Critical Pathways to Success in CNS Drug Development

fm BLBK249-Cutler Trim: 6in X 9in January 19, 2010 7:51 Char Count= Critical Pathways to Success in CNS Drug Deve...

15 downloads 1214 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

fm

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

Critical Pathways to Success in CNS Drug Development

Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

i

fm

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

Critical Pathways to Success in CNS Drug Development Neal R. Cutler,

M.D.

President and CEO, Worldwide Clinical Trials Beverly Hills, CA, USA

John J. Sramek,

Pharm. D.

Director of Clinical Research, Worldwide Clinical Trials Beverly Hills, CA, USA

Michael F. Murphy,

M.D., Ph.D.

Chief Medical and Scientific Officer, Worldwide Clinical Trials King of Prussia, PA, USA

Henry Riordan,

Ph.D.

Sr. VP Medical and Scientific Affairs, Worldwide Clinical Trials King of Prussia, PA, USA

Peter Bieck,

M.D., Ph.D.

Vice President of Clinical Research, Worldwide Clinical Trials Indianapolis, IN, USA

Angelico Carta,

M.D.

President, Worldwide Clinical Trials Europe London, UK

A John Wiley & Sons, Ltd., Publication

iii

fm

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

c 2010 by NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck This edition first published 2010, and A Carta Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has been merged with Wiley’s global Scientific, Technical and Medical business to form Wiley-Blackwell. Registered office: John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial offices: 9600 Garsington Road, Oxford, OX4 2DQ, UK The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 111 River Street, Hoboken, NJ 07030-5774, USA For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting a specific method, diagnosis, or treatment by physicians for any particular patient. The publisher and the author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness for a particular purpose. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. Readers should consult with a specialist where appropriate. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the publisher nor the author shall be liable for any damages arising herefrom. Library of Congress Cataloging-in-Publication Data Critical pathways to success in CNS drug development / Neal R. Cutler . . . [et al.]. p. ; cm. Includes bibliographical references and index. ISBN 978-1-4443-3064-9 1. Neuropsychopharmacology. 2. Drug development. 3. Critical path analysis. 4. Central nervous system–Effect of drugs on–Research–Methodology. I. Cutler, Neal R. [DNLM: 1. Central Nervous System Agents–pharmacology. 2. Clinical Trials as Topic–methods. 3. Drug Evaluation–methods. 4. Drug Evaluation, Preclinical–methods. QV 76.5 C934 2010] RM315.C715 2010 615 .78–dc22 2009035146 A catalogue record for this book is available from the British Library. R Set in 9.5/12pt Palatino by Aptara Inc., New Delhi, India Printed in Singapore

1 2010

iv

fm

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

Contents

Acknowledgments, vi Foreword, vii 1 The impending crisis in CNS drug development, 1 2 Animal models and procedures for CNS disorders, 14 3 Preclinical antecedents to early human clinical trials, 70 4 Biomarkers and surrogate markers in drug development, 101 5 Neuroimaging and cognitive assessments in early drug

development, 167 6 Bridging and CSF studies, 187 7 A case study from preclinical to early clinical trials, 230

Index, 254

v

fm

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

Acknowledgments

The authors are grateful to Andrew Kuhlman for his invaluable assistance in helping them research the literature, format tables, and write the present book. He was ever vigilant in organizing large sections of the text, keeping track of countless references, and integrating comments from all the authors with a unified writing style. We also wish to thank Tricia Long for her assistance with Chapter 6, and William Nowatzke, Ph.D., for his contribution on laboratory biomarkers in Chapter 4.

vi

fm

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

Foreword

The process of drug development has never been as complex and costly as it has become in the twenty-first century. Encouraged by sensationalistic and inaccurate reporting in the lay press, many citizens are convinced of the unrealistic fantasy that prescription pharmaceutical products should be risk-free. When effective medications cause adverse reactions—as they inevitably do— the response from the public and from politicians is outrage, with accusations of greed and malfeasance on the part of the pharmaceutical manufacturers and incompetence on the part of regulatory agencies. The outcome has been a development process driven by avoidance of liability and blame, with an increasing maze of paperwork and regulation that work against true innovation. For developers of new drugs in the area of neuropsychopharmacology, the already difficult process is further complicated by features intrinsic to the discipline. Mood, affect, thought, behavior, and cognition are not easily measured. Outcome instruments are “soft,” and responses potentially modulated by research setting, experience, expectations, and practice. Experimental animal models and surrogate markers of disease and treatment are less well developed than in other disciplines. Placebo response rates in trials of neuropsychiatric agents are uncomfortably high, leading to many “failed” studies in which even an established active comparator does not separate from placebo. I have known Dr. Neal Cutler and his work for more than three decades. His career represents a remarkable blend of academic achievement and entrepreneurial success. A physician and board-certified psychiatrist, Dr. Cutler spent his early professional career as an academic psychiatrist, both at the University of California and at the National Institutes of Health. In 1987 he started a successful clinical trial site, and subsequently a contract research organization (CRO) in Beverly Hills, specializing in clinical trials of marketed and candidate drugs in psychiatry and neurology. His work in the CRO industry has brought him to his present position as President and Chief Executive Officer of Worldwide Clinical Trials in Beverly Hills. The accomplishments of Dr. Cutler and his associates demonstrate that a financially successful CRO can not only serve the drug development objectives of the pharmaceutical industry, but also advance basic and clinical knowledge in neuropsychopharmacology. Over the years their work has resulted in hundreds of publications in the peer-reviewed medical and scientific literature. The vast experience of the group ranges from strategic scientific and conceptual aspects of drug development in psychopharmacology to practical issues

vii

fm

BLBK249-Cutler

viii

Trim: 6in X 9in

January 19, 2010

7:51

Char Count=

Foreword

of protocol design and execution, avoidance of typical problems and pitfalls, and compliance with regulatory requirements. This book collates and summarizes the experiences of Dr. Cutler and his professional colleagues. It is not an anthology of chapters from different authors at different institutions—each chapter is written by Dr. Cutler and his collaborating scientists at Worldwide Clinical Trials, Inc. It is their experience, expertise, and achievement that are represented on the printed page. For those determined enough to face the hazardous landscape of drug development in neuropsychopharmacology, this book will help. David J. Greenblatt, MD Boston, MA

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

CHAPTER 1

The impending crisis in CNS drug development

Introduction A plethora of information has been gathered across the fields of neuroimaging, genetics/genomics, proteomics, neurobiology, and epidemiology that have greatly enhanced our basic knowledge of the pathophysiological and genetic underpinnings of many common central nervous system (CNS) disorders such as schizophrenia, Alzheimer’s disease (AD), Parkinson’s disease, depression, and anxiety disorders. In fact, most of what is currently known about these CNS disorders has been discovered in the past decade. However, these breakthroughs in the CNS basic sciences have too often failed to translate into more effective, more affordable, and safer pharmaceutical products for patients suffering from these disorders. Much of the information resulting from these innovations has had little clinical relevance, and despite the newly acquired knowledge gained throughout the past few years, CNS drug development has been characterized by relative stagnation. In fact, the number of approvals for CNS drugs over the past several years has actually declined! Given the current economic climate in the US and around the world, this situation appears to be only getting worse—there are reduced resources available for drug development and reduced capital to fund this development. In addition, the drug development pathway is typically cumbersome and expensive, requiring fresh ideas and streamlined procedures to make development programs run faster and cheaper, as well as updated regulations from the Food and Drug Administration (FDA) to simplify the drug approval process. Many of these innovations already exist, and are beginning to be integrated into the drug development pipeline. Others are being validated and may soon become vital components of this pathway. The pharmaceutical and biotechnology industries are on the cusp of a major revolution in technology, procedures, and regulations regarding drug development—not because they wish to improve upon an already successful system, but because they need to replace a flawed and broken system if they wish to be relevant in coming decades, and create new, effective treatments for the patients who need them the most.

Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

1

c01

BLBK249-Cutler

2

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

Chapter 1

It is with this idea in mind that we have decided to write this book. We have compiled the latest advances in early CNS drug development from a vast body of literature, from clinical studies, and from our own experience. We have explained these advances in sequential, clearly organized chapters, beginning with preclinical models and going through first-in-man clinical trials. We have provided concise, relevant summaries and reviews of the newest techniques, markers, and models being used and introduced into the CNS drug development pipeline, and determined how they can be best utilized and what further validation is required. We have also reviewed the latest FDA regulations and guidelines, and discussed how each of these affects the drug development industry for better or for worse. However, before we discuss the many innovations and regulations designed to address the problems the drug development industry faces, we wish to briefly explain the problems themselves; this way you will have a better understanding of what requires fixing, why it needs to be fixed, and just how serious the problems really are.

Stagnation in CNS drug development The current stagnation in CNS drug development is evidenced by the lack of novel treatments across a number of neurologic and psychiatric disorders, with two of the most representative indications from the therapeutic area of CNS (AD and schizophrenia) serving as compelling illustrations of this stagnation. For example, the lack of approvable therapies that would hope to modify disease progression of AD has been truly frustrating—not only for physicians and family members who are caregivers for patients with AD, but also for society as a whole, given the looming financial and healthcare crisis associated with the ever-increasing prevalence of the disease. To date, all the drugs approved to treat AD, including the N-methyl-d-aspartic acid (NMDA) antagR (memantine), as well as all of the cholinesterase inhibitors, onist Namenda R R (galantamine/previously known as Reminyl ), including Razadyne R R Exelon (rivastigmine), and Aricept (donepezil), are prescribed for the treatment of the symptoms of AD and carry the label that there is “no evidence that any of these drugs alter the course of the underlying dementing process.” As an example, despite an explosion of publications advancing our understanding of the diagnostics, pathophysiology, genetics, and imaging associated with AD, there have not been any drugs that have successfully been shown to act as “disease modifiers.” This is certainly not for a lack of effort: In early 2007, there were approximately 12 drugs in US phase III clinical trials for AD, all of which showed great promise to slow or stop the progression of the disease based on their mechanisms of action. Additionally, the European Medicines Agency (EMEA) reported consulting on at least 15 different AD drugs, with 79% of their advice stemming from queries surrounding disease modification (20% were on symptomatic treatment and

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

The impending crisis in CNS drug development

3

1% were on diagnostics) [1]. Despite some early signals to the contrary, there R , are no development programs to date (including Neurochem’s Alzhemed TM Myriad Genetics’ Flurizan , and Wyeth/Elan’s bapineuzumab) that have unequivocally shown positive trial results, although some interesting trends were noted that will be discussed below. The lack of disease modifiers is especially dire when considering that the Alzheimer’s Association report from March of 2007 concluded that there were over 5 million people in the United States living with AD [2]. This number includes 4.9 million people over the age of 65 and between 200,000 and 500,000 people under the age of 65 with early-onset AD and other dementias. Equally alarming are the projections for the future. The prevalence of AD is predicted to increase 27% by 2020, an astonishing 70% by 2030, and nearly 300%, to approximately 13.2 million people, by 2050—unless a way can be found to slow the progression of the disease or prevent it [3]. Remarkably, it has been suggested that even a “5-year delay in onset could reduce the prevalence of AD by almost 50%” [4], underscoring the need for a drug that will delay the onset or progression of dementia. The prevalence projections for AD are in stark contrast to other indications such as heart disease and cancer, which are projected to remain stable or actually decline over time. From 2000 to 2005, death rates have declined for most major diseases—including heart disease, breast cancer, and prostate cancer—while deaths from AD continue to trend upward, and are expected to increase 44% by 2025 [5]. Another regrettable example of stagnation in the field of CNS therapies comes from psychiatry, and is evidenced by the lack of novel antipsychotic drugs to treat schizophrenia and other psychotic disorders. The three-phase Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study, one of the longest drug trials ever conducted in psychiatry, began in 2000 and data were analyzed beginning in 2005. In that time period, only two new antipsyR R and Zeldox ] in 2001 and chotic drugs (ziprasidone [marketed as Geodon R Abilify in 2002) were approved. Since then, only one other drug has been approved for the treatment of schizophrenia—Janssen’s InvegaTM (paliperidone), which was approved in 2006. Invega is an oral extended-release (ER) major active metabolite of risperidone and would not be considered to be a novel molecular entity or new molecular entity (NME). In the first phase of the CATIE study, 1493 patients with schizophrenia were recruited at 57 US sites and were randomly assigned to receive the antipsychotics olanzapine (7.5–30 mg/day), perphenazine (8–32 mg/day), quetiapine (200–800 mg/day), or risperidone (1.5–6.0 mg/day) for up to 18 months. Ziprasidone (40–160 mg/day) was added in 2002 following its approval by the FDA. The study concluded that the majority of patients in each treatment group discontinued due to inefficacy or intolerable side effects or for other reasons. Patients on olanzapine had the best record for continuing treatment, but this treatment was associated with greater weight gain and increases in measures of glucose and lipid metabolism. Overall, 74% of patients

c01

BLBK249-Cutler

4

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

Chapter 1

discontinued the study medication before 18 months. Surprisingly, the efficacy of the conventional antipsychotic agent perphenazine appeared similar to that of the atypical antipsychotics quetiapine, risperidone, and ziprasidone, suggesting no difference between first- and second-generation antipsychotics [6]. In the second phase of the CATIE study, 543 participants were selected who did not benefit from the first phase of the study. Patients were divided into two groups. In one group, patients were randomly assigned to get one of four medications: clozapine, olanzapine, quetiapine, or risperidone. In the other group, clozapine was not included, and ziprasidone, the newest of the atypical medications available in the early stages of CATIE, was compared with the other three (olanzapine, quetiapine, or risperidone). Clozapine, one of the earliest atypical antipsychotics, was found to be remarkably effective and substantially better than all the other, newer atypical medications in the study [7]. Unfortunately, clozapine is associated with serious side effects, including lifethreatening blood and heart complications, requiring careful monitoring of the patients taking this medication. It is often underprescribed because of this reason [8]. In the third phase of the CATIE study, 270 patients who had discontinued antipsychotic treatment in both first and second phases were enrolled. Patients and their doctors selected one of nine antipsychotic regimens (aripiprazole, clozapine, olanzapine, perphenazine, quetiapine, risperidone, ziprasidone, the long-acting injectable fluphenazine decanoate, or a combination of any two of these treatments). Symptoms showed modest improvement for most patients [9]. Clozapine was underprescribed due to safety issues, although it had been recommended as the only treatment consistently shown to be effective when others were not [10]. Predictions of better efficacy and safety of second-generation antipsychotics over conventional antipsychotics were not realized in the CATIE study, leaving CNS drug developers wondering what the second-generation antipsychotics actually added to treatment options. Importantly, during the 8 years since the start of the CATIE initiative, there has been a lack of novel “thirdgeneration” antipsychotic treatment for psychotic disorders. True third-generation antipsychotic medications would essentially be a new class of antipsychotic medications that differs substantially from older agents in terms of clinical effectiveness, reduced side effects, basic mechanisms, or some combination of these factors [11]. Based on these criteria, none of the approved antipsychotic medications would be considered third generation. Eli Lilly and Company has a promising third-generation antipsychotic with its compound LY2140023, but this agent is only in phase 2 of development and is far from approval. Thus, despite innumerable advances in basic research improving our understanding of the pathophysiology of schizophrenia, practicing psychiatrists anxiously await the arrival of a true third-generation antipsychotic compound that has novel therapeutic mechanisms of action.

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

The impending crisis in CNS drug development

5

Drug development stagnation: an industry-wide problem Regrettably, the stagnation in drug development is not restricted to the development of drugs for schizophrenia or AD or even limited to CNS drug development, but rather appears to be the prevailing trend across a number of therapeutic areas. In fact, despite the great increase in broad scientific/ medical developments, the number of new drug and biologic applications submitted to FDA has declined significantly over the past decade. The FDA approved 17 new molecular entities (NMEs) and 2 biologic license applications (BLAs) in 2007—the lowest number recorded since 1983, a year that had 14 approvals [12]. Of these 17 NMEs approved in 2007, only 2 were in CNS indications—New Rivers’ VyvanseTM (lisdexamfetamine) for attention–deficit/hyperactivity disorder and Schwarz BioSciences’ dopamine R ) for early stage idiopathic Parkinson’s receptor agonist rotigotine (Neupro disease. Of the 18 NMEs approved by the FDA in 2006 (the same number as in 2005), only 3 could be considered to fall under the CNS therapeutic R R (varenicline) for smoking cessation, Teva’s Azilect realm: Pfizer’s Chantix R (rasagiline) for Parkinson’s disease, and Janssen’s Invega (paliperidone) for schizophrenia. Of note, there was only one psychiatric drug approved in R (duloxetine HCL), and no psychiatric drugs at all 2004—Lilly’s Cymbalta approved in 2003 or 2005. 2008 has fared better with two psychiatric drug approvals, but there is still a dearth of NMEs. Approvals included Biovail’s AplenzinTM (bupropion hydrobromide) for the treatment of major depressive disorder and Banner Pharmacaps’ StavzorTM (valproic acid delayed release) for the treatment of bipolar manic disorder, seizures, and migraine headaches. There have been an additional five neurologic drug approvals (not R (tetrabenazine) for the just NMEs), including Prestwick Pharma’s Xenazine treatment of chorea due to Huntington’s disease; Eisai’s Banzel (rufinamide) for the treatment of seizures associated with Lennox-Gastaut syndrome R (fospropofol disodium), a sedativein pediatrics and adults; Lusedra hypnotic agent indicated for monitored anesthesia care sedation; Sirion R (difluprednate) for the treatment of inflammation Therapeutics’ Durezol R and pain associated with ocular surgery; and Schwarz Pharma’s Vimpat (lacosamide) for the treatment of partial-onset seizures in adults with epilepsy. An inspection across multiple years in other indications does not make the picture any brighter. The FDA approved more than 30 NMEs in only 1 year in the present decade. This is in stark contrast to the second half of the 1990s, in which the FDA approved more than 30 NMEs every year [13]. Stated another way, over the past 3 years, the FDA has approved a total of 53 NMEs—the same number approved in 1996 alone. The EMEA is also approving fewer products than the FDA, even though processing speed and volume have been significantly improved. Both the FDA and EMEA are asking for more data in drug applications that pose heightened safety concerns, although there have been no official changes in overall drug approval standards.

c01

BLBK249-Cutler

6

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

Chapter 1

Additionally, there does not appear to be much distinction in success between the pharmaceutical and biotechnology sectors, despite the fact that products originating from the biotech industry account for approximately two-thirds of all new drug applications. Henry Grabowski from Duke University reported that biotech drugs are slightly more successful in early-phase clinical trials than those of traditional pharmaceutical companies, but are more likely to fail in the larger phase III trials [14]. This is an important distinction as the basic questions concerning efficacy are largely determined in phase IIa and IIb studies. On average, 24.2% of biotech drugs are scrapped after phase III trials versus 12.6% of traditional drugs. Thus, biotech drugs account for more than 90% of phase III failures. These facts support the longheld views of biotechnology companies being more skilled at innovation and pharmaceutical companies being more skilled in the drug development process, and suggest that an alliance of these two might result in higher drug approval rates. Unfortunately, these alliances have not historically produced such results [15]. The decline in drug approvals is in harsh disparity to the ever-burgeoning costs and timelines required in drug development. The costs of developing the types of new drugs that have been pursued by traditional large pharmaceutical firms have been estimated in a number of studies, but the most widely cited figures come from DiMasi and colleagues [16]. This group utilized published cost estimates along with information on success rates and trial durations from a publicly available data set. They proposed that the cost associated with a new drug entering human clinical trials for the first time between 1989 and 2002 was estimated at $800 million, and for several years this figure has been widely used by the lay public and by politicians seeking to understand the cost of prescriptions. Estimates have generally supported this figure, citing costs that vary from around $500 million to more than $2000 million, depending on the type of therapy [17]. For example, it has been generally suggested that drugs for neurologic and psychiatric conditions tend to be the most expensive drugs to develop. In contrast, drugs targeted for infectious diseases and analgesia indications tend to be the least expensive to develop.

Specific difficulties with CNS drug development Despite the advent and acceptance of biological psychiatry and the abundant awareness campaigns by patient advocacy groups, the National Institutes of Health (NIH) and other government agencies often still inaccurately view CNS drugs as somehow less important than other drugs. This is because these drugs are often viewed as treatments for people who are fundamentally healthy but are seeking to improve a lifestyle problem, such as sadness, anxiety, addiction, or phobias. While CNS drugs may make it through the approval process at the FDA as quickly as their counterparts in other divisions, they often start off in a worse position by being viewed as having a relatively higher risk and lower priority due to the indications that they treat.

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

The impending crisis in CNS drug development

7

Furthermore, CNS drug developers are simply not utilizing the regulatory tools available to them, such as Priority Review or Fast Track designation, to have these drugs designated differently from the start. There are many CNS conditions (i.e., suicide) that would be considered to be serious or life threatening. In addition, given the historically poor treatment response to psychiatric drugs, the growing number of treatment refractory patients, and the degree of intolerable side effects, many CNS development programs would be considered to address an unmet medical need. In fact, the FDA Guidance for Industry on Fast Track Drug Development Programs—Designation, Development, and Application Review cites several CNS examples of whether the drug development plan addresses an unmet medical need: Effect(s) on serious outcomes of the condition not known to be affected by the alternatives (e.g., progressive disability in multiple sclerosis when the alternative treatments have shown an effect on exacerbations but have not shown an effect on progressive disability). Ability to provide benefit(s) in patients who are unable to tolerate or are unresponsive to alternative agents (e.g., an antipsychotic agent that is effective in people failing standard therapy), or an ability to be used effectively in combination with other critical agents that cannot be combined with available therapy. [18]

Products that receive Priority Review or Fast Track designation are not necessarily more likely to be approved by the FDA than products that do not receive any such designation. However, the Fast Track designation enables early interaction with the FDA that can help clarify elements of clinical study design and data presentation, whose deficiency upon New Drug Application (NDA) submission could delay approval decisions. Although the FDA makes similar interactions available to any sponsor who seeks their consultation throughout the stages of drug development, these meetings are not always guaranteed. A unique option within the Fast Track designation is the opportunity to submit sections of an NDA to the FDA when they become ready, rather than the standard requirement to submit a complete application at one time. Thus, many CNS development programs miss out on some very important advantages associated with special designations. The disparity in trial sample size is yet another manifestation of stigma or bias against CNS drug development. Charles B. Nemeroff, MD, PhD, noted large differences in number of patients in cardiovascular trials versus psychiatry trials [19]. Cardiovascular trials are strikingly larger than psychiatry trials. It would not be unusual to have 10,000 patients in a single cardiovascular study, while most psychiatry studies have less than one-tenth of that number (300–500 patients). Even the relatively large psychiatry trials sponsored by the NIH, such as the CATIE trial, have no more than a thousand patients, which results in a relative reduction of statistical power for CNS trials compared to

c01

BLBK249-Cutler

8

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

Chapter 1

cardiovascular trials. Nemeroff suggests that this is an artifact of the pharmaceutical industry’s reluctance to invest in psychiatry trials, and cites two main reasons for this. One reason is that psychiatry as a field is forced to deal with active “antipsychiatry” movements that do not believe in the benefit of psychiatric treatment. There appears to be no such movement for other disorders seen as purely “physical.” He also noted that the pharmaceutical industry is reluctant to get involved in psychiatry trials because the difference between active drug and placebo is often more difficult to demonstrate statistically. The general notion that drug–placebo differences are more difficult to discern in CNS trials is well acknowledged, and the conduct of CNS trials are often regarded as being as much art as science. It is certainly true that compound development in the CNS areas is more costly and often more difficult than compound development in other therapeutic areas, such as infectious disease. Some of the more salient reasons for this increased difficulty include: 1. The lack of correspondence between animal models and early patient studies in which very sophisticated and well-accepted animal models have failed to predict patient response to CNS therapeutics; 2. The difference seen in absorption, distribution, metabolism, and excretion (ADME) between normal healthy volunteers and patients, especially in AD, Parkinson’s disease, and schizophrenia; 3. The uncoupling of pharmacokinetic and pharmacodynamic measures, often seen across a multitude of psychiatry trials; 4. The lack of accepted biomarkers and surrogates by regulatory authorities and the scientific community, even in areas such as cerebral structural and functional imaging; 5. The use of subjective investigator and patient-rated diagnostic scales and endpoints, leading to rater inflation and regression to the mean following randomization (this is seen across a variety of psychiatry trials, but is especially problematic in trials of depression and anxiety); 6. The related issue of heightened placebo response (which is endemic in studies of depression and anxiety and is becoming more commonplace and troublesome in studies of schizophrenia and AD); 7. The often mismatch between clinical meaning and statistical significance across a variety of indications, especially in studies of analgesia; 8. Very high attrition rates with upward of 60% attrition in substance abuse and AD and schizophrenia/bipolar trials, making trials cumbersome, biasing treatment effects, and reducing statistical power and generalizability; 9. Very high levels of comorbid substance abuse rates, especially in trials of psychotic patient populations such as schizophrenia and bipolar mania; and 10. The large number of failed trials (not just nonsignificant trials) in which an already approved active comparator failed to differentiate from placebo, thus requiring a larger number of trials in order to secure two adequate and well-controlled studies. Although most companies entering into a CNS development program are aware of these issues, they tend to ignore them in favor of the potential

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

The impending crisis in CNS drug development

9

payoff of a CNS drug approval. In short, there is a great deal of money to be made by marketing to an ever-growing CNS customer base, especially in neurodegenerative disorders. The number of patients with CNS disorders far surpasses those with cardiovascular disorders, and given the population trend (in which those who are 85 years and older will quadruple by 2050), this difference and growth in CNS disorders that affect patients later in life (such as AD and Parkinson’s disease) is only likely to expand. No matter what the explanation for the lack of CNS studies (whether societal stigma, trial complexity, or difficulties in study conduct), most drug developers agree that there is room for expansion in the CNS marketplace. The potential size of the untreated CNS markets is so large that the future growth of the global neuropharmaceutical market could outpace the growth in the other sectors of the pharmaceutical industry. This fact alone makes CNS development attractive to the pharmaceutical and biotech sectors. In addition, the prospect of reducing patient suffering, prolonging life, and responding to important public health problems all demand greater efficiency in the clinical trial process, including a greater ability to secure approval for CNS drugs in a more timely and less costly manner.

A new outlook on CNS drug development The industry and the FDA must rethink and improve upon the typically cumbersome and expensive path of drug development, and come up with creative solutions that expedite the process and reduce costs while still producing effective new therapies. We view the drug development process as a creative opportunity that must be approached by drug developers cognizant of the entire multivariate processes involved. This can be seen as somewhat analogous to the role of a contractor constructing a new skyscraper. The contractor may not have degrees in the chemistry and physics of the materials, nor in engineering and geology, but he or she must fully comprehend the application of all physical principles in constructing a solid foundation. Once the foundation and first few floors have been prepared and properly constructed, the rest of the floors can rise mechanically and repetitively. If there are any problems encountered with the structural integrity as the floors rise, it is difficult if not impossible to correct it. Similarly, once the foundations of CNS development have been properly laid, and one understands early on the pharmacology of the compound and its potential for efficacy, the latter development stages can be executed with aplomb. The most elementary information about the compound, particularly the dose and regimen that will be used to maximal effect in the latter developmental stages, must be acquired in the early developmental stages. The CNS drug developer must fully understand the preclinical programs that have brought the compound to the point of human studies. He or she must be able to glean any important data that could impact the clinical development. For example, the animal toxicity data provide important indications about the

c01

BLBK249-Cutler

10

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

Chapter 1

underlying pharmacology and toxicity of a compound, which may be encountered in man, either initially or as long-term, late-onset events, or differences in gender, etc. Thus, the clinical program should be responsive to the appearance of such signals in man. If medicinal chemistry has provided a number of potential candidates, such data, in combination with animal models, can help the developer choose the most appropriate compound to proceed in human development. Ultimately, given the tremendous costs and resources associated with clinical trials, it is vital to determine in early human studies whether the potential drug candidate is worth continued development. In the CNS area, a thorough understanding of the available potential biomarkers, combined with novel designs and strategies for determining the safety and tolerance of the compound, as well as determining a useful dose range early on in the critical patient population, will enable the later stages to proceed on a solid foundation. For example, there may be biomarkers that can be employed to help confirm the mechanism of action, the best dose range, or even markers of unwanted side effects early in development. The latter stages of CNS development have their own complexities too, which will only be compounded if the early stages have not yielded useful information. For example, later stages of CNS drug development involve testing the often subjective effects of a compound in relation to placebo, and one does not need to have the added complexity of determining the proper dose and regimen in these stages—this should already have been determined early on. Every creative effort must be employed in the initial human program to understand the critical factors upon which the rest of the program will be built. The recent surge in development of biotechnology compounds has impacted the CNS area as well, adding layers of complexity to the drug development process due to interactions between these products and other biological processes. In addition, compounds produced from biotechnology often have unique patterns of administration, absorption and metabolism, and end organ effects. On the positive side, biotechnology compounds are often able to target specific underlying pathological processes, such as the pathways to formation of abnormal β-amyloid. This offers truly exciting potential for disease-modifying effects that cannot be duplicated by the more traditional small molecule therapies, which typically offer symptomatic improvement at best. The early development process can be accelerated by getting the compound into the target population as soon as possible and understanding its pharmacology (including adverse event profile) and dose range as quickly as possible. Application of new technologies in these early studies, particularly cerebrospinal fluid sampling of pharmacokinetic and pharmacodynamic endpoints, can greatly enhance the information we gleam about the compound at an early stage, and build a better foundation for future development. Most biomarkers will not provide definitive information that will direct subsequent

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

The impending crisis in CNS drug development

11

development, but nonetheless they can provide important clues and offer the opportunity for hypothesis generation that can be tested in other novel designs. In this book we will outline the known preclinical structures that are important to the early human studies, and elaborate on these early studies and the opportunities they yield in speeding the development process. In addition, in the last chapter of this book, we will provide a fictional case study of the drug development process, demonstrating how preclinical data of a novel compound can be used to construct early clinical studies. A solid foundation of these studies allows the first critical efficacy studies to provide a confident decision about the value of continuing the process. If the compound has failed at this point, there may be little point to expending further resources. If there is success, however, then the pathway for developing the program, including the time and commitment of hundreds of investigators and thousands of patients, will be justified. In this book, we take as our starting point the availability of a viable compound that enters preclinical development. A compound arrives at this point by often differing routes. Some will be identified for specific targets (such as receptor binding or enzyme inhibition) or by the screening of large chemical libraries of compounds. In other cases, new compounds will be generated by combinatorial chemistry. Yet in other cases, medicinal chemists will synthesize compounds, generally in series, based on alterations to an existing molecule through structure–activity relationship studies, in order to improve or strengthen its activity for a given target, such as receptor fit or binding. A new neuroscience discovery of a promising biochemical pathway linked to a disease state can rapidly push discovery efforts to screen for compounds with the potential to affect that pathway. After such efforts at synthesis and screening of compounds, preclinical studies are then undertaken mainly in the areas of pharmacology and toxicology to further screen these chemical candidates for activity. A full receptor screen for putative CNS compounds will be undertaken to assess the binding potential at all the major receptors and ion channels. For example, inhibition of reuptake or binding on the major monoamines (dopamine, serotonin, and norepinephrine) can be assessed in animal and human cells, as well as in animal brain structures. Potential candidates that have the desired pharmacological profile for a given disorder will then typically enter testing in animal models for efficacy (discussed in Chapter 2). Promising candidates will then often enter toxicity screening at an early stage (discussed in Chapter 3). Initially, because of the high cost of animal toxicity testing, in vitro toxicology tests will be conducted on the promising candidates before live animal studies. Thus, the activities discussed in the next two chapters will often progress somewhat concurrently, as promising candidates need to be quickly identified in today’s competitive environment and screened for safety prior to consideration for human trials. There are numerous potential CNS indications for promising compounds, but in this book the focus is on four of the major CNS indications which comprise the lion’s share of CNS drug sales today—depression, anxiety, schizophrenia, and

c01

BLBK249-Cutler

12

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

Chapter 1

Alzheimer’s Disease—as a grounding, and appreciation of these indications is fundamental to drug development issues for other CNS indications.

References 1. Sampaio C. (2008) Establishing plans for long term therapeutics benefit for treatment of Alzheimer’s disease: a European academic regulatory reviewer’s perspective. ISCTM meeting presentation; October 6–7, 2008; Toronto, Canada. 2. Alzheimer’s Association. (2007) Alzheimer’s disease facts and figures [document on the Internet]. Chicago: Alzheimer’s Association. Available from: www.alz.org. Cited February 12, 2009. 3. Hebert LE, Scherr PA, Bienias JL, et al. (2004) Alzheimer disease in the US population: prevalence estimates using the 2000 census. Arch Neurol. 60(8):1119–22. 4. Duara R. (2008) Onset of Alzheimer’s disease seen earlier in heavy drinkers, smokers. 60th Annual Meeting of the American Academy of Neurology; April 12–19, 2008; Chicago. 5. Hebert LE, Scherr PA, Bienias JL, et al. State-specific projections through 2025 of Alzheimer disease prevalence. Neurology. 62(9):1645. 6. Lieberman J, Stroup T, McEvoy J, et al. (2005) Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med. 353(12):1209–23. 7. National Institute of Mental Health. (2006) Studies offer new information about treatment choices for schizophrenia—phase 2 results [document on the Internet]. Available from: http://www.nimh.nih.gov/science-news/2006/studies-offer-newinformation-about-treatment-choices-for-schizophrenia-phase-2-results.shtml. Cited July 27, 2009. 8. National Institute of Mental Health. (2006) Questions and answers about the NIMH clinical antipsychotic trials of intervention effectiveness study (CATIE)—phase 2 results [document on the Internet]. Available from: http://www.nimh.nih.gov/health/trials/ practical/catie/phase2results.shtml. Cited July 26, 2009. 9. Stroup TS, Lieberman JA, McEvoy JP, et al. (2009) Results of phase 3 of the CATIE schizophrenia trial. Schizophr Res. 107(1):1–12. 10. Chakos M, Lieberman J, Hoffman E, et al. (2001) Effectiveness of second-generation antipsychotics in patients with treatment-resistant schizophrenia: a review and metaanalysis of randomized trials. Am J Psychiatry. 158(4):518–26. 11. Lohr JB, Braff DL. (2003) The value of referring to recently introduced antipsychotics as “second generation.” Am J Psychiatry. 160(8):1371–72. 12. Hughes B. (2007) FDA Drug approvals: a year of flux. Nat Rev Drug Discov. 7(2):107–9. 13. Hughes B. (2008) 2007 FDA drug approvals: a year of flux. Nat Rev Drug Discov. 7(2):107–9. 14. Greenhose P. (2009) Developing drugs is a costly business [document on the Internet]. Globe Newspaper Company. Available from: www.boston.com/business/technology/ biotechnology/articles/2007/05/09/developing drugs is a costly business. Cited May 8, 2009. 15. Martin B. (2008) Pharma vs. biotech: FDA-approval rates [document on the Internet]. Available from: http://bmartinmd.com/2008/03/pharma-vs-biotech-fdaapproval. html. Cited February 13, 2009. 16. DiMasi JA, Hansen RW, Grabowski HG. (2003) The price of innovation: new estimates of drug development costs. J Health Econ. 22(2):151–85.

c01

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

20:14

Char Count=

The impending crisis in CNS drug development

13

17. Adams CP, Brantner VV. (2006) Estimating the cost of new drug development: is it really $802 million? Health Affairs. 25(2):420–8. 18. FDA. (2006) FDA guidance for industry: fast track drug development programs—designation, development, and application review [document on the Internet]. Available from: http://www.fda.gov/downloads/Drugs/GuidanceCompliance RegulatoryInformation/Guidances/UCM079736.pdf Cited December 12, 2008. 19. Nemeroff C. (2008) Stigma in psychiatry, Part 1 of 2 [document on the Internet]. Available from: www.medscape.com. Cited September 17, 2008.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

CHAPTER 2

Animal models and procedures for CNS disorders

Introduction There are a large number of animal models and behavioral procedures used to test the efficacy of novel compounds in treating central nervous system (CNS) disorders. These models work under the premise that there is consistency among the physiological and behavioral characteristics of various species; therefore, extrapolation from animals to humans is possible. Animal models are used to screen compounds before they are tested in humans, and can provide insight into the mechanism of action and also provide the rationale for clinical trials. In this chapter, we will be primarily concerned with animal models that are predictive of a compound’s efficacy rather than safety or toxicity (these areas are discussed in Chapter 3). While an ideal animal model would reproduce the neuropathology, etiology, and symptoms of the condition it is modeling, this is not feasible for many CNS disorders because the neurological basis of the disorder is unknown, and animal species differ markedly in their ability to express an equivalent human condition. For example, disorders involving thought dysfunction, including anxiety, schizophrenia, and depression, are not possible to fully mimic in animals. Therefore, most animal models are only capable of reproducing certain aspects of the neurological condition they are attempting to model. Because it is not always possible to evaluate an animal model based on its faithful replication of a human CNS disorder, many models are evaluated based on their sensitivity to existing drugs which treat that disorder; i.e., using the known clinical utility of an agent in current use to infer clinical utility of a new chemical biological entity. A major drawback to this approach is that established animal models tend to work for compounds that are similar to preexisting drugs, but not for distinct and novel compounds [1, 2]. Validating an animal model provides an estimate of how relevant it is to the clinical condition it is modeling. The more limited a model is at representing the underlying complexity of a clinical condition, the more limited the range of compounds it can identify which treat that condition. A model that better approximates the neuropathology of a clinical condition will be able to identify more therapeutic compounds with a greater diversity of structure Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

14

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

15

Table 2.1 Types of validity Predictive validity

How well the results of an action (i.e., administration of a drug) on a model predict the results of the same action on the disorder

Face validity Construct validity Convergent validity

How well a model recreates the phenomenology of a condition How well a model recreates the underlying pathology of a condition How well a model’s results correlate with those from other models of the same condition

Discriminant validity

How well a model measures components of a condition that are unique from other models of the same condition

and mechanism of action because it provides more pathways and targets for treatment. It modulates processes that are more “upstream” from the clinical expression of the illness. There are many different criteria used to evaluate animal models, including predictive, face, construct, convergent, and discriminant validity, as well as general reliability (see Table 2.1 for a list of validity terms). As these concepts are commonly cited within the literature, each is separately reviewed. A model has good predictive validity if the results in the model predict the results in the clinical disorder [3]. For example, a drug which is known to alleviate the symptoms of a particular disorder in humans should also have a similar effect in the model of that disorder. In addition, there should be a low incidence of false negatives (drugs that do not work in the model but do work in humans) and false positives (drugs that work in the model but not in humans), as well as a similar potency of any observed effect. For certain classes of drugs, interspecies differences in drug metabolism, pharmacokinetics, and pharmacodynamics make predictive validity difficult. Face validity refers to how well an animal model appears to recreate the phenomenology of a clinical condition. It evaluates a model based on the expression of observable characteristics or symptoms that are similar to the human disorder it is modeling. Because of the superficial similarities it measures, and the frequent lack of correlation to underlying etiology, many consider face validity to be of limited use in determining the ability of an isolated animal model to mirror the clinical condition [4, 5]. Construct validity of a test is commonly defined as the accuracy with which an animal model recreates the disorder it is intended to model, typically in terms of the underlying neuropathology [6]. A model that accurately reproduces the neural origins of a human condition or disease has good construct validity. This is a fundamental principle of any sound animal model, but it is very difficult to determine because the pathology behind many CNS disorders is still not well understood, and biological systems as well as their interrelatedness can differ appreciably by species. Convergent and discriminant validity base their analysis of a model on comparisons with other tests of the same disorder. Convergent validity refers to how well a model’s results correlate with the results from other tests of the

c02

BLBK249-Cutler

16

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

same disorder. Discriminant validity refers to the uniqueness of a model; how well a model measures components of a disorder that are unique from those measured in other tests of the same disorder. Both forms of validity are necessary in order to have an accurate model of a disorder which does not measure the exact same parameters as a preexisting model [7]. The reliability of a model refers to the consistency and stability with which the measurements of interest are observed. Results should be consistent with each test animal used, across trials with the same animal, under similar testing conditions, and after similar manipulations [8].

The selection algorithm Every animal model has inherent strengths and weaknesses. Multiple animal models are needed for each neurological or psychiatric disorder to allow investigation into the different components of the condition and to provide a more complete assessment of the pharmacological profile of a compound designated to treat that disorder. These models complement each other without significant overlapping to provide the fullest representation of the underlying disorder. In addition, new models that address unique components of the disorders, expand on more rudimentary screens, or amplify insights regarding a particular mechanism of action may be developed. Determining the best set of models to be used for testing a novel CNS therapeutic compound can be a daunting task and the process is colloquially assumed to be “sponsor’s risk.” Specifically, since the pathophysiological processes of many neuropsychiatric conditions remain fundamentally speculative, the credence and weight given to the results of any one paradigm fundamentally reflect the belief system of the organization sponsoring that study, and the state-of-the-art methodology in existence for that therapeutic target at the time of the paradigm initiation. The evolution of therapeutic agents for the treatment of Alzheimer’s disease, for example, provides a classic example of increasingly sophisticated preclinical models suggesting efficacy as the biological data regarding Alzheimer’s disease became more extensively defined. Even here, fundamental concepts regarding the genesis of episodic disease remain argumentative in the absence of a specific therapeutic agent that would abort disease progression. The following sections describe the most widely used animal models for four prominent CNS disorders, including their strengths, weaknesses, design characteristics, output measurements, and correlations to their associated human condition.

Animal models of anxiety Many behavioral animal models of anxiety have been developed and used for the screening of compounds with anxiolytic activity. Numerous neurotransmitters and neuromodulators have been implicated in anxiety disorders, including the ␥ -aminobutyric acid (GABA)/benzodiazepine chlorine ionophore complex, serotonin and serotonin receptors (5-HT1A , 5-HT2 , and 5-HT3 ),

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

17

N-methyl-d-aspartate (NMDA), cholecystokinin, and others [9]. In animals, the distinctions between fear and anxiety are blurred, and behavioral tests assumed to be a proxy for fear are used to measure anxiety. Two categories of fear are measured; conditioned fear (i.e., fear that is learned by association with an aversive stimulus) and unconditioned fear (i.e., fear that is spontaneous and not learned). Anxiolytic compounds are evaluated for their ability to reduce fear-associated behaviors in animals compared to control groups in the absence of other adverse pharmacological effects, using models which have historically produced reliable estimates of drug efficacy in human clinical trials. Anxiety presents as an emotion in a spectrum of disorders, which includes generalized anxiety disorder, panic attacks, and obsessive–compulsive disorder (OCD). Tests that are sensitive to certain categories of anxiety may not be sensitive to others. This section summarizes some of the most widely used unconditioned and conditioned fear tests in animals. Anxiolytic compounds are listed that produce the most consistent and representative results. Test limitations with the major rodent strains will be discussed. See Table 2.2 for a review of the animal models and Table 2.3 for a review of the genetic models of anxiety covered in this section.

Open field and light/dark transition tests Rodents typically have a desire to explore novel environments, yet avoid brightly lit, open spaces, which are believed to induce anxiety [10]. The open field test takes advantage of these two characteristics by placing a rodent in a brightly lit, open environment where it cannot return to its home cage. Each rat is left in a corner of the environment and its behavior is recorded for 3–15 minutes. Measurements include total distance traveled in the field, average speed, rearing/elongation behavior, and time spent in various parts of the environment. Increased anxiety is associated with a decrease in locomotor activity and rearing, more activity in the periphery versus the central part of the testing area (thigmotaxis), and defecation [11]. Anxiolytic compounds are evaluated based on their ability to reduce these behaviors. The paradigm has been shown to be generally sensitive to classical treatments for anxiety disorder, as mentioned above [12]. It does not appear to work for selectiveserotonin reuptake inhibitors (SSRIs) that have a different spectrum of therapeutic efficacy in anxiety disorders such as panic attacks, generalized anxiety disorder, or OCD [13]. The light/dark transition test is similar to the open field test, and also capitalizes on an animal’s fear of brightly lit spaces. The test apparatus consists of a dark compartment connected to an adjacent illuminated compartment. Measurements include the latency to leave the dark compartment, the time spent and distance walked in the light compartment, and the number of crossings between the two compartments. Anxiolytic compounds are evaluated based on their ability to increase the amount of time the animal spends in the light compartment. Classic anxiolytics (benzodiazepines) as well as the

c02

BLBK249-Cutler

18

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

Table 2.2 Animal models of anxiety Major responsive drug classes

Limitations

Fear-associated behaviors due to bright, open environments can be ameliorated with anxiolytics

BZDs1 , 5-HT1A agonists and 5-HT2 antagonists

Certain SSRIs unresponsive or even anxiogenic

Elevated maze tests

Avoidance of elevated, unenclosed maze arms reduced by anxiolytics

BZDs, 5-HT1A agonists

Strain/species specific; BZD produces attenuation

Predator exposure tests

Escape/risk assessment behavior in response to a predator reduced by anxiolytics

BZDs, 5-HT1A agonists, 5-HT2 antagonists

Reliance on olfactory cues may be confounding; SSRI results contradictory

Fear-potentiated startle (FPS) test

Startle reflex in response to a shock-associated tone reduced by anxiolytics

BZDs, 5-HT1A agonists, SSRIs

Low predictive validity: responsive to some nonanxiolytics

Punishment/ conflict test

A pleasurable behavior attenuated by punishment is increased with anxiolytics

BZDs, 5HT2A agonists, SSRIs

5-HT1A agonists, 5-HT3 antagonists less responsive

Passive avoidance test

Exploratory behavior attenuated by shock administration is increased by anxiolytics

5-HT agonists, 5-HT2A antagonists, SSRIs

Limited response to BZDs; responsive to some nonanxiolytics

Social interaction test

Decreased social behavior under a stress-inducing environment is corrected by anxiolytics

BZDs (acute only), 5HT1A agonists, and 5-HT2 antagonists

Limited response to chronic BZD administration; SSRIs can be anxiogenic

Test type

Description of test

Open field test

1

BZD, benzodiazepines.

newer anxiolytic-like compounds (e.g., serotonergic drugs or drugs acting on neuropeptide receptors) can be detected using this paradigm.

