal radiation due to I3'I (from both inhalation and ingestion), and 3) internal (1321and 1331), becausethesethree radiationduetotheshort-livediodineisotopes components may have differing potential to cause thyroid cancer. For example, 1321and 1331,which decay more rapidly than1311,deliver their radiation dose overa shorter time interval and could theoretically have a carcinogenic effect on thyroid tissue similar to that of x-rays (101). It has been estimated (94) that the majority of thyroid exposure (85%) was from internally concentrated I3lI derived from ingestion. Approximately 15%was estimated to have been derived from inhalation of short-lived isotop Following the accident, direct measurements of thyroid radioactive iodine content were made in Belarus, Russia, and the Ukraine. From these, one can extrapolate the exposure to l3II, but the measurements were made too late to give useful information on 1321 and 1331.Furthermore, the direct measurements were made on only a small proportion of the affected population. Thyroid dose reconstruction is needed to estimat the exposure for the rest of the population. Many factors may account for variability in thyroid doses received by different individuals in the same geographic area. For example, many families grew their own vegetables and obtained milk from their ow
Radiation and Thyroid Cancer
99
Contamlrwtlonzo11(w Iodine-131
185-370 ki3q/rn2 l!!F!!3 3706,550 kBq/d r"17 5,550-18,500 k B q / d 1"""118,50O-37,000 kBq/rn2 above 37,000 kBq/rn2
0
=
Fig. 2. Map showing the distribution of I3'I ground contaminationin Belarus. Annual incidence ratesofthyroid cancer inchildren in differentgeographicdistrictsareshownper 100,000 children (basedon data in Reference127). (From Reference96 with permission of the American Association for the Advancement of Science.)
cow. Many individuals were outdoors mostof the day at the time of the accident and slept with the windows open at night, thereby maximizing their exposure to I3'I by inhalation. The thyroid dose is known to be inversely related to thyroid mass. Thus, for a given uptakeof I3'I, children will achieve a higher thyroid dose than adults. The level of iodine in the diet will also influence the efficiencyof uptake of 13'I.Southern Belarus suffers from mild iodine deficiency, with some relatively isolated pockets of severe iodine deficiency (102,103). Iodine supplementation measures had lapsed by 1985. The implications of this are that individuals living in iodine-deficient areas wou have a greater thyroid uptake of radioiodine than those living in iodine-replete areas. An effective prophylaxis program utilizing potassium iodide, as was administered in Poland (99), could have limited radioiodine exposure. Since exposed inhabitants were
zoo
Gerasimov Figge, Jennings,and
not immediately informed of the accident, and there was no immediate effort to ically prophylax the population, potassium iodide was not administered early enough (if at all) in Belarus and Ukraine to be effective.
Belarus Direct measurements of thyroid I3IIcontent were made during May and June, 1986, in approximately 300,000 individuals living in the contaminated areas of Belarus. Ab 200,000 records were verified and form the basis of a database for the calculation of individual thyroid dosesof Belarussian residents(104-109). About 150,000 individuals in the database were interviewed regarding lifestyle and diet. Thyroid dose estimates have been completed for 130,000 residents of the Gomel and Mogilev oblasts and Minsk City who had direct thyroid measurements completed before June 6. Estimates were based on the direct measurements plus information on lifestyle and diet (e.g., I3II intakebyinhalationandby level ofmilkconsumption).Calculationsassumed ingestion of fresh milk following a single depositionof fallout on pasture grass(104). Average thyroid doses have also been estimated for individuals living in 800 rural settlements who did not have direct thyroid measurements. These reconstructions are calculated using the above-noted database, taking into account theoflevel consumption of fresh cow's milk. I3II in Belarussian children living in different Reported average thyroid doses of of the Gomel and Mogilev oblasts ranged contaminated raions (administrative districts) from 0.15 to 4.7 Gy (104-109). Young children (age 7 and below) in these districts generally received thyroid doses that were 3- to 5-fold higher than those recorded in adults living in the same district. Several hundred children in Belarus received doses of 10 Gy or more to the thyroid. The highest thyroid dose did not exceed 60 Gy.
Russian Federation In addition to the 130,000 direct measurements in Belarus, there were 28,000 ments made in the Kaluga oblast and 2000 measurements in the Bryansk oblast of Russia (110-112). These oblasts also suffered from mild to moderate iodine deficiency (103). The mean thyroid dose due to iodine radionuclides in children in Bryansk was 0.5 Gy, but inthe more heavily contaminated zones it was 2.2 Gy. In the Kaluga obla the mean dose in children was 0.25 Gy. In the more heavily contaminated areas the mean dose was 0.5 Gy, with individual doses as high as 10 Gy.
Ukraine Direct measurements of thyroid I3II content were made in 150,000 people in Ukraine in May-June 1986, including 108,000 children and adolescents aged0-18 years (113117). The measurements were conducted in four of the northern oblasts: Chernigov, Kiev, Zhitomir, and Vinnytsia. Large-scale thyroid dose reconstructions were carried out using the direct measurements in combination with environmental data and inf tion on personal behavior and intakeof milk and leafy vegetables. Empirical relations of I3'I intake and the level of 137Cs soil contami were developed between parameters and the distance and direction from the nuclear plant. These relations allowed est of thyroid I3lIcontent in territories without direct measurements, such as the Cherkas and Rovno oblasts. In different administrative raions of northern Ukraine, average thyroid doses from I3'I in children and adolescents ranged from 0.03 to 1.6 Gy.
202
Radiation and Thyroid Cancer Table 3 Incidence of Thyroid Cancer in Children (Under Age 15 at Diagnosis) Rate 1991-1994 Location
990
1986-1
Belarus Gomel Oblast Ukraine Kiev, Chernigov, Cherkassy, ROMO, Oblastsand Zhitomir Russia and Brvansk Oblasts Kaluga
1981-1985
0.3 0.5 0.5
10.5 1.1
30.6 96.4 3.4
0.1
2.0
11.5
0
1.2
10.0
4.0
Annual incidence rates per million children under age 15 are given.
Data from Reference 94.
Thyroid Cancer Incidence in Children Following the Chernobyl accident, Prisyazhiuk and colleagues (118) reported a small increase in thyroid cancer cases in children from three districts in the northern Ukraine, within 80 km of the nuclear plant. Another report from the Ukraine followed (119). Local physicians had simultaneously detected a marked increase in the rate of child thyroid cancer in Belarus, starting in 1990 and primarily affecting the Gomel oblast (92,120-123).Whereas only 1 or 2 cases of thyroid cancer were seen annually in the Gomel oblast from 1986 through 1989, there were 14 cases in 1990 and 38 cases in 1)were reported to be papillary carcinomas. 1991. Most ofthe cases from Belarus (128/13 The initial reports were greeted with some skepticism by the international scientific community. Therefore, a team of international scientists, under the auspices of the WHO and the Swiss government, visited Belarus in July, 1992, to verify the ofaccuracy the histologic diagnosesof thyroid cancer. The international team studied the histologic specimens from 104 children in whom the diagnosis of thyroid cancer had been made (124).The team also reported since 1989 and agreed on the diagnosis in 102 cases that there was a marked increase in the incidence of childhood cancer (age 14 and under) in Gomel from 1990 onward, on the order of 80 cancers per million children 1 case of per year by 1992, as compared with the usual background rate of around cancer per million children per year. Subsequent data (94)has shown a continued increase in thyroid cancer incidence in children from Belarus, northern Ukraine (Kiev, Chernigov, Cherkassy, Rovno and Zhitomir oblasts), and southwestern Russia (Bryansk and Kaluga oblasts) since the accident (Table 3). As shown, rates are expressed as cases of pediatric (age 14 and under) thyroid cancer per million children per year. Belarus The annual pediatric (age 14 and under) thyroid cancer incidence rate in Belarus increased from 0.3 per million in 1981-1985 to 30.6 per million in 1991-1994, a 100fold increase.A total of 333 casesof pediatric thyroid cancer were diagnosed in Belarus 7 months of 1995 57 additional from 1986 through 1994 (94,125,126). During the first cases were diagnosed. By contrast, there were only 7 pediatric cases in Belarus for 9
202
Figge, Jennings, and Gerasimov
age at
operation
' 0 90
91
92
93
94
95
date of operation Fig. 3. Graph showing the age of Belarussian children at the time of thyroid surgery versus the date of surgery. The bold line corresponds to the ageof a child born on November 26, 1986. Note that very few cases fall below the bold line. (From Reference 130, courtesy of the
European Commission.)
In the Gomel oblast, the incidence rate years preceding the accident (1977-1985). increased nearly 200-fold, up to 96.4 per million. Of the 390 pediatric cases reported in Belarus through mid-1995,54.3% were fro the Gomel oblast, and 21.8% were from the Brest oblast. Only 1.8% were from the Annual Vitebsk oblast, which was not contaminated following the Chernobyl accident. childhood thyroid cancer incidence rates for different geographic zones in Belarus shown in Figure 2 (in cases per 100,000 children per year for 1990-1991). There is a 13'1contamination levels (Fig strong correlation between these incidence rates and soil (127-129). As noted by those 2), as documented by the study of Abelin and colleagues authors, the higher incidence rates occurred along the two paths taken by the initial clouds of volatile radioisotopes, one pathway to the west and one to the northeast. The highest annual incidence rate (130.8 per million) was reported in the southern part o the Gomel oblast, adjacent to Chernobyl, whereI3II thecontamination level was highe 13'1deposition argues The correlation between childhood thyroid cancer incidence and that radioactive isotopes of iodine played an etiological role in the pathogenesisof the thyroid cancers. The ratio of affected girls to boys in Belarus was 151.0. The majority of affected children (3861390) were borneither before the accident or near the time ofthe accident only four of the children were born after 1986. The rate of thyroid cancer in children born after 1986 is low and approximates baseline levels before the accident. Figure 3 shows data from 298 children diagnosed with thyroid cancer at the Pathology Ins in Minsk from 1990-1994 (130). Note that there is a sharp cutoff age (Fig. 3, bold line) below which very few young children have presented with thyroid cancer, and thecutoffageincreaseswithtime.TheboldlineinFigure3representschildren who were born on November 26, 1986. Children born on this date would have been approximately 10 weeks gestational age at the time of the Chernobyl accident. Since
103
Radiation and Thyroid Cancer
501
Y. of birth 1982-66
46
Y . of birth 1D77-81
0 Y . of blrth 1972-76 0Y . of birth 1867-71 ~~
1992 1993 1994
Fig. 4. Graph showing the number of new thyroid cancer cases during each of the years 1986 to 1994 in cohortsof Belarussian children defined by year of birth. (From Reference131, courtesy of the European Commission.)
the fetal thyroid gland can concentrate iodine by 12 weeks, these children could theoretically have sustained significant thyroid exposure toI3’Iin utero during the first month following the accident. These data strongly suggest that intrathyroidal accumulation of radioactive iodine isotopes, either in utero or after birth, was an important factor in the pathogenesis of the pediatric thyroid cancers in Belarus. An analysis of thyroid cancer cases in Belarus by cohorts, defined according to the patient’s dateof birth, is shown in Figure4 (128,131). It is clear that increasing numbers of cases have occurred in each cohort at least through 1993. The largest number of new thyroid cancer cases has occurred in individuals who were age 4 and younger at the time of the accident (birth date 1982-1986), followed by those who were age 5 through 9 (birth date 1977-1981); however, individuals as old as19 at the time of the accident were still at risk. These data suggest that younger children are most su to the carcinogenic effects of radioactive iodine isotopes. Russian Federation In the contaminated oblastsof the Russian Federation, an increase in the incidence (94,111,132,133). of thyroid cancer in children and adolescents has been registered The annual incidencein children (age 14 and under) in the Bryansk and Kaluga oblas has increased from background to 10 per million. The major increase has been in the Bryansk oblast where 21 cases have been reported between 1986 and 1994. Ukraine Between 1986 and 1994 there were211children (age 14 and under) who underwent surgery for thyroid cancerin Ukraine (94,113,114,134,135). The incidence in children increased from 0.4 to 0.6 per million pre-Chernobyl to 4 per million by 1992-1994. The ratio of girls to boys1.4: was 1.O. In the five most northerly oblasts (Kiev, Chern Zhitomir,Cherkassy,Rovno),whichwereheavilycontaminatedbytheChernobyl accident, the incidence was much higher,11.5 per million children. About60% of the 5 oblasts, out of25 oblasts in the country. Only cases in Ukraine originated from these two children who presented with thyroid cancer were born after 1986, equivalent to 1986. In the an incidence of less than 1 per million per year in children born after
2 04
Figge, Jennings, and Gerasimov
town of Pripyat, located 3.5 km from the Chernobyl plant, the incidence in children and adolescents who were age 0-18 at the time of the accident was 137 per million by 1990-1992. Throughout the Ukraine, there was a 30-fold gradient in thyroid can incidence rates in individuals aged 0-18 at the time of the accident, corresponding l3lI exposure (114).This relation directly to the gradient in thyroid doses resulting from between cancer incidence and thyroid I3lI dose strongly supports a role for radioactive iodine isotopes in the pathogenesis of the cancers.
PathoZogic and Biologic Features of the Pediatric Thyroid Cancers The pathologic features of the thyroid cancers arising after the Chernobyl accident (136in children from Belarus, the Ukraine, and Russia have been well characterized 145). With few exceptions, all of the cases have been papillary carcinomas. Several histological subtypes have been noted (136-139), including classical papillary architecture, often with mixed papillary/follicular elements (approximately 11%); a mixture of (8%). Primary solid and follicular structures (73%); and the diffuse sclerosing type tumors were 1 cm or larger in diameter in the vast majority of cases (79-88.5% in three series) (125,140,141). Thyroid tumors arising in children are typically more aggressive than those that arise in adults (146151). This phenomenon was also true in the Chernobyl-related cases. The tumors were commonly widely invasive within the thyroid gland [33% in one series (141);59% of cases in another series (136)l. There was direct invasion of extrathyroidaltissue(stageT4)inahighproportion of cases(48-63%) (125,126, 134,136,140,144).Lymphatic invasion was present in 77% of cases, and blood vessel invasion in 15-32% (136,140,141).Regional lymph node metastases (stage N1) were presentin 5948% ofcases (125,134,140,141,144). Distantmetastases(stage M1, usuallytolung)werepresentin 5 9 % of cases (125,140,144). Onlyafewcases showed features of “occult” or microcarcinoma. Taken together, these pathological and biological features argue strongly against the cancers being incidental findings (152-154). In nearly all cases, the cancers represented clinically significant disease; only 9% of the children in one series from Belarus were staged at T1 NO MO (140). MOLECULAR CHARACTERIZATION OF CHERNOBYL-ASSOCIATED PAPILLARY THYROID CARCINOMAS
Ret Oncogene Activation of the ret oncogene (for review, see Chapter 6) by chromosomal rea ment was initially reported in four of seven Chernobyl-associated pediatric cases by It0 and coworkers (155). Subsequent studies(156-159) identified retPTC3 as the most prevalentformof ret rearrangement in early post-Chernobyl papillary carcinomas, presenting prior to April, 1996 (Table 4). Several atypical forms of ret rearrangements have also been identified in a few post-Chernobyl cases (159-163). In contrast with the earlier studies, Pisarchik and colleagues (164) found a higher prevalence of retPTC1 rearrangements (29%) in 31 post-Chernobyl papillary thyroid carcinomas presentingin 1996. However, the prevalence of retPTC3 was found to be quite low (7%) in a subset of 15 of these cases (165). Pisarchik and colleagues (165)
Radiation and Thyroid Cancer
2 05
Table 4 Studies of Ret Rearrangements in Post-Chernobyl Papillary Thyroid Cancers Presenting Prior to April, 1996 Authors (re$)
N
Dates of Diagnosis
6 1991-1992 Fugazzola et al. (156) Nikiforov et al. (157) 1991-199238 Klugbauer et al. (158) 1993-199512 Rabes & Klugbauer (159) 59 Prior to April, 1996
ReVPTCI RetPTC2 RetPTC3 0 (0%) 6 (16%) 2 (17%) 8 (14%)
1(17%) 1(3%) 0 (0%) 0 (0%)
3 (50%) 22 (58%)
6 (50%)
19 (32%)
suggested that there was a switch in the ratio retPTC3 of to retPTC1 rearrangements in late(1996)versusearly(1991-1992)post-Chernobylpapillarythyroidcancers. Smida and colleagues(166) independently arrived at a similar conclusion after studying 51 Chernobyl-relatedcases: 26 casesdiagnosedin1996and1997versus25cases from1993through1995.Inthecasesoriginatingfrom1996and1997, retPTC1 rearrangements were found in 31%, and retPTC3 in only 15%, in keeping with the ret/ data of Pisarchik and coworkers (164,165). In the earlier cases from 1993 to 1995, PTC1 appeared in only 16%, whereas retPTC3 was found in 36%. These authors (166) suggested that retPTC3 may be typical for radiation-associated childhood papillary thyroid carcinomas with a short latency period, whereas retPTC1 may be a marker for carcinomas appearing after a longer latency period. The results regarding ret rearrangements are particularly interesting in view of the recent demonstration that X-irradiation(50-100 Gy) in vitro can induce ret oncogene rearrangements in undifferentiated human thyroid carcinoma cells (167). Furthermore, of ret rearrangements (primarBounacer and colleagues(168) reported a high frequency ily retPTC1) in papillary thyroid carcinomas originating from patients with external ret rearrangements are radiation history. These results, taken together, suggest that important in the pathogenesis of radiation-induced papillary thyroid carcinomas, and the particular type of molecular rearrangement (retPTC1 versus retPTC3) may influence the biology of the cancer (e.g., the latency period).
Other Genetic Loci Other genetic loci have been investigated (130,136,169-173) including p53, the TSH receptor (TSH-R), and the three ras genes (H-rus, K-rus, N-rus). Nikiforov and colleagues (169) used single-strand conformation polymorphism (SSCP) analysis and found a p53 missense mutation in one of 33 Chernobyl-associated papillary thyroid carcinomas (3%) involvingcodon 160. Hillebrandtandcoworkers (170,171) used temperature gradient gel electrophoresis("GGE) and identified only one p53 missence mutation (involving codon 258) out of 70 post-Chernobyl papillary thyroid carcinomas. . Smida and colleagues (172) foundfivecasesof a silent mutation in p53 codon 213 out of 24 Chernobyl-related papillary carcinomas. Suchy and colleagues (173) studied 34casesofpost-Chernobylpapillarycarcinomas,butfoundnomutationsinp53. Alterations of the TSH-R and rus genes are rare, suggesting that mutations in these genes do not play a significant role in the pathogenesis of Chernobyl-associated thyroid cancers.
106
Figge, Jennings, and Gerasimov
EPIDEMIOLOGICAL CONSIDERATIONS
Following the initial reports of thyroid cancer cases in the regions surrounding Chernobyl, there were many questions about whether the cases were related to the (82,I74,175).The data reviewed accident or simply represented increased ascertainment of the cancer cases were correctly in this chapter support the contention that nearly all diagnosed, and the majority represented clinically important disease, not incidental cases found by screening. Some oblasts that received little radiation (Vitebsk) were subjected to intensive screening but yielded very few cases. Thus, increased ascertai ment cannot explain the dramatic and sustained increase in incidence that has been documented. In addition, data reviewed above suggest that radioiodine isotopes are implicated in the pathogenesis of the cancers. A small case-control epidemiological study has provided some additional support for this point by demonstrating a doseresponse relationship at the level of the individual thyroid dose (176). Further largescale epidemiological studies are planned (176). Other questions to be addressed are of tellurium the exact contributions of1311 vs short-lived radioiodines. The contribution also needs to be sorted out. The possible contribution of other environmental factors (industrial pollution, iodine deficiency), and host factors (such as increased genetic also needs to be considpredisposition to increased sensitivity to radiation effects) ered ( I 77).
Studies in the Tula Oblast of Russia G. Gerasimov(TheRussianEndocrinologyResearchCenter)and J. Figge have conducted field surveys between 1991 and 1995 in the Arsenyevo districtof the Tula Oblast in Russia, an area contaminated with13’Cs at a density of 5-15 C W 2 , and in the Yasnagorsk district, a noncontaminated area. Both regions had mild iodine urinary iodine levels of 7-9 pg/dL. The distribution of benign thyroid lesions from both regions was similar. A papillary thyroid cancer was diagnosed in 1991 in one As pointed out by Williams female from Arsenyevo who was 12 years old in 1986. (154), it is not known why areas contaminated with fairly high levels of fallout in An SouthwestRussiaappeartohavefewerthyroidcancercasesthaninBelarus. answer to this question will be important to understand the factors involved in thyro cancer pathogenesis. REFERENCES 1. Hanson GA, Komorowski R A , Cerletty J M , Wilson SD. Thyroid gland morphology in young adults: normal subjects versus those with prior low-dose neck irradiation in childhood. Surgery 1983; 94:984-988. 2. Spitalnik PF, Straus FH.Patterns of human thyroid parenchymal reaction following lowdose childhood irradiation. Cancer 1978; 41:1098-1105. 3. Freedberg AS, Kurland GS, Blumgart HL. The pathologic effects of 1-131 on the normal thyroid gland of man. J Clin Endocrinol 1952; 121315-1348. 4. Can R F , LiVolsi VA. Morphologic changesin the thyroid after irradiation for Hodgkin’s and non-Hodgkin’s lymphoma. Cancer 1989; W.825-829. 5. Quimby EH, Werner SC. Late radiation effects in roentgen therapy for hyperthyroidism. JAMA 1949; 140:1046-1047.
Radiation
107
6. Winship T, Rosvoll RV. Thyroid carcinoma in childhood: final report on a 20 year study. Clin Proc Child Hosp Washington, DC 1970; 26:327-349. 7. Roudebush CP, Asteris GT, DeGroot LJ. Natural history of radiation-associated thyroid cancer. Arch Intern Med 1978; 138:1631-1634. 8. Samaan NA, Schultz PN, Ordonez NG, Hickey RC, JohnstonADA. comparison of thyroid carcinoma in those who have and have not had head and neck irradiation in childhood. J Clin Endocrinol Metab 1987; 64:219-223. 9. Mehta MP, Goetowski PG, Kinsella TJ. Radiation induced thyroid neoplasms 1920 to 1987: a vanishing problem? Int J Radiat Oncol Biol Phys 1989; 16:1471-1475. 10. Akslen LA, Haldorsen T, Thoresen SO, Glattre E. Incidence pattern of thyroid cancer in Norway: influence of birth cohort and time period. Int J Cancer 1993; 53:183-187. 11. Harach H R , Franssila KO, Wasenius V-M. Occult papillary carcinoma of the thyroid: a “normal” finding in Finland-a systematic autopsy study. Cancer 1985; 56531-538. 12. Sampson RJ, Woolner LB, Bahn RC, Kurland LT. Occult thyroid carcinoma in Olmsted County, Minnesota: prevalence at autopsy compared with that in Hiroshima and Nagasak Japan. Cancer 1974; 342072-2076. 13. Wilson SD, Komorowski R, Cerletty J, Majewski JT, Hooper M. Radiation-associated thyroid tumors: extent of operation and pathology technique influence the apparent incidence of carcinoma. Surgery 1983; 94:663-667. 14. Robbins J. Thyroid cancer: a lethal endocrine neoplasm-NIH conference. Ann Intern Med 1991; 115:133-147. 15. Mettler FA, Upton AC. Carcinogenesis at specific sites. In Medical effects of ionizing radiation. Philadelphia: WB Saunders, 1995130-139. thyroid a 16. Aldinger K A , Samaan NA, Ibanez M, Hill CS. Anaplastic carcinoma of the reviewof 84 cases of spindle and giant cell carcinoma ofthethyroid.Cancer1978; 41~2267-2275. 17. Duffy BJ, Fitzgerald PJ. Cancer of the thyroid in children: a report of 28 cases. J Clin Endocrinol Metab 1950; 10:1296-1308. 18. Simpson C L , Hempelmann LH. The association of tumors and roentgen-ray treatmentof the thorax in infancy. Cancer 1957; 10:42-56. 19. Saenger EL, Silverman FN, Sterling TD, Turner ME. Neoplasia following therapeutic irradiation for benign conditions in childhood. Radiology 1960; 74889-904. 20. Thompson DE, Mabuchi K, Ron E, Soda M, Tokunaga M, Ochikubo S, Sugimoto S, Ikeda T, Terasaki M, IzumiS, Preston DL. Cancer incidence in atomic bomb survivors. Part 11: Solid tumors, 1958-1987. Radiat Res 1994; 137:S17-S67. 21. Parker LN, Belsky JL, Yamamoto T, Kawamoto S, Keehn RT. Thyroid carcinoma after exposure to atomic radiation: a continuing survey of a fixed population, Hiroshima and Nagasaki, 1958-1971. Ann Intern Med 1974; 80:600-604. 22.SchullWJ.Atomicbombsurvivors:patternsofcancerrisk.ProgCancerResTher 1984; 26:21-36. 23. Fjalling M, Tisell L E , Carlsson S , Hansson G, Lundberg L ” , OdenA. ’Benign and malignant thyroid nodules after neck irradiation. Cancer 1986; 58:1219-1224. 24. Furst CJ, Lundell M, Holm LE, Silfversward C. Cancer incidence after radiotherapy for skinhemangioma: aretrospectivecohortstudyinSweden.JNatlCancerInst1988; 80~1387-1392. 25. Lundell M, Hakulinen T, Holm L-E. Thyroid cancer after radiotherapy for skin hemang in infancy. Radiat Res 1994; 140:334-339. 26. Lindberg S, Karlsson P, Arvidsson B, Holmberg E, Lundberg LM, Wallgren A. Cancer incidence after radiotherapy for skin haemangioma during infancy. Acta Oncol 1995; 34~735-740. 27. Schneider A B , Ron E, Lubin J, Stovall M, Gierlowski TC. Dose-response relationships
108
Figge, Jennings, and Gerasimov
for radiation-induced thyroid cancer and nodules: evidence for the prolonged effects of radiation on the thyroid. J Clin Endocrinol Metab 1993;77:362-369. 28. Schneider AB, Recant W, Pinsky SM, RyoUY, Bekerman C, Shore-Freedman E. RadiaAnn tion-induced thyroid carcinoma: clinical course and results of therapy in 296 patients. Intern Med 1986; 105:405-412. 29. Viswanathan K, Gierlowski TC, SchneiderAB. Childhood thyroid cancer: characteristics and long-term outcome in children irradiated for benign conditions of the head and neck. Arch Pediatr Adolesc Med 1994; 148:260-265. 30. Schneider AB, Shore-Freedman E, Weinstein RA. Radiation-induced thyroid and other head and neck tumors: occurrence of multiple tumors and analysisof risk factors. J Clin Endocrinol Metab 1986; 63:107-112. 31. Paloyan E, Lawrence AM. Thyroid neoplasms after radiation therapy for adolescent acne vulgaris. Arch Dermatol 1978; 11453-55. RA, Ames WR. Neoplasms in persons 32. Hempelmann LH, Hall WJ, Phillips M, Cooper J NatlCancerInst1975; treatedwithx-rays in infancy:fourthsurveyin20years. 55:519-530. 33. Shore RE, Hildreth N, Dvoretsky P, Andresen E, Moseson M, Pasternack B. Thyroid in infancy for an enlarged thymus gland. Am cancer among persons given x-ray treatment J Epidemiol 1993; 137:1068-1080. JD. Thyroid neoplasia 34.Ron E, Modan B, Preston D, AlfandaryE,StovallM,Boice following low-dose radiation in childhood. Radiat Res 1989;120:516-531. 35. ShoreRE, Albert RE, Pasternack BS. Follow-up study of patients treated by x-ray epilati for tinea capitis: resurvey of post-treatment illness and mortality experience. Arch Env Health 1976; 31:17-24. 36. Ron E, Modan B. Thyroid and other neoplasms follwing childhood scalp irradiation. P Cancer Res Ther 1984; 26:139-151. 37. Tucker MA, Morris Jones PH, BoiceJD, Robison LL, Stone BJ, Stovall M, Jenkin RDT, Lubin JH, Baum ES, Siegal SE, Meadows AT, Hoover RN, Fraumeni JF. Therapeutic radiationatayoungage is linkedtosecondarythyroidcancer.CancerRes1991; 5 1:2885-2888. 38.HancockSL,CoxRS,McDougall IR. Thyroid diseases after treatment of Hodgkin’s disease. N Engl J Med 1991; 325:599-605. 39. Boice JD, Day N E , Andersen A, Brinton LA, Brown R, ChoiNW, Clarke EA, Coleman MP, Curtis, RE, Flannery JT, Hakama M, Hakulinen T, Howe GR, Jensen OM, Klein RA, Magnin D, Magnus K, Makela K, Malker B, Miller AB, Nelson N, Patterson CC, Pettersson F, Pompe-KimV, Primic-Zakelj M, Prior P, Ravnihar B, Skeet RG, Skjerven JE, SmithPG, Sok M, Spengler RF, Storm HH, Stovall M, Tomkins GWO,WallC. Second cancers following radiation treatment for cervical cancer: an international collaboration among cancer registries. JNCI 1985; 74:955-975. 40. Boice JD, Engholm G, Klienerman RA, Blettner M, Stovall M, Lisco H, Moloney WC, Austin DF, Bosch A, Cookfair DL, Krementz ET, LatouretteHB, Merrill JA, PetersLJ, Schulz MD, Storm HH, Bjorkholm E, Pettersson F, Bell CMJ, Coleman M P , Fraser P, Neal FE, Prior P, ChoiNW, Hislop TG, Coch M, Kreiger N, Robb D, Robson D, Thomson KE, Rimpela A, Pejovic MH, DH, Lochmuller H, von Fournier D, Frischkorn R, Kjorstad Kim W, StankusovaH, Berrino F, Sigurdsson K, Hutchison GB, MacMahon B. Radiat dose and second cancer risk in patients treated for cancer of the cervix. Radiat Res 1988; 116:3-55. 41. Arai T, Nakano T, F h h i s a K, Kasamatsu T, Tsunematsu R, Masubuchi K, Yamauchi K, Hamada T, Fukuda T, Noguchi H, Murata M. Second cancer after radiation therapy for cancer of the uterine cervix. Cancer 1991; 67:398-405. 42. Hay JH,Duncan W, Kerr GR. Subsequent malignanciesin patients irradiated for testicula tumours. Br J Radio1 1984; 57:597-602.
Radiation
43. Fossa SD, Langmark F, Aass N, Andersen A, Lothe R, Borresen AL. Second non-germ cell malignancies after radiotherapy of testicular cancer with or without chemotherapy. Br J Cancer 1990; 61:639-643. 44. Kendall GM, Muirhead CR, MacGibbon BH, O’Hagan JA, Conquest AJ, Goodill AA, Butland BK, Fell TP, JacksonDA,Webb MA, HaylockRGE,Thomas JM, Silk TJ. Mortality and occupational exposure to radiation: first analysis of the National Registry for Radiation Workers. BMJ 1992; 304:220-225. 45. Beral V, Inskip H, Fraser P, Booth M, Coleman D, Rose G. Mortality of employees of the United Kingdom Atomic Energy Authority, 1946-1979. BMJ 1985; 291:44W7. 46. Wilkinson GS, Tietjen GL, Wiggs LD, Galke WA, Acquavella JF, Reyes M, Voelz GL, Wazweiler RJ. Mortality among plutonium and other radiation workers at a plutonium weapons facility. Am J Epidemiol 1987; 125:231-250. 47. Checkoway H, Pearce N, Crawford-Brown DJ, Cragle DL. Radiation doses and causeJ Epidemiol specific mortality among workers at a nuclear materials fabrication Am plant. 1988;127:255-266. 48. Wing S, Shy C M , Wood JL, Wolf S, Cragle DL, Frome EL. Mortality among workers at Oak Ridge Nationaal Laboratory: evidence of radiation effects in follow-up through 1984 JAMA 1991; 265~1397-1402. 49. Wang J-X, Boice JD, Li B-X, Zhang J-Y, Fraumeni JF.Cancer among medical diagnostic x-ray workers in China. J Natl Cancer Inst 1988; 80344-350. 50. Boice JD, Mandel JS, DoodyMM, Yoder RC, McGowan R.A health survey of radiologic technologists. Cancer 1992; 69586-598. 51. Smith PG, Doll R. Mortality from cancer and all causes among British radiologists. Br J Radio1 1981; 54:187-194. 52. Matanoski GM, Seltser R, Sartwell PE, Diamond EL, Elliott EA. The current mortality rates of radiologists and other physician specialists: specific causes of death. Am J Epidemiol 1975; 101:199-210. 53. Forman D, Cook-Mozaffari P, Darby S, Davey G, Sratton I, Doll R, Pike M. Cancer near nuclear installations. Nature 1987; 329:499-505. 54. Jablon S, Hrubec Z, Boice JD. Cancer in populations living near nuclear facilities: a survey of mortality nationwide and incidence in two states. JAMA 1991; 2651403-1408. 55. Stewart A, Kneale GW. Radiation dose effects in relation to obstetric x-rays and childhood cancers. Lancet 1970; 1:1185-1188. 56. Mole RH. Childhood cancer after prenatal exposure to diagnostic x-ray examinations in Britain. Br J Cancer 1990; 62152-168. 57. Oppenheim BE, Griem ML, Meier P. Effects of low-dose prenatal irradiation in humans: analysis of Chicago Lying-In data and comparison with other studies. Radiat Res 1974; 57:508-544. 58. Harvey EB, BoiceJD, Honeyman M, Flannery JT. Prenatal x-ray exposure and childhood cancer in twins. N Engl J Med 1985; 312541445. 59. Rodvall Y, Pershagen G, Hrubec Z, Ahlbom A, Pedersen NL, Boice JD. Prenatal x-ray exposure and childhood cancer in Swedish twins. Int J Cancer 1990; 46:362-365. 60. Yoshimoto Y, Kat0 H, Schull WJ. Risk of cancer among children exposed in utero to A-bomb radiations, 1950-1984. Lancet 1988; 2:665-669. 61. Lee W,Chiacchierini RP, ShleienB,TellesNC.Thyroidtumorsfollowing1-131or localized X irradiationtothethyroidandpituitaryglandsinrats.RadiatRes1982; 92:307-319. 62.HolmL-E,Wiklund KE, LundellGE,BergmanNA,BjelkengrenG,Cederquist ES, Ericsson UBC, Larsson LG, Lidberg ME, Lindberg RS, WicklundH V , Boice JD. Thyroid cancer after diagnostic dosesof iodine-131: a retrospective cohort study. J Natl Cancer Inst 1988; 80:1132-1138. 63.RonE,Doody MM, BeckerDV,Brill AB, Curtis RE, Goldman MB, HarrisBS,
Gerasimov Figge, Jennings, and
110
Hoffman DA, McConahey WM, Maxon HR, Preston-Martin S, Warshaue ME, Wong FL, Boice JD. Cancer mortality following treatment for adult hyperthyroidism. JAMA 64.
1998;280~347-355.
Holm L-E. Malignant disease following iodine-131 therapy in Sweden. Prog Cancer Res Ther 1984;26:263-271. 65. Hoffman DA, McConahey W M , Fraumeni JF, Kurland LT. Cancer incidence following treatment of hyperthyroidism. Int J Epidemiol 1982; 11:218-224. 66. Holm L-E, Dahlqvist I, Israelsson A, Lundell G. Malignant thyroid tumors after iodi therapy: a retrospective cohort study.N Engl J Med 1980; 303:188-191. 67. Hoffman DA, McConahey WM, Diamond EL, Kurland LT. Mortality in women treated for hyperthyroidism. Am J Epidemiol 1982; 115:243-254. 68. Goldman MB, Maloof F, Monson RR, Aschengrau A, CooperDS, Ridgway EC. Radioactive iodine therapy and breast cancer: a follow-up study of hyperthyroid women. Am J Epidemiol 1988;127:969-980. in children 69. Safa AM, Schumacher OP, Rodriguez-Antunez A. Long-term follow-up results and adolescents treated with radioactive iodine (‘-131) for hyperthyroidism. N Engl J Med 1975;292:167-171. 70. Johnson CJ. Cancer incidence in an area of radioactive fallout downwind from the N test site. JAMA 1984; 2513230-236. 71. Rallison ML, Dobyns BM, Keating FR, Rall E,Tyler FH. Thyroid disease in children: Am J Med 1974; 56:457463. a survey of subjects potentially exposed to fallout radiation. 72. Machado SG,Land CE, McKayFW.Cancer mortality and radioactive fallout in southwe ern Utah.-Am J Epidemiol 1987;125:44-61. 73. Conard RA, Paglia DE, Larsen PR, SutowWW, Dobyns BM, Robbins J, Krotosky WA, Field JB, Rall JE, Wolff J. Review of medical findingsa Marshallese in population twenty six years after accidental exposure to radioactive fallout. Brookhaven Natl Lab 1980; Rep BNL 51261. 28 years ago. 74. Conard RA. Late radiation effects in Marshall Islanders exposed to fallout Prog Cancer Res Ther 1984;26:57-71. 75. Hamilton TE, van Belle G, LoGerfoJP. Thyroid neoplasia in Marshall Islanders expose to nuclear fallout. JAMA 1987; 258:629-636. 76. Wiklund K, Holm L-E, Eklund G. Cancer risks in Swedish Lapps who breed reindeer. Am J Epidemiol 1990; 132:1078-1082. 77. Shore RE. Human thyroid cancer induction by ionizing radiation: summary of studies In Karaoglou A, Desmet G, Kelly GN based on external irradiation and radioactive iodines. Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:669-675. 78. Shore RE.Issues and epidemiological evidence regarding radiation-induced thyroid Radiat Res 1992;131:98-111. 79. Ron E,Lubin JH, Shore RE, Mabuchi K, Modan B, Pottern LM, Schneider AB, Tucker MA, Boice JD.Thyroid cancer after exposure to external radiation: a pooled analysis of seven studies. Radiat Res 1995;141:259-277. 80. Shore RE, Woodard E, Hildreth N, Dvoretslq P, Hempelmann L, Pasternack B. Thyroid tumors following thymus irradiation. JNCI1985; 74:1177-1184. 8 1. National Academy of Sciences. Committee on the biological effects of ionizing rad health effects of exposure to low level of ionizing radiation (BEIRV). Washington, DC: National Academy Press, 1990. 82. Ron E, Lubin J, Schneider AB. Thyroid cancer incidence. Nature 1992; 360:113. 83. USSR State Committee on the Utilization of Atomic Energy. The accident at the C nuclear power plant and its consequences. Information compiled for the International Atomic Energy Agency Experts Meeting, August25-29, 1986, Vienna. Moscow: USSR State Committee on the Utilization of Atomic Energy, 1986.
Radiation
111
84. Travis J. Inside look confirms more radiation. Science1994; 263:750. 85. Wannan EA. Paper presentedat theNew York Chapter Health Physics Society symposium
on the effectsof the nuclear reactor accident at Chernobyl, Brookhaven National Laboratory, Upton, N Y , April 3, 1987. 86. Sich AR. Chernobyl thesis. Science 1994; 266:1627-1628. 87. Stone R. The explosions that shook the world. Science 1996; 272:352-354. 88. Sobotovich E,Bondarenko G, Petriaev E. Geochemistry of Chernobyl radionuclides. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:477-483. 89. Izrael YA, De Cort M, JonesAR,Nazarov IM,Fridman SD, Kvasnikova EV,Stukin ED, KellyGN,Matveenko 11, Pokumeiko YM, Tabatchnyi LY, Tsaturov Y. The atlas of caesium-137 contamination of Europe after the Chernobyl accident. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, eds. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996: 1-10. 90. Balonov M, Jacob P, LikhtarevI, Minenko V. Pathways, levels and trends of population exposure after the Chernobyl accident. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: Europea Commission, 1996:235-249. 91. Tsaturov YS, De Cort M, Dubois G, Izrael YA, Stukin ED, Fridman SD, Tabachnyi LY, Matveenko 11, Guermenchuk MG, Sitak VA. The need for standardization in the analysis, sampling and measurement of deposited radionuclides. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:425-433. 92. Kazakov VS, Demidchik EP, Astakhova LN. Thyroid cancer after Chernobyl. Nature
1992;359:21. 93. World Health Organization. International programme on the health effects of the Chem accident. Geneva: World Health Organization, 1993. 94. Stsjazhko VA, TsybAF, Tronko N D , Souchkevitch G, Baverstock KF. Childhood thyroid cancer since accident at Chernobyl.BMJ 1995; 310:801. 95. Williams N, Balter M. Chernobyl research becomes international growthindustry. Science 1996; 272~355-356. 96. Balter M. Children become the first victimsof fallout. Science 1996; 272:357-360. 97. Goldman M, Catlin R, Anspaugh L. Health and environmental consequences of the Ch
by1 nuclear power plant accident. Washington, DC: U.S. Department of Energy Report DOE/ER 0332, 1987. 98. Lange R, Dickerson MH, Gudiksen PH. Dose estimates from the Chernobyl accident. Nucl Techno1 1988; 82:311-323. 99. Nauman J, Wolff J. Iodine prophylaxis in Poland after the Chernobyl reactor accident: benefits and risks. Am J Med 1993; 94524-532. 100. Dubina W, Schekin YK, Guskina LN. Systematisation and verification of the spectromet131rical analysis data of soil, grass, milk and milk products samples with results of iodine measurements. IJn Russian]. Minsk. 1990. 101. Becker DV, RobbinsJ, Beebe GW, Bouville AC, Wachholz BW. Childhood thyroid cancer following the Chernobyl accident. Endocrinol Metab Clin North Am 1996; 25:197-211. 102. Mityukova T, Astakhova L, Asenchyk L, Orlov M, Van Middlesworth L. Urinary iodine excretion in Byelarus children. Eur J Endocrinol1995; 133:216-217. 103. GerasimovG,AlexandrovaG,ArbuzovaM,Butrova S , KenzhibaevaM,Kotova G, Mishchenko B, Nazarov A, Platonova N, Sviridenko N, Chernova T, Troshina E, Dedov I. Iodinedefficiencydisorders(IDD)inregionsofRussiaaffectedbyChernobyl.In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:813-815. 104. Gavrilin Y, Khrouch V, ShinkarevS , DrozdovitchV, Minenko V, Shemyakina E, Bouville
112
Figge, Jennings, and Gerasimw
A, Anspaugh L. Estimation of thyroid doses received by the population of Belarus as a result of the Chernobyl accident. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:lOll-1020. 105. Gavrilin YI, Gordeev KI, Ivanov VK, Ilyin LA, KondrusevAI, Margulis W, Stepanenko V F , Khrouch VT, Shinkarev SM. The process and results of the reconstruction of inte thyroid doses for the population of contaminated areas of the Republic of Belarus. [In Russian]. Vestn Acad Med Sci 1992; 2:35-43. 106. Ilyin LA, Balonov MI, Buldakov LA, Bur’yak V N , Gordeev KI, Dement’ev SI et al. of the accident at the Radiocontamination patterns and possible health consequences Chernobyl nuclear power station.J Radiol Rot 1990; 10:13-29. 107. Gavrilin YI, Gordeev KI, Ilyin LA et al. Results of thyroid dose assessment for contaminated territories of Belarussia. FRussian]. Bull Acad Med Sci USSR 1991; 8:35. 108. Gavrilin YI, Khrouch V T , Shinkarev SM. Internal thyroid exposure of the residents in several contaminated areas of Belarus. [In Russian]. J Med Radiol 1993; 6:15-20. 109. Khrouch VT, Gavrilin M,Shinkarev SM, MargulisW, Samokhin IV, Soldatenkov VI, Ivanova OE. Generalization of results of individual thyroid dose reconstruction: determ tion of connections between parameters of contamination of people residences and levels of irradiation on thyroid glands. [In Russian]. Final Report of Institute of Biophysics, Moscow Contract N 7-17/93 with the Ministry of Public Health, Minsk, Belarus. Moscow, 1994. 110. Stepanenko V, GavrilinY,Khrousch V, ShinkarevS, Zvonova I, Minenko V, Drozdovich V, Ulanovsky A, Heinemann K, Pomplun E, Hille r, Bailiff I, Kondrashov A, Yaskova E, Petin D, Skvortsov V, Parshkov E, Makarenkova I, Volkov V, Korneev S , Bratilova A, Kaidanovsky J.The reconstructionof thyroid dose following Chernobyl. In Karaoglo A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission,1996:937-948. 11 1. Tsyb A F , Parshkov EM, ShakhtarinW, StepanenkoVF, Skvortsov W, and Chebotareva IV. Thyroid cancer in children and adolescents of Bryansk and Kaluga regions. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:691-697. 112. Zvonova I, Balonov MI. Radioiodinedosimetryandprediction of thyroideffectson MI, inhabitants of Russia following the Chernobyl accident. In Merwin SE, Balonov editors. The Chernobyl papers, vol. I: Doses to the Soviet population and early health effects studies. Richland, WA. Research Enterprises, 1993:71-126. 113. Sobolev B, Likhtarev I, Kairo I,Tronko N,OleynikV,Bogdanova T. Radiation risk assessment of the thyroid cancer in Ukrainian children exposed due to Chernobyl. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission,1996:741-748. 114. Likhtarev IA, Sobolev BG, Kairo IA, Tronko N D , Bogdanova TI, Oleinic .VA, Epshtein EV, Beral V. Thyroid cancer in the Ukraine. Nature 1995; 375365. 115. Likhtarev IA, Shandala NK, Gulko GM, Kairo IA, Chepurny NI. Ukrainian thyroid doses after the Chernobyl accident. Health Phys1993; 64594-599. 116. Likhtarev I, Sobolev B, Kairo I, Tabachny L, Jacob P, Prtjhl G, Goulko G. Results of in Ukraine. In Karaoglou A, Desmet G, Kelly GN, large scale thyroid dose reconstruction Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels European Commission, 1996: 1021-1034. 117. Likhtarev IA, Gulko GM, Sobolev BG, Kairo IA, Chepurnoy M, Pr6hl G, Henrichs K. Thyroid dose assessment for the Cherginov region (Ukraine): estimation based on 13’1 thyroid measurements and extrapolation of the results to districts without monitoring. Radiat Environ Biophys 1994; 33:149-166.
Radiation
113
118. Prisyazhiuk A, Pjatak OA, Buzanov VA, Reeves GK, Beral V. Cancer in the Ukraine, post-Chernobyl. Lancet 1991; 338:1334-1335. 119. Oleynic VA, Cheban AK. Thyroid cancer in children of Ukraine from 1981 to 1992. In of a workshop Robbins J, editor. Treatmentof Thyroid Cancer in Childhood. Proceedings No. DOE/ on September 10-11, 1992, at the National Institutes of Health. Publication EH-0406 1992; 35. 120. Astakhova LN. Condition of the thyroid system and peculiarity of forming its pathology in the BSSR population under influence of the iodine-radionuclides in connection with Chernobyl nuclear accident.IJn Russian]. Zdravoohranenie Belorussi1990; 6:ll-15. 121. Astakhova LN, Demidchuk EP, Davydova EV, Arinchin AN, Gres NA, Zelenko SM, Drozd V D , Poliakova TI, Dardynskaia IV,Bazyl’chik SV, et al. Health status of Byelorusas consequenceof the Chernobyl atomic sian children and adolescents exposed to radiation energy station accident. [In Russian]. Vestn Akad Med Nauk USSR 1991; 11:25-27. 122. Okeanov A E , Averkin YI. Analysis of malignant neoplasms in population of the Republic of Belarus before and after the Chernobyl accident. In Matyukhin VA, Astakhova LN, Konigsberg YE, Nalivko SN, editors. Catastrophe at the Chernobyl atomic energy station and estimationof health stateof population of the Republic of Belarus. Minsk: Research Institute of Radiation Medicine, 1991:25-33. 123. Astakhova LN, Vorontsova TV, Drozd VM. Thyroid nodule pathology in children of Belarus following the Chernobyl accident. In Robbins J(ed.). Treatmentof Thyroid Cancer in Childhood. Proceedings of a workshop on September 10-11, 1992, at the National Institutes of Health. Publication No. DOEEH-0406 1992; 35. 124. Baverstock K, Egloff B, Pinchera A, Ruchti C, Williams D. Thyroid cancer after Che Nature 1992; 359:21-22. 125. Demidchik EP, Kazakov VS, Astakhova LN, et al. Thyroid cancer in children after the Chernobyl accident: clinical and epidemiological evaluation of 251 cases in the Republic of Belarus. In NagatakiS, editor. Nagasaki symposium on Chernobyl: update and future. Amsterdam: Elsevier, 199421. 126. Demidchik EP, Drobyshevskaya IM, Cherstvoy ED, Astakhova LN, Okeanov AE, Vorontsova TV, Germenchuk M. Thyroid cancerin children in Belarus. In Karaoglou A, Desmet of theChernobyl G,KellyGN,MenzelHG,editors.Theradiologicalconsequences accident. Brussels: European Commission, 1996:677-682. B, Furmanchuk AW, GurtnerF, Korotkevich JA, 127. Abelin T, Averkin JI, Egger M, Egloff Marx A, Matveyenko 11, Okeanov AE, et al. Thyroid cancer in Belarus post-Chernobyl: improved detection or increased incidence? Soz Praventivmed1994; 39:189-197. 128. Averkin JI, Abelin T, Bleuer JP. Thyroid cancer in children in Belarus: ascertainment bias? Lancet 1995; 346:1223-1224. 129. Abelin T, Egger M, RuchtiC. Belarus increase was probably caused by Chernobyl. BMJ 1994; 309:1298. 130. Williams ED, Cherstvoy E, EgloffB, Hofler H, Vecchio G, Bogdanova T, Bragamik M, Tronko ND. Interaction of pathology and molecular characterization of thyroid cancers. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequen of the Chernobyl accident. Brussels: European Commission,1996:699-714. 131. Abelin T,Averkin JI, Okeanov AE, Bleuer JP.Thyroid cancer in Belarus: the epidemiologic situation. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological 1996:727-730. consequencesof the Chernobyl accident. Brussels: European Commission, 132. Tsyb A F , ParshkovEM,Ivanov V K , Stepanenko VF, Matveenko EG, Skoropad YD. Disease indicesof thyroid and their dose dependence in children and adolescents affected as a result of the Chernobyl accident. In Nagataki S, editor. Amsterdam: Elsevier Science, 1994:9-19. 133. Remennik LV, Starinsky W, Mokina VD, Chissov VI, Scheplyagina LA, Petrova GV,
114
Gerasimov Figge, Jennings, and Rubtsova MM. Malignant neoplasms on the territories of Russia damaged owing to the Chernobyl accident. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission,
1996~825-828. 134. TronkoN,BogdanovaT,KomissarenkoI,BolshovaE,OleynikV,TereshchenkoV,
Epshtein Y, Chebotarev V. Thyroid cancer in children and adolescents in Ukraine after the Chernobyl accident. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. Th radiological consequences of the Chernobyl accident. Brussels: European Commission,
1996:683-690. 135. Tronko N, Epstein Y,Oleinik V, et al. Thyroid gland in children after the Chernobyl In Nagataki S , editor. Nagasaki symposium on Chernoby accident (yesterday and today). update and future. Amsterdam: Elsevier,1994:31-46. 136. Williams ED, Tronko ND. Molecular, cellular, biological characterization of childhood thyroid cancer. Brussels: European Commission. 1996. 137. Cherstvoy E, Pozcharskaya V, HarachHR, Thomas GA, Williams ED. The pathology o
childhood thyroid carcinoma in Belarus.In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: Europ Commission, 1996:779-784. 138. Bogdanova T, Bragarnik M, Tronko N D , Harach HR, Thomas GA, William ED. The pathologyofthyroidcancerinUkrainepostChernobyl. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl acc Brussels: European commission, 1996:785-789. 139. Abrosimov AY, Lushnikov EF, Tsyb AF, Harach HR, Thomas GA, Williams ED. The pathology of childhood thyroid tumours in the Russian Federation after Chernobyl. In Karaoglou A, Desmet G, Kelly GN, Menzel HG, editors. The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996:791-793. 140. Furmanchuk AW, Averkin JI, Egloff B, Ruchti C, Abelin T, Schappi W, Korotkevich EA. Pathomorphological findings in thyroid cancers of children from the Republic of 86cases occurring between 1986(“post-Chernobyl”) and1991. HistopaBelarus: a study of thology 1992; 21:401408. 141. Nikiforov Y, Gnepp DR. Pediatric thyroid cancer after the Chernobyl disaster. Cancer
1994; 74748-766. 142. Williams ED. Thyroid cancer in United Kingdom children and in children exposed out from Chernobyl. In Nagataki S , editor. Nagasaki symposium on Chernobyl: update and future. Tokyo: Elsevier, 1994. 143. Nikiforov Y, Gnepp DR, Fagin JA. Thyroid lesions in children and adolescents after the
Chernobyl disaster: implications for the study of radiation tumorigenesis. J Clin E Metab 1996;81:9-14. 144. Pacini F, Vorontsova T, Demidchik EP, Delange F, Reiners C, Schlumberger M, Pin A. Diagnosis, surgical treatment and follow-up of thyroid cancers.In Karaoglou A, Desme G,KellyGN,MenzelHG,editors.TheradiologicalconsequencesoftheChernobyl accident. Brussels: European Commission, 1996:755-763. 145. NikiforovYE, Heffess CS,Korzenko AV, Fagin JA, Gnepp DR. Characteristics of f tumors and nonneoplastic thyroid lesions in children and adolescents exposed to ra as a resultof the Chernobyl disaster. Cancer 1995; 76:900-909. 146. Sierk A E , Askin FB, Reddick RL, Thomas CG. Pediatric thyroid cancer. Pediatr Path01
1990;10:877-893. 147. Harness JK, Thompson N W , Nishiyama RH. Childhood thyroid carcinoma. Arch Surg 1971;102:278-284. 148. Richardson JE, Beaugie JM, Brown CL, Doniach I. Thyroid cancer in young patients in Great Britain. Br J Surg 1974; 61:85-89.
Cancer Radiation and Thyroid
115
149. Tallroth E, Backdahl M, Einhorn J, Lundell G, Lowhagen T, Silfversward C. Thyroid carcinoma in children and adolescents. Cancer 1986; 58:2329-2332. 150. Mizukami Y,Michigishi T, Nonomura A, HashimotoT, Noguchi M, Matsubara F, Watanabe K. Carcinoma of the thyroid at young age: a review of 23 patients. Histopathology 1992; 20~63-66. 151. Robbins J. Characteristics of spontaneous and radiation induced thyroid cancers in child In NagatakiS, editor. Nagasaki symposium on Chernobyl: update and future. Amsterdam: Elsevier, 199491. 152. Williams ED. Radiation-induced thyroid cancer. Histopathology 1993; 23:387-389. 153. Willams ED, Chernobyl, eight years on. Nature1994; 371556. 154. Williams D. Thyroid cancer and the Chernobyl accident. [Editorial]. J Clin Endocrinol Metab 1996;81:6-8. 155. Ito T, SeyamaT, Iwamoto KS, Mizuno T, TronkoN D , Komissarenko IV, et al. Activated
RET oncogene in thyroid cancers of children from areas contaminated by Chernobyl accident. Lancet 1994; 344:259. 156. Fugazzola L,Pilotti S, Pinchera A, Vorontsova TV, Mondellini P, Bonganone I, et al. Oncogenic rearrangements of the RET proto-oncogene in papillary thyroid carcinomas 1995; 555617-5620. children exposed to the Chernobyl nuclear accident. Cancer Research 157. Nikiforov YE, Rowland JM, Bove KE, Monforte-Munoz H, Fagin JA. Distinct patternof ret oncogene rearrangementsin morphological variants of radiation-induced and sporadic thyroid papillary carcinomas in children. Cancer Res 1997; 57:1690-1694. 158. Klugbauer S, Lengfelder E, Demidchik EP, Rabes HM. High prevalence of RET rearrangement in thyroid tumors of children from Belarus after the Chernobyl reactor accident. Oncogene 1995;11:2459-2467. 159. Rabes HM, Klugbauer S. Radiation-induced thyroid carcinomas in children: high prevalence of RET rearrangement [German]. Verh Dtsch Ges Path01 1997; 139-144. HM. A new form of RET rearrangement 160. Klugbauer S, Lengfelder E, Demidchik EP, Rabes in thyroid carcinomas of children after the Chernobyl reactor accident. Oncogene 1996; 13: 1099-1 102. 161. Klugbauer S, Demidchik EP, Lengfelder E, Rabes HM. Molecular analysis of new subtypes
of ELERET rearrangements,theirreciprocaltranscriptsandbreakpointsinpapillary thyroid carcinomas of children after Chernobyl. Oncogene 1997; 16:671-675. 162. Fugazzola L, Pierotti M, Vigano E, Pacini F, Verontsova T, Bonganone I. Molecular and biochemical analysis ofRETPTC4, a novel oncogeneic rearrangement between RET and ELEl genes, in a post-Chernobyl papillary thyroid cancer. Oncogene 1996; 13:10931097. 163. Klugbauer S, Demidchik EP, LengfelderE, Rabes HM. Detection of a novel type of RET
of the involved rearrangement (PTCs) in thyroid carcinomas after Chernobyl and analysis RET-fused gene RFG5. Cancer Research 1998; 58:198-203. 164. Pisarchik AV, Ermak G, Fomicheva V, Kartel NA, Figge J. TheretPTC1 rearrangement is a common featureof Chernobyl-associated papillary thyroid carcinomas from Belarus. Thyroid 1998;8:133-139. 165. PisarchikAV,ErmakG,DemidchikEP,MikhalevichLS,KartelNA,FiggeJ.Low prevalence of the redPTC3rl rearrangement in a series of papillary thyroid carcinomas presenting in Belarus ten year post-Chernobyl. Thyroid 1998; 8:1003-1008. 166. Smida J, Salassidis K, Hieber L, Zitzelsberger H, Kellerer A, Demidchik EP, et al. Distinct frequency of ret rearrangements in papillary thyroid carcinomas of children and adults from Belarus. Int J Cancer 1999; 80: 32-38. 167. It0 T, Seyama T, Iwamoto KS, HayashiT, Mizuno T, Tsuyama N, et al. In vitro irradiation is able to cause RET oncogene rearrangement. Cancer Research 1993; 5332940-2943. 168. Bounacer A, Wicker R, Caillou B, Cailleux AF, Sarasin A, Schlumberger M, Suarez HG.
116
Figge, Jennings, and Gerasimov
High prevalence of activatingret proto-oncogene rearrangements,in thyroid tumors from patients who had received external radiation. Oncogene 1997; 15:1263-1273. 169. Nikiforov YE, Nikiforova MN, Gnepp DR, Fagin JA. Prevalenceof mutations of ras and p53 in benign and malignant thyroid tumors from children exposed to radiation after the chernobyl nuclear accident. Oncogene 1996; 13:687-693. 170. Hillebrandt S , Streffer C, ReinersC, Demidchik E. Mutationsin the p53 tumor suppressor gene in thyroid tumors of children from areas contaminated by the Chernobyl accident. Int J Radiat Biol 1996; 69:39-45. 171. Hillebrandt S, Streffer C, Demidchik EP, Biko3, Reiners C. Polymorphisms in the p53 areas in Belarus. Mutation gene in thyroid tumors and blood samples of children from Research 1997; 381:201-207. 172. Smida J, Zitzelsberger H, Kellerer AM, Lehmann L, Minkus G, Negele T, et al. P53 mutations in childhood thyroid tumors from Belarus and in thyroid tumors without history. Int J Cancer 1997; 732302-807. 173. Suchy B, Waldmann V, Klugbauer S , Rabes I". Absence of RAS and p53 mutations in thyroid carcinomasof children after Chernobyl in contrast to adult thyroid tumors. British J Cancer 1998; 77: 952-955. 174. Beral V, Reeves G. Childhood thyroid cancer in Belarus. Nature 1992; 359:680-681. 175. Shigematsu I, Thiessen JW. Childhood thyroid cancer in Belarus. Nature 1992; 359:681. 176. Beebe GW. Epidemiologic studies of thyroid cancer in the CIS. In KaraoglouA, Desmet (eds.).The radiological consequences of the Chernobyl accident. G, Kelly GN, Menzel HG Brussels: European Commission, 1996: 731-740. 177. Cardis E, OkeanovAE. What is feasible and desirable in the epidemiologic follow-upof Chernobyl. In Karaoglou A, Desmet G, Kelly GN, Menzel HG (eds.). The radiological consequences of the Chernobyl accident. Brussels: European Commission, 1996: 835-8
9 Classification of Thyroid Malignancies James Oertel and Yolanda Oertel We follow the WHO Histological Classification of Thyroid Tumors( I ) and that of the AFIP Atlas of Tumor Pathology (2) as follows: PRIMARY MALIGNANT TUMORS Malignant tumors of follicular cells Follicular carcinoma Papillary carcinoma Poorly differentiated carcinoma Undifferentiated (anaplastic) carcinoma
Malignant tumors of C cells Medullary carcinoma Malignant tumors of mixed follicular and C cells
Miscellaneous epithelial tumors Squamous cell carcinoma, adenosquamous carcinoma, mucin producing carcinom Hyalinizing trabecular neoplasms (predominantly adenomas) Neoplasms associated with familial intestinal adenomatous polyposis Mucoepidermoid carcinoma Thymic and related neoplasms Teratomas Malignant nonepithelial tumors Malignant lymphoma Sarcomas SECONDARY TUMORS Metastatic melanoma Metastatic renal cell carcinoma Metastatic mammary carcinoma Metastatic pulmonary carcinoma
From: Thyroid Cancer:A Comprehensive Guideto Clinical Management Edited by: L. Warfofiky0 Humann Press Inc., Totma, NJ
127
118
OerteZ and Oertel
GENERAL COMMENTS
Many thyroid cancers arise in essentially normal thyroid tissue. Most grow slowly and are amenable to appropriate treatment. The majority are papillary cancers, espe in those parts of theworldinwhichadequateiodidesarepresentinthedietand the environment. Proper handling of the tissues is essential to produce good histological sections for accuratediagnosis.Incompletefixationofanythyroidtissuemayproducelossof cellular details and pale nuclei in the sections (thus a superficial resemblance to the nuclei of papillary carcinoma). The pathologist must provide the following: weight of the specimen, exact size of of a capsule the neoplasm,its relation to the bordersof the thyroid gland, the presence around the tumor, whether the neoplastic cells extend directly beyond the border of the thyroid (and if so, which tissues are involved), histological diagnosis of the tumor (including the histologic patterns and mitotic activity),and whether any lymph nodes in the specimen contain metastatic tumor. Also needed are brief descriptions of the so, nonneoplastic thyroid parenchyma and any other thyroid tumors present, and if their histological diagnoses. The results of any special procedures (e.g., immunoperoxidase stains, analysis of nuclear ploidy, in situ hybridization of nucleic acids) should be provided, if these are available at reasonable cost. Critical assessment of any special laboratory procedures is essential if these are given much weight in the pathologic interpretation. The reagents must be of hig the technical assistanceskillful, and all of the personnel involved must be experienced
SPECIAL STUDIES
Evaluation of nuclear DNA content by flow cytometryor imaging photometry suggests that aneuploidy in the differentiated carcinomas may have adverse effects on (3-5). survival in patients who do not have metastases at the time of initial diagnosis Apparently aneupoidy does not have diagnostic significance. Studies of H-ras, K-ras, and N-ras protooncogennes demonstrate mutations in some follicular carcinomas as opposed to adenomas (6).N-ras mutation in papillary carcinoma increased the chance of death (7). The p53 protein is a tumor suppressor substance with a rapid turnover. If it is inactivated or present in a mutant form, it accumulates in the nuclei. Detecting this protein suggests a loss of differentiation,isand associated with unfavorable prognosisfactors. It hasbeenfoundinnumerousexamplesofpoorlydifferentiatedand undifferentiatedcarcinomas,incontrastto its absence or infrequentoccurrencein differentiated carcinomas (8-10). Conversely, bcl-2 expression is rare in undifferentiated carcinoma, butis common in well-differentiated carcinoma and poorly differentiated carcinoma (11). REFERENCES 1. Hedinger C, Williams ED,Sobin LH. Histological typingof thyroid tumours. World Health Organization International Histological Classificationof Tumours, 2nded. Berlin: SpringerVerlag,1988. 2. Rosai J, Carcangiu ML, DeLellis RA. Tumors of the thyroid gland. In Washington, DC: A.F.I.P, 1992. Rosai J. Sobin H, editors. Atlas of tumor pathology, 3rd Ser, Fasc 5.
Classification
119
3. Hay ID. Papillary thyroid carcinoma. Endocrinol Metab Clin North Am 1990; 19:545-576. 4. Pasieka L, Zedenius J, Auer G, et al. Addition of nuclear DNA content to theAMES riskgroup classification for papillary thyroid cancer. Surgery 1992; 112:1154-1160. M. Overexpression of p53 protein 5. Nishida T, NakaoK, Hamaji M, Nakahara M, Tsujimoto andDNAcontent areimportantbiologicprognosticfactorsforthyroidcancer.1996; 119~568-575. 6. SciacchitanoS, Paliotta DS, Nardi F, Sacchi A, Andreoli M, Pontecorvi A. PCR amplifica and analysis of RAS oncogenes from thyroid cytologic smears. Diagn Mol Path01 1994; 3~114-121. 7. Hara H, Fulton N,Yashiro T, It0 K, DeGroot LJ,Kaplan EL.N-Ras mutation: an independent prognosticfactorforaggressiveness of papillarythyroidcarcinoma.Surgery1994; 116:lOlO-1016. HP. High prevalenceof 8. Fagin JA, Matsuo K, Karmaker A, Chen DL, Tang S-H, Koeffler mutations of the p53 gene in poorly differentiated human thyroid carcinomas. J Clin Invest 1993; 91~179-184. MA. Gene p53 mutations 9. Donghi R, Longoni A,Pilotti S, Michieli P, Della Porta G, pierotti are restricted to poorly differentiated and undifferentiated carcinomas of the thyroid gland. J Clin Invest 1993; 91:1753-1760. 10. Soares P, Cameselle-Teijeiro J, Sobrinho-Simoes M. Immunohistochemical detection of p53 in differentiated, poorly differentiated and undifferentiated carcinomas of the thyroid. Histopathology 1994; 24205-210. 11.Pilotti S, ColliniP,Del Bo R, Cattoretti G, Pierotti MA, Rike F. Anovelpanel of antibodies that segregates immunocytochemically poorly differentiated carcinoma from undifferentiated carcinomaof the thyroid gland. Am J Surg Path01 1994; 18:1054-1064.
10 Thyroid Cancer in Children and Adolescents Merrily Poth GENERAL CONSIDERATIONS Although thyroid cancer is less common in (:hildren 1than in adults, approximately 10% of all cases of thyroid cancer are diagnosed before the age of 21 (I). Thyroid cancer is the most common endocrine tumor in children. It comprises 0.5-1.5% of childhood tumors and is the most common malignant tumor of the head and neck in young people (2). In addition, even though thyroid canceris not common in children and adolescents, a mass identified in the thyroid before age 21 is much more likely to be malignant than when the same finding occurs in an older patient (3). The disease also tends to be more advanced at diagnosis with local and even distant metastasis, it and continues to behave more aggressively with more frequent recurrence (45). In spite of this aggressive behavior, when thyroid cancer in children is appropriately treated, it has an excellent prognosis. Thus,it is important for those caring for children to have adequate understanding of its presentation so that valuable timeis not lost before evaluation can occur and treatment can begin. EPIDEMIOLOGY AND GENERAL RISK FACTORS The incidence of thyroid cancer varies from 1 to 6.0 per 100,000 in individuals under the age of 21 (6,7). While in some reported series this incidence appears to be increasing over time (8,9), in other studies it is found to be relatively constant, with the exceptionof episodic increases in specific geographic areas, usually associated with exposure to radiation (7,10,11). Careful and comprehensive epidemiological studies of the potential factors influencing the incidence of thyroid cancer in children are not available. The sparse available data examining the incidence in racial groups appear to show a greater incidence of papillary cancer in Caucasians than in blacks and with either an equal or an increased incidence of follicular tumors in blacks (12). The incidence of thyroid cancer is greater in females than in males, in children as it is in adults, with ratios reported ranging from 2.5-6.0 to 1 (12,13). However, this relatively greater prevalence is seen only in older children and adolescents. In very
From: Thyroid Cancer: A Comprehensive Guide to clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, NJ
222
122
Poth
young children the gender ratiois closer to unity. The interaction of other risk factor particularly radiation, with genderis not clear. One investigator reported an extremely increased risk in male children treated with radiation for Hodgkin’s disease (14), while others have not found such a difference in the relative risk after radiation between male and female children (11,15). In general, there is a relatively constant and low incidence of thyroid cancer in young children, with steady increase in the incidence beginning at the time of puberty (16,17). Analysis of the effect of iodine content of diet on thyroid cancer has shown an increased incidence of papillary cancer in areas where iodine intake is high and of follicular lesions where iodine intake is low (18). However, these data have not been separately analyzed for children. of other thyroid disease, particularly Hashimoto’s thyroid The possible interaction on the incidenceof thyroid cancer is somewhat controversial. Some studies imply tha the incidence of malignancy in the presence of thyroiditis is increased (19,21), while other authors flatly refute such an association. The topic is complicated by the general failure of those writing about the potential association to clearly define their criteria (22). Recent studies examining pathologica for the diagnosis of Hashimoto’s thyroiditis specimens of thyroid cancer for the presence of lymphocytic infiltration and correlatin this with prognosis(23) add to the uncertainty aroundthis issue. There are no separate reports of pediatric patients analyzing the questions of the effect of lymphocytes in of potential effects of thyroiditis on the incidence of thyroid c tumor specimens or even though autoimmune thyroid disease is common in this age group. The syndrome of familial adenomatous polyposis or Gardner syndrome,is associated this may present in adolescence with an increased risk of papillary thyroid cancer, and (24). There are other cases of reported familial thyroid cancer. Some of these are associated with other syndromes (25), while others appear to represent an isolated propensity for papillary cancer (26). The vast majority of these patients present for diagnosis in young adulthood.It would seem logical to institute increased surveillanc during adolescence for individuals identified as at risk based on family history. RADIATION AS A RISK FACTOR
Of all the potential risk factors for the development of thyroid cancer, by far the most important is exposure to ionizing radiation. It is such a portentous issue that, as a matter of policy, all children with a history of significant exposure to radiation, including radiation therapy for malignancy, should be prospectively monitored for thyroid dysfunction and for the occurrence of thyroid nodules and cancer. While radiation exposure is clearly a risk factor for the development of thyroid cancer in adultsas well as in children, the effects are exaggerated in children. Childr show both an increased sensitivityto radiation effects, with increases in the occurren of thyroid cancer after even relatively small doses of radiation, and a pronounced of thyroid decrease in latency in the time between radiation exposure and the occurrence neoplasm. The younger the child at the time of radiation exposure, the more vu he or she appears to be to radiation effects. There are multiple populations where th relationships have been shown.
r
Thyroid
in Children
123
The first of such studies on the relationship of radiation and thyroid cancer in chil was that ofD u m and Fitzgerald(27) in 1950, who reported on 28 children with thyroid cancer, of whom 10 had received radiation treatment for an enlarged thymus. Winship and Rosvoll (28), in a 1961 review offindingson 562 cases of thyroid cancer in children, found that almost 80% reported previous irradiation for enlarged thymus, hypertrophied tonsils and adenoids, nevi, or angiomas. Further analysis of the effects of the radiation therapy for benign conditions continued to be published by a number of investigators and all reports confirmed the association. A study analyzing the effects of therapeutic radiation given for ringworm of the scalp compared 10,834 irradiated persons, using 5392 siblings as controls (29). They found a relative risk for thyroid cancer of 4.0, and an excess risk of 1.2 per 10,000 persons per year. They reported a linear dose-response curve with an average absolute excess risk of 12.5 person-years per cGy. They also found the risk for children irradiated before 5 years of age to be increasedwhencomparedtochildrenirradiatedatolderages. A study of infants irradiated for skin hemangiomas(30) found an excess relative risk of 7.5 per Gy, and a report on infants irradiated for enlarged thymus found an excess relative (31).risk of A comprehensive paper by Ron and colleagues in 1995 (32) summarized the data, adding a study of atomic bomb exposure, a childhood cancer study, and two different studies of children irradiated for enlarged tonsils and adenoids, to the studies of tinea capitusandthethymusirradiationthataredescribedabove.Theirsummarynoted excess relative risks, which varied from 1.1 to 32.5 per Gy with excess absolute risks of 2.6 to 7.6. They reported a linear relationship between radiation dose and risk, with 0.1 Gy. They also emphasized that the increased increased risk even with doses asaslow risk continues for up to 40 years after irradiation. Children no longer receive radiation therapy for treatment of benign disease; however, treatment for childhood cancer often includes radiation therapy. As survival of patients with childhood cancer continues to improve, there is a large population of surviving patients who have sustained significant radiation exposure to the thyroid. Based on the previous experience with thyroid cancer after radiation for benign conditions, it would be expected that these patients would also be at risk for thyroid cancer. This question is discussed briefly in the paper by Ron’s group(32), and numerous other studies involving patients receiving radiation therapy for malignant conditions have confirmed this relationship. One study, which included a group of patients with a variety of childhood tumors, reported a 4.6% incidence of thyroid cancer, after a mean follow-up period of only11 years (33). Another group reported on 9170 patients who survived childhood cancer for at least 2 years, and found that the risk of thyroid cancer increased by a factor of 53 (34). This study found the risk highest in patients treated for neuroblastoma and Wilms tumors, and the authors attributed this to the fact that these were the patients who were the youngest when they received their radiation (35) therapy. A study of patients who received radiation therapy for Hodgkin’s disease found, in addition to thyroid nodules and cancers, a very high incidence of autoimmune thyroid disease and hypothyroidism. Based this on report, which indicated hypothyroidism frequently occurred after a dose of 30 Gy to the gland, it seemed appropriate to followupsuchpatientsprospectively. This follow-upshouldprobablyinclude, in addition to careful monitoring for nodules, adding replacement thyroid hormone as
124
Poth
soon as TSH levels were noted to increase. Another group reporting on patients with Hodgkin’s lymphoma after radiation therapy recommended yearly ultrasound eval to help in the earliest identification of nodules and malignancy(36). The newest and most devastating data on radiation and thyroid cancer have been reported following the 1986 tragic accident in Chernobyl, where an unprecedented40 million to 50 million Ci of 13’1 was released into the atmosphere (37). The resulting radioactive material was widely dispersed, entered the food chain, and was ingested by both children and adults. The subsequent impressive increase in childhood thyroid cancer has been widely reported. Early attempts to attribute even of thispart “epidemic” to increased surveillance, have been effectively rebutted (38,39). The occurrence of cancer has been shown to be correlated with distance from the event (40,41) and to show an unexpectedly short latency 3ofyears. Risk of thyroid cancer was much high in younger children, and the largest proportion of affected children were those who were less than 1 yr of age at the time of their exposure; more than 55% of the thyroid (40,41). cancers were reported in children who were less4 years than of age at exposure Essentially all the thyroid tumorsin this population were papillary carcinomas, and the female : male ratio of affected children reported was 1.15:1. The behavior of the tumors has been consistent with thyroid cancer in other (nonirradiated) children with thyroid cancer, with a high percentage of tumors showing capsular invasion at the of diagnosis and most tumors presenting with lymph node metastases (41). It also seems clear that the full magnitude of this event has yet to be felt and that continued surveillance will lead to the continued confirmation of increased numbers of lesions. l3II is commonly used for both diagnostic studies of thyroid function and for t of hyperthyroidism. There have been concerns about whether such exposure, pa in children, might be associated with an increase in the development of thyroid c However, several studies of I3’Iused for diagnostic studies or for treatment of thyroid disease have failed to show an increase in malignancy afterthis exposure (42,43). With all these datain mind, it might be logical to worry that in very young childre even “diagnostic” levels of radiation, such as might be experienced by infants with chronic lung or congenital heart disease, might be associated with an increase in the of large groupsof children incidence of thyroid cancer. There are no published studies followed for the subsequent development of thyroid cancer after extensive doses of radiation for diagnostic studies. However, there are case reports of thyroid cancer in such children, suggesting a possible relationship (44). The search for genetic markers in tumors of patients developing thyroid cancer a radiation exposure has been fruitful. A 1996 paper reported that 4 of 22 patients in p53 gene mutations whomthyroidcancer developed after childhood radiation had 18 thyroid cancer patients without radiation exposure (45). Another compared to none of study found no p53 mutations in 15 thyroid cancer specimens obtained from children (46). Recent attemptsto look for such markers in thyroid cancers following the Chern RET oncogene rearrangements in these tumor by1 accident are ongoing, but activating have been reported by two different research groups (47,48). These studies did not include a control group of thyroid cancers from patients without radiation exposure. Thus the importance of genetic mutations in thyroid cancers occurring after radiation exposure remains under active investigation.
Children Thyroid Cancer in
125
CLINICAL PRESENTATION OF THYROID CANCER IN CHILDREN
Both the presentation and behavior of thyroid cancer in children differs somewhat from that of adults. The extent of disease at diagnosis is often greater than in adults, and the disease often persists or recurs after initial treatment.In spite of this the longterm prognosis for eventual cure is excellent, and mortality from disease is low with appropriate follow-up and treatment. In children the incidence of local invasion oftumororspreadtolymphnodes approaches 90%, and more than half the cases of papillary cancer present as a neck (49-54). The importance of considering mass or without a palpable thyroid lesion thyroid cancer in the diagnosis of neck mass in children can scarcely be overempha This neck mass or solitary thyroid nodule often is observed on a routine physical examination for school or participation in a sports program. Alternatively, the patient or a family member may detect a mass and request an evaluation. Other symptoms, such as dysphagia, hoarseness,or pain, are rarely noted in children with thyroid cancer at presentation. Since the disease is relatively rare, there are no systematic prospective studies of the effects of specific approaches to treatment and the effects on either the disease or on morbidity. There are several long-term retrospective studies underway using a large clinical database and molecular biology techniques performed on tissue blocks to try to develop better ways to predict the relative aggressiveness of individual tumors and to examine the outcomes of therapy in a more systematic(55). wayCurrent approaches to therapy along with data regarding both long- and short-term issues of each thera approach are discussed below. PATHOLOGICAL DIAGNOSES The distribution of pathological types of thyroid cancer in children does not differ markedly from that in younger adults. The most common form seen is papillary, which composes70-90%of all thyroid cancers in this age group, with follicular cancers making up most of the rest (50-57). The larger numbers of tumors characterized as papillary in more recent series are a result of the change in classification to include all of the former “follicular variant of papillary” tumors as papillary. Upon analysis, all of the tumors with some papillary characteristics are felt to behave as papillary cancers, leading to the newer classification system. Anaplastic cancer is extremely rare in children and when it occurs has the same poor prognosis asit does in older patients. Fortunately, anaplastic tumors make up less in children, than 1%of the total in all reported series. Medullary thyroid cancer occurs, MEN syndromes; this disease is considered in great as in adults, in association with the detail in Chapter 42-47. REFERENCES 1. Buckwalter JA, Gurll NJ, Thomas Jr. CG. Cancer of the thyroid in youth. World J Surg 1981; 5~15-25. 2. Clark R M , Rosen IB, Laperriere NJ. Malignant tumors of the head and neck in a young population. Am J Surg 1982; 144:459-462.
126
Poth
3. Newman KD. The current management of thyroid tumorsin childhood. Semin Pediatr Surg 1993; 2~69-74. 4. Zohar Y,Strauss M, Laurian N. Adolescent versus adult thyroid carcinoma. Laryngoscope 1986; 96555-559. 5. McClellan DR, FrancisGL. Thyroid cancerin children, pregnant women, and patients wi Graves’ disease. Endocrinol Metab Clin North Am 1996; 25:27-48. of 6. Zimmerman D, Hay I, Bergstralh E. Papillary thyroid carcinoma in children. Treatment
thyroid cancer in childhood, workshop held at the National Institutes of Health, Bethesda MD, Sept 1992, pp. 3-10. 7. Harach H R , Williams ED. Childhood thyroid cancer in England and Wales. Br J Cancer,
1995; 72:777-783. 8. Sala E,Olsen JH. Thyroid cancerin the age group0-19: time trends and temporal change in radioactive fallout. Eur J Cancer1993; 29A.31443-1445. 9. Zheng T, Holford TR, Chen Y, Ma JZ, Flannery J, Liu W.Time trend and ageperiodof thyroid cancer in Connecticut,1935-1992. Int J Cancer1996; cohort effect on incidence 67(4):504-509. 10. Thoresen S , Akslen LA, Glattre E, Haldorsen T. Thyroid cancer in children in Norway 1953-1987. Eur J Cancer 1993; 29A:365-366. 11. Mangano JJ. A post-Chernobyl rise in thyroid cancer in Connecticut,USA. Eur J Cancer Prevent 1996; 575-81. 12. Correa P, Chen VW. Endocrine gland cancer. Cancer1995; 75:338-352. I, SwerdlowAJ. Sex differences in the risks of hormone-dependent cancers. 13. dos Santos Silva Am J Epidemiol 1993; 138:lO-28. 14. Sankila R,Garwicz S, Oslen JH, Dollner H,Kreuger A, Langmark F, et al. Risk of subsequent malignant neoplasms among 1,641 Hodgkin’s disease patients diagnosed in
childhoodandadolescence:apopulationbasedcohortstudyinfiveNordiccountries. J Clin Oncol 1996; 141442-1446. 15. Thompson DE, Mabuchi K, Ron E, Soda M, Tokunaga M, Ochikubo S, et al. Cancer incidence in atomic bomb survivors. Part II, Solid tumors, 1958-1987. Radiat Res 1994;
137zS17-67. 16. Ceccarelli C, Pacini F, et al. Thyroid cancer in children and adolescents. Surgery 1988; 104: 1143-1 148. 17. Zimmerman D,Jay ID, Gough IR, Goellner JR, Ryan JJ, Grant CS, McConahey W M . Papillary thyroid carcinoma in children and adults: long-term follow-upof 1039 patients conservatively treated at instiation one during three decades. Surgery1988; 104: 1157-1 166 18. Belfiore A, Giuffrida, et al. High frequencyof cancer in cold thyroid nodules occurring a young age. Acta Endocrinol 1989;121:197-202. 19. Ott RA, Calandra DB, McCall A, Shah KH, Lawrence AM, Paloyan E. The incidence of
thyroid carcinoma in patients with Hashimoto’s thyroiditis and solitary cold nod
1985; 1202-1206. 20. Mauras N, Zimmerman D, Goellner
JR. Hashimoto thyroiditis associated with thyroid cancer in adolescent patients. J Pediatr1985; 106895-898. 21. Okayasu I, Fujiwara M, Hara Y, Tanaka Y,Rose NR. Association of chronic lymphocytic thyroiditis and thyroid papillary carcinoma: a study of surgical cases among Japanese, an white and African Americans. Cancer 1995; 76:2312-2318. T. Lymphocytic infiltration in juvenile thyroid carcinoma. Ca 22. Kamma H, Fujii K, Ogata
1988; 62~1988-1993. 23. Matsubayashi S, Kawai K, MatsumotaY, Mukuta T, Morita T, Hirai K, et al. The correlatio
between papillary thyroid carcinoma and lymphocytic infiltration in the thyroid gland. J Clin Endocrinol Metab 1995; 80:3421-3424.
Thyroid
in Children
127
24. Bell B, Mazzafem EL. Familial adenomatous polyposis (Gardner’s syndrome) and thyroid carcinoma. Digest Dis Sci 1993; 38:185-189. 25. Kwok CG, McDougall IR. Familial differentiated carcinoma of the thyroid: report of five pairs of siblings. Thyroid 1995; 5:295-297. 26. Lote K, Andersen K, Nordal E, Brennhovd IO. Familial occurrence of papillary thyroid carcinoma. Cancer 1980; 46:1291-1297. 27. Duffy Jr BJ, Fitzgerald PJ. Thyroid cancer in childhood and adolescence; report on 28 cases. J Clin Endocrinol 1950; 10:1296-1308. 28. Winship T and Rosvoll RV. Childhood thyroid carcinoma. Cancer 1961; 14:734-743. E, Stovall M, Boice JD. Thyroid neoplasis following 29. Ron E, Madon B, Preston D, Alfandary low-dose radiation in childhood. Radiat Res1989; 120:516-531. Cancer incidence 30. Lindberg S, Karlsson P, Arvidsson B, Holmberg E, Lindber LM,A.Wallgren 34735-740. after radiotherapy for skin haemangioma during infancy.Oncoll995; Acta 31. Shore R E , Hildreth N, Dvoretsky P, Andresen E, Moseson M, Pasternack B. Thyroid cancer
among persons given X-ray treatment in infancy for an enlarged thymus gland. Am J Epidemiol 1993;137:1068-80. 32. Ron E, Lubin JH, Shore RE, Mabuchi K, Modan B, Poiern LM, et al. Thyroid cancer after 1995; exposuretoexternalradiation:apooledanalysis ofsevenstudies.RadiatRes 141~259-277. 33. Vane D, King DR, BolesJr ET. Secondary thyroid neoplasms in pediatric cancer patients: increased risk with improved survival. J Pediatr Surg1984; 109:855-860. Jr Robison L, Stone BJ, Stovall M, et al. Therapeutic 34. TuckerMA, Moms Jones PH, BoiceJD, radiation at a young age is linked to secondary thyroid cancer. Cancer 1991; Res 51:28852888. IR. Thyroid disease after treatment of Hodgkin’s disease. 35. Hancock SL, Cox RS, McDougall N Engl J Med 1991; 325:599-605. 36. Healy JC, Shafford EA, Reznek RH, Webb JA, Thomas JM, Bomanji JB, Kington JE. 37. 38. 39. 40. 41.
Sonographic abnormalities of thethyroidglandfollowingradiotherapyinsurvivorsof childhood Hodgkin’s disease. Br J Radio1 1996; 69:617-623. Becker DV, RobbinsJ, Beebe GW, Bouville AC, Wachholz BW. Childhood thyroid cancer II 1996;. 25:197-211. following the Chernobyl accident. Thyroid Cancer Abelin T, Averkin JI, Egger M, Egloff B, Furmanchuk AW, Gurtner F, et al. Thyroid cancer in Belarus post-Chernobyl: improved detection or increased incidence? Soz Praventimed 1994;39:189-197. Baverstock KF. Thyroid cancer in children in Belarus after Chernobyl. Word Health Stat Q 1993; 46~204-208. Nikiforov Y E , Gnepp DR, Fagin JA. Thyroid lesions in children and adolescents after the Chernobyl disaster: implications for the study of radiation tumorigenesis. J Clin Endocrinol Metab 1996;81:9-14. Nikiforov YE, Gnepp DR. Pediatric thyroid cancer after the Chernobyl disaster: pathomorphologicstudyof 84 cases (1991-1992) fromtheRepublic ofBelarus.Cancer 1994;
74~748-766. 42. Holm LE, Wiklund KE, Lundell GE, Bergman NA, Bjelkengren G, CederquistES, et
al.
Thyroid cancer after diagnostic doses of iodine-131: a retrospective cohort study. J Natl Cancer Inst 1988; 80:1132-1138. 43. Shore RE. Issues and epidemiological evidence regarding radiation-induced thyroid cancer. Radiat Res 1992;131:98-111. 44. Pillay R, Graham-Pole J,.Miraldi F, Yulish B, Newman A, Liebman J. Diagnostic x-jrradiation as a possible etiologic agent in thyroid neoplasmsof childhood. J Pediatr 1982; 101: 566-568.
128
Poth
45. Fogelfield L, Bauer TK, Schneider AB, Swartz JE, Zitman R.p53 gene mutation in radiati induced thyroid cancer. J Clin Endocrinol Metab 1996; 81:3039-3044. I, Itoh Y, Wakasugi E, et al. Clinicopatholog 46. Kobayashi T, Nakanishi H, Yana I, Nishisho findings and p53 expression of thyroid cancer in children. Surg Today1995; 25:217-212. 47. Ito T, Seyama T, Iwamoto KS, Mizuno T, TronkoN D , Komissarenko IV, et al. Activated RET oncogene in thyroid cancers of children from areas contaminated by Chernobyl dent. Lancet 1994; 344:259. 48. Fugazzola L, Pilotti S , Picnhera A, Vorontsova T V , Mondellini P, Bongarzone I, et al. Oncogenic rearrangements of the RET proto-oncogene in papillary thyroid carcinomas 1995; 55:5617-5620. children exposed to the Chernobyl nuclear accident. Cancer Res 49. Jocham A, Joppich I, Hecker W, Knorr D, Schwarz HP. Thyroid carcinoma in childhood management and follow-up of 11 cases. Eur J Pediatr 1994; 153:17-22. 50. Samuel A M , Sharma SM. Differentiated thyroid carcinomas in children and adolescents Cancer 1991; 2186-2190. 51. Ceccarelli C, PaciniF, Lippi F, Elisei R, ArganniM, Miccoli, PincheraA. Thyroid cance in children and adolescents. Surgery 1988; lO4:1143-1148. 52. Viswanathan K, Gierlowski TC, Schneider AB. Childhood thyroid cancer: characteristics and long-term outcome in children irradiated for benign conditions of the head and nec Arch Pediatr AdolescMed 1994; 148:260-263. 53. Harness JK, ThompsonN W , McLeod MK, PasiekaJL, Fukuuchi A. Differentiated thyroi carcinoma in children and adolescents. World J Surg 1992; 16:47-54. 54. Schlumberger M, De VathaireF, Travagli JP, Vassa G, Lemerle J, Parmentier C, Tubia M. Differentiated thyroid carcinoma in childhod: long term follow-up of 72 patients. J Cl Endocrinol Metab 1987; 65:1088-1094. 55. Welch-Dinauer CA, TuttleRM, Robie DK, McClellan DR, SvecRL, Adair D, Francis GL Clinical features associated with metastasis and recurrence of differentiated thyroid canc in children, adolescents and young adults. Clin Endocrinol1998; 49:619-628. 56. Fassina AS, RupoloM, Pelizzo MR, Casara D. Thyroid cancer in children and adolescen Tumori 1994; 80:257-262. 57. Lamberg BA, Karkinen-Jaaskelainen M, Franssila KO. Differentiated follicle-derived th roid carcinoma in children. Acta Pediatr Scand 1989; 78:419-425.
11 Imrnwnologic Aspects of Thyroid Follicular Neoplasms James R. Baker, Jr. IMMUNE INTERACTIONS WITH THYROID CANCER
Introduction This chapter highlights immunological aspects of cancers arising from thyroid folli lar cells, and examines the interplay between the immune system and abnormal thyroid follicular cells that could suppress transformed follicular cells from developing into carcinomas. The role of cytokines secreted by immune cells and how they modulate the function of thyroid carcinoma cells in a manner that could inhibit their growth is discussed, and potential difficulties in the immune response to thyroid carcinoma that could lead to the escape of immune containment are examined. Understanding these difficulties may lead to methodsto augment or modify the immune response that could have therapeutic importance. Finally, in this chapter we identify potential routes of immune augmentation or modification of the immune response that could be used therapeutically to suppress thyroid carcinoma. Despite the excellent therapy that has been designed for thyroid carcinoma, many patients still suffer significant morbidity and mortality. Potential immunotherapeutic approaches are not merely conjectural but could have significant clinical importance. Therefore the examination of the immune interaction between effector cells and thyroid carcinoma is an important task to accomplish to define new therapeutic initiatives. Does the immune system monitor and control neoplastic thyrocytes to prevent the development of clinically significant thyroid cancer? The concept of immune surveillance suggests that transforming events occur frequently in cells from many organs, but transformed cells resulting from these events are eliminated because these cells have unique antigens that are recognized by immune effector molecules, which leads to the destruction of the transformed cells and prevents the development of a tumor. Several lines of investigation support this concept in the thyroid. Due to its unique propensity for iodine (and therefore radioisotope) concentration, DNA damage would appear to be a frequent occurrence in the thyroid, iswhich supported by the observation that thyroid cancer is present in situ in approximately 10% of all autopsy specimens (I). The transformed cells lack a second transforming genetic event, such as a mutation
From: Thyroid Cancer: A Comprehensive Guide toClinical Management Edited by: L. Wartofiky 0 Humana Press Inc., Totowa, NJ
129
130
Baker
in p53, which allows for less regulated cell growth and a clinically apparent tumo reviewed in Chapter 6). After years of inactive disease without apparent alterationin tumor phenotype(1,2), some patients with thyroid cancer manifest a dramatic spr metastases. Thus, a breakdown in the immune control of thyroid cancer has occurred either through environmental factors that overwhelm the immune response or th ation of the tumor that leads to progression of disease. However, events such as p53 this scenario mutations are uncommon even in undifferentiated thyroid tumors, making less likely. Thyroid cancer of certain typesis also frequently observed as having a (3,4). Patients with thyroid cancer and this localized thyroiditi oid infiltrate in the tumor (4-6). Therefore, have better survival rates than those without a lymphocytic infiltrate immune control of thyroid cancer may be important in limiting the disease.
Immune Response Genes Predisposing to Thyroid Cancer A number of changes in the host immune response lead to progression of thyroid cancer. HLA antigens have been implicated as a factor in the development of variou types of malignant tumors (7,8). These antigens present tumor-specific antigens to th immune system to induce antitumor responses. If certain allotypesof these HLA antige cannot present tumor antigens, a failureof the antitumor response may result. Sever investigators have looked at allotypes of HLA class I and class 11 antigen and found (9)examined 137 Span an association with thyroid cancer. Rigopoulou and colleagues patients with papillary cancer and found an increase in the class I antigen B35 and th class 11antigen DR11, but the correlation did not hold for follicular carcinoma, w suggests that the etiopathogenesisof the two tumors may be different. An association between DR5 HLA polymorphisms and papillary cancer is also reported in German patients with papillary cancer (10) and an HLA B25 association has been suggested in Italian patients (11). In studies examining Hungarian and American patients with papillary cancer (12,13), HLA associations were observed but were unique to each 11 population. Sridama and coworkers (13) found a significant increase in the class antigen HLA-DR7 with nonradiation-associated follicular and mixed papillarythyroid in cancer patients treated at the University of Chicago (20147 (42.5%) cases vs 22.8% of 979 normal controls). In the Hungarian patients with carcinoma arising from thyroid follicular cells, an a association with class11MHC was also reported as HLA-DR1 was present atsignificantly higher than normal rate in those patients with either follicular, papillary, or mixed papillary-follicular patterns. There was an association of follicular and, to a this latter lesser measure, mixed papillary-follicular carcinoma with HLA-DR3 in study but HLA-DR 1,3 heterozygotes showed a greater relative risk for follicular and mix 11 HLA DR3 alone (12). thyroid carcinoma than thyroid cancer patients with class Studies on 37 thyroid epithelial carcinoma patients in Japan found a strongly s association with class11HLA-DR1 [80% of 37 cancer patients vs 18.3% of 120 he controls (11)]. The biological significance of observed class11HLA-DR associationsin the pathoge esis of thyroid cancer is not clear. Discrepancies between HLA and tumor associati are likely due to the fact that disease susceptibility factors that are associated with it is possible that these different HLA haplotypes in different ethnic groups. While factors are immune-related and influence activities such as cell-mediated immunity,
132
Baker
cytokine production, and the triggering of natural killer (NK) cell activity, they may also be nonimmune-related. Thus, while it can be concluded that immunogenetic fa are important in the response to thyroid cancer, theof failure an HLA-dependent immu response predisposing to the development of thyroid cancer has not been docume POTENTIAL PROBLEMSWITH THE IMMUNE RESPONSE TO CANCER
If immune surveillanceis in general effective in controlling thyroid carcinoma, might be the abnormalities that develop either in thyroid cancer or cells in the immun response that allow the escape of these tumors? Generally, these difficulties fall into three specific areas: 1) inability of the immune response to recognize the carcinoma cells, 2) inability of the immune response to effectively remove the carcinoma cells 3) substances or activities of the thyroid carcinoma cells thems once recognized, and that alter immune function in a way that prevents immune clearance of the cancer.
Ways in Which theImmune System Fails to Recognize Thyroid Cancer There are a number ofinways which the immune system might fail to recogn cancers. The first and most obvious, given the immunogenetic predisposition to is the lack of recognition of thyroid and thyroid cancer-specific antigens by This could be dueto either the inability of class 11MHC molecules to present antigen peptides from thyroid antigens (as discussed above) or the inability of an indiv cell a and P chain antigen receptors to recognize these peptides in association is markedly curtailed, and the end MHC types.In either way, the immune response is a lack of immune recognition. As outlined previously, immunogenetic predis that is, specific HLA allotypes that cannot present peptides, could be important inthis process. However, equally important are differences in T-cell antigen receptors tha to an absence of thyroid cancer responses. One manner in which this potential problem might occur would be deletion of certain subsets of VPTcells through exposure to b superantigen. This could delete wholeVP families in a manner that could remove effector CD4 cells to recognize thyroid cancer peptides. At a more fundamental leve lack of recognition could occur because the cancer cells themselves do not express CD ofMHC classmol molecules. In thyroiditis, gamma interferon causes an upregulation 11 If in less-differentiated thyroid cancers this cules thatmay lead to thyrocyte destruction. cannot occur, there may be a significant problem in generating an immune immune recognition may be failed either because of an absence or genetic differenc MHC T-cell interactions. Another important point to remember is that man ated thyroid cancers evolve from differentiated cancers that appear to be so in scope.It is thought that the undifferentiated cancer may grow faster and th in a more significant and rapidly growing neoplasm. However another option i undifferentiated cancer, failing to express many of the thyroid-specific antigens, escap immune recognition.This may also be true for the concept of immune targeting cussed below.
Ways in WhichImmune Targeting Fails to Kill Thyroid Cancer Cells A clear understanding of CD8 cytotoxicity, the central mechanism for clearance transformed cancer cells by the immune system (8), has led to an understanding of
Thyroid
133 Perforins
cytotoxic Cell
Target
Cell
Fig. 1. Documentation of how cytotoxic T cells kill target cells. The cytotoxic cell attaches to the target cell because of TCWCD8 recognition of theMHC Vantigen complex on the target of perforin molecules, which cause osmotic cell. It then induces death either through the release lysis by forming channels in the membrane of the target cell, or through the induction of an apoptotic signal by interaction with the FAS antigen on the target cell.
how alterations in neoplastic cells might prevent the destruction of these cells by the immune system. Clearly CD8 lymphocytes kill target cells by two mechanisms (Fig. 1). The first identified involves a secreted protein called perforin. These molecules are somewhat analogous to the membrane attack complex of complement and are released by CD8 cells and insert into the membranes of target cells (14). This causes lysis through osmotic changes to the cell. If changes in the carcinoma cell’s membrane prevents the effect of the perforins as sometimes occurs in cancer cells, then it would be likely that these cells would not be lysed by this pathway. The pathway that is more important to cell lysis by CD8 is through the induction of apoptosis in target cells (15). This recently identified pathway is mediated through a number of lymphocyte ligand on the CD8 cells (16). It interacts with a specific receptor (CD95, APOA1, or Fas) on the surface of the cancer cell. When the target cells have their Fas antigen interact with ligand, it induces an apoptotic signal through unique pathways in the cancer cell.This pathway is somewhat analogous to the signal induced by tumor necrosis factor (TNF) but actually occurs independently through a different set of kinase enzymes that have high sequence homology with the enzymes involved in the TNF-induced cell death pathway. Several ways in which thyroid cancer may subvert an immune response have been identified (Fig. 2). It has been shown in several cell lines that a lack of expressionof CD95 or alteration and different metabolic pathways (such as upregulation BCL2) can block the cell death in these pathways, thus preventing the induction of apoptosis even
134
Releaseof Blocking Antigen
Baker
Loss of C e l l - S p e c m C Antigens
Loss of Immune RecognitionMolecules
Fig. 2. Methods in which thyroid cancer cells can evade immune surveillance. One may be the release of blocking antigens or viral superantigens that missdirect or suppress immune recognition. A second potential mechanism involves the of thyroid-specific loss antigen such as thyroid peroxidase. This may be important since immune responses to thyroid canc may be primarily directed toward these antigens. A third mechanism might involve the lossof immune recognition molecules on the cells. Molecules such asHLA and CD95 on the surface of the cells are importantin inducing cytotoxic death andmay not be present on cancer cells Therefore, even though an immune response to the cancer exists, may it be ineffective.
if CD8 cells are formed. Carcinoma cells can also stop expressing class I MHC moleIn this case, cules, and this prevents the CD8 cells from targeting the cancer cells. even though an effective CD8 immune response is produced, cancer cells cannot be killed. The immune response to thyroid cancer can also be subverted as a result of actions of the cancer cell that actively suppress the immune response. It has been show that some virally induced tumors produce superantigens that destroy T cells reacting to the tumor(17). In other types of tumors, there are a number of actions that can turn off the immune response. One is the secretion of TGF-P by the cancer cells, which suppresses immune responsiveness and prevents the production of cytotoxic imm to the tumor. Several investigators have shown that thyroid cells have the ability to produce this cytokine and therefore this may be a potential pathway of suppressing immune responsiveness. Tumors can also produce an excess of soluble antigen that can be released into the immune system, and can bind up immune response elemen or be taken up by other cells misdirecting the immune response. Thyroglobulinis an example of such an antigen found in thyroid cells, which can be released at very to thyrogobulin could be high concentrations from cancer cells; immune responses misdirected and ineffective in this instance.
Attempts atImmunotherapy of Thyroid Cancer In considering the role of immunity and the potential value of immunotherapy fo differentiated thyroid cancers, it is of interest that these tumors were among the first neoplasms for which specific immunotherapy was attempted. These trials, which w conducted in the 1970s before modern insight into the immune system was available, werebasedonexperiencewiththyroidantigensgainedfromautoimmunethyroid disease. Amino and associates (18) used a homogenized extract of autologous thyroid cancer tissue to immunize three patients who had widely metastatic thyroid cancer
Thyroid Follicular Neoplasms
135
(2 with papillary cancer and1 with papillary-follicular carcinoma). Two of the immunized patients were in the terminal stage of illness and were anergic; they showed no response to either the thyroid cancer or control antigens. The one patient who was not anergic did demonstratein vitro evidence of specific antitumor cellular immunity after vaccination, and had a 33% decrease in tumor mass that persisted for over a year. LoGerfo and coworkers(19) treated eight patients with either papillary or follicular cancer. He immunized modified human thyroglobulin suspended in Freund’s adjuvant and found evidence of antithyrogobulin antibodies developing in five of the patients. Thyroiditis was diagnosed in two of the patients with thyroglublin antibodies; one of these patients showed a30% regression of tumor mass. He also reported that one other patient had stabilizationof presumed lung metastases; however, biopsies were not done on the lung lesions. The antithyroglobuin antibody titers in these patients were not high (1512 to 1:1024), and interference by circulating thyroglobulin made it difficult to evaluate the humoral response to the immunization. Despite their failings, these studies indicate that thyroiditis can be induced in humans through immunization. It may therefore be possible that the induction of thyroiditis may have an antitumor effec on thyroid cancer.
CYTOKINES AND THYROID CANCER CELLS Cytokine Effects on Thyroid CeZZs Cytokines are a major element in the human immune response and are responsible for cellular recruitment, chemotaxis, and the expressionof adhesion molecules. Cytokines have also been shown to have profound antitumor effects in certain cance Therefore it is important to examine the role cytokines play in immune responses to thyroid cancer cells to determine if they may be useful in immunotherapeutic applications.
Tumor Necrosis Factor and the Interleukins As its name implies, tumor necrosis factor (TNF) was first characterized as an agent
that has the ability to kill tumor, but not normal cells (20). TNF exists in many forms; TNF-p areproducedfromdifferentgenesandhavebothsolubleand membrane-bound forms. TNF-ais a cytokine produced mainly by monocytes and by large granular lymphocytes and enhances the cytolytic activity of NK cells, possibly this mechanism (21). TNF-p, also known as lymphomediating tumor regression through toxin, is released from stimulated lymphocytes in a manner similar to TNF-abut is less studied and understood. Studies have examined the specific biophysiological action of TNF-aon rat thyroid cells (22). It showed negative regulatory effects in vivo, including reduced circulating thyroid hormones and TSH, decreased thyroid response to TSH, and reduced thyroid iodine uptake. In vitro studies examining the effectsof TNF-aon F’RTL-5 rat thyroid cells demonstrated both multiple activities.TNF-ainhibited iodine-trapping functions in either the presence or absenceof TSH, but did not significantly stimulateF’RTL-5 cell growth (i.e., did not affect [3H]thymidine incorporation and cellular DNA content). However, TNF-astimulated RNA synthesis of the treated thyroid cells (as shown by increased [3H] uridine incorporation). These findings suggest that TNF-amay play a
TNF-a and
136
Baker
role in thyroid cell growth and function.TNF-amay also alter the immunogeneity of II HLAthyroid cells. Either aloneor synergistically withy interferon it enhances class DR expression in human thyroid cancer cell lines. Since thyroid cancer cells lack 11HLA-DR and HLA ICAMl expression and have decreased expression of some class class I antigens (23-25), these deficiencies might be overcome by the local production of TNF-a. Therefore, thyroid cell function and immunogeneity may be regulated by TNF-a. Interleukin 1 (IL-1) is a molecule with a wide range of inflammators and immunologic activities. Itis a cofactor for lymphocyte activation, mediates fever, and induces cytokine synthesis in neutrophils and vascular endothelial cells (26). IL-1 is synthesized by many cell types, most commonly by activated monocytes and macrophages and the effect of this locally produced IL-1 may be important for thyroid cell function (27). Human thyroid cells are known to have IL-1 receptors and have been documented to secrete E-1-like activity (27). Human recombinant IL-1 was fou to be a potent inhibitor of thyroglobulin and cyclic AMP production and/or release from human thyroid cell cultures (28), and has been reported to inhibit TSH-induced TPO gene expression in human thyroid cells (29). Pang and colleagues (22) have studied the in vitro action of IL-1 on FRTL-5 rat thyroid cells. E - 1 inhibited iodine uptake in the absence of TSH, while in the presence ofE -TSH, 1 showed its inhibitory FRTL-5 cell growth as effect only at a higher concentration. IL-1 also stimulated demonstrated by increased [3H] thymidine incorporation and increased cellular DNA content. Kimura and colleagues (27) also studied the effect of IL-1 on human pap and follicular thyroid cancer cells. They found an inhibitory effect of E - 1 on cell growth of papillary cancer cells but not follicular cancer cells. Thus, while IL-1 may play an important role in host defense against malignant progression, especially in the aspect of antitumorigenic action, its inhibitory effects on thyroglobulin and TPO expression may decrease the antigenicity of cancers. T cell growth factor now known as interleukin 2 (IL-2) is a polypeptide primarily produced by activated T cells of the CD4+ phenotype. IL-2 has pleiotrophic activity, it stimulates the proliferation of cytotoxic T lymphocytes (CTL), as well as Thl and Th2helpercells.IL-2inducesCTLboth in vivoandin vitro andstimulatesthe NK cells and the productionof TNF and IFNy (30). IL-2 is a key stimulator of tumoricidal activity of monocytes. The vital role of IL-2 in tumor surveillance and control is clear. IL-2 also plays a key role in inducing therapeutic antitumor immune responses, as discussed below. Interleukin 6(IL-6)is a cytokine synthesized by various cells, including monon phagocytes, vascular endothelial cells, fibroblasts and other cells in response to L-1 and, to a lesser extent, TNF. IL-6 is a pleiotropic cytokine which stimulates hepatic synthesis of acute phase proteins, and exhibits a number of unique functions in host defense, including a major role in T-cell activation and B-cell differentiation. It has been suggested serum concentrations of IL-6 might constitute a marker of thyroid destructive processes. IL-6 concentrations have been elevated in conditions associa as the percutaneous intranodular ethanol injection (P with thyroid cell damage, such radioactive iodine (RAI)administration andFNA (29). Although it was not examined these results raise the possibility that serum IL-6 levels may be a marker for thyroid lysis associated with thyroid cancer.
Thyroid Follicular Neoplasms
137
Interjerons Interferon 0a,IFNp, and IFNy are inducible proteins that are important in defense against viruses. These molecules also help to regulate the immune response, have intiproliferative activity against tumor cells when present in high concentrations. Expression of class I MHC gene products and NK cell activity are also enhanced by these proteins, and these activities have been exploited therapeutically in the treatment of certain tumors and viral infections. The effectof IFNy on MHC class 11antigen expression andTPO content in normal or thyrocytes or follicular cancer cells has been studied (31). IFNy was shown to induce the expressionof HLA-DR antigens, reduce theTPO content of thyrocytes, and inhibit the increase in TPO content induced by TSH (31). TSH actually enhanced the IFNyinduced expression of class 11 HLA-DR antigen suggesting that IFNy may play an important role in the modulation of thyrocyte antigen expression. Chemo kines Chemokines and interleukins that have chemoattractant activity have been termed “chemokines” (32). Interleukin 8 (IL-8) is one of this family of chemotactic factors that are produced by many cells, including monocytes, hepatocytes, endothelial cells, fibroblasts, epithelial cells, and neutrophils (33-38). Both IL-1 and TNF have been found to stimulate the expressionof IL-8 (34-36) and IL-8 plays an important role in the chemotaxis of inflammatory cells (33,3538). Thyroid epithelial cells have also been shown to produce IL-8, in vitro, and this production is enhanced by IL-1 and reduced by IFNy (39). A functional neutrophil chemotactic factor-like activity also has been identified, in vitro, in a cell line derived from an undifferentiated thyroid (40). carcinoma It is possible that the local production of IL-8 and other chemotactic cytokines could induce local inflammation, potentially preventing the spread of localized thyroid carcinoma. This concept is in need of further investigation.
Transforming Growth Factor p Transforming growth factors p (TGF-p) are a family of compounds synthesized by a wide variety of cell types including activated macrophages and platelets. They are involved in a wide range of activities associated with tissue remodeling, inflammation and cellular regulation (41). TGF-P inhibits a number of immune activities including T & B cell proliferation,IL-2receptor expression,IL- 1induced thymocyte proliferation, 11 antigen expression production of IFN-y and TNF, natural killer cell activity and Class (42-44). In vivo, many tumors may escape an immune response by secreting large quantities of TGF-p. However, the possible role of TGF-P in the pathogenesis of thy carcinoma has not been investigated and the ability of thyroid cells to produce TGF-fl in response to various stimuli is unknown. ANTIGENICITY OF THYROID EPITHELIAL CANCERS There are two different types of molecules that may serve as antigen targets in the are differentiated proteins immune response to thyroid cancer. One group of molecules
138
Baker
specific to thyroid cells, such as thyroglobulin, thyroid peroxidase and the TSH rec A second form of antigen includes antigens associated with cellular transformation progression to neoplasm that are specific for cancer cells. Both classes may serve as potential targets for immunotherapy.
Diflerentiated Thyroid Antigen Expression in Thyroid Cancer Autoimmune thyroid disease has identified thyroid specific proteins that are a as antigens to thyroid cancer (45), in humans. Several of these proteins also present including thyroglobulin, and thyroid hormone, TPO and the TSH receptor. Monitorin levels of some of these proteins in the serum of patientsis important in the diagnosis and treatment of thyroid cancer. More importantly, these proteins have implications for future immunotherapy of thyroid cancer. The TSH receptor is an autoantigen in Graves’ disease (46). Several investigators have identified a higher than expected incidence of thyroid cancer in patients with Graves’ disease, suggesting a possible link between TSH stimulating antibodies and thyroidcancer (47-49). This suggeststhatselectivelyblockingsignaltransduction through the TSH receptor on thyroid cancer cells could possibly alter tumorigenesis, a concept currently being investigated in thyroid cancer cell lines (50). No evidence as has definitely implicated the TSH receptor in thyroid tumorigenesis. TSH can act a growth factor for differentiated thyroid carcinomas (47-49,51). Some investigators have also recommended that TSH suppression should be carried out following me or surgical ablation of differentiated thyroid cancer (52). TPO is thekeythyroidenzymethatcatalyzesboth of theenzymaticreactions as the responsible for the production of thyroid hormone and previously identified thyroid microsomal antigen(24). It is a 104 kDa protein consisting of933 amino acids principally located intracellular and on the apical surface of thyroid epithelial cells. Sequence homologies exist with a number of different peroxidase enzymes, includin myeloperoxidase and cytochromeC oxidase (54-56). PAX-8, a thyroid-specific activa tion factor, activates theTPO promoter (57), and TSH also enhancesTPO transcription and translation. TPO is a complex autoantigen and the immune response it isto varied. TPO epitopes recognized by autoantibodies are both conformational and linear (58). This has been demonstrated by studies of autoantibodies using TPO in native and reduced forms.A strong correlation exist between the titers of these antibodies and the histological le of Hashimoto’s disease (59), suggesting that this antigen may play an important role in the pathogenesis of thyroiditis. TPO also is present on most follicular carcinomas and approximately half of papillary cancers(60,61).On inflamed or neoplastic thyroi follicular cells, TPO has been observed to be present diffusively on the plasma me brane. This differs from normal thyroid tissue, where the antigen is only present on TPO the follicular cell apical membrane, not in contact with the vasculature. Therefo may be a unique target for immune attack in thyroid cancer becauseit is accessible to the immune system. Thyroglobulin (Tg) is one of the most abundant proteins of the thyroid gland, and it provides the matrix for the synthesis of thyroid hormone. In humans, the Tg gene resides on the long arm of chromosome 8, distal to the c-myc oncogene (62). It is 660 kDa. Human Tg has many made up of two monomeric polypeptide chains of
Thyroid Follicular Neoplasms
139
autoantibody epitopes and high-titers of anti-Tg antibodies are foundin about 70% of patients with Hashimoto's thyroiditis, 60% of those with newly diagnosed idiopathic hypothyroidism, 30% of those with Graves' disease, as well as a smaller percentage of patients with thyroid carcinoma (46). TSH elevates the levels of Tg by enhancing the transcriptional activity of its promoter. Thyroglobulin is used as a monitor for recurrent or metastatic disease following treatment becauseit isreleased from cancer cells.An elevated serum Tg concentration during thyroid hormone suppression usually reflects metastatic disease or recurrent primary tumor, The measurement of Tg in fluid obtained by fine needle aspirates of neck masses has also been reported to be a simple and reliable procedure for the (63). AlthoughTg is an important tool for diagnosisofthyroidcancermetastases monitoring thyroid cancer,it varies inits antigenicity in both normal and thyroid cancer patients. This is due, in part, to iodine content and this may make thyroglobulin an unreliable target for the immunotherapy of thyroid cancer. Several studies have documented the loss of the presence or function of thyroid antigens in thyroid carcinoma cells. Decreases in thyroid peroxidase function have bee reported by several groups (64-68), and the loss of TPO in follicular and papillary carcinoma also has been observed utilizing immunofluorescence and electron-microscopic techniques(64). About 50% of follicular and papillary carcinoma lacked thyroid peroxidaseactivityelectronmicroscopically or histochemicallyinanotherstudy. Changes in TPO structure, such as a decrease in solubility of "0, have also been [6%in cancer vs.50% in normal thyroid(38)].Other thyroid reported in thyroid cancer For proteinshavealsoreportedtohavedecreasedexpressiononthyroidcancers. example, thyroid cancer cells on the average have quantitatively decreased expression levels of TSH-receptor mRNA and decreased level of 1251-TSH binding to the receptors (67). All this may indicate a loss of antigenicity in thyroid cancers as these tumors become progressively differentiated. Other factors may alter the immunogenicity of thyroid antigens on tumors, and could possibly aid in the induction of tumor-specific tolerance in the patient's immune syste Class 11 MHC antigen expression on the cancer cells may induce tolerance through incomplete stimulationof CD4 cells due to the absence of costimulating molecules on cancer cells. This might include molecules such as ICAM and B7 (2). As previously mentioned, the loss of Class I MHC antigens on thyroid cancer cells may lead to a defect in specific cellular cytotoxicity (8). Part of the reasons for these events is the dedifferentiation of cell type. Differentiated cellular phenotypeis due in large part to transcriptional activation factors. These factors are considered the main regulators of gene expression, and can be classified according to their localization of expression. Transcriptional factors that are tissue specific are very important in expressing the differentiated phenotype of each cell, because they are limited in their function, only regulating transcription of genes in specific cells (57). Thyroidfollicularcellexpressestwospecifictranscriptionalfactors-thyroid transcriptional factor-1 ("TF-l),and PAX-8 (41,42). Although these two factors can be found independently expressed at low levels in lung and kidney, respectively, they are only present together in thyroid cells (57). This unique combination and the high levels observed in follicular cells suggest that they play a central role in thyroid-specif phenotypic expression. This view is substantiated by the fact that both 'ITF-1 and
140
Baker
PAX-8 bind and activate the promoters for thyrogobulin and thyroid peroxidase. follicular neoplasms displaying varied histology express these transcriptional factors (69). However, there is heterogeneity of expression even among different tumors of the same histological type (70). ‘ITF- 1, PAX-8, TG, andTPO mRNA levels have been studied in normal and mali TTF-1appears to be a necessary factor, although not suffici nant human thyroid cells. (57). Cotransfection for the full expression ofthethyroiddifferentiatedphenotype experiments have shown that PAX-8 is a dose-dependent activator to the TPO promoter (and to a lesser extent the TG promoter), and its presence in cancer cells is a major determinant of TPO gene expression. As expected, TTF-1, PAX-8, TG, and TPO are not expressed in anaplastic carcinoma, but these studies suggested correlations be low levels of expression of PAX-8, TG, and TPO in “high-risk” papillary carcinoma (57). The heterogeneity of expression of these factors in thyroid cancer suggests that clinical and histopathological staging may fall short of predicting the aggressiveness of a thyroid neoplasm. It is also possible that replacement of these factors, through gene therapy, may lead to a more differentiated phenotype for some thyroid tumors.
Molecular Markers of Poorly DifSerentiated Thyroid Cancer Poorly differentiated thyroid carcinoma (PDC) may have unique antigens different from thyroid antigens. There appear to be significant associations between specific genetic alterations and histological types, with ofsome these mutations yielding protei with unique structures (71). Mutations of all three rus family genes can be found in follicular carcinoma (FC) (72,73) while N-rus mutations were found to be associated with aggressive PC. Mutations of these genes in other carcinomas has yielded uniq antigenic proteins that are tumor antigens. The expression of bcl-2 and p53 in PDC may also yield unique antigens. Bcl-2 expression is restricted to PDC and not prese in the majority of undifferentiated tumors and thus may be important for blocking apoptosis inthis tumor. By contrast, p53 mutations in PDCare restricted to only areas of active tumor infiltration, but are seen in almost all of the tumor cells in tumors (71). Thus,whilepoorlydifferentiatedandundifferentiatedcarcinomalack expression of thyroid autoantigen, suchas thyroglobulin andTPO,they may have othe unique antigens that could serve as targets of immunotherapy. Other recently identified antigens may have a potential role as immunotherap in differentiated thyroid cancer. Epithelial membrane antigen (EMA) is a glycoprotein which is expressed by malignant lesion of epithelial origin (74). In thyroid neoplasia 80-100% ofpapillarycancersandupto70%of EMAhasbeendemonstratedin follicular cancers (75). Cheifetz and coworkers (76) found a statistically significant difference in the expressionof EMA by follicular carcinomaas compared to follicular adenomas. In another investigation, EMA was linked to prognosis in papillary cancer being expressed in47% of metastatic tumors while absentin all localized tumors(48). Further investigations on larger series of patients will be needed to define the exact incidence of EMA, and its potential as a target for immunotherapy. Another antigen expressed on both epithelial and nonepithelial tumors is Leu-7 (78). This antigen is expressed by natural killer cells as well as both papillary and follicu cancers. A recent study was able to differentiate papillary cancer from benign lesion with pseudopapillae using Leu-7 expression (761, and this may be useful because many
Thyroid
141
benign goiters contain multiple pseudopapillae making them difficult to differentiate as a tumor target may be limited, it is extremely from papillary carcinoma. While its use helpful in the diagnosis of malignant thyroid neoplasms. POTENTIAL METHODS TO AUGMENT THE IMMUNE RESPONSE TO THYROID CANCER
Current therapeutic options for the treatment of thyroid cancer include surgery, radioactive iodine, radiation therapy, and chemotherapy. These modalities control the disease in the majority of patients with thyroid cancer. However, approximately 10% of patients die from this neoplasm.Thus,there is a need for additional treatment modalities. This might include immunotherapy. Immunotherapy to augment host defense using nonspecific immune stimulants has failed to improve survival rates. Therefore practical applications of this approach require new agents that elicit cellular immune responses and the use of different biological response modifiers may offer improved results. Currently, there are extremely few clinical trials utilizing any of these techn for thyroid neoplasm. The systemic administration of cytokines to boost the biologic responseof T cells has had limited usefulness as therapy in humans(30). IL-2 that was used in the induction of antitumor T-cell immune responses has limited the role of immunotherapy, primarily due to its short half-life and severe dose-related toxicity. The toxicity commonly i fever, chills, headache, diarrhea, anorexia, nausea, and vomiting. Less frequent but more serious complications include mental status changes, hypotension, respiratory distress syndrome, renal, hepatic and cardiac dysfunction, and even death in approximately 1 4 % of patients treated (79). Systemic IL-2 administration demonstrated that while a minority of patients respond clinically (about20%),most responses have been in patients with renal cell carcinoma and multiple melanoma. In a phase II pilot study, 14 patientswithadvancedendocrinetumors,whichincluded 5 metastaticthyroid tumors (3 anaplastic, 1 papillary, and 1 medullary) were treated with low-dose IL-2 in combination with the pineal hormone melatonin. A partial response was observed in 3 of 12 evaluable patients; however, none of the patients with thyroid cancer had a reduction in tumor size (80).
Lymphokine-Activated Killer Cells or Tumor-Inflltrating Lymphocytes and IL-2 Lymphokine-activated killer (LAK) cells are mononuclear cells from the peripheral blood which are generated by incubation in IL-2. In vitro, LAK cells kill a variety of In clinical trials tumor cells in a nonspecific manner but are not toxic to normal cells. LAK cells in combination with IL-2 resulted in a clinical response in 21% of solid 8% (81). Again, tumors that responded the best tumors and a complete response in were melanoma and renal cell carcinoma, but there are no reported trials in thyroid tumors. By contrast, killer cells can also be generated from tumor-infiltrating lymphocytes (TIL). These cells are presumably enriched for tumor reactivity, and therefore are more specific in their killing. Adoptive transfer of TIL in several murine models has shown that these cells can mediate the regression of tumors 50-100 times better than LAK cells (82). The TIL cytolysis is HLA class I restricted for the majority of
142
Bakw
tumors. Results from trials treating metastatic melanoma in humans have been promising, with response rates of 40-60% (83). The toxicity associated with TIL therapy is lower because the dose of IL-2 is reduced. Given the intense lymphocytic infiltrate observed in many thyroid tumors, which could be a source of thyroid-specific TILS, this would appear to be a potential therapy for thyroid cancer.
Cytokine-Secreting Tumor Cell Preparations Gene therapy using modified tumor cells that secrete cytokines as a “tumor vacc involves the removal of cancerous tissue from the body, transduction of tumor cells ex vivo, and reintroduction of the genetically altered cells back into the patient. This system attempts to make nonimmunogenic cancer stimulate the immune system to develop an immune response that will destroy all the tumor cells. Certain cytokines, amongthem IL2 and 6MCSF, havedemonstrated the abilitytoaugmentimmune responses to tumor cells in some animal models. Therefore, tumor cells genetically altered to secrete these cytokines might serve as an effective vaccine. However of t currentlyavailablegenedeliverysystems,replication-deficientretrovirushasbeen used to incorporate cytokine genetic material into tumors. Investigators have obtaine antitumor responses using gene-modified tumor cells secreting IL-2, IL-4, TNF, and GM-CSF (84). The advantage to this system is the potential for greater tumor selecti than systemic cytokine therapy.
CONCLUSION Thyroid cancers have a demonstrated antigenicity. Many of these antigens are for thyroid cells or thyroid cancer cells, and can induce intensive immune responses insomeindividuals.Withmanipulation of thecellularimmuneresponsetothese antigens, possible through the use of effector elements like cytokines, it now seems possible to use these antigensas immunotherapeutic targets. While the current therap of differentiated thyroid cancer is quite effective for many patients, 10% of papillary and 15-50% offollicularcancerpatientsdevelopdistantmetastasisandrecurrent disease. An even higher percentage of patients with less differentiated cancers fail conventional therapy, resulting in substantial mortality and morbidity. Thus, immun therapy, using an augmented or active specific cellular response against a number of potential target antigens could provide an additional therapeutic option in the trea of thyroid cancer. ACKNOWLEDGMENTS This work was supported by the following Grants: R 0 1 AI 37141-01 and 1 R 0 3 TWOO192-01 to JRB and Center Grants 2 P60 AR20557 and M01 RROOO4,2.
REFERENCES
1. Bisi H, Fernandes VSO, De Camargo RYA, Koch L, Abdo A H , De BritoT. The prevalence of unsuspected thyroid pathology in 300 sequential autopsies with special reference to the incidental carcinoma. Cancer 1989; 64:1888. 2. Baker Jr JR, Fosso CK. Immunological aspects of cancers arising from the thyroid follicula cells. Endocr Rev 1993; 146.
Thyroid
Neoplasms
143
3. Shull JH, Sharon N, Victor TA, Scanlon EF. Thyroid carcinoma: immunology, irradiation, and lymphocytic infiltration. Arch Surg 1979; 114729. 4. Maceri DR, SullivanMJ, McClatchney KD. Autoimmune thyroiditis: pathophysiology and relationship to thyroid cancer. Laryngoscope 1986; 96:82. 5. McConahey WM, Hay ID, Woolner LB, van Heerden JA, Taylor WF. Papillary thyroid cancer treated at the Mayo Clinic, 1946 through 1979: initial manifestations, pathologic findings, therapy, and outcome. Mayo Clin Proc 1986; 61:978. 6. Clark OH, Greenspan FS, Dunphy JE. Hashimoto's thyroiditis and thyroid cancer: indications for operation. Am J Surg 1980; 140:65. 7. Klitz W. Viruses, cancer and the MHC. Nature 1992; 35617-8. J (editor). The molecular basis of cancer. 8. HerbermanRB. Cellular immunity. In Mendelsohn Philadelphia: WB Saunders, 1995. 9. Rigopoulou D, Martinez-Laso J, Martinez-Tello F, Alcaide JF, Benmamar D, Hawkins F, Amaiz-Villena A. Both class I and class11HLA antigens are thyroid cancer susceptibility factors. Tissue Antigens 1994; 43:281-285. 10. DralleH, Robin-Winn M, Reilmann L, Laue A, Torok M. HLA und Schilddrusencarcinom. Klin Wochenschr 1986; 64:522. 11. Panza N, De Vecchio L, Maio M, De Felice M, Lombardi G, Minozzi M, Zappacosta S. Strong association between an HLA-DR antigen and thyroid carcinoma. Tissue Antigens 1982; 20:155. 12. Juhasz F, Boros P, Szegedi G,Balazs G, Suranyi P, Kraszits E, Stenszky V, Farid NR. Immunogenetic and immunologic studies of differentiated thyroid cancer. Cancer 1989; 63:1318. 13. Sridama V, Hara Y, Fauchet R, DeGrootLJ. Association of differentiated thyroid carcinoma with HLA-DR7. Cancer 1985; 56:1086. 14. Carter LL, Dutton RW. Relative perforin- and Fas-mediated lysis in T1 T2 and CD8 effector populations. J Immunol 1995; 155:1028. KJ, Levine H, Ritz J. Functional 15. Robertson MJ, ManleyTJ, Pichert G, Cameron C, Cochran consequences of APO-Was (CD95) antigen expression by normal and neoplastic hematopoietic cells. Leukoc Lymphom 1995; 17:51. 16. Nagata S, Golstein P. The Fas death factor. Science 1995; 267:1449. 17. SchererMT, IgnatowiczL, Winslow GM, KapplerJ W , Marrack P. Superantigens: bacterial and viral proteins that manipulate the immune system. Annu Rev Cell Biol 1993; 9:lOl. LJ. Immunologic aspects of human thyroid cancer: 18. Amino N, PysherT, Cohen EP, DeGroot humoral and cell-mediated immunity and a trial of immunotherapy. Cancer 1975; 36:963. 19. LoGerfo P, Feind C, Weber C, TingW. Immunotherapy of thyroid cancer by induction of autoimmune thyroiditis. Surgery 1983;94959. 20. Ruggiero V, Latham K, Baglioni C. Cytostatic and cytotoxic activity of tumor necrosis factor on human cancer cells. J Immunol 1987; 138:2711. of 21.Ostensen ME, Thiele DL, Lipsky PE. Tumor necrosis-a enhances cytolytic activity human natural killer cells. J Immunol 1987; 138:4185. 22. Pang X-P, Hershman JM, Smith V, Pekary AE, Sugawara M. The mechanismof action of tumour necrosis factor-a and interleukin 1 on FRTL-5 rat thyroid cells. Acta Endocrinol (Copenh) 1990; 123:203. 23. Liaw K-Y, Yao C-C, Chen Y-C, Deng J-S. Impaired cell-mediated immunity function in thyroid cancer. Cancer 1980; 46:285. 24. Pujol-Borrell R, Todd I, Doshi M, Bottazzo GF, Sutton R, Gray D, Adolf GR, Feldmann by interferon-y plus tumour necrosis factor M. HLA class 11induction in human islet cells or lymphotoxin. Nature 1987; 326:304. 25. Weetman A P , Cohen S, Makgoba MW, Borysiewicz LK. Expression of an intercellular adhesion molecule,ICA"1, by human thyroid cells. J Endocrinol 1989; 122:185. 106. 328: 26. Dinarello CA, Wolff SM. The role of interleukin-1 in disease. N Engl J Med 1993;
144
Baker
27. Kimura H, Yamashita S, Namba H, Tominaga T, Tsuruta M, Yokoyama N, I& M, Nagataki S. Interleukin-1 inhibits human thyroid carcinoma cell growth. J Clin Endocrin Metab 1992; 75596. 28. Rasmussen AK, Bech K, Feldt-Rasmussen U, Poulsen S, Siersbaek-Nielsen K, Friis T, in vitro cultured human Bendtzen K. The influence of interleukin-1 on the function of thyroid cells in monolayers. Acta Endocrinol (Copenh) Suppl 1991; 281:93. m i M, 29. Ashizawa K, Yamashita S, Tobinaga T, Nagayama Y, Kimura H, Hirayu H, h Nagataki S. Inhibition of human thyroid peroxidase gene expression by interleukin1. Acta Endocrinol (Copenh) 1989; 121:465. 30. Foa R, Guarini A, GansbacherB. IL2 treatment for cancer: from biology to gene therapy Br J Cancer 1992; 66:992. 3 1. Asakawa H, Hanafusa T, Kobayashi T, Takai S-I, Kono N, Tarui S. Interferon-y reduces the thyroid peroxidase content of cultured human thyrocytes and inhibits its incre by thyrotropin. J Clin Endocrinol Metab 1992; 74: 1331. 32. Strieter Rh4, Standiford TJ, Hufiagle GB, Colletti LM, Lukacs NW, Kunkel SL. “The good, the bad, and the ugly”: the of role chemokines in models of human disease. J Immunol 1996; 156:3583. 33. Standiford TJ, Kunkel SL, BashaMA, Chensue SW, Lynch JP IJJ, Toews GB, Westwick J, Strieter RM. Interleukin-8 gene expressionby a pulmonary epithelial cell line: a model for cytokine networks in the lung. J Clin Invest 1990; 86:1945. 34. StrieterRM, Phan SH, ShowellHJ, Remick DG, Lynch JP, Genord M, Raiford C, Eskand M, MarksRM, Kunkel SL. Monokine-induced neutrophil chemotactic factor gene ex in human fibroblasts. J Biol Chem 1989; 264:10621. 35.Strieter Rh4, KunkelSL,Showell HJ, RemickDG,Phan SH, WardPA,Marks RM. Endothelial cell gene expression of a neutrophil chemotactic factor by TNF-a,LPS, and IL-lp. Science 1989; 243:1467. 36. Thornton AJ, Strieter RM, Lindley I, Baggiolini M, Kunkel SL. Cytokine-induced gene J Immunol expression of aneutrophilchemotacticfactor/IL-8inhumanhepatocytes. 1990; 144:2609. 37. Strieter RM, Kasahara K, Allen RM, Standiford TJ,Roue MW, Becker FS, Chensue SW, Kunkel SL. Cytokine-induced neutrophil-derived Interleukin-8. Am J Path01 1992; 1 V M , Elner SG, StrieterRM. 38. Koch Polverini PJ, Kunkel SL, Harlow LA, DiPietro LA, Elner Tnterleukin-8 as a macrophage-derived mediatorof angiogenesis. Science 1992; 258:1798 39. Weetman AP, Bennett GL, Wong WLT. Thyroid follicular cells produce Interleukin-8. J Clin Endocrinol Metab 1992; 75328. K, Mizushima S, KawakitaM, 40.YoshidaM,MatsuzakiH,SakataK,TakeyaM,Kat0 Takatsuki. Neutrophil chemotactic factors produced by a cell line from thyroid carcinom Cancer Res 1992; 52:464. 41.LawrenceDA.Transforminggrowthfactor-beta:anoverview.Kidney IntSuppl1995; 49:S19. 42. Ranges GE, Figari IS, Espevik T, PalladinoMA Jr. Inhibitionof cytotoxicT cell developmen by transforming growth factor p and reversal by recombinant tumor necrosis factor a. J Exp Med 1987; 166:991. is a strong inhibitor 43. Ristow HJ. BSC-1 growth inhibitodtype b transforming growth factor of thymocyte proliferation: Proc Natl Acad Sci USA 1986; 835531. 44. Wrann M, Bodmer S, de Martin R, Siep C, Hofer-Warbinek R, FreiK, Hofer E, Fontana A. T cell suppressor factor from human glioblastoma cells is a 12.5-kd protein closely related to transforming growth factor-p.EMBO J 1987; 6:1633. 45. Haapala AM, SoppiE,MorskyP, Salmi J, Laine S, Mattila J. Thyroidantibodiesin association with thyroid malignancy: simultaneous occurrence of thyroiditis and thyroid malignancy. APMIS 1994; 102:390-4.
Thyroid
145
46. Barnett PS, McGregor AM. Immunologic factors. In: Wheeler MH, Lazarus JH, editors. Diseases ofthe thyroid. London: Chapman & Hall, 1994;86-103. 47. Mazzaferri EL. Thyroid cancer and Graves’ disease. J Clin Endocrinol Metab 1990; 70:826. 48. Filetti S, Belfiore A, Amir SM, Daniels GH, Ippolito 0, Vigneri R, Ingbar SH. The role of thyroid-stimulating antibodiesof Graves’ disease in differentiated thyroid cancer. N Engl J Med 1988; 318:753. 49. Belfiore A, GarofaloM R , Giuffrida D, Runello F, FilettiS, Fiumara A, Ippolito 0, Vigneri R. Increased aggressiveness of thyroidcancer in patients with Graves’ disease. J Clin Endocrinol Metab 1990; 70:830. 50. Ossendorp FA, Bruning PF, Schuuring EMD et al. Thyrotropin dependent and independent FRTL-5 tumors grown innudemice.Endocrinology thyroidcelllinesselectedfrom 1990; 127:419-430. 51. Brabant G, Maenhaut C, KohrleJ, Scheumann G, DralleH, Hoang-Vu C, Hesch RD, von zur Muhlen A, Vassart G, Dumont E.Human thyrotropin receptor gene: expression in thyroid tumors and correlation to markers of thyroid differentiation and dedifferentiation. Mol Cell Endocrinol 1991; 82:R7. 52. Dunn JT.Thyroid suppression and medical ablation for differentiated thyroid cancer. Arch Otolaryngol Head Neck Surg 1986; 112:1207. 53. Weetman AP. Thyroid peroxidaseas an antigenin autoimmune thyroiditis. Clin Exp Immuno1 1990; 80:l. 54. Kimura S , Hong Y-S, Kotani T, Ohtaki S, Kikkawa F. Structure of the human thyroid peroxidase gene: comparison and relationship to the human myeloperoxidase gene. Biochemistry 1989; 28:U81. 55. Yokoyama N, Taurog A,Doms ML, Klee GG. Studies with purified human thyroid peroxidase and thyroid microsomal autoantibodies. J Clin Endocrinol Metab 1990; 70:758. 56. Elisei R, MariottiS , Swillens S, Vassart G, Ludgate M. Studies with recombinant autoepitopes of thyroid peroxidase: evidence suggesting an epitope shared between the thyroid and the gastric parietal cell. Autoimmunity 1990; 8:65. 57. Fabbro D,Di Loreto C, Albert0 Beltrami C, Belfiore A, Di Lauro R, Damante G. Expre of thyroid-specific transcription factors ‘ITF-1 and PAX-8 in human thyroid neoplasms. Cancer Res 1994; 54:4744. 58. Portolano S , Chazenbalk GD, Set0 P, Hutchinson JS, Rapoport B, McLachlan SM. Recogn tion by recombinant autoimmune thyroid disease-derived Fab fragments of a dominant conformational epitopeon human thyroid peroxidase. J Clin Invest 1992; 90:720. B. Carbohydrate moieties in recombinate human thyroid peroxidase: role i 59. Foti D, Rapport recognition by antithyroid peroxidase antibodies in Hashimoto’s thyroiditis. Endocrinology 1990; 126:2983. in benignandmalignthyroid 60. FraguP,NatafBM.Humanthyroidperoxidaseactivity disorders. J Clin Endocrinol Metab1977; 451089. 61. De Micco C, RufJ, Chrestian M-A, Gros N, Henry J-F, Carayon P. Immunohistochemical study of thyroid peroxidase in normal, hyperplastic, and neoplastic human thyroid tissues. Cancer 1991; 67:3036. 62. Charreire J. Immune mechanisms in autoimmune thyroiditis. Adv Imrnunol1989; 46263. 63. Plachov D, Chowdhury K, Walther C, Simon D, Guenet JL, Gruss P. Pax-8,a murine paired box expressed in the developing excretory system and thyroid gland. Development 1990; 110:643. 6 4 . Pontius K, Hawk WA. Loss of microsomal antigen in follicular and papillary carcinoma of the thyroid. Am J Clin Path01 1980; 74:620. 65. Yamashita H, Noguchi S, Murakami N, Yokoyama S, Nakayama I. Loss of intracellular peroxidase and anaplastic change of differentiated carcinomaof human thyroid gland. Acta Pathol Jpn 1987; 37:425.
146
Baker
66. Neary JT, Nakamw C, Davidson B, Soodak M, Vickery AL,Jr, Maloof F. Studies on the membrane-associated nature of human thyroid peroxidase: a difference in the solu enzyme from benign and malignant thyroid tissues. J Clin Endocrinol Metab 1978; 67. Ohta K, Endo T, Onaya T. The mRNA levels of thyrotropin receptor, thyrogobulin and thyroid peroxidase in neoplastic human thyroid tissues. Biochem Biophys Res Commun 1991;174:1148. LJ, Jhiang SM. Thyroid peroxidase expression and DNA polymorphis 68. Smanik PA, Fithian in thyroid cancer. Biochem Biophys Res Commun 1994; 198:948. 69. Williams ED. The aetiology of thyroid tumors. Clin Endocrinol Metab 1979; 8:193. 70. Brabant G, Maenhaut C, Korle J, Scheumann G, Dralle H, Hoang-Vu C, Hesch RD, von zur Muhlen A, Vassart G, Dumont JE. Human thyrotropin receptor gene: expression in thyroid tumors and correlation to markers of thyroid differentiation and dedifferentiation. Mol Cell Endocrinol 1991; 82:R7. 71. Pilotti S , Collini P, Romualdo DB, Cattoretti G, Pierotti MA, Rilke F. A novel panel of antibodies that segregates immunocytochemically poorly differentiated carcinoma from undifferentiated carcinomaof the thyroid gland.Am J Surg Path01 1994; 18:1054. 72. Lemoine N R , Mayall ES, WylieFS, Williams ED, Goyns M, Stringer B, Wynford-Thomas of human thyroid tumorigenesis D. High frequencyof ras oncogene activation in all stages Oncogene 1989; 4159. 73. Suarez HG, du Villard JA, Severino M, et al. Presence of mutations in all three ras genes in human thyroid tumors. Oncogene 1990; 5565. 74. Pinkus GS, Kurtin PJ. Epithelial membrane antigen-a diagnostic discriminant in surgical pathology. Hum Path01 1985; 16:929. 75. Wilson N W , Pambakian H, Richardson TC. Epithelial markers in thyroid carcinoma: an immunoperoxidase study. Histopathology 1986; 10:815. 76. Cheifetz RE, Davis NL, Robinson BW, BereanKW, LeRiche JC. Differentiation of thyro neoplasms by evaluating epithelial membrane antigen, leu-7 antigen, epidermal growth factor receptor and DNA content. Am J Surg 1994; 167531. 77. Yamamoto Y, Izumi K, Otsuka H. An immunohistochemical studyof epithelial membrane antigen, cytokeritin and vimentinin papillary thyroid carcinoma. Cancer 1992; 70:2326. 78. Si L, WhitesideTL. Tissue distributionof human NK cells studied with anti-Leu-7 monoclonal antibody. J Immunol 1983; 1302149. 79. Ravikumar TS, Steele GD Jr. Modem immunotherapy of cancer. Adv Surg 1991; 2441. 80. Lissoni P, Barni S , Tancini G, Mainini E, Piglia F, Maestroni GJM, Lewinski A. Immu docrine therapy with low dose subcutaneous IL-2 plus melatonin of locally advanced or metastatic endocrine tumors. Oncology 1995;52163. 8 1. G r i m EA, Mazumade RA, Zang H Z . Lymphokine activated killer cell phenomenon:lysis of natural killer resistant fresh solid tumor cells by interleukin-2 activated autologo peripheral blood lymphocytes. J ExpMed 1982; 1551823. 82. Rosenberg SA, Spiess PJ,La Freniere R. A new approach to the adoptive immunotherapy of cancer with tumor infiltrating lymphocytes. Science 1986; 223:1318. 83. Rosenberg SA, Packard BS, Aebersold PM et al. Use of tumor infiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic melanoma. N Engl J Med 1988; 319:1676. 84. ViewegJ,Gilboa E. Considerations for the use of cytokine-secreting tumor cell prep for cancer treatment. Cancer Invest 1995; 13:193.
12 Radioiodine Therapy of Thyroid Cancer General Considerations"I Gerald Johnston and Diane Sweeney
Differentiatedthyroidcancer,when it hastheabilitytoconcentrateradioactive iodine, lends itself nicely to postthyroidectomy radioiodine therapy (Fig.1). The term radioiodine therapy is used to designate the treatment of residual recurrent, or metastat thyroid cancer (1-6). Radioiodine ablution, on the other hand, is used to describe the removal of noncancerous thyroid tissue, particularly that which is left after thyroidectomy. The distinction between these terms is frequently blurred, for when thyroid ca has been diagnosed, and near-total thyroidectomy performed, one cannot assume that there is no residual cancer. Many thyroid cancers are multicentric(8-10). Others show evidence of extrathyroidal extension or capsular penetration. It may be prudent to consider that residual thyroid cancer is present (4) and that treatment is not complete until all vestiges of thyroid tissue are gone. Radioiodine ablatiodtherapy after thyroidectomy provides for the following: 1. Treatment of the multifocal, multicentric, and microscopic thyroid cancer found in 24% (7) to 58% (10) of cases. 2. The athyroidal patient being more manageable. a. Residual thyroid tissue, being more avid for radioiodine, may prevent visualization of thyroid cancer sites (11). b. Thyroid hormones from residual thyroid tissue may suppress thyroid-stimulating hormone and impede imaging of less active thyroid cancer (1,U). c. Thyroglobulin from residual thyroid tissue reduces the usefulness of thyroglobulin as a tumor marker to monitor thyroid cancer (12,14-17).
THE THERAPY/ABLATION DOSE Two philosophies prevail for administering radioiodine therapy/ablation: the lowdose approach and the high-dose approach. Advocates of low-dose treatment hold that of outpatient therapy outweigh the benefitsof lower radiation exposure and convenience the increase in unsuccessful ablations. In the low-dose treatment, I3*Iis administered serially in doses of 30 mCi or less. In the high-dose approach, the inpatient receives 80-150 mCi of I3'I. Such doses may be considered as radiation therapy for residual thyroid cancer.
From: Thyroid Cancer: A Comprehensive Guide toClinical Management Edited by: L. Wartofsky 0 Humana Press lnc., T o t m , NJ
147
148
Johnston and Sweeny
Fig. 1. Postthyroidectomy diagnostic metastatic survey scans. Anterior projections from
to pelvis. (A) 3 days following10 mCi I3II dose: Uptakein the thyroid bed(-) and physiological activity in the nasal mucosa (l),salivary glands (2), stomach (3), and colon (4). (B) 10 days following 150 mCi “‘1 dose: Uptake in the thyroid bed (-) and physiological activity in nasal (4). Liver is visualize mucosa (l),salivary glands(2), liver (3), and minimal activity in the colon
owing to the incorporationof radioactive iodine into endogenously produced thyroid hormo which is subsequently metabolizedby the liver.
Experience with radioiodine therapy indicates a high level of success (approx 90%) in achieving total ablation of uptake confined to the thyroid bed using a single dose of 100 to 200 mCi of I3II (I).Lower doses of I3’I in the 25- to 29-mCi range were successful in8% (7) to 27% (18) to 61% (I1,12)in achieving ablation after near total thyroidectomy. From these results, we believe that the rate of successful ablation with a therapy dose of up to 29 mCi is too low despite the convenience. The approach we favor is to consider the ablation dose to be adjunctive therapy and to give approximately 150 is cumulative and that multiple mCi of l3IIin a single dose. The concept that radiation (five or six) smaller doses of 29 mCi are equivalent to a single 150-mCi dose is not borne out in practice with thyroid ablation. Whetherthis biological observation is the of subsequent result of thyroid tissue “stunning” or a decrease in the biological half-life radioiodine dosesor some other physiological alteration, the effectiveness of sequential radioiodine therapies is decreased (20). The regulation (Nuclear Regulatory Commission, Part 35.75), that isolation of a patient in a private monitored roomis required for body burdens of over 30 mCi has
Radioiodine
Therapy--I149
been the compelling argument for low-dose ablation attempts. Allen and Zielinski (21) have demonstrated that high-dose ablation therapy can be practical in an outpatient setting. With prior approval of the Texas NRC, the 430 patients they treated in this way were confined to their homes following regulations for protecting family members from radiation exposure. The safety and cost-effectiveness of this approach were demonstrated. Such an approach may be soon approved by the NRC. For now, however, this demonstration provides insight to a more practical approach to radioiodine highdose therapy. ARE ALL THYROID CANCER PATIENTS CANDIDATES FOR ABLATION?
Both the recurrence rate and the likelihood of cancer death following thyroidectomy are reduced by half in I3*Itreated patients compared with those treated with thyroid hormonereplacementaloneorwithexternalradiation (7). However,patientswith tumors smaller than1.5 cm completely confined to the thyroid were not found to benefi from total thyroid ablation. The use of l3IItherapy may be questioned 1) if the primary tumor is smaller than 1 cm without metastases(22), 2) if the tumor is not differentiated (papillary or follicular), and 3) if only a lobectomy or lumpectomy is performed (2). Many would support ablation of all residual thyroid tissue in the interest of improving patient management in an athyroidal state. Opinions differ widely on the efficacy of radioiodine ablation therapy following thyroidectomy for differentiated thyroid cancer. Cady and Rossi (23) at Lahey Clinic have stated their reluctance to use radioiodine in their “low-risk group.” This group includes all younger patients without distant metastases (men under 41 years and wo under 51). Low riskis also extended to older patients with intrathyroidal differentiated 5 carcinoma with no or minor capsular involvement and the primary cancer less than cm in diameter and no distant metastases. On the other hand, Samaan and colleagues (24) noted significantly fewer recurrences and deaths in low-risk patients using Lahey Clinic criteria. They recommend radioiodine therapy for all patients whose scans show lesions after surgery. Beienvaltes and associates believe that residual tissue should be ablated as part of the treatment of well-differentiated thyroid cancer. We believe that the weight of the evidence favors the ablation of detectable residual thyroid tissue in all patients with well-differentiated cancer using 150 pCi of 13’1as the standard dose. Patients with partial thyroidectomy should have as much of the remaining thyroid tissue, asis reasonable, removed before radioiodine ablation therapy is attempted. h a d and colleagues (25) reported an ablation rate of 28% with single-dose therapy (mean dose: 141.3 mCi) in patients following hemithyroidectomy. Large remnants are not amenable to ablation therapy. Conversely, vanishingly small (or vanished) evidence for residual thyroid tissue should be regarded cautiously when presented for radioiodine ablation therapy. These are usually a distinct group composedof treated differentiated thyroid cancer patients with positive or rising serum thyroglobulin levels and negative radioiodine scans. A positive thyroglobulin level in these patients is a value greater than2 ng/ml. Clark and this clinical dilemma: diffuse metastases Hoelting (26) suggest possible explanations for too small for detection, thyroid cancer unable to take up enough iodine for detection,
150
Johnston and Sweeney
high levels of "cold" iodine blocking radioiodine uptake, normal thyroid tissue h visualization of metastatic disease and a false-positive elevation of thyroglobulin. Four studies (11,14,27,28) reported 51 patients with elevated thyroglobulin levels In 46 of the 51 patients post and whose I3II scans were negative for metastatic disease. therapy scans that were positive for metastatic lesions or decreased thyroglobulin levels after I3II therapy was evidence of their documented metastatic or persistent thyroid cancer. The benefit of therapy with l3II in this situation has not been proven. Such patients should have thyroglobulin levels confirmed in two laboratories to show that In addition, measurement ofurinary iodine antibodies are not interfering with the assay. shouldbeperformedtoexcludeartifactualsuppression of 13'1 uptake.Finally,the patients should be counseled in detail about their situation. It is doubtful that13'1therapy in the absence of l3IIuptake will be of benefit. RADIATION SAFETY CONSIDERATIONS IN THE TREATMENT OF THYROID CANCER WITH 1311
The rules governing the release of patients receiving radioactive pharmaceu changed in 1997 dueto revisions in regulations by the Nuclear Regulatory Commissi (NRC). Previously, regulations restricted licensees from releasing patients from the hospital until the dose rate from the patient was less than 5 mremh at 1 m distance, or the radiopharmaceutical content of the patient was less than 30 mCi. The usuallyrequiredahospitalstayofapproximately 36-48 hours in aprivate, wellcontrolled, and monitored hospital room before discharge of l3II-treated thyroid cancer patients. NRC issued a final rule which shifted the However, effective May 29 1997, the focus to the potential dose to individuals coming into contact with the treated patien Medical Use of Byproduct The amended regulations, Section 35.75 from 10CFR35 Material from the NRC,includes several provisions as follows (29):
1. The release from control of any individual who has been administered radiopharmaceut if the total effective doseequivalent to any other individual from exposure to the treated patient is not likely to exceed 5 mSv (0.5 rem); 2. That the treated individual must be given written instruction on actions recommended to adhere to the ALARA concept (aslow as reasonably achievable), if total effective dose equivalent to other individuals is likely to exceed 1 mSv (0.1 rem); 3. That thelicensee must maintain arecord of the basis for authorizing the release of the patient.
For 13*1specifically, patients may be released if the activity administered is less than 33 mCi (1.2 GBq), if the dose rate at 1 m distance is less than 7 mRem/h (0.07 mSvl h), or if the patient is unlikely to expose other individuals to greater than 0.5 rem of the patient with patient (mSv). In the last case specifically, the basis for the release specific dose calculations must be recorded. staff, The risk of radiation exposure to nuclear medicine clinic personnel and nursing although small,is the primary guidepost in developing a workable method for treatin patients with 13*I.It has been estimated that the mean exposure rate of thyroid cance patients measured at 0.3 m is 15.7 mR/h on days 0-1,2.6 mRih on days 2-4, and 1.7 mRk on days 5-7 (30). A person, if continuously exposed for 7 days at 0.3 m from 1.1 rem. Therefore, patients are isolated the patient, would receive a calculated dose of
Radioiodine
Therapy-1151
in private rooms with limited close contact with hospital personnel, and are discharged from the hospital usually within 24 hours. Culver and Dworkin (30)proposed guidelines for close contact with posttherapy patients as follows: 1. Day 0-1following discharge: Restricted time in close contact with patientat 0.3, 0.6, and 1m. 2. Days 2-4 following discharge: Restricted contact with small children and pregnant women 0.3 at m. 3. Days 5-7 following discharge: No restrictions.
The new NRC regulations require that instructions be given to patients at discharge, including recommendations to limit exposure to others. These should include (31): 1. Maintaining distancefrom other persons 2. Minimizing time in public places 3. Reducing the risk of contamination 4. Guidelines on time periodsof the recommendations
PRACTICAL MANAGEMENT OF 1311 TREATED PATIENTS Informed consentis obtained in all patients, including a discussion of the alternatives to treatment with radioiodine. A pregnancy testis obtained in all premenopausal patients. A notation is made of a current CBC with emphasis on the white blood cell count.In our own laboratory, we usually treat with capsular I3lI; however, liquid "'I is used if the patient cannot swallow easily, orif a gastrostomy tubeor nasogastric tube is to be used for the delivery of the radiopharmaceutical. Use of liquid I3'I is not encouraged because it is volatile and can pose a hazard to the nuclear medicine staff in attendance. Following administration of the dose, all workers involved in thi dispensation must undergo a thyroid bioassay count in order to rule out inadvertent exposure. Following discharge of the patient from the hospital, the hospital room must be cleared by a TSH Radiation Safety officer before another patient can be assigned to the room. suppressive therapy with thyroxine or thyroxine and temporary triiodothyronine supplementation is usually started on the day of discharge.An outpatient posttherapy wholebody scan should be scheduled 7-10 days after administration of the inpatient dose. REFERENCES 1. Beierwaltes WH, Rabbani R, Dmuchowski C, et al. An analysis of "ablation of thyroid of Michigan. remnants" with1-131in 511patients from1947-1984 experience at University J Nucl Med 1984-1993; 25:1287-1293. 2. Freitas JE, Gross MD, Ripley S , et al. Radionuclide diagnosis and therapy of thyroid cancer: current status report. Semin NuclMed 1985; 15:106131. 3. Harbert JC.Radioiodine.therapyof differentiated thyroid carcinoma.In: Nuclear medicine therapy. New York Thieme, 1987. 4. Hurley J R , Becker DV. Treatment of thyroid carcinoma with radioiodine. In Gottschalk A, Hoffer PB, Potchen EJ, Berger HJ, editors. Diagnostic nuclear medicine, 2nd ed. Baltimore: Williams & Wilkins, 1988: 792. 5. Leeper R D , Shimaoka K. Treatment of metastatic thyroid cancer. Clin Endocrinol Metab 1980; 9:383.
152
Johnston and Sweeney
6. Maxon HR, Smith HR. Radioiodine-131 in the diagnosis and treatment of metastatic we A m 1990; 19:685-718. differentiated thyroid cancer. Endocrinol Metab Clin North 7. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. A m J Med 1994; 97:418-28. 8. Clark OH, Levin K, Zeng Q, et al. Thyroid cancer: the case for total thyroidectomy. Eu J Cancer Clin Oncol 1988; 24:305-13. 9. Tollefsen HR, DeCosse JJ, Hutter RVP. Papillary carcinoma of thethyroid a clinical an pathological study ofj70 fatal cases. Cancer 1964; 17:1035. 10. Clark RL, White E, Russell WO. Total thyroidectomy for cancer of the thyroid. Ann Surg 1959; 149:858. 11. Ronga G, FiorentinoA, Paserio E, et al. Can iodine-131 whole body scan be replaced thyroglobulin measurement in the postsurgical follow-up of differentiated thyroid carci noma? J Nucl Med 1990; 31:1766. 12. Sisson JC. Applying the radioactive eraser: 1311 to ablate normal thyroid tissuein patient from whom thyroid cancer has been resected. J Nucl Med 1983; 24:743. 13. Iida Y, Hidaka A, Hatabu H, et al. Follow-up study of postoperative patients with thyr cancer by thallium-201 scintigraphy and serum thyroglobulin measurement. J Nucl Med 1991; 32:2098. 14. Pacini F, Lippi F, Formica N, et al. Therapeutic doses of iodine-131 reveal undiagnose metastases in thyroid cancer patients with detectable serum thyroglobulin levels. J Nucl Med 1987; 28:1888. 15. Ramanna L, Waxman A, BraunsteinG. Thallium-201 scintigraphy in differentiated thyro cancer: comparison with radioiodine scintigraphy and serum thyroglobulin determinat J Nucl Med 1991; 32:441. 16. Schlumberger M. Can iodine-131 whole-body scan be replaced by thyroglobulin meas mentinthepostsurgicalfollow-upofdifferentiatedthyroidcarcinoma?JNuclMed 1992; 33:172. 17. Kuni CC, Klingensmith WC m. Failure of low doses of 13'1to ablation residual thyroid tissue following surgery for thyroid cancer. Radiology 1980; 137:773-774. 18. Commtois R, Theriault C, Del Vecchio P. Assessment of the efficacy of iodine-131 for thyroid ablation. J Nucl Med 1993; 34:1927-1930. 19. McCowan KD, Adler RA, Ghaed N,et al. Low-dose radioiodide thyroid ablation in p gical patients with thyroid cancer.Am J Med 1976; 61:52. 20. Rawson RW, Rall JE, Peacock W. Limitations and indications in the treatment of cance of the thyroid with radioactive iodine. J Clin Endocrinol 1951; 11:1128. 21. Allen HC Jr, Zielinski JD. 480 Non-hospitalized thyroid cancer patients treated wit doses 50-4OOmCi [Abstract]. J Nucl Med 1990; 3 1:784. 22. Davis NL, Gordon M, German E,et al. Efficacyof 1-131 ablation following thyroidecto in patients with invasive follicular thyroid cancer.Am J Surg 1992; 163:472. An expandedviewofriskgroupdefinitionindifferentiatedthyroid 23.CadyB,RossiR. carcinoma. Surgery 1988; 104:947-953. 24. Samaan NA, Schultz PN, Hickey R, et al. The results of various modalities of treatme of well-differentiated thyroid carcinoma: a retrospective review of 1599 patients. J Clin Endocrinol Metab 1992; 75:714-720. 25. Arad E, O'Mara RE, Wilson GA. Ablation of remaining functioning thyroid lobe with radioiodine after hemithyroidectomy for carcinoma. Clin NuclMed 1993; 18:662-663. 26. Clark OH, Hoelting T. Management of patients with differentiated thyroid cancer w positive serum thyroglobulin levels and negative radioiodine scans. Thyroid 1994; 4: 27. Pineda JD, Lee T, Reynolds J, et al. 131-1 therapy for thyroid cancer with elevat ulin and negative diagnostic scan [Abstract]. Thyroid 1992; 2:S16. 28.Robbins J(moderator).Thyroidcancer:alethalendocrineneoplasm. Ann Intern Me 1991;115:133.
Radioiodine Therapy -1
153
29. U.S. Nuclear Regulatory Commission. Release of individuals containing radiopharmaceutic a l ~or permanent implants. In: Part 35. Medical use of byproduct material. Washington, DC: US. Regulatory Commission, 1997: Paragraph 3575,3516. HJ:Radiation safety considerations for post-iodine-13 1thyroid cancer 30. Culver CM, Dworkin therapy. J Nucl Med 1992; 33:1402. of patients adminis31. U.S. Nuclear Regulatory Commission. Regulatory guide 8.39: Release tered radioactive material. Washington, DC: U.S. Nuclear Regulatory Commission, April 1997.
13 Radioiodine Treatment of Thyroid Cancer General Considerations-11 Side Effects of Radioiodine Therapy for Thyroid Cancer Diane Sweeney and Gerald Johnston
A 150-mCi treatment dose of 1311will result in a dose of 39,000 cGy to the thyroid and 36 cGy to the total body( I ) . Therefore, a discussion of side effects of13'1therapy for thyroid cancer must be included in the preliminary workup and treatment consider tions of all patients. The nuclear medicine physician, endocrinologist, primary care short- and longphysician, patient, and the patient's family should be aware of possible term radiation-related side effects and complications. However, this discussion is not as straightforward as we might like, due to the unknown significance of low-level radiation exposure and the lack of large prospective studies with matched patient populations. Yet, by careful analysis of the case reports and retrospective studies, we can make a fairly accurate assessment of the risks. SALIVARY GLAND DYSFUNCTION Radiation sialoadenitis appears to be a direct result of radiation injury due to the concentration of iodine by the glands. Salivary glands concentrate iodide 30-40~the plasma level (2). Spiegel and colleagues(3)conducted a prospective quantitative study using salivary gland scintigraphy to evaluate radiation risk of iodine-131 therapy. They found a dose-dependent reductionin salivary gland function following 13'1therapy. The parotid glands are more often affected than the submandibular glands, and the authors surmise that complete salivary gland failure may occur with doses greater than 500 mCi I3'I. Creutzig (4) suggests that xerostomia will be a common finding after 1 Ci 1311. Many patients will experience the clinical symptoms of acute or chronic sialoadenitis after I3'I therapy. Allweis and coworkers (2) found that 10 of 87 patients (11.5%) who had received a therapeutic dose of 1311(median dose: 100 mCi) reported symptoms of dry mouth, gland tenderness, and swelling; 9 of the 10 patients with symptoms had received prior13'1therapy. Onset of symptoms generally began within 1week of therapy and lasted 3 weeks to 2% years (duration of study). Treatmenthas included antibiotic therapy (for suppurative sialoadenitis), duct dilatation, and,in one case, parotidectomy
From: Thyroid Cancer:A Comprehensive Guide to Clinical Management Edited by: L. Wartofiky 0 Humana Press Inc., Totowa, NJ
155
156
Sweeney and Johnston
for intractable discomfort. Van Nostrand and colleagues (5) found salivary gland tend ness, pain and swelling, or dry mouth in 10 of 15 patients treated with 13’I.All but 2 13’1therapy. of these patients received doses greater than 200 mCi and had previous Of 258 patients in an earlier study, 10% were found to have salivary gland problems and had received total doses of 8-41 GBq (216-1108 mCi) (6). Radiation sialoadenitis occurs in approximately 10% of patients who receive 131 therapy and most frequently in those patients receiving multipleor high-dose therapy The most important preventive therapy is to increase salivation, through good hydr 13’1therapy. and the frequent therapeutic use of sour candies in the first days following Sucking lemons may be a bit extreme, but will also be effective. NECK PAIN, TENDERNESS, AND SWELLING
Radiation thyroiditis presenting as neck and facial swelling or pain can occur af I3’Itherapy (57). However, this has not been documented in treatments following total thyroidectomy or total thyroidectomy, in the absence of extensive neck metasta Painless neck swelling can also occur, but again, has been well-documented only in cases of less than total thyroidectomy, usually lobectomy (8). These patients usually experience painless swelling of the neck bilaterally, with a “tight feeling” or choki sensation. These symptoms usually occur within 48 hours of treatment and are consi ered to be distinct from radiation thyroiditis, which occurs 3-4 days after therapy. Bo painful and painless swelling can be treated with corticosteroids, if necessary. Ab of an entire lobe with13*1is difficult (9)and, therefore, the possibilityof neck swelling should be considered an additional caveat to attempting lobar ablation. GASTROINTESTINAL SYMPTOMS
Nausea is the most common gastrointestinal symptom seen in l3lI therapy patients In a recent prospective study of 50 patients receiving 150 mCiof I3’I,50% complained of nausea (10). Van Nostrand and coworkers (5) found that 67% of their patients experienced nausea, starting as early as 2 hours following therapy and lasting up to days after therapy. Antiemetics, usually given intramuscularly, can be used in of patients with persistent or debilitating symptoms and to prevent vomiting, which otherwise will decrease the patient’s effective dose and contaminate the room is a far less common side effect than nausea, occurring in 8% to 15% of patients in two small series (510). It is estimated to occur in less than5% of patients if doses d (11). The latter estimate (less than 5%) is closer to notexceed200 cGy to blood our experience. Acute radiation sickness with fatigue, headache, nausea, and vomiting has been (12), but this syndrome has not been reported in recent reported in earlier studies considering doses under200 mCi and the restriction of exposure to under 200 cGy the blood. TASTE DYSFUNCTION
Abnormalities andor loss of taste can occur with1311 therapy. Varma and coworke
(13) found loss of taste with or without taste distortion (described as phantom, m
Radioiodine Therapy-11
157
or chemical taste) in 41 of 85 patients who received 150-200 mCi of 1311 for welldifferentiated thyroid cancer (13).Persistent taste abnormalities were found in the absence of salivary gland dysfunction. The symptom of an "unpleasant taste" following therapy has been noted in several other reports (5,14). However, the persistent loss of taste or disconcerting long-term changes in taste has not been reported. Long-termdirected follow-up of these patients is warranted, since oral dosing and the presence of radioactive saliva in those patients puts the taste buds at risk. PARATHYROID GLAND DYSFUNCTION The parathyroid glands can be considered somewhat radioresistant. In addition, the location of the glands may shield them from radiation damage due to the limited range 13'1therapy have occurred in of beta particles. Most cases of hypoparathyroidism after patients treated for thyrotoxicosis, with doses of 4-40 mCi (15). However, they did find diminished parathyroid reserve in58% of patients studied prospectively who had received 80-150 mCi of I3II. Overt hypoparathyroidismis rare as a consequence of1311 therapy. However, long-term follow-up of calcium levels in these patients is prudent, especially with the appearance of muscle cramps or paresthesias.
BONE MARROW DEPRESSION
l3II therapy causes a transient drop in the white blood cell, red blood cell, and platele count (5,16).At the University of Michigan between1947 and 1960,152 patients were treated with1311and followed with serial complete blood counts (16).Mean hemoglobin levels dropped from 13.6 to 12.7 and mean white cell count fell from 7750 to 6600 after 3months. There also may be a persistent mild decrease in1year WBCafter therapy (5). Factors that may worsen bone marrow depression following 1311therapy include: 1. Doses yielding greater than 200 cGy whole-body radiation (14). 2. Priorx-raytherapy (16). 3. Extensive bone metastatic disease (14).
Blood counts should be monitored before and after therapy. The counts appear to if the CBC is reach a nadir 5-9 weeks after therapy. Therapy should be withheld abnormal,andhematologic investigation should ensue before 1311 dosing. With the availability of colony-stimulating factors, the risks have diminished. TESTICULAR FUNCTION AND MALE FERTILITY The standard treatment dose ofI3'I for thyroid cancer (150 mCi) exposes the testes cGy (0.5-1.0 in a hypothyroid man to an estimated absorbed radiation dose 75-100 of cGy/mCi) (17J8). Because thyroid cancer often strikes people in their childbearing years, the consequences of this radiation level must be evaluated, discussed, and limited as much as possible. Lushbaugh and Casarett (19) reviewed the data from experimental animal studies and retrospective clinical studies. They estimate the gonadal tolerance levels in human testis by cell type. They found that the LDso for spermatogonia is 15-33 cGy with the 500 cGy. Therefore, it is expected dose needed for permanent sterility to be at least
Johnston 158
and
Sweeney
that there will be some effect on testicular function, and subsequently, sperm cou in most young men treated with therapeutic doses of I3II. Testicular germinal-cell damage, decreased sperm counts, and elevated gonad I3lI (18,20,21). Handelsman an levels have been well documented in men treated with Turtle (18) found a dose-related deviation in FSH levels in 12 men treated with I3II in doses ranging from 50 to 540 mCi. Sperm density was also inversely correlated wit total I3'Idose. Severe spermatogenic depression, suggested by marked elevation and decreased sperm density, occurred in 2 patients who had received 400 mCi and 350 mCiI3II.However, 22- and 26-month follow-up in these patients suggested of testicular function and ultimately, reversibility of testicular damage. Ahmed and I3'I with Shalet (20) report 4-year follow-upof a 13-year-old boy treated with 350 mCi no suggestion of recovery of spermatogenesis, with persistently elevated basal FSH levels and azoospermia. In a comparatively large (103 patients) study reported in 1994, increased seru Higher cumulativ concentration occurred in one-third of male patients treatedI3lI. with doses, received by repeated treatments over months or years appear to be extremely damaging. Four patients treated with several doses of 13'1 (cumulative doses ranging from 520to 800 mCi) had an additive and persistent increase in FSH levels indicati permanent testicular damage. in these men does not ap Despite the risk to spermatogenesis, subsequent fertility to differ significantly from the general population. Sarkar and colleagues (22) fo 40 men and women treated with 13'1 for thyroid cancer at age 20 or under. Overall, the incidence of infertility was 12%, which matched the prevalence in the general population. Two men who had received doses of 210 mCi and97 mCi I3'I were deeme infertile. However, one of these men was married only 3 years of 23 years of follow up and had spent most of his time in mental institutions. Thereis also an underlying be malformed or genetically fear thatthe progeny of these irradiated young people will is evidence tha altered. This fear has never been substantiated and more likely, there radiation-damaged spermatogonia are self-destructive (19). There is a documented, detrimental effect I3lI of on germinal cell function. Radia damage to the testes appears to result from free radioactive iodine in the blood and (20). Therefore bladder urine and also from radioiodinated thyroglobulin and thyroxine radiation exposure can be decreased by good hydration and frequent urination, larly in the first 3 days after therapy. In all young male patients, but especially in thos with metastatic disease or pelvic disease or both, the long-term storage of sem be addressed prior to therapy. OVARIAN FUNCTION AND FEMALE FERTILITY
Subsequent fertility is an important consideration in treating a thyroid canc tion that includes many young women. The exposure to the ovaries following a 15 mCi I3II dose is estimated at 21 cGy ( I ) . Despite this high dose, significant risk to long-term ovarian function and fertility has not been found (6,22,23).Raymond and associates (24) retrospectively studied ovarian function by history and serum g I3'Itherapy. Of 66 women, 18 had tempor pin concentration during the first year after amenorrhea, all resolving within 12 months after radioiodine. These women had high
y
Radioiodine
-II
159
serum FSH-LH levels after therapy. However, several small series fail to find a long(23) compared the fertility of 814 term effect on fertility. Dottorini and colleagues women who received 1311with 187 untreated women. They found no significant difference in fertility rate, birth weight, or prematurity. A detailed accountof the subsequent l3II was fertility and birth historiesof 40 young patients treated for thyroid cancer with (22). Theincidenceofinfertilitywas 12%, not reportedbySarkarandcoworkers (25) did find significantly different from the general population. Einhorn and associates an increase in chromosomal “breaks” in patients given radioiodine therapy, but no No consistent chromosignificant differencein their children’s chromosomes was found. somal aberration or malformation was noted in of the anychildren of patients previously given I3lI. The risks to female fertility appear to be small but can be further minimized by In addition, good hydration and frequent urination during and in the week after therapy. the subspecialty of infertility management could be consulted by concerned patients interested in ensuring future fertility and in their options before therapy. But, therapy should not be compromised by these concerns. PULMONARY FIBROSIS
Pulmonaryfunctionimpairmentfollowing 13’1 therapy for lung metastases from thyroid cancer is a serious side effect. Early studies at Memorial Sloan-Kettering Canc 5 of 59 patients treated with 1311, Center found radiation pneumonitis occurring in resulting in 2 deaths (14). Form these early studies, dose limitations have been established. They recommended that the 48 hour retention in the whole-body not exceed 80 mCi in patientstreatedwith I3II withfunctioninglungmetastases.Leeperand Shimaoka (11)report that following this guideline over the past20 years, no cases of radiation pneumonitis have been seen at Memorial Hospital. Respiratory impairment in these patients may not be caused by irradiation but, rather, the extensive and persi presence of the pulmonary metastases (6). It appears to be prudent and effective to limit the dose lS1Iof in the presence of lung metastases. This still permits high-dose therapy of pulmonary metastases while lessenin as an the risk of respiratory impairment. Steroids may also be used in special cases additional precaution (25A). ANAPLASTIC TRANSFORMATION Theconversion of awell-differentiatedthyroidcancertoanaplasticcancerhas been documented (26-28). However, a causal relationship between 1311 therapy and subsequent transformation has not been established and the conversion may be the natural history of the disease. It may also be attributed to the therapeutic success of 1311in destroying most differentiated thyroid cells and leaving the undifferentiated cells to multiply. This is unproven speculation, and no causal relationship between1311and anaplastic transformation of the thyroid cancer has been proven. LEUKEMIA The association of ionizing radiation and leukemia has been reported in various circumstances from Hiroshima and Chernobyl to controlled external radiation therapy
Johnston 160
and
Sweeney
and occupational exposures (29).A recent report of 2510 patients with thyroid cance 834 of whom were treated with 1311(481-50,320 MBq) found no increase in leukemia risk after a mean observation period of 14 years (30). Other retrospective series also I3' show no significant reporting of an increased incidence of leukemia following therapy (7,26). There have been several reports of individual cases of leukemia following '"I therapy 365patients and found 3cases of leukemia, each occurri (6,31,32). Edmonds reviewed within 4 years of the firstI3'I treatment with patients receiving 42-65 GBq (1100-1700 mCi) (6). This is extremelyhigh-dosetherapy,usuallyreservedforpatientswith 4 of 250 patients treated with I3I advanced metastatic disease. Pochin (32) found developed leukemia. However, his patients were treated with a very aggressive reg using 150 mCi doses, every 2-3 months for four or five or more cycles. With cumulative doses approaching 1 mCi or more, the incidence of leukemia ma increase, and patients must be informed of the risk. However, many of these patients requiring those doses have severe, aggressive, or widespread disease, and the benef of further treatment with I3lImay outweigh the risks. The routine dosing with I3'I for ablation and therapy with doses ranging from 100to 200 mCi, given at intervals great than 6 months with less than 800 mCi I3lI total dose, probably does not confer an increased risk of leukemia.
OTHER SOLID TUMORS There has been reported an increased incidence of salivary gland tumors in patients treated with l3lI(23,301. Two cases of lymphoma of the parotid glands have also b reported in consequence of I3'I therapy (33). The authors believe that this could be coincidental, but the fact remains that the salivary glands receive a high radiatio during therapy. Most case series are small, which makes evaluation of possible and acceptable risk This is compoundedbythesuggestionof an increased difficultinthesepatients. incidence of breast cancer (23) and adrenal gland tumors (30) in thyroid cancer treated with or without l3II. A small but significant increase in deaths from bladder cancer was also noted in a study of 258 patients treated with I3lI (6). These patients may be at a slightly increased risk for the development of secondary tumors. Our limited knowledge of the potential carcinogenic risks of therapeutic I3lIdoses should be admitted to the patient, along with a respect for its benefits. CENTRAL NERVOUS SYSTEM
Although brain metastases in thyroid cancer are rare (34,351, the treatment of int cerebral metastases can cause life-threatening complications. Datz (36) reports acute cerebral edema diagnosed afterI3'Itreatment in a patient with intracerebral metastas fromthyroidcancer.Thepatientexperiencedseizures,slurredspeech,andmuscle weakness 12hours after therapy. Sudden hemorrhage into an intracerebral metastasis has also been reported after I3'I therapy (37). Corticosteroids are used before external beam radiation to decrease the incidence of cerebral edema (38). Glycerol has also been used as an alternative therapy an be preferable in I3II-treated patients because corticosteroids may decrease iodine
Radioiodine Therapy-11
161
in thyroid tissue (36,38,39). Most importantly, brain metastases must be considered before therapy. Metastatic survey scans should include views of the head, and if there is advanced disease or neurological symptoms, an MRI or CT scan without contrast 1311therapy. Corticosteroidsor glycerol can then be administered should be done before appropriately. As always, the possibility of surgical removal of the thyroid metastasis should also be explored before 13'1therapy.
REFERENCES
1. MIRD Dose Estimate Report No. 5. Summary of current radiation dose estimates to humans from 1-123, 1-124, 1-126, 1-130, 1-131, 1-132 as sodium iodide. J Nucl Med 1975; 16:857. 2. Allweiss P, Braunstein GD, Kate A,et al. Sialoadenitis following 1-131 therapy for thyroid carcinoma: concise communication. J Nucl Med 1984; 25:755. 3. Spiegel W, Reiners C, BornerW. Sialoadenitis following iodine-131 therapy for thyroid carcinoma. [Letter]. J Nucl Med 1985; 26:816. 4. Creutzig H. Sialoadenitis following iodine-131 therapy for thyroid carcinoma. [Letter]. J Nucl Med 1985; 26:817. 5. Van Nostrand DV, NeutzeJ, Atkins F. Side effects of "rational dose" iodine-131 therapy for metastatic well-differentiated thyroid carcinoma. J Nucl Med 1986; 27:1519. 6. Edmonds C J , Smith T. The long-term hazards of the treatment of thyroid cancer with radioiodine. Br J Radiol 1986; 59:45. 7. Simpson W, Panzarella T, Carmthers J, et al. Papillary and follicular thyroid cancer: impact of treatment in 1578 patients. Int J Radiat Oncol Biol Phys 1988;141063. 8. Goolden AWG, Kam KC, Fitzpatrick MC, et al. Oedema of the neck after ablation of the thyroid with radioactive iodine. Br J Radiol 1986; 59583. 9. Arad E, O'Mara RE, Wilson GA. Ablation of remaining functioning thyroid lobe with radioiodine after hemithyroidectomy for carcinoma. Clin NuclMed 1993; 18:662. 10. Kahn S , Waxman A, Ramanna L, Ashok G, Nagaraq N, Braunstein G. Transient radiation effects following high dose I3II therapyfor differentiated thyroid cancer PTC). [Abstract]. J Nucl Med 1994; 35:15P. 11. Leeper RD, Shimaoka K. Treatment of metastatic thyroid cancer. Clin Endocrinol Metab 1980; 9:383. 12. AbbattJD, Brown WMC, Farran HEA. Radiation sickness in man following the administration of therapeutic radioiodine: relationship between latent period, dose rate and body size. Br J Radiol 1955; 28:358. 13.Varma V M , Dai WL, Henkin FU. Tastedysfunctioninpatientswiththyroidcancer following treatment withI3lI [Abstract]. J Nucl Med 1992; 33:996. 14. BenuaRS, Cicale N R , Sonenberg M, et al. The relation of radioiodine dosimetry to results and complications in the treatmentof metastatic thyroid cancer.AJR 1962; 87:171. 15. Glazebrook GA. Effect of decicurie doses of radioactive iodine-131 on parathyroid function. Am J Surg 1987; 154:368. 16. Haynie T, Beierwaltes W. Hematologic changes observed following therapy for thyroid carcinoma. J Nucl Med 1963; 4:85. 17. Werner S, Ingbar SH. The thyroid; 4th ed. Philadelphia: Lippincott-Raven, 1996. 18. Handlesman DJ, Turtle JR. Testicular damage after radioactive iodine ( 9 ) therapy for thyroid cancer. Clin Endocrinol 1983; 18:465. 19. Lushbaugh CC, Casarett GW. The effects of gonadal radiation in clinical radiation ther a review. Cancer 1976; 37:llll. 20. Ahmed SR, Shalet SM. Gonadal damage due to radioactive iodine (I-131) treatment for thyroid carcinoma. Postgrad Med J 1985; 61:361.
162
Sweeney and Johnston
21. Allweiss P, Braunstein GD, Katz A, et al. Sialadenitics following 1-131therapy for thyroi carcinoma: concise communication. J Nucl Med 1984; 25:755-788. 22. Sarkar SD, Beierwaltes WH, Gill SP, et al. Subsequent fertility and birth histories of children and adolescents treated with l3II for thyroid cancer. J Nucl Med 1976; 17:460. 23. Dottorini ME, Lomuscio G, Mazzucchelli L, et al. Assessment of female fertility and carcinogenesis after13'1therapy for differential thyroid carcinoma. J Nucl Med1995; 36:21. 24. Raymond JP, Izembart M, Marliac V et al. Temporary ovarian failure in thyroid cancer
J Clin Endocrinol Meta patients after thyroid remnant ablation with radioactive iodine.
1989; 69:186-190. 25. Einhorn J, Hulten M, Lindsten J, Wicklund H, Zetterqvist P. Clinical and cytogenetic
investigation in children of parents treated with radioiodine. Acta Radio1 Ther Phys
1972; 11:193. 25A. Edmonds CJ. Treatment of thyroid cancer. Clin Endocrin Metab1979; 8:223-242. 26. Beierwaltes WH. The treatmentof thyroid carcinoma with radioactive iodine. Semin Med 1978;8:79. 27. Crile G, Wilson DH. Transformation of a low-grade papillary carcinomaof the thyroid to an anaplastic carcinoma after treatment with radioiodine. Surg Gynecol Obstet 1959; 108:357. TP, et al. Pulmonary metastasis of differentiated thyr 28. Samaan NA, Schultz PN, Haynie carcinoma: treatment results in 101patients. J Clin Endocrinol Metab 1985; 65:376. 29. Hall P, Boice J, Berg G, et al. Leukemia incidence after iodine exposure. Lancet 1991; 340:1. 30. Hall P, Holm LE, Lundell G, et al. Cancer risks in thyroid cancer patients. BrJ Cancer 1991; 64:159. 31. Brincker H, Hansen HS, Andersen AP. Induction of leukaemia by 131-1treatment of thyroid carcinoma. Br J Cancer 1973; 28:232. 32. Pochin EE. Radioiodine therapy of thyroid cancer. Semin Nucl Med 1971; 1:503. 33. Wiseman JC, HalesIB, Joasoo A. Two casesof lymphomaof the parotid gland followi ablative radioiodine therapy for thyroid carcinoma. Clin Endocrinol1982; 17:85. 34. Maheshwari YK,Hill CS, HaynieTP, et al.I3'I therapy in differentiated thyroid carcin M.D. Anderson Hospital experience. Cancer 1981; 47:664. 35. Mazzaferri EL, Young E. Papillary thyroid carcinoma: a 10-year follow-up report of t impact of therapy in 576 patients. Am J Med 1981; 70:511. 36. Datz FL. Cerebral edema followingiodine-131therapy for thyroid carcinoma metasta to the brain. J Nucl Med 1986; 27:637. 37. Holmquest DL, Lake P. Sudden hemorrhage in metastatic thyroid carcinoma of the br during treatment with iodine-131. J Nucl Med 1976; 17:307. 38. Bedikian AY, Valdivieso M, Heilbrun LK, et al. Glycerol: a successful alternative to
dexamethasone for patients receiving brain irradiation for metastatic disease. Cancer T Rep 1978;62:1081. 39. Datz FL. Gamuts in nuclear medicine, 3rd ed. St. Louis: Mosby-Year Book, 1995.
14 Recombinant Human Thyrotropin Matthew D. Ringel
Initial management of patients with thyroid cancer generally includes surgical thyro or malignant) with radioactive iodine, ectomy, eradication of iodine-avid tissue (benign and long-term treatment with L-thyroxine at doses sufficient for suppression of pituitary production of thyrotropin (TSH) (1,2). Thyroid cancer will recur in 20% to 40% of patients, requiring long-term monitoringtumor for recurrence or progression (3).Similar to most other malignancies, monitoring is performed primarily by three procedures; physical examination, measurement of tissue or tumor-specific serum markers, and radiographic imaging. Measurements of serum thyroglobulin concentrations and radioiodine whole-body imaging are used most frequently to monitor thyroid cancer patients (I$). Both of these modalities measure relatively thyroid-specific functions. However, the sensitivities of iodine scanning and thyroglobulin measurement are limited by the small relative amount of thyroid tissue present in patients treated by thyroidectomy and dedifferentiation of tumor cells compared to normal thyrocytes. Therefore, for optimal sensitivity, both radioiodine imaging and serum thyroglobulin measurement require stimulation of thyroid tissue by elevated levels of TSH. Moreover, elevated serum concentrations of TSH are also required for radioiodine therapy. To attain the elevated serum TSH concentrations required for accurate monitoring, protocols have been designed to stimulate endogenous pituitary TSH production and secretion. Most commonly, L-thyroxine is withdrawn 4-6 weeks before radioiodine scanning and serum thyroglobulin measurement. To limit the duration of symptomatic hypothyroidism,patients are frequently treated with triiodothyronine (T3), an agent 2 to 3for weeks following discontinwith a shorter circulating half-life than L-thyroxine, (>30 uation of thyroxine. Most patients attain an adequate serum TSH concentration mUA) 2 to 3 weeks after discontinuation of T3,allowing for scanning and thyroglobulin measurement ( I ) . Several days after scanning and/or therapy, one or both types of thyroid hormone are restarted. Usingthis paradigm, patients are clinically hypothyroid for approximately4-8 weeks, which resultsin substantial morbidity, including lethargy, depression, irritability,and limitation in ability to work (5-7). Moreover, elevated TSH levels for extended periods of time have been associated with rapid growth of metastat tumor tissue, resulting in clinical compromise, particularly among patients with central nervous system metastases (8,9).For these reasons, effective alternative methods for (10,11) or do thyroid cell stimulation that require limited thyroid hormone withdrawal From: Thyroid Cancer: A ComprehensiveGuide to Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, NJ
163
164
Ringel
not require thyroid hormone withdrawal at all, have been sought for many decades (12).In this chapter, the history of alternative forms of thyroid cell stimulation using exogenous thyroid-stimulating agents is reviewed with particular emphasis on the re development of recombinant human TSH (rhTSH). EXOGENOUS THYROTROPIN-RELEASING HORMONE
Initial attempts to stimulate endogenous production of pituitary TSH production without thyroxine stimulation were performed using exogenous thyrotropin-releasing hormone (TRH) administered eitheras intramuscular (IM) or intravenous injections, or as an oral preparation. When administered N,TRH is rapidly inactivated, with a half-life of 4-5 minutes.TSHpeaksapproximately 20 to 30 minutes following administration of TRH in normal individuals, but this response is blunted in patients (13-15). Repeated doses and infusi with hyperthyroidismor on thyroxine suppression have been shown to enhance the TSH response to TRH, but this agent has proven to be too cumbersome for clinical use (16). Several groups subsequently evaluated oral TRH as an adjunct to standard thyroxin withdrawal or as a method to elevate the serum TSH concentration while patients remain on their L-thyroxine suppression (16-18). Longer periods of TSH elevation were observed with oralTRH, compared to IV or IM TRH administration, particularly when usedto augment the TSH rise of thyroxine withdrawal. However, TRH administration alone was less effective in stimulating iodine uptake than standard thyroid h (17).When used in withdrawal, despite similar rises on serum TSH concentrations combination with thyroxine withdrawal and lithium carbonate,TRH oraladministration did not enhance the iodine uptake compared to withdrawal only (19). More recent data suggest that the glycosylated forms of TSH secreted after acute stimulation with intravenousTRH may differ from the usual circulating forms of TS (20).Human TSH contains three asparagine-linked oligosaccharide chains that, when fully processed, terminate either with sialic acid linked to galactose, or with sulfate attached to N-acetylgalactosamine. Two of these oligosaccharide chains are attached to the a subunit and one is attached to the p-subunit. The biological importance of theseglycosylatedformsremainstobecompletelydetermined;however,different forms exert different cellular effects in vitro, have unique metabolism and serum lives, and have specific affinities for association with the a-subunit (21-27).Therefore it is possible that the formsof TSH released followingTRH stimulation may not hav equivalent biological activity to those present with a more gradual development of hypothyroidism, or in the absence of T3. This may explain the apparent dissociation between iodine uptake and serum TSH concentration following TRH administration Studies evaluating the importance of glycosylation pattern on in vivo and in vitro function of rhTSH are detailed below.
w)
BOVINE THYROTROPIN
Seidlin and colleagues(28)and Stanley and Astwood(29)first reported administra tion of bovine TSH to stimulate radioiodine uptake in humans. Benua and colleagues (30) subsequentlyreportedtheir18-yearexperiencewithbovineTSHtoaugment iodine uptake in patients already hypothyroid following thyroidectomy and ablation
Recombinant Human Thyrotropin
165
usinga2-daydosingregimen.Twentypatientswerestudied,andonlyaminimal increase in iodine uptake was seen; however, these patients were already hypothyroid at the time ofthebovineTSHadministration.Theadministration of bovine TSH as analternativetothyroidhormonewithdrawalin duringthyroxinetherapy preparation for radioiodine scanning was first reported in 1953 by Sturgeon et al. (31); and again by Catz et al. (32,33). in 1959These reports suggestedthat bovine TSH administration may be an acceptable alternative to thyroid hormone withdrawal in selected ca In addition, a cellular rationale for use of bovine TSH in patients was obtained when similar binding and activity was seen for human and bovine TSH in the chick bioassay (34). Schneider and coworkers (35) subsequently showed similar enhancement of thyroid iodine uptake in normal subjects following injection of either bovine or human pituitary TSH. The similar activitiesof bovine and human TSH provided a basis for clinical studies designed to evaluate the efficacy of bovine TSH-stimulated radioiodine scanning and treatment in patients with thyroid cancer during thyroid hormone therapy and after a period of thyroxine withdrawal. Pharmacokinetic studies showed a peak serum TSH concentration 4 hours after IM administration of bovine TSH, and that by 10 hours, 50% (36). Initial resultsinthyroidcancer serumconcentrationshaddecreasedby patients suggested that administrationof bovine TSH was effective, although it did not stimulate iodine uptake as well as thyroid hormone withdrawal (36,37). Local and systemic adverse events were associated with bovine TSH administration, (38,39). These including local induration, nausea, vomiting, urticaria, and anaphylaxis (37). Because of these allergic were particular evident in patients treated multiple times reactions and the diminished effectiveness of the agent with repeated doses, patients were studied for the development of neutralizing antibodies to bovine TSH. Detectable circulating neutralizing antibodies developed in the majorityof patients who received multiple dosesof bovine TSH(4042). These anti-bovine TSH antibodies also interfered with measurement of endogenous TSH, hindering the ability to monitor patients for (4347). With the development of efficacy of thyroid hormone suppression therapy specific immunoassays against human TSH, it was determined that these antibodies were either directed against bovine TSH alone, or cross-reacted with bovine and hum TSH (42,44). Therefore, there was concern that these antibodies may modulate the bioactivity of endogenous TSH as well as bovine TSH, limiting subsequent use of radioiodine therapy with either methodof stimulation. A series of in vitro bioassay studies confirmed that the antibodies generated by bov (40,42). However, the TSH were partially neutralizing to both bovine and human TSH effect on bovine TSH bioactivity was more pronounced. Due to the combination of relative ineffectiveness with multiple dosing and the development of antibodies, bovine this agent is currently not available in the United TSH use subsequently diminished, and States for clinical use.
HUMAN THYROTROPIN Human Pituitary TSH Human pituitary TSH was proposed to be useful in preparing patients for radioiodine scanning. Studies reporting kinetics in humans showing effective stimulation of thyroid
166
Ringel
hormone production and iodine uptake (48-51). In the early 1980s several cases of Creutzfeld-Jakob syndrome were reported in patients treated with human pituitary growth hormone (52). In addition,thepurityofthehumanTSHpreparationwas questioned. Although useful in the laboratory as a standard, human pituitary TSH is unlikely to be clinically utilized secondary to the unacceptable potential risk of slo virus transmission.
Recombinant Human TSH In Vitro Studies The cloning of the gene encoding the human TSH-P subunit (53,54) raised the possibility of producing recombinant human TSH using molecular techniques. After several years, bioactive recombinant human TSH was successfully manufactured by several groups by cotransfecting mammalian cells with complementary DNAs e (21,23,55both the common human commona subunit and the human TSH-P subunit 58). As noted above, the human TSH proteinis glycosylated at three sites, two on the a subunit and one on the p subunit. The glycosylated forms of TSH found in the pituitary are heterogeneous and may differ from the predominant circulating forms. Because bacterial cells do not possess the enzymes necessary for protein glycos the use of more labor-intensive mammalian gene expression systems were required. In this chapter, rhTSH derived from transfected Chinese hamster ovary (CHO) cells is discussed, since it has been manufactured commercially in large amounts and has been used in clinical studies. The recombinant protein was isolated from the cotransfected CHO cells and pu by several methods (55-58).In vitro activity and the chemical structure of rhTSH compared to the international human pituitary TSH standards utilized in clinical a BindingstudiesrevealedthatrhTSHhadhighaffinityforhumanTSHreceptors (59) and for human TSH recept expressed endogenously on human fetal thyroid cells expressedonChinesehamsterovarycellstransfectedwithTSHreceptorcDNA (25,26,55-58). Moreover, rhTSH binding was not species-specific, displaying re high affinities for both endogenous rat TSH receptors expressed on Fisher rat thyr cells (FRTL-5) and mouse TSH receptors (25,26,62). Recombinant human TSH was also functionally active in all TSH receptor-expressing types cell analyzed both in vitr and in vivo (25,26,5562). These in vitro studies led to several in vivo animal studies in mice and rats(6062) and primates (63) that revealed enhanced serum TSH concentrations and radioiodine of the differencesin activity uptake following administration of rhTSH. Careful analysis between the batches of recombinant human TSH both in vitro and in vivo were the performed. In those studies,it became apparent that there was a poor correlation in vitro and in vivo activity and that differences in the glycosylation patterns of the batches were responsible. As noted above, three oligosaccharide chains are attached to the endogenousa and p subunits of TSH that terminate either with a sialic acid bound to a galactose resi or a sulfate bound to N-acetylgalactosamine. Because the sulfotransferases and GalN transferases are found only in pituitary cells, rhTSH produced in cells lacking these enzymes, such as CHO cells, contains only the sialylated form. Both human pituita TSH and rhTSH contain a heterogeneous mixture of glycosylated forms. The di
Recombinant
167
in bioactivity between the different batches of rhTSH appeared to vary with the pH of (61). thereactionconditionsinthebioreactorusedformanufacturingtheprotein Specifically, the in vitro bioactivity was higher for the more basic, sulfated forms of rhTSH compared to the more acidic, sialylated forms. However, the in vivo bioactivity of the sialylated form was greater than sulfated form, presumably related to its longer serum half-life. The sulfated formis excreted in the kidneys and has a relatively short half-life, while the sialylated form is hepatically metabolized, resulting in a longer (20) identified sialylated serum half-life(22-26,61). In addition, Magner and colleagues TSH as the predominant circulating TSH glycoprotein. Therefore, in addition to its may be more similar to circulating endogenous greater in vivo activity, sialylated rhTSH TSH than either sulfated rhTSH or human pituitary extract. Clinical Studies SWIES IN NORMAL SUBJECTS While several studies of recombinant human TSH were performed in euthyroid animals, only one published study has been reported in normal human subjects. Ram et al. (64) evaluated six euthyroid subjects with no prior history of thyroid disease, normal thyroid physical examinations, and no biochemical evidence of thyroid disease. 3 Thesubjectsreceived 0.6 units ofrecombinanthumanTSHintramuscularlyon consecutive days. Serum TSH, T4, T3, free thyroxine index, and thyroglobulin were monitored every 4 hours for the first 12 hours, at 24,72, and 96 hours, and 7 days after administration of the dose. The development of antibodies against human TSH 1.3 U/ml following the injections was also assessed. Serum TSH rose from a of baseline to a mean of 40 U/mlin 4 hours and peaked after 24 hours. It decreased to below baseline 7 days after the injection. Serum T3 and T4 concentrations showed similar patterns except the peak occurred after48 hours with continued elevation (still within the normal range) after 1 week. Serum thyroglobulin also rose following recombinant TSH administration, but the maximal rise did not occur until48 to 72 hours after the dose. The medication was well tolerated and no patients developed anti-TSH antibodies. Radioiodine uptake was not measured in this study of normal subjects. SWISS IN PATIENTS WITH THYROID CANCER Several studies and case reports of use of rhTSH in patients with thyroid cancer (9,6567). The focus of this section is on the have been published in the literature Phases I, II, and 111 clinical trials that address rhTSH administration for diagnostic scans and measurements of serum thyroglobulin. It should be noted that the second Phase 111 clinical trial was not yet published at the time of this chapter; however, the investigators generously provided data and preliminary analysis from that study. In these three studies, patients were prepared first by recombinant human TSH dur thyroid hormone therapy and second by thyroid hormone withdrawal because of ethical considerations. Randomization of scan order would subject some patients to a second period of hypothyroidismin preparation for radioiodine therapy. Therefore, the possibility of reduced sensitivity of the withdrawal scan secondary to “stunning” by the first scanning dose must be considered as a potential confounding factor. Several studies, and clinical experience, suggest that the incidence of “stunning” is low, particularly when using low (2-5 mCi) scanning doses; thus the effects of this bias are likely to be minimal (1,7).
2 68
Ringel
In 1994, Meier and coworkers (67) performed a Phase I/II clinical trial designed to comparetheefficacyandpharmacokineticsofvariousdosingregimensofrhTSH administration on iodine uptake and serum thyroglobulin concentrations in patients with thyroid cancer. In addition, they also compared the efficacy of the various rhTSH preparation regimens with standard thyroid hormone withdrawal. They evaluated 19 patients with differentiated thyroid cancer. All patients were treated with triiodothy nine (T3) for an average of 37 days before receiving recombinant human TSH. Suppressed serum TSH concentrations were documented in these patients, and they were randomized to receive a single intramuscular injection of recombinant TSH (10, 20, 30, or 40 units) or multiple doses (2 or 3) of 10 units or 2 doses of 20 units at 24-hour intervals while they remained on T3. Laboratory evaluation included serum concentrations of TSH, thyroglobulin, free T4, total T3, and antithyroglobulin an 1311 were performed 48 Diagnostic whole-body radioiodine scans using 1-2 mCi of hours after the last rhTSH injection. After the 13'1 scan, patients were withdrawn from T3 foran average of 29 days until the serum TSH concentration was above 30 U/ml. Patients then received a second diagnostic whole-body 13'1 scan. Patients were treated as clinically indicated based upon the results of the scans and serum thyroglobulin concentrations. Diagnostic scans using the two preparations were compared by dent, blinded nuclear medicine physicians, and then later as paired samples in which the reviewers were blinded to the order and dates of the two scans. The pharmacokinetic study revealed that serum TSH concentrations were ma elevated with higher doses of recombinant human TSH, but that the lower, 10-unit dose resulted in mean serum TSH concentrations similar to withdrawal (127 U/ml versus 77 mUh, respectively) after one dose with a greater peak after the second d (mean value: 220 mull). The TSH elevation was maintained for a longer period of time with a multiple injection schedule. In the blinded review of scans, radioiodine scans were read as equivalent in 17 of 19 (89%) patients. In two patients, the withdrawal scans were considered superior.In the paired evaluation, scans were of equivalent quality in 12 of 19 cases;in four cases the rhTSH scan was superior and in three cases, the withdrawal scan was superior. to withdrawal preparatio The iodine uptake was lower in the rhTSH scans compared in 72% of patients regardless of dosing regimen. The uptake was similarin the group prepared with one or two doses of 10 units and one dose of 20 units. There was no correlation between degree of TSH elevation and the percentage uptake between the rhTSH groups. Retention of the 1311dose in the neck was measured in seven patient A twofold greater dose retention was demonstrated after thyroid hormone withdrawa This difference was corrected by controlling for w than after rhTSH administration. body retention. Thus, the likely cause of the longer retention time in the thyr in withdrawal scans was thought to be reduced metabolism and clearance of iodine the hypothyroid subjects. Serum thyroglobulin concentrations also increased in response to the recombinant human TSH. Similar to the response in normal euthyroid subjects, maximal serum concentrations in the thyroid cancer patients occurred 4872and hours after administra patients tion. Serum thyroglobulin concentration increased more than twofold inof79% after thyroid hormone withdrawal compared to 58% o f patients after rhTSH. No data
Recombinant Human Thyrotropin
Day 1
Day 2
SerumTSH, Tg, HCG
169
Day 3 1311 Dose,
SerumTSH
Day 4
Day 5 Serum TSH, Tg, 1311 Scan
0.9 mg rhTSH IM
Fig. 1. Recommended dosing regimen for rhTSH: 0.9 mg of rhTSH (bioequivalence is 10 U/mg protein, Second World Health Organization International Reference Preparation, Thyrotro pin, Human, for Bioassay) is administered on two consecutive days. Based on prior studies, the maximal rise in serum TSH occurs24 and 48 hours after the last doseof rhTSH, and the maximal rise in serum thyroglobulin (Tg) concentrations occurs72 hours after the last dose. Pregnancy tests (serum HCG) should be obtained from all women of childbearing age before rhTSH administration.
are provided about the frequency of lesser elevations of thyroglobulin or the correlation between withdrawal and rhTSH-induced elevations; 4 of 19 patients with circulating antithyroglobulin antibodies were included in the study. None of the patients in the study showed detectable levels of circulating antibodies against human TSH. Quality of life assessment using both the Billewicz Scale (6) to assess hypothyroid symptoms and the Profile of Mood State Comparison (68) to assess changes in mood and other psychological symptoms revealed more frequent abnormal scores during thyroid hormone withdrawal. Thus, this Phase I/II study showed that, in most patients, rhTSH preparation for diagnosticwhole-bodyscans was asefficaciousaswithdrawalscanning.However, lower neck uptake and retention of isotope and lower rises in serum thyroglobulin were seen following rhTSH preparation. Patients tolerated the medication well and had far fewer symptoms compared to withdrawal preparation. The two-injection 10-unit regimen produced similar rises in TSH to higher dose regimens and was well tolerated. this Phase I/II trial, a PhaseIII trial was initiated to further Based upon the results of compare the diagnostic utility of rhTSH with standard withdrawal scanning. Ladenson and colleagues (7) reported the results of a similarly designed study of 152 patients with thyroid cancer who received rhTSH, 0.9 mg intramuscularly, on 2 consecutive days during thyroid hormone suppression therapy with either/or L- T~and T3, followed 4 to 6 weeks later (Fig. 1). Thyroxine supby a thyroid hormone withdrawal scan pression was confirmed by serum TSH concentrations. Patients received 2- to 4-mCi doses of radioiodine 1 day after the second dose of rhTSH and were scanned 2 days In 35 patients, later. Serum concentrations of TSH and urinary iodine were measured. serum thyroglobulin and antithyroglobulin antibodies were also measured. Whole-body
170
Ringel
radioiodine scans were interpreted by three independent reviewers in a blinded andtheresultswere compared. Hypothyroid symptoms and mood alterations were (6) and the Profiles of Mood States Comparison measuredbytheBillewiczScale (68),respectively. Of the initial 152patients enrolled, 127 were included in the study evaluation. The majority of patients not included in the analysis were excluded for undefined protocol violations. Mean serum TSH concentrations were 132 mUA 24 hours after the second rhTSH dose and 101mUA following thyroid hormone withdrawal. In 51% of patients, scans revealed no uptake in both the withdrawal and rhTSH prepared scans. Among the 62 patients with uptake identified on one or both scans, 45 had thyroid bed uptake, 10 had cervical metastases, and 7 had distant metastases. Scans were considered discor if additional areas of uptake were seen on one scan compared to the other, even if 66% change in tumor stage occurred. RhTSH and withdrawal scans were concordan of the patients with positive scans. The rhTSH scan was superior in 5%, and the withdrawal scan was superiorin 29%. Tumor stage was altered by the scan discord in 6 or 17 patients with metastases. Including the concordant negative scans, the concordance rate for the 127 patients was 83%. rhTSH scans were superior in 3% of cases and withdrawal scans were superior in 14% of cases. J/II study, local neck uptake was lower after rhTSH prepa Similar to the Phase but when normalized for the differences in whole body retention ofI3lI, no differenc was noted. Symptoms of hypothyroidism were significantly less common at the time of the rhTSH administration than after thyroid hormone withdrawal. Serum cho triglyceride, uric acid, and creatinine concentrations were also lower at the time of rhTSH stimulation than following withdrawal of thyroid hormone. Serum thyroglobulin concentrations were measured in 35of the patients. After rhT administration, thyroglobulin values were highest 72 to 96 hours after the first dose. Thyroglobulin rose to a value greater than 5 n g / d in 13 patients after rhTSH and in 14 patients after withdrawal. No patients developed anti-TSH antibodies, including seven patients who previo 48in of 152subjects received rhTSHin the PhaseI/IIstudy. Adverse events were noted The most frequent adverse effect was nausea, which occurred in 25 patients and was generally mild and self-limited. This Phase III trial using a two-dose regimen demonstrated that among patient recurrent or residual thyroid tissue, rhTSH preparation of patients for radioiodine scanning resulted in inferior scans in 29% of cases. This frequency of inferior scans was concerning and several of these patients were treated differently based upon the discordant scan. However, many of them may have also been identified as requiring 1311 treatment based on their rhTSH-stimulated serum thyroglobulin concentrations. Measurement of an rhTSH-stimulated thyroglobulin appeared to be quite sensitive concordant with thyroid hormone withdrawal-stimulated thyroglobulin. Unfortunat this laboratory test was obtained from only 35 of the 127 patients in this study. Most patients tolerated the rhTSH well, and symptoms of hypothyroidism were drama reduced with the use of rhTSH. Several reasons could account for the greater sensitivity of withdrawal scans compared to rhTSH scans: 1)reduced clearance of theI3'I in hypothyroidism present afte
Recombinant Human Thyrotropin
171
withdrawal results in a higher bioavailability for the iodine-avid tissue, and 2) the longer duration of the elevated TSH levels after withdrawal may be important for maximally stimulating iodine uptake (11). To further define a potential role for rhTSH as a monitoring agent and to reevaluate the dosing regimen, a second Phase 111 clinical trial comparing the two-injection regimen to a three-injection regimen in 226 patients was recently completed. This study also was designed to better evaluate the sensitivity of rhTSH-stimulated thyroglobulin concentrations. The protocol was designed in a similar manner to the study of Ladenson or and coworkers (7) except that patients were randomized to receive either two three 0.9-mg doses of rhTSH intramuscularly. Patients received 1311 24 hours after the last dose of rhTSH, were scanned using2-4 mCi I3’I2 days later, and serum thyroglobulin 72 hours after the last dose of rhTSH. Following concentrations were measured 48 and this scan, patients were withdrawn from thyroid hormone for diagnostic scans, serum thyroglobulin measurements, and treatment as needed. In this study, scan discordance was defined as uptake on one scan that altered the stage of disease. Using this definition, the overall concordance rate between rhTSH and withdrawal 13’1whole-body scans was 89%. Withdrawal whole-body scans were interpreted as superior in 8% of cases and rhTSH scans were superior in3% of cases. No statistically significant difference between the accuracy of rhTSH-stimulated scans and withdrawal scans were identified. Moreover, no statistical differences were seen between the two rhTSH preparation regimens. Combined data from the two Phase III trials comparing the utility of the two- and three-injection regimens for radioiodine scanning versus withdrawal scanning are summarized in Table 1A. Serum thyroglobulin measurements were measured 48 and 72 hours after the last dose of rhTSH and following thyroid hormone withdrawal using a highly sensitive radioimmunoassay. Serial samples from individual patients were measured on the same assay. Analysis using different values of thyroglobulin to identify disease presence (detectable iodine-avid tissue on diagnostic andor posttherapy scan) was performed for both basal and stimulated values. The lowest concentration that provided greatest accuracy for stimulated thyroglobulin concentrations using rhTSH or thyroid hormone 3 ng/ml or greater, the sensitivity withdrawal was determined to 3 beng/ml. At values of and specificity were 72% and 95% for rhTSH-stimulated thyroglobulin and 71% and 100%for the withdrawal-stimulated thyroglobulin measurements. Patients with circulat ing antithyroglobulin antibodies were excluded from this analysis. In general, serum thyroglobulin rose to similar levels following rhTSH stimulation and thyroid hormone withdrawal. The interpretation of these thyroglobulin data are dependenton the reproducibility of the thyroglobulin assay at lower values, a factor that varies greatly betwe different commercial laboratories.This is particularly critical when interpreting values or 10 ng/ml as a “positive in the2 to 10 ng/ml range. Using a TSH-stimulated value5 of value,” the combination of rhTSH stimulated thyroglobulin and scan was 94% sensitiv and 93% specific in predicting iodine-avid tissue on subsequent withdrawal and/or post23 ng/ml was usedin combination therapy scan. When a stimulated thyroglobulin value 32 patients with cervical or distant metaswith scanning, rhTSH stimulation identified all IIIstudy evaluating the accuracy of combining rhTSH tases. Data from the second Phase scan and thyroglobulin measurement to identify metastases are summarized 1B. in Table
Recombinant Human Thyrotropin
173
Summary of Recombinant Human TSH in Thyroid Cancer Based upon the three controlled clinical studies performed in a limited number of patients, it appears that rhTSH is a safe alternative to thyroid hormone withdrawal in the detection of recurrent or residual thyroid cancer. There also appears to be no significant advantage of the three-injection regimen as compared to the two-injection regimen. RhTSH preparation for I3lI scanning appears to be less sensitive than thyroid hormone withdrawal; however, the combined use of scanning and measurements of thyroglobulin improves the sensitivity of rhTSH monitoring. RhTSH preparation avoids the severe, transient hypothyroidism that occurs with thyroid hormone withdrawal. The advantages must be balanced against the risk of recurrence in each individual diagnostic patient, and itis important to realize that the rhTSH has been studiedinonly testing and not radioiodine therapy. Therefore, patients with high-risk tumors, or pa you are preparing for postthyroidectomy ablation, may not be appropriate for rhTSH screening. In addition, reliance on serum thyroglobulin measurements depends on the absence of antithyroglobulin antibodies and the sensitivity and reproducibility of the p ticular thyroglobulin immunoassay, factors that vary greatly between different laboratories. Larger clinical studies are needed to adequately address several questions regarding the clinical use of rhTSH. First, what is the sensitivity of rhTSH in patients with poorly differentiated tumors; and second, is there a role for rhTSH stimulation1311for treatment in selected individuals.A critical issueis standardization of thyroglobulin assays, since the most effective use of rhTSH relies heavily on an assay with high reproducibility A suggested algorithm for use of rhTSH asa diagnostic agent while in the lower range. monitoring patients for recurrence of thyroid cancer based upon the current data is 2. Similar algorithms were recently published (69,70), and determinadepicted in Figure tion of the optimal use of rhTSH in thyroid cancer requires further study and more extensive clinical use. Recombinant Human TSH in Other Conditions No clinical trials have been reported using rhTSH to facilitate therapy for thyroid diseases other than thyroid cancer. Clinical trials of using rhTSH as preparation for radioiodine therapy of toxic and nontoxic, large goiters is ongoing. RhTSH may be of particularly useful if the overall iodine uptakeof the goiter is low. In addition, it may prove to be a useful adjunct ‘=I to scanning in patients with poor quality scans perhaps related to iodine exposure. In the determinationof serum TSH concentrations, rhTSH may be a more standardized and easily replenishable sourceof concentration controls than human pituitary TSH (71,72). Future Directions More clinical experience and clinical trials are needed for a full assessment of the utility of rhTSH in monitoring patients with thyroid cancer. Recent reports by Skudli (74) raise the possibility of developing and colleagues(73) and Grossman and colleagues al. TSH receptor superagonists with enhanced effects on iodide uptake. Grossman et inserted four mutations in the common a subunit and three mutations in the TSH-P subunit in locations based on the crystallographic structure of HCG and homology w TSH. This mutatedTSHproteindisplaysa1000-foldgreateraffinityfortheTSH in vivo activity than wild-type TSH. Although this receptor and a 100-fold greater
:r
n
fm c
174
Recombinant Human Thyrotropin
175
agent has not been tested in humans, similar superagonists of the TSH receptor may improve the sensitivity of rhTSH preparation for diagnostic testing in thyroid cancer and other disorders (75). Summay The development of recombinant human TSH as an adjunct for diagnostic testing in patients with thyroid cancer may enable physicians to limit the morbidity associated with monitoring techniques. Although there are some concerns regarding sensitivity of the agent, for many patients with low-risk tumors, screening for recurrence by rhTSH stimulated thyroglobulin and1311scanning may be appropriate. Mutated formsof TSH with enhanced bioactivity may improve the sensitivity of monitoring techniques that do not require thyroid hormone withdrawal. Further studies and more extensive clinical this exciting new agent in the experience are needed to fully delineate the role for detection and treatment of thyroid cancer and other thyroid conditions.
REFERENCES
1. Singer PA, Cooper DS, Daniels GH, et al. Treatment guidelines for patients with thyroid nodules and well-differentiated thyroid cancer. Arch InternMed 1996; 156:2165-2172. 2. Solomon BL, Wartofsky L, Burman KD.Current trends in the management of well differentiated papillary thyroid carcinoma. J Clin Endocrinol Metab1996; 81:333-339. 3. Mazzaferri EL, Jiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular cancer.Am J Med 1994; 89:418-428. 4. Ozata M, Suzuki S , Miyamoto T, et al. Serum thyroglobulin in the follow-upof patients with treated differentiated thyroid cancer.J Clin Endocrinol Metab 1994; 778:188-196. 5. Dow KH, Ferrell BR, Anne10 C. Quality of life changes in patients with thyroid cancer after withdrawal of thyroid hormone therapy. Thyroid 1997; 7:613-619. 6. Billewicz WZ, Chapman RS, CrooksJ, et al. Statistical methods applied to the diagnosis of hypothyroidism. Q J Med 1969; 38:255-266. Mazzafeni EL, et al.Comparisonofadministrationof 7. LadensonPW,BravermanLE, recombinant human thyrotropin with withdrawal of thyroid hormone for radioactive iodine scanning in patients with thyroid carcinoma. N Engl J Med 1997; 337:888-896. 8. Stakianakis GN, Stillman TG, GeorgeJM. Thyroxine withdrawal in thyroid cancer. Ohio State Med J 1975; 71:79-82. 9. GoldbergLD,Ditchek NT. Thyroidcarcinomawithspinalcordcompression.JAMA 1981; 245~953-954. 10. Guimaraes V, Degroot LJ. Moderate hypothyroidism in preparation for whole body 1311 scintiscans and thyroglobulin testing. Thyroid1996; 6:69-73. 11. Maxon HR.Recombinant human thyrotropin symposium: Detection of residual and recurre thyroid cancer by radionuclide imaging. Thyroid1999; 9:443-446. 12. Robbins J. Recombinant human thyrotropin seminar: Pharmacology of bovine and human thyrotropin: an historical perspective. Thyroid 1999; 9:451-453. 13. Wide L, Dahlberg PA. Quality requirements of basal S-TSH assays in predicting S-TSH an response to TRH. Scand J Clin Lab Invest 1981; 40:lOl. 14. Sawin CT, Hershman JM. The TSH response to thyrotropin-releasing hormone(TRH) in young adult men: intra-individual variation and relation to basal serum TSH and thyroid hormones. J Clin Endocrinol Metab 1976; 425309-816. 15. Spencer CA, Schwarzbein D, Guttler R B , LoPresti JS, Nicoloff JT. Thyrotropin (TSH)releasing hormone stimulation test responses employing third and fourth generation TSH assays. J Clin Endocrinol Metab 1993; 76:494498. 16. Fairclough PD, Cryer RJ, McAllister J, et al. Serum TSH responses to intravenously and
176
17. 18. 19. 20.
Ringel
orally administered TRH in man after thyroidectomy for carcinoma of the thyroid. Clin Endocrinol 1973;2:351-359. Wenzel KW, Meinhold H, Bogner U, Adlkofer F, Schleusener H. Serum TSH levels in thyroidectomized patients after withdrawal of thyroid hormone therapy or oral administr of TRH. Acta Endocrinol 1973; 173(Suppl):15. Eissner D, Hahn K, Grimm W. Oral TRH stimulation in patients with thyroid carcinoma. ROFO 1983;138:95-100. Ang ES, TheHS, Sundram F X , Lee KO. Effect of lithium and oral thyrotrophin-releasing hormone (TRH) on serum thyrotrophin (TSH) and radioiodine uptake in patients with we differentiated thyroid carcinoma. SingaporeMed J 1995; 36:606-608. Magner JA, Kane J, Chou ET. Intravenous thyrotropin (TSH)-releasing hormone releases human TSH that is structurally different from basal TSH.J Clin Endocrinol Metab 1992;
74:1306-1311. 21. Schaaf L, Leiprecht A, Saji M, Hubner U, Usadel KH, Kohn LD. Glycosylation variants of human TSH selectively activate signal transduction pathways. Mol Cell Endocrinol
1997;132:185-194. 22. Szkudinski MW, Thokatura N R , Weintraub BD. Subunit-specific functions of N-linked 23.
24. 25. 26. 27. 28. 29.
oligosaccharides in human thyrotropin: role of terminal residues of alpha and beta subun oligosaccharides in clearance and bioactivity. Proc Natl Acad USA 1995; Sci 92:9062-9066. Cannone C, Papandreou MJ, Medri G, Verrier B,Ronin C. Biological and immunochemical characterization of recombinant human thyrotrophin. Glycobiology 1995; 5473481. SzkudlinskiMW, ThokaturaN R , Bucci I, et al. Purification and characterization of recom nant human thyrotropin (TSH) isoforms produced by Chinese hamster ovary cells: the r of sialylation and sulfation in TSH bioactivity. Endocrinology1993; 133:1490-1503. Thokatura N R , Desai RK, BatesLG,Cole ES, WeintraubBD.Biologicalactivityand metabolic clearanceof a recombinant human thyrotropin produces in Chinese hamster cells. Endocrinology 1991;128:341-348. Thokatura N R , SzkudlinskiM W , Weintraub BD. Structure-function studies of oligosaccharides of recombinant human thyrotropin by sequential deglycosylation and resialylation. Glycobiology 1994; 4:525-533. Matzuk MM, Kornmeier CM, Whitfield GK, Kourides IA,Boime I. The glycoprotein asubunitiscriticalforsecretionandstability of thehumanthyrotropinP-subunit.Mol Endocrinol 1988; 2:95-100. SeidlinSM, Oshry E,Yalow AA. Spontaneousandexperimentallyinduceduptake of J Clin Endoradioactive iodinein metastases from thyroid carcinoma: a preliminary report. crinol 1948; 8:423432. Stanley MM, Astwood EB. The response of the thyroid gland in normal human subjects 1-131. Endocrinology totheadministration of thyrotropin, as shownbystudieswith
1949; W.49-60 30. Benua RS, Sonenberg M, Leeper RD, Rawson RW. An 18 year study of the use of beef J NuclMed 1964; thyrotropin to increase 1-131uptakeinmetastaticthyroidcancer. 5796-801. 31. Sturgeon C T , Davis FE, Catz B, Petit D, Starr P. Treatment of thyroid cancer metastases with TSH and 1-131during thyroid hormone medication. J Clin Endocrinol Metab 1953; 13~1391-1407. of thyrotropic hormone and 32. Catz ByPetit D, StarrP. The diagnostic and therapeutic value
heavy dosage scintigrams for the demonstrationof thyroid cancer metastases. A m J Med Sci 1959;237:158-164. 33. Catz B, Petit DW, Schwartz M, DavisF, McCammon C, Starr P. Treatment of cancer of the thyroid postoperatively with suppressive thyroid medication, radioiodine, and thyro stimulating hormone. Cancer 1959; 12:371-383. 34. Reichert LE Jr. On the relationship between human thyrotrophin research standard A, th
177
Recombinant Human Thyrotropin
Unites States Pharmacopeia thyrotrophin standard (bovine) and the International Standard for thyrotrophin (bovine). J Clin Endocrinol1970; 31:331-333. 35. Schneider PB, Robbins J, Condliffe PG. Thyroid response to human thyrotropin in man. J Clin Endocrinol Metab 1965; 25:514-517. (TSH) levels after thyroid ablation compared 36. Hershman J M ,Edwards C L . Serum thyrotropin with TSH levels after exogenous bovine TSH implications for 13'1 treatment of thyroid carcinoma. J Clin Endocrinol 1972; 342316818. 37. Hays MT, Solomon DH, Werner SC. The effect of purified bovine thyroid-stimulating hormone in man.II.Loss of effectiveness with prolonged administration. J Clin Endocrinol Metab 1961; 21:1475-1482. 38. Krishnamurthy GT. Human reactive to bovineTSH. concise communication. J Nucl Med 1978;19:284-286. 39. Sherman WB, WernerSC.Generalizedallergicreactiontobovinethyrotropin.JAMA 1964,190:2&245. 40. Hays MT, Solomon DH, Pierce JG, Carsten ME. The effect of purified bovine thyroidstimulating hormone in man. I. Dose-response characteristics studies with 1132. J Clin Endocrinol Metab 1961; 21:1469-1474. JM, Krishnamurthy GT, Blahd Neutralizing antibodies 41. Melmed S, Harada A, Hershman
WH.
to bovine thyrotropin in immunized patients with thyroid cancer. J Clin Endocrinol Metab
1980; 51~358-363. 42. Hays MT, Solomon DH, Beall GN. Suppression of human thyroid function by antibodies to bovine thyrotropin. J Clin Endocrinol Metab1967; 27:1540-1549. 43. Greenspan FS, Lowenstein JM, West MN, Okerlund MD. Immunoreactive material to 1972; bovine TSH in plasma from patients with thyroid cancer. J Clin Endocrinol Metab 35~795-798.
44. Frohman LA, Baron MA, SchneiderAB. Plasma immunoreactive TSH spurious elevation due
to antibodies to bovine TSH which cross-react with human TSH. 1982;Metabolism 312334-840. MD, et al. Falsely positive bovine TSH radioimmunoassay 45. Greenspan FS, Lew W, Okelund 1974; Metab 38:1121responses in sera from patients with thyroid cancer. J Clin Endocrinol 1122.
46. Sain A, Sham R, Singh A, al. et Erroneous thyroid-stimulating hormone radioimmunoassay
results due to interfering anti-bovine thyroid-stimulating hormone antibodies. Am J Clin Path01 1979; 71:540-542. 47. Chaussain JL, Binet E, Job JC. Antibodies to human thyrotropin in the serum of certain hypopituitary dwarfs. Rev Eur Etudes Clin Biol1972; 17:95-99. 48. Uller RP, van Herle AJ, Chopra IJ. Comparison of alterationsin circulating thyroglobulin, triiodothyronine and thyroxine in response to exogenous (bovine) and endogenous (huma thyrotropin. J Clin Endocrinol Metab1973; 37:741-745. TR. Human thyrotrophic hormone kinetics and 49. Kuku SF, Harsoulle P, Kjed M, Fraser effects in euthyroid males. Horm Metab Res 1975; 7:54-59. 50. Law A, Jack GW, Tellez M, Edmonds CJ. In vivo studies of a human thyrotrophin prepa tion. J Clin Endocrinol 1986; 110:375-378. 51. RidgewayEC,WeintraubBD, Maloof F. Metabolic clearance and production rates of human thyrotropin. J Clin Invest1974; 53:8895-903. 52. Brown P, Gadjusek DC, Gibbs Jr CJ, Asher DM. Potential epidemic of Creutzfeld-Jakob disease from humangrowth hormone therapy. N Engl J Med 1985; 110:375-378. 53. HayashizakiY, Miyai K, Kat0K, Matsubara K. Molecular cloning of the human thyrotropin beta-subunit gene. FEBS Lett 1985; 188:394-400. 54. Wondisford FE, Radovick S, Moates JM, Usala SJ, Weintraub BD. Isolation and character1988; 263:12538-12542. ization of the human thyrotropin-beta subunit gene. J Biol Chem 55. WatanabeS, HayashizakiY, Endo Y, et al. Production of human thyroid stimulating hormone in Chinese hamster ovary cells. Biochem Biophys Res Commun 1987;149:1149-1155.
178
Ringel
56. Hussain A, Z m e r m a n CA, Boose JA, et al. Large scale synthesis of recombinant human
thyrotropin using methotrexate amplification: chromatographic, immunological, and b 1996; 8 1:1184-1 188. cal characterization. J Clin Endocrinol Metab 57. Cole ES, Lee K, Lauziere K, et al. Recombinant human thyroid-stimulating hormone: development of a biotechnology product for detection of metastatic lesions of thyroid carcinoma. Biotechnology 1993;11:1014-1024. of the human thyrotropin p58. Wondisford FE, Usala SJ, DeCherney GS, et al. Cloning subunit gene and transient expressionof biologically active human thyrotropin after gene transfection. Mol Endocrinol 1988;2:32-39. 59. Huber GK, Fong P, Concepcion ES, Davies TF. Recombinant human thyroid-stimulating hormone: initial bioactivity assessment using humanfetal thyroid cells. J Clin Endocrinol Metab 1991; 72:1328-1331. 60. Leitolf H,SzkudlinskiMW, Thotakura N R , von zur Muhlen A, Brabant G, Weintraub BD. Effects of continuous and pulsatile administration of pituitary rat thyrotropin and recombi 1995;Res 27: 173-178. nant human thyrotropin in a chronically cannulated rat. Horm Metab 61. East-Palmer J, Szkudlinski MW, Lee J, Thotakura NR,Weintraub BD. A novel nonradioactive in vivo bioassayof thyrotropin (TSH). Thyroid1995; 555-59. 62. Colzani R M , Alex S, Fang S-L, Braverman LE, Emerson CH. The effect of recombinant 1998; 8:797-801. human thyrotropin (rhTSH) on thyroid function in mice and rats. Thyroid 63. Braverman LE, Pratt BM, Ebner S , Longcope C. Recombinant human thyrotropin stimulat thyroid function and radioactive iodine uptake in the rhesus monkey. J Clin Endocrinol Metab 1992;74:1135-1139. 6 4 . Ramirez L, Braverman LE, White B, Emerson C E . Recombinant human thyrotropin is a potent stimulator of thyroid function in normal subjects. J Clin Endocrinol Metab 1997; 82:2836-2839. 65. Ringel MD, Ladenson PW. Diagnostic accuracy of I3'I scanning with recombinant human
thyrotropin versus thyroid hormone withdrawal in a patient with metastatic thyroid carc noma and hypopituitarism. J Clin Endocrinol Metab1996; 81:1724-1725. 66. Rudavsky A Z , Freeman LM. Treatment of scan-negative, thyroglobulin-positive metastati 131 1and recombinant human thyroid stimulating horm thyroid cancer using radioiodine J Clin Endocrinol Metab 1997; 82:9-10. 67. Meier CAY Braverman LE, Ebner SA, et al. Diagnostic of recombinant use human thyrotropi inpatientswiththyroidcarcinoma(phase I/II study).JClinEndocrinolMetab 1994;
78~188-196. 68. Albrecht RR,Ewing SJ. Standardizing administration of the Profile of Moods States (POM development of alternative word lists. J Personal Assess 1989; 53:31-39. for thyrotropin use 69. Ladenson PW. Recombinant human thyrotropin symposium: Strategies to monitor patients with treated thyroid carinoma. Thyroid 1999; 9:429433. An overview of the manage70. Mazzafeni EL. Recombinant human thyrotropin symposium: ment of papillary and follicular thyroid carcinoma. Thyroid1999; 9:421427. 71. Ribela MT, Bianco AC, Bartolini P. The use of recombinant human thyrotropin produced
by Chinese hamster ovary cells for the preparation of immunoassay reagents. J Clin Endocrino1 Metab 1996; 81:249-256. 72. Morgenthaler NG,Pampel I, Aust G, Seissler J, Scherbaum WA. Application ofa bioassay with CHO cells for the routine detection of stimulating and blocking autoantibodies to the TSH-receptor. Horm Metab Res 1998; 30:162-168. 73. Szkudlinski MW, Teh NG, Grossman M, et al. Engineering human glycoprotein hormone superactive analogues. Nature Biotechnol 1996;14:1257-1263. 74. Grossmann M, Leitolf H, Weintraub BD, et al. A novel strategy for rational design of protein hormone superagonists. Nature Biotechnol 1998; 16:871-875. MW.Recombinant human thyrotropin symposium: developm 75. Weintraub BD, Szkudlinski and in vitro characterization of human recombinant thyrotropin. Thyroid1999; 9:447450.
15 Chemotherapy for Thyroid Cancer Lawrence S. Lessin and My0 Min
Chemotherapy has been used as a single modality treatment or part of combined modality therapy in metastatic or locally advanced thyroid cancer when other conventional treatments such as surgery and radiation therapy have failed. Cytotoxic chem apy is predominantly employed in anaplastic carcinomas, and may be employed in the 20% of differentiated (papillary, follicular, and mixed) thyroid carcinomas which do not concentrate iodine. Although chemotherapy may induce a tumor response and provide palliation of troublesome symptoms, there is no established evidence that it prolongs survival. Numerous reports on use of chemotherapy in a variety of thyroid cancers have been published, but there are few controlled studies that compare the efficacy of different drug regimens. Although, in general, the response to chemotherapy is only modest, some investigators have stated that patients who respond to the first chemotherapeutic agent are more likely to respond to a second agent when relapse occurs ( I ) . CHEMOTHERAPEUTIC AGENTS USED IN THYROID CANCER Individual chemotherapeutic agents with known (or proposed) antitumor activity against thyroid cancer are listed in Table 1. DOXORUBICIN Doxorubicin is an anthracycline derivative that has been the most widely used and studied chemotherapeutic agent in thyroid cancer. In 1974, one of the earliest studies (2) who treated 30 patients with on doxorubicin was reported by Gottlieb and Hill different types of refractory thyroid carcinoma. Of 30 patients, 11 (37%) achieved a partial response. Median survival was found to be significantly better in responders compared to nonresponders (1 1 months vs 4 months). Since then, many reports on use of doxorubicin in advanced thyroid cancer have been published with response rates varying from 30% to 45%. Currently, doxorubicin is considered the most effective single agent at a dose of60 mg/m2 every 3 weeks ( I ) . Lower doses of doxorubicin at Hill (2) to be inferior to 60 mg/mz, with no 45 mg/mz were found by Gottlieb and responses in3 patients. By contrast,3 of 13patients responded to60 mg/mz;3 additional Fmm:Thyroid Cancer: A Comprehensive Guide toClinical Management Edited by: L. Wmtofiky 8 Humana Press Inc., Totma, iiJ
179
2 80
Lessin and Min
Table 1 Chemotherapeutic Agents Usedin Thyroid Cancer Drug
Major toxicity
ChSS
Doxorubicin Antitumor antibiotic Cardiac; limit to
550 mg/m2 using intermittent bolus schedule
Bleomycin Antitumor antibiotic Pulmonary toxicity; follow pulmonary diffusion capacity Cisplatin Heavy metal; binds directly Nephrotoxicity; follow electrolytes to DNA Carboplatinum Heavy metal, analogue Myelotoxicity of cisplatin Myelosuppression Epipodophyllotoxin; Etoposide topoisomerase II inhibitor Dacarbazine Alkylating-like Myelosuppression; activity gastrointestinal Cyclophosphamide Paclitaxel Vincristine Methotrexate
Alkylating agent Taxane Vinca alkyloid Antimetabolite
closely
toxicity Myelosuppression; hemorrhagic cystitis Myelosuppression; neuropathy Peripheral neuropathy GI toxicity; liver fibrosis
responses were seen when the dose was escalated to75 mg/m2; and2 more responses 90 mg/m2. Doxorubicin-induced cardiomyopathy were seen with further increase to was found to be the limiting toxicity. Lack of response to low-dose doxorubicin was also found by Droz and colleagues(3).Nutrition and performance statusof the patient (4) and also seemed to influence response to doxorubicin. O’Bryan and coworkers to doxorubicin occurredin patients Benker and Reinwein(5) noted that poorer response with low performance status. Shimaoka and associates(6) in the only randomized trial published also found performance status a significant predictor of response. Dox has been used in combinations with other agents including cisplatinum, bleomycin, vincristine, and vindesine.
BLEOMYCIN
Bleomycin was the first chemotherapeutic agent to be used in metastatic diffe thyroid cancer. Although relatively ineffective as a single agent, when used in com tion with doxorubicin and vincristine(ABC) or with doxorubicin and cisplatin (BAP), response rates up to 30% are reported (7).
CISPLATIN Cisplatin has been used as monotherapy in heavily pretreated patients with a variety of thyroid cancers. Hoskin and Harmer( I ) reported 5 responders of 13 patients (38%) treated. Droz and colleagues (8)treated 18 patients with medullary carcinoma of thyroid and reported 3 responses (21%), including 1 patient with a complete response lasting 9 months. Along with doxorubicin, cisplatin is commonly used as part of a combination
Chemotherapy
181
chemotherapy regimen. Droz and colleagues (8) also reported a Phase II study of 44 cases with both differentiated and anaplastic thyroid cancer, utilizing doxorubicin and cisplatin as either monotherapy or in combination. No objective responses were seen in the 19 patients treated with single-agent cisplatin,13 of whom had previously failed doxorubicin therapy. Therefore, in patients refractory to doxorubicin, cisplatin may produce responses in patients with medullary carcinoma, but not in differentiated and anaplastic carcinoma. ETOPOSIDE Hoskin and Harmer (I)reported their experience with etoposide as a single agent in 22 heavily pretreated patients with 4 responses in a variety of thyroid cancers. (9)in medullary thyroid Etoposide was used as a single agent by Kelsen and coworkers carcinoma with a response rateof 14%. Related neuroendocrine tumors such as small cell cancer of the lung and peripheral neuroectodermal tumors are highly responsive to etoposide. CARBOPLATIN Hoskin and Harmer (I)reported limited experience with single agent carboplatin in
9 heavily pretreated patients; 2 of 9 showed partial responses.
METHOTREXATE
Methotrexate, the antifolate antimetabolite, was used in the early 1980s in combination with doxorubicin and lomustine (a nitrosourea). Because of poor response ra little usage of methotrexate has been recently reported in treatment of thyroid cancer. Application of the various chemotherapeutic agents to specific thyroid cancers is discussed for differentiated (papillary and follicular thyroid cancer, medullary thyroid cancer, and anaplastic carcinoma of the thyroid in Chapters 20, 37, and 47. REJXRENCES 1. Hoskin PJ, HarmerC. Chemotherapy for thyroid cancer. RadiotherOncoll987; 10187094. 2. Gottlieb JA, HillCS. Chemotherapy of thyroid cancer with Adriamycin: experience with 30 patients. N Engl J Med 1974; 290:193-197. 3. Droz JP, Charbord P, Rougier P, Parmentier C. Echec de la chimiotherapie des cancersde la thyroide. Bull Cancer 1981; 68:350. 4. O’Bryan RM, Baker LH, Gottlieb JJ3, Rividkin SE, BakerzakSP, Grumet GN, Salmon SE, Moon TE, Hoogstraten B. Dose response evaluation of Adriamycin in human neoplasia. Cancer 1977; 39:1940. 5. Benker G, Reinwein D. Ergegnisse der Chemotherapie des Schilddrusenkarzinoms. Dtsch Med Wochenschr 1983; 11:403-406. R, DeConti R.A randomizedtrial of doxorubi6. ShimaokaK,Schoenfeld D, DeWys W, Creech 1985; cin vs doxorubicin plus cisplatin in patients with advanced thyroid carcinoma. Cancer 56~2155-2160. 7. Harada T, Nishikawa Y, Suzuki T, et al. Bleomycin treatment for cancer of the thyroid. Am J Surg 1971; 2253.
182
Lessin and Min
8. Droz JP, Schlumberger M, Rougier P, Caillou B, Goddefroy W, Gardner P, Parmentier C. Phase 11 trials of chemotherapy with Adriamycin, cisplatin and their combination in thyroid cancers: a review of 44 cases. Int Congr Ser 1985; 684:203-208. 9. Kelsen D, Fiore J, Heelan R, ChengE, Magill G. Phase11 trial of etoposide inMUD tumors. Cancer Treat Rep 1987; 71:305-307.
I11 Differentiated Tumors of the Thyroid Gland A. Papillay Carcinoma
16 Papillary Carcinoma Clinical Aspects Leonard Wartofsky
Papillary carcinoma is the most common type of thyroid malignancy, accounting for 65-80% of all thyroid cancers (1-8). Papillary thyroid carcinoma(PTC) is a cancer of the thyroid follicular epithelium, and like follicular carcinoma it is the more highly differentiated of all of the classes of thyroid malignancy. The biological behavior of PTC varies widely, from small (4.0 cm) tumors found at autopsy with surprisingly high frequency and which show little evidence of invasion, to rapidly growing, locally invasive tumors that may be resistant to radioiodine therapy and eventually metastasi and can cause death. To date, it has not been possible in a given patient to predict which course the tumor may take until several years, or even decades, of follow-up have elapsed. Happily, the overwhelming majority of those tumors4 . 0 cm in diameter tend to behave in a more biologically benign manner, and can be completely cured with definitive therapy. The small so-called papillary thyroid “microcarcinomas” have a clearly different naturalhistorythandolargerlesions.Theformeraredescribedasoftendetected incidentally at either autopsy or during thyroidectomy for an indication other than tumor, and can range in size from 2 mm to 1.5 cm in diameter. They are usually (7,9). nonencapsulated but appear to remain biologically silent with minimal morbidity Of 535 microcarcinoma cases reviewed by Hay and colleagues (9),99% were histologically grade 1 tumors and no local invasion was apparent in 98% of the patients; 32% of recurrence, had positive nodes at presentation that correlated with higher likelihood but the prognosis was nevertheless excellent after near-total thyroidectomy alone, irrespective of whether radioiodine remnant ablation was done. Papillary thyroid cancer tends to occur in younger patients, most commonlyin the of all the varietiesof thyroid third and fourth decadesof life, and has the best prognosis malignancy. A number of clinical and pathological characteristics have been evaluated of as predictors of tumor behavior and ultimate patient prognosis. The more useful these parameters are described below. As described in Chapters 7 and 8, exposure to radiation is a risk factor for papillary thyroid cancer, the risk of malignancy increasing 5-10% up to 30-50% (7).Although PTC tends to affect women more often than men risk of cancer in thyroid nodules (-2: l), we are equallyif not more concerned about the From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wnrtofiky 0 Humana Press Inc., Totown, NJ
185
186
Wartojiky
in men because of the much lower frequency of any type of thyroid disease in men. This chapter deals with several clinical aspects of these tumors and descriptions of pathology and management appear in the immediately following chapters. 1 to 10 Annual incidence rates for well-differentiated thyroid cancer range from cases per 100,000 population, and a detailed description of the epidemiology of thes tumors appears in Chapter 7. One of the largest cohorts of patients with PTC was (5). Of 1500 patients, two-thirds followed up at the Mayo Clinic and reviewed by Hay were women (a 2:l ratio), and the ages at diagnosis ranged widely from only 5 years old to 93 years old. The majorityof patients presented between ages 30 and 60. There are major differences in frequencies in other ethnic groups. In Japanese, for example the female : male ratio has been reported as high as 13 : 1 (10). In most patients, the clinical presentation of papillary thyroid carcinoma is that of a thyroid nodule, discovered during routine physical examination by a physician o as a lump in the neck by a friend or relative. The diagnosticof evaluation the nodule (see Chapter 4) virtually always includes fine needle aspiration cytology which can dem strate the pathognomonic features of this tumor.A recent survey of clinical thyroid sp as to the diagnostic evaluation of such patients and the nee ists indicated fair consensus for subsequent thyroidectomy and radioiodine ablation, although there were se of controversy (11).As many as a third of patients will have some underlying thyroid disease such as Hashimoto’s disease or multinodular or adenomatoid goiter.In one series of 596 PTC patients from the Mayo Clinic (5), 40% of patients had other benign thyroi disease, 33% having coexistent thyroidnodules, and 20% having Hashimoto’s t Underlying Hashimoto’s disease appears to be a favorable prognostic factor for both reduced rates of recurrence and increased survival (12). Not infrequently, the patient presents as well with cervical lymphadenopathy, the (13). The evaluation of which then leads to detection of the primary thyroid tumor lymphadenopathy is due to local metastases of the tumor, and the most freque tion will be to nodes on the ipsilateral side of the neck as the tumor as well as down into the superior mediastinum. Metastasesto the contralateral cervical chainsof nodes occurwitheither more advanced or more aggressive disease. This may be related to initial intrathyroidal seeding via the thyroidal lymphatic system since micro- to macroscopic foci of papillary carcinoma are often found in the contralateral thyroid lobe. In the Mayo series(5,6) of 1500 patients, the primary tumor was confined to on lobe in 71%, bilateral in 19%, and with multicentric lesions in 26%. Cervical lymph node involvement was diagnosed in 38% (573/1500) (median number of nodes = 4), and 2% (32/500) had distant metastases at time of initial diagnosis. These findings ar comparable to the series reported by Mazzaferri and Jhiang(14) in which 32% of the tumors were multicentric and 43% had nodal involvement. When the papillary cance is more aggressive,as evident by widespread involvement throughout the thyroid gla the likelihoodof lymph node metastases is approximately 75%.This was seen frequent in the series of 129 patients reported by DeJong and associates (13) in whom there were metastases to the ipsilateral jugular nodes in 34%, bilateral jugular nodes and to the central compartment in 81%.In their total series of 243 patients with PTC, DeJong’s group found that 21% actually presented with lateral cervical nodes and no palpable thyroid nodule(s).
Papillary Carcinoma
187
Patients presenting with palpable cervical lymph nodes at varying intervals after their original thyroidectomy need to be evaluated for metastases. The extent of the lymphadenopathy can be assessed byCT or MRI scans of the neck and mediastinum, and such recurrent disease is usually, but not always (15), suggested by rising levels of serum thyroglobulin. Enlarged palpable lymph nodes canbe readily aspirated with or without ultrasound guidance for cytological examination and thyroglobulin measure ment. The use of polymerase chain reaction amplification of TSH receptor and thyroglob(16). ulin transcripts in the node aspirate to confirm metastasis has also been reported Future management of such patients is likely to be facilitated as well by the use of measurement of thyroglobulin mRNA in s e m (17), especially in those patients with thyroid autoantibodies that interfere with the thyrogobulin assay. Since these tumors are so frequently detected incidentally, it is obvious that most are asymptomatic. It is only becauseof increasing size and/or invasion that symptoms arise. Symptoms can include cough, dysphagia, or odynophagia, or, more commonly, a senseof fullness or pressure in the neck. On occasion, patients complain of an achin in the areaof the involved lobe. The differential diagnosis of pain in the thyroid gland is fairlyshort,consisting ofonlythreeentities:invasivethyroidcancer,subacute thyroiditis, or hemorrhage within a nodule or cyst. Young (e17 years old) patients may have metastases of papillary thyroid cancer to the lungs (18-20). Patients with lung metastases may present with hemoptysis or dyspnea at rest or on exertion. Pulmonary metastases were confirmed in one report by cytological examination after bronchoalveolar lavage (21). Although papillary thyroid cancer can metastasize to the lungs, follicular thyroid carcinomais more likely to invade blood vessels and appear in distant sites such as bone and lung and present with symptoms of local bone pain or dyspnea, respectively (4,8,22,23). Pulmonary metastases of PTC are typically difficult to completely eradicate with radioiodine therapy (24). Patients with PTC will only very rarely present with local thyroid or cervical pain, in contrast to the patients with less differentiated thyroid cancers. Involvementof the the ipsilateral vocal cord and hoarseness. recurrent laryngeal nerve will result in of palsy The texture or consistency of a malignant thyroid nodule may be no different from that of a benign follicular adenoma; the classic very firm to hard consistency of a nodule due to papillary cancer may be related to the nodule’s content of calcium (psammoma bodies). Medullary cancers (in which calcifications are frequent) or anap are the well-differentitic carcinoma are much more frequently hard on palpation than ated tumors. The diagnostic evaluation for the patient presenting with a mass in the thyroid is describedabove(Chapter 4) andmayincludefineneedleaspirationcytologyand ultrasound imaging. The tumors are almost always solid on ultrasound scans. While purely cystic nodules rarely contain malignancy, large mixed cysticholid nodules pos greater risk demanding closer management. Other imaging techniques, such as com tomography (CT) or magnetic resonance imaging (MRI), are rarely required, but can serve to demonstrate the extent of metastatic lymphadenopathy either before or after surgery. It is common for a patient to present with discovery of the thyroid mass because of a CT scan or MRI having been performed for another, unrelated, problem, Once the diagnosis is established, surgical thyroidectomy is the next step, and this is
188
Wartofsky
usually a near-total to total thyroidectomy with exploration for enlarged lymph nodes, especially on the side ipsilateral to the nodule, as described in Chapters 5 and 18. Total thyroidectomyis associated with greater risk of complications (recurrent lary geal nerve trauma with subsequent hoarseness or temporary to permanent hypoparathyroidism) but is more likely to result in a lower rate of recurrence, presumably due to removal of bilateral or multifocal foci of tumor. Patients who undergo only lobectom or subtotal thyroidectomy may have a 2.5-fold risk of death compared to those undergo ing total thyroidectomy(25). However, patients with tumors of1.5 cm diameter or less by and large will have an excellent prognosis after only a lobectomy with isth without postoperative radioiodine ablation of the residual contralateral thyroid lobe (7,9,26-31). This includestheso-calledincidental“microcarcinomas”(discussed above), which may be found in a thyroid gland resected for some other indication, such as Graves’ disease or nodular goiter. These tumors are often foundin the thyroid at postmortem examination. We believe that they have an excellent prognosis, and completion thyroidectomy or radioiodine ablation is rarely indicated. as a thyroid nodule Virtually all patients with papillary thyroid carcinoma presenting will be euthyroid, and hence there is little need for routine thyroid function tests. Occasional patients with underlying Hashimoto’s disease may have associated hy roidism which is coincidental and not related to the tumor per se, for it is only the much normally functionmore aggressive anaplastic forms of cancer that will so replace as to cause thyroid hypofunction. In patients undergoing thyroiding thyroid parenchyma ectomy for benign causes, there may be some advantage in knowing the preoperative levels of serum TSH and free thyroxine in order to have target levels for subsequent levothyroxine replacement therapy. Such information is rarely pragmatic for postoperative management of patients with malignant disease, however, because the dosing of thyroxine therapy is selected to be suppressive rather than for replacement, and the preoperative hormone levels are thus irrelevant. Staging of papillary thyroid cancer is based on both the extent of disease and the age of the patient at presentation(25,32). The extent of disease may be determined by clinical, radiological, and pathological examinations to include biopsy of suspicious lymph nodes. Awareness of distant metastases may not be obtained until a total body radioisotopic survey scan is performed postoperatively (33). ”‘I scans are generally more useful to detect bony metastases than technetium pyrophosphate bone scans, which often do not visualize tumor because the tumors do not elicit an osteoblastic reaction. However, combined scanning with technetium-99m hydroxymethylene (34). phonate and thallium-201 has been shown to be useful to detect bony metastases CT scanning may be very useful to image bone lesions, especially those of the lower spine and pelvis. The TNM (tumor-nodes-metastases) system of staging has been shown to be a usefulmethodofstaging in regardtocorrelationwithobservedoutcomesbased upon aretrospectiveanalysis of 700 patientsovera25-yearperiod byLohand coworkers (25). The TNM system is based on the following subgroups. Tumor size refers to the original single malignant nodule or the largest nodule in a gland with multifocal lesions. Regional lymph nodes are defined as bilateral cervical and upper mediastinal nodes.
189
Papillay Carcinoma
Primary tumor cannot be assessed No evidence of primary tumor Tumor 1 cm or less in greatest dimension limited to thyroid n Tumor >1 cm but e4 cm in greatest dimension and limited to the thyroid Tumor >4 cm in greatest dimension and limited to the thyroid T3 Tumor of any size extending beyond the thyroid capsule T4 Nodes: NX Regional nodes cannot be assessed No metastases to regional nodes NO Metastases to regional nodes are present N1 N1A -in ipsilateral cervical nodes N1B -in bilateral, midline, or contralateral cervical or mediastinal nodes Metastases: MX Presence of distant metastases cannot be assessed No distant metastases MO M1 Presence of distant metastases Tumor:
TX TO T1
The staging for papillary thyroid cancer is designated as Stage I, 11, III, or IV based upon the TNM status and the age of the patient as follows:
Under 45 years
of
age
Stage I any T,any N,MO Stage 11 any T,any N, M1 Stage 111 Stage IV
45 years old and
older
T1, NOM ,O "2,NOM ,O T3, NO, MO or any T, N1, MO Any T,any N, M1
The importance of age as a prognostic factor canbe appreciated by noting that all patients less than 45 years old without distant metastases are classified as StageI and those with distant metastases are not classified any higher than stage II. The prognosisof papillary thyroid cancer is discussed in Chapter 25. The overwhelming majority of patients will fall into stagesI and 11and have excellent prognosis with little risk for recurrenceor death from their disease. There was a remarkable 25-year survival of 97% in 1408 patients reviewed by Hay (5) who had complete surgical resection of their apparent disease. For such patients, 30-year survivalofrates 75-85% are not unusual. Stages 11 and 111 patients may have recurrences requiring additional therapy but remain at relatively low risk for death (3,5,14,28,29). One major clinical difficulty for both the patients and their physiciansis dealing with the indolent nature of thesetumors.Thyroidcancercellsareusuallyslow-growing,andwhile this is favorable for recurrence and death rates, it also implies the absolute necessity for longterm meticulous follow-up because there canbe recurrences in patients believed tobe free of disease as late as 15 to 20 years after their original presentation. Hay (5) and others have identifiedthree varieties of presentation that reflect a different prognostic 1) the presenceof postoperative local metastatic category in regard to tumor recurrence: nodes, 2) postoperative distant metastases, or3) local recurrence in the thyroid bedor adjacent tissue other than lymph nodes. The Mayo group incorporates these factors into their "MACIS" scoring system, which can reliably predict outcome based on data at initial presentation (35). Age of diagnosis appears to be the most important factor in terms of having an impact on prognosis, with clearly more aggressive tumor behavior likely after age
190
Wartofsky
40-45. In one large retrospective review of 15,698 cases of thyroid cancer, age was a stronger predictorof survival for patients with follicular carcinoma than for papill carcinoma (30). While not as important as age, the size of the original tumor is also important, with tumors less than 1.5 cm having the best prognosis and those that are larger than 4 cm having the worst prognosis. Finally, a worse is also prognosis associated with extensive local invasion, and even more so with distant metastases, especially those to bone. Some investigators have also incorporated the histological grade of the tumor into the prognostic score ($31). Following thyroidectomy for papillary lesions >1.5 cm, most workers employ dine in doses of 30-100 mCito ablate residual tissue and facilitate follow-up monit (33,36) as is discussed in Chapters 12,22, and 23. Early series of patients have indicated that 13*1ablation of thyroid remnants was followed by a significantly lower recurrence rate (27). However, the belief that such management is necessary and actually improves (5). A recent retrospective analysis of 700 patients suggested prognosis has been disputed that patients not treated with radioiodine ablation had a 2.1-fold greater risk of their malignancy (P= .OOOl), although no difference in death rates (25). in regard to thyroxine suppressive therapy, The follow-up of patients postoperatively and monitoring with periodic radioisotopic scans and serum thyroglobulin measure (37,38) has been employed is discussed belowin Chapter 22. External radiation therapy with variable success for tumors which do not trap radioiodine or are resistant to suchtherapy,asdiscussed in Chapter21.Chemotherapy is anotheralternative in such patients (39-41) and is discussed in Chapter 20. The availability of recombinant human TSH (42,43) has radically altered routine follow-up evaluations for residual or recurrent disease of patients after their initial management by thyroidectomy and radioiodine ablation (see Chapter 14).
REFERENCES 1. Ain KB. Papillary thyroid carcinoma: etiology, assessment, and therapy. Endocr Metab Clin N Am 1995; 24~711-760.
2. Clark OH, Duh Q-Y. Thyroid cancer. Med Clin North Am 1991; 75:211. FH. Natural history, treatment, and cours 3. DeGroot LJ, Kaplan EL, McCormick M, Straus of papillary thyroid carcinoma. J Clin Endocrinol Metab1990; 71:414424. 4. Goldman N D , Coniglio JU, Falk SA. Thyroid cancers I: Papillary, follicular, and Hlirthle cell. Otolaryngol Clin NorthAm 1996; 29593-609. 5. Hay ID. Papillary thyroid carcinoma. Endocrinol Metab Clin North Am 1990; 19545-576. 6. McConahey WM, Hay ID, Woolner LB, van Heerden JA, Taylor WF. Papillary thyroid cancer treated at the Mayo Clinic, 1946 through 1970: initial manifestations, pathologic findings, therapy, and outcome. MayoClin Proc 1986; 61:978-996. Ann Intern Med 1991; 7. Robbins J, et al. Thyroidcancer:alethalendocrineneoplasm. 115:133-147. 8. Schlumberger MJ. Papillary and follicular thyroid carcinoma. N Engl J Med 1998;338: 297-306. 9. Hay ID, Grant CS,van Heerden JA,et al. Papillary thyroid microcarcinoma: a study of 535 cases observed in a 50-year period. Surgery 1992; 112:1139-1147. 10. It0 J, Noguchi S, Murakami T, et al. Factors affecting the prognosis of patients with carcinoma of the thyroid. Surg Gynecol Obstet 1980; 150539. 11. Solomon BL, Wartofsky L, Burman KD. Current trends in the management of well differentiated papillary thyroid carcinoma.J Clin Endocrinol Metab1996; 81:333-339.