Elevated maze tests The elevated plus maze balances an animal’s fear of heights and open spaces with its natural tendency to explore novel environments [14, 15]. A mouse or a rat is put on an elevated, plus-shaped platform that has two enclosed arms and two open, unenclosed arms. The open arms are believed to trigger anxiety in rodents. This is supported by higher blood corticosterone levels (the equivalent of human cortisol levels), increased freezing of motion, and increased defecation in animals confined to this part of the maze [16]. Anxiolytic compounds are tested for their ability to increase the time the animal spends

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

19

Table 2.3 Genetic models of anxiety Most responsive drug classes

Model type

Characteristics of model

Limitations

CRH-OE mice

Increased heart rate, blood pressure, acoustic startle response, and freezing Facilitated conditioned fear, decreased exploration, and feeding. Excess fat accumulation, muscle atrophy, thin skin, increased heart rate, and hair loss

CRH1 receptor antagonists

Exhibit inconsistencies in different behavioral tests, suggesting that CRH overexpression does not produce a clear anxious phenotype

5-HT1A receptor knockout mouse

Increased anxiety-related behavior in several standard anxiety tests, less activity than wild-type mice under stress-inducing conditions

Sometimes responsive to benzodiazepines

Inconsistent results across different strains and different behavioral tests

RGS2

Increased anxiety on standard behavioral tests

Unknown

Unknown

exploring the open arms versus the enclosed arms. Measurements include the time spent in each arm, the number of entries into each arm, and the number of times the animal crosses through the center of the maze. This test has several notable limitations [17]. Different rat strains show varying sensitivities to anxiolytic compounds on the test, as do different habituation procedures. In addition, the anxiolytic effects of benzodiazepines are attenuated on the second trial of the test for both rats and mice, possibly due to the induction of a phobic state against which benzodiazepines are ineffective [18]. The elevated T-maze is a variant of the plus maze that combines both conditioned and unconditioned fear responses in one test [19]. The maze consists of three arms; one enclosed and two open. Conditioned fear is measured in terms of inhibitory avoidance, which is when an animal learns to avoid a fearinducing stimulus. It takes an animal longer to move from the enclosed arm of the maze to the open arms with repeated trials, suggesting that the animal has learned that the open arms represent a fearful stimulus [20]. Unconditioned fear is measured in terms of escape time, which is the time it takes an animal to move from the open arms to the enclosed arms. The time to leave the open arm does not undergo habituation over consecutive trials, thereby providing evidence of an unconditioned response. The test is responsive to benzodiazepines, 5-HT1A agonists, and 5-HT2 antagonists, among many other compounds. In addition, different compounds affect inhibitory avoidance and escape behavior independently on the test, suggesting that each represents a unique aspect of anxiety. Graeff et al. believe that the pattern of responsive drugs suggests that inhibitory avoidance in the elevated T-maze is related to generalized anxiety disorder, while the escape behavior is associated with

c02

BLBK249-Cutler

20

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

panic disorder [21]. The test appears to work much more consistently with rats than with mice [22, 23], but also varies with the emotional reactivity of the rat strain being used [24]. A drawback with this model’s predictive validity is that SSRIs appear to be anxiogenic on this test [25].

Predator exposure tests There are a large variety of tests that induce anxiety by exposing a rat or mouse to the odor and/or sight of a natural predator. The mouse defense test battery (MDTB) is designed to model both general anxiety disorder and panic disorder. A mouse is placed inside an oval runway, and a deeply anesthetized rat, held by the experimenter’s hand, is approached toward the mouse at a fixed speed. The mouse generally flees at a distance of 1 m in what is assumed to be a panicked response. The mouse undergoes risk assessment behavior characterized by abrupt movement arrest and orientation toward the approaching rat. This is believed to correlate with general anxiety disorder. The pharmacological profile of the MDTB evidences its high predictive value. Flight is reduced by benzodiazepines highly active on panic disorder, while a 5-HT1A agonist and a 5-HT2 antagonist, which are effective for general anxiety disorder, reduce risk assessment behavior but not flight behavior. SSRIs give contradictory results in this test [26, 27]. The fear/defense test battery and the anxiety/defense test battery are similar to the MDTB in design and evaluation, except that they both use a rat as the test subject and a cat or cat odor as the predatory stimulus [28, 29]. A potentially confounding aspect of this test is the animal’s reliance on olfactory cues, which has no correlate in humans. Fear-potentiated startle (FPS) In the fear-potentiated startle (FPS) paradigm, the animals are trained to associate a neutral conditional stimulus (such as a light or a tone) with an aversive stimulus (such as a shock). After a few pairings, the conditioned stimulus induces a state of fear that can be measured by a potentiation of the startle reflex. The startle reflex is a chained series of rapid extensor and flexor movements, which is strongly conserved across species, and can be measured in rats using specially designed cages [30, 31]. Many classes of anxiolytics, including benzodiazepines [32], 5-HT1A agonists [33], and SSRIs [34], have been shown to reduce the potentiated startle response, and the test is generally considered a predictive model of the effects of anxiolytic compounds in humans. False positives have been reported, including with the 5-HT3 agonist ondansetron [35], morphine, and haloperidol [36, 37]. In addition, benzodiazepines do not reduce FPS in humans [38]. Therefore, positive results on this test do not always imply that a compound will have anxiolytic properties in humans. Punishment/conflict Punishment/conflict models of anxiety contrast the motivation to perform a certain behavior (typically hunger, thirst, or exploration) with the fear of a punishment which follows that behavior. This paradigm is one of the most

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

21

commonly employed in the screening algorithms for anxiolytic agents. The frequency of response behavior is reduced in control animals following the introduction of the punishment. Anxiolytic compounds are tested for their ability to increase the frequency of response under these conditions in the absence of appreciable sedation. There are many different types of punishment/ conflict models. Typically, punishment/conflict models show good response with benzodiazepines and are less consistent with other classes of anxiolytics. The standard tests in this category are the Vogel conflict test and the conditioned conflict test. The Vogel conflict test measures the ability of anxiolytics to increase the drinking behavior of water-deprived rats exposed to mild foot shocks every time they initiated drinking. Drinking behavior was increased following administration of benzodiazepines and 5-HT2B agonists, while 5-HT1A agonists and 5-HT3 antagonists did not affect the response consistently [39–42]. SSRIs were also effective anxiolytics on this test [43]. The conditioned conflict test alternates punishment and nonpunishment conditions with food or water delivery [44]. During the nonpunishment conditions, animals pressing a lever receive food or water. During the punishment conditions, animals pressing a lever receive food or water accompanied by a foot shock. Well-trained animals learn to push the lever only during nonpunishment conditions. Anxiolytics are evaluated for their ability to restore lever pressing during the punishment condition. Similar to the Vogel conflict test, benzodiazepines work well on this test in most species, while 5-HT1A agonists do not [45]. There are numerous varieties of this test, and most show the same limitations in regard to nonbenzodiazepines.

Passive avoidance The passive avoidance model applies a mild foot shock to an animal when it steps into a foreign chamber, creating an attenuated exploratory response when the animal is allowed to enter the same chamber in subsequent trials. Anxiolytic compounds are tested for their ability to diminish this passive avoidance response. The model is typically sensitive to 5-HT agonists, 5-HT2A antagonists, SSRIs, and somewhat sensitive to benzodiazepines [46– 50]. A drawback with this test’s predictive utility is the wide range of nonanxiolytic compounds that can affect an animal’s performance. For example, drugs that affect the learning process underlying the acquisition of conditioned fear, such as apomorphine and scopolamine, can disrupt performance on this test without displaying anxiolytic properties [51, 52]. Social interaction test In the social interaction test, unfamiliar pairs of rats are placed in novel environments and their social interactions are recorded. Interactions include sniffing, nipping, grooming, following, mounting, kicking, boxing, and crawling under and over each other [53]. Social interactions are reduced in anxietypromoting conditions, such as under bright lights. In addition, fear-associated

c02

BLBK249-Cutler

22

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

behaviors, such as freezing and defecating, are increased. Anxiolytic compounds are tested for their ability to increase social interactions and reduce fear-associated behaviors under anxiety-inducing conditions. The social interaction test is believed to be a good model of human anxiety because it does not involve conditioning, deprivation, or punishment, but rather examines rat behavior during the course of their normal interactions [53]. The drawbacks of the model are that benzodiazepines only function as anxiolytics when administered acutely and SSRIs are anxiogenic on the test [54, 55].

Inbred/selectively bred rodent strains Numerous commercially available inbred mouse strains show heightened anxiety levels in comparison to wild-type controls, and these strains have frequently been used to test the efficacy of novel anxiolytic compounds in behavioral paradigms. For example, most of the 129 mouse substrains show heightened anxiety levels when compared to their C57BL/6J progenitor strain [56–58]. Other inbred strains showing heightened anxiety include DBA/2, BALB/c, LP, and BDP mice [59, 60]. Some mice and rat strains have been selectively bred to produce higher anxiety phenotypes. These include Louis, HAB, LAS, and SLOW rats [61–64]. These strains are believed to model anxiety disorder better than single gene knockouts because their phenotype arose spontaneously and likely does not reflect only a single gene [65]. A major caveat with utilizing these strains (or any animal) in preclinical trials of anxiolytic compounds is that even the best validated models can produce variable results. This is because there are many factors that can affect an animal’s anxiety level, including the prenatal environment, the type of maternal care, the animal provider, transport conditions to the laboratory, the specific test conditions, etc. [66–69]. Therefore, any mouse or rat strain receiving an anxiolytic compound should be tested in more than one behavioral model. Genetic models of anxiety Increased anxiety-like behavior has been exhibited in mice with genetic alterations to monoanergic, corticotropic, gabaergic, and peptidergic systems [70]. In addition, genetic manipulation of neurotrophic-type molecules and regulators of intracellular signaling and gene expression have also induced anxiety-like behaviors in rodents [71]. The complexity of anxiety, as with any mood disorder, makes it extremely difficult to replicate all facets of the disorder with a single genetic manipulation. The combination of numerous genes across diverse neurotransmitter systems and molecular pathways may better mimic the overall phenotype. Single-gene manipulations allow for animal models that represent at least facets of anxiety disorder, as indicated by their increased anxiety-associated phenotypes on behavioral tests in comparison with control animals. There is a large conservation of anxiety-associated genes between humans and mice, indicating that the neural basis of anxiety in rodents may be similar to that of humans [72, 73]. Despite this similarity, few genetic models of anxiety disorder have been validated enough to have

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

23

utility as markers of efficacy during preclinical trials. A selection of prominent genetic models of anxiety disorder is described in this section.

CRH-OE mice Corticotropin-releasing hormone (CRH) is a hormone and neurotransmitter involved in the body’s response to stress and implicated in anxiety-related behavior [74]. CRH administration in mice increases heart rate, blood pressure, and acoustic startle response. It also increases freezing, facilitates conditioned fear, and decreases exploration and feeding [75]. Transgenic mice overexpressing CRH (CRH-OE) exhibit elevated plasma corticosterone levels, a marked increase in plasma adrenocorticotropin hormone (ACTH), excess fat accumulation, muscle atrophy, thin skin, increased heart rate, and hair loss [76]. In addition, one line of CRH-OE mice has shown chronic stress-like neuroendocrine and autonomic changes [77]. A major problem with this model is that CRH-OE mice exhibit inconsistencies in different behavioral tests, suggesting that CRH overexpression does not produce a clear anxious phenotype. For example, CRH-OE mice showed significantly less freezing than wild-type (WT) mice in fear-conditioning tests, which is in direct contradiction to their increased freezing in avoidance tests [78]. In addition, CRH knockout mice show normal stress-induced behavior [79]. Currently, anxiolytics are being developed which target the CRH pathway, such as CRH1 receptor antagonists [80]. These compounds have been shown effective at reducing the symptoms of anxiety, yet their effects on the neuroendocrine pathway remain unclear [74]. While CRH appears to be involved in an animal’s response to stress, its role may be modulatory instead of critical. Further characterization of this transgenic model is needed before it can be used as a measure of efficacy for novel anxiolytics. 5-HT1A receptor knockout mouse The serotonergic system has been strongly implicated in anxiety disorder, specifically the 5-HT1A receptor [81]. 5-HT1A receptor dysfunction has also been linked to panic disorder in humans [82]. 5-HT1A knockout mice display increased anxiety-related behavior in several standard anxiety tests [83–85]. They also displayed less activity than WT mice under stress-inducing conditions. The 5-HT1A knockout mouse requires further validation before it can be used as a predictor of anxiolytic efficacy, because it has shown inconsistent results across different strains and different behavioral tests. One strain of 5-HT1A knockout mice did not respond when injected with either diazepam [86] or fluoxetine [87], while another strain responded to diazepam [88]. In addition, approach–avoidance models have produced false positives and false negatives when tested with 5-HT1A receptor agonists [65], suggesting that they may also be inconsistent with 5-HT1A knockout mice. The light/dark test, stress-induced hyperthermia, and a classical fear-conditioning test all failed to produce behavioral differences between WT mice and a strain of 5-HT1A

c02

BLBK249-Cutler

24

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

knockout mice, suggesting that these mice do not display a clear anxious phenotype [89, 90].

RGS2, CCK2, and NK1R There are several genes that have recently been linked to anxiety disorder, with transgenic mouse models showing increased or decreased levels of anxiety on standard behavioral tests, and observed correlations between these genes and anxiety-associated behaviors in humans. While none of these transgenic mouse models are currently being used to test the efficacy of novel anxiolytics, we believe they are worth mentioning because their protein products are being considered as drug targets for anxiolytics, and they could also serve a unique role as animal models of anxiety once validated. The RGS family member RGS2 is an immediate early-response immuneassociated gene upregulated in human lymphocytes after activation [91]. It also plays a role in anxiety, as male RGS2 knockout mice exhibit increased anxiety on standard behavioral tests [92, 93]. RGS2 is associated with intermediate phenotypes for anxiety disorder in humans, making it a good target candidate for novel anxiolytic compounds [94]. Cholecystokinin2 (CCK2 or CCKB ) is a neurotransmitter in the brain associated with anxiety disorder. CCK2 agonists are anxiogenic in rats, mice, guinea pigs, cats, and monkeys [95–98], and CCK2 antagonists attenuate anxious behavior in animals [99] and panic disorder in humans [100]. CCK2 peptide and nonpeptide antagonists have also been tested in preclinical and clinical trials as novel anxiolytics with encouraging results [100–102]. CCR2 knockout mice displayed reduced anxiety on the elevated plus maze and the light/dark transition test [103, 104]. In addition, CCK hyperactivity appears to be linked with submissive behavior, and CCK2 displays increased receptor expression and binding in suicide victims and animal models of anxiety [105]. A mouse model overexpressing CCK2 may be useful in validating the in vivo targets of novel CCK2 antagonists. Neurokinin 1 (NK1R) is a mammalian tachykinin receptor highly expressed in the limbic system that preferentially binds to neuropeptide substance P. Both NK1R and Substance P are linked with anxiety-related behaviors. Stress is associated with high levels of Substance P in animals and humans, and injection of Substance P into the brain influences anxiety levels [106, 107]. Additionally, NK1R antagonists show anxiolytic effects in animal models of anxiety, and knockout models of NK1R decrease anxiety-related behaviors in animals [108, 109]. Preclinical studies of NK1R antagonists demonstrated anxiolytic effects [110]. An overexpression model of NK1R could serve as a unique model of anxiety and help validate the targets of NK1R antagonists.

Animal models of depression Clinical depression is a disorder characterized by dysfunction in several neurotransmitter systems, including serotonin, norepinephrine, and dopamine.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

25

Given the relationships that exist between these systems, there are ample hypotheses to be addressed for innovative therapeutic agents (e.g., peptide antidepressants such as nemifitide). In addition, clinically depressed patients have shown abnormalities in regional cerebral blood flow and glucose metabolism (surrogates of neuronal function) in the limbic cortex, the prefrontal cortex, the hippocampus, the amygdala, and the anterior cingulate cortex, among others [111, 112]. This further supports the “overdetermined” or multifactorial nature of neurochemical abnormalities for this condition (an event is overdetermined if there exist more than one antecedent events, any of which would be a sufficient condition for the event occurring). Clinical depression differs from stress-induced, transient depression, because in the latter, these systems function normally and homeostatic mechanisms remain intact [113]. Animal models of clinical depression should exhibit similar imbalances in these known neurotransmitter systems and brain regions, as well as demonstrate behavior that correlates with a depressive state in humans. Animal models should also respond to treatment with antidepressants in a manner similar to humans. However, rejection of a putative compound based on lack of biochemical correlates would be imprudent if behavioral effects comparable to established antidepressants were demonstrated. This section reviews the major animal models and genetic models of depression, which are summarized in Tables 2.4 and 2.5, respectively.

Behavioral despair models The forced-swim test (or Porsolt test) is the most common behavioral despair model and the most widely used animal model in depression research [114, 115]. It has been found to be very responsive to antidepressant treatment. A rat or mouse is placed in a tank filled with water from which it cannot escape, and the time it takes for the animal to stop struggling and become immobile (besides movements to simply stay afloat), as well as the duration of immobility, is recorded. Rats can be tested in two trials separated by 24 hours (15 minutes for the first trial, 5 minutes for the second trial), with recording for immobility during the second trial. Mice can also be tested in one trial lasting 6 minutes, with recording for immobility done during the last 4 minutes of the trial [116]. The immobile state within the above model is believed to correlate with induced despair. Antidepressants are tested for their ability to increase the latency to immobility and to decrease the time spent immobile. They are hypothesized to work by restoring the animal’s active coping mechanisms and decreasing stress-induced freezing behavior [117]. Some studies have suggested that an animal’s immobility reflects its attempt to conserve energy, thus refuting the notion that the test is a measure of induced despair [118, 119]. Despite potentially low construct validity, the test has some predictive utility, and has been found to be responsive to 90% of clinically active antidepressants, including tricyclic antidepressants (TCAs), monoamine oxidase inhibitors (MAOIs), and atypical antidepressants, although it is less responsive

c02

BLBK249-Cutler

26

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

Table 2.4 Animal tests of depression Most responsive drug classes

Test type

Description of test

Limitations

Behavioral despair models: forced swim, tail suspension

A rodent placed in an inescapable, unpleasant situation exhibits signs of despair which can be reduced by antidepressants

TCAs, MAOIs, and atypical antidepressants

Less responsive to SSRIs; strain specific; false positives

Learned helplessness

Animals shocked without ability to escape will not attempt escape even when it’s made available in subsequent trials. Antidepressants increase escape attempts

TCAs, MAOIs, and SSRIs

May have poor construct validity

Chronic mild stress-induced anhedonia model

Chronic, mild stress reduces response to rewarding stimuli (anhedonia). Response can be increased with antidepressants

TCAs, SSRIs, NA reuptake inhibitors, MAOIs, atypical antidepressants

Takes up to 6 months; results often variable

Olfactory bulbectomized (OB) rat

Surgical ablation of olfactory bulbs in rats produces impaired acquisition of passive avoidance. This can be ameliorated with antidepressants

TCAs, 5HT1A agonists, SNRIs, NA reuptake inhibitors, atypical antidepressants

Not responsive to MAOIs; limited face validity

Social situation tests: defeat, hierarchy, separation, adult isolation

Rodents placed with dominant member of the same species display submissive behaviors that can be corrected with antidepressants

TCAs, SSRIs

Responsive to anxiolytics. Involves chronicity; complex exptl. design; large number of animals

to SSRIs [120, 121]. Results have been shown to vary widely with the strain of the animal selected, therefore selecting the strain with the largest available database for both active and inactive agents is essential for obtaining meaningful results [122–124]. False positives caused by generalized increases in mobility have been shown with caffeine, levodopa (L-dopa), and morphine (among others), but these can usually be distinguished by an additional locomotor activity test [125]. Antidepressants that increase serotonergic neurotransmission typically increase swimming behavior, whereas those that increase catacholaminergic neurotransmission typically increase climbing behavior [126]. The model has also been shown to be capable of differentiating between antidepressants and anxiolytics [114]. The tail suspension test is another behavioral despair model. This very simple test has been found to be highly responsive to the acute effects of

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

27

Table 2.5 Genetic models of depression Most responsive drug classes

Model type

Characteristics of model

Limitations

Flinders sensitive line of rats (FSL)

Reduced appetite and psychomotor functions, sleep and immune abnormalities, increased immobility, learning difficulties, cholinergic, and circadian rhythm dysfunction

TCAs, SSRIs, and PDE-4 inhibitors

Only addresses acetylcholine dysfunction

“Helpless” line of mice (HL)

Increased immobility, anhedonia, sleep/wakefulness alterations, neurochemical imbalances characteristic of depression

TCAs, SSRIs

Model still relatively new and untested

WAG/Rij mice

Reduced exploratory behavior, increased immobility, anhedonia

TCAs, dopamine receptor agonists

Dopamineric specific model; copresents with absence seizures

Fawn-Hooded rat

Increased immobility, decreased social interactions, high corticosterone levels

TCAs, MAOIs, SSRIs

Model confounding; copresentation of alcoholism and social phobia

antidepressants. A mouse or rat is suspended by its tail for 6 minutes and the amount of time it spends immobile and not struggling is recorded. Putative antidepressants are tested for their ability to decrease the amount of time an animal spends immobile, and this test is generally sensitive to the same category of antidepressants that is responsive on the forced-swim test [127, 128].

Learned helplessness Learned helplessness models of depression are based on the observation that animals exposed to a series of shocks which they cannot escape will cease attempting escape, even when an exit is later made available to them. This behavior is not dependant on the shocks but rather on the circumstances associated with the shocks, because animals provided with an exit while being shocked still attempt escape in later trials. It is believed that animals which are not given the option to escape learn to become helpless under similar circumstances [129]. This helplessness may be the result of an acquired deficit in motor activity as opposed to a generalized depressive disorder, because animals shocked without an exit are still capable of learning escape behavior if the task is simple (i.e., requires little motor activity). Whether this deficit is

c02

BLBK249-Cutler

28

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

specific to the motor system, or whether it is a generalized threshold effect that results from a depressive disorder, remains to be determined [130]. Regardless of this, this test possesses strong predictive validity: numerous categories of antidepressants are capable of promoting escape behavior in animals that have undergone learned helplessness, including subchronic administration of TCAs, MAOIs, and SSRIs [131–133]. The test also shows good specificity and can distinguish anxiolytic compounds from antidepressants [134].

Chronic mild stress-induced anhedonia model The chronic mild stress-induced anhedonia model is one of the most widely used and best validated animal models for identifying novel antidepressant compounds. It uses mild, chronic stressors (confinement, food/water deprivation, overcrowding, background noise, continuous illumination, etc.) to serve as trigger factors for depression in animals [135]. When applied over a period of several weeks, chronic stressors have been shown to reduce the sensitivity to rewarding stimuli in rats and mice (i.e., decreasing consumption of a normally consumed sucrose medium). This behavior is believed to relate to a core symptom of depression in humans called anhedonia—the inability to derive pleasure from normally pleasurable activities. Antidepressants are tested for their ability to restore a stressed animal’s sensitivity to a rewarding stimulus back to baseline levels (i.e., increase consumption of a sucrose medium to control group consumption) [136]. Most classes of antidepressants have been found effective on the chronic mild stress-induced anhedonia model, while stimulants, neuroleptics, and anxiolytics are ineffective in the test [137, 138]. The model displays good construct and face validity due to the similarity of the behavioral condition in humans (anhedonia) and induction method to human clinical depression [139]. Drawbacks of the model are that it takes a long period of time and a lot of effort to establish anhedonia in an animal (typically weeks to months). The results vary depending on the procedure used, the testing facility, and the strain and species of the animal used. In addition, the model sometimes produces results that are opposite of what it is attempting to measure; such as increased locomotor activity in the open field test [140–142]. Olfactory bulbectomized (OB) rat Surgical ablation of the olfactory bulbs in rats creates impaired acquisition of passive avoidance behavior. Passive avoidance is the inhibition of a previously exhibited response; e.g., displaying attenuation of movement when placed in a brightly lit, open field. OB rats do not show attenuation of movement, but instead show hyperactivity and heightened aggression [143]. Bulbectomy appears to cause a major dysfunction of the corticalhippocampal-amygdala circuit that underlies these behavioral abnormalities. These neuroanatomical areas also seem to be dysfunctional in patients with major depression, thus producing a model of depression with strong construct validity [144–146]. Administration of antidepressants largely corrects most of the behavioral, endocrine, immune, and neurotransmitter changes that occur

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

29

following bulbectomy, and restores passive avoidance behavior in these animals [147]. The OB model also possesses significant face and predictive validity. Many classes of antidepressants work on this model. In addition, it produces behavioral changes that respond to chronic, but not acute treatment with antidepressants, which closely mimics the time course of antidepressant treatment in humans [148]. SSRIs show less consistent effects on this model; some studies suggest that SSRIs work acutely to restore passive avoidance, while others suggest that chronic administration is effective [149, 150].

Social situation tests There are many tests that take advantage of the complex social dynamics in mice and rats to model depressive phenotypes. The social defeat model takes advantage of the similarities between human depression and animal submissive behaviors [151]. There is considerable evidence that depressed individuals see themselves as inferior and behave submissively, and assertiveness training is a major component of psychotherapy for depression [152]. A male mouse experiencing a single defeat by a dominant mouse of the same C57BL/6J strain exhibits a gradual increase in passive behavior in response to a stressor, as well as higher immobility times on the force swim test [153]. Likewise, male rats dominated by a more aggressive strain lose their dominant status to previously submissive rats of the same strain, implying wide-ranging changes in behavior [154]. Both of these defeat-induced phenotypes can be reversed by administering the tricyclic antidepressant imipramine, suggesting that these animals may be undergoing some form of depression. In other studies, defeated mice have undergone reduced growth, increases in core temperature and circulating corticosteroid levels, hypophagia, weight gain, and increased measures of anxiety and chronic stress, which can be reduced by SSRIs [155–157]. The social defeat model is still undergoing standardization for use in screens of putative antidepressants. In the social hierarchy model, rats housed in close groups undergo a social hierarchy which can be determined by measuring each member’s success during antagonistic encounters with other members of the same group. Chronic administration of antidepressants results in an increase in the ranking of the animal in the group, likely due to its greater assertiveness during social encounters [158]. Daily assessments allow for the time course of chronic antidepressant treatment to be evaluated. This test is problematic because there are many characteristics of subordinate animals that contribute to their social status, as evidenced by the fact that anxiolytics also increase social ranking [159]. Chronic depression is often reported after the loss of a loved one, which has led to models of depression based on social separation in animals. In nonhuman primates, and to some extent in cats, dogs, rodents, guinea pigs, and precocial birds (offspring born relatively mature and independent), the separation of infants from their parents (neonatal isolation) produces behaviors similar to anaclitic depression (relating to or characterized by a strong emotional dependence on others) observed in institutionalized children, including

c02

BLBK249-Cutler

30

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

decreases in activity, appetite, play, and social behaviors [160]. Antidepressant treatment in isolated nonhuman primates reverses some, but not all of these effects, likely due to the inherent complexities of primate social behavior [161– 164]. Guinea pigs and chickens have served as better models, with antidepressant treatment able to reverse the effects of isolation on the vocal calls of chicks and guinea pig pups [165, 166]. Paradoxically, it is the species that is most removed evolutionarily from humans who have the highest predictive utility. Adult isolation has also been used as an antidepressant screen in rats. Isolated adult rats show impaired social cooperative behavior, similar to depressed humans [167, 168]. Chronic treatment with the tricyclic antidepressant imipramine or the SSRI fluoxetine successfully ameliorated this impairment [169].

Genetic models of depression There are numerous genetic models of depression in rats and mice. Unfortunately, genetic models may have limited construct, face, and predictive validity in the assessment of antidepressant effects. Their utility in the discovery process may be easily supplanted by the more commonly available and generally less resource-intensive paradigms previously reviewed, although much research is going on in this area. This subsection reviews some of these genetic models of depression, focusing on the models with the highest validity rankings. The Flinders sensitive line (FSL) of rats was selectively bred for increased response to anticholinesterase agents, and has since become a model of depression [170]. The FSL rat has good face and construct validity. It displays reduced appetite and psychomotor functions, sleep and immune abnormalities, learning difficulties, cholinergic, and circadian rhythm dysfunctions similar to those seen in depression [171, 172]. The FSL rat demonstrates exaggerated immobility when exposed to stressors such as a foot shock and forced swimming. The chronic treatment of FSL rats with various antidepressants (including tricyclics, SSRIs, and PDE-4 inhibitors) successfully reverse this immobility [172, 173]. Treatments not typically used for depression (such as lithium and exposure to bright lights) have no effect on this model, showing good predictive validity. A drawback is that this model only addresses acetylcholine dysfunction, while clinical depression likely involves the interplay between several different neurotransmitters. Therefore, this model may only represent one component of this complex disorder. The “helpless” (HL) line of mice displays numerous behavioral and neurological characteristics similar to individuals with chronic depression. Like the FSL mice, this model shows good construct, face, and predictive validity. The HL mice are essentially immobile in the tail suspension test and the forcedswim test, and show reduced consumption of a sucrose solution in comparison to controls [174]. They exhibit sleep–wakefulness alterations resembling those classically observed in depressed patients, display higher basal corticosterone levels, and decreases in serotonergic tone similar to that seen in

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

31

depressed humans [175]. In addition, these behavioral dysfunctions and neurochemcial imbalances can be improved by chronic treatment with noradrenergic or serotonergic antidepressants [176]. WAG/Rij mice are similar to FSL and HL models of depression, exhibiting behaviors that are associated with a depressive phenotype. They show reduced investigative activity in the open field test, increased immobility in the forced swimming test, and decreased consumption and preference for a sucrose solution [177]. Chronic administration of the dopamine agonist parlodel reversed these depressive characteristics, suggesting that the WAG/Rij rat is a dopamine-dependent model of depression [178]. A confounding aspect of this model is that depressive symptoms usually copresent with absence of seizures (and possibly anxiety disorder), reducing the construct validity of this model [179]. The Fawn-Hooded rat is an inbred strain of rat that expresses many characteristics of depression, including increased immobility in the forced-swim test [180]. Chronic treatment with antidepressants successfully reduces this immobility. A major drawback with this rat strain is that it appears to model numerous disorders at once. It has been shown to be a useful model of alcoholism, because the rat exhibits high voluntary ethanol intake and experiences typical withdrawal symptoms when alcohol is removed [181]. It also shows enhanced social phobia and may be a model of anxiety [182]. Due to these confounding phenotypes, the Fawn-Hooded rat is not typically used to test antidepressants, but it can be a useful model for testing compounds believed to affect multiple disorders (i.e., alcoholism and depression). There are numerous other genetic models of depression at various stages of development, including hypothalamic-pituitary-adrenocortical (HPA) transgenics, 5-HT transporter knockouts, CRH receptor subtype knockouts, and tachykinin receptor knockouts, but the majority of these have no proven validity and are not routinely being used for testing putative antidepressants [183].

Animal models of schizophrenia Schizophrenia is extremely difficult to model in animals due to its many symptoms that require self-reports to identify, including delusions, hallucinations, and thought disorders. There is also concern that animals such as mice and rats are not complex enough biologically to accurately reproduce these symptoms, excluding the inability of the animal to provide a behavioral equivalent of a self-report. In addition, heterogeneity in clinical symptoms, course of the disorder, and speculative causative factors are significant hurdles in creating a definitive animal model of schizophrenia. Because of these difficulties, animal models of schizophrenia tend to focus on specific causative or mechanistic hypotheses regarding the disease. Many of these models involve administration of a drug to an animal to produce correlates of schizophrenialike symptoms.

c02

BLBK249-Cutler

32

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

There are two categories of symptoms generally associated with schizophrenia; positive and negative. Positive symptoms of schizophrenia are those that occur in addition to normal experiences, such as hallucinations, illusions, and delusions. Negative symptoms involve components of normal experiences that are lacking or inhibited in schizophrenia patients, such as delayed thinking, talking and moving, indifference to social contact, disrupted sleeping patterns, abnormal body language, an inability to experience pleasure, and increased indifference [184]. Schizophrenia patients can experience symptoms in predominantly one or both categories, and that presentation may be unstable through time. Animal models and genetic models of the disease try to recreate both categories with varying degrees of success. Because there isn’t one unifying theory on the pathogenesis of this disorder, a huge number of models exist. This section covers the animal and genetic models of psychosis with the best construct, face, and predictive validities, which are summarized in Tables 2.6 and 2.7.

Dopaminergic-induced psychosis Dopaminergic transmission defects have been hypothesized to be a major underlying cause of schizophrenia, particularly within the mesolimbic and mesocortical systems, and there is a large body of evidence that supports the use of animal models that amplify dopaminergic transmission within the central nervous system [185–189]. Studies have found increased L-dopa decarboxylase levels in schizophrenia, which transforms the precursor L-dopa into the neurotransmitter dopamine (DA) [190]. In addition, most typical antipsychotics are DA receptor antagonists (receptor type D2 specifically), and there is a strong correlation between clinical efficacy and the degree of DA receptor antagonism [191–193]. DA receptor density has been reported to be greater in untreated and drug-naive schizophrenia patients [194, 195]. In addition, D2 dopamine receptors receive increased stimulation by dopamine in schizophrenia patients versus normal controls, particularly during acute phase of the disease [194, 196]. Dopaminergic animal models involve the application of a chemical that produces abnormal transmission of dopamine (i.e., amphetamine, apomorphine, steroids, etc.) in the CNS. The best-known pharmacological model of schizophrenia involves systemic amphetamine injection into rats, which stimulates DA release and is associated with an increase in a variety of behavioral effects that may provide a correlate of psychotic symptoms [197, 198]. These symptoms include increases in locomotion and stereotyped behaviors (i.e., continuous sniffing or licking), shakes, parasitotic-like grooming episodes, and a variety of other forms of hallucinatory-like behaviors [199, 200]. Amphetamine also exacerbates psychotic symptoms in schizophrenic patients at levels that are nonpsychogenic in controls [201]. Antipsychotics that block the D2 dopamine receptor can alleviate the symptoms produced by amphetamine administration. The amphetamine model helped identify typical neuroleptics for the treatment of schizophrenia for first-generation antipsychotics, and is

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

33

Table 2.6 Animal models of psychosis Most responsive drug classes

Test type

Description of test

Limitations

Dopaminergicinduced psychosis -Amphetamine -Apomorphine to be used with other tests

Systemic amphetamine or apomorphine causes behavioral effects in rats characteristic of schizophrenia

Neuroleptics

DA dependent; test not predictive for drugs affecting other receptors

Serotonergicinduced psychosis -LSD -PCP -DMT

Produces stereotyped behaviors, locomotor hyperactivity, and latent inhibition characteristic of schizophrenia

Atypical antipsychotic; Metabotropic glutamate drugs (mGlu2/3)

Construct validity difficult to establish; behavioral tolerance after repeat LSD

Glutamatergicinduced psychosis -PCP -NMDA -Ketamine

In rodents negative and positive effects (social behavior, immobility, hyperlocomotion)

Atypical antipsychotics; Sarcosine; mGlu2/3 agonist LY404039

Construct validity difficult to establish; based on single drug Rx

GABAergicinduced psychosis -Picrotoxine

Injection in prefrontal cortex of rats causes hyperlocomotion, catalepsy, or stereotyped behaviors

Haloperidol

Limited face and predictive validity with possibly more robust construct validity

Lesion-induced psychosis -Prefrontal -Hippocampal

Neurotoxic agents (kainic acid) produce schizophrenia-like symptoms in animals

Haloperidol

Limited construct validity; atypical antipsychotics can worsen the symptoms

Developmental injury models -Neonatal ventral hippocampal lesion (NVHL)

Damage to a rat’s hippocampus during the neonatal stage can produce an array of schizophrenia-like symptoms with a delayed onset

Typical and atypical antipsychotic drugs

Good face, predictive, and construct validity

Behavior models for psychosis (1) -Latent inhibition (LI)

Delayed conditioning results from previous exposure to the CS without its associated US. LI is the result of an automatic inhibitory process

Typical and atypical antipsychotic drugs

Good predictive validy; LI disruption not seen in all studies

(Continued)

c02

BLBK249-Cutler

34

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

Table 2.6 (Continued) Most responsive drug classes

Test type

Description of test

Limitations

Behavior models for psychosis (2) -Prepulse inhibition

Reduction in startle amplitude when the startle stimulus is preceded by a weak prepulse; model sensorimotor gating defects

Typical and atypical antipsychotic drugs

Good face, predictive, and construct validity; results drug class dependent

Behavior models for psychosis (3) -Conditioned avoidance response (CAR)

Very useful preclinical animal model; involves placing an animal (typically a rat) in a two-chamber box, and presenting a neutral CS; i.e., a light or tone, followed by an aversive US (foot shock) in the chamber the rat is in

Animals treated with low doses of antipsychotics fail to run during CS, yet respond normally to US–effect seen with all classes of antipsychotics, correlated with clinical potency

Good predictive validity; unknown construct validity but very useful screen for novel antipsychotics

still commonly used for the assessment of novel antipsychotic compounds [202, 203]. Apomorphine is a DA receptor agonist that also creates schizophrenia-like symptoms when administered to animals. This model shows good face validity because it impairs preattentional sensorimotor gating, which is a type of information processing that is also impaired in schizophrenia. Antipsychotics can restore preattentional sensorimotor gating in apomorphine-treated rats [204]. Interestingly, the administration of small doses of apomorphine to human volunteers with schizophrenia may enhance the overall antipsychotic experience due to stimulation of presynaptic auto receptors at dopaminergic neurons [205]. A major drawback of dopaminergic animal models is that their method of action is entirely dependent on the DA neurotransmitter system. Therefore, these models have limited predictive validity with antipsychotic compounds that also affect nondopaminergic receptors (such as atypical antipsychotics, which are also significant serotonergic receptor antagonists) [206]. These models also have limited construct validity, because research has implicated numerous other neurotransmitter systems in the pathogenesis of schizophrenia, including serotonin, gamma-aminobutyric acid (GABA), glutamate and the NMDA receptor, aspartate, and acetylcholine [207–209]. In fact, there is little direct evidence that DA plays a primary role in the development of schizophrenia, even though it may mediate some of the more clinically obvious signs and symptoms [210, 211], and its importance may vary by the disease state. For example, some patients with schizophrenia, particularly those

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

35

Table 2.7 Genetic models of psychosis

Candidate genes

Description

Most responsive drug classes

Limitations

Downregulated NMDA receptor genes in mice

Display of schizophrenia-like symptoms

Ameliorated with typical and atypical antipsychotic drugs

Good predictive but lack of construct validity

Reduced neuroregulin1 (NRG1) expression in mice

Behavioral phenotype similar to mouse models of schizophrenia

Behavior partially reversible with clozapine

Tremor in some models; heterozygous mutants–homozygous knockout is fatal; suspected compensatory activity of other genes

Disrupted in schizophrenia1 (DISC1) mouse model

DISC1 binds proteins known to be involved in neuronal migration, neurite outgrowth, cytoskeletal modulation, and signal transduction

Unknown

Model still being developed and tested

with predominantly negative symptoms, respond poorly or not at all to treatment with DA antagonists [212]. While dopaminergic animal models have proven useful as a screen for dopamine-dependent antipsychotics, scientists recommend antipsychotic screening in conjunction with other schizophrenia models that involve different neurotransmitter systems.

Serotonergic-induced psychosis The serotonergic (5-HT) system has been frequently implicated in the pathogenesis of schizophrenia. The indoleamine hallucinogen lysergic acid diethylamide (LSD) is able to mimic and antagonize serotonin in smooth muscle cells and the CNS [213, 214]. The phenethylamine hallucinogen mescaline shows similar clinical effects and cross-tolerance, and both indoleamines and phenethylamines share affinity for the 5-HT2A receptors [215]. Since hallucinations are one of the major positive symptoms associated with schizophrenia, the 5-HT system was implicated in this disorder. Atypical antipsychotics, such as clozapine, also show a relatively high affinity for the 5-HT2A receptor [216, 217]. LSD has been shown to disrupt startle habituation and preattential sensorimotor gating in humans and rats through direct stimulation of 5-HT2A receptors [218, 219]. It also produces stereotyped behaviors, locomotor hyperactivity, and latent inhibition characteristic of schizophrenia animal models [220]. The schizophrenia-mimicking drug phencyclidine (PCP) is another hallucinogen that has been proposed to trigger its effects through indirect activation of 5-HT2A receptors [221]. Finally, dimethyl tryptamine (DMT) is another serotonergic-acting hallucinogen that, unlike LSD and mescaline, does not

c02

BLBK249-Cutler

36

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

induce tolerance, and therefore may have higher construct validity than both of these drugs [222, 223]. Atypical antipsychotics, which tend to be antagonists of the serotonin receptor, can alleviate symptoms of LSD-induced psychosis [224]. Interestingly, a recent study found that LSD acts simultaneously on both serotonin and glutamate receptors, and a study has shown that novel compounds which target glutamate receptors (such as the metabotropic mGlu2/3 selective ligands LY341495 and LY379268 from Eli Lily) can ameliorate the positive symptoms of schizophrenia induced by LSD [225, 226]. While the serotonergic model of schizophrenia holds some face and predictive validity, its construct validity remains difficult to establish. There is little evidence of a primary dysfunction in the serotonergic systems in schizophrenia. In addition, repeat administration of LSD in humans or animals leads to behavioral tolerance, which represents very different phenomenology compared to schizophrenia [227]. Furthermore, the hallucinations induced by LSD are primarily of a visual nature, while the hallucinations associated with schizophrenia tend to be more auditory. As with the dopaminergic system, serotonin dysfunction is likely associated with schizophrenia, but it is probably not the only dysfunctional system.

Glutamatergic-induced psychosis Patients with schizophrenia often have a decreased release of glutamate in their frontal and temporal cortices [228]. The street drug phencyclidine (PCP), in addition to its effect on the 5-HT system, acts as an antagonist for a glutamatergic receptor. It was found to induce a set of disturbances in healthy subjects involving attention, perception, and symbolic thinking indistinguishable from those induced in schizophrenia [228]. These included a flat affect, auditory hallucinations, the inability to maintain a particular mental state, and peculiar verbalizations [229]. In addition, the NMDA glutamate receptor antagonist ketamine precipitates psychoses in patients with schizophrenia and induces psychotomimetic symptoms in healthy subjects [230–233]. Ketamine belongs to the same drug class as PCP but is considered less dangerous, and therefore is used in human clinical trials. Both PCP and ketamine act as noncompetitive antagonists on glutamatergic NMDA receptors [234]. This has led to the hypothesis that schizophrenia may involve hypofunction of NMDA receptors [235–237]. Rodents treated with PCP repeatedly exhibit negative symptoms associated with schizophrenia, including social behavioral deficits, enhanced immobility in a forced swimming test, sensorimotor gating deficits, and cognitive dysfunctions [238]. They exhibit positive symptoms, including hyperlocomotion, stereotyped movements, circling, and ataxia. They also disrupt startle habituation, prepulse inhibition, and latent inhibition [239–241]. Some of these behavioral changes endure after withdrawal from PCP treatment. Furthermore, repeated PCP treatment induces some neurochemical and neuroanatomical changes, including prefrontal cortex (PFC) dysfunction and reduced DA

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

37

utilization, similar to those seen in schizophrenia [242]. Many of the symptoms of PCP-induced psychosis can be ameliorated by atypical antipsychotics [243]. Glutamatergic models of schizophrenia may also affect the DA system, because antipsychotics that work through the dopamine receptor, such as clozapine and haloperidol, ameliorate symptoms induced by PCP [244, 245]. Novel antipsychotics that target the glutamate receptors (such as the amino acid sarcosine or the compound LY404039 [246] have been shown to ameliorate the positive symptoms of schizophrenia in both animal models and schizophrenia patients [236, 247, 248]. Like the previously mentioned pharmacologic models, PCP-induced schizophrenia has good face and predictive validity, yet limited construct validity. Many of the studies to date have involved only single injections of PCP, which is believed to contrast the persistent disruptions of the glutamatergic systems induced by schizophrenia. Indeed, long-term PCP administration has been reported to produce differential electrophysiological and neurochemical effects compared with single injections [249, 250]. Whether these differences are significant enough to affect the preclinical utility of this model remains to be seen.

GABAergic-induced psychosis Alteration in ␥ -aminobutyric acid (GABA) neurotransmission is also believed to play a role in schizophrenia for a number of reasons. Evidence for the role of GABA dysfunction in schizophrenia includes: reduced GABA uptake sites in the temporal lobe [251], abnormalities in the distribution of GABAergic neurons in the cortex [252], increased GABAA receptor binding in superficial layers of cingulate cortex [253], and decreased gene expression and activity of glutamic acid decarboxylase (GAD; a precursor of GABA) in the prefrontal cortex [254]. The GABAA receptor antagonist picrotoxin has been shown to inhibit preattentional sensorimotor gating in rats when injected into the prefrontal cortex. The dopamine antagonist haloperidol is able to reverse this inhibition [255]. Effects may also include hyperlocomotion, catalepsy, or stereotyped behaviors [256]. The face and predictive validity of this model is limited due to the relatively few behavioral symptoms induced by blocking the GABAA receptor, and its construct validity remains unclear. This model shows increased baseline hippocampal function, which has been observed in schizophrenia patients in some studies [257, 258], while other studies have reported the opposite effect [259–260]. Lesion-induced psychosis There is evidence that schizophrenia could be due to either developmental or degenerative disorders, producing brain abnormalities which could potentially be recreated using lesion models [261–263]. Lesion models typically use toxic agents to damage neuronal tissue by stimulating glutamate release or

c02

BLBK249-Cutler

38

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

by acting as glutamate receptor agonists, although electrolytic and aspirationbased lesioning techniques are also employed. Patients with schizophrenia typically exhibit dysfunction in their prefrontal cortex (PFC), which is the seat of many higher cognitive functions that are often impaired in schizophrenia, such as attention, working memory, emotional expression, and social interaction [264–266]. Prefrontal lesion models produce schizophrenia-like symptoms in animals, including reduced prepulse inhibition (PPI; a measure of sensorimotor gating), hyperresponsiveness to stress, transient increases in locomotor exploration, and stereotypy, among others [267–269]. The PFC also receives dense dopaminergic innervations, further implicating it in the pathogenesis of schizophrenia. Symptoms in this model can be reduced by the administration of the typical antipsychotic haloperidol [270]. This model has limited construct validity because peripheral L-dopa and low apomorphine administration also reduce symptoms, which is not the case in schizophrenia patients [271, 272]. The hippocampus has been implicated in the pathogenesis of schizophrenia because this structure modulates PFC activity and controls the mesolimbic dopaminergic system, which is believed to be dysfunctional in this disorder [211, 273]. There are numerous animal models that use hippocampal lesions to model schizophrenia. Aspiration of the entire hippocampus, the ventral hippocampus, or kainic acid administration to the dorsal hippocampus produce behavioral deficits similar to those seen in schizophrenia, including impairments in recognition learning and complex task acquisition, stereotyped behavior, polydipsia, and hyperlocomotion [256, 274, 275]. Many of these deficits can be reversed with the administration of the typical antipsychotic haloperidol [276, 277], although atypical antipsychotics can worsen symptoms [278].

Developmental injury models There is a large body of evidence that the brains of people with schizophrenia show abnormalities during development, despite the fact that schizophrenia is typically an adult onset disorder. This evidence includes ventricular enlargement, reduced cerebral volume, and cytoarchitectural abnormalities [279, 280], as well as subtle neurological, neuropsychological, and craniofacial abnormalities in children who later develop schizophrenia [281–283]. Numerous papers have suggested that the limbic abnormalities observed in schizophrenia may be the result of an early developmental injury which does not manifest until adulthood [284–286]. This “developmental injury” theory is currently the most promising model for reconciling the discrepancies observed between the numerous studies of the pathogenesis of this disorder. The neonatal ventral hippocampal lesion (NVHL) model is a developmental model of schizophrenia that looks at how damage to a rat’s hippocampus during the neonatal stage can produce an array of schizophrenia-like symptoms later in its life. The ventral hippocampus was targeted because it projects directly to the prefrontal cortex, plays a major role in regulating subcortical DA, and represents an area of the hippocampus that shows abnormal

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

39

function in schizophrenia patients [211, 287], Before postnatal day 56, rats with hippocampal lesions show little behavioral deficits. After postnatal day 56 (postpuberty), rats begin to display an array of symptoms characteristic of schizophrenia. This includes deficits in PPI and latent inhibition, impaired social behaviors, working memory problems, increased locomotion, and stereotypy [288–291]. These deficits can be ameliorated to some extent with administration of typical and atypical antipsychotics [270, 292, 293]. A major advantage of this model is that it demonstrates a delayed onset of symptoms, similar to the clinical presentation of schizophrenia in humans. The symptoms correspond to those observed in the disorder and are responsive to antipsychotics. In addition, primates with neonatal hippocampal lesions produce symptoms similar to those observed in rats, with a comparable delay in the onset of symptoms, showing the model has good construct validity [294]. A major disadvantage with all experimentally induced lesion models is that they typically reflect much greater damage than what is observed in the brains of schizophrenia patients. Despite this limitation, they still possess good face, predictive and construct validity, and remain a viable model for testing novel antipsychotics.

Behavioral models of psychosis Latent inhibition (LI) refers to the delayed conditioning that results from previous exposure to the conditioned stimulus without its associated unconditioned stimulus. LI is the result of an automatic inhibitory process that allows an individual to disregard or even stop the formation of new memories, by preventing associated learning with stimuli it considers unimportant. This attentional filtering mechanism has been observed in many species, including humans, and is believed to play an integral role in observation and learning [295]. In acute schizophrenia patients, pre-exposure to the conditioned stimulus does not inhibit later conditioning, and may in fact accelerate it. This is believed to reflect the inability of acute schizophrenia patients to ignore irrelevant stimuli [296]. In both rats and humans, amphetamine disrupts LI and typical and atypical antipsychotics restore it [297]. In addition, typical and atypical antipsychotics can improve latent inhibition in untreated rats. This makes it a unique animal model for testing antipsychotics which do not require any prior manipulations to the animal. Very few compounds without antipsychotic activity are active in this model, giving it good predictive validity [298]. A weakness of this model is that LI disruption is not observed in all studies; it is typically observed in acute schizophrenia patients and it is usually transient [299]. In addition, some studies have suggested that latent inhibition is dependent on the learning paradigm involved, and this paradigm differs between humans and rats, thereby weakening the model’s construct validity [300]. Finally, some studies have suggested that LI disruption is a consequence of schizophrenia treatment and not a result of the original disorder

c02

BLBK249-Cutler

40

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

[301]. Despite these criticisms, this model remains a useful screen for novel antipsychotics. Pre-pulse inhibition (PPI) refers to the reduction in startle amplitude when the startle stimulus is preceded by a weak prepulse. PPI is significantly reduced in schizophrenia patients [302]. PPI can be reduced in animal models by a large number of different mechanisms, including stimulation of D2 dopamine or serotonin receptors; amphetamine, apomorphine, or PCP administration; blockade of NMDA receptors; hippocampal lesions; or isolation rearing [303, 293]. PPI can be restored in these models with both typical and atypical antipsychotics [218, 304]. While induced PPI deficits in animals do not recreate the full symptomology of schizophrenia, they are believed to model the sensorimotor gating deficits observed in schizophrenic patients, and have good face, predictive, and construct validity [204, 305]. It should be noted that each method of inducing PPI in animals creates a unique model which results in a different pattern of responses to various antipsychotics. Therefore, it’s important to select the model of PPI that best responds to the class of antipsychotics one is evaluating. Conditioned avoidance response (CAR) is a very useful preclinical animal model in the study of antipsychotic drugs [306, 307]. It involves placing an animal (typically a rat) in a two-chamber box, and presenting a neutral, conditioned stimulus (CS; i.e., a light or tone), followed by an aversive, unconditioned stimulus (US; i.e., a foot shock) in the chamber the rat is in. After a few CS–US pairs, the animal learns to run to the other chamber during the CS, avoiding the US altogether. Animals treated with low doses of antipsychotic drugs fail to run during the CS, yet still respond normally to the US. This effect is seen with all classes of antipsychotics and is closely correlated with clinical potency [308]. In addition, the model shows good predictive validity because it is specific to antipsychotics; neither anxiolytics nor antidepressants show this effect [309]. The dopamine D2 receptor is strongly implicated in antipsychotic-induced disruption of avoidance, but the underlying behavioral and psychological processes remain intensely debated. Theories range from antipsychotic-induced deficits in “motor initiation” to decreased motivation [310, 311]. Despite unknown construct validity, this model has proven to be a very useful screen for novel antipsychotics.

Genetic models of psychosis While genetic animal models are not currently being used for the regular screening of novel antipsychotics, there are many genes implicated in the pathogenesis of schizophrenia, and numerous genetic models are being developed which successfully recreate aspects of the disorder. A few promising candidate genes are described below. Mice with a downregulated NMDA receptor gene display a host of schizophrenia-like symptoms, including hyperlocomotion, stereotypy, social behavior problems, and sensorimotor gating deficits [312, 313]. Many of these symptoms can be ameliorated with both typical and atypical antipsychotics

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

41

[314]. This model shows good predictive validity, yet its construct validity is lacking because there is no evidence for abnormalities in gene expression levels for subunits of the NMDA receptor in schizophrenia patients [315]. Neuregulin 1 (NRG1) has a role in activating neurotransmitter receptors, including glutamate receptors, and NRG1 mutants with reduced NRG1 expression show a behavioral phenotype similar to mouse models of schizophrenia, including reduced PPI, hyperlocomotion, cognitive deficits, and social behavioral deficits [316]. NRG1 mutants have fewer functional NMDA receptors than wild-type mice. In addition, their behavioral phenotype is partially reversible with clozapine [317]. Disrupted in schizophrenia 1 (DISC1) is a hub protein that has been causally linked to working memory, cognitive aging, gray matter volume in the prefrontal cortex, and abnormalities in hippocampal structure and function. It has been strongly implicated in schizophrenia as well as a host of mental disorders [318]. In addition, DISC1 binds a number of proteins known to be involved in neuronal migration, neurite outgrowth, cytoskeletal modulation, and signal transduction [319]. Several groups are currently generating DISC1 mice and characterizing them [320, 321] . As new genes associated with schizophrenia are identified, new animal models can be developed that better match the etiology of the disorder, offering improved construct validity and a new route for testing putative antipsychotics.

Example: the preclinical profile of the antipsychotic aripiprazole Aripiprazole (AbilifyTM ) is an atypical antipsychotic approved by the FDA in 2002, developed by Otsuka Japan and jointly marketed by Otsuka America and Bristol-Myers Squibb. It is a good case study for the utility of multiple animal models during preclinical trials. Like other atypical antipsychotics, it has affinity for both the dopamine and serotonin receptors, although it is a partial agonist at the D2 receptor while most atypical antipsychotics display antagonism at this receptor [322]. It is a partial agonist at the 5-HT1A receptor, and similar to other atypical antipsychotics, an antagonist at the 5-HT2A receptor. Aripiprazole has proven to be relatively successful in treating the positive, negative, and cognitive symptoms of schizophrenia. Its preclinical profile in animals is highly predictive of its efficacy in schizophrenia patients and mirrors that of other atypical antipsychotics [323]. Investigators used the performance of this compound on various behavioral animal models of schizophrenia to guide its further development and to predict its clinical utility. The following is a brief description of its preclinical results compared to other atypical antipsychotics. Aripiprazole displayed properties of both atypical and typical antipsychotics during preclinical evaluation. It was able to reverse apomorphineinduced PPI deficits in rats along with other atypical antipsychotics, which normally display 5-HT1A receptor antagonism [324]. Aripiprazole reduced apomorphine-induced stereotyped behaviors to the same extent as typical

c02

BLBK249-Cutler

42

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

and atypical antipsychotics [325]. It also reversed PPI deficits induced by the dopamine agonist amphetamine, which is characteristic of both typical and atypical antipsychotics [326]. It was, however, unable to reverse deficits produced by NMDA antagonists (PCP and MK-801) or a serotonin agonist (DOI), which is more characteristic of typical antipsychotics [327]. Along with another compound with dual 5-HT1A /D2 activity, aripiprazole significantly reversed PCP-induced social interaction deficits [328], and also blocked the PCP-induced disruption of PPI better than the atypical antipsychotics clozapine or olanzapine [329]. It reduced locomotor hyperactivity induced by various psychomimetic drugs in a dose-dependent manner, including dopamine and serotonin agonists and NMDA receptor antagonists [327]. Since typical antipsychotics normally do not reverse NMDA-induced hyperactivity, aripiprazole most closely matched atypical antipsychotics in this model [303]. Finally, aripiprazole had comparably low catalepsy and ptosis induction rates in rodents when compared to other atypical antipsychotics, which correlates with its relatively low incidence of extrapyramidal side effects reported in clinical trials [323, 330]. The typical and atypical properties of aripiprazole in preclinical trials mirrored its efficacy profile in humans, which was similar to both classes of antipsychotics in regard to treatment response, efficacy, and tolerability [331]. The conflicting results for this compound in preclinical studies underscore the need for a battery of tests when evaluating the efficacy of any novel compound. By combining the results from multiple animal models, investigators can get a better idea of the properties of a novel compound before it is tested in humans.

Animal models of Alzheimer’s disease Alzheimer’s disease (AD) is a widespread, devastating neurodegenerative disorder, with limited options for treatment and a prognosis that results in dementia. The risk of developing AD increases with age, doubling every 5 years after age 65, and reaching nearly 50% by age 85 [331]. Present therapies for AD do little to reduce the course of the disease and mainly treat symptoms. In fact, there is still much debate regarding the underlying etiology of the disorder, and animal models do not exist which recapitulate all of its behavioral and pathological characteristics. Transgenic models can recreate some of the neuropathological and behavioral aspects of the disorder, but it remains unclear how much validity they have. The defining characteristic of AD is the development of dementia in correlation with the accumulation of β-amyloid plaques and neurofibrillary tangles in regions of the cerebral cortex and subcortical nuclei [332]. Whether these plaques are the root cause of the disorder, or whether they are the result of it, remains to be determined. In addition, it is unclear whether Alzheimer’s disease is truly a disease—many scientists believe that it may be a normal part of aging, since even cognitively normal elderly people show lesions characteristic of AD [333]. Despite these uncertainties, many animal models

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

43

have been developed which recreate aspects of this disorder, and these models can provide a fertile testing ground for novel compounds that may halt or even reverse its course. The major animal and genetic models of AD are described in the following section, and summarized in Table 2.8.

Aged animal models of AD Several animal species naturally develop age-related deficits similar to those observed in AD, including rodents, rabbits, dogs, and monkeys. Rodents demonstrate deficiencies in learning and memory with increased age, and these deficiencies can be ameliorated with cholinergic agents [334]. Unfortunately, rodent models do not develop the characteristic plaques and neurofibrillay tangles associated with the pathogenesis of AD, thus limiting their construct validity. Primates and canines have been the most extensively studied animal models because they have human-like β-amyloid sequences; they naturally develop senile plaques, tau abnormalities, and cerebral amyloid angiopathy (CAA; the deposit of β-amyloid in the blood vessels of the brain). These species have large, complex brains that allow for a detailed analysis of neurodegeneration and in vivo imaging, and they show age-related deficits in behavior [335–338]. Like humans, individual animals show variations in the rate and amount of amyloid plaque accumulation. Drugs used to treat age-related cognitive decline in humans also work in these models [339–341]. In addition, diets rich in antioxidants and mitochondrial cofactors, as well as behavioral enrichment, can ameliorate age-related deficits in cognition [342– 344]. A drawback of these models is that fully formed neurofibrillary tangles have never been detected in animal models, yet they are prevalent in AD patients. In addition, due to the limited lifespan of animals in comparison to humans, the full-blown course of the disease is difficult to recreate. Finally, the cost and difficulty of running drug studies with large animal models limits their scope and utility. Pharmacological models of AD The scopolamine-induced behavioral model has been used for decades in both animals and humans to model the symptoms of AD and to test the efficacy of novel therapeutics. Scopolamine is a muscarinic antagonist which causes temporary cognitive dysfunction and memory impairments characteristic of AD. This model is based on the hypothesis that degeneration and subsequent hypofunction of the cholinergic basal forebrain produces the cognitive deficits observed in AD [345]. Scopolamine-induced deficits can be reversed by cholinesterase inhibitors [346–348]. The face and construct validity of this model is limited because it only represents deficits related to the cholinergic system, while AD has a much more complex etiology and more pronounced behavioral deficits [349]. In addition, while this model works well for evaluating cholinesterase inhibitors, it has more limited predictive validity for evaluating other classes of compounds [350, 351].

c02

BLBK249-Cutler

44

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

Table 2.8 Animals tests and genetic models for Alzheimer’s disease Responsive drug classes

Test type

Description of test

Aged animal models of AD

Primates and canines have human-like β-amyloid sequences; they naturally develop senile plaques, tau abnormalities, and cerebral amyloid angiopathy

Cholinesterase inhibitors, ampakine, adrafinil

No fully formed neurofibrillary tangles; limited lifespan of animals; large animal cost

Pharmacological models of AD -Scopolamine -Hemicholinium

Temporary behavioral and cognitive deficits

Cholinesterase inhibitors

Face and construct validity limited because only cholinergic system deficits represented; hemicholinium does not pass BBB

Lesion-based models of AD (1)

Lesions produced with electrolytic, excitotoxic, and immunotoxic techniques cause region-specific degeneration of cholinergic brain regions and behavioral changes

Depends on lesion type/location and animal strain; cholinesterase inhibitors have shown efficacy [392]

Good face validity, but limited construct validity because of wide, nonspecific toxic damage and methodological issues

Lesion-based models of AD (2)

Injection of amyloid peptides directly into the brain of rats leads to fibrils similar to AD patients

NMDA glutamate receptor blocker have shown efficacy [393]

Good face and construct validity; construct validity not as good as transgenic models

Transgenic models of AD (1) -Recreation of A-␤ plaques in mice

Models represent only components of AD; plaque production and behavioral deficits; App23 transgenic good model for CAA; PS/APP produces more severe and earlier disease phenotype

Immunization with A-β; specific APP antibodies; screen for plaque preventing compounds; PS/APP is a useful screen for compounds that prevent plaque accumulation

PDAPP mice have good predictive value, but method is company-owned; CRND8 mice show no neuronal cell death; Tg2576-Hsiao relies on inbred strain with retinal degeneration; PS/APP has limited construct validity because the sporadic form of AD does not involve PS mutations

• PDAPP transgenic • Tg2576-Hsiao transgenic • App23 transgenic • CRND8 transgenic • PS/APP double transgenic mice

Limitations

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

45

Table 2.8 (Continued) Responsive drug classes

Test type

Description of test

Limitations

Transgenic models of AD (2) -Recreation of neurofibrillary tangles • JNPL3 transgenic

Well suited for Model represent only testing new AD Rx components of AD; NFTs in neurons of brain and spinal cord; motor and behavioral disturbances

Transgenic models of AD (3) -Recreation of A- ␤ plaques and neurofibrillary tangles • Tau/APP double transgenic

Model represents only components of AD; best model with cellular and behavioral changes

Well suited for testing new AD Rx

Limited face and construct validity

Neurodegeneration models in invertebrates

Quick, cheap, large scale for recreating tau-related changes, e.g., Drosophila

Well suited for testing new AD Rx

Further removed species makes extrapolation of results to humans difficult

Limited construct validity since humans have 6 Tau isoforms and JNP3 has only 1

The hemicholinium-induced behavioral model also targets the cholinergic system, yet it offers higher construct validity than the scopolamine-induced model because it primarily affects presynaptic cholinergic function, which is believed to be specifically dysfunctional in AD [352, 353]. It produces behavioral deficits which can be partially ameliorated with the administration of cholinesterase inhibitors [354]. A drawback of this model is that hemicholinium is unable to pass through the blood–brain barrier, and therefore needs to be injected directly into the brain.

Lesion-based models of AD There are many AD models based on the formation of lesions in the cholinergic basal forebrain of animals, an area believed to be especially affected in AD. Electrolytic, excitotoxic, and immunotoxic techniques are commonly used to induce lesions. This leads to region-specific degeneration of presynaptic cholinergic projections and some behavioral characteristics associated with AD [355–357]. These models are believed to have good face validity, yet their construct validity is limited because many of the methods used to create lesions also induce widespread, nonspecific toxic damage [358]. Immunotoxin administration offers the most precise lesions, and also induces local upregulation of APP, adding to the face validity of this model [359, 360]. However,

c02

BLBK249-Cutler

46

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

a drawback with this technique is that it is difficult to implement localized immunotoxic damage due to methodological issues. Another AD lesion model is based on the injection of amyloid peptides directly into the brain. In rats, this leads to fibrils similar to those seen in AD, reduced cholinergic activity, brain atrophy, ventricular enlargement, neuronal damage, and behavioral deficits [361]. This model is believed to possess good face and construct validity. However, because it produces enhancements on certain behavioral tests, it does not have as good construct validity as many transgenic models of AD [362].

Genetic models of AD Transgenic models attempt to recreate the pathology of AD in animals, typically mice. These models tend to fall into one of three categories; (1) the recreation of amyloid-β peptide (A-β) plaques, (2) the recreation of neurofibrillary tangles, or (3) the recreation of both pathological lesions in the same animal, typically using a double transgenic model. The verdict is still unavailable regarding which of these lesions is more central to the pathogenesis of the disorder and which is a precursor of the other. A-β models tend to overexpress the human β-amyloid precursor protein (APP) by coupling it to a mouse promoter gene. The APP is believed to give rise to the A-β peptide, which accumulates in plaques and vessels in AD [363]. The formation of these plaques is usually associated with neurodegeneration and behavioral deficits, some of which can be ameliorated with amyloid antibodies [364]. Neurofibrillary tangle models attempt to recreate the neurofibrillary tangles of AD, which are composed of the Tau protein. The Tau protein is a protein that binds to and regulates the assembly and stability of neuronal microtubules and is found in an abnormal form as the major component of neurofibrillary tangles. Tau is primarily expressed in the brain and regulates the orientation and stability of microtubules. Hyperphosphorylation of Tau may cause it to dissociate from microtubules and aggregate into neurofibrillary tangles, which can cause neuronal death [365]. While no transgenic model is able to fully recapitulate all the characteristics of the disease, many are able to model components of the disorder and guide therapeutic development. This section lists the models that have been most frequently used for testing novel compounds in each of the categories. The PDAPP (platelet-derived growth factor promoter expressing amyloid precursor protein) transgenic mouse model produces overexpression of the human APP by coupling it to the platelet-derived growth factor-β (PDGF-β) promoter. This model is characterized by A-β plaques deposited in organs and blood vessels, as well as behavioral memory deficits associated with the pathology of AD [366, 367]. The model shows good predictive validity and was used by the biotechnology company Elan to demonstrate that immunization with the A-β peptide could clear amyloid deposits in the brain, a promising avenue for AD treatment [368]. A drawback with this model is that it is

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

47

not generally available to the research community, and therefore is not as well characterized as other models. The Tg2576-Hsiao transgenic mouse model produces overexpression of APP by coupling it to a mouse prion promoter. Like the PDAPP model, it also creates amyloid plaques and behavioral deficits characteristic of AD [369]. It is one of the best characterized mouse models of AD and is commercially available. A major drawback of this model is that it relies on an inbred strain of mouse (SJL) that carries a retinal degeneration mutation, which can influence performance on vision-dependent behavioral tasks [370]. The App23 transgenic mouse model expresses APP751 under a neuronspecific promoter. APP751 is an isoform of APP that is not normally expressed in neurons and leads to amyloid plaques and neurodegeneration [371]. These mice have a propensity to accumulate amyloid in the vascular wall, and are therefore a good model for studying cerebral amyloid angiopathy (CAA) [372]. The CRND8 transgenic mouse model massively overproduces β-APP under a prion promoter. The massive amount of amyloid is associated with ADassociated behavioral deficits [373]. These deficits can be corrected with the administration of specific APP antibodies [374]. A drawback with this model is that it lacks the neuronal cell death typically seen in advanced AD patients. PS/APP double transgenic mice models have been developed to hasten plaque formation and to produce a more severe disease phenotype. APP expressing animals were crossed with presenilin mutants (PS; a protein whose mutation is known to cause early onset AD), and the result is an animal that produces amyloid plaques at an earlier age and a faster rate than most other mouse models of AD [375, 376]. This model is a useful screen for compounds that prevent plaque accumulation, yet its construct validity is limited because the sporadic form of AD does not involve PS mutations. Additionally, whether PS mutations affect other cellular processes besides amyloid plaque formation remains unknown. Finally, no APP overexpression model (including this one) produces as much neurodegeneration as which occurs during the pathogenesis of AD. Overexpression of human Tau in mice leads to weakness and hyperphosphorylated insoluble tau immunoreactivity in the spinal cord, brainstem, and cortex, as well as degeneration of motor axons, yet does not readily produce neurofibrillary tangles (NFT) [377–379]. The JNPL3 transgenic mouse has been successful at recreating NFTs. This mouse overexpresses a mutant form of Tau known as P301L that occurs in another Tau associated disorder in man [380]. In JNPL3 mice, NFTs appear in neurons of the brain and spinal cord, and this expression is coupled with motor and behavioral disturbances [381, 382]. A limitation to the construct validity of this model is that humans have six Tau isoforms while the P301L transgenic mouse overexpresses only one Tau isoform. Currently, double transgenic mice are being developed which combine overexpression of both APP and Tau in order to better model the pathology of

c02

BLBK249-Cutler

48

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

AD. These Tau/APP double transgenic mice display an increase in the formation of NFTs in comparison to JNPL3 mice, suggesting a disease-enhancing effect when these two proteins are overexpressed together [383]. Tau/APP double transgenic mice represent the cutting edge of AD models, recreating both cellular and behavioral characteristics of the disorder, and are well suited for testing future therapeutics. Finally, it is worth mentioning that numerous invertebrate models exist for recreating tau-related neurodegeneration. These models allow for a wide variety of genetic manipulations and can be conducted quickly, cheaply, and on a large scale due to their rapid breeding times. In addition, invertebrate models can recreate characteristics of AD surprisingly well, despite their simple nervous systems. For example, Drosophila melanogaster overexpressing mutant Tau displays increased lethality, adult onset progressive neurodegeneration, and increased Tau accumulation in nerve cells [384, 385]. These invertebrate models have also been useful for testing the efficacy of novel AD therapeutics.

Animal model implications for early drug development Animal models which suggest efficacy often play an important step in the early drug development process. Following the identification of suitable molecular candidates, based on receptor binding, modeling, or changes to existing known active compounds, there will hopefully be a number of candidates to choose from. Receptor binding; toxicological; and absorption, distribution, metabolism, and excretion (ADME) studies in animals will narrow this field to fewer suitable candidates. At this point, such compounds can be further screened in some of the animal models as described in this section. While activity in these models does not guarantee that a compound will be active in humans, it nonetheless provides confirmation at this early stage of development. For many smaller companies, this may provide the impetus for additional funding. Generally, one likes to identify a candidate that appears to have activity in more than one model for the disease or condition, and such activity may be a crucial decision point for going forward with a specific candidate. Thus, while these models often play an important role in the early development process, once a candidate has been selected, the focus shifts to other important goals such as safety/tolerance in man, and the animal models rarely influence the subsequent development process. An exception to this is that sometimes the dose identified as active in an animal model is carried forward with the expectation that it is also the active dose in humans. Many of the dose–response relationships that emerge from these animal models suggest an inverted U response, and often suggest that small doses will be effective in humans. However, in our experience, the vast majority of CNS compounds do not display such a clinical pattern, and the response is usually linear with dose until adverse events intervene. We will address this issue more fully in Chapter 6. For further studies in man, one should look for important signals

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

49

during the entire preclinical process, not just at one component. If the selected compound is an entirely new strategy for a disease/illness, then animal models may or may not be helpful. In such cases, more emphasis is placed on identifying efficacy signals in early human development, using biomarkers and/or other human models of the disease. Occasionally, efficacy signals in animals are predictive of those in humans. For example, the SSRI sertraline produced rapid increases in cerebrospinal fluid (CSF) 5-HT levels in primates [386, 387], and the monoamine oxidase inhibitors cimoxatone and clorgyline reduced deaminated metabolites of norepinephrine, dopamine, and serotonin in the CSF of primates [388, 389]. Both of these changes were consistent with biomarker changes in humans that correlated with drug efficacy [390, 391]. This is why compiling a robust preclinical profile of a compound, including biomarkers, can be so advantageous—having some predictive data about the safety and efficacy of a novel compound in preclinical studies can help shape clinical trials, reducing time and costs, and suggesting the most viable candidates for subsequent development. Unfortunately, a compound’s preclinical profile is not always predictive of its clinical profile. This is where human biomarkers are needed to determine a compound’s safety and efficacy in early clinical studies. These biomarkers will be discussed at length in Chapter 3.

References 1. Matthysse S. (1986) Animal models in psychiatric research. Prog Brain Res. 65:259– 70. 2. Segal DS, Geyer MA. (1985) Animal models of psychopathology. In: Judd Ll and Groves PM, editors. Psychobiological Foundations of Clinical Psychiatry. JB Lippincott, Philadelphia. 3. Willner P. (1984) The validity of animal models of depression. Psychopharmacology. 83:1–16. 4. Ahlenius S. (1991) The Clever Hans phenomenon in animal models of schizophrenia, or homology as an important factor in comparing behavioral functions across species. Perspect Biol Med. 34:219–25. 5. Hinde RA. (1976) The use of differences and similarities in comparative psychopathology. In: Serban G and Kling A, editors. Animal Models in Psychobiology. Plenum Press, New York, pp 187–202. 6. Cronbach LJ, Meehl PE. (1955) Construct validity in psychological tests. Psychol Bull. 52:281–302. 7. Campbell DT, Fiske DW. (1959) Convergent and discriminant validation by the multitrait-multimethod matrix. Psychol Bull. 56:81–105. 8. Markou A, Weiss F, Gold LH, et al. (1993) Animal models of drug craving. Psychopharmacology. 112:163–82. 9. Kulkarni SK, Reddy DS. (1996) Animal behavioral models for testing antianxiety agents. Meth Find Exp Clin Pharmacol. 18:219–30. 10. Valle FP. (1970) Effects of strain, sex, and illumination on open-field behavior of rats. Am J Psychol. 83(1):103–11. 11. Tarantino LM, Bucan M. (2000) Dissection of behavior and psychiatric disorders using the mouse as a model [review]. Human Mol Genet. 9(6):953–65.

c02

BLBK249-Cutler

50

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

12. Stefanski R, Palejko W, Kostowski W, et al. (1992) The comparison of benzodiazepine derivatives and serotonergic agonists and antagonists in two animal models of anxiety. Neuropharmacology. (12):1251–8. 13. Prut L, Belzung C. (2003) The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: a review. Eur J Pharmacol. 463(1–3):3–33. 14. Montgomery KC. (1955) The relation between fear induced by novel stimulation and exploratory behavior. J Comp Physiol Psychol. 48(4):254–60. 15. Pellow P, File SE. (1986) Anxiolytic and anxiogenic drug effects on exploratory activity in an elevated plus-maze: a novel test of anxiety in the rat. Pharmacol Biochem Behav. 24(3):525–9. 16. Pellow S, Chopin P, File SE, et al. (1985) Validation of open:closed arm entries in an elevated plus-maze as a measure of anxiety in the rat. J Neurosci Methods. 14(3):149–67. 17. Hogg S. (1996) A review of the validity and variability of the elevated plus-maze as an animal model of anxiety. Pharmacol Biochem Behav. 54(1):21–30. 18. File S, Zangrossi JH. (1993) “One-trial tolerance” to the anxiolytic actions of benzodiazepines in the elevated plus-maze, or the development of a phobic state? Psychopharmacology. 110(1–2):240–4. 19. Jardim MC, Nogueira RL, Graeff FG, et al. (1999) Evaluation of the elevated T-maze as an animal model of anxiety in the mouse. Brain Res Bull. 48(4):407–11. 20. Sanson LT, Carobrez AP. (1999) Long-lasting inhibitory avoidance acquisition in rats submitted to the elevated T-maze model of anxiety. Behav Brain Res. 101(1):59–64. 21. Graeff FG, Netto CF, Zangrossi H, Jr. (1998) The elevated T-maze as an experimental model of anxiety. Neurosci Biobehav Rev. 23(2):237–46. 22. Pinheiro SH, Zangrossi H, Jr, Del-Ben CM, et al. (2007) Elevated mazes as animal models of anxiety: effects of serotonergic agents. An Acad Bras Cienc. 79(1):71–85. 23. Carvalho-Netto EF, Nunes-de-Souza RL. (2004) Use of the elevated T-maze to study anxiety in mice. Behav Brain Res. 148(1–2):119–32. 24. Conde C, Costa V, Tomaz C. (2000) Effects of emotional reactivity on inhibitory avoidance in the elevated T-maze. Braz J Med Biol Res. 33:233–6. 25. Kshama D, Hrishikeshavan HJ, Shanbhogue R, et al. (1990) Modulation of baseline behavior in rats by putative serotonergic agents in three ethoexperimental paradigms. Behav Neural Biol. 54:234–53. 26. Adamec R, Bartoszyk GD, Burton P. (2004) Effects of systemic injections of vilazodone, a selective serotonin reuptake inhibitor and serotonin 1A receptor agonist, on anxiety induced by predator stress in rats. Eur J Pharmacol. 504(1–2):65–77. 27. Griebel G, Blanchard CD, Agnes RSBRJ. (1995) Differential modulation of antipredator defensive behavior in Swiss-Webster mice following acute or chronic administration of imipramine and fluoxetine. Psychopharmacology. 120:57–66. 28. Shepherd JK, Flores T, Rodgers RJ, et al. (1992) The anxiety/defense test battery: influence of gender and ritanserin treatment on antipredator defensive behavior. Physiol Behav. 51(2):277–85. 29. Blanchard RJ, Griebel G, Henrie JA, et al. (1997) Differentiation of anxiolytic and panicolytic drugs by effects on rat and mouse defense test batteries. Neurosci Biobehav Rev. 21(6):783–9. 30. Landis C, Hunt WA. (1939) The startle pattern. Int J Psycho-Anal. 21:370. 31. Seaman RL, Chen J. (1996) Sensing platform for acoustic startle responses from rat forelimbs and hindlimbs. IEEE Trans Biomed Eng. 43(2):221–5. 32. Davis M. (1979) Diazepam and flurazepam: effects on conditioned fear as measured with the potentiated startle paradigm. Psychopharmacology. 62:1–7.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

51

33. Mansbach RS, Geyer MA. (1988) Blockade of potentiated startle in rats by 5-hydroxytryptamine receptor ligands. Eur J Pharmacol. 156:375–83. 34. Santos JM, Martinez RC, Brand˜ao ML. (2006) Effects of acute and subchronic treatments with fluoxetine and desipramine on the memory of fear in moderate and high-intensity contextual conditioning. Eur J Pharmacol. 542(1–3):121–8. 35. Bell J and Deveaugh-Geiss J. Multicenter trial of a 5-HT3+ antagonist, ondansetron in social phobia. 33rd Annual Meeting of the American College of Neuropsychopharmacology; December 12–16, 1994; San Juan, Puerto Rico. 36. Davis M. (1979) Morphine and naloxone: effects on conditioned fear as measured with the potentiated startle paradigm. Eur J Pharmacol. 54(4):341–7. 37. Hijzen TH, Houtzager SWJ, Joordens RJE, et al. (1995) Predictive validity of the potentiated startle response as a behavioral model for anxiolytic drugs. Psychopharmacology. 118(2):150–4. ¨ 38. Baas JM, Grillon C, Bocker KB, et al. (2002) Benzodiazepines have no effect on fearpotentiated startle in humans. Psychopharmacology (Berl). 161(3):233–47. 39. Costall B, Naylor RJ. (1991) Pharmacological properties and functions of central 5-HT3 receptors. Therapie. 46(6):437–44. 40. Vanover KE, Robledo S, Huber M, et al. (1999) Pharmacological evaluation of a modified conflict procedure: punished drinking in non-water-deprived rats. Psychopharmacology (Berl). 145(3):333–41. 41. Griebel G, Rodgers RJ, Perrault G, et al. (2000) The effects of compounds varying in selectivity as 5-HT(1A) receptor antagonists in three rat models of anxiety. Neuropharmacology. 39(10):1848–57. 42. Kennett GA, Trail B, Bright F. (1998) Anxiolytic-like actions of BW 723C86 in the rat Vogel conflict test are 5-HT2B receptor mediated. Neuropharmacology. 37(12):1603– 10. 43. Handley SL, McBlane JW. (1992) Opposite effects of fluoxetine in two animal models of anxiety. Brit J Pharmacol. 107:446P. 44. Geller I, Seifter J. (1960) The effects of meprobamate, d-amphetamine and promazine on experimentally induced conflict in the rat. Psychopharmacologia. 1:482–92. 45. Gleeson S, Barrett JE. (1990) 5-HT1A agonist effects on punished responding of squirrel monkeys. Pharmacol Biochem Behav. 37(2):335–7. 46. Sanger DJ, Joly D. (1990) Performance of a passive avoidance response is disrupted by compounds acting at the 5-HT1A receptors. Behav Pharmacol. 1:235–40. 47. Sanger DJ, Joly D. (1985) Anxyiolytic drugs and the acquisition of conditioned fear in mice. Psychopharmacology. 85:284–8. 48. Patel JB, Ciofalo VB, Iorio LC. (1979) Benzodiazepine blockade of passive-avoidance task in mice: a state-dependent phenomenon. Psychopharmacology (Berl). 61(1):25–8. 49. Genkova-Papazova M, Lazarova-Bakarova M, Petkov VD. (1994) The 5-HT2 receptor antagonist ketanserine prevents electroconvulsive shock- and clonidine-induced amnesia. Pharmacol Biochem Behav. 49(4):849–52. ¨ ¸ mez S, Erden F, et al. (2006) Chronic administration of fluoxetine or ven50. Ulak G, Goc lafaxine induces memory deterioration in an inhibitory avoidance task in rats. Drug Dev Res. 67(5):456–61. 51. Doyle E, O’Boyle KM, Shiotani T, et al. (1996) Nefiracetam (DM-9384) reverses apomorphine-induced amnesia of a passive avoidance response: delayed emergence of the memory retention effects. Neurochem Res. 21(6):649–52. 52. Wilson LM, Riccio DC. (1976) Scopolamine’s effect on passive avoidance behavior in immature rats. Dev Psychobiol. 9(3):245–54.

c02

BLBK249-Cutler

52

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

53. File SE, Hyde JRG. (1979) A test of anxiety that distinguishes between the actions of benzodiazepines and those of other minor tranquilizers and of stimulants. Pharmacol Biochem Behav. 11:65–8. 54. de L Angelis, File SE. (1979) Acute and chronic effects of three benzodiazepines in the social interaction anxiety test in mice. Psychopharmacology (Berl). 64(2):127–9. 55. Dekeyne A, Denorme B, Monneyron S, et al. (2000) Citalopram reduces social interaction in rats by activation of serotonin (5-HT)(2C) receptors. Neuropharmacology. 39(6):1114–7. 56. Homanics GE, Quinlan JJ, Firestone LL. (1999) Pharmacologic and behavioral responses of inbred C57BL/6J and strain 129/SvJ mouse lines. Pharmacol Biochem Behav. 63:21–6. 57. Cook MN, Williams RW, Flaherty L. (2001) Anxiety-related behaviors in the elevated zero-maze are affected by genetic factors and retinal degeneration. Behav Neurosci. 115:468–76. 58. Voikar V, Koks S, Vasar E, et al. (2001) Strain and gender differences in the behavior of mouse lines commonly used in transgenic studies. Physiol Behav. 72:271–81. 59. Ohl F, Roedel A, Binder E, et al. (2003) Impact of high and low anxiety on cognitive performance in a modified hole board test in C57BL/6 and DBA/2 mice. Eur J Neurosci. 17:128–36. 60. Thompson WR. (1953) The inheritance of behavior: behavioral differences in fifteen mouse strains. Can J Psychol. 7:145–55. 61. Kulikov A, Aguerre S, Berton O, et al. (1997) Central serotonergic systems in the spontaneously hypertensive and Lewis rat strains that differ in the elevated plus-maze test of anxiety. J Pharmacol Exp Ther. 281:775–84. 62. Liebsch G, Montkowski A, Holsboer F, et al. (1998) Behavioral profiles of two Wistar rat lines selectively bred for high or low anxiety-related behaviour. Behav Brain Res. 94:301–10. 63. Hine B. (1995) Differential open-field reactivity in HAS and LAS rats. Physiol Behav. 57:301–6. 64. Mohapel P, McIntyre DC. (1998) Amygdala kindling-resistant (SLOW) or prone (FAST) rat strains show differential fear responses. Behav Neurosci. 112:1402–13. 65. Belzung C, Griebel G. (2001) Measuring normal and pathological anxiety-like behavior in mice: a review. Behav Brain Res. 125:141–9. 66. Poltyrev T, Keshet GI, Kay G, et al. (1996) Role of experimental conditions in determining differences in exploratory behavior of prenatally stressed rats. Dev Psychobiol. 29:453–62. 67. Chapillon P, Manneche C, Belzung C, et al. (1999) Rearing environmental enrichment in two inbred strains of mice: Effects on emotional reactivity. Behav Genet. 29:41–6. 68. Majercsik E, Haller J, Leveleki C, et al. (2003) The effect of social factors on the anxiolytic efficacy of buspirone in male rats, male mice, and men. Prog Neuropsychopharmacol Biol Psychiatry. 27(8):1187–99. 69. Bourin M, Hasco¨et M. (2003) The mouse light/dark box test. Eur J Pharmacol. 463(1–3):55–65. 70. Cl´ement Y, Calatayud F, Belzung C. (2002) Genetic basis of anxiety-like behavior: a critical review. Brain Res Bull. 57(1):57–71. 71. Wood SJ, Toth M. (2001) Molecular pathways of anxiety revealed by knockout mice. Mol Neurobiol. 23(2–3):101–19. 72. Flint J, Corley R. (1996) Do animal models have a place in the genetic analysis of quantitative human behavioural traits? J Mol Med. 74:515–21.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

53

73. International HGSC. (2001) Initial sequencing and analysis of the human genome. Nature. 409:860–921. 74. Holsboer F, Ising M. (2008) Central CRH system in depression and anxiety–evidence from clinical studies with CRH1 receptor antagonists. Eur J Pharmacol. 583(2–3):350– 357. 75. Koob GF, Heinrichs SC, Merlo PE, et al. (1993) The role of corticotropin-releasing factor in behavioural responses to stress. In: Chadwick DJ, Marsh J, Ackrill K, editors. Ciba Foundation Symposium—Corticotropin-Releasing Factor. Vol 172. Wiley, Chichester, UK, pp 277–95. 76. Stenzel-Poore MP, Cameron VA, Vaughan J, et al. (1992) Development of Cushing’s syndrome in corticotropin-releasing factor transgenic mice. Endocrinology. 130:3378–86. 77. Dirks A, Groenink L, Bouwknecht JA, et al. (2002) Overexpression of corticotropinreleasing hormone in transgenic mice and chronic stress-like autonomic and physiological alterations. Eur J Neurosci. 16(9):1751–60. 78. van GMM, Stenzel-Poore MP, Holsboer F, et al. (2002) Effects of transgenic overproduction of CRH on anxiety-like behaviour. Eur J Neurosci. 15:2007–15. 79. Weninger SC, Dunn AJ, Muglia LJ, et al. (1999) Stress-induced behaviors require the corticotropin-releasing hormone (CRH) receptor, but not CRH. Proc Natl Acad Sci. 96:8283–8. 80. Holsboer F. (1999) The rationale for corticotropin-releasing hormone receptor (CRH-R) antagonists to treat depression and anxiety. J Psychiatr Res. 33(3):181–214. 81. Olivier B, Soudijn W, Van WI. (1999) The 5-HT1A receptor and its ligands: structure and function. Prog Drug Res. 52:104–65. 82. Lesch KP, Wiesmann M, Hoh A, et al. (1992) 5-HT1A receptor-effector system responsivity in panic disorder. Psychopharmacology (Berl). 106(1):111–7. 83. Heisler LK, Chu HM, Brennan TJ, et al. (1998) Elevated anxiety and antidepressant-like responses in serotonin 5-HT1A receptor mutant mice. Proc Natl Acad Sci. 95:15049–54. 84. Parks CL, Robinson PS, Sibille E, et al. (1998) Increased anxiety of mice lacking the serotonin1A receptor. Proc Natl Acad Sci. 95:10734–9. 85. Ramboz S, Oosting R, Amara DA, et al. (1998) Serotonin receptor 1A knockout: an animal model of anxiety-related disorder. Proc Natl Acad Sci. 95:14476–81. 86. Sibille E, Pavlides C, Benke D, et al. (2000) Genetic inactivation of the Serotonin(1A) receptor in mice results in downregulation of major GABA(A) receptor alpha subunits, reduction of GABA(A) receptor binding, and benzodiazepine-resistant anxiety. J Neurosci. 20:2758–65. 87. Toth M. (2003) 5-HT1A receptor knockout mouse as a genetic model of anxiety. Eur J Pharmacol. 463(1-3):177–84. 88. Olivier B, Pattij T, Wood SJ, et al. (2001) The 5-HT(1A) receptor knockout mouse and anxiety. Behav Pharmacol. 12:439–50. 89. Groenink L, Pattij T, De JR, et al. (2003) 5-HT1A receptor knockout mice and mice overexpressing corticotropin-releasing hormone in models of anxiety. Eur J Pharmacol. 463(1-3):185-97. 90. Pattij T, Hijzen TH, Groenink L, et al. (2001) Stress-induced hyperthermia in the 5-HT(1A) receptor knockout mouse is normal. Biol Psychiatry. 49:569–74. 91. Siderovski DP, Heximer SP, Forsdyke DR. (1994) A human gene encoding a putative basic helix–loop–helix phosphoprotein whose mRNA increases rapidly in cycloheximide-treated blood mononuclear cells. DNA Cell Biol. 13(2):125–47. 92. Oliveira-Dos-Santos AJ, Matsumoto G, Snow BE, et al. (2000) Regulation of T cell activation, anxiety, and male aggression by RGS2. Proc Natl Acad Sci USA. 97(22):12272–7.