Papillay Carcinoma
191
12. Kashima K, Yokoyama S, Noguchi S, et al. Chronic thyroiditis as a favorable prognostic factor in papillary thyroid carcinoma. Thyroid1998; 8:197-202. 13. DeJong S, Demeter J, Jarosz H, et al. Primary papillary thyroid carcinoma presenting as cervical lymphadenopathy. Am Surg 1993; 59:172-177.
14. Mazzafem EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer.Am J Med 1994; 97:418-428. 15. Miller JH, Marcus CS. Metastatic papillary thyroid carcinoma with normal thyroglobulin level. Clin Nutr Med 1988; 9:652. R, Filetti S. Early diagnosis 16. Arturi F, Russo D, Giuffrida D, Ippolito A, Perrotti N, Vigneri by genetic analysis of differentiated thyroid cancer metastases in small lymph nodes. J Clin Endocrinol Metab 1997; 82:1638-1641. 17. Ringel MD, Ladenson PW, Levine MA. Molecular diagnosis of residual and recurrent thyroid cancer by amplification of thyroglobulin messenger ribonucleic acid in peripheral blood. J Clin Endocrinol Metab 1998; 83:4435-42. Am 18. Gorlin JI3, Sallan SE. Thyroid cancer in childhood. Endocrinol Metab Clin North
1990; 19549. 19. McClellan DR, Francis GL. Thyroid disease in children, pregnant women, and patients with Graves’ disease. Endocrin Metab Clin North Am 1996; 25:27-48. 20. Zimmerman D, HayID, Gough IR, et al. Papillary thyroid carcinoma in children and adults: follow-up of 1039 patients conservatively treated at one institution during three decades. Surgery 1988;104:1157-1166. 21. Mello CJ, VeronikisI, Fraire AE, Aronin N, Irwin RS, Braverman LE. Metastatic papillary thyroid carcinoma to lung diagnosed by bronchoalveolar lavage.J Clin Endocrinol Metab 1996; 81:406-410. 22. Cooper DS, Schneyer CR. Follicular and Hiirthle cell carcinoma of the thyroid. Endocr Metab Clin North Am 1990; 19577-592. 23. GrebeSKG,Hay ID. Follicularthyroidcancer.EndocrMetabClinNorth Am 1995; 241761402. 24. Sisson JC, Jamadar DA, Kazerooni et EA, al. Treatmentof micronodular lung metastases of
papillary thyroid cancer: are the tumors too small for effective irradiation from radioiod Thyroid 1998;8:215-221. 25. Loh K-C, Greenspan FS, GeeL, Miller TR, Ye0 PPB. Pathological tumor-node-metastasis (pTNM) staging for papillary and follicular thyroid carcinomas: a retrospective analysis of 700 patients. J Clin Endocrinol Metab 1997; 82:3553-3562. 26. Simpson WJ, McKinney SE, Carmthers JS, et al. Papillary and follicular thyroid cancer: prognostic factors in 1578 patients. Am J Med 1987; 83:479-488. 27. Mazzafem EL. Thyroid remnant 131-1ablation for papillary and follicular thyroid carcinoma. Thyroid 1997; 7:265-271. 28. MazzaferriEL, YoungRL, Oertel JE,Kemmerer WT, Page CP. Papillary thyroid carcinoma: the impact of therapy in 576 patients. Medicine 1977; 56:171-196. 29. Mazzaferri EL, YoungRL. Papillary thyroid carcinoma: A10 year follow-up reportof the impact of therapy in 576 patients. Am J Med 1981; 70:511-518. 30. Gilliland F D , Hunt WC, Moms DM, Key CR. Prognostic factors for thyroid carcinoma: a population based studyof 15698 cases from the surveillance, epidemiology and end results (SEER) program 1973-1991. Cancer 1997; 79564-573. 31. Samaan NA, Schultz PN, Hickey RC, Goepfert H, Haynie TP, Johnston DA, OrdonezNG. Well differentiated thyroid carcinoma and the results of various modalitiesof treatment: a retrospective review of 1599 cases. J Clin Endocrinol Metab1992; 75:714-720. 32. American Joint Committee on Cancer. The Thyroid In Gland. AJCC Cancer Staging Manual, 5th Edition, Lippincott-Raven, Philadelphia,1992; pp. 59-64. 33. Sweeney DC, Johnston GS. Radioiodine therapy for thyroid cancer. Endocr Metab Clin North Am 1995; 24803-840.
34. Alam MS, Takeuchi R, Kasagi K, i y ~ o t oS, Iida U, daka A, Konishi J. Value of combined t e c ~ e t i u m - 9 9hydroxy ~ methylene diphosphonate and th~lium-201 carc~oma.Thyroid 1997; 7:705-712. in detecting bone meta ant CS. Predicting outcome in papillary Bergstralh EJ, Goellner thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated in one insti~tionduring 1940 through 1989. Surgery 1993; 114:1050-1058. 36. Heufelder AEi, Gorman CA. Radioio~netherapy in the ~eatmentof differentiated thyroid cancer: guidelines and consideration docrinologist 1991; 1:273-280. gier P, S ~ a z i D. n E~ternal radiotherapy 37. Tubiana M, Haddad E, S c ~ u ~ b e r ~ e r in thyroid cancers. Cancer 1985; 55 38. Simpson WJ, Carruthers JS. The role of external radiation in the management of pap ill^ and follicular thyroid cancer. Am J Surg 1978; 136:457. J Endocrinol Invest 1987; 10:303. 39, Ahuja S, Ernst H. ~h~motherapy of , DeConti R. A r a n d o ~ z e dtrial of 40. Shimaoka K, Schoenfeld DA, Dew doxorubicin versus doxorubicin plus cisplatin in patients with advanced thyroid carcinoma. 1985; 562155-60. ,Leeper RJ2, Treatment of locally advanced thyroid carcinoma with a combi~ation of doxorubicin and radiation therapy. Cancer 1987; 60:237~-2375. 42. Mazzaferri EL. An overview of the m~agementof papillary and follicular thyroid cancer. Thyroid 1999; 9:421-27. 43. Ladenson PW, Strategies for thyrotropin use to monitor patients with treated thyroid carcinoma. Thyroid 1999; 9:429-33.
17 Papillary Carcinoma Cytology and Pathology James Oertel and Yolanda Oertel Papillary carcinoma is the most common thyroid cancer, constituting 75% to 85% of the malignant thyroid lesions in regions where iodine-deficiency goiteris no longer present (I).It represents most of the thyroid cancers that occurin children and young adults, whether idiopathicor radiation-related. A small proportionis familial. Generally, papillary carcinomas grow slowly and spread mostly by lymphatic vessels. The majority are infiltrative and without a capsule. Both gross and microscopic features are quite varied, depending on cellularity, amount and type of stroma, and the content of colloid (2,3). Nondiploid papillary cancers (1,4) and those having N-rus mutations are more likely to have metastases and to cause death (5).
NOT OTHERWISE SPECIFIED CLASSICAL PATTERN The not otherwise specified (NOS) papillary carcinoma (classical ispattern) a mixture of neoplastic papillae (Fig. 1) and follicles, sometimes with several tiny solid regions (3,6). Minor cystic changes are common. The gross appearance is of a firm, opaque mass, usually poorly defined and with a granular or finely nodular cut surface. Irregular scarring is common, and foci of calcification frequently can be found. If part of the tumor is rich in colloid, this is translucent and gelatinous with some resemblance to normal thyroid tissue or an adenomatoid nodule. If psammoma bodies are numerous, the tissue feels gritty, and it should be decalcified. In histologicalsectionsthecellsarelargerthannormalfollicularcellsandare cuboidal to low columnar (Fig.2). The cytoplasm is typically amphophilic to slightly eosinophilic. Nuclei are relatively large (but vary somewhat in size), ovoid, and subtly irregular in shape and in their positions in the cells. Nuclear indentations and round intranuclear inclusions of cytoplasm are common, although thesevary in number and degree in different tumors and in different parts of the same tumor. Nucleoli are usu close to the nuclear membranes, and so also is the heterochromatin, thereby causing the nuclear membrane to appear “thick” and much of the interior of the nucleus to be “pale,” “empty,” “clear,” or “ground glass” in appearance. Follicles may be colloidfilled or empty and range from tiny to large. Many are irregular in shape and may be
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, NI
193
Oer~eland Oer~el
.
~ a p i l carcinoma. l ~ ~apillaeof various sizes are evident. The tumor is unusual because most cells are of oxyphilic type. (W&E stain. x75).
.
pill^ carcinoma. The nuclei appear crowded together, and their variation in shape is apparent. Some have pale, seemingly empty centers (W&E stain. x400).
cytozogy and Pathology
195
Fig. 3. Papillary carcinoma. Elongated sinus follicles and papillae are crowded together. (H&E stain. x150).
elongated (almost tubular) (Fig. 3). Papillae vary greatly in size and complexity 1). (Fig. Each papilla consists of a fibrovascular core covered by a single layer of cuboidal to low columnar cells. The longer papillae are typically twisted and slightly irregular. Nearly all these cancers contain immunoreactive thyroglobulin and keratins. Psammoma bodies occur in about40% (2). Foci of irregular calcification (even ossification) are moderately common. The colloid may be dense or thin.Dense colloid often appears “stringy” or globular. Fibrosis occurs in an irregular pattern, often as trabeculae or nodules of dense collagenous tissue (7). A latticework of dense fibrous tissue is present frequently. Chronicinflammatorycells,mostlylymphocytes,maybenumerouswithinand around a papillary neoplasm. Sometimes the papillae are filled with lymphocytes and 4). Presumably this inflammatory response or histiocytes, often foamy histiocytes (Fig. is areflectionofthehighincidenceofthedepositionofimmunoglobulin G and complement factors on the cells of papillary carcinoma (8). Histiocytes in the tumor may contain lipofuscin and/or hemosiderin, particularly when hemorrhage and/or cystic change are present. Foreign-body-type giant cells (multinucleated histiocytes) are moderately frequent 5). Some may be closely associated to numerousin the classic papillary carcinomas (Fig. with psammoma bodies. Multiple foci of cancer cells are common in the thyroid, both as spread through the lymphatic vessels and as several simultaneous primary sites. Cervical lymph nodes are the most common sitesof metastatic foci. The cytologic smears contain large numbers of cells (“tumor cellularity”) (Fig. 6). Tight clusters of neoplastic cells, some in a papillary arrangement, and single cells are
196
Oertel and Oertel
c
‘“rr6?‘3
”
.
Fig. 4. Papillary carcinoma. Foamy histiocytes lie in thestroma of thepapillae. stain. x300).
(H&E
observed. These cells have enlarged nuclei (at least twice the size of red blood cells dense chromatin, sharp nuclear borders, and variable shapes (round, ovoid, triangu Nucleoliareseldomseen.Nuclear“grooves”arerarelyevidentinsmearsstained with
[email protected] cytoplasmic inclusions are seen frequently (Fig.7). The cytoplasm is usually dense and well demarcated. Colloid may be scant to abundant, appearing as thick strands (“ropy” or “bubble-gum” colloid) (9,10), irregular masses, or dense balls that stain bright pink to lavender. Psammoma bodies are seen frequ (Fig. 8). Multinucleated histiocytes are common and often are a conspicuous (11) feature
FOLLICULAR VARIANT
A diagnosis of the follicular variant of papillary carcinoma may be made when more than 70% of the histological patternis composed of neoplastic follicles (Fig. 9). Such neoplasms are often small and are usually less fibrotic or cystic than the papillary carcinomas NOS, but the amount of colloid present and the size of the follicles vary considerably. If follicles are small and contain little colloid, and if papillae are tiny and scarce, the tumor will appear fleshy and opaque on gross examination; it may be misclassified as a follicular adenomaor follicular carcinoma on microscopic examina tion. If most folliclesare medium-sized to large and thereis abundant colloid, then the tumor resembles an adenomatoid nodule or macrofollicular adenoma, both grossly microscopically; it may be mistaken for an adenomatoid nodule. This is the diffuse follicular variantor the macrofollicular variant (12). Psammoma bodies and multinucle ated histiocytes are usually sparse.
Cytology and PathoZogy
197
Fig. 5. Papillary carcinoma. Multinucleated histiocytes(giant cells) are surrounded by neoplastic cells, some with oxyphilic cytoplasm, a few with clear cytoplasm.A dense infiltrate of lymphocytes is present. (H&E stain. x250).
Fig. 6. Papillary carcinoma. Aspirate showing "tumor cellularity."@iff-Quik stain. x200).
198
Oertel and Oertel
Fig. 7. Papillary carcinoma. Aspirate showing neoplastic cells with variation in nuclearsizes. Note the "intranuclear cytoplasmic inclusion" in large nucleus on the left. (Diff-Quik stain. ~400).
Fig. 8. Papillary carcinoma. Multinucleated histiocyte partly surrounded by neoplasticcells. stain. x400). Many erythrocytes lie in the background of the smear. @iff-@&
cytozogy and PafhoZogy
199
Fig. 9. Papillary carcinoma,.follicular variant. Only neoplastic follicles are visible; some are filled with colloid, and others are empty. (H&E stain. x150).
Smearsfromfineneedleaspiration(FNA)aremarkedlycellular.Theenlarged 10) or lie in neoplastic cells form small follicles with well demarcated lumina (Fig. clusters forming rosettes and tubules (Fig. 11). Occasional papillary tissue fragments may be seen. The nuclei are dark-staining, have smooth contours, and vary in size and shape; a few are triangular, resembling arrowheads. Pink-staining colloid appears inside follicles or as balls and massesof variable shapes in close proximity to the neoplastic cells. Intranuclear cytoplasmic inclusions, psammoma bodies, and multinucleated histio cytes are less frequent than in the classic type of papillary carcinoma (13).
ENCAPSULATED VARIANT About 10%to 20% of the.papillary cancers that are not microcarcinomas are encapsulated variants. A moderate number are cystic as well as encapsulated, so the gross appearance is extremely varied. Capsules range from delicate to thick (Fig.12). These neoplasms have a lower incidenceof nodal metastases than the other types, but careful searchoftheperipheryofsuchatumorusuallyrevealsmicroscopicevidenceof just outside the penetration of its capsule orfoci of neoplastic cells in the thyroid tissue tumor’s capsule. Therefore, gross evidence of invasion is often absent, but microscopic evidence of aggressiveness is usually present. Many papillary carcinomas have irregular, dense fibrous tissue that accompanies their infiltrating cells, producing asort of pseudocapsule around partof the neoplasm. This should be differentiated from the well-organized, continuous true capsule that we
Oertel and Oerte
200
.
Papillary carcinoma, follicular variant. Aspirate showing multiple tiny, empty folli cles. (Diff-Quik stain. ~400).
.
Papillary Carcinoma, follicular variant. Aspirate showing rosettes and tubules. Vari tion in nuclear sizes and shapes is apparent. (Diff-Quik stain. ~400).
cytozogy and PuthoZogy
202
Fig. 12. Papillary carcinoma. A tiny carcinoma with a thick capsule is present. Thyroid tissue shows changes consistent with treated Graves’ disease. (H&E stain. x63).
on the right
have just described. Cytologic smears are diagnosed on FNA.
as classic papillary carcinoma
PREDOMINANTLY PAPILLARY
A few papillary carcinomas are composed almost exclusively of papillae, and neo tic follicles are rare. The cytologic smears are those of the classic papillary carcinoma. CYSTIC CARCINOMA
Tiny cystic foci are fairly common in papillary carcinoma, but a few cancers includ of tissue one or two cysts that occupy most of the lesion. The fluid often contains bits andflecksofcalcificmaterial.Papillaryfrondsmaybevisibletothenakedeye. as a fluid-filled sac. Sometimes there is so little epithelium that the tumor appears These cancers are often encapsulated, which suggests a reduced riskof metastases. Aspiration yields at least 1.0 ml of fluid, often a larger amount, typically thin and pale yellow, greenish, or brown, or thick and brown. The fluid may reaccumulate rapidly. After evacuation of the fluid and collapse of the cyst, if a residual mass can be detected, this should be aspirated. It is important to remember that most cystic thyroid masses are benign (14). Direct smears and smears from sediment after centrifugation of the fluid should be prepared. Large numbers of hemosiderin-laden histiocytes and considerable cellular debris are present. Sheets of intact follicular cellsmay be seen, which resemble those from cellular adenornatoid.nodu1es (15);.these may-be neoplastic cells, however. In a
Oertel and Oerte
202
3
Fig. 13. Papillarycarcinomawith cystic change.Aspirateshowingneoplastic cells with dense, well-demarcated cytoplasm on the left. They contrast with larger neoplastic cells wit pale cytoplasm mimicking histiocytes (on the right). @iff-@& stain. x400).
benign cystic lesion such epithelial cells should be shrunken and degenerate, so th presence of apparently well-preserved cellsis a warning of a possible neoplasm. Th cells are slightly larger than normal follicular cells, their cytoplasm is denser, they lac paravacuolargranules,andtheirnucleiarelargerthannormalnucleiandarenot pyknotic. Also seen are some groups of larger cells with clear cytoplasm and dense, convex cellular borders (Fig.13). The edges of these cellular clusters have a scallop appearance (16). In other cases the predominant cell type resembles a histiocyte wit clear cytoplasm andan enlarged, atypical nucleus; these are presumably partly de ated neoplastic epithelial cells (17). Dense globules of pink-staining colloid (“pink balls”) may be seen. EXTENSIVELY FIBROTIC CARCINOMA
Extensively fibrotic tumors are rare and typically infiltrative. Very little epitheliu (18), or fasciitislike (3). Th is present, and the stroma may be dense, myxomatous gross appearance depends on the amount and density of the stroma. We have no this diagnosis on cytological smears. DIFFUSE PAPILLARY CARCINOMA
In diffuse papillary carcinoma, most of a lobeor the entire thyroid gland is involved The involved tissue is firm, pale, and opaque. Usually no discrete mass can be foun The lymphatic vessels of the gland are permeated by the cancer, and typical au
CytoZogy and Pathology
203
Fig. 14. Papillary carcinoma (microcarcinoma). The tiny tumor is surrounded by normal thyroid parenchyma. Note the irregular shapes of the neoplastic follicles. (H&E stain. ~ 2 0 0 ) .
thyroiditis is usually present (19). Often there is diffuse fibrosis throughout, but not always. Psammoma bodies andfoci of squamous metaplasia in the cancer are common. Adjacent lymph nodes are involved often, and spread to the lungs is common, but these features do not necessarily portend short survival because many of the patients are young women. We have used cytological smears from these cases for a diagnosis of classic papillary carcinoma. MICROCARCINOMA Small papillary carcinomas were identified 1.5 as cm in diameteror less and considlarge to be occult, ered “occult” (1,20). Because this size is nowregardedastoo carcinomas may beso labeled if each tumoris 1.0 cm or less in diameter.A carcinoma only a few millimeters in diameter is palpable if it is fibrotic and on the anterior surface of the gland; therefore, it is not truly “occult.” These tiny lesions probably should be called “microcarcinomas” or “minimal carcinomas” (Fig. 14). OXYPHILIC PAPILLARY CARCINOMA Only a small number of papillary carcinomas with oxyphilic cells occur, and only 1). They may be infiltrative or encapsulated. Some a few have been studied (Fig. pathologists believe that this variant is considerably more aggressive than the usual papillary carcinoma (21).Others have not been able to detect any clear differences,
204
Oertel and Oertel
Fig. 15. Papillary carcinoma, tall cell variant. There is considerable variation in the locat of the nuclei, although the majority are basal or central. (H&E stain. x300).
allowing for the other prognostic factors also present(22). Aspirates may be misdiag nosed as follicular neoplasms of Hiirthle cell type because of the cytoplasmic characte tics. In the oxyphilic papillary carcinoma nucleoli are rarely visible, intranuclear cy plasmic inclusions are common, and multinucleated histiocytes are frequently seen. CLEAR CELL PAPILLARY CARCINOMA
A few cells with clear cytoplasm are present in a modest proportion of papillary carcinomas. Only a few cancers have many or most of their cells with clear cytoplas This feature does not have any known prognostic significance. We have limit ence with this type of neoplasm on aspirations. TALL CELL VARIANT
The tall cell variantof papillary carcinoma is uncommon and has a poor progn (3,23,24). The neoplastic cell has a with a greater tendency to recur or metastasize height twice (or greater) than its width (Fig. 15). Cytoplasm is usually eosinophilic Many tumors are large, are extensively papilliferous, and occur mostly in middle-a or elderly persons (unfavorable prognostic features). Trabecular patterns have been Leu M1,a reported (24). In a small series all tall cell examples were positive for myelomonocytic marker from cluster designation group 15 (CD 15) (24). The tumor may fit with the less well differentiated group. The prognosis may be better when th tumor is heavily infiltrated by lymphocytes and plasma cells (25).
Cytology and
205
Recently, some attempts have been made to establish diagnostic criteria for the recognition of this variant in cytologic smears (26-29). These reports are based upon very small numbers of cases, in retrospective studies, and therefore conflicting statements have been made. Harach and Zusman (26) described the presence of papillary fronds in their three cases, Bocklage and colleagues (29)found them in only one case, (27) Kaw did not find them in his case, and Gamboa-Dom’nguez and colleagues (28) did not even mention them as diagnostic features. Some common characteristics have been reported: larger cells with abundant oxyphilic cytoplasm, more frequent intranuclear cytoplasmic inclusions, and fewer psammoma bodies than in the classic papillary carcinoma. LESS WELL DIFFERENTIATED PAPILLARY CARCINOMAS
Less well differentiated papillary cancers are classified as grades 2 and 3 by the Mayo Clinic Broders’ classification( I ) , and are described as “moderately differentiated papillary carcinoma”(30),that is, papillary carcinoma with marked atypia (multilayeredepithelium, notable variations in size and shape of the cells andnuclei, and nuclei with hyp matism and abnormal chromatin distribution) (31). Other investigators include the tall cell and columnar cell cancers as less well differentiated papillary carcinomas (23). Extensive trabecular pattern of growth has been stated to indicate a worse prognosis (32),but such cases may overlap with the tall cell variant (24). Patterns of solid growth might not be significant, especially in young persons (3). Focal necrosis and invasion of blood vessels also may indicate a higher grade of malignancy (33). When follicles are empty and closed or when papillae or trabeculae are pressed together, a neoplasm may appear solid, without actually having a solid, diffuse pattern of growth. True solid regions are moderately common in papillary carcinoma, but are usuallyonlyaminorcomponent.Sometimesthesolidfociaretheresultoffocal squamous metaplasia, and typically this is not a significant feature. Only in a few instances is the entire neoplasm solid or predominantly so. Aspirates may show malignant features without specifically resembling a papillary furthyroid carcinoma. We have diagnosed a few such cases as “carcinoma, cannot ther classify.” COLUMNAR CELL CARCINOMA
Columnar cell papillary carcinoma is rare and occurs in adults of all ages. It is usually a solid, nodular, light-colored mass, either encapsulated or infiltrative, which contains tall, slender, columnar cells arranged in patterns that are papillary or trabecul (34-37) (Fig. 16). Solid regionsmay occur with small polygonal and/or spindled cells. Follicles of various sizes may be present. An alveolar pattern is sometimes suggested. Cytoplasm is usuallyclear,sometimeseosinophilicoramphophilic,and is scanty. Nuclei are hyperchromatic, rarely pale, are elongated in the tall cylindrical cells, and may contain longitudinal grooves; intranuclear cytoplasmic inclusions are rare. These elongated nuclei differ sufficiently in their positions in the cells to produce a stratified or pseudostratified appearance. Nucleoli are inconspicuous. Mitotic figures are numerous (37). The cells contain glycogen, thyroglobulin,and sometimes keratin. A few psammoma bodies may be found.
206
Oertel and Oerte
Fig. 16. Columnar cell carcinoma. There is a cribriform pattern with small papillae. Part of the tumor resemble tall-cell papillary carcinoma. No colloid is seen. (H&E stain. ~ 1 5 0 )
Neoplasms have been reported in which the columnar cell pattern was mixed w tall cell papillary carcinoma (38,391, as well as with solid regions of typical papillary carcinoma (36,39,40). Also, we have seen extensive insular and trabecular patterns adjacent to the columnar cell pattern. Reports suggest that the locally infiltrative tumors are usually fatal (34,38,39,41), but those that are encapsulated may be successfully resected (36,37). We have n personal experience with aspirates from this neoplasm. One report has described n ous papillary fragments composed of pseudostratified columnar cells crowded to (42). Their hypochromatic elongated nuclei were oriented perpendicular to the s of the papillae. Cytoplasm was pale. Empty spaces resembling follicles were noted, but no colloid was seen.
REERENCES
1. Hay ID. Papillary thyroid carcinoma. Endocrinol Metab Clin North Am 1990;19:545-576 2. LiVolsi VA. Surgical pathology of the thyroid. Major Probl Pathol 1990; 22:136-172. 3. Rosai J, Carcangiu ML, DeLellis RA. Tumors of the thyroid gland. In Rosai J, Sobin LJ editors. Atlas of tumor pathology, 3rd Ser; Fasc 5). Washington, DC: A.F.I.P., 1992. 4. Nishida T, Nakao K, Hamaji M, Nakahara M, Tsujimoto M. Overexpression of p53 and DNA content are important biologic prognostic factors for thyroid cancer. Surgery 1996; 119~568-575. 5. Hara H, Fulton N, Yashiro T, It0 K, DeGroot LJ, Kaplan EL. N-Ras mutation: an indepe
cytozogy and Pathology
207
prognostic factor for aggressiveness of papillary thyroid carcinoma. Surgery 1994;116:
1010-1016. 6. Cuello C, Correa P, Eisenberg H. Geographic pathology of thyroid carcinoma. Cancer 1969; 23:230-239. 7. Isarangkul W. Dense fibrosis: another diagnostic criterion for papillary thyroid carcinoma. Arch Path01 Lab Med 1993; 117545-646. B, et al. Tumor-specific deposition of immunoglobulin 8. Lucas SD, Karlsson-Parra A, Nilsson G and complement in papillary thyroid carcinoma. Hum Path011996; 27:1329-1335. P-0. Aspirationbiopsy 9. Mwhagen T, Willems J-S, LundellG,SundbladR,Granberg cytology in diagnosisof thyroid cancer. World J Surg 1981; 5:61-73. In 10. Abele JS, Miller TR. Fine-needle aspirationof the thyroid nodule: clinical applications. Clark OH, editor. Endocrine surgery of the thyroid and parathyroid glands, 1st ed. St. Louis: CV Mosby, 1985; 293-366. 11. Kini SR, Miller JM, Hamburger JI, Smith MJ. Cytopathology of papillary carcinoma of the thyroid by fine needle aspiration. Acta Cytol 1980; 24:511-521. 12. Albores-Saavedra J, Gould E, Vardaman C, Vuitch F. The macrofollicular variant of papillary thyroid carcinoma: a study of 17 cases. Hum Pathol 1991; 22:1195-1205. 13. Gallagher J, Oertel YC, Oertel JE. Follicular variantof papillary carcinoma of the thyroid: fine-needle aspirates with histologic correlation. Diagn Cytopathol 1997; 16:207-213. S , Cunningham JJ, Mazzafeni EL. Cystic thyroid 14. de 10sSantosET,Keyhani-Rofagha nodules: the dilemma of malignant lesions. Arch InternMed 1990; 150:1422-1427. 15. Busseniers A E , Oertel YC. “Cellular adenomatoid nodules” of the thyroid review of 219 fine-needle aspirates. Diagn Cytopathol 1993; 9:581-589. 16. Oertel YC. Fine-needle aspiration and the diagnosisof thyroid cancer. Endocrinol Metab Clin North Am 1996; 25:69-91. 17. Droese M. Atlas and manual: aspiration cytology of the thyroid, 2nd ed. Stuttgart: Schattauer, 1995. 18. Ostrowski MA, Asa C L , Chamberlain D, MoffatFL, Rotstein LE. Myxomatous change in papillary carcinoma of thyroid. Surg Path01 1989; 2:249-256. 19. G6mez-Morales M, Alvko T, Muioz M, et al. Diffuse sclerosing papillary carcinoma of the thyroid gland: immunohistochemical analysis of the local host immune response. Histopathology 1991;18:427-433. 20. Rosen IB, Azadian A, Walfish PG. Adverse aspects of small thyroid cancer and need for treatment. Head Neck 1995; 17:373-376. 21. Schr6der S. Pathological and clinical featuresof malignant thyroid tumours: classification, immunohistology, prognostic criteria. New York Gustav Fischer, 1988. 22. Beckner ME, Heffess CS, Oertel JE. Oxyphilic papillary thyroid carcinomas. Am J Clin Pathol 1995;103:280-287. E, Rilke F. Poorly differentiated forms of papillary 23. Pilotti S, Collini P, Manzari A, Marubini
thyroidcarcinoma:distinctiveentitiesormorphologicalpatterns?SeminDiagnPath01 1995;12:249-255.
24. Ostrowski ML, Merrino MJ. Tall cell variant of papillary thyroid carcinoma: a reassessment type of papillary carcinoma and immunohistochemical study with comparison to the usual of the thyroid. Am J Surg Path01 1996; 20:964-974. 25. Ozaki 0,It0 K, Mimura T, Sugino K, HosodaY. Papillary carcinoma of the thyroid: tallcell variant with extensive lymphocyte infiltration.Am J Surg Path01 1996; 20:695-698. tall variantof thyroid papillary carcinoma. 26. Harach HR,Zusman SB. Cytopathology of thecell Acta Cyt01 1992; 36~895-899. 27. Kaw YT. Fine needle aspiration cytologyof the tall cell variantof papillary carcinoma of the thyroid. Acta Cytol 1994; 38:282-283.
208
Oertel d
28. Gamboa-Dom'nguez A, Candanedo-GonzBilez F, Uribe-Uribe NO, Angeles Tall cell variant of papillary thyroid carcinoma: a cytohistologic correlation 1997; 411672-676. 29. Bocklage T, DiTomasso JP, Ramzy I, Ostrowski ML. Tall cell variant of pap carcinoma: cytologic features and differential diagnostic considerations. Diag 1997; 17~25-29. 30. Tscholl-Ducommun J, Hedinger CE. Papillary thyroid carcinomas: morphology sis. Virchows Arch A Pathol Anat Histopathol 1982; 396:19-39. 3 1. Tennvall J, Biorklund A, Moller T, Ranstam J, h e r m a n M. Prognostic factors follicular and medullary carcinomas of the thyroid gland: retrospective multiva of 216 patients with a median follow-up of 11 years. Acta Radio1 Oncol 198 32. Mizukami Y, Noguchi M, Michigishi T, et al. Papillary thyroid carcinoma i Japan: prognostic significance of histological subtypes. Histopathology 1992; 33. Akslen LA. Prognostic importance of histologic grading in papillary thyroid Cancer 1993; 72:2680-2685. 34. Evans HL. Columnar cell carcinoma of the thyroid: a report of two cases of a variant of thyroid carcinoma. Am J Clin Pathol 1986; 85:77-80. 35. Hwang TS, Suh JS, Kim YI, et al. Poorly differentiated carcinoma of the thyro tive clinical and morphologic evaluation. J Korean Med Sci 1990; 5:47-52. 36. Ferreiro JA, Hay ID, Lloyd RV. Columnar cell carcinoma of the thyroid: re additional cases. Hum Pathol 1996; 27: 1156-1 160. 37. Evans HL. Encapsulated columnar-cell neoplasms of the thyroid: a report o suggesting a favorable prognosis. Am J Surg Pathol 1996; 20: 1205-121 1. 38. Akslen LA, Varhaug JE. Thyroid carcinoma with mixed tall-cell and columnarAm J Clin Path01 1990; 94:442-445. 39. Mizukami Y, Nonomura A, Michigishi T, Noguchi M, Nakamura S, Hashimoto cell carcinoma of the thyroid gland: a case report and review of the literature. 1994; 25: 1098-1 101. 40. Gaertner EM, Davidson M, Wenig BM. The columnar cell variant of thyr carcinoma: case report and discussion of an unusually aggressive thyroid pa noma. Am J Surg Pathol 1995; 19:940-947. 41. Sobrinho-Simoes M, Nesland JM, Johannessen JV. Columnar-cell carcinoma: ant of poorly differentiated carcinoma of the thyroid. Am J Clin Pathol 1988; 42. Hui P-K, Chan JKC, Cheung PSY, Gwi E. Columnar cell carcinoma of the needle aspiration findings in a case. Acta Cytol 1990; 34:355-358.
18 Surgical Approach to Papillary Carcinoma Orlo H. Clark
General aspects of the approach to patients with thyroid nodulesare which proven to be malignant are discussed earlier in this volume. Absent prospective studies comparing various surgical or postsurgical therapies, the debate over which procedure might be best is likely to continue. Most thyroid surgeons agree that the minimal operation for a thyroid nodule suspicious for malignancy is a total thyroid lobectomy and isthmecto on the side of the nodule. The reasonthis forrecommendation is that if further surgery scararea tissue, and risks of complications, is needed one does not have to operate inofan such as hypoparathyroidism or recurrent laryngeal nerve injury, should be minimized ( I ) . Also, thyroid tissue' remaining after a partial thyroidectomy may be difficult to remove during a second procedure because its adherence of to the surrounding structures. This chapter addresses aspects of the surgical management of papillary carcinoma of the thyroid, and special considerations regarding follicular, medullary, and anaplastic carcinoma of the thyroid are discussed in the respective chapters later in this volume. One reason for the controversy concerning the extent of thyroidectomy required,is that most patients with papillary thyroid cancer have an excellent prognosis. Thus, patients with occult(e1 cm) papillary thyroid cancers without nodal involvement have a 6% to 8% recurrence rate but only0.2% mortality rate (2,3). It is, therefore, difficult to improve upon these numbers. When lymph node metastases are present or when there is angioinvasion within the occult papillary thyroid cancer, the recurrence rate and death rates are higher (2,3). We recommend total or near-total thyroidectomy for virtually all patients with papillary thyroid cancer larger than 1 cm. We realize that TNM, AGES, AMES, the mortality rateof patients considered to be at low risk by the or MACIS classifications is less than 5% and about 75% of all patients with thyroid (4). We believe, however, that if we can decrease cancer would be classified at low risk this mortality rate further it is advisable, unless the improvement in survival rate is countered by a high complication rate.It is also important to mention that the AGES, A M E S , and MACIS classifications are postoperative classifications. For example, local invasion,tumordifferentiation, resectability, and even distant metastases, often are 13*1 scan and serum not recognized until after the operation or after a post operative thyroglobulin determination. Grant and colleagues (5) also reported that recurrent cancer is less common after bilateral thyroid operations in both low- and high-risk patients,
From: Thyroid Cancer: A Comprehensive Guide to clinical Management Edited by:L. Wartofsky 0 Humanu Press Inc., Totowu, NJ
209
210
Clark
as determined by the AGES classification. DeGroot and colleagues (6)and Mazzaferri and Jhiang (7) also reported fewer recurrences and improved survival in patients afte total or near total thyroidectomy. The major reasons we recommend total thyroidectomy is that one can then make use of serum thyroglobulin levels and radioactive scanning to determine if all tumor has been removedor needs to be ablated with radioiodine(8-10). One might ask why not wait to see if patients will develop recurrent disease because most patients do n develop recurrent tumor. The problem with this approach is that once the recurrent tumor becomes clinically evident or is evident on a chest radiograph the chance of curative therapy with 13'1 decreases from about 70% to about 7% in tumors that take up radioiodine (9-11). Other reasons for total thyroidectomy include that it removes 30% to 87% of patients, it lowers the multifocal or bilateral disease that occurs in recurrence rate, and it probably improves survival, since one-third of tothe halfpatients who develop recurrent thyroid cancer die of thyroid cancer (8-10). The easiest time to do a total thyroidectomy is also at the initial operation. Near-total thyroidectomy, leaving less than 1 g of thyroid tissue, rather than total thyroidectomy should be done when the surgeon is concerned about the viability of the parathyroid glands or the recurrent laryngeal nerve during the initial dissection of normal thyroid tissue on the contr the sideof the tumor. Leaving a small remnant eral side to the tumor, that can subsequently be ablated with 13'1, is preferable in this situation. Numerous retrospective studies report that total or near-total thyroidectomy followed by I3II ablation and TSH suppressive therapy results in the fewest recurre and the best survival(6-8). Before thyroidectomy and during the initial thyroid ope tion,thesurgeonshouldcarefullylookforandpalpateforenlargedlymphnodes adjacent to the thyroid tumor and medial or lateral to the carotid sheath. All nodes i in the lateral nec the central neck should be removed and patients with palpable nodes benefit from an ipsilateral modified radical neck dissection. For most patients today the histologyof the thyroid tumoris known preoperatively because a fine needle biopsy is usually done.before the operative procedure. Needle biopsy is quite accurate for papillary, medullary, and anaplastic thyroid cancers, but cannot differentiate between a follicular or Hiirthle cell adenoma and a follicular or Hurthle cell carcinoma. After needle biopsy, one can usually plan the definit and discuss what will be done with the patient preoperatively. At operation in pat with follicular or Hurthle cell neoplasms by cytological examination, the surge look for lymph nodes and if present remove them for frozen section examina half of patients with follicular neoplasms by cytological examinations who are found to have thyroid cancer have a follicular variant of papillary thyroid cancer; nodal involvement is quite common in these patients, although only about 10% of patients with follicular cancer have nodal involvement (12). Frozen section examination is unfortunately not very effective in differentiating between benign or malignant follicular or Hiirthle cell neoplasms. One does know, however, that follicular and Hiirthle cell neoplasms thatare larger than 4 cm or occu In patients with follicular or Hiirth in older patients are more likely to be malignant. cell neoplasms, I usually do a thyroid lobectomy, as most patients will have benign and disease. I also discuss the various situations with the patient before surgery inform them that about10%of the time a second operation-completion total thyroidectomy-
Approach Surgical
to Papillay Carcinoma
211
may be necessary if cancer is only diagnosed by permanent histological examination. Patients who have solitary follicular adenomas usually do not require thyroxine po atively since recurrent follicular adenomas are rare.
REFERENCES 1. Levin KE, Clark AH, Duh QY, Demeure M, Siperstein AE, Clark OH. Reoperative thyroid surgery [see comments]. Surgery 1992;111:604-609. 2. Mazzafem EL, Jhiang SM. Differentiated thyroid cancer long-term impactinitial of therapy. Trans Am Clin Climatol Assoc 1994; 106:151-168; discussion 168-170. 3. MazzafeniEL.Papillarythyroidcarcinoma:factorsinfluencingprognosisandcurrent 1988; 15:xl. SeminOncol 1987; therapy[publishederratumappearsinSeminOncol 14~315-332. 4. Hay ID.Papillary thyroid carcinoma. Endocrinol Metabol Clin Am North 1990; 19:545-576. 5. Grant CS, HayID, Gough IR, Bergstralh EJ, Goellner JR, McConahey M. Local recurrence in papillary thyroid carcinoma: is extent of surgical resection important? Surgery 1988; 104~954-962. 6. DeGroot LJ, Kaplan EL, McCormick M, StrausFH. Natural history, treatment, and course of papillary thyroid carcinoma. J Clin Endocrinol Metab 1990; 71:414-424.
7. Mazzafem EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer [see comments] [published erratum appears in Am J Med 1995; 98:215]. Am J Med 1994; 97:418-428. 8. Clark OH. Total thyroidectomy: the treatment of choice for patients with differentiated thyroid cancer. Ann Surg 1982;196:361-370. 9. Clark OH, Levin K, Zeng QH, Greenspan FS, Siperstein A. Thyroid cancer: the case for total thyroidectomy. Eur J Cancer Clin Oncol 1988; 24:305-313. JP, Fragu P, 10. Schlumberger M, Tubiana M, De Vathaire F, Hill C, Gardet P, Travagli Lumbroso J, Caillou B, Parmentier C. Long-term results of treatment of283 patients with lung and bone metastases from differentiated thyroid carcinoma.J Clin Endocrinol Metab 1986; 63:960-967. 11. Casara D, Rubello D, Saladini G, Masarotto G, Favero A, Girelli ME, Busnardo B. Different
features of pulmonary metastases in differentiated thyroid cancer: natural history and multivariate statistical analysis of prognostic variables. J Nucl Med 1993; 341626-1631. 12. Emerick GT, Duh QY, Siperstein AE, Burrow GN, Clark OH. Diagnosis, treatment, and outcome of follicular thyroid carcinoma [see comments]. Cancer1993; 723287-3295.
Differentiated Thyroid Carcinoma Radioiodine Therapy-I Gerald Johnston and Diane Sweeney
Differentiated thyroid carcinomas can be subdivided on the basis of their histologi features into papillary and follicular varieties, as well as follicular variants of papillary carcinoma. Nevertheless, treatment with radioiodine is essentially the same for both papillaryandfollicular tumors. Aside from some Hiirthle cell and tall cell variant tumors, histology is not always a reliable way to characterize the future behavior of differentiated thyroid cancer. Since histologically well-differentiated thyroid carcinoma usually follow a relatively benign course, therapeutic regimens consisting of initial thyroid hormone suppressive therapy, nodulectomy, hemithyroidectomy, or total thyroidectomy with or without radioiodine ablation have all produced acceptable shortterm resultsin a number of studies. The more biologically benign tumors have respond to any orall of these treatments, and the more aggressive tumors require more aggress therapy. Even a majority of the latter will have a good outcome when the therapy includes total thyroidectomy or completion thyroidectomy with node dissection and follow-up radioiodine ablation with150-250 mCi I3*I.Some differentiated tumors that are apparently histologically identical to those having a benign course will inexplicabl show little or no response to therapy, andmay not even take up radioiodine. We have yet to find a way, with biological markers or on the basis of histological morphology, to identify which tumors will respond to therapy and which will not. Consequently, at this point in our incomplete understanding of the basis for differentiated thyroid cancers, we rely on a standard therapeutic approach for most of the histological typesof tumor which consists of near-total to total thyroidectomy, radioiodine ablation, and further radioiodine therapy if needed. As was stated above, the term “radioiodine therapy” is used for those patients with ablation” is the terminology reserved residual or recurrent thyroid cancer. “Radioiodine for removalof residual, normal thyroid tissue following thyroidectomy. Both situations require assumptions that may be difficult to establish absolutely. Freitasandcolleagues ( I ) list the indications for 1311 therapy for thyroid cancer as follows: 1. primary tumor is inoperable 2. Postoperative residual in the neck
From: Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited by: L. Wmtofsky 0 Humana Press Inc., Totowa, NJ
213
214 3. 4. 5. 6.
Johnston and Sweeney Distant metastases Invasion of the thyroid capsule Cervical or mediastinal node metastases Recurrent thyroid cancer
There is general agreement among practitioners with this listing. However, theories differ on the amounts of I3'I needed for proper therapy. A standard, empirical fixedA quantitated dosimetric approach based on calcu dose methodis most widely used. (2-5), which may be radiation doses delivered to cancer cells is an alternate approach gaining advocates. The most effective dose of I3'I can be calculated to bypass the variations in effective half-life and uptake from patient to patient. The calculated d 48 at hours is tailored to deliver200 cGy to blood with no more than 120 mCi retained (80 mCi limit at 48 hours in the presence of lung metastases). In one study of 44 patients given70 doses over5 years (6),the average dose was 304 mCi with the lowes of No severe side-effects were reporte dose being88 mCi andthe highest 541 mCi I3lI. for these patients. Hurley and Becker (7) use this approach but limit the dose to 300 mCi per therapy. (8-10) report achieving single Using a quantitative approach, Maxon and coworkers dose ablation in 84% of inpatient therapies and 79% of outpatient therapies, as well as successful treatment of 74% of lymph node metastases. The standard fixed dose approach toI3lItherapy (10-13) illustrated by Beierwaltes' treatment protocol, which varies only for the site of uptake, that is, not less than 100 mCi for uptake in the thyroid bed, not less than 150 mCi for uptake in cervical n and not less than 175 mCi for distant metastases. We have found the use of a stand dose regimen to be effective, safe, and efficientin time and cost. A small number of patients may do better with higher-dose therapy as provided by quantitated dosimetry. These include patients with nonresectable neck cancer or widespread metastases. These patients may benefit from a calculated maximal dose o 13'1 but this is yet to be proven (14). METASTATIC THYROID CARCINOMA
Lymph Node Metastases
Cervical or mediastinal lymph node metastases at the time of initial therapy for thyroid carcinoma have variedfrom 11% to42% in published series(15,16).A number of variables contribute to this including timing, aggressiveness of the treatment team and extensiveness of the lymph node dissection (17). In some series, the presence of cervical node metastases have not been found to affect life expectancy significantly (14,18,19). The recurrence rate has been significa (20,21). increased in patients observed to have cervical lymph node metastases initially However, radioiodine therapy has been observed to reduce both recurrence rate and death rate in retrospective studies of these patients (15,22).
Pulmona y Metastases Patients with lung metastases from differentiated thyroid cancer have also been studied in long-term retrospective reviews (19,23,24). Youth and positive radioiodine uptake are favorable prognostic criteria. The absence of bone metastases, treatment
Differentiated
215
Fig. 1. Anterior whole-body diagnostic scan done 3 days following 10 mCi 13'1 dose. There (2) andnasal is thyroid bed activity (arrow) and physiological uptake in the salivary glands mucosa (1). Activity in the left hemithordupper abdomen (3) provedto be activity in the stomach, not metastatic lung disease. This is a frequent false-positive finding.
with total thyroidectomy, and papillary histology were independent prognostic factors 10 for survival. Approximately halfof the patients meeting these criteria were alive at years (22,24,25). About25%wishbonemetastasessurvived10years (24). When both bone and pulmonary metastases were present, the 10-year survival was less than 15% (24). About half of these patients will have positive radioiodine uptake in pulmonary metastases seen on x-ray film. The mortality rate is lower in these patients than in those whose metastases do not concentrate radioiodine pigs. 1 and 2). The occurrence of late pulmonary metastases can be seen as an indicator of th 24 years after management. In patients whose pulmonary metastases occurred 1 to surgery:
Johnston and Sweeney
216
b
I
Fig. 2. Metastatic thyroid cancer to lungs and mediastinum-anterior chest images, ontwo different patients 3 days after 10 mCi I3lI. (A) Diffuse, bilateral iodine uptake in lung fields (arrows). This patient has had this pattern of uptake for 8 years following two courses of l3II She is only minimally symptomatic.(B) Focal uptake in mediastinum and both lungs (arrow This patient suffered right upper lobe collapse and postobstructive pneumonia.
1. Partial thyroidectomy was followed by an 11% incidence 2. Partial thyroidectomy with l3II therapy, a 5% incidence 3. Total thyroidectomy, a 3% incidence 4. Total thyroidectomy with I3'I therapy, a 1.3% incidence of late pulmonary metastases
Bone Metastases
Differentiated thyroid cancer will metastasize to bone on occasion (Figs. 3 and 4). Such metastases herald a poor prognosis. As such, bone metastases should be met w
Diferentiated Thyroid Carcinoma
21 7
Fig. 3. Three-hour delayed bone scan images utilizing 25 mCi 99"Tc methylene diphosphonate. (A) Posterior head and thorax showing at least two midline skull lesions (arrows) in a patient with metastatic follicular thyroid carcinoma. (B) Same patient, lateral head view. Three skull lesions (arrows) can be seen. These lesions were better visualized on bone scan than on I3'I survey scans.
218
~ o h n s ~and ~ nSweeney
.
metastatic survey scan and 99mTc MDP bone scan on the same patient, a 39-yearpreviously treated with subtotal thyroidectomy for follicular thyroid cancer who presented 12 years later with back pain, (A) Posterior chest and abdomen with makers placed anteriorly at s u p r a s t e ~notch ~ and xiphoid. Note a ~ n o ~l3II a laccumulation in the "12 and L2 vertebrae-biopsy proven to be metastatic thyroid cancer. ( ) Posterior whole-body bone scan: increased uptake in the same vertebrae. l3II
Thyroid Differentiated
219
a combined therapeutic approach with 1) surgery to remove as much of the tumor as This, of course, possible followed by 2)l3IItherapy and, finally, 3) x-ray beam therapy. following total thyroidectomy. The dose of l3II shouldbehigh,inthe200to250 mCi range or more. Chemotherapy does not appear to have a role in treating these bony metastases. Statistics available from two studies (2526) showed only 10% of bone metastases went into complete remission, and all of the cures receivedI3’Itherapy alone or with x-ray therapy. Over 50% of these patients were dead in a year. Of the 25% given only chemotherapy, none responded. Harness and coworkers(271, in a review, conclude that surgical resection is important to decrease tumor bulk, to resect solitary metastases, or for neurological and orthoped palliation. Debulking improves the response to radioiodine therapy, ranging from pal tion to cure. External radiation therapy offers benefits when used in conjunction with I3’I therapy (see Chapter 21 and 30).
Brain Metastases Brain metastases are rare in patients with thyroid carcinoma. Their early detection and treatment are crucial. The optimal approach is similar to the approach to bone tumors and combines prompt surgical removal followed by radioiodine therapy and, possibly, external radiation (28). LOCALLY INVASIVE THYROID CANCER Locally invasive surgically unresectable thyroid cancer is associated with a high cancer mortality and recurrence rate (15,29,30). In one study of 80 such patients, 35 (44%)died at a median follow-up of 2 years (30). The therapeutic approach is to use reasonable but aggressive surgery to remove as much tumor as possible. Radioiodine therapy is then used if its uptake can be demonstrated. REFERENCES 1. Freitas JE, Gross MD, Ripley S , et al. Radionuclide diagnosis and therapy of thyroid cancer: current status report. Semin NuclMed 1985; 15:106-131. of readioiodine dosimetry to results 2. Benua RS, Cicale NR. Sonenberg M, et al. The relation and complications in the treatment of metastatic thyroid cancer.AJR 1962; 87:171. 3. Maxon HR, Thomas SR, Boehringer A, et al. Low iodine diet in 1-131 ablation of thyroid remnants. Clin Nutr Med 1983; 8:123-126. 4. Van Nostrand DV, Neutze J,Atkins F. Side effects of “rational dose’’ iodine-131 therapy for metastatic well-differentiated thyroid carcinoma. J Nucl Med 1986; 27:1519. 5. Bushel1 D, Boles MA, Kaufman G,et al. Complications, sequela and dosimetry of iodine131 therapy for thyroid carcinoma. J Nucl Med 1992; 33:2214. 6. Leeper R D , Shimaoka K. Treatment of metastatic thyroid cancer. Clin Endocrinol Metab 1980; 9:383. 7. Hurley J R , Becker DV. Treatmentof thyroid carcinoma with radioiodine.In Gottschalk A, Hoffer PB, Potchen ET, Berger HJ, editors. Diagnostic nuclear medicine, 2nd ed. Baltimore: Williams & Willcins, 1988:792. 8. Maxon HR, Smith HR. Radioiodine-131 in the diagnosis and treatmentof metastatic welldifferentiated thyroid cancer. Endocrinol Metab Clin NorthAm 1990; 19:685-718.
220
Johnston an
9. Maxon HR, Thomas SR, Hertzberg VS, et al. Relation between effective ra and outcome of radioiodine therapy for thyroid cancer. N Engl J Med 1983; 10. Thomas SR, Maxon HR, Kereiakes JG, et al. Quantitative external counting enabling improved diagnostic and therapeutic decisions in patients with well-d thyroid cancer. Radiology 1977; 122:731. 11. Pacini F, Lippi F, Formica N, et al. Therapeutic doses of iodine-131 reveal metastases in thyroid cancer patients with detectable serum thyroglobulin le Med 1987; 28:1888. 12. Beierwaltes WH. The treatment of thyroid carcinoma with radioactive iodine. Med 1978; 8:79-94. 13. Beierwaltes WH, Nishiyama RH, Thompson NW, et al. Survival time and “cure” and follicular thyroid carcinoma with distant metastases: statistics following U Michigan therapy. J Nucl Med 1982; 23561-568. 14. Harbert JC. Radioiodine therapy of differentiated thyroid carcinoma. In: Nucle therapy. New York: Thieme, 1987. 15. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medica papillary and follicular thyroid cancer. Am J Med 1994; 97:418-428. 16. Massin JP, Savoie JC, Garnier H, et al. Pulmonary metastases in differenti carcinoma. Cancer 1984; 52:982-992. 17. Hay ID, Grant CS, van Heerden JA, et al. Papillary thyroid microcarcinoma 535 cases observed in a 50-year period. Surgery 1992; 112:1139-1147. 18. Cady B, Rossi R. An expanded view of risk group definition in differenti carcinoma. Surgery 1988; 104:947-953. 19. Rossi RL, Niedoda C, Cady B, et al. Malignancies of the thyroid gland. Surg Am 1985; 65:211. 20. Varma VM, BeierwaltesWH, Nofal MM, et al. Treatment of thyroid cancer: dea surgery and after surgery followed by sodium iodine 1131. JAMA 1970; 214 21. Harwood J, Diarck OH, Dunphy JE. Significance of lymph node metastasis in d thyroid cancer. Cancer 1979; 432310. 22. Samaan NA, Schultz PN, Hickey R, et al. The results of various modalities of well-differentiated thyroid carcinoma: a retrospective review of 1599 pat Endocrinol Metab 1992; 75:714-720. 23. Mazzaferri EL, Young RL, Oertel JE, et al. Papillary thyroid carcinoma: th therapy in 576 patients. Medicine 1977; 56:171. 24. Samaan NA, Schultz PN, Haynie TP, et al. Pulmonary metastasis of different carcinoma: treatment results in 101 patients. J Clin Endocrinol Metab 1985; 6 25. Schlumberger M, Tubiana M, De Vathaire F, et al. Long-term results of 283 patients with lung and bone metastases from differentiated thyroid carcin Endocrinol Metab 1986; 63:960. 26. Proye CAG, Dromer DHR, Carnaille BM, et al. Is it worthwhile to treat bon from differentiated thyroid carcinoma with radioactive iodine? World J Surg 1 27. Harness JK, Thompson NW, Sisson JC, et al. Differentiated thyroid carcinom of distant metastases. Arch Surg 1974; 108:410. 28. Pacak K, Sweeney DC, Wartofsky L, et al. Solitary cerebellar metastasis fro thyroid carcinoma: a case report. Thyroid 1998; 8:327-335. 29. Cody HS, Shah JP. Locally invasive, well-differentiated thyroid cancer. 1981; 142~480-483. 30. Rossi RL, Cady B, Silverman ML, et al. Surgically incurable well-different carcinoma. Arch Surg 1988; 123569.
20 Chemotherapy of Differentiated (Papillary or Follicular) Thyroid Carcinoma Lawrence S. Lessin and My0 Min SINGLE MODALITY TREATMENT WITH CHEMOTHERAPY
Chemotherapy is employed as a palliative measure in the 25% of recurrent inoperab or metastatic, follicular, or mixed thyroid cancers that do not concentrate I3'I. Doxorubicin, bleomycin, and cisplatin have been the mainstay of treatment. Since advanced thyroid cancer is rare, there are few meaningful clinical trials which compare single and multiple-agent chemotherapy for differences in efficacy and toxicity. The Eastern Cooperative Oncology Group designed and completed the only randomized study of doxorubicin vs doxorubicin plus cisplatin in inoperable, radioiodine-resistant advanced thyroid cancer in chemotherapy naive patients. In this study reported in 1985 ( I ) , 41 patients received doxorubicin alone and43 patients received the combination. Sixteen patients with differentiated thyroid cancers received single-agent doxorubicin, given at a dose of60 mg/m2intravenously every3 weeks. Nineteen comparable patients received the Combination of doxorubicin, 60 mg/m2 and cisplatin at 40 mg/m2 every 3 weeks. Treatment was discontinued when stable disease was achieved after 3 cycles of treatment, disease progressed after 2 cycles of treatment, or when total doseof doxorubicin exceeded 550 mg/m2. Suppressive thyroid hormone treatment was continued throughout the treatment in both groups of patients. The overall response rate forall patients was 21%. In the doxorubicin-alone arm, response rate was 17% compared to 26% in the combination arm; however, because of the small number of patients, this difference was not statistically significant. Moreover, neither the time to relapse nor overall su was found tobe statistically different among the two groups. Weight loss of more than lo%, presence of lung metastases, and poor performance status were found to be significant prognostic indicators. Treatment-related toxicity, predominantly hematological and gastrointestinal was worse in the combination group, but no fatalities were arm. Although the combination regimen was not statistically reported in either treatment superior to doxorubicin alone, for response rate and survival, all five complete res received the combination regimen. of Two orthe five complete responders had differentiated thyroid cancers.By contrast, a similar combination treatment tested by the Southeastern Cancer Study Group in 22 patients produced only two partial responses with
From: Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited b y : L. Wartofsky 0 Hulnana Press Inc., Totowa, NJ
221
222
Lessin and Min 100
6 80
E
8 60
a
2 40 3
I-
s
20 I
0
I
I
I
I
2
I
4 6 TIME AFTER TREATMENT (YEAR)
I
I
a
Fig. 1. Local tumor control rate as a function of time after combined treatment for patients with differentiated (group 1) versus anaplastic (group 2) thyroid cancers. (From Kim JH and Leeper R D , Cancer 602372-75, 1987, with permission.)
serious toxic reactions(2). In 1997, the St. Jude's group reported unsustained compl response of childhood papillary cancer to both doxorubicin and a combination of topotecan and carboplatin (3). CHEMOTHERAPY COMBINED WITH EXTERNAL BEAM RADIATION THERAPY
The combined modality approach for locally advanced, I3'I refractory, differentiated thyroid cancer using low dose doxorubicin combined with external beam radiatio prospectively studiedby Kim and Leeper(4). In their series,22 patients with histologi cally confirmed, well-differentiated papillary, follicular,or mixed thyroid cancer were of 10 mg/m2 per week by bolus injection with concomi given doxorubicin at a dose radiation therapy. Radiation was given at doses of 200 cGy/day for 5 days each week to a total tumor dose of5600 cGy. A 91% complete response rate was observed wit In patients with differentiated thyroid cance 77% long-term local control (see Fig. 1). overall survival was 50% at 5 years (Fig. 2). In this study, deaths of patients with differentiated thyroid cancers were due to distant metastases rather than local tumor invasion. All patients developed moderate, 3 to 4 weeks after initiation treatment-related pharyngoesophagitis and tracheitis within of combined chemoradiotherapy but none required cessation of treatment. Before this study, combined modality treatments for locally advanced thyroid cancer employing higher doses of doxorubicin were plagued by increased local tissue toxicity a systemic morbidity (5). This study provides an excellent and well-tolerated treatment option for locally advanced refractory differentiated (and anaplastic) thyroid cancer. In summary, single-agent or combination chemotherapy containing doxorubicin is effective in metastatic, refractory differentiated thyroid cancer, with improved sur in responders. For locally advanced cancer, combined low-dose chemoradiotherapy offers an effective means of palliation.
223
Chemotherapy of Diferentiated Thyroid Carcinoma 100
80
$
60
U
2 40 s
20
01 0
I
I
2
I
I
4
I
I
6
I
I
a
TIME AFTER TREATMENT (YEAR)
Fig. 2. The actuarial survival curvesof the two groups of patients after combined therapy. (From Kim JH and Leeper R D , Cancer 60:2372-75, 1987, with permission.)
REFERENCES 1. Shimaoka K, Schoenfeld D, De Wys W, Creech R, De Conti R.
A randomized trial of doxorubicin vs. Doxorubicin plus cisplatin in patients with advanced thyroid carcinoma. Cancer 1985; 56:2155-2160. 2. Williams SD, Birch R, Einhorn LH. Phase 11 evaluation of doxorubicin plus cisplatin in advancedthyroidcancer:aSoutheasternCancerStudyGroupTrial.CancerTreatRep
1986; 70:405-407. 3. Kuefer MU, Moinuddin M, Heideman RL, Lustig RH, Rose SR, Burstein S, VanMiddlesworth
L, Fleming I, Jenkins JJ, Shearer PD. Papillary thyroid carcinoma: demographics, treatment 1997; and outcome in eleven pediatrics treated at a single institution. Med Pediatr Oncol
28:433-440. 4. Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with combination doxorubicin and radiation therapy. Cancer1987; 60:2372-2375. 5. Tallroth E,Lundell G, Tennvall J, Wallin G. Chemotherapy and multimodality treatment in
thyroid carcinoma: disorders of the thyroid and parathyroidII.Otolaryngol ClinNorth Am 1990; 23523-527.
21 Management of Papillary Thyroid Carcinoma External Radiation Therapy Robert L. White A review of the useof external radiation therapy for the treatment of thyroid cancer has appeared recently ( I ) . The same group reported on their success with radioiodine and external radiotherapy in 382 patients with differentiated thyroid cancer,of whom 262 had papillary carcinoma of the thyroid(2). In general, external megavoltage irradiation plays a limited role in the management of the differentiated thyroid carcinomas. Generally speaking, papillary and mixed papillary-follicular tumors are more radiosensitive than follicular thyroid tumors (3). External megavoltage radiation can be used of theradioactiveiodinebut is inconjunctionwith 1311 tosupplementtheefforts especially important when the carcinoma does not accumulate1311. The primary indication for megavoltage external irradiation is bulky unresectable 1311or where1311may not be adequate thyroid carcinoma that either does not accumulate for local control of the tumor (4). Residual bulky tumor after surgery in many cases will not be controlled by 1311 alone, primarily because of circulation changes after surgery and surgical healing. Bulky mediastinal disease is of particular concern because of the difficulty in controlling the differentiated carcinoma 1311 by alone. When superior vena caval syndrome is present, external radiation will improve the of rapidity response 1311is not accumulated by the carcinoma. to 1311or may be the treatment of choice when are present, external radiation is indicated to prevent pathologiWhen skeletal metastases cal fractures, regardless of the concentration of 1311. When a patient develops brain metastasis from thyroid carcinoma, external radiation is indicated for a reliable and 1311may be altered rapid response. The ability of the brain metastasis to concentrate by the blood-brain barrier and thus the reliability of external irradiation is indicated in this clinical situation. External radiation is indicated for metastatic and locally recurrent thyroid carcinoma which occurs in spite of l3IIaccumulation or after maximalI3IItherapy. If the surgeon is concerned that the extent of the thyroid carcinoma is such that complete removalis not possible, then the use of preoperative external radiation alone or in conjunction withlS1Imay shrink or occasionally stabilize the tumor mass. Surgery following radiation may technically be easier and with less risk of operative blood loss when preoperative external radiation has been planned. Coordination between the
From: Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited by: L. Wartofsky6 Humana Press Znc., Totma, NJ
225
226
White
Fig. 1. Thetreatmentvolumeshouldinclude the entirethyroidgland,therightandleft cervical lymph nodes, right and left supraclavicular nodes and the superior mediastinum.
surgeon and the radiation oncologist is very important in the management of thyroid carcinoma to optimize the patient's treatment and timing and feeling of security. External irradiation may be used in sequence or in conjunction with chemotherap particularly where the thyroid carcinoma is anaplastic or poorly differentiated(5). Since and systemic chemotherapy optimal time dose relationships between external radiation have not been optimized, local agreement between the medical oncologist and r oncologist is important to help patients understand the importance of coordinating their treatment. For curative treatment for thyroid carcinoma with external megavoltage irradiatio there are many technically demanding details. The definitive dose for residual or bu thyroid carcinoma is 6500 cGy in 7 weeks with a daily dose of 180-200 cGy daily 5 days a week. The treatment volume (Fig. 1) should include the entire thyroid gland, the right and left cervical nodes, right and left supraclavicular nodes, and the superior mediastinum (6). It is necessary to pay particular attention to the spinal cord dose. Special blocking techniques with a cerrobend blocking system should limit the ra other radiation-sensitive structures. All of the treatm dose to the spinal cord as aswell of cancer couldbe present should be treated areas where microscopic or small deposits with doses of 5000 cGy over 5 to 6 weeks time. The spinal cord is shielded after 4500 cGy in 4.5 to 5 weeks time. Where tissue thickness results in doses of less than 5000 cGy in 5 to 6 weeks, boosting techniques must be employed to ensure that the dose is as uniform as possible. There are several methods of radiation beam arrangements and portals that allow adequate doses to be delivered to the neck and mediastinum.In most cases an anterior to posterior and posteriorto anterior set of portals with 'To, 4 or 6 MV photons will allow 4500 to 5000 cGy to be delivered in 4.5 to 6 weeks time Boosting techniques utilizing electron portsof 8 to 14 MeV can supplement the areas treated to 4500 to 5000 cGy to definitive doses of6500 to 7000 cGy in 5 to 8 weeks time. To avoid the spinal cord, in addition to cerrobend blocking, oblique anterior portals with wedges are occasionally utilized. Some of the newer treatment techniq 2) to optimize external include arching or rotational fields with flying wedges (Fig. irradiation to the treatment volume while minimizing treatment to the spinal cord or other critical structures. Clinical experience has documented that external irradiation with or without I3'I can produce long-term local control in patients with differentiated
Management of Papillary Thyroid Carcinoma
227
Fig. 2. Some of the newer treatment techniques include arching or rotational fields flying with wedges to optimize external irradiation tothe treatment volume while minimizing treatment to the spinal cord or other critical structures.
thyroid carcinomas who have microscopic residual or gross disease after surgery for up to 25 years (7). Skeletal, brain, hepatic, pulmonary, or subcutaneous metastasis of differentiated thyroid carcinoma may be treated with external megavoltage irradiation with or with 13*I.Obviously, if the metastatic thyroid cancer does not accumulate l3II, then external irradiation alone becomes the treatment of choice. Dose levels of 3500 to 4500 cGy
228
White
in 3 to 4.5 weeks are recommended for optimal palliation of metastasis to soft tissue or bone.Whenthere is a possibility of pathological fracture in the case of bone metastasis, stabilization with an intramedullary rod or other orthopedic procedur precede the external radiation. Patients who receive systemic chemotherapy and external irradiation concurrentl sequentially should not be treated with daily doses to exceed 180 cGy because of the possibility of undesirable dose potentiating side effects. In patients receiving che 500 cGy. apy, the dose to the spinal cord should be reduced by Daily management for t patient receiving combinations of chemotherapy and external irradiation is difficul of oral mucositis, requires close surveillance and observation. Usually the side effects are worse for patients treated with combined modalit esophagitis, and skin erythema and patients need to be carefully and cautiously observed regularly (8). The treatment of radiation-induced thyroid cancer would depend on the in findings as well as the clinical presentation. Overall the treatment is the same as for thyroid cancers not induced by radiation. The treatment modalities include surgery, thyroid hormone therapy, 1311therapy, external irradiation, interstitial irradiation, and chemotherapy. Interstitial irradiation is helpful and valuable in the treatment of primary thyroid carcinomas as well as metastatic carcinoma to the thyroid from other primary sites. lZsIhave been utilized in the clinical sett Removable 1921rand permanently implanted In addition, Ig2Ir has been implanted into mediastinal masses metastatic from thyroid carcinomas and sarcomas. Since thereis minimal general experience and few patients have been treated, the interstitial treatment has not been widely publicized. In experi enced hands, the interstitial irradiation techniques have produced long-term diseasefree survivalinpatientsandimprovedlocalcontrol. The advantage of interstitial irradiation includes minimal side effects and complications and improved local res siveness, but the clinical experience is limited.
REFERENCES
1. Brierley JD, Tsang RW. External radiation therapy in the treatmentof thyroid malignancy. Endocrinol Metab Clin N Am 1996; 25141-157. 2. Tsang RW, Brierley JD, Simpson WJ, Panzarella T, Gospodarowicz MK, Sutliffe SB. The effects of surgery, radioiodine, and external radiation therapy on the clinical outcome of patients with differentiated thyroid carcinoma. Cancer 1998; 82:375-388. 3. Greenfield LD.Radiation therapy in the management of thyroid carcinoma. In Greenfield LD, editor. Thyroid cancer. Boca Raton, FL:CRC; 1978:177-187. 4. Lindberg RD. Externalbeamirradiation in thyroidcarcinomas. In Hechor GH, editor. Textbook of radiotherapy, 3rd ed. Philadelphia: Lea 8z Febiger, 1980:384-388. 5. Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with combination doxorubicin and radiation therapy. Cancer 1987; 602372-2375. 6. Moss WT, Brand WN, Battifora H. The thyroid. In:Radiation oncology: rationale, technique results, 5th ed. St. Louis: CV Mosby, 1979:233-242. 7. Simpson WJ, McKinney SE, Carruthers JS, Gospodarowicz MK, Sutcliffe SB, Panzarella in 1578 patients.h e r J Med T. Papillary and follicular thyroid cancer. Prognostic factors 1987; 83~479-88. 8. Greenfield LD. Thyroid tumors. In Perez CA, Brady LW, editors. Principles ofpractice of radiation oncology, Philadelphia: JB Lippincott, 1987:1126-1156.
22 Papillary Thyroid Cancer Henry B. Burch STRATEGY
Effective surveillance for recurrent papillary thyroid cancer begins with an asses of the likelihood of recurrence or death from disease, based on individual character of the patient and their tumor. This information is used to determine an appropriate level of follow-up, which may vary from as little as an annual neck examination on replacement thyroid hormone therapy for occult lesions to annual or semiannual who body scanning (WBS) off thyroid hormone for high-risk patients. Effective follow-up is also contingent upon a current understanding of the strengths and limitations of the tools available for thyroid cancer surveillance. This chapter focuses on the rationale used to determine the method and frequencyof follow-up for papillary thyroid cancer patientsandgivesanoverviewofthecurrenttechniquesavailabletoaccomplish this objective.
WHAT LEVEL OF SURVEILLANCE? The factors used to determine an appropriate level of surveillance include individual patient characteristics, such as age and sex; tumor features, such as size, histological grade, and presence of extrathyroidal extension or distant metastases; and the extent (see Chapter25).Whileagreatdealof ofpriorsurgeryandradioiodinetherapy variability exists in the level and frequency of follow-up in patients with papillary thyroid cancer ( I ) , this evaluation should be individualized, with more rigorous and frequent monitoring in those patients deemed likely to experience a recurrence or dea from disease, and less intense surveillance for patients having a low likelihood of an adverse outcome. My approach to patients with papillary thyroid cancers larger than 1.5 cm in diameter is to recommend near-total thyroidectomy followed by radioiodine ablation with 50-100 mCi of 13'I. Patients with one or more poor prognostic factors (see Chapter 25) undergo WBS every 6 months for 18 months, and then annually for 5 years. Thereafter, WBS and serum thyroglobulin (TG) levels are obtained at 3- to 5-year intervals. Patients with no poor prognostic indicators are subjectedWB toS and at to 5-year intervals thereafter. serum TG measurement annually for3 years, and then 3From: Thyroid Cancer:A Comprehensive Guide toclinical Management Edited by: L. Wartofiky 0 Humana Press lnc., Tofowa,N1
229
230
Burch
SERUM THYROGLOBULIN MEASUREMENT
Molecular Biology, Physical Properties, and Immunogenicityof TG
The human TG gene is located on chromosome8 and comprises more than 300 kb with 42 exons, making it one of the largest known human genes. TSH regulates TG gene transcription through the adenylate cyclase pathway. The TG molecule is a 660kDa glycosylated protein consisting of two identical 330-kDa subunits. Approxima 10% of T G s weight can be attributed to three different carbohydrate moieties, which are covalently bound to asparagine, serine, and threonine residues (2). TG is the key substratefor the biosynthesis and storage of thyroid hormone. Amo a total of 134 tyrosyl residues contained within TG, less than one in five is iodinated six to eight become iodothyronines (3,4), determined per molecule. Of these, only largely by the secondary and tertiary protein structure of (5). TG Once synthesized, TG is stored in the colloid space until retrieved through pinocytosis. Proteolysis su results in a rapid release of thyroxine and triiodothyronine into the circulation, and degradation of the majority of the remaining TG molecule. This entire process is stimulated by TSH. TG, which is released to the serum, consists of a heterogeneous population-largely a result of mRNA splicing (6). TG is cleared from the circulation primarily by the liver, with a half-life ranging from 3 to 30 hours depending on the degree of posttranslational modification and the population under study(7,8).Table 1 lists the various causes of an elevated serum TG. TG autoantibodies (TGAb) have a profound influence on the clinician’s ability to accuratelymeasureandfollowserumTGlevels.Unfortunately,TGAbarewidely prevalent, being demonstrable in 4% to 27% of the general population, 51% to 97% of patients with Graves’ disease or Hashimoto’s thyroiditis, and 15% to 30% of patients with thyroid cancer (9).Although 40 different epitopes have been identified within th TG molecule, only a few of these are recognized by the majority of naturally occurrin TG autoantibodies (10). The presence and titerof TGAb is independent of serum TG concentration (IO).The immunoreactivityof TGparallels its iodine content( I ] ) ,and TG autoantibodies are capableof distinguishing between structural differences introduced through variable iodination (12). Variable glycosylation, on the other hand, does not (12). TGAb recognize distinct conformational seem to affect TG antibody binding epitopes in patterns that may be characteristic for particular disorders (13-16). Conversely, TGAb found in the general population do not recognize specific epitop In patients with a family history of autoimmune thyroid disease, the tendency TGAb appears to be inherited (17). TG Assay Methods, Relative Cost
Most commercial laboratories use immunometric assay technology to measure s TG levels, while a few still use a radioimmunoassay (RIA)methodology (9,18). The lower limit of detection for serumTG ranges from 0.5 to 1.5ng/ml for immunometric assays, to 1.0 to 3.0 ng/ml for RIA (18). The cost for serum TG assay in commercial laboratories currently ranges from $50.00 to $120.00 and the turnaround time varies from 1-4days (18).
Papillary Thyroid Cancer Follow-Up
231
Table 1 Causes of Increased Serum Thyroglobulin Concentration Thyrotropin mediated Acute and transient TSH administration Protirelin (TRH) administration Neonatal period (1-96 hrs postpartum) Chronic stimulation Iodine deficiency Endemic goiter Goitrogens Reduced thyroid reserve Thyroxine-binding globulin deficiency Resistance to thyroid hormone TSH-producing pituitary adenoma Iodine block of the thyroid Non-thyrotropin-mediated Substances with TSH-like bioactivity Thyroid stimulating immunoglobulins (Graves’ disease) Chorionic gonadotropins; trophoblastic disease Pregnancy Direct trauma to the thyroid gland Percutaneous needle aspiration Thyroid surgery Iodine 131 therapy Abnormal release Autonomous toxic nodules Solitary nontoxic nodules Multinodular goiter Subacute thyroiditis (early phase) Differentiated none medullary thyroid carcinoma. Neonatal period Abnormal clearance Renal failure
Adapted from Torrens JI, Burch HB. Clinical application of serum thyroglobulin testing. Endocrinologist 1996; 6:125-144.
TG Assay Problems Table 2 summarizes the problems shared to varying degrees by most current TG assays. These may be divided into assay-dependent and assay-independent problems (18). ThelackofaninternationalTGstandardandthedifferingsensitivitiesand specificities of the TGAb used in different methodologies each serve to preclude the comparison of serial TG values obtained by different laboratories. The best way to circumvent this problem is to either use the same laboratory for all measurements or to store frozen serum for future simultaneous assay (19). Likewise, assay imprecision, leading to increased interassay variation despite using the same methodology, limits the ability to detect small serial increments in TG. Storage of frozen serum to be run
232
Burch Table 2 Factors Known to Affect the Reliability and Reproducibility of Serum Thyroglobulin Measurement Assay dependent Poor interassay precision Lack of an international Tg standard Different specificitiesof TgAb used Suboptimal functional sensitivity Assay independent Heterophilic antibodies Monoclonal gammopathies End-stage renal disease TgAb Extremely high levels of Tg (the “hook” effect) Immunologically inactive Tg
simultaneously with the patient’s current sample also circumvents this problem (9). The functional sensitivity of an assay depends on the methodology chosen, with chemiluminometric assays (ICMA) being the most sensitive and RIA being the least sensitive among available techniques. TGAb are by far the greatest impediment to the measurement and interpretation o serum thyroglobulin levels. These antibodies may spuriously increase or decrease the measured serum TG level, depending largely on the methodology chosen. For ex ICMA methods double antibodyRIA methods tend to overestimate serum TG, whereas characteristically underestimate TG levels (18). The degree of antibody interference also depends on such factors as the affinity and specificity of the antibody employed of serum used, and characteristics of the interfering autoantibod in the assay, the amount (18,20-22). No method is totally free of interference from TGAb (19,20). Conseque TG antibody detection employing a sensitive technique is critical. Unfortunately, many reference laboratories continue to use hemoagglutination assays for screening (1619). RIA measurements which use a high-affinity first antibody and a species-specifi antibody appear to be the most resistant to TG antibody interference(9). The presence of TGAb does not completely negate the of value serum TG measure ment. For example, if it is known that a particular assay underestimates TG levels in the presence of antibodies, and serial testing using the same methodology reveals an upward trend in a particular patient, disease recurrence or progression shouldbe suspected (18). It should be noted, however, that alternative explanations exist for such a pattern, including a change in TG antibody affinity or capacity and changes in the metabolic clearance of TG-TGAb complexes. is Another assay-independent problem that may occur with serum TG measureme a falsely low TG reading due to the presence of extremely high (10 to 10,000 times (19). The the upper detectable range) serum TG levels, the so-called “hook effect” hook effect occurs when the amount of TG presentin the sample exceeds the capacit of the capture antibody on the solid support (19). Double-antibody immunoassays, such
Papillary Thyroid Cancer Follow-Up
233
Table 3 Causes of an Elevated Thyroglobulin Level anda Negative Scan Diffuse small metastases, too small to be detected despite radioiodine uptake Intact TG synthesis with decreased or impaired iodine trapping Iodine contamination Serum TSH adequateto induce TG synthesis but not sufficient to stimulate radioiodine uptake TGAb Thyroid tissue remnant increasing the TG synthesis and decreasing the radioiodine uptake in the metastatic deposits False positive elevationof TG
Adapted from Torrens JI, Burch HB. Clinical application of serum thyroglobulin testing. Endocrinologist 1996; 6:125-144.
as the immunoradiometric assay(IRMA) and ICMA assays, are particularly susceptible to this effect (23). Dilution (1-10) of the serum in TG antibody-positive patientsmay be used to exclude the hook effect(19,23).
TG-Based Decision Analysis in Patients With Thyroid Cancer Several considerations are noteworthyin the applicationof serum TG measurement to the follow-up of patients with thyroid cancer. 1. Serum TGis more sensitive than the whole-body scan in detecting recurrent or metastatic of TG levels in the disease (2425). Other possible explanations for a discordant elevation face of a negative WBS are summarized in Table3. 2. The histological grade of tumor differentiation is not an indication of TG synthesizing
capacity. Hence poorly or moderately differentiated thyroid cancer may produce high levels of serum TG, while highly differentiated carcinomasmay have unexpectedly low serum levels (26,27). 3. Since TG synthesis and radioiodine uptake reflect different functions, the serum TG level does not predict radioiodine uptake nor does uptake predict TG level(18). 4. A low( 4 ng/ml) or undetectable TG level on thyroid suppressive therapy does not exc recurrent or metastatic disease(18).
Serum TG is a reliable marker of persistent, recurrent, or metastatic disease only afterneartotal or totalthyroidectomyfollowedbyradioiodineablationtherapy (25,27,28-30). The measurement of serum TG during the hypothyroid state increases the sensitivity of the test for the detection of recurrent thyroid cancer, whereas me ment while on suppressive therapy increases the specificity of an elevated TG level (18). Therefore, measurement of TG both before and after thyroid hormone withdrawal provides the most clinically useful information (31). Theoretically, all patients who undergo near-total or total thyroidectomy followed by radioiodine ablation would be expected to have an undetectable TG level (25), but this is not always the case. One study found that40% of patients without evidence of disease during prolonged followor total thyroidectomy followed by up had detectable TG levels despite near total radioiodine ablation(29). Most patientsin clinical remission will have a gradual decreas in TG level suggesting possible atrophy and death of residual TG-producing(29). tissue
V
L
234
PapillayFollow-Up Thyroid Cancer
235
4 Fig. 1. Postablative managementof thyroid cancer patients. Patients are classified into either low or high risk for recurrence categories. Low-risk patients are further categorized based on the level of clinical suspicion and TG value off suppressive therapy.(From Torrens JI, Burch HB. Serum thyroglobulin measurement:utility in clinical practice. Endocrinologist 1996; 6:125144,with permission.)
Theclinician’sresponsetoaparticularTGlevelshouldbeindividualized.For example, a patient with one or more poor prognostic indicators (discussed below) and 5 ng/ml on thyroid hormone suppression (29,32-36), or greater a TG level greater than than 10 ng/ml off thyroid hormone suppression (29,37-42), should be suspected as having recurrent or persistent disease. These low thresholds for action increase the sensitivity of the assay since the pretest probability of active disease in these patients is high. Conversely, patients with no poor prognostic factors and a benign biological behavior of the tumor under observation, should be evaluated if the TG level is higher (35,4345) or higher than 30 ng/ml off suppressive than 10ng/ml on suppressive therapy therapy (3544). An approach to serum TG testing in the follow-up of patients with thyroid cancer is shown in Figure 1.
TG Testing in Patients With Thyroid Remnants (See Chapter 2 4 ) Although reliance on serum TG levels in patients with thyroid remnants is fraught with hazard, based on the above observations, it seems reasonable to investigate and treat patients with a TG level greater than 30 n g / d or a serial rise in TG while on suppressive therapy (18). The author’s bias is to proceed with surgical or radioiodine ablation of the remnant lobe in this circumstance, followed in 2-3 months by wholebody scanning and repeat TG measurement, at which time a decision may be reached regarding the need for further1311therapy. A TG value of 10 ng/ml or less in patients with thyroid remnants on suppressive therapy significantly reduces but does not elimi the possibility of recurrent disease (44).
TG Testing After Injection of Recombinant Human TSH (See Chapter 14) The diagnostic use of recombinant human TSH (rhTSH) is likely to revolutionize are followed up.In an analysis comparing the manner in which thyroid cancer patients the diagnostic utilityof radioiodine scanning (and serum TG measurement) using either exogenous rhTSH or T3 withdrawal, rhTSH gave smaller increments in TG levels than did conventional T3 withdrawal (46). Specifically, after rhTSH administration, serum TG increased twofold or more in only 59% of 19 patients, compared to 79% of the (46).Further optimizationof both the rhTSH regimen same patients after T3 withdrawal and the timing of serum TG measurement after rhTSH is needed to fully assess the utility of TG measurement after rhTSH administration. REFERENCES 1. Solomon BL, Wartofsky L, Burman KD. Current trends in the management of well differentiated papillary thyroid cancer. J Clin Endocrinol Metab 1996; 81:333-339. 2. Van Herle A J , Vassart G, Dumont J. Control of thyroglobulin synthesis and secretion. N Engl J Med Part one: 1979; 301:239-249. Part two: 1979; 301:307-314.
236
Burch
3. Izumi M, LarsenPR. Triiodothyronine, thyroxine, and iodine in purified thyroglobulin from patients with Graves’ disease.J Clin Invest 1977; 59:1105-1112. in thyroglobulin iodin4. Ogawara H, Bilstad JM, Cahnmann HJ.Iodoamino acid distribution ated in vivo and in vitro. Biochim Biophys Acta 1972; 257:339-349. 5. Edelhoch €I.The propertiesof thyroglobulin. VIII. The iodinationof thyroglobulin. J Biol Chem 1962; 237:2778-2787. 6. Bertaux F, Noel M, MalthieryY, Fragu P. Demonstration of a heterogeneous transcription pattern of thyroglobulin mRNA in human thyroid tissue. Biochem Biophys Res Commun 1991; 178:586-592. m i M, Kubo I,T a m M, Yamashita S, et al. Kinetic study of immunoreative human 7. h thyroglobulin. J Clin Endocrinol Metab 1986; 62:410-412. 8. Van HerleAJ. Thyroglobulin.In:Werner and Zngbar’s The Thyroid,6th ed. 1991:493-505. 9.SpencerCA,WangCC.Thyroglobulinmeasurement:techniques,clinicalbenefits,and pitfalls. Endocrinol Metab Clin NorthAm 1995; 24:841-864. 10. Volpe R. Immunologyof human thyroid disease.In:Autoimmune diseasesofthe endocrine system. Boca Raton, FL: CRC Press, 1990:76-239. 11. Saboori AM, Rose M , Kuppers RC, et al. Immunoreactivity of multiple molecularforms of human thyroglobulin. Clin Immunol 1994; 72:121-128. 12. Kiso Y, Furmaniak J, Morteo C, Smith BR. Analysis of carbohydrate residues on human of deglycosylation, reduction thyroid peroxidase(TPO) and thyroglobulin (Tg) and effects and unfolding on autoantibody binding. Autoimmunity 1992; 12:259-269. 13. Saboori A M , Caturegli P, Rose N R , et al. Tryptic peptides of human thyroglobulin: II. Immunoreactivity from patients with thyroid diseases. Clin Exp Immunol1994; 98:459463. 14. Caturegli P, Mariotti S , Kuppers RC, et al. Epitopes on thyroglobulin: a study of patients with thyroid disease. Autoimmunity 1994;18:41-49. M, Salhi SL, et al. Antigenic domains on the human thyroglobulin 15. Piechaczyk M, Bouanani moleculerecognized by autoantibodies in patients’seraand by naturalautoantibodies of healthy subjects.Clin Immunol Immunopathol 1987; 45:114-121. isolated from the sera 16. Prentice L, Kiso Y, Fukuma N, et al. Monoclonal thyroglobulin autoantibodies: variable region analysis and epitope recognition.J Clin Endocrinol Metab 1995;80:977-986. 17. PhillipsD, McLachlan S , Stephenson A,et al. Autosomal dominant transmission of autoan J Clin Endocrinol Metab 1990; 70:742-746 bodies to thyroglobulin and thyroid peroxidase. 18. Torrens JI, Burch HB. Clinical applicationof serum thyroglobulin testing. Endocrinologis 1996; 6:125-144. 19. Spencer CA, Takeuchi M, Kazarosyan M. Current status and performance goals for serum thyroglobulin assays. Clin Chem 1996; 42164-173. 20. Feldt-Rasmussen U, Rasmussen K. Serum thyroglobulin (TG) in presence of thyroglobul autoantibodies (TgAb): clinical and methodological relevanceof the interaction between Tg and TgAb in vitro and in vivo. J Endocrinol Invest 1985; 8:571-576. 21. Schneider AB, Pervos R. Radioimmunoassayof human thyroglobulin: effects of antithyro globulin autoantibodies.J Clin Endocrinol Metab 1978;47:126-137. 22. Ross DS. Long-term managementof differentiated thyroid cancer. EndocrinolMetab Clin North Am 1990; 19~719-739. 23. Cole TG, JohnsonD, Eveland BJ, Nahm MH. Cost effective method for detection of “hook effect” in tumor marker immunometric assays. Clin Chem 1993; 39:695-696. 24. Aiello DP,Manni A. Thyroglobulin measurementvs iodine 131 total-body scan for follow up of well-differentiated thyroid cancer. Arch InternMed 1990; 150:437-439. 25. Lindegaard MW, Paus E. Thyroglobulin in patients with differentiated thyroid carcinoma. Scand J Clin Lab Invest Suppl 1991; 206:79-84. 26. Dralle H, SchwarzrockR, LangW, et al. Comparisonof histology and immunohistochemis try with thyroglobulin serum levels and radioiodine uptake in recurrences and metastase of differentiated thyroid carcinomas. Acta Endocrinol (Copenh) 1985; 108504-510.
Pupillay Thyroid Cancer Follow-Up
237
27. Schlumberger M, FraguP, Parmentier C, TubianaM. Thyroglobulin assay in the follow-
up of patients with differentiated thyroid carcinomas: comparison of its value in patients with or without normal residual tissue. Acta Endocrinol (Copenh) 1981; 98:215-221. of serum thyroglobulin measurements 28. Ashcraft MW, Van HerleAJ. The comparative value and iodine131total body scans in the follow-up study of patients with treated differentiated thyroid cancer. Am J Med 1981; 71906-814. 29. Ozata M, Suzuki S , Miyamoto T, et al. Serum thyroglobulin in the follow-up of patients with treated differentiated thyroid cancer. J Clin Endocrinol Metab1994; 79:98-105. PS, Rodger AB. Measurement of serum thyroglobulin 30. Harvey R D , Matheson NA, Grabowski is of value in detecting tumor recurrence following treatment of differentiated thyroid carcinoma by lobectomy. Br J Surg 1990; 77:324-326. 31. Girelli ME, Busnardo B, Amerio R, et al. Critical evaluationof serum thyroglobulin (Tg) levels during thyroid hormone withdrawal and total body scan: results 291inpatients with thyroid cancer. Eur J Nucl Med 1986; 11:333-335. 32. Black EG, SheppardMC, Hoffenberg R. Serial serum thyroglobulin measurementsin the management of differentiated thyroid carcinoma. Clin Endocrinol 1987; 27:115-120. 33. Black EG, SheppardMC. Serum thyroglobulin measurements in thyroid cancer: evaluation of “false” positive results. Clin Endocrinol1991; 35:519-520. 34. Baskin HJ. Effect of postoperative I3lI treatment on thyroglobulin measurements in the follow-up of patients with thyroid cancer. Thyroid 1994; 4:239-242. 35. Mazzaferri EL. Treating high thyroglobulin with radioiodine: a magic bullet or a shot in the dark? [Editorial]. J Clin Endocrinol Metab 1995; 80:1485-1487. T M D , Maisey MN, et al. Serum thyroglobulin in thyroid cancer. Lancet 36. Black EG, Gimlette 1981;1:443-445. of serum thyroglobulin measurements 37. Ashcraft MW, Van Herle AJ. The comparative value in the follow-up study of patients with treated differentiated and iodine131total body scans thyroid cancer. Am J Med 1981; 71906-814. 38. Robbins J, Merino MJ, Boice JD, et al. Thyroid cancer: a lethal endocrine neoplasm. Ann Intern Med 1991;115:133-147. J M , Robbins J. Sequential serum thyroglobulin determina39. Schneider A B , Line B, Goldman
tions, I3lI scans, and l3]Iuptakes after triiodothyronine withdrawal in patients with thyroid cancer. J Clin Endocrinol Metab 1981; 53:1199-1206. 40. Httffner M, Stumpf I”, Grussendorf M, et al. A comparison of the effectiveness of I3lI whole body scans and plasma Tg determinations in the diagnosisof metastatic differentiated 1984; carcinomaofthethyroid: aretrospectivestudy.ActaEndocrinol(Copenh) 104~327-332. 41. Barsano CP,Skosey C, DeGroot LJ, Refetoff S. Serum thyroglobulin in the management of patients with thyroid cancer. Arch InternMed 1982; 142:763-767. et al. Correlation of thyroglobulin measurements 42. Ramanna L, WaxmanAD, Brachman ME%,
and radioiodine scansin the follow-upof patients with differentiated thyroid cancer. Cancer
1985; 55~1525-1529. 43. Ericsson U B , Tegler L, Lennquist S , et al. Serum thyroglobulin in differentiated thyroid carcinoma. Acta Chir Scand 1984; 150:367-375.
44. Schlumberger M, Parmentier C, de Verthaire F, TubianaM. l3lI and external radiation in the treatment of local and metastatic thyroid cancer. In Thyroid Disease: Endocrinology, Surgery, Nuclear Medicine, and Radiotherapy. Raven Press, 1990, pp. 537-552. 45. Lubin E, Mechlis-Frish S , Zatz S , et al. Serum thyroglobulin and iodine-131 whole-body scan in the diagnosis and assessment of treatment for metastatic differentiated thyroid carcinoma. J Nucl Med 1994; 35:257-262. Daniels GH, Ross DS, et al. Diagnostic 46. Meier CA, Braverman LE, EbnerSA, Veronikis I, in patients with thyroid carcinoma (phase I/IIstudy). use of recombinant human thyrotropin J Clin Endocrinol Metab 1994; 78:188-196.
23 Radioiodine Treatment of Thyroid Cancer-I1 Maximizing Therapeutic and Diagnostic 2311Uptake Diane Sweeney and Gerald Johnston Prior planning for the administration of radioiodine therapy is important. Lowering iodine levels, sufficiently stimulating the thyroid, checking baseline laboratory levels, ordering diagnostic procedures and gaining informed consent from patients must be done before therapy. This requires several months of concerted effort between the surgeon, endocrinologist, primary care practitioner, nuclear medicine physician, nutritionist, and the patient. THYROTROPIN STIMULATION
Radioiodine therapyof residual, functional thyroid tissue or thyroid cancer metastases is best accomplished following full and adequate thyrotropin (TSH) stimulation. Follow ing total or near-total thyroidectomy, TSH values will reach maximal inlevels approximately 4-5 weeks following surgery (I).Circulating T4 secretedby the gland prior to surgery delays TSH response. These patients can be maintained on L-triiodothyronine (T3) replacement therapy (25 pg BID or TID) for approximately 4 weeks. This dose shouldthenbewithheld and a TSH determination performed in approximately14 days (2). TSH concentrations of greater than 30 pIU/rnl are considered adequate for stimulation of radioiodine uptake in metastatic lesions (3). Athyreotic patients who to reach have previously been treated with thyroidectomy and thyroid ablation appear profound hypothyroidism more quickly than patients recently undergoing "total" thyroidectomy. Following withdrawal of T3 replacement, TSH levels can be drawn in approximately 10 days (I). The symptoms of prolonged hypothyroidism are minimized by maintenance on T3 replacement therapy. However, prolonged TSH stimulation to the cancer cells is a significant concern in initiating this course of therapy. Historically, bovine TSH administration was not found to stimulate radioiodine uptake in residual thyroid tissue as well as endogenous TSH stimulation (4). Moreover, significant allergic reactions to bovine TSH often occurred, with antibody mediated inhibition of the effects of endogenous TSH (5).
From: Thyroid Cancer:A Comprehensive Guide toClinical Management Edited by: L. Wartofiky 0 Humana Press Inc., Tofma, NJ
239
More recently, recombinant human TS has been proven to increase T4 and iodine a1 studies (6).Recombinant human TSH has been studied in an encouragase II trial involving 19 patients (7): 63% of patients had concurrent nostic scans done following recombinant human TSH stimulation and following T3 withdrawal; 16% of patients had additional foci of uptake following exogenous stimulation,and 16%had additionalfoci visu d after endogenousstimulation following T3 withdrawal. Recombinant human caused nausea in 16% of patients but in many patients is certain to replace hypoth~oidisminduced by thyroxine withdrawal. The most recent more large scale clinical trials of recombinant human TSH are reviewed in Chapter 14. However, it is currently only available for routine diagnostic but not therapeutic use.
IC The iodine-concentrat~gability of thyroid cancer cells is poor compared to normal thyroid tissue. Therefore, i o d ~ e - s e e ~ nability g should be optimized in these cells before therapy. Iodine depletion before dosing may maximize uptake. colleagues (8) designed a diet and diuretic regimen to induce iodine depletion before radioiodine therapy. They found that the ~ o u noft radioiodine taken up and retained reased by 146% in three studies. es, the daily iodine intake has increased from 160-1250 pg/day to y due to iodine additivesin prepared foods (9). at strict adherence to a complex low-iodine 1311therapy or scannin~will decrease iodine excretion from a mean value of 346.9 to 42.8 pg iodine/g creatinine per day. They also calculated that the radiation dose to residual thyroid tissue (but not n e c e s s ~ l ymetastases) increased more than twofold. A l o ~ - i o ~ diet n e appears to be an important adjunct to optimizing diagnosis and I3lI therapy. Yet, in modem society, it can be quite burdensome on patients. They must forego takeout food, restaurant food, and many prepared foods for 1-2 weeks before therapy. The best way to ensure compliance and success is to orchestrate a team approach with a quali~ednut~tionistor dietitian, the physician? and the patient. The patient must have ready access to sample menus and recipes and i n s ~ c t i o n son how to prepare these foods. ost patients are motivated and will strictly adhere to guidelines on diet if they are understood. A s u m a r y of typical diet guidelines and rest~ctions to ensure low-iodine intake appears in an Addendum at the end of this chapter. Of note, all iodinated contrast agents must be avoided before radioiodine administration. These agents, both ionic and nonionic, contain €ree iodides which can inhibit radioiodine uptake (11). Many patients will not recall their ad~nistrationand many physicians may not realize that these must be avoided. Rigorous history taking and investigation should e ~ ~ n athe t e possibility that radiographic contrast agents have been used in these patients for 4-6 weeks before radioiodine dosing. Spot urine iodine testing may be helpful in selected, puzzling cases.
a metastatic survey scan should be performed for several reasons. efore I3lIt~erapy? Although a surgical total thyroidectomy often results in a s i g ~ ~ c aresidual nt of thyroid
Maximizing Therapeutic and Diagnostic ‘jlI Uptake
241
tissue in the neck, this should be proved before the administration of high doses of radioactivity. Whole-body scans also allow a search for metastatic disease, possibly affecting the therapeutic dose subsequently given to the patient. The timing and size of the diagnostic dose given before a therapeuticis presently dose controversial. The dose should be large enough to image all local and distant metastati deposits while small enough to lessen the sublethal effects of “stunning.” “Stunning” was first observed in1951by Rawson and colleagues(12) who reported that a noncancericidal doseof I3’Icould impair the ability of thyroid tumors to concentrate subsequent therapeutic doses. This finding has been observedin more recent reports(13,14). Park and coworkers (13) found that prior diagnostic dosing with 3-10 mCi of 13’1resulted in decreased uptake in 20 of 26 patients receiving subsequent larger ablative doses. The reduction was dose-related. Jeevranram and associates (14) found that a diagnostic 2-3 mCi I3II diagnosticdoses) dose of 1750-3500 cGytothethyroid(estimated decreased thyroid uptake to 37.48% (k21.24%) in 15 patients receiving subsequent therapeutic dose (diagnostic dose uptake taken to be 100%). There is also evidence thatdecreasingthetimebetweendiagnosticandtherapeuticdosingmaydiminish the effects of “stunning” (13,14). Therapeutic dosing should be performed promptly following diagnostic scanning. These are striking results, which indicate that the thyroid cell uptake and function are depressed following the “stunning” diagnostic radiation dose and argues convincingly for limits to prior diagnostic scanning. However, the diagnostic dose must be large enough to image local and metastatic thyroid disease. Studies have shown that given the minimal radioiodine uptake of 0.05-0.5% per gram) higherdosesof 1311 allow thyroidcancers(ontheorderof visualization of potentially treatable lesions (15,16). 10 mCi of 1311allows more lesion visualization than 2 mCi and 30 mCi is more accurate than 10 mCi. In fact, Waxman and coworkers (16) consider a dose of less than 2 mCi l3II inadequate for evaluating ablation in patients with thyroid cancer. There are two clinical scenarios that require consideration in the decision-making regarding diagnostic dosing. The first is diagnostic dosing before therapeutic dosing to cautious administration (Table 1).The evidence for possible “stunning” should lead of a diagnostic dose. Several authors have advocated the use of 1231to scan the thyroid remnant in these cases, allowing that more accurate high dose1311 scanning will occur following the therapeutic dosing (13,17). Advantages include decreased radiation exposure, decreased cost and most importantly, avoidance of “stunning” prior to ablation therapy. Disadvantages include a decrease in imaging of small metastatic deposits which would be demonstrated with dosesI3lI. ofHowever, this has not been documented (13). 1231or smaller dosesof 1311(2-5 mCi) should be to impact on patient management 13’1dosing is advocated if patient management used for pre-therapy diagnostic scanning. decisions hinge upon the diagnostic scan or if metastatic deposits are expected. Ho inmostcases,clinical,pathological,andradiologicalresultswillhavedetermined early treatment. The second clinical situation requiring a diagnostic dose of l3’I is the follow-up metastatic survey scan (Table2). This study is most commonly performed 1year after 5 years. The diagnostic therapeutic dosing and annually or biannually for approximately a paucity dose inthis setting must be large enough to image metastatic disease. isThere of data regarding conventional dosingof these patients. Published data on dose levels
242
Sweeney and Johnston
Table 1 Timetable for Patients Receiving 1311 Therapy After Near-Total or Total Thyroidectomy Week 0 Week 4 Week 4-5 Week 6
surgery Start Cytomel(25 pg BID or TID) Stop Cytomel Begin low-iodine diet Draw TSH, P-HCG,” WBC, thyroglobulin
If TSH > 30 plU/ml, then Diagnostic dose Whole-body metastatic survey Therapeutic dose Start thyroid replacement Posttherapy scan
2-5mCi I3lI or 300-500 pCi lUI 48-72 hlater or 24 h later As soon as possible 1 day after therapeutic dose 7-10 days after therapeutic dose
“In women of child-bearing age.
Table 2 Timetable for Patients After Surgery or Radioiodine Therapy for Metastatic Survey Diagnostic Scan Week 0 Week 4 Week 5-6
Stop thyroid replacement Start triiodothyronine (Cytomel9 replacement (25 pg BID or TD) Stop Cytomel Begin low-iodine diet Draw TSH, P-HCGP thyroglobulin
If TSH > 30 pIU/ml, then .5-10 mCi I3lI Diagnostic dose 48-72 h later Whole-body metastatic survey For abnormal scans or unexplained activity Delayed radioiodine scan images Thallium or sestamibi scans MEU or CT scanning Possible therapeutic dosing “In women of child-bearing age.
are scarce, although 5 mCi 13’1 appears to be used at several institutions (18,lP). We advocate a dose of 5-10 mCi in the follow-up diagnostic scans of these patients. A significant amount of data indicate that higher doses willmore allow accurate visualization of metastases and recurrence, which is the goal of the diagnostic scan (15,16). “Stunning” should not be a concern on diagnostic studies, following ablation or There is no consensus on follow-up of these patients with thyroid cancer.A recent article that surveyed members of the American Thyroid Association, regarding the follow-up of patients with papillary cancer, found that only 59% of clinicians would
Maximizing Diagnostic Therapeutic and
'jlI Uptake
243
performatleastonefollow-upradioiodinescan,andonly 85% woulduseserum thyroglobulin levels for monitoring disease recurrence (20). We believe that l3II scans should be performed annually until they are negative and then routinely every 1-2 years for 5 years along with routine serum thyroglobulin monitoring. The intensityof surveillance should be somewhat dependent on the initial 13'1 diagnostic scan within pathology and prognostic factors. We hesitate to perform an 6 months of the initial therapeutic dose because of the risks of not allowing enough time for the initial therapeutic effect and the risk of frequent high-dose I3lI therapy. 20 Maxon and Smith (21) in a recent review, advocate every 5-year follow-up for years after the last known evidence of disease.
TECHNIQUES OF SCANNING 1311 diagnostic imaging should be performed 48-72 hours after oral dosing (22). However, iodine-depletion diets cause a reduction in iodine clearance, which may (8). Delayed imaging(72 hours or longer) may be necessary increase background activity in these patients to improve target-background activity ratios. Renal failure patients on dialysis have been found to have a shorter effective I3'I half-life, most likely due to dialysis clearance of iodine (23). These patients should be dialyzed before dosing, with the next dialysis postponed until after imaging, if possible. Imaging should be performed using a high-energy collimator and a modern highquality gamma camera. Anatomic landmarks must be acquired using markers, particularly at the suprasternal notch and xiphoid, and the bladder should be emptied before imaging. Adequate count statistics are very important, requiring long imaging times to acquire a minimum of 50,000 counts per image of the neck, chest, abdomen, and pelvis (22). Physiological accumulationof radioactive iodine occurs in the salivary glands, nasal mucosa, bladder, gastric mucosa, small bowel, and colon (24,251. In the presence of functional thyroid tissue, vague or readily apparent diffise liver activity is usually due of 13'I-thyroxine (26). Positive results are defined as uptake to physiological localization of iodine within residual thyroid tissue in the thyroid bed or concentration in metastatic deposits. False positives occur in a variety of entities, including other malignancies, (25) (Table 3). contaminationbysecretionsandinmultiplemiscellaneousentities Delayed imaging, as well as utilization of other diagnostic procedures such as CT scanning or MRI may be helpful in differentiating true positive and false positive results. There are several other techniques which have been promoted to increase the uptak of 13'1in thyroidtissueormetastaticdiseaseinordertoenhancedetectabilityon diagnostic scanning or to promote ablation or treatment. Hydrochlorothiazide has been used to further enhance iodine uptake by stimulating iodine excretion(8).Lithium may slow the rate of iodine release and thereby prolong I3'Iretention in tumor (27). However, its action has not been proven and lithium toxicityis a concern. Radiosensitizers, such as low-dose Adriamycin (doxorubicin) used with1311therapy, are being tested in highriskpatientstopromotethetumoricidialeffect of theradioiodine (28). Euthyroid patients undergoing recombinant human TSH stimulated diagnostic studies will have significantly more rapid renal clearance of l3II with reduced body retention than will hypothyroid patients after thyroxine withdrawal.
Johnston 244
and
Sweeney
Table 3 Common Findings on Diagnostic “I Metastatic Survey Scans ~~
Findings
Physiological Nasal mucosa, salivary glands @OPharynX secretions nasal and saliva to Gastric and intestinalactivity Bladder and urinary tract Main route Common areas of uptake Usually normal hepatic concentration (diffuse) Liver toDue abnormal appear Stomach May
~
~
~
~~~
~~~~
Mechanism a d o r features
High iodine concentrating ability Due Due to concentration by gastric mucosa and excretion; can be increased with constipation of iodine excretion
of T4produced by endogenous thyroid tissue; usually not obvious on early (24-72 h) scans to hiatal in position due hernia
Usually abnormal ly foci macroscopic and Microscopic Lungs representing metastases nottaken beshould Care Bone
thyroid outside Uptake Neck
to thoracic confuse and lumbar spine uptake with physiological gastric or bowel activity; also clavicular or sternal uptake can appearas thyroid bed activity differentibe bed must ated from normal salivary gland activity
Uncommon areas of uptake usually not associated with thyroid cancer metastases May be seen in young patients usually seen Normal thymus only after therapeutic (large) doses Due to retained saliva Ectopic thyroid tissue, ectopic gastric mucosa, esophageal abnormalities Due to poolingof urine as seen with Urinary tract abnormalities obstruction or dilatation May be due to increased vascular permeability Inflammation that Particularly tumors tissue Other neoplasms from arising concentrates iodine EffusionsandserousfluidcollectionsParticularlypleuralandpericardialeffusions Adapted from Referema 24 and 25.
POSTABLATMYPOSTTHERAPYSCAN Following therapeutic administrationof I3*I,a posttherapy scan should be perform 7-10 days later. Maxon and colleagues (29) studied 92 posttherapy scans and identifie in 10%of patients. additional fociof disease (not seen on lower dose diagnostic scans) This is not surprising since it has been reported that the sensitivity of whole-body 13’1 scanning increased with increasing dosesof the radiopharmaceutical (15,16,30). Sherman and colleagues (304 have argued that posttreatment scans “rarely yielded new
MaximizingDiagnostic Therapeutic and
l3lI
Uptake
245
information that would potentially alter the patient's prognosis"; however, 10% of posttreatment scans in their study revealed new locations of metastatic disease. This is new information gained with no additional radiation burden to the patient. Therefore, the opportunity to image a patient following therapeutic dosing should not be missed.
THERAPEUTIC ADMINISTRATION OF 1311 IN PATIENTS WITH A NEGATIVE DIAGNOSTIC 1311 SCAN AND A DETECTABLE SERUM THYROGLOBULIN LEVEL (SEE CHAPTER 24) There is a management dilemma that has evolved out of the increased utilization and high sensitivity of serum thyroglobulin (Tg) measurements. AfterT4withdrawal, 10 ng/ml are often indicative of metastatic or persistent thyroglobulin levels higher than disease (18). However, the diagnostic I3lI scan is often negative in these patients. In fact, Pineda and coworkers (31) found 17 patients with thyroglobulin levels of 8 ng/ ml or higher after thyroid hormone withdrawal, with negative 5 mCi I3'I diagnostic 150-300 mCi ofI3lIas therapeutic doses. Posttherscans (31). These patients were given 16 of these patients(94%).Within several years, apy scans showed abnormal uptake in thyroglobulin levels decreased after these 13'1 therapeutic doses. There is reasonable and compelling evidence that these patients with elevated thyroglobulin levels will often have a positive posttherapy scan after high-dose I3lI therapy despite a negative low-dose diagnostic scan (31-33). Whether this empirical therapeutic dosing results in decreased morbidity and mortality has not been addressed with long-term followup or in a prospective study. However, there is some evidence that there is a clinical response after therapeutic dosing (31). An important caveat to this discussion is that patients who have not undergone thyroidectomy (subtotal, near-total,or total), followed by radioiodine ablation or therapy cannot be accurately assessed with periodic thyroglobulin levels, andor diagnostic radioiodine scans. The iodine uptake and thyroglobulin production in the remaining remnant makes follow-up with these diagnostic modalities insensitive. Follow-up with both radioiodine scans and serum thyroglobin levels is advantageous (34) examined the use of thyroglobulin level (during after surgery. Ronga and associates L-thyroxine therapy) and diagnostic scans. Taken together, sensitivity reached 95.7%, specificity loo%,and accuracy 96.7%in the discrimination between patients with and without metastases. Neither alone was adequate. Although serum thyroglobulin levels (off L-thyroxine) are extremely sensitive, they may lack specificity without a concurrent radioiodine scan. DIAGNOSTIC ALTERNATIVES TO RADIOIODINE IN THE FOLLOW-UP OF THYROID CANCER PATIENTS
Magnetic Resonance Imaging and Computed Tomography Other methods of patient follow-up in thyroid cancer have been studied with great interest in the hope that cessation of thyroid hormone replacement therapyin order to detect tumor would become unnecessary. However, no other diagnostic modality ap (MRI) to yet stand alone. Computed tomography (CT) and magnetic resonance imaging appear to have a role in follow-upof local neck recurrence (33,35). MFU is preferred
246
Sweeney and Johnston
due to better differentiation between scar tissue and recurrentor persistent tumor and for eliminating the need for iodinated contrast material. CT and MRI are preferable to ultrasonography, in most cases, for their ability to detect disease in the mediastinum. See also Chapters 2 and 24.
Thallium and Sestamibi Radionuclide studies such as thallium and sestamibi hold promise as adjunctive diagnostic studies. 201Tl and %Tc methoxyisobutyl isonitrite (%Tc-MIBI) imagi the advantage of not requiring thyroid hormone cessation, immediate imaging after l3II. Several investigators have found injection, and more favorable dosimetry than thallium imagingto be more sensitive thanI3'I diagnostic imaging for the detectionof thyroid cancer and effective in showing more extensive disease (36-38). However, thalliumscanningappearstolackspecificitywhencomparedto 1311 imaging (38). Thallium is known to accumulate in a variety of inflammatory and malignant lesions (39).Uptake in tumor cells most likely involves the ATPase sodium potassium tr system and is influenced by tumor viability and blood flow (39). Therefore, lack of specificity can be expected. Sestamibi imaging has been shown to be comparable to thallium studies but may offer more favorable imaging characteristics due to its %Tc label (40,4I). Most authors reviewing retrospective studies of thallium, I3lI, %Tc-MIBI imaging and thyroglobulin determinationin the follow-up of thyroid cancer find a combinatio (36,38,4043). Thalliumandsestamibi of modalitiesyieldsthegreatestreliability imaging is very useful in patients who have discordance in the findings on 13'1diagnostic of thyroid scans and thyroglobulin determination and in patients for whom the cessation hormone would be imprudent.
Somatostatin Receptor Scintigraphy Other radionuclide diagnostic scans have also been utilized in the follow-up of thyroid cancer patients. Somatostatin receptor scintigraphy (SRS), using "'In-DTPAPhe-octreotide, has been studied by a group in France (44,451 with favorable results in identification of unknown tumor sites. In 16 patients with negative I3'I whole-body scans and elevated thyroglobulin levels,12 were found to have a turnor site identified by SRS (45). They also report that it may show a specific use in identification of site of insular thyroid carcinoma(44). Although differentiated thyroid cancers in vitro ha (16),there is a recent report of bindi not been found to contain somatostatin receptors tosomatostatinreceptorsinmembranesandcelllines of thyroidcarcinoma (47). Like thallium and sestamibi, SRS does not require withdrawal of T4 treatment. Most significantly, positive SRS imaging may indicate the efficacy of octreotide therapy This certain aggressive tumors that have not responded to other therapeutic modalities. has not yet been explored.
FDG PET Imaging
2-~uoro-2-deoxy-~-g~ucose (F'DG)is accumulated preferentiallyin the heart, brain, of glycolysis (48). Therefore, it is not surprising and malignant tissue due to the rate high that FDG labeled with the positron-emitting 18F can be utilized in the detection of
Maximizing Therapeutic and Diagnostic
l3II
Uptake
247
metastatic thyroid cancer. However, I8Fhas a half-life of only 109 minutes, requiring a nearby cyclotron facility. In addition, positron emission tomography (PET) is not readilyavailableatmostnuclearmedicinefacilities.Butrecently,technological advances in imaging have made it possible to transform single photon emission com tomography (SPECT) equipment for use in positron imaging, with the use of special computer software and high-energy collimators. A recent prospective study compared 41 patients who had previously undergone thyroidectomy and radiotherapy, after examination byVDG PET and 13’1 whole-body imaging (49). Combined T D G and 13’1 imagingresultedinasensitivityof95%. However, there was an “alternating” pattern of uptake. Of the 41 patients, 34 had increased thyroglobulin levels, 19 had positive I8FDG scans and negative I3’I scans, and 6 had positive 1311 studies and negative I8FDG scans, with 5 of these having a in “mixedtype”pattern.Most importantly is a 94%sensitivityinI8FDGscanning detection of metastases in patients with high thyroglobulin levels and no I3’Iuptake. tumors represent functionally The authors believe that ‘3’I-positive and 18FDG-negative better differentiated low-grade tumor cells, whereas the opposite pattern of uptake (I8FDG positive and l3II negative tumors) represent a lower functional differentiation and higher malignancy. This conclusion has not been proven but has been hypothesized by other groups as well (48). As PET scanning becomes more readily available, FDG imaging may provide an important concurrent screening examination when used with I3lI imaging. ADDENDUM. LOW-IODINE DIET AVOID Iodized salt Milk and dairy products Eggs Seafood (including fish, shellfish, kelp, and seaweed) Breads made with iodate dough conditioners Red food dyes (found in cereals, candies, and vitamins) Restaurant food (including “fast” food restaurants) Foods containing any of the following ingredients: iodized salt, sea salt, iodates, iodides, algin, alginates, agar agar Adapted from Lakshmanan M, Schaffer A, Robbins J, et ak A simplified low iodine diet in 1-131 scanning and therapy of thyroid cancer. Clin Nutr Med 1988; 13:866.
REFERENCES 1. Hilts SV, HellmanD, Anderson J,et al. Serial TSH determination after T3 withdrawal or thyroidectomy in the therapy of thyroid carcinoma. J Nucl Med 1979; 20:928. 2. Goldman J M , Line BR, Aamodt RL, et al. Influenceof triiodothyronine withdrawal time JClin Endocrinol Metab1980; 50734. on l3’Iuptake post-thyroidectomy for thyroid cancer. 3. Edmonds CJ, Hayes S , Kermode JC, etal.MeasurementofserumTSHandthyroid
hormones in the management of treatment of thyroid carcinoma with radioiodine. Br J Radio1 1977; 50:799. 4. Hershman J M ,Edwards CL. Serum thyrotropin (TSH) level after thyroid ablation compare
248
Sweeney and Johnston
with TSH levels after exogenous bovine TSH implications for I3’I treatment of thyroid carcinoma. J Clin Endocrinol 1972; 34814. 5. Hays MT, Solomon DH, Beall GN. Suppression of human thyroid function by antibod to bovine thyrotropin. J Clin Endocrinol 1967; 27:1540. S, etal.Recombinanthumanthyrotropinstimulates 6.BravermanLE,PrattBM,Ebner thyroid function and radioactive iodine uptake in the rhesus monkey.J Clin Endocrinol Metab 1992; 74:1135. 7. Meier CA, Braverman LE, Ebner SA, et al. Diagnostic use of recombinant human I/IIstudy). J ClinEndocrinolMetab pininpatientswiththyroidcarcinoma(phase 1994; 78:188. 8. Maruca J, Santner S, Miller K, et al. Prolonged iodine clearance with a depletion regim for thyroid carcinoma: concise communication.J Nucl Med 1984; 25:1089. 9. Lakshmanan M, Schaffer A, Robbins et al. J, A simplified low iodine in diet 1-131 scannin and therapy of thyroid cancer. Clin Nutr Med 1988; 13966. 10. Maxon HR, Thomas SR, Boehringer A, et al. Low iodine diet in 1-131 ablation of thy remnants. Clin Nutr Med 1983; 8:124. on radioactive 11. Laurie AJ, Lyon SG, Lasser EC. Contrast material iodides: potential effects iodine thyroid uptake. J Nucl Med 1992; 33:237-238. 12. Rawson RW, Rall JE, Peacock W. Limitations and indications in the treatment of cance of the thyroid with radioactive iodine. J Clin Endocrinol 1951; 11:1128. 13. Park HM, Perkins OW, Edmondson J W , et al. Influence of diagnostic radioiodineson the uptake of ablative dose of iodine-131. Thyroid 1994; 4:49. 14. Jeevanram RK, Shah DH, Sharma M, et al. Influence of initial large dose on subseque Med Biol 1986; 13:277 uptake of therapeutic radioiodine in thyroid cancer patient. Nucl 15. Arnstein N B , Carey JE, Spaulding SA, et al. Determination of iodine-131 diagnostic dos for imaging metastatic thyroid cancer.J.Nucl Med 1986; 27:1764. 16.Waxman A, RamannaL,ChapmanN,etal.Thesignificanceof1-131scandosein patients with thyroid cancer: determination of ablation: concise communication. J Nucl Med 1981; 22961. 17. Naddaf S, Young I, Rapun R, et al. Comparison between iodine-123 0-123) and iodineinthyroid cancer patients. [Abstract]. J N 131 (1-131) sodiumiodide total body scanning Med 251P; 37:1996. of serum thyroglobulin measureme 18. AshcraftMW, Van Herle AJ. The comparative value and iodine 13 1total body scans in the follow-up study of patients with treated dif thyroid cancer. Am J Med 1981; 71906-814. 19. Becker D, Charkes N D , Dworkin H, et al. Procedure guideline for extended scintigrap for differentiated thyroid cancer: 1.0.J Nucl Med 1996; 37: 1269-1271. L, Burman KD. Current trends in the management of well20.SolomonBL,Wartofsky differentiated papillary thyroid carcinoma.J Clin Endocrinol Metab 1996; 81:333-339. 21. Maxon HR, Smith HR. Radioiodine-131 in the diagnosis and treatment of metastatic wel differentiated thyroid cancer. Endocrinol Metab ClinNorth Am 1990; 19:685. 22. Hurley J R , Becker DV. Treatment of thyroid carcinoma with radioiodine. In Gottschalk A, HofferPB,Potchen EJ, Berger HJ, editors. Diagnostic nuclear medicine, 2nd ed. Baltimore: Williams & Wilkins, 1988:792. 23. Morrish DW, Filipow W, McEwan AJ, et al. 13’1treatment of thyroid papillary carcinom in a patient with renal failure. Cancer 1990; 66:2509-2513. 24. Sutter CW, Masilungan BG, Stadalnik RC. False-positive results of 1-131 whole-body scans in patients with thyroid cancer. Semin Nucl Med 1995; 25:279-282. 25. Geattia 0, Shapiro B, Orsolon PG, Mirolo R, DiDonna A. An unusual false-positive sca in a patient with pericardial effusion. Clin NuclMed 1994; 19:678-682.
MaximizingDiagnostic Therapeutic and
'jlI Uptake
249
26. Ziessman HA, Bahar H, Fahey F H , Dubiansky V. Hepatic visualization on iodine-131 whole-body thyroid cancer scans. J Nucl Med 1987; 28:1408-1411. 27. Harbert JC. Radioiodine therapy of differentiated thyroid carcinoma. In HN, Wagner Szabo Z, Buchanan J W , editors. Principles of nuclearmedicine, 2nded.Philadelphia: WB Saunders, 199597-1019. 28. Reynolds J. Future prospectsfor treatmentof differentiated thyroid carcinoma. Ann Intern Med 1991; 115133. 29. Maxon HR, Englaroee, Hertzberg VS, ChenLS. Chest x-rays, bone scans and immediate post treatment 1-131 scans: utility in well-differentiated thyroid cancer. [Abstract]. J Nucl Med 1992; 33:894. 30.Nemac J, Rohling S, SamrazilV, et al. Comparison of the distribution of diagnostic and thyroablative 1-131 in the evaluation of differentiated thyroid cancers. J Nucl Med 1979; 20:92. S, et al. Clinical utilityof post-treatment radioiodine scans 30a. Sherman SI, Tielens ET, Sosre in the management of patient thyroid carcinoma. J Clin Endocrinol Metab 1994; 78:629. 1therapy in thyroid cancer patients with high thyroglobu31. PinedaJD et al. Value of iodine-13 lin and negative diagnostic scan. J Clin Endocrinol Metab 1995; 801488-1492. 32. Pacini F, Lippi F, Formica N, et al. Therapeutic doses of iodoine-131 reveal undiagnosed metastases in thyroid cancer patients with detectable serum thyroglobulin levels. J Nucl Med 1987; 28:1888. 33. Clark OH, Hoelting T Management of patients with differentiated thyroid cancer who havepositiveserumthyroglobulinlevelsandnegativeradioiodinescans.Thyroid 1994; 4501. 34. Ronga G, Fiorentino A, Paserio E, et al. Can iodine-131 whole body scan be replaced by thyroglobulin measurement in the postsurgical follow-up of differentiated thyroid carcinoma? J Nucl Med 1990; 31:1766. 35. Aufferman W, Clark OH, Thurmers S, Galante M, Higgins CB. Recurrent thyroid carcinoma: characteristics on MR images. Radiology 1988; 168:753-757. 36. Ramanna L, Waxman A, Braunstein G. Thallium-201 scintigraphy in differentiated thyroid cancer: comparison with radioiodine scintigraphy and serum thyroglobulin determination. J Nucl Med 1991; 32:441. 37.Burman KD, Anderson JH, Wartofsky L, et al. Managementofpatientswiththyroid carcinoma: application of thallium 201 scintigraphy and magnetic resonance imaging. J Nucl Med 1990; 31:1958. 38. Hoefnagel CA, Deprat CC, Marcuse HR, devijlder JJM. Role of thallium-201 total-body scintigraphy in follow-up of thyroid carcinoma. J Nucl Med 1968; 27:1854-1857. 39. Waxman AD, Ramanna L, Memsic LD, et al. Thallium scintigraphy in the evaluation of mass abnormalities of the breast. J Nucl Med 1993; 34:18-23. 40. Dadparvars S , Chevres A, Tulchinsky M, Krishna-Badrinath L, Khan AS, Slizofski WJ. Clinical utilityof technetium-99m methoxisobutylisonitrite imaging in differentiated thyroid carcinoma: comparison with thallium-201 and iodine-l31Na scintigraphy and thyroglobulin quantitation. Eur J Nucl Med 1995; 22:1330-1338. 41. KosudaS, Yokoyama H, Katayama M, Yokokawa T, Kusano S, Yamamoto 0.Technetium99mtetrofosminandtechnetium-99msestamibiimagingofmultiplemetastasesfrom differentiated thyroid carcinoma.Eur J Nucl Med 1995; 22:1218-1220. 42. Brendel AJ, Guyot M, Jeandot R, Lefolt G, Manciet G. Thallium-201 imaging in the follow-up of differentiated thyroid carcinoma. J Nucl Med 1988; 29:1515-1520. 43. Iida Y, Hidaka A, Hatabu H, Kasagit, Konishi J. Follow-up study of post-operative patients with thyroid cancer by thallium-201 scintigraphy and serum thyroglobulin measurement. J Nucl Med 1991; 32:2098-2100.