c02

BLBK249-Cutler

54

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

93. Yalcin B, Willis-Owen SA, Fullerton J, et al. (2004) Genetic dissection of a behavioral quantitative trait locus shows that Rgs2 modulates anxiety in mice. Nat Genet. 36(11):1197–202. 94. Smoller JW, Paulus MP, Fagerness JA, et al. (2008) Influence of RGS2 on anxiety-related temperament, personality, and brain function. Arch Gen Psychiatry. 65(3):298–308. 95. Mannisto PT, Lang A, Harro J, et al. (1994) Opposite effects mediated by CCKA and CCKB receptors in behavioural and hormonal studies in rats. Naunyn Schmiedebergs Arch Pharmacol. 349:478–84. 96. Singh L, Lewis AS, Field MJ, et al. (1991) Evidence for an involvement of the brain cholecystokinin B receptor in anxiety. Proc Natl Acad Sci USA. 88:1130–3. 97. Blommaert AG, Weng JH, Dorville A, et al. (1993) Cholecystokinin peptidomimetics as selective CCK-B antagonists: design, synthesis, and in vitro and in vivo biochemical properties. J Med Chem. 36:2868–77. 98. Harro J, Vasar E, Bradwejn J. (1993) CCK in animal and human research on anxiety. Trends Pharmacol Sci. 14:244–9. 99. Costall B, Domeney AM, Hughes J, et al. (1991) Anxiolytic effects of CCK-B antagonists. Neuropeptides. 19(Suppl):65–73. 100. Bradwejn J, Koszycki D, Couetoux dTA, et al. (1994) The panicogenic effects of cholecystokinin-tetrapeptide are antagonized by L-365, 260, a central cholecystokinin receptor antagonist, in patients with panic disorder. Arch Gen Psychiatry. 51:486–93. 101. Noble F, Roques BP. (1999) CCK-B receptor: chemistry, molecular biology, biochemistry and pharmacology. Prog Neurobiol. 58(4):349–79. 102. Kalindjian SB, McDonald IM. (2007) Strategies for the design of non-peptide CCK2 receptor agonist and antagonist ligand. Curr Top Med Chem. 7(12):1195–204. 103. Raud S, Runkorg K, Veraksits A, et al. (2003) Targeted mutation of CCK2 receptor gene modifies the behavioural effects of diazepam in female mice. Psychopharmacology. 168:417–25. 104. Horinouchi Y, Akiyoshi J, Nagata A, et al. (2004) Reduced anxious behavior in mice lacking the CCK2 receptor gene. Eur Neuropsychopharmacol. 14:157–61. 105. Harro J. (2006) CCK and NPY as anti-anxiety treatment targets: promises, pitfalls, and strategies. Amino Acids. 31(3):215–30. 106. Ebner K, Singewald N. (2006) The role of substance P in stress and anxiety responses. Amino Acids. 31(3):251–72. 107. Teixeira RM, Santos AR, Ribeiro SJ, et al. (1996) Effects of central administration of tachykinin receptor agonists and antagonists on plus-maze behavior in mice. Eur J Pharmacol. 311(1):7–14. 108. Santarelli L, Gobbi G, Debs PC, et al. (2001) Genetic and pharmacological disruption of neurokinin 1 receptor function decreases anxiety-related behaviors and increases serotonergic function. Proc Natl Acad Sci USA. 98(4):1912–7. 109. Bilkei-Gorzo A, Zimmer A. (2005) Mutagenesis and knockout models: NK1 and substance P. Handb Exp Pharmacol. (169):143–62. 110. Herpfer I, Lieb K. (2005) Substance P receptor antagonists in psychiatry: rationale for development and therapeutic potential. CNS Drugs. 19(4):275–93. 111. Kalia M. (2005) Neurobiological basis of depression: an update. Metabolism. 54(5 Suppl 1):24–7. 112. Nemeroff CB. (2002) Recent advances in the neurobiology of depression. Psychopharmacol Bull. 36(Suppl 2):6–23. 113. Richardson JS. (1991) The olfactory bulbectomized rat as a model of major depressive disorder. In: Boulton A, Baker G, Martin-Iverson M, editors. Neuromethods: Animal Models in Psychiatry. Vol 19. Humana Press, Clifton, pp 61–79.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

55

114. Porsolt RD, Bertin A, Jalfre M. (1977) Behavioral despair in mice: a primary screening test for antidepressants. Arch Int Pharmacodyn Ther. 229(2):327–36. 115. Tarazi FI, Schetz JA. (2005) Neurological and Psychiatric Disorders: From Bench to Bedside. Humana Press. 116. Boulton AA, Baker GB, Bateson AN. (1998) In Vivo Neuromethods. Humana Press, Totowa, New Jersey, pp 114–15. 117. Porsolt RD, Brossard G, Hautbois C, et al. (2003) UNIT 8 10A: Rodent Models of Depression: Forced Swimming and Tail Suspension Behavioral Despair Tests in Rats and Mice. In: Taylor G, editor. Current Protocols in Neuroscience. Wiley. 118. West AP. (1990) Neurobehavioral studies of forced swimming: the role of learning and memory in the forced swim test. Prog Neuropsychopharmacol Biol Psychiatry. 14(6):863–77. ´ S, et al. (1999) Learned immobility is also involved 119. Parra A, Vinader-Caerols C, Monleon in the forced swimming test in mice. Psicothema. 11(2):239–46. 120. D’Haenen HAH, den Boer JA, Willner, P. (2002) Biological Psychiatry. Wiley-Blackwell, West Sussex, England, pp 708–709. 121. Borsini F. (1995) Role of the serotonergic system in the forced swimming test. Neurosci Biobehav Rev. 19(3):377–95. 122. Lahmame A, Armario A. (1996) Differential responsiveness of inbred strains of rats to antidepressants in the forced swimming test: are Wistar Kyoto rats an animal model of subsensitivity to antidepressants? Psychopharmacology (Berl). 123(2):191– 8. 123. Petit-Demouliere B, Chenu F, Bourin M. (2005) Forced swimming test in mice: a review of antidepressant activity. Psychopharmacology (Berl). 177(3):245–55. 124. David DJ, Renard CE, Jolliet P, et al. (2003) Antidepressant-like effects in various mice strains in the forced swimming test. Psychopharmacology (Berl). 166(4):373–82. 125. Borsini F, Meli A. (1988) Is the forced swimming test a suitable model for revealing antidepressant activity? Psychopharmacology (Berl). 94(2):147–60. 126. Cryan JF, Valentino RJ, Lucki I. (2005) Assessing substrates underlying the behavioral effects of antidepressants using the modified rat forced swimming test. Neurosci Biobehav Rev. 29(4–5):547–69. 127. Steru L, Chermat R, Thierry B, et al. (1985) The tail suspension test: a new method for screening antidepressants in mice. Psychopharmacology (Berl). 85(3):367–70. 128. Chermat R, Thierry B, Mico JA, et al. (1986) Adaptation of the tail suspension test to the rat. J Pharmacol. 17(3):348–50. 129. Seligman ME, Rosellini RA, Kozak MJ. (1975) Learned helplessness in the rat: time course, immunization, and reversibility. J Comp Physiol Psychol. 88(2):542–7. 130. Weiss JM, Glazer HI. (1975) Effects of acute exposure to stressors on subsequent avoidance-escape behavior. Psychosom Med. 37(6):499–521. 131. Kazuaki T, Shigeru Y, Shigeru O. (2001) Availability of learned helplessness test as a model of depression compared to a forced swimming test in rats. Pharmacology. 63:147–53. 132. Martin P, Soubrie P, Simon P. (1987) The effect of monoamine oxidase inhibitors compared with classical tricyclic antidepressants on learned helplessness paradigm. Prog Neuropsychopharmacol Biol Psychiatry. 11(1):1–7. 133. Arturo Z, In´es A, Luis L, et al. (2007) Reversal of learned helplessness by selective serotonin reuptake inhibitors in rats is not dependent on 5-HT availability. Neuropharmacology. 52(3):975–84. 134. Sherman AD, Sacquitne JL, Petty F. (1982) Specificity of the learned helplessness model of depression. Pharmacol Biochem Behav. 16(3):449–54.

c02

BLBK249-Cutler

56

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

135. Willner P, Muscat R, Papp M. (1992) Chronic mild stress-induced anhedonia: a realistic animal model of depression. Neurosci Biobehav Rev. 16(4):525–34. 136. Willner P, Towell A, Sampson D, et al. (1987) Reduction of sucrose preference by chronic unpredictable mild stress, and its restoration by a tricyclic antidepressant. Psychopharmacology (Berl). 93(3):358–64. 137. Papp M, Moryl E, Willner P. (1996) Pharmacological validation of the chronic mild stress model of depression. European Journal of Pharmacology. 296(2):129–36. 138. Briley M, Montgomery SA. (1999) Antidepressant Therapy at the Dawn of the Third Millennium. Informa Healthcare, London England, p 95. 139. Moreau JL, Bos M, Kenck F, et al. (1996) 5HT2c receptor agonist exhibit antidepressantlike properties in the anhedonia model of depression in rats. Eur Neuropsychopharmacol. 6:169–75. 140. Willner P. (1997) Validity, reliability and utility of the chronic mild stress model of depression: a 10-year review and evaluation. Psychopharmacology (Berl). 134(4):319–29. 141. Willner P. (2005) Chronic mild stress (CMS) revisited: consistency and behaviouralneurobiological concordance in the effects of CMS. Neuropsychobiology. 52(2):90– 110. 142. Harris RB, Zhou J, Youngblood BD, et al. (1997) Failure to change exploration or saccharin preference in rats exposed to chronic mild stress. Psychopharmacology (Berl). 134(4):319–29. 143. Skolnick P. (1997) Antidepressants: New Pharmacological Strategies. Humana Press, Totowa, New Jersey, p 226. 144. Bremner JD, Narayan M, Anderson ER, et al. (2000) Hippocampal volume reduction in major depression. Am J Psychiatry. 157:115–8. 145. Sheline YI, Gado MH, Price JL. (1998) Amygdala core nuclei volumes are decreased in recurrent major depression. Neuro Report. 9:2023–8. 146. Bremner JD, Vythilingam M, Vermetten E, et al. (2002) Reduced volume of orbitofrontal cortex in major depression. Biol Psychiatry. 51:273–9. 147. Song C, Leonard BE. (2005) The olfactory bulbectomized rat as a model of depression. Neurosci Biobehav Rev. 29(4–5):627–47. 148. Harkin A, Kelly JP, Leonard BE. (2003) A review of the relevance and validity of olfactory bulbectomy as a model of depression. Clin Neurosci Res. 3(4–5):253–62. 149. Lucas G, Rymar VV, Du J, et al. (2007) Serotonin(4) (5-HT(4)) receptor agonists are putative antidepressants with a rapid onset of action. Neuron. 55(5):712–25. 150. Song C, Leonard BE. (2000) Fundamentals of Psychoneuroimmunology. Wiley, West Sussex, England, p 87. 151. Henry JP. (1982) The relation of social to biological processes in disease. Soc Sci Med. 16:369–80. 152. Gilbert P, Allan S. (1998) The role of defeat and entrapment (arrested flight) in depression: an exploration of an evolutionary view. Psychol Med. 28(3):585–98. 153. Korte SM, Smit J, Bouws GAH, et al. (1991) Neuroendocrine evidence for hypersensitivity in serotonergic neuronal system after psychosocial stress of defeat. In: Olivier B, Mos J, Slangen J, editors. Animal Models in Psychopharmacology. Birkhauser, Basel, pp 199–203. 154. Willner P, D’Aquila PS, et al. (1995) Loss of social status: preliminary evaluation of a novel animal model of depression. J Pharmacol. 9:207–13. 155. Keeney AJ, Hogg S. (1999) Behavioral consequences of repeated social defeat in the mouse: preliminary evaluations of a potential animal model of depression. Behav Pharmacol. 10:753–64.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

57

156. Keeney AJ, Hogg S, Marsden CA. (2001) Alterations in core body temperature, locomotor activity, and corticosterone following acute and repeated social defeat of male NMRI mice. Psysiol Behav. 74:177–84. 157. Berton O, Durand M, Aguerre S, et al. (1999) Behavioral, neuroendocrine and serotonergic consequences of single social defeat and repeated fluoxetine pretreatment in the Lewis rat strain. Neuroscience. 92(1):327–41. 158. Mitchell PJ, Redfern PH. (1992) Chronic treatment with clomipramine and mianserin increases the hierarchical position of subdominant rats housed in triads. Behav Pharmacol. 3(3):239–47. 159. Joly D, Sanger DJ. (1991) Social competition in rats: a test sensitive to acutely administered anxiolytics. Behav Pharmacol. 2(3):205–13. 160. Henn FA, McKinney WT. (1987) Animal models in psychiatry. In: Meltzer HY, editor. Psychopharamcology: The Third Generation of Progress. Raven, New York, pp 697–704. 161. Hrdina PD, von KP, Stretch R. (1979) Pharmacological modification of experimental depression in infant macaques. Psychopharmacology (Berl). 64(1):89–93. 162. Lewis JK, McKinney WT. (1976) Effect of electroconvulsive shock on the behavior of normal and abnormal rhesus monkeys. Behav Psychiatr. 37:687–93. 163. McKinney WT, Jr, Young LD, Suomi SJ, et al. (1973) Chlorpromazine treatment of disturbed monkeys. Arch Gen Psychiatry. 29(4):490–494. 164. Suomi SJ, Seaman SF, Lewis JK, et al. (1978) Effects of imipramine treatment of separation-induced social disorders in rhesus monkeys. Arch Gen Psychiatry. 35(3):321–5. 165. Lehr E. (1989) Distress call reactivation in isolated chicks: a behavioral indicator with high selectivity for antidepressants. Psychopharmacology (Berl). 97(2):145–6. 166. Molewijk HE, Hartog K, Van Der Poel AM, et al. (1996) Reduction of guinea pig pup isolation calls by anxiolytic and antidepressant drugs. Psychopharmacology (Berl). 128(1):31–8. 167. Berger BD, Schuster R. (1982) An animal model of social interaction: implications for the analysis of drug action. In: Spiegelstein MY and Levy A, editors. Behavioral Models and the Analysis of Drug Action. El Sevier, Amsterdam, pp 415–28. 168. Lewinsohn PM. (1974) A behavioral approach to depression. In: Friedman RJ and Katz MM, editors. The Psychology of Depression: Contemporary Theory and Research. Winston/Wiley, New York, pp 157–85. 169. Willner P, Sampson D, Phillips G, et al. (1989) Effects of isolated housing and chronic antidepressant treatment on cooperative social behaviour in rats. Behav Pharmacol. 1(1):85–90. 170. Overstreet DH, Russell RW, Helps SC, et al. (1979) Selective breeding for sensitivity to the anticholinesterase, DFP. Psychopharmacology. 65:15–20. 171. Overstreet DH, Janowsky DS. (1991) A cholinergic supersensitivity model of depression. In: Boulton A, Baker G, Martin-Iverson M, editors. Neuromethods: Animal Models in Psychiatry. 19th ed. Humana Press, Clifton, pp 81–114. 172. Overstreet DH, Friedman E, Math´e AA, et al. (2005) The Flinders Sensitive Line rat: a selectively bred putative animal model of depression. Neurosci Biobehav Rev. 29(4–5):739–59. 173. Overstreet DH. (1993) The Flinders sensitive line rats: a genetic animal model of depression. Neurosci Biobehav Rev. 17(1):51–68. 174. El YM, Bouali S, Popa D, et al. (2003) Behavioral, neurochemical, and electrophysiological characterization of a genetic mouse model of depression. Proc Natl Acad Sci USA. 100(10):6227–32.

c02

BLBK249-Cutler

58

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

175. Daniela P, Malika EY, Jean-Marie V, et al. (2006) Homeostatic regulation of sleep in a genetic model of depression in the mouse: effects of muscarinic and 5-HT1A receptor activation. Neuropsychopharmacology. 31:1637–46. 176. Crowley JJ, Lucki I. (2005) Opportunities to discover genes regulating depression and antidepressant response from rodent behavioral genetics. Curr Pharm Des. 11:157–69. 177. Sarkisova KY, Midzianovskaia IS, Kulikov MA. (2003) Depressive-like behavioral alterations and c-fos expression in the dopaminergic brain regions in WAG/Rij rats with genetic absence epilepsy. Behav Brain Res. 144(1–2):211–26. 178. Sarkisova KY, Kulikov MA, Midzyanovskaya IS, et al. (2008) Dopamine-dependent nature of depression-like behavior in WAG/Rij rats with genetic absence epilepsy. Neurosci Behav Physiol. 38(2):119–28. 179. Sarkisova KY, Kulikov MA. (2006) Behavioral characteristics of WAG/Rij rats susceptible and non-susceptible to audiogenic seizures. Behav Brain Res. 166(1):9–18. 180. Rezvani AH, Parsian A, Overstreet DH. (2002) The Fawn-Hooded (FH/Wjd) rat: a genetic animal model of comorbid depression and alcoholism. Psychiatr Genet. 12(1):1–16. 181. Overstreet DH, Rezvani AH, Cowen M, et al. (2006) Modulation of high alcohol drinking in the inbred Fawn-Hooded (FH/Wjd) rat strain: implications for treatment. Addict Biol. 11(3–4):356–73. 182. Kantor S, Anheuer ZE, Bagdy G. (2000) High social anxiety and low aggression in Fawn-Hooded rats. Physiol Behav. 71(5):551–7. 183. D’Haenen HAH, den Boer JA, Willner P. (2002) Biological Psychiatry. Wiley-Blackwell, West Sussex, England, pp 716–8. 184. Andreasen NC, Olsen S. (1982) Negative v positive schizophrenia. Definition and validation. Arch Gen Psychiatry. 39(7):789–94. 185. Seeman P. (1987) Dopamine receptors and the dopamine hypothesis of schizophrenia. Synapse. 1(2):133–52. 186. Carlsson A. (1995) The dopamine theory revisited. In: Hirsch SR and Weinberger DR, editors. Schizophrenia. Blackwell Science, Oxford, pp 379–400. 187. Dworkin RH, Opler LA. (1992) Simple schizophrenia, negative symptoms, and prefrontal hypodopaminergia. Am J Psychiatry. 149(9):1284–5. 188. Abi-Dargham A, Gil R, Krystal J, et al. (1998) Increased striatal dopamine transmission in schizophrenia: confirmation in a second cohort. Am J Psychiatry. 155(6):761–7. 189. Breier A, Su TP, Saunders R, et al. (1997) Schizophrenia is associated with elevated amphetamine-induced synaptic dopamine concentrations: evidence from a novel positron emission tomography method. Proc Natl Acad Sci USA. 94(6):2569–74. 190. Laruelle M. (1998) Imaging dopamine transmission in schizophrenia: a review and meta-analysis. Q J Nucl Med. 42(3):211–21. 191. Seeman P, Lee T, Chau-Wong M, et al. (1976) Antipsychotic drug doses and neuroleptic/dopamine receptors. Nature. 261:717–9. 192. Creese I, Burt DR, Snyder SH. (1976) Dopamine receptor binding predicts clinical and pharmacological potencies of antischizophrenic drugs. Science. 192:481–3. 193. Seeman P, Kapur S. (2000) Schizophrenia: more dopamine, more D2 receptors. Proc Natl Acad Sci USA. 97(14):7673–5. 194. Abi-Dargham A, Rodenhiser J, Printz D, et al. (2000) Increased baseline occupancy of D2 receptors by dopamine in schizophrenia. Proc Natl Acad Sci USA. 97(14):8104–9. 195. Crawley JC, Owens DG, Crow TJ, et al. (1986) Dopamine D2 receptors in schizophrenia studied in vivo. Lancet. 2(8500):224–5. 196. Laruelle M, Abi-Dargham A, Gil R, et al. (1999) Increased dopamine transmission in schizophrenia: relationship to illness phases. Biol Psychiatry. 46(1):56–72.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

59

197. Randrup A, Munkvad I. (1965) Special antagonism of amphetamine-induced abnormal behavior. Psychopharmacologia. 7:416–22. 198. Angrist BM, Gershon S. (1970) The phenomenology of experimentally induced amphetamine psychosis—preliminary observations. Biol Psychiatry. 2(2):95–107. 199. Ellison GD, Eison MS. (1983) Continuous amphetamine intoxication: an animal model of the acute psychotic episode. Psychol Med. 13(4):751–61. 200. Eichler AJ, Antelman SM, Black CA. (1980) Amphetamine stereotypy is not a homogeneous phenomenon: sniffing and licking show distinct profiles of sensitization and tolerance. Psychopharmacology. 68(3):287–90. 201. Lieberman JA, Alvir J, Geisler S, et al. (1994) Methylphenidate response, psychopathology and tardive dyskinesia as predictors of relapse in schizophrenia. Neuropsychopharmacology. 11(2):107–18. 202. Featherstone RE, Kapur S, Fletcher PJ. (2007) The amphetamine-induced sensitized state as a model of schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry. 31(8):1556–71. 203. Tenn CC, Fletcher PJ, Kapur S. (2003) Amphetamine-sensitized animals show a sensorimotor gating and neurochemical abnormality similar to that of schizophrenia. Schizophr Res. 64(2–3):103–14. 204. Swerdlow NR, Braff DL, Taaid N, et al. (1994) Assessing the validity of an animal model of deficient sensorimotor gating in schizophrenic patients. Arch Gen Psychiatry. 51(2):139–54. 205. Tamminga CA, Schaffer MH, Smith RC, et al. (1978) Schizophrenic symptoms improve with apomorphine. Science. 200(4341):567–8. 206. Bruhwyler J, Chkeide E, Houbeau G, et al. (1993) Differentiation of haloperidol and clozapine using a complex operant schedule in the dog. Pharmacol Biochem Behav. 44(1):181–9. 207. Stone JM, Morrison PD, Pilowsky LS. (2007) Glutamate and dopamine dysregulation in schizophrenia—a synthesis and selective review. J Psychopharmacol. 21(4):440– 52. 208. Aghajanian GK, Marek GJ. (2000) Serotonin model of schizophrenia: emerging role of glutamate mechanisms. Brain Res Brain Res Rev. 31(2–3):302–12. 209. Stone JM, Pilowsky LS. (2007) Novel targets for drugs in schizophrenia. CNS Neurol Disord Drug Targets. 6(4):265–72. 210. Cowan WM, Harter DH, Kandel ER. (2000) The emergence of modern neuroscience: some implications for neurology and psychiatry. Annu Rev Neurosci. 23:343–91. 211. Grace AA. (2000) Gating of information flow within the limbic system and the pathophysiology of schizophrenia. Brain Res Brain Res Rev. 31(2–3):330–41. 212. Crow TJ. (1987) Two syndromes of schizophrenia as one pole of the continuum of psychosis: a concept of the nature of the pathogen and its genomic locus. In: Henn FA and DeLisi LE, editors. Handbook of schizophrenia: the neurochemistry and neuropharmacology of schizophrenia. Elsevier, Amsterdam, pp 17–48. 213. Woolley DW, Shaw W. (1954) A biochemical and pharmacological suggestion about certain mental disorders. Proc Natl Acad Sci USA. 40:228–31. 214. Shaw E, Woolley DW. (1956) Some serotonin-like activities of lysergic acid diethylamide. Science. 124:121–3. 215. Titeler M, Lyon RA, Glennon RA. (1988) Radioligand binding evidence implicates the brain 5-HT2A receptor as a site of action for LSD and phenylisopropanylamine hallucinogens. Psychopharacology. 94:213–6. 216. Meltzer HY. (1996) Preclinical pharmacology of atypical antipsychotic drugs: a selective review. Br J Psychiatry Suppl. 29:23–31.

c02

BLBK249-Cutler

60

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

217. Arranz MJ, Munro J, Sham P, et al. (1998) Meta-analysis of studies on genetic variation in 5-HT2A receptors and clozapine response. Schizophr Res. 32(2):93–9. 218. Geyer MA, Braff DL. (1987) Startle habituation and sensorimotor gating in schizophrenia and related animal models. Schizophr Bull. 13(4):643–68. 219. Geyer MA. (1998) Behavioral studies of hallucinogenic drugs in animals: implications for schizophrenia research. Pharmacopsychiatry. 31(Suppl 2):73–9. 220. Ouagazzal A, Grottick AJ, Moreau J, et al. (2001) Effect of LSD on prepulse inhibition and spontaneous behavior in the rat: a pharmacological analysis and comparison between two rat strains. Neuropsychopharmacology. 25(4):565–75. 221. Yamada S, Harano M, Annoh N, et al. (1999) Involvement of serotonin 2A receptors in phencyclidine-induced disruption of prepulse inhibition of the acoustic startle in rats. Biol Psychiatry. 46(6):832–8. 222. Weil-Malherbe H, Szara SI. (1971) The Biochemistry of Functional and Experimental Psychoses. Charles C. Thomas Publisher, Springfield. 223. Strassman RJ, Qualls CR, Berg LM. (1996) Differential tolerance to biological and subjective effects of four closely spaced doses of N,N-dimethyltryptamine in humans. Biol Psychiatry. 39(9):784–95. 224. Geyer MA. (1998) Behavioral studies of hallucinogenic drugs in animals; implications for schizophrenia research. Pharmacopsychiatry. 2:73–9. 225. Gonz´alez-Maeso J, Ang RL, Yuen T, et al. (2008) Identification of a serotonin/glutamate receptor complex implicated in psychosis. Nature. 452(7183):93–7. 226. Winter JC, Eckler JR, Rabin RA. (2004) Serotonergic/glutamatergic interactions: the effects of mGlu2/3 receptor ligands in rats trained with LSD and PCP as discriminative stimuli. Psychopharmacology (Berl). 172(2):233–40. 227. Braff DL, Geyer MA. (1980) Acute and chronic LSD effects on rat startle: data supporting an LSD—rat model of schizophrenia. Biol Psychiatry. 15(6):909–16. 228. Sherman AD, Davidson AT, Baruah S, et al. (1991) Evidence of glutamatergic deficiency in schizophrenia. Neurosci Lett. 121(1–2):77–80. 229. Steinpreis RE. (1996) The behavioral and neurochemical effects of phencyclidine in humans and animals: some implications for modeling psychosis. Behav Brain Res. 74:45– 55. 230. Breier A, Adler CM, Weisenfeld N, et al. (1998) Effects of NMDA antagonism on striatal dopamine release in healthy subjects: application of a novel PET approach. Synapse. 29(2):142–7. 231. Malhotra AK, Pinals DA, Adler CM, et al. (1997) Ketamine-induced exacerbation of psychotic symptoms and cognitive impairment in neuroleptic-free schizophrenics. Neuropsychopharmacology. 17(3):141–50. 232. Javitt DC, Zukin SR. (1991) Recent advances in the phencyclidine model of schizophrenia. Am J Psychiatry. 148(10):1301–8. 233. Lahti AC, Holcomb HH, Gao XM, et al. (1999) NMDA-sensitive glutamate antagonism: a human model for psychosis. Neuropsychopharmacology. 21:S158–S69. 234. Lodge D, Aran JA, Church J, et al. (1987) Excitatory amino acids and phencyclidine drugs. In: Hicks TP, Lodge D, McLennan H, editors. Excitatory amino acid transmission. Alan Liss, New York, pp 83–90. 235. Coyle JT. (1996) The glutamatergic dysfunction hypothesis for schizophrenia. Harv Rev Psychiatry. 3(5):241–53. 236. Deutsch SI, Mastropaolo J, Schwartz BL, et al. (1989) A “glutamatergic hypothesis” of schizophrenia: rationale for pharmacotherapy with glycine. Clin Neuropharmacol. 12(1):1–13.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

61

237. Olney JW, Farber NB. (1995) Glutamate receptor dysfunction and schizophrenia. Arch Gen Psychiatry. 52(12):998–1007. 238. Mouri A, Noda Y, Enomoto T, et al. (2007) Phencyclidine animal models of schizophrenia: approaches from abnormality of glutamatergic neurotransmission and neurodevelopment. Neurochem Int. 51(2–4):173–84. 239. Geyer MA, Segal DS, Greenberg BD. (1984) Increased startle responding in rats treated with phencyclidine. Neurobehav Toxicol Teratol. 6:1–4. 240. Swerdlow NR, Geyer MA. (1998) Using an animal model of deficient sensorimotor gating to study the pathophysiology and new treatments of schizophrenia. Schizophr Bull. 24:285–301. 241. Turgeon SM, Auerbach EA, Heller MA. (1998) The delayed effects of phencyclidine (PCP) disrupt latent inhibition in a conditioned taste aversion paradigm. Pharmacol Biochem Behav. 60(2):553–8. 242. Jentsch JD, Redmond DE, Jr, Elsworth JD, et al. (1997) Enduring cognitive deficits and cortical dopamine dysfunction in monkeys after long-term administration of phencyclidine. Science. 277:953–5. 243. Malhotra AK, Adler CM, Kennison SD, et al. (1997) Clozapine blunts N-methyld-aspartate antagonist-induced psychosis: a study with ketamine. Biol Psychiatry. 42:664–8. 244. Sams-Dodd F. (1998) A test of the predictive validity of animal models of schizophrenia based on phencyclidine and d-amphetamine. Neuropsychopharmacology. 18(4):293–304. 245. Kargieman L, Santana N, Mengod G, et al. (2007) Antipsychotic drugs reverse the disruption in prefrontal cortex function produced by NMDA receptor blockade with phencyclidine. Proc Natl Acad Sci USA. 104(37):14843–8. 246. Fell MJ, Svensson KA, Johnson BG, et al. (2008) Evidence for the role of metabotropic glutamate (mGlu)2 not mGlu3 receptors in the preclinical antipsychotic pharmacology of the mGlu2/3 receptor agonist (−)-(1R,4S,5S,6S)-4-amino-2-sulfonylbicyclo 3 1 0 hexane-4,6-dicarboxylic acid (LY404039). J Pharmacol Exp Ther. 326(1):209–17. 247. Lane HY, Liu YC, Huang CL, et al. (2007) Sarcosine (N-methylglycine) treatment for acute schizophrenia: a randomized, double-blind study. Biol Psychiatry. 63(1):9–12. 248. Patil ST, Zhang L, Martenyi F, et al. (2007) Activation of mGlu2/3 receptors as a new approach to treat schizophrenia: a randomized phase 2 clinical trial. Nat Med. 13:1102– 7. 249. Matsuzaki M, Dowling KC. (1985) Phencyclidine (PCP): effects of acute and chronic administration on EEG activities in the rhesus monkey. Electroencephalogr Clin Neurophysiol. 60(4):356–66. 250. Hori T, Subramaniam S, Srivastava LK, et al. (2000) Behavioral and neurochemical alterations following repeated phencyclidine administration in rats with neonatal ventral hippocampal lesions. Neuropharmacology. 39(12):2478–91. 251. Simpson MD, Slater P, Deakin JF, et al. (1989) Reduced GABA uptake sites in the temporal lobe in schizophrenia. Neurosci Lett. 107(1–3):211–5. 252. Kaplan HI, Sadock BJ. (1995) Comprehensive Textbook of Psychiatry. Williams & Wilkins, Baltimore. 253. Benes FM, Vincent SL, Alsterberg G, et al. (1992) Increased GABAA receptor binding in superficial layers of cingulate cortex in schizophrenics. J Neurosci. 12(3):924–9. 254. Volk DW, Austin MC, Pierri JN, et al. (2000) Decreased glutamic acid decarboxylase67 messenger RNA expression in a subset of prefrontal cortical gamma-aminobutyric acid neurons in subjects with schizophrenia. Arch Gen Psychiatry. 57(3):237–45.

c02

BLBK249-Cutler

62

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

255. Japha K, Koch M. (1999) Picrotoxin in the medial prefrontal cortex impairs sensorimotor gating in rats: reversal by haloperidol. Psychopharmacology (Berl). 144(4):347–54. 256. Boulton AA, Baker GB, Martin-Iverson MT. (1991) Animal Models in Psychiatry. Humana Press, Clifton, New Jersey. 257. Soares JC, Gershon S. (2003) Handbook of Medical Psychiatry. Informa Health Care, London England, p 287. 258. Gisabella B, Bolshakov VY, Benes FM. (2005) Regulation of synaptic plasticity in a schizophrenia model. Proc Natl Acad Sci USA. 102(37):13301–6. 259. Zhou Y, Shu N, Liu Y, et al. (2008) Altered resting-state functional connectivity and anatomical connectivity of hippocampus in schizophrenia. Schizophr Res. 100(1–3):120–32. 260. Gothelf D, Soreni N, Nachman RP, et al. (2000) Evidence for the involvement of the hippocampus in the pathophysiology of schizophrenia. Eur Neuropsychopharmacol. 10(5):389–95. 261. Waddington JL, Lane A, Scully PJ, et al. (1998) Neurodevelopmental and neuroprogressive processes in schizophrenia: antithetical or complementary, over a lifetime trajectory of disease? Psychiatr Clin North Am. 21(1):123–49. 262. DeLisi LE. (1999) Regional brain volume change over the life-time course of schizophrenia. J Psychiatr Res. 33(6):535–41. 263. McGlashan TH, Fenton WS. (1993) Subtype progression and pathophysiologic deterioration in early schizophrenia. Schizophr Bull. 19(1):71–84. 264. Holmes AJ, MacDonald A, III, Carter CS, et al. (2005) Prefrontal functioning during context processing in schizophrenia and major depression: an event-related fMRI study. Schizophr Res. 76(2–3):199–206. 265. Davis KL, Charney D, Coyle JT, et al. (2002) Neuropsychopharmacology: The Fifth Generation of Progress: An Official Publication of the American College of Neuropsychopharmacology. Lippincott Williams & Wilkins, Philadelphia, Pennsylvania, p 747. ¨ 266. Soyka M, Koch W, Moller HJ, et al. (2005) Hypermetabolic pattern in frontal cortex and other brain regions in unmedicated schizophrenia patients: results from a FDG-PET study. Eur Arch Psychiatry Clin Neurosci. 255(5):308–12. 267. Lazar NL, Rajakumar N, Cain DP. (2008) Injections of NGF into neonatal frontal cortex decrease social interaction as adults: a rat model of schizophrenia. Schizophr Bull. 34(1):127–36. 268. Jaskiw GE, Karoum FK, Weinberger DR. (1990) Persistent elevations in dopamine and its metabolites in the nucleus accumbens after mild subchronic stress in rats with ibotenic acid lesions of the medial prefrontal cortex. Brain Res. 534(1–2):321–3. 269. Braun AR, Jaskiw GE, Vladar K, et al. (1993) Effects of ibotenic acid lesion of the medial prefrontal cortex on dopamine agonist-related behaviors in the rat. Pharmacol Biochem Behav. 46(1):51–60. 270. Lipska BK, Jaskiw GE, Braun AR, et al. (1995) Prefrontal cortical and hippocampal modulation of haloperidol-induced catalepsy and apomorphine-induced stereotypic behaviors in the rat. Biol Psychiatry. 38(4):255–62. 271. Swerdlow NR, Lipska BK, Weinberger DR, et al. (1995) Increased sensitivity to the sensorimotor gating-disruptive effects of apomorphine after lesions of medial prefrontal cortex or ventral hippocampus in adult rats. Psychopharmacology (Berl). 122(1):27–34. 272. Brozoski TJ, Brown RM, Roswold HE, et al. (1979) Cognitive deficit caused by regional depletion of dopamine in prefrontal cortex of Rhesus monkeys. Science. 205:929–32. 273. Sesack SR, Pickel VM. (1990) In the rat medial nucleus accumbens, hippocampal and catecholaminergic terminals converge on spiny neurons and are in apposition to each other. Brain Res. 527(2):266–79.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

63

274. Mittleman G, Bratt AM, Chase R. (1998) Heterogeneity of the hippocampus: effects of subfield lesions on locomotion elicited by dopaminergic agonists. Behav Brain Res. 92(1):31–45. 275. Bardgett ME, Jacobs PS, Jackson JL, et al. (1997) Kainic acid lesions enhance locomotor responses to novelty, saline, amphetamine, and MK-801. Behav Brain Res. 84(1–2):47–55. 276. Devenport LD, Devenport JA, Holloway FA. (1981) Reward induced stereotypy: modulation by the hippocampus. Science. 212(4500):1288–9. 277. Oades RD, Isaacson RL. (1978) The development of food search behavior by rats: the effects of hippocampal damage and haloperidol treatment. Behav bio. 24:327– 37. 278. Levin ED, Christopher NC. (2006) Effects of clozapine on memory function in the rat neonatal hippocampal lesion model of schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry. 30(2):223–9. 279. Harrison PJ. (1999) The neuropathology of schizophrenia: a critical review of the data and their interpretation. Brain. 122:593–624. 280. Weinberger DR, Lipska BK. (1995) Cortical maldevelopment, anti-psychotic drugs, and schizophrenia: a search for common ground. Schizophr Res. 16:87–110. 281. Jablensky A. (1997) The 100-year epidemiology of schizophrenia. Schizophr Res. 28:111–25. 282. MacDonald C, Murray RM. (2000) Early and late environmental risk factors for schizophrenia. Brain Res Rev. 31:130–7. 283. Waddington JL, O’Callaghan E, Buckley P, et al. (1995) Tardive dyskinesia in schizophrenia: relationship to minor physical anomalies, frontal lobe dysfunction, and cerebral structure on magnetic resonance imaging. Br J Psychiatry. 167(1):41–4. 284. Fish B, Marcus J, Hans SL, et al. (1992) Infants at risk for schizophrenia: sequelae of a genetic neurointegrative defect, a review and replication analysis of pandysmaturation in the Jerusalem Infant Development Study. Arch Gen Psychiatry. 49(3):221–35. 285. Weinberger DR, Berman KF, Zec RF. (1986) Physiologic dysfunction of dorsolateral prefrontal cortex in schizophrenia: I regional cerebral blood flow evidence. Arch Gen Psychiatry. 43(2):114–24. 286. Stevens JR. (1992) Abnormal reinnervation as a basis for schizophrenia: a hypothesis. Arch Gen Psychiatry. 49(3):238–43. 287. Lang MV. (2004) Trends in Schizophrenia Research. Nova Publishers, New York. 288. Lipska BK, Jaskiw GE, Weinberger DR. (1993) Postpubertal emergence of hyperresponsiveness to stress and to amphetamine after neonatal excitotoxic hippocampal damage: a potential animal model of schizophrenia. Neuropsychopharmacology. 9(1):67– 75. 289. Al-Amin HA, Shannon WC, Weinberger DR, et al. (2001) Delayed onset of enhanced MK-801-induced motor hyperactivity after neonatal lesions of the rat ventral hippocampus. Biol Psychiatry. 49(6):528–39. 290. Becker A, Grecksch G, Bernsteinn HG, et al. (1999) Social behavior in rats lesioned with ibotenic acid in the hippocampus: quantitative and qualitative analysis. Psychopharmacology. 144:333–8. 291. Flores G, Barbeau D, Quirion R, et al. (1996) Decreased binding of dopamine D3 receptors in limbic subregions after neonatal bilateral lesion of rat hippocampus. J Neurosci. 16(6):2020–6. 292. Lipska BK, Weinberger DR. (1994) Subchronic treatment with haloperidol and clozapine in rats with neonatal excitotoxic hippocampal damage. Neuropsychopharmacology. 10(3):199–205.

c02

BLBK249-Cutler

64

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

293. Le PG, Moreau JL. (2002) Disruption of prepulse inhibition of startle reflex in a neurodevelopmental model of schizophrenia: reversal by clozapine, olanzapine and risperidone but not by haloperidol. Neuropsychopharmacology. 27(1):1–11. 294. Bachevalier J, Alvarado MC, Malkova L. (1999) Memory and socioemotional behavior in monkeys after hippocampal damage incurred in infancy or in adulthood. Biol Psychiatry. 46(3):329–39. 295. Rascle C, Mazas O, Vaiva G, et al. (2001) Clinical features of latent inhibition in schizophrenia. Schizophr Res. 51(2–3):149–61. 296. Swerdlow NR, Braff DL, Hartston H, et al. (1996) Latent inhibition in schizophrenia. Schizophr Res. 20(1–2):91–103. 297. Russig H, Murphy CA, Feldon J. (2002) Clozapine and haloperidol reinstate latent inhibition following its disruption during amphetamine withdrawal. Neuropsychopharmacology. 26(6):765–77. 298. Moser PC, Hitchcock JM, Lister S, et al. (2000) The pharmacology of latent inhibition as an animal model of schizophrenia. Brain Res Brain Res Rev. 33(2–3):275–307. 299. Gray NS, Pilowsky LS, Gray JA, et al. (1995) Latent inhibition in drug naive schizophrenics: relationship to duration of illness and dopamine D2 binding using SPET. Schizophr Res. 17(1):95–107. 300. Swerdlow NR, Braff DL, Hartston H, et al. (1996) Amphetamine-induced disruptions of latent inhibition are reinforcer mediated: implications for animal models of schizophrenic attentional dysfunction. Schizophr Res. 20(1–2):91–103. 301. Williams JH, Wellman NA, Geaney DP, et al. (1998) Reduced latent inhibition in people with schizophrenia: an effect of psychosis or of its treatment. Br J Psychiatry. 172:243–9. 302. Van Den Buuse M, Garner B, Koch M. (2003) Neurodevelopmental animal models of schizophrenia: effects on prepulse inhibition. Curr Mol Med. 3(5):459–71. 303. Geyer MA, Krebs-Thomson K, Braff DL, et al. (2001) Pharmacological studies of prepulse inhibition models of sensorimotor gating deficits in schizophrenia: a decade in review. Psychopharmacology (Berl). 156(2–3):117–54. 304. Mansbach RS, Geyer MA, Braff DL. (1988) Dopaminergic stimulation disrupts sensorimotor gating in the rat. Psychopharmacology (Berl). 94(4):507–14. 305. Geyer MA, Markou A. (1991) Animal models of psychiatric disorders. In: Kupfer DJ, editor. Psychopharmacology: The Fourth Generation of Progress. Raven Press, New York, pp 787–98. 306. Kilts CD. (2001) The changing roles and targets for animal models of schizophrenia. Biol Psychiatry. 50(11):845–55. 307. Wadenberg ML, Hicks PB. (1999) The conditioned avoidance response test reevaluated: is it a sensitive test for the detection of potentially atypical antipsychotics? Neurosci Biobehav Rev. 23(6):851–62. 308. Arnt J. (1982) Pharmacological specificity of conditioned avoidance response inhibition in rats: inhibition by neuroleptics and correlation to dopamine receptor blockade. Acta Pharmacol Toxicol (Copenh). 51(4):321–9. 309. Reynolds GP, Czudek C. (1995) New approaches to the drug treatment of schizophrenia. Adv Pharmacol. 32:461–503. 310. Aguilar MA, Mar´ı-Sanmill´an MI, Morant-Deusa JJ, et al. (2000) Different inhibition of conditioned avoidance response by clozapine and DA D1 and D2 antagonists in male mice. Behav Neurosci. 114(2):389–400. 311. Beninger RJ. (1989) The role of serotonin and dopamine in learning to avoid aversive stimuli. In: Archer T, Nilsson L, editors. Aversion, Avoidance, and Anxiety: Perspective on Aversively Motivated Behavior. Lawrence Erlbaum Associates, Hillsdale, pp 265–84.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

65

312. Enomoto T, Noda Y, Nabeshima T. (2007) Phencyclidine and genetic animal models of schizophrenia developed in relation to the glutamate hypothesis. Methods Find Exp Clin Pharmacol. 29(4):291–301. 313. Duncan GE, Moy SS, Perez A, et al. (2004) Deficits in sensorimotor gating and tests of social behavior in a genetic model of reduced NMDA receptor function. Behav Brain Res. 153:507–19. 314. Duncan GE, Moy SS, Lieberman JA, et al. (2006) Typical and atypical antipsychotic drug effects on locomotor hyperactivity and deficits in sensorimotor gating in a genetic model of NMDA receptor hypofunction. Pharmacol Biochem Behav. 85(3):481–91. 315. Nishiguchi N, Shirakawa O, Ono H, et al. (2000) Novel polymorphism in the gene region encoding the carboxyl-terminal intracellular domain of the NMDA receptor 2B subunit: analysis of association with schizophrenia. Am J Psychiatry. 157:1329–31. 316. Savonenko AV, Melnikova T, Laird FM, et al. (2008) Alteration of BACE1-dependent NRG1/ErbB4 signaling and schizophrenia-like phenotypes in BACE1-null mice. Proc Natl Acad Sci USA. 105(14):5585–90. 317. Stefansson H, Sigurdsson E, Steinthorsdottir V, et al. (2002) Neuregulin 1 and susceptibility to schizophrenia. Am J Hum Genet. 71(4):877–92. 318. Chubb JE, Bradshaw NJ, Soares DC, et al. (2008) The DISC locus in psychiatric illness. Mol Psychiatry. 13(1):36–64. 319. Hennah W, Thomson P, Peltonen L, et al. (2006) Genes and schizophrenia: beyond schizophrenia: the role of DISC1 in major mental illness. Schizophr Bull. 32(3):409–16. 320. Hikida T, Jaaro-Peled H, Seshadri S, et al. (2007) Dominant-negative DISC1 transgenic mice display schizophrenia-associated phenotypes detected by measures translatable to humans. Proc Natl Acad Sci USA. 104(36):14501–6. 321. Hikida T, Jaaro-Peled H, Seshadri S, et al. (2008) How has DISC1 enabled drug discovery? Mol Cell Neurosci. 37(2):187–95. 322. Burris KD, Molski TF, Xu C, et al. (2002) Aripiprazole, a novel antipsychotic, is a high-affinity partial agonist at human dopamine D2 receptors. J Pharmacol Exp Ther. 302(1):381–9. 323. Fleischhacker WW. (2005) Aripiprazole. Expert Opin Pharmacother. 6:2091–3101. 324. Auclair AL, Kleven MS, Besnard J, et al. (2006) Actions of novel antipsychotic agents on apomorphine-induced PPI disruption: influence of combined serotonin 5-HT1A receptor activation and dopamine D2 receptor blockade. Neuropsychopharmacology. 31(9):1900–9. 325. Bardin L, Kleven MS, Barret-Gr´evoz C, et al. (2006) Antipsychotic-like vs cataleptogenic actions in mice of novel antipsychotics having D2 antagonist and 5-HT1A agonist properties. Neuropsychopharmacology. 31(9):1869–79. 326. Andersen MP, Pouzet B. (2001) Effects of acute versus chronic treatment with typical or atypical antipsychotics on d-amphetamine-induced sensorimotor gating deficits in rats. Psychopharmacology (Berl). 156(2–3):291–304. 327. Nordquist RE, Risterucci C, Moreau JL, et al. (2008) Effects of aripiprazole/OPC-14597 on motor activity, pharmacological models of psychosis, and brain activity in rats. Neuropharmacology. 54(2):405–16. 328. Bruins SLA, Kleven MS, Newman-Tancredi A. (2005) Effects of novel antipsychotics with mixed D(2) antagonist/5-HT(1A) agonist properties on PCP-induced social interaction deficits in the rat. Neuropharmacology. 49(7):996–1006. 329. Fejgin K, Safonov S, P˚alsson E, et al. (2007) The atypical antipsychotic, aripiprazole, blocks phencyclidine-induced disruption of prepulse inhibition in mice. Psychopharmacology (Berl). 191(2):377–85.