Johnston 250
and
Sweeney
44. Tenenbaum F, Lumbroso J, Schlumberger M, CaillouB, Fragu P, ParmentierC. Radiola
beled somatostatin analog scintigraphy in differentiated thyroid carcinoma. J Nucl Med 1995; 36~807-810. 45. Baudin E, Schlumberger M, Lumbrosa J, et al. Octreotide scintigraphy in patients wit differentiated thyroid carcinoma: contribution for patients with negative radioiodine s J Clin Endocrinol Metab 1996; 81(7):2541-2544. 46. Krenning E, Kwekkeboom DJ, BakkerWH, et al. Somatostatin receptor scintigraphy w ["'In-DTPA-D-Phel- and ["'WI- octreotide: the Rotterdam experience with more 1000 patients. EurJ Nucl Med 1993; 20:716-731. 47. Ain KB, Taylor KD. Somatostatin analogs affect proliferation of human thyroid carcinom cell lines in-vitro. J Clin Endocrinol Metab 1994; 78:1097-1102. 48.Joensuu H, AhonenA.Imagingofmetastasesofthyroidcarcinomawithfluorine-18 fluorodeoxyglucose. J Nucl Med 1987; 28:910. 49. Feine V, Lietzenmayer R, Hanke Jp, Held J, Wohrle H, Muller-SchauenbergW. Fluorine WFDG and iodine-131-iodine uptake in thyroid cancer. J Nucl Med 1996; 37:1
24 An Approach to the Management of Patients with Scan Negative, Thyroglobulin Positive, Differentiated Thyroid Carcinoma Alternative Imaging Procedures Leonard Wartofsky Elsewhere inthis volume, authorities have described general aspects of the diagnosis (DTC) with recommendations and management of differentiated thyroid carcinoma largely consistent with those in recent reviews ( I ) and published guidelines for management (2). Yet, it remains clear that controversy continues to plague our ability to develop specific evidence-based practice guidelines for issues related to initial radioI3II diagnostic and therapeutic interventions, due largely iodine ablation and subsequent to the broad heterogeneityof the clinical characteristicsof our patients and to the lack (3-7). In our earlier survey (8) of sufficient data from well-controlled prospective studies of management practices by clinical thyroidologists DTC, for postoperative radioiodine ablation was recommended for a 2-cm well-encapsulated lesion without evidence of tissue invasion by 61% of respondents. 69% of respondents obtained a pretreatment l3lI whole-bodyscan ( W B S ) and 87% aposttreatmentscan.59%wouldobtaina subsequent follow-up scan and 85% monitored serum thyroglobulin (Tg), whereas of other variations of the index case described. management varied widely on a number While DTC remains one of the most curable of all cancers, occasional patients with aggressive disease are seen, and outcomes have been clearly related to a number of variables (6,7,9-13). Given then the numerous variables existing in any single individual or group of patients, I believe the designof, and adherence to, an algorithmic approach of patients withDTC to be both treacherous and possibly to the follow-up management overly simplistic. Rather, management of each case should be individualized. Several series suggest that excellent prognosis with cure attends DTC of ~ 1 . 5cm diameter, even when treated with less than a near-total thyroidectomy(3,6,7,12,13). I would like to focus this discussion on higher risk lesions which are 2 cm or greater and which may have been associated with either regional node metastases or distant metastases. For such patients, the usual management would be total thyroidectomy followed by radioiodine ablation.This initial management has beenalmost universally adopted since the publication by Mazzaferri and colleagues (14) demonstrating that
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press he., Totowa, NI
251
252
Wartofsky
less aggressive surgery without radioiodine ablationis associated with higher rates of recurrence and death. Some authorities would have us adopt a more “selective to the use of radioiodine, challenging whether ablation is actually of benefit and pointi out that the reduced recurrence rates seen in the Mazzafemi series after radioiodine ablation may have been the result of less than complete total thyroidectomies. This conclusion is based on the finding of comparable recurrence rates with or without radioiodine ablation at the MayoCl i i c where an ostensibly more complete thyroidec tomy was performed. The resultsof Simpson and colleagues(15) can be citedas supporting this contention because these workers found that radioiodine ablation benefited survival only withresidualmicroscopicdisease.But,curiously,theMayogrouphadpreviously published data indicating that frequency of local recurrence was no different between (16),and survival was no patients having total versus bilateral subtotal thyroidectomy improved by total thyroidectomy in either minimalor higher risk patients with papill carcinoma (17). Thus, the cited explanation for the varianceof the Mayo results with those of Mazzaferri would appear unlikely or at least incomplete, and there must be I wouldinfera otherdifferencesbetweentherespectivepatientgroupsanalyzed. somewhat different conclusion from the observationsof Simpson’s group, and thatis a resultant clear benefit of radioiodine ablation. The issue is related to the degree of certainty of the treating physician that thereis no residual microscopic disease. Many of the reported series do not provide clear evidence of the extent of residual tissue as might be inferred fromRAI uptakes or serum Tg levels. I would agree that a low- to 2 ng/ml and a postoperativ moderate-risk patient with a baseline Tg level of less than radioactive iodine (RAI) uptake of less than 0.5% wouldgot require ablative therapy However, if we place ourselves in the shoes of even a “low-risk” patient, would we 30-ato 60-mCi ablative dose 13’1 ofin exchang not willingly accept the consequence of forthecertaintyandpeace ofmindprovidedby asubsequentnegativescanand undetectable serum Tg level? That radioiodine therapy “wipes the slate clean” and 13’1WBS or monitoring provides the abilityto more readily detect recurrence by either serum Tg is a somewhat compelling argument to me. Others may feel that Tg levels adequately monitor such patients, with 93% and 80% having immeasurably low lev while on and off levothyroxine suppression, respectively. These are impressive p ages on a statistical basis but not on an individual patientI have basis;difficulty settlin for an 80-90% average if radioiodine ablation could bring my awareness of the 100%. It can be argued that any patients orabsenceofresidualdiseasecloserto to rise, and that they coul “missed” could be caught at a later time as serum Tg began if necessary at that time. Such management w then be further evaluated and treated likely suffice and achieve a “cure”in most patients, but againI have concern for thos In such patients, cure appe who unpredictably may manifest more aggressive disease. to be best achieved when their disease is caught and treated as early as possible. Mazzaferri and Jiang(11)found that delays in treatment were associated with a hig mortality. Moreover, a serum Tg-based system of follow-up for detection of residual disease may be misleading due to false negatives; Schlumberger’s group(18) reported patients with little elevationin Tg even in the faceof pulmonary metastases, and the have described 20% of patients with known lymph node metastases who may have
cedures Alternative Imaging
253
low serum Tg. That Tg may be undetectable in the face of known metastatic disease while patients are on levothyroxine has been known for quite some time (19). Administration of an ablative dose of radioiodine permits a follow-up scan that can detect the presence of unrecognized metastases. This fact also bears on our ability to achieve the salutary effects of treatment by earlier diagnosis. Casara and coworkers (20) realized remission rates of 78% in patients with negative chest radiographs but positive lung uptake on scan, in contrast to a remission rate of less than 4% in those patients who already had evidence of lung metastases on chest x-ray. Such studies confirm the reciprocal relationship between success of cancer therapy and size and duration of lesions. Several recent reports have confirmed improved survival or reduced rates of recurrence after radioiodine ablation (33-7). Certainly few would argue with taking advantage of the administration of a treatmen dose of I 3 I I by performing a posttreatment scan, particularly with large(75-150 mCi) treatment doses, which may disclose previously undetected metastases. But we can see patients with low serum Tg and significant metastatic disease even when no longer of falseTSH-suppressed (21). This phenomenon is related to the whole dilemma negative or low-serum Tg levels in the face of known residual malignancy. The most likely explanationsfor this include the presenceof interfering anti-Tg antibodies, which can be present in as many as40% of patients with DTC (22), and to dedifferentiation of the tumor cells such that they can still trap iodide but can no longer make and release Tg. It should be mandatory for laboratories to measure anti-Tg antibodies in every sample in which Tg is being measured. While assays for serum Tg are improving (22) point and becoming more standardized (albeit very slowly), Spencer and associates out that the presence of antibodies continues to be a problem, and can be associated with either falsely low- or falsely high-serum Tg measurements. Hence a largely Tgbased strategy may be problematic at times, given the frequency of antibodies in the DTC population. The need for levothyroxine withdrawal for the purpose of a follow-upTBS a year this procedure requires allowing after therapy has also been questioned. At present, the patient to become sufficiently hypothyroid to raise serum TSH to A 0 mUA and thereby facilitate 13’1 uptake and imaging of either residual thyroid bed tissueor malignancy. Such scans are done with doses l3II of which range from3 to 10 mCi. The larger doses are associated with better images and improved sensitivity of tumor detection but may also be associated with “stunning” or a lower fractional uptake of 1311 with the (23). Some have advocated abandonment of these TBS doses subsequent treatment dose as either unnecessary in the low-risk patient with low-serum Tg, or problematic in the higher risk patient because of stunning. In the latter group, they would have us opt instead for a much larger dose that would suffice for both imaging and treatment. We Iz3I or technetium have found stunning to be much less problematic by employing postoperatively rather than 13’I.In this context, I am drawn again to the importance of I believe that the aggresindividualizing management in the decision-making process. siveness of further diagnostic and therapeutic approaches should be based upon the patient’s risk factors, clinical situation, serum Tg on and off levothyroxine therapy, and other nonisotopic imaging techniques to identify residual or recurrent disease such as ultrasonographyor MRI. With the imminent availability of recombinant human TSH
254
Wartofiky
(rhTSH) (24) (see Chapter 14), we will be able to evaluate patients without having to render them hypothyroid. Indeed, in the absence of anti-Tg antibodies, failure of Tg to increase after rhTSH administration would be a compelling argument for the being free of malignancy and therefore having no indication for scanning or further isotope treatment. There will also likely be a role for rhTSH in facilitating the thera of patients with metastatic disease. I have personally treated one patient with metasta disease after rhTSH preparation and another has been reported with salutary results (20). The ability to identify residual or recurrent diseaseby other scanning modalities which also would not require discontinuing TSH-suppressive levothyroxine therapyis discussed below. Most endocrinologists dealing with thyroid cancer patients would agree with the therapeutic approach of combination 1311 and surgery. In one case report, I3’Itherapy was administered preoperatively to effectively reduce tumor bulk (26). Some centers have innovatively employed a preoperative dose 13’1to ofidentify lesions by an intra ative detector probe (27,281. Many centers in the USA would operate first on any palpable or otherwise accessible tumor recurrence and follow the surgery with ra dine therapy. This has the benefit of “debulking” the tumor mass and rendering the radioiodine dose as more effective therapy, for small neoplastic foci respond better 1311than larger ones. I think that it might be preferable to employ preoperative a tion of Iz3Ito guide the intraoperative gamma probe of the surgeon, an isotope that would be associated with less radiation exposure and less potential for stunning. Th following surgery, the patient could be treated with a large doseof 1311,particularly if (27). Surgicalexcisionmaynotbewarrantedforthe serum Tg was still elevated appearance of clinically detectable cervical lymph nodes in the presenceof low serum Tg and negative anti-Tg antibodies. Rather, one would want to first demonstrate the presence of DTC metastases in the lymph nodes, which could be done by fine needle aspiration biopsy with or without ultrasound guidance and subjecting the aspirate to either cytologic examination (29), polymerase chain reaction (PCR)-based gen sis (30), or measurement of Tg (31). Schlumberger was the first investigator to advocate empirical high-dose13’1therapy for patients who are “scan-negative, thyroglobulin-positive” (32). Others have advis caution in applying high-dose therapy in such patientsin the absence of data confirmin I must agree with them. With this efficacy and an acceptable risk:benefit ratio, and scenario, one should first attempt to uncover a cause for a possibly false-negative or a false-positive elevation of serum Tg. As mentioned above, the latter can be due to interfering anti-Tg antibodies (22). Explanations for a false-negative radioiodine scan include inadequate TSH elevation, stable iodine contamination (e.g., history of recentiodinecontrastradiography),dispersedmicroscopicmetastasestoosmallto visualize, or dedifferentiation of the tumor such that it can still produce Tg but has lost its iodide trapping ability. Recent attempts to “redifferentiate” such tumors with retinoic acid are of considerable interest may and be associated with restoration of bo radioiodine uptake and thyroglobulin production (33,3#). To rule out iodine con tion, serum or urinary iodide can be measured and a repeat TBS 4-6 weeks after an iodide depletion regimen can be considered (35). In this setting faced with a decision as to how to proceed, Iwould again look atthe patient in terms of risk factors, evidence of prior metastatic or aggressive diseas
Alternative
255
arguments for employing other imaging tools such as MRI by or ultrasound to visualize possibly occult disease. For example, given this scenario of “positive Tg and negative TBS” in a 58-year-old man with a history of a 4-cm papillary or a 2-cm follicular lesion, I would consider the Schlumberger approach and treat with high-dose radioiodine. On the other hand, with a history of a 2-cm papillary cancer with negative nodes and only marginally measurable or slightly elevated Tg (e.g., -5 ng/ml), I might favor a more is whether conservative approach.Of significant utility in the decision-making process serum Tg levels are stable or rising. Of course, the patient must be brought into the decicollective knowledge, experision-making process and informed fully of the our extent of ence, and biases in regard to their specific situation. We likewould to avoid treatment of patients with aggressive high-dose radioiodine for uncertain indications, and which (36). result in troubling sequelae such as xerostomia and/or azoospermia Absent evidence of progressive disease, the risks of aggressive radioiodine therapy may not be warranted given ill-defined goals.In addition to the experienceof Schlumberger and collaborators (32), another oft-cited experience in support of empirical treatment ofthe thyroglobulin-positive, scan-negative patient is that of Pineda and coworkers (37). These workers reported 17 such patients who all had prior total thyroidectomy and I3’I,16/17 radioiodine ablation. After treatment with 150-300 mCi of had visualization of metastases on their posttreatment scan. Tg levels decreased in 81% of patients after their first treatment dose, and in 90% and 100% of those patients who received second and third doses, respectively. While these results sound impressive as expressed, examination of the individual patient’s Tg level responses is less so. Mean Tg decreased from 74 to 62 to 32 over1-2 years of follow-up, and only 6/29 positive scans became negative. The cogent issues raised byMcDougall(38) and Mazzaferri (39), reflect the fact that many of these patients have minimal if any disease that would affect their life expectancy and we may be exposing them to unwarranteddosesofradiation exposure, unwarranted at least until we obtain sufficient data from well-controlled this studies that confirm efficacy of therapy. Certainly another important aspect of empirical therapyis the cost to the patient in regard to the morbidity of hypothyroidism and its negative impact on productivity, as well as the cost in health care dollars relat to hospitalization and the associated expensive technological procedures. Increasing scrutinybywatchdogagenciesmaychallengetheindications for this therapyand possible denial of reimbursementmay cause additional problems for both the patients and their physicians. Finally, I would mention the additional or alternative imaging procedures that are being developed and evaluated for patients with thyroid cancer. Given a negative I311 TBS, are there other scanning modalities that might provide useful information, even though the I3’I TBS is generally considered to be the gold standard for detection of metastases? Once we have eliminated the various causes for false-positive serum Tg As reviewed above, many authorities or false-negative TBS, what is the clinician to do? question the risk:benefit ratio of arbitrary high-dose13’1therapy as has been employed by Schlumberger and coworkers (32). Alternative therapeutic approaches to metastatic deposits of thyroid cancer include surgical excision or localized external radiation MN therapy (6,40), but the location of the metastases would need to first be identified. and ultrasound have been employed for this purpose. In addition, I suggest that alternative scanning agents may play a very important role in this regard, for several recent
256
Wartofiky
reports have documented their potential utility in identlfying lesions which are not visualized with traditional I3lI WBS. One of the first to be employed was 2oLTl, and absence or presence of 2 0 1 T l uptake has been shown to correlate with success or failure of prior I3'I therapy and thus have predictive prognostic utility(41). In one recent study of patients with bone metastase documented with positive I3'I scans (42), 2 0 1 T l was compared to the bone agent, 9 9 m T hydroxymethylenediphosphonate(99mTc-HMDP).The two agentshadacombined I3IIscans and other evidenc sensitivity of 93.5%. In a groupof 14patients with negative of thyroid malignancy, 2 0 1 T l was positive in 10/14 and Vc-HMDP was positive in all 14. Carril and associates (43) found that z o l T l enjoyed a sensitivity and specificity that was higher than that 13'1 forfor recurrentor persistent disease. Lesions were dete in 31/116 patients by 2 0 1 T l but not by l3II TBS. In patients who have been ablated and show no further I3'Iuptake, the investigators propose continuing management with n additional I3'I scans; since 2 0 1 T l scanning does not require levothyroxine withdrawal, follow-up would be guided only by "'Tl scanning and by monitoring serum Tg. Dadparvar and colleagues (44) compared ' O I T l and scanning with 99mTc-methoxyisobuty1 isonitrile (%Tc-MIBI) and found that I3II TBS alone was satisfactory as a preablation study, but that the addition of either alternative agent increased the diagnostic yield postablation, particularly when the13'1TBS was negative. These results have not been (45) found 13'1TBS to be both universal, however, because Lorberboym and colleagues 2 0 1 T l , with the latter giving several false-positive sca more sensitive and specific than Ugur and colleagues(46) noted a 70% overall concordance between"'T1, 99"Tc-MIBI, and I3IITBS, but observed false negatives with both alternative agents and conclude that they should not be used in lieu of 13'1 TBS. In a recent case report of a Hiirthle cell oxyphilic carcinoma that neither took up I3'I nor released thyroglobulin, 2 0 1 T l TBS successfully demonstrated multiple metastatic sites in bone and mediastinum (47). =Tc has been similarly useful and Elser and coworkers (48) noted a 94% sensitivity for the detection of positive lymph nodes and local recurrence with %Tc-sestamibi; they detected 32/40 metastases with Sestamibi compared to only 18/40 with l3II TBS. More recently, investigators have attempted detection of thyroid cancer with =Tc-tetrafosmin, a cationic agent employed pre for myocardial perfusion imaging (49-51). For 12 patients with elevated serum Tg (four of whom had negative 13'1TBS), tetrafosmin was slightly superior to either *"Tl or*"Tc-MIBI. This same group of workers (51) reported that tetrafosmin successfully identified 21/21lesions which were positive byI3'I TBS but an additional 17/23 lesions that were negative by 1311 TBS. The agent had 86% sensitivity for distant metastases, was positive in four patients with 1311 negative proven pulmonary metastases, and the findings correlated with other modalities identifying tumor such as CT or ultrasound scans. A follow-up study of a larger group of patients indicated that tetrafosmin was in lung, significantly more sensitive than I3'I scanning for detection of metastases mediastinum, and lymph nodes, but less sensitive for thyroid remnants or bone ses (52). It is also significant that these alternative agents are logistically both more con and more expedient than scanning with I3'I. In addition to being able to scan patients while they are still taking TSH-suppressive doses of levothyroxine, the time required for evaluation is much reduced. Instead of scanning 48-72 hours after a dose of I3'I
Alternative Imaging Procedures
257
the gh"Tc-tetrafosmin planar scan is performed 20 minutes after injection of the isotope with additional images taken by SPECT of any suspicious lesions. gh"Tc-tetrafosmin scans were negative in all 68 patients studied by Lind and colleagues (51) who were free of disease on the basis of 13'1TBS and serum Tg. In a comparison of 201Tl, I3'I, andtetrafosmin,Unalandcoworkersfoundrelativelycomparablesensitivitiesfor detection of distant metastases of 0.85, 0.78, and 0.85, respectively (53). Iodine was much more sensitive (1.00) for detection of residual tissue postoperatively than either 2 0 1 T l (0.33) or tetrafosmin (0.33). Another agent,18Ffluorodeoxyglucose (FDG)is employed with PET scanning (FDGPET) with uptake of the agent related to glucose utilization by tumor tissue. The gre uptake sensitivity has been noted with the fastest growing undifferentiated tumors.In fact, time-activity curvesof FDG uptake suggest the ability of scanning with this agent to distinguish benign from malignantnoduleswiththemalignantlesionsshowing increasing uptake with time while uptake in benign lesions decreases with time (54). Grunwald and associates (55) compared FDG-PET to 99mTc-sestamibi and I3lITBS. Of 29 studies, 11/29 had disease detected only with FDG-PET, 8/29 were detected only with I3II TBS, and 10/29 were detected by both. Five sites were detected by FDG-PET and not by %Tc-sestamibi. FDG-PET may be useful in patients in whom I3lI TBS is not feasible due to a history of iodine exposure and similarly, its use would not preclude as an additional means of imaging tumor. A CT scanning with contrast if desired drawback is the lack of widespread availability of PET scanners due to their high cost. Fridrich and colleagues (56) compared FDG-PET to 9h"Tc-MIBI and 1311 TBS and I3'I TBS with a slight edge in favor of V c found both to be more sensitive than MIBI. In addition to the benefit of having good uptake independent of the patients' se TSH level, FDG-PET or MIBI did not have the propensity to have high background in 1311, and could be employed more effectively the neck, mediastinum, and chest, as does to detect small metastases in these areas. On the other hand, liver and brain will demonstrate high uptake of FDG and the ability to pick up metastases in these areas will be limited withthis agent. Indeed, Feine and colleagues(57) were able to localize in six patients with elevated and identify positive neck metastases with FDG-PET serum Tg levels. A more conservative view to the utility of FDG-PET scanning has beenproposedbyDietleinandcoworkers (58). They observed positive FDG-PET 1311 TBS; images in 7 of 21 patients with positive lymph node metastases but negative sensitivity was 82%in patients with high-serum Tg but negative TBS. They concluded l3II TBS, but would serve as a useful that FDG-PET should not be used instead of complement to evaluation, particularly when the 1311 TBS was negative in the face of a rising or elevated levelof serum Tg. Early results with thyroid tumor detection with another recently employed technetium scanning agent, 99mTc-furifosmin, have not s it to be at all as sensitive as 2 0 1 1 1 or FDG-PET (59). Finally, imaging of DTC by somatostatin receptor scintigraphy (SRS) with octreotide has been reported by Baudin and colleagues (60). Of 25 patients with DTC and elevated serum Tg levels, 16 had negative 1311 TBS and SRS was positive in 12 of these 16 patients, and in 8 of 9 patients with positivel3IITBS. While confirmatory studies will be required, SRS with labeled octreotide may represent another useful alternative to 1311 TBS with the advantage of not having to withdraw TSH-suppressive levothyroxine therapy.
258
Wartofiky
In conclusion,howshouldonemanagethescan-negative,thyroglobulin-positive patient with no underlying reason to suspect either a false-negative scan or a falsepositive serum Tg level? Schlumberger (1,32)would empirically treat with 100 mCi 13’1any patient with a Tg level higher than10 ng/ml while off levothyroxine, and only repeat the 1311 WBS every 2-5 years when the Tg level is in the range of 1-10 ng/ml. Given clearly measurable Tg levels, I would encourage alternative imaging procedure For papillary thyroid carcinoma with a propensity to regional recurrence, that could include ultrasound, CT, MRI, 9h”Tc”IBI, or FDG-PET. For follicular thyroid cancer with its propensity for distant metastases (especially to bone and lung), 99mTcor %Tc-HMDPorcouldbeemployed.Identificationofisolateddistantlesions by these methods would allow earlier intervention by surgical excision or external radiotherapy, rather than delaying further treatment until a subsequent I3lITBS might become positive or serum Tg levels might increase further as a result of tumor g In patients with higher risk disease following early total thyroidectomy and high-do radioiodine ablation,this approach should permit effective management until such ti as more target-specific tumoricidal therapies become available.
REFERENCES
1. Schlumberger MJ. Papillaryandfollicularthyroidcarcinoma. N EnglJ Med 1998; 338~297-306. FS,Levy EG, et al. Treatm 2. Singer PA, Cooper DS, Daniels GH, Ladenson PW, Greenspan guidelines for patients with thyroid nodules and well-differentiated thyroid cancer. Arch Intern Med 1996; 156:2165-2172. 3. Taylor T, SpeckerB, Robbins J, Sperling M, HoM, Ain K, et al. Outcome after treatmen of high-risk papillary and non-Hurthle-cell follicular thyroid carcinoma. Ann Intern Med 1998; 129:622-627. 4. Cady B. Presidential address: Beyond risk groups-A new look at differentiated thyroid cancer. Surgery 1998; 124:947-957. 5. Schlumberger M, Challeton C, De Vathaire F, Travagli J-P, Gardet P, Lumbroso J-D, et al. Radioactive iodine treatment and external radiotherapy for lung and bone metastases from thyroid carcinoma. J Nucl Med 1996; 37598-605. 6. Tsang RW, Brierley JD, Simpson WJ, Panzarella T, Gospodarowicz MK, Sutcliffe SB. The effects of surgery, radioiodine, and external radiation therapy on the clinical outc of patients with differentiated thyroid carcinoma. Cancer 1998; 82:375-388. 7. Mazzaferri EL. Thyroid remnant 131-1 ablation for papillary and follicular thyroid carcinoma. Thyroid 1997; 7:265-271. 8. Solomon BL, Wartofsky L, Burman KD. Current trends in the management of well different ated papillary thyroid carcinoma.J Clin Endocrinol Metab 1996;81:333-339. 9. Ruegemer JJ, Hay ID,Bergstralh ET, Ryan JJ, Offord KP, Gorman CA. Distant metastase in differentiated thyroid carcinoma: a multivariate analysis of prognostic variables. J Clin Endocrinol Metab 1988; 67501-508. 10. Dulgeroff AJ, Hershman JM. Medical therapy for differentiated thyroid carcinoma. Endo Rev 1994; 15500-515. 11. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 1994; 97:418428. 12. Samaan NA, Schultz PN, Hickey RC, Goepfert H, HaynieTP, Johnston DA, Ordonez NG The results of various modalities of treatment of well differentiated thyroid carcinoma: a retrospective review of 1599 patients. J Clin Endocrinol Metab 1992; 75714-720. 13. Hay ID,Bergstrdh ET, Goellner JR, Ebersold JR, Grant CS. Predicting outcome in pap
rocedures Alternative Imaging
259
thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of1779 patientssurgicallytreatedatoneinstitutionduring 1940 through 1989. Surgery 1993; 114:1050-1058. RL, Oertel JE, KammererWT, Page CP. Papillary thyroid carcinoma: 14. Mazzaferri EL, Young the impact of therapy in 576 patients. Medicine 1977; 56:171-196. 15. Simpson WJ, Panzarella T, CarruthersJS, Gospodarowicz MK, Sutcliffe SB. Papillary and follicular thyroid cancer: impact of treatment1578 in patients. IntJ Radial Ankle Bill Phys 1988; 14~1063-1075. 16. Grant CS, Hay ID, Gogh IR, Bergstralh ET, Goellner J R , McConaheyW M . Local recurrence 1988; in papillary thyroid carcinoma: is extent of surgical resection important? Surgery 104:954-962. 17. Hay ID, Grant CS, Taylor WF, McConahey WM. Ipsilateral lobectromy versus bilateral lobar resection in papillary thyroid carcinoma: a retrospective analysis of surgical outcome using a novel prognostic scoring system. Surgery 1987; 102:1088-1095. 18. Schlumberger M, Tubiana M, DeVathaire F, Hill C, Gardet P, Travagli JP, et al. Longterm results of treatment of283 patients with lung and bone metastases from differentiated thyroid carcinoma. J Clin Endocrinol Metab 1986; 63:960-967. 19. Grant S, Luttrell B, Reeve T, WisemanJ, Wilmshurst E, Stiel J, et al. Thyroglobulin may be undetectable in the serum of patients with metastatic disease secondary to differentiated thyroid carcinoma: follow-up of differentiated thyroid carcinoma. Cancer 1984; 5416251628. Girelli ME, BusnardoB. 20. CasaraD,Rubellow D, Saladini G, Masarotto G, Favero A, Different features of pulmonary metastases in differentiated thyroid cancer: natural history J Nucl Med1993; 34: 1626-1631. and multivariate statistical analysis of prognostic variables. 21. Brendel A J , Lamber B, Guyot M, Jeandot R, Dubourg H, Roger P, et al. Low levels of serum thyroglobulin after withdrawal of thyroid suppression therapy in the follow-up of differentiated thyroid carcinoma. EurJ Nucl Med 1990; 16:35-38. RB, Singer PA, etal. Serum 22. Spencer CA, Takeuchi M, Kazaroxyan M, Wang CC, Guttler thyroglobulin autoantibodies: prevalence, influence on serum thyroglobulin measurement, and prognostic significance in patients with differentiated thyroid carcinoma. J Clin End no1 Metab 1998; 83:1121-1127. 23. Park I", Perkins OW, EdmondsonJM, Schnute RB, Manatunga A. Influence of diagnostic radioiodine on the uptake of an ablative dose of iodine. Thyroid 1994; 449-54. 24. Ladenson PW, Braverman LE, Mazzaferri EL, Brucker-Davis F, Cooper DS, Garber J R , et al. Comparison of administration of recombinant human thyrotropin with withdrawal of thyroid hormone for radioactive iodine scanning in patients with thyroid carcinoma. N En J Med 1997; 337:888-896. 25. Rudavsky A Z , Freeman LM. Treatment of scan-negative, thyroglobulin-positive metastatic 131-1 and recombinant human thyroid stimulating hormone. thyroid cancer using radioiodine J Clin Endocrinol Metab 1997; 82:ll-14. 26. Shingu K, KobayashiS, Yokoyama S, Shimizu T, Kasuga Y, Fujimori M,et al. Effectiveness of preoperative radioactive iodine (131-1) therapy for locally advanced papillary thyroid cancer: a case report. Thyroid 1998; 8:1113-1116. 27. Gallowitsch HJ, Fellinger J, Mikosch P, Kresnik E, Lind P. Gamma probe-guided resection of a lymph node metastasis with 1-123 in papillary thyroid carcinoma. Clin Nucl Med 1997; 22591-592. 28. Travagli JP, Cailleux AF, Ricard M, Baudin E, CaillouB, Parmentier C, Schlumberger M. Combination of radioiodine (131-1) and probe-guided surgery for persistent or recurrent thyroid carcinoma. J Clin Endocrinol Metab 1998; 83:2675-2680. 29. Boland GM, Lee MJ, Mueller PR, Mayo-Smith W, Dawson SL, Simeone JF. Efficacy of Am Jnodes. Roentgen01 sonographically guided biopsy of thyroid masses and cervical lymph 1993; 161~1053-1056.
260
Wartofiky
30. Arhui F, Russo D, Giuffrida D, Ippolito A, Perrotti N, Vigneri R, Filetti S. Early diagnosis by genetic analysis of differentiated thyroid cancer metastases in small lymph nodes, Endocrinol Metab 1997; 821638-1641. 31. Pacini F, Fugazzola L, Lippi F, Ceccarelli C, Centoni R, Miccoli P, et al. Detection of thyroglobulin in fine needle aspiratesof nonthyroidal neck masses: a clue to the diagnosis of metastatic differentiated thyroid cancer.J Clin Endocrinol Metab 1992; 74:1401-1404. 32. Schlumberger M, Mancusi F, Baudin E, PaciniF. 131-1 therapy for elevated thyroglobulin levels. Thyroid 1997; 7:273-276. 33. Grunwald F, Pakos R, Bender H, Menzel C, Otte R, Palmedo H, et al. Redifferentiation therapy with retinoic acid in follicular thyroid cancer. J Nucl Med 1998; 39:1555-1558. 34. Grunwald F, Menzel C, BenderH, Palmedo H, Otte R, Fittuners R,et al. Redifferentiation therapy-induced radioiodine uptake in thyroid cancer. J Nucl Med 1998; 39:1903-1906. 35. Maxon HR, Thomas SR, BoehringerA, Drilling J, Sperling MI, Sparks JC, ChenIW.Low iodine diet in 1-131 ablation of thyroid remnants. Clin Nucl Med 1983; 8:123-126. 36. Alexander C, Bader JB, Schaefer A, Finke C, Kirsch C-M. Intermediate and long-term J Nucl Med 1998; side effects of high-dose radioiodine therapy for thyroid carcinoma. 39~1551-1554. 37. Pineda JD, Lee T, Ain K, Reynolds J, Robbins J. Iodine-131 therapy for thyroid cancer JClin Endocrinol Metab patients with elevated thyroglobulin and negative diagnostic scan. 1995; 801488-1492. 38. McDougall IR. 131-1 treatment of 131-1 negative whole body scan, and positive thymg lin in differentiated thyroid carcinoma: what is being treated? Thyroid 1997; 7:669-672. 39. Mazzaferri EL. Treating high thyroglobulin with radioiodine: a magic bullet or a shot in the dark? [Editorial]. J Clin Endocrinol Metab 1995; 80:1485-1487. 40. Tubiana M, Haddad E, SchlumbergerM, Hill C, Rougier P, Sarrazin D. External radiotherapy in thyroid cancers. Cancer 1985; 55:2062-2071. T, et al. Thallium-201 41. Nakada K, Katoh C, Kanegae K, Tsukamoto E, Shiga T, Mochizuki scintigraphy to predict therapeutic outcome of iodine-131 therapy of metastatic thyroid carcinoma. J Nucl Med 1998; 39:807-810. 42. Alam MS, Takeuchi R, Kasagi K, Misaki T, Miyamoto S, Iida Y, et al. Value of combined technetium-99m hydroxy methylene diphosphonate and thallium-201 imaging in detecti bone metastases from thyroid carcinoma. Thyroid 1997; 7:705-712. 43. Carril JM, Quirce R, Serrano J,Banzo I, Jimenez-Bonilla JF, Tabuenca 0, Barquin RG. Total body scintigraphy with thallium-201 and iodine-131 in the follow-up of differe thyroid cancer. J Nucl Med 1997; 3tk686-692. 44. Dadpawar S, Chevres A, Tulchinsky M, Krishna-Badrinath L, Khan AS, Slizofski WJ. Clinical utilityof technetium-99m methoxisobutylisonitrileimaging in differentiated thyr 1 scintigraphy and serum thyroglo carcinoma: comparison with thallium-201 and iodine13 ulin quantitation. Eur J Nucl Med 1995; 22:1330-1338. 45. Lorberboym M, Murthy S, Mechanick JI, Bergman D, Morris JC, Kim CK. Thallium201 and iodine-131 scintigraphy in differentiated thyroid carcinoma. J Nucl Med 1996; 37~1487-1491. 4 6 . Ugur 0, Kostakoglu L, Caner B, Guler N, Gulaldi NC, Ozmen M, et al. Comparison of in the follow-upof patients with well differentiate 201-Tl, 99mTc-MIBI and 131-1 imaging thyroid carcinoma. Nucl Med Commun 1996; 17:373-377. 47. Harder W, Lind P, Molnar M, MikoschP, Gomez I, Gallowitsch H-J, et al. Thallium-201 uptake with negative iodine-131 scintigraphy and serum thyroglobulin in metastatic oxyphilic papillary thyroid carcinoma. J Nucl Med 1998; 39:236-238. 48. Elser H, Henze M, Hermann C, Eckert W, Mende W. Wm-Tc-MIBI for recurrent and metastatic differentiated thyroid carcinoma. Nuklearmedizin 1997; 36:7-12.
Alternative Imaging Procedures
261
49. Lind P, Gallowitsch HJ. The use of non-specific tracers in the follow up of differentiated thyroid cancer: results with Tc-99m tetrofosmin whole body scintigraphy. Acta Med Austr 1996; 23~69-75. 50. Gallowitsch HJ, Kresnik E, Mikosch P, Pipam W, Goez I, Lind P. Tc-99m-tetrafosmin scintigraphy: an alternative scintigraphic method for following up differentiated thyroid carcinoma: preliminary results. Nuklearmedizin 1996; 35:230-235. 51. Lind P, Gallowitsch HJ, Langsteger W, Kresnik E, Mikosch P, Gomez I. Technetium-99mtetrafosmin whole-body scintigraphy in the follow-up of differentiated thyroid carcinoma. J Nucl Med 1997; 38~348-352. 52. Gallowitsch HJ, Mikosch P,Kresnik E, Unterweger 0, Gomez I, Lind P. Thyroglobulin 1 and technetium-99m-tetrafosmin whole-body scintigraphy in difand low-dose iodine- 13 ferentiated thyroid carcinoma. J Nucl Med 1998;393870-875. 53. Unal S , Menda Y, Adalet I, BoztepeH,OzbeyN,AlagolF,Cantez S. Thallium-201, Technetium-99m-tetrofosmin and iodine-13 1 in detecting differentiated thyroid carcinoma metastases. J Nucl Med 1998; 39:1897-1902. 54. Uematsu H, Sadato N, Ohtsubo T, Tsuchida T, NakamuraS , Sugimoto K,et al. Fluorine18-gluorodeoxyglucose PET versus thallium-201 scintigraphy evaluation of thyroid tumors. J Nucl Med 1998; 39:453459. 55. Grunwald F, Menzel C, Bender H, Palmedo H,Willkomm P, Ruhlmann J, et al. Comparison of 18FDG-PET with 131-iodine and 99m-Tc-sestamibi scintigraphy in differentiated thyroi cancer. Thyroid 1997; 7:327-335. 56. Fridrich L, Messa C, Landoni C, Lucignani G, Moncayo R, Kendler D, et al. Whole-body scintigraphy with 99m-TC-MIB1, 18F-FDG and 131-1 in patients with metastatic thyroid carcinoma. Nucl Med Commun 1997; 18:3-9. 57. Feine U, Lietzenmayer R, HankeJP, Held J, Wohrle H,Muller-Schauenburg W. Fluorine18-FDG and iodine-131 uptake in thyroid cancer. J Nucl Med 1996; 37: 1468-1472. 58. Dietlein M, Scheidhauer K, Voth E,Theissen P, Schicha H. Fluorine-18 fluorodeoxyglucose positron emission tomography and iodine-131 whole-body scintigraphy in the follow-up of differentiated thyroid cancer.Eur J Nucl Med 1997; 241342-1348. 59. Brandt-Mainz K, Muller SP, Sonnenschein W, Bockisch A. Technetium-9%-furifosmin in the follow-up of differentiated thyroid carcinoma.J Nucl Med 1998; 39:1536-1541. 60. Baudin E, Schlumberger M, Lumbroso J, Travagli JP, Caillou B, ParmentierC. Octreotide scintigraphy in patients with differentiated thyroid carcinoma: contribution for patients with negative radioiodine scan.J Clin Endocrinol Metab 1996; 81:2541-2544.
25 Papillary Thyroid Cancer Prognosis Henry B. Burch DETERMINANTS OF PROGNOSIS IN PAPILLARY THYROID CANCER
Prognostication in papillary thyroid cancer has been facilitated by the recognition of clinical and pathological features which correlate with the risk of recurrence and death from disease (Table 1). Numerous retrospective analyses have identified tumor size larger than 4 cm, advanced tumor grade, patient age older than 50 years, male as sex, local tumor invasion beyond the thyroid capsule, and distant metastatic disease having a negative impact on survival (reviewed in reference I ) . An extensive review of 1500 consecutive cases of papillary thyroid cancer seen at the Mayo Clinic over a of such an analysis(2). The 20-year cancer-specific 40-year period provides an example mortality inthis cohort was 0.8% for patients less than50 years of age, 7%for patients 50 to 59 yearsof age, 20% for patients 60-69 yearsof age, and 47% for patients aged 70 or more. Likewise, mortality from thyroid cancer increased with the size of the tumor, with a 20-year mortality of 0.8% for patients with tumors less than 2.0 cm in 50% for diameter, 6% for tumors 2.0 to 3.9 cm, 16% for tumors 4.0 to 6.9 cm, and tumors greater than 7 cm. Patients with tumors extending through the thyroid capsule had a 20-year mortality of 28%, compared to only 1.9% of patients with tumor confine tothethyroid.Theworstoutcomeoccurredinpatientswithdistantmetastasesat presentation, who experienced a 10-year mortality rate of 69%, compared to 3% in patients with tumors confined to the neck. Overall mortality from thyroid cancer was 9% for men and 4%for women. Althoughthis discussion has focused on cause-specific mortality from thyroid cancer, many of the same prognostic indicators cited in this section are also predictive of local recurrences and distant metastases (2). EFFECT OF TREATMENT ON OUTCOME
The extent of initial therapy for papillary thyroid cancer has value for predicting tumor recurrence and cancer-related death. In a study including 1077 patients with papillary thyroid cancer and 278 with follicular thyroid cancer, patients with tumors greater than 1.5 cm and no distant metastases had a 30-year recurrence rate of 26% and a cancer-related mortality rate of 6% when treated with total or near-total thyroidectomy,
From: Thyroid Cancer: A Comprehensive Guide toClinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totown, NJ
263
264
Burch Table 1 Poor Prognostic Factors for Differentiated Thvroid Cancer Patient age > 40 at diagnosis Large tumors (> 4.0 cm) Male sex Advanced tumor grade
Tumors with local extension Bilateral cervical or mediastinallymph node metastases* Distant metastases Extensive vascular and camular invasion? *Applies mainly to papillary thyroid cancer. ?Applies mainly to follicular thyroid cancer.
compared to rates of40% and 9%, respectively, for patients treated with less complet surgery (3). These same authors found that despite having more advanced disease, patients receiving postoperative radioiodine ablation for tumors greater than 1.5 cm 3) hadsignificantlylowerratesoftumor andnodistantmetastases(stages2and recurrence (16% versus 38%, P < .001) and cause-specific mortality (3% versus 9%, P = .03) than patients not receiving radioiodine ablation (3).Another study including 269 patients with papillary thyroid cancer followed for an average of 12 years found 1 cm in diameter had a lower incidence of that patients with tumors greater than recurrenceand death when a total or near-total thyroidectomy was performed (4). Likewise, this study determined that postoperative radioiodine ablation resulted in from thyroid cancer, although this was of marginal rates of recurrence and death statistical significance and limited to patients with tumors larger 1 than cm and confined to the thyroid, or metastatic only to cervical lymph nodes (4). Not all studies have supported the use of prophylactic radioiodine ablation following surgery for papilla 1500 papillarythyroidcancerpatients,no thyroid. In theMayoClinicreviewof difference in recurrence or cause-specific mortality was found between 946 patients treatedwithsurgeryaloneand220patientstreatedwithsurgeryplusradioiodine ablation (2). This disparity likely reflects the limitations imposed by the application retrospective data to judge treatment efficacy for thyroid cancer. Patients treated m aggressively are likely to have been deemed at a higherrisk for recurrence and death fromdisease. This confoundingeffectwouldtendtounderestimatethebenefitof treatment for known therapy. Conversely,the inclusion of patients receiving radioiodine residual disease in an analysis of remnant ablation would tend to overestimate the efficacy of this therapy.
EFFECT OF TUMOR SUBTYPE
Although papillary thyroid cancer as a whole has an excellent prognosis, it has recently become evident that certain rare subtypesthis ofdisease havea distinctly poor prognosis. These include thetall cell variant, the columnar variant, and insular patte (5).The follicular variant of papillary thyroid carcinomas as has recently been reviewed thyroid is a subtype having a microfollicular histological pattern but nuclear features and biological behavior similar to typical papillary thyroid cancer(6).
Papillay Thyroid Cancer Prognosis
265
REFERENCES 1. Ain KJ3. Papillary thyroid carcinoma:etiology, assessment, and therapy. Endocrinol Metab Clin North Am 1995; 24:711-760. 2. Hay ID. Papillary thyroid cancer. Endocrinol Metab Clin NorthAm 1990; 19:545-576. 3. Mazzafem EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer.Am J Med 1994; 97:418-428. 4. DeGroot W,Kaplan EL, McCormick M, et al: Natural history, treatment, and course of papillary thyroid carcinoma.J Clin Endocrinol Metab 1990; 71:414-424. 5. Buman KD, Ringel MD, Wartofsky L. Unusual types of thyroid neoplasms. Endocrinol Metab Clin North Am 1995; 25:49-68. 6. Rosai J, Zampi G, Carcangiu ML. Papillary carcinoma of the thyroid a discussion of its several morphologic expressions, with particular emphasis on the follicular variant.Am J Surg Path01 1983; 7:809-817.
26 Papillary Cancer Special Aspects in Children Merrily Poth
As stated in Chapter 10, the most common presentation for thyroid cancer in childr and adolescents is that of a solitary mass lesion, either in the thyroid itself or in the neck whereit represents a lymph node metastasis (1-6). The patientis usually otherwise 30-70% asymptomatic, but may be being followed for other thyroid disease. Although of patients have palpable lymph nodes at presentation (2-6), palpable cervical lymphadenopathy is common in children in general and does not necessarily imply that a thyroid nodule is malignant. Thyroid cancer in children has a high rateof spread at the time of diagnosis. Up to 90% of lesions which turn out at surgery to be papillary cancer, are already locally invasive and or have invaded local lymph nodes (2-9). Metastasis to lung at the time of diagnosis is also relatively common, with 6-20% of tumors havinglung metastases at the time of presentation. This number varies depending on the technique used to determine theorpresence absence of lung metastases. Chest radiographs and even whole-body scans after low doses l3II of are insensitive in detecting lung metastases. This is particularly true during the initial 1311 scan after surgery, when residual normal thyroid tissue is present and competing for uptake. Typically only half of the total number of patients with lung metastases noted on postablation scans are diagnosed with these less sensitive methods (3,8).In patients with only lymph node metastases at diagnosis, lung metastases may occur later(3-9). It is important to note that evenin the presence of lung metastases most children with Less than 10%of children thyroid cancer will eventually be able to be cured of disease. andadolescentswith thyroid cancer in most series die of their disease, even with extensive or recurrent disease (2-9). It should be noted that while spread to nodes, to local tissue, and to lungs is com and long-term prognosis is good, bone metastases are rare in children, occurring in less than1%in all reported series. When they do occur, they cany may a poor prognosis. EVALUATION The recommended process for evaluating a or child adolescent with an thyroid nodule has changed somewhat with the use of fine needle biopsy (FNB) and the availability From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by:L. Wmtofiky Q Humma Press Inc., Totowa, NI
267
268
Poth
of this technique for children. This technique, proven to be so effective in selecting patients for surgeryin adult populations, has now been instituted as a primary tool younger patients (10,11). The usefulness ofFNB in the evaluation of thyroid nodules in adults has clearly been shown by multiple authors. There are also well-developed guidelines for this procedure, the most stringent and perhaps appropriate are those He clearly articulates the need to acquire sufficient m detailed by Hamburger(12). and recommends the use of six separate aspirates in a given biopsy procedure as a technique for ensuring sufficient clinical material. The use of local anesthesia before aspiration and even the use of sedation in younger or more anxious patients enable this process. He and others also note the critical needfor experienced and competent cytopathologists to ensure that the pathological interpretation of materialis accurate. It is important to note that in adults with a lower incidence of malignancy in thyroid nodules, it may be easier to follow a nodule after negative or even an equivocal biopsy result. In younger children and adolescents, the much higher incidence of malignanc FNBresults will allow the endocrinologist means that only the most benign and definitive to follow the patient expectantly. More relevantly, the finding of clearly malignant cells will allow the opportunity for appropriate preoperative planning of a “cancer” operation and thoughtful counseling of parents and child before the operation(13). After radiation exposure there are often multiple thyroid lesions, which may be benign adenomas, multicentric cancers, or a mixture of these (14-16).Therefore it may be argued that the appearance of a thyroid nodule in a child with a history of significant radiationis sufficient to indicate the need for surgery, without any additio preoperative evaluation, including FNB. Diagnostic results from preoperative FNB in a child with a thyroid nodule may obviate the need for frozen-section examination of the tissue during the surg of the mass (lobectomy). Many surgeons feel that frozen sections add little to the da FNB (17). If the FNB yields cells consistent with “follic obtained with a preoperative neoplasm,” it is rare that the frozen-section evaluation will give more definitive re and most pediatric surgeons would rather remove the affected lobe and perhaps the isthmus, and then await the final pathology report prior to deciding whether to prothis issue remains under ceed with a completion subtotal thyroidectomy. However, debate (18). In contrast to the increasing use of FNB in the evaluation of thyroid nodules in children, the routine use of scanning procedures to characterize a nodule as “hot,” “warm,” or “cold” is decreasing. Whereas the thyroid scan was once a standard part of the initial evaluation of a thyroid nodule it has proven to be less useful and many practitioners no longer use thyroid scans as a routine part of the evaluation of thy primarily to rule out agenesis of the remaining t lesions. Those who do maysodo as a cause of the mass if this cannot be determined by palpation (19). There aretwo reasons for the perceived lack of value of scanning in children before is the useof F N B ,which, when available deciding whetherto proceed with biopsy. One yields much more valuable and specific information, as discussed above. In addition, the reported incidence in children of malignant lesions in thyroid nodules reported t be hot or warm is not insignificant(20-22).With the potential for malignancy in eve warm or hot lesions in children, the information regarding activity on thyroid scan is usually not useful in making further decisions about management.
Special Aspects in Children
2 69
It is important to emphasize that the effective diagnosis of thyroid cancer also entai the recognition that as many as 50% of thyroid cancer patients present with cervical adenopathy as their initial symptom (3,5,6). Several of our worst cases in terms of the extent of disease at diagnosis were referred only after relatively extended evaluation of what was felt to be cervical adenitis. This evaluation included several courses of antibiotic therapy and repeated examinations. It was only when eventual biopsy of the lymph node revealed thyroid cancer that the thyroid was carefully examined and the primary lesion identified. Thus education must be provided and emphasized to primary of the thyroid care providers and surgeons regarding the need for careful examination gland as part of the evaluation of cervical adenopathy.
APPROACH TO TREATMENT
The initial therapy of thyroid cancerin children, as in adults, involves surgery. The extent of surgery recommended does remain somewhat controversial, however. In the later discussion of follicular neoplasms of the thyroid in children (Chapter 33), there is further discussion of this issue. In the case of papillary cancer, which make up the vast majority of childhood thyroid cancer, the general recommendation for surgery includes a near-total or even total thyroidectomy, along with removal of local lymph this procedure is dependent on the experience of the nodes (23-29). The safety of surgeon. In the hands of an inexperienced surgeon there may be a high morbidity; however, in the hands of surgeons experienced with this operation in children, there is a very low incidence of complications and essentially no mortality (23,24). Thus it cannot be overemphasized that this procedure should only be performed by someone who has extensive and successful experience with the disease. In the past there has been controversy regarding the need to remove the contralateral lobe of the thyroid. Those that championed a more limited procedure than a subtotal or near-total thyroidectomy argued that since the more extreme operation carried a higher risk of complications and thyroid cancer in children had a good prognosis it was not worth the additional risk and that a simple lobectomy was sufficient for the initial operation (30). As noted above, in the hands of experienced surgeons, the near-total operation carr a low and probably acceptable risk, and careful analysis of the long-term follow-up of children and adolescents with thyroid cancer confirms that. Although greater 90% than of patients are eventually cured of their disease, this good prognosis applies only with the use of I3'I to treat residual or recurrent disease (28,29). After less than a near total thyroidectomy, the presence of large amounts of normal thyroid tissue makes the effectiv use of I3'I to eradicate thyroid cancer cells, with their less efficient iodine uptake me nism, difficultif not impossible. Scanning with I3lIto search for the presence of residual or recurrent tumor and the use of serum thyroglobulin measurements to follow the of residual normal thyroid tissue. of ablation therapy are also impeded by the presence Another justification for the use of a subtotal or near-total thyroidectomy is the frequent occurrence of multifocal diseasein patients with thyroid cancer;30% to 40% of children and adolescents with papillary thyroid cancer have multiple foci of disease (7,29). This percentage is probably larger in thyroid cancer occurring after radiation exposure.
270
Path
In summary, in spite of the obvious potential for an increase in surgical complic with more extensive surgery in children, the author believes that removal of the affected lobe and the isthmus with a subtotal resection of the contralateral lobeis the operation of choice for papillary thyroid cancer in children and adolescents. All suspicious lymp nodes should also be removed, but usually a more radical dissection to remove nodes that are not readily apparent is not indicated andmay add significantly to the operation Ensuring thatthis surgery is only performed by surgeons which extensive experien with the procedure results in a drastically decreased rate of both minor and major complications. In most centers at the present time with this policy the incidence of complications is well under 5% with the most common being mild temporary hypoca cemia (23-29).The incidenceof significant recurrent laryngeal nerve damage (not d to compression of the nerve by tumor before surgery) is low and the incidence of bilateral damage during surgery is uncommon. Even authors who seem to be arguing against the routine use of near-total thyroidectomy note excellent results and low rates of complications in first operations when the procedure is performed by more experienced surgeons (30). 1311
ABLATION
The second therapeutic modality for papillary thyroid cancer involves theofuse I3II for ablation of thyroid remnants and any malignant tissue remaining after the primary operation. There are advantages to this therapeutic approach, in addition to the destru tion of malignant cells, which are complementary to the ofusemore extensive thyroid allows for the effecti surgery. Destruction of all residual normal thyroid tissue1311 with use of surveillance methods, includingl3II scanning and serum thyroglobulin measure ments as markers for the presence of persistent or recurrent disease. Current recommendations are to administer an ablative dose of 1311to patients with differentiated thyroid cancer, usually4 to 6 weeks after surgery. Patients should be off thyroid hormone replacement during this time and should be on a reduced iodine diet for at least the last weeks before receiving I3*I.A serum TSH should be performed before the administration of I3lI, and, if sufficient thyroid was removed at surgery, it should be at least 30 pU/ml. This will ensure effective uptake of iodine into both normal residual tissue and malignant cells. There is some controversy about how much 13'1 should be givenfor this initial dose. A relatively small dose of just less than 30 mCi can be administered as an outpatient procedure, and will result in ablation of residual thyroid tissue93% in of patients who at the time of this dose have an uptake of less than 0.3% on follow-up scan and in Unfortunately, in patients with high whom this uptake is only in the thyroid bed(31). this low dose results in successfu percentages of uptake and/or with metastatic disease, 20 were ablation in only 59% of patients. In one study where four patients under initially treated with this low dose, only two of them were successfully ablated. In addition, some ofthe patients initially thought to be successfully ablated had later recurrences (31). Therefore it is probably preferable to give a larger doseI3lI, of usually 80-150 mCi, sufficient to ablate normal tissue in almost all patients. This higher dose also serves
Special Aspects in Children
271
as an initial treatment for metastatic disease, and may allow identification and localiza7 to 10 days tion of metastaticdisease in apostablationwholebodyscandone postablation (32). There are multiple areas of concern regarding potential long-term toxicity 1311 of in children. Becauseof the relatively smaller total body mass in most children and adoles for the treatment cents relative to adults, there has been some hesitancy in of the1311 use this age group. These include the of thyroid disease, both benign and malignant, in already alluded to increased sensitivity of the immature individual to effects of radiati and their relatively longer life expectancy. Overall most published data are relatively reassuring regarding toxicity. However, it is important to be familiar with potential long-term effectsof radioiodine therapy and to address them with the patient and fam in the context of the risks and benefits. Issues of particular concern include effects on bone marrow, effects on reproductive In addition, function in both males and females, and potential carcinogenic effects. there are the relatively acute toxic effects of 1311 of thyroiditis and inflammation of salivary glands, as well as acute nausea and vomiting (32,33). Salivary gland effects may be decreased by the use of careful attention to hydration and measures to increase salivary flow during the exposure. Acute suppression of bone marrow can be followed and hematologic parameters are usually normalizedby 60 days. Untoward long-term consequences of this level of acutebonemarrowsuppressionareextremelyrareandthereforeshouldnotbea cause of concern. However, it is important to allow recovery of bone marrow before retreatment with 1311to ensure that full recovery will take place. Current recommenda6 monthsapart.Therearereported tions are fortreatmentstobespacedatleast incidences of leukemia after multiple high doses of 1311 given over a short period of time (32). Other potential long-term toxic effects of 13'1 include toxicity to the gonads with azoospermia (33). Effects on ovarian function and effects on future offspring and on fertility after I3lItreatment of female children and adolescents have not been shown to occur. The increase in miscarriage rate within the year after large dose of1311has led totherecommendationtoavoidconceptionduringtheyearafter 1311 (34).Males receiving 1311 for treatment of thyroid cancer have been reported to have an acute decrease in spermatogenesis as well as increases in FSH. After acute doses the values tend to return to normal. However, with high cumulative doses FSHmay values remain elevated and fertility may be affected (35). This phenomenon should be explained to patients receiving doses in this range, and consideration should be given to freezing 1311therapy in adolesand storage of sperm before the institution of repeated high-dose cents and young adult males who desire subsequent fertility. Data on the vulnerability of testicular function to the toxic effects of radiation relati to the process of sexual maturation seem to indicate that more mature testicles are more vulnerable to toxic effects and thus testicular effects may not be a major source of concern when treating prepubertal children. There are reports of breast cancerin young women who were treated during adolescence with 13'1 for thyroid cancer (36). Other malignancies have been reported after high dose 1311 therapy, including cancer of the bladder and leukemia (32). Vigorous
272
Poth
hydration and frequent urination and emptying of the bowel are recommended after high doses of 13’1 to decrease colon, bladder, and total body exposure. There is an increased riskof subsequent development of thyroid nodules in patients 1311 for toxic nodules or Graves’ disease. This does not treated with small doses of generally apply to the doses used for the treatmentof malignant disease (greater than 20 mCi), particularly since there is usually complete or almost complete thyroidectom before treatment with 1311. There are no reports of subsequent increases in thyroid disease after I3lI treatment for thyroid cancer. THYROID HORMONE REPLACEMENT
Irrespective of whether I3II ablation was performed, patients ultimately need to be placed on thyroid hormone therapy. The agent of choice is Gthyroxine and the only issue becomes whether the LT, dosageis for “replacement” (associated with a measur able serum TSH) or “suppressive” (associated with an immeasurably low TSH). The choice often relates to the nature of prognostic factors associated with the original tumor, such as size, stage, and grade. The theory behind suppressive dosage therapy includes the fact that TSH is a theoretical “cocarcinogen” and is believed to increase of TSH to the growth of microscopic foci of thyroid cancer cells. Thus suppression below the lower limit of deduction of the modern assays for TSH is the standard of care for patients with thyroid cancer. The smallest dose of thyroxine able to ac this should be used, and freeT4 and TSH shouldbe regularly monitored to ensure that TSH suppression without needless hyperthyroidism is accomplished. Hyperthyroidism resulting from the relatively high levels of free T4 needed for TSH (37,38). suppression may result in potential negative effects on bone density long-term This is particularly problematic because children and adolescents treated for thyroid cancer will be placed on these suppressive doses of thyroxine during the time when they are still in the process of acquiring their peak bone mass. Long-term studies are needed on the effects on bone densityof doses of thyroxine high enough to suppress TSH, and on potential interventions to prevent bone loss. However, at the present ti the recommendation is for treatment of children and adolescents with a history of thyroid cancer with the minimum dose of thyroxine sufficient to suppress the serum level of TSH. PROGNOSTIC FACTORS
At the time of diagnosis of thyroid cancer in children the vast majority of children 40-90% of children atinitial surgical already have disease in local lymph nodes. Indeed, diagnosis will have local spread of papillary cancer to lymph nodes, compared to 20% of adult patients (1-9,39,40). The incidence of metastatic disease at diagnosis is also relatively high, with 10-20% of children and adolescents showing spread of the disease to lung at diagnosis, compared to5%only in adults. A small number of childre less than 2% in all published series, have metastases to bone at the time of diagnosi It is important to be familiar with the relatively high incidence of disease spread is so at diagnosis in younger patients. Even though metastatic disease at diagnosis common in children, with appropriate surveillance and treatment for metastatic and
Special Aspects in Children
2 73
recurrent disease the prognosis remains excellent. Less than 10% of children die of their disease. After initial therapy with surgery and I3'I, 1540% of patients may be expected to have recurrent disease (6-9,2890-42). Some authors make a point of the fact that recurrence is higher in younger children than in older children and adolescents(42). In spite of the common finding of extensive disease at the time of diagnosis and the increased incidence of recurrent disease after treatment in children, the long-term prognosis for children with thyroid cancer is better than that of adults. In one large study of patients with thyroid cancer, of140 patients under20 at diagnosis, there were 48 recurrences after treatment (37%) but only 5 patients died of the disease. They an approximately 20% compared this to the group of older patients who had only recurrence rate, but a death rate of 13% (41). The disease-free survival of children with papillary thyroid cancer is reported in various studies to be 80-90% and the long term mortality from PTC to be less than 10% in multiple series(2-9,3942). Because of these excellent statistics there are still controversies as to the appropriateness of aggressive therapy in the child with this disease. There is a school of thought that states that with the excellent prognosis even after recurrent disease, there is no need for aggressive therapy. Some would even advocate less than aggressive thyroid resection, perhaps only removal of the affected lobe and the isthmus, to be followed only by observation for recurrent diseaseif there is no obvious or palpable nodal involvement. Since the incidence of local spread to lymph nodes is well over 50% it would appear to be prudent to at least perform a resection of those nodes which may be easily identified. However, it should be noted that all studies to date on the effects of therapy, inclu the extent of surgery and the use of 13'1 and even the need for suppressive doses of thyroxine after treatment, are retrospective. The data are therefore not really conclusiv in regard to sorting out the effects of specific treatment modalities on the long-term outcome of the disease. While optimism is clearly justified, as the vast majority of patients attain long-term disease-free survival, the current treatment modalities are In addition, the good outcomes long-term nonetheless onerous and carry some morbidity. are predicated on both careful monitoring for recurrent disease and appropriate trea It is hoped that current studies characterizing thyroid cancers using molecular bi cal techniques will lead to better predictive ability. If it were possible to better determine the likelihood of a given tumor behaving aggressively at the time of initial diagnosis, then therapy might be able to be better tailored to treat the individual tumor patient while minimizing morbidity. With the present state of knowledge, it is recommended that all papillary thyroid tumors in children and adolescents be treated and followed aggressively. The longterm follow-up of these patients may be complicated by transition of the care of these patients from pediatric providers to internists. The possibility of recurrent disease many years after apparently negative evaluations mandates that patients once diagnosed with 20 years. Reports of the thyroid cancer should be followed expectantly for at least recurrence of disease in some patients during pregnancy after many yearsof negative evaluations (43) suggests that pregnancy is a period of time with a higher than usual rate of recurrence.
274
Poth
In summary, even thoughthe long-term prognosis is excellentin children and adoles-
cents with papillary thyroid cancer, it is important to ensure that all patients continue to be followed closely for extended periods of time, and perhaps for life. Recurrences may occur years after the initial diagnosis and treatment, even when appropriately treated and d e r what appears to be disease-free survival. More specific recommendations regarding prognosis and need for more or less aggressive treatment may evolve are available.Atthe as studies to characterizetumorsusingmoleculartechniques present time the recommendation for therapy include at least subtotal thyroidectomy, of thyroxine.Serumthyroglobulin followedby I3'I ablationandsuppressivedoses measurements (44) and follow-up 13'1scans to monitor for recurrent or persistent disease and repeat treatment of disease with I3'I should resultin an excellent long-term prognos in the vast majority of cases.
REFERENCES 1. Jocham A, JoppichI, Hecker W, Knorr D, Schwartz HP. Thyroid carcinoma in childhood: Management and follow up of 11 cases. Eur J Pediatr 1994; 153:17-22. 2. Samuel A M , Sharma SM. Differentiated thyroid carcinomas in children and adolescents. Cancer 1991; 67:2186-2190, 3. Ceccarelli C, Pacini F, LippiF, Elisei R, Arganni M, Miccoli, Pinchera A. Thyroid can in children and adolescents. Surgery 1988; 104:1143-1148. 4. Viswanathan K, Gierlowski TC, Schneider AB. Childhood thyroid cancer: characteristics and long-term outcome in children irradiated for benign conditionsof the head and neck. Arch Pediatr Adolesc Med 1994; 148:260-263. 5. Harness JK, ThompsonN W , McLeod MK, PasiekaJL, Fukuuchi A. Differentiated thyroid carcinoma in children and adolescents. World J Surg 1992; 16:47-54. 6. Schlumberger M, De VathaireF, Travagli JP, Vassal G, LemerleJ, Parmentier C, Tubiana M. Differentiated thyroid carcinomain childhood: long term follow-up 72 ofpatients. J Clin Endocr Metab 1987; 651088-1094. 7. Welch-Dinauer CA, TuttleRM, Robie DK, McClellan DR, SvecRL,, Adair C, FrancisGL. Clinical features associated with metastasis and recurrence of differentiated thyroid c in children, adolescents and young adults. Clin Endocrinol 1998; 49519-28. 8. Fassina AS, Rupolo M, Pelizzo MR, Casara D. Thyroid cancer in children and adoles Tumori 1994; 80:257-262. 9. Lamberg BA, Karkinen-Jaaskelainen M, Franssila KO. Differentiated follicle-derived thyroid carcinoma in children. Acta Pediatr Scand 1989; 78:419425. 10. Degnan BM, McClellan DR, Francis GL. An analysis of fine-needle aspiration biopsy of the thyroid in children and adolescents. J Pediatr Surg 1996; 31:903-907. 11. Raab SS, Silverman JF, Elsheikh TM, Thomas PA, Wakely PK. Pediatric thyroid nodules: diseasedemigraphicsandclinicalmanagement as determinedbyfineneedleaspiration biopsy. Pediatrics 1995; 95:46-49. 12. Hamburger JI. Diagnosis of thyroid nodules by fine needle biopsy: use and abuse. J Clin Endocrinol Metab 1994; 79:335-338. 13. De Keyser LF, Van Herle AJ. Differentiated thyroid cancer in children. Head Neck Surg
1985; 8:lOO-114. 14. Shore RE, Hildreth N, Dvoretsky P, Andresen E, Moseson M, Pasternack B. Thyroid c amongpersonsgivenx-raytreatmentininfancyforanenlargedthymusgland.AmJ Epidemiol 1993; 137:1068-1080. 15. Nikiforov YE, Gnepp DR, Fagin JA. Thyroid lesions in children and adolescents after the Chernobyl disaster: implications for the studyof radiation tumorigenesis.J Clin Endocrinol Metab 1996; 81:9-14.
Special Aspects in Children
275
16. NikiforovYE, GneppDR. Pediatric thyroid cancer after the Chernobyl disaster: pathomorp logic study of 84 cases (1991-1992) fromthe Republicof Belarus. Cancer 1994; 74:748-766. 17. RodriguezJM, Panilla P, SolaJ, Bas A, Moreno A, Soria T. Comparison between preoperative cytology and intraoperative frozen-section biopsy in the diagnosis of thyroid nodules. Br J Surg 1994; 81:1151-1154. 18. Gibb GK, PasiekJL. Assessing the need for frozen sections: still a valuable tool in thyroid surgery. Surgery 1995; 118:1005-1010. 19. De Roy van Zuidewijn DB, Songun I, Hamming J, Kievit J, van de Velde CJH,Veselic M. Preoperative diagnostic tests for operable thyroid disease. World J Surg 1994; 18:506-510. 20. Croom RD III, Thomas CG Jr, ReddickRL, Tawil MT. Autonomously functioning thyroid nodules in childhood and adolescence. Surgery 1987; 102:llOl-1108. FM, Foley Tp Jr. Functioning thyroid masses in childhood 21. Hopwood NJ, Carroll RG, Kenny and adolescence. J Pediatr 1976; 89:710-718. 22. Smith M, McHenry C, Jarosz H, Lawrence AM, Paloyan E. Carcinoma of the thyroid in patients with autonomous nodules. Amer Surgeon 1988; 54:44849. 23. Farahati J, Bucsky P, Parlowsky T, Mader U, Reiners C. Characteristics of differentiated thyroid carcinoma in children and adolescents with respect to age, gender, and histology. Cancer 1997; 80(11):2156-62. 24. Stael A P , Plukker JT, Piers DA, Rooouwe C W , Vermey A. Total thyroidectomy in the treatment of thyroid carcinoma in childhood. Br J Surg 1995; 82:1083-1085. 25. Patwardhan N, Cataldo T, Braverman LE. Surgical managementof the patient with papillary cancer. Surg Clin North Am 1995; 75:449464. 26.Shindo ML. Considerationsinsurgeryofthethyroidgland.OtolarynClinNorthAm 1996; 29:629-635. 27. Vassilopoulou-Sellin R, Goepfert H, Raney B, Schultz PN. Differentiated thyroid cancer in children and adolescents: clinical outcome and mortality after long-term follow-up. Head & Neck 1998; 20(6):549-55. 28. FrankenthalerR 4 , Sellin RV, Cangir A, Goepfert H. Lymph node metastasis from papillary follicular thyroid carcinoma in young patients.Am J Surg 1990; 160:341-343. 29. Massimino M, Gasparini M, BalleriniE, Del Bo R. Primary thyroid carcinoma in children: a retrospective studyof 20 patients. Med Pediatr Oncol 1995;24:13-17. 30. Cohn KH, Backdahl M, Forsslund G, Auer G, Zetterberg Am Kundell G, et al. Biologic considerations and operative strategy in papillary thyroid carcinoma: arguments against the routine performance of total thyroidectomy. Surgery 1984; 96:957-971. 31. Van Wyngaarden M, McDougall IR. What is the roleof 1lOOMBq (c30 mCi) radioiodine 1311 in the treatment of patients with differentiated thyroid cancer? Nucl Med Commun 1996; 17:199-207. 32. Maxon HR III, Smith HS. Radioiodine-131 in the diagnosis and treatment of metastatic Am 1990; 19:685-715. well differentiated thyroid cancer. Endocrinol Metab Clin North 33.Eklmonds CJ, Smith T. The long-term hazards of the treatment of thyroid cancer with radioiodine. Br J Radio1 1986; 59:45-51. 34. Casara D, Rubello D, Piotto A, Pelizzo MR, Girelli ME, Busnardo. Pregnancy after high therapeutic dosesof iodone-131 in differentiated thyroid cancer: potential risks and recommendations. E m J Nucl Med 1993; 20:192-194. R, et al. Testicular 35.Pacini F, Gasperi M, Fugazzola L, Ceccarelli C, Lippi F, Centoni function in patients with differentiated thyroid carcinoma treated with radioiodine. J Nucl Med 1994; 35:1418-1422. 36. Green DM, Edge SB, Penetrante RB, Bakshi S , Shedd D, Zevon MA. In situ breast carcinoma treatmentduringadolescenceforthyroidcancerwithradioiodine. Med PediatrOncol 1995; 24182-86. 37. Solomon BL, WartofskyL, Burman KD. Prevalence of fractures in postmenopausal women with thyroid disease. Thyroid 1993; 3:17-23.
276
Poth
38.Radetti G, Castellan C, Tat6 L, Platter K, Gentili L, Adami S. Bonemineraldensity of L-thyroxine.HormRes in children and adolescent females treated with high doses 1993; 39~127-131. 39. Feinmesser R, LubinE, Segal K, Noyek A. Carcinoma of the thyroid in children-a review. Journal of Pediatric Endocrinology & Metabolism 1997; 10(6):561-8. 40. Zimmerman D, Jay ID, Gough IR, Goellner JR, Ryan JJ, Grant CS, McConahey WM. Papillary thyroid carcinoma in children and adults: long-term follow-upof 1039 patients conservatively treated at one institution during three decades. Surgery 1157-1 1988; 166. 104: 41. Samaan NA, Schultz PN, HickeyRC, Goepfert H, HaynieTP, Johnston DA, Ordonez NG. The results of various modalities of treatment of well differentiated thyroid carcinoma: a retrospective review of 1599 patients. J Clin Endocrinol Metab 1992; 75714-720. 42. Travagli P,Schlumberger M, De Vatharie F, Francese C, Parmentier C. Differentiated thyroid carcinoma in childhood. J Endocrinol Invest 1995; 18:161-164. 43. Merrick Y and Hansen HS. Thyroid cancer in children and adolescents in Denmark. Eur J Surg Oncol 1989; 15:49-53. 44. Kirk JMW, Mort C, GrantDB, Touzel RJ, Plowman N. The usefulnessof serum thyroglobu lin in the follow-up of differentiated thyroid carcinoma in children. Med Pediatr Oncol 1992; 20:201-208.
IT7 Differentiated Tumors of the Thyroid Gland B. Follicular Carcinoma
27 Follicular Thyroid Carcinoma Clinical Aspects Leonard Wartofsky
CLINICAL PRESENTATION
Follicular carcinoma may typically present as a single, painless thyroid nodule in an older (>55 years of age) male, although it is more common in women by twofold or more. Lymphadenopathy due to involved cervical nodes is uncommon, but distant metastases will be presentin lung or bone in10-20% of patients at the time of initial presentation (15-19). At presentation, all routine blood thyroid function tests will be within normal limits, including the serum TSH (except in the presence of severe iodine deficiencyandendemicgoiter).Serumthyroglobulin(Tg)maybeelevated,buta diagnosis should not be inferred from serum Tg levels. Utility of Tg monitoring is this volume (Chapters22 and 3l), and maybe adversely affected discussed elsewhere in by the presence of interfering antithyroglobulin antibodies, which usually falsely lower serum Tg levels.This may be more problematic with immunoluminometric assays than with highly specific radioimmunoassays for thyroglobulin (20). Unfortunately, as many as 25-35% of thyroid cancer patients may have underlying Hashimoto's disease with positive thyroid autoantibodies. Future management of such patients may be facilitated by measurement of thyroglobulin mRNA in serum (21). Other techniques may allow distinction between circulating thyroglobulin derived from benign versus malignant thyroid tissue (22). Patients with known metastatic or residual thyroid cancer should be followed up by an endocrinologistlthyroid specialist in addition to their primary care physician. The physician should ensure that serum Tg is being measured only in this should be in the same laboratory at a laboratory of the highest quality. Ideally, each follow-up time interval, and the laboratory should provide companion Tg levels In the postoperative on a remeasurement of stored serum from the prior venapuncture. state, a clearly measurableor rising serum Tg while the patientis TSH-suppressed on levothyroxine may be a definite clue to recurrence, but serum Tg levels are usually mostusefulwhenmeasured while the patient is hypothyroid, for example, during preparation for follow-up scanning. The current availability of recombinant human TSH has facilitated monitoring Tg before and after rhTSH stimulation (see 14). Chapter
From: Thyroid Cancer:A Comprehensive Guideto Clinical Management Edited by: L. Wnrtofiky 0 Humann Press Inc., Totown, NI
279
280
Wartofiky
The problem presented by patients with negative radioiodine scan surveys but ele serum thyroglobulin is discussed in Chapter 24. In regard to nuclear medicine examinations, the radioiodine uptake will be normal, and an isotopic scan will disclose a “cold” nodule corresponding to the palpable lesion In a multinodular gland, there may be other “hot” or autonomously hyperfunctional nodules, but malignancy in the latter is very rare. Thyroid ultrasound imaging is useful to confirm the presenceof the nodule or nodules detected on physical examination, to size the nodule(s), and to determine if there is multinodularity. Ultrasonography canno reliably distinguish between benign and malignant lesions, although a purely cystic (anechoic) lesion only rarely harbors a malignancy, and a clear area around a solid (echogenic) lesionmay represent an intact capsule (“halo” sign) suggesting benign While the diagnostic sine qua non for evaluation of thyroid nodules is usually the cytological examination derived from a fine needle aspiration of the nodule, this technique also does not reliably distinguish between benign and malignant follicular neo plasms (see below and Chapters 3 and 28). A number of investigators are currently exploring different potential tumor markers and genetic analysis to distinguish benig frommalignantlesions (23,24). In one recent report (141, estimatesoftelomerase activity appeared to show some promise in this regard. Follicular carcinoma tends to occur in endemic (iodide-deficiency) goiters and in TSH preexisting adenomatous goiters, and either iodide deficiency or the secondary stimulation associated with it appear to be etiologically related to the development of these tumors (26). In one recent series from Norway, decreased risk was associated with use of iodized salt and increased risk was shown in areas of endemic goiter(27). In the western hemisphere in areas of iodide sufficiency, papillary carcinoma seems to be more common. Thus, while the total number of thyroid cancers does not ap be significantly increased, a smaller proportion of theis due total to follicular carcinom In addition to iodide deficiency and endemic goiter, there are a number of other possible predisposing factors for follicular thyroid cancer. These include advancing age, female sex, and radiation exposure to the head and neck. The greater frequency in women and the somewhat increased presentation of thyroid cancer during preg implies an association with higher endogenous estrogen levels. The high levels of that occur in early pregnancy could be another etiological or permissive factor because TSH receptor and can constitute a stimulus to both hormone pro hCG binds to the and thyroid hypertrophy. Nevertheless, pregnancy does not appear to have an adverse impact on ultimate outcome (28). A decline in the reported frequency of pure follicular carcinomaof the thyroid may be due also in part to more rigid pathological diagnostic criteria. In many hospitals, follicular tumors are more often misdiagnosed (i.e., false-positive diagnoses) due to confusion with other lesions such as benign follicular adenomas, adenomatoid goiter, or the follicular variants of either papillary or medullary carcinoma (29). The not-uncommon occurrence of follicular thyroid cancerin an adenomatous goite has suggesteda pathogenetic evolution of these cancers from lesions that were o benign. Both follicular adenomas and carcinomas appear to be of monoclonal origin, and evolution of an adenoma into a malignant lesion could occur via mutational or translocational activation of oncogenes, particularly therus oncogene which has been specifically identifiedin follicular tumors (30,31). Evolution of adenomainto carcinom
Follicular Thyroid Carcinoma
281
could occur through genetic loss of tumor suppressor genes, which, taken together with the rus oncogene activation, would lead to clonal expansion and growth of a malignant subclone of cells. Certain cytogenetic alterations have been linked to more aggressive tumor behavior (32). Much current investigation also is focusing on mutations of the TSH receptor in a variety of thyroid disorders, and there well may be mutations of either the TSH receptor or the a subunit of the stimulatory G protein, which could also lead to tumorigenesis (31,33). A patient with an activating mutation of the TSH receptor in a tumor metastatic to lungs and lymph nodes sufficient to cause thyro has been described (34). CLINICAL ISSUES RELATED TO PATHOLOGY AND CYTOLOGY
There are two major types of true follicular carcinoma: a minimally invasive enc lated type and a more aggressive and invasive form, which often presents with distant metastases. Histologically, the encapsulated type may closely resemble a benign follicular adenoma. (See Chapter 28 for a full description of the pathological features.) On rare occasions, even a small, well-encapsulated follicular carcinoma may present with distant metastases in bone or lung, presumably due to poorly understood differences in host factors. do notdemonstrateanyofthetypical or Pathologically,follicularcancercells pathognomonic features of papillary cancer such as crowded, overlapping cells with nuclear cleftsor grooves, and large intranuclear clear inclusions (“Orphan Annie eyes”). Invaginations of cytoplasm into the nuclei,which are also common in papillary cancer, can be seen rarely in follicular carcinoma. These tumors are usually composed of follicular elements but not the papillated structures typical of papillary thyroid carcinoma. In contrast to papillary thyroid cancer, both lymph node invasion and thyroidal multicentricity are uncommon. Possible confusion with a papillary cancer would occur in those tumors with the greatest degree of potential overlapping characteristics, that is, the “follicular variant” of papillary carcinoma (35,361.However, these tumors will exhibit a biological behavior much more similar to papillary than to follicular carc including a pattern of metastasis to regional nodes rather than hematogenous spread to distant sites, as well as a better prognosis. Because ofthe very close similarity of the appearance of the follicular cells in benign adenomata and those of follicular carcinoma, it is not usually possible to distinguish between the two by cytological examination after fine needle aspiration. Some cytopathologists may detect a greater degree of nuclear “atypia” and an increased rate of so difficult as to be treacherous. mitoses in malignant lesions, but the distinction may be The diagnosis is based instead on histological rather than cytologic criteria, including evidence for either capsular or vascular invasion. While cytological differentiation of follicular adenoma from carcinomais difficult, frozen-section analysis is even worse in (37) most but not all (38)hands. An experienced pathologist should haveno problem in making the diagnosisin the aggressive invasive type of follicular carcinoma, which will often present with distant metastases to bone or lung. While local lymph node invasion can be discovered at thyroidectomy in pe 5-10% of follicular cancers, it is much less common than in the papillary variety, which may present with involved cervical nodes in3545% of patients. In addition to
282
Wartofiky
the usually higher frequencyof metastases to bone (spine, skull, pelvis) and the lun other less common sites for distant metastases include the brain, and rarely the liver The initial evaluation of a patient proven to have a follicular thyroid carcinoma should include a chest radiograph, but even if interpreted as negative, the lungs may *3lI still demonstrate radionuclide uptake indicating metastases on the postoperative Scan. Should pulmonary metastases be evident, a chest computed tomography (CT) scan will provide excellent anatomical imaging as a baseline study for future comparisons after therapy. Pulmonary metastases may be particularly difficult to fully e as has been seen in children and young adults (39d1). The mainstay of follow-up, (l3II)scans. however, will be periodic serial monitoring of serum Tg levels and isotopic The frequency of both will depend on the staging of the tumor and the specific cl circumstances, for example, low- vs high-risk,of each patient, andis discussed below in Chapters 12,13, 23, and 31. Recurrences of cancer after initial thyroidectomy and radioiodine ablation and therapy are most likely to occur within the first 18 to 60 months. Patients with known tumor based on rising serum levels of thyroglobulin bu negativeradioiodinescansmaybestudiedwithotherimagingtechniquessuchas 2 0 1 T l , sestamibi, or FDG-glucose (see Chapter 24). Hiirthle cell variants often do not concentrate radioiodine. Tumor stagingis important to establish prognosis but staging methods remain s what controversial (6,42,43). Of the several staging systems for thyroid carcinoma in use, the TNM system remains the most widely applied (44,45). This system allows determination of the category of relative risk, for example, low vs high, and provides prognostic indicators for recurrence and death from the tumor. Prognostic indicators have been incorporated into systems which consist of a scale of risk. The “AGES” system, devised and advocated bythe Mayo Clinic(46), incorporates the risks contrib uted by patient _age, tumor grade, gxtracapsular invasion, and tumor &e. Such syste provide useful parameters upon which to discuss prognosis with patients in reaso precise terms, given the wide variability and uncertainty underlying prognosis of an malignancy. In the review of differing staging systems by DeGroot and colleagues (47), the TNM system was felt to best provide risk stratification, at leastfor papillary thyroid cancer. TheHiirthlecellvariantoffollicularcarcinoma is atumorcomposedoflarge acidophilicoroncocyticcells (14,48). Likeotherfollicularneoplasms, it is more common in women than men, but the patients tend to be older than those with thyroid carcinoma. When patients are stratified as to low versus high risk, there does not appear to be a significantly poor prognosis for Hiirthle cell than for follicular carcinoma (49; see Chapter 32). Htirthle cells are present to a lesser extent in many benign thyroid disorders, particularly in Hashimoto’s thyroiditis. Hiirthle cell neop may be benign or malignant, the distinction being based upon demonstration of or capsular invasion, metastatic capacity,and rate of growth, just as in other follicula neoplasms (50). In contrast to other follicular carcinomas, they have a higher rate of bilaterality or multicentricity. One recent report suggested that the distinction cou 4 cm being inferred from the size of the lesion, those tumors that are larger than invariably malignant (51). The Hiirthle cell malignancy tends to have a worse prognosis than other follicula tumors in many (52,53) but not all series (54). This may be due in part to their greate
noma
Follicular Thyroid
283
tendency to be locally invasive, and their propensity to concentrate radioiodine less avidly, thereby rendering them more difficult to manage with isotopic scanning and to the management of follicular therapy (55). Recent studies have explored approaches cancers that have lost their ability to either trap iodide (and be treated with radioiodin or to synthesize and release thyroglobulin. The ability of both normal and malignant thyroid cells to concentrate iodide is dependent upon expression of a Na+/I- symporter gene (56). The loss of this gene during tumor dedifferentiation can account for the tumor's failure to concentrate iodide. Redifferentiation therapy with retinoic acid holds promise for the restoration of both radioiodine uptake and thyroglobulin production (57-59). Growth of Hiirthlecell tumors has been described to reflect of theproliferanet tive versus apoptotic indices and these characteristics may distinguish benign from malignant lesions (60). Hiirthle cell cancer is discussed at length below in Chapters 48 and 49. These and other follicular cancers that do not concentrate radioiodine may 15 and 20) or by external radiation therapy be treated with chemotherapy (see Chapters (61,62) (seeChapter 30), or redifferentiationtherapywithretinoicacidcouldbe attempted. However, although external radiation therapy may cause apparent tumor regression, the effect may be transitory with little improvement in survival rate (63).
SUMMARY While most primary thyroidal cancers arise from follicular epithelium, the most common type of thyroid malignancy is the well-differentiated papillary carcinoma, which accounts for about 70-80% of thyroid tumors, with true follicular carcinomas accounting for only 5-15% of all thyroid cancers. Follicular cancer is more common in older patients and spreads by blood vessel invasion, often presenting with metastase in lungs or bone. The mass of functioning metastatic cancer may rarely beso great as to cause thyrotoxicosis. Papillary cancer occurs more commonly in younger patients, is slowly growing and less aggressive with a more favorable prognosis. Certain characteristics are associated with a worse prognosis with follicular tumors, including a more male sex, and larger size, especially highly invasive or metastatic tendency, age 50, above with lesions larger than 4 cm diameter. Unlike papillary carcinoma, follicular carcinoma is a much less likely tumor to occur as a result of prior radiation exposure to the head and neck. The management of follicular thyroid cancer differs from that of papillary carcinoma in one important way, and that relates to the requirement for early operative ( I ) . However, one management to consist of a total rather than subtotal thyroidectomy of surgery review of 82 patients with follicular thyroid carcinoma found that the extent did not affect the rate of disease-free survival which was more directly related to radioiodine therapy (2). Thyroidectomy is then followed by radioiodine ablation of any remnant tissue as a prerequisite to the potential need to more effectively treat distant metastases in lung or bone with radioiodine by removing all residual thyroid tissue that might compete for radioiodine (3,4). Someworkershaveadvocatedradioiodineablationoflarger remnants as an alternative to the risks inherent with a completion thyroidectomy (5). Patients with follicular thyroid carcinoma are more likely to have advanced disease (Stage IIIor IV)at presentation, placing them at higher risk than patients with papillar thyroid cancer (6). The 10-yr overall survival rate for follicular cancer is 85% (7),
284
Wartofsky
whereas papillary cancers tend to have a more favorable prognosis with a 93% 10-yr survival (7),especially if less than 1.5 cm in size, and may not require aggressive management withradioiodine. Follow-up evaluations with 13'1scanning require allowing TSH to rise following discontinuation of L-thyroxine therapy, and monitoring serum thyroglobulin as a tumor marker for recurrence. Several authoritative review articles have appeared in recent years (8-14).
REFERENCES
1. Schwartz AE, Clark OH, Ituarte P, LoGerfo P. Thyroid surgery-the choice. JClin Endocrino1 Metab 1998; 83:1097-1105. 2. Taylor T, Specker B, Robbins J, Sperling M, Ho M, Ain K. Outcome after treatment of Ann Intern Med high-risk papillary and non-Htirthle cell follicular thyroid carcinoma. 1998; 129:622-627. 3. Lin JD, Kao PF, Chao TC. The effects of radioactive iodine in thyroid remnant ablation and treatment of well differentiated thyroid carcinoma. Brit J Radio1 1998; 71:307-313. 4. Pelikan DM, Lion HL, Hermans J, Goslings BM. The role of radioactive iodine in the Clin 47:713-720. treatment of advanced differentiated thyroid carcinoma. Endocrinoll997; 5. Lin JD, Chao TC, Huang MJ, Weng HF, Tzen KY. Use of radioactive iodine for thyroid remnant ablation in well-differentiated thyroid carcinoma to replace thyroid reoperation. Amer J Clin Oncol 1998; 21:77-81. 6. Sherman SI, Brierley JD, Sperling M, Ain KB, Bigos ST, Cooper DS, et al. Prospective multicenter study of thyroid carcinoma treatment: Initial analysis of staging and outcome. Cancer 1998; 83~1012-1021. 7. Hundahl SA, Fleming ID,Fremgen AM, Menck HR. A national cancer data base report on 53,856 cases of thyroid carcinomatreated in the U.S., 1985-1995. Cancer 1998 2648. 8. Grebe SKG, Hay ID. Follicular thyroid cancer. Endocrinol Metab Clin North Am 1995; 24~761-801. 9. Goldman N D , Coniglio JU,Falk SA. Thyroid cancers I Papillary, follicular, and Htirthle cell. Otolaryngol Clin North Am 1996; 29:593-609. 10. Dulgeroff A J , Hershman JM. Medical therapy for differentiated thyroid carcinoma. Endo Rev 1994; 15500-515. 11. Emerick GT, Duh Q-Y, Siperstein AE, Burrow GN, Clark OH. Diagnosis, treatment, and outcome of follicular thyroid carcinoma. Cancer 1993; 72:3287-3295. 12. Robbins J, Merino MJ, Boice JD, Ron E, Ain KB, Alexander HR, et al. Thyroid cancer: a lethal endocrine neoplasm. Ann Intern Med 1991; 115133-147. 13. Schlumberger MJ. Papillaryandfollicularthyroidcarcinoma. N Engl J Med1998; 338~297-306. 14. Cooper DS, Schneyer CR. Follicular and Hiirthle cell carcinoma of the thyroid. Endocr Metab Clin North Am 1990; 19:577-591. 15. Jensen MH, Davis RK, Derrick L. Thyroid cancer: a computer-assisted review of 5287 cases. Otolaryngol Head Neck Surg 1990; 102:51-65. 16. Ruegemer JJ,Hay ID, Bergstralh ET, Ryan JJ,Offord KP, Gorman CA. Distant metastases in differentiated thyroid carcinoma: a multivariate analysis of prognostic variables. J Clin Endocrinol Metab 1988; 67:501-508. 17. Schlumberger M, Tubiana M, de Vathaire F,H ill C, Gardet P, Travagli JP, et al. Longterm resultsof treatment of 283 patients with lung and bone metastases from differen thyroid carcinoma. J Clin Endocrinol Metab 1986; 960-966.
Follicular
285
18. Young RL, Mazzafem EL, Rahe AJ, Dorfman SG. Pure follicular carcinoma: impact of therapy in 214 patients. J Nucl Med 1980; 21:733-737. 19. Simpson WJ, McKinney SE, Carruthers JS, Gospodarowicz MK, Sutliffe SB, Panzanella T. Papillary and follicular thyroid cancer: prognostic factors in 1578 patients. Am J Med 1987; 83:479438. RB, Singer PA, et al. Serum 20. Spencer CA, Takeuchi M, Kazaroxyan M, Wang CC, Guttler thyroglobulin autoantibodies: prevalence, influence on serum thyroglobulin measurement, and prognostic significance in patients with differentiated thyroid carcinoma. J Clin Endoc no1 Metab 1998; 83:1121-1127. 21. Ringel MD, Ladenson PW, Levine MA. Molecular diagnosis of residual and recurrent thyroid cancer by amplification of thyroglobulin messenger ribonucleic acid in peripheral blood. J Clin Endocrinol Metab 1998; 83:4435-4442. Editorial comment by Haber RS, The diagnosis of recurrent thyroid cancer-anew approach. J Clin Endocrinol Metab 83:4189-4190. 22. Maruyama M, Kat0 R, Kobayashi S, Kasuga Y.A method to differentiate between thyroglobulin derived from normal thyroid tissue and from thyroid carcinoma based on anlysis of reactivity to lectins. Arch Path Lab Med 1998; 122:715-720. S. Early diagnosis 23. Arturi F, Russo D, Giuffrida D, Ippolito A, Perrotti N, Vigneri R, Filetti by genetic analysisof differentiated thyroid cancer metastases in small lymph nodes. J Clin Endocrinol Metab 1997; 82:1638-1641. 24. Winzer R, Schmutzler C, Jakobs TC, Ebert R,J, Reiners Rend1 C, et al. Reverse transcriptasepolymerase chain reaction analysis of thyrocyte-relevant genes in fine-needle aspiration bioDsies of the human thyroid. Thyroid 1998; 8:981-987. 25. Umbricht CB, Saji M, Westra V&, Udelsman R, Zeiger MA, Sukumar S. Telomerase activity: a marker to distinguish follicular thyroid adenoma from carcinoma. Cancer Res 1997; 57:2144-2147. 26. Franceschi S. Iodine intake and thyroid carcinoma-a potential risk factor. Exper Clin Endocrinol Diab 1998; 106 (Suppl 3):S38-S44. 27. Galanti MR, Hansson L, Bergstrom R, Wolk A, Hjartaker A, Lund E, et al. Diet and the risk of papillary and follicular thyroid carcinoma: a population-based case-control study in Sweden and Norway. Cancer Causes Control 1997; 8:205-214. 28. Moosa M, Mazzaferri EL. Outcome of differentiated thyroid cancer diagnosed in pregnant women. J Clin Endocrinol Metab 1997; 82:2862-2866. 29. LiVolsi VA, Asa SL. The demise of follicular carcinoma of the thyroid gland. Thyroid 1994; 4~233-236. 30. Farid N R , Shi Y, Zou M.Molecular basisof thyroid cancer. Endocr Rev 1994; 15:202-232. 31. Challeton C, Bounacer A, DuVillard JA, Caillou B, DeVathaire F, Monier R, et al. Pattern of ras and gsp oncogene mutations in radiation-associated human thyroid tumors. Oncogene 1995; 11:601-603. 32. Roque L, Clode A, Belge G, Pinto A, Bartnitzke S, Santos J R , et al. Follicular thyroid carcinoma:chromosomeanalysisof 19 cases.Genes,Chromosomes & Cancer1998; 21:250-255. S, Suarez HG. Activating 33. Russo D, Aruri F, Schlumberger M, Caillou B, Monier R, Filetti mutations of the TSH receptor in differentiated thyroid carcinoma. Oncogene 1995; 11:19071911. 34. Russo D, Tumino S, Arturi F, Vigneri P, Grass0 G, Pontecorvi A, et al. Detection of an in a caseof an autonomously hyperfunctioning activating mutation of the thyrotropin receptor thyroid insular carcinoma. J Clin Endocrinol Metab 1997; 82:735-738. 35. Tielens ET, ShermanSI, Hruban RH, Ladenson PW. Follicular variant of papillary thyroid carcinoma: a clinical pathologic study. Cancer 1994; 73:424431.
286
Wartofiky
36. Baloch ZW, Gupta PK, Yu GH, Sack MJ, LiVolsi VA. Follicular variant of papillary carcinoma. Cytologic and histologic correlation. AmerJ Clin Path 1999; 111:216-222. 37. Collins SL. Thyroid cancer: controversies and etiopathogenesisof thyroid cancer. In Falk S, editor. Thyroid disease: endocrinology, surgery, nuclear medicine, and radiothera 2nd ed. New York: Raven Press, 1997:495-564. et al. 38. Paphavasit A, Thompson GB, Hay ID, GrantCS,vanHeerdenJA,IlstrupDM,
Follicular and Hlirthle cell thyroid neoplasms: is frozen section evaluation worthwhile? Arch Surg 1997;132:674-678. 39. Vassilopoulou-SellinR,Klein MJ, Smith TH, Samaan HA, Frankenthaler R A , Goepfert H, et al. Pulmonary metastases in children and young adults with differentiated thyroid Cancer 1993; 71:1348-1352. 40. SamuelA M , Rajashekharrao B, Shah DH. Pulmonary metastases in children and adolesce with well-differentiated thyroid cancer.J Nucl Med 1998; 39:1531-1536. 41. Feinmesser R, Lubin E, Segal K, Noyek A. Carcinoma of the thyroid in children-a J Ped Endocrinol Metab 1997; 10:561-568. 42. Cady B. Staging in thyroid carcinoma. Cancer 1998; 83:844-847. 43. Sherman SI. Editorial: Staging in thyroid carcinoma-a reply. Cancer 1998; 835348-850. 44. Brierley JD, Panzarella T, Tsng RW, Gospodarowicz MK, O’Sullivan B.A comparison of different staging systems predictability of patient outcome: thyroid carcinoma as an example. Cancer 1997; 79:2414-2423. 45. Loh K-C, Greenspan FS, GeeL, Miller TR, Yeo PPB. Pathological tumor-node-metastasi @TNM) staging for papillary and follicular thyroid carcinomas: a retrospective analy of 700 patients. J Clin Endocrinol Metab 1997; 82:3553-3562. 46. Hay ID,Bergstralh ET, Goellner J R , Ebersold JR, Grant CS. Predicting outcomein papillary thyroid carcinoma: Development of a reliable prognostic scoring system in a cohort of 1779 patients treated surgically at one institution during 1940 through 1989. Surgery 1993;
1141050-1058. 47. DeGroot LJ, Kaplan EL, Straus FH, Shukla MS.Doesthemethodofmanagementof J Surg 1994;18: papillarythyroidcarcinomamakeadifferenceinoutcome?World 123-130. 48. Watson RG,Brennan MD, Goellner JR, van Heerden JA, McConahey W M , Taylor W F.
Invasive Htirthle cell carcinoma of the thyroid: natural history and management. Ma Proc 1984; 59~851-855. 49. Sanders LE,Silverman M. Follicular and Htirthle cell carcinoma: predicting outcome and directing therapy. Surgery 1998; 124:967-974. 50. Bronner M p , LiVolsi VA. Oxyphilic (AskenasykXtirthle cell) tumors of the thyroid microscopic features predict biologic behavior. Surg Pathol1988; 1:137-150. 5 1. Chen H, NicolTL, Zeiger MA, Dooley WC, Ladenson PW, Cooper DS,et al. Htirthle cell neoplasmsofthe thyroid are there factors predictive of malignancy? Ann Surg 1998; 227:542-546.
52. Samaan NA, SchultzPN, Haynie TP, Ordonez NG. Pulmonary metastasis of differentiate thyroid carcinoma: treatment results in 101 patients. J Clin Endocrinol Metab 1985; 60:
376-380. 53. Samaan NA, Schultz PN, Hickey RC, Goepfert H, Haynie TP, Johnston DA, et al. The results of various modalitiesof treatment of well differentiated thyroid carcinoma: a ret spective review of 1599 patients. J Clin Endocrinol Metab 1992; 75714-720. 54. Har-El G, Hadar T, Segal K, Levy R, Sidi J.HUrthle cell carcinoma of the thyroidgland a tumor of moderate malignancy. Cancer 1986; 57:1613-1617. 55. Thoresen SO, Akslen LA, Glattre E, Haldorsen T, Lund EV, Schoultz M. Survival and prognostic factors in differentiated thyroid carcinoma: a multivariate analysis of 1055 cases Br J Surg 1989; 59:231-235. 56. Arturi F, Russo D, Schlumberger M, duvillard JA, Caillou B, Vigneri P, et al. Iodide
Follicular Thyroid Carcinoma
287
symportergeneexpression in humanthyroidtumors. J ClinEndocrinolMetab1998; 83:2493-2496. 57. Gfunwald F, Pakos R, Bender H, Menzel C, Otte R, Palmedo H, et al. Redifferentiation therapy with retinoic acid in follicular thyroid cancer. J Nucl Med 1998; 39:1555-1558. 58. Grunwald F, Menzel C, BenderH, Palmedo H, Otte R, FimmersR, et al. Redifferentiation therapy-induced radioiodine uptake in thyroid cancer. J Nucl Med 1998; 39:1903-1906. 59. Schmutzler C, Wnzer R, Meissner-Weigl J, Kohrle J. Retinoic acid increases sodiudiodide symporter mRNA levels in human thyroid cancer cell lines and suppresses expression of functional symporter in nontransformedFRTL-5 rat thyroid cells. Biochem Biophys Res C O 1997; ~ 240~832-838. 60. Lazzi S , Spina d, Als C, Tosi P, Mazzucchelli L, Kraft Rainer, et al. Oncocytic (Hiirthle cell) tumors of the thyroid Distinct growth patterns compared with clinicopathological features. Thyroid 1999; 9:97-103. 61. Simpson WJ, PanzarellaT, Carruthers JS, GospodarowiczMK, Sutcliffe SB. Papillary and follicular thyroid cancer: impact of treatment in 1578 patients. Int J Radiat Oncol Biol Physiol 1988; 14:1063-1075. 62. Tubiana M, Haddad E, Schlumberger M, Hill C, Rougier P, SarrazinD. External radiotherapy in thyroid cancer. Cancer 1985; 55(Suppl):2062-2071. 63. Lin JD, Tsang NM, Huang MJ, Weng HF. Results of external beam radiotherapy in patients J Clin Oncol 1997; 27:244-247. with well differentiated thyroid carcinoma. Jpn
28 Pathology of Follicular Cancer James Oertel and Yolanda Oertel
Follicular carcinomas are rather rare in the industrialized nations today(I,2). They do not have the nuclear features of papillary carcinoma, usually have no papillae, lack amyloid and calcitonin, anddo not contain the numerous spindle cells, giant cells, and mitotic figuresof undifferentiated (anaplastic) carcinoma. Most published classifications are based upon the degree of invasiveness of the cancer, but histological patterns of the neoplasm also may provide clues to likely behavior (3). At present, evaluatingthe relationship of the neoplasm to the surrounding tissues has proved be the to most useful guide to categorizing these tumors (4-7). Inspection of the tissues usually reveals a single, spherical, solid, fleshy neoplasm, with pink to tan cut surfaces (if fresh) or pale tan to pale gray surfaces (if fixed in formaldehyde) (8). Tumors composed of oxyphilic cells (AskanazyMirthle cells) are brown. If the tumor contains considerable colloid, the cut surface may appear translu and gelatinous. Small hemorrhages may be present, and there may be focal scarring (especially in the center). A few cancers present as multiple neoplastic nodules, with “daughter nodules” around the one with the thickest capsule. Cystic change and focal necrosis sometimes occur. The tumors are usually encapsulated, but if a tumor is quite invasive, only remnants of the capsule can be detected. Capsules vary in thickness (often are thick), and when a small tumor has a thick capsule, the pathologist should suspect carcinoma rather than adenoma(9,IO). A moderate proportion of follicular carcinomas occurs in association with multiple adenomatoid nodules (or adenomas). Sometimes it is difficult to decide which of the so systematic sectioningof such a specimen tumors is malignant on gross examination, is essential. Follicular carcinomas canbe considered as minimally invasive oras widely invasive (I,7 , I I ) . Such assessmentis performed after surgical resection of the tumor (or occasio ally at autopsy) and requires multiple sections of the periphery of the neoplasm to exclude an adenoma. A total of 10 tissue blocks from the periphery of the tumor is desirable (4)”more if the tumor is particularly cellular or contains numerous mitotic figures. For small tumors, the entire neoplasm should be embedded in such a way that multiple views of its periphery are obtained (12). “Minimally invasive carcinoma” (or “encapsulated carcinoma”) is one with scattered tiny foci of vascular and capsular invasion at its periphery (Figs.1, 2). Very rarely, extension through its capsule in one From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, N1
289
Oertel and Oertel
290
Fig. 1. Follicular carcinoma, minimally invasive. The cancer extends into a thick capsule. (H&E stain; x75).
.""...
.
.-
-
vessel in the
.
Fig. 2. Follicular carcinoma, minimally invasive. There is subtle infiltration of the capsule of the tumor (arrows). (H&E stain; x75).
Pathology
291
Fig. 3. Follicular carcinoma, widely invasive. Vessels are distended by the carcinoma (superior part of the field). (H&E stain; ~ 1 6 ) .
or two places without evident vascular invasion is found (56). When only capsular penetration appears to be present, the patient probably will not have distant spread (13), but the pathologist should search vigorously for vascular invasion (56).Therefore, if a tumor is not evaluated systematically, it may be mistaken for an adenoma (12). Differentiating capsular invasion from invasion of small vessels in the capsule may be so discussion of such phenomena must be evaluated with caution difficult or impossible, (2,5,13). ‘Widely invasive carcinoma” is uncommon, with extensive protrusion into surrounding tissues andlor extension into multiple vessels (often large vascular spaces) (Fig. 3). A “moderately invasive carcinoma” could occupy the region between these extremes and is difficult to define exactly; therefore, such neoplasms are categorized with the widely invasive tumors. in histological sections, Cells of follicular carcinomaare often small and monotonous with uniform round nuclei, stippled chromatin, and central nucleoli. The nucleoli vary considerably in size from one carcinoma to another. Mitotic figures vary greatly in number from one tumor to the next; atypical mitoses are rare. Scattered large or bizar nuclei may occur, but they appear in atypical adenomas as well as in the carcinomas, and their prognostic significance is uncertain (14,151.Cytoplasm is lightly eosinophilic or amphophilic, rarely clear. Usually papillae are absent, but if such structures are present, they are few, small, and simple (I).Psammoma bodies are usually absent; when a few are present, they lie in the colloid of the neoplastic follicles (I). Assessing differentiation by examining routine histological sections be of can interest, especially when combined with immunohistochemical staining with antithyroglobulin
292
Oertel and Oertel
(3,7,I5,I6).Patterns within a neoplasm may be uniform or notably heterogeneous,so biopsies are of limited value in determining the degree of differentiation. Well-differentiated follicular carcinomas are composed entirely or almost entirelyof follicles, either empty or colloid-filled (16).These may vary from minute microfollicl (easily visualized with the PAS technique showing the tiny droplets of colloid in the ( I ) . Most cancers follicles) to follicles even larger than those of normal thyroid tissue (2).Large amounts of immu with a follicular pattern are predominantly microfollicular noreactive thyroglobulin are present in the cells and in the follicles of many of these tumors. Some investigators report that a predominantly follicular pattern has a more favorable prognosis (3,7). Moderately differentiated follicular carcinomas are in those which follicular elemen of any size are mixed with solid islands of cells and/or cordsof cells (trabeculae) (16). Both the solid regions and the trabeculae may contain some microfollicles. Considerab thyroglobulin is present in some regions, but it is sparse or absent in others. Some neoplastic cells may be elongated, even spindled, especiallyin regions where the cells form trabeculae. Such parts are different from undifferentiated carcinomas by being better organized, lacking numerous mitotic figures, and being devoid of necrosis. Poorly differentiated follicular carcinoma is a solid and/or trabecular neoplasm wi some microfollicles, lacks cellular characteristicsof papillary carcinoma, is devoid of the usual features of anaplastic thyroid carcinoma, and may have focal production of thyroglobulin (Id). Obviously, such a cancer overlaps (or is the same as) the poorly differentiated carcinoma (insular carcinoma, ‘Wuchernde struma”). Metastases to cervical lymph nodes are rare(4,8,17).This often is accompanied by direct extrathyroidal extensionof the cancer. The presence of such nodal involvemen should provoke review of the histological features of the resected tissues. If nodes contain follicular carcinoma, the prognosis is probably worse (18,19). Assessing nuclear ploidy has not provided a reliable means of differentiatin adenomas from follicular carcinomas(18,20,21),and the prognostic valueis uncertain (20,221.Argyrophilic staining of nucleolar organizing regions might be useful in r (23,24),but so far it is just one of a variety nizing the malignant follicular neoplasms special techniques that havenot yet proven sufficiently to consider reliable using routine Aspirates from these lesions are diagnosed as follicular neoplasms, which include both follicular adenomas and follicular carcinomas. In our reports we state that “to differentiate between an adenoma and a carcinoma multiple sections through t are required.” On aspiration, these neoplasms bleed of the surgically excised specimen easily, so many specimens are diluted by blood and therefore may be interpreted as unsatisfactory. If the physician performingthe aspiration is experienced and exception ally careful, a specimen with “tumor cellularity” may be obtained. In the “hypocellular~’ smears, the presence of a few microfollicles with inspissated Also present may be some colloid should raise the possibility of a follicular neoplasm. follicular cells arranged in rosettes and tubules. The “hypercellular” smears contain many follicular cells arranged in rosettes and tubules (Fig. 4), microfollicles (often with inspissated dark blue colloid) (Fig.5), and tissue fragments. The neoplastic follicular cells are enlarged and have delicate, pale pink or bluish cytoplasm (scant to moderate in amount), with poorly demarcated borders. The nu
,j I
Fig. 4. Follicular neoplasm. Hypercellular smear with neoplastic cells arranged predominantly in rosettes. Resected specimen revealed a follicular adenoma.(Diff-Quik stain; x400).
Fig. 5. Follicular neoplasm. Aspirate contains rosettes and three follicles with inspissated colloid (arrows).Resected specimen revealed a follicular adenoma. @iff-Quik stain; x400).
294
g. 6, Follicular neoplasm with oxyphilic cells. Smear shows neoplastic cells with a b u n d ~ dense cytoplasm and well demarcated borders. iff-~uik stain; x400).
are enlarged, the chromatin varies in density, and it usually has a mottled appearance; the nuclear borders are slightly irregular, and nucleoli may be visible. In some tumors (both benign and malignant) the variation in nuclear size may be marked (25). Colloid is usually absent, except for the droplets of inspissated colloid observed in some of the neoplastic ~ c r o f o ~ i c l e s . Follicular carcinoma with oxyphilic cells ( A s k a n a z y ~ u r ~cells) e are composed mostly or completely of these distinctive cells. Recogni~ingthe malignant potential of a tumor depends upon the evidence of aggressive behavior at its periphery (~,26-28) Trabecular patterns are common. Bizarre, large, andor h ~ e r c h r o ~ a tnuclei ic may be a striking histological feature, but these are more c o ~ o inn the benign proliferations of oxyphilic cells. Proliferative cell nuclear antigen (PCNA) is reported as present at higher levels in i n d e t e ~ n a t eand m a l i g n ~ oxyphilic t cell neoplasms in c o m p ~ s o to oxyphilic cell adenomas (29).Metastases to cervical lymph nodes are more common than with the usual follicular carcinoma, especially after the patient has undergone surgery for the cancer. Some studies suggest that oxyphilic follicular carcinomas are more aggressive than the usual nonoxyp~licfollicular carcinomas. The presence of nondiploid cells in an oxyphilic carcinoma indicates a poorer prognosis than for one with diploid nuclei (30). The cytological smears show “tumor cellularity” and commonly, a monotonous cell po~ulation.In most cases the cells are large (occasion~ythey are small), have generous amounts of grayish-pink to grayish-blue cytoplasm, large round nuclei, and p r o ~ n e n nucleoli. ~inucleationis common. They are arranged in large tissue fragments, small
Pathology
295
1.
?
Fig. 7. Follicular neoplasm with oxyphilic cells. Smear shows large neoplastic cells with abundant cytoplasm and conspicuous nucleoli. Three empty follicles are visible in superior hal of the field. (Diff-Quik stain; x400). clusters, or singly (25). Frequently the cellular borders are well demarcated (Fig. 6). The neoplastic follicles are common but appear empty (Fig. 7). Some follicles with inspissated blue colloid may be seen.
REFERENCES 1. Franssila KO,Ackerman LV, BrownC L , Hedinger CE. Follicular carcinoma. Semin Diagn Path01 1985; 2:lOl-122. 2. LiVolsi VA. Surgical pathology of the thyroid. Major hob1 Path01 1990; 22:173-212. 3. Mueller-Gaertner H-W, Brzac HT, Rehpenning W. Prognostic indices for tumor relapse and tumor mortality in follicular thyroid carcinoma. Cancer 1991; 67:1903-1911. 4. Lang W, Georgii A, Stauch G, Kienzle E. The differentiation of atypical adenomas and encapsulated follicular carcinomas in the thyroid gland. Virchows Arch A Path01 Anat Histopathol 1980; 385125-141. 5. Kahn NF, Perzin KH. Follicular carcinomaof the thyroid an evaluation of the histologic criteria used for diagnosis. Path01 Annu 1983; 18(Part 1): 221-253. 6.Lang W, ChoritzH,Hundeshagen H. Riskfactorsinfollicularthyroidcarcinomas:a retrospective follow-up study covering a 14-year period with emphasis on morphological findings. A m J Surg Pathol 1986; 10:246-255. 7. Hdie J, Stenwig AE. Long-term survival in patients with follicular thyroid carcinoma-the Oslo experience: variations with encapsulation, growth pattern, timeof diagnosis, sex, age, and previous thyroid surgery. J Surg Oncol 1992; 49:226-230. 8. Rosai J, Carcangiu ML, DeLellis RA. Tumors of the thyroid gland.In Rosai J, Sobin LH, editors. Atlas of tumor pathology, 3rd Ser, Fasc 5. Washington, E:A.F.I.P., 1992.
296
Oertel and Oerte
9. Evans HL. Follicular neoplasms of the thyroid: a study of 44 cases followed for a m i ~ m u of 10 years, with emphasis on differential diagnosis. Cancer 1984; 54535-540. 10. Yamashita T, Fujimoto Y, Kodama T, et al. When is total thyroidectomy indicated as a treatment of “follicular carcinoma”? World J Surg 1988; 12:559--564. 11. Grebe SKG, Hay ID. Follicular thyroid cancer. Endocrinol Metab Clin North Am 1995 24:761-801. 12. Yamashina M. Follicular neoplasms of the thyroid. Total circu~erentialevaluation of the fibrous capsule. Am J Surg Pathol 1992; 16:392--400. 13. van Heerden JA, Hay ID, Goellner JR, et al. Follicular thyroid carcinoma with capsula invasion alone: a non~eateningmalignancy. Surgery 1992; 112:1130-1 138. 14. Hazard JB, Kenyon R. Atypical adenoma of the thyroid. Arch Pathol Lab Med 1954 58:554-563. 15. Jorda M, Gonzalez-Campora R, Mora J, Herrero-ZapateroA, Otal C, Galera H. Prognosti factors in follicular carcinoma of the thyroid. Arch Path01 Lab Med 1993; 117:631-635. 16. Harach HR, Franssila KO. Thyroglobulin i ~ u n o s t a i n i n gin follicular thyroid carcinoma relationship to the degree of differentiation and cell type. ist to pathology 1988; 13:43-54. 17. Schroder S, Pfannschmidt N, Dralle H, Arps H, Bocker W. The encapsulated follicula carcinoma of the thyroid: a clinicopathologic study of 35 cases. Virchows Arch A Path01 Anat Histopathol 1984; 402:259-273. 18. Schroder S. Pathological and clinical features of malignant thyroid tumors: Classification i~unohistology , prognostic criteria. New York: Gustav Fischer, 1988. 19. Segal K, Arad A, Lubin E, Shpitzer T, Hadar T, Feinmesser R. Follicular carcinoma of the thyroid. Head Neck 1994; 16533-538. 20. Grant CS, Hay ID, Ryan JJ, Bergstralh EJ, Rainwater LM, Goellner JR. Diagnostic and prognostic utility of flow cytometric DNA measurements in follicular thyroid tumors. World J Surg 1990; 14:283-290. 21. Oyama T, Vickery Jr AL,Preffer FI, Colvin R€3. A comparative study of flow cytometry and histopathologic findings in thyroid follicular carcinomas and adenomas. Hum Path0 1994; 25 ~ 21-275. 7 22. Hruban RH, Huvos AG, Traganos F, Reuter V, Liebeman PH, Melamed MR. Follicula neoplasms of the thyroid in men older than 50 years of age: a DNA Aow cytometric study Am J Clin Path01 1990; 94527-532. 23. Ruschoff J, Prasser C, Cortez T, Hohne HM, Hohenberger W, Hofsttidter F. Diagnosti value of AgNOR staining in follicular cell neoplasms of the thyroid: comparison of evalua tion methods and nucleolar features. Am J Surg Path01 1993; 17:1281-1288. 24. Shem-Tov Y, Straus M, Talmi YP, Rath-Wolfsom L, Zohar Y, Gal R. Nucleolar organize regions in follicular tumors of the thyroid. Head Neck 1994; 16:420-423. 25. Droese M. Cytological aspiration biopsy of the thyroid gland, 1st ed. Translated by A Burt. Stuttgart: FK Schattauer, 1980. 26. Grant CS, Barr D, Goellner JR, Hay ID. Benign Hiirthle cell tumors of the thyroid: a diagnosis to be trusted? World J Surg 1988; 12:488-495. 27. Bronner MP, LiVolsi VA. Oxyphilic ( A s k a n ~ y ~ u r t hcell) l e tumors of the thyroid: micro scopic features predict biologic behavior. Surg Path01 1988; 1:137- 150. 28. Carcangiu ML, Bianchi S, Savino D, Voynick IM, Rosai J. Follicul~Hurthle cell tumor of the thyroid gland. Cancer 1991; 68:1944--1953. 29. Tateyama H, Yang Y-P, Eimoto T, et al. Proliferative cell nuclear antigen expression in follicular tumours of the thyroid with special reference to oxyphilic cell lesions. Virchow Arch A Pathol Anat Histopathol 1994; 424533-537. 30. Ryan JJ, Hay ID, Grant CS, Rainwater LM, Farrow GM, Goellner 3R. Row cytometri DNA measurements in benign and malignant H W e cell tumors of the thyroid. World Surg 1988; 12:482-487.
29 Surgical Management of Follicular Cancer Orlo H. Clark
Follicular thyroid cancers are derived from follicular epithelium within the thyroid gland. They account for about 10% of all thyroid cancers, and this percentage seems to be decreasing( I ) . Follicular thyroid cancers differ from the more common follicular adenomas because the follicular cells in the cancers invade the vessels or into the capsule or both. Most follicular thyroid cancers have a microfollicular histological In contrast to papillary pattern. These tumors are usually unifocal and encapsulated. thyroid cancers that often metastasize to regional lymph nodes, follicular thyroid can infrequently involve the lymph nodes (less than 10% of patients), but more frequently (2). Follicular thyroid tumors that contain metastasize hematogenously to lung and bones papillary elements are considered to be papillary thyroid as cancer are follicular variants of papillary thyroid cancer (3). In fact, when a young patient is reported to have a follicular thyroid cancer with numerous regional lymph node metastases, this tumor on review is usually a follicular variant of papillary thyroid cancer. Crile and Hazard (4) also stated that follicular thyroid cancers in children behave like papillary thyroid cancer and lymph node metastasesare common. It is likely that some of these tumors were actually follicular variants of papillary thyroid cancer. Patients with follicular thyroid cancers are generally considered to have a worse prognosis than patients with papillary thyroid cancers (5). Most of the difference in prognosis, however, is related to the patients’ older age and more advanced tumor stage at presentation (6).The survival rates of patients with follicular and papillary thyroid cancer when compared at comparable age and disease stageare similar (7-9). Patients will follicular cancers that are small, with minimal capsular invasion, have an excellent prognosis(9).Patients with follicular cancers larger 4than cm, with angioinvasion or with extensive capsular invasion, and who are older have a poor prognosis (3,5,7,10-12). Hiirthle cell cancer is included within the category of follicular thyroid cancer by the WHO classification. Both tumors are judged to be malignant when there is angioinvasion, capsular invasion, or distant metastases. Both Htirthle cell cancers and follicularthyroidcancersoriginatefromfollicularthyroidepithelium,andusually increase CAMP and thyroglobulin production in response to thyrotropin (TSH) (13). HUrthle cell cancers, however, are more likely to be multifocal, more likely to involve
From: Thyroid Cancer:A Comprehensive Guide to Clinical Management Edited b y :L. Wartofsky Q Humana Press Inc., Totowa, NJ
297
298
CZurk
regional lymph nodes, more likely to occur after radiation exposure, to recur locally (14). DeGroot and colleagues (15) recently reported a and more likely to be lethal 12.5% in patient mortality rate of24% in patients with Hiirthle cell carcinoma versus with follicular carcinoma. Others have also reported a higher mortality in patients with (16) Hiirthle cell cancer than in patients with other well-differentiated thyroid cancers Only about 9% of Hiirthle cell carcinomas take up radioiodine, whereas about 75% of follicular cancers take up radioiodine (17J8). Of interest, all deaths and recurrences in patients with follicular carcinoma, reported by DeGroot and colleagues (15) occurred within 13 years, whereas recurrences and deaths due to papillary thyroid cancer co ued during the 40 years of follow up. Surgical management of follicular and Hiirthle cell thyroid cancers like that for papillary thyroid cancers, is controversial. However, more “experts” agree that total or near total thyroidectomy should be done because of the presumed more aggressiv behavior of these tumors. The problem of managing most patients with follicular or Hiirthle cell cancers is that this diagnosis is usually not made preoperatively by FNA cytology or by frozen-section examination, but only after permanent histologic are available. The surgeon must, therefore, have a plan of action in patients with follicular neoplasms. In patients with follicular neoplasm by fine needle aspiration (FWA), I usually recommend a sensitive TSH test and a radioiodine scan if the lesio is smaller than 3 cm. When the scan demonstrates a hot nodule, the patient can be 1%).When the observed as these tumors are rarely ever thyroid cancer (approximately nodule is cold I recommend thyroid lobectomy and isthmusectomy. As mentioned and percentage increases in older pat about 20% of such nodules will be cancer, this or when the tumor is larger than 4 cm in maximal diameter (11J9). In about half of the 20% of patients with cancer, the diagnosis can be made intraoperatively because of regional nodal involvement (usually in patients with follicular variants of papillar thyroid cancer) or because of local invasion. The diagnosis in such patients should b confirmed by frozen section of the enlarged node or of tissue at the site of apparent invasion. For most patients with follicular or Hllrthle cell neoplasms, frozen-section examination is a waste of time and money, since the pathologist usually cannot dis guish between benign and malignant lesions until the permanent sections are obtain Before surgery, I discuss the potential situation withmy patients, and tell them tha the pathologists cannot determine whether a tumoris benign or malignant during the operative procedure. I prefer to do a thyroid lobectomy. In the 10% of patients who have cancer on permanent section, I recommend a completion total thyroidectomy. Some surgeons recommend a near-total thyroidectomy for all patients with follicular or H W e cell neoplasms. I am against this recommendation because it subjects all patients to bilateral procedures and the need for lifelong thyroid hormone replace as good therapy; also near-total or subtotal thyroidectomy is probably not of an opera as a total thyroidectomy for follicular thyroid cancer, because the remnant normal thyroid tissue usually has to be ablated before possible distant metastases c with radioiodine scanning. (9) Many patients with follicular thyroid cancer have minimal capsular invasion These patients have an excellent prognosisso that thyroid lobectomy usually provide definitive treatment. Some pathologists, however, use different definitions of “mini capsular invasion” (12). I consider thyroid cancer to be minimally invasive when i
Surgical Management
299
just invades into the capsule. When there is invasion through the capsule, I consider this “regular invasion” versus where there is extensive invasionI consider these tumors to be “widely invasive.” Kahn and Penin (12) reported the presence of metastatic disease in 14% of patients with capsular invasion versus about 50% for patients with angioinvasion, and in 75% where there is both angioinvasion as well as invasion into extrathyroidal tissue. In contrast to patients with follicular thyroid cancer, in patients with HUrthle cell neoplasms, I look very closely for regional nodal metastases in the central neck and in the tracheoesophageal groove. As mentioned, patients with HUrthle cell neoplasms 30%) and these metastases cannot are more likely to have nodal metastases (about usually be ablated by radioiodine (15-17). When I know a Hiirthle cell neoplasmis a HtirthIe cell cancer, I treat the patient similarly to a patient with a medullary thyroid cancer; I therefore, do a total thyroidectomy and a thorough ipsilateral central neck dissection to avoid tumor recurrence in this area. Postoperatively, I manage patients with follicular thyroid cancer similarly to those with papillary thyroid cancer. In brief, I obtain a serum thyroglobulin to be sure this is less than 3 ng/Inl. I repeat the serum thyroglobulin when the patient is rendered hypothyroid in preparation for radioiodine scanning or therapeutic treatment with 13’I. For low-risk patients I would recommend treatment with an outpatient dose I3’I of (230 mCi). I recommend hospitalization and treatment with 100 to 200 mCi of 13*1for highrisk patients (older than45 years) or patients whose tumors are angioinvasive or have extensive capsular invasion, or both, or are larger than larger 4 cm or have distant metastases. Clinically, solitary distant metastases should be removed surgically and radioiodine should be used to destroy and ablate any residual microscopic disease.
REFERENCES 1. LiVolsi VA, Asa SL. The demise of follicular carcinoma of the thyroid gland. Thyroid 1994; 4233-236. G, A, Girelli ME, Busnardo B. Different 2. Casara D, Rubello D, Saladini G, MasarottoFavero features of pulmonary metastases in differentiated thyroid cancer: natural history and multivariate statistical analysisof prognostic variables. J Nucl Med 1993; 34:1626-1631. 3. Evans HL. Follicular neoplasmsof the thyroid: a study of 44 cases followed fora minimum of 10 years, with emphasis on differential diagnosis. Cancer 1984; 54535-540. 4. Crile G Jr, Hazard JB. Relationship of the age of the patients to the natural history and prognosis of carcinoma of the thyroid. Ann Surg 1953; 138:33-38. 5. Grebe SK, Hay ID. Follicular thyroid cancer. Endocrinol Metab Clin North Am 1995; 24~761-801. 6. Donohue JH, Goldfien SD, Miller TR, Abele JS, Clark OH. Do the prognosesof papillary and follicular thyroid carcinomas differ? Am J Surg 1984; 148:168-173. 7. Brennan MD, Bergstralh El, van Heerden JA, McConahey W M . Follicular thyroid cancer treated at the Mayo Clinic, 1946 through 1970: initial manifestations, pathologic findings, therapy, and outcome [see comments]. Mayo Clin Roc 1991; 66:ll-22. 8. Mazzafem EL. Treating differentiated thyroid carcinoma: where do we draw the line? [Editorial; Comment]. Mayo Clin Roc 1991; 66:105-111. 9. van Heerden JA, Hay ID, Goellner J R , Salomao D, Ebersold JR, Bergstralh El, Grant CS. Follicular thyroid carcinoma with capsular invasion alone: a nonthreatening malignancy. Surgery 1992; 112:1130-1136; discussion 1136-1138. Weghorst CM, Clark OH, et al. Ret mutation screening in MEN2 10. Jhiang SM, Fithian I,
300
Clark
patients and discovery of a novel mutation in a sporadic medullary thyroid carcinoma. Thyroid 1996; 6:115-121. 11. Emerick GT, Duh QY,Siperstein A E , Burrow GN, Clark OH. Diagnosis, treatment, and outcome of follicular thyroid carcinoma. [Comments]. Cancer 1993; 72:3287-3295. 12. Kahn NF, Penin KH. Follicular carcinomaof the thyroid. Path01 Ann 1996; 18:221-253. 13. Clark OH, Gerend PL. Thyrotropin receptor-adenylate cyclase system in Httrthle cell ne plasms. J Clin Endocrinol Metabol 1985; 61:773-778. 14. Grossman RF, Clark OH. HUrthle cell carcinoma. Cancer ControlJ Moffitt Cancer Center 1997; 4:13-17. 15. DeGrootLJ, Kaplan EL, Shukla MS,Salti G, Straus FH.Morbidity and mortality in follic thyroid cancer. J Clin Endocrinol Metab 1995; 802946-2953. 16. Azadian A, Rosen IB, Walfish PG, Asa SL. Management considerations in Htlrthle cell carcinoma. Surgery 1995; 118:711-714; discussion 714-715. of the thyroid. Endocrino 17. Cooper DS, Schneyer CR. Follicular and HUrthle cell carcinoma Metab Clin NorthAm 1990, 19:577-591. 18. Har-El G, Hadar T, Segal K, Levy R, Sidi J. Httrthle cell carcinomaof the thyroid gland a tumor of moderate malignancy. Cancer 1986; 57:1613-1617.