c02

BLBK249-Cutler

66

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

330. Hirose T, Uwahodo Y, Yamada S, et al. (2004) Mechanism of action of aripiprazole predicts clinical efficacy and a favourable side-effect profile. J Psychopharmacol. 18(3):375–83. 331. El-Sayeh HG, Morganti C. (2006) Aripiprazole for schizophrenia. Cochrane Database Syst Rev. (2):CD004578. 332. Hauw JJ, Duyckaerts C. (2001) Alzheimer’s Disease. In: Duckett S, De La Torre JC, editors. Pathology of the aging human nervous system. Oxford University Press, New York, pp 207–63. 333. Fukumoto H, Asami-Odaka A, Suzuki N, et al. (1996) Amyloid beta protein deposition in normal aging has the same characteristics as that in Alzheimer’s disease: predominance of A beta 42(43) and association of A beta 40 with cored plaques. Am J Pathol. 148; 259–65. 334. Ingram DK, Spangler EL, Iijima S, et al. (1994) Rodent models of memory dysfunction in Alzheimer’s disease and normal aging: moving beyond the cholinergic hypothesis. Life Sci. 55(25–26):2037–49. 335. Walker LC, Cork LC. (1999) The neurobiology of aging in nonhuman primates. In: Terry RD, Katzman R, Bick KL, Sisodia SS, editors. Alzheimer Disease. Williams and Wilkins, Philadelphia, Pennsylvania, pp 233–43. 336. Voytko ML, Tinkler GP. (2004) Cognitive function and its neural mechanisms in nonhuman primate models of aging, Alzheimer disease, and menopause. Front Biosci. 9:1899–914. 337. Tapp PD, Siwak CT, Gao FQ, et al. (2004) Frontal lobe volume, function, and betaamyloid pathology in a canine model of aging. J Neurosci. 24(38):8205–13. 338. Wegiel J, Wisniewski HM, Dziewiatkowski J, et al. (1995) The origin of amyloid in cerebral vessels of aged dogs. Brain Res. 705(1–2):225–34. 339. Studzinski CM, Araujo JA, Milgram NW. (2005) The canine model of human cognitive aging and dementia: pharmacological validity of the model for assessment of human cognitive-enhancing drugs. Prog Neuropsychopharmacol Biol Psychiatry. 29(3):489–98. 340. Rupniak NM, Tye SJ, Field MJ. (1997) Enhanced performance of spatial and visual recognition memory tasks by the selective acetylcholinesterase inhibitor E2020 in rhesus monkeys. Psychopharmacology (Berl). 131(4):406–10. 341. Rupniak NMJ. (1992) Profile of cholinomimetic drugs in primates: status of screens for potential Alzheimer therapies. Drug Dev Res. 27(2):77–88. 342. Milgram NW, Head E, Zicker SC, et al. (2005) Learning ability in aged beagle dogs is preserved by behavioral enrichment and dietary fortification: a two-year longitudinal study. Neurobiol Aging. 26(1):77–90. 343. Siwak CT, Tapp PD, Head E, et al. (2005) Chronic antioxidant and mitochondrial cofactor administration improves discrimination learning in aged but not young dogs. Prog Neuropsychopharmacol Biol Psychiatry. 29(3):461–9. 344. Walker LC, Ibegbu CC, Todd CW, et al. (2005) Emerging prospects for the diseasemodifying treatment of Alzheimer’s disease. Biochem Pharmacol. 69(7):1001–8. 345. Perry EK. (1986) The cholinergic hypothesis—ten years on. Br Med Bull. 42(1):63–9. 346. Buccafusco JJ, Terry AV, Jr, Webster SJ, et al. (2008) The scopolamine-reversal paradigm in rats and monkeys: the importance of computer-assisted operant-conditioning memory tasks for screening drug candidates. Psychopharmacology (Berl). 199(3):481–94. 347. Prohovnik I, Arnold SE, Smith G, et al. (1997) Physostigmine reversal of scopolamineinduced hypofrontality. J Cereb Blood Flow Metab. 17(2):220–8. 348. Capsoni S, Giannotta S, Stebel M, et al. (2004) Ganstigmine and donepezil improve neurodegeneration in AD11 antinerve growth factor transgenic mice. Am J Alzheimers Dis Other Demen. 19(3):153–60.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

67

349. Ebert U, Kirch W. (1998) Scopolamine model of dementia: electroencephalogram findings and cognitive performance. Eur J Clin Invest. 28(11):944–9. 350. Sarter M, Hagan J, Dudchenko P. (1992) Behavioral screening for cognition enhancers: from indiscriminate to valid testing: Part I. Psychopharmacology (Berl). 107(2–3):144–59. 351. Sarter M, Hagan J, Dudchenko P. (1992) Behavioral screening for cognition enhancers: from indiscriminate to valid testing: Part II. Psychopharmacology (Berl). 107(4):461–73. 352. Salamone JD. (1991) Strategies for drug development in the treatment of dementia. In: Wilner P, editor. Behavioral Models in Psychopharmacology. Cambridge University Press, Cambridge, UK, pp 491–36. 353. Herholz K, Weisenbach S, Kalbe E. (2008) Deficits of the cholinergic system in early AD. Neuropsychologia. 46(6):1642–7. 354. Hagan JJ, Jansen JH, Nefkens FE, et al. (1990) Therapeutic effect of THA on hemicholinium-3-induced learning impairment is independent of serotonergic and noradrenergic systems. Psychopharmacology (Berl). 101(3):376–83. 355. Scattoni ML, Adriani W, Calamandrei G, et al. (2006) Long-term effects of neonatal basal forebrain cholinergic lesions on radial maze learning and impulsivity in rats. Behav Pharmacol. 17(5–6):517–24. 356. Li L, An X, Qiao J. (1998) Pathological characteristic of Alzheimer’s disease produced by lesion in nucleus basalis of Meynert in rats. Chin Med J (Engl). 111(7):638–40. 357. Pepeu G, Casamenti F, Pedata F, et al. (1986) Are the neurochemical and behavioral changes induced by lesions of the nucleus basalis in the rat a model of Alzheimer’s disease? Prog Neuropsychopharmacol Biol Psychiatry. 10(3–5):541–51. 358. Dunnet SB, Barth TM. (1991) Animal models of Alzheimer’s disease and dementia with an emphasis on cholinergic systems. In: Willner P, editor. Behavioral Models in Psychopharmacology. Cambridge University Press, Cambridge, UK, pp 359– 418. 359. Book AA, Wiley RG, Schweitzer JB. (1994) 192 IgG-saporin: I, specific lethality for cholinergic neurons in the basal forebrain of the rat. J Neuropathol Exp Neurol. 53(1):95–102. 360. Leanza G. (1998) Chronic elevation of amyloid precursor protein expression in the neocortex and hippocampus of rats with selective cholinergic lesions. Neurosci Lett. 257(1):53–6. 361. Nakamura S, Murayama N, Noshita T, et al. (2001) Progressive brain dysfunction following intracerebroventricular infusion of beta(1-42)-amyloid peptide. Brain Res. 912(2):128–36. 362. Meziane H, Dodart JC, Mathis C, et al. (1998) Memory-enhancing effects of secreted forms of the beta-amyloid precursor protein in normal and amnestic mice. Proc Natl Acad Sci USA. 95(21):12683–8. 363. Wilson CA, Doms RW, Lee VM. (1999) Intracellular APP processing and A beta production in Alzheimer disease. J Neuropathol Exp Neurol. 58(8):787–94. 364. Bard F, Cannon C, Barbour R, et al. (2000) Peripherally administered antibodies against amyloid beta-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nat Med. 6(8):916–9. 365. Lace GL, Wharton SB, Ince PG. (2007) A brief history of tau: the evolving view of the microtubule-associated protein tau in neurodegenerative diseases. Clin Neuropathol. 26(2):43–58. 366. Games D, Adams D, Alessandrini R, et al. (1995) Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature. 373(6514):523–7.

c02

BLBK249-Cutler

68

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Chapter 2

367. Dodart JC, Meziane H, Mathis C, et al. (1999) Behavioral disturbances in transgenic mice overexpressing the V717F beta-amyloid precursor protein. Behav Neurosci. 113(5):982–90. 368. Schenk D, Barbour R, Dunn W, et al. (1999) Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature. 400(6740):173–7. 369. Hsiao K, Chapman P, Nilsen S, et al. (1996) Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science. 274(5284):99–102. 370. Wong AA, Brown RE. (2006) Visual detection, pattern discrimination and visual acuity in 14 strains of mice. Genes Brain Behav. 5(5):389–403. 371. Bornemann KD, Staufenbiel M. (2000) Transgenic mouse models of Alzheimer’s disease. Ann NY Acad Sci. 908:260–6. 372. Calhoun ME, Burgermeister P, Phinney AL, et al. (1999) Neuronal overexpression of mutant amyloid precursor protein results in prominent deposition of cerebrovascular amyloid. Proc Natl Acad Sci USA. 96(24):14088–93. 373. Hyde LA, Kazdoba TM, Grilli M, et al. (2006) Age-progressing cognitive impairments and neuropathology in transgenic CRND8 mice. J Clin Invest. 116(1):193– 201. 374. Levites Y, Das P, Price RW, et al. (2006) Anti-Abeta42- and anti-Abeta40-specific mAbs attenuate amyloid deposition in an Alzheimer disease mouse model. J Clin Invest. 116(1):193–201. 375. Matsuoka Y, Picciano M, Malester B, et al. (2001) Inflammatory responses to amyloidosis in a transgenic mouse model of Alzheimer’s disease. Am J Pathol. 158(4):1345– 54. 376. Shen J, Kelleher RJ, III. (2007) The presenilin hypothesis of Alzheimer’s disease: evidence for a loss-of-function pathogenic mechanism. Proc Natl Acad Sci USA. 104(2):403–409. 377. Lee VM, Goedert M, Trojanowski JQ. (2001) Neurodegenerative tauopathies. Annu Rev Neurosci. 24:1121–59. 378. Ishihara T, Hong M, Zhang B, et al. (1999) Age-dependent emergence and progression of a tauopathy in transgenic mice overexpressing the shortest human tau isoform. Neuron. 24(3):751–62. 379. Boutajangout A, Leroy K, Touchet N, et al. (2002) Increased tau phosphorylation but absence of formation of neurofibrillary tangles in mice double transgenic for human tau and Alzheimer mutant (M146L) presenilin-1. Neurosci Lett. 318(1):29–33. 380. Aoyagi H, Hasegawa M, Tamaoka A. (2007) Fibrillogenic nuclei composed of P301L mutant tau induce elongation of P301L tau but not wild-type tau. J Biol Chem. 282(28):20309–18. 381. Lewis J, McGowan E, Rockwood J, et al. (2000) Neurofibrillary tangles, amyotrophy and progressive motor disturbance in mice expressing mutant (P301L) tau protein. Nat Genet. 25(4):402–5. 382. Sahara N, Lewis J, DeTure M, et al. (2002) Assembly of tau in transgenic animals expressing P301L tau: alteration of phosphorylation and solubility. J Neurochem. 83(6):1498–1508. 383. Lewis J, Dickson DW, Lin WL, et al. (2001) Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau and APP. Science. 293(5534):1487–91. 384. Feany MB. (2000) Studying human neurodegenerative diseases in flies and worms. J Neuropathol Exp Neurol. 59(10):847–56. 385. Wittmann CW, Wszolek MF, Shulman JM, et al. (2001) Tauopathy in Drosophila: neurodegeneration without neurofibrillary tangles. Science. 293(5530):711–4.

c02

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:5

Char Count=

Animal models and procedures for CNS disorders

69

386. Anderson GM, Bennett AJ, Weld KP, et al. (2002) Serotonin in cisternal cerebrospinal fluid of rhesus monkeys: basal levels and effects of sertraline administration. Psychopharmacology (Berl). 161(1):95–9. 387. Anderson GM, Barr CS, Lindell S, et al. (2005) Time course of the effects of the serotoninselective reuptake inhibitor sertraline on central and peripheral serotonin neurochemistry in the Rhesus monkey. Psychopharmacology (Berl). 178(2–3):339–46. 388. Garrick NA, Seppala T, Linnoila M, et al. (1985) Rhesus monkey cerebrospinal fluid amine metabolite changes following treatment with the reversible monoamine oxidase type-A inhibitor cimoxatone. Psychopharmacology (Berl). 86(3):265–9. 389. Garrick NA, Scheinin M, Chang WH, et al. (1984) Differential effects of clorgyline on catecholamine and indoleamine metabolites in the cerebrospinal fluid of Rhesus monkeys. Biochem Pharmacol. 33(9):1423–7. 390. Javors MA, Houston JP, Tekell JL, et al. (2000) Reduction of platelet serotonin content in depressed patients treated with either paroxetine or desipramine. Int J Neuropsychopharmacol. 3(3):229–35. 391. Major LJ, Murphy DL, Lipper S, et al. (1979) Effects of clorgyline and pargyline on deaminated metabolites of norepinephrine, dopamine and serotonin in human cerebrospinal fluid. J Neurochem. 32(1):229–31. 392. Peruzzi P, von Euw D, Lacombe P. (2000) Differentiated cerebrovascular effects of physostigmine and tacrine in cortical areas deafferented from the nucleus basalis magnocellularis suggest involvement of basalocortical projections to microvessels. Ann NY Acad Sci. 903:394–406. 393. Nakamura S, Murayama N, Noshita T, et al. (2006) Cognitive dysfunction induced by sequential injection of amyloid-beta and ibotenate into the bilateral hippocampus; protection by memantine and MK-801. Eur J Pharmacol. 548(1–3):115–22.

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

CHAPTER 3

Preclinical antecedents to early human clinical trials

Introduction Drug candidate selection is a multivariate process that requires the ability to measure the drug candidate in biological matrices, confirm reasonable metabolic stability as well as bioavailability and systemic persistence, determine the mutagenic or cytotoxic potential of the test substance, and estimate dose-limiting and specific target organ toxicity. The ability to formulate the test substance for animal toxicology studies and early human trials, and the creation of a pathway that results in the synthesis of an active pharmaceutical ingredient at reasonable cost, also provide formidable barriers to “first-inman” studies [1]. Therefore, preclinical testing of new chemical or biological entities using in vivo animal models and in vitro systems is one component of a layered process that is essential for inferring potential beneficial and deleterious effects of exogenous compounds on the human body. The results of these evaluations provide information regarding the organ systems that may be affected by the local or systemic administration of these products and permit conservative estimates regarding the safe starting dosage for initial clinical trials. Requirements for animal testing of proposed therapeutics have been in place since the passage of the Federal Food, Drug, and Cosmetic Act in 1938, although they have gone through major revisions and expansions as public policies and healthcare technology advanced. Today, preclinical studies offer an enormous variety of animal and in vitro models that vary appreciably in methods of exposure and endpoints. These paradigms can provide useful data regarding pharmacokinetics and pharmacodynamics, yielding insights regarding toxicity and potential efficacy of novel compounds in humans (see Table 3.1 for examples of important preclinical data). Considerable experience, however, is mandated in the selection and sequence of a preclinical testing procedure in order to facilitate a discovery process that is both efficient and predictive. Extrapolation of results of preclinical data into human trials is considered one of the more challenging interfaces in the discovery-development process, with both falsely positive and falsely negative conclusions commonly encountered. While it is not possible to account for every variable and Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

70

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

71

Table 3.1 Examples of important preclinical data Pharmacology

In vitro receptor-binding profile Activity in in vivo animal models

Pharmacokinetics (PK)

Bioavailability after oral dosage administered Linear vs nonlinear PK Differential exposure (AUC) in male vs female animals Metabolism in various species (including main route of excretion) In vitro identification of p450 Isoenzymes in human liver microsomes

Safety pharmacology

Cardiovascular and respiratory studies in dogs; renal and GI effects in rats CNS effects on arousal, motor coordination, and behavior in animals

Toxicology

Mutations in Ames test or in cultured human lymphocytes Fetal malformations in rodents Organ effects in multiple-dose toxicity studies Definition of NOAEL dose

difference in pharmacokinetics and pharmacodynamic properties between animal and human subjects, similarities in responses are acknowledged with limitations, and preclinical data can be extremely useful in guiding the planning and expected outcomes of a clinical trial. This chapter provides an overview of preclinical testing paradigms for central nervous system (CNS) drugs, including the most advanced model systems for a number of major CNS diseases, their advantages and limitations, and a few relevant examples from recent discovery efforts.

An overview of preclinical evaluation schemes in the twenty-first century A candidate agent for clinical evaluation represents the survivor of a process in which in vitro and in vivo testing is completed in order to determine a compound’s acceptability for human exposure. Prototypical phases include target discovery (target identification and validation), drug discovery (identification of a hit, a lead compound, then optimization to a potential drug candidate), and traditional preclinical drug development processes in which biodisposition, safety pharmacology, and toxicological properties are defined. The attrition process leading to selection of a candidate compound is considerable. It varies appreciably by therapeutic target, the complexity of the medicinal chemistry effort driving this process, and the algorithm used to define chemical as well as biological properties considered “optimal”. However, all discovery algorithms are characterized by a tiered decision process, in which in vitro data generally are available before in vivo evaluations. Physicochemical

c03

BLBK249-Cutler

72

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

characteristics strongly influence the selection of an individual compound, and in vivo models with the most direct clinical relevance are generally reserved for the later phase of the discovery process. Supporting activities include a wide array of disciplines that range from medicinal chemistry through bioinformatics. Discovery-development processes in the genomics era have been characterized by Whittaker in 2003 [2]. He emphasizes the concept of “discovery in development.” A discovery effort that extends into the development phase is a tacit recognition that biological properties of clinical importance are commonly not discerned until first-in-man trials have been completed, and that clinical information can inform the discovery initiative with an impact exceeding that obtained exclusively from animal data. Indeed, innovative trial design methodology increasingly is facilitated by the existence of regulatory guidance both in the US and Europe that will permit compound evaluation in humans, with certain restrictions, following submission of a comparatively limited portfolio of preclinical studies [3]. For example, limited “exploratory investigational new drug (IND) investigations” can be initiated with less or different preclinical supporting information than would be required for a traditional IND application. Such proposed clinical trials present fewer potential risks to subjects than traditional trials that specifically look for dose-limiting toxicities. An exploratory IND program permits an understanding of a compound’s mechanism of action relevant to a proposed indication and can enable the selection of a lead candidate from a group of novel chemical entities that modulate a proposed therapeutic target. In addition, a limited clinical investigation can adequately characterize the biodisposition or pharmacokinetic profile of a candidate drug, obviating in part the need for extensive preclinical evaluation in multiple species. Should a compound prove “interesting” in a limited clinical data set obtained through innovative trial designs, then safety and toxicological studies in animals that enable large-scale clinical trial testing would be completed. This strategy might be appropriately utilized by those pharmaceutical companies with limited resources and an “outlicensing” business strategy, or by an organization that has a very robust discovery pipeline in which selection of a compound from a series of similar candidates for advancement into clinical testing cannot be based solely on animal information. Because of the importance of preclinical evaluation paradigms within the traditional IND process, the sections that follow elaborate on basic tenants and well-established concepts in safety and toxicological disciplines.

Toxicity testing using in vitro systems and animals Toxicology is the study of the adverse effects of chemicals on living organisms. Toxicity testing for either investigational or classic regulatory purposes is an essential discipline that any new drug candidate needs to address during the

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

73

preclinical evaluation process. Indeed, because toxicological studies may exaggerate pharmacological properties of a candidate compound through either species selection or aggressive dosing strategies, they provide an important component of the portfolio that influences subsequent clinical testing—the initial dosage level, the maximal dosage level, the duration of exposure, and potential target organ sensitivity. They also provide inferences regarding dose–response relationships through toxicokinetic data. In vitro and animal studies are used to provide acute, subacute, chronic, reproductive, developmental, and administration-specific toxicity of any new compound intended for human use. The sequence of these investigations and their durations are critically important aspects of the discovery-development process. In addition, recent advances, including the development of new biomarkers and the burgeoning field of genotoxicity (examining genetic damage on a chromosomal level) have opened the door for an efficient screening process for a larger number of candidate agents. This can be accomplished using only limited in vivo data, with studies that are either in vitro or ex vivo. The guidelines for toxicity testing that would be generally required for those compounds eventually exposed to humans have become increasingly standardized with the creation of the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) in 1990, which brought together regulatory authorities in Europe, Japan, and the US [4], as well as the guidelines for the testing of chemicals set by the global Organization for Economic Cooperation and Development (OECD) [5, 6]. This section provides a summary of standard toxicology designs used in preclinical studies and discusses the recent advances in technology, protocols, and regulations affecting this field.

Dose–response curves The concept of “dose–response” is based on receptor pharmacology [7]. The dose–response curve is a depiction of the observed effect of a drug as function of its concentration in the receptor compartment (see Figure 3.1 for sample dose–response curves) [8]. The dose response reaches its maximal value when the drug occupies all the receptor sites (Figure 3.1a). Most dose–response curves are constructed with the logarithm of concentrations in order to depict the full range of the relationship between dose and effect. The concentration that produces 50% of the maximal response quantifies drug activity and is designed as EC50 (effective concentration for 50% response). Many nonlinear dose–response curves level off with increasing dose (Figure 3.1c). Some drugs show U-shaped relationships; they display hormesis (Figure 3.1d) [9]. Among drugs displaying this property are dopaminergic and serotonergic agonists. Every new chemical or biological entity submitted for clinical evaluation may have its own unique dose–response curve that is species, dose, and target organ dependent. An appreciation of this relationship permits an interpretation of toxicological data that will influence subsequent experimental designs.

BLBK249-Cutler

74

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

(b)

Dose

Dose

(a)

Response

Response

(c)

(d) Asymptote

Dose

Threshold

Dose

c03

Response

Inverted U shaped curve

Response

Figure 3.1 Sample dose–response curves and their properties. (a) linear dose–response curve; (b) sigmoidal dose–response curve; (c) threshold asymptote; and (d) non-monotonic dose–response curve. [Note: Figures 3.1a—c are monotonic does–response curve.]

Toxicity is a reflection of exposure, and dosage levels producing limiting toxicity should be larger than those associated with putative efficacy (“the therapeutic index”). The antipsychotic aripiprazole can show an inverted U-shaped dose–response curve for its antipsychotic effects, consistent with partial agonist activity at the D2 receptors. However, these well-established observations do not constrain the dose range evaluated in toxicological programs. The endpoint of toxicological studies may be entirely “off target,” and therefore the putative mechanism of action suggesting efficacy has less impact on the selection of doses, the design of the study, the organ-specific effects examined, and the method of interpreting the data. The role of toxicological investigations represents a conceptually fundamental departure from efficacy-based considerations. It is more concerned with probing the limitations of a compound’s biological effects, commonly exaggerating pharmacological properties already known, and just as commonly detecting biological effects that are unexpected and dose limiting. Indeed, the international harmonization effort that has occurred for toxicological initiatives in part had its genesis in profound differences in the objectives that had been employed across the pharmaceutical industry in these studies. Methodological differences frequently suggested safety margins that could not be confirmed subsequently by clinical data. The international codification of this process has helped to establish the limitations as well as the use of animal studies in the prediction of human toxicity, describing the organ systems where concordance between animals

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

75

and humans has been highest (hematological, gastrointestinal, and cardiovascular) or inadequate (e.g., cutaneous). In addition, it has helped to identify the species or combination of species and the durations of exposure that would permit reliable inferences [10].

IND-enabling toxicology programs A standard preclinical toxicity program that provides a required antecedent to a new drug application permitting human dosing will usually include acute, subchronic, and chronic toxicity tests (see below for detailed explanations). Often short-term repeated dose studies are also included. Contingent upon the phase of clinical testing, genotoxicity, reproductive toxicity, and carcinogenicity trials would either be completed or initiated. The goals of toxicity studies include characterizing adverse effects in regard to target organs, dose dependence, exposure routes, and potential reversibility [11]. Studies are designed to estimate the margin of safety between a potentially clinically efficacious dose and a toxic dose, and to permit the selection of a safe starting dose for humans [12]. By definition, “toxicity” must be identified with the organ system most likely affected by the biological properties of a compound. A number of useful measurements can be obtained from preclinical studies to guide the setup of human clinical trials (see Table 3.2). For example, the US Food and Drug Administration (FDA) and the pharmaceutical industry have jointly agreed to a scaling algorithm to estimate the maximum recommended starting dosage (MRSD) for humans in initial clinical trials, based on the “No Observed Adverse Effect Level” (NOAEL) obtained in various species based on toxicological endpoints in preclinical studies [13]. Obtaining an agreement on the NOAEL provides a critical threshold for subsequent clinical development (see the section “Dose scaling from animal models to humans”). Studies typically use two species (one non-rodent; both unanesthetized), which are selected to give the best correlate to human responses based on the compound’s (a chemical being evaluated in preclinical studies may be “drug

Table 3.2 Some measurements derived from preclinical studies useful for planning human clinical trials NOEL

No observed effect level

NOAEL

No observed adverse effect level

NTEL

No toxic effect level

MTD

Maximum tolerated dose; the dose that produces the minimum toxic effect

HTD

Highest tested dose

ADME

Absorption, delivery, metabolism, and excretion information

ED50

Median effective dose; the dose at which half the animals experience an effect of the drug as predefined by a given model

LD50

Median lethal dose; the dose at which half the animals die

c03

BLBK249-Cutler

76

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

like,” but is not considered a true “drug” until it is being used for therapeutic purposes, and will therefore be referred to as a “compound” in this book) putative mechanism and route of administration. The use of a rodent and nonrodent species permits an interpretation of the result in the context of a very large database across a number of compounds that have had human exposure. This is important, as a number of potentially deleterious observations can be observed that are either not compound related, or not clinically relevant due to unusual species sensitivity. Drug dosage and the number of animals used are scaled to provide a quantifiable basis for risk assessment, and drug administration occurs via the same route as clinically anticipated for humans. This last requirement reflects the established observation that the route of administration directly affects the rate and extent of exposure and can change the conclusions regarding local or systemic toxicity. Study measurements typically include body weight, clinical signs, clinical chemistries, hematology, and histopathology, as well as an evaluation of which organs and systems are the targets of toxicity. The duration of chronic toxicity studies must be the same or greater than the duration of corresponding clinical trials for humans, and this varies depending on the stage of clinical trials. The major classes of toxicity tests will be discussed in more detail in the following sections.

Acute toxicity studies Acute toxicity is the toxicity produced when a drug is administered in one or more doses in a period of 24 hours or less. Acute toxicity studies in animals are useful for identifying target organs of toxicity, starting doses for limited duration phase 1 clinical trials, dose ranging for chronic and subchronic toxicity tests, and information on effects that can be observed during acute overdosing. Testing is done to identify doses causing no observed effects (NOEL), no observed adverse effects (NOAEL), or causing life-threatening toxicity (e.g., LD50 ). Testing is recommended via the same route as clinically anticipated for humans as well as intravenously (when feasible). Studies using rodents employ three to five animals per sex per dose, and fewer animals may be used for non-rodent species (e.g., dogs). Animals are carefully observed for 14 days after drug administration, recording all clinical signs, mortalities, morbidity, and the onset, duration, and reversibility of any observed problems in tolerance. Gross necropsy is conducted to determine cellular-level damage in various organ systems, and pharmacokinetic information may also be obtained when necessary (e.g., toxicokinetics) [14]. The reversibility of testsubstance–related cellular-level damage in organ systems is a key component of many toxicology studies, and a cohort of animals can be discontinued from dosing and followed for a period of time before sacrifice and necropsy in order to examine that effect. One measurement that can be derived from acute toxicity tests is the median lethal dose (LD50 ); the dose at which half the animals of a population receiving a drug die. The LD50 is usually expressed as the mass of a substance

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

77

administered per unit mass of the test subject (i.e., gram of a substance/ kilogram of body weight). This allows for comparisons across species, although this estimate does not always hold true since many factors besides mass contribute to an animal’s response to a substance. The LD50 usually varies, depending on the route of drug administration; therefore, the derived value often contains a reference to the route of administration (e.g.: LD50 i.v. = 730 mg/kg). This measurement can assist in estimating the dose range to be used in subsequent toxicology studies and, depending on therapeutic area, may have implications for the dose range employed in clinical trials. The method for obtaining this estimate has gone through major revisions due to limitations in the extrapolation that can be made to humans using these data, the wide variability in results across species and even within species, and the inherent inefficiencies involved in exposing a sufficiently large sample of animals to permit this estimate [15]. The standard LD50 test for determining acute oral toxicity was phased out by the OECD in 2001 and replaced with other tests such as the fixed dose procedure (FDP), the acute toxic class method, and the up-and-down procedure, all of which allow for reliable estimates of the LD50 without sacrificing nearly as many test animals [16]. The FDP uses about 10 to 20 animals to find the dose producing toxicity signs but not death, and from there predicts the lethal dose [17]. The acute toxic class method exposes three animals of a single sex to fixed dosages, which are raised or lowered until a mortality difference is established and a LD50 range can be predicted [18]. The up-and-down procedure tests one animal at a time, beginning with a dosage below the LD50 (as best estimated), and then raising and lowering the dosage until a confidence interval for the LD50 is established [19].

Short-term, subchronic, and chronic dose repeated toxicity studies Short-term, subchronic, and chronic repeated-dose toxicity tests examine the toxicity of a compound over a longer period than acute toxicity tests, using a lower, sublethal dose that still produces toxic effects. The typical short-term repeated-dose toxicity test generally involves three study endpoints (e.g.; 14, 28, and 90 days in rodents) [20]. Subchronic toxicity tests usually refer to a dosing period consisting of 10% of a test animal’s lifespan or less, although sometimes dosing periods between 24 hours and 10% of a test animal’s lifespan are referred to as “subacute” or short-term repeated-dose toxicity studies. Chronic toxicity tests require dosing periods of at least 12 months and can be combined with carcinogenicity tests or reproductive/developmental tests [21]. However, certain segments of reproductive/developmental testing results are required early in clinical development, and prior to the initiation of chronic toxicology studies. Chronic, subchronic, and repeated-dose toxicity studies are used to detect general toxicity, including neurological, physiological, biochemical,

c03

BLBK249-Cutler

78

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

hematological, and pathological effects, as they may be influenced by repetitive dosing at levels that exceed anticipated clinical exposure for a prolonged period of time. They are also used to establish the NOEL, which can predict an acceptable initial dose range, or Reference Dose (RfD) for humans [22]. Testing is usually performed using both sexes of two mammalian species, one rodent and one nonrodent, with enough animals to provide a thorough biological evaluation at the end of the study. Three doses are typically used—the highest doses should elicit signs of toxicity without causing lethality, and the lowest dose should not produce any toxicity. Traditional routes of administration for most CNS agents are orally, subcutaneously, and intravenously. Other routes of exposure are common (dermal, inhalation, and intrathecal), specific to the formulation in development. Currently, non-animal supplements for chronic toxicity tests are being developed. These include human- and animal-perfused organs, tissue slices, suspended cells, primary cultured and genetically engineered cell lines, threedimensional cell cultures, and computational systems, among others [23–27]. It remains to be seen how well these disparate, in silico (performed on a computer or via a computer simulation), and in vitro systems can be organized into a predictive, combined testing system that rivals the robustness and utility of toxicity tests with live animals. Rather than supplanting traditional whole animal paradigms that must be completed prior to dosing in humans, it is likely that their maximum utility will occur during lead optimization in discovery programs in order to eliminate compounds prior to the allocation of resources that would be required for traditional safety and toxicology studies.

Genotoxicity studies Genotoxicity tests are in vitro and in vivo methods for detecting compounds that induce genetic damage directly or indirectly. This damage includes genetic mutations, larger-scale chromosomal damage, recombination, and numerical chromosome changes; all of which can lead to cancer or heritable defects. A comprehensive program acknowledges the recognition that tumor promoters and co-carcinogens exist that do not themselves directly interact with genetic material, and that lesions are possible in both mitochondrial as well as the nuclear genome. The extreme sensitivity of the mitochondrial genome to an environmental insult, for example, is best illustrated by trial results with fialuridine (FIAU) in 1993 in which 7 of 15 patients experienced severe drug-related toxicity as a result of mitochondrial injury and respiratory inhibition [28]. Therefore, comprehensive genotoxicity testing is required for any new pharmaceutical and is usually carried out as a battery of complementary tests, each designed to measure a different aspect of genetic damage [29]. Such standardized genotoxicity tests are now a mandatory regulatory barrier for new chemical entities prior to first-in-man studies. The general components of the standard three-test battery are summarized below; in vitro studies are completed early in the preclinical development process and commonly

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

79

before any investment occurs in safety pharmacology, biodisposition, or toxicological studies, as previously described. 1. Bacterial Reverse Mutation (AMES) Test: Designed to test the ability of a compound to produce genetic mutations in bacterial systems. The AMES test [30] establishes a direct relationship between mutagenic activity in a simple bacterial system and the carcinogenicity potential in humans. This assay measures genetic damage at the single base level in DNA by using five or more tester strains of bacteria, and is sensitive to a wide range of mutagenic and carcinogenic chemicals [31–33].The Escherichia coli reverse mutation assay is also commonly used for these measurements [34]. 2. Mammalian Cell System Test: Designed to evaluate DNA damage not adequately measured in bacterial systems. These include tests for gross chromosomal damage, genetic mutations, and clastogenic effects (chromosome breakage). Commonly used tests are the in vitro mammalian chromosome aberration test, the in vitro mammalian cell gene mutation test, and the in vitro micronucleus test [35–37]. 3. In vivo Test for Chromosomal Damage: Uses an animal model to create a complex genotoxic profile for a drug, taking into account the effects of absorption, distribution, metabolism, and excretion. Examples include the tests for chromosomal damage in rodent hematopoietic bone marrow cells and rodent peripheral blood erythrocytes [29]. In addition, there are a large number of in vivo tests which directly assess the heritable genotoxicity of a drug, including predictive genotoxicity studies with somatic cells, germ cell mutagenicity tests, and tests that assess genetic damage to an animal’s progeny [38–44]. New in vivo genotoxicity tests are currently being developed and validated for use in regulatory decisions that are extremely versatile and sensitive to DNA damage. These include the in vivo Comet assay [45], which can detect small levels of DNA damage and repair in any eukaryotic cell type, and the in vivo transgenic mutation assay [46], which can evaluate the mutagenic potential of a drug in any target organ in vivo. A recent, retrospective study of the current FDA regulatory battery of genotoxicity tests used to predict carcinogenicity concluded that two tests had good predictive value (gene mutation in Salmonella and in vivo micronucleus) and two tests had poor predictive value (mouse lymphoma gene mutation and in vitro chromosome aberrations) [47].

Carcinogenicity studies A carcinogen is a substance which induces cancer or increases its incidence. The FDA and both the Centers for Drug/Biologics Evaluation and Research (CDER and CBER) recommend carcinogenicity testing for all new drugs prior to commercialization [48]. It is vitally important to determine whether a new drug will cause cancer in humans during preclinical studies, yet this determination is often confounded by many variables, including the complexity of carcinogenesis and the inherent differences in tumor kinetics between animal

c03

BLBK249-Cutler

80

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

and human models. A typical carcinogenicity study involves subjecting three groups of animals to a low, medium, and high dose of a test compound (one group per dose) for a majority of their normal lifespan, observing all clinical and pathological signs, then performing a necropsy and microscopic examination for tumors and lesions at the end of the study [49]. As the investment in time and resources for this paradigm is considerable, and not practically subject to replication, these are one of the more carefully considered preclinical programs with the initiation critically timed to the progress of clinical development. The maximally tolerated dose (MTD) in animals was traditionally used to calculate the highest dose selection for carcinogenicity studies, yet this was problematic because exposures in rodents greatly in excess of the intended human dosage often had little relevance for calculating the human risk. For example, nongenotoxic substances in large enough doses could potentially induce tumors in rodents by altering their normal physiology. A famous example of this was the controversy with the ubiquitous artificial sweetener saccharin, which sweetens everything from soft drinks to toothpaste. In two generation studies carried out in the mid-1980s, second-generation laboratory rats developed bladder cancers significantly more than controls when their diet was composed of 4% saccharin [50]. This caused the FDA to propose a ban on saccharin and Canada to enforce a ban. The carcinogenic risk was later deemed negligible when the National Institute for Environmental Health Sciences determined that the cancer-inducing mechanism in rats does not apply to humans. The high osmolarity of rodent urine enhances the precipitation of calcium phosphate-containing crystals that are cytotoxic to the bladder epithelium, leading to hyperplasia and tumors. This hypersensitivity was supported by the ability of ascorbic acid (vitamin C) to cause bladder cancer in rats when given in similar doses as saccharin, yet the same administration of vitamin C has no carcinogenic effects in humans [51]. The ICH subsequently proposed several criteria for selecting the appropriate dose for carcinogenicity studies. These included dosing criteria based on toxicology, pharmacokinetic or pharmacodynamic endpoints, saturation and absorption characteristics, or the maximum feasible dose, among other characteristics [52]. Carcinogenicity studies are usually designed after genotoxicity, pharmacodynamic, and some toxicity studies, and take the results of these studies, as well as patient and dosage information, into account. They can also be combined with chronic toxicity studies to reduce the cost and number of animals used [53]. Dose selection is generally calculated from a 90-day preliminary study using the route and method of drug administration to be used in the carcinogenicity study. Carcinogenicity studies usually consist of two tests: a long-term rodent study (usually for the duration of the animal’s lifespan, or around 2 years) combined with a short- or medium-term rodent study focusing on models of tumor development, or two long-term studies using different rodent species. The species should include both sexes and be selected based on

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

81

pharmacological, toxicological, metabolic, and administration parameters that best approximates the drug’s properties in humans. The standard is 50–100 animals per sex per dose group [54]. The rat is often used for these studies because it has been shown to be more sensitive to carcinogenic compounds than mice [55]. The end of the study requires a full pathological analysis of the animal’s tissues and organs. Follow-up mechanistic studies may be justified to determine the risk that tumors in animal models pose to humans. Carcinogenicity tests may not be required for compounds administered infrequently or for a very short duration of time, unless there is a cause for concern. In addition, highly genotoxic compounds are also assumed to be carcinogens, obviating a need for further testing. Physiological changes that could influence the interpretation of the study, such as a greater than 10% loss in body weight, target organ toxicity, and significant changes in clinical pathology should be avoided when possible by selecting doses that cause minimum toxic effects. A number of transgenic mouse models are being considered for use in carcinogenicity studies, but none of these models is currently sufficient as a stand-alone assay or serves as a major improvement over existing genotoxicity tests [56]. In addition, similar to developments in chronic toxicology, a number of in silico and in vitro methods are being developed to complement carcinogenicity studies, including cell-based assays and computational prediction models [57–60]. These methods are significantly faster and less expensive than in vivo systems, but they are not robust enough to replace animal models at this time. A retrospective study reviewed standard genetic toxicity tests, reproductive and developmental toxicity tests, and rodent carcinogenicity bioassays to determine which were the most accurate predictors of carcinogenicity. Results revealed that carcinogenicity was well correlated with tests for gene mutation, in vivo clastogenicity, unscheduled DNA synthesis assay, and reproductive toxicity [47]. While genotoxicity studies are highly suggestive of the carcinogenic potential of a drug, they do not obviate the requirement for long-term carcinogenicity studies in animals.

Reproduction and developmental toxicity studies Reproductive toxicity is defined as the adverse effects of a compound on reproductive ability or capacity, including fertility, parturition, and lactation. Developmental toxicity is defined as adverse effects induced during pregnancy or as a result of parental exposure which can appear at any time in the offspring’s lifespan. These include mortality, dysmorphogenesis (structural alterations), alterations to growth, and functional toxicities [61, 62]. The OECD has developed guidelines for a number of animal-based reproductive and developmental toxicity tests. The standard tests for reproductive toxicity are the one-generation reproductive toxicity study, the two-generation reproductive toxicity study, the reproduction/developmental toxicity screening test, and the prenatal developmental toxicity study. These

c03

BLBK249-Cutler

82

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

standard regulatory studies in the pharmaceutical industry typically proceed in three segments—”segment 1” (male and female preconception, fertility, and preimplantation studies in rodents); “segment 2” (prenatal development/embryotoxicity and organogenesis studies in a rodent and nonrodent species); and “segment 3” (peri- and postnatal development studies in rodents) [63]. This segment terminology has been dropped from the most recent ICH guidelines for determining reproductive toxicity [64], but continues in common use since the most popular study designs still feature this threesegment strategy [65]. The timing for the completion of these trials in relationship to the clinical development program is also codified by ICH regulation. The one-generation reproduction toxicity study examines the effects of a compound on male and female reproductive performance as well as at offspring viability. Reproductive test measurements include gonadal function, estrous cycle, mating behavior, conception, parturition, lactation, and weaning. Progeny test measurements include neonatal morbidity, mortality, behavior, and teratogenesis. The test compound is administered to male and female animals in three doses. The species used is typically mouse or rat, and enough animals are used to produce 20 pregnant females. Males should be dosed during growth and at least one spermatogenic cycle before mating. Females are dosed for at least two complete estrous cycles before mating. The test substance is administered during the mating period and subsequently only to females during pregnancy and weaning. The entire test takes approximately 23 weeks in mice from the beginning of the spermatogenetic cycle in adult male mice to sexual maturity of the first generation progeny. Animals and offspring are observed daily for behavioral changes and any signs of toxicity. Gross necropsy and microscopic examination of all target organs is carried out at the end of the study [66]. The two-generation reproduction toxicity study is a more comprehensive version of the one-generation reproduction toxicity study, examining the parental (P) generation for reproductive toxicity effects, their progeny (F1 generation) for growth, developmental, and reproductive toxicity effects, and the second generation of progeny (F2) for growth and developmental toxicity effects. The procedure for compound administration and monitoring of the parental generation is the same as the one-generation reproduction toxicity study, except that the dosing continues with the F1 and the F2 generations. The parental generation begins dosing at between 5 and 9 weeks of age, and the F1 generation begins dosing at weaning. Animals should be mated with a member of their same dose range group (using a different litter for the F1 generation mating). All animals are sacrificed and examined as described above [67]. The reproduction/developmental toxicity screening test is a faster, more limited test for detecting postnatal manifestations of prenatal exposure to a compound. It is used at early stages of toxicological assessment of a compound and can provide initial information on reproductive/developmental toxicity or as a dose-range-finding study. Males are dosed for at least 4 weeks

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

83

(2 weeks prior to mating, during the mating period, and 2 weeks post mating). Females are dosed 2 weeks prior to mating until at least 4 days after delivery (including the day before the scheduled sacrifice). Animals are observed and mated as described in the previous two studies, and are similarly sacrificed and examined at the termination of the study. Three doses of the compound are used with at least 10 animals of each sex [68]. The prenatal developmental toxicity study is designed primarily to examine a compound’s effect on the developing fetus. A test substance is administered in three doses to groups of pregnant animals from implantation (5 days post mating) to sacrifice, which should be as close as possible to the date of delivery. After sacrifice, the uterine contents are extracted and examined and the fetus is examined for soft tissue and skeletal changes [69]. Other in vivo reproductive toxicity tests include the developmental neurotoxicity study, designed to look at the developmental neurological effects of a test substance in utero and early postnatal development [70], and the uterotrophic bioassay, designed to measure how a compound disrupts the estrogen system, the uterus, and the embryo [71]. Numerous in vitro tests are currently being developed that model various components of the reproductive process, but due to this process’s complexity, these tests will remain at most supplementary to animal-based studies, at least for the near term [72–74]. The FDA has made five categories to describe the level of potential risk to human pregnancy based on human and animal studies. The categories are A, B, C, D, and X. Category A represents substances where well-controlled studies in pregnant humans failed to demonstrate any risk to the fetus. Category B represents substances that have not been tested in humans but demonstrate no risk to pregnant animals, or demonstrate a small risk to pregnant animals but no risk to humans. Category C represents substances where there are no controlled studies in pregnant humans but studies in pregnant animals demonstrate a risk, or animal studies have not been conducted. This is the most common category for new pharmaceuticals—even the slightest demonstrated risk in animal studies can warrant this categorization. Category D represents substances where there is a demonstrated developmental risk to humans, yet the benefits of use may outweigh the risks. An example is the angiotensinconverting enzyme (ACE) inhibitors used to treat hypertension, which may pose a risk to a fetus (fetopathy) but whose benefit to the mother (prevention of eclampsia) outweighs the risk [75]. Category X represents substances with demonstrated human developmental toxicity whose risks do not outweigh their benefits. Therefore, this drug is contraindicated for women who are or who may become pregnant.

System and population-specific toxicity studies There are additional specific areas of preclinical testing that have been briefly described in this section due to their importance in the selection process for

c03

BLBK249-Cutler

84

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

new drug candidates, and the potential information they provide regarding clinical experimental design.