30 Follicular Carcinoma of the Thyroid External Radiation Therapy Robert L. White and Leonard Wartofsky
It has been observed by DeGroot and colleagues ( I ) that the mortality from follicular thyroid carcinoma may be double that of the papillary variety of cancer with death at in men over age 45 withinitial tumors a younger age. The poor prognosis, particularly larger than 2.5 cm led these workers to propose that measures more vigorous than simply thyroidectomy followed by radioiodine should be considered, to include external radiotherapy and prophylactic chemotherapy. There are limitations, however, even in the efficacy of external megavoltage irradiation in the management of follicular thyroid carcinomas just as was described earlier in this volume for papillary carcinoma. In general, follicular tumors are less radiosensitive than papillary or mixed papillaryfollicular tumors (2). External megavoltage radiation has its greatest application to follicular tumors which have lost their ability to accumulate radioiodine but it can also A general review be used as an adjunct to supplement the effects of radioactive iodine. of the useof external radiation therapy for the treatment of thyroid cancer has appeared recently (3). The same group reported on their success with radioiodine and external 120 had radiotherapy in 382patientswithdifferentiatedthyroidcancer,ofwhom follicular carcinoma(4). It is often difficult to assess the responses to radiation therapy when the patients selected to receive such therapy are often those with the most ext disease. This was the case in the latter study, in which the use of external radiation was associated with the most advanced local disease. Notwithstanding this caveat and after adjustment for prognostic factors, there was no significant difference in cause specific survival from the cancer between patients who received external radiation and those who did not. Patients with either residual microscopic or macroscopic papillary carcinoma tended to do better than the patients with follicular carcinoma (4). Similar results were seen in a retrospective analysis by Lin and associates (5) of 72 patients with differentiated cancer who received external radiation therapy postoperatively. While radiation therapy appeared to cause temporary tumor regression, no significant effect was seen in survival rate. Arguably the largest series of patients, a Canadian survey of 504 follicular thyroid cancers demonstrated better results(6).Improved local control rates and cause-specific survival rates were seen, although some patients received combination radiotherapy
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L Wartofsky 0 Humma Press Inc., Totowa, NJ
301
302
White and Wartofiky
with radioiodine treatment. Improvement was best in those patients with local in soft tissues. The salutary results led Simp into the trachea, esophagus, or peritracheal (7) to recommend that all patients with extrathyroidal invasion and no uptake of l3l should receive irradiation to the thyroid bed. A more recent report on the efficacy of prophylactic irradiation on survivalin 56 patients with follicular thyroid cancers indi cated that postoperative irradiation would reduce local or regional relapse but was effective in prolonging survival as was shown for the same management of papillary thyroid cancer (8). As with aggressive papillary cancers, the primary indication for megavoltage irradiation is the presence of bulky unresectable thyroid carcinoma that either does accumulate 13’1 or where 1311 may not be adequate for local control of the tumor (9) Residual bulky follicular tumor after surgery would be unusual in the neck (asmay occur in papillary, medullary, or anaplastic cancer), and rather may present as distan metastases to lung or bone. When residual or metastatic follicular carcinoma in the lower neck or mediastinum does occur and is not controlled by 1311 alone, external radiation therapy may be effective. Salutary effects have been seen, for example, for the amelioration of signs and symptoms of a superior vena caval syndrome. In the presence of metastatic follicular carcinoma to bone, external radiation is indicated to prevent pathologic fractures, regardless of the ability of the metastases to concentrate 13’I.In general, the use of external radiation therapy should be considered w and locally recurrent follicular thyroid carcinoma occurs in spite of I3lIaccumulation or after maximal I3lI therapy. Just because a tumor readily may trap radioiodine and be imaged on scintiscan does not necessarily imply that the tumor will be rad to the I3lI. When a patient develops brain metastasis from thyroid carcinoma, extern radiation is indicated for a potentially reliable and rapid response. The ability of the brain metastasis to concentrate I3’Imay be altered by the blood-brain barrier and thus external irradiation could be more effective this in clinical situation. However, patien withbrainmetastasesfromfollicularcarcinomashouldbeconsideredtohavean extremely poor prognosis, and responses to radiation therapy are not necessarilyseem. In one recent series of 47 patients with thyroid carcinoma metastatic to brain, there was no apparent evidence of efficacy of either radioiodine or external radiation therap (10). In another recent series of 15 patients with brain metastases, six of whom were six months after treated with external radiation, the average survival was less than detection of the brain metastases (11). With some focal deposits of tumor such as is not possible metastases to bone or other sites where complete removal by a surgeon the use of preoperative external radiation alone or in conjunction with 1311 may serve be to shrink or stabilize the tumor mass such that surgery following radiation may technically easier and with less risk of operative blood loss.In such cases, coordinatio between the surgeon and the radiation oncologist is very important inthe management ofthyroidcarcinoma to optimize the timing of treatment and the patient’s feeling of security. External irradiation may be used in sequence or in conjunction with chemotherap particularly where the thyroid carcinoma is anaplastic or poorly differentiated. Since optimal time and dose relationships between external radiation and systemic c apy have not been optimized, local agreement between the medical oncologist and
External
Rx
303
radiation oncologistis important to help patients understand the importance of coordin Kim andLeeper (12) reported 41 patientsgivencombination ingtheirtreatment. Adriamycin (doxorubicin) and external-beam radiation therapy,ofhalf whom had welldifferentiated thyroid carcinoma. The initial complete tumor response was 91% with 77% by 2 yr of follow-up, and a median survival time4of years. local tumor control at For curative treatment for thyroid carcinoma with external megavoltage irradiation, therearemanytechnically demanding details (13). The typical definitive dose for residual or bulky thyroid carcinoma is 6500 cGy in 7 weeks with a daily dose of 180-200 cGy daily 5 days a week. In patients in whom there are metastases to the spine or spinal cord, it is mandatory to pay particular attention to the spinal cord dose to avoid irreversible neurologic damage. Special blocking techniques with a cerrobend as well as other blocking system should limit the radiation dose to the spinal cord radiation sensitive structures. All of the treatment areas where microscopic or small deposits of cancer could be present are treated with doses to 5000 cGy over 5 to 6 weeks time. The spinal cord is shielded after 4500 cGy in 4.5 to 5 weeks time. If chemotherapy is employed in conjunction with radiation, the cumulative spinal cord dose should be reduced by 5OOcGy. Where tissue thickness results in doses of less than 5000 cGy in 5 to 6 weeks, boosting techniques must be employed to assure that thedose is as uniform as possible.Thereareseveralmethodsofradiationbeam arrangements and portals which allow adequate doses to be delivered to the neck and set of mediastinum. In most cases an anterior to posterior and posterior to anterior portals with 'To, 4 or 6 M V photons will allow 4500 to 5000 cGy to be delivered in 4.5 to 6 weeks time. Boosting techniques utilizing electron ports of 8 to 14 MeV can supplement the areas treated to4500 to 5000 cGy to definitive doses of6500 to 7000 cGy in 5 to 8 weeks time. To avoid the spinal cord, in addition to cerrobend blocking, oblique anterior portals with wedges are occasionally utilized. Some of the newer treatment techniques include arching or rotational fields with flying wedges to optimize external irradiation to the treatment volume while minimizing treatment to the spinal cord or othercritical structures.Clinical experience has documented that external irradiation with or without 13'1 can produce long term local control for upto 25 years in patients with differentiated thyroid carcinomas who have microscopic residual or gross disease after surgery (14). Skeletal, brain, hepatic, pulmonary or subcutaneous metastasis of differentiated follicular thyroid carcinoma may be treated with external megavoltage irradiation with or without 13'I.Obviously, if the metastatic thyroid cancer does not accumulate13'1, then 3500of to 4500 external irradiation alone becomes the treatment of choice. Dose levels cGy in 3 to 4.5 weeks are recommended for optimal palliation of metastasis to soft tissue or bone. When thereis a possibility of pathological fracture in the caseof bone metastasis, stabilization with an intramedullary rod or other orthopedic procedure sh precede the external radiation. Patients who receive systemic chemotherapy and external irradiation concurrently or sequentially should not be treated with daily doses to exceed 180 cGy because of the possibility of undesirable dose potentiating side effects. Daily management for the patient receiving combinationsof chemotherapy and external irradiation is difficult and requires close surveillance and observation. Usually the side effects of oral mucositis,
304
White and Wartofiky
esophagitis, and skin erythema are worse for patients treated with combined modalit and patients need to be carefully and cautiously observed regularly (15). To turn from radiation as therapy to radiation as a causeof thyroid neoplasia, it is well known that there is a linear relationship between radiation doses up to1800 cGy and the incidence of thyroid nodules and cancer. The increased risk of thyroid cancer is primarily a problem after radiotherapy to the head and neck in children. Among individuals in the U.S. receiving head and neck irradiation in childhood, palpable of these nodules (16,17). nodules are found in 16%to 29% and carcinoma in one third Most of these tumors will present within10 to 20 years of exposure, but the risk may 35 years. The vast majority will be papillary and not follicular ca exist for over and the treatment of radiation-induced thyroid cancer will depend on the initial physical is the same findings as wellas the clinical presentation. However, overall the treatment as for thyroid cancers not induced by radiation. The treatment modalities incl thyroid hormone therapy, 13'1therapy, external irradiation, interstitial irradiation, and chemotherapy when required. Interstitial irradiation is helpful and valuable in the treatment of primary thyroid carcinomas as well as metastatic carcinoma to the thyroid from other primary sites. lZ5Ihave been utilized in this clinical setting Removable 1921rand permanently implanted and may have application when thereis no uptake of radioiodine(18,19). In addition, IgZIr has been implanted into mediastinal masses metastatic from thyroid carcinomas and sarcomas. In one study of 155 patients with differentiated thyroid tumors, five with marked vascular and/or capsular invasion received 400-1000 cGy intraoperative brachytherapy with lair coupled with percutaneous radiation and tumor control was obtained for the thyroid bed inall five (19). Since thereis minimal general experience andfewpatientshavebeentreated,theinterstitialtreatmenthasnotbeenwidely publicized. In experienced hands, the interstitial irradiation techniques have produce long-term disease-free survival in patients and improved local control. The advantag of interstitial irradiation includes minimal side effects and complications and imp local responsiveness, but the clinical experience is limited.
REFERENCES
1. DeGroot LJ,Kaplan EL, Shukla MS, Salti G, Straus FH. Morbidity and mortality in follicular thyroid cancer. J Clin Endocrinol Metab 1995; 802946-2953. 2. Greenfield LD: Radiation therapy in the management of thyroid carcinoma. In Greenfield DL, editor. Thyroid cancer. Boca Raton,FL: CRC, 1978; 177-187. of thyroid malignancy 3. Brierley JD, Tsang RW. External radiation therapy in the treatment Endocrinol Metab ClinNorth Am 1996; 25141-157. SB.The 4. Tsang RW, Brierley JD, Simpson WJ, Panzarella T, Gospodarowicz MK, Sutliffe
effects of surgery, radioiodine, and external radiation therapy on the clinical outcome of patients with differentiated thyroid carcinoma. Cancer1998; 82:375-388. 5. Lin JD,Tsang NM, Huang MJ, Weng HF. Results of external beam radiotherapy in patien with well differentiated thyroid carcinoma. JpnJ Clin Oncol 1997; 27:244-247. 6. Simpson WJ, Panzarella T, Carmthers JS, et al. Papillary and follicular thyroid cancer: impact of treatment in 1578 patients. Int J Radiat OncolBiol Phys 1988; 14:1063-1075. 7. Simpson WJ. Radioiodine and radiotherapy in the management of thyroid cancers. go1 Clin North Am 1990; 23509-521.
External Radiation RX
305
8. Esik 0,Nemeth G, EllerJ. Prophylactic external irradiation in differentiated thyroid cancer: a retrospective study over a 30-year observation period. 9. Lindberg RD. External beam irradiation in thyroid carcinomas. In Fletcher GH, editor. Textbook of radiotherapy, 3rd ed. Philadelphia: Lea & Febiger, 1980; 384-388. 10. Chiu AC, Delpassand ES, Sherman SI. Prognosis and treatment of brain metastases in thyroid carcinoma. J Clin Endocrinol Metab 1997; 82:3637-3642. 11. Samuel A M , Shah DH. Brain metastases in welldifferentiated carcinomaof the thyroid. Tumori 1997; 83:608-610. 12. Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with combination doxorubicin and radiation therapy. Cancer1987; 60:2372-2375. 13. Moss WT, Brand WN, Battifora H. The thyroid.In Radiation oncology: rationale, technique, results, 5th ed. St. Louis: CV Mosby, 1979; 233-242. 14. Simpson WJ, McKinney SE, Carruthers JS, Gospodarowicz MK, Sutcliffe SB, Panzarella T. Papillary and follicular thyroid cancer. Prognostic factors in 1578 patients. Amer J Med 1987; 83:479-88. 15. Greenfield, LD. Thyroid tumors. In Perez CA, Brady LW, editors. Principles and practice of radiation oncology. Philadelphia: JB Lippincott, 1987; 1126-1156. 16. Favus MJ, Schneider AB, Stachura ME, Arnold JE, Ryo W, Pinsky SM, Colman M, Arnold MJ, Frohman LA. Thyroid cancer occuning as a late consequence of head-andneck irradiation. N Engl J Med 1976; 294:1019-1025. of 17. DeGroot LJ, Reilly M, Pinnameneni K, Refetoff S. Retrospective and prospective study radiation-induced thyroid disease.Am J Med 1983; 74:852-862. 18. Kanitz W, Kopp J, Hamper1 WD, Heidenreich P, Wagner T. Interstitial radiotherapy with 125-1seeds in non-operable and non-radioiodine retaining local recurrences of differentiated and undifferentiated thyroid cancers. WienKlin Wochensch 1990; 102277-280. 19. Wolf G, Kohek P, Geyer E, Pakisch B, Langsteger W, Ramschak S, Passath A. Introperative radiation therapy, endotracheal hyperthermia, and IR-191 brachytherapy in patients with advanced thyroid cancer. Acta Med Austr 1996; 23:76-79.
31 Follicular Thyroid Cancer Follow-Up Henry B. Burch STRATEGY
As is the case with papillary thyroid cancer, the effort applied to the search for recurrent follicular thyroid canceris determined by the likelihood of tumor recurrence and death from disease. Patients with follicular cancer have a higher incidence of cancer-related death than patients with papillary thyroid cancer ( I ) . A recent review two malignancies cited 16 studies showing improved of survival statistics for these survival in papillary thyroid cancer compared to follicular thyroid cancer, 5 studies showing no difference in survival, and 1 study showing improved survival in patients with follicular thyroid cancer(2). Recently, patients with minimally invasive follicular thyroidcancerhavebeenscrutinizedwithrespecttooutcomeandfoundtobeat (3-6). As is discussed below relatively lowrisk for recurrence and cancer-related death (see Chapter 32), many of the same determinants of prognosis in papillary thyroid cancer are applicable to patients with follicular thyroid cancer. The presence or absen of poor prognostic indicators is used to tailor the frequency and intensity of surveillance for tumor recurrence. This chapter reviews the rationale used to determine appropriate follow-up for patients with follicular thyroid cancer and provides a current overview of the tools available to assist in this objective.
WHAT LEVEL OF SURVEILLANCE? The higher rate of cancer-related death associated with follicular thyroid cancer mandates a higher index of suspicion for recurrent disease than with papillary thyroid cancer. The propensity for early hematogenous spread also directs attention to distant sites such as the lung, bones, brain, and liver in cases of suspected recurrent disease (7). The usual approachto patients with widely invasive follicular thyroid cancer is to recommend near-total thyroidectomy followed by radioiodine ablation with 100-150 mCi of I3lI (1,2,8). Patients with one or more poor prognostic factors(see chapter 32) undergo whole-body scanning every 6 months for 18 months, and then annually for 5 years. Thereafter, whole-body scan (WBS) and serum thyroglobulin (TG) levels are obtainedat3-yearintervals.Themethodschosentofollowpatientswithsmaller,
F m : Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited b y :L. Wartofsky 0 Humana Press Inc., Totowa,NJ
307
308
Burch
minimally invasive tumors are dependent on the extent of initial therapy. Patients treatedwithsubtotalthyroidectomywithoutradioiodineablation maybefollowed with a careful neck examination and serum thyroglobulin levels on thyroid hormone suppressive therapy. Patients treatedinitially with total thyroidectomy and radioiodine ablation, as is my practice, are subjected toWBS and serum Tg measurement annually for 3 years, and at 3- to 5-year intervals thereafter. Adverse events appear to occur earlier in patients with follicular thyroid cancer compared to those with papillary thyroid cancer (2,4,8). One extensive review of the follicular thyroid cancer literature found that most recurrences and cancer-relat occur in the first 5 years after diagnosis, with most studies showing that 5040% of adverse events occur in the first 2 years after diagnosis (2). A recent study including 49 patients with follicular thyroid cancer and Hiirthle cell carcinomas found that all recurrences and deaths occurred within 13 years of diagnosis, whereas papillary thyr cancer patients at the same institution continued to experience adverse events 40 years of observation (8).
SERUM THYROGLOBULIN MEASUREMENT IN PATIENTS WITH FOLLICULAR THYROID CANCER Patients with follicular carcinomas generally have higher levels of TG than those with papillary carcinomas (9,lO). The utility and limitations of serum thyroglobulin measurement as reviewed in Chapters 22 and 24 also apply to patients with follicular thyroid cancer. OTHER TUMOR MARKERS
Persistent or increasing TGAb titers may serve as a tumor marker for persistent thyroid cancer (11-14). When rendered free of disease, most thyroid cancer patients of these antibodies over with positive TGAb experience a gradual decrease in the titer time (11,12). Conversely, TGAb titers often remain positive or increase further in patients with persistent disease (11,12). One study, involving 32 patients with posi 5 patients with persistent or progressive TGAb before therapy, found that each of disease had persistently positive Tg antibody levels, whereas only 6 of the 27 patient (13). Another study, involving 43 deemed free of disease remained TGAb positive thyroid cancer patients with positive TGAb followed found 5that of 19 (26%) patients 0 of 23 patients in whom with persistent TGAb had residual disease, compared to TGAb decreased after therapy(14). It appears that the presence of functioning thyro tissue, metastatic or otherwise, is necessary to perpetuate TG antibody synthesis.
REFERENCES 1. Robbins J, Merino UT, Boice JD, Ron E, Ain K B , Alexander HR, et al. Thyroid cancer:
a lethal endocrine neoplasm. Ann Intern Med 1991; 115:133-147. 2. Grebe SKY Hay ID. Follicularthyroidcancer.EndocrinolMetab Clin North Am 1995; 24~761-801. 3. Davis NL, Bugis SPY McGregor GI, Gennann E. An evaluation of prognosticscoring systems in patients with follicular thyroid cancer. Am J Surg 1995; 170:476-480. 4. Jorda M, Gonzalez-Campora R, Mora J, Herrero-Zapatero A, Otal CyGalera H. Prognostic factors in follicular carcinoma of the thyroid. Arch Path01 Lab Med 1993; 117:631-635.
5. Segal K, Arad A, Lubin E, Shpitzer T, Hadar T, Feinmesser R. Follicular carcinoma of the thyroid. Head Neck 1994; 16:533-538. 6. van Heerden JA, Hay ID, Goellner JR, Salomao D, Ebersol J R , Bergstralh ET, Grant CS. Follicular thyroid cancer with capsular invasion alone: a nonthreatening malignancy. S 1992;112:1130-1138. 7. Cooper DS, Schneyer Follicular and Hiirthle cell carcinoma of the thyroid. Endocrinol Metab Clin North Am 1990; 19577-591. 8. DeGroot LJ, KaplanEL,ShukulaMS, Salti G, Straw FH. Morbidity and mortality in follicular thyroid cancer.J Clin Endocrinol Metab 1995; 80:2946-2953. S, Geerlings H. Compari9. Dralle H,SchwaraockR, LangW, Backer W, Ziegler H, Schroder son of histology and immunohistochemistry with thyroglobulin serum levels and radioiodin uptake in recurrences and metastases of differentiated thyroid carcinomas. Acta Endocrinol (Copenh) 1985; 108:504-510. 10. ShahDH,Dandekar SR, Jeevanram RK, KumarA,SharmaSM,Ganatra RD. Serum thyroglobulin in differentiated thyroid carcinoma: histological and metastatic classification. Acta Endocrinol (Copenh) 1981; 98:222-226. 11.PaciniF,Mariotti S, FormicaN,Elisei R, Anelli S , CapotortiE,PincheraA.Thyroid autoantibodies in thyroid cancer: incidence and relationship with tumor outcome. Acta Endocrinol (Copenh) 1988; 119:373-380. U, Sharma SM. Significance of 12. Kumar A, Shah DH, Shrihari U, Dandekar SR, Vijayan antithyroglobulin antibodiesin differentiated thyroid carcinoma. Thyroid 1994; 4:199-202. of the 13. Rubello D, Girelli ME, Casara D, Piccolo M, Perin A, Busnardo B. Usefulness combined antithyroglobulin antibodies and thyroglobulin assay in patients with differentia thyroid cancer. J Endocrinol Invest 1990; 13:737-742. 14. Rubello D, Casara D, Girelli ME, Piccolo M, Busnardo B. Clinical meaningof circulating antithyroglobulin antibodies in differentiated thyroid cancer: a prospective study. J Nucl Med 1992; 33:1478-1480.
CR.
32 Follicular Thyroid Cancer Prognosis Henry B. Burch
DETERMINANTS OF PROGNOSIS IN PATIENTS WITH FOLLICULAR THYROID CANCER
A large number of retrospective analyses have been performed to determine patient and tumor characteristics associated with a poor prognosis in patients with follicular thyroid cancer (reviewed in reference I ) . Factors which have consistently been shown to negatively affect prognosis in follicular thyroid cancer include patient age greater than 45 years, tumor size larger than4 cm diameter, local tumor extension beyond the thyroid, extensive capsular and vascular invasion, and the presence of distant metastases is illustrated in a series of 100 (Table 1). The impact of these factors on survival patients with pure follicular thyroid cancer receiving treatment at the Mayo Clinic ove a 35-year period (2). The overall cancer-related mortality was 29% at 20 years inthis study. However, patients with only one negative prognostic indicator had a 20-year mortality of only14%, while patients withtwo or more predictors had a 92% likelihood 20 years (2). In addition, certain variants of of having died from thyroid cancer at follicular thyroid cancer, including Hiirthle cell carcinoma(3) and insular carcinomas (4) are associated with a generally worse prognosis.
MINIMALLY INVASIVE FOLLICULAR THYROID CANCER
An important prognostic considerationin patients with follicular thyroid carcinoma is thedegree of capsularandvascularinvasion.Patientswithminimallyinvasive follicular thyroid cancer have generally been shown to have lower cause-specific m (5-8). In fact, patients with minimal ity rates than patients with widely invasive tumors capsular invasion alone have a survival rate which approximates that of the general population (8).This finding has led some authors to recommend avoidance of a “cancer diagnosis in these patients due to socioeconomic concerns such as employability and insurability (9).
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wmtofiky 0 Humana Press lnc., Totowa, NJ
311
312
Burch Table 1 Poor Prognostic Factors for Differentiated Thyroid Cancer Large tumors (> 4.0 cm) ~~
Male sex Advanced tumor grade Tumors with local extension Bilateral cervical or mediastinal lymph node metastases* Distant metastases Extensive vascular and capsular invasion? *Applies mainly to papillary rhyroid cancer. ?Applies mainly to follicular thyroid cancer.
EFFECT OF THERAPY ON PROGNOSIS
As is the case with papillary thyroid cancer, retrospective assessment ofthe effect of therapy on prognosis is hampered by the fact that patients selected to receive more extensive therapy are likely to have more advanced disease. Therefore, a finding of no difference in survival between patients receiving or not receiving a therapy, such as radioiodine ablation, might actually indicate a beneficial effect since these patients would have been expected to have a shorter survival than those with less advanced disease. Looking at the effect of the extent of surgery on survival, one study found thatamong 214 patients operated upon for follicular thyroid cancer, those treated with total thyroidectomy had survival rates similarto patients receiving less extensive procedures (10). Another study, after adjusting for otherrisk factors in a multivariate analysis, found no difference in survival between 19 patients undergoing lobectomy and 81 patients treated with a bilateral procedure (2). Conversely, most studies have shown that patients with complete removal of their tumor have improved survival ov those with known residual macroscopic or microscopic disease after surgery( I ) . The use of radioiodine for remnant ablation after thyroidectomy has been found to have variable effects on survival, with a definite beneficial effect found in some studies (10,11), a marginal effect in another study (12), and no effect in still another study (13). Despite this controversy, effective follow-up of patients with follicular thyroid cancer is greatly facilitated by the use of both near-total thyroidectomy and radi ablation therapy (1,11).
REFERENCES
ID.
1. Grebe SK, Hay Follicular thyroid cancer. Endocrinol Metab Clin North A m 1995; 24761-801. 2. Brennan MD, Bergstralh ET, van Heerden JA, McConahey W M . Follicular thyroid cance treated at the Mayo Clinic,1946 through 1970: initial manifestations, pathologic finding therapy, and outcome. Mayo Clinh o c 1991; 66:ll-22. 3. Carcangiu ML, Bianchi S, Savin0 D, Voynick IM, Rosai J. Follicular HUrthle cell tumors of the thyroid gland. Cancer 1991; 68:1944-1953. 4. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (insular) thyroid carcinoma. A reinterpretation of Langhans’ wuchernde Struma. Am J Surg Pathol 1984; 8:655-668. 5. Davis NL, BugisSP,McGregorGI,Germann E. An evaluation of prognostic scoring systems in patients with follicular thyroid cancer. Am J Surg 1995; 170:476-480.
Prognosis
313
6. Jorda M, Gonzalez-Campora R, Mora J, Hemro-Zapatero A, Otal C,Galera H. Prognostic
factors in follicular carcinomaof the thyroid. Arch Path01 Lab Med 1993; 117:631-635. 7. Segal K, Arad A, Lubin E, Shpitzer T, Hadar T, Feinmesser R. Follicular carcinoma of the thyroid. Head Neck 1994; 16:533-538. 8. van Heerden JA, Hay ID, Goellner J R , Salomao D, Ebersol JFt, Bergstralh El, Grant CS. Follicular thyroid cancer with capsular invasion alone: a nonthreatening malignancy. S 1992;112:1130-1138. 9. Feind C. Discussion following: van HeerdenJA,Hay ID, Goellner JR, Salomao D, Ebersol J R , Bergstralh ET, Grant CS. Follicular thyroid cancer with capsular invasion alone: a nonthreatening malignancy. Surgery 1992; 1121130-1 138. 10. Young RL, Mazzafeni EL, RaheAJ, Dorfman SG.Pure!follicular thyroid carcinoma: impact of therapy in 214 patients. J Nucl Med 1980; 21:733-737. 11. Samaan NA, Maheshwari YK,Nader S, Hill CS, Shultz PN, HaynieTF', Hickey RC, Clark RL, Goepfert H, Ibanez ML, Litton CE. Impact of therapy for differentiated carcinoma of the thyroid: an analysis of 706 cases. J Clin Endocrinol Metab 1983; 56:1131-1138. 12.DeGroot LJ, KaplanEL,ShukulaMS,SaltiG,Straus FH. Morbidity and mortality in follicular thyroid cancer.J Clin Endocrinol Metab 1995; 80:2946-2953. 13. Jensen MH, Davis RK, Derrick L. Thyroid cancer: a computer-assisted review of 5287 cases. Otolaryngol Head Neck Surg 1990; 10251-65.
33 Follicular Thyroid Cancer Special Aspects in Children and Adolescents Merrily Poth Any discussion of follicular thyroid cancer in this population is severely handicapped by an almost total lack of definitive data. While there are a few published studies that concentrate on follicular cancer, in none of these are the small number of children separated in the discussion from the larger group of “younger” patients, which usually refers to all patients under the age40. of In these reports all of the young patients have an excellent prognosis, particularly if the initial tumor is relatively small. Small tumors, 4 cm (1-4). with good prognoses in these reports are those with diameters of less than Likewise there are multiple reports analyzing the presentation and outcome of differentiated thyroid cancer in children and adolescents but in the vast majority of these, most of the tumors are papillary and thereis no separate analysisof the few follicular lesions. The overall treatment presentation and outcome data from these studies are covered in Chapters 16, 22, and 25. The implication of these studies is that follicular and papillary thyroid cancers in children share an excellent prognosis (5-9). In spite of this lack of definitive and specific data, there are a few observations which are probably worth making about follicular tumors in children. The presentation of these tumors is similar to that of papillary cancer. That is, most of them present as a solitary thyroid nodule. The difference is that there may be a lower presentation that includes enlarged local lymph nodes and in fact the incidence of local spread at is less than in papillary cancer. The evaluationof the solitary nodule will usually include a fine needle(FNB). biopsy When the FNB report is suspicious of follicular neoplasm, the initial operation will usually be a lobectomy and further surgery will await the final pathology report. In this circumstance it is unlikely that examination of tissue at frozen section will yield useful data and any discussion regarding further surgery will await the final pathology report. The need for completion thyroidectomy in the caseof a small lesion reported as a follicular thyroid cancer is also controversial. Since the long-term prognosis is excellent in lesions smaller than 4 cm, and there are no studies that report that the extent of surgery or the use of I3’Iin these cases improves the prognosis, some experts would
From: Thyroid Cancer: A Comprehensive Guideto clinical Management Edited by:L. Wartofsky 0 Humana Press Inc., Totowa, NJ
315
31 6
Poth
simply follow such patients after a lobectomy, without adjunctive therapy. Most, however, would treat the condition with thyroxine at a dose sufficient to keep the senun TSH at the lower extremeof the normal range. Again, there are no data from carefully controlled studies that support either the need for thyroxine therapy or establish the degree of TSH suppression required in such patients.
REFERENCES 1.Emerick GT, Duh
QY,Siperstein AE, Burrow GN, Clark OH. Diagnosis, treatment, and outcome of follicular thyroid carcinoma. Cancer 1993; 723287-3295. 2. de Souza FM.Role of subtotal thyroidectomy in the management of the follicular neoplas of the thyroid. Laryngoscope 1993; 103:477-493. 3. Segal K, Arad A, LubinE, Shpitzer T, Hadar T, Feinmesser R. Follicular carcinoma of the thyroid. Head Neck 1994; 16(6):533-538. 4. Lamberg BA, Karkinen-Jaaskelainen M, Franssila KO. Differentiated follicle-derived carcinoma in children. Acta Pediatr Scand 1989; 78:419-425. 5. Fassima AS, Rupolo M, PelizzoM R , Casara D. Thyroid cancer in children and adolescen Tumori 1994; 80:257-262. 6. Moir CR, Telander RL. Papillary carcinomaof the thyroid in children. Semin PediatrSurg 1994; 3:182-187. 7. Zimmerman D, Hay ID, Gough IR,Goellner JR, Ryan JJ, Grant CS, McConahey W M . Papillary thyroid carcinoma in children and adults: long-term follow-up of 1039 patients three decades. Surgery1988; 101:1157-1166. conservatively treated at one institution during 8. Welch-Dinauer CA, TuttleR M , Robie DK, McClellan DR,Svec RL, Adair C, Francis GL.
Clinical features associated with metastasis and recurrence of differentiated thyroid cancer in children, adolescents and young adults. Clin Endocrinol1998; 49:619-628.
V Undifferentiated Cancers A. Anaplastic Carcinoma
34 Anaplastic Carcinoma Clinical Aspects Steven I. Sherman Anaplastic carcinoma describes an undifferentiated malignancy derived from more well-differentiated thyroid follicular epithelium. In contrast to the generally indolent nature of differentiated thyroid carcinoma, anaplastic carcinoma represents one of the most aggressive human neoplasms, with a disease-specific mortality of at least 90%. Early recognition of the disease is essential to allow prompt initiation of therapy and to maintain hope for a significant tumor response.
VARIANTS Traditional descriptions of undifferentiated thyroid carcinomas divided anaplastic lesions into two categories, based upon histological features (see Chapter 35). The spindle cell, giant cell, and squamoid tumors belonged to the typically more aggressive subtype, occurring in older patients and rapidly leading to death. These histologies continue to be classified as anaplastic carcinomas. The “small cell” histology, on the other hand, was thought to be associated with relatively improved survival ( I ) . Later studies with electron microscopy and immunocytochemical markers for lymphoid and neuroendocrine cell lineage demonstrated that mostof these “small cell” tumors were in factlymphomas or medullarycarcinomas.Forexample,ofthe three long-term survivors with “small-cell carcinoma” reported from the Mayo Clinic 1985 in ( I ) , two were later reported to have lymphoma and one medullary carcinoma (2). Therefore, current nosology for undifferentiated thyroid carcinomas does not include small cell variants (3). Histopathological subtypes continue to be described, but the biological and clinical relevance of such subdivisions is unclear. CLINICAL ASPECTS
Epidemiology
The age-adjusted annual incidence of anaplastic carcinoma is about 2 per million persons in the United States(4), accounting for only2-5% of all thyroid malignancies. A similarly low incidence and frequency among thyroid malignancies has been de
From: Thyroid C a n c e r : A Comprehensive Guideto Clinical Management Edited by: L. Wartofiky 6 Humana Press Inc., Totowa, NJ
319
320
Sherman Table 1 Demographic Features of Patients With Anaplastic Carcinoma in Nine Recent Series ~
~
~~~~
~
~~
No. of Patients
Female (W
Mean Age at Diagnosis (Years)
1 8 9 92 93 26 21 10 45
82 70 12132 20 48 33 21 46
41 76 55 19 80 63 76 52 76
65 67 61 58 75 66 74 65 71
Total or weighted mean
473
60
66
Source
from a national cancer registry in Norway (5). Higher incidence may existin iodinedeficient areasof the world (61, with reduction subsequently reported following iodin supplementation. The mechanism behind this effect of iodine deficiency is not clear. Given the diagnostic confusionin older series that combined anaplastic and medullar carcinomas along with lymphoma as one disease entity, earlier estimates of 20-30% frequency among all thyroid malignancies are probably inaccurate. Patients with anaplastic carcinoma typically present two decades older than those with differentiated carcinomas. Among nine recent series describing 473 patients with anaplastic carcinoma, the overall weighted meanat diagnosis age was66 years (Table 1) In the five studies that provided sufficient age data, only 7.5% of the patients were diagnosed before the age50of years. Women comprised 60%of all anaplastic carcinoma patients in these series, a frequency generally lower than that reported for differ 15,700 patients with thyroid carcinomas. However, in one recent report describing carcinoma in the United States 68% of the 251 anaplastic carcinomas occurred in women (7). In this same study, 87% of the patients with anaplastic disease were non79% Hispanic whites, a significantly higher frequency than thefor papillary and follicu lar carcinomas (p e .005) (7). No other large study has systematically reported on the ethnic distribution of anaplastic carcinoma.
Previous Thyroid Disease In 20-30% of cases of anaplastic carcinoma, a coexisting differentiated carcinoma can be readily identified(1,8-10); a higher frequency of coexisting disease was rep following more extensive pathological examination (11). The great majority of these differentiated tumors are of papillary histology, but follicular tumors have also been reported. Nearly 10% of patients with oxyphilic (Hiirthle cell) carcinomas may d anaplastic foci(12). In most of these cases, the undifferentiated malignancy represent the larger tumor within the thyroid, andis generally the histology found in metastases (8). Cases have also been described of patients with a primary differentiated carci
321
Anaplastic Carcinoma 100 90 80 70 Q
>
‘E
60 50 40
30 20
1
lo 0
1
2
3
4
6
7
8
9
Month Fig. 1. Product-limit survival for 46 patients with anaplastic thyroid carcinoma. (Adapted from Reference 45.)
within the thyroid, with subsequent development of anaplastic foci in distant metastase (13). In 20% of patients, a clinical history of antecedent thyroid neoplasia can be obtained. It has been suggested that anaplastic carcinoma develops from more differen ated tumors as a result of one or more dedifferentiating steps, particularly the loss of the tumor suppressor protein p53 (14-17). The role of radiation to the thyroid gland as inducing anaplastic transformation has been controversial. Both I3’I and external beam radiotherapy have been implicated as associated with a greater likelihood of developing anaplastic carcinoma (18-20). However, multiple studies describing the long-term follow-upof patients treated withI3lI for differentiated carcinoma have failed to identify a higher frequency of subsequent development of anaplastic carcinoma. Evidence for a triggering event or environmental exposure leading to dedifferentiation remains elusive. Therefore, the exact mechanism leading to anaplastic transformation of differentiated carcinoma is uncertain.
CZinicaZ Presentation The clinical manifestations of anaplastic carcinoma reflect the mass effects due to growth of primary tumor in the thyroid as well as metastatic sites. Nearly all patients present with symptoms and/or signs of anenlarging primary tumor in the neck. Metastases to cervical and mediastinal lymph nodes are common, with fewer than 5%of (11).Direct extrathyroidal patients presenting with disease limited to the thyroid gland (10,21). invasion into surrounding structures can be documented 90% in upof patients to Potential sites of direct invasion can include perithyroidal fat and muscles, larynx,
322
Sherman
trachea, esophagus, great vessels of the neck and mediastinum, sternum, and vertebr 15-5096 of column. Distant metastases are found at initial disease presentation in for distant patients (1,8,9).As with differentiated carcinoma, the most common location metastases from anaplastic carcinoma is pulmonary, identified in up to 90% of patients with distant disease (8,9).Both intrapulmonary mass lesions and pleural involvement can be seen. Less commonly found are metastases to bone (5-15%) and brain (5%), and more rarely skin, liver, kidneys, stomach, pancreas, heart, tonsil, small bowel, mesentery, and adrenal glands (8-10,22-26~). In rare instances, patients have presen of evaluation without clinical evidenceof a thyroid tumor, but are identified as a result (either pre- or postmortem) for a metastatic undifferentiated carcinoma of unknown primary (11). The primary clinical manifestation of anaplastic carcinoma is generally a rapidly enlarging neck mass, reported by about 75% of patients. Often, the speed of tumor growth can be documented by marking the cutaneous outlineof the tumor on a daily basis. In up to half of patients, a goiter has been previously recognized, A history of previous thyroid surgery, either for differentiated carcinoma or an apparently benign tumor, can often be elicited. Due to the enlarging goiter, patients often complain of symptoms from compression or invasion of the upper aerodigestive tract. Dyspneais reported by about36% of patients, followed by dysphagia in 30%, hoarseness in 28%, cough in26%, and neck pain in 17%. One caseof dyspnea has been described seconda to an intralaryngeal metastasis functioning like a ball-valve, obstructing airflow duri inspiration (27). Lesscommonly,patientsnotehemoptysis,chestpain,bonepain, headache, confusion, or abdominal pain from metastases (22,26a,28). Constitutional symptoms can include weight loss, fatigue, and fever of unknown origin (28-31). Rarely, rapid growth of the primary tumor can cause a nonspecific thyroiditis, with symptoms of thyrotoxicosis due to follicular disruption andofrelease preformed thyroid hormone (22,32,33). On physical examination, the goiter is typically quite hard, often nodular, and gen ally enlarged bilaterally. Softer, fluctuant masses have been associated with focal necrosis (11).However, anaplastic carcinoma can also present as a solitary nodule or a diffuse nonnodular goiter. Often, the neck mass is noted to be fixed to surrounding or underlying structures and does not move with deglutition. Byofthe presentation, time 5 cm in diameter, but exact measurement the primary tumor is commonly larger than are oftendifficulttoobtaingivenindistinctborders of thetumor.Metastatic adenopathy may be detected on physical examination in either the neck or axillae, although half of patients will not have palpable adenopathy. Other findings of local diseaseinvolvementcanincludestridorduetotrachealcompressionorinvasion, trachealdeviation,vocalcordparalysisduetolaryngealinvasionorinvolvement of recurrent laryngeal nerves, and venous dilatation and superior vena cava syndrom from retrosternal tumor growth. Cutaneous findings can include ulceration, atrophy, or erythema of the skin overlying the primary tumor, as well as metastatic nodules (11,34). Focalneurologicalabnormalitiesmay onthechestandabdominalwalls be indicative of brain metastases. The diagnosis of anaplastic carcinoma is usually established by tissue examination, and is often available from cytopathological and electron-microscopic review of fine 35). Laboratory testingis rarely of diagnostic needle aspiration specimens (see Chapter
Anaplastic
Carcinoma323
value in patients with anaplastic carcinoma. Serum thyroid hormone and TSH levels are generally normal except in the rare cases of thyrotoxicosis due(22,32,33). to necrosis Increased serum levels of thyroglobulin can be seen, although it is likely that coexisting differentiated tumor cells are the source rather than the anaplastic component of the tumor. By contrast, serum concentrations of the neuroendocrine markers calcitonin, carcinoembryonic antigen, and neuron-specific enolase are usually(35). normal Nonspecific markers of systemic illness can be mildly abnormal, such as an elevated erythro sedimentation rate, elevated serum C-reactive protein, anemia, and hypoalbuminemia. Hypercalcemiainthe absence of obvious osseous metastases can be secondary to (36,37).Marked leukocytosis has been secretion of parathyroid hormone-related peptide of granulocyte macrophagereported in several patients in whom elevated serum levels colony stimulating factor (GM-CSF) or granulocyte colony stimulating factor (G-CSF) were found (25,31,36,38,39). Diagnostic imaging is useful for defining the extent of disease, planning therapy, and monitoring response to treatment. Computed tomography (CT)of the neck and mediastinum can accurately identify tumor invasion of great vessels and upper aerodigestive tract structures and is superior to palpation for detection of pathological adenopathy (40). Typical findings include masses that are isodense or slightly hyperdense relative to skeletal muscle, dense calcifications, and frequent necrosis. Similarly, neck ultrasonography can accurately identify pathological involvement of locoregional nodes; both ultrasound imaging and CT can help to guide fine needle aspiration to solid, nonnecrotic tumor for diagnosis (41,42). Although ultrasound imaging cannot distinguish benign from malignant intrathyroidal tumors, as both tendto produce hypo(43). echoic lesions, extrathyroidal invasion can support the diagnosis of carcinoma Routine chest radiographs can readily diagnose most instances of pulmonary metastases, given the typical macronodular appearance of these lesions( I ) . In patients with bony metastases, skeletal radiographs demonstrate lytic lesions. Scintigraphic imaging with radioiodine or pertechnetate usually reveals hypofunctioning or “cold” foci corresponding to palpable tumor ( I ) . In the setting of thyrotoxicosis due to necrotic thyroiditis, depressed radioiodine uptake has been reported(32). 67Ga imaging has been reported to demonstrate positive uptake in 16 of 19 cases of anaplastic. carcinoma, with one false negative and two equivocal images (44).Marked positivity was also noted in lymphoma, chronic thyroiditis, and metastases to the thyroid from other malignancies, but was absent from all 19 differentiated carcinomas. However, 67Ga uptake in meta foci may not be sufficiently sensitive to supplant radiographs and CT (44).
REFERENCES 1. Ne1 CJ, van Heerden JA, Goellner J R , Gharib H, McConahey W M , Taylor WF, Grant study of 82 cases. Mayo Clin CS. Anaplastic carcinomaof the thyroid a clinicopathologic Proc 1985; 6051-58. 2.Rosai J, Saxn EA, Woolner L. Session 111: undifferentiatedandpoorlydifferentiated carcinoma. Semin Diag Patholl985; 2:123-136. 3. Rosai J, Carcangiu ML, DeLellis RA. Tumors of the thyroid gland. In Rosai J, editor. Atlas of tumor pathology, 3rd Ser, Fasc 5. Washington DC: Armed Forces Institute of Pathology, 1992. 4. Mazzaferri EL. Undifferentiated thyroid carcinoma and unusual thyroid malignancies.In
324
Sherman
Mazzafem EL, SamaanNA,editors.Endocrinetumors.Boston:BlackwellScientific Publications, 1993; 378-398. 5. Akslen LA, Haldorsen T, Thoresen S, Glattre E. Incidence of thyroid cancer in Norway 1970-1985. A P M I S 1990; 98549-558. 6. Harach HR, Escalante DA, Onativia A, Lederer Outes 3, Saravia Day E, Williams ED. Thyroid carcinoma and thyroiditis in an endemic goitre region before and after iodine prophylaxis. Acta Endorcinol (Copenh) 1985; 108:55-60. 7. Gilliland FD, Hunt WC, Morris DM, Key CR. Prognostic factors for thyroid carcinoma: a population-based study of 15,698 cases from the Surveillance, Epidemiology and End Results (SEER) program 1973-1991. Cancer 1997; 79:564-573. 8. Carcangiu ML, Steeper T, Zampi G, Rosai J. Anaplastic thyroid carcinoma: a study of 70 cases. Am J Clin Path01 1985; 83:135-158. Samaan Anaplastic 9. VenkateshYSS, Ordonez NG, Schultz PN, Hickey RC, Goepfert H, NA. carcinomaof thethyroid a clinicopathologic study of 121cases. Cancer 1990; 66:321-330 10. Tan RK, Finley RK, Driscoll D, Bakarnjian V, Hicks WL, Jr., Shedd DP. Anaplastic carcinoma of the thyroid a 24-year experience. Head Neck 1995; 17:41-47. 11. Aldinger KA, Samaan NA, Ibanez ML, Hill CSJ. Anaplastic carcinoma of thethyroid a reviewof 84 cases of spindle and giant cell carcinoma of the thyroid. Cancer 1978; 41:2267-2275. AA, Schultz PN, Ordonez NG, Sherman SI. Prognostic clinicopat 12. Chiu AC, Oliveira features in Htirthle cell neoplasia. Thyroid 1996;6529. 13. Moore Jr. JH, Bacharach B, Choi HY. Anaplastic transformation of metastatic follicular carcinoma of the thyroid. J Surg Oncol 1985; 29:216-221. E, Karakawa K, Fujita S, et al. p53 gene mutations 14. Nakamura T, Yana I, Kobayashi T, Shin associated with anaplastic transformation of human thyroid carcinomas. Jpn J Cancer R 1992; 83~1293-1298. 15. It0 T, Seyama T, Mizuno T, TsuyamaN, Hayashi T, Hayashi Y, et al. Unique association of p53 mutations with undifferentiated but not with differentiated carcinomas of the thyroid gland. Cancer Res 1992; 52:1369-1371. 16. It0 T, Seyama T, Mizuno T, Tsuyama N, Hayashi Y, Dohi K, et al. Genetic alterations in thyroid tumor progression: association with p53 gene mutations. Jpn J Cancer Res 1993; 84:526-531. 17. Moretti F, Farsetti A, Soddu S, Misiti S, Crescenzi M, Filetti S, et al. p53 re-expression inhibits proliferation and restores differentiationof human thyroid anaplastic carcinoma cells. Oncogene 1997; 14:729-740. S, Zeman V. Early manifestation of anaplastic 18. Nmec J, Nierdle B, Cenkova V, Vana carcinomaafterradioiodinetreatmentfortoxicnodulargoiter.Neoplasma1971; 18~325-333. 19. Gtaz EP, Shimaoka K. Anaplastic carcinomaof the thyroid in a population irradiated for Hodgkin disease, 1910-1960. J Surg Oncol 1979; 12:181-189. 20. Komorowski RA, Hanson GA, Garancis JC. Anaplastic thyroid carcinoma following lo dose irradiation. Am J Clin Pathol 1978; 70:303-307. 21. TennvallJ, Lundell G, Hallquist A, Wahlberg P, Wallii G, TibblinS. Combined doxorubi cin, hyperfractionated radiotherapy, and surgery in anaplastic thyroid carcinoma: report on two protocols. Cancer 1994; 74:1348-1354. 22. Nishiyama R H , Dunn EL, Thompson NW. Anaplastic spindle-cell and giant-cell tumors of the thyroid gland. Cancer 1972; 30:113-127. 23. Hadar T, Mor C, Har-El G, Sidi J. Anaplastic thyroid carcinoma metastatic to the tons J Laryngol Otol 1987; 101:953-956. 24. Phillips DL, Benner KG, Keeffe EB, TraweekST. Isolated metastasis to small bowel fr anaplastic thyroid carcinoma: with a review of extra-abdominal malignancies that sprea to the bowel. J Clin Gastroenterol 1987; 9563-567.
Carcinoma325
Anaplastic
25. Murabe H,Akamizu T,Kubota A, Kusaka S. Anaplastic thyroid carcinoma with prominent cardiac metastasis, accompaniedby a marked leukocytosis with a neutrophilia and high 1992; 3 1:1 107-1 11 1. GM-CSF level in serum. Intern Med 26. Hadar T,Mor C, Shvero J, Levy R, Segal K. Anaplastic carcinomaof the thyroid. Eur J Surg Onc01 1993; 19511-516. 26a. Chiu AC, Delpassand ES, Sherman SI. Prognosis and treatment of brain metastases in thyroid carcinoma. J Clin Endocrinol Metab 1997; 82:3637-3642. 27. Lee WC, Walsh R M . Anaplastic thyroid carcinoma presenting as a pharyngeal mass with . ball-valve type obstruction of the larynx. J Laryngol Otol 1996; 110:1078-1080. 28. Lip GY, Jaap A J , McCruden DC. A presentation of anaplastic carcinomaof the thyroid with symptomatic intra-abdominal metastases. Br J Clin Pract 1992; 46:143-144. 29. Glikson M, Feigin RD, Libson E, Rubinow A. Anaplastic thyroid carcinoma in aretrosternal goiter presenting as feverof unknown origin. Am J Med 1990; 88:81-82. 0,Baglin A. Anaplastic cancerof the 30. Hanslik T, Gepner P, Franc B, Baglin AC, Bletry thyroid gland disclosed by prolonged fever or hyperleukocytosis: two cases. [Letter]. Ann Med Interne (Paris) 1996; 147:122-124. 31. ChangTC,LiawKY,KuoSH,ChangCC,Chen FW. Anaplastic thyroid carcinoma: review of 24 cases, with emphasis on cytodiagnosis and leukocytosis. Taiwan I Hsueh Hui Tsa Chih 1989; 88551-556. 32. Murakami T,Noguchi S, Murakami N, Tajiri J, Ohta Y. Destructive thyrotoxicosis in a patient with anaplastic thyroid cancer. Endocrinol Jpn1989; 36:905-907. 33. Oppenheim A, Miller MyAnderson GH, Jr, Davis B, SlagleT. Anaplastic thyroid cancer presenting with hyperthyroidism. Am J Med 1983; 75702-704. 34. Barr R, Dann F. Anaplastic thyroid carcinoma metastatic to skin. J Cutan Path01 1974; 1:201-206. B. Miscellaneous tumorsof the thyroid. In Braverman LE, Utiger 35. Schlumberger M, Caillou RD, editors. Werner and Ingbar’s the thyroid, 7th Philadelphia: Lippincott-Raven, 1996~961-965. 36. Yazawa S , Toshimori H, Nakatsuru K, Katakami H, Takemura J, Matsukura S. Thyroid anaplastic carcinoma producing granulocyte-colony-stimulating factor and parathyroid hormone-related protein. Intern Med 1995; 34584-588. 37. Takashima S , Morimoto S, Ikezoe J, Kokado Y, Kozuka T. Occult anaplastic thyroid carcinoma associated with marked hypercalcemia. J Clin Ultrasound 1990; 18:438-441. I, Nonomura A, et al. Anaplastic thyroid 38. Iwasa K, Noguchi M, Mori K, Ohta N, Miyazaki carcinoma producing the granulocyte colony stimulating factor (G-CSF): report of a case. Surg Today 1995; 25158-160. 39. Oka Y, Kobayashi T, FujitaS , Matsuura N, Okamoto S , Asakawa H, et al. Establishment of a human anaplastic thyroid cancer cell line secreting granulocyte colony-stimulating factor in response to cytokines.In Vitro Cell Dev Biol Anim 1993; 29A:537-542. S , Kobayashi T, Koyama H, et al. CT evaluation 40. Takashima S , Morimoto S, Ikezoe J, Takai of anaplastic thyroid carcinoma.AJR 1990; 154:1079-1085. 41. Gatenby RA, Mulhern CB, Jr., Richter WP, MoldofskyPJ.CT-guidedbiopsy for the detection and stagingof tumors of the head and neck.AJNR 1984; 5:287-289. J W , James EM, Grant CS, Hay ID. US-guided biopsy 42. Sutton RT, Reading CC, Charboneau of neck masses in postoperative management of patients with thyroid cancer. Radiology 1988; 168~769-772. 43. Leisner B. Ultrasound evaluationof thyroid diseases. Horm Res 1987; 26:33-41. 44. Higashi T, It0 K, Nishikawa Y, Everhart FR, Ozaki 0, Manabe Y, et al. Gallium-67 imaging in the evaluationof thyroid malignancy. Clin Nucl Med 1988; 13:792-799. 45. Sherman SI, Brierley JD, Sperling My AinKB, Bigos ST, Cooper DS, et al. Prospective multicenter study of treatment of thyroid carcinoma: Initial analysis of staging and outcom Cancer 1998; 83(5):1012-1021.
ed.
35 Anaplastic Carcinoma Pathology James Oertel and Yolanda Oertel
Anaplastic carcinoma (undifferentiated carcinoma)is now uncommon, is extremely malignant, and is usually fatal (1-4). The thyroid gland often has been enlarged for years, containing multiple nodules or a low-grade, well-differentiated carcinoma that has grown slowly. These cancers usually infiltrate the thyroid parenchyma and the juxtathyroidal tis Metastases to regional lymph nodes and the lungs are common. The neoplastic tissue is pale, firm to hard, and opaque. Fociof hemorrhage and necrosis are frequent; these parts are soft. Extensive dense fibrosis may be evident grossly. Foci of calcification are rare. Occasionally there are regions of metaplastic cartilageandor bone. Varied histological patterns are present: 1) rounded to irregular medium-sized to giant cells with eosinophilic cytoplasm and large or giant nuclei (often bizarre), 2) fusiform (spindle) cells in a fascicular or storiform pattern (Fig. l), and 3) mediumsized to large cells with squamoid characteristics (Fig. 2). Some of these cells may have clear cytoplasm. The neoplastic giant cells may have a single nucleus or may be multinucleated. These various cellular types may be mixed together, and transitional forms can be seen. Suggestions of an alveolar or trabecular pattern may be evident (5). Bizarre nuclei, often vesicular, are common; large nucleoli may be present. Mitotic figures are numerous, and some are atypical. “Osteoclast-type” giant cells of histiocyti origin are present in a few of the tumors (6) (Fig. 1). Neoplastic cellsmay replace portionsof vessel walls, and small clusters of neoplastic cells may extend into individual thyroid follicles. Polymorphonuclear leukocytes may infiltrate the tumor and sometimes are numerous near the necrotic regions. A paucicellular variant has been reported(7). Coagulative necrosis and fibrosis are extensive.Thesparseneoplasticcellshaveatypicalnuclei.Suchtumorsresemble Riedel’s fibrous thyroiditis, especially when chronic inflammatory cells are scattered through the neoplasm. Regardless of the types of cells present, some undifferentiated carcinomas are associated with large amounts of hyalinized fibrous tissue, sometimes as dense nodules (8). Immunoreactive thyroglobulin typically is absent, and when present is evident only in someof the larger “epithelioid” cells. Immunoreactive keratin may be demonstrated, ($9).Immunoreacand is the most common marker suggesting epithelial characteristics From:Thymid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wnrtofsky 8 Humnnn Press Inc., Totmn, N]
327
328
Oertel and Oertel
Fig. 1. Undifferentiated carcinoma. Spindled cells and osteoclast-type giant cells are pre (H&E stain; x150.)
Fig. 2. Undifferentiated carcinoma. Part of the neoplasm has a “squamoid” appearance. (H&E stain; x150.)
Pathology
329
Fig. 3. Undifferentiated carcinoma. Hypercellular aspiratewith marked variation in the size of the neoplastic cells. Multinucleated cells are evident. (Diff-Quik’ stain; x200.)
tive vimentin often is detected and may be expressed in the same cells as keratin. Interpretation of such findings can be difficult because both normal thyroid epithelium and remnants of well differentiated carcinoma (or a benign nodule) may be trapped within the aggressive neoplasm.Also, the neoplastic cellsmay absorb these substances nonspecifically from neighboring thyroid tissue. Evidence of a previous nodular goiter or a follicular or papillary carcinoma often may be found if multiple sections of the neoplasm are taken (3,10). The tumors with substantial spindle-cellor giant- and spindle-cell components may as be mistaken for soft tissue neoplasms, although usually they are not recognizable one of the well-characterized sarcomas. Most of the so-called “small cell carcinomas” diagnosed in the past were malignant lymphomas.Small-cellcarcinomaalmostcertainlyexists,but it is rare.Extensive study of such a lesion may demonstrate foci of well-differentiated carcinoma, poorly differentiated (insular) carcinoma, or medullary carcinoma. Therefore, critical analysis suggests that only a few small cell cancers exist and belong with the undifferentiated carcinomas (11-14). Inmiddle-aged or elderly patients a portion of an otherwise well-differentiated carcinoma may be anaplastic carcinoma. This has grave prognostic implications. If such a focus is only a few millimeters in diameter, it may have little effect on the patient’s long-term survival, but in some patients this is not true, unfortunately (15). The same applies for a tiny anaplastic carcinoma discovered in a thyroid removed for multinodular goiter. or necrosis and hemorrhage (Fig. 3), Cytological smears may show marked cellularity depending on the part of the mass sampled (16,17). Leukocytes may be numerous.
Oertel and Oerfel
330
: iI” .-
I
Fig. 4. Undifferentiated carcinoma. Hypercellular aspirate with osteoclast-type cell onthe right. @iff-Quik@ stain;x400.)
Spindled cells and giant cells are present (18);the latter may be multinucleated histio cytes(Fig. 4) or (moreoften)bizarreneoplasticcellswithoneorseveralnuclei. Abnormal mitotic figures may be seen. The smears may show cells from a follicular neoplasm or a papillary carcinoma if one coexists with the anaplastic carcinoma (19), and therefore this possibility illustrates the requirement that several aspirations should be performed when a fast-growing mass is present. Also, these neoplasms may be hemorrhagic or fibrotic,so the epithelial cells may be sparse or may be diluted by bloo
REFERENCES
1. Nishiyama R H , Dunn EL, Thompson NW. Anaplastic spindle-cell and giant-cell tumors of the thyroid gland. Cancer 1972; 30:113-27. 2. LiVolsi VA. Surgical pathology of the thyroid. Major Rob1 Path01 1990; 22:253-274. 3. Rosai J, CarcangiuML, DeLellis RA. Tumors of the thyroid gland.(In Rosai 3, Sobin LH editors. Atlas of tumor pathology. 3rd Ser Fasc 5). Washington, DC: AFIP, 1992. 4. Tan R K , Finley III RK, Driscoll D, Bakamjian V, Hicks Jr WL, Shedd DP. Anaplastic carcinoma of the thyroid: a 24-year experience. Head Neck 1995; 17:41-48. 5. Hurlimann JH, Gardiol D, Scazziga B. Immunohistology of anaplastic thyroid carcinoma a study of 43 cases. Histopathology 1987; 11567-580. 6. Gaffey MJ, Lack EE, Christ ML, Weiss L. Anaplastic thyroid carcinoma with osteoclas like giant cells: a clinicopathologic, immunohistochemical, and ultrastructural study. Am J Surg Pathol 1991; 15160-168. 7. Wan S-K, Chan JKC, Tang S-K. Paucicellular variant of anaplastic thyroid carcinoma: a mimic of Riedel’s thyroiditis. Am J Clin Path01 1996; 105388-393. 8. Chetty R, Mills AE, LiVolsi VA. Anaplastic carcinomaof the thyroid with sclerohyaline nodules. Endocr Path01 1993; 4:llO-114.
Pathology
331
9. Ord6liezNG,El-Naggar
AK, HickeyRC,SamaanNA.Anaplasticthyroidcarcinoma: immunocytochemical study of 32 cases. Am J Clin Path01 1991; 96:15-24. JL, Tsang RW, Asa SL. The associationof well-differentiated 10. van der Laan BFAM, Freeman thyroid carcinoma with insular or anaplastic thyroid carcinoma: evidence for dedifferentiation in tumor progression. Endocr Path011993;4:215-221. 11. Cameron RG, Seemayer TA, Wang N-S, Ahmed MN, Tabah ET. Small cell malignant study. Patholl975; 6:73 1-740. tumors of the thyroid: a light and electron microscopic Hum 12. Luna MA, Mackay B, Hill CS, Hussey DH, Hickey RC. The quarterly case: malignant small cell tumor of the thyroid. Ultrastruct Path01 1980; 1:265-270. cell neoplasmsof the thyroid: an immunoperoxidase 13. Mambo NC, Irwin SM. Anaplastic small study. Hum Pathol 1984; 15:55-60. 14. Wolf BC, Sheahan K, DeCoste D, Variakojis D, Alpern HD, Haselow RE. Immunohistochemical analysisof small cell tumors of the thyroid gland:an eastern cooperative oncology group study. Hum Path01 1992; 23:1252-1261. 15. Aldinger KA, Samaan NA, Ibanez M, Hill Jr CS. Anaplastic carcinoma of the thyroid a reviewof 84 casesofspindleandgiantcellcarcinomaofthethyroid.Cancer 1978; 41:2267-2275. 16. SchneiderV, Frable WJ. Spindle and giant cell carcinoma of the thyroid: cytologic diagnosis by fine needle aspiration. Acta Cytol 1980; 24184-189. 17. Brooke PK, Hameed M, Zakowski MF. Fine-needle aspirationof anaplastic thyroid carci-
nomawithvariedcytologicandhistologicpatterns:acasereport.DiagnCytopathol
1994; 11~60-63. 18. Us-Krasovec M, Golouh R, Auersperg M, BesicN, Ruparcic-Oblak L. Anaplastic thyroid carcinoma in fine needle aspirates. Acta Cytol1996; 40:953-958. 19. Vinette DSJ, MacDonald LL, YazdiHM. Papillary carcinoma of the thyroid with anaplastic 1991; transformation: diagnostic pitfalls in fine-needle aspiration biopsy. Diagn Cytopathol 7~75-78.
36 Anaplastic Carcinoma Management Orlo H. Clark
Thetreatment of patientswithanaplasticthyroidcancer-likethetreatmentof patients with papillary thyroid cancer-is controversial. The reason for the controversy is that anaplastic carcinomais one of the most aggressive malignancies. Most patients with anaplastic thyroid cancer have poor prognosis regardless of treatment, and usually die of suffocation due to local tumor invasion; the median survival time is about 6 months (1,2) and the overall mortality rateis about 97% (1-3). At initial examination, (5-10 cm) fixed mass and about 30% already have distant, patients usually have a large usually pulmonary, metastases (4). Most patients with anaplastic thyroid cancer also have well-differentiated thyroid cancer (4-7). It appears that1%of differentiated thyroid cancers transform to anaplastic cancers (8). Someof these tumors demonstrate progression from well-differentiated to insular to anaplastic cancer (9).Serial transplantation of differentiated thyroid tumors are more also leads to anaplastic transformation (10,11). Anaplastic thyroid cancers likely to have p53 and PDGF mutations than do differentiated thyroid cancers (12-14). Anaplastic thyroid cancers occur most often in older patients especially in areas of endemic goiter (15). Iodine deficiency is an important factor, since anaplastic cancer is decreasing in the United States despite our aging population. Radiation has also been implicated as a causative agent, although in more than 70% of patients with anaplastic (16,17). Once anaplastic thyroid cancer has become cancer no radiation has been given recognized, curative treatment is unlikely, so that prevention of endemic goiter, and diagnosis by fine needle aspiration biopsy and removal of suspicious differentiated thyroidnodules is recommended.Evenwhenanaplasticthyroidcancersarefound incidentally when removing a differentiated thyroid cancer, the outcome is guarded. Patients who have tumors that can be completely resected, however, have a slightly better prognosis (18-20). Most patients with anaplastic thyroid cancer are not difficult to diagnose. Theyare usually older patients, and80% report a long history of goiter or a thyroid nodule (1,2). The thyroid goiter or nodule suddenly begins to grow rapidly and patients develop pain, dysphagia, and/or hoarseness. Some patients may experience symptoms of hyperthyroidism with pseudothyrotoxicosis and can be misdiagnosed as having subacute
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., Totown, NJ
333
334
CZark
thyroiditis. Fine needle biopsyis usually definitive, althoughtumor cells may be scant in large tumors because of hemorrhagic necrosis. Because of the large size of these tumors at presentation, CT or MRI scanning is recommended to document the extent of the disease. It can also determine if there is TSH receptors or intratracheal growth. Although most of these tumors do not have take up radioiodine or make thyroglobulin, some tumors do as they originated from differentiated tumors, so that documenting thyroid function and serum thyroglobulin levels is also recommended. Once the diagnosis has been made by cytological examination or by open biopsy, treatment with multimodality therapy seems indicated because the results of other (19), treatments are dismal.A cooperative prospective study has been done in Sweden where 33 patients were treated for a rapidly enlarging thyroid mass with Adriamycin to both enhance the external radiation therapy and alsoto perhaps limit the growthor speed of disseminated cancer. After 4 weeks of combined radiation and chemotherapy as much thyroid and tumor should be removed as can be done safely. Complications should be avoided as most patients are receiving palliative rather than curative ther From a technical point of view when doing thyroidectomy in these patients, in c to other patients with thyroid cancer, one should usually remove the least involved this portion of the thyroid lobe first, since this orients the surgeon to the trachea. Once gland has been removed, removing the more involved side often become easier. In the Swedish study of patients with anaplastic thyroid cancer, 29 of the 33 patients (19). had the diagnosis establishedby fine needle aspiration and cytologic examination No patients failed to complete the protocol because of toxicity. Definitive resection or debulking was donein 23 of the 33 or 70% of patients. After thyroidectomy, radiation, 2 more weeks. To date, complete local and chemotherapy were readministered for control was obtainedin 16 of 33 (48%)of the patients and four patients had no evid of disease at2 years. Only 8 of 33 (24%) patients diedof local failure. The researchers reported report that debulking surgery appeared to be a prerequisite for local as control, in other studies (20).
REErERENCES 1. Aldinger KA, Samaan NA, Ibanez M, Hill CS Jr. Anaplastic carcinoma of the thyroid a
review of 84 cases of spindleandgiant cell carcinoma of thethyroid.Cancer1978; 41:2267-2275. 2. Samaan NA, Ordonez NG. Uncommon types of thyroidcancer.EndocrinolMetabClin North Am. 1990; 19:637-648. 3. Junor EJ, Paul J, Reed NS. Anaplastic thyroid carcinoma: 91 patients treated by surgery and radiotherapy. Eur J Surg Oncol 1992; 18:83-88. 4. Nicolosi A, Addis E, Massidda B, Malloci A, MuraE,Esu S. [Anaplastic carcinoma of the thyroid our experience]. Minerva Chir 1992; 47:1161-7. 5. Ne1 CJ, van Heerden JA, Goellner JR, Gharib H, McConahey W M , Taylor WF, Grant CS. Anaplastic carcinoma of the thyroid: a clinicopathologic study of 82 cases. Mayo Clinic ROC 1985; 6051-58. 6. Carcangiu ML, Steeper T, Zampi G, Rosai J. Anaplastic thyroid carcinoma: a study of 70 cases. Am J Clin Pathol 1985; 83:135-158. 7. Nishiyama RH, DunnEL,Thompson NW. Anaplastic spindle-cell and giant-cell tumors of the thyroid gland. Cancer 1972; 30:113-127.
Surgey
335
8. Cohn KH, Backdahl M, Forsslund G, Auer G, Zetterberg A, Lundell G, Granberg PO,
LowhagenT, Willems JS, Cady B. Biologic considerations and operative strategy in papillar Surthyroid carcinoma: arguments against the routine performance of total thyroidectomy. gery 1984;96:957-971. 9. Vanderlaan BFAM, FreemanJL, Tsang RW, Asa SL. The association of well-differentiated thyroid carcinoma with insular or anaplastic thyroid carcinoma: evidence for dedifferentiation in tumor progression. Endocr Path011993; 4:215-221. 10. Ito T, SeyamaT, Mizuno T, Tsuyama N, Hayashi T, Hayashi Y,et al. Unique association of the thyroid of p53 mutations with undifferentiated but not with differentiated carcinomas gland. Cancer Res 1992; 52:1369-1371. 11. Farid N R , Shi Y,Zou M. Molecular basisof thyroid cancer. Endocr Rev 1994; 15:202-232. 12. Fagin JA, Matsuo K, Karmakar A, Chen DL, Tang SH, Koeffler HP. High prevalence of mutations of the p53 gene in poorly differentiated human thyroid carcinomas. J Clin Invest 1993; 91:179-184. 13. Heldin N E , Gustavsson B, Claesson-WelshL, Hammacher A, MarkJ, Heldin CH, Wester-
mark B. Aberrant expressionof receptors for platelet-derived growth factor in an anaplastic thyroid carcinoma cell line. Proc Nat Acad Sci USA 1988; 859302-9306. 14. Fagin JA. Tumor suppressor genes in human thyroid neoplasms: p53 mutations are associated undifferentiated thyroid cancers. J Endocrinol Invest1995; 18:140-142. 15. Williams ED. Thyroidcancer:pathologicandnaturalhistory.RecentResCancerRes 1980; 73~47-55. 16. Baker HW. Anaplasticthyroidcancertwelveyearsafterradioiodinetherapy.Cancer 1969; 23~885-890. 17. Samaan NA, Schultz PN, Haynie TP, Ordonez NG. Pulmonary metastasis of differentiated 101 patients.JClinEndocrinolMetab 1985; thyroidcarcinoma:treatmentresultsin 60:376-380. 18. Levendag PC, De Porre PM, van Putten WL. Anaplastic carcinoma of the thyroid gland treated by radiation therapy. Int J Radiat Oncol Biol Phys 1993; 26:125-128. 19. TennvallJ, Lundell G, Hallquist A, Wahlberg P, Wallin G,S. Combined Tibblin doxorubicin,
hyperfractionated radiotherapy, and surgery in anaplastic thyroid carcinoma: report on two protocols-the Swedish Anaplastic Thyroid Cancer Group. Cancer 1994; 741348-1354. JE, Droz JP, Sarrazin D. Combination 20. SchlumbergerM,Parmentier C, Delisle MJ, Couette therapy for anaplastic giant cell thyroid carcinoma. Cancer1991; 67564-566.
37 Chemotherapy of Anaplastic Thyroid Cancer Lawrence S. Lessin and Myo Min
Anaplastic carcinoma of the thyroid is among the most aggressive human cancers with a median survival of 4-6 monthsafterdiagnosis.Itisrelativelyresistantto chemotherapy alone. Shimaoka and associates ( I ) reported responseof anaplastic thyroid of doxorubicin vs doxorubicin plus cancer to chemotherapy in a randomized trial Of the 39 cisplatin in a study conducted by the Eastern Cooperative Oncology Group. patients with anaplastic thyroid cancer enrolled in the study, 21 were treated with doxorubicin alone and18 were treated with a combination of doxorubicin and cisplatin. Only 1patient showed partial response to doxorubicin alone, compared to3 complete responses and 3 partial responses in the combination arm. However, patients treated with combination chemotherapy did not have a statistically longer duration of response nor time to relapse. The investigators concluded that combination chemotherapy has an advantage over single-agent treatment in anaplastic carcinoma with higher response rate but no survival advantage. More recently, Tamura and colleagues(2) reported a study of the Japanese Societyof Thyroid Surgery in which17 patients with anaplastic carcinoma were treated with a regimen of infusional cisplatin, and bolus doxorubicin 1-5, on day 1, bolus etoposide days1-3 and peplomycin (a bleomycin analogue) days with G-CSF support. Of 10patients with measurable lesions, 2had brief partial responses lasting 2-3 months. All patientsexperiencedmajorneutropeniadespiteG-CSF treatment. Because single modality treatment with either radiotherapy, surgery or chemotherap does not affect survival, a number of investigators have assessed the potential synergy of chemotherapy and radiation, often combined with surgical debulking. Studies of chemoradiotherapy for anaplastic thyroid carcinoma, usually given concomitantly, are summarized in Table 1. COMBINED CHEMOTHERAPY AND RADIATION THERAPY Combined chemoradiotherapy for the treatment of anaplastic thyroid cancer has been reported since the 1970s. Tennvall and Tallroth and their collaborators(3-7) reported aseries of papersrelatingtheirexperiencewithcombinedmodalitytreatmentfor anaplastic thyroid cancer. In the early 1970s, combined chemotherapy and radiation therapy employing single-agent methotrexate was tested, achieving 7 responses in 8
From: Thyroid Cancer: A Comprehensive Guide to Cliical Management Edited by:L. Wartofsky 0 Humana Press Inc., Totowa, NJ
337
Chemotherapy
339
patients, but none survived their disease. Severe side effects (mucositis, cytopenias) were encountered, and thus the chemotherapy regimen was changed to a combination 5FU (BCF) (5). With this regimen, 7 of 9 patients of bleomycin, cyclophosphamide, and had aresponseand1was operated on after completion of chemoradiotherapy and survived for 14 years without recurrence. Remissions observedin other patients were transient. In view of this experience, surgery was incorporated early in the treatment protocol using the same combination chemotherapy and hyperfractionated (twice daily) radiotherapy (5). Twenty patients were treated and 75% achieved a response of which 3 remained in complete remission for more 6than years. Again, combination chemotherapy with BCF combined with hyperfractionated radiation therapy produced severe local and systemic toxicity. In 1983, Kim and Leeper (8) reported a combined modality regimen with low-dose doxorubicin (10 mg/m2/week) as bolus injection before hyper8 out of 9 patients treated fractionated radiation therapy. Their initial report stated that achieved complete remission at primary tumor sites and 6 out of 8 patients remained free of local disease in the neck through the time of death from distant metastases.In thisstudy,only1 of 9patientshadcompletesurgicalresectionandtheresthad only biopsy or partial resection. Due to the low dose of doxorubicin used and the hyperfractionation of radiation therapy, both acute and late-phase normal tissue toxicity (9) began a prospective study of therapy of were low. After 1983, Kim and Leeper anaplastic thyroid cancer and published an updated report of their data in 1987 after treatment of 19 patients. Complete response was achieved in 84% and local control rate was 68% at 2 years. Median survival was about 1 year which was significantly longer than the 4-month median survival observed in historical controls. These authors also noted that if tumor volume exceeded more than 200 cm3 at the time of start of radiation, there was no significant response to combined modality treatment.Kim After and Leeper published their findings in 1983, the Tennvall-Tallroth group reported 16 patients treated with low-dose doxorubicin and radiation insteadof the BCF regimen, used both before and after surgery, and confirmed the improved efficacy and reduced toxicity (3,5). DRUG RESISTANCE
Anaplastic thyroid cancer is relatively resistant to chemotherapy. The mechanism (10) who assessed expresof drug resistance was studied by Yamashita and coworkers sion of the MDR-1tumor resistance gene and its p-glycoprotein gene product in relatio to chemotherapy response. Anaplastic thyroid carcinoma showed low expression of MDR-1 and no relationship between response to chemotherapy and MDR-1 expression orp-glycoprotein was found. Asakawa and colleagues (11)utilized an in vitro chemosen14 patients. These assays demonstrated sitivity assay on anaplastic thyroid cancers from chemoresistance to doxorubicin, cisplatin, etoposide, cyclophosphamide and carboplat in the majority of tumors. Only one patient had in vitro sensitivity to doxorubicin, no clinical response was seen; none of the in vitro-resistant patients had a clinical respo to chemotherapy. The investigators suggest that in vitro chemosensitivity testing may prevent administration of ineffective chemotherapy. Recently, paclitaxel (Taxol) has shown significant antineoplastic activity against anaplastic carcinoma cell lines and tumor xenografts (12). (See Chapter 39.)
340
Less
In summary, anaplasticthyroid cancer, although relatively chemoresistan response to combined modality therapy, including surgery and chemorad in patients with distant metastases, local control with combined treatment distressing upper airway obstruction and improve the quality of life.
REFERENCES
1. Shimaoka K, Schoenfeld D, De Wys W, Creech R, De Conti R. A random doxorubicin vs. doxorubicin plus cisplatin in patients with advanced thyroi Cancer 1985; 56:2155-2 160. 2. Tamura K, Shimaoka K, Mimura T, Sugenoya A, Noguchi S. Intensive c for anaplastic thyroid carcinoma: combination of cisplatin, doxorubicin, e peplomycin with granulocyte-colony stimulating factor. Jpn J Clin Oncoll995; Also, Proc Annu Meet Am SOCClin Oncol 1996; 15A:906. 3. Tennvall J, Lundell G, Hallquist A, Wahlberg P, Wallin G, Tibblin S, and Anaplastic Thyroid Cancer Group. Combined doxorubicin, hyperfractionated and surgery in anaplasticthyroid carcinoma: report on two protocols. Cancer 19 1354. 4. Tennvall J, Anderson T, Aspengren K, et al. Undifferentiated giant and spind noma of the thyroid: report on two combined treatment modalities. Acta R 1979; 18:408. 5. Tennvall J, Tallroth F, Hassan E, Lundell G, Akerman M, Biorklund A, et a thyroid carcinoma-doxorubicin, hyperfractionated radiotherapy and surgery 1990; 29:1025-1028. 6. Tennvall J, Tallroth F, Hassan E, Lundell G, Akerman M, Biorklund A, et a thyroid carcinoma-doxorubicin, hyperfractionated radiotherapy and surgery 1990; 29:1025-1028. 7. Tennvall J, Lundell G, Hallquist A, Wahlberg P, Wallin G, Tibblin S, and Anaplastic Thyroid Cancer Group. Combined doxorubicin, hyperfractionated and surgery in anaplasticthyroid carcinoma: report on two protocols. Cancer 19 1354. 8. Kim JH, Leeper RD. Treatment of anaplastic giant and spindle cell carci thyroid gland with combination adriamycin and radiation therapy: a new appr 1983; 52~954-957. 9. Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with doxorubicin and radiation therapy. Cancer 1987; 60:2372-2375. 10. Yamashita T, Watanabe M, Onodera M, Shimaoka K, Ito K, Fujimoto Y, et resistance gene and p-glycoprotein expression in anaplasticcarcinoma of the th Detect Prevent 1994; 18:407-413. 1 1. Asakawa H, Kobayashi T, Komoike Y, Maruyama H, Nakano Y, Tamaki Y, et a sitivity of anaplasticthyroid carcinomaand poorly differentiatedthyroid carcino cer Res 1997; 17:2757-2762. 12. A n KB, Tofiq S, Taylor KD. Antineoplastic activity of Tax01 against hum thyroid carcinoma cell lines in vitro and in vivo. J Clin Endocrinol Metab 19 3653. 13. Schlumberger M, Parmentier C, Delisle MJ, Couette JE, Droz JP, Sarrazin D. therapy for anaplastic giant cell thyroid carcinoma. Cancer 1991; 67564-566
38 Management of Anaplastic Carcinoma External Radiation Therapy Robert L. White and Leonard Wartofsky
An excellent recent review of anaplastic carcinoma includes a brief overview of management of this highly aggressive tumor with external radiation therapy( I ) . Anaplastic carcinoma is the least radiosensitive of the thyroid neoplasms. The giant cell variety shows little response to external irradiation, while the small-cell variant is more radiosensitive (2). Megavoltage external radiation may be utilized after needle aspiration or simple biopsy to establish the diagnosis. Surgical excision is recommended to remove or debulk as much of the neoplasm as is possible. However, it is extremely difficult for the surgeon to leave a thyroid area completelyfree of this tumor. In one series of 43 patients with distant metastases, it was noted that only one had virtually all of the (3).Maximal control of any local disease tumor removed at the original thyroidectomy in the neck by a .combination of extensive surgery followed by external radiation is criticaltosurvival. A tracheostomymaybenecessarybeforestartingtheexternal irradiation to provide an adequate airway during therapy, although tracheostomy may be associated with poor local wound healing, which can lead to postponement of the radiotherapy (4). Unfortunately even with 6500 to 6000 cGy to the primary lesion, the neck and the superior mediastinum, control of the tumor is almost never accomplished, and some series of patients have indicated no increase in survival, even in the face of apparently initial good responses to radiation therapy (5). However, some very slight survival benefit may be gained by treatment, with3000 cGy or more being associated with 6 months survival in one series (6),whereas survival only averaged 2 months after lower doses. Because of the marked dedifferentiation of these tumors, they do not trap radioiodine, and as a result 13*1has not generally been used to treat anaplastic carcinoma. Usually, even the addition of chemotherapyto the treatment options has not helped in the managementof these anaplastic thyroid carcinomas, but there have been occasional as a radiosensuccesses. The use of combination therapy with adriamycin (doxorubicin) (7) of the MD Anderson sitizing agent was proposed earlier by Rogers and colleagues Clinic and was employed by Kim and Leeper as reported both in an initial trial (8) and a subsequent larger series of 19 patients with anaplastic carcinoma (9). Patients are given 5760 cGy 3 days weekly over 40 days with 320 cGy daily in a divided dose From: Thyroid Cancer: A Comprehensive Guideto c l i c a l Management Edited by: L. Wartofsky 0 Humana Press Inc., Totowa, NJ
341
342
White and Wartojiky
of 160 cGy 4 hours apart. The Adriamycinis given in a dosage of 10 mg/mz only once weekly 90 minutes before the initiation of radiotherapy. We reported a single patient with spindle cell anaplastic carcinoma using a similar protocol prior to the series published by Kim and Leeper but with Actinomycin D, and obtained a full cure (10). Indeed, the patient was reported as a possible cure at 4 years, and is currently alive and free of cancer, over 30 years later. Unfortunately, all of the patients in the series of Kim and Leeper died of their disease, but there was good local control aand median survival of 1year. Much of the discussion of external irradiation for differentiated cancer elsewhere this volume generally is applicable for these poorly differentiated thyroid tumors. Th techniques of treatment and the doses utilized are the same as for differentiated thyroid cancers, although higher doses have been given. For curative treatment for thyroid carcinoma with external megavoltage irradiation, there are many technically dem details. The definitive dose for residual or bulky thyroid carcinomais 6500 cGy in 7 weeks with a daily dose of 180-200 cGy daily 5 days a week. The treatment volume should include the entire thyroid gland, the right and left cervical lymph nodes, right and left supraclavicular nodes and the superior mediastinum (11). It is necessary to pay particular attention to the spinal cord dose. Special blocking techniques with a as well cerrobend blocking system should limit the radiation dose to the spinal cord as other radiation-sensitive structures.All of the treatment areas where microscopic o small deposits of cancer could be present should be treated with doses to 5000 cGy over 5 to 6 weeks time. The spinal cord is shielded after 4500 cGy in 4.5 to 5 weeks time. Where tissue thickness results in doses of less than 5000 cGy in 5 to 6 weeks, as uniform as possible. boosting techniques must be employed to ensure that the isdose There are several methods of radiation beam arrangements and portals that allow In most cases an anterior adequate doses to be delivered to the neck and mediastinum. to posterior and posterior to anterior set of portals with@'CO,4 or 6 MV photons will allow 4500 to 5000 cGy to be delivered in 4.5 to 6 weeks time. Boosting techniques utilizing electron ports of 8 to 14 MeV can supplement the areas treated to 4500 to 5000 cGy to definitive doses of6500 to 7000 cGy in 5 to 8 weeks time. To avoid the spinal cord, in addition to cerrobend blocking, oblique anterior portals with wedges are occasionally utilized. Some of the newer treatment techniques include arching or rotational fields with flying wedges to optimize external irradiation to the treatment volume while minimizing treatment to the spinal cord or other critical structure experience has documented that external irradiation with or without 13'1 can produce long-term local control in patients with differentiated thyroid carcinomasfor up to 25 years who have microscopic residual or gross disease after surgery (12). Obviously, since metastatic anaplastic thyroid cancer does not accumulate I3II, the only therapeutic approaches remaining are either surgical debulking or externa tage irradiation. Dose levels of3500 to 4500 cGy in 3 to 4.5 weeks are recommended for optimal palliation of metastasis to soft tissue or bone. When thereis a possibility of pathologic fracturein the caseof bone metastasis, stabilization with an intramedu rod or other orthopedic procedure should proceed the external radiation. The pro in patients with distant metastases of anaplastic thyroid carcinoma often may be measured in weeks rather than months or years. In one series of patients with metastatic disease treated with less than 3000 cGy, the survival was less than a month with no
External Radiation Rx
343
one alive at 12 months, compared to a median survival of over 3 months in those treated with > 3000 cGy, with 10% of patients alive at 12 months (3). Patients who receive systemic chemotherapy and external irradiation concurrently or sequentially generally should not be treated with daily doses to exceed 180 cGy because of the possibilityof undesirable dose potentiating side effects. Note, however, that the technique of Kim and L4eper (8,9) described above, provides 320 cGy daily in divided dosage for three days weekly and achieved fair success. Daily management for the patient receiving combinations of chemotherapy and external irradiation is difficult and requires close surveillance and observation. Usually the side effects of oral mucositis, esophagitis and skin erythemaare worse for patients treated with combined modalities and patients nekd to be carefully and cautiously observed regularly (13). Administering the radiotherapy in more frequent and smaller doses by a hyperfractionation protocol is designed to reduce potential radiation toxicity. However, Wong and 3000-4500 coworkers (14) employed a hyperfractionated regimen giving a totalofdose cGy as 100 cGy every 3 hours for 4 doses per day, 5 days/week, and noted significant two occurrence of mucositis and related toxicity as well as radiation myelopathy in patients which developed at8 and 13 months after total spinal cord doses of3990 and 4830 cGy. Tennvall and colleagues(15,16)also employed a hyperfractionated regimen togetherwithdoxorubicinradiosensitization,butadministered it in two separate sequences, one preoperatively (3000 cGy) and one course postoperatively (1600 cGy) with the doxorubicin given once weekly 60-90 minutes prior to the startof radiation treatments for the week. The results were notas promising as those seen by Kim and Leeper (91,but half of the patients achieved local control of disease, and they proposed that more complete surgical resections were facilitated by the preoperative therapies. There have been several other reports of series of patients with anaplastic carcinoma in whom various combination regimens of external radiation therapy with chemotherapeuticagents(includingdoxorubicin,cisplatin,methotrexate,bleomycin,5FU,and cyclophosphamide) have been attempted but none have either exceeded or achieved the response rates seen with the regimen of Kim and Leeper (9). Interstitial irradiation is helpful and valuable in the treatment of primary thyroid carcinomas as well as metastatic carcinoma to the thyroid from other primary sites. Removable I9*Ir and permanently implanted have been utilized in the clinical setting. In addition, '%Irhas been implanted into mediastinal masses metastatic from thyroid carcinomas and sarcomas. Since thereis minimal general experience and few patients have treated, the interstitial treatment has not been widely publicized. In experienced hands,theinterstitialirradiationtechniqueshaveproducedlongtermdiseasefree survival in patients and improved local control. The advantage of interstitial irradiation includes minimal side effects and complications and improved local responsiveness, but the clinical experience is limited,
REFERENCES 1.Ain KB. Anaplasticthyroidcarcinoma:behavior, biology, andtherapeuticapproaches. Thyroid 1998; 8:715-726. 2. Hill CS Jr, Aldinger KA: Management of anaplastic cancer of the thyroid. In Greenfield LD,editor. Thyroid cancer. Boca Raton, FL: CRC, 1978:165-176.
344
White and Wartofsk
3. Levendag PC, DePorre PMZR, van Putten WLJ. Anaplastic carcinoma of the thyroid g treated by radiation therapy. Int J Radiation Oncol Biol Phys 1993; 26:125-128. 4. Hotling T, Meybier H, Buhr H. Stellenwert der h-acheotomie in der behandlung des
torischennotfallsbeimanaplastichenschilddrusenkarzinom.WienKlinWochenschr
1990;102~264-266. 5. Junor EJ, Paul J, Reed NS. Anaplastic thyroid carcinoma: 91 patients treated by surgery and radiotherapy. Eur J Surg Oncol 1992; 18:83-88. 6. Kobayashi T, Asakawa H, Umeshita K, Takeda T, Maruyama H, Matsuzuka F, Monde M. Treatment of 37 patients with anaplastic carcinoma of the thyroid. Head Neck 1996 18:36-41. 7. Rodgers JD, Lindberg RD, Hill CS Jr, et al: Spindle giant cell carcinoma of the thyroid a different therapeutic approach. Cancer1974; 34:1328. 8. Kim JH, Leeper RD. Treatment of anaplastic giant and spindle cell carcinoma of th
thyroid gland with combination adriamycin and radiation therapy: a new approach. C
1983; 52:954-957. 9. Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with combinatio doxorubicin and radiation therapy. Cancer1987; 60:2372-2375. 10. Casterline PF, Jaques D, Blom H, and Wartofsky L. Anaplastic Giant and Spindle Cell 1980; 45:1689-1692 Carcinoma of the Thyroid A Different Therapeutic Approach. Cancer 11. Moss WT, Brand WN, Battifora H. The thyroid.In Radiation oncology: rationale, techniq results, 5th ed., St. Louis, MO: Mosby, 1979:233-242. 12. Simpson WJ, McKinney SE, Carruthers JS, Gospodarowicz MK, Sutcliffe SB, Panzarell T. Papillary and follicular thyroid cancer. Prognostic factors1578 in patients. Amer J Me 1987; 83~479-88. 13. Greenfield LD. Thyroid tumors. In Perez CA, Brady LW, editors. Principles and practic of radiation oncology. Philadelphia, JB Lippincott, 1987: 1126-1 156. 14. Wong CS, VanDyk J, Simpson WJ. Myelopathy following hyperfractionated accelerated radiotherapy for anaplastic thyroid carcinoma. Radiother Oncol1991; 20:3-9. 15. Tennvall J, Tallroth E, El Hassan A, Lundell G, Akermann M, Bjorklund al. Anaplast A, et
thyroid carcinoma: doxorubicin, hyperfractionated radiotherapy and surgery. Acta Oncol
1990; 29:1025-1028. 16. TennvallJ, Lundell G, Hallquist A, Wahlberg P, Wallin G, TibblinS. Combined doxorubic 1994 hyperfractionated radiotherapy, and surgery in anaplastic thyroid carcinoma. Cancer 7411348-1354. 17. SchlumbergerM, Parmentier C, DeLisle MJ, Couette J-E, Droz J-P, S m i n D. Combinatio therapy for anaplastic giant cell thyroid carcinoma. Cancer 1991; 67564-566. 18. Werner B, AbeleJ, Alveryd A, Bjorklund A, Franzen S, Granberg P-0, et al. Multimod therapy in anaplastic giant cell thyroid carcinoma. World J Surg 1984; 854-70. 19. Auersperg M, Us-Krasovec M, Petric G, Pogacnik A,N.Besic Results of combined moda
treatment in poorly differentiated and anaplastic thyroid carcinoma. Wien Klin Wo
1990;102~267-70. 20. Tallroth E, Wallin G, Lundell
G, Lowhagen T, Einhorn J. Multimodality treatment in anaplastic giant cell thyroid carcinoma. Cancer1987; 60:1428-1431.
39 Anaplastic Carcinoma Prognosis Steven I. Sherman
All studies that have examined the outcome of patients with anaplastic carcinoma have this disease. Product-limit estimates of demonstrated thegrim prognosis associated with 3 to 7 months, and the1- and 5-year survival median survival from diagnosis range from probabilities are 20-35% and 5-lo%, respectively (1-6). The cause of deathis related to upper airway obstruction and suffocation in 50-60% (often despite the presence of a tracheostomy) and a combination of complications of local and distant disease in the remaining patients (2,7). Examination of survival curves from these studies reveals two distinct components: a sharp initial decline for the first 18-24 months, followed 1). However, in at least one by a slower rate of death over the ensuing years (Fig. study, mostof the long-term survivors actually had lymphoma or medullary carcinoma rather than anaplastic tumors( I ) . Nonetheless, perhaps 5% of patients with anaplastic carcinoma may survive many years after initial diagnosis and treatment without evide of recurrent disease (2,6). Several important prognostic clinical parameters have been identified in retrospective studies. Among clinical features that can be evaluated at disease presentation, various univariate analyses have suggested that greater extent of disease and larger primary tumor size increase the risk of dying from anaplastic carcinoma. In the most recent M.D. Anderson Cancer Center, patients with disease series from the University of Texas initially confined to the neck had a mean survival of 8 months, with compared 3 months if disease extended beyond the neck (P e .OOl) (8). These results are consistent with other reports that patients with disease either confined to the thyroid in locoregional or (I,IO,II). Patients whose metastases survived longer than those with distant metastases primary tumor was less than 6 cm in maximum diameter have been reported to have a 25% 2-year survival, compared with 3-15% for those with tumors larger than6 cm (I,Z). Other prognostic variables that may also predict worse prognosis include older age at diagnosis, male gender, and dyspnea as a presenting symptom. In one study, patients who survived longer than 2 years after diagnosis of anaplastic carcinoma had an average age at diagnosis of only 54 years, significantly younger than the 64 years for the group who died before 24 months ( P e .Ol) (8), but no significant effect of age was noted in other studies (I,Z,IZ,I3). A 3-fold longer survival was noted for
F m : Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofshy 0 Humana Press Inc., Totaw. rJI
345
346
Sherman
100% 90% 80%-
70%:
-> 60%: (II
'2
50%-
=I
0
40%30%-
1,
20%-
I
0
12 24
36 48 60 72 84 96
Months Fig. 1. Product-limit Survival for 46 patients with anaplastic thyroid carcinoma. (Adapted from Reference 6.)
women compared with men in one study (2), but no effect of gender was found in others, including two recent large epidemiologic surveys(1,8,10-13). Only one stud addressed the predictive valueof presenting symptoms, finding a relative mortality or hoarseness (13).Patients who previou of 2 for dyspnea but no impact of dysphagia had been treated for differentiated carcinoma and subsequently developed anaplastic disease had outcomes similar to those without a previously treated differentiated can cer (8,12). Several approaches to prognostic classification have been applied to anaplastic noma. In both the TNM-based staging approach, adopted by the Union Internationale Contre Cancer and the American Joint Commission on Cancer, and the staging s used by the National Thyroid Cancer Treatment Cooperative Study Registry, a as stage IV disease, the highest stage (6,14).The prognos tic carcinomas are classified scoring system introduced by the European Organization for Research on Treatmen of Cancer gives45 points for a diagnosis of anaplastic carcinoma(15);given addition points for advanced age and extrathyroidal invasion, virtually all patients are staged as either4 or 5 (6).In an earlierM.D.Anderson series, patients with anaplastic disea were divided into four groupings (16): Stage I: disease confined to the thyroid Stage 11: disease in locoregional nodes Stage III: disease extending to soft tissues in the neck Stage IV: distant metastases
347
Anaplastic Carcinoma
Using this classification, advancing stage was significantly associated with shorter survival (8). In the absence of a comparisonof the predictive value of these differing approaches to staging, it is impossible to recommend one approach over another (6). The impact of treatment on survival is unclear. Except for patients whose tumors are small and confined entirely within the thyroid, attempts at total thyroidectomy and (1,8,13). External complete tumor resectionare not associated with prolonged survival beam radiotherapy, administered in conventional doses, also does not appear to affect survival. Although a complete response may be obtained in up to 40% of patients irradiated, most relapse locally with or without distant disease (13). Treatment with single-agent chemotherapy alone does not appear to improve survival or local control of neck disease, though perhaps 20% may have some degree of response in distant metastases (17). The introduction of hyperfractionated radiotherapy, combined with radiosensitizing doses of doxorubicin, may improve the local response rate to about 80%, with subsequent median survival of 1 year, but distant metastases remain the leading cause of death (18). Similar improvement in local disease control has been reported with the combination of hyperfractionated radiotherapy with radiosensitizing doxorubicin, followed by debulking surgery in responsive patients (3). However, the addition of larger doses of doxorubicidcisplatin, mitoxantrone, or bleomycidcyclopho phamide/5-fluorouracil is not associated with improved control of distant disease or improved survival (7,19). In summary, improved survival has only been demonstrated for patients with disease localized to the thyroid who receive aggressive local intervention. Improved therapies for distant metastases are needed that can be combined with multimodality treatment of local neck disease. Some activity against anaplastic carti11 studies of taxol as monotherapy or noma has been noted with taxol, and Phase combined with hyperfractionated radiotherapy are underway(20).
REFERENCES 1. Ne1 C J , van Heerden JA, GoellnerJ R , Gharib H, McConahey W M , Taylor WF, Grant CS. Roc Anaplastic carcinomaof the thyroid a clinicopathologic study of 82 cases. Mayo Clin 1985; 6051-58. 2. Tan RK, FinleyRK,DriscollD,Bakamjian V, Hicks WL, Jr.,SheddDP.Anaplastic carcinoma of the thyroid a 24-year experience. Head Neck 1995; 17:4147. 3. TennvallJ, Lundell G, Hallquist A, Wahlberg P, Wallin G, Tibblin S. Combined doxorubicin, twoon hyperfractionated radiotherapy, and surgery in anaplastic thyroid carcinoma: report protocols. Cancer 1994; 74: 1348-1354. 4. Hadar T,Mor C, Shvero J, Levy R, Segal K. Anaplastic carcinoma of the thyroid. Eur J Surg Oncol 1993; 19511-516. 5. SpiresJR, SchwartzM R , Miller RH. Anaplastic thyroid carcinoma: association with differentiated thyroid cancer. Arch Otolaryngol Head Neck Surg 1988; 114:40-44. 6. Sherman SI, Brierley J, Sperling M, Maxon III Initial analysisof staging and outcomes 1996; 6:S39. from a prospective multicenter of study treatmentof thyroid carcinoma. Thyroid 7. Tallroth E, Wallin G, Lundell G, Lowhagen T, Einhorn J. Multimodality treatment in anaplastic giant cell thyroid carcinoma. Cancer1987; 60:1428-1431. 8. Venkatesh YSS, Ordonez NG, Schultz PN, Hickey RC, Goepfert Sarnaan H, NA. Anaplastic carcinoma of the thyroid:a clinicopathologic study of 121cases. Cancer1990; 66:321-330. 9. Nishiyama R H , Dunn EL, Thompson N W . Anaplastic spindle-cell and giant-cell tumors of the thyroid gland. Cancer 1972; 30:113-127. 10. Gilliland F D , Hunt WC, Moms DM, Key CR. Prognostic factors for thyroid carcinoma:
HR.
348
Sherman
A population-based study of 15,698 cases from the Surveillance, Epidemiology and End Results (SEER) program 1973-1991. Cancer 1997; 79564-573. 11. Akslen LA, Haldorsen T, Thoresen S , Glattre E. Survival and causes of death in thyroid 1991; 51:1234cancer: a population-based study of2479 cases from Norway. Cancer Res
1241. 12. Carcangiu ML, Steeper T, ZampiG, Rosai J.Anaplastic thyroid carcinoma: A study of 70 cases. Am J Clin Path01 1985; 83:135-158. 13. Junor EJ, Paul J, Reed NS. Anaplastic thyroid carcinoma: 91 patients treated by surgery and radiotherapy. Eur J Surg Oncol 1992; 18:83-88. for stagingof cancer, American Joint Comm 14. Beahrs OH, Henson DE, Hutter RVP. Manual sion on Cancer, 3rd ed. Philadelphia: Lippincott, 1988. 15. Byar DP, Green SB, Dor P, Williams ED, Colon J, van Gilse HA, et al. A prognostic ind
for thyroid carcinoma. A study of the E.O.R.T.C. thyroid cancer cooperative group. Eur J Cancer 1979;151033-1041. 16. Aldinger KA, Samaan NA, Ibanez ML, Hill CSJ. Anaplastic carcinoma of the thyroid: a of thethyroid.Cancer 1978 reviewof 84 cases of spindleandgiantcellcarcinoma
41:2267-2275. 17. Ahuja S, ErnstH.Chemotherapyofthyroidcarcinoma. J Endocrinol Invest 1987;10: 303-310. 18. Kim JH, Leeper RD. Treatment of locally advanced thyroid carcinoma with combination doxorubicin and radiation therapy. Cancer1987; 60:2372-2375. 19. Schlumberger M, Parmentier C, Delisle MJ, Couette JE, Droz JP, Sarrazin D. Combination therapy for anaplastic giant cell thyroid carcinoma. Cancer1991; 67564-566. 20. Ain KB, Tofiq S , Taylor KD. Antineoplastic activity of taxol against human anaplastic J Clin Endocrinol Metab1996; 81:3650thyroid carcinoma cell lines in vitro and in vivo. 3653.
VI Undifferentiated Cancers B. Lymphoma
40 Thyroid Lymphoma Steven I. Sherman
primary thyroidal non-Hodgkin’s lymphoma, although rare, is an important component of the differential diagnosis for thyroid nodules or malignancy, mainly because 2% of of the significantly different prognosis and treatment approach. Only about extranodal lymphomas arise as primary malignancies within the thyroid gland, and 5% of all thyroid malignancies( I ) . In a Danish epidemiothese represent no more than as 2.1 per million persons, with logical survey, the annual incidence rate was estimated a 4:l female predominance (2). Most other retrospective series have confirmed this markedly higher frequencyof disease in women (3-7). The mean and median ages at at a diagnosis are between 65 and 75 years, with a suggestion that women present significantly older age than men(2-7); like anaplastic carcinoma, presentation before age 40 is extremely rare. Preexisting Hashimoto’s thyroiditis is the only significant risk factor for primary thyroidal lymphoma, as patients with Hashimoto’s have at least 60-fold relative risk for developing non-Hodgkin’s lymphoma of the thyroid (2,8,9).Worldwide, the frequency of thyroid lymphoma appears higher in areas with greater prevalence of thyr tis. Along with an increasing frequency of thyroiditis, lymphoma may also occur more commonly following iodine supplementation (IO). The development of lymphoma in the setting of Hashimoto’s thyroiditis has not been adequately explained. One potential mechanism may be the result ofchronic antigenic stimulation in thyroiditis, enhancing the probabilityof neoplastic transformation(11).There appears be to no clear association although between exposure to ionizing radiation and the development of (121, lymphoma individual cases have been described. Karyotypic chromosomal abnormalities have been rarely reported. primary thyroidal lymphomais almost always of B-cell lineage (2,6,7).In contrast, occasional T-cell lymphomas have been reported, particularly in areas endemic for (I3,I4).Diffuse, large-cell histoloHTLV-I-associated adult T-cell IeukemiaAymphorna gies (formerly describedas histiocytic lymphomas) generally predominate, accounting for about 70-80% of tumors (3,5,6,15).Less frequent histologies include follicular or nodular patterns, mixed lesions, lymphomas with plasmacytoid features, signet cell lymphomas, and lymphocytic lymphomas. Differing classification schemes have been proposed, adding considerable confusion to the literature on thyroid lymphomas. Using
F m : Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited by: L. Wartofsky 0 Humana Press Inc., T o t m ,
351
352
Sherman
the NCI Working Formulation, about 70% of thyroid lymphomas are intermediategrade, With the remaining cases evenly divided among low-grade, high-grade, and undefined histologies (4416). By contrast, using the Kiel classification, about 65% are low-grade, 30% high-grade, and 5% undefined (17). The predictive value of tumo grading is unclear; whereas one study suggested that Working Formulation high-gr tumors have a far worse prognosis than low- or intermediate-grade tumors (17), other studies have failed to demonstrate a significant difference based upon either Workin Formulation (3,s) or Kiel classification (4). It has been suggested that the B-cell thyroid lymphomas should be grouped with mucosa-associated lymphoid tissue(MALT) lymphomas, given histologic and progno tic similarities (18). Classically, MALT lymphomas contain small to medium-sized centrocytelike cells, often with plasmacytoid features, associated with reactive g centers and lymphoepithelial lesions(6). It is likely that high-grade lesions arise from transformation of low-grade MALT lymphomas, given a high frequency of tumors with both histologies sharing identical immunoglobulin light-chain restriction (2,191. Molecular abnormalities reportedly associated with MALT lymphomas (including roid) have included a loss of expression of bcl-2 and an increase in p53 inactivation with higher grade disease (20,21). Although several studies have indicated that as many as 70-80% of thyroid lymphomas have histologies consistent with MALT neoplasms (2,22), amulticenterimmunohistochemicalanalysisfromtheEasternCooperative Oncology Group reported a far smaller frequency of less than 10% (6). Consistent with the concept of thyroid lymphoma as a MALT lesion is the reported 10-60% frequency of concomitant or metasynchronous gastrointestinal tract lymphomas, considerably higher than is typically found in nodal non-Hodgkin’s lymphomas (23,24); but, this finding has been disputed in other studies. Similar to anaplastic carcinoma, symptoms of thyroid lymphoma typically are of a rapidly expanding bilateral goiter, occurring in90-100% of patients (2,3,5,7,22). Symptoms and signs due to compression of surrounding structures are also common, such as dysphagia, dyspnea, stridor, hoarseness, neck pain, and venous dilatation due to superior vena cava obstruction. Although these symptoms are usually present for only a few months, longstanding goiter is reported in 10-20% of patients, generally as a solitary in association with hypothyroidism. Occasionally, lymphoma can present nodule or with unilateral involvement, rather than as a diffuse firm or hard goiter. On examination, the thyroidis commonly fixed to underlying structures and does not m well with deglutition. Substernal extensionis common. When a distinct intrathyroidal 5 cm, but often the exact bord mass is identified, the diameter is typically larger than of the goiter cannot be appreciated. Regional adenopathy in cervical or supraclav in about half of patients.In the setting of stridor and hoarsene chains can be palpated (25). Besides the local manifestatio laryngoscopy can often identify vocal cord paresis up to 10% of patients may report “B” symptoms, including fever, sweats, and weight loss. Symptoms due to hypothyroidism may be present in up to 10%of patients, due to coexisting thyroiditis rather than destructive infiltrationby lymphoma (16). In rare cases, thyrotoxic findings occur in association with rapid destruction of follicles and release ofpreformedhormone intocirculation (26) orduetopreexistingGraves’ disease (27).
Thyroid Lymphoma
353
There are no laboratory abnormalities specific or to diagnostic of thyroid lymphoma (2). Primary hypothyroidism, when present,is associated with an elevated serum TSH, 50%. of lactate with minimal increases commonly found in up to A serum concentration dehydrogenase greater than 500 U/L is seen in about 25%, increased levels of serum uric acid in about 15%, and increased serum concentrations of IgA, IgM, or IgG in about 30% of patients (2). Elevated titers of anti-thyroid antibodies (antithyroglobulin and antimicrosomal) have been reported in up to95% of patients (2,16). Radiographic and scintigraphic imaging studiesare useful in defining the extent of are unable to disease, planning therapy, and monitoring response to treatment, but (28). A common distinguish lymphoma from other thyroid malignancies or thyroiditis CT findingis the so-called “donut sign,” given the tendency of lymphoma to completely encircle the trachea.CT and MRI are superior to ultrasound given their greater ability to detect malignant invasion of the trachea, substernal extension of disease, and involvement of mediastinal and abdominal nodal groups. Radioiodine scanning has no role in thyroid lymphoma, given the lack of iodine concentrating ability in lymphocytes. By contrast, 67Ga imaging detects abnormal uptake in about 90% of patients with thyroid lymphoma (29). However, it is quite nonspecific, with frequent uptake noted in Hashi20’Tl moto’s thyroiditis. Other radionuclides that may detect thyroid lymphoma include and VC-MIBI, but sensitivity and specificity are unknown (30). Lymphoma should be part of the working differential diagnosis of a solitary thyroid nodule, a dominant nodule in a multinodular goiter, or any patient with Hashimoto’s thyroiditis whose chronic goiter enlarges or produces new symptoms. The diagnosis of lymphoma can often be established by cytologic examination of material obtained by fine needle aspiration, particularly the common large-cell histologies. Given the frequent coexistenceof chronic autoimmune thyroiditis, small cell lymphomas are more difficult to diagnose cytologically, and immunohistochemical staining to demonstrate lymphocyte monoclonality may be necessary. On occasion, large-bore needle biopsy or surgical excision maybe needed to obtain sufficient material for immunohistochemical staining. On the other hand, the presence of lymphoma cannot be completely ruled out by an aspirate showing thyroiditis, and clinical judgement, open surgical biopsy, and follow-up are often necessary (see Chapter 41). The Ann Arbor stage classification is most widely used for primary thyroid lym (16,31). Abouthalfofpatientspresentwithdiseaselimitedtothethyroidgland, designated stageE.Another 45% have disease limited to the thyroid and locoregional nodes, classifiedas stage IIE and occasionally subdivided based upon disease in mediastinal nodes. Only about5% of patients present with additional disease located in nodal groups on both sidesof the diaphragm (stageLIE) or with diffuse organ involvement (stage IV). Among the extranodal sites that have been reported are bone marrow, (2,15,25,32). Given these potengastrointestinal tract, lungs, liver, pancreas, and kidney tial sites of disease involvement, the initial staging work-up for a patient diagnosed with primary thyroid lymphoma should probably include a complete blood count; se chemistries includingTSH, lactate dehydrogenase, and uric acid; chest radiograph; and CT of the neck, chest and abdomen. Consensus does not exist regarding the routine use of bone marrow examination, gallium scintigraphy, or lymphangiography, and these studies should probably be used selectively.
354
Sherman
As with anaplastic disease, airway management is often a primary focus of the in therapy of thyroid lymphoma. Severe airway compromise requiring tracheostomy can be anticipated in up 25% to of patients. Given the rapidly growing nature of the di it is often necessary to perform an emergent procedure, with an attendant increased risk of complication. Therefore, early consideration should be given to elective tomy in patients at risk for airway obstruction. Surgery is also occasionally requir establishment of the diagnosis. Beyond airway management and diagnosis, contro exists about the appropriate extent of further surgery for thyroid lymphoma. Several recent studies have failed to demonstrate a significant survival advantage from more extensive surgery such as total thyroidectomy in stages IE and IIE,particularly when concurrent prognostic factors such asinitial tumor bulk or extrathyroidal invasion are considered (3,7,15). However, for stage IE patients who have disease truly confined within the gland, without invasion across the thyroid gland capsule, thyroidectomy followed by adjuvant radiotherapy may be appropriate treatment(4,33). The difficulty in making this distinction lies in the fact that thyroidectomy is often required to determ of the gland capsule (33). Radical procedure whether the disease in fact extends outside that increase the risk to surrounding structures such as recurrent laryngeal nerves or the upper aerodigestive tract should be avoided. IE External beam radiotherapy has been the traditional primary therapy for stage and IIE disease, administered alone or in combination with other modalities in 70% of all patients (16). Disagreement exists regarding the appropriate amount and dur of therapy as well as the optimal extentof the radiation fields. Decreased relapse-fre 30 Gy compared with survival has been reported for patients treated with less than those who received more than 30 Gy (7).However, decreased doses have often been prescribed for patients presenting with the most advanced disease or who are too sick to tolerate full radiation dosing, biasing the analysis against lower doses. External radiotherapy delivering 40 Gy over a 4- to 5-week period may be associated with a 55-70% 5 year survival, with 90% survival reported for stage IE disease (25,34,35). Some authors advocate irradiating the thyroid bed and bilateral neck, emphasizin most treatment failures tend to occur in distant sites (3,7,25). Others routinely recom mend including the superior mediastinum in the radiation field (15). More advanced lymphoma (stage IIIE or IV)or disease presenting in older patientsis often treated with chemotherapy. For patients who develop distant recurrence fol primary therapy for stage E or IIE,chemotherapy is also indicated. Treatment regim that have been used in recent reports include CHOP (cyclophosphamide, doxorubici vincristine, and prednisone), ProMACE-CytaBOM (prednisone, methotrexate, doxor bicin, cyclophosphamide, etoposide, cytosine arabinoside, vincristine, and bleomycin BACOP(bleomycin,doxorubicin,cyclophosphamide,vincristine,andprednisone), C-MOPP(cyclophosphamide,vincristine,prednisone,andprocarbazine),and CW (cyclophosphamide,vincristine,andprednisone).Becausethenumbersofpatients treated with any given regimen have been small, it has not been possible to identify a superior drug combination. The results of multiple studies that incorporated combined multimodality therapy (CMT) for stage IE and IIE disease, using both chemotherapy and radiotherapy, ha (36). Distant recurrence occurred significantly less frequ been recently summarized following CMT than either radiotherapyor chemotherapy alone. The benefit appeare
Thyroid Lymphoma
355
limited to patients with disease extending into the mediastinum. Given the limitations of detecting distant micrometastases by current imaging modalities, the addition of a brief course of chemotherapy to local radiation may improve long-term outcome (5,15) The disease-specific 5-year survival of all patients with primary thyroid lymphoma is 45-65% (2,4,7,15,16). For patients with stage IE or IIE disease, the corresponding five-year estimates are 55-80% and20-50%, respectively (2-5,7,15). In contrast, fiveyear disease-specific survival for patients with either stage IIIE or IV disease is only 15-35% (2,4). Following initial therapy, about 85% of stageE or IIE patients achieve complete remission ($15). However, at least half relapse within 5 years, whichis more likely to occur outside of the initial radiation fields and in distant sites. Relapse-free (15). survival may be increased to 80% following initial combined modality therapy Oncerelapseoccurs,occasionalcompleteresponseandlong-termsurvivalcanbe obtained with chemotherapy and/or radiotherapy, but the median survival is about 7 months following relapse (7).
REFERENCES 1. Freeman Cy Berg J W , Cutler SJ. Occurrence and prognosis of extranodal lymphomas. Cancer 1972; 39:252-260. 2. Pedersen RK, Pedersen NT. Primary non-Hodgkin’s lymphoma of the thyroid gland a population based study. Histopathology1996; 28:25-32. 3. Junor EJ, Paul J, Reed NS. Primary non-Hodgkin’s lymphoma of the thyroid. Eur J Surg Oncol 1992;18:313-321. 4. Pyke CM, Grant CS, Habermann TM, Kurtin PJ, van Heerden JA, Bergstralh El, et al.
Non-Hodgkin’s lymphoma of the thyroid: is more.than biopsy necessary? World J Surg
1992;16:604-609. 5. Skarsgard ED, Connors JM, Robins RE. A current analysis of primary lymphoma of the thyroid. Arch Surg 1991;126:1199-1203. 6. Wolf BC, Sheahan K, DeCoste D, Variakojis D, Alpern HD, Haselow RE. Immunohistochemical analysis of smalltumors cell of the thyroidgland an Eastern Cooperative Oncology Group study. Hum Path01 1992; 23:1252-1261. 7. Logue JP, Hale RJ, Stewart AL, Duthie MB, Banerjee SS. Primary malignant lymphomaof the thyroid: a clinicopathological analysis. Int J Radio1 Oncol Biol 1992; Phys 22:929-933. 8. HyjekE,IsaacsonPG.Primary B celllymphomaanditsrelationshiptoHashimoto’s thyroiditis. Hum Path01 1988; 19:1315-1326. 9. Holm LE, Blomgren H, Lowhagen T. Cancer risks in patients with chronic lymphocytic thyroiditis. N Engl J Med 1985; 312:601-604. 10. Harach H R , Williams ED. Thyroid cancer and thyroiditis in the goitrous region of Salta, Argentina, before and after iodine prophylaxis. Clin Endocrinol (Oxf) 1995; 43:701-706. 11. Burke JS, Butler JJ, Fuller ML. Malignant lymphomas of the thyroid: a clinical pathologic study of 35 patients including ultrastructural observations. Cancer1977; 39:1587-1602. 12. Matsuzuka F, Miyauchi A, Katayama S , Narabayashi I, Ikeda H, Kuma K, Sugawara M. Clinicalaspects ofprimarythyroidlymphoma:Diagnosisandtreatmentbasedon our experience of 119 cases. Thyroid 1993; 3:93-99. 13. Mizukami Y,Michigishi T, Nonomura A, Nakamura S, Hashimoto T, Katsuda S, et al. Primary lymphoma of the thyroid: a clinical, histological, and immunohistochemical study of 20 cases. Histopathology 1990; 17:201-209. T14. Ohsawa MyNoguchi S , Aozasa K. Immunologic type of thyroid lymphoma in an adult cell leukemia endemic area in Japan.Leuk Lymphoma 1995; 17:341-344. 15. Tsang RW, Gospodarowicz MK, Sutcliffe SB, Sturgeon E,Panzarella T, Patterson BJ.
356
Sherman
Non-Hodgkin’s lymphomaof the thyroid gland: prognostic factors and treatment outcom The Princess Margaret Hospital Lymphoma Group. Int J Radiat Oncol Biol Phys 1993; 27599-604. 16. Mazzafeni EL, Oertel YC. Primary malignant thyroid lymphoma and related lymp ative disorders. In Mazzafem Samaan EL, NA, editors. Endocrine tumors. Boston: Blackw Scientific Publications, 1993: 348-377. 17. Aozasa K, Inoue A, TajimaK, Miyauchi A, Matsuzuka F,Kuma K. Analysis of79 patients with emphasis on histologic prognostic factors. Cancer 1986; 58:lOO-104. 18. Isaacson P, Wright DH. Extranodal malignant lymphoma arising from mucosa-associated lymphoid tissue. Cancer 1984; 53:2515-2524. 19. Chan JKC, Ng CS,Isaacson PG. Relationship between high-grade lymphoma and lowgrade B-cellmucosa-associated lymphoid tissue lymphoma (MALToma) of the stomach. Am J Path01 1990; 136:1153. 20. Ashton-Key M, Biddolph SC, Stein H, Gatter KC, Mason DY. Heterogeneity of bcl-2 expression in MALT lymphoma. Histopathology 1995; 26:75-78. 21. Du M, Peng H, Singh N, Isaacson PG, Pan L. The accumulation of p53 abnormalities is associated with progressionof mucosa-associated lymphoid tissue lymphoma. Blood 199 86:45874593. 22. Laing RW, Hoskin P, Hudson BV, Hudson GV, Harmer C, BennettM H , MacLennan KA. The significance of MALT histology in thyroid lymphoma: a review of patients from the BNLI and Royal Marsden Hospital.Clin Oncol (R Coll Radiol). 1994; 6:300-304. 23. Anscombe AM, Wright DH. Primary malignant lymphoma of the thyroid-a tumour of mucosa-associated lymphoid tissue: review of seventy-six cases. Histopathology 1985; 9: 81-97. 2 4 . Hemnann R, Panahon AM, Barcos M p , Walsh D,Stutzman L. Gastro-intestinal involveme in non-Hodgkin’s lymphoma. Cancer 1980; 46:215-222. 25. Tupchong L, Hughes F, Harmer CL. Primary lymphoma of the thyroid clinical features, prognostic factors, and results of treatment. Int J Radiat Oncol Biol Phys 1986; 12:18131821. 26. Jennings AS, Saberi M. Thyroid lymphoma ina patient with hyperthyroidism.Am J Med. 1984; 76551-552. 27. Zeki K, Et0 S, Fujihira T, MasudaM, Oda S, Chiba S , Suzuki H.Primarymalignant lymphoma of the thyroid in a patient with longstanding Graves’ disease. Endocrinol Jpn 1985; 32:435-440. 28. Podoloff DA.Is there a place’for routine surveillance using sonography, C T , or MR imaging for early detection (notably lymphoma) of patients affected by Hashimoto’s thyroiditis? AJR Am J Roentgen01 1996; 167:1337-1338. 29. Higashi T, It0 K, Mimura T, Ohi T, Nishikawa Y.Clinical evaluation of 67-Ga scanning in the diagnosis of anaplastic carcinoma and malignant lymphoma of the thyroid. Ra 1981; 141:491497. 30. Scott AM, Kostakoglu L, O’Brien JP, Straus DJ, Abdel-Dayem H M ,Larson SM. Compariso of technetium-99m-MIBI and thallium-201-chloride uptake in primary thyroid lymphom J Nucl Med 1992; 33:1396-1398. 31. Carbone PP, Kaplan HS, Musshoff K, Smither DW, Tubiana M. Report of the committee on Hodgkin’s disease staging classification. Cancer Res 1971; 31:1860-1861. C L . Primarynon-Hodgkin’s 32. Evans TR, Mansi JL, BevanDH,DalgleishAG,Harmer lymphoma of the thyroid with bone marrow infiltration at presentation. Clin Oncol (R Coll Radiol) 1995; 754-55, 33. Friedberg “I, Coburn MC, Monchik JM. Role of surgery in stage IE non-Hodgkin’s lymphoma of the thyroid. Surgery 1994; 1161061-1066. 34. Vigliotti A, Kong JS, Fuller LM, Velasquez WS. Thyroid lymphomas stages IE and IIE:
Thyroid Lymphoma
357
comparative results for radiotherapy only, combination chemotherapy only, and multimodality treatment. Int J Radiat Oncol Biol Phys 1986; 12:1807-1812. 35. Compagno J, Oertel E.Malignant lymphoma and other lymphoproliferative disordersof the thyroid gland. Am J Clin Pathol 1980; 74:1-1 1. 36. Dona R, Jekel JF, Cooper DL. Thyroid lymphoma. The case for combined modality therapy. Cancer 1994; 73:200-206.
41 Thyroid Lymphoma Pathology James Oertel and Yolanda Oertel
The thyroid may be involved secondarily by lymphoma from other sites in the body or by leukemic infiltrates ( I ) . Most primary thyroidal lymphomas areB cell types, but both Hodgkin’s lymphoma (2,3) and T-cell lymphoma (4) may occur. Gross examinationrevealshomogeneous,pale, firm tissuethathasirregularly replaced the thyroid. If advanced autoimmune thyroiditis is also present, the lymphoma probably cannot be distinguished from the inflammatory infiltrate without histologic examination. Microscopically, the regions of autoimmune thyroiditis reveal either the usual benign lymphoplasmacytic infiltrates or the common alterations of the follicular epithelial cells (such as oxyphilic cell metaplasia). The lymphoma usually consistsof a monotonous infiltrate of abnormal lymphoid cells which replace the thyroid parenchyma, fill and distend some thyroid follicles (Figs. 1 and 2), permeate the walls of some of the larger vessels, and may extend into the juxtathyroidal tissues. Antikeratin antibodies can demonstrate displaced and distorted follicular epithelium, is visible which against the lymphomatous infiltrate. Cervical lymph nodes may be involved. Smears from the aspirates may show predominantly chronic lymphocytic thyroiditis, and it is necessary to perform multiple aspirates until the lymphomatous regions are sampled. The latter show a monotonous lymphoid population, readily observed mitotic figures, and a conspicuous absence of follicular epithelial cells. The lymphoma may be diffuse or follicular (nodular) in type. Separating the lymphoma from an adjacent infiltrate of autoimmune thyroiditis requires careful appraisal of the abnormal cells in the routine histologic sections and the use of immunohist cal and/or molecular genetic techniques. These demonstrate the abnormal phenotypes of the lymphoma and the varied, nonneoplastic cells of any autoimmune thyroiditis prese Some thyroid lymphomas present as aggressive neoplasms, often in the elderly, typically with a short history of thyroid enlargement and with extension to cervical tissues outsideof the gland. The histological featuresin a moderate proportion of these cases have been interpreted as showing a high grade of malignancy, but this did not influence outcome in the treated patients (5). A considerable proportion of the thyroid lymphomas are of intermediate or lowgrade malignancy, sometimes having a history consistent with longstanding autoimmun
From: Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited by: L. Wartofsky 0 Humann Press Inc., Totowa, EJI
359
Oertel and Oertel
360
Fig. 1. Malignantlymphoma. A numberofthyroidfolliclesareformedbymetaplastic epithelial cells, presumably the result of previous Hashimoto’s thyroiditis. Both the intersti (H&E stain; ~100) tissue and the altered follicles are extensively infiltrated by the lymphoma.
E
Thyroid Lymphoma
362
thyroiditis. A few have been discovered in glands resected because of Hashimoto’s thyroiditis (6). Cases of this type, combined with careful evaluation of many other examples of lymphoma, have resultedin the concept that a thyroid involved by autoimmune diseaseis comparable to mucosa-associated lymphoid tissue (MALT)-for example,thePeyer’spatches of theintestine (7). Thusthesuggestionhasbeenmade that the majority of thyroidal lymphomas are MALT-lymphomas (6-8), composed of centrocytelike cells (cells similar to the cells just outside the lymphoid follicles, the parafollicular lymphoid cells). These lymphomas may tend to be localized for lengthy periods, possibly explaining why some thyroid lymphomas have been cured by surgery alone. Thyroid lymphomas may spread to other sites of MALT. The centrocytelike cells are varied in their morphology, which may be the sourceof some of the different cellular types described in various reports of thyroid lymphomas. They may undergo (2,6,8).Such cells plasmacytic differentiation, a frequent finding in thyroid lymphomas are usually monotypic with immunoglobulin light-chain restriction. These lymphoma cells may extend into the reactive lymphoid follicles of the autoimmune thyroiditis, thereby explaining the follicular (nodular) pattern of some of these lymphomas (9). Also, persisting lymphoid follicles have been reported in the rare plasmacytomas of the thyroid ( I O ) , perhaps supporting the concept that plasmacytomas of the thyroid are “mature” MALT-lymphomas. When a high grade lymphoma is present, often there is evidence that it has arisen from a low-grade MALT-lymphoma (11).
REFERENCES 1. Naylor B. Secondary lymphoblastomatous involvement of the thyroid gland. Arch Pathol Lab Med 1959; 67:432438. 2. Compagno J, Oertel JE. Malignant lymphoma and other lymphoprdiferative disordersof the thyroidgland a clinicopathologic studyof 245 cases. Am J Clin Path01 1980; 74:l-11. 3. Feigen GA, Buss DH, Paschal B, WoodruffRD, Myers RT. Hodgkin’s disease manifested as a thyroid nodule. Hum Pathol 1982; 13:774-776. 4. Mizukami Y, Michigishi T, Nonomura etA,al. Primary lymphoma ofthe thyroid a clinical, histological and immunohistochemical study of 20 cases. Histopathology 1990; 17:201-209. 5. Tennvall J, Cavallin-SM E, herman M. Primary localized non-Hodgkin’s lymphoma of Eur J Surg Oncoll987; 13:297-302. the thyroid a retrospective clinicopathological review. 6. Hyjek E, Isaacson PG. Primary B cell lymphoma of the thyroid and its relationship to Hashimoto’s thyroiditis. Hum Path01 1988; 19:1315-1326. 7. Isaacson PG. Lymphomas of mucosa-associated lymphoid tissue(MALT).Histo@thology 1990; 16~617-619. 8. Anscombe AM, Wright DH. Primary malignant lymphoma of the thyroid-a tumor of mucosa-associated lymphoid tissue: review of seventy-six cases. Histopathology 1985; 9: 81-97. L, Wright DH. Follicular coloniza9. Isaacson PG,Androulakis-PapachristouA, Diss TC, Pan tion in thyroid lymphoma.Am J Pathol 1992; 141:43-52. 10. Aozasa K, Inoue A, Yoshimura H, Miyauchi A, Matsuzuka F, Kuma K. Plasmacytoma of the thyroid gland. Cancer 1986;58:105-110. 11. Pedersen RK, Pedersen NT. Primary non-Hodgkin’s lymphoma of the thyroid gland a population based study. Histopathology 1996; 28:25-32.