Safety pharmacology studies Safety pharmacology studies assess a compound’s effects on vital functions, including cardiovascular, central nervous, and respiratory systems [76]. While general toxicity studies identify target organs and describe toxic responses relying heavily on microscopic pathology, safety pharmacological studies reveal broader harmful effects on critical organ functions and are more relevant for the clinical testing environment as they occur at levels of exposures that can be easily attained in humans [77]. Measurements include cardiovascular, respiratory, and electrocardiogram (ECG) parameters, neuromuscular and autonomic functions, gastric pH, intestinal transit and emptying time, body temperature, and animal activity, among others. Many of these parameters can be measured during toxicity studies, although not with sufficient detail to support subsequent development decisions. Separate safety pharmacology studies are always mandated as a prerequisite to first-in-man dosing, with additional paradigms developed if the compound is suspected of having adverse effects on a particular organ or system, thus necessitating a more focused approach. These include electroretinography studies in dogs for compounds that may affect the visual system, and special assessments to clarify drug effects on electrocardiographic indices such as a cardiac ion channel screen or a Langendorff or Purkenji cell preparation. Such additional studies will often be dictated when a compound belongs to a class of agents for which specific concerns have already been identified in humans and are already included in the labeling of approved drugs (i.e., class labeling). With antipsychotics, for example, the preclinical work for a novel atypical compound should include ophthalmological studies even if the compound is not a phenothiazine derivative. Cardiac electrophysiology Many in vivo and in vitro methods exist for measuring a compound’s effect on cardiac function during preclinical studies, and several methods are often used together to complement one another. Typical measurements include ionic current fluctuations, action potential or ECG recordings, and proarrhythmic effects. Model systems include isolated cardiac preparations, human cardiac myocytes, cardiac cell lines, and live animals [78]. An in vivo QT interval prolongation study in conscious, fully ambulatory beagle dogs and the in vitro hERG (IKr ) potassium-repolarizing channel antagonism study are considered routine tests for detecting arrhythmic-inducing compounds regardless of therapeutic class, and are always completed prior to first introductions in humans [79]. The QT interval is an electrical measure of heart muscle activity, beginning with depolarization of the interventricular septum and ending with the repolarization of the ventricles. The faster the heart beats, the shorter the QT

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

85

interval. The QT interval is a very sensitive measure of cardiac dysfunction, with numerous studies suggesting a link between drug-induced prolonged QT intervals and cardiac arrhythmias [80–84], which could lead to cardiac arrest. The QT interval is currently one of the most scrutinized preclinical outcomes, due to the fact that numerous FDA-approved drugs have been subsequently removed from the market due to reports of them inducing potentially fatal cardiac arrhythmias [85–91]. A mandatory formal QT investigation in humans is required with limited exceptions prior to the initiation of pivotal studies, leading to product registration [92]. Among in vivo studies, the surface ECG has been the gold standard for determining the effects of pharmaceutical compounds on cardiac electrophysiology. Guinea pigs, rabbits, and especially dogs are the most commonly used species for cardiac electrophysiology studies due to their extensive similarities with humans in regard to cardiac structure and function [93, 94]. They can be tested awake and unrestrained, awake and restrained, or anesthetized, depending on the type and sensitivity of the measurement required. During these studies, the heart rate and a standardized six-lead (at DI, DII, DIII, aVF, aVR, and aVL derivations) ECG is recorded before and after dosing in animals of both sexes and over a range of escalating doses [95, 96]. The dog is the most predictive preclinical species with regard to human electrocardiography [97]. This is due to both the large historical canine database that facilitates interpretation of results and the similarities between canine ventricular tissue and the human heart [98]. The dog’s prediction of human cardiovascular toxicities has been found to be among the highest of animal species studied, and the beagle dog is known to be the most predictive model of the species [98, 99]. Modern in vitro studies such as unicellular cardiac preparations are detailed enough to allow the identification of the specific ion channels likely to be affected by a compound [100, 101] and can also model complicated risk factor scenarios [102]. Compounds that increase the QT interval tend to block highly specific ionic repolarizing currents, typically delayed rectifying potassium currents [103, 104]. This information is commonly used early in the discovery process to eliminate individual compounds or series of compounds from further consideration, or to prompt additional investigations. The standard assay for modeling drug-induced ionic current inhibition is the IKr repolarizing channel antagonism study. This study uses cell lines transfected with the human IKr channel to detect inhibition of repolarizing currents by a compound of interest, and can identify most compounds capable of inducing arrhythmias in humans [105].

Immunotoxicity It is important to evaluate the adverse effects that novel compounds have on the immune system, and measurements of these effects are typically incorporated into standard preclinical testing procedures. Suppression of the immune system by a compound can lead to decreased host resistance to infectious agents or tumors, while enhancement of the immune system can produce

c03

BLBK249-Cutler

86

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

autoimmune disease, hypersensitivity, or an immune response to the compound itself. Compounds can interfere with the immune system because they are designed to do so (i.e., immunosuppressive drugs designed to prevent organ transplant rejection), or because they damage and inhibit immune cells, or interact with receptors involved with the immune system as a side effect of their normal function (i.e., the antiproliferative function of some cancer drugs also inhibits bone-marrow-derived immune cells). Standard toxicity studies should include leukocyte counts, globulin levels, gross pathology of lymphoid organs, histopathological examinations of the spleen and thymus, and stress-related effects (i.e., cortisol level changes). Additional, immune-specific toxicity studies may be required, depending on the results of these standard toxicity studies. Observations that usually prompt additional inquiries include hematological changes (i.e., leukocytopenia or lymphocytopenia), changes in immune system organ weight or histology, changes in serum globulin levels, increased infection rate (in those studies capable of detecting those events), or greater tumor incidence. Additional immunotoxicity studies are also justified if the pharmacological, structural, or metabolic properties of a compound suggest that it affects the immune system, if the intended patient population is immunocompromised, or if ongoing clinical trials in humans show signs of immunotoxicity that was not previously detected [106]. Immunotoxicity studies can measure changes in the function of specific immune components in animal models being treated with a compound. These measurements include T-cell-dependent antibody responses (TDAR), natural killer cell activity, macrophage/neutrophil function, and cell-mediate immunity. Host resistance studies look at the ability of an animal to fight off an infection (bacterial, fungal, viral, or parasitic) or tumor cells, assessing overall immune function. Immunophenotyping studies identify changes in the amount and variety of white blood cell subsets [107]. These studies allow for a detailed analysis of the nature and extent of any observed immunotoxicity.

Neurotoxicity Neurotoxicity refers to an adverse change in the normal structure or function of the nervous system caused by a foreign compound. Neurotoxicity tests are typically a component of general toxicity tests because the nervous system is one of the main target organs for analysis in these studies. Additional neurotoxicity studies are justified in order to confirm or gather more information about any neurotoxicity observed during general toxicity testing, or if the structure or function of a compound indicates that it may have a neurotoxic effect. In vivo neurotoxicity studies can be carried out using either a single dose or repeat doses of a test compound, with repeat dosing regimens ranging from 28 days to a year or longer. Behavioral and neurological abnormalities are observed during the study, and sections of the brain, spinal cord, and peripheral nerves are prepared and examined at the study’s conclusion. Typically rats are

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

87

used, with dosing beginning as soon as possible after weaning (before 6 weeks of age) via the same administration route as intended for humans. At least 20 animals should be used (10 of each sex) for each of three dosing groups; high, medium, and low [108]. Physical observations, bodily functions, behavioral changes, and coordination of movement should be noted during the study. Functional tests should also be conducted, including sensory activity assessments, limb grip strength, motor activity, and cognitive functions, among others [109–113]. A detailed list of animal-based tests measuring motor, sensory, cognitive, and functional changes, as well as cellular and electrical properties of neurons, can be found in the OECD Guidance Document for Neurotoxicity Testing [114]. Tissues from at least five animals/sex/group should be examined for neuropathological findings, and the remaining animals can be used for any supplemental neurobehavioral, neuropathological, neurochemical, or electrophysiological studies that would aid in the characterization of any observed neurotoxicity [115–122]. Delayed neurotoxicity is a characteristic observed with organophosphorus esters [123, 124], and the OECD has subsequently written guidelines to test a compound’s ability to induce delayed neurotoxicity after acute and chronic exposure [125, 126]. The acute delayed neurotoxicity test exposes a group of domestic hens to a single dose of a suspected compound and observes them for behavioral abnormalities, ataxia, and paralysis for 21 days. The chronic delayed neurotoxicity test exposes a group of domestic hens to a daily dose of a suspected compound for 28 days (90 days is preferred for slow-acting compounds). Biochemical measurements are taken from members of each group, and at the end of both studies the remaining hens are sacrificed and examined for neuropathological findings, including neuropathy target esterase (NTE) activity (a neural membrane protein made toxic by organophosphorus esters) [127–133] and acetylcholinesterase (AChE) inhibition (an enzyme that produces neurotoxicity when inhibited) [134, 135]. The dose level should be set as high as possible without being lethal, based on observations from a preliminary dosing study. The results of the acute delayed neurotoxicity study will determine whether a chronic delayed neurotoxicity study is necessary. Developmental neurotoxicity studies can be incorporated into reproductive toxicity, adult neurotoxicity, or developmental toxicity studies, or can be conducted separately. The test compound is administered to animals (typically rats) in three doses during gestation and lactation, and offspring are randomly selected for evaluation at different points in their lifetime [70]. Behavioral and histopathological observations should include gross neurologic, physical, and behavioral abnormalities [64, 136], behavioral ontogeny [137–139], motor activity [113, 140, 141], motor and sensory functions [142–145], learning and memory changes [146–154], brain weight changes [155], and any signs of neuropathology [156–158]. Many in vitro procedures are being developed for neurotoxicity testing to complement in vivo studies and to provide more accurate results when these studies are inconclusive [26]. The goal of in vitro neurotoxicity tests is not to

c03

BLBK249-Cutler

88

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

model the complexity of the entire nervous system (which is currently not possible), but to identify specific mechanisms and targets of neurotoxicity using cell lines, cultures, and brain slices [159]. The most promising approach is a two-tiered testing battery that first distinguishes between neurotoxic and cytotoxic chemicals, and then investigates the cellular mechanisms of toxicity [24]. While in vitro neurotoxicity tests continue to grow more useful, none of them have currently been accepted by the OECD as a replacement for live animal tests.

Seizures The identification of convulsant potential is another important preclinical safety goal for CNS compounds, as seizures can be life-threatening. While many approved drugs (including antidepressants and antipsychotics) can lower the seizure threshold, so that effective drugs may not be completely devoid of this potential, it is important to try and develop effective drugs that have minimal or no convulsant activity. No single preclinical test is completely reliable for predicting convulsant activity in humans. Therefore, a battery of tests is often conducted, including a comparison of a compound’s receptor and ion channel profile with known proconvulsants, followed by in vitro assays such as electrophysiological studies in rat brain slices. A usually sensitive in vivo test assesses whether a compound can modify the seizure threshold in rodents by administering it prior to infusion of the CNS stimulant pentylenetetrazole [160]. Additional studies may then be conducted in rats or mice to assess for electroencephalogram (EEG) or behavioral changes, including the Irwin test which assesses locomotor function, muscle strength, and coordination [161]. Later toxicological studies in two species will seek to identify seizure activity, which may often occur with CNS compounds at very high doses.

Dose scaling from animal models to humans Extrapolating a safe yet effective starting dose of a novel compound from an animal model to a human is a difficult task, yet decades of experience with this critical step have produced standardized and predictive guidelines. Prior to 1997, the LD50 and the LD10 (the doses of a compound which result in 50% and 10% lethality of an animal group, respectively) were the basis for calculating the starting dose in humans. After 1997, the ICH and the FDA adopted the no observable adverse effect level (NOAEL; the highest dose that produces no observed toxicity in the animal species most relevant to humans or most sensitive to the compound) as the basis for calculating the starting human dose [162]. In 2002, the FDA (working through the Center for Drug Evaluation and Research [CDER] and the pharmaceutical industry) released a guidance document that provided an algorithm explaining how to calculate the human maximum recommended starting dose (MRSD) from the NOAEL,

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

Determine most appropriate animal model

Calculate no observed adverse effect level (NOAEL) in toxicity tests

Divide NOAEL by conversion factor (BSA-CF) to get human equivalent dose (HED)

89

Divide HED by safety factor to get maximum recommended starting dose (MRSD)

Sample Calculation 1.3 mg/kg/day

0.21 mg/kg 1.3 mg/kg = 0.021 mg/kg = 0.21 mg/kg 10 6.2 Using the animal-specific body surface area conversion factor (BSA-CF), a dose of 1.3 mg/kg in a rat is equivalent to 0.21 mg/kg in man1.

1

Surface area increases less than mass from rats to humans. Because the dose is calculated based on surface area, the smaller surface area/mass ratio in humans requires a lower dose/kg to be equivalent to the dose in rats, which is factored in by the BSA-CF.

Figure 3.2 Conversion of animal dose to human dose based on body surface area.

and updated this algorithm in 2005 [13]. This recommendation addresses initial human dosing, but not the limitations for a maximal clinical dose that ultimately might be evaluated. In general, the FDA’s algorithm requires selecting the appropriate animal species, determining the NOAEL in that species, converting the NOAEL to a human equivalent dose (HED), and then applying a safety factor (see Figure 3.2 for a sample calculation). While other data may factor into the algorithm, such as pharmacological results, only the NOAEL should be used directly in the equation. The NOAEL is calculated based on overt toxicity, surrogate markers of toxicity, and exaggerated pharmacodynamic effects. Conversion to a HED should generally be done by normalizing doses to body surface area (unless other normalizations are deemed appropriate, e.g., when NOAELs occur at similar mg/kg doses across species, body weight can be used for normalization), based on historical evidence showing the accuracy of this approach at predicting toxicity endpoints across species [163, 164]. Normalization is accomplished by dividing the NOAEL in each of the animal species studied by the appropriate body surface area conversion factor (BSA-CF). The conversion factor is a unitless number that converts mg/kg dose for each animal species to the mg/kg dose in humans, using surface area (mg/m2 ) as the basis for conversion. The resulting dose is the HED. The FDA has issued a guidance document which lists a number of common BSA-CFs [13]. The species that generates the lowest HED is called the most sensitive species, and this species should generally be used unless other factors make them a poor model (i.e., historical differences in pharmacokinetic, pharmacodynamic, or toxicity profiles in comparison to humans). A safety factor of at least 10 should be applied to the HED to decrease the risk of adverse effects in humans. The MRSD is obtained by dividing the HED by the safety factor and then assuming

c03

BLBK249-Cutler

90

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

that a typical weight for a human is 60 kg. There are many additional factors that may raise or lower the MRSD; for example, if the pharmacologically active dose (PAD) is lower than the MRSD, it might be prudent to lower the dose to avoid exaggerated pharmacologic effects.

Therapeutic implications Despite decades of experience with toxicity testing and dose scaling in animals and in vitro systems, it remains incredibly difficult to account for all the possible risks a novel compound can bring to clinical trials. A dramatic, recent example of these shortcomings involves the monoclonal antibody TGN1412, also known as CD28-SuperMAB, produced by the former German company TeGenero Immuno Therapeutics in 2006. CD28-SuperMAB was created as an immunomodulatory drug originally intended for the treatment of B-cell chronic lymphocytic leukemia and rheumatoid arthritis. It was believed to work by preferentially activating regulatory T cells. Toxicity testing was carried out on animal models, including nonhuman primates, and the drug appeared to be well tolerated. A minor side effect noted was a transient increase in the size of the lymph nodes in primates, but no severe adverse effects in any species were reported. First-in-man clinical trials were conducted at 1/500th the MRSD, as determined in animal models. Six healthy volunteers were injected with the compound on March 6, 2006. All six of the volunteers began to feel ill a few minutes after injection, experiencing nausea, fever, and vomiting. By the end of the day, all subjects were in intensive care with multiple organ failure. One of them fell into a coma which lasted for weeks. They were reported to have experienced a cytokine release syndrome and indiscriminate cell destruction, triggered by the nonspecific activation of natural killer T cells [165]. One theory for this extreme, unanticipated reaction is that a component of the antibody tail was specifically designed for humans; therefore, the reaction observed in animals was far less potent and falsely reassuring [166]. While the study met the appropriate safety standards, it has since had farreaching consequences for the design of preclinical safety trials. Today, a more detailed analysis of minor adverse events is required and a greater reliance on in vitro toxicity studies using human cell lines [167–169]. The areas of preclinical toxicity testing and dose scaling are continuously evolving to minimize risks while maximizing the utility of subsequent clinical trials. As the above example demonstrates, it is still a work in progress.

References 1. Pritchard JF, Jurima-Romet M, Reimer ML, et al. (2003) Making better drugs: decision gates in non-clinical drug development. Nat Rev Drug Discov. 2(7):542–53. 2. Whittaker PA. (2003) What is the relevance of bioinformatics to pharmacology? Trends Pharmacol Sci. 24(8):434–9.

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

91

3. US Department of Health and Human Services, Federal Drug Administration, Center for Drug Evaluation and Research. (2005) Guidance for Industry, Investigators, and Reviewers. Exploratory IND Studies. [document on the Internet]. Available from: http://www.acciumbio.com/Exploratory%20IND%20Draft%20Guidance.pdf. Cited May 7, 2009. 4. Catalano J. (2008) The International Conference on Harmonization (ICH) and Its Relevance to Cell Therapy [document on the Internet]. Available from: http://www.bcgusa.com/regulatory/docs/FDA Presentations Publications/SP9D06.pdf. Cited June 2008. 5. Auletta CS. (2001) Acute, subchronic and chronic toxicology. In: Derelanko MJ, Hollinger MA, editors. Handbook of Toxicology. 2nd ed. CRC Press, Boca Raton, pp 69– 126. 6. Limbird LE. (2005) Cell Surface Receptors: A Short Course on Theory and Methods. 3rd ed. Springer-Verlag, New York. 7. Limbird LE. (2004) The receptor concept: a continuing evolution. Mol Interv. 4(6): 326–36. 8. Brunton LL, Laxo JS, Parker KL. (2005) Goodman & Gilman’s: The Pharmacological Basis of Therapeutics. 11th ed. McGraw-Hill Companies, Columbus, Ohio. 9. Calabrese, EJ, Baldwin LA (2003) Hormesis: the dose–response revolution. Annu Rev Pharmacol Toxicol. 43:175–97. 10. Olson SC, Bockbrader H, Boyd RA, et al. (2000) Impact of population pharmacokineticpharmacodynamic analyses on the drug development process: experience at ParkeDavis. Clin Pharmacokinet. 38:449–59. 11. ICH. (1997) Guidance for Industry—M3 Nonclinical Safety Studies for the Conduct of Human Clinical Trials for Pharmaceuticals [document on the Internet]. Available from: http://www.fda.gov/downloads/RegulatoryInformation/Guidances/ucm129173.pdf. Cited June 19, 2009. 12. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. (2006) Guidance for industry, investigators, and reviewers: exploratory IND studies. [document on the Internet]. Available from: http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm078933.pdf. Cited June 19, 2009. 13. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. (2005) Estimating the maximum safe starting dose in initial clinical trials for therapeutics in adult healthy volunteers. [document on the Internet]. Available from: http://www.fda.gov/downloads/Drugs/ GuidanceComplianceRegulatoryInformation/Guidances/UCM078932.pdf. Cited July 5, 2009. 14. Center for Drug Evaluation and Research, Food and Drug Administration. (1996) Single dose acute toxicity testing for pharmaceuticals. [document on the Internet]. Available from: http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatoryInformation/Guidances/ucm079270.pdf. Cited July 7, 2009. 15. Rowan A. (1983) Shortcomings of LD50-values and acute toxicity testing in animals. Acta Pharmacol Toxicol (Copenh). 52(Suppl 2):52–64. 16. Organization for Economic Cooperation and Development (OECD). (2001) OECD Test Guideline 401 was deleted in 2002: a major step in animal welfare – OECD Reached Agreement on the Abolishment of the LD50 Acute Toxicity Test [document on the Internet]. Available from: http://www.oecd.org/document/52/0,3343,en 2649 34377 2752116 1 1 1 1,00.html. Cited July 1, 2009.

c03

BLBK249-Cutler

92

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

17. Van Den Heuvel MJ, Clark DG, Fielder RJ, et al. (1990) The international validation of a fixed-dose procedure as an alternative to the classical LD50 test. Food Chem Toxicol. 28(7):469–82. 18. Organization for Economic Cooperation and Development. (2008) Guidelines for testing of chemicals: acute oral toxicity—up-and-down procedure [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K= 5LMQCR2K7MS0&DS=Test-No.-425-Acute-Oral-Toxicity-Up-and-Down-Procedure. Cited July 6, 2009. 19. Organisation for Economic Co-operation and Development. (1997) Test No. 471: acute oral toxicity—acute toxic class method [document on the Internet]. Available from: http://www.ingentaconnect.com/content/oecd/1607310x/2001/00000001/00000004/ 9742301e. Cited July 9, 2009. 20. Organisation for Economic Co-operation and Development. (1997) Summary of considerations in the report from the OECD expert groups on short term and long term toxicology [document on the Internet]. Available from: http://www.ingenta connect.com/content/oecd/1607310x/2001/00000001/00000004/9740001e. Cited July 8, 2009. 21. Organization for Economic Cooperation and Development. (1981) OECD guidelines for the testing of chemicals. Test 452: chronic toxicity studies. [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K= 5LMQCR2K7MLW&DS=Test-No.-452-Chronic-Toxicity-Studies. Cited July 7, 2009. 22. Organization for Economic Cooperation and Development. (2002) Guidance notes for analysis and evaluation of chronic toxicity and carcinogenicity studies.[document on the Internet]. Available from: http://www.ingentaconnect.com/ content/oecd/1607310x/2001/00000001/00000005/9750351e?crawler=true. Cited July 7, 2009. 23. Spielmann H, Bochkov NP, Costa L, et al. (1998) 13th meeting of the scientific group on methodologies for the safety evaluation of chemicals (SGOMSEC): alternative testing methodologies for organ toxicity. Environ Health Perspect. 106(Suppl 2):441–51. 24. European Centre for the Validation of Alternative Methods. (2002) Target organ and target system toxicity. Altern Lab Anim. 30(Suppl 1):71–82. 25. Pfaller W, Balls M, Clothier R, et al. (2001) Novel advanced in vitro methods for longterm toxicity testing; ECVAM Workshop Report 45. Altern Lab Anim. 29:393–426. 26. Prieto P, Clemedson C, Meneguz A, et al. (2005) Subacute and subchronic toxicity. Altern Lab Anim. 33(Suppl 1):109–16. 27. Prieto P, Baird AW, Blaauboer BJ, et al. (2006) The assessment of repeated dose toxicity in vitro: a proposed approach: the report and recommendations of ECVAM workshop 56. Altern Lab Anim. 34:315–41. 28. McKenzie R, Fried MW, Sallie R, et al. (1995) Hepatic failure and lactic acidosis due to fialuridine (FIAU), an investigational nucleoside analogue for chronic hepatitis B. N Engl J Med. 333(17):1099–105. 29. Food and Drug Administration. (1997) S2B Genotoxicity: a standard battery for genotoxicity testing of pharmaceuticals [document on the Internet]. International Conference on Harmonization and the Food and Drug Administration. Available from: http://www.fda.gov/downloads/Drugs/GuidanceCompliance RegulatoryInformation/Guidances/ucm074929.pdf. Cited July 7, 2009. 30. Ames BN, Gurney EG, Miller JA, et al. (1972) Carcinogens as frameshift mutagens: metabolites and derivatives of 2-acetylaminofluorene and other aromatic amine carcinogens. Proc Natl Acad Sci USA. 69(11):3128–32.

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

93

31. McCann J, Ames BN. (1976) Detection of carcinogens as mutagens in the Salmonella/ microsome test: assay of 300 chemicals: discussion. Proc Natl Acad Sci USA. 73:950–4. 32. McCann J, Choi E, Yamasaki E, et al. (1975) Detection of carcinogens as mutagens in the Salmonella/microsome test: assay of 300 chemicals. Proc Natl Acad Sci USA. 72:5135–9. 33. Maron DM, Ames BN. (1983) Revised methods for the Salmonella mutagenicity test. Mut Res. 113:173–215. 34. Organization for Economic Co-operation and Development. (1997) Guidelines for the testing of chemicals. Test 471: bacterial reverse mutation test. [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K= 5LMQCR2K7MG0&DS=Test-No.-471-Bacterial-Reverse-Mutation-Test. Cited July 7, 2009. 35. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 473: in vitro mammalian chromosome aberration test [document on the Internet]. Available from: http://www.ingentaconnect.com/ content/oecd/1607310x/2001/00000001/00000004/9747301e. Cited July 7, 2009. 36. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 476: in vitro mammalian cell gene mutation test [document on the Internet]. Available from: http://www.ingentaconnect.com/content/ oecd/1607310x/2001/00000001/00000004/9747601e. Cited July 7, 2009. 37. Organization for Economic Co-operation and Development. (2007) Draft proposal for a new guideline. Test 487: in vitro mammalian cell micronucleus test (MNvit) [document on the Internet]. Available from: http://www.oecd.org/document/55/ 0,2340,fr 2649 34377 2349687 1 1 1 1,00.html. Cited July 9, 2009. 38. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 474: mammalian erythrocyte micronucleus test [document on the Internet]. Available from: http://www.ingentaconnect. com/content/oecd/1607310x/2001/00000001/00000004/9747401e;jsessionid=2ok50b 4cp3k3m.alexandra. Cited July 7, 2009. 39. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 475: mammalian bone marrow chromosome aberration test [document on the Internet]. Available from: http://www.ingentaconnect. com/content/oecd/1607310x/2001/00000001/00000004/9747501e. Cited July 7, 2009. 40. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 486: unscheduled dna synthesis (UDS) test with mammalian liver cells in vivo. Available from: http://www.ingentaconnect.com/content/ oecd/1607310x/2001/00000001/00000004/9748601e. Cited July 7, 2009. 41. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 484: genetic toxicology: mouse spot test [document on the Internet]. Available from: http://www.ingentaconnect.com/content/ oecd/1607310x/2001/00000001/00000004/9748401e. Cited July 9, 2009. 42. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 483: mammalian spermatogonial chromosome aberration test [document on the Internet]. Available from: http://www.ingentaconnect. com/content/oecd/1607310x/2001/00000001/00000004/9748301e. Cited July 9, 2009. 43. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 485: genetic toxicology, mouse heritable translocation assay [document on the Internet]. Available from: http://www.ingentaconnect. com/content/oecd/1607310x/2001/00000001/00000004/9748501e;jsessionid= 1793486o29sti.alexandra. Cited July 7, 2009.

c03

BLBK249-Cutler

94

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

44. Organization for Economic Co-operation and Development. (1997) OECD guidelines for the testing of chemicals. Test 478: genetic toxicology: rodent dominant lethal test [document on the Internet]. Available from: http://www.ingentaconnect.com/ content/oecd/1607310x/2001/00000001/00000004/9747801e. Cited July 9, 2009. 45. Brendler-Schwaab S, Hartmann A, Pfuhler S, et al. (2005) The in vivo comet assay: use and status in genotoxicity testing. Mutagenesis. 20(4):245–54. 46. Dean SW, Brooks TM, Burlinson B, et al. (1999) Transgenic mouse mutation assay systems can play an important role in regulatory mutagenicity testing in vivo for the detection of site-of-contact mutagens. Mutagenesis. 14(1):141–51. 47. Matthews EJ, Kruhlak NL, Cimino MC, et al. (2006) An analysis of genetic toxicity, reproductive and developmental toxicity, and carcinogenicity data: identification of carcinogens using surrogate endpoints. Regul Toxicol Pharmacol. 44(2):83–96. 48. U.S. Interagency Staff Group on Carcinogens. (1986) Chemical Carcinogens: a review of the science and its associated principles. Environ Health Perspect. 67:201–82. 49. Organization for Economic Cooperation and Development. (1981) OECD guidelines for testing of chemicals. Test 451: carcinogenicity studies [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K= 5LMQCR2K7MQ4&DS=Test-No.-451-Carcinogenicity-Studies. Cited July 7, 2009. 50. Squire RA. (1985) Histopathological evaluation of rat urinary bladders from the IRDC two-generation bioassay of sodium saccharin. Food Chem Toxicol. 23:491–7. 51. Cohen SM, Anderson TA, de OLM, et al. (1998) Tumorigenicity of sodium ascorbate in male rats. Cancer Res. 58:2557–61. 52. Food and Drug Administration. (2008) Guideline for industry: dose selection for carcinogenicity studies of pharmaceuticals. International Conference on Harmonisation and the Food and Drug Administration [document on the Internet]. Available from: http://www.fda.gov/downloads/RegulatoryInformation/Guidances/UCM129112. pdf. Cited July 7, 2009. 53. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 453: combined chronic toxicity/carcinogenicity studies [document on the Internet]. Available from: http://www.ingentaconnect.com/ content/oecd/1607310x/2001/00000001/00000004/9745301e. Cited July 9, 2009. 54. Jena GB, Kaul CL, Ramarao P. (2005) Regulatory requirements and ICH guidelines on carcinogenicity testing of pharmaceuticals: a review on current status. Indian J Pharmacol. 37:209–22. 55. European Medicines Agency. (1998) Guidance for industry: S1B testing for carcinogenicity of pharmaceuticals. International Conference on Harmonization and the European Medicines Agency [document on the Internet]. Available from: http://www.emea.europa.eu/pdfs/human/ich/029995en.pdf. Cited July 9, 2009. 56. Goodman JI. (2001) A perspective on current and future uses of alternative models for carcinogenicity testing. Toxicol Pathol. 29(Suppl):173–6. 57. Maurici D, Aardema M, Corvi R, et al. (2005) Carcinogenicity. Altern Lab Anim. 33(Suppl 1):177–82. 58. Maurici D, Aardema M, Corvi R, et al. (2005) Genotoxicity and mutagenicity. Altern Lab Anim. 33(Suppl 1):117–30. 59. Combes R, Balls M, Curren R, et al. (1999) Cell transformation assays as predictors of human carcinogenicity: the report and recommendations of ECVAM workshop 39. Altern Lab Anim. 27:745–67. 60. Cronin MTD, Jaworska JS, Walker JD, et al. (2003) Use of QSARs in international decision-making frameworks to predict health effects of chemical substances. Environ Health Perspect. 111:1391–401.

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

95

61. United Nations Economic Commission for Europe. (2009) Globally Harmonized System of Classification and Labeling of Chemicals (GHS). Part 3: Health and environmental hazards. Chapter 3.5: Germ cell mutagenicity [document on the Internet]. Available from: http://www.unece.org/trans/danger/publi/ghs/ghs rev02/02files e.html. Cited July 9, 2009. 62. Center for Drug Evaluation and Research and Food and Drug Administration. (2001) Integration of study results to assess concerns about human reproductive and developmental toxicities: reviewer guidance [document on the Internet]. Available from: http://www.fda.gov/downloads/Drugs/GuidanceCompliance RegulatoryInformation/Guidances/UCM079240.pdf. Cited July 7, 2009. 63. Hayes AW. (2007) Principles and Methods of Toxicology. 5th ed. CRC Press, Philadelphia, PA, p336. 64. ICH. (2005) Guideline S5(R2): detection of toxicity to reproduction for medicinal products and toxicity to male fertility. International Conference on Harmonization [document on the Internet]. Available from: http://www.ich.org/LOB/media/ MEDIA498.pdf. Cited July 23, 2009. 65. Barrow PC. (2009) Reproductive toxicity testing for pharmaceuticals under ICH. Reprod Toxicol. 28(2):172–9. 66. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 415: one-generation reproduction toxicity study. [document on the Internet]. Available from: http://www.ingentaconnect.com/ content/oecd/1607310x/2001/00000001/00000004/ 9741501e;jsessionid=2ok50b4cp3k3m.alexandra. Cited July 9, 2009. 67. Organization for Economic Cooperation and Development. (2001) OECD guidelines for the testing of chemicals. Test 416: two-generation reproduction toxicity. [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K= 5LMQCR2K7NLV&DS=Test-No.-416-Two-Generation-Reproduction-Toxicity. Cited July 9, 2009. 68. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 421: reproduction/developmental toxicity screening test [document on the Internet]. Available from: http://www.ingentaconnect. com/content/oecd/1607310x/2001/00000001/00000004/9742101e. Cited July 7, 2009. 69. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 414: prenatal development toxicity study [document on the Internet]. Available from: http://www.ingentaconnect.com/content/ oecd/1607310x/2001/00000001/00000004/9741401e. Cited July 7, 2009. 70. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 426: developmental neurotoxicity study [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K= 5L4FG25MNKXS&DS=Test-No.-426-Developmental-Neurotoxicity-Study. Cited July 7, 2009. 71. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 440: uterotrophic bioassay in rodents: a short-term screening test for oestrogenic properties [document on the Internet]. Available from: http://www.ingentaconnect.com/content/oecd/1607310x/2001/00000001/00000004/ 9744001e;jsessionid=182n5dsnhgshp.alexandra. Cited July 9, 2009. 72. Scholz G, Genschow E, Pohl I, et al. (1999) Prevalidation of the embryonic stem cell test (EST)—a new in vitro embryotoxicity test. Toxicol in Vitro. 13(4–5):675–81. 73. New DA. (1990) Experimental teratology: whole embryo culture, teratogenesis, and the estimation of teratologic risk. Teratology. 42(6):635–42.

c03

BLBK249-Cutler

96

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

74. Fort DJ, Stover EL, Farmer DR, et al. (2000) Assessing the predictive validity of frog embryo teratogenesis assay—zenopus (FETAX) teratogenesis, carcinogenesis, and mutagenesis. Teratog Carcinog Mutagen. 20(2):87–98. 75. Piper JM, Ray WA, Rosa FW. (1992) Pregnancy outcome following exposure to angiotensin-converting enzyme inhibitors. Obstet Gynecol. 80(3 Pt 1):429–32. 76. Food and Drug Administration. (2001) Guidance for industry. S7A: safety pharmacology studies for human pharmaceuticals. International Conference on Harmonization and the Food and Drug Administration [document on the Internet]. Available from: http://www.fda.gov/RegulatoryInformation/Guidances/ucm129150.htm. Cited July 7, 2009. 77. Kinter L, Johnson D. (1998) Safety pharmacology testing in general toxicity studies. J Toxicol Sci. 23(4):267. 78. Food and Drug Administration. (2005) Guidance for industry. S7B: the nonclinical evaluation of the potential for delayed ventricular repolarization (QT interval prolongation) by human pharmaceuticals. International Conference on Harmonisation and the Food and Drug Administration [document on the Internet]. Available from: http://www. fda.gov/RegulatoryInformation/Guidances/ucm129121.htm. Cited July 7, 2009. 79. Hanton G. (2007) Preclinical cardiac safety assessment of drugs—a review article. Drugs R&D. 8(4):213–28. 80. Fu EY, Clemo HF, Ellenbogen KA. (1998) Acquired QT prolongation: mechanisms and implications. Cardiol Rev. 6(6):319–24. 81. Koh KK, Rim MS, Yoon J, et al. (1994) Torsade de pointes induced by terfenadine in a patient with long QT syndrome. J Electrocardiol. 27(4):343–6. 82. Oberg KC, Bauman JL. (1995) QT interval prolongation and torsades de pointes due to erythromycin lactobionate. Pharmacotherapy. 15(6):687–92. 83. Paris DG, Parente TF, Bruschetta HR, et al. (1994) Torsades de pointes induced by erythromycin and terfenadine. Am J Emerg Med. 12(6):636–8. 84. Roden DM. (1993) Torsade de pointes. Clin Cardiol. 16(9):683–6. 85. Granberry MC, Gardner SF. (1996) Erythromycin monotherapy associated with torsade de pointes. Ann Pharmacother. 30(1):77–8. 86. Katapadi K, Kostandy G, Katapadi M, et al. (1997) A review of erythromycin-induced malignant tachyarrhythmia—Torsade de pointes: a case report. Angiology. 48(9):821–6. 87. Albengres E, Le LH, Tillement JP. (1998) Systemic antifungal agents: drug interactions of clinical significance. Drug Saf. 18(2):83–97. 88. Sakemi H, VanNatta B. (1993) Torsade de pointes induced by astemizole in a patient with prolongation of the QT interval. Am Heart J. 125(5 Pt 1):1436–8. 89. Drolet B, Khalifa M, Daleau P, et al. (1998) Block of the rapid component of the delayed rectifier potassium current by the prokinetic agent cisapride underlies drug-related lengthening of the QT interval. Circulation. 97(2):204–10. 90. Babe KS, Serafin WE. (1996) Histamine, bradykinin, and their antagonists. In: Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, editors. Goodman & Gilman’s The Pharmacological Basis of Therapeutics. 9th ed. McGraw-Hill, New York, p592. 91. O’Brien JM, Rockwood RP, Suh KI. (1999) Haloperidol-induced torsade de pointes. Ann Pharmacother. 33(10):1046–50. 92. Patterson S AM, Anziano R, Burgess T, et al. (2005) Investigating drug-induced QT and QTc prolongation in the clinic: a review of statistical design and analysis considerations. Drug Inf J. 39:243–66. 93. Gerhardy A, Scholtysik G, Schaad A, et al. (1998) Generating and influencing torsades de pointes-like polymorphic ventricular tachycardia in isolated guinea pig hearts. Basic Res Cardiol. 93(4):285–94.

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

97

94. Hondeghem LM, Hoffmann P. (2003) Blinded test in isolated female rabbit heart reliably identifies action potential duration prolongation and proarrhythmic drugs: importance of triangulation, reverse use dependence, and instability. J Cardiovasc Pharmacol. 41(1):14–24. 95. Vandenhende F. (2001) Heart rate-specific reference ranges for QT-interval in beagle dogs. Drug Inf J. 35:1179–88. 96. Gad S. (2004) Safety Pharmacology in Pharmaceutical Development and Approval. 1st ed. CRC Press, Boca Raton, FL. 97. Detweiler DK. (1997) Electrocardiography in toxicological studies. In: Sipes IG MC, Gandolfi AJ, editors. Comprehensive Toxicology. 6th ed. Pergamon Press, New York. 98. Gralinski M. (2003) The dog’s role in the preclinical assessment of QT interval prolongation. Toxicol Pathol. 31(Suppl):11–6. 99. Olson H, Betton G, Robinson D, et al. (2000) Concordance of the toxicity of pharmaceuticals in humans and in animals. Regul Toxicol Pharmacol. 32:56–67. 100. Crumb W, Cavero I. (1999) QT interval prolongation by non-cardiovascular drugs: issues and solutions for novel drug development. Pharm Sci Technol Today. 2:270– 80. 101. Taglialatela M, Castaldo P, Pannaccione A, et al. (1998) Human ether-a-gogo related gene (HERG) K+ channels as pharmacological targets: present and future implications. Biochem Pharmacol. 55:1741–6. 102. Shah R. (2002) The significance of QT interval in drug development. J Clin Pharmacol. 54:188–202. 103. Roy M, Dumaine R, Brown AM. (1996) HERG, a primary human ventricular target of the nonsedating antihistamine terfenadine. Circulation. 94(4):817–23. 104. Woosley RL, Chen Y, Freiman JP, et al. (1993) Mechanism of the cardiotoxic actions of terfenadine. JAMA. 269(12):1532–6. 105. Redfern W, Carlsson L, Davis A, et al. (2003) Relationships between preclinical cardiac electrophysiology, clinical QT interval prolongation and torsade de pointes for a broad range of drugs: evidence for a provisional safety margin in drug development. Cardiovasc Res. 58(1):32–45. 106. Food and Drug Administration. (2006) Guidance for industry. S8: immunotoxicity studies for human pharmaceuticals. International Conference on Harmonization and the Food and Drug Administration [document on the Internet]. Available from: http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInfor mation/Guidances/ucm074965.pdf. Cited July 9, 2009. 107. Luster MI, Portier C, Pait DG, et al. (1992) Risk assessment in immunotoxicology: sensitivity and predictability of immune tests in immunotoxicology. Fund Appl Toxicol. 18:200–210. 108. Organization for Economic Cooperation and Development. (1997) OECD guidelines for the testing of chemicals. Test 424: neurotoxicity study in rodents. [document on the Internet]. Available from: http://www.ingentaconnect.com/content/oecd/ 1607310x/2001/00000001/00000004/9742401e;jsessionid=2g6t7ia1ml6ub.alexandra. Cited July 9, 2009. 109. Tupper DE, Wallace RB. (1980) Utility of the neurologic examination in rats. Acta Neurobiol Exp. 40:999–1003. 110. Gad SC. (1982) A neuromuscular screen for use in industrial toxicology. J Toxicol Environ Health. 9:691–704. 111. Moser VC, McDaniel KM, Phillips PM. (1991) Rat strain and stock comparisons using a functional observational battery: baseline values and effects of amitraz. Toxicol Appl Pharmacol. 108:267–83.

c03

BLBK249-Cutler

98

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

112. Meyer OA, Tilson HA, Byrd WC, et al. (1979) A method for the routine assessment of fore- and hind-limb grip strength of rats and mice. Neurobehav Toxicol. 1:233–6. 113. Crofton KM, Howard JL, Moser VC, et al. (1991) Interlaboratory comparison of motor activity experiments: implications for neurotoxicological assessments. Neurotoxicol Teratol. 13:599–609. 114. Organization for Economic Cooperation and Development. (2004) OECD guidance document. Test 20: neurotoxicity testing. [document on the Internet]. Available from: http://www.oecd.org/LongAbstract/0,3425,en 2649 34377 34605367 1 1 1 37407,00.html. Cited July 9, 2009. 115. Tilson HA, Mitchell CL. (1992) Neurotoxicology Target Organ Toxicology Series. Raven Press, New York. 116. Chang LW. (1995) Principles of Neurotoxicology. Marcel Dekker, New York. 117. Broxup B. (1991) Neuropathology as a screen for neurotoxicity assessment. J Amer Coll Toxicol. 10:689–95. 118. Moser VC, Anthony DC, Sette WF, et al. (1992) Comparison of subchronic neurotoxicity of 2-hydroxyethyl acrylate and acrylamide in rats. Fund Appl Toxicol. 18:343–52. 119. O’Callaghan JP. (1988) Neurotypic and gliotypic proteins as biochemical markers of neurotoxicity. Neurotoxicol Teratol. 10:445–52. 120. O’Callaghan JP, Miller DB. (1988) Acute exposure of the neonatal rat to triethyltin results in persistent changes in neurotypic and gliotypic proteins. J Pharmacol Exp Ther. 244:368–78. 121. Fox DA, Lowndes HE, Birkamper GG. (1982) Electrophysiological techniques in neurotoxicology. In: Mitchell CL, editor. Nervous System Toxicology. Raven Press, New York, pp 299–335. 122. Johnson BL. (1980) Electrophysiological methods in neurotoxicity testing. In: Spencer PS, Schaumburg HH, editors. Experimental and Clinical Neurotoxicology. Williams and Wilkins Company, Baltimore, pp 726–42. 123. Davis CS, Richardson RJ. (1980) Organophosphorus compounds. In: PS Spencer, HH Schaumberg, editors. Experimental Clinical Neurotoxicology. Williams and Wilkins Company, Baltimore, pp 527–44. 124. Johnson MK. (1975) Organophosphorus esters causing delayed neurotoxic effects: mechanism of action and structure/activity studies. Archiv Toxicol. 34:259–88. 125. Organization for Economic Cooperation and Development. (1995) Guideline for testing of chemicals. Test 418: delayed neurotoxicity of organophosphorus substances following acute exposure [document on the Internet]. Available from: http:// www.oecdbookshop.org/oecd/display.asp?K=5LMQCR2K7NF6&DS=Test-No.-418Delayed-Neurotoxicity-of-Organophosphorus-Substances-Following-Acute-Exposure. Cited July 9, 2009. 126. Organization for Economic Cooperation and Development. (1995) Guidelines for testing of chemicals. Test 419: delayed neurotoxicity of organophosphorus substances: 28-day repeated dose study [document on the Internet]. Available from: http://www.oecdbookshop.org/oecd/display.asp?K=5LMQCR2K7NBQ&DS=TestNo.-419-Delayed-Neurotoxicity-of-Organophosphorus-Substances-28-day-RepeatedDose-Study. Cited July 9, 2009. 127. Johnson MK. (1982) The target for initiation of delayed neurotoxicity by organophosphorus esters: biochemical studies and toxicological applications. Rev Biochem Toxicol. 4:141–212. 128. Johnson MK. (1977) Improved assay of neurotoxic esterase for screening organophosphates for delayed neurotoxicity potential. Archiv Toxicol. 37:113–5.

c03

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Preclinical antecedents to early human clinical trials

99

129. Zech R, Chemnitius JM. (1987) Neurotoxicant sensitive esterase: enzymology and pathophysiology of organophosphorus ester-induced delayed neuropathy. Prog Neurobiol. 29:193–218. 130. Kayyali US, Moore T, Randall JC, et al. (1991) Neurotoxic esterase (NTE) assay: optimized conditions based on detergent-induced shifts in the phenol/4-aminoantipyrine chromophore spectrum. J Anal Toxicol. 15:86–9. 131. Carrera V, D´ıaz-Alejo N, Sogorb MA, et al. (1994) In vivo inhibitionby mipafox of soluble and particulate forms of organophosphorus neuropathy target esterase (NTE) in hen sciatic nerve. Toxicol Lett. 71:47–51. 132. Moretto A, Capodicasa E, Peraica M, et al. (1991) Age sensitivity to organophosphateinduced delayed polyneuropathy. Biochemical and toxicological studies in developing chicks. Biochem Pharmac. 41(10):1497–504. 133. Tormo N, Gimeno JR, Sogorb MA, et al. (1993) Soluble and particulate organophosphorus neuropathy target esterase in brain and sciatic nerve of the hen, cat, rat and chick. J Neurochem. 61(6):2164–8. 134. Johnson CD, Russell RL. (1975) A rapid, simple, radiometric assay for cholinesterase, suitable for multiple determinations. Anal Biochem. 64:229–38. 135. Ellman GL, Courtney KD, Andres V, Jr, et al. (1961) A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem Pharmacol. 7:88–95. 136. Conolly RB, Beck BD, Goodman JI. (1999) Stimulating research to improve the scientific basis of risk assessment. Toxicol Sci. 49:1–4. 137. Spear LP. (1990) Neurobehavioral assessment during the early postnatal period. Neurotoxicol Teratol. 12:489–95. 138. Altman J, Sudarshan K. (1975) Postnatal development of locomotion in the laboratory rat. Anim Behav. 23:896–920. 139. Adams J. (1986) Methods in behavioral teratology. In: Riley Ep VCV, editor. Handbook of Behavioral Teratology. Plenum Press, New York, pp 67–100. 140. Reiter LW, MacPhail RC. (1979) Motor activity: a survey of methods with potential use in toxicity testing. Neurobehav Toxicol. 1:53–66. 141. Robbins TW. (1977) A critique of the methods available for the measurement of spontaneous motor activity. In: Iverson LL, Iverson DS, Snyder SH, editors. Handbook of Psychopharmacology. Plenum Press, New York, pp 37–82. 142. Ross JF, Handley DE, Fix AS, et al. (1997) Quantification of the hindlimb extensor thrust response in rats. Neurotoxicol Teratol. 19:405–11. 143. Handley DE, Ross JF, Carr GJ. (1998) A force plate system for measuring lowmagnitude reaction forces in small laboratory animals. Physiol Behav. 64:661–9. 144. Koch M. (1999) The neurobiology of startle. Prog Neurobiol. 59:107–28. 145. Mattsson JL, Boyes WK, Ross JF. (1992) Incorporating evoked potentials into neurotoxicity test schemes. In: Tilson H, Mitchell C, editors. Target Organ Toxicology Series: Neurotoxicity. Raven Press, New York, pp 125–45. 146. Peele DB, Allison SD, Crofton KM. (1990) Learning and memory deficits in rats following exposure to 3,3 -iminopropionitrile. Toxicol Appl Pharmacol. 105:321– 32. 147. Bushnell PJ. (1988) Effects of delay, intertrial interval, delay behavior and trimethyltin on spatial delayed response in rats. Neurotoxicol Teratol. 10:237–44. 148. Green RJ, Stanton ME. (1989) Differential ontogeny of working memory and reference memory in the rat. Behav Neurosci. 103:98–105. 149. Kucharski D, Spear NE. (1984) Conditioning of aversion to an odor paired with peripheral shock in the developing rat. Develop Psychobiol. 17:465–79.