VI1 Undifferentiated Cancers C. Medullary Carcinoma
42 Medullary Thyroid Carcinoma Douglas W. Ball
Medullary thyroid carcinoma (MTC), an uncommon neoplasm stemming from the calcitonin-producing thyroid parafollicular C cells, accounts for approximately 3% to 5% of cases of thyroid cancer. Unique among all types of thyroid canceris the strong association of MTC with inherited tumor syndromes in approximately 20% of cases. The molecular basis of inherited MTC remained obscure until two groups provided evidence that a susceptibility locus for inherited MTC was located on human chromosome 10 (1,2). In 1993, Mulligan and colleagues (3), as well as several other groups (4,5),discovered that characteristic mutations inretthe protooncogene were responsible for the inherited formsof MTC. This seminal finding has had a tremendous impact on the clinical diagnostic evaluation of MTC and continues to shape the direction of research in MTC biology.
CLASSIFICATION MTC is traditionally classified as sporadic versus hereditary. There are three princip autosomal dominant hereditary MTC syndromes, described in Table 1. The most frequent hereditary MTC syndrome is multiple endocrine neoplasia type 2A (MEN 2A) comprising MTC in 95% of affected individuals, pheochromocytoma in approximately 50%, and hyperparathyroidism in 15-30%. The second most common syndrome is MEN 2B, including MTC, pheochromocytoma, ganglioneuromas of the oral mucosa and gastrointestinal tract, a characteristic elongated facies, a marfanoid body habitus, and no increase in hyperparathyroidism. The least common hereditary MTC syndrome is familial MTC (FMTC) without known extrathyroidal manifestations. Each of the hereditary syndromes is associated with dominantly acting, germline mutations of the ret gene. As described below, specific mutation sites inretthe gene are strong predictors of the phenotype of each syndrome. In addition to the three principal inherited MTC syndromes, there aretwo minor variantsof MEN 2A, MEN 2A associated with Hirsch(6),and MEN2A associated sprung’s disease (hypoplasia of intestinal myenteric plexus) with the skin disorder cutaneous lichen amyloidosis (CLA) (7).
From: Thyroid Cancer: A Comprehensive Guideto Clinical Management Edited by:L. Wartofsky 0 Humana Press Inc., Totown, A7
365
Ball
366 Table 1 Classification of Medullarv Thvroid Carcinoma ~~
Lesion
Associated Type
Biological Behavior ~
Somatic None Sporadic FMTC Germline None MEN 2A MEN 2B
Pheochromocytoma Hyperparathyroidism Pheochromocytoma . Ganglioneuromas Marfanoid habitus
~~~
Ret Mutation Gene ~
(Met918Thr) (Cys609,611,618, 620,634+Any; 768,804). Germline (Cys609,61I , 618, 620,634+Any) Germline (Met918Thr)
~~
Intermediate Less aggressive Intermediate More aggressive
DEVELOPMENTAL BIOLOGY
MTC arises from the thyroid parafollicularC cells, clustered in the upper two-third
of the thyroid lobes.In contrast to thyroid epithelial cells which derive from endod
the C cells are believed to originate in the neural crest and enter the developing when the ultimobranchial body from the fourth pharyngeal pouch fuses with thyroid epithelium (8,9).In the adult,C cells are restricted to the upper two-thirds of the t lobes (IO).Recent advances in understanding the developmental biology of the neural crest are now being applied to further elucidate MTC pathophysiology. For example, criticalneurotrophicgrowthfactors,includingglial-derivedneurotrophicfactor [(GDNF) a natural ligand for the ret receptor] as well as nerve growth factor (NGF) and other neurotrophins, appear to have essential roles in promoting the survival and (11).Ret functions as a tyrosine kinase rec differentiation of neural crest derivatives which, along with an accessory molecule, GDNFR-a, can transduce the GDNF signal via a pathway involving rus and mitogen-activated protein kinase (MAPK). BIOCHEMISTRY
The characteristic secreted product of parafollicular C cells, and the most useful circulating marker for MTC is the polypeptide hormone calcitonin. The mature 32amino acid polypeptide is synthesized as a large 135-amino acid precursor, which is processed by prohormone convertases associated with the C cell. Calcitoninis encoded by a multiexonic gene on chromosome llp, which produces two distinct messenger RNA species.In addition to calcitonin itself, alternative splicing ofthe primary calcitonin transcript yields calcitonin gene-related peptide (CGRP). The resulting calcitonin and CGRP polypeptides are unique and interact with distinct receptors. Calcitonin secr predominates in normal thyroid C cells, whereas CGRP predominatesin neural tissue (12). In MTC, abnormalRNA splicing permitsan approximately equal ratio of calcito nin : CGRP. A variety of nonthymidal tissues produce modest levels of calcitonin including pulmonary neuroendocrine cells, adrenal medulla, and gastroenteropancre of calcitonin are virtually diagno endocrine cells(13). Although substantial elevations ticforMTC,modestelevationscanbeseeninpulmonaryinflammatorydiseases, small-cell lung cancer, gastrinoma, carcinoidtumors, and renal failure(14), as well as nonneoplastic causes of C-cell hyperplasia discussed below.
Medullary Thyroid Carcinoma
367
Innormalparafollicular C cells,calcitoninsecretion is positivelycoupledwith extracellular ionized calcium concentrations, via the adenylyl cyclase-coupled calcium sensing receptor (CSR) (1516). Defects in calcium signaling have been shown in several established MTC cell lines, which are characterized by constitutive calcitonin secretion (15). In addition to calcium, a useful calcitonin secretogogueis pentagastrin. MTC cells express biochemical markers typifying secretoryofcells the diffuse neuroendocrine system. Polypeptide hormones produced by MTC cells include somatostain (171, adrenocorticotrophic hormone (18), gastrin-releasing peptide (19), substance P (20), andvasoactiveintestinalpeptide (21). Other neuroendocrine markers include neuronspecificenolase,neuralcelladhesionmolecule(NCAM),chromograninA, prohormone convertases, synaptophysin, and the amine synthetic enzyme, L-dopa decarboxylase (22). In addition, many MTC tumors express two surface markers that have been exploited for imaging purposes, carcinoembryonic antigen and the receptor for somatostatin.
DIAGNOSIS In both sporadic and familial contexts, there are complex issues facing the clinician who is evaluating patients for possible MTC. Because thefamilial natureof the disorder may be occult or unclear, an additional section discusses the evaluation of patients with uncertain hereditary status.
Sporadic MTC Presentation The diagnosis of MTC, outside of the20% of cases where there is known heritable disease, most commonly begins with the palpation of an asymptomatic thyroid nodule, typically in the third to sixth decades of life. In most instances, the history and physical examination do not offer any distinctive information compared to typical patients with is warranted to detect the presence thyroid nodules.A sufficiently detailed family history of thyroid cancer, pheochromocytoma, or hyperparathyroidism in first degree relatives. Approximately 20% of patients present with locally advanced disease with symptoms related to recurrent laryngeal nerve invasion, dysphagia, or painful lymph node metastasis. Rare individualsmay present with a paraneoplastic manifestation such as flushing, secretory diarrhea,or symptoms relatedto hypercortisolism. Eachof these paraneoplastic manifestations is essentially confined to individuals with large MTC tumor burdens. Minimal calcitonin elevation, in the presence of these syndromes, is usually not indicative of MTC. Biopsy The diagnostic procedure of choice for sporadicis MTC thyroid fine needle aspiration biopsy. The sensitivity of thyroid cytopathology is theoretically equivalent to that in of papillary cancer. In practice, however, the cytopathologist needs an adequate index suspicion for atypical, thyroglobulin-poor specimens, in order to employ the diagnostic calcitonin immunostain. Thus many specimens are misdiagnosed as atypical follicular cancers, or poorly differentiatedor even anaplastic thyroid cancers. Other patients may be referred for thyroidectomy without F N A B , thus raising the potential of adverse outcomes related to unsuspected pheochromocytoma. The correct preoperative diagno
368
BaIl
is essential, not only to screen for pheochromocytoma and hyperparathyroidism, but especially to plan an appropriate staging workup and a suitably extensive neck e tion by an experienced surgeon. The important differences in surgical approach for MTC versus papillary and follicular cancer are detailed in the treatment section later in this chapter.
Role of Calcitonin Testing Virtually all patients with clinically evident MTC have elevated basal levels of calcitonin. The most recent generation of calcitonin two-site immunoradiometric assay (IRMA), now available through many commercial laboratories, appears superior to previous one-site radioimmunoassays. These earlier assays were subject to artifactual addition, recognition of larger calcitonin precursors seen rarely outside of MTCIn(23). the greater sensitivity of the IRMA may allow better separation of normal subjects from patients with mild calcitonin elevations compared to older radioimmunoassays (RIA). Because of the low prevalence of MTC among thyroid nodule or goiter pa routine use of calcitonin determinations inthis setting traditionally was not advocated (24). However, recent data from the French medullary study group as well as two o in patients large prospective series suggest a potential role for calcitonin screening with thyroid nodules or goiter. In these large screened populations, the prevalenceof significant calcitonin elevations (>50 pg/ml) ranged from 0.6% to 1% among patients presenting with nodular thyroid disease (25-27). Significantly, all of the patients with athyroidmass and significant hypercalcitoninemia proved to have microscopic or macroscopic MTC.The French medullary study group prospectively performed ca nin determinations and thyroidectomy in 1167 patients with thyroid nodule or goiter (25). The prevalence of MTC in this series was surprisingly high, 1.37%, compared to previous estimates of 0.6% to 0.8%, based on less complete surgical ascertainment (26,27). In this series, the sensitivity of a basal calcitonin value >35 pg/ml was 75% with no false positives, compared to a sensitivity of less than 50% for FNAB. A lower MTCs and calcitonin threshold would have lead to improved detection of microscopic a higher yield of C-cell hyperplasia associated with autoimmune thyroiditis or p this approach is needed before or follicular cancers. Although further experience with a definitive recommendation is possible, a screening calcitonin level by IRMA now to and TSH in the evaluation of nodular thyro appears to be a rational addition FNAB disease (28).
Staging Following detection of MTC, a prudent preoperative staging evaluation includes imaging of the neck, mediastinum, and chest by a combination of CT scans, ultraso imaging, orMRI. A significant percentage of MTC patients presenting with an asy tomatic thyroid nodule have detectable metastases at these sites, particularly in regi lymph nodes in the lateral neck and central compartment, especially along the trach is the liver. The conve esophageal groove.An additional important metastatic problem tional imaging approaches listed above, as well as octreotide and dimercaptosuccinic acid(DMSA) scintigraphy, have poor sensitivity for detection of liver metastases. While microscopic liver metastases would not generally obviate the need for an initia thyroidectomy and neck exploration, some investigators advocate laparoscopic liver
Medullay Thyroid Carcinoma
369
biopsy prior to proceeding with more involved procedures such as repeat neck micr section (29). In addition to imaging studies, a basal calcitonin and CEA level, serum calcium, and 24-hour urine collection for catecholamines or metanephrines should be completed before surgery. The roleof ret gene testing in this setting is discussed in a following section.
Heredita y MTC
Historically, diagnosis of individuals at risk for inherited MTC had been accomplished by periodic calcitonin provocative testing in order to identify asymptomatic subjects with C-cell hyperplasia or microscopic MTC. Although concerns about the specificity of this testing approach have subsequently arisen, the approach was largely successful in selecting at-risk subjects for early thyroidectomy, resulting in improved survival compared to unscreened subjects (30). This testing approach has been superceded, in ret gene mutations, most instances, by the availability of highly specific tests for germline now available in reference laboratories and referral centers in North America, Europe, (PCR)amplifiand Japan. These testing procedures employ polymerase chain reaction 10, 11, and 16. Mutation detection cation of the commonly mutated ret exons, typically schemesmayemployautomatedDNAsequencing,restrictionenzymeanalysis, or procedures that detect electrophoretic mobility changes in annealed DNA strands containing mismatched bases. The three inheritedMTC syndromes, described in Table 1, are all characterized by ret protooncogene. Ret is a tyrosine kinase relatively stereotyped mutations in the receptor for the neurotrophic growth factor GDNF,is important which for the differentiation, survival, and proliferation of a variety of neuronal cell types (11,311. Structurefunction studies indicate that these alterations in ret are activating mutations, which enhance the intrinsic tyrosine kinase enzyme activity of the receptor. There are fascinating parallels between the activating mutations of ret seen in the MEN 2 syndromes and the ret gene rearrangements which.characterize a significant minority of papillary cancers (see Chapter 6 and 50 for details). These hybrid oncoproteins, like the MEN 2A mutant forms of ret, are capable of spontaneous dimerization in the absence of ligand and activation of the receptor tyrosine(32,33). kinaseAn international consortium of centers treating inherited MTC recently published an analysis of over 400 families studied worldwide (34). Figure 1 illustrates the location of these mutation sites within the ret protein, organized according to disease type.
MEN 2A The most common syndrome, MEN 2A, is associated with mutation 1 of 5 cysteine in sites located in the immediate extracellular region of the receptor (codons 609, 611, 618, 620 in exon 10 and codon 634 in exon 11). Altogether, approximately 98% of MEN 2A families have a detectableret mutation at oneof these five sites. Codon 634 is by far the most common of this group, accounting for 85% of MEN 2A families (34). Although these five cysteine residues may be mutated to any other amino acid, a single change at position 634, cysteine to arginine, accounts for 52% of all MEN Less 2A families, and is strongly associated with the incidence of hyperparathyroidism. commonmutationsatmoreproximalcysteineresiduesaccountfortheremaining MEN 2A families. Rare MEN 2A families which exhibit combined inheritance with
370
Ball
... ... .... .... ....
. .
... .. a
.
... ..
o n 0 0
0
=MENU
0
=m
0
0 =MEN28
Fig. 1, Distribution of ret gene mutations in 306 families worldwide with inherited MTC. Shown is a schematic representation of the structure of the ret tyrosine kinase receptor with numbers indicating the position of known mutation sites. TM indicates the transmembrane domain, K, the tyrosine kinase domain. In the nomogram below, each symbol represents an individud family, indicating its phenotypeand reported mutation site. (Codons609,611,618, and 620 are in ret exon 10, 634 in exon 11,768 in exon 13,804 in exon 14, and 918 in exon 16.) [Data are adapted from the InternationalRET Consortium (341.1
Hirschsprung Disease have ret mutations at positions 618 or 620 (35). The syndrome of MEN 2A with cutaneous lichen amyloidosis appears to have a pattern of ret similar to more typical cases of MEN 2A (34). In addition to ret mutation testing, preoperative screening for pheocromocytoma is mandatory, usually employing 24-hour urine collections for catecholamines or metais also indicated. nephrines. Additional testing with an albumin-corrected serum calcium
FMTC In FMTC (familial MTC without other associated endocrinopathy), the distribution of ret gene mutations overlaps with MEN 2A. Several important distinctions can be 634, observed, however. FMTC families less frequently have mutations at position particularly the characteristic cysteine to arginine mutation, compared to MEN 2A families (34). FMTC families are more likely thanMEN 2A families to have mutations at positions618,620, and two rare noncysteine sites within the tyrosine kinase doma (codons 768 and 804). A small percentage of FMTC families, approximately 12%, appear to have no detectable ret mutations despite persistent study (34). MEN 2B In MEN 2B,the great majority of families exhibit a single mutation at codon 918 (methionine to threonine), resulting in an alterationof the substrate recognition pocket of the tyrosine kinase enzyme (32). Unlike MEN 2A, MEN 2B gennline mutations
inoma Medullary Thyroid
371
are frequently de novo in the presenting individual, eg. not detectable in either parent. The de novo mutation is noted at a much greater frequency in the allele inherited from the patient’s father (36).
Family Screening Recommendations For families with a known pattern of inheritance, children of affected parents are typically tested beginning at ages5 or 6. As illustrated in Figure 1, analysis is initially targeted atret exons 10 and 11for these disorders. Following the discovery of atypical ret mutation, most investigators currently recommend prophylactic thyroidectomy, without resorting to calcitonin provocative testing(37). Similarly, a negative DNA test, in the setting of a known mutation in affected family members, is highly reassuring. Abnormal calcitonin provocative testing results in this setting are likely to be false positive, re to idiopathic rather than neoplastic C-cell hyperplasia (38,39).In the case of inherited 1 may be screened. Since a FMTC, additional sites including those listed in Figure significant minority ofFMTCfamilieshavenodetectable ret mutation, calcitonin ret mutationprovocativetestingremainstheprincipaldiagnostictoolforthese negative families. In the case of MEN 2B, ret testing usually is recommended in the first 2 years of life, owing to the generally more aggressive course of MTCin this disorder. A diagnosis is often possibleby physical examination in early childhood, based on visible mucosal neuromas of the lips and tongue (see Fig. 2).
Uncertain Hereditay Status Two problematic scenarios relating to uncertain hereditary status occasionally confront physicians who treat MTC patients. The first scenario concerns MTC patients who have a negative family history for MEN 2 but actually harbor cryptic germline mutations and are capable of passing the disorder to their offspring. When MTC patien ret testing, approximately with a negative family history are investigated with germline 3-6% are found to harbor such mutations (4546).Clinical and genetic features occasionally associated with this scenario include the following: 1. Bilobar MTC or background C-cell hyperplasia 2. Concurrent occult pheochromocytomaor hyperparathyroidism 3. De novo ret mutation 4. Misdiagnosis or early death of an affected parent 5 . Small family size and incorrect assumptions about paternity
Although no cost-benefit analysis is available for germline ret mutation screening in apparently sporadic patients, such mutation detection frequently has a “multiplier” effect allowing identification of other family members at risk. Therefore, it appears prudent to offer germline ret analysis to patients presenting with apparently sporadic MTC, especially when there are first degree relatives potentially at risk. If the individ with MTC tests negative, then no further testing is usually recommended for additional relatives. A second problematic scenario concerns families in which a single patientMTC with undergoes negative DNA testing, but additional relatives have borderline or frankly abnormal calcitonin provocative tests. The existence of MEN 2 families without known
372
Ball
Fig. 2. Photograph of apatientwith MEN 2B, showingthetypicalmucosalneuromas associated with a marfanoid habitus.(A) Lips. (B) Tongue. (C) Hyperextensibilityof the hands.
ret mutations makesthis scenario troubling. However, false-positive calcitonin prov tivetests(associatedwithidiopathicratherthanneoplasticC-cellhyperplasia)are increasingly being identified (39). In families in which no clinical manifestations of MEN 2 can be detected otherthan a singleMTC patient who is ret mutation-negative, abnormal calcitonin provocative testing in siblings appears more likely to be false positive than attributable to occult familial disease. Careful clinical and biochemical follow-up is recommended. Further refinements in mutation detection are necessary to satisfactorily resolve this issue.
Carcinoma Medullay Thyroid
373
Somatic ret Mutations in SporadicMTC In contrast to inherited cases of MTC in which the patient’s gemline DNA (e.g., ret mutation, patients with sporadic DNA in cells throughout the body) contain the MTCs may have ret mutations that are somatic, that is, limited to the tumor cells themselves. The prevalence of somatic ret gene,mutationsis controversial, with a range (5,42).Some, but not all investigative groups, of 23%to 70% reported in different series ret mutation is a negative prognostic indicatorin sporadic have reported that a somatic MTC (42). The vast majority of these alterations are MEN 2B-type mutations found at position 918. Fewer than 10%of sporadic tumors harbor mutations elsewherein ret including codons 768 and 883 in the tyrosine kinase domain (43). Individual MTC ret gene mutations, potentially accounting tumors may in fact be heterogeneous for for discrepancies reported in the literature (44). Eng and associates showed that at least 80% of individuals with sporadic MTC have at least one tumor cell population with a mutation in ret codon 918. Even MEN 2A-associated tumors, bearing a germline (45). Hence mutation at position 634, can harbor an additional mutation at position 918 detection of a somatic codon 918 mutation is not necessarily a reliable indicator of of performing mutation analysis on sporadic, versus heritable MTC. The clinical value MTC tumors is not currently well-defined. CLINICAL FEATURES
C-Cell Hyperplasia With the adventof molecular diagnosisof inherited MTC, there has been substantial progress in characterizing the natural historyof these disorders, especially the earliest stages of C-cell neoplasia. In inherited MEN 2, affected individuals may show evidence of C-cell hyperplasia and microscopic carcinoma as early as age 3 to 4 in MEN 2A 2 years of life in the case of MEN 2B (37,46). There has been considerable and in the first controversy regarding the definition of C-cell hyperplasia. In the classic definition of 50 Ccellsperhigh-powerfieldconstituteC-cell Dekllis andWolfe,morethan hyperplasia (47). Recent reviews have emphasized that normal autopsy specimens or (IO), hyperparathyroidism, as specimens from patients with autoimmune thyroiditis well as with benign and malignant thyroid epithelial neoplasms (48), may frequently meet criteria for C-cell hyperplasia. The incidence of idiopathic C-cell hyperplasia in the general population is greater than wasinitially appreciated prior to the availability 2 kindreds who areknown to lack their of ret mutation testing. Individuals from MEN family’s characteristic germlineret mutation (and thus be unaffected with the inherited disorder)nonethelesshaveasignificantincidence of calcitoninhypersecretionin (39) response to the secretogogues calcium and pentagastrin. Marsh and collaborators demonstrated substantial overlap between mutation carriers and noncarriers in MEN 2A kindreds with false-positive peak stimulated values exceeding 600p g / d in some male nonmutation carriers. When subjected to thyroidectomy, these individuals exhibit C-cell hyperplasia but no frank MTC. Males appear to have a higher incidence of (IO),and a brisker response to calcitonin secretogogues. C-cell hyperplasia than females SeveralinvestigatorshavearguedthattheearliestlesionintheinheritedMTC in situ, rather than syndromes should properly be termed medullary thyroid carcinoma in situ reflects a preinvasive carcinoma that C-cell hyperplasia(49).The term carcinoma
374
Ball
has not penetrated follicular basement membranes but has the potential to Inme MEN 2A, for example, distant metastases have been reported as early as age 5 (46). Many cases of early C-cell disease associated with MEN 2 contain substantial nodular clusters with a hundredor more cells. These clusters are recognizable on hematoxy and eosin staining and stain floridly for calcitonin. By contrast, C-cell hyperplasia to secondary or idiopathic causes lacks such nodular clusters and is generallynot distinguishable from normal thyroid, in the absence of calcitonin immunostaining(50). Based on a number of autopsy studies, secondary or idiopathic C-cell hyperplasia does not appear to be associated with a risk of later progression to subclinical or frank MTC ( I 0,50).
Microscopic MTC Unlike papillary thyroid cancer, it is distinctly unusual to find incidental foci of MTC in thyroidectomy specimens. The vast majority of microscopic MTCs( 4 . 5 cm) are found in prophylactic thyroidectomies from individuals who have screened po for heritable MTC. In general, isolated microscopic MTCs generally carry an excell prognosis and frequently require no further therapy beyond thyroidectomy. Not infre quently,however, patients may present with neck lymphadenopathy and a normal thyroid examination, onlyto have MTC diagnosed by calcitonin immunostaining of MTC found on subsequent thyroidectom resected lymph nodes with a clinically occult Because of the propensity of MTC tumors to metastasize prior to developing into a palpable thyroid nodule, early diagnosis of stage 11 tumors is important but clinically difficult.
Mucroscopic MTC Macroscopic MTC lesions are firm thyroid nodules that may appear well-demarcat
or grossly invasive. Routine neck radiographs occasionally reveal a dense calcificatio pattern. Histologically, MTC tumors are highly pleiomorphic with marked variation To date, the presence of these variant architecture even within individual tumor nodules. has not provided useful prognostic information. The most useful adjunct to routine histology is calcitonin immunostaining. Positive calcitonin staining, combined with an absence of thyroglobulin, and positive staining for other neuroendocrine markers such as chromogranin and neuron specific enolase, characterize most MTC specimens. M tumors may stain homogeneously for calcitonin or with a patchy, heterogenous pat This latter pattern is associated with a poorer prognosis in multivariate analysis (see below). As noted previously, most patients presenting with palpableMTC tumors also harbor microscopic or gross lymphadenopathy. The extent of lymph node involvem is frequently underestimated by available radiographic and nuclear medicine studies.
Tumor Progression
Because of the limited means for detecting microscopic or in situ MTC in the absenc of a positive family history, the vast majority of sporadicMTC patients present with palpable nodules, and greater than 40% have macroscopic regional lymphadenopathy at the time of initial detection (51). Surgical studies employing ipsilateral and central compartment lymph node dissections reveal significantly higher rates of lymph node
Carcinoma Medullay Thyroid
375
micrometastasis. The tendency for early regional lymph node metastasis is a primary factor in the relatively low surgical cure rates for sporadic MTC. Of typical patients with a palpable sporadic MTC who have undergone thyroidectomy and neck exploration, greater than80% have persistent elevations of calcitonin (52). Even patients considered to be free of residual disease at the time of surgery have a very strong likelihood of persistent hypercalcitonemia. In the absence of overt adenopathy or extensive distant metastases, the clinical outcome is usually characterized by slow disease progression. Patients with minimal calcitonin elevations after primary surgery and no further radiologically detectable disease have an 86% 10 year overall survival rate and relatively little tumor-associated morbidity (53). Similarly, a Memorial Sloan-Kettering series reported 94% 5-year survival in patients with nodal disease alone versus 41% in stage IV disease (54). In addition to regional lymphadenopathy, the most common metastatic sites include the liver and lung. Both of these metastases occur in a diffise, hematogenous pattern, usuallywithslow growth. Fortunately, modest metastatic burdens in the lung and liver can be compatible with lengthy survival. Standard imaging modalities are quite (29). Occasionally, liver metastases can insensitive for detecting early liver metastases become bulky and painful. Extensive liver metastases are also frequently associated w diarrhea. Although liver resectionis not routinely advocated forMTC liver metastases because of the multifocality ofthis process, surgical debulking of symptomatic masses occasionally can provide useful palliation.
Complications The most frequent serious complications observed in advanced MTC relate to local tumor invasion into the thyroid bed, trachea, and carotid sheath, or progressive m in the upper mediastinum and lung. Recurrent laryngeal nerve paresis, tracheal and esophageal invasion, superior vena cava syndrome, aspiration-related and postobstructive pneumonia, and hemoptysisall may be seen in patients with advanced disease and contribute to disease mortality. Disseminated metastasis, including abdominal viscera, bone, and CNS, may occur as a terminal manifestation of the illness. The principal paraneoplastic humoral complications ofMTC are flushing, diarrhea, and less commonly, the ectopic ACTH syndrome. The etiology of flushing in MTC patients is still a cause of some debate. One likely candidate mediator is CGRP, a potent vasodilator capableof inducing prolonged cutaneous erythema with intradermal administration (55). Symptomatic flushing frequently can be improved by subcutaneous octreotide injection(56).Unfortunately, octreotide has little efficacy in MTC-associated diarrhea; some patients paradoxically worsen. Like flushing, the pathophysiology of diarrhea in MTC requires further clarification. There is no consistent evidence for either malabsorption or a secretory abnormality in the small intestine. Instead, patients exhibit colonic hypermotility and a decreased ability to absorb water (57). Intravenous CGRP can increase colonic outputof water and electrolytes(58), although the relative importance of other mediators including vasoactive inhibitory peptide, histamine, and prostaglandins, remains unclear. Patients can be treated symptomatically with loperamide and diphenoxylate to lengthen colonic transit time. Calcitonin excess, per se, is not associated with any clinically significant changes in bone or mineral metabolism.
376
Ball
Table 2 Clinical Staging of Medullary Thyroid Cancer STAGE I STAGE 11
STAGE III STAGE IV
C-cell hyperplasia. Tumorless than 1 cm and negativelymphnodes. Tumors 1 cm or more or tumor of any size with positive nodes. Tumors of any size with metastases outside the neck or with extrathyroidal extension.
*Clinical Tumor Staging, National Thyroid Cancer Treatment Cooperative Study.
PROGNOSIS Medullary thyroid cancer occupies an intermediate position among thyroid cancer histologic types with respect to biologic behavior and long term prognosis. Although there is intrinsic variability in patients’ clinical course, prognostic factors apparent at the time of diagnosis and initial surgery have important utility in predicting long-term outcomes of MTC. An accurate understanding of the influence of prognostic factors in both sporadic and hereditary contextsis essential for selecting appropriate levels o therapeutic intervention.
Sporadic MTC The most comprehensive reviews of prognostic factors in MTC to date are based with follow-up to30 years (59,60).These on nationwide cancer surveillance in Sweden, studies have revealed important several important predictors of survival. Among all sporadic MTC patients, relative survival (the ratio between observed and expected survival) was 63% at 10 years and 50% at 20 years. The most important prognostic factor was initial clinical stage (see Table 2). Stage III (nodal disease) and stage IV (distant metastases) were associated with relative hazards of3.3 and 4.1 compared to patients with no known nodal or distant disease. Initial clinical stage remained highly predicative of future mortality, even up to 20 years after diagnosis. Other important > 3 cm, capsular invasion, weak o negative prognostic indicators included tumor size heterogeneous calcitonin staining, male gender, and older age. By contrast, patients with a tumor measuring < 1 cm without known metastases did not differ in survival from the general population even 20 years or longer after diagnosis (61).
Hereditay MTC The significant improvement in outcome seen over the last three decades for ofpresymptomaticscreening with heritable MTC can be attributed to the success programs, first with calcitonin secretogogues and more recently with genetic testing. The success of such programs has been well-documented (30), with survival rates of MEN 2A subjects identified at screening now indistinguishable from the general tion (60). An unsettled question is whether the MTC in MEN 2A behaves intrinsically less aggressively than sporadic tumors when matched for clinical stage. Swedish MTC MEN 2Aand registrydatasuggeststatisticallysimilaroutcomesfornonscreened FMTC sporadic patients (60). At either extremes of the hereditary MTC spectrum, appears significantly less aggressive than MEN 2A in terms of disease latency and survival (61);MEN 2B is generally more aggressive.
Carcinoma Medullay Thyroid
377
The age-specific likelihood that anMEN 2A gene carrier would present with detectable calcitonin hypersecretion or with symptomatic MTC has been studied extensiv Ponder and colleagues. Approximately65% of obligate gene carriers exhibit calcitonin hypersecretion at age 20 years.By age 35, fully 95% of gene carriers have a positive provocative test.In contrast, the likelihoodof a clinical presentation withMTC is only 25% at age 35 and only approximately 60% at age 70 (62).
FOLLOW-UP Routine Strrdies Since follow-up surgeryor systemic therapies forMTC are usually not curative, the goal of follow-up should be to anticipate complications that have a significant impact on morbidityor survival in a cost-effective manner over intervals that frequently extend beyond 10 years. History and examination should be targeted to possible indicators of disease recurrence including dysphagia, neck mass, hoarseness, shortness of breath, chest pain, abdominal pain, bone pain, diarrhea, flushing, or symptoms of pheochromocytoma. Follow-up testing, for typical patients with modest calcitonin elevations after primary surgery, utilizes a combination of biochemical tumor markers and radiologic studies to screen for disease recurrence. In one commonly used scheme, serum calcitonin, are obtainedatapproximately6-monthintervals. CEA,andthyroidfunctiontests For low-risk patients with isolated postoperative hypercalcitonemia, imaging studies generally can be reserved for instances of significant progression of biochemical and 'clinical findings. CT, MRI, ultrasound, octreotide, and DMSA scans all have limited sensitivity in the setting of modest calcitonin elevations (650 pg/ml). For patients with known radiographic abnormalities, follow-up scans are reasonable at more freque intervals. The value of routine bone scans is controversial, given the relative infrequency of asymptomatic bone metastases in MTC. Many patients become unduly alarmed with progressivein their rises serum calcitonin and are concerned that thereis an inverse correlation between the calcitonin level and their life span. Although the absolute calcitonin level correlates roughly with tumor burden and increases with disease progression, the calcitonin level se has per not proved (59). In addition, there is significant biological to be a useful indicator of survival variability in basal calcitonin secretion over time. Thus, patients should be dissuaded from gauging their progress based solely on the calcitonin level. MEN 2A, follow-up examinations for pheochromocytoma For patients with inherited shouldbecarriedoutonanannual or biannualbasis,withroutine24-hoururine studies for catecholamines or metanephrines. Imaging studies, most frequently MRI, are reserved for patients with abnormal urine tests. Most investigative groups currently advocate unilateral resection where possible, to avoid or postpone the morbidity assoc ated with primary adrenal insufficiency (63). The recent development of laparoscopic adrenalectomy can facilitate this approach (64).
Decision t o Reoperate Following identification of new lymphadenopathy on radiographic scans or by phy cal examination, the decision to surgically intervene must be made on an individual basis.Patientswhohaveundergonepreviouslymphnodedissections mayhave a
378
Ball
significantly higher rate of nonspecific nodal enlargement, relatedto disrupted lymph in normalization drainage patterns.In general, repeatMTC operations infrequently result of calcitonin levels or surgical cure. Where such normalization is achieved, the surgica approach is generally a systematic nodal dissection in a favorable prognosis patient, rather than an isolated “node-picking” procedure (65). When the rationale is not an anticipated cure, follow-up surgery should be targeted to lesions that are symptomat or to those that are expected to cause significant morbidity with further growth. Suc high risk sites include the mediastinum adjacent to the great vessels, tracheoesopha groove, carotid sheath, and lateral nodes associated with the brachial plexus.
Outlook
With the elucidation of the genetic basis of inherited MTC, these disorders have become a paradigm case for the value of genetic screening in the prevention of an inherited cancer syndrome. The principal challenge is now to achieve a comparable impact in the more common sporadicform. Further understanding of the pathophysiol ogy of established MTC,stemming from basic advancesin molecular genetics, offers the best chance for truly effective systemic therapies for this cancer.
REFERENCES
1. Mathew CG, Chin KS, Easton DF, et al. Linked genetic markers for multiple endocrine neoplasia type 2A on chromosome 10. Nature 1987; 328527428. 2. Simpson N E , Kidd KK, Goodfellow PJ, et al. Assignment of multiple endocrine neoplas
type 2A to chromosome 10 by linkage. Nature 1987; 328528-530. 3. Mulligan LM, Kwok JBJ, Healey CS, et al. Germline mutations of the ret proto-oncogen in multiple endocrine neoplasiatype 2A. Nature 1993; 363:458-460. 4. Donis Keller H, Dou S, Chi D, et al. Mutations in the ret proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet 1993; 2951-856. 5. Hofstra RMW, Landsvater RM, Ceccherini I, et al. A mutation in the ret proto-oncogene associated with multiple endocrine neoplasia type 2B and sporadic medullary thyroid noma. Nature 1994; 367:375-376. 6. Frilling A, Becker H, Roehr H-D. Unusual features of multiple endocrine neoplasia. Henr Ford Hosp Med J 1992; 40:233-235. 7. Gage1 RF, Levy ML, Donovan DT, et al. Multiple endocrine neoplasiatype 2a associated with cutaneous lichen amyloidosis. Ann Intern Med 1989; 111:802-806. 8. LeDouarin N,Le Lievre C. Demonstration de l’origine neurale des cellules a calcitonin corps ultimobranchial chez l’embryon de poulet. C R Seances Acad Sci Paris 1970; 270:2857 9. Ericson LE, Fredriksson G. Phylogeny and ontogeny of the thyroid gland. In Greer MA, editor. The thyroid gland. New York: Raven Press, 1990. 10. Guyetant S, Rousselet M-C, Durigon M, et al. Sex-related C cell hyperplasiain the normal human thyroid: a quantitative autopsy study. J Clin Endocrinol Metab 1997; 82:42-47. 11. Robertson K, Mason I. The GDNF-ret signalling partnership. Trends Genet 1997; 13:l-3. 12. Amara SG, Jonas V, Rosenfeld MG, Ong ES, Evans RM. Alternative RNA processing in calcitoningeneexpressiongeneratesmRNAsencodingdifferentpolypeptideproducts. Nature 1982; 298:240-244. 13. Becker KL, Snider RH, Moore CF, et al. Calcitonin in extrathyroid tissue&mnan. Ack Endocrinol 1979; 92:746-751. 14. Becker KL, Nash D, Silva OL, et al. Increased serum and urinary calcitonin in patients with pulmonary disease. Chest 1981; 79:211-216.
Carcinoma Medullay Thyroid
379
15. Garrett JE, Tamir H, Kifor 0, et al. Calcitonin-secreting cells of the thyroid express an extracellular calcium receptor gene. Endocrinology1995; 13652024211. 16. Freichel M, Zink-hrenz A, Holloschi A, et al. Expression of a calcium sensing receptor in a human medullary thyroid carcinoma cell line and its contribution to calcitonin secret Endocrinology 1996;137:3842-3848. 17. Roos BA, Lindall AW, Ells J, et al. Increased plasma and tumor somatostatin-like immunoreactivity in medullary thyroid carcinoma and small cell lung cancer. J Clin Endocrinol Metab 1981;52:187-194. 18. Melvin KE, Tashjian AH Jr, CassidyC E , Givens JR Cushing's syndrome caused by ACTHand calcitonin-secreting medullary carcinoma of the thyroid. Metabolism 1970; 192331-838. 19. KameyaT,BesshoT,TsumurayaM, et al.Productionofgastrin-releasingpeptidein medullary carcinoma of the thyroid. Virchows Arch [A] 1983; 401:99-107. 20. Skrabanek P, Cannon D, Dempsey J, et al. Substance P in medullary carcinoma of the thyroid. Experientia 1979; 35: 1259-1260. 21. Said SI. Evidence for secretion of vasoactive intestinal peptide by tumours of pancreas, adrenal medulla, thyroid and lung. Clin Endocrinol1976; 5:201S-204S. 22. Baylin SB, Mendelsohn G. Medullary thyroid carcinoma: a model for the studyof human In Owens AH Jr, Coffey DS, Baylin SB, editors. tumor progression and cell heterogeneity. Tumor cell heterogeneity, origins and implications. New York Academic Press, 1982: 12. 23. Body JJ, Heath HIII. Nonspecific increases in plasma immunoreactive calcitonin in healthy individuals: discrimination from medullary thyroid carcinoma new extraction by a technique. Clin Chem 1984; 30511-514. 2 4 . Singer PA, Cooper DS, Daniels GH, et al. Treatment guidelines for patients with thyroid nodules and well-differentiated thyroid cancer. Arch InternMed 1996; 156:2165-2172. 25. NiccoliP,Wion-BarbotN,CaronP,et al. Interest of routinemeasurementofserum calcitonin: study in a large series of thyroidectomized patients. J Clin Endocrinol Metab 1997; 82~338-341. 26. Pacini F, Fontanelli M, Fugazzola L, et al. Routine measurement of serum calcitonin in nodular thyroid disease allows the pre-operative diagnosis of unsuspected sporadic med thyroid carcinoma. J Clin Endocrinol Metab 1994; 782326429. 27. Rieu M, Lame MC, Richard A,et al. Prevalence of sporadic medullary thyroid carcinoma: the importanceof routine measurement of serum calcitonin in the diagnostic evaluation of thyroid nodules. Clin Endocrinol (Oxf) 1995; 42:453-460. 28. Horvit PK, GagelRF. The goitrous patient withan elevated serum calcitonin: what to do? [Editorial]. J Clin Endocrinol Metab 1997; 82:335-337. 29. Tung WS, VeselyTM, Moley JF. Laparoscopic detection of hepatic metastases in patients with residual or recurrent medullary thyroid cancer. Surgery1995; 118:1024-1029. etThe clinical outcome of prospective screening 30. Gagel RF, Tashjian AH Jr, Cummings T, al. for multiple endocrine neoplasia type2a. N Engl J Med 1988; 318:478-484. 31. Trupp M, Arenas E, Fainzilber M, et al. Functional receptor for GDNF encoded by the c-ret proto-oncogene. Nature 1996; 381:785-789. 32. Santoro M, CarlomagnoF, Romano A,et al. Activationof RET as a dominant transforming gene by germline mutations of MEN2A and MEN2B. Science 1995; 267:381-383. 33. Bonganone I, Monzini N, Borrello MG, et al. Molecular characterization of a thyroid tumor-specific transforming sequence formed by the fusion of ret tyrosine kinase and the regulatory subunit RI alpha of cyclic AMP-dependent protein kinase A. Mol Cell Bioi 1993;13:358-366. 34. Eng C, Clayton D, Schuffenecker I, et al. The relationship between specific RET prototype 2: internaoncogene mutations and disease phenotype in multiple endocrine neoplasia tional RET mutation consortium analysis. JAMA 1996; 276:1575-1579. 10 mutations 35. Mulligan LM, Eng C, AttieT, et al. Diverse phenotypes associated with exon of the RET proto-oncogene. Hum Mol Genet 1994; 3:2163-2167.
380
Ball
36. CarlsonKM, Bracamontes J, Jackson CE,al.etParent-of-origin effects in multiple endoc neoplasia type 2B. Am J Hum Genet 1994; 55:1076-1082. 37. Wells SA Jr,Chi DD, ToshimaK, et al. PredictiveDNA testing and prophylactic thyroide tomy in patients at risk for multiple endocrine neoplasia type 2A. Ann Surg 1994; 220 237-247. 38. Lips CJ, Landsvater RM, Hoppener J W , et al. Clinical screening as compared with DNA analysis in families with multiple endocrine neoplasia type 2A. N Engl J Med 1994; 331: 828-835. of false positive responsesto 39. Marsh DJ, McDowall D, Hyland VJ, et al. The identification the pentagastrin stimulation test in RET mutation negative members of MEN 2A fami Clin Endocrinol(0xf) 1996; 44:213-220. 40. Wohllk N, Cote GJ, Bugalho MM, et al. Relevance of RET proto-oncogene mutations in sporadic medullary thyroid carcinoma. J Clin Endocrinol Metab 1996; 81:3740-3745. 41. Eng C, Mulligan LM, Smith DP, et al. Low frequency of germline mutations in the RET proto-oncogene in patients with apparently sporadic medullary thyroid Carcinoma.Clin Endocrinol (Oxf) 1995; 43:123-127. proto42. Zedenius J, Larsson C, Bergholm U, et al. Mutations of codon 918 in the RET oncogene correlate to poor prognosis in sporadic medullary thyroid carcinomas. J Clin Endocrinol Metab 1995; 80:3088-3090. point mutation in the tyrosine kinase doma 43. Eng C, Smith DP, Mulligan LM, etA novel al. of the RET proto-oncogene in sporadic medullary thyroid carcinoma and in a family w FMTC. Oncogene 1995; 10509-513. 44. Eng C, Mulligan LM, Healey CS,al.etHeterogeneous mutation of the RET proto-oncog in subpopulationsof medullary thyroid carcinoma. Cancer Res 1996; 56:2167-2170. 45. Marsh DJ, Andrew SD, Eng C, et al. Gennline and somatic mutations in an oncogene: RET mutations in inherited medullary thyroid carcinoma. Cancer Res 1996; 56:1241 4 6 . Gill J R , Reyes-Mugica M, Iyengar S, et al. Early presentation of metastatic medullary carcinoma in multiple endocrine neoplasia, type IIA: implications for therapy. J Pediatr 1996; 129:459-464. 47.Rosai J, Carcangiu ML, Dehllis R4. Tumorsofthethyroidgland.3rdSer;Fasic 5. Washington, DC: A F I P , 1984. 48. Albores-Saavedra J, Monforte H, Nadji M, Morales AR. C-cell hyperplasia in thyroid tis adjacent to follicular cell tumors.Hum Path01 1988; 19:795-799. 49. LiVolsi VA. C cell hyperplasidneoplasia. [Editorial]. J Clin Endocrinol Metab 1997;82: 39-41. 50. Peny A,Molberg K, Albores-Saavedra J. Physiologic versus neoplastic C-cell hyperpl the thyroid: separation of distinct histologic and biologic entities. Cancer 1996; 77: etClinical characteristics in sporadic and familial 51. Bergholm U, Adami HO, Bergstrom R, al. medullary thyroid carcinoma: a nationwide study of 249 patients in Sweden from 1959 through 1981. Cancer 1989; 63:1196-1204. 52. Wells SA Jr, DilleyWG, Farndon JA,bight GS, Baylin SB. Early diagnosis and treatme of medullary thyroid carcinoma. Arch Intern Med 1995; 145:1248-1252. 53. van Heerden JA, Grant CS, Gharib H, et al. Long-term course of patients with persistent hypercalcitoninemia after apparent curative primary surgery for medullary thyroid carci noma. Ann Surg 1990; 212:395-400. 54. Ellenhorn JD, Shah JP, BrennanMF. Impact of therapeutic regional lymph node dissect for medullary carcinoma of the thyroid gland. Surgery 1993; 114:1078-1081. 55. Brain SD, Williams TJ, TippinsJR. Calcitonin gene-related peptide is a potent vasodila Nature 1985; 313:54-56. 56. Modigliani E, Guliana JM, Maroni M, et al. Effects of subcutaneous administration of of thyroid medullary cancer.Ann Endocrinol (Paris sandostatin (SMS 201.995) in 18 cases 1989; 50:483-488.
Medullary
381
57. Rambaud JC, Jian R, Flourie B, et al. Pathophysiological study of diarrhoea in a patient with medullary thyroid carcinoma: evidence against a secretory mechanism and for the role of shortened colonic transit time. Gut 1988; 29537-543. 58. Rolston RK, Ghatei MA, Mulderry PK, Bloom SR. Intravenous calcitonin gene-related peptide stimulates net water secretion in rat colon in vivo. Dig Dis Sci 1989; 34612-616. 59. Bergholm U, Adami HO, Auer G, et al. Histopathologic characteristics and nuclear DNA content as prognostic factors in medullary thyroid carcinoma: a nationwide study in Sweden. The Swedish MTC Study Group. Cancer 1989; 64:135-142. 60. Bergholm U, Bergstrom R, Ekbom A. Long-term follow-up of patients with medullary carcinoma of the thyroid. Cancer 1997; 79:132-138. 61. Farndon JR, Leight GS, Dilley WG, et al. Familial medullary thyroid carcinoma without associated endocrinopathies: a distinct clinical entity. Br J Surg 1986; 73:278-281. 62. Ponder BA, Ponder MA, Coffey R, et al. Risk estimation and screening in families of patients with medullary thyroid carcinoma. Lancet 1989; 1:397-401. 63. Lairmore TC, Ball DW, Baylin SB, Wells SA Jr. Management of pheochromocytomas in patients with multiple endocrine neoplasia type 2 syndromes.Ann Surg 1993; 217595-601. 6 4 . Brunt LM, Doherty GM, Norton JA, et al. Laparoscopic adrenalectomy compared to open adrenalectomy for benign adrenal neoplasms. J Am Coll Surg 1996; 183:l-10. 65. Moley JF,Wells SA, Dilley WG, Tisell LE. Reoperation for recurrent or persistent medullar thyroid cancer. Surgery 1993; 114:1090-1095.
43 Medullary Thyroid Cancer Pathology James Oertel and Yolanda Oertel
In the 1950s the physicians of the Cleveland Clinic defined medullary carcinoma During the same decade a few were of the thyroid as a clinicopathologic entity (1,2). recognized independently as unusual tumors different from the majority of thyroid neoplasms (3).Subsequently, a thorough searchof the literature revealed several other probable medullary carcinomas (4). The neoplasm consists of solid masses of rounded, polygonal, and/or spindled neoplastic cells of various sizes (Fig. l), often mixed with (5), rarelyintheisthmus (4). Insular amyloid,nearlyalwaysinthelaterallobes Less (micronodular, nesting) and trabecular (cordlike, ribbonlike) patterns are common. obvious in many tumors are arrangements described as rosettes, pseudorosettes, and glandular, tubular, and follicular structures (4). Many medullary carcinomas contain microacini (6).Other characteristics include pseudopapillary and papillary patterns, sol “small cell,” “oat-cell,” and “neuroblastomalike” features (7), and more pleomorphic anaplastic patterns. “Plasmacytoid” cells (81, binucleated cells (4,7), and cells with very large nuclei, even giant nuclei (4) have been noted. Multinucleated neoplastic cells occur (4). Similaritiesto Askanazy/Hurthle cells have been illustrated or reported (9), as have squamous cell characteristics(10,11) and clear cells. Some tumors produce mucus, and rarely a few form melanin granules. Deposits of amyloid may be accompanies by giant cells of the foreign body type (29). A minority lack amyloid (4). Some are well-encapsulated (9). Within the carcinomas there may be focal necrosis (7,9), (49). The pathologist cystic change (7), irregular calcification, and psammoma bodies should search for necrosis, because its presence may be an indicator of a worse progno sis (12). Medullarycarcinoma may occur in association with autoimmune thyroiditis (9). Other proliferative processes also may be present in the gland: adenomatoid nodules and adenomas (2,9), and other thyroid cancers (7,s). A few are small, and thus are discovered during autopsies, as aofresult an operation for another thyroid disorder, during an evaluation for hypercalcemia, by finding elevated levels of calcitonin, or by the appearance of metastases. Such small tumors are unlikely to be encountered during aspiration.
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wurfo+ky 0 Humuna Press Inc., Totowu, NJ
383
384
Owtel and Owtel
Fig. 1. Medullary carcinoma. The tumorinfiltrates the normal thyroid tissue. (H&E stain; x75).
Most medullary carcinomas are readily recognized on histologic examination of the absence of the usual features of papillary or follicular carcinomas and the pres of amyloid. Any unusual thyroid carcinoma of uncertain classification should be ated for calcitonin because it may be medullary carcinoma (Fig. 2). A cytological diagnosis of medullary carcinoma is often a “diagnosisof exclusion.’’ Examining the wet smears under the microscope after performing the aspiration, the pathologist is surprised at the findings: the pattern is “unexpected,” not fitting the usua non-neoplastic findings (adenomatoid nodule or chronic thyroiditis) or c o m o d y seen neoplastic patterns (follicular neoplasms or papillary carcinoma). As these are rare tumors, usually several years are required to see more than one or two. Our first 10 cases were all different. Not until we encountered additional ca did some of the patterns start repeating themselves. Having diagnosed 25 of them, we of any thyroid suggest that these neoplasms have the most varied cytological patterns (13-15):frequent lack of neoplasm. However, they have a few aspects in common cellular cohesiveness (Fig. 3), pleomorphic multinucleated neoplastic cells scattered among the predominant cell population (Fig.4), lack of prominent nucleoli, cells with plasmacytoid features, spindled cells (Fig. 5), and frequent binucleation. Multinucl histiocytes (usually associated with the presence of amyloid) are seen rarely. Ca stains are used) are seen in cytoplasmic granules (that stain pink if hematological approximately 30% of cases. A very few neoplasms have been reported in which there is differentiation toward C cells(calcitoninproduction)andfollicularcells(thyroglobulinproduction).The follicular elements have been describedas similar to follicular carcinoma(16-18) and
385
CancerMedulla y Thyroid
”
,
.
.
L
.
. ..
Fig. 2. Medullary carcinoma. Calcitoninis demonstrated in the cancer and in the normal C cells in the right side of the field. (Anticalcitonin stain;x75.)
Oertel and Oertel
Fig. 4. Medullary carcinoma. Smear shows pleomorphic cells with cytoplasmic vacuoles. (Diff-QuW stain; ~400.)
Fig. 5. Medullary carcinoma. Smear shows spindled cells with tenuous cytoplasm. (DiffQ
u
i
k
@
stain; x200.)
Medullay Thyroid Cancer
387
papillary carcinoma (18,19). Critical examination of such a neoplasm is needed to exclude collision tumors and nonspecific uptake of thyroglobulin by a medullary carcinoma.
REFERENCES
1. Hazard JB, Crile Jr G, Dinsmore RS, Hawk WA, Kenyon R. Neoplasms of the thyroid. Classification, morphology, and treatment. Arch Path01 Lab Med 1955; 59502-513. 2. Hazard JB, Hawk WA, Crile Jr G. Medullary (solid) carcinoma ofthyroid the a clinicopathologic entity. J Clin Endocrinol Metab 1959; 19:152-161. 3. Horn Jr RC. Carcinoma of the thyroid. Description of a distinctive morphological variant and report of seven cases. Cancer 1951; 4:697-707. 4. Ljungberg 0. On medullary carcinoma of the thyroid. APMIS Suppl 1972; [A]80:1-57. 5. Papotti M, Sambataro D, Pecchioni C, Bussolati G . The pathology of medullary carcinoma of the thyroid: review of the literature and personal experience 62 oncases. Endocr Path01 1996; 7:l-20. 6. Sobrinho-Simoes M, Sambade C, Nesland J M , Holm R, DamjanovI. Lectin histochemistry and ultrastructure of medullary carcinoma of the thyroid gland. Arch Path01 Lab Med 1990; 114:369-375. 7. Williams ED, Brown CL, Doniach I. Pathological and clinical findings in a series of 67 cases of medullary carcinoma of the thyroid. J Clin Path01 1966; 19:103-113. RL. Solid carcinoma of the thyroid gland analysis 8. Ibanez ML, Cole VW, Russell WO, Clark of 53 cases. Cancer 1967; 20:706-723. 9. Freeman D, Lindsay S. Medullary carcinoma of the thyroid gland a clinicopathological study of 33 patients. Arch Path01 Lab Med 1965; 80575-582. 10. Uribe M, Grimes M, Fenoglio-Preiser CM, Feind C. Medullary carcinoma of the thyroid gland clinical, pathological, and immunohistochemical features withofreview the literature. Am J Surg Path01 1985; 9577-594. 11. Dominguez-Malagon H, Delgado-Chavez R, Torres-Najera M, Gould E, Albores-Saavedra J. Oxyphil and squamous variants of medullary thyroid carcinoma. Cancer 1989; 63:11831188. G, Colombo L. Multivariate analysis 12. Dottorini ME, Assi A, Sironi M, Sangalli G, Spreafico of patients with medullary thyroid carcinoma: prognosis significance and impact on t of clinical and pathologic variables. Cancer 1996; 77:1556-1565. 13. SWerstrom N, Telenius-Berg M, herman M. Diagnosis of medullary carcinoma of the thyroid by fine needle aspiration biopsy. Acta Med Scand 1975; 197:71-76. W. Cytological findings in medullary carcinoma 14. SchWer R, Muller H-A, PfeiferOrmanns U, of the thyroid. Path01 Res Pract 1984; 178:461-466. 15. Mendonqa ME, Ramos S, Soares J. Medullary carcinoma ofthyroid a re-evaluationof the cytological criteria of diagnosis. Cytopathology 1991; 2:93-102. 16. Ljungberg 0, Ericsson U-B, Bondeson L, Thorell J. A compound follicular-parafollicular cell carcinoma of the thyroid a new tumor entity? Cancer 1983; 52:1053-1061. 17. Pfaltz M, Hedinger CE, Mtihlethaler JP. Mixed medullary and follicular carcinoma of the thyroid. Virchows Arch A Path01 Anat Histopathol 1983; 40053-59. 18. Ljungberg 0, Bondeson L, Bondeson A-G. Differentiated thyroid carcinoma, intermediate type: a new tumor entitywith features of follicular and parafollicular cell carcinoma. Hum Path01 1984; 15218-228. 19. Albores-SaavedraJ, Gorraez de la MoraT, de la Torre-RendonF, Gould E. Mixed medullarypapillary carcinoma of thethyroid a previously unrecognized variant of thyroid carcinoma. Hum Path01 1990; 21:1151-1155.
44 Medullary Carcinoma of the Thyroid Nuclear Medicine Imaging and Treatment Diane Sweeney and Gerald Johnston
Early total thyroidectomy with resection of proven metastases is the treatment of choice for medullary thyroid cancer (MTC)(1-3). Therefore, localization of involved lymph nodes, soft tissue and bony metastasesis important during the initial diagnostic workupandforrecurrencesurveillance.High-resolutionultrasound,computerized tomography (CT), and magnetic resonance imaging (MRI) have been used with some is often characterizedby distant metastasuccess (3-5). However, more advanced MTC ses, involving lungs, liver and the skeletal system, making whole-body surveillance, of the type afforded by nuclear medicine techniques, the most useful for long-term follow-up. A variety of scintigraphic tracers, including1231,1311, 99"TcDMSA, 201Tl, 1311MIBG, 99mTc MIBI, and "'In pentetreotide (Octreoscan) have been utilized for diagnosis and treatment of medullary thyroid cancer. Newer agents such as labeled antibodies may offer increased detection. IODINE-123 AND TECHNETTO"99M PERTECHNETATE In the detection ofprimarymedullarythyroidcancer,themostcommonlyused nuclear medicine technique is a routine lBI or *Tc pertechnetate thyroid scan. MTC most often presents as a "cold" or hypofunctioning thyroid nodule, indistinguishable from other thyroid tumors and colloid cysts. The difficulty rests in the follow-up and detection of recurrence in MTC.
INDIUM-111 PENTETREOTIDE
Perhaps the most useful radiotracer for detection of recurrent or metastaticMTC has become "'In pentetreotide or Octreoscan, also referredasto somatostatin receptor scintigraphy (SRS). Krauz reports pathological uptake was detected in 9 of 10 patients with persistent or recurrent MTC, with5 of these foci not seen by other diagnostic modalities by octreotide (CT or MRT) (6).In another series, tumor localization was demonstrated of 18 patients scanning in11of 17patients (7). In a recent comparative, prospective study with MTC, patients were divided into those with macroscopic and minimal disease, and (SRS) (8). (See Figs1-5.) each underwentMRI and somatostatin receptor imaging
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofihy 0 Humana Press Inc., Totma, NJ
389
390
Sweeney and Johnston
I
Fig. 1. A 23-year-old man with medullary carcinoma, elevated calcitonin and CEA levels. Imaging with indium-1 11 pentetreotide (octreoscan) performed before (left) andafter (right) surgery.Thewhite arrows pointtothetumorwithmetastasesinregionalcervicalnodes. (Courtesy Mallinckrodt Medical, Inc., Nuclear Medicine Division, St. Louis, MO)
Histological confirmation was available for 19 “macroscopic” lesions.MRI indicated true positives in 13 lesions and SRS was true positive in 18. In cases of minimal residualdisease,diagnosed by persistentevaluationofcalcitonin, MRI performed poorly compared to SRS. Each imaging modality was compared with selective venous catheterization (SVC).MRI and SVC concurred on only 1 of 7, while SRS detected recurrence in5 of 7. The low sensitivity ofMRI for minimal diseaseis due to the slow
Nuclear
391
Fig. 2. Same patientas Figure 1. This follow-up scan was performed2 months after surgery. On the left, the 24-hour imageof the anterior whole body shows faint uptake in the operative (right). region, but no abnormal uptake, this andis confirmed on the posterior whole-body image
The colon activity isseen becasue no antecedent bowel prep was done. (Courtesy Mallinckrodt Medical, Inc., Nuclear Medicine Division, St. Louis, Mo.)
progression of MTC lymph node metastases, which may result in infiltrated but not enlarged lymph nodes. Don and coworkers (8) also found increased sensitivity of SPECT imaging utilizing octreoscan in patients with recurrent neck disease. SRS appears tobe insensitive for the detection of liver metastases and intrathyroidal tumor (8A).Failure to visualize some lesions of MTC maybe due to insufficient number and density of somatostatin receptors. In vitro studies by Reubi and colleagues (9)
Johnston 392
and
Sweeney
Fig. 3. A patient with known medullary carcinoma of thyroid. Imaging with indium-111 pentetreotide (octreoscan) performed at 24 hours (Zefl) and at 48 hours (right) demonstrate multiple sites of metastasis. (Courtesy Mallinckrodt Medical, Inc., Nuclear Medicine Division St. Louis, MO)
found that lz3ITyr 3-octreotide detected only 6 or 26 tumors (9). The somatostatin receptor density and number in MTC appears to be lower than other neuroendocrine tumors. Well-differentiated tumors and calcitonin-positive tumor sites appear to have a greater number of somatostatin receptors. MTC tumors may also produce s which may compete for receptor binding (10). TECHNETIU"99M (V)DIMERCAPTOSUCCINIC ACID
Pentavalent *Tc dimercaptosuccinicacid(DMSA)appearstolocalizeinMTC due to increased turnover of calcium and phosphate ions and has shown satisfactory tumor:blood,tumor:thyroid,andtumor:muscleratiostopermitgoodscintigraphic images (11).However, experience with DMSA has been limited by difficulties in the preparation and storage of the radioisotope (12). Guerra and coworkers (13) studied 26 patients with medullary thyroid cancer gh"rc with (V) DMSA and found84% overall sensitivity with no false positive results. They suggest that %Tc (V) DMSA should be the first imaging approach inthe follow-up of patients with MTC with persistently (14) found gh"rc (V) DMSA imaging increased calcitonin levels. Clarke and associates to havean overall sensitivityof 95% in detection of85 bone and soft tissue metastase superior to both 1311 MIBG and %Tc MDP in detection of metastatic disease. In the United States, %Tc (V) DMSA is not available commercially but must be prepared
g
NucZear Medicine
393
Fig. 4. Same patient as Figure 2. Metastases demonstrated on left lateral and anterior view of head and chest at 24 hours (Zeff) and right lateral and posterior view of head and chest
at 24 hours (right). (CourtesyMallinckrodtMedical,Inc.,NuclearMedicineDivision, Louis, MO)
St.
in each laboratory by adding sodium bicarbonate to standard DMSA kits. The ratio of gn”rc (IV) DMSA to *Tc (V) DMSA mustbe monitored usingthinlayer chromatography. Small changes in bicarbonate/DMSA levels cause dramatic worsening of image quality and sensitivity. TC-99M METHYLENE DIPHOSPHONATE gn”rc methylene diphosphonate (MDP) has been used for years for nonspecific but extremely sensitive detection of bone metastases of a variety of tumors, including medullary thyroid cancer (14). 25% of MTC patients will present with bone and/or (15). liver metastases, therefore MDP should be utilized in a limited but important role Nonskeletal metastases of MTC have also been imaged using T c MDP. This may be (15). related to the propensity of MTCto calcify or the association with amyloid deposits
THALLIUM-201 CHLORIDE AND TECHNETIUM-99M 2-METHOXYISOBUTYL ISONITRITE The tumor-seeking properties of 2 0 1 T l have been utilized in the diagnosis and followup of patients with differentiated papillary and follicular thyroid cancer (16). Both 99”rc MIBI and ‘ O I T l have an affinity for a variety of neoplasms and appear to localize due to increased blood flow or increased potassium content in tumors. Thereis some
Johnston 394
and
Sweeney
I
Fig. 5. Same patient as Figure 2. Metastases demonstrated on anterior abdominalview (Zef) and posterior abdominalview (right) at 48 hours. (Courtesy Mallinckrodt Medical, h . , Nuclear Medicine Division, St. Louis, MO)
indication that uptake also occurs in medullary thyroid cancer (17). Hoefnagel and associates (Id), found positive thallium uptake in 910ofpatients with elevated calcito 20 minutesafterinjection)maybesuperiorto nin/CEAlevels. Early scanning (at delayed scanning (at 3 hours) (17). More recently, %Tc sestamibi has been utilized in a small number of patients with superior localization due to the addition of SPE images (18). IODINE-131 (OR IODINE-123) METAIODOBENZYLGUANIDINE
Metaiodobenzylguanidine (MIBG)is a radiolabeled norepinephrine analogue. MIB is taken upby cells through a specific, energy-dependent transport mechanism in w the tracer competes with norepinephrine. Although the overall sensitivity of 1311MIBG imaging for localization of pheochromocytoma has been reported as 87%, the same (19). Several results have not been proven in detection of medullary thyroid cancer case reports confirm accurate localization of primary and recurrent MTC in sporadic and familial cases using 1311 MIBG (20-22). However, a literature review conducted by Skowsky and Wilf (12), including 12 series of patients, suggests an overall truepositive incidence of only 30% and a false-negative incidence of 52% in 97 patients undergoing MlBG imaging.In a comparative study, Clarke and colleagues(14) found 61% with that only 12% of metastatic lesions were identified with MIBG compared to =Tc MDP and 95% with 9h”rc DMSA.
Nuclear Medicine Imaging
395
Table 1 Medications that Inhibit the Uptakeof 1311 MIBG Antihypertensive agents-including reserpine, labetalol, calcium channel blockers Tricyclic antidepressants-including amitriptyline, imipramine, doxepine, amoxepine, loxapine Sympathomimetic agents-phenylephrine, phenylpropanolamine, pseudoephedrine, ephedrine (included in many OTC decongestants) Cocaine Adrenergic blockers-including bretylium and guanethidine
Imaging with l3II MIBG is fraught with technical difficulties, due to the I3'I label. The thyroid gland must be blocked with cold iodine administration prior to the study. This can be performed with 1 drop of saturated solution of potassium iodide (SSKI) given orally 3 times a day for a totalof 5 days, beginning the day before injection of theradiopharmaceutical.Thehigh-energygammarays(364keV) of alsomake imaging technically difficult and patient dosimetry unfavorable. In addition, several common medications can blockMIBG uptake and should be avoided in these patients before imaging. These are listed in Table 1. MONOCLONAL ANTIBODY IMAGING Two recent studies have evaluated the use of immunoscintigraphy with "'In-labeled F (ab')z anti-CEA monoclonal antibodies in a total of13 patients (23,24).The reported sensitivities in each small study was in the range of 80-83%. SPECT imaging is reported to increase sensitivityin one of the studies(24). However, in both groups of patients, anti-CEA antibody imaging is hampered by the nonspecific uptake of the of radiolabel in the spleen, kidneys, bone marrow, and liver, thereby limiting imaging the abdomen. Another recent report studied 26 patients with known or occult MTC with diagnostic imaging using %TC-, lZ3I,13'I-labeled anti-CEA antibodies (25). They report sensitivity for detection of known sites of disease ranging from 76% to loo%, when compared with CT,MRI,bone scan, and other imaging modalities. The antibody scan also identified sites of disease in 7 of 9 patients with previous occult disease. A newer agent, 99mTc monoclonal anti-CEA antibody BW431/26 does not bind to bone marrow and blood cells and has also shown promise (26). TREATMENT
lS1Ihas not been considered useful as therapy for medullary thyroid cancer due to the inability of parafollicular cells to concentrate iodine. However, there have been several case reports describing the use of l3II therapy in selected patients with recurrent or persistent disease in the region of the thyroid bed. I3'I is trapped by the remaining follicular cellsof the thyroid, thereby locally irradiating the adjacent residual medullar (28) measured thyrocalcitonin levels thyroid carcinoma(27,28). Hellman and colleagues before and afterthis therapy and concluded that all local residual disease was eradicated by this procedure. Distant metastases and neck metastases located beyond the short range of beta radiation could not be treated inthis manner.
396
Sweeney and Johnston
Lack of localization of medullary thyroid cancer with MIBG is discouraging because of the potential for treatment of these tumors with '3'I-labeled MlBG. Patients with malignant pheochromocytoma treated with I3'I MIBG have shown a subjective and objective response rate of greater than 50% (29). Fuzy and Kornyei (30) in Hungary have reported regression and palliation in several patients treated with 3700 MBq of I3'I MIBG for MTC. In another small case report, two patients were reported to have 0 0mCi therapeutic doses of I3'I marked improvement in pain and diarrhea following1 MIBG but showed no significant long-term change in calcitonin levels (31).Tangible uptake of MIBG by the tumor must be established before a therapeutic trial. Studies of the useof monoclonal antibody immunoscintigraphy in medullary thyroid cancer have been limited but these agents offer another possible treatment applicatio I3'I(24,32,33). Another recent report describes the use of an int when labeled with "'In pentetreotide ative nuclear probeto localize tumor tissue in the neck labeled with (SRS) preoperatively (34). SUMMARY
Overall, the most sensitive and specific single agent for detection of recurrent or metastatic MTC at this time appears to be " * I n pentetreotide, octreoscan. It is readily available and easily imaging using conventional methods, However, a multimodality approach to detection of MTC is most prudent due to the lack of an optimal imaging method. MRI and ultrasound are particularly useful in detecting cervical disease and recurrences, while 99mTc-MDPis sensitive for bone metastases. 99"rc MIBI or '''Tl chloride have a role in whole-body surveillance but the false positive rates are high. Monoclonal antibody imaging agents and 99mTc (V) DMSA are not readily available I3'I MIBGhavebeenverydisappointing,in for routine imaging. The results with addition to its unfavorable imaging characteristics and dosimetry.
REFERENCES
1. Wohllk N, Cote GJ, Evans DB, Goepfert H, Ordonez NG, GagelRF. Application of genetic screening information to the management of medullary thyroid carcinoma and multiple endocrine neoplasiatype 2. Endocrinol Metab Clin NorthAm 1996; 25:l-26. 2. Simpson WJ, Palmer JA, Rosen IB, Mustard RA. Management of medullary carcinoma of the thyroid. Am J Surg 1982; 144:420-422. 3. Sisson JC. Medical treatmentof benign and malignant thyroid tumors. Endocrinol Metab Clin North Am 1989; 18:359-387. 4. Gorman B, Charboneau J W , James EM, et al. MeduUary thyroid carcinoma: role of highresolution US. Radiology 1987; 162:147-150. 5. Dorr U, Sautter-Bihl ML, Bihl. The contribution of somatostatin receptor scintigraphy to the diagnosisof recurrent medullary carcinoma of the thyroid. Semin Oncol1994;21 :42-45. 6. Krausz Y,Ish-Shalom S, DeJong RBJ, et al. Somatostatin-receptor imagingof medullary thyroid carcinoma. Clin Nucl Med 1994; 19:416421. 7. Kwekkeboom DJ, Reubi JC, Lamberts SWJ, et al. In vivo somatostatin receptor imaging in medullary thyroid carcinoma. J Clin Endocrinol Metab1993; 761413-1417. 8. Dorr U,Wurstlin S, Frank-Rave K, et al. Somatostatin receptor scintigraphy and magnet resonance imaging in recurrent medullary thyroid carcinoma: a comparative study. Ho Metab Res (Suppl) 1993; 27:48-55.
Nuclear
397
8A. Frank-Rave K, Bihl H, Dorr V, Buhr H, Ziegler R, Rave F. Somatostatin receptor imaging in persistent medullary thyroid carcinoma. Clin Endocrinol 1995; 42:31-37. 9. Reubi JC, Chayvialle A, Franc B, Cohen R, Calmettes C, Modigliani E. Somatostatin receptors and somatostatain content in medullary thyroid carcinomas. Lab Invest 1991; 64:567-573. 10. Pacini F, Elisei R, Arelli S, Bas010 F, Cola A, Pinchera A. Somatostatin in medullary thyroid cancer: in vitro and in vivo studies. Cancer 1989; 63:1189-1195. 11. Miyauchi A, Endo K, Ohta H, et al. 99m-Tc(V)-Dimercaptosuccinic acid scintigraphy for medullary thyroid carcinoma. World J Surg 1986; 10:640-645. 12. Skowsky WR, Wilf LH. Iodine 131 Metaiodobenzylguanidine scintigraphyof medullary carcinoma of the thyroid. Southern Med J 1991; 84:636-641. 13. Guerra UP, Pizzocaro C, Terzi A, et al. New tracers for the imaging of the medullary thyroid carcinoma. Nucl Med Commun 1989; 10:285-295. 14. Clarke SEM, Lazarus C R , Wraight P, Sampson C, Maisey MN. Pentavalent [99m-Tc] DMSA, [131-11 MIBG,and [99m-Tc] MDP: an evaluation ofthree imaging techniques in patients with medullary carcinomaof the thyroid. J NuclMed 1988; 29:33-38. 15. Johnson DG, Coleman RE, McCook TA, Dale JK, Wells SA. Bone and liver images in medullarycarcinomaofthethyroidgland:concisecommunication. J NuclMed 1984; 25:419422. H R , deVijlder JJM. Role of thallium-201 total body 16. Hoefnagel CA, Delprat CC, Marcuse scintigraphy in follow-upof thyroid carcinoma. J Nucl Med 1986; 27:1854-1857. 17. Koizumi M, YamadaY, Nomura E, et al. Scintigraphic detection of recurrenceof medullary thyroid cancer. Ann Nucl Med 1995; 9:lOl-104. 18. Lebouthillier G, Morais J, Picard M, Picard D, Chartrand R, D’Amour P. Tc-99m sestamib and other agents in the detection of metastatic medullary carcinoma of the thyroid. Clin Nucl Med 1993; 18:657-661. J, Beierwaltes WH. Iodine-131 metaiodo19. Shapiro B, Copp JE, Sisson JC, Exre PL, Wallis benzylguanidine for the locating of suspected pheochromocytoma: experience in 400 cases. J Nucl Med 1995; 26:576-585. 20. Sone T, Fukunaga M, Otsuka N, et al. Metastatic medullary thyroid cancer: localization with iodine-131 metaiodobenzylguanidine.J Nucl Med 1985;26: 604-608. 21. Ansari AN,Siege1ME, DeQuattroV, GazarianLH. Imaging of medullary thyroid carcinoma and hyperfunctioning adrenal medulla using iodine-13metaiodobenzylguanidine. 1 J Nucl Med 1986; 27~1858-1860. 22. Endo K, Shiomi K, Kasagi K, etal.Imagingofmedullarythyroidcancerwith131-1MIBG. [Letter]. Lancet 1984; 2:233. JP, Chetanneau A, Saccavini JC, Chatal JF.Immunoscintigra23. Vuillez JP, Peltier P, Caravel of anticarcinoembryonic antigen monoclonal phy using 111-In-labelled F (ab’)* fragments antibody for detecting recurrences of medullary thyroid carcinoma. J Clin Endocrinol Meta 1992; 74~157-163. 24. O’Byme KT, Hamilton D, Robinson I, Sweeney E, Freyne PG, Cullen MJ. Imaging of medullary carcinoma of the thyroid using111-In labelled anti-CEA monoclonal antibody fragments. Nucl Med Commun 1992; 13:142-148. 25. Juweid M, Sharkey RM, Behr JM, et al. Improved detectionof medullary thyroid cancer with radiolabeled antibodies to carcinoembryonic antigen. J Med Nucl 1996; 37(suppl):9P. 26. Fritzsche H. Immunoscintigraphy in medullar thyroid cancer with Tc-99m-labelled monoclonal anti-CEA antibodies (BW431/26). Proc Int Thyroid Conf 1991; 10:41. 27. Deftos LJ, Stein MF. Radioiodine as an adjunct to the surgical treatment of medullary thyroid carcinoma. J Clin Endocrinol Metab 1980; 50:967-968. R. Radioio28. Hellman DE, Kartchner M, Van Antwerp JD, Salmon SE, Patton DD, O’Mara dine in the treatment of medullary carcinoma of the thyroid. J Clin Endocrinol Metab 1979; 48:451455.
398
Sweeney and Johnston
29. McEwan AJ, Shapiro B, Sisson JC, Beierwaltes WH, Ackery DM. Radioiodobenzylguanidine for the scintigraphic location and therapy of adrenergic tumors. Semin Nucl Med 1985; 15~132-153. 30.FuzyM,KornyeiJ. 131-1”IBG therapy of widespread medullary thyroid carcinoma. [Abstract]. Eur J Nucl Med 1994; 21:743. 31. Clarke SEM, Lazarus CR, Edwards S , et al. Scintigraphy and treatment ofmedullary carcinoma of the thyroid with iodine-131 metaiodobenzylguanidine. J Nucl Med 1987; 28:1820-1825. 32. Zanin DEA, van Dongen A, Hoefnagel CA, Bruning PF. Radioimmunoscintigraphy using iodine-131 anti-CEA monoclonal antibodies and thallium-201 scintigraphy in medullary thyroid carcinoma: a case report. J Nucl Med 1990;31:18541855. 33. Juweid KM, Sharkey RM, Behr TM, et al. Treatment of medullary thyroid cancer (MTC) withradiolabelledmonoclonalantibodies (MABs) againstcarcinoembryonicantigen (CEA). [Abstract]. J Nucl Med 1996; 37(suppl):243P. 34. Waddington WA, Kettle AG, Heddle RM, Coakley AJ. Intraoperative useof indium-1 11 pentetreotide and a nuclear surgical probe. Eur J Nucl Med 1994; 2363-364.
45 Management of Medullary Carcinomaof the Thyroid su*ge?Y Orlo H. Clark Medullary thyroid cancer accounts for about 6-10% of all thyroid malignancies. About 70% of patients with medullary thyroid cancer have sporadic disease and30% have either familial MTC, MEN 2A or MEN 2B ( I ) . The different clinical types of MTC are associated with different but specific ret point mutations. About 50% of patients with sporadicMTC also have somaticret mutations in their medullary cancers that appear to correlate with specific mutations (2). For example, the918 mutation that is associated withMEN 2B when found as a somatic mutationin the tumor of patients with medullary thyroid cancer is associated with a worse prognosis (2). Most patients with MTC are detected by fine needle biopsy and in patients with possible familial disease by testing blood samples for a ret mutation or by basal and pentagastrin or calcium-stimulated calcitonin levels (2-5). The diagnosisof MTC when suspected by cytological examination should be confirmed by testing blood CT and CEA levels and or by doing CT stains of the cytological specimen. All patients with suspected MTC by cytological examination should have a serum calcium level to determine whether hyperparathyroidism is present and more importantly musthavea%-hoururine for catecholamine and metanephrine levels to rule out a pheochromocytoma. If present, the pheochromocytoma takes precedence over the medullary cancer and after appropriate alpha blockade, hydration and often beta blockade (for tachycardia) the pheochromocytoma should be removed. The treatmentof choice for patients with medullary thyroid cancer is total thyroidecor without removing the upper tomy and a meticulous central neck dissection with thymus (6).I request an ultrasound examination of the thyroid and nodes preoperatively and my central neck dissection is more vigorous on the side of any focal defects in the thyroid. Parathyroid glands should be marked but not removed unless the patient has primary hyperparathyroidismor the parathyroid gland or glands appear abnormal. However, any parathyroid gland that appears possibly devascularized should be autotransplanted into the neck musclein familial, sporadicMTC, and in MEN 2B patients, and into the forearmof patients with MEN 2A. An ipsilateral modified neck dissection is recommended for: a) patients with palpable cervical nodes; b) patients with central neck nodes; c) patients with primary thyroid tumors greater than 2 cm in size. From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by:L. Wnrtofsky 0 Humana Press Inc., T o t m , NJ
399
400
Clark
Studies by the Wells group (7) and at the Mayo Clinic (8) document that although about 70% of patients who present with nodal metastases are alive at 10 years, very few of these patients are calcitonin-negative. Patients with more than occult familial or bilateral disease need bilateral modified radical neck dissections. The parathyroid glands are at more risk in patients with medullary thyroid cancer than in patients w papillary or follicular thyroid cancer because of the required meticulous central neck dissection. As previously mentioned, when a parathyroid gland cannot be saved onits vascular pedicle it should be removed, washed in saline, biopsied to confirm it is parathyroid and immediately autotransplanted either to the sternocleidomastoid mu on the side with less or no tumor, and in patients with MEN 2A autotransplanted to the forearm since subsequent hyperplasia and hyperfunction may occur. ret oncogene positiveor who have elevated basal calciton Family members who are in response to pentagastrin or calcium should have totalathyroideclevels that increase tomy and central neck dissection after age5. As mentioned above, I recommend doing an ultrasound to document whether there are any intrathyroidal nodules or adjacent lymph nodes preoperatively. This helps with planning the operation. Postoperatively, patients are followed with calcitonin and CEA levels. MRI or CT scans are useful for documenting recurrent disease in the neck and/or mediastinum. Selective venous is useful for detecting distant metasta catheterization of the hepatic and cervical veins that are usually situated in the liver, lungs, or bone (9). Radiation therapy may be palliative in some patients.
REFERENCES
1. Az4dian A, Rosen IB, Walfish PG, Asa SL. Management considerations in Hlirthle cell carcinoma. Surgery 1995; 118:711-714; discussion 714-715. 2. Zedenius J, Larsson C, Bergholm U, Bovee J, Svensson A, Hallengren B, et al. Mutations of codon 918 in the ret proto-oncogene correlate to poor prognosisin sporadic medullary thyroid carcinomas. J Clin Endocrinol Metab 1995; 80:3088-3090. 3. Eng C. Seminars in medicineof the Beth Israel Hospital, Boston: theret proto-oncogenein multipleendocrineneoplasiatype 2 andHirschsprung’sdisease. N EnglJMed, 1996; 335~943-951. 4. Wells SA Jr, Chi DD, Toshima K,Dehner LP, Coffin C M , Dowton SB, et al. Predictive DNA testing and prophylactic thyroidectomy in patients at risk for multiple endocrine neoplasia type 2A. Ann Surg 1994; 220:237-247; discussion 247-250. 5. Lips CJ, Landsvater RM, Hoppener J W , Geerdink RA, Blijham G, van Veen JM, et al. Clinical screening as compared with DNA analysis in families with multiple endocrine neoplasia type 2A [see comments]. N Engl J Med 1994; 331:828-35. S, Salander H. Reoperation in the treatmentof asymptomatic 6. Tisell LE, Hansson G, Jansson metastasizing medullary thyroid carcinoma. Surgery1986; 9950-66. 7. Lairmore TC, Wells SA Jr.Medullarycarcinomaofthethyroid:currentdiagnosisand management. Semin Surg Oncol 1991; 7:92-99. 8. Gharib H, McConaheyWM, Tiegs RD, BergstralhET,Goellner J R , Grant CS, al. et Medullary thyroid carcinoma: clinicopathologic features and long-term follow-up65 ofpatients treated during 1946 through 1970. Mayo Clin hoc 1992; 67:934-940. 9. Gautvik K M , Talle K, Hager B, Jorgensen O G , Aas M. Early liver metastases in patients with medullary carcinoma of the thyroid gland. Cancer 1989; 63:175-180.
46 Medullary Carcinoma Management E x t m l Radiation Therapy Robert L. White and Leonard Wartofsky
In addition to the extensive sections on medullary cancer this in volume, other recent reviews have addressed advances in diagnosis and management (1,2) and the role of external radiation therapy(3). The key to successful external radiation therapyis prior total surgical resection, to include disease in the cervical and mediastinal lymph nodes of patients. Failure of serum calcitonin that are involved at surgery in half to three-fourths levels to fall to within the normal range is an indicationof residual disease. The location of tumor deposits may be identified by a variety of imaging techniques such as CT, MRI, or the isotopic methods described above. Should tumor masses be visualized, a better result with radiation therapy will accrue following a second surgical procedure to resect all bulky disease. Should no tumor be identified in the presenceof elevated serum calcitonin concentrations, empirical radiation therapy to the cervical and med nal nodes is indicated. If the surgeon is concerned that the extent of the thyroid carcinoma is such that complete removal is not possible, then the use of preoperative external radiation may shrink or occasionally stabilize the tumor mass. Surgery following radiation may technically be easier and with less risk of operative blood loss when preoperative external radiation has been planned. Coordination between the surgeon and the radiation onco gist is very importantin the management of thyroid carcinoma to optimize the patient’s treatment and timing and feeling of security. Medullary thyroid canceris moderately radiosensitive butis generally less sensitive than differentiated thyroid carcinoma. Fortunately, medullary carcinoma is more responsive than anaplastic cancer. Elective radiation therapy after thyroidectomy and node dissection which normalizes serum calcitonin is not recommended on a prophylactic basis. Moreover, the efficacy of radiation therapy this for tumor continues to be argued, with some authorities claiming the treatment plays a very small role(4), while others adhere to the position that treatment combined with appropriate and complete prior surgical resection offers the best hope(5). Indeed, the latter workers reported a 5-year survival rate of 97% in patients with medullary carcinoma treated with external radiati after surgery, whereas survival averaged62% in those managed without postoperative
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited b y :L. WmtOFky 0 Humana Press Inc., Totown, NJ
402
402
White and Wartofsky
radiation therapy. After surgery, the primary indication for megavoltage external irra tion is bulky unresectable thyroid carcinoma, particularly because medullary carcin does not accumulate l3II,and radioiodine should not be used in an attempt to achieve local control of the tumor (6). This is so even though there were a few early case reports (7,8), suggesting a beneficial effectof radioiodine, presumably due to death of C cells by the I3'I concentrated by surrounding thyroid follicular cells. When distant metastases are present, external radiation is indicated to prevent patho logical fractures in bone and for palliation of symptoms. When there is a possibility of pathological fracturein the case of bone metastasis, stabilization with an intramed lary rod or other orthopedic procedure should precede the external radiation. Relativ high doses have been employed for bone metastases, for example, in 40-50 fractions Gy of 2 Gy. Other common sites for medullary carcinoma include lymph nodes, lung, and occasionally brain. When a patient develops brain metastasis, external radiation is indicated for an often reliable and rapid response, but is problematic to use for lung metastases because of the risk of radiation fibrosis and respiratory compromise. Liver metastasis may be amenable to surgical resection. Dose levels of 3500-4500 cGy in 3 to 4.5 weeks are recommended for optimal palliation of metastasis to soft tissue. Chemotherapy of medullary carcinoma has not been impressively successful, but some benefit may be gained by employing external irradiation in sequence or conjunction with chemotherapy, particularly where the thyroid carcinoma is poorly differentiated and/or aggressive. Since time dose relationships between external radiation and sy chemotherapy have not been optimized, local agreement between the medical on and radiation oncologist is important to help patients understand the importance of coordinating their treatment. For curative treatment for thyroid carcinoma with external megavoltage irradiatio there are many technically demanding details. The definitive dose for residual or bulky thyroid carcinoma is 6500 cGy in 7 weeks with a daily dose of 180-200 cGy daily 5 days a week. The treatment volume should include the entire thyroid gland, the right and left cervical lymph nodes, right and left supraclavicular nodes, and the superior mediastinum (9). It is necessary to pay particular attention to the spinal cord dose. Special blocking techniques with a cerrobend blocking system should limit the dose to the spinal cord as well as other radiation sensitive structures. All of the trea areas where microscopic or small deposits of cancer could be present shouldbe treated with doses to 5000 cGy over5 to 6 weeks time. The spinal cord is shielded after 45 cGy in 4.5 to5 weeks time. If chemotherapy is used in conjunction with radiation, the cumulative spinal cord dose should be 500 cGy less. Where tissue thiclmess results doses of less than 5000 cGy in 5 to 6 weeks, boosting techniques must be employed to ensure that the dose is as uniform as possible. There are several methods of radiation beam arrangements and portals that allow In most cases, an anterior adequate doses to be delivered to the neck and mediastinum. to posterior and posterior to anterior set of portals with @'Co, 4 or 6 MV photons will allow 4500-5000 cGy to be delivered in 4.5 to 6 weeks time. Boosting techniques utilizing electron ports of 8-14 MeV can supplement the areas treated to 4500-5000 cGy to definitive doses of 6500-7000 cGy in 5 to 8 weeks time. To avoid the spinal cord,inadditiontocerrobendblocking,obliqueanteriorportalswithwedgesare
Medulla Management y Carcinoma
403
occasionally utilized. Someof the newer treatment techniques include arching or rotational fields with flying wedges (Fig. 2) to optimize external irradiation to the treatmen volume while minimizing treatment to the spinal cord or other critical structures. For local or regional control postoperatively in patients who have elevated calcitonin levels but no evidenceof anatomically resectable disease, Brierly and Tsang(3) recommend two phases of therapy. First, a total dose of 4000 cGy given over four weeks in 20 fractions of 200 cGy to the cervical, supraclavicular and mediastinal nodes; followed by an additional 1000 cGy to the thyroid bed given to the area of the original thyroid bed in five 200 cGy fractions. Some reports have indicated improved local control of tumor but little effect on ultimate survival (10-12). Brierley and colleagues (12) noted good local control of 86% at 10 years in 25 patients treated with surgery and radiation, while 15 patients The difference treated with surgery alone demonstrated a 52% rate of local(12). control was more impressive in the group of patients reported by Mak and colleagues (10) who achieved local control with postoperative radiationin 84% compared to only13% of those patients managed without postoperative radiation therapy. In view of these data, Brierley and coworkers(12) recommend external radiation therapyin all patients of regional or local recurrenceof disease, while recognizing deemed to be at high risk that there may be little or no survival benefit. Patients who receive systemic chemotherapy and external irradiation concurrently or sequentially should not be treated with daily doses to exceed 180 cGy because of the possibility of undesirable dose-potentiating side effects. Daily management for the patient receiving combinations of chemotherapy and external irradiation is difficult and requires close surveillance and observation. Usually the side effects of oral mucositis, esophagitis, and skin erythema are worse for patients treated with combined modalities and patients need to be carefully and cautiously observed regularly (13). Interstitial irradiation might be helpful and valuable in the treatment of medullary thyroidcarcinomasinbothprimaryandmetastaticsites.Removableandperma*=Irhas been nently implantedlZI have been utilizedin the clinical setting. In addition, implanted into mediastinal masses metastatic from thyroid carcinomas and sarcomas, Since thereis minimal general experience and few patients have treated, the interstitial In experienced hands, the interstitial irradiation treatment has not been widely publicized. techniques have produced long term disease free survival in patients and improved local control. The advantage of interstitial irradiation includes minimal side effects and complications and improved local responsiveness, but the clinical experience is limited.
REFERENCES 1. Heshmati I", Gharib H, van Heerden JA, Sizemore GW. Advances and controversies in the diagnosis and management of medullary thyroid carcinoma.Am J Med 1997; 103:60-9. 2. Marsh DJ,Learoyd DL, Robinson BG. Medullary thyroid carcinoma: recent advances and management update. Thyroid 1995; 5:407424. 3. Brierley JD, Tsang RW. External radiation therapy in the treatmentof thyroid malignancy. Endocrinol Metab Clin N Am 1996; 25:141-157. 4. Samaan N,Schultz P, Hickey R. Medullary thyroid carcinoma: prognosis of familial versus sporadic disease and the role of radiotherapy. Medicine 1988; 67:801-805.
404
White and Wartofiky
5. Jensen MH, Davis RK, Derrick L. Thyroid cancer: a computer-assisted review of 5287 cases. Otolaryngol Head NeckSurg 1990; 10251-65. 6. Lindberg RD. External beam irradiation in thyroid carcinomas. In Flechor GH, editor. Textbook of radiotherapy, 3rd ed. Philadelphia: Lea & Febiger, 1980; 384-388. 7. Nusynowitz ML, Pollard E, Benedetto A R , et al. Treatment of medullary carcinomaof the thyroid with 1-131. J Nucl Med 1982; 23:143. 8. ParthasarathyKL,, ShimaokaK, Bakshi SP, et al. Radiotracer uptakein medullary carcinoma of the thyroid. Clin Nub: Med 1980; 545. 9. Moss WT, Brand WN, Battifora H. The Thyroid.In Radiation oncology: rationale, techniqu results, 5th ed. St. Louis: CV Mosby, 1979: 233-242. 10. Mak A, Morrison W, Garden A, et al. The valueof postoperative radiotherapyfor regional medullary carcinoma of the thyroid. Int J Radiat Oncol Biol Phys 1994; 30:234. in medullary 11. Nguyen T,Chassord J, Lagarede P. Results of postoperative radiation therapy carcinoma of the thyroid. Radiother Oncol 1992; 23:l-5. 12. Brierley J, Tsang R, Simpson WJ, Gospodarowicz M, SutcliffeS, Panzarella T. Medullary of radiation therapy thyroid cancer: analysisof survival and prognostic factors and the role in local control. Thyroid 1996; 6:305-310. 13. Greenfield LD.Thyroid tumors. In Perez CA, Brady LW, editors. F’rinciples and practice of radiation oncology. Philadelphia: JB Lippincott, 1987: 1126-1 156.
47 Medullary Carcinoma of the Thyroid Chemotherapy Lawrence S. Lessin and My0 Min
Like other neuroendocrinetumors, medullary carcinoma of thyroid runs a protracted course. Early detection and surgery provides the only curative approach. Extension of the tumor beyond the thyroid capsule is the most significant prognostic indicator and For local and distant whenpresentthediseasecannotbecuredbysurgeryalone. or external metastases of medullary carcinoma which cannot be treated by surgery beam radiation therapy, chemotherapy is used. Doxorubicin used either singly or in combination is the most widely applied chemotherapeutic agent. Controversy exists regarding the chemosensitivity of medullary carcinoma. Gottlieb and Hill ( I )treated a variety of thyroid cancers with doxorubicin, including medullary thyroid cancers, and found 3 partial responses of 5 patients treated. They noted that disease-related diarrhea improved not only in tumor responders but also in one patient who did not achieve a response. (See Table 1.) Benker and Reinwein(2) and De Besi and colleagues(3) treated medullary thyroid carcinoma with Doxorubicin alone and found a higher response rate than in differ and anaplastic carcinomas. By contrast, Scheruble and colleagues (4) treated 10 patients with advanced medullary thyroid carcinoma with combination chemotherapy using doxorubicin, bleomycin and vindesine; response rate was poor with only one partial response. Although 6 patients had stable disease, calcitonin and CEA tumor markers continued to rise. Similarly, Athanassaides and associates(5) treated six patients with doxorubicin and cisplatinum and found no response. Droz and coworkers (6)reported their experience over a 10-year period using five different protocols with both single agent and combination chemotherapy.Of 41 treatments, only2 partial responses were noted. Both partial responses occurred in patients receiving doxorubicinat a dose rate 3 months in both of 60 mg/m* every 4 weeks. Response duration was brief lasting only (7) treated a patient with medullary thyroid cancer by lowcases. Porter and Ostrowski dose doxorubicin at15 mg/m2 per week and achieved a complete response; the patient remained in remission for 18 months. 111trial data, we can conclude that combination Although there are no definitive Phase chemotherapy appears to have no advantage over single-agent doxorubicin in treatmen of advanced medullary thyroid carcinoma. Massart and colleagues (8) showed over
From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofiky 0 Hurnana Press lnc., Totown, NJ
405
406
Lessin and Min
Table 1 Chemotherapeutic Trialsin Treatment of Medullary Thyroid Cancer Response:
(Ref.) AuthorPatients
Used
Agent ~
Gottlieb and Hill, 1974 (1) Shimaoka, 1985 (15)
~~
~~~
Doxorubicin 60-75 mg/mz q3 weeks Doxorubicin 60 mg/m243 wks; doxorubicin 60 mg/mz plus cisplatin40 mg/m2/q3 wks Benker and Reinwein, 1983 (2) Doxorubicin at various dose schedule Hoskin and Harmer, 1987 (9) Doxorubicin, bleomycin, and vincristine Droz et al., 1984 (16) Doxorubicin 60 mg/mz q4 wks Scheruble et al., 1990 (4) Doxorubicin 50 mg/m2, cisplatin60 mg/mz, and vindesine 3 mg/mz Athanassaiade et al., 1988 (5) Doxorubicin 50 mg/m2 and cisplatin70 mg/ mz q3 wks Frame et al., 1988 (17) Doxorubicin 20 mg/m2 and streptozotocin 1000 mg/m2 De Besi et al., 1991 (3) Doxorubicin 60 mg/mz bleomycin 30 U x 3 days, and cisplatin 60 mg/m2 Burgess et al., 1995 (12) Doxorubicin 45-70 mg/m2, carboplatin 600-750 mg/mz, DTIC600-800 mg/m2, vincristine, 2 mg, all q21 days
3 of5 1 Of4 2 of 6 8 of 20 5 of 13 2 of 18 1 of 10 0 of 6
4 of 9
expression of MDR-1 gene in a human medullary thyroid carcinoma cell line which could explain its multidrug resistance. They also showed that in vitro resistance to 1 pump with cyclospor doxorubicin can be partially reversed by blockageMDRof the A or verapamil. Etoposide has been used as single agent chemotherapy by Hoskin and Harmer (9) and by Kelsen and associates ( I O ) , based onits activity in other neuroendocrine tumors; however only minor responses were seen, A complete response was also reported by Sigurdur and Petursson (11) using darcarbazine and 5-fluorouracil; pulmonary and subcutaneous metastases remainedin remission for 10 months along with suppression (12)reportingfor of CEA andcalcitonintumormarkers.Burgessandcolleagues the M.D. Anderson group, treated 9 patients with metastatic MCT with a four-drug combination including doxorubicin, carboplatin, imidazole carboxamide @TIC), and vincristine.Fivepatientsrespondedforamedian of 5 monthsatthetimeofthe report, with sustained decreases in calcitonin and/or CEA. The principal toxicity was myelosuppression and the authors consideredthis to bean effective, tolerable regimen New regimens have begun to assess the role of high-dose chemotherapy with etic stemcellrescue(autologoustransplant)inmetastaticMCT.Whileanecdotal responses have been reported, this approach is investigational and benefit remains to be determined. Because medullary thyroid carcinoma is a neuroendocrine tumor, octreotide has (13)reported reduction in tumor employed since 1977 when Muller and colleagues (14) reported markersinapatienttreatedwithoctreotide.Mahlerandcolleagues symptomatic improvement of diarrhea with octreotide and reduction in tumor markers,
Chemotherapy
407
but only minimal or no decrease in tumor mass. Tachyphylaxis, which occurs with continueduseofoctreotide, can be temporarily overcome by increasing the dose. Since medullary carcinomais often indolent, quality of life assessment is important in treatment evaluation. Octreotide produces relief of diarrhea with avoidance of electrolyte of lifewithminimalsideeffects. depletion,gaininweightandenhancedquality Interferon a hasbeen combined with octreotide to control symptoms, but without measurable response in tumor mass. In summary, chemotherapyis active against medullary thyroid cancer. Monotherapy with doxorubicin is equivalent to and less toxic than combination chemotherapy with response rates ranging up 40%. to Octreotide can be effectively used to control diarrhea related to this cancer.
REFERENCES
1. Gottlieb JA, Hill CS. Chemotherapy of thyroid cancer with Adriamycin: experience with 30 patients. N Engl J Med 1974; 290193-197. 2. Benker G, Reinwein D. Ergegnisse der Chemotherapie des Schilddrusenkarzinoms. Dtsch Med Wochenschr 1983; 11:403-406. 3. DeBesi P, Busnardo B, Toso S , Girelli ME, Nacamulli D, Simioni N, Casara D, %rat P, Fiorentino MV. Combined chemotherapy with bleomycin, Adriamycin and platinum in advanced thyroid cancer. J Endocrinol Invest 1991; 14:475-480. 4. Scherubl H, Rane F, Ziebler R. Combination chemotherapy of advanced medullary and differentiated thyroid cancer. J Cancer Res Clin Oncol1990; 116:21-23. S. Serial serum calcitonin 5. Athanassiades P, Piperingos G, Pandos P, Koutras D, Moulopoulos concentrations to evaluate response to therapy of patients with medullarty thyroid carcinoma. Chemioterapia 1988; 7:195-197. C. Chemotherapy 6. Droz JP, Schlumberger M, Rougier P, Ghosn M, Garder P, Parmentier in metastatic non-anaplastic thyroid cancer: experience at the Institut Gustave-Roussy. Tumori 1990; 76:480-483. MJ. Medullary carcinomaof the thyroid treated by low-dose Adriamy7. Porter AT, Ostrowski cin. Br J Clin Pract 1990; 44517418. 8. Massart C, GibassierJ, Raoul M, Pourquier P, Leclech G, Robert J, Lucas C. Cyclosporin A,verapamiland S9788 reverse doxorubicin resistance in a human medullary thyroid carcinoma cell line. Anti-cancer Drugs1995; 6:135-146. 9. Hoskin PJ, Harmer C. Chemotherapy for thyroid cancer. Radiother Oncol1987;10:187-194. 10. Kelsen D, Fiore J, Heelan R, Cheng E, Magill G. Phase II trial of etoposide in APUD tumors. Cancer Treatm Rep 1987; 71:305-307. 11. Sigurdur R, Petursson. Metastatic medullary thyroid carcinoma complete response to co nation chemotherapy with dacarbazine and 5-fluorouracil. Cancer 1988; 62:1899-1903. 12. Burgess MA, Sellin RV, Gage1RF. Chemotherapy for medullary carcinomaof the thyroid with doxorubicin, imadazole carboximide, vincristine and cyclophosphamide.Proc Annu Meet Am SOC Clin Oncol1995; 14417. 13. Muller OA, Landgraf R, Zeigler R, ScaribaPC. Effects of somatostatin on calcitonin and ectopic ACTH release in a patient with medullary thyroid carcinoma. Acta Endocrinol 1977; 84 (suppl):49-50. 14. Mahler C, Verhelst J, DeLongueville M, Harris A. Longterm treatment of metastatic medullary thyroid carcinoma with the somatostatin analogue octreotide. Clin Endocrinol 1990; 33~261-269. 15. Shimaoka K, Schoenfeld D, De Wys W, Creech R, De Conti R. A randomized trial of doxorubicin vs doxorubicin plus cisplatin in patients with advanced thyroid carcinoma. Cancer 1985; 56:2155-2160.
408
Lessin and Min
16. Droz JP, Rougier P, Goddefroy V, Schlumberger M, Garder P, Parmentier C. Chemotherap for medullary cancer of thyroid: phase11 trials with adriamycin and cisplatinum administere as monochemotherapy. Bull Cancer 1984; 71:195-199. 17. Frame J, Kelsen D, Kemeny N, Cheng E, Niedzwiecki D, Heelan R, Lippermann R. A phase 11 trial of streptozotocin and Adriamycin in advanced APUD tumors. Am J Clin Oncol 1988; 11:490-495.
VI11 Miscellaneous and Unusual Cancers of the Thyroid
48 Pathology of Miscellaneous
and Unusual Cancers of the Thyroid James Oertel and Yolanda Oertel METASTATIC (SECONDARY) NEOPLASMS
The most common metastatic neoplasms in the gland that may mimic primary tum (1,2). They have the usual range of histological are those from lung, breast, and kidney patterns depending upon the primary sites. Involvement of an existing adenomatoid nodule or adenoma is likely, thereby complicating the morphological features. Immunohistochemical proceduresare helpful in separating such metastatic lesions from primary thyroid neoplasms when there is uncertainty about the interpretation. The materials of thyroglobulin must be evaluated carefully, however, because some nonspecific uptake may occur by the metastatic cells. Malignant melanoma involves the thyroid with moderate frequency, but the patient’ clinical history makes this metastasis rather easy to diagnose. SQUAMOUS CELL CARCINOMA, ADENOSQUAMOUS CARCINOMA, AND MUCIN-PRODUCING CARCINOMA
These are rare, aggressive neoplasms, usually occurring in middle-aged or elderly patients, often in glands containing a well-differentiated carcinoma (especially papillary carcinoma), an adenoma, a multinodular goiter, or (occasionally) chroNc autoimmune thyroiditis (1,3,4). Because of the intimate relationship to neoplastic glandular elements, some squamous cancers have been called adenosquamous carcinomas. Undifferentiated (5). Squacarcinoma may be evident along with the predominant squamous carcinoma mous carcinoma with extensive spindled cell change has been reported in association (6).An occasional squamous carcinoma of the thyroid with tall-cell papillary carcinoma has been associated with hypercalcemia and leukocytosis(7). Mucin-producing carcinoma has been reported and has been associated with squais a complex problem, mous carcinoma(1,8).The presenceof mucosubstances in thyroid and uncertainty about such neoplasms continues (9,lO). Bland focal squamous metaplasia may occur in both follicular and papillary carcinomas, but usually this does not behave as squamous carcinoma.
From: Thyroid Cancer: A ComprehensiveGuide to Clinical Management Edited by: L. Wartofiky 8 Humana Press Inc., Tofowa, NJ
411
412
Oertel and Owfel
Fig. 1. Poorly differentiated carcinoma. Both insular and cribriform patterns are evident. (H&E stain; x30.)
POORLY DIFTERENTIATED CARCINOMA
Poorly differentiated carcinoma is a concept proposed to include carcinomas of follicular thyroid epithelium that retain sufficient differentiation to produce scattered small follicular structures and some thyroglobulin, but which generally lack the usual morphologic characteristics of papillary carcinoma and follicular carcinoma (11-14). Instead, the histological patterns are described as solid, insular of(islands cells separated l), with by connective tissue and artifactual spaces), trabecular, and alveolar (Fig. scattered tiny follicles containing colloid (Fig. 2) within the solid, insular, and regions. These patterns may mix with one another, and small foci of characteristic follicularandpapillarycarcinomaalsomaybefound,evenbothwithinthesame neoplasm. The Bcl-2 protein (a suppressor of apoptosis) has been describedin a large proportion of these tumors, in contrast to undifferentiated carcinoma (15). Much of the cancer may be composed of small, uniform cells with pale, scanty cytoplasm and small, spherical nuclei having uniformly dark, finely clumped ch (Fig. 2). Nuclear contours are smooth. These have been labeled “primordial cells” (14). PAS-positiveglobulesof becauseoftheirresemblancetofetalthyroidcells thyroglobulin may lie next to the nuclei. Such cells may encompass a large proporti also may be of the neoplasm. Medium-sized to large cells with more varied nuclei present and may dominate some parts. Their cytoplasm may be eosinophilic or clear. Although most of their nuclei are regular and round with smooth contours, some are as spindled or large and pleomorphic. Such irregular cells are isolated, not present
Miscellaneous Unusual Cancers and
Thyroid of the
413
Fig. 2. Poorly differentiated carcinoma. Thesolid part has small to medium-sized cells. Two small follicles are filled with dense colloid. (H&E stain; x300.)
regions of anaplastic carcinoma. Oxyphilic cell variants have been described (16). In both “small cell” and “large cell” regions necrosis may be present, either as large field or as tiny foci (especially within the insular parts). Preservation of the neoplastic cells nexttovesselsintheregions of necrosis may cause a “peritheliomatous” pattern. Mitotic figures are variable in number. Usually nucleoli are tiny, but a few tumors have been described with conspicuous nucleoli. Although the majority of the nuclei of these tumors are rounded and have rather evenly distributed heterochromatin (thereby resembling those of follicular carcinoma), portions of these tumors may contain nuclei resembling those of papillary carcinoma. Poorly differentiated carcinoma has been considered a controversial entity because of the varied histologic patterns present and the evidence of well differentiated papillar or follicular carcinoma that can be found as small portions of most of these neoplasms. This is understandable, but does not account for the presence of the “primordial cells” in many of these neoplasms and the fact that they share many characteristics different from both well differentiated cancers and anaplastic cancers (17). We do not have much experience with the cytological appearance of these neoplasm theloose cellsclusters, Reports (18-20) state that the smears are notably cellular with in small sheets or large sheets. Little colloid is evident. Necrosis and hemorrhage are common in the background. Microfollicles, rosettes, trabeculae, and papillae have been noted. Single cells may be present. The cells usually have scant, poorly defined cytoplasm and darkly stained, rounded nuclei. Nuclear grooves may be seen and there are occasional intranuclear cytoplasmic inclusions. Nuclear overlapping can occur.
414
Oerfel and Oerfel
ANGIOSARCOMA
It has been regarded as a rare neoplasm, even in endemic goiter regions (21,221; several tumors with histological features consistent with angiosarcoma have express immunoreactive keratins, as well as endothelial features such as factor VIII-associat of these neoplasms antigen and Weibel-Palade bodies. Some authors suggest that some might be consideredangiomatoid carcinomas (a form of undifferentiatedor anaplastic carcinoma) (231, expressing epithelial markers as well as endothelial characteristics. Most patients are middle-aged or elderly, and most tumors are extremely aggressive, thereby resembling typical anaplastic carcinomas. LEIOMYOSARCOMA
Leiomyosarcoma has been reported, but is extremely rare. Not only is a characteristic histological pattern required, but immunohistochemical and ultrastructural techniques are needed to separate it from anaplastic carcinoma (24). THYROGLOSSAL DUCT CANCER
Nearly all of the tumors arise in the thyroid tissue that accompanies the duct or cyst (25,26). Most are papillary carcinomas (26,27), a very few are follicular carcinomas (281, and rarelyan anaplastic carcinomais present (29). Squamous carcinoma has bee reported (30), presumably arising from the respiratory epithelium of the thyroglossal duct or cyst. (31). Fine needle aspiration(FNA) is useful in diagnosing thyroglossal abnormalities If the aspirate consistsof more than the usual hypocellular specimen from a cyst, the the standard criteria are applied for recognizing thyroid tumors. HYALINIZING TRABECULAR NEOPLASMS
These rare neoplasms are solid masses, often less than 3.0 cm in diameter and well circumscribed (usually encapsulated) (32,33). The cut surfaces are light-colored, and vessels and small foci of fibrosis may be visible. Microscopically, the tumors consist of trabeculae and lobules of elongated, oval, or polygonal cells, usually of medium of cells are surroundedby capillaries size and with poorly defined borders. The groups and variable amounts of eosinophilic, hyalinized material. This consists of clumps of type IV collagenandlaminin. Also contributingtotheeosinophiliczonesarethe of the numerous cytoplasmic microfilaments (presumably keratin) present in many epithelial cells. The neoplastic cells have been described as eosinophilic, amphoph or clear with fine granules apparentin the cytoplasm. Nuclei appear rounded, oval,or elongated, often grooved. They may contain cytoplasmic inclusions and clear zones; these zones in turn contain tiny rods composedof bundles of minute filaments (visible on electron microscopy) (34). Variable numbers of small follicles occur in the trabeculae, some with colloid, some empty. Electron microscopy shows intercellular spaces surrounded by microvi Psammoma bodies may bescattered presumablyrepresentingdevelopingfollicles. through the tumor. Most cells contain immunoreactive thyroglobulin and keratin. C tonin has never been demonstrated.
Miscellaneous Unusual Cancers and
Thyroid of the
415
Rarely, the tumors have been invasive and have involved cervical lymph nodes. An alteration resembling this hyalinizing tumor has been described in the adenomatoid nodules of a nodular goiter (34). Aspirates of these lesions have been confused with follicular neoplasms, papillary carcinoma, and medullary carcinoma (35). Moderate to marked cellularity is evident, with the cells forming clusters and follicles. Colloid is not present, but irregular masses of hyaline material are adjacent to the clusters of cells. This material has been described as having fringed or granular margins with a suggestion of a fibrillary structure. It is purplish-red or magenta with May-Grunwald Giemsa stain (36) and pink to grayishblue with Papanicolaou stain (35). Cells are rounded, polygonal, or elongated with cytoplasm of variable density. Nuclei often contain grooves (in Papanicolaou-stained material)andcytoplasmicinclusions.Smallpsammomabodieshavebeenseenin some cases. NEOPLASMS ASSOCIATED WITH FAMILIAL INTESTINAL ADENOMATOUS POLYPOSIS
These rare neoplasms have trabecular, solid, papillary, and cribriform patterns form by spindled, polygonal, and tall columnar cells and are different from the usual papi and follicular carcinomas (37-39). Small whorls of cells may be found, but these are not squamous foci. Cytoplasm is oxyphilic to amphophilic, sometimes clear. Nuclei are hyperchromatic, slightly irregular, and medium-sized. Nucleoli vary in size and visibility. Cytoplasmic inclusions in the nuclei vary in number and size, and some nuclear grooves may be seen. Focal positivity for thyroglobulin can be found, but colloid is absent or present in minimal amounts. These neoplasms are usually small and multiple, and the majority have occurred in girls and young women. MUCOEPIDERMOID CARCINOMA
These neoplasms are rare, occur mostly in women, and are usually of a low grade of malignancy, sometimes associated with papillary carcinoma (or even present as a metaplastic focus in a papillary carcinoma) (40-43). An associated undifferentiated (anaplastic) carcinoma has been reported (44). They are typically solid, firm, lightcolored masses, not encapsulated, sometimes cystic, sometimes with mucus visible on the cut surfaces. Microscopic examination reveals islands of epithelial cells anastomosing with one another.Thecellsalsoformpseudovascularandcribriformpatternsandirregular glandular spaces (Fig. 3). Mucous cells are scattered among the numerous cells with squamous characteristics. Many mucous cells line the glands and ducts. Squamous cells have variable cytoplasm, ranging from pale to eosinophilic; some are keratinized Keratin pearls may be present (Fig. 4), and some groups of cells have intercellular bridges by light microscopy. Mucous cells have pale, foamy, or granular cytoplasm, and intermediate forms between squamous and mucous cells may be recognized. Nuclei are round or ovoid, medium-sized, generally regular (occasionally atypical), andsomecontaingroovesandcytoplasmicinclusions.Centralnucleolivaryfrom inconspicuous to large. Mitotic figures vary in number depending on the case. Some
41 6
Oertel and Oertel
Fig. 3. Mucoepidermoid carcinoma. Dilated cystic spaces are visible. (H&E stain; x75.)
Fig. 4. Mucoepidermoid carcinoma. Nests of squamous cells and two keratin pearls lie in a heavy infiltrate of lymphocytes. (H&E stain; x100.)
Miscellaneous Unusual Cancers and
Thyroid of the
41 7
of the cells have been reported to contain thyroglobulin (dl), but many do not (42). Virtually all cells contain keratins. Ciliated cells may line the spaces. Within the spaces are colloid-like PAS-positive material, sulfated mucus stained with Alcian blue and mucicarmine, and cellular debris Small PAS-positive hyaline bodies maybefound in the nests and cords of cells. Psammoma bodies may occur (41). Fibrous stroma, often dense, is present, and the tumor may be infiltrated by lymphocytes and both neutrophilic and eosinophilic granulocytes. Lymphocytic thyroiditisis usually present in the remainder of the thyroid gland. Metastases to lymph nodes in the neck have been reported.
THYMIC AND RELATED NEOPLASMS Thymic, parathyroid, and salivary gland tissues may be infound the thyroid(45-47), and therefore it is not surprising that occasional neoplasms occur in the thyroid and inferior partof the neck that resemble the thymus (48-51). Such tumors maybe benign or malignant.
TERATOMA Benign teratomasin newborns and infants may cause various obstruction phenomena, especially when cystic, and must be resected promptly. They do not or spread recur (52). Malignant teratomas have been seen in adults andare composed of primitive epithelial, mesenchymal, and neurectodermal elements(5334). They spread locally and may have metastases. REFERENCES
RA.
Tumors of the thyroid gland.In Rosai J, Sobin LH, 1. Rosai J, Carcangiu ML, DeLellis eds. Atlas of tumor pathology, 3rd Ser, Fasc5. Washington, DC: A.F.I.P., 1992. 2. Nakhjavani MK, Gharib H, GoellnerJ R , van Heerden JA. Metastasisto the thyroid gland: a report of 43 cases. Cancer 1997; 79:574-578. 3. Huang T-Y, Assor D. Primary squamous cell carcinoma of the thyroid gland a report of four cases. A m J Clin Path01 1971; 55:93-98. 4. Harada T, Shimaoka K, Yakumaru K, It0 K. Squamous cell carcinoma of the thyroidgland Transition from adenocarcinoma. J Surg Oncol1982; 19:36-43. 5. Harada T,It0 K, Shimaoka K, Hosoda Y, Yakumaru K. Fatal thyroid carcinoma. Anaplastic transformation of adenocarcinoma. Cancer 1977; 39:2588-2596. 6. Bronner M P , LiVolsi VA. Spindle cell squamous carcinoma of the thyroid: an unusual anaplastic tumor associated withtall cell papillary cancer. Mod Path011991; 4:637-643. 7. Riddle PE, DincsoyHp. primary squamous cell carcinomaof the thyroid associated with leukocytosis and hypercalcemia. Arch Path01 Lab Med 1987; 111:373-374. 8. LiVolsi VA. Surgical pathology of the thyroid. Major Probl Path01 1990; 22253-74. 9. GherardiG.Signetringcell‘mucinous’thyroidadenoma:afolliclecelltumourwith
abnormal accumulation of thyroglobulin and a peculiar histochemical profile. Histopathology 1987;11~317-326. 10. Rigaud C, Bogomoletz W.“Mucin secreting” and “mucinous” primary thyroid carcinoma pitfalls in mucin histochemistry applied to thyroid tumours. J 1987; 40:890-895. Clin Path01 11. Sakamoto A,Kasai N, Sugano H. Poorly differentiated carcinoma of the thyroid: a clinico1983; pathologic entityfor a high-risk groupof papillary and follicular carcinomas. Cancer 52~1849-1855.
41 8
Oertel and Oertel
12. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (“insular”) thyroid carcinoma: a reinterpretation of Langhans’ “wuchernde Struma.” Am J Surg Path01 1984; 655-668. BY, et al. Poorly differentiated 13.HwangTS, Suh JS,Kim YI, ParkSH,KohCS,Cho carcinoma of the thyroid remspective clinical and morphologic evaluation. J Korean Sci 1990; 5:47-52. 14. Papotti M, Botto Micca F, FaveroA, Palestini N, Bussolati G. Poorly differentiated thyroi carcinomas with primordial cell component: a groupof aggressive lesions sharing insular, trabecular, and solid patterns. Am J Surg Path01 1993; 17:291-301. 15.Pilotti S, ColliniP,Del Bo R,CattorettiG,Pierotti MA, Rilke F. A novelpanelof antibodiesthat segregatesimmunocytochemically poorly differentiated carcinoma from undifferentiated carcinoma of the thyroid gland.Am J Surg Path01 1994; 18: 1054-1064. 16. Papotti M, Torchio B, Grassi L, Favero A, Bussolati G. Poorly differentiated oxyphilic (HUrthle cell) carcinomasof the thyroid. Am J Surg Path01 1996; 20:686-694. 17.Sobrinho-SimoesM.Poorlydifferentiatedcarcinomasofthethyroid.EndocrPath01 1996; 7~99-102. 18. Pietribiasi F, Sapino A, Papotti M, Bussolati G. Cytologic features of poorly differentiated “insular” carcinoma of the thyroid, as revealed by fine-needle aspiration biopsy. Am J Clin Pathol 1990; 94:687-692. 19. Sironi M, ColliniP, Cantaboni A. Fine needle aspiration cytology of insular thyroid carc noma: a report of four cases. Acta Cytol 1992; 36:435-439. 20.Pereira EM, Maeda SA, Alves F, Schmitt FC. Poorly differentiated carcinoma (insular carcinoma) of thethyroiddiagnosedbyfineneedleaspiration(FNA).Cytopathology 1996; 7~61-65. 21. Ruchti C, GerberHA, SchaffnerT. Factor Vm-related antigen in malignant hemangioen thelioma of the thyroid additional evidence for the endothelial origin of this tumor. Am J Clin Pathol 1984; 82:474-480. 22. Tijtsch M, Dobler G, Feichtinger H,Sandbichler P, Ladurner D, SchmidKW. Malignant hemangioendotheliomaof the thyroid: its immunohistochemical discrimination from un ferentiated thyroid carcinoma. Am J Surg Pathol 1990; 14:69-74. 23. Mills SE, Gaffey MJ, Watts JC, Swanson PE, WickM R , LiVolsi VA, et al. Angiomatoid carcinoma and “angiosarcoma” of the thyroidgland a spectrum of endothelial differentia tion. Am 3 Clin Path01 1994; 102:322-330. 24. Iida Y, Katoh R, Yoshioka M, Oyama T, Kawaoi A.primary leiomyosarcoma of the thyroid gland. Acta Pathol Jpn 1993; 43:71-75. 25.EllisPDM,vanNostrand A m . Theappliedanatomyofthyroglossaltractremnants. Laryngoscope 1977; 87~765-770. 26. LiVolsi VA,Penin KH, SavetskyL. Carcinoma arisingin median ectopic thyroid (includin thyroglossal duct tissue). Cancer 1974; 34:1303-1315. 27. Jaques DA, Chambers RG, Oertel JE.Thyroglossal tract carcinoma: a review of the lit and addition of eighteen cases. A m J Surg 1970; 120439446. 28. Case WG, Ausobsky J, Smiddy FG, High AS. Primary follicular adenocarcinoma arising in the thyroglossal tract. J R Coll Surg Edinb 1987; 32:250-251. 29. Woods RH, Saunders Jr J R , Pearlman S, Hirata RM, Jaques DA. Anaplastic carcinoma arising in a thyroglossal duct tract. Otolaryngol Head Neck Surg 1993; 109:945-949. 30. Deshpande A, Bobhate SK.Squamous cell carcinomain thyroglossal duct cyst. J Laryngo Otol1995; 109:1001-1004. 31. Shaffer MM, Oertel YC, Oertel JE. Thyroglossal duct cysts: diagnostic criteria by .fineneedle aspiration. Arch Pathol Lab Med 1996; 120:1039-1043. Am 32. Carney JA, Ryan J, Goellner JR.Hyalinizing trabecular adenoma of the thyroid gland. J Surg Pathol 1987; 11:583-591. 33. Sambade C, FranssilaK, Cameselle-Teijeiro J, NeslandJ, Sobrinho-Shoes M. Hydinizing
Miscellaneous Unusual Cancers and
7'hyroid of the
419
trabecular adenoma: a misnomer for a peculiar tumor of the thyroid gland. Endocr Path01
1991; 2~83-91. 34. Chan JKC, Tse CCH, Chiu HS.Hyalinizing trabecular adenoma-like lesion in multinodal goitre. Histopathology 1990; 16:611-614. 35. Goellner J R , Carney JA. Cytologic features of fine-needle aspirates of hyalinizing trabecular adenoma of the thyroid. Am J Clin Path01 1989; 91:115-119. 36. Bondeson L, Bondeson A-G. Clue helping to distinguish hyalinizing trabecular adenoma from carcinomaof the thyroid in fine-needle aspirates. Diagn Cytopathol1994; 10:25-29. 37. Chan JKC, Loo KT. Cribriform variant of papillary thyroid carcinoma. Arch Path01 Lab Med 1990;114622-624. 38. Harach HR, Williams GT, Williams ED. Familial adenomatous polyposis associated thyroid carcinoma: a distinct type of follicular neoplasm. Histopathology1994; 25:549-561. 39. Mizukami Y,Nonomura A, Michigishi T, Noguchi M, Ishizaki T. Encapsulated follicular
thyroid carcinoma exhibiting glandular and spindle cell components: a case report. Path01 Res Pract 1996;192:67-71. 40. Rhatigan R M , Roque JL, Bucher RL. Mucoepidermoid carcinoma of the thyroid gland. Cancer 1977;39:210-214. 41. Sambade C, Franssila K, Basso-de-Oliveira CA, Sobrinho-Shoes M. Mucoepidermoid carcinoma of the thyroid revisited. Surg Path011990; 3:271-280. J, Battifora H, Carcangiu ML, Rosai J. Sclerosing mucoepider42. Chan JKC, Albores-Saavedra moid thyroid carcinoma with eosinophilia: a distinctive low-grade malignancy arising from the metaplastic folliclesof Hashimoto's thyroiditis. Am J Surg Path01 1991; 15:438-448. 43. Wenig BM, Adair CF, Heffess CS. Primary mucoepidermoid carcinoma of the thyroid gland: a report of six cases and a review of the literature of a follicular epithelial-derived tumor. Hum Path01 1995; 26:1099-1108. 44. Cameselle-Teijeiro J, Febles-Pkrez C, Sobrinho-Simoes M. Papillary and mucoepidermoid carcinoma of the thyroid with anaplastic transformation: a case report with histologic and immunohistochemical findings that support a provocative histogenetic hypothesis. Path01 Res Pract 1995;191:1214-1221. 45. Russell WO, IbanezML, Clark RL, White EC. Thyroid carcinoma: classification, intraglandular dissemination, and clinicopathological study based upon whole organ sections of 80 glands. Cancer 1963;16:1425-1460. in the thyroid and cervical region: an explana46. LiVolsi VA. Branchial and thymic remnants tion for unusual tumors and microscopic curiosities. E n d m Path01 1993; 4 115-119. 47. Mizukami Y,Nonomura A, MichigishiT, Noguchi N, Nakamura S. Ectopic thymic tissue in the thyroid gland. Endocr Path01 1993; 4162-164. 48. Harach HR, Saravia Day E, Franssila KO. Thyroid spindle-cell tumor with mucous cysts: an intrathyroid thymoma? Am J Surg Path01 1985; 9:525-530. 49. Chan JKC, Rosai J. Tumors of the neck showing thymic or related branchial pouch dif ation: a unifying concept. Hum Path01 1991; 22:349-367. 50. Mizukami Y, Kurumaya H, YamadaT, MinatoH, Nenomura A, Noguchi M,et al. Thymic carcinoma involving the thyroid gland report of two cases. Hum Path011995; 26:576-579. 51. Shek TWH, Luk ISC, Ng IOL, Lo CY. Lymphoepithelioma-like carcinoma of the thymid gland: lack of evidence of association with Epstein-Barr virus. Hum Path01 1996;27: 851-853. 52. Bale GF. Teratoma of the neckin the regionof the thyroid gland a review of the literature and report of four cases. Am J Path01 1950; 26:565-579. 53. Kimler SC, MuthWF. Primary malignant teratoma of the thyroid case report and literature review of cervical teratomas in adults. Cancer 1978; 42:311-317. 54. Bowker CM, Whittaker RS. Malignant teratoma of the thyroid case report and literature review of thyroid teratoma in adults. Histopathology1992; 21:81-83.
49 Clinical Aspects of Miscellaneous and Unusual Types of Thyroid Cancers Matthew D. Ringel, Kenneth D. Burman, and Barry M. Shmookler
The majorityof epithelial thyroid tumors maintain some degree of thyroid follicular cell function, as evidenced by thyroglobulin production and the ability to concentrate as those seen in papillary iodine. They also have typical histological appearances, such and follicular carcinomas.In this chapter we discuss a group of unusual primary thyroid of differentiated thyroid cellular function neoplasms characterized by limited or absence and structure; they have, in general, more aggressive clinical courses than differentiated carcinomas. We also discuss tumors that metastasize to the thyroid gland. These thyroid (WHO) underthe tumorshavebeenclassifiedbytheWorldHealthOrganization headings of “other”thyroidcarcinomas,nonepithelialtumors,andinthecaseof severalhistologicaltypestobediscussed,variants of papillaryandfollicular carcinoma (1,Z). These tumors represent 5-15% of all thyroid tumors (34 (the majority originate from thyroid follicular epithelium), the most common of which is anaplastic carcinoma (see Chapters 9 and 35). Based upon the latest recommendation of the WHO, the previously described variantsof spindle cell and giant cell carcinoma now are included under anaplastic carcinoma, whereas the small-cell carcinoma variant is not included because nearly all of these tumors have been reclassified as non-Hodgkin’s lymphomas (5-7). In addition, many of the tumors classified as sarcomas in the past have been reclassified as anaplastic carcinomas, although some sarcomas of the thyroid clearly exist. We review the important clinical and histological diagnostic features of these rare, often aggressive tumors, and discuss the therapeutic options.