c03

BLBK249-Cutler

100

Trim: 6in X 9in

January 19, 2010

8:11

Char Count=

Chapter 3

150. Morris R. (1984) Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods. 11:47–60. 151. Voorhees CV. (1997) Methods for detecting long-term CNS dysfunction after prenatal exposure to neurotoxins. Drug Chem Toxicol. 20:387–99. 152. Akaike M, Tanaka K, Goto M, et al. (1988) Impaired Biel and Radial arm maze learning in rats with methyl-nitrosurea induced microcephaly. Neurotoxicol Teratol. 10:327–32. 153. Crofton KM, Peele DB, Stanton ME. (1993) Developmental neurotoxicity following neonatal exposure to 3,3 -iminodipropionitrile in the rat. Neurotoxicol Teratol. 15:117–29. 154. Cory-Slechta DA, Weiss B, Cox C. (1983) Delayed behavioral toxicity of lead with increasing exposure concentration. Toxicol Appl Pharmacol. 71:342–52. 155. De GDMG, Hartgring S, van dHL, et al. (2005) 2D and 3D assessment of neuropathology in rat brain after prenatal exposure to methylazoxymethanol, a model for developmental neurotoxicity. Reprod Toxicol. 20:417–32. 156. Hyman BT, Gomez-Isla T, Irizarry MC. (1998) Stereology: a practical primer for neuropathology. J Neuropathol Exp Neurol. 57:305–10. 157. Jensen KF, Catalano SM. (1998) Brain morphogenesis and developmental neurotoxicology. In: Slikker W, Jr, Chang LW, editors. Handbook of Developmental Neurotoxicology. Academic Press, New York, pp 3–41. 158. Gavin CE, Kates B, Gerken LA, et al. (1994) Patterns of growth deficiency in rats exposed inutero to undernutrition, ethanol, or the neuroteratogen methylazoxymethanol (MAM). Teratology. 49:113–21. 159. Harry GJ, Billingsley M, Bruinink A, et al. (1998) In vitro techniques for the assessment of neurotoxicity. Environ Health Perspect. 106(Suppl 1):131–58. 160. Loscher W. (2009) Preclinical assessment of proconvulsant drug activity and its relevance for predicting adverse events in humans. Eur J Pharmacol. 610(1–3):1–11. 161. Easter A, Bell ME, Damewood JR, Jr, et al. (2009) Approaches to seizure risk assessment in preclinical drug discovery. Drug Discov Today. 14(17–18):876–84. 162. Schacter BZ. (2006) The New Medicines: How Drugs Are Created, Approved, Marketed, and Sold. Greenwood Publishing Group, Westport, CT, p70. 163. Environmental Protection Agency. (1992) A cross-species scaling factor for carcinogen risk assessment based on equivalence of mg/kg 3/4/day. Federal Register. 57:24152–73. 164. Lowe MC, Davis RD. (1998) The current toxicology protocol of the National Cancer Institute. In: Hellman K, Carter SK, editors. Fundamentals of Cancer Chemotherapy. McGraw-Hill, New York, pp 228–35. 165. Mae-Wan H, Cummins J. (2006) London drug trial catastrophe—collapse of science and ethics. Institute of Science in Society Report [document on the Internet]. Available from: http://www.i-sis.org.uk/LDTC.php. Cited 20 August, 2009. 166. Hopkin M. (2006) Can super-antibody drugs be tamed? Nature. 440:855–6. 167. Kenter MJ, Cohen AF. (2006) Establishing risk of human experimentation with drugs: lessons from TGN1412. Lancet. 368(9544):1387–91. 168. Stebbings R, Findlay L, Edwards C, et al. (2007) Cytokine storm in the phase I trial of monoclonal antibody TGN1412: better understanding the causes to improve preclinical testing of immunotherapeutics. J Immunol. 179(5):3325–31. 169. Moberly T. (2006) TGN1412: lessons learnt the hard way. Pharm J. 277:730.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

CHAPTER 4

Biomarkers and surrogate markers in drug development

Introduction While animal models provide a good starting point for identifying a novel compound’s putative safety and efficacy, there is no clear and easy path that leads from preclinical trials to clinical trials. Often results in animals do not mirror those in humans, and invasive or harmful techniques used in animal models cannot be carried over to clinical trials. Early indicators of efficacy and toxicity in humans are needed in order to minimize the cost and time required to run full-term clinical trials; companies can focus their attention on the compounds that are potentially efficacious, saving time and money in clinical trials. Such an approach would also prevent patients from being exposed to potentially inefficacious or toxic compounds in early clinical trials. Biological markers, or biomarkers, are used to help bridge the gap between preclinical and clinical trials. Biomarkers provide early data on whether a compound is reaching its intended target or modifying the intended disease pathway or process it was designed to modify. They can also provide preliminary information on the efficacy, dose range, side effects, and toxicity of a novel compound, among other measures [1–3]. Biomarkers can be surrogate indicators of a disease state; valid biomarkers should change in a predictive and reliable manner with changes to the underlying disorder [4]. In order to provide useful information, a biomarker should demonstrate the following: 1. A reasonable reflection of overall disease severity, with changes that correspond longitudinally with clinical deterioration 2. Active changes in response to current disease activity 3. A good temporal association between drug administration and changes in the biomarker 4. Eventual correspondence with changes in a primary clinical endpoint measure Biomarkers must be validated to ensure their predictive value. Validation criteria are specific for each biomarker, the underlying disorder, and the task the biomarker is intended to measure (see Table 4.1 for a list of some biomarker definitions and validation criteria) [5, 6]. A biomarker that is fully validated in its ability to predict clinical endpoints is referred to as a surrogate Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

101

c04

BLBK249-Cutler

102

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

Table 4.1 Some biomarker definitions and validation terms Biological marker (biomarker)

A proposed indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention

Clinical endpoint

An indicator that measures how a patient feels, functions, or survives

Surrogate marker/endpoint

A biomarker that is intended to substitute for a clinical endpoint and is able to predict clinical outcomes. A biomarker that has been fully validated becomes a surrogate marker—i.e., high blood pressure for cardiovascular disease

Sensitivity

Ability to identify all patients with the disease

Specificity

Ability to identify all individuals without the disease

Accuracy

How well the biomarker reflects the underlying condition

Precision

The reproducibility of the biomarker across trials and subjects

Prior probability

Background prevalence of disease in the population

Positive predictive value

Presence of biomarker predicts disease diagnosis

Negative predictive value

Absence of biomarker predicts no disease diagnosis

Dynamic range

The range of disease state that the biomarker varies with

Assay linearity

The ability of an assay to obtain results (within a given range) that vary in a manner directly proportional to changes in the concentration of the biomarker

Analytical range

Analytical range of the biomarker over which it shows acceptable performance

Analytical specificity

Ability of the method to measure the biomarker with reacting with other related substances

Calibration/standard curve

Relationship between known quantities of biomarker material and instrument response

Reproducible range

The range of reproducible results over which the method is validated

Sample stability

Stability of the biomarker under various conditions

Ruggedness

The ability of the biomarker to produce similar results under different analytical condition

marker for that disorder. Examples of surrogate markers include CD4+ T-cell count as an indicator of progression to AIDS and blood pressure for cardiovascular disease.

CNS biomarkers Biomarkers for central nervous system (CNS) disorders present a unique challenge beyond those for nonpsychiatric conditions. Many neuropsychiatric disorders present clinically in a heterogeneous manner—symptoms vary from patient to patient, and it is not always clear how symptom presentation

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

103

Table 4.2 Categories of biomarkers Brain imaging techniques

Computed tomography (CT), regional cerebral blood flow (rCBF), magnetic resonance imaging (MRI), positron emission tomography (PET), single photon emission computed tomography (SPECT), magnetic resonance spectroscopy (MRS), magnetoencephalography (MEG)

Cell-based imaging

Fluorescent resonant energy transfer, confocal imaging in brain slices

Electrophysiological marker

Electroencephalogram (EEG), pupillometry, saccadic eye movements

Small molecule marker

Concentrations of catecholamines, hormones, enzymes, proteins, drugs, and drug metabolites

Immunological marker

Immunoglobulin, lymphocyte responses, lymphokine, cytokine, interleukin, interferon; viral serology; Alz-50; anticardiolipin antibodies (ACA)

Neuroendocrine marker

Dexamethasone-suppression test (DST), thyrotropin-releasing hormone stimulation test (TRHST), growth hormone (GH) challenge test, cortisol (saliva and plasma), ACTH

Provocative anxiety tests

Lactate infusion, carbon dioxide (CO2 ) challenge, cholecystokinin (CCK) challenge

Genetic markers

DNA banking, genotyping, restriction fragment length polymorphisms (RFLPs), microarrays

Proteomic and metabolomic markers

Nuclear magnetic resonance (NMR), lipoprotein fractions and subfractions, matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS)

correlates with the underlying pathology. The most common CNS disorders develop over time and show highly subjective and context-dependent signs and symptoms. In addition, there is a fundamental lack of understanding about the neurochemical and biological basis for many of these conditions, which highlights the need for reliable diagnostic indicators to act as surrogate markers for disease progression and drug efficacy. There are many different types of biomarkers for CNS disorders (see Table 4.2 for a breakdown of biomarker categories). The explosion of brain imaging technology and genetic profiling have allowed for a unique view of CNS dysfunction and a compound’s therapeutic properties. Positron emission tomography (PET) studies of radiolabeled compounds can be used to trace a compound’s path, its interactions, and its effects in the brain very early in clinical trials, even before a full toxicology profile is established. PET microdosing studies involve administration of a radiolabeled compound at very low dose without toxicity concerns. These studies provide invaluable information about a compound’s pharmacokinetics, including whether it passes through the blood–brain barrier, its relative concentration in the brain versus plasma, the rate of exchange between brain and plasma, and its accumulation throughout the body. In addition, labeling of neurotransmitter receptors

c04

BLBK249-Cutler

104

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

and functional magnetic resonance imaging (fMRI) studies can visualize how a compound alters brain function on a small and large scale. The microarrays of various putative biomarkers are being used to examine the genetic profile of a patient and track how this profile changes after administration of a compound. For example, the early clinical screening of compounds for toxicity utilizes microarrays to detect expression changes in toxicity-associated genes. Genetic differences in pharmacodynamic and pharmacokinetic markers can account for variations in drug metabolism, response and clinical outcomes, as well as adverse events. In the future, gene expression profiling of blood, peripheral blood mononuclear cells (PBMCs), and cerebrospinal fluid (CSF) might be used to monitor the efficacy and safety of novel CNS compounds in early clinical trials. The measurements of secreted biomarkers in peripheral fluids, such as serum, plasma, CSF, saliva, or urine, can also give a helpful readout of a compound’s effect on the CNS. Protein and peptide biomarkers, while largely unknown for psychiatric disorders, are very useful in Alzheimer’s disease [7]. In addition, many biomarkers are derivatives of the same metabolic pathway, such as catecholamines, which can function as biomarkers for several CNS indications at various metabolic steps (see Figure 4.1 for an overview of catecholamine biomarker metabolism). Antibody biomarker assays are being developed to evaluate a compound’s toxicity and efficacy. In addition, proteomic and metabolomic assays are being developed to identify global differences in endogenous groups of proteins, metabolites, and other small molecular weight molecules in individuals with psychiatric disorders versus healthy controls [8]. In this chapter, we focus on CNS biomarkers that have relevance to the potential efficacy of a compound, but drug developers should also keep foremost in mind the potential to use biomarkers as safety indicators based on preclinical data. For example, γ-secretase inhibitors being developed for Alzheimer’s disease not only work on amyloid precursor protein (APP), but they can also cleave other proteins, including Notch (important for cell signaling in many pathways) and can potentially affect goblet cells in the gastrointestinal (GI) tract. Fortunately, there are several specific blood messenger ribonucleic acid (mRNA) and protein assays that can be used to track the effects of a γ-secretase inhibitor on these other critical systems at an early stage in clinical development. Currently, the use of safety biomarkers is primarily an exploratory measure and must be considered only as an adjunct to a thorough clinical evaluation of the patient. This chapter reviews the most promising biomarkers of efficacy for four major CNS conditions. Alzheimer’s disease, anxiety disorder, depression, and schizophrenia. It also reviews some of the newest, most cutting-edge biomarkers being developed for these conditions. While there are no fully validated biomarkers for these CNS disorders yet, these examples can provide important information regarding a compound’s mechanism of action and pharmacodynamic effects early in development, and can be of great utility in a drug

Methoxyhydroxyphenylglycol (MHPG)

Homovanillic acid (HVA)

Figure 4.1 Multiple biomarkers from one metabolic pathway: catecholamines.

Dihydroxyphenyl glycol (DHPG)

Dihydroxyphenylacetic acid (DOPAC)

Vanilllymandelic acid (VMA)

Inactivation by reuptake through 5-HT transporter (SERT)

Serotonin 5-Hydroxytryptamine (5-HT)

5-Hydroxyindoleacetic acid (5-HIAA)

21:32

Adrenaline (A)

Norepinephrine (NE) Noradrenalinc(NA)

Dopamine (DA)

5-Hydroxytryptophan

December 29, 2009

Inactivation by reuptake through NE transporter (NET)

Inactivation by reuptake through DA transporter (DAT)

Tryptophan

Trim: 6in X 9in

DOPA

Tyrosine

BLBK249-Cutler

Catecholamines

c04 Char Count=

Biomarkers and surrogate markers in drug development 105

c04

BLBK249-Cutler

106

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

development program. For example, the ability to detect sufficient concentrations of a compound in CSF, as well as changes in pharmacodynamic endpoints, such as brain monoamine levels, provides important early information to developers. The conclusion of this chapter discusses the Critical Path Initiative of the US Food and Drug Administration (FDA) and how it relates to biomarker qualification and development, as well as bioanalytical considerations for the laboratory measurements of biomarkers.

Alzheimer’s disease As in many CNS indications, the most valuable endpoint in Alzheimer’s disease (AD) studies is a therapeutic outcome measure, while the most practical pharmacodynamic endpoint is a valid biomarker. For example, amyloidbeta (Aβ) and APP levels may be indicative of disease progression in AD, but do not reveal relevant information about the quality of life of a patient with AD. Expert consensus conferences have established certain guidelines that biomarkers of AD should meet. These include that the marker reflects a neuropathologic characteristic of the disease, can be validated in patients with a neuropathologic diagnosis, changes with disease progression, and is reproducible. In addition, AD biomarkers are evaluated for their sensitivity, specificity, prior probability, positive predictive value, and negative predictive value. AD biomarkers should have at least 85% sensitivity, 75% specificity, and a positive predictive value of over 80% [9, 10]. The biomarkers of AD include neurochemical, neuroimaging, PET, and genetic markers. At present, none of these markers are robust, affordable, and standardized enough to be routinely used in the clinic, but they have all undergone significant development and have the potential to become promising diagnostic and therapeutic indicators. Biomarkers for AD could be potentially used in the preliminary clinical trials of a novel compound to identify correct dosage, improve safety, demonstrate pharmacologic activity, and document onset of activity or even efficacy [11]. The following section reviews the major biomarkers for AD, summarized in Table 4.3.

Cholinergic markers The destruction of nerve cells during the pathogenesis of AD results in an impairment in neurotransmitter systems, with components of the cholinergic system being the most severely affected. Acetylcholine (ACh) is an essential neurotransmitter in the central and peripheral nervous system, with roles in selective attention, sensory processing, and associative thinking, among many others. Studies have shown that ACh receptors (AChR) are significantly reduced during the course of AD [12], and choline acetyltransferase activity (an enzyme that helps form ACh), ACh release, and high affinity choline uptake are all impaired in the basal forebrain of AD patients [13–15]. Choline acetyltransferase activity deficits were also correlated with the onset of clinical dementia [16]. In addition, insoluble plaques formed from Aβ pathology have

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development Table 4.3 Biomarkers for AD

107

Cholinergic markers Acetylcholine (ACh) ACh receptors (AChR) Choline acetyltransferase activity Acetylcholinesterase (AChE) Amyloid-beta (A␤) Aβ42 Aβ40 Aβ42/Aβ40 Tau protein Total tau (t-tau) Hyperphosphyorylated tau (p-tau) Protease enzymes β-secretase (BACE1) γ-secretase α-synuclein Amyloid precursor protein (APP) Platelet APP Soluble APP (sAPP) Isoprostanes Sulfatides Inflammatory biomarkers Prostaglandins Complement components Cytokines(IL-1, IL-6, and TNF-α) Chemokines Proteases Protease inhibitors Free radicals Electroencephalography (EEG) Quantitative EEG (QEEG) Overall pattern changes, deterioration, and covariance changes α-, β-, δ-, and θ-wave changes Salivary amylase CD-69 Genetic markers of AD Cholinesterase-metabolizing gene CYP2D6 Epsilon4 (ε4) isoform of the apolipoprotein E (ApoE) gene

been linked to the degeneration of cholinergic neurons in studies of AD patients [17, 18]. The upregulation of acetylcholinesterase (AChE; an enzyme that breaks down ACh) around amyloid plaques further inhibits ACh transmission in AD [19]. The evidence of cholinergic deficits in AD has led to the development of new biomarkers based on this system, with the goal being to track disease progression and to evaluate novel therapeutics, many of which target the cholinergic system directly. Boosting cholinergic function has been the most commonly used treatment for AD, with acetylcholinesterase inhibitors (AChEIs)

c04

BLBK249-Cutler

108

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

being the most successful therapeutic strategy to date [20]. AChE activity has been used as an early clinical biomarker for efficacy of these compounds. For example, rivastigmine, an inhibitor of AChE and butylcholinesterase, has been shown to produce moderate but significant improvements in AD patients on tests of cognitive performance [21]. It also induces a significant decrease in AChE activity [22]. Donepezil is another selective AChE inhibitor that demonstrates both short- (up to 6 months) and long-term (up to 1 year) modest positive effects on cognitive function [23]. While one study showed it reduced AChE activity in the cerebral cortex of AD patients [24], another study showed it produced a significant increase in CSF AChE activity, possibly due to the induction of tolerance or a unique regulatory mechanism [25]. Peripheral AChE activity has also shown conflicting results in regard to predicting AD drug efficacy and cognitive improvements in AD patients. One study found that amiridine and gliatiline both inhibited red blood cell (RBC) AChE in AD patients, and this inhibition was significantly correlated with the clinical efficacy of the compounds [26]. On the other hand, a 2000 review by the authors found inconsistent results regarding the use of RBC AChE as a predictor of clinical response to AD therapeutics that functioned as AChE inhibitors. The degree of inhibition yielding maximum cognitive improvements was highly variable using different compounds (30–80%) [27]. Further, investigators did not prove a relation between central and peripheral pharmacodynamics (PD), or demonstrate an advantage over dose in the ability of RBC AChE inhibition to predict clinical response. Another study compared CSF and RBC AChE activity after chronic treatment of AD patients with the AChE inhibitor metrifonate [28]. They found that CSF AChE recovered faster than RBC AChE, and RBC AChE inhibition did not necessarily reflect CSF or brain AChE inhibition. These studies suggest that peripheral (RBC) AChE inhibition may be an inconsistent indicator of AD drug efficacy. The results with CSF AChE inhibition are more promising, but this method is more invasive and therefore harder to justify in a clinical setting. A better understanding of the relationship between AChE inhibition and AD drug efficacy is required before AChE inhibition can be a useful AD biomarker.

Amyloid-␤ The amyloid-beta (Aβ) peptide makes up the main insoluble component of AD-associated senile plaques, which are deposits of amyloid in the gray matter of the brain linked to neural degeneration and cognitive deficits. Aβ forms from cleavage of the amyloid precursor protein (APP), which is excreted from cells and exists in numerous isoforms. An isoform with a length of 42 amino acids (Aβ42) is primarily indicated in the pathogenesis of AD, and is the first and most prominent component of senile plaques. Another isoform with a length of 40 amino acids (Aβ40) is deposited later in the disease and is prominent in vascular amyloid deposits [29]. Numerous studies have shown that Aβ42 in CSF is actually reduced in AD patients by around 50% in

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

109

comparison to normal controls. The diagnostic sensitivity and specificity is 80–90% when compared to normal controls but only 60% when differentiating between other forms of dementia [30, 31]. The reasons for this reduction are still unknown. Comparing the ratio between Aβ42/Aβ40 in CSF is another potential marker for disease diagnosis and progression. While Aβ42 decreases in the CSF of AD patients, Aβ40 appears to remain constant in comparison to normal controls [32, 33]. Therefore, there is a decreased Aβ42/Aβ40 ratio in patients with AD compared to both non-AD dementia and normal controls [34]. Sampling from the CSF is a more intensive and invasive procedure than measuring peripheral blood, plasma, or serum levels of Aβ. Unfortunately, no serum biomarkers possess the sensitivity and specificity of these CSF biomarkers, and studies examining the relationship between serum Aβ levels and disease state have often been contradictory [35–37]. In addition, the concentration of Aβ in CSF is 50–100 times higher than in plasma, and changes in plasma Aβ seem to have no effect on CSF Aβ [38, 39]. New techniques are allowing for direct imaging of amyloid in the brain using PET combined with multiphoton microscopy. Many fluorescent radioligands are thought to label fibrillary Aβ [40]. The most extensively studied ligand is called Pittsburgh Compound-B (PIB), which appears to be relatively selective for Aβ plaques and shows enhanced uptake in AD patients compared with healthy controls [41, 42]. Because these imaging techniques are still relatively new, it remains to be determined how well they compare to other, more established biomarkers. An important advance in this field has been the quantification of synthesis and clearance rates of Aβ in the CSF. The amino acid leucine, a building block of Aβ, was labeled with an isotope and given intravenously to subjects. The spinal taps of their CSF at various intervals showed that this isotope-labeled leucine was being incorporated into Aβ in as little as 5 hours, and all of the labeled Aβ was cleared from the CSF in approximately 12 hours [43]. This technique could be coupled with measurements of unique biomarkers based on PD changes in Aβ (or any other compound for that matter) which may occur in AD patients when compared to healthy subjects. An interesting observation is that Aβ40 and Aβ42 levels are significantly correlated in healthy subjects, indicating they may share similar production and clearance rates [44]. The observed decrease in Aβ42 relative to Aβ40 in AD patients could represent a disease-induced alteration in the PD parameters of this compound. An important point to consider for any study involving Aβ-based biomarkers is the dramatic diurnal fluctuations Aβ exhibits. Aβ CSF levels fluctuate as much as fourfold, depending on the time of day and the activity level of the subject, typically in a sinusoidal pattern [44]. Therefore, diagnostic and therapeutic trials should take pains to conduct measurements of Aβ at similar time points and under similar activity levels to minimize variability. Biomarkers based on Aβ have been employed in numerous clinical trials of putative therapeutic compounds for AD. For example, the γ-secretase

c04

BLBK249-Cutler

110

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

inhibitor LY450139 from Eli Lily underwent dose optimization in clinical trials by comparing how various doses altered Aβ levels in plasma [45]. The Aβ-binding compound 3-amino-1-propanesulfonic acid (3APS) had its mechanism of action verified in clinical trials with AD patients by showing that the compound could reduce CSF Aβ levels over a 3-month period [46]. The therapeutic AD vaccine AN-1792, made of synthetic Aβ-42 peptide, demonstrated an ability to reduce Aβ burden in AD patient’s postmortem brain tissue before clinical trials were halted due to side effects [47]. As more AD therapeutic compounds are developed, Aβ biomarkers will increasingly be employed to validate their mechanism of action, select the appropriate dose, and determine initial parameters of efficacy.

Tau protein Tau is a microtubule-associated protein that stabilizes and promotes microtubule assembly through interactions with tubulin. The phosphorylation of tau results in the disruption of microtubules, and hyperphosphorylation of tau results in the self-assembly of tangles of paired filaments, which are involved in the pathogenesis of AD. The total amount of tau (t-tau) in the CSF is approximately 300% greater in AD patients compared with nondemented elderly subjects [48]. T-tau has a sensitivity and specificity between 80 and 90% [30] when comparing AD patients to healthy controls, but it provides a poor differential diagnosis with other neurodegenerative dementias, even in combination with Aβ [31]. A better tau biomarker is hyperphosphyorylated tau (p-tau). The tau protein can be hyperphosphorylated at threonine 231 (p-tau231P ) [49], threonine 181 (p-tau181P ) [50], and serine 199 (p-tau199P ) [51]. P-tau is consistently increased in the CSF of AD patients versus healthy controls [52, 53], and shows good sensitivity and specificity at distinguishing AD from healthy controls as well as other forms of dementia [54]. In addition, p-tau is a good predictor of AD progression in patients with mild cognitive impairment (MCI), both alone and in combination with Aβ [55, 56]. Because of its high negative predictive value, a normal value of p-tau rules out the presence of AD with almost 90% certainty [57]. Currently, flow-cytometry-based assays are being tested which allow for the determination of Aβ, t-tau, and p-tau levels in one sample using a relatively small amount of CSF. Preliminary results in multiple studies show these assays are highly predictive at identifying which patients with MCI will go on to develop AD [58–60]. One large German study recruited 223 MCI patients and used monoclonal antibodies coupled to fluorescent dyes to identify the amount of Aβ42 , t-tau, and p-tau181P in CSF samples [61]. Investigators were able to separate patients with MCI of Alzheimer’s disease type (MCIAD) from patients with MCI of other dementia (MCI-O) based on the levels of these biomarkers. Another study was able to distinguish what subjects with MCI would go on to develop AD within 2–3 years by looking at these markers [62].

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

111

␤-Secretase (BACE1) β-Secretase (also called BACE1) is one of the key enzymes responsible for the cleavage of the APP to form Aβ, and therefore is believed central to the pathogenesis of AD. The genetic knockout of BACE1 in mice has been shown to abolish brain Aβ production without altered behavioral or pathological phenotypes [63, 64]. BACE1 CSF levels appear to be able to distinguish AD from other forms of dementia, yet not Creutzfeldt–Jakob disease (CJD) [65]. BACE1 concentration and activity level appears to deviate from healthy controls in patients with MCI and AD, although the nature of this change remains disputed as studies have produced contradictory results. Several studies have found a significant increase in BACE1 concentration and activity in the CSF of MCI subjects compared with healthy controls [65–68], and also as an indicator of progression to AD [69]. Another study found increased BACE1 concentration and activity levels in the frontal cortex of postmortem patients with sporadic AD, which correlated with amyloid accumulation [70]. Contradictory to this, a recent study found a significant decline in age-adjusted CSF BACE1 activity levels for AD patients compared to healthy controls [71]. The exact nature of the BACE1 enzymatic activity in CSF also remains controversial; one study concluded that the full-length 70kDa protein produces activity [68], while another study claims it is the truncated form of the protein that is active [67]. In addition, a standard measure of BACE1 activity has been the hydrolysis of an internally quenched fluorescent peptide substrate; this method does not use a standard curve and has relatively low sensitivity and specificity [72]. Recently, a much more sensitive assay for BACE1 levels has been developed, which may resolve some of the discrepancies observed in enzyme activity level and concentration in AD patients versus healthy controls [67]. Numerous β-secretase inhibitors are currently being developed that have proven effective at inhibiting Aβ production in the brain of transgenic mouse models [73], and a reliable BACE1 biomarker will be needed in order to prove these compounds are working as intended during early clinical trials. While promising, more studies with BACE1 are needed to determine precisely how it changes with disease progression, what form of it is enzymatically active, and what kind of differential diagnosis it can provide in regard to AD. ␥-Secretase Another key enzyme in the processing of Aβ is called γ-secretase. This enzyme can cleave APP in multiple sites to generate a peptide from 39 to 42 amino acids long, with Aβ42 being most implicated in AD pathogenesis. Numerous γ-secretase inhibitors are currently in preclinical and clinical development. Potential side effects involving the GI tract, thymus, and spleen are common, possibly due to the additional involvement of γ-secretase in Notch processing [74, 75]. There has not been substantial development of γ-secretase biomarkers, partially due to the fact that complete downregulation of this

c04

BLBK249-Cutler

112

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

enzyme is impractical due to toxic side effects; partial inhibition currently is being pursued as a safer alternative [76, 77].

␣-Synuclein α-Synuclein (αSN) is a major component of Lewy bodies associated with Parkinson’s disease (PD) and dementia with Lewy bodies (DLB). αSN has also been found to be a constituent of AD amyloid plaques and it was identified as the precursor of the non-Aβ component of AD amyloid plaques (NACP) [78]. The relationship between Aβ and αSN remains unclear; some studies have found an association suggesting codependent aggregation [79], while other studies have found them to aggregate independently in plaques [80]. αSN has been detected in Lewy bodies in the amygdala and other brain structures of patients with both familiar and sporadic AD [81–83]. It has also been postulated to be involved in aberrant synapse formation [84, 85]. Phosphorylated αSN is a common feature of advanced AD and is believed to occur independently of tau phosphorylation [86]. αSN was originally believed to exist only as an intracellular protein because it lacks an endoplasmic reticulum (ER) targeting signal. This was found to be untrue, as numerous studies have detected it in the culture medium of neuronal cells, in normal human blood, and in CSF [87–90]. In addition, αSN CSF levels have been shown to be elevated in PD patients [91], and studies have shown that αSN can function as a biomarker for PD [92–94]. αSN was found to aggregate more extensively in AD CSF versus CSF from other neurologic disorders, indicating its potential use as an AD biomarker [95]. Further studies are needed to examine the predictive potential of αSN in tracking AD progression and drug efficacy. Amyloid precursor protein (APP) Amyloid precursor protein (APP) is cleaved to form Aβ, which makes up the main insoluble component of AD-associated senile plaques. Therefore, APP is believed to be a central component of AD pathogenesis. Platelets contain more than 95% of the circulating APP. In addition, they contain the enzymatic machinery necessary for APP metabolism, producing soluble APP fragments and Aβ peptides. The pattern of platelet APP fragments was found to be altered in patients with AD in comparison to healthy elderly controls [96], but not in patients with MCI [97] or in other forms of dementia [98]. This altered pattern involves a decrease in the amount of the higher (130-kDa) APP band compared to the lower (110-kDa) APP band on western blots in AD patients versus healthy controls [97]. Platelet APP differences track with AD progression and severity [99], show good specificity and sensitivity with AD diagnosis [100], and correlate with cognitive improvements induced by treatments such as cholinesterase inhibitors [101, 102]. In addition, platelet APP ratios have been shown to predict conversion from MCI to AD [103]. This measure has the advantage of being minimally invasive, requiring only a blood test. Platelet APP ratios hold a lot of promise as potential biomarkers for AD

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

113

progression and as markers for the efficacy of new AD therapeutics, yet a better understanding of the relationship between peripheral APP metabolism and neuronal AD pathogenesis is required. Studies have shown that the soluble form of APP (sAPP) is also reduced in the CSF of patients with sporadic AD, which usually occurs later in life [104, 105] and in patients carrying the Swedish APP mutation [106, 107]. Additionally, decreased levels of sAPP have been shown to correlate with reduced performance on clinical rating scales [106]. Reduced levels of sAPP in AD are consistent with evidence suggesting that sAPP plays a protective role in the brain [108]. Reductions in sAPP in AD patients could also reflect an increased cleavage of APP favoring the amyloidogenic pathway, resulting in greater production of Aβ.

Isoprostanes Isoprostanes are the novel markers of oxidative injury thought to be important in the pathogenesis of AD. Isoprostanes are prostaglandin-like compounds formed from the free-radical-catalyzed peroxidation of essential fatty acids such as arachidonic acid (AA). Numerous studies have shown evidence of oxidative damage in AD [109–111], and specific isoprostanes have been shown to be elevated in the urine, blood, and CSF of patients with AD [40, 112, 113]. These values correlate with memory impairments and CSF tau levels, suggesting isoprostanes could be a reliable biomarker for AD. They are also able to distinguish between patients with MCI and healthy controls [114]. More studies are needed with AD and MCI patients in order to validate this marker. Sulfatides Sulfatides are a class of sulfated lipids synthesized primarily in the oligodendrocytes in the CNS and found in trace amount in other tissues. Studies have shown that a substantial and specific depletion of sulfatide is present in neural tissue from patients at the earliest stages of AD [115]. This deficiency was not observed in patients with PD or dementia with Lewy bodies, suggesting that sulfatide depletion may be specific for AD [116]. Further research is needed to determine how this depletion is related to the pathogenesis of the disorder. Inflammatory biomarkers There is a wealth of evidence implicating inflammatory processes in the pathogenesis of AD. Many neuroinflammatory mediators have been found to be upregulated in affected brain regions of AD patients, including prostaglandins, complement components, cytokines, chemokines, proteases, protease inhibitors, and free radicals, among others [117]. Polymorphisms of the inflammatory cytokines interleukin-1 (IL-1), interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α) have been associated with an increased risk of developing AD [118–120]. IL-1 and TNF-α blood levels have also been shown to be elevated in patients with AD [121, 122], and higher levels of PBMC production of IL-1 or TNF-α were associated with an increased risk of developing

c04

BLBK249-Cutler

114

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

AD [123]. There is even evidence that these inflammatory mechanisms occur before fibrillar amyloid deposits are present and before the onset of dementia, making them good candidates for predictive biomarkers as well as targets for preventative therapy [124, 125]. Results from clinical studies using nonsteroidal anti-inflammatory drugs (NSAIDs) to treat AD have been mixed. Two pilot clinical trials with full therapeutic doses of indomethacin or diclofenac showed moderate protection against cognitive decline in AD patients versus healthy controls [126, 127], while one trial with a subtherapeutic dose of naproxen failed [128]. In addition, not all studies have shown consistent results with inflammatory markers predicting AD risk; some studies have shown cytokine levels to be elevated [129, 130], while others have shown them to be unaltered [131] or even decreased in AD patients [132, 133]. Much like other AD biomarkers, the lingering discrepancies in inflammatory biomarkers need to be resolved before their predictive potential can be realized.

Electroencephalography (EEG) Electroencephalography (EEG) is the measurement of the electrical activity produced by the brain, recorded from electrodes placed in various locations on the scalp. Quantitative EEG (QEEG) is a statistical method of analyzing EEG activity for abnormalities by converting it to a digital signal and comparing this signal to a database of healthy individuals without neurological disorders. QEEG evaluations in normal elderly were found to be highly sensitive and specific predictors for future cognitive decline to MCI or dementia over a 7-year period, with future decliners characterized by increased ␪ power (a characteristic EEG waveform), slowing of mean frequency, and changes in covariance [134]. In addition, deterioration in normal EEG patterns has been found to predict which patients with MCI would progress to AD within 25 months [135]. The slowing of EEG activity in AD patients has been observed by numerous studies, and the synchronization between EEG waves is impaired in both AD and MCI patients [136]. EEG patterns in AD patients are characterized by higher ␦ and ␪, and lower α and β powers than normal elderly subjects, and can be discriminated with an average sensitivity of around 75% [137–140]. Improved data-analysis techniques, such as artificial neural networks (ANNs) and learning algorithms, have allowed for better resolution of EEG pattern differences: one study was able to discriminate AD patients from non-AD patients with a sensitivity of 80% and a specificity of 100% [141]; another was able to discriminate between healthy subjects and MCI patients with 91–93% accuracy [142]. EEG may also be able to predict AD long before the onset of any cognitive symptoms; e.g., healthy middle-aged subjects carrying the ApoE ε4 allele (allele of the epsilon4 [ε4] isoform of the Apolipoprotein E [ApoE] gene), a common genetic risk factor for AD, showed a significantly greater decrease in α power than in healthy middle-aged ε4 noncarriers under hyperventilating conditions [143].

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

115

EEG is noninvasive, which would make it an ideal candidate biomarker for AD. In addition, unlike biomarkers derived from the serum or CSF, EEG changes represent direct changes in brain activity levels. EEG has been used as a biomarker for efficacy and toxicity during the early clinical trials of numerous compounds designed to treat AD. For example, the EEG of AD patients taking the cholinesterase inhibitor donepezil for 8 weeks showed a trend toward renormalization of α, ␦, and ␪ waves that had been altered by disease, and this renormalization was coupled with increased frontal lobe activity on fMRI scans [144]. In addition, EEG was able to classify mild AD patients (Mini Mental State Examination [MMSE] range 17–24) taking donepezil for 1 year into “responders” versus “nonresponders” based on the partial restoration of aberrant EEG rhythms, which correlated with stable MMSE scores during treatment [145]. Interestingly, this study was able to use EEG to classify “responders” versus “non-responders” even before administration of the drug, despite these two groups having no obvious behavioral or cognitive differences. Likewise, the acetylcholine esterase inhibitor rivastigmine produced an EEG ␪ power decrease in AD patients within a week or two of administration, which predicted the therapeutic efficacy of the drug [146]. The changes in EEG patterns were different for each drug; donepezil affected α, ␦, and ␪ sources while rivastigmine affected only ␪ sources. These effects are not seen with every AD therapeutic drug class; e.g., in the same study the AD therapeutic memantine, a noncompetitive NMDA receptor antagonist, did not produce any EEG alterations in AD patients. While imperfect, EEG will continue to be a useful biomarker for predicting efficacy as more novel AD therapeutics are brought to clinical trials.

Salivary amylase Documented cholinergic deficits in AD patients [147, 148] have led to treatments that attempt to boost cholinergic activity in the brain. While AChEIs are the most recognized and widely used drug class for this therapeutic category, muscarinic acetylcholine receptor (mAChR) agonists are also being developed to boost cholinergic activity, and many new compounds are currently in preclinical and clinical trials. These agonists are designed to target the M1 muscarinic receptor because of its major role in cognitive processes relevant to AD [149, 150], its predominance in the cerebral cortex and hippocampus [151], and its preserved integrity during the pathogenesis of the disorder [152, 153]. Because the cholinergic system is inhibited during AD progression, typical AChE inhibitors may lose efficacy over time, and muscarinic agonists could become viable replacement therapies for middle and late-stage AD patients originally treated with AChE inhibitors. A major problem with this drug class has been the nonspecificity of M1 agonists; negative side effects were often reported due to the activation of M3 receptors in the intestine, bladder, and lungs [154]. Novel M1 agonists show increased specificity, lower side effects, and demonstrate efficacy in preclinical trials [155–158]. Biomarkers are needed to

c04

BLBK249-Cutler

116

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

assess both the efficacy and safety of these compounds in early clinical trials. Safety biomarkers are especially needed due to the propensity of these compounds to illicit nonspecific M3 receptor activation in the periphery, causing significant intestinal side effects. Because peripheral M3 receptor activation increases intestinal and salivary secretions, salivary amylase, a component of these secretions, could be used as an early biomarker for the nonspecific side effects of M1 agonists. Indeed, salivary amylase was found to be increased in rats exposed to the M1 agonist xanomeline tartrate in preclinical trials (Eli Lily, unpublished data), and a trend toward increasing salivary amylase levels was also reported in early clinical trials of this compound conducted by the company [159]. Based on the subsequent clinical trials of xanomeline with high discontinuation rates, salivary amylase appears in retrospect to be predictive of the negative side effects of this compound [160]. Therefore, it could be a useful biomarker in predicting peripheral effects of novel M1 agonists, although more clinical trials are needed to verify its utility.

CD-69 The expression of the T-lymphocyte cell surface marker CD-69 on peripheral blood lymphocytes following a mitogenic stimulus was found to differentiate AD from normal controls and other dementias [161]. In addition, CD-69 positive monocyte/macrophages were elevated in patients with AD in comparison to normal controls [162]. The elevated expression of CD-69 is believed to reflect systemic immune system upregulation resulting from cellular damage incurred by the pathogenic progression of AD. While these initial findings are promising, this biomarker requires significantly more validation before it can be reliably used with AD clinical drug development. For example, it remains unclear whether changes in CD-69 levels correlate with the efficacy of AD therapeutics, or how peripheral levels of immune cell upregulation correlate with CNS damage incurred by AD progression. It is also unclear how early this marker shows up in the pathogenesis of AD. This is important because if peripheral CD-69 upregulation only occurs in the later, more debilitating stages of the disease, it will not be useful for judging the efficacy of early acting therapeutics. Genetic markers of AD There are many genetic markers which may predispose individuals to acquiring AD, modify the onset and progression of the disease, and even determine the patient’s response to therapeutics. These genetic markers may be able to guide the course of treatment and categorize AD drugs into efficacious groups based on patient-specific genotypes. In addition, the proteins that these genes code for may change expression levels in direct response to AD drug administration. Studies have found that the response of therapeutics in AD is genotype-specific, with more than 200 genes associated with AD pathogenesis and neurodegeneration, and approximately 1400 genes accounting for 20–95% of the variability in drug disposition and metabolism [163].

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

117

For example, the cholinesterase metabolizing gene CYP2D6 exists in various forms, including extensive metabolizers (EM), intermediate metabolizers (IM), poor metabolizers (PM), and ultrarapid metabolizers (UM). AD patients with EM and IM versions of CYP2D6 show the best response to therapies, including cholinesterase inhibitors, while patients with PM and UM versions of the gene show the poorest response [164]. In addition, carrying both alleles of the ApoE gene, which codes for a protein that catabolizes triglyceride-rich lipoproteins, increases the risk of developing AD in comparison to elderly adults without an e4 allele [165]. In addition, the ApoE e4 gene may affect patient response to certain treatments. For example, AD patients who carry the ApoE e4 allele respond more favorably to donepezil [166] or atorvastatin [167] treatment than noncarriers, while ApoE e4 carriers showed poorer response to rosiglitazone treatment than noncarriers [168]. This effect does not hold true for all therapeutics; a study examining patient response to rivastigmine treatment found no difference in response between ApoE e4 carriers or noncarriers [169]. The ApoE protein can also change in response to drug therapy. For example, the plasma expression levels of ApoE protein changed with rosiglitazone administration that correlated with therapeutic efficacy of the drug, as measured by the Alzheimer’s Disease Assessment Scale-Cognitive (ADAS-Cog) [170].

Anxiety disorder The following section covers the major and most promising biomarkers for anxiety disorder, which are also summarized in Table 4.4.

Table 4.4 Biomarkers for anxiety disorder

Cortisol Salivary amylase Corticotrophin-releasing factor (CRF) Glyoxalase-1 (GLX1) Catecholamines Epinephrine (EP) Norepinephrine (NE) Noradrenaline (NA) 3,4-dihydroxyphenylacetic acid (DOPAC) Challenge paradigms Lactate infusions CO2 administrations CCK-4 peptide Saccadic peak velocity (SPV) Brain imaging PET fMRI MRI Neurotransmitter receptor binding studies

c04

BLBK249-Cutler

118

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

Cortisol Individuals with generalized anxiety disorder (GAD) have been shown to have elevated basal salivary cortisol levels in comparison to nonanxious subjects, and severity of GAD correlates with increasing cortisol levels [171]. In addition, the anxiolytic drug diazepam reduces plasma cortisol both acutely and during chronic treatment [172]. There was a significant reduction in urinary cortisol levels in anxious surgery patients receiving diazepam versus placebo [173]. Cognitive therapy of GAD also reduces plasma cortisol levels [174]. Cortisol levels have been used as a marker for judging the efficacy of anxiolytics. A drawback with the use of cortisol as a biomarker for GAD is that cortisol levels vary based on numerous factors. For example, anxious females produce lower levels of salivary cortisol than anxious males [175]. In addition, cortisol levels fluctuate diurnally and spike in the morning [176]. Individual stressors can modulate salivary cortisol levels [177], and even chewing gum can reduce salivary cortisol [178]. This variability must be accounted for during efficacy measurements of novel anxiolytics based on cortisol levels. Combining several measurements of cortisol bracketing a particular time point (i.e., −10, 0, and 10 minutes) is one possible strategy to account for this inherent variability. The high variability, perhaps due to susceptibility to varying levels of stress, makes it a less-than-perfect marker. Salivary amylase The State Anxiety Inventory score (STAI-s) has been significantly correlated to salivary α-amylase levels [179]. In addition, salivary amylase has been strongly associated with psychological stress levels using the Trier Social Stress Test (TSST) [180]. Salivary amylase is not closely related to other stress biomarkers, making it a potentially unique measure of anxiety. It was found to increase more significantly and react more rapidly to psychological stress than cortisol, suggesting that it is also a more responsive marker of anxiety [181]. The specificity of both amylase and cortisol as biomarkers of GAD and other anxiety-associated disorders remains questionable since both respond to any form of psychological stress, which is not necessarily an indicator of GAD. Corticotrophin-releasing factor (CRF) Several studies have shown that elevated levels of corticotrophin-releasing factor (CRF; also known as corticotrophin-releasing hormone, or CRH) are linked to anxiety-related disorders in mice [182, 183] as well as nonhuman primates [183]. They have also been found to be elevated in the CSF of patients with posttraumatic stress disorder (PTSD) [184] and female victims of abuse who show signs of anxiety [185]. Several CRH-1 receptor antagonists are currently undergoing development to treat anxiety disorders, and have shown promising results [186, 187]. A problem with this hormone is that it is also elevated in patients with depressive disorders, demonstrating poor

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

119

specificity [188]. In addition, some studies show no difference in CRF levels in patients with anxiety disorders [189, 190], showing this molecule requires further study before it can be considered for a biomarker of anxiolytic efficacy.