SQUAMOUS CELL CARCINOMA OF THE THYROID Demographics primary squamous cell carcinoma of the thyroid is a rare disorder comprised of cells of uncertain origin. The current WHO classification defines squamous cell carcinoma as a tumor “comprised entirely of cells showing so-called intracellular bridges andor
From: Thyroid Canm. A Comprehensive Guide to Clinical Management Edited b y : L. Wartofsky 0 Humma Press Inc., Totowu, NJ
421
422
Ringel, Burman and Shmookler
Table 1 Demographics of Patients With Squamous Cell Carcinoma of the Thyroid (n)
Reference Number
Prakash (14) Bahuleyan (15) Harada (11,201 White and Talbert (12)* Kapoor (16) Misonou (17) Tsuchiya (18) Kampsen (19) Sarda (21) Theander (22) Chaudhary (23) Budd (24)
Simpson (10) Huang (25) Bukachevsky (26)t Korvonin (27) Riddle (28)
Shimaoki (29) Saito (13)
TOTAL
1 1 2
1 1 1 3 2 7 1 1 2 8 4 1
4 1 3 1 45
Age (years, mean)
Female (n)
Mule (n)
38 35
1 1
0 0 0 1 0 0 1 0
71 (at death) 61 45 61 63 65 46 72 76 59
60
58 73 62 66 60 71 59
2
0 1 1 2
2 5 1
1 1 3 2 1 3 0 1 1 29
2 0 0
1 5 2 0 1 1 2 0 16
*Thyroglossal duct tumor. ?Lingual thyroid cancer.
forming keratin.” (I) This definition of squamous cell carcinoma is quite important 43% of papillary carcinomaswill contain regionsof squamous cell metaplas since up to and many anaplastic carcinomas will be comprised,in part, of squamoid regions (8,9) (see Chapter35). Adenosquamous cell carcinomas and adenoacanthomas, tumors com prised of regions of squamous cell carcinoma and adenocarcinoma (usually papillary are also excluded by this WHO definition. Using these strict criteria, the incidenceof squamous cell carcinoma of the thyroid is less than 1% of all thyroid malignancies (10J1).Squamous cell carcinoma appears to have a predilection toin thyroglosdevelop sal duct remnants, accounting for an estimated 7% of thyroglossal duct tumors (12). Care must also be taken to exclude local extension or metastasis from aorlaryng other to head and neck carcinoma. Squamous cell carcinoma of the lung can also metastas the thyroid gland. Demographics are difficult to determine secondary to the rarity of pure squamous cell thyroid carcinomas. We have reviewed 45 case reports of pure squamous cell carcinoma published in the English literature since ,1970 (10-31). Cases reported as squamous cell carcinoma that do not clearly meet the criteriaby as the WHO defined were excluded. The excluded tumors were generally papillary carcinomas, adenosquamous carcinomas, or anaplastic carcinomas with a squamoid element. Table 1 summarizes the demographic information on pure squamous cell carcinoma. Similar to anaplastic
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
423
carcinoma, the tumors usually presentedin the fifth,sixth, or seventh decades of life, 35 years of age. The female:male but cases have been described in patients as young as ratio is 1.7:l.
Clinical Characteristics While several cases of squamous cell carcinoma have been described to be associa with thyroiditis and squamous cell metaplasia, the etiologyof this rare variant remain obscure. One patient with adenosquamous thyroid carcinoma has been described after radiation therapy (321, but thus far none has been described in several large cohorts of patients with Hodgkin’s disease followed up longitudinally, making a relationship with prior radiation therapy unlikely. The origin of squamous cells within the thyroid remains controversial, and several hypotheses have been proposed. Squamous cell carcinoma has been described within (12,25). Thesetumorswere thyroglossalductremnantsandlingualthyroidglands believedtoderivefromsquamousepithelialcellsinthewallsoftheserestsof cells. A few cases of pure squamous cell carcinoma appear to have developed from this pattern of squamous metaplasia. However, as Klinck and Menk observed (33), squamousmetaplasialeadingtosquamouscellcarcinomahasnotbeenidentified (11) didnotidentifyanyareas in otherorgans.Moreover,Haradaandcolleagues ofsquamousmetaplasia in their series of squamous cell carcinomas. If squamous cellcarcinomacommonlyarosefromsquamousmetaplasia,onewouldexpecta higher occurrence of these cancers in clinical conditions associated with squamous metaplasia, such as thyroiditis, papillary thyroid carcinoma, and adenomatoid nodules, but this has not been reported. (34) recently reported a variant thyroid epithelial cell population Bond and colleagues of thyroglobulin staining, and positive characterized by a squamoid appearance, absence immunostaining for cytokeratin and vimentin. These cells had a higher proliferative capacity than the follicular cellsin primary culture. The authors suggest that these cells may represent areas of squamous metaplasia within the thyroid gland, although they might represent a small population of normal squamoid thyroid cells that grow well in the cell culture environment. In comparison to presquamous cell carcinoma, this rare neoplasm also occurs in association with a well-differentiated carcinoma (particularly papillary carcinoma), an adenoma,amultinodulargoiter, or (occasionally)chronicautoimmunethyroiditis (1,3,4). Because of the intimate relationship to neoplastic glandular elements, some squamous cancers have been called adenosquamous carcinomas. Undifferentiated carcinoma may be evident alongwith the predominant squamous carcinoma(5). Squamous carcinoma with extensive spindled cell change has been reported in association with tall-cell papillary carcinoma (6). Mucin-producing carcinoma has been reported and (1,351.The presenceof mucosubstances has been associated with squamous carcinoma inthyroid is acomplexproblem,anduncertaintyaboutsuchneoplasmscontinues (36,37). Bland focal squamous metaplasia may occur in both follicular and papillary carcinomas, but usually this does not behave as squamous carcinoma. Most patients reported with squamous cell carcinoma of the thyroid present with the rapid growth of a firm mass in a previously existing multinodular goiter. The
424
Ringel, Burman and Shmookler
symptoms generally begin over several weeks to months. The rapid growth is often associated with pain, weight loss, night sweats, and local symptoms such as dysp and dysphonia. Similar to some cases of anaplastic carcinoma, a syndrome of sis and non-PTH-mediated hypercalcemia has been described (13,28,38,39). At least one cell line derived from a patient with squamous cell carcinoma of the thyroid has been characterized to make an interleukin la-like factor and a colony stimulating fac (13). Although many patients develop distant metastases during the course of their illness, the majority of patients present with local neck complaints. The diagnosisof squamous cell carcinoma is usually made on fine needle aspira or at the time of surgery for progressive local symptoms. These le biopsy (FNAB) typically do not concentrate radioiodine and are "cold" on radionuclide scanning. C of metastases from local head and neck tum must be taken to exclude the possibility orlungcarcinoma.Neckandchestcomputedtomographic(CT)scansaswellas bronchoscopy and endoscopy should usually be performed preoperatively to ensure that the tumor originated from the thyroid.
Treatment and Clinical Course Squamous cell carcinoma of the thyroid has a clinical course resembling anaplast carcinoma. Complete surgical resectionis the primary curative therapy in combinatio with postoperative external beam radiation therapy. Radioiodine scanning and are of limited value, since nearly all of these tumors are not iodine-avid.A variety of chemotherapeutic regimens have been attempted to cure individual cases including bleomycin, Adriamycin, and cisplatin, all with disappointing results. Thyroid hormo suppression is usually initiated, although the clinical utility of this treatment has not been well documented. It is likely that these cancers, similar to anaplastic cancers, growth. have sustained sufficient genetic alterations to leadnon-thyrotropin-mediated to Palliative surgery and radiation therapy is appropriate in selected patients to avoid airway compromise and inability to swallow. 16 Nearly all of the patients reported with squamous cell carcinoma died within months of diagnosis. These deaths were caused by distant metastases or local co tions of disease. The rare long-term survivors of these tumors are those who prese with earlier stage disease who had a near-complete of complete surgical resection. After surgery, these patients were generally treated with external beam radiation.
Summary Puresquamouscellcarcinoma is ararethyroidcancerwithanextremelypoor prognosis. Similar to anaplastic carcinoma, these tumors present as rapidly enlarging masses in older persons and are generally not responsive to radioiodine or conventiona chemotherapy. Direct extension or metastases from other squamous cell carcinomas the head, neck, and lungs must be ruled out before choosing a treatment plan. Cure seems possible only in those rare patients who present with surgically resect Postoperative radiation therapy should generally be prescribed in such patient this disease dine andthyroidhormonesuppressiontherapyhavelimitedutilityin Control of local disease is important to preserve the quality of life in patients with squamous cell carcinoma.
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
425
POORLY DIFFERENTIATED (“INSULAR”) CARCINOMA
Demographics Insular thyroid carcinoma was originally described by Langhans in 1907 (40) and named wuchentde Struma (see Chapters 9 and 17). Subsequently, in1984Carcangiu and colleagues (41)renamed this variant poorly differentiated carcinoma (insular carcinoma) this tumor. Insular because of the solid clusters of polygonal cells characteristic of WHO (1) and appears carcinoma is included as a variant of follicular carcinoma by the to be more aggressive than well-differentiated follicular carcinoma, but less aggressive than anaplastic carcinoma. Poorly differentiated carcinoma defines a group of carcinomas of follicular thyroid epithelium that retain sufficient differentiation to produce scattered small follicular structures and some thyroglobulin, but which generally lack the usual morphological (42,431. Instead, the characteristics of papillary carcinoma and follicular carcinoma as solid, insular (islands of cells separated by connechistological patterns are described tive tissue and artifactual spaces), trabecular, and alveolar, with scattered tiny follicles containing colloid within the solid, insular, and trabecular regions. These patterns may mix with one another, and small foci of characteristic follicular and papillary carcinoma also may be found, even both within the same neoplasm. The Bcl-2 protein (a suppre of apoptosis) has been described in a large proportion of these tumors, in contrast to undifferentiated carcinoma (44). In a review of the English language literature from1966 to 1996, we identified 228 (41,4564). The median age at presentation well-documented cases of insular carcinoma for the group is 54years with a range 12ofto 77 years. The incidence of this histological variant appears low; Mizukami and associates (55) reported identifying only 3 cases of insular carcinomaof 800 thyroid tumors resected at their institution during 20 years. Higher rates have been described in other populations; however, many of these groups solid, trabecular, included tumors with other “poorly differentiated” appearancesas such this group of tumors described as and alveolar patterns. It remains unclear whether “poorly differentiated”is a spectrum of one tumor type identified during the dedifferent type represents a specific entity. In all ating process, or whether each histological probability, the former is correct. Clinical Characteristics The prognosis of insular carcinoma has been controversial because of the rather varied histological pattern present and the evidence of well-differentiated papillary or follicular carcinoma that can be found within most of these neoplasms. Although the controvery is understandable, one must account for the presence of the “primordial cells” in these neoplasms, and the fact that they display many characteristics that differ from well-differentiated cancers (65-67). We have compared the presenting characteristics and clinical outcomes of patients (68)of 1355 with insular carcinoma reported in the literature by Mazzaferri and Jhiang patients with well-differentiated thyroid cancer (Table 2). Many of the larger reports are summarized in detail below. Patients with insular carcinoma tended to be older, had larger primary tumors, and were more likely to present with metastases when
426
Shmookler
and
Ringel, Burman
Table 2 Clinical Presentation and Outcomeof Insular Thyroid Cancer Compared With Differentiated Thyroid Carcinoma Presentation Insular Differentiated Characteristic Insular
Age (Yr)36 Size (cm) Intrathyroidal Regional 50% Distant
Dlferentiated"
54
4.7
34% 36% 30%
2.5 48%
2%
Disease Outcome
(%)
(%o)
Dead of disease Alive with disease No disease Died with disease
20
8 29
30
44 6
63 0
Adapted from Burman and Ringel (210) with permission of W.B. Saunders Co. *Data on differentiated cancer reportedby Mazzafem and Jhiang (134).
compared to patients with differentiated carcinoma. Thereis often a preceding history of goiter in adults with insular carcinoma. The male:female ratio is similar to welldifferentiated carcinoma. In the three largest series', the female:male ratio was 2.1: 1. The most common presenting complaint is an enlarging mass, although metastatic presentation in the neck, mediastinum, and femur have been reported. The duratio of these tumors are slow-growing nodules of enlargement appears variable, but most enlarging for months or years before clinical diagnosis. These lesions are not typi FNAl3 is usually iodine- or technetium-avid relative to normal thyroid tissue, and suggestive of malignancy. Children with this entity tend to present with early nodal (48,55,58). Aswithanaplasticand metastases,either in theneckormediastinum squamous cell carcinomas, insular carcinoma frequently develops in the setting multinodular goiter, underscoring the importance of both long-term follow-up and rapid recognition of changes in growth pattern.
Treatment and Clinical Course The clinical course of insular thyroid carcinomainitially was described by Carcang 1984 (41). In their series of25 patients, 11(44%)presented with intrathyroidal diseas 11 (44%)with neck metastases or invasion of local neck structures, and 4 (12%) with metastatic disease. With a mean follow-up time of 3.5 years (range: 1-8 years), 11 (44%)died of disease and7 (28%) were alive with disease. Of the 4 patients presentin with distant metastases, 3 died of disease in follow-up. The authors concluded that insular carcinoma was more aggressive than well-differentiated carcinoma, but less aggressive than anaplastic carcinoma. Ashfaq and colleagues (57) reported 41 patients with insular carcinoma with mea follow-up of 4 years (range: 1 month to 12 years). Clinical outcome information was available on 28 patients; 18% died of disease, 21% were alive with disease at the tim of the study, and 50% were alive with no evidence of disease; 11% died of other 41% of causes, but had known disease. Similar to Carcangiu and colleagues, only patients presented with intrathyroidal lesions. They reported no difference in clinical (1040%)or a predominant histological com outcome between patients with a minor nent (50-90%) of insular carcinoma. The majority of these patients were treated with
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
427
thyroidectomyfollowedbyradioiodine.Papottiandcoworkers (46) addressedthe treatment of thyroid cancers containing a “primordial cell” component by considering (n = 31) two groups of tumors, those comprised predominantly of insular carcinoma and those with predominant trabecular or solid patterns with a minor component of insular carcinoma (n = 32). The only presentation or outcome difference between the two groups was a higher recurrence rate among tumors with a predominant insular histology with a mean follow-upof 4.6 years. After surgery, 46 (72%) of the patients were free of disease, while 16 (28%) had evidence of metastases or invasion of local structures at presentation. All patients other than those with incidentally discovered malignancies at thyroidectomy received near-total thyroidectomy followed byI3lI and subsequent L-thyroxine suppression therapy. Of the 46 patients initially rendered free of disease at surgery, 27 were free of disease at follow-up; 83% of the tumors were iodine-avid and produced measurable serum levels of thyroglobulin, which indicated differentiated follicular cell function.Of the 63 patients, 35 had persistent or recurrent disease, and 30 of the 35 patients with recurrence had evidence of iodine uptake on diagnostic scan and were treated with I3’I.Iodine scan and measurements of serum thyroglobulin showed 17% of those patients to be cured,27% were alive with disease, while 56% died of disease.Five patients with non-iodine-avid recurrence or metastases were treated with chemotherapy and/or external beam radiation therapy, three of whom were alive with disease 1to 15 years after presentation. Similar data were recently reported by Sasaki and colleagues (59), who identified 44 cases of papillary or follicular carcinoma with an insular component. For the time period reviewed, this number represented 1.8% of thyroid carcinomas. Although the 18 patients were free of follow-upwasvariable, 17 patients died of their disease, disease, and2 were alive with disease. Multivariate analysis revealed that the presence of insular carcinoma,in addition to tumor size, the absence of a tumor capsule, vascula all independently associated with a worse invasion, and necrosis within the tumor were I3*Itherapy in patients with prognosis. Many other groups have reported successful metastatic disease. In addition, metastatic insular carcinoma has been detected utilizing ?c (45,54). Serum levels of thyroglobulin are also useful in monitoring patients for tumor recurrence. Table 2 also compares the outcome of patients with insular carcinoma to that of those with well-differentiated carcinoma as reported by h4azzafen-i and Jhiang (68). The follow-up of the patients with insular carcinoma was shorter, and the treatment tumors. Among modalities were variable, compared to the cohort with well differentiated patientspresentingwithmetastaticinsularthyroidcarcinomawithrelativelyshort 60% died of disease, and 20% were alive with disease. follow-up periods (several years), The remaining20% were believed to be cured from their metastatic carcinoma following treatment with I3II. Treatment with L-thyroxine therapyis prudent in patients with insular thyroid car& noma. Since many of these patients have tumors that display differentiated epithelial function, treatment with dosesof L-thyroxine to fully suppress pituitary production of TSH is recommended. We believe the goal TSH level in these patientsis a value less than 0.01 pU/ml, if tolerated andif the patient has no contraindications to such therapy. Of course, this TSH level should be achieved gradually and the dosejust sufficient to suppress the TSH level to this range should be given. The average dose of exogenous
428
Ringel, Burman and Shrnookler
L-thyroxine required to suppress TSH is approximately 2.1 pg/kg. Because the recurtumor are higher than for differentiated carcin rence rates and mortality ratesthisfor and since these carcinomas generally remain iodine-avid, aggressive surgical and therapy are appropriate, followed by frequent monitoring for tumor recurrence with radioiodine scanning and measurements of serum thyroglobulin.We also believe that the adjunctive use of radiological procedures, such as neck sonogram,andor CT, MRI may be very useful in diagnosing recurrent disease as early as possible.
Summay Insular thyroid carcinomais a histological variant with a clinical course more a sive thanwell-differentiatedcarcinomabutlessaggressivethananaplasticthyroid carcinoma. Insular carcinomamay therefore represent an important intermediate stag in the dedifferentiation of thyroid carcinoma. Patients with insular carcinoma are likely to present with metastases and to develop tumor recurrence than those wi differentiated tumors, but the majority present with either intrathyroidal lesions o regional disease. These tumors generally maintain differentiated thyroid follicular cell function, allowing for 13'1 therapy and scanning and measurement of serum levels of thyroglobulin. The majority of patients with local disease are successfully treated surgically and with I3lI therapy followed by L-thyroxine suppression. Tumor recurrence rates appear to be higher among patients with local (intrathyroidal or regional disease) insular carcinoma than well-differentiated tumors. Patients with distant metastases treated l3II and thyroid hormone suppression appear to have a cure rate of approximately 20%, justifying aggressive treatment with surgery, 13'1, and thyroid hormone suppression. TALL-CELL VARIANT OF PAPILLARY CARCINOMA
Demographics The tall-cell variant of papillary carcinoma (TCV) was initially reported by Hazard in 1964 (69), who defined a group of papillary tumors with the following characteristic in at least 30% of the tumor: papillary structures, epithelial cell height at least twice the cellular width, oxyphilic cytoplasm, and hyperchromic basilar nuclei. Hawk and this definition. They also Hazard (70) found that 9% of their papillary tumors met described these tumorsas being grossly larger and more locally invasive than non-ta cell papillary carcinoma, affecting an older group of patients (mean ageof 57 years). 510% of thyroid carcinomas. Most studies report that TCV represents approximately The female predominance typical of other forms of thyroid canceris preserved. Since 1976, we have identified 163 patients with TCV in the literature with follow-up data reported in 148 patients (70-87).
Clinical Characteristics At least 30% of the tumor must be composed of tall cells to be considered a TC tumor (see Chapter 17). However, we are concerned about the inherent subjectivity involved in this diagnosis. For example, no studies have been published in which identical slides were sent, in a blinded fashion, to multiple pathologists to determine how frequently the diagnosisof TCV wouldbe made. TCV must be differentiated f
%
429
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer Table 3 Clinical Presentation of Tall-Cell Variant Comuared With Differentiated Thvroid Cancer Tall-Cell Variant Differentiated Thyroid Carcinoma* (n = 163)
Characteristic ~
Age (years) Size (cm) Intrathyroidal Regional (neck) Distant Female
(n = 1355)
~~
51.8 3.2 33% 19%
36
74%
67%
2.5
50% 2%
Adapted from Burman and Ringel (210) with permission of W.B. Saunders Co. *Data on differentiated cancer reported by Mazzafeni and Jhiang (134).
the less common columnar cell variant characterized by ‘‘tall‘‘ cells, but with nuclear stratification as opposed to basilar nuclei and scant, nonoxyphilic, but often vacuolated, cytoplasm. These two closely related variants of papillary carcinoma were recently described in the same patient, suggesting a similar pathogenesis (87). TCVhasbeenfound in combination with cell types other than usual forms of papillary carcinoma. A recent report describes 5 cases of TCV in tumors with regions of a variant of anaplastic carcinoma described as spindle cell squamous carcinoma (88). These cases, in addition to the general occurrence of TCV within tumors containing usual papillary carcinoma and a high incidence of p53 mutations (89), suggest that this lesion might represent a transition between papillary carcinoma and anaplastic carcinoma, similarto the intermediate stage assigned to insular carcinoma. The similarities between TCV and the columnar cell variant have been described above. These two variants have been classified separately, but may represent similar “transition” tumors. The consistency and reliability in diagnosing TCV also requires further study. The interobserver variability is unknown, but may be quite important when reviewing the frequency and clinical course of TCV. Table 3 compares the presenting characteristics of163 patients with TCV compiled from the literature to the cohort of patients with usual differentiated papillary carcinoma (68).Patients with TCV tend to be older, have larger reported by Mazzafem and Jhiang lesions, and are more likely to have metastases outside of the neck than are patients with well-differentiated tumors. The aggressivenessof TCVhas been debated in the literature since its initial description, particularly in younger patients (see below) (71,77).
Treatment and Clinical Course Johnson and associates(77) attempted to define the prognosis of TCV by comparing 12 patients with TCV to a similar group of age- and sex-matched patients with the usual papillary thyroid carcinomas. The groups were different in extent of disease at presentation. Allof the patients with usual papillary carcinoma had either intrathyroidal (58%) or regional lymphatic spread (42%); 25% of the patients with TCV had intrathyroidal disease, 33% had cervical lymphatic spread and 42% had invasion into cervical
430
and
Burman
RingeZ,
ShmookZer
soft-tissue or distant metastases. Tumor size correlated with recurrence and tumorrelated mortality. Clinical outcome was different between the TCV and usual pap 50 years of age, although the author thyroid carcinoma only among patients older than concluded that TCV was more aggressive than usual papillary thyroid carcinom and colleagues (71)described 19 patients with TCV and compared them to patients with usual papillary carcinoma. The follow-up times were similar for both groups (62 and 93 months, respectively). Patients with TCV had larger tumors than those with papillary carcinoma (4.2 cm vs 2.8 cm). Patients older than 50 years with TCV had larger tumors than younger patients with TCV (5.6 cm vs 2.7 cm) and had a higher incidence of distant metastases and locally invasive disease. They performed a ate analysis on their data comparing TCV to usual papillary carcinoma and found 4.0 cm were associated with an patients older than 50 and tumor size larger than increased riskof tumor recurrence, but not mortality. Similar to Johnson and coll (77) they could find no difference in the prognosis of patients younger than 50 with TCV compared to usual papillary carcinoma. Moreno Egea and colleagues(72) recently reported aseries of 5 patients with TCV aged 58 to 73 years. All of these patients presented with large tumors with extrathy disease. When compared 85 to patients with the usual papillary carcinoma, a statist significant increase in tumor recurrence and mortality were reported. The patients in all of which included these series’ were treated utilizing a variety of protocols, nearly thyroidectomy andI3’Itherapy. Some patients received palliative external beam r therapy to control local recurrence. When compared to the large cohort of Mazzafe and Jhiang (Table 4), the clinical outcomes of patients under the age of 50 were indistinguishable from those with usual forms of papillary carcinoma, but older were more likely to have recurrence of their tumors. The patients with TCV were followed for a shorter period of time, and treatment may have been more variab this reference group. Mortality rates and recurrence rates may rise as the follow-up periodlengthens.ThedatasuggestapoorprognosisforolderpatientswithTCV presenting with larger primary tumors. This tumor variant most commonly occurs in patients older than 50, but the distinction between younger and older patients may b important when determining the prognosis of a younger patients with a small TCV thisatpoint there are no convinci tumor. Although our opinion may be controversial, data to suggest that small primary TCV tumors presenting in patients younger than50 years of age have a worse prognosis than the usual papillary carcinomas of similar size. Surgical therapy followed byl3II treatment and L-thyroxine suppressionis approLoss of these priate. These tumors typically produce thyroglobulin and are iodine-avid. characteristics of differentiation should be taken as further dedifferentiation of
Summay The tall-cell variant of papillary carcinoma typically presents in older patients w a higher frequencyof local adenopathy or invasion and distant metastases. It has been reportedinassociationwithwelldifferentiatedpapillarycarcinomaandanaplastic carcinoma, suggesting it represents a “transition” histology between these two types. These tumors generally maintain thyroglobulin production and iodine-avidity. Amon older patients with thyroid cancer, TCV appears to have a higher recurrence rate an perhaps, a higher mortality than the usual formof papillary thyroid carcinoma.In the
431
432
Shmookler Ringel, Burman and
unusual younger patient who presents with a locally confined small TCV tumor, the prognosis appears similar to usual papillary carcinoma. Patients with TCV, probably regardless of size or invasive features, should be treated aggressively with thyroidectomy, 1311, and long-term L-thyroxine suppression. Follow-up should include periodic 1311 scanning and determinations of serum levels of thyroglobulin. Periodic cervical to help detect evidenceof recurrence sonogram, CT scans, andor MRI also can be used as soon as possible. When there are suggestive changes with these techniques, guided biopsies can be performed to confirm recurrent cancer rather than a benign reactive process. COLUMNAR CELL VARIANT OF PAPILLARY CARCINOMA
Demographics The columnar cell variantof papillary carcinoma is a very rare tumor with only 2 reported cases (87-97) (see Chapter 17). Patients with columnar cell carcinoma had mean age of 45 years with a range of16 to 76 years and a female:male ratio of 1.4:1. As described above, the pathology is similar to TCV in that the height of the cells should be at least three times the width. The nuclei are stratified, not basilar, and the cytoplasm is clear as opposed to pink (87,89-92). These tumors form large, distinct papillae. The follicular cells are immunoreactive for thyroglobulin. The tumors are typically large, with a mean longest dimension of 5.4cm.Cytopathology is most commonly consistent with papillary carcinoma. Columnar cell carcinoma has been described in an otherwise unremarkable well differentiated papillary carcinoma (91), along with anaplastic carcinoma (89), or in combination with tall-cell carcinoma (87). Initial reports of the columnar cell variant ofpapillarycancer suggested a highly aggressive neoplasm. However, Wenig and coworkers (92) recently analyzed 16 cases of this rare tumor. They concluded that of presentation, rather than cell type, was predi extrathyroidal extension at the time of a more aggressive clinical course.
Clinical Characteristics Neoplasms have been reported in which the columnar cell pattern was mixed with tall cell papillary carcinoma (87,91), as well as with solid regions of typical papillary carcinoma (91,96).Also, we have seen extensive insular and trabecular patterns a to the columnar cell pattern. Reports suggest that the locally infiltrative tumors are usually fatal (87,239-91), but those which are encapsulated may be successfully resected (94,95).
Treatment and Clinical Course Most of the patients presented with large masses: metastatic disease was present at presentation in 29%, includingsix to the lungs, one to the adrenal glands, one to b and two to bone. The patients were treated with thyroidectomy and radioactive iodin Of the 24 patients described, 58% were free of disease at follow-up. Recurrences all, 38% (9/24) of the patien typically occurred within2 years of the initial surgery. In died of their columnar cell carcinoma.
Clinical Aspects of MisceZZaneous and
Unusual Thyroid Cancer
433
Treatment should be directed at early diagnosis byFNAB followed by a complete surgical resection when possible. Treatment with 1311is appropriate along with subsequent L-thyroxine suppression. If the primary tumor is large and a complete surgical resection is notpossible,localpalliativecontrolwithexternalbeamtherapyhas been utilized.
Summay Columnar cell carcinoma is a rare variant of papillary thyroid carcinoma that appears tallthe cell variant, consistent with dedifferentiation compared to have a course similar to to well-differentiated papillary thyroid carcinoma. Early metastases and local tumor invasionarecommon;thekeytocurativetherapy is earlydiagnosisfollowedby complete surgical resection.
DIFFUSE SCLEROSING VARIANT OF PAPILLARY CARCINOMA Approximately 63 cases ofthis unusual variant have been described in the literature, most commonly in younger patients with a mean age of 31 years (98-106). Thirteen patients (21%) were younger than 20 years of age. Pathologically, diffuse sclerosing papillarycarcinoma is characterized by pronounced fibrosis, numerous psammoma bodies, and extensive lymphocytic infiltrates (see Chapter 17). Mucin may be present, thus these tumors may be misclassified as mucoepidermoid or anaplastic carcinomas. Although the prognostic significance of this variant is not clear, these tumors most commonly present as diffuse enlarging lobes or entire thyroid glands. The clinical course of these tumors is not yet well-characterized, but it appears that they behave similarly to well differentiated papillary carcinomas (98) and should be treated with surgery, radioiodine, and thyroxine suppression. SOLID VARIANT OF PAPILLARY THYROID CARCINOMA
Untilrecently, this variantwaspoorlydescribedintheliterature;however,the apparent unique association between the solid variant of papillary carcinoma and the Chernobyl nuclear accidentin 1986 has stimulated renewed interest in its pathogenesis and prognosis (107-110). While rare in adults, this histological pattern (see Chapter 17) is frequently identified in tumors in children, but usually comprises a small amount of an otherwise well-differentiated tumor. The prognostic significance of these small regions is unclear but is not reported to impact on survival. The role of radiation exposure and the high prevalence of retlpTC3 gene rearrangethis tumor typeis a unique thyroid ments in solid varianttumors support the notion that of thyroid cancer malignancy. Nikiforov and colleagues(110) compared the histologies in tumors removed from children who were exposed to the Chernobyl accident versus a control group of sporadic thyroid tumors removed from children from the United States. These investigators found an that 37% of the “radiation-induced” tumors had solid variant as the predominant growth pattern compared to 4% of the “sporadic” tumors. About 79% of the solid variant tumors had retPTC3 gene rearrangements. These data support the hypothesis that rearrangements of this gene occurred in response to the radiation exposure and may be involved in the tumorigenesis. Increased p53
434
Ringel, Burman and Shm
immunohistochemical staining has been reported in some of the solid regions o tumors, raising the concern of more aggressive clinical behavior ( I 08). There are no data to confirm the clinical implications of this variant. Lon follow-up studies of these children are planned. The impact of the solid variant hi in adults may differ from the impact in children as well.
MIXED MEDULLARY-FOLLICULAR CELL CARCINOMA These variants of medullary carcinoma of the thyroid (MTC) display the micr features of both MTC and carcinomas of follicular cells ( I ) . Regions of the tum immunoreactive with calcitonin while other regions have thyroglobulin prod Normal follicles may be “trapped” within any MTC and seemingly cause thyrog immunoreactivityin MTC. This pattern differs from the mixed variants that have of follicular cancer or papillary cancer adjacent to regions of MTC (111-117). medullary-papillary cancers have been reported (118-120). The pathogenesis of these tumors is uncertain. Care must be taken to exclude carcinoma of the thyroid, which may have a similar histological appearance t but will express only thyroglobulin. Paragangliomas may also have an app similar to MTC, but these are rare in the thyroid. The appearance of these mixed suggests the presence of a progenitor thyroid cell that differentiates into follic C cell lineages, although most researchers believe these cell types derive from s lineages. Pappoti and associates (11I ) recently reported the identification of ra expressing both thyroglobulin and calcitonin in 2 of 11 cases. However, in instances the medullary-follicular cell carcinoma may represent a “collision tum the thyroid. Clinical recommendations are difficult to formulate for such rare As with most cases of MTC, aggressive surgical therapy is appropriate. Tre with I3lI and L-thyroxine suppression, not performed for usual forms of MT recommended for these rare lesions. Serum calcitonin and thyroglobulin measur are helpful. The prevalence of mutations in the ret gene is not known, although f occurrence of the medullary-follicular and medullary-papillary variants hav reported (115,118).
MUCOEPIDERMOID CARCINOMA Demographics Mucoepidermoid carcinoma is a rare variant of thyroid carcinoma of uncert lineage. A recent report of 6 new cases of mucoepidermoid carcinoma also id 31 cases in the literature (121). They are more common in women than men with the majority of patients presenting with a solitary “cold” nodule. There is n relationship with risk factors such as a prior history of radiation therapy or a history of thyroid cancer. The mean presenting age of patients with mucoepid carcinoma is 42 years (range: 10-71 years) with 4 patients under 20 years These neoplasms are usually of a low-grade of malignancy, and their histoge uncertain. Sometimes the tumor is associated with papillary carcinoma (or even as a metastatic focus in a papillary carcinoma) (121-127). An adjacent undiffer (anaplastic) carcinoma has been reported (124). The tumors are typically soli
Clinical Aspects
of Miscellaneous and Unusual Thyroid Cancer
435
light-colored masses, not encapsulated, sometimes cystic, sometimes with mucus visi on the cut surfaces (see Chapter48). A variety of sizes (1-8 cm) have been reported.
Clinical Presentation and Diagnosis Most of the 31 reported patients presented with a painless neck mass or a slowly growing thyroid nodule (121-127). When obtained, radioiodine scanning revealed a FNA has been reported to be diagnostic photopenic region corresponding to the nodule. of carcinoma, but is not specific for mucoepidermoid carcinoma (125). The diagnosis is usually made or confirmed histologically at the timeof thyroidectomy. Both partial and near-total thyroidectomies have been performed, the former only for small, wellcircumscribed tumors. At presentation, 17/31 (55%) had extrathyroidal disease, however. All but one patient had disease confined to the neck. Patients with cervical node metastases or invasion of local structures were treated with surgery by followed external No reports of postoperative 1311 scanning after the usual preparabeam radiation therapy. tion with thyroid hormone withdrawal have been described. Clinical Course and Treatment Among patients with local adenopathy, complete remission rates are quite high, although the duration of follow-up is variable. Thesetumors appear to be more indolent than many of the other forms of dedifferentiating thyroid carcinoma. Similar to most other forms of thyroid cancer, cure rests on early diagnosis and surgical intervention. (4.0 cm) confined to the thyroid appear to do well Patients with small primary tumors (121). Therapy for those patients with locally advanced disease (utilizing surgery and external beam radiation) appears to induce remission in the majority of patients. Several patients have either presented with or developed distant metastases. Treatment has not been reported; however, the metastatic lesions tend to be indolent, similar to welldifferentiated thyroid carcinoma. The utility of thyroid hormone therapy and radioiodine scanning and treatment have not yet been evaluated in these tumors.
Summay Mucoepidermoid carcinoma of the thyroid is a rare tumor affecting adults. It is commonly coincident with lymphocytic thyroiditis and shares some features with pap lary thyroid carcinoma. It has been identified adjacent to well-differentiated papillary carcinoma and anaplastic carcinoma. The so-called adenosquamous carcinoma could represent an aggressive variant of mucoepidermoid carcinoma.In all, these factors raise this tumor represents a slowly growing variant of papillary carcinoma, the possibility that although this is an unproven hypothesis. Cure seems possible with a complete surgical resection (when feasible) and external 1311 therapy in mucoepiderbeam radiation therapy for residual local disease. The role of moid carcinoma requires further evaluation.
SARCOMAS OF THE THYROID True primary sarcomas of the thyroid gland are exceedingly rare. Some cases of sarcoma have been reclassified as anaplastic carcinoma, and in a few cells display mixed immunohistochemical markers of epithelial and mesenchymal lineages (so-called “carcinosarcomas”). Careful ultrastructural and immunohistochemical analyses have
436
Ringel, Burman and
convincingly described several cases of leiomyosarcoma (128-131), os (132,133),chondrosarcoma (134, fibrosarcomas (131), liposarcomas (134 commonly, angiosarcomas (131,135-137). Angiosarcomas (malignant he theliomas) have been found mostly in European Alpine regions known to deficient. In general, sarcomas present in older patients, often with a long-standin a goiter. Three thyroid sarcomas have been described in patients with a p of external beam radiation therapy, including in a 23-year-old patient, a cases were reported among a large group of previously irradiated patients with disease (134). These tumors resemble sarcomas arising in other locations. Angiosarcom have features of endothelial differentiation with immunoreactivity for factor V antigen, CD34 and CD31. Keratin immunoreactivity has been reported in Because of this characteristic, some authors prefer to consider these neoplasm toid carcinomas (135).There may be little importance in differentiating sar anaplastic carcinomas. Most of the patients presented with large primary tumo local structures and having lymphatic spread. The majority of the patien aggressive local or metastatic disease. Similar to anaplastic carcinoma, possible only with complete surgical resection. Local control with radiat seems advisable if the patient is clinically stable following surgical res utility of chemotherapy for thyroid sarcomas has not been reported. Sarcomas may rarely metastasize to the thyroid gland. A primary organ thyroid should be excluded in any patient presenting with thyroid sarco primary thyroid sarcomas are identical to sarcomas in other organs. Kapos (KS), a well-recognized secondary disease in patients with AIDS, has be to infiltrate the thyroid. One case of infiltration of the thyroid by KS causing h ism has been reported (138).
TERATOMAS OF THE THYROID
The diagnosis of teratoma, whether benign or malignant, requires demo various cells with characteristics of +e three germ-cell layers. Teratomas of are rare, usually occur in childhood, and are most commonly benign. M teratomas are found in infancy and are generally quite large, often larger t Buckley and colleagues (139)identified 139 cases of childhood thyroid terat all of which were benign, usually presenting as a mass causing local c symptoms. Thyroidal origin is inferred by identifying the blood supply as a the thyroidal vessels. Among adults, teratomas are even more unusual than among children, b commonly malignant. Bowker and Whittaker (140)recently reported a case o teratoma in a 17-year-oldpatient and reviewed nine other cases reported in t (139-145). Adults with malignant teratoma of the thyroid had a mean years ( 17-68). There were no specific risk factors associated with maligna Most of them were quite large (up to 17 cm in diameter). Patients were t thyroidectomy, radiation, and chemotherapy. Cervical and/or distant meta reported in all cases. No cases of long-term survival have been reported, a
Clinical Aspects of Miscellaneous and
Unusual Thyroid Cancer
437
to radiation therapy and chemotherapy appears to be transient. Only the patient reported by Bowker and Whitaker was free of disease at follow-up (7 months). The remainder 22 monthsofdiagnosis.Childrenwithraremalignant ofthepatientsdiedwithin teratomas of the thyroid have a similarly poor prognosis.
RARE, MOSTLY BENIGN, TUMORS OF THE THYROID
There are case reports of rare histological types of benign tumors involving the thyroid. We mention a few nonepithelial varieties. Benign leiomyomas (128,146,147) and neurilemomas (146-149) have presented as slowly growing, palpable masses that were“cold”onradioiodine imaging. They were composed of “spindle” cells with abundant eosinophilic cytoplasm but no atypia or evidence of increased mitotic activity. Immunoperoxidase staining confirmed the neural or smooth muscle nature of these tumors. The lack of extrathymidal invasion and the absenceof recurrent disease after 1 to 6 years of follow-up support a benign diagnosis. One case of granulosa cell tumor of the thyroid in a girl treated with relatively high doses of ethinyl estradiol (0.1 mg daily) and medroxyprogesterone (10 mg) for short stature for several years was also recently reported(150). This patient has also done well after surgical resection. Microscopically, this tumor resembled Hiirthle cell adenomas because of the abundant eo philic cytoplasm. HYALINIZING TRABECULAR NEOPLASMS, USUALLY ADENOMAS
These rare neoplasms are solid masses, often less than 3.0 cm in diameter and well circumscribed (usually encapsulated) (151-155). Psammoma bodies may be scattered through the tumor. Most cells contain immunoreactive thyroglobulin and keratin. Calcitonin has never been demonstrated. Colloid is not present, but irregular masses of hyaline material are adjacent to the clusters of cells. Nuclei often contain grooves and cytoplasmic inclusions, which,in addition to the psammoma bodies, are reminscent of papillary thyroid carcinoma. Thus, aspiratesof these lesions have been confused with papillary neoplasms. Rarely, the tumors have been invasive and have involved cervical (156). lymph nodes and has been termed the cribriform variant of papillary carcinoma An alteration resembling this hyalinizing tumor has been described in adenomatoid nodules in multinodular goiter (155) (see Chapter 48).
Thymic and Related Neoplasms Thymic, parathyroid, and salivary gland tissues may be foundin the thyroid (157159), and therefore it is not surprising that occasional neoplasms occurin the thyroid and inferior part of the neck that resemble the thymus (160-163). Such tumors may be benign or malignant. EMBRYOLOGIC THYROID REMNANTS
Thyroglossal Duct Cysts The thyroglossal tract in adultsmay be a vestigial remnant or, may be a more fully developed structure, composed of thyroid follicles, a duct (usually lined by ciliated pseudostratifiedcolumnarepithelium),connectivetissue,andlymphoidtissue.The
438
Ringel, Burman and Shmookler
thyroid follicles in this tract may undergo any of the changes that occur in the gland type (164,165). When thyroid proper, even thyroid cancer, papillary or follicular in cancer occurs at the more proximal portion of this duct it may actually involve the base of the tongue. Generally, the thyroglossal tract resides in the midline, and a thyroglossal tumor may move cephalad when the tongue is protruded because of the continuing connection between the mass and the tongue. When a tumor is present in the thyroglossal tract,it may or may not be associated with a similar tumor within th thyroid gland itself. Thyroglossal tumors must be differentiated from thyroglossal branchial cleft cysts, and cystic hygromas (fluid-filled multiloculated lymphangiomas which are present at birth). Thyroid cancers which arise exclusively in the thyroglossal tract are rare.
Clinical Presentation The proper approach to thyroglossal duct tumors is largely unknown, but our approa is to perform a FNA and base our therapeutic decision to a large extent on these findings (165,166).If the FNAis positive or suspicious for thyroid cancer, we genera recommend removal of the entire thyroglossal tract from the base of the tongue to t thyroid gland. The thyroid gland may be removed at the initial surgery if the FNA is diagnostic of thyroid cancer, and if the FNA is suspicious for thyroid cancerwe may recommend, in conjunction with discussions with the patient, removal of the thyroid gland at the initial operation. Some prefer a subsequent thyroidectomy when th sis of thyroid cancer in the thyroglossal tract has been confirmed. Carcinoma residing this area or, within the thyroglossal tract may emanate from the thyroid epithelium in alternatively, could arise within the thyroid gland and metastasize to the thyroglossal tract (rare). The diagnostic and therapeutic approach to thyroglossal duct tumors is controversial, as long-termcontrolledstudiesassessingvariousoptionshavenot been performed. The natural history of thyroid carcinomas arising within the thyroglossal duct is (167)retrospectively reviewed thyroglossa poorly understood. Heshrnati and coworkers carcinomain 12 patientsseenovera44-yearperiodattheMayoClinic.Ageat presentation ranged from 17 to 60 years with a mean of 40 years. The patients were equally divided between men and women. The most common presenting complaint was a midline neck mass. In all 12 cases, papillary thyroid cancer was found, and 3 patients also had involvement of the thyroid gland. Nine patients had a subtotal or neartotal thyroidectomy and, despite the fact that only three patients received postopera of 13 years, no patient had radioactive iodine therapy, during a mean follow-up period recurrence, distant metastases, or disease-specific mortality. The usual surgical included a Sistrunk procedure in conjunction with a thyroidectomy. Because these of patients were seen over a long time period, especially before our better unders thyroid cancer, we do not necessarily concur with their recommendations that iodine is not necessary and that these patients have an excellent prognosis. Tew and coworkers (168) found 90 thyroglossal duct cysts or nodules over a 30-year period. Four patients had thyroid cancer in the thyroglossal duct cyst, an incidence similar to that of carcinoma arising in an intrathyroidal location. Mahnke and colleagues (169) estimate that approximately 150 cases of thyroid carcinoma arising within a thyrogl tract have been reported. We tend to treat patients having a thyroglossal papillary
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
439
thyroid cancer with a Sistrunk procedure, a total or near-total thyroidectomy, and most frequently we recommend radioiodine therapy, based upon the clinical findings. At present, there is noreasontoexpectthesetumorswillbehavedifferentlythanan soft tissue intrathyroidal papillary cancer. We suspect that size, capsular invasion, invasion and vascular or nodal invasion should be considered in the decision how to treat a patient with a thyroglossal papillary thyroid cancer. In addition to papillary thyroid cancer, squamous cell carcinoma and lymphoma may arise within this tissue (170,171).Deshpande and Bobhate (172)believed that only 9 cases of squamous cell carcinoma arising within a thyroglossal tract have been reported. This tumor is difficult to diagnose, and we suggest that a Sistrunk procedure conjunction with a total or near-total thyroidectomy be performed for these rare tum METASTATIC CANCERS IN THE THYROID GLAND
Metastatic tumors in the thyroid gland occur in as many as 24% of patients when examined at autopsy, although the clinical manifestations of these metastases certainly as a solitary are uncommon. Generally, a metastatic tumor in the thyroid gland presents nodule, which usually is hypofunctioning on radioisotope scans. Involvement of an existing adenomatoid nodule or adenoma is likely, thereby complicating the morphological features. The most common primary sitesof such tumors are breast, kidney, lung, (173,174).Usuallythere is widespreadmetastatic andskin(malignantmelanoma) disease present, and the manifestations in the thyroid gland are clinically unimportant. Nevertheless, solitary thyroid metastasis to the thyroid may glandbe the initial evidence of disease or perhaps the first presentation of recurrent disease. For example, we have seen a patient with acute myelogenous leukemia who had been treated earlier, and the first evidenceof recurrent disease was in the thyroid gland. Nakhjavani and colleagues (174)recently reported a total of 43 patients (23 women and 20 men) with tumors metastatic to the thyroid gland. Solitary thyroid nodules or a multinodular gland was the presentation in 40 patients, while the remaining three had tracheal compression necessitating thyroid surgery. Renal cell carcinoma was found in 14 patients, lung cancer in 7, and breast cancer in 7. More rarely, parathyroid cancer, salivary gland tumors, ovarian or uterine cancer, skin cancer, and esophageal cancer was found. In some instances the source of the tumor was identified essentially concurrently with the thyroid gland metastases, whereas renal cell carcinoma within the thyroid gland was 15% of subjects found as long as 26 years after the original diagnosis of the tumor; had evidence of thyroidal involvement before the diagnosis of metastasis to other sites. Although the investigators suggest that a thyroidectomy was associated with enhanced survival compared to a nonsurgical approach (to the thyroid gland), we are skeptical about this observation and believe that more information is required addresssing this issue. The diagnostic evaluation revolves around the performance of a fine needle asp and examination of the cytological sample. In most instances, is abundant there cellularity and the cells may be typical of the original site, especially when specific immu chemical stains are performed. Obviously, diagnostic evaluation for the site of the of other metastatic sites is important before approachoriginal tumor and for the presence ing the thyroid tumor. Occasionally it may be difficult to determine if the cytological
440
and
Ringel, Burman
Shmookler
specimen is originating from the thyroid gland, such as an anaplastic thyroid carcinoma or the unusual clear variant of follicular carcinoma( I ) , or represents metastatic disease in the thyroid gland. The therapeutic approach depends upon the clinical context and the cytological examination. For example, if there is widespread metastasis from an obvious extrathyroidal site and the thyroid nodule cytology examination is supportive of metastasis from the same site, it may not be appropriateto remove the thyroid gland. On the other hand, if the patient had a renal cell carcinoma treated successfully several years e and nowthe patient presents with a thyroid nodule that cytologically appears to resem renal cell carcinoma, then it might be useful to perform a lobectomy for diagnostic reasons. In fact, if an evaluation confirms thatthis single thyroid nodule might be the only evidence of metastatic disease, some clinicians might want to approach the lesion for diagnostic and even therapeutic reasons. Our general opinion is that it is desirable to obtainas much information as possible from the thyroid cytological speci men,including specific staining, and that radiological evaluation for other sites of metastasis may be useful. Once as much information as possible is obtained, a frank discussion of the prognosis should be made with the patient and their family. In the outlook of patients with metastasis to the thyroid gland is poor, but individual tumors or circumstances may require a different approach to treatment. Tumors may also invade the thyroid gland by local extension. Most commonlythis occurs with laryngeal, pharyngeal, and esophageal tumors, and such invasion may in around the thyroid gland. Cervical lymph node enlarge present as a neck mass or as CT (generally without contrast) or may also be noted. Radiological studies, such MRI, and direct visualization of the larynx or esophagus may be useful. On thyroid isotope scanning, such invasion may present as a hypofunctioning area.
m T H L E CELL CARCINOMA
Hiirthle cells (Askanazy cells, oxyphilic cells, oncocytes) are follicular cells that have many mitochondria (which are the basis for the abundant, eosinophilic, granular 28). They are cytoplasm), frequently eccentric nuclei and visible nucleoli (see Chapter altered follicular cells.For example, Hiirthle cells bind TSH and have TSH receptors, as do other types of follicular thyroid cells (175,176).The basic genetic or environme factors that allow a thyrocyte to differentiate into a Hiirthle cell are unknown. Inde Hilrthle cells may occurin a variety of thyroid disorders. Solitary thyroid nodules ma have a predominance of Hiirthle cells, to the exclusion of more typical thyrocytes; in such cases, the amountof colloid is limited and there are only scattered macrophage When such a lesion is aspirated, the interpretation would be suggestive of a Hiirthle cell neoplasm, although, as with follicular tumors, a definite diagnosis would be difficult or impossible. However, solitary thyroid nodulesmay also have a more varied picture in which Hiirthle cells are intermingled with thyrocytes, macrophages, and lymp with moderate amounts of colloid. In such circumstances the fine needle aspiration might be more difficult to interpret and may not have sufficient Hiirthle cells to be characterized as a Hiirthle cell neoplasm. Furthermore, Hiirthle cells may be found in in circumthe thyroid glandsof patients with Hashimoto's thyroiditis, but usually this stance the Hiirthle cells are scattered and are not the predominant celltype (176).
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
441
These neoplasms are composed mostlyor completely of these distinctive cells, and most seem to be follicular carcinomas. Recognizing the malignant potential of a tumor 77-183). Bizarre, depends upon the evidence of aggressive behavior its at periphery (1 large, andor hyperchromatic nuclei may be a striking histological feature, but these are more common in the benign proliferations of oxyphilic cells. Metastases to cervical lymph nodes are more common than with the usual follicular carcinoma, especially after the patient has undergone surgery for the cancer. Some studies suggest that oxyphilic follicular carcinomas are more aggressive than the usual nonoxyphilic follicular carcinomas. The presence of nondiploid cells in an oxyphilic (184). Papillary carcinoma indicates a poorer prognosis than for one with diploid nuclei carcinomas may also be composed of these distinctive cells. Whether they are more aggressive than a nonoxyphilic cancer with otherwise similar characteristics remains uncertain.
Clinical Presentation Hiirthle cell carcinomasare thought to represent about3-5% of all types of thyroid carcinomas. Our view is that most Hiirthle cell carcinomas are a slightly more aggress variety of follicular carcinoma, with more frequent recurrences, higher morbidity, and mortality, but this is controversial (177,178). The tumors are frequently multifocal and bilateral. They are generally less avid for radioactive iodine and, therefore, respond less frequently than the usual follicular carcinoma. Several groups are currently eval in Htirthle cell ing the thyroid gland sodium iodide symporter function and expression carcinomas relative to follicular thyroid cancer (175-178). Thompson and associates (178-180) suggested that it is difficult to differentiate of their studies is that benign from malignant Hiirthle cell tumors. The implication even experienced pathologists may not be able to make a reliable distinction. Carcan and coworkers (181) and Grant and associates (182) support the concept that strict histological criteria, with adequate sampling, may be able to differentiate Hiirthle cell carcinoma from adenoma, in nearly all cases. Grant and associates(182) reviewed the world literature and observed that only 6 of 642 patients with Htirthle cell adenomas 1%.Gosain and Clark(183) were found to have a recurrence, an incidence of less than also found no patients with Hiirthle cell adenoma in whom recurrences were observed. Finally, Bondeson and coworkers (183) studied 42 patients diagnosed with Hiirthle cell adenoma over a 2 to 20 year period and found no recurrences. These articles, in our view, support the contention that Htirthle cell adenomas can be accurately diagnosed by experienced pathologists. The major histological criteria that separate a H W e cell adenoma from a Hiirthle cell carcinomaare vascular and/ or capsular invasion. Subtleties do remain, however. For example, does the capsular invasion have to be completely through the capsule is invasion or into, but not through, the capsule sufficient to make the diagnosis? In sum, we believe that experienced pathologists can reliably make this differential diagnosis. Because these tumors concentrate radioiodine less well than usual follicular carcinothis therapy. Soh and Clark (183) suggest that it is mas, they respond less well to reasonable to approach Hiirthle cell carcinomas as if they were medullary carcinomas, m ea that is, with more aggressive diagnostic procedures and treatment. PatientsHwith cell neoplasm diagnosed byFNA should undergo surgery promptly. We recommend a
442
Ringel, Burman and Shmookler
near-total thyroidectomy by an experienced thyroid surgeon.It is important to discuss with the patient the alternative approaches of anear-totalthyroidectomyversusa lobectomy with isthmusectomy. If only a lobectomy and isthmusectomy are perform and if the lesionis found tobe carcinoma, then a subsequent completion thyroidectom must be performed. This completion thyroidectomy frequently causes mental and psy if the requirementfor a completion thyroidec chosocial distress to the patient, especially 20% of Htirthle cell neoplasms tomy is unexpected by the patient. However, only about diagnosed by fine needle aspiration will be found to be malignant. If a total thyroidectomy is performed initially then, of course, 80% of the time this procedure would be unnecessarily aggressive and exposes the patient to a higher risk of temporary and permanent hypocalcemia and recurrent laryngeal nerve paralysis. The decision as to which operation to perform in the setting of a patient with a solitary thyroid nodule is difficult. A frozen section interpreand anFNA consistent with Htirthle cell neoplasm tation is often problematic and therefore not helpful(185). We think it is important to candidly discuss the advantages and disadvantages of each approach with the patient and the family and arrive at a mutual decision. The initial operation should include ipsilateral central node dissection. Obviously, the surgeon must be allowed to exerc judgment at the time of surgery regarding the precise operative procedures. Soh and Clark ( I 75,186)also suggest that a routine modified radical neck dissection be when tumor is found in the central compartment or cervical nodes. McDonald and coworkers(187) reviewed 40 cases of Htirthle cell carcinoma noting that this represented 4% of all thyroid cancers in their experience. Follow-up after thyroidectomy had a median of 8.5 years. Vascular or capsular invasion was observe in 32 patients, extrathyroidal invasion in 11, and regional lymph node involvement in 2. One patient had distant metastases at presentation. Only9 patients received I3lI. Of 34 subjects analyzed, 5 died of the thyroid cancer, 9 died of nonthyroidal causes, 4 were alive with disease, and 16 were alive without evidence of disease. At about a median of 4 years, 9 patients had recurrences and 5 had distant disease. Recurrent disease was associated with mortality in half of these patients. The assessed risk fa at presentation could help predict recurrence. Low-risk tumors did not recur (e.g., tumor less than 5 cm diameter, lack of distant metastases, age less than 41 years for men and less than 51 years for women). Although it is difficult to compare individual reports with mortality and morbidity rates from the literature, it appears that this study support the view that Htirthle cell carcinoma is associated with a poorer prognosis than the usual follicular or papillary thyroid cancer. Following appropriate surgery for Htirthle cell carcinoma, we recommend ra scanning and therapy. The preparation for the scan would be routine and is usually performed about 6 weeks after surgery. We think a radioiodineisscan important before therapy, as this scan helps determine the avidity of the remaining thyroid cells for radioiodine and helps to define the nature and extent of remaining disease. We then follow the scan with radioiodine therapy, usually with 100-150mCi l3*I.Wealso routinely perform a scan about 7-10 days after therapy. It is reported that only about 10% of H W e cell cancers trap and respond to radioiodine.This number seems low in our experience, and, of course, depends to some degree on the dose of I3*Iused for scanning, the time the patient did not receive thyroid hormone, the extent of TSH
Clinical Aspects
of Miscellaneous and
Unusual Thyroid Cancer
443
elevation, and, perhaps, on the assiduous adherence to a low-iodine diet. In published reports, it may be difficult to assess these factors adequately, and perhaps in some cases lackof apparent iodine-avidityby the tumor may not be an accurate representation of the tumor's properties. Over the subsequent 5 years we recommend following the patient with physical examinations and thyroid function tests and thyroglobulin levels about every 3-6 months for the first several years, and perhaps every4-6 months for the next several years if there has been no evidence of disease recurrence. The desired TSH in most patients would be pU/ml 0.1 or lower, depending upon the clinical context. Thyroglobulin levels are analyzed at the same time as thyroid function tests, and the 2 ng/ml thyroglobulin level during L-thyroxine suppression must be less than (depending 13*1scanin 1 yearandthen3-5 ontheassay).Wesuggestarepeatwholebody years later. Given the aggressive nature of this tumor, we also may obtain occasional radiographs of the chest. An imaging study of the neck, such as an MRI, is desirable, especially if the tumor is not iodine-avid, if the thyroglobulin levelis increasing or if palpable cervical abnormalities become manifest. In women, especially those who are postmenopausal, suppressive L-thyroxine thera should be accompanied by measures to prevent osteoporosis (daily oral ingestion of calcium 1-1.5 g, 400 units of vitamin D, and exercise against gravity) (188).
REFERENCES
1. HedingerC,WilliamsED,SobinLH.Histologicaltyping of thyroidtumors.Berlin: Springer-Verlag, 1988. 2. Hedinger C, Williams ED, Sobin LH. The WHO histological classification of thyroid tumors: a commentary on the second edition. Cancer 1989; 63:908-911. 3. Mazzaferri EL. Undifferentiated thyroid carcinomas and unusual thyroid malignancies.In MazzaferriEL, Samaan NA, editors.Endocrinetumors.Boston:BlackwellScientific Publications 1993; 378-398. 4. Samaan NA, Ordoflez NG. Uncommon types of thyroid cancer. Endocrinol Metab Clin North Am 1990; 19~637-648. 5. HiSlting T, Mbller P, Tschahargane C, Meybier H, Buhr H, Herfarth C. Immunohistoche cal reclassificationof anaplastic carcinoma reveals small and giant cell lymphoma. World J Surg 1990; 14~291-295. 6. Schmid KW, Krbll M, Hofsttider F, Ladurmer D. Small cell carcinomaof the thyroid A reclassification of cases originally diagonsed as small cell carcinomas of the thyroid. Path01 Res Pract 1986; 181:540-543. 7. Wolf BC, Sheahan K, DeCoste D, VariakojisD, Alpern HD, Haselow RE. Immunohistotumors of the thyroidgland an eastern cooperative oncology chemical analysis of small cell group study. Hum Path01 1992; 23:1252-1261. 8. LiVolsiVA,Merino MU. Squamouscellsinthethyroidgland. Am JSurgPath01 1978; 2~133-140. Kasai N, Yagawa K. Squamous differentiation in thyroid carcinoma 9. Katoh R, Sakamoto A, with special reference to histogenesis of squamous cell carcinoma of the thyroid. Acta Pathol Jpn 1980; 39:306-312. 10. Simpson WJ, Carmthers J. Squamous cell carcinoma of the thyroid gland. Am J Surg 1988; 156:44-46. 11. Harada T, Shimaoka K, Katagiri M, Shimizu M, Hosoda Y,It0 K. Rarity of squamous cell carcinoma of the thyroid: autopsy review. World J Surg 1994; 18:542-546.
444
Ringel, Burman and Shmookler
12. White
E,Talbert W M . Squamous cell carcinoma arising in thyroglossal duct remnant cyst epithelium. Otolaryngol Head Neck Surg1982; 9025-31. 13. Saito K, Kmtomi Y , Yamamoto K, et al.primary squamous cell carcinoma ofthe thyroid associated with marked leukocytosis and hypercalcemia. Cancer 1981; 48:2080-2083. 14. Prakash A, Kukreti SC, Sharma MP. primary squamous cell carcinoma of the thyroid. Intern Surg 1968; 50:538-541. 15. Bahuleyan CK, Ramachandran P. Primary squamous cell carcinoma of the thyroid. Indian J Cancer 1972; 9:89-91. 16. Kapoor VK, Sharma D, MukhopadhyayAK, Chattopadhyay TK. Primary squamous cell carcinoma of the thyroid gland a case report. Jpn J Surg 1985; 1560-62. Y, Motohara T. Pure squamous cell carcinoma 17. Misonou J, Aizawa M, Kanda M, Uekita of the thyroid gland-report of an autopsy case and review of the literature. Jpn J Surg
1988; 18:469474. R, Katayama S. Squamous cell carcinom 18. Tsuchiya A,Suzuki S, Nomizu T, Yamaki Y, Abe of the thyroid a report of three cases. Jpn J Surg 1990; 20:341-345. 19. Kampsen EB, Jager N, Max MH. Squamous cell carcinoma of the thyroid a report of two cases. J Surg Oncol 1977; 9567-578. 20. Harada T, Katagiri M, Tsukayama C, Higashi Y, Shimaoka K. Squamous cell carcinom with cyst of the thyroid. J Surg Oncol 1989; 42136-143. 21. Sarda AK, BalS, Arunbh, SinghMK, Kapur MM. Squamous cell carcinoma of the thyroid. J Surg Oncol 1988; 39:175-178. 22. Theander C, LiidCnB, Berglund J, Seidal T. Primary squamous cell carcinoma of the thyroid: a case report. J Laryngol Otol1993; 107:1155-1158. 23. Chaudhary RK,Barnes EL, Myers EN. Squamous cell carcinoma arising in Hashimoto’s thyroiditis. Head Neck 1994; 16582-585. 24.
Budd DC, Fink DL, Rashti M Y , Woo TH. Squamous cell carcinoma of the thyroid. J Med SOCNJ 1982; 79~838-840. 25. Huang T-Y, Assor D. Primary squamouscell carcinomaof the thyroid gland: areprt of four cases. Am J Clin Path01 1970; 5593-98. 26. Bukachevsky RP, Casler JD, Oliver J, Conley J. Squamous cell carcinoma and lingual thyroid. Ear Nose Throat J 1991; 70505-507. 27. Korovin GS, Kuriloff DV, Cho HT, Sob01 SM. Squamous cell carcinomaof the thyroid a diagnostic dilemma.Ann Otol Rhino1 Laryngol 1989; 9859-65. 28. Riddle PE,Dinesoy HP. primary squamous cell carcinomaof the thyroid associated with leukocytosis and hypercalcemia, Arch Pathol Lab Med 1987; 111:373-374. 29. Shimaoka K, Tsukada Y.Squamouscellcarcinomasandadenosquamouscarcinomas originating from the thyroid gland. Cancer1980; 46:1833-1842. 30. Renard TH, Choucair RJ, Stevenson WD, Brooks WC, Poulos E. Carcinoma of the thyroglossal duct. Gynecol Obstet 1990; 171:3005-3008. 31. Bakri K, Shimaoka K, Rao U, Tsukada Y.Adenosquamous carcinoma of the thyroid afte 1983; 52:465-470. radiotherapy for Hodgkin’s disease: a case report and review. Cancer 32. Bakri K, Shimaoka K, RaoU, Tsukada Y. Adenosquamous carcinoma ofthe thyroid after 1983; 52:465-470. radiotherapy for Hodgkin’s disease: a case report and review. Cancer 33. Klinck GH, Menk KF. Squamous cells in the thyroid gland. Mil Surg1951; 109:406. 34. Bond JA, Wyllie FS, Ivan M, Dawson T, Wynford-Thomas D. A variant epithelial subvitro.inMol Cell Endocrino population in normal thyroid with high proliferative capacity 1993; 93~175-183. 35. Carcangiu ML, Steeper T, Zampi G, Rosai J. Anaplastic thyroid carcinoma: a study of 70 cases. A m J Clin Pathol 1985; 83:135-158. 36. Hadar T, Mor C, ShveroJ, Levy R, Segal K. Anaplastic carcinomaof the thyroid. Eur J Surg Oncol 1993;19:511-516.
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
445
37. Tan RK, Finley RK, Driscoll D, Bakamjian V, Hicks WL, Shedd DP. Anaplastic carcinoma of the thyroid: a 24 year experience. Head Neck 1995; 17:41-48. 38. Saito K, Fujii Y, On0 M, Nomura H, Shizume K. Production of interleukin 1 alpha-like factor and colony-stimulating factor by a squamous cell carcinoma of the thyroid (T3M-5) derived from a patient with hypercalcemia and leukocytosis. Cancer Res 1987; 4754746480. 39. Okabe T, Nomora H, Oshawa N. Establishment and characterization of a human colonystimulating factor-producing cell line from a squamous cell carcinoma of the thyroid gland. J Natl Cancer Inst 1982; 69:1235-1243. 40. Langhans T. m e r die Epithelialen formen der malignen Struma. Virchows Arch Pathol Anat 1907; 385:125-141. 41. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (“insular”) thyroid carcinoma. Am J Surg Path01 1984; 8:655-668. H. Poorly differentiated carcinoma of thyroid a clinicopath42. Sakamoto A, Kasai N, Sugano ologic entity for a high-risk group of papillary and follicular carcinomas Cancer 1983; 52:1849-1855. 43. Hwang TS, Suh JS, Kim YI, et al. Poorly differentiated carcinoma of the thyroid: retrospective clinical and morphologic evaluation. J KoreanMed Sci 1990; 5:47-52. 44. Pilotti S, Collini P, Del Bo R, Cattoretti G, Pierotti MA, Rilke F. A novel panel of antibodies that segregates immunocytochemically poorly differentiated carcinoma from undifferentiated carcinomaof the thyroid gland. Am J Surg Path01 1994; 18:1054-1064. 45. Yen T-C, King K-L, Yang A-H, Liu R-S, Yeh S-H. Comparative radionuclide imaging of metastatic insular carcinoma of the thyroid: value of technetium-99m-(V)DMSA. J Nucl Med 1996; 37:78-80. 46. Papotti M, Botto MicaF, Favero A, Palestini N, Bussolati G. Poorly differentiated thyroid carcinomas with primordial cell component: a group of aggressive lesions sharing insular, trabecular, and solid patterns. Am J Surg Path01 1993; 17:291-301. 47. KilleenRh4, Barnes L, Watson CG, WL, Marsh Chase DW, Schuller DE. Poorly differentiated (“insular”) thyroid carcinoma, Arch Otolaryngol Head Neck Surg 1990; 116: 1082-108 JE, Robinson RA, Walker WP, Gurll NJ, Hawes DR. Insular carcinoma: 48. Justin EP, Seabold a distinct thyroid carcinoma associated with iodine-131 localization. J Nucl Med 1991; 32:1358-1363. 49. Sakamoto A, Kasai N, Sugano H. Poorly differentiated carcinoma of the thyroid: a clinicopathologic entity for a high risk group of papillary and follicular carcinomas. Cancer 1983; 52~1849-1855. 50. Ljungberg 0, Bondeson L, Bondeson A-G. Differentiated thyroid carcinoma, intermediate type: a new tumor entity with features of follicular and parafollicular cell carcinoma.Hum Pathol 1984; 15:218-228. 51. Pietribiasi F, Sapino A, Papotti M, Bussolati G. Cytologic features of poorly differentiated “insular” carcinoma of the thyroid, as revealed by fine-needle aspiration biopsy. Am J Clin Path01 1990; 94:687-692. 52. Flynn SD, Foreman BH, Stewart EF, KinderBK. Poorly differentiated (“insular”) carcinomaof the thyroid gland: an aggressive subset of differentiated thyroid neoplasms. Surgery 1988; 104:963-970. 53. Chandrasekhar B, Padhy AK, Panda S , Kumqr L, Basu AK. “Insular” carcinoma of the thyroid: a subset of anaplastic thyroid malignancy with a less aggressive clinical course. Clin Nutr Med 1993; 18:1056-1058. 54. Zk lT, Seabold JE, Gurll NJ. Tc-99m MIBI scintigraphic detection of metastatic insular thyroid carcinoma. Clin Nucl Med 1995; 20:31-36. T, et al. Poorly differentiated (“insular”) carcinoma 55. Mizukami Y, Nonomura A, Michigishi of the thyroid. Path Intern 1995; 45:663-668.
446
Ringel, Burman and Shmookler
56. Dominguez-Malagon H, Guerrero-Medrano J, Suster S. Ectopic poorly differentiated (insu lar) carcinomaof the thyroid: reportof a case presentingas an anterior mediastinal mass. Am J Clin Path01 1995; 104:408-412. 57. Ashfaq R, Vuitch F, Delgado R, Albores-Saavedra J. Papillary and follicular carcinomas with and insular component. Cancer 1994; 73:416-423. 58. Kotiloglu E, GUlsev K, Senocak ME. Follicular thyroid carcinoma with a predominant insular component in a child: a case report. Tumori 1995; 81:296-298. 59. Sasaki A, Daa T, Kashima K, Yokoyama S, Nakayama I, Noguchi S. Insular component as a risk factor of thyroid carcinoma. Path Intern 1996; 46:939-946. 60. Pereira EM, Maeda SA, Alves F, Schmitt FC. Poorly differentiated carcinoma (insular carcinoma) of the thyroid diagnosed by fine needle aspiration (FNA). Cytopathology 1996; 7~61-65. GS. Insular (poorly differentiated) carcinoma of the thyroid an ultrastruc61. Begin LR, Allaire tural and immunohistochemical study of two cases. J Submicrosc Cytol Path01 1996; 28~121-131. 62. Paik SS, Kim WS, Hong EK, Park MH, Lee JD. Poorly differentiated ("insular") carcinom of the thyroid gland: two cases report. J Korean Med Sci 1997; 12:70-74. 63. Russo D, Tumino S, Vigneri P,et al. Detection of an activating mutation of the thyrotropin receptor in a caseof an autonomously hyperfunctioning thyroid insular carcinoma. J Cl Endocrinol Metab 1997; 82:735-738. 6 4 . Hassoun AAK, Hay ID, Goellner J R , Zimmerman D.Insular thyroid carcinoma in adole cents: a potentially lethal endocrine malignancy. Cancer 1997; 79: 1044-1048. 65. Papotti M, Torchio B, Grassi L, Favero A, Bussolati G. Poorly differentiated oxyphilic (Hurthle cell) carcinomasof the thyroid. Am J Surg Pathol 1996; 20:686-694. 66.Sobrinho-SimoesM.Poorlydifferentiatedcarcinomasofthethyroid.EndocrPathol 1996; 7~99-102. 67. Sironi M, Collini P, Cantaboni A. A fine needle aspiration cytology of insular thyroid carcinoma. A report of four cases. Acta Cytol 1992; 36:435-439. 68. Mazzafeni EL, Jhiang SM. Long term impactof initial surgical and medical therapy on papillary and follicular thyroid cancer.Am J Med 1994; 97:418-428. 69. Hazard JR. Neoplasia. Acad Path01 Monogr 1964,5:239-255. 70. Hawk WA, Hazard JB. The many appearances of papillary carcinoma of the thyroid. Cleve Clin Q 1976; 45:207-216. 71. Terry JH,St. John SA, Karkowski FJ,al.etTall cell papillary thyroid carcinoma: inciden and prognosis. Am J Surg 1994; 168:459-461. 72. Egea AM, Gonzales J M R , Perez JS, CogollosTS, Parico PP. Prognostic valueof the tall cell variety of papillary cancer of the thyroid. Eur J Surg Oncol 1993; 19:517-521. 73. Harach HR, Zusman SB. Cytopathology of thetall cell variantof thyroid papillary carcinoma. Acta Cytol 1992; 36:895-899. 74. Kaw YT. Fine needle aspiration cytology of the tall cell variant of papillary carcinoma of the thyroid. Acta Cytol 1994; 38:282-283. 75. Hicks MJ, Batsakis JG. Tall cell carcinoma of the thyroid gland.Ann Otol Rhino1 Laryngol 1993; 102:402-403. 76. Robbins J, Merino MJ, Boice JD, et al. Thyroid cancer: a lethal endocrine neoplasm. Ann Int Med 1991; 115~133-147. 77. Johnson TL, Lloyd RV, ThompsonN W , Beierwaltes WH, Sisson JC. Prognostic implications of the tall cell variant of papillary thyroid cancer. Am J Surg Pathol 1988; 12:22-27. 78. Ain KB. Papillary thyroid carcinoma: etiology, assessment, and therapy. Endocrinol Clin North Am 1995; X711-760. 79. OstrowskiML, Merino MJ. Tall cell variant of papillary thyroid carcinoma: a reassessm type of papillary carcinoma and immunohistochemical study with comparison to the usual of the thyroid. Am J Surg Path01 1996; 20:964-974.
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
447
80. Ozaki 0, It0 K, Mimura T, Sugino K, Hosoda Y. Papillary carcinoma of the thyroid tall cell variant with extensive lymphocyte infiltration. Am J Surg Path01 1996; 20:695-698. 81. Segal K, Fridental R, Lubin E, ShveroJ, Sulkes J, Feinmesser R. Papillary carcinoma of the thyroid. Otolaryngol Head Neck Surg 1995; 113:356-363. 82. Rtiter A, Nishiyama R, Lennquist S. Tall-cell variant of papillary thyroid cancer: disregarded entity? World J Surg 1997; 21:15-21. 83. Rtiter A, Dreifus J, Jones M, Nishiyama R, Lennquist S. Overexpression of p53 in tall cell variants of papillary thyroid carcinoma. Surgery 1996; 120:1046-1050. 84.vandenBreckel MWM, Hekkenberg RJ, AsaSL,TomlinsonG,Rosen IB, Freeman JL. Prognostic features on tall cell papillary carcinoma and insular thyroid carcinoma. Laryngoscope 1997; 107:254-259. 85. Gamboa-Dominquez A, Candanedo-GonzalezF, Uribe-Urib NO, Angeles-Angeles A. A tall cell variant of papillary thyroid carcinoma: a cytohistologic correlation. Acta Cytol 1997; 41~672-676. 86. Bocklage T, DiTomasso JP, Ramzy I, Ostrowski ML. Tall cell variantof papillary thyroid carcinoma: cytologic features and differential diagnostic considerations. Diagn Cytopathol 1997;17:25-29. JE. Thyroid carcinoma with mixed tall-cell and columnar-cell features 87. Asklen LA, Verhaug Am J Clin Pathol 1990; 94442445. 88. Bronner M P , LiVolsi VA. Spindle cell squamous carcinoma of the thyroid an unusual anaplastic tumor associated withtall cell papillary cancer. Mod Path01 1991; 4:637-643. 89. Evans HJ. Columnar-cell carcinoma of thethyroid a reportof two cases of an aggressive variant of thyroid carcinoma. Am J Clin Pathol 1986; 8577-80. 90.Sobrinho-SimoesM,Nesland JM, Johannessen JV. Columnar-cellcarcinoma:another variant of poorly differentiated carcinoma of the thyroid. Am J Clin Path01 1988; 89: 264-267. 91. Mizukami Y, Nonomura A, Michigishi T, Noguchi M, NakamuraS, Hashimoto T. Columnar cell carcinoma of the thyroid gland a case report and reviewof the literature. Hum Pathol 1994; 251098-1101. 92. Wenig BM, Thompson LDR, Adair CF, Shmookler B, Heffess CS. Thyroid papillary carcinoma, columnar cell type: a clinicopathologic study of 16 cases. [Abstract]. Roc Meet US Can Acad Path01 1995. 93. Hwang TS,Suh JS, Kim YI et al. Poorly differentiated carcinoma of the thyroid. retrospective clinical and morphologic evaluation. J Korean Med Sci 1990; 547-52. 94. Ferreiro JA, Hay ID, Lloyd RV. Columnar cell carcinoma of the thyroid report of three additional cases. Hum Path01 1996; 27:1156-1160. 95. Evans HL. Encapsulated columnar cell neoplasms of the thyroid a report of four cases suggesting a favorable prognosis. Am J Surg Path01 1996; 20:1205-1211. 96. Gaertner EM, Davidson M, Wenig BM. The columnar cell variant of thyroid papillary carcinoma: case report and discussion of an unusually aggressive thyroid papillary carcinoma. Am J Surg Path01 1995; 19:940-947. 97. Hui P-K, Chan JKC, Cheung PSY, Gwi E. Columnar cell carcinoma of the thyroid fine needle aspiration findings in a case. Acta Cytol 1990; 34:355-358. 98. Caplan RH, Wester S, Kisken WA. Diffuse sclerosing variant of papillary thyroid carcinoma: case report and review of the literature. Endocr Pract 1997; 3:287-292. 99. Chan JKC, Tsui MS, Tse CH. Diffuse sclerosing variant of papillary thyroid carcinoma of the thyroid: a histological and immunohistochemical of study three cases. Histopathology 1987; 11:191-201. 100. Soares J, Limbert E, Sobrino-Simoes M. Diffuse sclerosing variant of papillary thyroid carcinoma: a clinicopathologic studyof 10 cases. Path01 Res Pract 1989; 185:200-206. 101. Wu PS-C, Lesie PJ, McLaren KM, Toft AD. Diffuse sclerosing papillary carcinoma of the thyroid: a wolf in sheep’s clothing. Clin Endocrinol 1989; 31535-540.
448
and
Burman
Ringel,
Shmookler
102. Carcangiu ML, Bianchi S. Diffuse sclerosing variant of papillary thyroid carcinoma: a clinicopathologic studyof 15 cases. Am J Clin Path01 1989; 13:1041-1049. 103. FugimotoY, ObaraT, It0Y, Kodama T, AibaM, Yamaguchi K. Diffuse sclerosing varian of papillary carcinoma of the thyroid clinical importance, surgical treatment, and follow up study. Cancer 1990; 66:2306-2312. 104. Hayashi Y, Sasao T, Takeichi N, Kuma K, Katayama S. Diffuse sclerosing variant of papillary carcinoma of the thyroid:a histopathologic studyof four cases. Acta Pathol Jpn 1990; 40~193-198. 105. Mizukami Y, Nonomura A, Michigishi T, et al. Diffuse sclerosing variant of papillary carcinoma of the thyroid a report of three cases. Acta Path01 Jpn 1990; 40476-482. JM, Sola Perez J, Soria T, Parilla Paricio P. Cli 106. Moreno Egea A, Rodriguez Gonzales pathological study of the diffuse variety of papillary cancerof the thyroid presentation of 4 new cases and reviewof the literature. Eur J Surg Oncol 1994;20:7-11. in thyroid 107. Fermanchuk AW, Averkin JI, Egloff B, et al. Pathomorphological findings cancers of children from the Republicof Belarus: a study of 86 cases occurring between 1986 (post-Chernobyl) and 1991. Histopathology 1992; 21:401-408. 108. Nikoforov YE, Nikiforov MN, Gnepp DR, Fagin JA. Prevalence of mutations of ras and p53 in benign and malignant thyroid tumors from children exposed to radiation after Chernobyl nuclear accident. Oncogene 1996; 13:687-693. 109. KlugbauerS, Lengfelder E, Demidchki EP, Rabes HM. High prevalence ofRET rearrangement in thyroid tumors of children from Belarus after the Chernobyl reactor accident. Oncogene 1995; 11:2459-2467. 110. NikiforovYE, Rowland JM, Bove KE, Monforte-MunozH, Fagin JA. Distinct patternof ret of radiation-induced thyroid papillary oncogene rearrangements in morphological variants carcinomas in children. Cancer Res 1997; 55:5617-5620. 111. Papotti M,NegroF,Carney JA, Bussolati G, Lloyd RV. Mixed medullary-follicular carcinoma of the thyroid a morphological, immunohistochemical and in situ hybridiza analysis of 11 cases. Virchows Arch Path01 Anat 1997; 430:397-405. 112. Hales M, Rosenau W, Okelund MD, Galante M. Carcinoma of the thyroid with a mixed medullary and follicular pattern. Cancer 1982; 501352-1359. 113. Kashima K, Yokoyama S, Inoue S, Dao T, Kaduma M, Nakayuma I, Noguuni S. Mixed medullary and follicular carcinoma of the thyroid: report two of cases with and immunoh tochemical study. Acta Path01 Jpn 1993; 43:428-433. 114. Ljundberg0, Ericsson U-B, Bondeson L, Thorell J. A compound follicular parafollicula cell carcinoma of the thyroid: a new tumor entity. Cancer 1983; 52:1053-1061. 115. Mizdcami Y, Michigishi T, Nonomura A, Nakamura S, Noguchi M, Hashimoto T, Itoh N. Mixed medullary-follicular carcinoma of the thyroid occurring in the familial form. Histopathology 1993; 22:284-287. N, Fujii A, Sakata S, Tokimitsu N. Simultaneous occurrence 116. TanakaT, Yoshimi N, Kanai medullary and follicular carcinoma in the same thyroid lobe. Hum Patholl989; 20:83-86. 117. Tanda F, Massarelli G, Mingioni V, Bonsicu L, Moroni R, Cossu A. Mixed follicularparafollicular carcinoma of the thyroid a light, electron microscopic and histoimmunol study. Surg Pathol 1990; 3:65-74. 118. Lamberg BA, Reissel P, StenmanS, Koiuunicmi A, Ekbolm M, MukinenJ, Franssila K. Concurrent medullary and papillary thyroid carcinoma in the same thyroid lobe and in siblings. Acta Med Scand 1981; 209:421-424. 119. Lax SF, Beham A, Kronberg-Schonecker D, Lansteger W, DenkH. Coexistenceof papillary and medullary thyroid carcinoma of the thyroid gland: mixed or collision tumor?: cl pathologic analysis of three cases. Virchows Arch Path01 Anat 1994; 424:441-447. SL. Concurrent medullary and papillary carcinomas of thyroid 120. PastoleroGC,Coire CI, Asa Am J SurgPath01 1996;20:245-250. with lymph node metastasis: a collision phenomenon. 121. Wenig BM, Adair CF, Heffess CS. Primary mucoepidermoid carcinoma of the thyroid
Clinical Aspects of Miscellaneous and Unusual
Trlyroid Cancer
449
gland a report ofsix cases and a review of the literatureof a follicular epithelial-derived tumor. Hum Path01 1995; 26: 1099-1 108. 122. matigan RM, R q u e JL, Bucher RL. Mucoepidermoid carcinoma of the thyroid gland. Cancer 1997; 39:210-214. 123. Samba& C, Franssila K, Basilio-de-Oliveira CA, Sobrinho-SimoesM. Mucoepidermoid carcinoma of the thyroid revisited. Surg Path011990; 3:271-280. 124. Cameselle-Teijeiro J, Febles-perez C, Sobrinho-SimoesM.Papillary and mucoepidermoid carcinoma of the thyroid with anaplastic transformation: a case report with histologic and immunohistochemical findings that support a provocative histogenetic hypothesis. Path01 Res Pract 1995;191:1214-1221. 125. Franssila KO, Harach HR, Wasenius V-M. Mucoepidermoid carcinoma of the thyroid. Histopathology 1984;8:847-860. 126. Chan JKC, Albores-Saavedra J, Battifora H, Carcangiu ML, Rosai J. Sclerosing mucoepidermoid thyroid carcinoma with eosinophilia: adistinctive low-grade malignancy ar from the metaplastic follicles of Hashimoto’s thyroiditis. Am J Surg Path01 1991;15: 438-448. 127. Katoh R, Sugai On0 T, S, TakayamaK, TomichiN, Kurihara H, Takamatsu M. Mucoepidermoid carcinomaof the thyroid gland. Cancer 1990; 652020-2027. 128. Thompson LDR, Wenig BM, Adair C F , Shmookler BM, Heffess CS. Primary smooth muscle tumors of the thyroid gland. Cancer1997; 79579-587. 129. Kawahara E, Nakanishi I, Terahata S, Ikegaki S. Leiomyosarcoma of the thyroid gland a comparative studyof five casesof anaplastic carcinoma. Cancer1988; a case report with 62~2558-2563. 130. Chetty R, Clark SP, Dowling JP. Leiomyosarcoma of the thyroid immunohistochemical and ultrastructural study. Pathology1993; 25:203-205. 131. Iida Y, Katoh R, Yoshioka M, Oyama T, Kawaoi A. Primary Leiomyosarcoma of the thyroid gland. Acta Path01 Jpn1993; 43:71-75. 132. Lindahl F. Sarcoma of the thyroid gland twenty-two cases in Denmark1943-1968. Dan Med Bull 1976; 23~103-119. 133. Syrjiinen KJ. An osteogenic sarcomaof the thyroid gland (reportof a case and survey of the literature). Neoplasma 1979; 26:623-628. 134. Griem KL, Robb PK, Caldarelli DD, Templeton AC. Radiation-induced sarcoma of the thyroid. Arch Otolaryngol Head Neck Surg 1989; 115991-993. 135. Mills SE, Gaffey MJ, Watts JC,et al. Angiomatoid carcinoma and “angiosarcoma” of the thyroid gland a spectrum of endothelial differentiation. Am J Clin Path01 1994;102: 322-330. 136. Lamovec J, Zidar A, Zidanik B. Epithelioid angiosarcoma of the thyroid gland: reportof two cases. Arch Path01 Lab Med 1994; 118542-646. 137. Chan YF,Ma L, Boey JH, Yeung HY.Angiosarcomaof the thyroid an immunohistochemical and ultrastructural studyof a case in a Chinese patient. Cancer 1986; 57:2381-2388. 138. Mollison LC, MijchA, McBride G, Dwyer B. Hypothyroidism due to destructionof the thyroid by Kaposi’s sarcoma. Rev Inf Dis 1991; 13:826-827. 139. Buckley NH, Burch W, bight GS. Malignant teratoma in the thyroid glandanofadult: a case report and a review of the literature. Surgery 1986; 100:932-937. 140. Bowker CM, Whittaker RS. Malignant teratoma of the thyroid: case report and literature review of thyroid teratoma in adults. Histopathology1992; 21:81-83. 141. Stone HH, Henderson W D , Guidio FA. Teratomas of the neck. Am J Dis child 1967; 113~222-224. 142. Buckwalter JA, Layton JM. Malignant teratoma in the thyroid gland of an adult. Ann Surg 1954;139:218-223. 143. Kinder SC, Muth WF. Primary malignant teratomaofthe thyroid: case report and literature review of cervical teratomas in adults. Cancer1978; 42311-317.
450
Ringel, Burman and Shmookler
144. Hadju SI, Hadju EO. Malignant teratoma of the neck. Arch Pathol Lab Med 1967; 83: 567-570. 145. OHiggins N, Taylor S. Malignant teratoma in the adult thyroid gland. Br J Clin Pract 1975; 291237-238. 146. Andrion A, Bellis D, DelsedimeL, Bussolati G, MazzuccoG. Leiomyoma and neurilem-
oma: report of two unusual non-epithelial tumoursof the thyroid gland. Virchows Arch Pathol Anat 1988; 413:367-372. 147. Hendrick JW. Leiomyoma of the thyroid gland, Surgery 1957; 42597-599. Ann Surg 1964; 1601014-1016. 148. Delaney WE, Fry KE. Neurilemoma of the thyroid gland. 149. Goldstein J, Tovi F, Sidi J. Primary schwannomaof thethyroidgland.InternSurg 1982; 67:433-434. 150. Mahoney CP, Patterson SD, Ryan J. Granular
cell tumor of the thyroid gland in a girl receiving high-dose estrogen therapy. Ped Pathol Lab Med 1995; 15:791-795. 151. Carney JA, Ryan J, Goellner JR. Hyalinizing trabecular adenoma of the thyroid gland. Am J Surg Path01 1987; 11:583-591. 152. Sambade C, Franssila K, Cameselle-Teijero, NeslandJ, Sobrino-Simoes M. Hyalinizing trabecular adenoma: a misnomer for a peculiar tumorof the thyroid gland. Endocr Path01
1991; 2~83-91. 153. Goelher JR, Carney JA. Cytologic features of fine-needle aspirates’ of the hylainizing trabecular adenomaof the thyroid. Am J Clin Path011989; 91:115-119. 154. Bondeson L, Bondeson A-G. Clue helping to distinguish hyalinizing trabecular adenoma from carcinoma of the thyroid in fine-needle aspirates. Diagn Cytopathol 1994; 10:25-29. 155. Chan JKC,Tse CCCH, Chiu HS. Hyalinizing trabecular adenoma-like lesion in multin lar goitre. Histopathology 1990;16:611-614. 156. Chan JKC, Loo KT. Cribiform variant of papillary thyroid carcinoma. Arch Pathol Lab Med 1990; 114:622-624. 157. Russell Wo, Ibanez ML, Clark RL, White EC. Thyroid carcinoma: classification, intraglan dular dissemination, and clinicopathological study based upon whole organ sections of 80 glands. Cancer 1983;16:1425-1460. 158. LiVolsi VA. Branchial and thymic remnants in the thyroid gland and cervical region: anexplanationforunusualtumorsandmicroscopiccuriosities.EndocrPathol 1993; 4~115-119. 159. Mizukami Y, Nonomura A, Michigishi T, Noguchi N, NakamuraS . Ectopic thymic tissue in the thyroid gland. Endocr Path01 1993; 4:162-164. 160. Harach HR, Saravia DayE, Franssila KO. Thyroid spindle-cell tumor with mucous cysts an intrathyroid thymoma?Am J Surg Path01 1985; 9525-530. or related branchial pouch 161. ChanJKC,RosaiJ.Tumorsoftheneckshowingthymic differentiation: a unifying concept.Hum Pathol 1991; 22:349-367. 162. Mizukami Y,Kurumaya H, Yamasa T, Minato H, Nonomura A, Noguchi M, Matsubara
F. Thymiccarcinomainvolvingthethyroid 1995; 26576-579.
gland report of two cases.HumPathol
163. Shek TWH, Luk ISC, Ng IOL,Lo CY. Lymphoepithelioma-like carcinomaof the thyroid gland lack of evidence of associationwithEpstein-Barrvirus.HumPathol 1996; 27:815-853. 164. Pollock WF, Stevenson EO. Cystsandsinuses ofthethyroglossalduct. A m JSurg 1996; 112225-232. 165. Jaques DA, Chambers RG, Oertel JE. Thyroglossal tract carcinoma. Am J Surg 1970; 120:439-446. in median ectopic thyroid (includ 166. LiVolsi VA,Perzin KH, Savetsky L. Carcinoma arising ing thyroglossal duct tissue). Cancer1974; 34:1303-1315. 167. Heshmati H M , Fatourechi V, Van Heerden JA, Hay ID, Goellner JR. Thyroglossal duct carcinoma: report of 12 cases. Mayo Clin Proc 1997; 72:315-319.
Clinical Aspects of Miscellaneous and Unusual Thyroid Cancer
451
168. Tew S, Reeve TS, Poole AG, DelbridgeL. Papillary thyroid carcinoma arising in thyroglossal duct cysts: incidence and management.NZ J Surg 1995; 65717-718. U, WernerJA, Rudert H.primary papillary carcinoma of the thyroglossal 169. Mahnke CG, Janig duct: case report and reviewof the literature. Auris, Nasus, Larynx 1994; 21:258-263. 170. Kwan W B , Liu FF, Banerjee D, Rotstein LE, Tsang RW. Concurrent papillary and squaJ Surg 1996; 39:328-332. mous carcinomain a thyroglossal duct cyst: a case report. Can 171. Udoji WC. Thyroglossal duct cyst mass with Hashimoto’s thyroiditis and non-Hodgkin’s lymphoma. J Tenn Med Assoc 1996; 89:113-114. in thyroglossal duct cyst. J Laryngol 172. Deshpande A, Bobhate SK.Squamous cell carcinoma Otol 1995;109:1001-1004. 173. Czech JM,Lichtor TR, Carney JA, Van Heerden JA. Neoplasms metastaticto the thyroid gland. Surg Gynecol Obstet 1982; 155503. 174. Nakhjavani MK, GharibH, Goellner JR, van HeerdenJA. Metastasis to the thyroid gland a report of 43 cases. Cancer 1997; 79:574-578. 175. Young Soh E, Clark OH. Surgical considerations and approach to thyroid cancer. Endocrino1 Metab Clin North A m 1996; 25:115-139. 176. Cooper DS, Schneyer CR. Follicular and Htirthle cell carcinoma of the thyroid. Endocrin Metab Clin North Am 1990; 19:577-591. 177. Azadian A, Rosen IB, Walfish PG, Asa SI. Management considerations in HUrtle cell carcinoma. Surg 1995; 118:711-715. 178. Gundry SR, Burney RE, Thompson N W , Lloyd K. Total thyroidectomy for Htirthle cell neoplasms of the thyroid. Arch Surg 1983; 118:529-533. 179. Thompson N W , Dunn EL, Batsakis JG, Nishiyama RH. Htirthle cell lesions of the thyroid gland. Surg Gynecol Obstet 1974; 139:555-560. 180. McLeod MK, Thompson NW. HUrthle cell neoplasms of the thyroid. Otolaryngol Clin North Am 1990; 23:441-452. 181. Carcangiu ML, Bianchi S, Savino D, Voynick IM, Rosai J. Follicular Htirthle celltumors of the thyroid gland. Cancer 1991; 68:1944-1953. 182. Grant CS, Barr D, Goellner J R , Hay ID. Benign Htirthle cell tumors of the thyroid a diagnosis to be trusted? World J Surg 1988; 12:488-495. 183. Gosain AK, Clark OH. HUrthle cell neoplasms. Arch Surg 1984, 118:529-532. 184. Bondeson L,Bondeson A, Ljungberg 0, Tibblin S. Oxyphil tumorsof the thyroid: followup of 42 surgical cases. Ann Surg 1981; 194:677-680. 185. Chen H, Nicol TL, Udelsman R. Follicular lesions of the thyroid does frozen section evaluation alter operative management.Ann Surg 1995; 222101-106. 186. Clark OH,Hoelting T. Management of patients with differentiated thyroid cancer who have positive serum thyroglobulin levels and negative radioiodine scans. Thyroid1994; 4~501-505. 187. McDonald M P , Sanders LE, SilvermanML, Chan HS, BuyskeJ. Htirthle cell carcinoma of the thyroid gland: prognostic factors and results of surgical treatment. Surgery 1996; 120:1000-1004. 188. Burman KD. How serious are the risks of thyroid hormone over-replacement? Thyroid Today 1995;18: Oct-Dec., pp. 1-6.
Jx Future Directions
50 Thyroid Cancer DNA Ploidy, Tumor M a r k s , and Cancer-Causing Genes Michael McDermott At the time of conception, the human organism is a single-cell zygote. During the this cell expands into a complex mass of approxicourse of development into an adult, mately 100 trillion cells having an enormous variety of shapes, sizes, and functions. Normal tissue growth and development require prolific cell division, exquisitely regulated cell differentiation, and appropriately timed cell death or apoptosis. Neoplastic transformation of tissue generally occurs when abnormal regulatory mechanisms promote excessive cell division, impaired cell differentiation, and/or failure of apoptosis. In most tumors types, this aberrant control originates at the genetic level. Intensive study of these regulatory mechanisms has led to significant progress in to understand the basic our ability to diagnose, to predict biological behavior and molecular pathophysiology of thyroid neoplasms. Inthis chapter we explore the major advances in the areas of DNA ploidy analysis, tumor marker measurement, and the exciting developments in the studyof thyroid oncogenes and tumor suppressor genes. Since our discussion will rely on an understanding of some basic concepts, we will first briefly review the essential elements of the cell cycle and of gene function. CELL CYCLE
The life cycle of a cell can be viewed as consisting of two alternating stages: 1). Interphase, the longer stage, is composedofthree interphaseandmitosis(Fig. substages or phases: Gap 1 (Gl), DNA Synthesis ( S ) and Gap 2 (G2). During G1, cells useDNA asa template to transcribe messenger RNA (mRNA) and then to trans mRNA into proteins. In theS phase, DNA is replicated, resulting in a doubling of the cellular DNA content. During G2, DNA repair corrects any mutations that occurred during S phase as the cell makes final preparations for entry into mitosis ( I ) . Mitosis is a much shorter stage, during which cell division occurs; it consists of four sequential phases. In prophase, sister chromatids pair by attaching at their centromeres, the nuclear membrane disappears, and cytoplasmic spindle fibers begin to form. During metaphase, the chromosomes condense and line up along the equator of the cell attached to the spindles. Anaphase is characterized by separation of sister chromatids
F m : Thyroid Cancer: A Comprehensive Guide to Clinical Management Edifed by:L. Wartofiky 0 Humana Press lnc., Totowa, NJ
455
456
McDermott
n
Fig. 1. The cell cycle. The cell cycle consists of two recurring stages: interphase and mitos The first substage or phaseof interphase is G1, the period during which most gene expression takes place. Following this is the S phase, when DNA replicates resulting in doubling of the number of chromosomes. During G2, DNA repair occurs and the cell prepares to divide. The form two cell thenentersintothemitosis,orcelldivision,stagewhenthecellsplitsto daughter cells.
and migration of individual chromatids along the spindles to opposite ends of the cell. In telophase, two new nuclear membranes form, and cytoplasmic division (cytokines set23 chromooccurs. The end resultis two daughter cells, each having a complete of some pairs ( I ) . Meiosis is a more complicated typeof cell division that occurs only in germ cells. This process consists of two consecutive cell divisions; the second division, however is not preceded byDNA replication. In contrast to mitosis, meiosis produces 4 daughter cells, each having only 23 single chromosomes ( I ) .
A WORD ABOUT GENES
The basic blueprintsof life are contained within our genes.A gene is a segment of DNA that carries the information necessary for a cell to produce a specific protein. There are approximately50,000 to 100,000genes in every human cell.To be successful a gene must perform a number of essential functions such as expression, replication, and repair. Gene expression (Fig. 2) takes place predominantly during the G1 period of interphase. Genes havetwo general regions, termed the regulatory and the structural (c regions. Nuclear proteins,known as transcription factors, bind to the regulatory regi and govern the rate at which the structural region is transcribed into mRNA. mRNA molecules then travel to the cytoplasm where theyare translated into the proteins that are characteristicof that particular cell phenotype. These proteins carry out the activities of the cell ( I ) .
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
G ene
Structural Region
Transcription Factor
Regulatory Region
457
P
t
1
RNA Pol II
Heterogeneous
RNA
1
mRNA
1
Protein Fig. 2. Gene expression. Genes have both regulatory and structural (coding) regions. Gene expression begins when nuclear transcription factors bind to regulatory regions of target genes. an enzyme, RNA polymerase II (RNA Pol 11), These proteins modulate the rate at which transcribes the structural region into heterogeneous RNA. Introns are then removed, leaving which exits the nucleus and attaches to ribosomes where the mature messenger RNA (-A) nucleotide sequenceis translated into proteins.
Gene replication (Fig. 3) occurs during the S phase. At this time enzymes known as helicases unwind the DNA double helix, leaving two single strands of unpaired DNA. Using these as templates, DNA polymerases then promote the assembly offree nucleotide bases intotwo new complementary chains that bind to each of these single strands, resulting in the formation of two new identical DNA double helices ( I ) . Gene repair, which takes place during G2, utilizes a complex set of enzymes collectively referred to as the DNA repair system. The human genome, contained within each cell, consistsof approximately 3 billion nucleotide pairs that are replicated with each cell division. It is estimated that spontaneous mutations occur at a rate of about 2 per million base pairs during each S phase; thus as many as 6,000 mutations may appear every time a cell divides. The proteins of the DNA repair system rapidly and efficiently scan along the chromosomes to detect and repair or all ofmost these mutations before the cell proceeds into mitosis ( I ) . DNA PLOIDY ANALYSIS Most cells spend the majority of their lives in the G1 phase expressing their specific proteins. During this time, all human somatic cells have 23 chromosome pairs, or 46 total chromosomes, and are referred to as being diploid. Beginning in S phase and extending throughout mitosis, cells have 23 chromosome quadruplets,92or total chrotetraploid. Since diploid and tetraploid quantities of DNA mosomes, and are said to be euploid or normal. By are found in normal cells, these states are considered to be when they contain anything other than a euploid contrast, cells are said to aneuploid be amount of DNA. Because malignant cells undergo abnormal, excessive cell division,
458
McDemzott Parent Gene
w
PP Daughter Genes Fig. 3. Gene replication. During theS phase, enzymes knownas helicases unwind the DNA double helix, separating it into two single strands. DNA polymerases then utilize the singlestrand templates to assemble free nucleotide bases into complementary strands that anneal to these templates, producingtwo identical molecules of double-stranded DNA that subsequently reacquire their helical structures.
tumors are often found to harbor an abnormally high proportionof tetraploid cells or to have a population of cells that is frankly aneuploid. The quantification of DNA in tissue samples, known as DNA ploidy analysis, has been investigated extensively in the evaluation of thyroid tumors. Two basic technique have been developed. The first was slide cytophotometry in which DNA content was assessed by microscopically examining individual cells on slides of cytological specimens. More recently, flow cytometryhas been employed to measure theDNA present in nuclear suspensions of tissue homogenates. These techniques may be utilized with (FNA) either pathological specimens or material obtained by fine needle aspiration biopsy. The two main applications of DNA ploidy analysis have been to aid in the differential diagnosis of thyroid neoplasms and to assist in predicting the biological behavior, and hence the prognosis, of individual thyroid carcinomas (2,3). The use of DNA analysis in the diagnosis of thyroid tumors has, in general, been disappointing. In the major published series, the percentages of thyroid malignancies that have exhibited aneuploidy have varied from 6% to 60% for papillary carcinoma (4-10), from 5% to 93% for follicular carcinoma (5-17), from 25% to 100%for (10,lPanaplastic carcinoma(7-10,18) and from23% to 43% for medullary carcinoma 21). In the same studies, aneuploidy rates ranged from 0%to 36% in follicular adenoma and from 4% to 18% in other benign lesions. These degrees of overlap are clearly too great for current methods of ploidy analysis to be useful in establishing an accurate, specific diagnosis. Despite these limitations as a diagnostic tool, DNA ploidy analysis has proved to be helpful as an adjunct in assessing the prognosis of thyroid malignancies (3). Most published studies have reported that aneuploidy in thyroid carcinomas correlates with
DNA PZoidy, Tumor Markers,
Cancer-Causing and
Genes
459
patient age and the degree of tumor dedifferentiation (5,9,22), with more aggressive clinical behavior (5,8,11) and with decreased survival (4,7,10,17,23-30). The reported premature death rates for patients with differentiated thyroid carcinomas exhibiting aneuploidy have varied from 26% to 100% and have approached 100%for aneuploid undifferentiated carcinomas (4,10,28,29). A strong correlation with mortality has also been foundin Hiirthle cell carcinomas, where tumor related deathoccurs almost exclusively in patients with aneuploidy; the reported mortality rate for patients with aneup Hurthle cell tumorsis approximately 50% (31-35). Similar results have been reported formedullarycarcinoma (7,10,19-21,23,24,26,36-38), includingafourfoldhigher 10-year mortality rate for patients with aneuploid tumors compared to those with eu tumors (20). The potential clinical utility of DNA ploidy analysis thus relates to its positive correlation with tumor behavior and with clinical outcome. Accordingly, it may be feasible to use DNA ploidy analysis to predict those tumors that are most likely to recur, invade, or metastasize. Decisions regarding clinical management issues such as the extent of surgery, the subsequent administration of radioactive iodine, the aggressi ness of levothyroxine suppression therapy and the frequency of future surveillance scans might then be based on a tumor’s predicted clinical behavior. However, wellcontrolled prospective studies need to be performed to validate such an approach befo decision-making algorithms of this sort can be recommendedfor clinical use.
TUMOR MARKERS A tumor marker is any measurable entity that can be used to assess some aspect of or clinical behavior (39). Like DNA ploidy abnormal tumor biology such as tumor type analysis, marker detection has been evaluated as both a diagnostic and a prognostic tool. A list and brief description of some of the more commonly measured thyroid tumor markers is presented in Table 1. Most tumor markers are proteins that are qualitatively abnormal or quantitatively excessive. Thus, commonly employed methods for marker analysisare immunohistochemistry and immunocytochemistry in which labeled antibodies are used to detect, localize and quantitate specific protein markers in tissue or cytological specimens. Alternatively, the mRNA from which a protein was translated may be isolated and analyzed using the Northern blot technique. Another approach is to convert mRNA into complementary DNA (cDNA) by exposure to the enzyme reverse transcriptase (RT). The cDNA can then be amplified by the polymerase chain reaction (PCR) and examined for mutations with procedures such as single-strand conformation polymorbe phism (SSCP) analysis and direct DNA sequencing. Finally, the gene itself can isolated, selectively amplified by PCR, and then subjected to SSCP andor a direct sequencing procedure. Papillary thyroid carcinomas usually exhibit sufficiently distinctive features that the preoperative FNA cytological diagnosis and postoperative histological diagnosis ordinarily do not pose significant problems for the experienced pathologist. Nonethe specific tumor markers may prove to be useful in atypical cases. The RetPTC protein, forexample,appearstobeseenexclusively in papillarycarcinoma;although this oncogene product is expressed in only about 2040% of papillary carcinomas, it has
460
McDemzott
Table 1 Tumor Markers That Have Been Investigatedin Thyroid Carcinomas Marker Amyloid BRST- 1 Bcl-2 CA19-9 CA50 Calcitonin CEA CGRP Chromogranin A Cytokeratin E-cadherin EGF EGF-R EMA HSP-90 IGF-1 Leu 7 Leu-M1 MYC NSE PCNA Ras RET Synaptophysin Telomerase Thyroglobulin
TGF-cc TGF-p TPO
TR-P
TSH-R Vimentin
Description
Breakdown product of parafollicular C-cell proteins Glycoprotein presentin glandular tissues Protein inhibitorof cellular apoptosis Sialylated protein determinantof Lewis blood group Sialylated protein determinantof Lewis blood group Protein productof parafollicular Ccells Carcinoembryonic antigen Calcitonin gene-related peptide Soluble protein componentof secretory granules Cytoskeletal filamentous proteinin epithelial cells Cell-cell adhesion molecule Epidermal growth factor Epidermal growth factor receptor Epithelial membrane antigen Heat shock protein 90 Insulinlike growth factor1 Antigenic marker on killer lymphocytes Antigenic marker on monocyte-macrophages Nuclear transcription factor Neuron specific enolase Proliferating cell nuclear antigen Signaling protein for tyrosine kinase coupled receptor Tyrosine kinase coupled receptor Membrane protein in presynaptic veiicles Enzyme that restores telomeric ends of chromosomes primary synthetic productof thyroid follicular cells Transforming growth factor alpha Transforming growth factor beta Thyroid peroxidase enzyme Thyroid hormone receptor beta Thyrotropin (TSH) receptor Mesenchymal cell protein
consistently been undetectable in other malignant and benign thyroid tumor types (40,41). Similarly, the diagnosis of follicular carcinoma in surgical specimens is relatively disstraightforward. Difficultyis frequently encountered, however, in the preoperative tinction between follicular adenomas and carcinomas by FNA cytology. The measurement of tumor markers could prove to be particularly helpful in this regard. Markers of tissue differentiation such as the thyrotropin receptor (TSH-R), thyroid peroxidase (TPO),thyroglobulin (TG),thyroid hormone receptor beta (TR-p), heat shock protein 90 (HSP 90) and E-cadherin are generally expressed in follicular adenomas (42-48). Carcinomas, by contrast, frequently overexpress markers of dedifferentiation such as
DNA Ploidy, Tumor Markers, Cancer-Causing and
Genes
461
proliferating cell nuclear antigen (PCNA), CA50, CA19-9, Leu-7, Myc proteins, Ras @MA), ceruloplasmin, and lactoferproteins, p53 protein, epithelial membrane antigen rin (9,16,35,49-56). These trends have indeed been documented in numerous clinicopathological studies but the data has been somewhat inconsistent among the different investigating groups. The most promising results to date have been with PCNA, CA50 and EMA, which have been reported as positive in 73%, 60%, and 75%, respectively, of follicular carcinomas and in 6%, 4%, and 0%, respectively, of follicular adenomas (9,16,50). TPO, conversely, is present in virtually all adenomas (43-45) and has been detected in as few as 3% of carcinomas in one study (43), but in a considerably larger number in other reports(44,45). Measurement of Myc, Ras, insulinlike growth factor1 (IGF-1), transforming growth factor alpha (TGF-a) and TGF betamay also be useful for distinguishing follicular neoplasms from their Hiirthle cell variants (57). Additional of markers can be recomresearch will be required before any single marker or panel mended for general clinical use. Anaplastic or undifferentiated thyroid carcinomas are thought to arise, in many cases, from preexisting differentiated carcinomas. Although the histological distinction between frankly anaplastic tumors and their well-differentiated precursors is not usually difficult, detecting tumors in transition may be facilitated by finding reduced expression (45,5863) and the appearance of markers of dedifferentiation, of differentiation markers in particular Ras (64), Myc (64) and p53 (65-72). Another diagnostic dilemma stems from the fact that anaplastic thyroid tumors often closely resemble thyroid lymphomas and medullary carcinomas of the thyroid. As an adjunct to careful histological study, this is perhaps the area where tumor markers have thus far shown their greatest di utility. Anaplastic thyroid carcinomas very frequently express epithelial andor mesen(60,61,73-80), vimentin (74-78) and EMA chymalmarkerssuchascytokeratin (60,61,76,77,81), and some may belie their thyroid follicular origin by staining positive for thyroglobulin(6063,73,75-77,82). Lymphomas, by contrast, bear leukocyte markers like common leukocyte antigen (78,79,81,82) and surface immunoglobulins (80). Medullary carcinomas of the thyroid also express cytokeratin and vimentin but nearly always stain negative for thyroglobulin and positive for C-cell or neuroendocrine cellspecific markers such as calcitonin (83-88), calcitonin gene-related peptide (CGRP) (88), synaptophysin (89-91), chromogranin A (88), neuron-specificenolase (NSE) (87-89,92), and carcinoembryonic antigen (CEA) (87). Tumor markers, like DNA ploidy, may have greater value as prognostic than as diagnostic tools. Thyroid cancers that are positive for differentiation related markers like TSH-R (45,46,58), TPO (45,46), thyroglobulin (45,46,5963), E-cadherin (48), and S-100 protein (93,94) are associated with significantly lower rates of distant metas ses and tumor-related deaths than are tumors lacking these markers. By contrast, tumors that are positive for proliferation and dedifferentiation-related markers, such as c-Myc (45,46,58,95), p21 Ras (96,97), PCNA (98), epidermal growth factor (EGF) (99,100), Leu-M1 (101), hexokinase (102), p53 (65-72), and an increased mitotic rate (103), tend to have a more aggressive clinical course and a much higher tumor specific mortality rate than marker negative tumors. Leu-M1 positivity, for instance, has been reported to be associated with a 17-fold increase in papillary cancer related mortality (101). The p53 protein is often undetectable or expressed at very low levels in welldifferentiated thyroid carcinomas but becomes increasingly overexpressed in poorly
462
McDmott
Cell Cell Division Differentiation Death +
111,
Cell
DNA Repair
Fig. 4. The three transitional stagesin the life of acell. Stem cells initially proliferate rapidly, greatly expanding their cell numbers. Later they differentiate into mature cells that perform the functions characteristic of their phenotype. Eventually they grow senescent and undergo programmed cell death or apoptosis. DNA that is damaged during cell division is repaired by the DNA repair system and returns to the cycle or, if the damageis irreparable, thecell soon dies.
differentiated and anaplastic carcinomas (65-72). Abnormal expression ofthis marker, either as an initial finding or as a later development, usually portends a significantly worse prognosis. For medullary carcinomas, signs of increased tumor aggressiveness and greater risk of mortality include negative staining for calcitonin, amyloid, CGRP, NSE, chromogranin A, CEA, cytokeratin, vimentin and Bcl-2, and positive staining for Leu-M1, N-Myc and BRST-1 (19,37,38,104-107). Clearly the application of tumor marker methodology to the clinical management of thyroid carcinomas should be a fertile ground for future research. The measurem oftumor-specific markers should eventually be of significant value in the routine preoperative and postoperative diagnosis of distinct tumor types and subtypes. Measuring tumor markers to assess prognosis should also prove to be an important guide to the subsequent management of thyroid malignancies by indicating which patients are likely to benefit least and most from aggressive therapy. Finally, as we shall discuss in the next section, marker detection can help to pinpoint the underlying molecular mechanismsoftumordevelopment,providingvaluableinsightsonwhichtobase innovative new therapies for the treatment of thyroid cancer. ONCOGENES AND TUMOR SUPPRESSOR GENES
Throughout their lifespans, somatic cells can be thought of as progressing through 4). Stem cells initially proliferate by undergothree overlapping transitional stages (Fig. ing repetitive cell division resulting in a rapid expansion of immature tissue mass. Subsequently, these cells differentiate into mature cells that carry out the functions characteristic of their particular phenotype. Later they grow senescent and undergo programmed cell death or apoptosis. Tumor development, or neoplasia, results from stimuli that augment cellular proliferation or that impair cell differentiation andor apoptosis. A diverse set of signaling and effector proteins is involved in the precise of events. Mutations in the genes encod regulation of this enormously complex series (108). these proteins have been found to underlie the majority of human malignancies Genes that encode the proteins which promote normal cell proliferation are called protooncogenes. Protooncogenessometimesdevelopactivating or gain-of-function or mutations that result in the production of proteins which are qualitatively overactive quantitatively excessive and which thereby promote overly robust cellular prolife
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
463
These mutated protooncogenes are known oncogenes as (108-114). Oncogene mutations tend to be dominantly expressed and thus become clinicallyinapparent the heterozygous state. Other genes, termedtumor suppressor genes(114-118), encode the proteins that serve to restrain excessive cellular proliferationor to promote cell differentiationand or apoptosis. Inactivatingor loss-of-function mutationsof these tumor suppressor genes can also lead to neoplasia; these tend to be recessive and thus of clinical are consequence only when they are presentin the homozygous state. Cells that are undergoing unregulated proliferation as a result of an activated oncogene or an inactivated tumor sup gene are said to be trunsfonned. Cancer-causing mutations may be either somatic or germline. Somatic mutuh’ons are those that develop, usually in a single cell, at any time in life after fertilization. The transformed cell, through some survival advantage conferred by the mutation, expands monoclonally into a solitary tumor mass that may eventually invade or metastasize. Germline mutations, by contrast, originate in a parent and are passed to offspring through a germ cell. Affected offspring have the mutation present diffusely and may thus be susceptible to the development of multipletumors within a given organ or to tumors in multiple organs throughout the body. Most known inherited cancer syndrom result from germline mutations in tumor suppressor genes. Accordingly, individuals initially unaffected because of the are born heterozygous at a critical locus but are normal gene at the homologous locus. If, however, a somatic mutation later in life inactivates the normal homologous locus, the individual is rendered unable to make any of the normal suppressor protein and begins to develop cancer. The complex system that regulates cellular proliferation, differentiation, and apopt has many checks and balances. Although a single genetic mutation may initially transform a cell permitting the monoclonal expansion of its progeny, it is unlikely that a singlemutationalonecouldresultinthedevelopmentofhighlymalignanttumor behavior. However, it appears that the unregulated proliferation of a transformed cell predisposes it to develop additional mutations. These, t inurn,provide further selective survival advantages by promoting ever-accelerating cell proliferation, tissue invasion and distant metastases. Indeed, experimental evidence indicates that multiple activated oncogenes and inactivated tumor suppressor genes are often found in highly malignant and metastatic tumors (119-120). Having reviewed these basic concepts, we will now examine a general model of cellular growth signal transduction and then apply this model to thyroid growth and function and current concepts of thyroid oncogenesis. NORMAL GROWTH SIGNAL TRANSDUCTION: AMODEL
Although the precise mechanisms of signal transduction vary considerably among the different tissues of the body, a simple and general model is proposed in Figure 5. Accordingly, in response to a need for cellular change, an extracellular molecular sig is generated and binds to a specific cell membrane receptor. This results in the generation of intracellular messengers that relay the message to the nucleus by activating appro nuclear transcription factors. These proteins bind to the promoter regions of specific genes to modulate production of the cell cycle regulatory proteins that direct cells to proliferate, to differentiate, or to undergo apoptosis (108,118).
464
McDemzo tt
Messengers
~
y/
Regulators Cell Cell Cell Division Differentiation Death
J
r
x DNA Repair fl
Fig. 5. A model for growth signal transduction.An extracellular signal molecule binds to a cell membrane receptor resulting in the generation of intracellular messengers that relay the bind to signal to the nucleus by activating transcription factors. Active transcription factors targeted genes to modulate their productionof the cell cycle regulatory proteins that ultimately direct the cell to proliferate, to differentiate, or to die.
CANCER CAUSING GENETIC MUTATIONS IN THE THYROID GLAND
Much has been learned about the molecular mechanisms of tumor formation in th thyroid gland. Thyroid cancers, like malignancies elsewhere in the body, appear to result from oneor more aberrations in the genes that code for proteins involved in regulation of cellular proliferation, differentiation, and apoptosis. In this section we will use our simple model ofgrowth signal transduction (Fig.5 ) to classify theknown thyroid oncogenes and tumor suppressor genes into those whose protein products se as extracellular signals, as signal receptors, as intracellular messengers,as transcription factors, and as cell cycle regulatory proteins(120-127). We will further discuss how abnormalities of maintenance proteins can help to sustain tumorgrowth and foster the development of tissue invasion and metastases. After exploring these mechanisms we will discuss how the study of tumor-causing genes may be applied clinically to the diagnosis, prognosis prediction, and treatment of specific thyroid malignancies.
Extracellular Signaling Proteins Thyrotropin (TSH) is secreted into the circulation by the pituitarygland and subsequently binds to specific receptors on thyroid follicular cell membranes. itThere serves as the predominant extracellular regulator of thyroid function and as one of the signals for thyroid growth (128). The latter is also strongly influenced by other systemic and
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
465
Growth
twnction Fig. 6. Extracellular signals regulating thyroidgrowth and function. TSH is secreted by the pituitary gland and subsequentlybinds to specificreceptorsonthyroidcellmembranes;it activates an intracellular messengerpathway that promotes both thyroid cellgrowth and function. Other growth factors also bind to their specific receptors and initiate signaling cascades that
stimulate primarily thyroid growth.
local growth factors (Fig.6) such as IGF-1, EGF, basic fibroblast growth factor (FGF), platelet-derivedgrowthfactor(PDGF), TGF-a, TGF-P,interleukin-1 (E-1) and (129,130). Some of these are undoubtedly others that have not yet been identified produced by thyroid cells themselves or by local struma and bind to specific thyroid cell receptors, acting in an autocrine or paracrine fashion to stimulate signaling pathways involved in thyroid cell proliferation or growth. The role of TSH in the development and progression of thyroid neoplasia is, at present, somewhat uncertain (131,132). Endemic iodine deficiency, with its attendant chronic TSH hypersecretion, clearly fosters goiter development and is associated with (133). TSH hypersea significantly increased incidence of follicular thyroid carcinomas cretion, inthis instance, is not due to a genetic mutation; the association does, however, This principle suggest a mitogenic effect of excess TSH stimulation on thyroid tissue. is the basis for the widespread clinical use of L-thyroxine therapy to suppress TSH secretion in most patients with differentiated thyroid carcinoma. Other thyroid growth factors also appear to be important in thyroid tumorigenesis. IGF-1, EGF, FGF, PDGF, TGF-a, TGF-P, and IL-1, as well as receptors for IGF-1, EGF, and PDGF, have all been reported to be overexpressed to various degrees in manythyroidtumors (134-164). The mechanisms responsible for their heightened expression have not been determined. Furthermore, it is not yet known if these growrh signals are capable of providing a primary stimulus for thyroid neoplasia. It seems
466
McDemzott
Function Growth
>&
Eig. 7. TSH signal transduction in thyroid cells.TSH binds to the extracellular domainof the TSH receptor (TSH-R) causing dissociationof the guanine nucleotide binding stimulatory protein (GsP) into its a, p, and y subunits and release of an a-bound guanosine diphosphate (GDP) molecule. Alpha then attaches to a guanosine triphosphate (GTP)molecule, forming an a subunitT active dimer that stimulates cyclic adenosine monophosphate (c-AMP) production. then utilizes intrinsic GTPase activity to deactivate itself by converting the GTP back to GD This pathway promotes both thyroid cellular proliferation and thyroid hormone production.
likely, however, that one or more growth factors, acting alone or in concert, serve at least a supportive role in the development, maintenance and progression of thyroid tumor growth (134,135). Thyroid-stimulating immunoglobulins (TSI) are yet another aberrant extracellular in patients with Graves’ diseas thyroid growthsignal. Produced by altered B-cell genes theyordinarilypromotethyroidhyperplasiaandhyperfunction.There is evidence, however,thatpatientswithGraves’diseasehave an increased risk of developing differentiated thyroid cancer (165-168) and further that these tumors may be more (169-1 71). Presumably these aggressive than thosein patients without Graves’ disease effects are related to the growth stimulating properties of the TSI (172).
TSH ReceptotLGsP a Signaling Pathways
The TSH receptor (TSH-R) is a protein composed of a large extracellular domain, a transmembrane domain with seven membrane-spanning segments and a short in lular domain (Fig. 7). It is coupled to a guanine nucleotide binding stimulatory protein (GsP) that is composed of a,p, and y subunits and an alpha-bound guanosine diphosp (GDP)molecule. The a subunit has several significant properties: it is inactive when
467
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
GDPGTP
1
-P
4
Hyperfunctioning Follicular Adenoma
A AM PI
Fig. 8. Activating TSH receptor (TSH-R) mutations. Some TSH-R gene mutations produce a TSH-R that is constitutively active, without a need for TSH binding. Somatic mutations of this type may lead to the development of an autonomously functioning follicular adenoma.
bound to GDP; it is active when bound to guanosine triphosphate (GTP); it possesses and intrinsicGTPaseactivitythatconvertsGTPtoGDP. TSH bindingtothe TSH-R extracellular domain stimulates the transmembrane and intracellular domains to initiate an intracellular messenger cascade that begins with the release of GDP and dissociation of GsP into its subunits. Once liberated, the free a subunit attaches to GTP, forming an active dimer that stimulates adenylate cyclase to generate cyclic adenosine monoits intrinsic GTPase activity, the a subunit converts phosphate (c-AMP). Then, utilizing the bound GTP back to GDP, thereby deactivating the a-GTP dimer. The burst of c-AMP stimulates protein kinaseA, which enlists additional proteins into a messenger cascade that ultimately stimulates thyroid cells to proliferate and to produce thyroid hormone (128). Autonomously functioning follicular adenomas are benign thyroid neoplasms that grow and produce thyroid hormone without any apparent requirementTSH for stimulation. Most of these tumors have been found in (173-178), some but not other(179-181), geographic locations to harbor activating point mutations in TSH-R the gene, resulting in the presence or absence of TSH binding. in TSH-Rs that are constitutively overactive Since theTSH-R is normally involved in stimulating both thyroid growth and function, these constitutively active TSH-Rs promote the neoplastic expansion of a clone of hormone-producing thyroid cells (Fig. 8). Activating mutations of the gene encoding the aGsP subunit havealso been detected (180-183). The mutations in up to 25% of autonomously functioning follicular adenomas described produce ana subunit that has normal GTP binding ability but lacks intrinsic a-GTP dimers GTPase activity (Fig. 9). These abnormal alpha proteins form active
468
McDermott
GDP GTP
I' E' "
Hyperfunctioning Follicular Adenoma
i
Fig. 9. Activating GsP-a subunitmutations.Mutations in thegeneencodingthe GsP-a subunit produce an a protein that retains the ability to bindGTPbut has deficient intrinsic GTPase activity. Once activated by GTP binding,it is unable to deactivate itself by degrading a-GTP dlimer generates a continuous excess ofc-AMP that GTP to GDP. The persistently active promotes thyroidcell proliferation and function leading to the development ofan autonomously functioning follicular adenoma.
that generate c-AMP continuously and excessively because they have no intrinsic GTPase to deactivate themselves. Since GsP-a is a downstream component of the TSH-R signaling pathway, these activating GsP-a mutations also predispose to the development of benign neoplasms that autonomously proliferate and secrete thyroid hormone. Neither TSH-R nor GsP-alpha mutations, however, to appear play a significan role in the development of malignant thyroid neoplasms (180,181,184).
Growth FactotcRas Signaling Pathway Thyroid growth factors other than TSH also bind to their specific cell membrane receptors and initiate a diverse array of growth signaling pathways. Some of these utilize Ras proteins as intracellular messengers(Fig. 10). Like the GsP-a subunit, Ras proteins are inactive when bound to GDP,are active when bound to GTP, and possess intrinsic GTPase activity.In the basal state, Ras is tethered to the cell membrane, in an inactive dimeric complex with GDP. The binding of an extracellular ligand to membrane receptor causes receptor dimerization and phosphorylation of tyrosine residues on the receptor's intracellular domain. The activated receptor then engages two proteins, Grb-2 andSos, which cooperate to dissociate Ras from GDP. Ras immed attaches to GTP, forming an active Ras-GTP dimer that initiates a multistep protein phosphorylation cascade. Subsequently Ras utilizes its intrinsic GTPase activity to disable itself by converting the bound GTP back to GDP. This Ras-associated pathway
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
-P grb2 sos
1 - 1 1 ,
Growth
469
P21 ras
GDP GTP
p21
U ras
GDP
'+-
GTP
1 I I MAP Kinase Cascade
Fig. 10. The Ras signaling proteins. Ras proteins are bound to GDP molecules as inactive dimers attached to the inner aspect of the cell membrane. When a ligand binds to a Rasassociated receptor, the intracellular portion of the receptor becomes phosphorylated and engages two proteins, Grb-2 andSos, into a complex that dissociatesRas from GDP. Ras then binds to GTP, forming an active dimer that initiates a protein phosphorylation cascade. Ras then uses its intrinsic GTPase activity to deactivate itself by converting the GTP back to GDP. This pathway promotes primarily thyroid cell proliferation.
eventually activates nuclear proteins that promote primarily thyroid cellular proliferat but have little or no known effects on thyroid function. Activating Ras gene mutations have been found in20% to 80% of nonfunctioning (35,52-55,57,64,96,97,185follicular adenomas and differentiated thyroid carcinomas 187). In a situation analogous to GsP-a, these mutated Ras proteins can bind to GTP but lack the GTPase activity necessary to deactivate the Ras-GTP dimer (Fig.11).In contrast to GsP-a, however, Ras mutations result in overstimulation of a signaling cascade that encourages the growth of nonfunctioning thyroid tumors.
Ret Receptor-Tyrosine Kinase Signaling Pathway The Ret receptoris a component of a different signaling cascade that appears to be involved primarily in the proliferation of cells of neural crest origin (Fig. 12). Ret is normally expressed in the calcitonin producing parafollicular C cells but not in the thyroid hormone-producing follicular cells. Ret has an extracellular ligand binding domain, a single transmembrane segment and an intracellular portion that possesses low-level intrinsic tyrosine kinase activity. The ligand that binds to the Ret receptor (188-191). wasrecentlyidentifiedasglial-cellderivedneurotropicfactor(GDNF)
470
McDermott
c
-P sos
“ 1 1
ras
ras GDP
I
I
GDP GTP
4
4
GTP
-P
Growth4-
Fig. 11. Activating Ras mutations. Mutations in Ras genes result in the production of Ras proteins that can bind to GTP but that lack intrinsic GTPase activity. The persistently active Ras-GTP dimer excessively stimulatesa cascade of protein phosporylations that promote prima ily thyroid cell proliferation. This predisposes to the development of nonfunctioning follicular neoplasms.
Binding ofGDNF results in dimerization of the Ret receptor and significant enhance of the receptor’s tyrosine kinase activity. This results in activation of a Ras-signaling pathway of downstream proteins that relay the message to the nucleus to promote cell division. Medullary carcinoma of the thyroid(MCT)is a malignancy of the parafollicular C cells. Approximately 10% are familial, while the remainder appear to be sporadic. FamilialMCT is anautosomaldominantdisorderthatoccurs in three recognized forms: familial isolated MCT, MCT associated with the MEN IIA syndrome and MCT associated with the MEN IIB syndrome. Familial MCT of all three types has been found, in over 90% of cases studied thus far, to harbor activating-point mutations in the Ret gene regions that encode the transmembrane and intracellular domains of the a Ret receptor (192-202). These result in constitutively active Ret receptors with high level of basal tyrosine kinase activity that sets in motion downstream growth-signa cascades eventually leading to diffuse C-cellhyperplasia and multifocal MCT (Fig. 13). The Ret/MCT oncogenehas also been detected in some sporadic MCTs (192,195198). This suggests that sporadic MCT, generally a unifocal tumor, may sometimes arise from a somatic Ret mutation in a single cell; familial MCT, by contrast, almost always results from a germline Ret mutation affecting all cells. Not surprisingly, the Ret/MCT oncogene has also been detected in pheochromocytomas from patients wi MEN IIA and MEN IIB syndromes (192-194,197,199).
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
Tyrosine Kinase
-
471
receptor I I I
Cell Division Fig. 12. The Ret receptor. Binding of the ligand, glial cell-derived neurotropic factor(GDNF), to the extracellular domain of the Ret receptor enhances the intrinsic tyrosine kinase activity of the intracellular domain, resulting in the phosphorylation of downstream proteins thatparticipate in a pathway that promotes cell proliferation.Ret receptors arenormally present in thyroid parafollicular C cells but not in the more abundant follicular cells.
Somatic activating Ret mutations of a different type have been discovered in approximately 2040% of papillary thyroid carcinomas (40,41,203-205). These mutations involve a gene truncatiordremangement that deletes the coding region for the recepto extracellular domain and interposes a promoter sequence adjacent to the coding region for the intracellular domain. This results in expression and constitutive activation of the intracellular portion of the Ret receptor in thyroid follicular cells14),(Fig. generating a continuous excess of tyrosine kinase activity, which stimulates downstream growth signals to promote thyroid follicular cell neoplasia. Why the Ret/PTC oncogene causes thyroid follicular cell tumors while the Ret/ MCT oncogene selectively transforms parafollicular C cells is now becoming clear. As mentioned, the wild-type Ret receptor is normally expressed in the neural crest derived C cells, although it is not normally present in branchial cleft-derived thyroid follicular cells. A simple point mutation is sufficient to constitutively activate Ret in C cells whereas a more drastic mutation such as the Ret/PTC rearrangement must this receptor in cells where it is be required to cause expression and activation of normally repressed. Met and TRK are other membrane receptors with intrinsic tyrosine kinase activity that can be enhanced by cognate ligand binding. Constitutively activating mutations
472
McDermott
(Tyrosine Kinase
-
1 I
Cell Division
Fig. 13. Activating Ret receptor mutations (ReVMCT) in medullary carcinoma of the GermLine point mutations in the gene regions that code for the intracellular domain of the Ret receptor produce receptors with enhanced basal tyrosine kinase activity. This results in excessiv activationofdownstreamproteinsthatpromoteC-cellproliferationpredisposingtoC-cell hyperplasia and multifocal medullary carcinoma.
v Cell Division
Fig. 14. Activating Ret receptor mutations (RetPTC) in papillary thyroid carcinoma. deletiodrearrangement mutations of the Ret receptor gene in thyroid follicular cells produ basal tyrosine kinase activity. This causes excessive activation truncated receptor with enhanced of a pathway that promotes thyroid follicular cell proliferation and leads to the development of papillary thyroid carcinoma.
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
473
$. Cell Cycle
Fig. 15. Normal p53 protein function. The normal p53 protein enters the nucleus and binds to targeted genes where it modulates the production of the cell cycle regulatory proteins that inhibit cell proliferation and promotecell differentiation, DNA repair and age appropriate apoptosis.
of the genes that code for these receptors are also reportedly involved in the pathogenesis of some papillary carcinomas (204,206,207). Investigation in this area is ongoing and it is hoped will be enlightening in the near future.
Thyroid Transcription Factors Transmission of messages along most growth-signaling pathways eventually results Fos are two such transcription in the activation of nuclear transcription factors. Myc and
the manufacture of the cell factors that bind to targeted genes where they enhance cycle regulatory proteins that ultimately stimulate thyroid cell division (208). Activating Fos may cause mutations in the promoter regions of the genes encoding Myc and excessive synthesis of these factors with resultant overproduction of growth promoting regulatory proteins. Overexpression of both Myc and Fos has, in fact, been detected in a large but variable proportion of thyroid tumors and tends to be associated with more aggressive tumor behavior (45,46,51,57,58,64,95,99,209-211). The p53 protein,in contrast, is a multifunctional antiproliferative transcription factor (Fig. 15) that augments the production of regulatory proteins that inhibit cell division and others that promote DNA repair and apoptosis (212,213). Inactivating mutations of the p53 gene result in the production of nonfunctional p53 proteins that predispose to malignant transformation and progressionby multiple mechanisms (Fig. 16). Since abnormal p53 proteins are often catabolized more slowly than normal p53, their prese of total p53 protein. p53 is frequently discovered by finding increased tissue levels gene mutations are among the most common genetic abnormalities in human cancer,
474
McDermott I
”-
=
Fig. 16. Inactivating p53 mutations. Inactivating mutations of the p53 genea nonfuncproduce tionalp53protein that leads to progressive malignant tumorgrowth by failing torestrain excessive proliferation or failingto induce differentiation,DNA repair or apoptosis in thyroid cells that have been previously transformed by a primary genetic mutation.
being detected in up to 50% of all malignant tumors(214-216). They are rarely found in differentiated thyroid carcinomas but have been detected in 20% to 100%of poorly (65-72). This indicates that most p53 ge differentiated and anaplastic thyroid cancers mutations are not primary, but instead develop in cells transformed by a prior genetic abnormality and, further, that they portend progression to a more aggressive tumor 18% phenotype. Point mutations of the p53 gene were recently detected in of radiationinduced differentiated thyroid carcinomas and were associated with spread to cervic lymph nodes but, paradoxically, not with tumor recurrences or distant metastases(217). This suggests that the genetic events leading to radiation induced thyroid neoplasia may differ from those that produce other thyroid tumors. Bax and Bcl-2 belong to a different family of transcription factors that modulate the production of proteins involved primarily in the regulation of apoptosis (218,219). Alterations in the pattern of expression of these peptides have recently been discove in various thyroid cancer specimens (105,219), underscoring the importance of failed apoptosis in the pathogenesis of thyroid neoplasia.
Cell Cycle Regulatoy Proteins Cell cycle regulatory proteins are the end products or effectors of the growth-signali cascades (Fig. 5) (108,220,221). Mutations of the genes encoding these proteins also predispose to neoplasia by. a variety of mechanisms (108,222,223). The best-known protein in this category is pRB, a master regulator that normally serves to halt the cycle by preventing cells from progressing past the G1 phase. Inactivating pRB gene
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
475
mutations, initially described in childhood retinoblastomas, have now been detected in in the development of thyroid carcinomultiple tumor types(224) and may be involved mas (126). Cyclins are a separate class of regulatory proteins that promote the passage of cells through the cell cycle. Cyclin D, for example, binds to a cyclin-dependent kinase (CDK), forming a complex that inactivates pRl3, thereby allowing the cell to progress intoS phase (108). Activating mutations of the cyclin genes have been reported (126). in various tumors (222,223), but have yet to be demonstrated in thyroid cancers The DNA repair systemis another critical groupof regulatory proteins. Inactivating mutations of the genes encoding one or more proteins in this system are the cause of an inherited skin cancer syndrome (xeroderma pigmentosum) andare associated with hereditary nonpolyposis colon cancer and an increased incidence of other visceral malignancies. To date, however, they have not been linked to the development of thyroid neoplasms. Telomerase is a unique type of regulatory protein that is normally present only in germ cells. Telomeres are the chromosomal caps that function to prevent chromosomes from sticking together or forming otherwise unstable configurations. Human telomeres consist of approximately2000 repeats of the nucleotide sequence TI'GGGG. Each time a cell divides,it loses some of its telomeric sequences. Once telomeres are reduced to a critical length, the cell stops dividing and eventually undergoes apoptosis. Germ cells, however, avoid this fate by producing an enzyme known as telomerase that rebuilds this allowsgermcellstoproliferatecontinuously telomeresaftereachcellcycle; 90% of human throughout the life of the host. Telomerase has been detected in nearly cancers from various sites (225,226). It has been theorized that it may play a role in the immortalization of malignant cells. Telomerase has recently been demonstrated in over 60% of papillary thyroid carcinomas (227); curiously, it was not found in other thyroid tumor types. Clarification of the significance of these findings and the mech by which the normally silent telomerase gene becomes activated into an oncogene in human neoplasms will await further investigation.
Maintenance Proteins As an organ grows it must manufacture ancillary proteins to maintain the integrity of existing tissue and to serve as a support for further tissue growth. Although a vast array of proteins belong in this category, for the purposes of this discussion we shall briefly consider only two groups: the angiogenesis factors and the cell adhesion molecules. Angiogenesisfactorsareproteinsproducedbygrowingtissuestostimulatethe development of vascular networks to ensure an adequate supply of oxygen and nutri Since tumors frequently require a robust vascular supply and many develop hypervasc larity, it is likely that overexpression of one or more angiogenesis factors plays a key supportive role at some point in tumor development(228). Basic FGFis one angiogenic factor that is present in normal thyroid tissue and has been shown to be increased in several thyroid tumors (229,230). The contribution of FGF and other vascular growth factors to the development of thyroidtumors is not yet fully understood but certainly warrants further investigation (230). Cell adhesion molecules fall into two general categories: cell-cell adhesion proteins and cell-matrix adhesion proteins. Cell-cell adhesion proteins, such as the cadherins,
476
McDemzott
Table 2 Genetic Abormalities in Thyroid Neoplasms Classified According to Their Gene Products
oduct Gene Category Product Gene
Signals
TSH
PTC, FC
Growth factors
All types
TSI Signal receptors
Intracellular messengers
TSH-R Ret/MCT RetPTC Met Trk
AFA
GsP-a
AFA M A , FC, PTC
Ras Transcription factors
Regulatory proteins
PTC, FC
MCT PTC PTC PTC
Fos P53
Bm, Bcl-2
FC, PTC, AC FC, PTC, AC AC All types
Telomerase
PTC
MYC
TSH,thyrotropin; PTC, papillary thyroid carcinoma;FC, follicular carcinoma;TSI,thyroid stimulating MCT, immunoglobulins;TSH-R,thyrotropin receptor;M A , autonomously functioning follicular adenoma; medullary carcinoma of the thyroid; NFA, nonfunctioning follicular adenoma; AC, anaplastic carcinoma,
maintain tissue integrity by causing the cells of a given organ to stick together. Cellmatrix adhesion proteins, known as integrins, anchor cells to the extracellular matrix; not only does this provide structural integrity, but these cell-matrix links serve as a scaffolding that is required by some cell types for continued cell division (anchorage dependence). Benign neoplasms that remain localized to their tissue of origin proba retain these adhesive proteins, whereas they may belost or altered in tumors that invad adjacent tissuesor metastasize to distant sites(230-233). In support of this hypothesis, the expression of E-cadherin, a cell-cell adhesion protein, is significantly reduced in malignant compared to benign thyroid neoplasms and its absence is associated with an increased risk of metastatic spread (48).
Multiple Mutations Contribute to Thyroid Cancer Several different oncogenes and at least one inactivated tumor suppressor gene have thus been found in various types of thyroid neoplasia (Table 2). Some are associated primarily with benign functioning follicular adenomas (TSH-R andGsP-a) and others are present in nonfunctioning tumors, both benign and malignant (Ras). Some may b seen in multiple carcinoma types (Myc,Fos) while others are detected predominantly or exclusively in specific tumor types (RetPTC in papillary carcinomas, Ret/MTC in medullary carcinomas and p53 in undifferentiated and anaplastic carcinomas). Once a cell has been transformed by a primary mutation, it becomes increasingly prone to develop additional mutations. Therefore it is likely that thyroid.cancers generally resu
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
477
Thyroid Follicular Cell Gsa
Functioning Follicular Adenoma
Transformed Thyroid Cell
Nonfunctioning Follicular Adenoma
RevpTcl
Telomerase
Papillary Carcinoma
Fos
Follicular Carcinoma
Anaplastic Carcinoma
Fig. 17. Multiple genetic abnormalities in thyroid neoplasia. Activating mutations of the TSH-Rand GsP-a genesstimulateboththyroid growth andfunction,predisposingtothe developmentof autonomously functioning follicular adenomas with a low potential for mal degeneration. Ras mutations, by contrast, enhance only growth, promoting the formation of nonfunctioning follicular tumors that are prone to incur further genetic alterations. Mutations that affect the Ret receptor or, possibly, that induce telomerase expression may then lead to the development of papillary carcinoma; those that cause overexpression of Myc and/orFos may transform follicular adenomas into follicular carcinomas. In any transformed cell line, an inac vating p53 mutation may then cause progression to a highly malignant anaplastic carcinoma.
from multiple sequential genetic abnormalities. A current proposed scheme for the molecular events underlying the development of various thyroid typestumor is illustrated in Figure 17. CLINICAL APPLICATIONS
The study of oncogenes and tumor suppressor genes has provided scientists and physicians with extremely valuable insights into the pathophysiology of thyroid cancer development and progression. Many of these discoveries have or will soon have important clinical applications in the areas of diagnosis, prognosis, and treatment.
Diagnosis Since specific oncogenes and tumor suppressor genes are characteristic of certain tumor types (Table 2), tissue obtained by FNA, at surgery, or from archived samples might soon be screened with an oncogene panel to facilitate the determination of an accurate diagnosis. This could be particularly helpful in the preoperative distinction between follicular adenomas and follicular carcinomas.
478
McDemzott
Preclinical screening for familial MCT is another area where oncogene testing has great promise. Since this disorder is autosomal dominant and will be transmitted, on average, to 50% of an affected individual’s offspring, family members of an MCT patient must be screened for the disease. Traditionally, family screening has involved the measurement of basal and stimulated serum or plasma calcitonin levels to detect members with early involvement. However, these tests do not become positive until either C-cell hyperplasiaor frank MCT is present and therefore may not detect disease early enough to effect a cure. Furthermore, when negative, they must be repeated annually until around age35. By contrast, Ret/MCT genetic testing can be performed of over 99%, can on a sample of peripheral blood mononuclear cells, has a sensitivity detect the disorder at birth and needs to be done only once in each family member (194,202).Though not yet widely available, testing for the Ret/MCT oncogene will soon likely eliminate the need for calcitonin testing when screening family members for the presence of preclinical or subclinical MCT.
Prognosis The capability of accurately forecasting the probable future behaviorof a tumor is particularly important in the management of thyroid cancer. The majority of thyroid malignancies are indolent and do not substantially affect life expectancy or lifestyle but a significant minority recur or metastasize, resulting in morbidity and premature mortality. Prognosis estimates are currently based on clinical features such as patient age, tumor size, histologic grade and the presence of local tissue invasion or distant metastases. Oncogene screening of tissue from the primary tumor or from recurrent lesions may soon allow more accurate prognostic estimation based on the types or numbers of genetic mutations detected.
Treatment Thyroid cancers are usually treated with a combination of surgery, radioactive i administration, and L-thyroxine suppression. In the near future, oncogene screening may play a prominent role in therapeutic decision-making. It could help physicians a more aggresdecide, based on prognosis, whether an individual patient should receive sive or more conservative course of treatment. In this regard it may serve as a guide to decisions such as the optimal extent of surgery, the dose of radioactive iodine and the degree of L-thyroxine suppression. Furthermore, by more fully illuminating the molecular factors underlying tumor behavior,it may lead to the development and use of chemotherapeutic and/or immunological agents that specifically retard thyroid cel proliferation, that promote differentiation into more mature cell types, that restore th normal mechanisms of DNA repair and/or apoptosis, that impair tumor angiogenesis or that maintaincell-cell and cell-matrix adhesiveness(234-236).Such advances coul usher in a new era of safer and more effective treatment.
REFERENCES 1. Jorde LB, Carey JC, White RL. Basic cell biology: structure and function of genes and chromosomes. In: Medical genetics. St Louis: Mosby-Yearbook, 19957-29. 2. Wallin G,Backdahl M, Auer G. Analysis of the DNA content in themanagement of thyroid tumors. Thyroidology 1991; 3:25-29.
DNA PZoidy, Tumor Markers, Cancer-Causing and
Genes
479
3. Hay ID. Prognostic factors in thyroid carcinoma. Thyroid Today1989; 12:l-9. 4. Cusick EL, MacIntosh CA, Krukowski ZH, Ewen SW, Matheson NA. Comparison of flow cytometry with static densitometry in papillary thyroid carcinoma. Br J Surg1990; 77~913-916. of cellular DNA content on survival 5. Joensuu H, Klemi P, Eerola E, Tuominen J. Influence in differentiated thyroid cancer. Cancer1986; 58:2462-2467. T, Noguchi M, NakamuraS , Hashimoto 6. Mizukami Y,Nonomura A, Michigishi T, Kosaka T. Flow cytometric DNA measurement in benign and malignant human thyroid tissues. Anticancer Res 1992; 12:2213-2217. 7. Tsuchiya A, Sekikawa K, Ando Y, Suzuki S , Kimijima I, Abe R; Flow cytometric DNA analysis of thyroid carcinoma. Jpn J Surg 1990; 20:510-514. 8. Schelfhout LJ, Cornelisse CJ, Goslings BM, Hamming JF, Kuipers-Dijkshoorn NJ, van de Velde C J , Fleuren GJ. Frequency and degree of aneuploidy in benign and malignant thyroid neoplasms. Int J Cancer 1990; 45:16-20. 9. Cheifetz RE, Davis NL, Robinson BW, Berean KW, LeRiche JC. Differentiation of thyroid neoplasms by evaluating epithelial membrane antigen, Leu-7 antigen, epidermal growth factor receptor and DNA content. Am J Surg 1994; 167:531-534. 10. Jonasson JG, Hrafnkelsson J. Nuclear DNA analysis and prognosis in carcinoma of the thyroid gland: a nationwide study in Iceland on carcinomas diagnosed 1955-1990. Virchows Arch 1994; 425:349-355. 11. Grant CS, Hay ID, Ryan JJ, Bergstrah El, Rainwater LM, GoellnerJR. Diagnostic and prognostic utility of flow cytometric DNA measurements in follicular thyroid tumors. World J Surg 1990; 14:283-289. 12. Hruban RH,Huvos AG, Traganos F, Reuter V, Lieberman PH, Melamed M R . Follicular neoplasms of the thyroidin men older than50 years of age: a DNA flowcytometric study. Am J Clin Path01 1990; 94527-532. 13. Cusick EL, Ewen SWB, Krukowski ZH, Matheson NA. DNA aneuploidy in follicular thyroid neoplasia. Br J Surg 1991; 78:94-96. 14. Harlow S , Duda RB, Bauer KD. Diagnostic utility of DNA content flow cytometry in follicular neoplasmsof the thyroid. J Surg Oncol 1992; 50:1-6. 15. Zedenius J, Auer G, Backdahl M, Falkmer U, GrimeliusL, Lundell G, Wallin G. Follicular tumors of the thyroidgland diagnosis, clinical aspects and nuclear DNA analysis. World J Surg 1992; 16:589-594. 16. Czyz W,Joensuu H, PylkkanenL, Klemi PJ. p53 protein, pcnastaining, andDNA content in follicular neoplasmsof the thyroid gland. J Pathol 1994; 174;267-274. 17. Lukacs GL, Balazs G, Nagy I, Miko T. Clinical meaning of DNA content in the long term follow up. EurJ Surg1994; 160:417-423. behaviourof follicular thyroid tumours:12-year a 18. Galera-Davidson H, BibboM, Dytch HE, Gonzalez-Campora R, Fernandez A, Wied GL. Nuclear DNA in anaplastic thyroid carcinoma with a differentiated component. Histopathology 1987; 11~715-722. 19. Schroder S , Bocker W, Baisch H, et al. Prognostic factors in medullary thyroid carcinomas: and content. survival in relation to age, sex, stage, histology, immunocytochemistryDNA Cancer 1988; 61:806-816. 20. Hay ID, Ryan JJ, Grant CS, Bergstralh El, van Heerden JA, Goellner JR. Prognostic significance of nondiploid DNA determined by flow cytometry in sporadic and familial medullary thyroid carcinoma. Surgery 1990; 108:972-980. 21. el Naggar AK,Ordonez NG, McLemore D, Schultz P, HickeySamaan RC, N. clinicopatho1990; logic and flow cytometric DNA study of medullary thyroid carcinoma. Surgery 108~981-985. de Tolosa EMC, Ferreira EABF. DNA image cytometric analysis 22. Camargo RS, Scafuri AG, of differentiated thyroid adenocarcinoma specimens.Am J Surg 1992; lM640-645. 23. Backdahl M, Carstensen J, Auer G, Tallroth E. Statistical evaluation of the prognostic
480
McDemzott
value of nuclear DNA content in papillary, follicular, and medullary thyroid tumor J Surg 1986;10:974-980. 2 4 . Backdahl M, Wallin G, Auer G,et al. Cellular DNA content in thyroid tumors: a reliabl factor for grading and prognosis. Prog Surg1988; 19340-53. 25. Hamming JF, Schelfhout LJDM, Cornelisse C J , et al. Prognostic value of nuclear DNA content in papillary and follicular thyroid cancer. World J Surg1988; 12503-508. M, Niendorf A, Schroder S. DNA cytophotometry in malignan 26. Arps H, Sablotny B, Dietel thyroid tumors: use of different evaluation schemes for prognostic statements. Virchows Archiv A Path01 Anat 1988; 413:319-323. 27. Smith SA, Hay ID, Goellner J R , Ryan JJ, McConahey W M . Mortality from papillary thyroid carcinoma: a case-control study of 56 lethal cases. Cancer 1988; 62:1381-1388. 28. Pasieka JL, Zedenius J, Auer G, Grimelius L, Hoog A, Lundell G, Wallin G, Backdahl M. Addition of nuclear DNA content to the AMES risk-group classifcation for papillary thyroid cancer. Surgery 1992;12:1154-1160. 29. Yamashita H, NoguchiS , Murakami N, Tsuji K, Yamaoka N, Sakamoto A. DNA ploidy and stromal bone formation as prognostic indicators of thyroid papillary carcinoma in aged patients: a retrospective study. Acta Path01 Jpn1993; 43:22-27. 30. Schurmann G, Mattfeldt T, Feichter G, Koretz K, Moller P, Buht H. Stereology, flow cytometry, and immunohistochemistry of follicular neoplasms of the thyroid gland. H Path01 1991;22:179-184. 31. Ryan JJ, Hay ID, Grant CS, Rainwater LM, Farrow GM, Goellner JR. Flow cytometric DNA measurements in benign and malignant Httrthle cell tumors of the thyroid. World J Surg 1988; 12:482-487. 32. el-Naggar A K , Batsakis JG, LunaMA, Hickey RC. Httrthle cell tumors of the thyroid a flow cytometric DNA analysis. Arch Otolaryngol Head Neck Surg1988; 114520-521. 33. Bronner MP, Clevenger CV, Edmonds PR, Lowell DM, McFarland MM, LiVolsi VA. Flow cytometric analysis of DNA content in Hiirthle cell adenomas and carcinomas of the thyroid. Am J Clin Pathol 1988; 89:764-769. 34. McLeod MK, ThompsonN W , Hudson JL, et al. Flow cytometric measurements of nucle DNAandploidyanalysisinHiirthle cell neoplasms of the thyroid. Arch Surg 1988; 123:849-854. 35. Schark C, Fulton N, Tashiro T, Stanislav G, Jacoby R, Straus
FH, 2nd. The value of measurement of RAS oncogenes and nuclear DNA analysis in the diagnosis of Hurthle cell tumors of the thyroid. World J Surg 1992; 16:745-752. 36. Ekman ET, Bergholm U, Backdahl M, Adami HO, Bergstrom R, Grimelius L, Auer G. Nuclear DNA content and survival in medullary thyroid carcinoma. Swedish Medullary Thyroid Carcinoma Study Group. Cancer 1990; 6531-517. 37. Pyke CM, Hay ID, Goellner J R , Bergstralh El, van Heerden JA, Grant CS. Prognostic significance of calcitonin immunoreactivity, amyloid staining, and flow cytometric DNA measurements in medullary thyroid carcinoma. Surgery 1991; 110964-971. 38. Riddell DA, Lampe HB, Cramer H, Troster M. Medullary thyroid carcinoma: prognostic factors. J Otolaryngol 1993; 22:180-183. 39. Hurwitz M, Sawicki M, Samara G, Passaro E. Diagnostic and prognostic molecu in cancer. Am J Surg 1992; 164:299-306. 40. Santoro M, Carlomagno F, Hay ID, Herrmann MA, Grieco M, Melillo R. Ret oncogene activation in human thyroid neoplasms is restricted to the papillary cancer subtype. J Invest 1992;89:1517-1522. 41. Jhiang SM, Mazzafem EL. The retPTC oncogene in papillary thyroid carcinoma.J Lab Clin Med 1994;123:331-337. of thyrotropin 42. BronnegardM, Tomng 0, Boos J, Sylven CM, Wallin G. Expression receptor and thyroid hormone receptor messenger ribonucleic acid in normal, hy and neoplastic human thyroid tissue. J Clin Endocrinol Metab1994; 79:384-389.
DNA Ploidy, Tumor Markers,Cancer-Causing and
Genes
481
43. de Micco C, Ruf J, Chrestian MA, Gros N, Henry JF, Carayon P. Immunohistochemical study of thyroid peroxidase in normal, hyperplastic and neoplastic human thyroid tissues. Cancer 1991; 67:3036-3041. 44. Yamasaki Y, Mori K, Naito M, Akagi M, Takahashi K. Histochemical determination of iodine peroxidase activity in various thyroid tissues.Am J Surg 1990; 160:271-276. 45. Brabant G, Maenhaut C, Kohrle J, Scheumann G, Dralle H, Hoang-Vu C, Hesch RD, von zur Muhlen A, Vassart G, Dumont JE. Human thyrotropin receptor gene: expression in thyroid tumors and correlation to markers of thyroid differentiation and dedifferentiation. Mol Cell Endocrinol 1991; 82:R7-R12. 46. Hoang-Vu C, Dralle H, Scheumann G, Maenhaut C, Horn R, von zur Muhlen A, Brabant G . Gene expressionof differentiation and dedifferentiation markers in normal and malignant human thyroid tissues. Exp Clin Endocrinol 1992; 100:51-56. 47. Wallin G, Bronnegard M, Grimelius L, McGuire J, Torring 0. Expression of the thyroid hormone receptor, the oncogenes c-myc and H-ras, and the 90 kD heat shock protein in normal, hyperplastic and neoplastic human thyroid tissue. Thyroid 1992; 2:307-313. 48.ScheumannGFW,Hoang-VuC,CetinY,Gimm 0, Behrends J, von Wasielewski R, Georgii A, BirchmeierW, von zur Muhlen A, Dralle H, Brabant G. Clinical significance of E-cadherin as a prognostic marker in thyroid carcinomas. J Clin Endocrinol Metab 1995; 80~2168-2172. 49. Vierbuchen M, Schroder S , Uhlenbruck G, Ortmann M, Fischer R. CA50 and CA199 antigen expression in normal, hyperplastic, and neoplastic thyroid tissue. Lab Invest 1989; 60~726-731. 50. Ghali VS, Jimenez US, Garcia RL. Distribution of Leu-7 antigen (HNK-1) in thyroid for follicular and papillary carcinomas. Hum tumors: its usefulnessas a diagnostic marker Path01 1992; 23:21-25. 51.Auguste LJ, Masood S , WesterbandA,BellucoC,ValderamaE,AttieJ.Oncogene expression in follicular neoplasmsof the thyroid. Am J Surg 1992; 164:592-593. 52. Namba H, GutmanRA, Matsuo K, Alvarez A, Fagin JA. H-Ras protooncogene mutations in human thyroid neoplasms. J Clin Endocrinol Metab 1990; 71:223-229. 53. Namba H, Rubin SA, Fagin JA. Point mutations of ras oncogenes are an early event in thyroid tumorigenesis. Mol Endocrinol 1990; 4: 1474-1479. Ras oncogene mutations 54. Karga H, Lee J-K, Vickery AL, Thor A, Gaz RD, Jameson in benign and malignant thyroid neoplasms. J Clin Endocrinol Metab 1991; 73:832-836. RF, Westbrook CA, StrausFH, Kaplan EL. N-ras 61 oncogene 55. Schark C, Fulton N, Jacoby mutations in HIwlle cell tumors. Surgery 1990; 108:994-1000. 56. Tuccari G, Barresi G. Immunohistochemical demonstration of ceruloplasmin in follicular adenomas and thyroid carcinomas. Histopathology 1987;11:723-73 1. 57. Masood S, Auguste LJ, Westerband A, Belluco C, Valderama E, Attie J. Differential A m JSurg oncogenicexpressioninthyroidfollicularandHiirthlecellcarcinomas. 1993; 166:366-388. 58. Shi Y, Zou M, Farid NR. Expression of thyrotrophin receptor gene in thyroid carcinoma is associated with a good prognosis. Clin Endocrinol Oxf 1993; 39:269-274. 59. Logmans SC, JobsisAC. Thyroid-associated antigens in routinely embedded carcinomas. Cancer 1984; 54:274-279. M R , MakinCA,HeydermanE. 60.Wilson N W , PambakianH,RichardsonTC,Stokoe Epithelial markers in thyroid carcinoma: an immunoperoxidase study. Histopathology 1986; 10:815-829. 61. Beltrami CA, Barbatelli G, Criante P, Paliage A, Amadi CE. An immunohistochemical study in thyroid cancer. Appl Path01 1987; 5:229-245. 62. Schroder S , Kloppel G. Carcinoembryonic antigen and nonspecific cross-reacting antigen in thyroid cancer: an immunocytochemical study using polyclonal and monoclonal antib ies. Am J Surg Path01 1987; 11:lOO-108.
JL.
482
McDmott
63. de Micco C, Ruf J, CarayonP; Chrestian MA, Henry JF, Toga M. Immunohistochemical study of thyroglobulin in thyroid carcinomas with monoclonal antibodies. Cancer 1987; 59:471476. 6 4 . Hashimoto T, Matsubara F,Mimkami Y, Miyazaki I, Michigishi T, Yanaihara N. Tumor markers and oncogene expression in thyroid cancer using biochemical and immunohist chemical studies. Endocrinol Jpn 1990; 37:247-254. 65. Dobashi Y, Sakamoto A, Sugimura I, Mernyei M, Mori M, Oyama T, Machinami R. Overexpression of p53 as a possible prognostic factor in human thyroid carcinoma. Am J Surg Pathol 1993; 17:375-381. 66. Dobashi Y, Sugimura H, Sakamoto A, Mernyei M, Mori M, Oyama T, Machinami R. Stepwise participation of p53 gene mutation during dedifferentiation of human thyroid carcinomas. Diagn Mol Path01 1994; 3:9-14. 67. Nakamura T, Yana I, Kobayashi T, Shin E, Karakawa K, Fujita S, Miya A, Mori T, Nishisho I, Takai S. p53 gene mutations associated with anaplastic transformation of human thyroid carcinomas. Jpn J Cancer Res 1992; 83:1293-1298. of p53 mutations with undifferentiate 68. Ito T, Seyma T, Mizuno T, et al. Unique association but not with differentiated carcinoma of the thyroid. Cancer Res 1992; 52:1369-1371. 69. Fagin JA, Matsuo K, Karmakar A, Chen DL, Tang S-H, Koeffler HP. High prevalence of mutations of the p53 gene in poorly differentiated human thyroid carcinomas. J Clin Invest 1993; 91:179-184. 70. Donghi R, Longoni A, Pilotti S, Michieli P, Porta GD, Pierotti MA. Gene p53 mutations are restricted to poorly differentiated and undifferentiated carcinomas of the thyroid gland. J Clin Invest 1993; 91:1753-1760. 71. Zedenius J, Larsson C, Wallin G, Backdahl M, Aspenblad U, Hoog A, Borresen A-L, Auer G. Alterations of p53 and expression of WAFl/p21 in human thyroid tumors. Thyro 1996; 6:l-9. 72. Zou M, Shi Y, Farid NR. p53 mutations in all stages of thyroidcarcinomas.J Clin Endocrinol Metab 1993; 77: 1054-1058. 73. Carcangiu ML, Steeper T, Zampi G, Rosai J. Anaplastic thyroid carcinoma: a study of 70 cases. Am J Clin Path01 1985; 83:135-158. Franke W. Intermediate74. SchroderS, Dockhom-Dworniczak B, Kastendieck H, Bocker W, W filament expression in thyroid gland carcinomas. Virchows Archiv (A) 1986; 409:7 75. Hurlimann J, Gardiol D, Scazziga B. Immunohistology of anaplastic thyroid carcinoma: a study of 43 cases. Histopathology 1987; 11567-580. tumors: immunohistology. 76. LiVolsiVA, BrooksJJ, Arendash-Durand B. Anaplastic thyroid Am J Clin Pathol 1987; 87:4344l2. 77. VenkateshYSS, Ordonez NG, Schultz PN, Hickey RC, Goepfert H, Samaan NA. carcinoma of thethyroid a clinicopathologic study of 121 cases. Cancer 1990; 66321-33 E, Mason DY. The value of 78. Ralfkiaer N, Gatter KC, Alcock C, Heryet A, Ralfkiaer immunocytochemical methods in the differential diagnosis of anaplastic thyroid tumors. Br J Cancer 1985; 52:167-170. 79. Kendall CH. Distinguishing lymphoma and small cell anaplastic carcinoma of the thyr by immunocytochemistry. J Clin Pathol 1986; 39:231. 80. Holting T, Moller P, Tschahargane C, Meybier H, Buhr H, C. Herfarth Immunohistochemical reclassificationof anaplastic carcinoma reveals small and giant cell lymphoma. W J Surg 1990; 14:291-294. 81. Burt AD, Ken DJ, Brown IL, Boyle P. Lymphoid and epithelial markers in small cell anaplastic thyroidtumors. J Clin Path01 1985; 38:893-896. 82. Shvero J, Gal R, AvidorI, Hadar T, KesslerE. Anaplastic thyroid carcinoma: a clinical histologic, and immunohistochemical study. Cancer 1988; 62:319-325. 83. Sundler F, Alumets J, Hakanson R, Bjorklund L, Ljundberg 0. Somatostatin immunoreac tive cells in medullary carcinomaof the thyroid. Am J Path01 1977; 88:381-386.
DNA PZoidy, Tumor Markers, Cancer-Causing and 84.
Genes
483
Mendelsohn G, Eggleston JC, Weisburger WR, Gann DS, Baylin SB. Calcitonin and histaminase in C-cell hyperplasia and medullary thyroid carcinoma: a light microscopic and immunohistochemical study.Am J Path01 1978; 92:35-52. 85. Capella C, Bordi C, Monga G, Buffa R, Fontana P, BonfantiS. Multiple endocrine cell 5HT, types in thyroid medullary carcinoma: evidence for calcitonin, somatostatin, ACTH, and small granule cells. Virchows Archiv (A) 1978;377:lll-128. of medullary thyroid 86. Deftos LJ, Bone HG 111,Parthemore JG. Immunohistological studies carcinoma and C cell hyperplasia. J Clin Endocrinol Metab 1980;51:857-862. 87. Krisch K, Krisch I, Horvat G, NeuholdN, Ulrich W. The valueof immunohistochemistry in medullary thyroid carcinoma: a systematic study of 30 cases. Histopathology 1985; 9~1077-1090. 88. Sikri KL, Varndell IM, Hamid QA, Wilson BS, Kamaya T, Ponder BA. Medullary carcinoma of the thyroid. An immunocytochemical and histochemical study of 25 cases using eight separate markers. Cancer 1985; 56:2481-2491. 89. Wiedenmann B, Franke WW, Kuhn C, Moll R, Gould VE. Synaptophysin: a marker protein for neuroendocrine cells and neoplasms. Roc Natl Acad Sci USA 1986; 8333500-3504. 90. Gould VE, Wiedenmann B,Lee I, Schweehheimer K, Dickhorn-Dwoniczak B, Radasevich JA. Synaptophysin expression inneuroendocrine neoplasms as determinedby immunocytochemistry. Am J Path01 1987; 126;243-257. 91. Buffa R, Rindi G, Sessa F, Gini A, Capella C, Jahn R. Synaptophysin immunoreactivity and small clear vesicles in neuroendocrine cells and related tumours. Mol Cell Probes 1988; 2~367-381. 92. Tapia FJ, PolakJM,Barbosa MA, Bloom SR, Marangos PJ, Dermody C. Neuron-specific ; 808-81 enolase is produced by neuroendocrine tumours. Lancet 198 1 1. 93. Schroder S , Schwarz W, Rehpenning W, LoningT, Bocker W. Dendriticflangerhans cells and prognosis in patients with papillary thyroid carcinomas: immunocytochemical study J Clin Path01 1988; 89:295-300. of 106 thyroid neoplasms correlated to follow-up Am data. 94. Schroder S , Bay V, Dumke K, KremensB, Muller-Gartner H-W, Bocker W, Kastendieck H. Diffuse sclerosing variantof papillary thyroid carcinoma, S-100 protein immunocytochemistry and prognosis. Virchows Archiv A Path01 Anat 1990; 416;367-371. 95. Romano MI, Grattone M, Karner M P , Moiguer S, Tetelbaum F, Romano LA, IllescasE, Padin R, Cueva F, Burdman JA. Relationship between the level of c-myc mRNA and histologic aggressiveness in thyroid tumors. Horm Res 1993;39:161-165. 96. Basolo F, Pinchera A, Fugazzola L, Fontanini G, Elisei R, Romei C, Pacini F. Expression of p21rasprotein as aprognosticfactorinpapillarythyroidcancer.EurJCancer 1994; 30Az171-174. 97. Goretzki PE, Lyons J, Stac-Phipps S, Rosenau W, Demeure M, Clark OH. Mutational activation of RAS and GSP oncogenes in differentiated thyroid cancer and their biological implications. World J Surg 1992; 16:576-582. 98. Shimizu T, Usuda N, Yamanda T, Sugenoya A, Iida F. proliferative activity of human thyroid tumors evaluated by proliferating cell nuclear antigenlcyclin immunohistochemic studies. Cancer 1993; 1:2807-2812. 99. Mizukami Y, Nonomura A, HashimotoT, Michigishi T, Noguchi M, Matsubara F, Yanaihara N. Immunohistochemical demonstration of epidermal growth factor and c-myc oncogeneproductinnormal,benignandmalignantthyroidtissues.Histopathology1991; 18:ll-18. 100. Mizukami Y, Nonomura A, Michigishi T, Noguchi M, Makamura S , Hashimoto T. Tumor proliferation-related markers in papillary thyroid carcinomas: correlation with histologic subtypes. Anticancer Res 1993; 13:267-271. 101. Schroder S, Schwarz W, Rehpenning W, LoningTH, Bocker W. Prognostic significance of Leu-M1 immunostaining in papillary carcinomas of the thyroid gland. Virchows Arch A 1987; 411:435-439.
484
McDermott
102. Verhagen JN,Van der Heijden MCM, Rijksen G, der Kinderen PJ, Van Unnik JAM, Staal
GET. Determination and characterization of hexokinase in thyroid cancer and benign neoplasms. Cancer 1985; 55:1519-1524. 103. Lee TK, Myers RT, MarshallRB, Bond G, Kardon B. The significance of mitotic rate: a retrospective study of 127 thyroid carcinomas. Hum Path01 1985; 16:1042-1046. 104. Saad MF, Ordonez NG, Guido Samaan JJ, NA. The prognostic value of calcitonin immuno1984; 59:850-856. staining in medullary carcinoma of the thyroid. J Clin Endocrinol Metab 105. Viale G, Roncalli M, Grimelius L, Graziani D, Wilander E, JohanssonH, Bergholm U, Coggi G. Prognostic value of bcl-2 immunoreactivity in medullary thyroid carcinoma. Hum Pathol 1995; 26:945-950. 106. SchroderS, Schwarz W, Rehpenning W, DralleH, Bay V, Bocker W. Leu-Ml immunoreactivity and prognosis in medullary carcinomas of the thyroid gland. J Cancer Res Clin Oncol 1988; 114291-296. 107. Roncalli M, Viale G, Grimelius L, Johansson H, Wilander E, Alfano RM, Springall D, Battezzati PM, Pol& JM, CoggiG.PrognosticvalueofN-mycimmunoreactivityin medullary thyroid carcinoma. Cancer 1994; 74:134-141. 108. Weinberg RA. How cancer arises. Sci Am 1996; 275:62-70. 109. Cline MJ, Slamom DJ, Lipsick JS. Oncogenes: implications for the diagnosis and treatm of cancer. Ann Intern Med 1984; 101:223-233. 110. Gordon H. Oncogenes. Mayo Clin Proc 1985; 60:697-713. 11 1. Druker BJ, Mamon HJ, Roberts TM. Oncogenes, growth factors, and signal transduction. N Engl J Med 1989; 321:1383-1391. 112. Krontiris TG. Molecular medicine: oncogenes. N Engl J Med1995; 333:303-306. 113. Latchman DS.Transcription-factor mutations and disease. N Engl J1995; Med334:28-33. 114. Friend SH, Dryja TP, Weinberg RA. Oncogenes and tumor-suppressing genes. N Engl J
Med 1988; 3183618422. 115. Weinberg RA. Tumor suppressor genes. Science1991; 2541138-1145. 116. Marshall CF. Tumor suppressor genes. Cell 1991; W313-326. USA90: 10914117. Knudson AG. Antioncogenes and human cancer. Proc Natl Acad Sci1993; 10921. 118. Hartwell LH, Kastan MB. Cell cycle control and cancer. Science1994; 266:1821-1828. of cancer. Trends Genet 1993; 9:138-141. 119. Vogelstein B, KinzlerKW. The multistep nature 120. Bishop JM. Cancer: the rise of the genetic paradigm. Genes Develop 1995; 9:1309-1315. 121. Melmed S. Oncogenes and the thyroid. Thyroid Today1988; 11:l-7. 122. Frauman AG, Moses AC. Oncogenes and growth factors in thyroid carcinogenesis. E no1 Metab Clin North Am 1990; 19:479-492. 123. Fagin JA. Genetic basis of endocrine disease 3: molecular defects in thyroid gland neopla J Clin Endocrinol Metab1992; 5:1398-1400. 124. Fagin JA. Molecular pathogenesis of human thyroid neoplasms. Thyroid Today 1994; 17:14. 125. Farid N R , Shi Y, Zou M.Molecular basisof thyroid cancer. Endocr Rev 1994; 15:202-232. 126. Farid N R , Zou M, Shi Y. Genetics of follicular thyroid cancer. Endocrinol Metab Clin North Am 1995; M.865-883. 127. Williams ED. Mechanisms and pathogenesis of thyroid canceranimals in and man. Mutat Res 1995; 333:123-129. 128. Vassart G, Dumont JE. The thyrotropin receptor and the regulation of thyrocyte functi and growth. Endocrine Rev 1992; 13:596-611. 129. Egg0 MC, Pratt MAC, Becks G, Burrow GN. Regulationof growth and of differentiation in thyroid follicular cells. Adv ExpMed Biol 1990; 261:327-340. 130. Goretzki PE, Frilling A, Simon D, Roeher HD. Growth regulationof normal thyroids and thyroid tumors in man. Recent Results Cancer Res 1990; 118:48-54. 131. Derwahl M, Broecker M, MeyerK, Schatz H, Studer H. Molecular mechanisminvolved
DNA Ploidy, Tumor Markers,
Cuncer-Causing and
Genes
485
in thyroid tumorigenesis: possible role of the human TSH receptor. Acta Endocrinol (Copenh) 1992; 126(suppl4):50. OH. Biphasic effectsof thyrotropin 132. HoeltingT, Tezelman S , SipersteinA E , Duh Q-Y, Clark oninvasionandgrowth of papillaryandfollicularthyroidcancerinvitro.Thyroid 1995; 535-40. 133. Cuello C, Correa P, Eisenberg H. Geographic pathology of thyroid carcinoma. Cancer 1969; 233230-238. 134. Dumont JE, Maenhaut C, Pirson I, Baptist M, Roger PP. Growth factors controlling the thyroid gland. Balliere Clin Endocrinol Metab 1991; 5727-754. 135. van Der Laan BFAM, Freeman JL, Asa SL. Expression of growth factors and growth factor receptors in normal and tumorous thyroid tissue. Thyroid 1995; 567-73. 136. Egg0 M, Bachrach LK, Burrow GN. Interaction of TSH, insulin and insulin-like growth factors in regulating thyroid growth and function. Growth Factors 1990; 2:99-109. 137. TramontanoD, Cushing GW, Moses AC, Ingbar SH. Insulin-like growth factor-1 stimulate the growth of rat thyroid cells in culture and synergizes the stimulator of DNA synthesis induced by TSH and Graves’-IgG. Endocrinology 1986; 119: 940-945. 138. Bachrach LK, Egg0 MC, Hintz RL, Burrow CN. Insulin-like growth factors in sheep thyroid cells: Action, receptors and production. Biochem Biophys Res Commun 1988; 154:861-867. 139. Minuto F, Barreca A, Del Monte P,Cariola G, Torre GC, Giordano G. Immunoreactive insulin-like growth factor1 (IGF-1) and IGF-1 binding protein content in human thyroid tissue. J Clin Endocrinol Metab 1989; 68:621-626. 140. Williams DW, Williams ED, Wynford-Thomas D. Evidence for autocrine production of IGF-1 in human thyroid adenomas. Mol Cell Endocrinol 1989; 61:139-143. 141.VannelliGB,Barni T, ModiglianiU,etal.Insulin-likegrowthfactor-1receptorsin nonfunctioning thyroid nodules.J Clin Endocrinol Metab 1990; 71:1175-1182. 142. Maciel RMB, Moses AC, Villone G, Tramontano D, Ingbar SH. Demonstration of the production and physiological role of insulin-like growth factor 2 in rat thyroid follicular cells in culture.J Clin Invest 1988; 82:1546-1553. 143. Westermark K, Karlsson FA, Westermark B. Epidermal growth factor modulates thyroid growth and function in culture. Endocrinology 1983; 112:1680-1686. 144. Waters MJ, Tweedale RC, Whip TA, Shaw G, ManleySW, Bourke JR. Dedifferentiation of cultured thyroid cells by epidermal growth factor: some insights into the mechanism. Mol Cell Endocrinol 1987; 49: 109-1 17. AE, Miller R 4 , Sancho JJ, Demeure MJ, Clark OH. Epidermal growth 145. Duh QY, Siperstein factor receptors in normal and neoplastic thyroid tissue. Surgery 1985; 98:1000-1007. 146. Kanamori A, Abe Y, Yajima Y, Manabe Y, It0 K. Epidermal growth factor receptors in J Clin Endocrinol Metab plasma membranes of normal and diseased human thyroid glands. 1989; 68~899-903. 147. Myamoto M, Sugawa H, Mori T, Hase K, Kuma K, Imura H. Epidermal growth factor receptors in cultured neoplastic human thyroid cells andofeffects epidermal growth factor and thyroid stimulating hormone on their growth. Cancer Res 1988; 48:3652-3656. A, Kobayashi S, Kasuge Y,Iida F. Epidermal growth factor receptor 148. Masuda H, Sugenoya on human thyroid neoplasms. World J Surg 1988; 12:616-622. 149. HoeltingT, SipersteinAE, Clark OH, Duh Q-Y. Epidermal growth factor enhances proliferation, migration and invasion of follicular and papillary thyroid cancer in vitro and in vivo. J Clin Endocrinol Metab 1994; 79:401408. L, et al. Aberrant expression of recepters for 150. Heldin N-E, Gustavsson G, Claesson-Welsh platelet-derived growth factor in an anaplastic thyroid carcinoma cell line. Proc Natl Acad Sci USA 1988; 859302-9306. 151. Matsuo K, Tang S , Sharifi B, Rubin SA, Schreck R, FaginJA. Growth factor production byhumanthyroidcarcinomacells:abundantexpressionof aplatelet-derivedgrowth
486
McDmott factor-B-like proteinby a human papillary carcinoma cell line. J Clin Endocrinol Metab
1993; 77:996-1004. 152. Emoto M, Isozaki 0, Arai M, Murakami H, et al. Identification and characterization of basic fibroblast growth factorin porcine thyroids. Endocrinology 1991; 12858-64. 153. Logan A, Black EG, Gonzalez A-M, Buscaglia M, Sheppard MC. Basic fibroblast gro 1992; 130:2363factor: an autocrine mitogen of rat thyroid follicular cells. Endocrinology 2372. 154. Taylor A H , Millatt LJ, Whitley GStJ, Johnstone A P , Nussey SS. The effect of basic fibroblast growth factor on the growth and function of human thyrocytes. J Endocrinol 1993;136~339-344. 155. Egg0 MC,Hopkins JM,Franklyn JA, JohnsonGD, Sanders SA, Sheppard MC. Expression 1995; 80:1006-1011. of fibroblast growth factors in thyroid cancer. J Clin Endocrinol Metab 156. Tsushima T, Arai M, Saji M, Ohba Y,Murakami H, Ohmura E. Effects of transforming growth factor-beta on deoxyribonucleic acid synthesis and iodine metabolism in porcin thyroid cells in culture. Endocrinology1988; 123:1187-1194. 157. Coletta G, Cirafici AM, DiCarlo A. Dual effects of transforming growth factor beta on of thyroidrat thyroid cells: Inhibition of thyrotropin-induced proliferation and reduction specific differentiation markers. Cancer Res1989; 49:3457-3462. 158. Grubeck-Loebenstein B, Buchan G, Sadeghi R, Kissenerghis M, Londei M, Turner M, e al. Transforming growth factorbeta regulated thyroid growth: rolein the pathogenesisof nontoxic goiter. J Clin Invest 1989; 83:764-770. 159. Holting T, Zieke A, Siperstein AE, Clark OH, Duh Q-Y.Transforming growth factor growth, migration, beta 1is a negative regulator for differentiated thyroid cancer:ofstudies invasion and adhesion of cultured follicular and papillary thyroid cancer cell lines. J C Endocrinol Metab 1994; 79:806-813. WW, Ingbar SH.Interleukin-1 stimulates thyroid cell grow 160. Mine M, Tramontano D, Chin and increases the concentration of the c-myc proto-oncogene mRNA in thyroid follicular cells in culture. Endocrinology 1987; 120:1212-1214. 161. Zakarija M,McKenzie JM. Influence of cytokines on growth and differentiated function of FRTL5 cells. Endocrinology 1989; 1251260-1265. 162. Kawabe Y, Eguchi K, Shimomura C, Mine M, Otsabo T, Ueki Y,et al. Interleukin-1 production and action in thyroid tissue. J Clin Endocrinol Metab 1989; 68:1174-1183. 163. Inoue H,OshimoK, Miki H, Kawano M, Monden Y. Immunohistochemical study of estrogen receptors and the responsiveness to estrogen in papillary thyroid carcinoma. Cancer 1993; 72:1364-1368. 164. Hoelting T, Siperstein A E , Duh Q-Y,Clark OH.Tamoxifen inhibits growth, migration and invasion of human follicular and papillary thyroid cancer cells in vitro and in vivo. J Clin Endocrinol Metab1995; 80:308-313. 165. Pemberton JD, Black BM,The association of carcinoma of the thyroid gland and exopth mic goiter. Surg Clin NA 1948; 28:935-952. 166. Farbota LM, Calandra DB, Lawrence AM, Paloyan E. Thyroid carcinoma in Graves’ disease. Surgery 1985; 98:1148-1152. 167. Filetti S, Belfiore A, Amir SM, Daniels GH, Ippolito 0, Vigneri R, Ingbar SH. The role N Engl of thyroid-stimulating antibodies of Graves’ disease in differentiated thyroid cancer. J Med 1988; 318:753-759. e l l i S , Macchia E, Concehi R, et al. Thyroid carcinoma 168. Pacini F, Elisei R, DiCoscio bGC, in thyrotoxic patients treatedby surgery. J Endocrinol Invest 1988; 11:107-112. 169. Ozaki 0,It0 K, Kobayashi K, et al. Thyroid carcinoma in Graves’ disease. WorldJ Surg 1990; 1443741. 170. Belfiore A, Garofalo MR, Giuffrida D, Runello F, Filleti S, Finmane A, et al. Increased aggressivenessof thyroid cancerin patients with Graves’ disease. J Clin Endocrinol 1990; 70:830-835.
DNA Ploidy, Tumor Markers, and Cancer-Causing Genes
487
171. Hales IB, McElduff A, Crummer P, Clifton-Bligh P, Delbridge L, Hoschl R, et al. Does
Graves’ disease or thyrotoxicosis affect the prognosis of thyroid cancer. J Clin Endocrinol Metab 1992;75:886-889. 172. HuberGK,SafirsteinR,NeufeldD,Davies TF. Thyrotropinreceptorautoantibodies induce human thyroid cell growth and c-fos activation. J Clin Endocrinol Metab 1991; 72: 1142-1 147. J, Parma J, Vassart G. Identification and functional 173. Paschke R, Tonacchera M, Van Sande
characterizationof two new somaticmutationscausingconstitutiveactivation of the thyrotropin receptor in hyperfunctioning autonomous adenomas of the thyroid. J Clin Endocrinol Metab 1994; 79:1785-1789. 174. Porcellini A, Ciullo I, Laviola L, Amabile G, FenziG, Awedimento VE.Novel mutations of thyrotropin receptor gene in thyroid hyperfunctioning adenomas: rapid identification by fine needle aspiration biopsy. J Clin Endocrinol Metab1994; 79:657-661. et al. 175. Russo D, Arturi F, Wicker R, Chazenbalk GD, Schlumberger M, DuVillard JA, Genetic alterations in thyroid hyperfunctioning adenomas. J Clin Endocrinol Metab 1995;
80~1347-1351. 176. Ohno M, Endo T, Ohta K, Gunji K, Onaya T. Point mutations in the thyrotropin receptor in human thyroid tumors. Thyroid 1995; 5:97-100. 177. Parma J, Duprez L, Van Sande J, Paschke R, Tonacchera M, Dumont J, et al. Constitutively Mol CellEndocrinology 1994; activereceptors as adisease-causingmechanism. 100:159-162. 178. Van Sande J, Parma J, Tonacchera M, Swillens S , Dumont J, Vassart G. Somatic and
J Clin Endocrinol Metab germline mutations of the TSH receptor gene in thyroid diseases.
1995;80:2577-2585. 179. Takeshita A, Nagayama Y, Yokoyama N, Ishikawa N, It0 K, Yamashita T, et
al. Rarity of oncogenic mutations in the thyrotropin receptorof autonomously functioning thyroid nodules in Japan. J Clin Endocrinol Metab 1995; 80:2607-2611. is not an 180. Matsuo K, Friedan E, Gejman PV, Fagin JA. The thyrotropin receptor (TSH-R) oncogene for thyroid tumors: structural studies of the TSH-R and the alpha subunit of Gs in human thyroid neoplasms. J Clin Endocrinol Metab 1993; 76:1446-1451. 181. Esapa C, FosterS, Johnson S , Jameson JL, Kendall-Taylor P, Harris PE, etal. G protein and thyrotropin receptor mutations in thyroid neoplasia. J Clin Endocrinol Metab 1997; 82:493496. 182. Lyons J, Landis CA, Harsh G, Vallar L, Grunewald K, Feichtinger H,et al. Two G protein oncogenes in human endocrine tumors. Science 1990; 249:655-658. 183. Dumont JE. Thyroid adenoma, Gsa expression and the cyclic adenosine monophosphate 1995; mitogenic cascade: a complex relationship. [Editorial]. J Clin Endocrinol Metab 80:1518-1520. 184. Spalmbarg D, Sharifi N, Elisei R,Gross JL,, Medeiros-Netog G, Fagin JA, et al. Structural
studies of the thyrotropin receptor and Gsa in human thyroid cancers: low prevalence of mutations predicts infrequent involvement in malignant transformation. J Clin Endocrinol Metab 1996;81:3898-3901. 185. Lemoine N R , Mayall ES, Wyllie FS, et al. Activated ras oncogenes in human thyroid cancers. Cancer Res 1988; 48:4459-4463. 186. Suarez HG, Du Villard JA, Calliou B, et al. Detection of activated ras human thyroid carcinomas. Oncogene 1988; 2:403406. 187. Ezzat S , Zheng L, Kolenda J, Grigorian A,Freeman JL,Asa SL. Prevalence of activating ras 1996; 5:409-416. mutations in morphologically characterized thyroid nodules. Thyroid 188. Tmpp M, Arenas E, Fainzilber M, Nilsson AS, Sieber BA, Grigoriou M, et al. Functional receptor for GDNF encoded by the c-ret proto-oncogene. Nature 1996; 381:785-789. p, et 189. Durbec P, Marcos-Gutierrez CV, Kilkenny C, Grigoriu M, Warti0waa-a K, Suvanto 1996; 381:789-793. al. GDNF signalling through the ret receptor tyrosine kinase. Nature
488
McDmott
190. Jing S, Wen D, Yu Y,Holst PL, Luo Y,Fang M, et al. GDNF-induced activation of the ret protein tyrosine kinase is mediated by GDNF-a,a novel receptor for GDNF. Cell 1996; 85: 11 13-1 124. et al. Character191. Treanor JJ, Goodman L, de SauvageF, Stone DM, PoulsenKT, Beek 0, ization of a multicomponent receptor for GDNF. Nature1996; 382:80-83. 192. Hofstra R M , LandsvaterRM, CeccheriniI, Stulp Rp,StelwagenT, Luo Y,et al. A mutation 2B and in the RET protooncogene associated with multiple endocrine neoplasia type sporadic medullary thyroid carcinoma. Nature 1994; 367:375-376. RET proto-oncogene in the multiple 193. Smith DP, Eng C, Ponder BAJ. Mutations of the J CellSci 1994; endocrineneoplasia type 2 syndromesandHirschsprungdisease. 18(Suppl):43-49. 194. Lips CJ,Landsvater RM, Hoppener J w , Geerdink RA, Blijham G, van Veen JM, et al. Clinical screening as compared with DNA analysis in families with multiple endocrine neoplasia type 2A. N Engl J Med 1994; 31:828-835. 195. Zedenius J, Larsson C, Bergholm U, Bovee J, Svensson A, Hallengren B, Grimelius L, Backdahl M, Weber G, Wallin G. Mutations of codon 918 in the RET proto-oncogene correlate to poor prognosisin sporadic medullary thyroid carcinomas. J Clin Endocrinol Metabol 1995; 80:3088-3090. 196. Romei C, Elisei R, Pinchera A, CeccheriniI, Molinaro E, Mancusi F, Martino E, Romeo G, Pacini F. Somatic mutations of the ret protooncogene in sporadic medullary thyroid carcinoma are not restricted to exon16 and are associated with tumor recurrence.J Clin Endocrinol Metab 1996; 81:1619-1622. 197. Jhiang SM, FithianL, Weghorst CM, Clark OH,Falko JM, O’Dorisio T, Mazzaferri EL. RET mutation screening in MEN2 patients and discovery of a novel mutation in a sporad medullary thyroid carcinoma. Thyroid 1996; 6115-121. 198. Wohllk N, Cote G, Bugalho MM, Ordonez N, Evans DB, Goepfert H, et al. Relevance of ret proto-oncogene mutations in sporadic medullary thyroid carcinoma. JClin Endocrinol Metab 1996; 81:3740-3745. 199. Frank-Raue K, Hoppner W, Frilling A, KotzerkeJ, Dralle H,Haase R, Mann K, Seif F, Kirchner R, Rendl J,Deckart HF, Ritter MM, Hampel R, Klempa J,Scholz GH, Raue F, and the German Medullary Thyroid Cancer Study Group. Mutations of the ret protoon German multiple endocrine neoplasia families: relation between genotype and pheno J Clin Endocrinol Metab 1996; 81:1780-1783. Rate M, Nunziata V, et al. 200. Quadro L, Panariello L, Salvatore D, Carlomagno F, Del Frequent RET protooncogene mutations in multiple endocrine neoplasia type 2A. J Clin Endocrinol Metab 1994; 79590-594. 201. Mulligan LM, Ponder BAJ. Genetic basis of endocrine disease: multiple endocrine n sia type 2. J Clin Endocrinol Metab 1995; 80:1989-1995. type 2 and Hirschsprung’s 202. Eng C. Theret proto-oncogene in multiple endocrine neoplasia disease. N Engl J Med 1996; 335943-951. 203. Grieco M, Santoro M, Berlingieri MT, Melillo RM, Donghi R, Bongarzone I, Pierotti MA, Della-Porta G, Fusco A, Vecchio G.PTC is a novel rearrangedform of the ret protoin vivo in human thyroid papillary carcinomas. Cell oncogene and is frequently detected 1990; 60557-563. 204. Bongarzone I, Fugazzola L, Vigneri P, Mariani L, Mondellini P, Pacini F, Bas010 F, Pinchera A,Pilotti S, Pierotti MA. Age-related activation of the tyrosine kinase receptor protooncogenes ret and ntrkl in papillary thyroid carcinoma. J Clin Endocrinol Metab 1996; 81:2006-2009. 205. Sugg S, Zheng L, RosenIB, Freeman JL, Ezzat S, Asa SL.ret/PTC-1, -2, and -3oncogene JClin rearrangements in human thyroid carcinomas: implications for metastatic potential? Endocrinol Metab 1996; 81:3360-3365.
DNA Ploidy, Tumor Markers,Cancer-Causing and
Genes
489
206. Di Renzo MF, Oliver0 M,Fern S , Prat M, BongamoneI, Pilotti S , et al. Overexpression of the c-METMGF receptor gene in human thyroid carcinomas. Oncogene 1992; 7:2549 2553. 207. Belfiore A, Gamgemi P, Santomocito MG, LaRosa GL, Constantino A, Fiumara A, et al. Prognostic value of c-MET expression in papillary thyroid carcinoma. Thyroid 1995; ~ ( S U P1):S-133. P~ 208. Kelly K, Siebenlist U. The role of c-myc in the proliferation of normal and neoplastic cells. J Clin Immunol 1985; 565-77. 209. Del Senno L, GambariR, Degli Uberti E, Barbieri R, Bernardi F, Buzzuni D, etal. C-myc oncogene alterationsin human thyroidcarcinomas. Cancer Detect Prevent 1987; 10: 159 M, Caillou B, Travagli JP, et al. Structure 210. Temer P, Sheng 2-M, Schlumberger M, Tubiana and expression of c-myc and c-fos proto-oncogenes in human thyroid carcinomas. Br J Cancer 1988; 57:43-47. ED, Wynford-Thomas D. N-myc expression in neoplasia 211. Boultwood J, Wyllie FS, Williams of human thyroid C-cells. Cancer Res 1988; 48:40734077. 212. Lane DP. p53, guardian of the genome. Nature 1992; 358:15-16. 213. Marx J. How p53 suppresses cell growth. Science 1993; 262:1644-1645. 214. Frebourg T. Cancer risks from germline p53 mutations. J Clin Invest 1992; 90:1637-1641 215. Harris CC.Medicalprogress: clinical implications of the p53 tumor suppressor gene. N Engl J Med 1993; 329:1318-1327. 216. Greenblatt MS, Bennett WP, Hollstein M, Harris CC. Mutations in the p53 tumor suppressor gene: clues to cancer aetiology and molecular pathogenesis. Cancer Res 1994; 54:48554878. 217. Fogelfeld L, Bauer TK, Schneider AB, Swartz JE, Zitman R. p53 gene mutations in radiation-induced thyroid cancer. J Clin Endocrinol Metab 1996;81:3039-3044. the regulation ofprogrammed cell death. J Cell Biol 1994; 124 218. Reed JC. Bcl-2 and 1-6. 219. Branet J, Brousset P, Krajewski S, Schlaifer D, Selves J, Reed JC, Caron P. Expression of the cell death-inducing gene bax in carcinomas developed from the follicular cells of the thyroid gland. J Clin Endocrinol Metab 1996; 81:2726-2730. of cell division. 220. Doree M, Galas S. The cyclin-dependent protein kinases and the control FASEB J 1994; 8:1114-1121. 221. Morgan DO. Principles of CDK regulation. Nature 1995; 374:131-134. 222. Hartwell L. Defects in the cell cycle checkpoint may be responsiblefor genomic instability of cancer cells. Cell 1992; 71543-546. 223. KambA, Gravis S , Weaver-FeldheusJ, Liu Q,Hmhman K, Tavtigian SV, et al. A cell cycl regulator potentially involved in genesis of many tumor types. Science 1994; 2a436-440. 224. Cyrns V L , Thor A, Xu H-J, Hu SX,Wiernian ME, Vickery AL, Jr, et al. Loss of the retinoblastoma tumor-suppressor gene in parathyroid carcinoma. N Engl J Med 1994; 330:757-761. 225. Haber DA. Telomeres, cancer and immortality. N Engl J Med 1995; 332:955-956. 226. Greider CW, Blackburn EH. Telomeres, telomerase and cancer. Am Sci 1996; 274:92-97. 227. Haugen BR, Nawaz S , Markham N, Hashizume T, Shroyer KR. Telomerase activity in benign and malignant thyroid tumors. Paper presented at the 69th Annual Meeting of the American Thyroid Association 1996, San Diego, CA, Nov 14-17. 228. Folkman K. Fighting cancer by attacking its blood supply. SciAm 1996; 275150-154. 229. Cuevas P, Gonzalez A-M, Carceller F, Baird A. Vascular response to basic fibroblast growth factor when infused onto normal adventitia or into the injured media of the rat carotid artery. Circ Res 1991; 693360-369. 230. Bernstein LR, Liotta LA. Molecular mediators of interactions with extracellular matrix components in metastasis and angiogenesis. Curr Opin Oncol 1994; 6:106-113.
McDermott
231. Ruoslahti E, Reed JC. Anchorage dependence, integrins and apoptosis. Cell 1994;77: 477-478. 232. Akiyama SK, Olden K, Yamada K M .Fibronectin and integrins in invasion and metastas Cancer Metast Rev 1996; 14:173-189. 233. Ruoslahti E. How Cancer Spreads. Sci A m 1996; 27572-77. Am 234. Oliff A, Gibbs JB, McCormick F. New molecular targets for cancer therapy. Sci 1996; 275144-149. 235. Old W.Immunotherapy for cancer. Sci A m 1996; 275136143. 236. BassiV,Vitale M, Feliciello A, de Riu S, RossiG,FenziG.Retinoicacidinduces
intercellular adhesion molecule-1 hyperexpression in human thyroid carcinoma cell lines J Clii Endocrinol Metab 1995; 80: 1129-1 135.
51 New Approaches to Chemotherapy for Thyroid Cancer Lawrence S. Lessin and My0 Min BIOLOGICAL RESPONSE MODIFIERS Differentiated (Papillary, Follicular, and Mixed) Thyroid Carcinoma In 1986, McConahey and colleagues( I ) reported their observation that patients with
thyroid cancer who had concomitant lymphocytic thyroiditis at time of surgery were found to have better prognosis. Attempts at inducing autoimmune thyroiditis were m by Amino and colleagues (2) and LoGerfo and colleagues (3) by immunization of patients with saline homogenates of thyroid cancer and chemically altered thyroglobulin, minirespectively; poor inflammatory response was obtained and tumor response was mal. Mouse monoclonal antibodies have been produced against human thyroglobulin and membrane proteins such as TSH receptor and thyroid peroxidase expressed on cells of both normal thyroid follicles and differentiated cancers. These antibodies may be potentially employedas humoral immunotherapy in patients where minimal residual disease remains after primary treatment, including tumor cells which remain afterI3’I therapy. Another approach involves induction of “redifferentiation” of differentiated 13’1 refractory thyroid cancers with 13-cis-retinoic acid. Oral therapy with 13-cis-retino acid (Accutane), at a dose of 1.5 mgkg per day,in 10 patients with advanced, differentiated tumors, induced renewed uptake of radioiodine in4 of 10 patients (4).
Anaplastic Thyroid Carcinoma Anaplastic thyroid carcinomas do not concentrate iodine and thus, unlike well diff entiated thyroid carcinomas, radioactive iodine is not used to treat systemic disease. Response rates to chemotherapy are limited. Immunotherapy therefore was viewed as a theoretical alternative for treatment of systemic disease. However, these same tumors are least likely to express thyroid-specific antigens as such TSH-receptor, thyroglobulin and thyroid peroxidase which could be used as target antigens for immunotherapy with directed monoclonal antibodies. Lissoni and associates (3) reported use of interleukin-2 subcutaneously along with melatonin in three patients with anaplastic thyroid carcinoma months with survival of 14 months. achieving stable disease in one patient, lasting4 for As recently reviewed by Ain (6), investigators have been searching for specific mutational proteins which might provide the key to the understanding of resistance to From: Thyroid Cancer: A Comprehensive Guide to Clinical Management Edited by: L. Wartofsky 0 Humana Press k c . , T o t m , NJ
491
492
Lessin and Min
chemotherapeutic agents. Expression of various gene products has been identified in anaplastic thyroid cancer suchas P-glycoprotein (MDR-l), major vault protein (LRP), DNA topoisomerase11-a(TOP-II), and multidrug resistance-associated protein ("-1) (7,8). The membrane proteins,MDR-1 and MRP-1, are believed to be associated with (9). chemotherapyresistanceduetotheirabilitytoexcludedrugsfromcellentry Individual anaplastic cancers have been found to variably express several of these different gene products, andthis ability for drug transport or extrusion differs with th (8). As suggested by different gene products for the different chemotherapeutic agents Ain (6), inactivationoftheseproteinscouldrestoreorpromotechemotherapeutic effectiveness.
Medulla y Thyroid Carcinoma Combination chemotherapy has been employed for medullarycarcinomaofthe thyroid (10). Immunotherapy using low-dose interferon-alfa (3 million units 3 times two patientswithmedullarythyroidcarcinoma,with perweek)wasemployedin improvement of diarrhea and reduction of calcitonin level; similar to octreotide, no reduction in tumor size was seen (11). Interleukin-2 generates antitumor cytotoxicity by the activation of killer T- lymphocytes. It has been tested in patients with melanoma in 7-15% of patients. The same cytokin and renalcell carcinoma with clinical responses was used to treat a patient with metastatic medullary thyroid carcinoma without cli improvement. Lissoni and coworkers (5) recently reported use of interleukin-2 and melatonin in one patient with medullary thyroid carcinoma; stable disease was noted 15 months. Recently, l3II-labeled anti-CEA murin for 6 months and the patient survived monoclonal antibodies (MN-14 immunoglobulin G and h4N- 14 Fab fragment) have I trials are underway, accruing patients with become available for clinical study. Phase metastatic medullary thyroid carcinoma and elevated CEA levels, relapsed after s of140-267 andlor chemotherapy (12-14). Fourteen patients, treated at dose levels cGy, experienced myelosuppressionas the major dose-limiting toxicity. Seven patient 11 of 12 radiologically had a mean 55% transient decrease in tumor markers and evaluable patients had stable disease ranging 3-26 months with a median of twelve months (13). A current Phase I/II trial utilizes high doses of the same 1311-anti-CEA antibody followed by autologous bone marrow (stem cell) rescue in advanced MCT; response data are not yet available (14).
CHEMOTHERAPY TRIALS ShcIies With Paclitaxel flaxol)
Based on the antineoplastic activity of paclitaxel against human anaplastic thyroid (15) several Phases I and 11 carcinoma (ATC) cell lines and nude mice xenografts clinical trials are now studying the efficacy of paclitaxel, both as single agent and in combination with other drugsand/or radiation therapy, in patients with anaplastic t carcinoma. One multiinstitutional trial is evaluating paclitaxel given as a prolonged, continuous 96-hour infusion in ATC patients without prior chemotherapy (16).Two other phase I studies are assessing feasibility of combining either paclitaxel (17) or paclitaxel with carboplatin (18) along with concurrent radiation therapy in advanced ATC and other head and neck tumors.
New Approaches to Chemotherapy
493
REFERENCES
1. McConahey WM, Hay ID, Woolner LB, Van Heerden JA, Taylor WF. Papillary thyroid cancer treated at the Mayo Clinic, 1946 through 1970: initial manifestations, pathologic findings, therapy, and outcome. Mayo Clinic Proc 1986; 61:978-999. LJ. Immunologic aspects of human thyroid cancer: 2. Amino N, Pysher T, cohen EP, DeGrppt humoraland cell-medicatedimunity and atrial of immunotherapy. Cancer 1975; 36:963-973. 3. hGerf0 PL, Feind C, Weber C, Ting W. Immunotherapy of thyroid cancer by induction of autoimmune thyroiditis. Surgery 1983; 94:959-965. 4. Simon D, KohrleJ, Schmutzler C, Mainz K, Reiners C, Roher HD. Redifferentiation therapy of differentiated thyroid carcinoma with retinoic acid basics and first clinical results. Experimental & Clinical Endocrinology & Diabetes 104 Suppl4:13-5, 1996. 5. Lissoni P, Barni S, Tancini G, Mainini E, Piglia F, Maestroni GJM, Lewinski A. Immunoendocrinetherapywithlow-dosesubcutaneousinterleukin-2plusmelatonininlocally advanced or metastatic endocrine melatonin in locally advanced or metastatic endocrine tumors. Oncology 1995; 52:153-166. 6.Ain KB. Anaplasticthyroidcarcinoma:behavior,biology,andtherapeuticapproaches. Thyroid 1998; 8:715-726. 7. Satake S, Sugawara I, Watanabe M, Takami H. Lack of a point mutation of human DNA topisomerase II in multidrug-resistant anaplastic thyroid carcinoma cell lines. Cancer Lett 1997;116:33-39. 8. h e DW, Deeley RG, Cole SPC. Biology of the multi-drug resistance-associated protein, MRP. Eur J Cancer 1996; 32A945-957. 9. Lehnert M. Clinical multidrug resistancein cancer: a multifactorial problem. Eur J Cancer 1996; 32Az912-920. 10. Scherubl H, Rane F, Ziebler R. Combination chemotherapy of advanced medullary and differentiated thyroid cancer. J Cancer Res Clin Oncol 1990; 116:21-23. 11. Grohn P, Kumpulainen E, Jakobsson M. Response of medullary thyroid cancer to low dose alpha Interferon therapy. Acta Oncol 1990; 29:950. I/II therapy trialof 131 I-labelled 12. Juweid M, SharkeyRM, Dunn R, Goldenberg DM. Phase MN-14 F (ab)2 anti-CEA monoclonal antibody (MAB) in patients with metastatic thyroid cancer. hoc Annu Meet Am Soc Clin Oncol 1998; 17:392a. 13. Juweid M, Rubin A. Phase I study of 131 I-labelled MN-14 IgG or 131 I-labelled MN-14 F (ab)2 anti-CEA antibody in advanced cancer. NCLV93-0306 (status: active). I/II studyofhigh-dose131I-labelledanti-CEAantibody 14.*JuweidM,RubinA.Phase fragment MN- 14F (ab)2 followed by autologous bone marrow rescue in advanced med thyroid cancer. Natl Cancer Inst PDQ 1998; 5:284(11034). 15. Ain KB, Tofiq S, Taylor KD. Antineoplastic activity of Tax01 against human anaplastic thyroid carcinoma cell lines in vitro andin vivo. J Clin Endocrinol Metab 1996; 81:365& 3653. 16. *Chandler AB, et al. phase II study of a 96 hour continuous infusion of paclitaxel for anaplastic carcinomaof the thyroid previously untreated with chemotherapy: current clinica trials: oncology. Natl Cancer Inst PDQ 1998; 5:379(11530). 17. *Maniglia k T , et al. Phase I study of radiotherapy and concurrent paclitaxel following resection of locallyadvancedheadandneckcancers.NatlCancerInstPDQ1998; 5:754(13061). 18. *Davis TH, et al. PhaseI study of concurrent paclitaxel and carboplatin with radiotherapy in patients with head and neck cancer. Natl Cancer Inst PDQ 1998; 5:657(12787).
*Available (with updates) on Internet at http://www.cuncernet.ncinih.gov/clinicultrials/
52 Advances in Radiotherapy for Thyroid Cancer R. Larry White External megavoltage irradiation has been utilized alone or in combination with surgery, 1311therapy, interstitial radiation, chemotherapy and hyperthermia to produce cure or improved local-regional control. The optimal sequencing of these effective trials and clinical practice. New methods modalities continues to be evaluated in clinical of time-dose relationships including hyperfractionation, hypofractionation, and highinterstitial irradiation, dose fractions are currently under clinical evaluation. Theofuse in experienced hands, as an addition to standard treatment modalities has produced good results in a limited numberof patients with thyroid carcinoma. Hyperthermiais potentially applicable to all cancers, including thyroid carcinoma. The clinical research is withhyperthermia so farhasdemonstratedthebestresultswhenhyperthermia combined with external irradiation. The area of computer assisted three-dimensional treatment planning has recently allowed radiation oncologists to treat cancers with much higher doses of irradiation while offering better protection to the surrounding tissue and vital structures. These planning techniques offer exciting opportunities to give external irradiation doses that will obliterateany cancer, including thyroid carcinomas, while protecting the surrounding sensitive structures such as the spinal cord. Over the next five years, we will continue to see the technical aspects of external irradiation alone andin combination with other effective treatment modalities improve the chanceof local control and the chance for curing thyroid carcinoma.
From: Thyroid C an= A Comprehensive Guide to CliNcalManagement Edifed by: L. Wartofsky Q Humnn Press Inc., T o t m , Nf
495
Index A
molecular markers of poor differentiation,
ABCC study,see Atomic Bomb Casualty Commission study Acetylcholine, as tumor growth stimulator, 58,61 Adenminoma, see Papillary or follicular carcinoma Adenoma hyalinizing trabecular,437 ras mutation in,65 Adenosine, as tumor growth stimulator,
metastases, 322 p53 mutations, 62,67,333 ,pathology (see also Histopathology),
58,61
Adhesion molecules, cell surface,63, 67,68,475,476
Ad~iamycin,see Ckmo-, doxorubicin Age, incidenceo f solitary nodules and,3 Airway, managementin lymphoma, 354 Anaplastic carcinoma E-cadherin expression in,68 chemotherapy in,334,337,338,341-
140
327-33 1
immunoreactive keratin in,327 immunoreactivevimentin in, 327,328 platelet derived growth factor (PDGF) in, 333 precursor thyroid lesions, 320,32 1,329 prognosis, 321,342,343,345-347 radiation exposure and,87,32 1 radiation therapy, external,341-343 combination with doxorubicin, 337,338,341,343,347
dosage, 343 interstitial irradiation,343 portals, 343 spinal cord dosage,343 survival, 342,343,347 technique, 343 ras mutations in,66 signs and symptoms,322 staging, 346 survival, 321,342,343,345-347 thyroglobulin levels and,323 tracheal compression,322 treatment, chemotherapy, 334 radiation, 341-343 variants, 319 paucicellular, 327 Angiosarcoma (angiomatoid carcinoma),
343,49 1,492
in children, 125 clinical presentation,321,322 demographics, 320 diagnosis, 322,323,333 epidemiology, 79,3 19 fine needle aspiration,334 histopathology giant cell,319 small cell,319,329 patterns, 327 spindle cell, 319,328 squamous, 319,328 imaging, 323 incidence, 79
414
497
498
Antigenicity of thyroid cancer, 137-1 40 PAX-8 activation factor, 138-140 thyroglobulin as antigen, 138, 139 thyroid peroxidase as antigen, 138 thyrotropin receptor as antigen, 138 Apoptosis (programmed cell death and p53), 62, 133,473 Atomic Bomb Casualty Commission study (ABCC study) death rates, 87 follow-up data, 87 occult thyroid cancer incidence in, 87 Azkanazy cell (Hurthle cell) 294
Index
in medullary carcinoma, 39, 50, 367 tumor marker,460462 Carcinogenesis, multiple hit theoryof, 63 CD44, see ICAM’s or cell adhesion molecules CEA, see Carcinoembryonic antigen Cell cycles,455,456,462 maintenance proteins, 475,476 regulatory proteins, 474,475 Cell surface adhesion molecules, 63,67, 68,475,476 Chemokines, 137 Chemotherapy, 179-1 8 1,221-223 B adverse side effects, 222 biologic response modifiers,491,492 B-cell lymphoma, Hashimoto’s thyroiditis bleomycin, 180,339,343,354 and, 80 carboplatin, 180, 18 1 bcl-2 expression, 118, 133, 140,352, cisplatin, 180, 18 1, 343,354,492 412,460-462,474 combination therapy with external Becquerel, 86 radiation, 222,223, Belarus, radiation exposure(see also for anaplastic carcinoma,337,338, Chernobyl) 34 1,343 radiation exposure in children, 101-1 03 for lymphoma, 354 Benign tumors of the thyroid, 437 cyclophosphamide, 180,339,343,354 Bone metastases,see specific cancer dacarbazine, 180,406 Bq (Becquerel), 86 doxorubicin, 179, 180,221,343,354 Bradykinin, as tumor growth stimulator, for anaplastic carcinoma, 334, 337 58,61 for medullary carcinoma,405407 drug resistance,339,406 C chemosensitivity testing, 339 Calcification and MDR-1 or p-glycoprotein in papillary carcinoma, psammomatous, expression 339 see Psammoma bodies etoposide, 180, 181,354,406 Calcitonin, 366, 367,460 5-Fluorouracil, 339,343,406 diagnostic sensitivity, 20 interleukin-2,49 1,492 serum levels lymphonqseeLymphoma,chemod.lerapy in anaplastic carcinoma, 323 melatonin, 491 in medullary carcinoma, 50 methotrexate, 180, 181,343 with nodule, 39 monoclonal antibodies, 492 Calcitonin gene-related peptide, 366, octreotide, 406,407 375,460-462 paclitaxel, 180,347,492 Calmodulin, 60 redifferentiation therapy,254,283,49 1 CAM kinase, 60 single modality therapy, 22 1,222 Carcinoembryonic antigen (CEA) survival curves, 223 in anaplastic carcinoma, 323 vincristine, 180,406
499
Index Chernobyl nuclear accident, 4,65,96-106, 124,433 dose to thyroid,98 geographic distribution,96,98-104 isotopes released, 96-98 cesium-l37,97 iodine isotopes, 97,98 comparison to Three Mile Island, 98 Columnar cell variant papillary carcinoma, 432,433 clinical presentation,432 demographics, 432 treatment, 432,433 Computed tomography (CT), 1 1 in anaplastic carcinoma,334 advantages, 22 donut sign,353 and iodine content,2 1-23 in lymphoma, 353 of nodule, 1 1 principles and method,21 of thyroid masses,1 1 utility in cancer without known residual tumor, 29 in cancer with known residual tumor, 30,3 1 in diagnosis of cancer post-operatively, 28 for sectional images,24 in undiagnosed, suspected cancer, 27,245 vs. magnetic resonance imaging,2 1-23 Cowden’s disease, thyroid cancer and, 49,68,80 CT, see Computed tomography Cytokines effects on thyroid cancer cells, 13 1, 135437,141 tumor-secreting, 142
D Diagnostic imaging,9-34 (see also specific imaging modalities) 240,247 Diet, thyroid cancer and,
Differentiated thyroid carcinoma (DTC), see Follicular carcinoma; Papillary thyroid carcinoma Diffuse sclerosing variant papillary carcinoma, 433 DNA ploidy analysis,441,457459 as tumor marker,40 Doxorubicin, for thyroid carcinoma,see Chemotherapy DTC, see Differentiated thyroid carcinoma
E Eastern Cooperative Oncology Group 22 1, (ECOG), chemotherapy trial, 353 E-cadherin, 68,460-462,476 ECOG (Eastern Cooperative Oncology Group), chemotherapy group,337 EMA (epithelial membrane antigen), 140,460-462 Epidermal growth factor (EGF),460 Epidermal growth factor-receptors, in thyroid cancer, 460 Epithelial membrane antigen (EMA),140 External radiation therapy,see Radiation therapy, external
F Familial adenomatous polyposis of colon, papillary thyroid carcinoma and, 49, 68,80,122,415 Familial medullary thyroid carcinoma, see Medullary thyroid carcinoma, familial Familial multiple endocrine neoplasia, see Multiple endocrine neoplasia FDG (2-deoxy-2-fluoro-~-glucose)-PET, 246,247,257 Fine-needle aspiration (FNA),35-37, 39,40 advantages, 40 with biopsy, see Fine-needle biopsy (FNB 1 in children,268 DNA analysis, 40,254,457-459
500
equipment needed, 35 PCR amplification genetic analysis, 254 of TSH receptor or thyroglobulin, 40 51 radiation-related tumors, utility for, repeat FNA, indications, 43 smear preparation,36 technical hints, 37 technique, 35-37 ultrasound-assisted, 19,20 Fine-needle biopsy (FNB) of lymph nodes, 20,254 Follicular adenoma, ras mutation in, 65, 66 (see also Follicular carcinoma, 279-3 17 Hurthle cell carcinoma) benign vs. malignant,280,281 clinical presentation,279-28 1,297, 298 in children, 15,3 3 16 diagnosis, 3 15 encapsulated, 289,298 and endemic goiter, 280 fine needle aspiration,280,28 1,292295,315 fOllOW-Up, 282,307-3 10 level of surveillance, 307 strategy, 307 thyroglobulin and antithyroglobulin, 308 frozen section in, 298 Hurthle cells, in, 294 Hurthle cell variant,282,283,297299,441-443 incidence, 280 iodine deficiency, and, 280 management principles,283,284,298 MDM2 overexpression in, 67 metastases, 299 capsular involvement,298,299 lymph nodes, 299 minimally invasive,289,290,3 11 moderately invasive, 29 1 widely invasive, 299
Index
mortality, 283,284 pathology, 289-296 PCNA (proliferative cell nuclear anti gen) in, 294,462 poorly differentiated, 292 postoperative management, 299 prognosis, 283,284,297,3 1 1-3 13 AGES scoring system,51,282 AMES scoring system,5 1 influencing factors for,311, 312 MACIS scoring system,5 1 therapy, effect of, 3 12 TNM, 282 radiation-induced tumors, 304 radiation therapy, external,301-304 bone metastases, 303 brain metastasis, 302 combination chemotherapy, 302, 303 effect on survival,302,3 12 interstitial radiation, 304 megavoltage therapy, 495 boosting technique, 303 indications, 302 dosage, 303 radiosensitivity, 302 spinal cord shielding, 303 radioiodine therapy,147-1 5 1,213-2 19 bone metastases, 16-2 2 19 brain metastases, 160, 161,219 doses for, 147, 148,214 for distant metastases, 14-2 2 19 objectives of treatment, 147 practical issues, 15 1 pregnancy tests, 15 1 posttherapy guidelines, 15 1 posttherapy scan, 151 for pulmonary metastases, 14,2 2 15 safety considerations, 150, 15 1 single high-dose method, 147,214 stunning, 148, 167 ras mutation in,65,66,460 staging 282(see ako Follicular carcinoma, Prognosis)
Index surgical treatment, 49-53,297-299 extent of surgery and, 5 1,52,298, 315,316 postoperative radiotherapy and, 299 thyroidectomy for, 298 survival, 312 TSH receptors, 460,462 tumor cellularity in, 294 vs. follicular adenoma,28 1 Follicular variant of papillary carcinoma, see papillary carcinoma, follicular variant) Frozen section,52,53 utility for papillary carcinoma, 10 utility for follicular carcinoma, 298 G
502
activating in hyperfunctioning adenoma (gsp), 61 in thyroid cancer,61,464477, 476 (table) role ofTSH in, 465 repair, 457 replication, 457,458 ret rearrangements ELEl gene, 64 ret/PTC 1, PTC2, PTC3 rearrangement, 64,65,459 trk, 64 somatic mutations, 463 structural organization of, 62 tumor suppressor genes (p53),61,62, 462,463 Glial cell derived neurotrophic factor (GDNF), 64,366,469,470 G protein-coupled receptor kinases (GPKs), 60-62 G-protein-mediated second messenger systems, 60-62 G-proteins, TSH receptor signal transduction and, 57,60,61,66 Gq protein, 61 Granulocyte-macrophagecolony stimulating factor (G-CSF)in anaplastic carcinoma, 323 Graves’ disease, thyroid cancer in, 80 Gray (GY), 86 Grb2 (adapter molecule),58,59 Growth signal transduction,5743,464467 GsP (guanine nucleotide stimulatory protein), 61,66 Gs subunit, 61,66 GTPase activity of ras protein, 58 Guanine nucleotide exchange factor (mSOS), 58 GY (gray), 86
Gardner’s syndrome, thyroid cancer and, 49,68, 80, 122,415 Gene expression regulation,5742,456, 457 Genes basal transcriptional apparatus, 1,62 6 DNA, see DNA ploidy analysis germline mutations, 463 HLA associations with cancer, 130 immune response genes and cancer, 130-1 32 MDM2,62 molecular aberrations in cancer, 6249 cell surface receptors, 63 c-erbB, 63 c-erbB2/neu, 63 PDGF-receptor (platelet derived growth factor), 63 DNA methylation, 63 growth factors, 58,63,465,466 EGF, 58,63,460 IGF-l,58,63,460 TGF-a, 58,63,460 H extracellular signaling proteins, 464 Hashimoto’s thyroiditis, see Thyroiditis, growth signal transduction,463,464 mutations, 463,464477 (see also Hashimoto’s Oncogenes and protooncogenes) hCG, see Human chorionic gonadotrophin
502
Hepatocyte growth factor,64 HGF, see Hepatocyte growth factor HLA associationswith thyroid cancer, 130 Hodgkin’s disease radiotherapy and thyroid cancer, 90, 122 Homer’s syndrome, 6 Human chorionic gonadotrophin,as tumor growth stimulator,58 Hurthle cell carcinoma,282,283,440-443 clinical behavior,297,298,44 1,442 diagnosis, by fine-needle biopsy, 442 follow-~p,442,443 prognosis, 442 and diploid nuclei, 441 radioiodine concentration, 441 surgical treatment,298,299,441,442 thyroglobulin monitoring, 443 Hyalinizing trabecular carcinoma,14,4 4 15 I ICAM’s (intracellular adhesion molecules), 63,67,68,460-462 IGF- 1,see Insulin-like growth factor- 1 Imaging, non-isotopic,9-34 Immune response genes predisposing to cancer, 130-132 failure of immune targeting,132-134 CD8 cytotoxicity, 132-1 34 Immunologic aspects of thyroid cancer, 129-1 42 antigenicity of thyroid cancer, 137-140 PAX-8 activation factor, 138-140 thyroglobulin as antigen, 138, 139 thyroid peroxidaseas antigen, 138 thyrotropin receptor as antigen, 138 mechanisms of reduced immunogenicity, 139 methods to augment immune response, 141,142 Immunotherapy of thyroid cancer, 134, 135,141,142 Incidentalomas, 3,10, 15, 188 Insular carcinoma,see Poorly differentiated carcinoma
Index
Insulin-like growth factor-1 (IGF-1)as tumor growth stimulator,58,63 Interleukin 1 (IL-1)as tumor growth stimulator, 58 Interferons, a,j3,y, and immune response ofcancer, 131,132,137 Interleukins-2,-4,-6, -8, - 10, -12 and immun responsesofcancer, 131,136,137,142, 49 1,492 Iodine deficiency endemic goiter and, 280 and follicular carcinoma 280 thyroid cancer and low iodine die 240,247 radioactive, see Radioiodine symporter gene, 283 Iodine- 123,see Radioiodine Iodine- 13 see 1, Radioiodine l 3 ‘I-meta-iodobenzylguanidine(MIBG) scanning, 42
L
LAK (Lymphokine-activated killer) cells, 141, 142 Leiomyosarcoma, 4 14 Leukemogenesis, of radioiodine therapy, 59 Leu M1 myelomonocyte marker, 204, 460-462 Levothyroxine therapeutic assessment, TSH assay for, 42 therapy adverse effects, 42 for nodule,42,43 risk of cancer and size reduction, 4 risk of osteopenia, 42 thyrotropin serum levels and, 42 Li-Fraumeni syndrome, 62 Lymph node biopsy,18,20 Lymphocytes, tumor infiltrating (TIL), 141,142 Lymphokine-activated killer (LAK) cells, 141, 142
503
Index Lymphoma antithyroid antibodies,353 B-cell, 351,359 classification of tumor type,359 NCI formulation, 352 Kiel classification,352 cytology, 353 diagnosis, 5,353 donut sign, 353 epidemiology, 79,35 1 Hashimoto’s thyroiditis and, 80, 122, 351,360
histopathology, 351,352,359-361 imaging, 353 incidence, 79 lymphadenopathy, 352 mucosa-associated lymphoid tissue (MALT), 352,361 pathology, 359-361 radionuclide imaging, 353 risk factors,351 staging, 352,353,354 surgical treatment,354 survival, 355 symptoms, 352 T-cell, 35 1,359 treatment by combined modalities,354, 355 chemotherapy, 354,355 BACOP, 354 CHOP, 354 C-MOPP, 354 CVP, 354 ProMACE-CytaBOM, 354 radiotherapy, 354 surgery, 354 LyrnphoGxL, see Tumor necrosis factor-a
M MACIS prognostic scoring system,51 Magnetic resonance imaging,see MRI MALT lymphoma, 352,361 MAPK (radmitogen-activated protein kinase), 59,65,366 MAPKK (MAP kinase kinase; MEK),59
MBq (megabequerel), 86 MDM2 gene or protein, 62,67 MDR- 1 expression in anaplastic carcinoma, 339,491 in medullary carcinoma,406 Medullary thyroid carcinoma (MTC), 39,49,365-408
associated disorders, 365,366 adrenal medullary disease,377 multiple endocrine neoplasia,6, 39,49
biochemistry, 366,367 calcification in, 17, 18 calcitonin in, 39, 50,366,367,375, 384,399
immunostaining, 374 prognosis and, 377,400 screening, 368 calcitonin gene-related peptide (CGRP), 366,375,460-462 carcinaembyonic antigenin, 39,50,367 C-cell hyperplasia, 373,374 chemotherapy, 402,403,405-408 combination chemotherapy,405 doxorubicin, 405-407 drug resistance and MDR-1 gene, 406 DTIC, 406 etoposide, 406 interleukin-2,492 monoclonal antibodies,492 octreotide, 406,407,492 classification, 365,366, 376 clinical course, 375,399 cutaneous lichen amyloidosis (CLA) and, 365,370 complications, 375,376 cytology, 384,386,399 developmental biology, 366 diagnosis, 49,50,399,400 by radionuclide imaging, 368,377, 389-396
RET protooncogene,39,49,68,69, 366,369,370,373,399,478
epidemiology, 79 etiology
504
genetic inheritance,49,68, 69 oncogenes in, 39,68,69 familial medullary thyroid carcinoma, 6,370 genetic abnormalities,39,68,478 family screening, 37 1,400 follow-Up, 376 gene mutations in,366,369,370,373 hereditary, see Familial medullary thyroid carcinoma Hirschsprung’s disease and, 365,370 histopathology, 374,383,384 hormones produced, 367 incidence, 79,373 lymphadenopathy, 374,375 MEN-2A, see Multiple endocrine neoplasia) MEN-2B, see Multiple endocrine neoplasia) metastases, 375,402 microscopic orin situ lesions, 373,374 mixed follicular-medullary type, 434 nuclear medicine imaging in, 389-396 indium pentetreotide (octreoscan), 389-391 iodine- 131, iodine- 123metaiodobenzylguanidine (MIBG), 394,395,396 isotopic agents employed, 389 monoclonal antibody imaging, 395 SPECT,39 1,395 technetium-99m dimercaptosuccinic acid (DMSA),392,393 technetium-99m methylene diphosphonate, 393 technetium-99m 2methoxyisobutyl isonitrite (MIBI), 393 thallium chloride, 393 paraneoplastic, 367,375 pathology, 383-387 pheochromocytoma and,365,370,377 prognosis and,374,375,376,400,403 radiation therapy, external, 40 1404 brain and spinal cord, 402
Index
combination chemotherapy, 402 empiric, 40 1 interstitial radiation, 403 metastatic disease, 402 postoperative, 401,402 preoperative, 40 1 technique, 402,403 survival, 403 radioisotope treatment, 395,396 RET receptor in,39,49,366,369-371 sex differences, 373 sporadic clinical courseof, 367,375 diagnosis of, 367 familial screening for,39,400 genetic abnormalities in, 366 management of, 377,399 paraneoplastic manifestations, 367,375 prognosis in, 376 screening tests for,370,371 surgical treatment, 377,378,399,400 in, 370 adrenal gland considemtions laparoscopic, 377 parathyroid gland considerations in, 400 ret mutations as negative prognostic factor, 373 total thyroidectomy for, 400 staging, 368,369,376 thyroglobulin in, 384 treatment with radioiodine, 395, 396 Megabequerel (Mbq), 86 MEK, see MAPKK MEN, see Multiple endocrine neoplasia Metastases in children, 104, 125 follicular carcinoma, 292 Hurthle cell carcinoma, 44 1443 to thyroid, 411,439,440 common primary tumors, 439 diagnostic evaluation, 439 met protooncogene, 64 MIBG (metaiodobenzylguanidine), I3II, 42, 394,395
Index
505
type IIA, 39,49,68,69,79,365,369, treatment of medullary carcinoma, 370,374,399 396 genetic abnormalities in, 49 Microcarcinoma, papillary, see Papillary hyperparathyroidism and, 39,49, microcarcinoma 3 69 Micropapillary carcinoma, see Papillary medullary thyroid carcinoma and, microcarcinoma 17,18 Mixed follicular-medullary carcinoma, prognosis, 366,367 434 type IIB, 39,49,68,69,79,365,370MRI, 11,2 1-24 374,399 in anaplastic carcinoma, 334 genetic abnormalities in, 49 advantages, 22,32 medullary thyroid carcinoma and, 39 carcinoma characteristics, 24 prognosis, 376,377 comparison to CT,22,23 Mutations substernal adenopathy, 26 activating, of TSH receptor, 65 T1 and T2 weighted images, 22 c-myc oncogene, 67,460-462,463, tracheal invasion, 25 473 utility germline, 463 in cancer without known residual somatic, 463 tumor, 29 in cancer with known residual tumor, 30,3 1,245,246 N in diagnosis ofcancer pst-operatively, Nerve growth factor(NGF) 28 thyroid cancer and, 64 for sectional images, 24 in medullary carcinoma, 366 in undiagnosed, suspected cancer, Neural cell adhesion molecule (NCAM) 27 in medullary carcinoma, 367 Molecular pathogenesis of thyroid cancer, Neuron specific enolase, in anaplastic 57-75 carcinoma, 323 growth stimulating factors,58 Nodules, solitary,3-7, receptor/phospholipase Uprotein aspiration, fine-needle, 35-37,39,40 kinase C pathway, 57,60,61 autonomous, 4 (see also Hyperfuncsignal transduction pathways,57-61 tioning, hot, toxic) TSH receptor/adenylate cylclase/protein benign, suppressive therapy for, 42,43 kinase A pathway, 57,60,61 calcified, 16, 18 Mucoepidermoid carcinoma, 145-4 17, cancer in, 4,10,49 434-435 clinical considerations, 5, 6 clinical presentation, 435 coldhonfunctioning, 40,43 pathology, 415,417 diagnosis, 8, 11,44 Mucosa-associated lymphoid tissue cytologic, 19,20,39,40 lymphoma, see MALT cost of evaluation,9, 10,40,44 Mucosal neuroma syndrome,see Multiple CT imaging, 1 1 endocrine neoplasia(MEN), type IIB diagnostic algorithm, 8 Multiple endocrine neoplasia (MEN), 6, diagnostic procedures, 10, 1 1 49 differential, 4,5 type II,39,372,373
506
Index frozen section,44 imaging methods for,9-34 MRI imaging, 1 1 physical examination for,5 , 6 radionuclide scanning for,40-42 241-americium fluorescent scanning, 41 131-cesium,42 67-gallium,42 Iz3-iodine,41 l3I-MIBG
(metaiodobenzylguanidine), 42 75-selenomethionine,42 99mTc-pertechnetate,41 201'thallium,42 by ultrasonography, 11-20,40 fine-needle biopsy,35-37,39,40 needles for, 35,36 radiation-related tumors, utility for, 51 repeat procedure, inadequate sample and,40,43 technical hints, 37 technique for, 35-37 ultrasound-guided, 16, 19,20 hotdhnctioning, 4,40 (see also Toxic; Hypefinctioning; Autonomous) infarction, 4 1 radionuclide imaging of,40,4 1 risk of malignancy,40,41 hyperhnctioning 4,43 (see also Nodule, hot and Nodule, toxic) radioisotope imaging of,40,41 incidence, 3,49 incidentalomas, 3, 10, 15 laboratory evaluation,8 , 3 9 malignant incidence of,49 ultrasonography of,15-2 1,40 management, 8,39-48 medical treatment ethanol intralesional injections for, 41 thyroxine suppression therapy for, 42,43
papillary thyroid cancer and,4, 10 pathogenesis, 3,4 prevalence, 3 in radiated children,4, 122-124 risk, after radiation exposure,4,44, 122-1 24 risk of cancer in,49,50 size, as indication for surgery, 41 surgical treatment, 44,45,49-53 choice of procedure, 5 1,52 frozen section,44 indications for,41 near total vs. total thyroidectomy, 51,52 technique for,5 1,52 thyroid cancer risk and,4-6,9, 10 toxic, 4,43 clinical features of,4 cancer in,4,5,40,41 pathogenesis of,4 warm, 43 ultrasound, utility of,9-34,40 Non-isotopic imaging, 9-34 role in diagnosis of cancer postoperatively, 28 role in cancer without known residu tumor, 29 known residual role in cancer with tumor, 30 role in undiagnosed, suspected cancer, 27 Nuclear proteins controlling genomic integrity, 67 Nuclear Regulatory Commission, regulation 148,149,150,151 Nuclear transcription factors,67 0
Occult papillary carcinoma, see Papillary microcarcinoma Oncogenes, 5%1,462,463 activation, 61 bcl-2, 118,460-462 cell surface receptors,63 c-erbB, 63 c-erbB2/neu, 63
Index
507
thyroid hormone therapy, 272 cytology, 193-206 death rates,78,79,263,264 diagnosis, 186, 187 in children,267,268 diffuse sclerosing variant, 433 epidemiology, 77-82, 186 etiology diet and, 95 radiation exposure and,4,65,77, 85-1 06 (see also Chernobyl) follicular variant, 196, 199,200,28 1 follow-up, 190,229-235 histopathology, 79 clear cell carcinoma, 204 columnar cell variant,205,206 P cystic carcinoma, 20 1,202 diffuse sclerosis variant, 202,203 Papillary m i m i n o m a (see also Papillary encapsulated variant, 199,200 thyroid carcinoma) fibrotic carcinoma, 202 diagnosis of, 185, 188 follicular variant, 196, 199,200 histology, 79,203 less well differentiated, 205 incidence of, 86 microcarcinoma, 203 prevalence of, 78,86 oxyphilic papillary carcinoma, 203 prognosis of, 263-265 tall cell variant,204,205,428-432 treatment for, 188 incidence, 186 Papillary thyroid carcinoma (PTC), lymphadenopathy, 186,187 185-1 90 MDM2 overexpression in, 67 age and, 78, 189, 190 metastases in children, 267 CD44 protein in, 68 occult, see papillary microcarcinoma columnar cell variant papillary oncogenes, 5748,462,463 carcinoma, 432,433 (see also Papillary pathology, 193-206 clinical presentation, 432 thyroid carcinoma, histopathology) demographics, 432 prognosis, 189,229,263-265 treatment, 432,433 age at diagnosis and, 189, 190 in children, 12 1-1 28,267-276 in children, 272,273 clinical presentation, 125,267-269 effect of treatment and, 263,264 diagnosis, 267,268 effect of tumor subtype and, 264 fine needle aspiration, 268 influencing factors for, 189,263,264 management, 269,270,273,274 pediatric, 121-128,267-276 pathology, 125 psammoma bodies, 195, 196, 199,203 after radiation, 10 1-1 04 radiation exposure radioiodine ablation clinical presentation in children, 104 adverse effects, 271 dose response relationship,96 dosage, 270 exposure types choice of surgical approach, 269,270
c-fos, 59 Gsa, 6 1,66 gsp, 6 1366 c-jun, 59 growth signal transduction and, 57-61 in thyroid cancer diagnostic utility,477,478 mutations, activating, 61 mutations, cancer, 61 screening and prognosis, 478 Raf- 1,59 ras, 61 (see also ras) ret, see ret rsk, 59 Oxyphil carcinoma, 203
508
Index external radiation therapy for acne, 89 for cervical tuberculous adenitis, 87, 88 in childhood, thyroid cancer and, 77, 80, 87 in children after irradiation, 104
for cutaneous hemangioma, 88,89
for other malignancies,90 for thymus, 89,90 for tonsils and adenoids, 89 internal radiation exposure,92,93 from nuclear fallout(see also Chernobyl), 87,93,96106
risk in children and adolescents, 93,94,101-104 in Belarus, 101, 102 in Russian Federation, 103 in Ukraine, 103, 104 from diagnostic/therapeutic radioiodine, 92 near nuclear facilities,91 occupational, 91 gene mutations, 104, 105 modifying factors on risk, 94-96 age at irradiation, 94 iodine deficiency,95, 122 race, 95 sex, 94 temporal pattern, 95 molecular characterization,104106 prenatal, 91,92 pathology, 85-87 risk assessment,94-96 radiation therapy, external,225-228 with brain metastasis,225 combination therapywith radioiodine, 225
dosage and technique,226-228 indications, 225 interstitial irradiation,228 preoperative, 225
side effects,228 spinal cord shielding,228 technical factors,226 radioactive iodine therapy (see also Radioiodinethempy), 147-151,213-219 bone metastases,216-219 brain metastases, 160, 16 1,219 in children,270,271 doses for, 147, 148,214 for distant metastases, 214-2 19 objectives of treatment,147 practical issues, 151 pregnancy tests, 151 posttherapy guidelines,151 posttherapy scan, 151 for pulmonary metastases, 214,2 15 safety considerations, 150, 151 single high-dose method, 147,214 stunning, 148,167 ras mutations in,65,66, 105 Ret receptor in,49,433,46973 (see also ret) risk factors,94,95 sclerosing variant,see Difise sclerosing variant solid variant papillary carcinoma, 433,434
role of radiation exposure, 433 staging, 188, 189,209,210 surgical treatment, 49-53 approach, 188,209-21 1 in children,269,270 extent of surgery and, 51,52 lymph node dissection in,52 postoperative radiotherapy and, 190 (see also Radioiodine) surveillance, 229 symptoms, 187 tall cell variant,204,205,428-432 clinical presentation,428,429 comparison to differentiated papillary 429,43 1 demographics, 428 outcome, 431 prognosis, 430
lndex treatment, 429,430 PAX-8,138-140 PCNA (proliferative cell nuclear antigen), 294,460 Pendred's syndrome, 6 PET (positron emission tomography), see FDG-PET Poorly differentiated carcinoma (insular), 412,413,425-427 clinical presentation,425,426 cytology, 4 1 3 histologic pattern, 425 outcome, 426,427 prognosis, 427 treatment, 426,427 with thyroid hormone,427,428 see Positron-emission tomography (PET), FDG-PET Prostacyclin, as tumor growth stimulator, 58 Prostaglandins as tumor growth stimulators, 58 Protein kinaseA (PK-A), 61 Protooncogenes, 59-62,64,65,462,463 bcl-2, 1 18 C-fos, 67 c-myc, 67 ELEl gene, 64 mutated, see Oncogenes in thyroid cancer, 61 ret, c-ret,see ret ret/PTCl rearrangement, 64,65,104, 105,433 trk, 64 Psammoma bodies in follicular carcinoma, 291 in papillary thyroid carcinoma, 17,18 p53 tumor suppressor gene,61,62, 105, 118,124,140,433,434,461,473 increased inactivationin lymphoma, 352 mechanisms of inactivation,62,473 mutations of, 6 1,62,67
R Radiation exposure, 85-106 (see also Pap illary carcinoma, radiation exposure;
509
Thyroid cancer, radiation-induced) clinical presentationin childrg 104,125 dose response relationship, 96 exposures near nuclear facilities, 9 1 occupational, 9 1, 150 prenatal, 91,92 external radiation therapy for acne, 89 for cervical tuberculous adenitis, 87,88 in childhood, thyroid cancerand, 77, 80, 87 for cutaneous hemangioma,88,89 for other malignancies, 90 for thymus, 89,90 for tonsils and adenoids, 89 internal radiation exposure,92,93 from nuclear fallout(see also Chernobyl), 87,93,96-106,433 from diagnostichherapeutic radioiodine, 92 risk in children and adolescents, 93,94, 101-104 in Belarus, 10 1,102 in Russian Federation, 103 in Ukraine, 103, 104 gene mutations, 104, 105 metastases, 86 modifying factors on risk, 94-96 age at irradiation, 94 iodine deficiency, 95 race, 95 sex, 94 temporal'pattern, 95 molecular characterization,104-1 06 papillary thyroid carcinoma and, 86 pathology, 85-87, 125 in children after irradiation,104, 122-1 24 risk assessment, 94-96, 121-1 24 thyroid abnormalities and,85-1 06 thyroid cancer risk, 90,93-96,433 Radiation therapy, external for nonthyroid cancer
510
thyroid cancer risk, 77,85-106,121-124 Radiation therapy, external for thyroid cancer, 225-228,341-343 with brain metastasis,225 combination therapy with radioiodine, 225 dosage and technique,226228,343 follicular carcinoma, see Follicular carcinoma, radiation therapy future directions,495 indications, 225 interstitial irradiation, 228, 343 medullary carcinoma, see Medullary carcinoma, radiation therapy papillary carcinoma, see Papillary carcinoma, radiation therapy preoperative, 225 side effects,228 spinal cord shielding,228,343 technical factors,226 Radiation thyroiditis, 156 Radioactive iodine,see Radioiodine Radioiodine candidates for ablation, 149, 150 candidates for therapy,213,2 14 in children, 270,271 complications, 155-1 62 acute side effects, 155, 156 anaplastic transformation, 159 bone marrow depression, 157 in children, 27 1 breast cancer,271 carcinogenesis, 160 cerebral edema, 160, 161 cerebral metastasis swelling,160,16 1 genetic defects,see Gene, mutations) infertility, 157-1 59 leukemogenesis, 159 ovarian function and infertility, 158, 159 parathyroid gland hypofunction, 157 pulmonary fibrosis, 159 radiation sickness, 156 salivary gland dysfunction,155,156
Index
sialoadenitis, 155, 156 taste dysfunction, 156, 157 testicular dysfunction and infertility, 157,158 in children,271 thyroiditis, 156 diagnostic studies, 124 dosimetry, 2 14 dosage schemes, 147, 148,214, 240-243 treatment goals and, 147,149 exposure, 92-94 iodine- 123,4 1 imaging iodine-123 for, 41 posttherapy scans, 15 1 postoperative, 148 posttherapy guidelines, 151 radiation dosimetry,see Radioiodine, dosimetry radiation safety and, 150, 15 1 stunning, 148, 167,241,253 therapy for thyroid cancer,147-1 5 1, 2 13-2 19,239-247 bone metastases,2 16-2 19 brain metastases,160, 161,2 19 in children,270,271 doses for, 147, 148,214 ALARA concept, 150 for distant metastases, 2 14-2 19 empirical high dose therapy,245, 254 for local invasiodrecurrence, 2 19 lymph node metastases,2 14 methods for,2 14 in negative scan, positiveTg patient, 245,254 objectives of treatment,147 for papillary carcinoma,2 13-2 19, 270,271 practical issues, 151 pregnancy tests, 151 posttherapy guidelines, 15 1 posttherapy scan, 15 1 for pulmonary metastases,2 14,215
511
Index safety considerations, 150, 151 single high-dose method,147,214 stunning, 148, 167,241,253 thyroid hormone withdrawal,239 timetable for metastatic survey scan, 242 timetable for therapy, post-surgery, 242
TSH stimulation,239,240 in thyrotoxicosis and cancer risk, 92 Radionuclide imaging alternatives to 131-I for imaging, 255-258
doses for scan,24&243 metastatic survey,242-244 radiopharmaceuticals for 241hericium fluorescent scanning, 41
131Cesium,42 FDG-PET, 246,247,257 99mTc-furifosmin,257 67Gallium,42 12310dine,41 13’-MIBG (metaiodobenzylguanidine),42 75Selenomethionine,42 Se~tar nibi-~~~Tc, 246,256 somatostatin receptor,246,257, 258
99mTc-pertechnetate,41 99mTc-tetrafosmin,256,257 *OIThallium,42,246,256 scintigraphy low iodine diets,240,247 metastatic survey scan common findings,244 timetable, 242 post-surgical scan,242 post-therapy/ablative scan,244 scan positive, thyroglobulin negative patients, 150,233,245, 251-258 techniques, 243
to increase uptake,243,244 stunning, 148, 167,241,253
Radiotherapy, see Radiation therapy, external Raf-l,59 Ras gene, point mutations in thyroid tumors, 66, 105, 118,193,468 Radmitogen-activated protein kinase (MAPK), 59,65,366 ras oncogene family, in thyroid cancer, 59,61
Ras protein, mutations,58,61,65,66, 105, 140,366,460462
Ras signaling system,58, 59, 65 Recombinant human thyrotropin,see Thyrotropin, recombinant human Redifferentiation therapy,254,283,491 Replacementh t q(see also underspm~c thyroidhornones),w h t ib w a l of,prior to radioisotope imaging, 239,240 Ret protooncogene, blood screening,369,370 mutations in hereditary medullary thyroid carcinoma, 49,366,369,370 after radiation exposure,104, 105, 124,433 point mutations,49,366,370,460,470 retPTC1 rearrangement, 64,65, 104, 105,459-462,471,472 Ret receptor, 366,370 signaling pathway,469-473 Rsk oncogene, 58
Russia, radiation exposure,see Chernobyl S
Sarcoma, thyroid,435,436 as distinct from anaplastic carcinoma, 436
metastases, 436 Sievert (Sv),86 Signal transduction proteins,57,63,65, 463-467
Sipple’s syndrome,see Multiple endocrine neoplasia (MEN) type IIA squamouscell ( a d e n ~ 0 u s carcinoma, ) 411,421-424
512
clinical presentation, 423 demographics, 421,422 thyroglossal duct and, 439 treatment, 424 Staging of papillary thyroid cancer, 188,189 Stunning, 148, 167,241,253 Sv (Sievert), 86
T Tall cell variant papillary carcinoma, 204,205,428-432 clinical presentation, 428,429 comparison to differentiated papillary, 429,43 I demographics, 428 outcome, 431 prognosis, 430 treatment, 429,430 T4, see Thyroxine Telomerase, 460,475 Teratomas of thyroid,417,436,437 Tg, see Thyroglobulin TGF-P and immune suppression, 134, 137 tumor marker, 460-462 Thymic tumors, 417,437 Thyroglobulin (Tg) antibodies, 230,232,308 antigenicity, 230 molecular biology, 230 physical properties, 230 serum levels assay methods, 230-233 detection limits, 230 effect of antibodies, 230-232 on RIA, 232 on ICMA, 232 factors affecting assay, 232 in follicular carcinoma, 308 hook effect, 232 cost, 230,23 1 1 elevated levels, causes, 23 negative scan with elevatedTg,150, 233,251-258
Index
in thyroid cancer anaplastic carcinoma, 323 antibodies, 230,232,308 decision analysis based upon Tg, 233,234 diagnostic sensitivity, 20 for follow-up, 168-171,230-235 with thyroid remnant,234,235 in follicular carcinoma, 308 immunogenicity, 139, 140,230 as tumor marker, 230-235 Thyroglobulin-mRNA, 187,279 Thyroglossal duct carcinoma associatedwith, 414,437439 clinical presentation, 438 cytologic examination, 414 diagnosis and imagingof, 438 squamous cell carcinoma and, 439 treatment of, 438 cysts, 437 carcinoma in,438,439 Thyroid cancer (see d o p @ ccarcinomas) age and, 77-79,12 1,122 antigenicity of thyroid cancer, 137-140 PAX-8 activation factor, 138-140 thyroglobulin as antigen, 138, 139 thyroid peroxidase as antigen, 13 8 thyrotropin receptor as antigen, 138 associated disorders endemic goiter, 79 iodine deficiency, 79 nodules, 49 thyroid lymphoma, 80 in children, 121-128,267-276 clinical presentation, 125, 276 after radiation,101-1 04 chemotherapy, 179-1 8 1,22 1-223 adverse side effects, 222 bleomycin, 180,406 carboplatin, 180, 181 cisplatin, 180, 181 combination therapy with external radiation, 222,223 cyclophosphamide, 180
513
lndex dacarbazine, 180,406 doxorubicin, 179, 180,22 1,405407 etoposide, 180, 181,406 methotrexate, 180, 18 1 octreotide, 406,407 paclitaxel, 180 single modality therapy,221,222 vincristine, 180,406 classification, 1 17 death rate, 79 diagnosis by fine-needle biopsy, 35-37, 39-41
epidemiology, 77-83, 121, 122 follicular thyroid carcinoma, see Follicular Thyroid Carcinoma follow-up, thyroglobulin levels for, 168-171
genetic aberrationsin, 6042,476 (table) histopathology, 79 immunology of, 129-1 42 mechanisms of reduced immunogenicity, 139 immunotherapy, 134,135,141, 142 incidence, 49,77,78 age and, 77,79, 121,122 ethnicity and,78 gender differences in,77-79 geographic patterns of,77-79 iodine deficiency/sufficiency,79 medical treatment suppressive therapy for,42,43 metastatic invasion,see Metastases and specific carcinomas molecular pathogenesis, 57-75 growth stimulating factors,58,63 markers of poor differentiation, 140,141
receptor/phospholipase Uprotein kinase C pathway,57,60,61 signal transduction pathways,5761,63,463467
TSH receptodadenylate cylclase/ protein kinaseA pathway, 57, 60,61,466,468
monitoring for recurrence, algorithm, 174 mortality, 78,79 occult, see Papillary microcarcinoma pathology (seealso individual tumors) angiosarcoma, 4 14 hyalinizing trabecular tumors, 4 14 leiomyosarcoma, 4 14 metastatic to thyroid, 4 1 1,439,440 mucoepidermoid, 415,4 16 pathologic interpretation, 117, 118 poorly differentiated,412 squamous cell,4 11 teratoma, 417 thymic tumor,417 thyroglossal duct cancer,414 prevalence, 78 prognosis, 263-266,3 11-3 14,345-349 radiation-induced, 50, 5 1,77 radioisotope imaging of(seealso specific tumors) 241Americium fluorescent scanning, 41
131Cesium,42 67Gallium,42, 353 12310dine,41 13'-MIBG (metaiodobenzylguanidine), 42 99mTc-Sestamibi,353 75Selenomethionine,42 99mTc-pertechnetate,4 1 20'Thallium,42, 353 recurrence, algorithm for monitoring, 174
redifferentiation, 254,283,491 risk, 79,80 surgical treatment, 49-53 with intraoperative probe,254 radical neck dissection for,5 1,52 survival, see Thyroid cancer, prognosis and specific carcinomas, prognosis thyroglobulin in,23&235 antibodies, 230,232,253 antigenicity, 230
514
Index decision analysis based upon Tg, 233,234
diagnostic sensitivity,20 for follow-up, 168-1 71,230-235 with thyroid remnant,234,235 immunogenicity, 139,140,230 molecular biology,230 physical properties,230 serum assay methods,230-233 detection limits,230 effect of antibodies, 230-232
on RIA, 232
factors affecting assay,232 cost, 230,23 1 elevated levels, causes,23 1 negative scan with elevated Tg, 150,233,25 1-258 as tumor marker,230-235
thyroglossal duct-associated cancer, 414,437439
undifferentiated,seeAnaplasticcarcinoma Thyroidectomy (seealso Subtotal thyroidectomy) in children, 269,270 near-total, 51, 52, 2 10 advantages of,51,52 for follicular and papillary carcinoma, 210
for nodule, 50-53 technique for,51,52 for papillary carcinoma, 210,269,270 frozen section examination, utility of,210 total, 52 advantages of, 52 for papillary carcinoma,210 risks from, 52 selection of,2 10 technique for,210 tumor identificationwith intraoperative
Pk254
tumor markers,see Tumor markers Thyroid function tests(seeaZso specific tests) utility in thyroid noduleskancer, 188 Thyroid hormones(seealso specific hormones)
dosage, 43 in children,272 Thyroiditis antitumor effects, 135 Hashimoto’s thyroiditis cancer and, 122 with lymphoma, 80,122,351,360 Thyroid lymphoma,see Lymphoma Thyroid sarcoma,see Sarcoma and individual tumors Thyroid-stimulating antibodies (TSAbs) in thyroid canceras tumor growth stimulator, 58 Thyroid stimulating hormone,see Thymtmpin Thyroid stimulating hormone receptor, see TSH receptor Thyroid-stimulating immunoglobulin (TSI) in thyroid cancer,58,80,466 Thyrotropin (TSH) bovine, for isotope scanning, 164, 165 growth signal transduction and,57, 58,60,6 1,465-467
human pituitary TSH(seealso Thyrotropin, recombinant), 165, 166 monitoring protocols in thyroid cancer 163
neoplasia, role in,465 serum levels, 163 suppression (seealso Thyroxine suppression therapy), 42,43 in thyroid hormone replacement therapy, 42,43 Thyrotropin receptor(seeTSH receptor) Thyrotropin, recombinant human,163175,240,279
algorithm for monitoring for recurrence, 174 antibodies to, 170 dosing regimedschedule, 169 future directions, 173 hormone withdrawal, comparison to, 168-1 73
quality of life,169 in vitro studies, 166, 167 studies in normal subjects,167 studies in thyroid cancer,167-1 73
515
Index synthesis, 166 thyroglobulin response, 168, 169, 170,171 use in conditions other than thyroid cancer, 173 Thyrotropin-releasing hormone(TRH) for isotope scanning, 164 Thyroxine (T4) dosage, 43 replacement therapy in children, 272 TSH-suppressive therapy, 42,43 use of adrenergic blockers, 42 probability of malignancy, 42 Thyroxine suppression therapy, for nodules, 42,43 TNF, see Tumor necrosis factor Total-body scans non-iodine scanning agents, 1,42 4 Trachea, invasion by anaplastic carcinoma, 322 by papillary carcinoma, 25 Transcriptional factors (TTF-1; PAX-8), 139 Transforming growth factor-p, see TGF-P TSH, see Thyrotropin TSH receptor activating mutations, 57, 61,65,467, 468 radiation-induced mutation, 105 signaling pathways, 466,467 tumor marker, 460 TSI, see Thyroid stimulating immunoglobulin Tumor growth factor-alpha(TGF-a), as tumor growth stimulator, 58 Tumor infiltrating lymphocytes (TIL), 141, 142 Tumor markers (see also specific tumor markers)
correlation with prognosis, 46 1,462 descriptive table, 460 immunocytohistochemical methods for, 460,461 ploidy, see DNA ploidy analysis thyroglobulin, see Thyroglobulin Tumor necrosis factors(TNF-a;TNF-P) , 135,136 Tumor staging,188-1 89 Tumor suppressor genes,130-1 34,462, 463 oncogenesis and, 463
U Ukraine radiation exposure,see Chernobyl Ultrasonography, 11-2 1 color flow Doppler,15,21 correlation with histopathology, 15, 16 cost effectiveness,25,26 echogenicity as index of disease, 16, 18 gray scale, 1 1-14 halo sign, 16,2 1,280 indications, 11 of lymph nodes,18,21 of nodule, 1 1-21,280 utility in cancer without known residual tumor, 29 in cancer with known residual tumor, 30,31 in diagnosis of cancer post-operatively, 28 in follow-up of nodules, 18 in undiagnosed, suspected cancer, 27
W Wuchernde struma, 292,425