Glyoxalase-1 Proteomic analysis of red and white blood cells in mice selectively bred for anxiety-related phenotypes demonstrates that low-anxiety mice have higher levels of glyoxalase-1 (GLX1) than high-anxiety mice [191]. This enzyme has been demonstrated to be a robust, reliable biomarker for the nonanxious versus anxious phenotypes in mice, yet testing in humans remains to be carried out [192]. While this enzyme shows promise as a biomarker of anxiety, GLX1 has also been shown to be upregulated in mice models of AD and depression [193, 194]. It remains to be determined whether it is a risk marker or risk factor for anxiety-associated phenotypes [195]. Catecholamines Patients with generalized anxiety disorders have higher plasma catecholamine levels than normal controls, and increasing anxiety levels correlate directly with greater plasma catecholamine concentrations [196, 197]. PTSD is associated with higher urine catecholamine levels, including epinephrine (EP) and norepinephrine (NE) [198]. Catecholamine levels are affected differently by different anxiolytics in animals. The anxiolytic buspirone increases plasma catecholamine levels in rats [199] while diazepam and adinazolam significantly reduce plasma catecholamines [200, 201]. Catecholamine levels are also affected by aerobic fitness, psychological stress, and blood sugar levels [202]. Because of this variability, more validation is necessary before catecholamines can become a reliable biomarker for anxiolytic efficacy. Preoperative patients receiving the anxiolytic diazepam had smaller CSF concentrations of noradrenaline (NA) and of the dopamine metabolite, 3,4-dihydroxyphenylacetic acid (DOPAC) than untreated controls [203], although a similar study found that plasma levels but not CSF levels of these catecholamines changed between diazepam treatment and control groups [204]. A later study found CSF and plasma catecholamine measurements to be of little use in determining the clinical effects of different anxiolytics [205]. The utility of using plasma and CSF measurements of catecholamines to determine anxiolytic efficacy remains unclear; further investigation is required to determine this. Challenge paradigms Patients at risk for panic disorders may have deficits in the regulatory mechanisms of the hypothalamic-pituitary-axis associated with an abnormal response to stress. These patients appear to be more sensitive than normal controls regarding the number and intensity of symptoms that they develop following a panic challenge paradigm [206]. The ability of lactate infusions to induce panic attacks in patients with panic disorder has been extensively replicated and verified [207, 208]. Patients prone to panic attacks are

c04

BLBK249-Cutler

120

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

significantly more susceptible to the effects of lactate infusion than healthy controls or even patients with other psychiatric disorders without panic attacks [209]. Lactate infusion serves as a viable biomarker for evaluating the ability of anxiolytics to reduce panic attacks. For example, the administration of the anxiolytic diazepam is able to attenuate this effect [210]. In addition, novel anxiolytics have been evaluated based on their ability to prevent panic attacks induced by lactate [211, 212]. Despite its utility, the sensitivity and specificity of the test for patients with various subtypes of panic disorder requires further characterization. Additionally, lactate infusion is a much less predictive biomarker for anxiety disorders without panic attacks. Like lactate infusion, CO2 administration produces dose-related increases in anxiety, somatic symptoms, vital signs, and plasma cortisol levels. In patients experiencing panic attacks, the frequency of attacks and the increases in anxiety and somatic symptoms were significantly higher than those induced by CO2 in healthy subjects [213]. While used frequently in the past, this anxiety biomarker has fallen out of favor in recent times. Cholecystokinin (CCK) is a peptide hormone of the GI system involved in stimulating the digestion of fat and protein. Short CCK peptides, such as cholecystokinin tetrapeptide (CCK-4), have been found to elicit panic attacks in humans and anxiety-associated characteristics in animal models [214, 215]. Many studies have examined the ability of novel anxiolytics and other compounds with suspected anxiolytic activity to attenuate CCK-4-induced panic attacks in both healthy subjects and panic disorder patients. The selective serotonin reuptake inhibitor (SSRI) citalopram was shown to attenuate CCK-4-induced panic attacks in patients with panic disorder [216]. In addition, the benzodiazepine alprazolam [217], the γ-aminobutyric acid (GABA) catabolism inhibitor vigabatrin [218], the anticonvulsant tiagabine [219], and the atrial natriuretic peptide (ANP) [220] all reduced CCK-4-induced panic attacks in healthy subjects. On the other hand, the SSRI escitalopram failed to attenuate CCK-4-induced panic attacks in healthy subjects, and actually increased panic symptoms in some subjects [221]. Panic levels in this challenge paradigm are typically measured according to standard scale questionnaires, such as the Acute Panic Inventory (API) and the Panic Symptom Scale (PSS), which can be highly variable due to a lack of standardized methodology and procedures for testing novel anxiolytics [222]. In addition, gender differences may produce different results for the same anxiolytic compound in this model [223]. Therefore, results from CCK-4 challenge paradigms should be verified with other challenge paradigms or biomarkers.

Saccadic peak velocity (SPV) Saccadic peak velocity (SPV) is the highest velocity reached during a fast movement, or “saccade” of the eye. All benzodiazepines caused an impairment in SPV at therapeutic doses [224]. This measure correlates with the onset and duration of drug efficacy, is noninvasive, and is relatively easy to carry out [225]. A major drawback with SPV as a biomarker is that it does not work as

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

121

consistently with nonbenzodiazepine anxiolytics. It remains unclear whether this effect is due to the sedative or anxiolytic properties of benzodiazepines.

Brain imaging The PET scans of brain glucose metabolism show that patients with generalized anxiety disorder tend to have increased metabolic rates, particularly in the occipital temporal lobes, the frontal lobes, and the cerebellum. They also show decreased metabolic activity in the area of the basal ganglion, the cingulate gyrus, the temporal lobes, the amygdala, and the hippocampus. When benzodiazepines (anxiolytics) were administered to these patients, cerebral glucose metabolism over the cortical surface diminished markedly [226]. In a recent study, fMRI was used to measure the response to fearful faces in the amygdala and rostral anterior cingulate cortex (rACC) of patients with generalized anxiety disorder before and after treatment with venlafaxine. The magnitude of treatment response was predicted by greater pretreatment reactivity to fearful faces in rACC and lesser reactivity in the amygdala [227]. In addition, the fMRI measurements of brain activity, while listening to worrying statements, showed reduced activity in generalized anxiety disorder patients after treatment with citalopram [228]. Brain morphology could also serve as an anxiety-related marker. MRI studies have shown a smaller hippocampal volume in drug-na¨ıve adults suffering from chronic PTSD [229]. Neurotransmitter receptor binding studies have also shown alterations in anxiety-related disorders. The benzodiazepine antagonist flumazenil showed decreased binding in benzodiazepine naive, drug-free patients with panic disorder, highlighting a binding defect in GABAA receptors that could potentially be ameliorated with anxiolytics [230].

Depression The following section covers the major and most promising biomarkers for depression, which are also summarized in Table 4.5.

G proteins The Gs α subunit (or Gs protein) is a heterotrimeric G protein subunit which activates adenylate cyclase. Numerous studies have looked at the association of Gs α with depression. A recent study found that depression localizes Gs α to lipid rafts in the cell membrane, where it is less likely to couple to adenylyl cyclase [231]. In addition, chronic treatment with tricyclic or SSRI antidepressants causes Gs α to migrate from a Triton X-100 (TX-100)-insoluble membrane domain (lipid raft) to a TX-100-soluble nonraft membrane domain, upregulating its signaling. In the future, a simple blood test could allow raft localization of Gs α to serve as a biomarker for depression and antidepressant responsiveness, although more studies of this protein are needed. In addition, the polymorphisms of the β3 subunit of G protein (GNB3) gene have been associated with the symptom severity and treatment response of

c04

BLBK249-Cutler

122

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

G proteins Gs α subunit (Gs α) β3 subunit of G protein (GNB3) gene IL-6 Cortisol Salivary cortisol Plasma cortisol Corticotrophin-releasing hormone (CRH) CRH CRH receptors Dexamethasone (dex) Dexamethasone (dex) suppression test (DST) Dex/CRH test Neurotransmitters, metabolites, transporters Norepinephrine (NE) Dihydroxyphenylglycol (DHPG) 3-Methoxy-4-hydroxyphenylglycol (MHPG) NE transporter (NET) Serotonin (5-HT) 5-Hydroxyindoleacetic acid (5-HIAA) Dopamine (DA) Homovanillic acid (HVA)

Table 4.5 Biomarkers for depression

major depressive disorders (MDDs) [232, 233]. Further investigation is required to determine how prevalent this polymorphism is in the depressed patient population, as well as what antidepressants affect this gene. For example, the antidepressant mirtazapine failed to show a change in the GNB3 polymorphism in MDD patients [234]. Studies have been unable to find a similar association between polymorphisms in the Gs α gene and depression [235].

IL-6 Patients with major depression have been found to exhibit all the hallmark symptoms of inflammation, including elevations in the levels of inflammatory cytokines and their receptors in both peripheral blood and CSF [236, 237]. Specifically, depression has been linked to the elevated levels of the proinflammatory cytokine interleukin-6 (IL-6) in a variety of populations. IL-6 has been linked to depression in continuous ambulatory peritoneal dialysis (CAPD) patients [238], in advanced cancer patients [239], in older people [240], in MDD patients, and in healthy controls [241]. It remains unclear whether IL-6 is a causative factor or a symptom of depression, but studies in animals suggest the former; IL-6 knockout mice exhibit resistance to depression-like symptoms induced by stressful conditions [242]. IL-6 levels show mixed results in regard to antidepressant treatment. IL6 levels are reduced by SSRI treatment in patients with major depression [243]. Additionally, tricyclic antidepressants inhibit IL-6 production in human

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

123

blood monocytes in culture [244]. In contrast, another study showed IL-6 levels to be elevated by imipramine, venlafaxine, 5-hydroxytryptophan (5-HTP), or fluoxetine treatment in patients with treatment-resistant depression (TRD) [245]. This discrepancy will need to be resolved before IL-6 can be used as a biomarker of antidepressant efficacy.

Cortisol Cortisol is a stress-related hormone secreted from the adrenal glands that has been associated with clinical depression. The nature of this association, as well as the directional shift in cortisol in depressed patients versus healthy controls, remains unclear. One study found that around 50% of patients with MDD hypersecrete cortisol [246]. High cortisol levels brought on by stressful situations may lower brain 5-HT function and lead to a depressive state [247], or they may impair feedback inhibition and cause HPA axis hyperactivity [248]. In contrast, lower salivary cortisol levels were found in suicide attempters compared to nonsuicidal psychiatric patients, and linked suicidal behavior with low salivary cortisol levels [249]. Because not all patients with MDD have altered cortisol levels, deviations in cortisol may be a risk factor but not a requirement for depressive disorders [250]. Antidepressants also appear to affect the availability of cortisol in the brain and the level of cortisol in the plasma, although this effect varies, depending on the class of antidepressant administered [251, 252]. One study found that the SSRI fluoxetine increased plasma cortisol levels in patients with major depression, while tricyclic antidepressants did not alter plasma cortisol levels [253]. Another study found that the tricyclic antidepressant clomipramine increased the responsiveness of cortisol to the 5-HT precursor, 5-hydroxytryptophan (5-HTP) [254]. Yet another study showed that the tetracyclic antidepressant mirtazapine reduced salivary cortisol levels in depressed patients [255]. While cortisol shows some promise as a marker for depressive disorders and the effects of novel antidepressants, the fact that its levels are not altered in all depressed patients nor by all classes of antidepressants marginalizes its clinical utility, at least in the near future. Corticotropin-releasing hormone (CRH) and dexamethasone (dex) Depressed patients have elevations of basal serum cortisol levels and CRH in their CSF [256]. In addition, postmortem studies show elevated levels of CSF CRH and a decreased density of CRH receptors in the frontal cortex of suicide victims [257]. Depressed patients have an exaggerated adrenocorticotropin hormone (ACTH) and cortisol response to CRH after pretreatment with dex. In addition, CRH hypersecretion may produce hypothalamus–pituitary–adrenocortical (HPA) hyperactivity and possibly impaired corticosteroid receptor signaling. HPA alterations are consistently observed in patients with acute affective disorders, including severe depression, and these systems typically normalize after successful treatment [251, 258, 259]. The tricyclic antidepressant imipramine and the SSRI fluoxetine

c04

BLBK249-Cutler

124

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

both inhibited CRH gene promoter activity, although the selective serotonin reuptake enhancer (SSRE) tianeptine did not [260]. Because CRH is believed to play a central role in depression, new antidepressants are being developed which directly antagonize CRH receptors [187,261,262]. While this new class of antidepressants is still undergoing clinical development, the presumed central role of CRH in the pathogenesis of depression bolsters its case as a potential biomarker for this disorder. The dex suppression test (DST), an established measure for detecting functional alterations in the HPA system, has been tested as a surrogate marker to predict the clinical course in depressed patients. It has also been utilized to predict the treatment response of novel antidepressants [263, 264]. The DST was combined with the CRH stimulation test to improve its sensitivity and performance [265, 266]. This test is sensitive to impaired glucocorticoid receptor signaling and increased secretions of CRH and vasopressin [267, 268]. The combined dex/CRH test has been shown to discriminate normal subjects from patients suffering from a major depressive episode, from bipolar disorder, and from chronic depression [269–271]. The dex/CRH test shows normalization of the neuroendocrine response after the administration of tricyclic antidepressants [272], SSRIs [273], SSREs [274], and noradrenergic and specific serotonergic antidepressants (NaSSAs) [275]. Despite the wide range of antidepressants the dex/CRH test responds to, improvements in this test did not always correspond to the clearing of depressive symptoms [276]. Patients with chronic depression do not always show dysregulation of the HPA system, limiting the clinical population sensitive to this biomarker. In addition, CRH varies differently, depending on the category of depression the patient is afflicted with; melancholic depression typically produces hyperactive central CRH systems, while atypical depression is characterized by hypoactive central CRH systems [277]. Finally, the specificity of the dex/CRH test is limited—the test responds to other disorders besides depression, including acute manic condition, anxiety disorders, schizophrenia, Cushing’s disease, and multiple sclerosis [278–282].

Norepinephrine, DHPG, and MHPG It has long been hypothesized that a deficiency in the neurotransmitter norepinephrine (NE) is implicated in the pathogenesis of depression, and there is a substantial body of evidence which supports this association [283]. Monoamine oxidase inhibitors (MAOIs) and serotonin–norepinephrine reuptake inhibitors (SNRIs) have a successful record in treating depression by facilitating and prolonging the release of neurotransmitters such as NE [284, 285]. In addition, drugs which deplete NE, either directly or indirectly, have been shown to induce depression or depression-like symptoms in humans [286– 289]. The NE transporter (NET) transports NE out of the synapse. Its altered function has been implicated in the pathogenesis of depression [290, 291]. NET is reduced in the CSF of deceased suicide victims, possibly due to a lack of availability of their substrate, NE [292].

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

125

Dihydroxyphenylglycol (DHPG) is the first metabolite of NE. It can be used as a measure of NE turnover. DHPG is found in CSF, plasma, and urine. Urinary DHPG excretion was significantly elevated in patients suffering from mixed anxiety and depressive disorder (MADD) who showed nonsuppression on the DST [293]. In addition, healthy volunteers who administered the antidepressant duloxetine showed a significant reduction in the DHPG/NE plasma and urine ratio after 2 weeks of treatment. This suggests that the degree of NET blockade can be assessed with the plasma or urine DHPG/NE ratio [294]. Many potent antidepressants block NET [295] or inhibit NET through downregulation [296, 297]. A portfolio of biomarkers is useful for the assessment of NET inhibition because there are substantial differences in the sensitivity with which the different downstream biomarkers are affected. For example, the DHPG/NE ratio is a more sensitive biomarker than the tyramine pressor test (which assesses NE reuptake capacity via the uptake of tyramine), and these biomarkers are more reliable when used in conjunction [298]. The assessment of such biomarkers in CSF, plasma, and urine during treatment with the potent NET reuptake inhibitor atomoxetine was more recently published [299]. The several-fold higher concentrations of DHPG in CSF indicate that this primary metabolite of NE originates in the central compartment. There is a statistically significant nonlinear relationship with the DHPG in CSF and plasma. The changes observed in the biomarker DHPG indicate a longlasting central effect of atomoxetine on the NE transporter. MAOIs are a class of antidepressant drugs that work by inhibiting the activity of monoamine oxidase (MAO), thus preventing the breakdown of monoamine neurotransmitters and increasing their availability. While they typically were considered a last resort behind SSRIs and tricyclic antidepressants due to their potent side effects, better formulations and safer delivery mechanisms have recently caused a resurgence in their use as a first-line depression treatment [300, 301]. MAO inhibition causes a greater than 98% reduction in plasma DHPG levels [302]. This plasma DHPG is derived almost exclusively from neuronal sources [303]. In addition, patients with genetically impaired function of the NET transporter have low plasma levels of DHPG [304], and patients with genetic deficiencies in MAO-A and MAO-B (the two functional forms of the MAO enzyme) had nearly undetectable plasma levels of DPHG [305]. This suggests that plasma DHPG could be a potent measure of efficacy for both novel MAOIs and antidepressants that affect NET function. The MAO-A inhibitor befloxatone demonstrated a dose-dependent reduction in plasma levels of DHPG, showing that such measurements are feasible [306]. In addition, the MAOI selegiline produced a reduction in CSF levels of DHPG, suggesting CSF DHPG could also track with MAOI efficacy [307]. A limitation of DHPG as a depression biomarker is that it is very sensitive to stress [308, 309] and the effects of anxiolytics [205] to catecholamineproducing tumors [310, 311]. Therefore, depression biomarkers with various sensitivity profiles should be used in conjunction to rule out such potentially confounding sources of variability.

c04

BLBK249-Cutler

126

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

3-Methoxy-4-hydroxyphenylglycol (MHPG), the second metabolite of NE degradation, is often reduced in the urine of patients suffering from bipolar depression in comparison to normal subjects [312]. This consistent pattern does not hold up for patients suffering from unipolar depression [313]. Patients suffering from major depression experienced significant reductions in CSF levels of MHPG after long-term administration of SSRIs [314, 315] or tricyclic antidepressants [316, 317]. Most MHPG is derived from DHPG in peripheral sympathetic nerves—not from the brain. MHPG sulfate reflects NE turnover in the GI tract and does not reflect turnover in the brain [303]. Therefore, DHPG should be used over MHPG when looking for drug-induced CSF changes in NE metabolites.

5-HIAA and HVA Abnormalities in serotonin levels have also been implicated in depression. 5-Hydroxyindoleacetic acid (5-HIAA) is the main metabolite of serotonin in the human body. Numerous studies have found that 5-HIAA concentration is reduced in the CSF of depressed patients versus healthy controls [318–320]. 5-HIAA-level abnormities also correlate with suicide risk in depressed patients [321]. Low 5-HIAA CSF levels were able to predict future early suicide in male, high-risk suicide attempters, and this predictive ability was better than the Beck Suicide Intent Scale (SIS) or the Beck Hopelessness Scale (BHS) [322]. CSF levels of 5-HIAA were reduced in depressed patients after chronic administration of tricyclic antidepressants [316, 317] and SSRIs [314, 315, 323– 325], but increased with acute administration of SSRIs [326]. The levels of 5HIAA were also reduced in rats administered with a triple-reuptake inhibitor [327]. One possible reason for reduced 5-HIAA and MHPG levels is inhibition of NA and 5-HT breakdown, caused by the increased synaptic availability of the transmitters during antidepressant treatment. While some studies show antidepressant-induced reductions in 5-HIAA levels, not all studies agree with these findings, and some have shown variable results in 5-HIAA levels with antidepressant treatment [328, 329]. In addition, animal-based models sometimes show conflicting results with human studies. For example, the administration of tricyclic antidepressants in rats did not alter 5-HIAA CSF levels [330], and long-term administration of the SSRI trazodone in rats produced significant increases in CSF levels of 5-HIAA [331]. A study found that depressed patients receiving SSRIs, tricyclic antidepressants, and MAOIs for an average of 15 weeks experienced reductions in CSF levels of 5-HIAA, but long-term administration (greater than 30 weeks) returned CSF levels of 5-HIAA back to pretreatment levels [332]. Therefore, caution should be taken when extrapolating the effect of antidepressants between studies with different time periods of drug administration. A disadvantage of using 5-HIAA as a biomarker for depression is that its levels are reduced in patients with suicidal thoughts and behaviors outside the context of depression [333]. Reduced serotonin levels may result in weakened control of impulsive behaviors in general [334]. Therefore, suicidal

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

127

behavior and impulsiveness may need to be accounted for in any study using 5-HIAA as a biomarker for the efficacy of antidepressants. Homovanillic acid (HVA) is a major catecholamine metabolite associated with dopamine levels in the brain. HVA/5-HIAA ratios have been shown to be reduced in depressed patients versus normal controls and have also been shown to be the markers of suicidal intent [335–337]. The low CSF levels of HVA have been shown to be the biomarkers of suicide intent, although not necessarily a marker of nonsuicidal depression [338, 339]. Severely depressed patients administered with SSRIs showed reductions in CSF HVA levels, although these reductions were less pronounced than 5-HIAA or MHPG reductions [314]. The treatment of depressed patients with tricyclic antidepressants did not produce reductions in CSF HVA [340, 341], suggesting that clinical studies using HVA alone are insufficient for evaluating a novel antidepressant’s efficacy.

Schizophrenia There are currently no validated biomarkers of schizophrenia that both establish diagnosis and reliably predict response to treatment for all classes of antipsychotics, but there are many that are being developed which show a great deal of promise. These have been reviewed in the following section and summarized in Table 4.6.

Dopamine There is a large body of evidence implicating dopamine dysfunction in the pathogenesis of schizophrenia. Amphetamines increase the levels of dopamine in the brain and can cause psychosis which is similar to the positive symptoms of schizophrenia [342, 343]. Studies have found a complex dysregulation of dopamine in the brains of schizophrenia patients, with an overall decrease in dopamine signaling in the prefrontal cortex (PFC) [344] and excess subcortical dopamine [345]. Functional neuroimaging studies have demonstrated an increase in amphetamine-triggered dopamine release in the brains of schizophrenia patients [346–348]. In addition, there is an excess release of dopamine from the striatum in individuals at high risk of psychosis or experiencing their first psychotic episode [349]. This is possibly due to a heightened dopaminergic transmission state that may be a fundamental characteristic of disease pathology [350]. The importance of dopamine in schizophrenia is evidenced by the fact that the D2 dopamine receptor remains the common target for nearly all typical and atypical antipsychotics on the market today [351]. Studies have shown that the plasma levels of the dopaminergic metabolite HVA correlate with the efficacy of neuroleptics in schizophrenic patients, particularly those that act as DA receptor blockers [352–356]. For example, the atypical antipsychotic risperidone reduced plasma HVA levels in schizophrenia patients in correlation with improvement on the Positive and Negative Syndrome Scale (PANSS), a measure of schizophrenia symptom severity [354].

c04

BLBK249-Cutler

128

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

Dopamine (DA) metabolites and receptors HVA Dihydroxyphenylacetic acid (DOPAC) D2 receptor Serotonin (5-HT) metabolites and receptors 5-Hydroxyindoleacetic acid (5-HIAA) 5-HT1A receptor 5-HT2A receptor Glutamatergic system Glutamine Glutamate-metabolizing enzymes (GMEs) Glutamate synthetase (GS) Glutamate synthetase-like protein (GSLP) Glutamate dehydrogenase (GDH) Glutamine/glutamate NMDA-type glutamate receptors CSF metabolic profile Glucose concentration CSF pH Acetate and lactate concentrations Glucose tolerance Neurophysiological signals Mismatch negativity Event-related potentials (ERPs) N100 P50 gating P300 N400 P3 Prepulse inhibition (PPI) Smooth-pursuit eye movements Ethane exhalation

Table 4.6 Biomarkers for schizophrenia

In addition, the typical antipsychotic haloperidol produced reductions in plasma HVA levels in schizophrenia patients that correlated with improvements in symptoms, measured by the Brief Psychiatry Rating Scale (BPRS) [357]. The baseline plasma levels of HVA in drug-na¨ıve schizophrenia patients were able to predict the effect of subsequent treatment as well as the patient’s long-term prognosis [358, 359]. Interestingly, HVA levels were also elevated during the prodromal phase of schizophrenia, suggesting that this marker may be useful for early detection of the disorder before the onset of psychotic symptoms [360]. The treatment of schizophrenia patients with antipsychotics produced significant increases in the CSF HVA/5-HIAA and HVA/MHPG ratios in comparison to normal controls [361]. A remaining caveat with HVA as a biomarker is that some changes induced by antipsychotics do not always track with drug efficacy or performance on cognitive tests [362]. For example, in one study the atypical antipsychotic olanzapine significantly increased HVA concentrations and the HVA/5-HIAA

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

129

ratio in the CSF of schizophrenia patients, but these changes did not track with clinical response [363]. Another study looked at CSF HVA changes after antipsychotic administration and also did not find a correlation with clinical response [364]. The levels of plasma HVA do not produce a consistent and linear return to baseline after schizophrenia patients discontinue drug treatment, making it difficult to track the effects of antipsychotic withdrawal [365, 366]. These inconsistencies will need to be resolved before HVA can become a reliable biomarker for antipsychotic efficacy. The dopamine metabolite dihydroxyphenylacetic acid (DOPAC) is another potential biomarker for schizophrenia. DOPAC was found to be decreased in postmortem samples of the anterior cingulate cortex in individuals with schizophrenia compared with normal controls [367]. In rat models, antipsychotics increased the interstitial levels of DOPAC (measured via microdialysis) in the nucleus accumbens, and the atypical antipsychotics clozapine, risperidone, and sertindole preferentially increased DOPAC in the PFC [368]. This could explain a proposed mechanism of action for some antipsychotics— correcting dopaminergic tone deficits induced during schizophrenia. Neither CSF nor plasma DOPAC levels show significant differences between treatment-na¨ıve schizophrenia patients and normal controls [369, 370]. Despite this, several studies show modulation of DOPAC levels in patients treated with antipsychotics. For example, the atypical antipsychotic clozapine increased plasma DOPAC levels in schizophrenia patients [371]. Another study found changes in plasma DOPAC levels of schizophrenia patients during treatment with the typical antipsychotic haloperidol [372]. While promising, more studies are required to resolve this metabolite’s reported inconsistencies and to evaluate its ability to serve as a biomarker for antipsychotic efficacy. It remains to be determined why changes in DOPAC levels cannot be detected in the CSF or plasma of unmediated schizophrenia patients. It also needs to be determined how DOPAC levels change for the various classes of antipsychotics, and whether these changes can be detected in the CSF of schizophrenia patients. Future studies will hopefully address these questions. DA receptors can be labeled using various compounds and imaging techniques, and this permits differential measurements of DA receptor binding and densities in the brains of schizophrenia patients versus healthy controls, as well as after antipsychotic administration. In addition to postmortem studies, PET and single photon emission computed tomography (SPECT) can be combined with radioligands that bind to the DA receptors, such as [3 H]/[11 C]Raclopride, [76 Br]Bromolisuride, [3 H]Emonapride, [123 I]IBZM, [11 CN]/[3 N]Methylspiperone, [3 H]Spiperone, [3 H]Flupenthixol, and [3 H]Haloperidol (among many others), to evaluate DA receptor binding and density changes in schizophrenia patients in vivo. For example, the D2 dopamine receptor binding potential (BP) was found to be reduced in the anterior cingulate cortex of drug-na¨ıve schizophrenia patients versus healthy controls [373]. This was determined using PET in conjunction with the

c04

BLBK249-Cutler

130

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

high-affinity D2 receptor radioligand carbon 11-labeled FLB 457. Additionally, the density of the D2 receptor was found to be elevated in postmortem brain samples [374] as well as in schizophrenia patients in several studies using PET and various radioligands [375–379], although this result was not replicated in all studies or patient populations [380–383]. This discrepancy may be because DA receptor density changes may be specific to certain subpopulations of schizophrenia patients, or may be more prominent with certain radioligands or in specific brain regions. Studies have also shown that schizophrenia patient D2 receptor density and binding affinity differences are further exaggerated from normal controls when these patients are treated with antipsychotics [384], suggesting that DA receptor density changes could serve as a marker for antipsychotic effects. There are a number of remaining issues with this potential biomarker. First of all, the cause of this exaggeration in D2 receptor density cannot be solely attributed to the medication, and may also be due to disease progression or patient aging. Additionally, DA receptor differences do not show up for all radioligands used [385] or all antipsychotics administered [386]. Further evaluation is required to determine how DA receptor changes in density and binding potential track with antipsychotic efficacy.

Serotonin There exists a large number of conflicting studies regarding how and even whether the serotonergic system is altered in schizophrenia. Several postmortem and PET studies have reported increases in 5-HT1A receptor density in frontal and temporal cortices in the brains of schizophrenia patients, suggesting a deregulation of 5-HT1A function in this disorder [387–389]. In contrast, while some studies show 5-HT1A receptor density decreases with schizophrenia [390], others show no changes at all [391, 392]. Similarly, various PET and postmortem studies have shown increases [393, 394], decreases [395], or no changes [396] in cortical 5-HT1A receptor binding in schizophrenia. A similar pattern of contradictory results exists for the 5-HT2A receptor. One PET study found decreased prefrontal 5-HT2A receptor binding in subjects at risk for schizophrenia [397], while another found no changes in cortical 5-HT2A receptor binding but increased binding in the caudate nucleus of early-stage schizophrenics [398]. Likewise, some PET studies have found 5HT2A receptor density to be decreased in neuroleptic-na¨ıve or prodromal schizophrenia patients [399, 400], while others have found no significant changes in these receptors when compared to normal controls [401, 402]. One of the main serotonin metabolites, 5-hydroxyindoleacetic acid (5-HIAA), also shows inconsistent results across studies of schizophrenia. The CSF levels of 5-HIAA in schizophrenia patients have been reported to be increased [403], decreased [404], or unchanged [405, 406] in comparison to healthy controls. One postmortem study reported an increase in 5-HIAA levels in subcortical areas [407] and another reported a decrease in 5-HIAA levels in cortical regions [408]. Serotonin transporters, an index of 5-HT innervation,

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

131

are generally reported to be decreased in the frontal cortex of patients with schizophrenia [409, 410]. Many atypical antipsychotic drugs are either direct or indirect 5-HT1A agonists, and have been shown to improve cognitive function in patients with schizophrenia [411]. Atypical antipsychotics also function as 5-HT2A antagonists and have been reported to downregulate the 5-HT2A receptor in patients [412]. The efficacy of antipsychotics that target the 5-HT receptors suggests that this system does play a role in the pathogenesis of the disorder [413]. Unfortunately, the heterogeneity of study results has not allowed for a consensus on what this role is, and the conflicting results preclude the use of a serotonergic biomarker to assess antipsychotic efficacy, at least until these discrepancies can be resolved. There are many possible reasons why studies have produced contradictory results regarding the role of the serotonergic system in schizophrenia. These include potential variability in the resolution or accuracy of the analytical techniques used in each study (i.e., postmortem tissue labeling versus in vivo PET imaging studies), variability regarding the reagents or methodology selected for each study (i.e., differing affinities of the various PET radioligands), variability in the brain region or regions analyzed, possible heterogeneity in regard to the patient subpopulation examined (age, sex, genetic characteristics, or disease-state differences), or the wide-ranging effects of medication, among many other variables. Hopefully, further investigation and new analytical tools will shed new light onto the role of the serotonergic systems in schizophrenia, and allow these lingering discrepancies to be resolved.

Glutamate There is a wide range of changes in the glutamatergic system implicated in the pathogenesis of schizophrenia and the administration of antipsychotics, some of which hold the potential to become useful biomarkers. There is evidence for a hyperglutamatergic PFC underlying psychotic symptoms in schizophrenia [414]. Proton magnetic resonance spectroscopy (MRS) has allowed for the detection of an increase in glutamine level in the medial PFC, the left anterior cingulate cortex, and the thalamus of drug-na¨ıve schizophrenic patients compared with healthy controls [415, 416]. Chronic antipsychotic administration appears to reverse this imbalance, because medicated schizophrenia patients had significantly lower levels of glutamine in their left anterior cingulate cortex when compared to healthy controls [416]. It remains to be determined how glutamine levels vary across the entire schizophrenic brain, as well as how their levels are modulated by a wide range of antipsychotics, in order to assess their potential as a biomarker. Significant changes in the amounts of glutamate-metabolizing enzymes (GMEs) have also been found in isolated blood platelets of patients with schizophrenia, and these changes were able to differentiate schizophrenia patients from healthy controls [417]. Specifically, glutamate synthetase (GS), glutamate synthetase-like protein (GSLP), and glutamate dehydrogenase (GDH)

c04

BLBK249-Cutler

132

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

levels were altered in schizophrenia patients. These differences in GME levels were also noted in several studies comparing the autopsied brains of schizophrenia patients with those from healthy subjects [418–420]. The administration of the atypical neuroleptic olanzapine produced GME-level changes in schizophrenia patients that tracked with drug efficacy. This suggests that GMEs could potentially serve as a biomarker for both disease progression and drug efficacy, although further validation is necessary in order to determine what classes of antipsychotics affect these enzymes. There are many other measures of glutamate function and concentration which are altered in schizophrenia patients. For example, a recent study found the ratio of glutamine to glutamate in the CSF of drug-na¨ıve schizophrenia patients to be significantly higher than that of normal controls [421]. In addition, several studies have found evidence of general hypofunction of NMDA-type glutamate receptors in schizophrenia [422–424]. A single dose of an NMDA receptor antagonist such as phencyclidine (PCP) or ketamine produces schizophrenia-like symptoms in healthy individuals and exacerbates preexisting symptoms in schizophrenia patients, primarily by blocking the receptor [425–427]. The agonists of the NMDA-type glutamate receptors were able to increase prepulse inhibition (PPI; a measure of sensorimotor gating typically impaired in schizophrenia) in mice models of schizophrenia equally to the atypical antipsychotic clozapine [428]. This suggests that the function of the NMDA-type glutamate receptors could serve as a potential biomarker for antipsychotic efficacy. While there are many examples of dysfunction and concentration variance in the glutamatergic system that track with schizophrenia progression and antipsychotic administration, all of these measures require more validation before they can be accepted as useful biomarkers for this disorder.

CSF metabolic profile Examining the CSF metabolic profile of schizophrenia patients using 1 H nuclear magnetic resonance spectroscopy (NMR) in conjunction with computerized pattern recognition analysis is a promising new biomarker for detecting disease onset, progression, and the efficacy of antipsychotics. Studies have found that the 1 H NMR spectra of CSF samples from drug-na¨ıve patients with first-onset schizophrenia show a different distribution of metabolites than samples from healthy volunteers. This includes an elevation in CSF glucose concentrations, as well as a significantly lowered CSF pH, lower acetate, and lower lactate concentrations, possibly due to disease-induced alterations in cellular respiration [429]. Fasting glucose tolerance is also impaired in these patients [430]. The metabolic profiles of schizophrenia patients became normalized following treatment with atypical antipsychotics, suggesting this profiling technique could potentially serve as a biomarker of novel antipsychotic efficacy. These differential metabolic profiles did not exist for non-CSF-evaluated metabolites [431]. A drawback with this potential biomarker is that some antipsychotics can disrupt glucose regulation by themselves, making it difficult to track drug

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

133

efficacy based on metabolic profile alone [432]. In addition, disturbed glucose regulation has also been associated with mood and psychotic disorders, although it is unclear whether the complete array of altered metabolites in schizophrenia can reliably distinguish it from other neurological disorders [433]. The further evaluation of this technique is needed in order to determine how the array of CSF metabolites in schizophrenia is distinct from other, similar disorders, and also whether efficacious antipsychotic administration produces predictable and reliable changes in the pattern of CSF metabolites when compared to the drug-na¨ıve metabolic state.

Neurophysiological signals Abnormal neurophysiological signals in schizophrenia include reduced mismatch negativity, abnormalities in event-related potentials (ERPs), reductions in prepulse inhibition (PPI), and deficits in smooth-pursuit eye movements, all of which appear to track with disease pathogenesis, and possibly antipsychotic treatment as well. Mismatch negativity (MMN) is a unique ERP response to an odd stimulus in an otherwise constant stimuli sequence. For example, a deviant sound in a constant sequence of sounds will produce a unique electrophysiological signal that can be recorded from scalp electrodes. MMN response is reduced in patients with schizophrenia [434]. There is some controversy regarding whether this reduction is dependent on disease state or even antipsychotic treatment. One study observed MMN deficiencies in patients with first episodes of schizophrenia [435], while another found MMN reduction in patients with chronic schizophrenia but not first-episode schizophrenia [436]. MMN reductions have been correlated with positive symptoms [437], negative symptoms [438], or no clinical symptoms at all [439]. In addition, the effect of antipsychotic administration on MMN reductions in schizophrenia patients remains unclear due to a lack of definitive studies. When given to schizophrenia patients for 4 weeks, olanzapine improved symptoms but had no significant effects on latencies and amplitudes of MMN [440]. Another study suggested that MMN reductions may be due to antipsychotic administration, not necessarily the disease pathogenesis [441]. Auditory-evoked potentials (AEPs) are a form of event-related potentials (ERPs)—very small electrical voltage potential changes originating from the brain and recorded from electrodes in the scalp. AEPs occur in response to an auditory stimulus, such as a tone. N100 is the largest component of the AEP, peaking between 80 and 120 ms after stimulus onset and maximal over frontocentral leads. N100 is generated by a complex network of cortical areas [442] and is associated with perceptual processing [443]. Patients with schizophrenia have been shown to demonstrate deficits in N100 generation, especially at long interstimulus intervals (ISIs) and extremely short ISIs [444]. The nature of this deficit appears to depend on antipsychotic administration and perhaps disease state. One study found that patients with high doses of antipsychotics had significantly smaller N100 amplitudes, as compared to patients with low doses of antipsychotics [445].

c04

BLBK249-Cutler

134

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

Other studies found reduced N100 amplitude in medicated, but not in unmedicated schizophrenia patients [446–448]. This result must be interpreted with caution because unmedicated subjects are typically less ill than medicated patients. In addition, several studies found reductions in N100 amplitude in unmedicated patients compared to healthy subjects [449–451], and other studies did not find changes in N100 levels with antipsychotic administration [452– 454]. These conflicting results could be due to the class of antipsychotic drug administered, the stage of the disease, or other unknown variables. Clearly, more studies are needed to determine the effect of antipsychotics on the N100 amplitude, as well as to differentiate between the effects of different kinds of antipsychotics. Schizophrenia patients show deficits in other components of the ERP compared to healthy subjects, some of which may be ameliorated by antipsychotics. For example, schizophrenia patients have deficits in P50 gating [455], which can be corrected with the administration of atypical antipsychotics such as clozapine [456, 457]. Schizophrenia patients also have reduced amplitude and latency of P300 [458, 459], reduced amplitude of N400 [460], and reduced amplitude of P3 [461], among many others. A remaining problem with all ERP deficits in schizophrenia is that not all studies find the same deficits, or the same magnitude of deficits, in patients. In addition, it is unclear how antipsychotic administration, disease state, and other factors affect these deficits. More studies are needed to answer these questions before ERPs can be reliably used as biomarkers for antipsychotic efficacy. PPI is the reduction in startle response typically elicited by a sudden auditory stimulus when it is preceded by a prepulse. Taken as a measure of sensorimotor gating, PPI is reduced in schizophrenia patients [462]. This deficit is more prominent earlier in the disease [463]. PPI deficits may be reversed by the administration of atypical antipsychotics such as clozapine but not by conventional antipsychotics [464, 465]. In addition, PPI deficits are not specific to schizophrenia and are seen in a variety of psychiatric disorders [466]. Therefore, caution should be exercised when measuring PPI as a schizophrenia biomarker. Eye-pursuit abnormalities are also common in schizophrenia patients. These include impaired smooth pursuit to a moving target [467] and the inability to suppress automatic reflexive glances toward an object, also known as antisaccade performance [468]. Both correlate with prefrontal abnormalities, which have been implicated in the pathology of schizophrenia [469]. The effect of antipsychotics on these deficits remains unresolved [470, 471], and more studies are required to determine whether the resolution of eye-pursuit abnormalities correlate with the efficacy of antipsychotics.

Ethane exhalation There is evidence pointing to free-radical-mediated damage and perturbation of the body’s defenses against such damage in patients with schizophrenia. Erythrocyte antioxidant enzyme activity has been reported to be altered

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

135

in chronic schizophrenia, with increased activity of the antioxidant superoxide dismutase (SOD) [472]. The raised levels of membrane lipid peroxidation products, indicators of cell damage from oxidative lipid degradation, have also been reported in schizophrenia patients, in both plasma [473] and possibly CSF [474], although this has been disputed [475]. The hydrocarbon ethane is produced as a terminal catabolite of polyunsaturated fatty acid oxidation and is excreted in the breath; therefore, it can be used as a measure of freeradical-mediated damage. It has been discovered recently that the level of ethane in the breath of schizophrenia patients has been found to be significantly elevated in comparison to healthy controls [476]. Because it is completely noninvasive, ethane exhalation would make an ideal biomarker for disease progression. It remains to be determined whether ethane levels are also reduced by antipsychotic administration.

Biomarkers and the FDA’s Critical Path Initiative In an effort to increase the number of innovative medical products submitted for approval, to promote greater efficiency in clinical development, and to help bridge the ever-widening chasm apparent between basic and clinical trial successes, the FDA launched the Critical Path Initiative in 2004. This initiative was designed to stimulate and facilitate a national effort to modernize the process through which potential drugs, biological discoveries, and medical devices are transformed from a concept into an actual medical product that could benefit humans. Specifically, the initiative identifies and prioritizes the most pressing clinical development problems across a number of therapeutic areas and defines the ones that may provide the greatest opportunity for rapid improvement and public health benefits. This is accomplished by directing research not only toward novel medical breakthroughs and discoveries, but also toward the creation of novel tools such as biomarkers. The hope is that these tools will facilitate the development of new treatments—all in an effort to ensure that patients will be able to benefit from more timely, affordable, and predictable access to drug therapies (more information on the initiative can be found at http://www.fda.gov/oc/initiatives/criticalpath/). In 2006, the Critical Path Initiative also called for a list of specific activities that define the areas of greatest opportunity for improvement in the drug and device development field. More than 70 concrete examples were provided that outline how new scientific discoveries might be applied to various stages of drug development in order to help predict safety/efficacy profiles and improve trial results while decreasing timelines and costs. Several of these opportunities/activities are of particular relevance to the domain of CNS drug development (for the complete list, refer to http://www.fda.gov/oc/ initiatives/criticalpath/reports/opp list.pdf). Four of these activities/opportunities deal with biomarkers, emphasizing practical approaches that address development issues efficiently and rationally. The first opportunity concerns a broad effort to promote Biomarker

c04

BLBK249-Cutler

136

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

Qualification and Standards. The second opportunity is specific to the development of Imaging Biomarkers in Neurocogntive Diseases such as Alzheimer’s and Parkinson’s disease. The third involves the establishment of Diagnostic Markers for Neuropsychiatric Conditions. The final opportunity falls under “Topic 2 Streamlining Clinical Trials,” and entails Improving Measurement of Patient Responses by Measuring Patient-Centered Endpoints.

Biomarker Qualification The Critical Path Opportunities list identifies its first topic (Topic 1) as Better Evaluation Tools: Developing New Biomarkers and Disease Models to Improve Clinical Trials and Medical Therapy. The authors note that the “process and criteria for qualifying biomarkers for use in product development should be mapped. Clarity on the conceptual framework and evidentiary standards for qualifying a biomarker for various purposes would establish the path for developing predictive biomarkers.” The implicit assumption is that currently there is a lack of process, criteria, framework, and standards for developing biomarkers and disease models, and this is contributing to the stagnation seen in drug development. The section begins by exploring the process and criteria for qualifying biomarkers as well as the intelligent use of biomarkers in CNS product development for both diagnostic and predictive purposes. While regulations permit the FDA to base the approval of a drug product on how that drug affects an unvalidated surrogate marker (i.e., one for which it is not known that the surrogate actually predicts the desired clinical benefit), there are a number of difficulties in interpreting trials that use surrogate markers as primary measures of drug effect. The actual language is the “FDA may grant marketing approval for a new drug product on the basis of adequate and well-controlled clinical trials establishing that the drug product has an effect on a surrogate endpoint that is reasonably likely, based on epidemiologic, therapeutic, pathophysiologic, or other evidence to predict clinical benefit, or on the basis of an effect on a clinical endpoint other than survival or irreversible morbidity.” Validated surrogate markers are those for which evidence has established that a drug-induced effect on the surrogate predicts (results in) the desired effect on the clinical outcome of interest. Unvalidated surrogates, on the other hand (and as the regulation describes), are “reasonably likely” to predict the clinical benefit of interest, but for which there is insufficient evidence to establish that such an effect definitively results in the desired clinical outcome [477]. Several validated surrogate markers for drug approval include CD4/CD8 cell counts in HIV/AIDS, glucose and hBA1c in diabetes, blood pressure and serum lipids in cardiovascular disease, intraocular pressure in glaucoma, and seropositivity in vaccine protection. However, the standards for validation of a surrogate endpoint are stringent and it is not sufficient to simply demonstrate that the biomarker value is correlated with a clinical outcome. There are numerous examples of biomarkers of disease that keep pace with disease

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

137

outcome but are not considered to be valid surrogate endpoints of clinical outcome. This is because validation requires analysis of a series of randomized clinical trials, showing that the differences in biomarker change between the randomized treatment groups is concordant with the differences in clinical outcome. Due to the relative strict requirements for establishing a biomarker as a valid surrogate endpoint, it is not surprising that sponsors feel that it is often best to perform pivotal totals using standard measures of clinical outcomes [478]. As stated earlier, the criteria for qualifying biomarkers for use in product development needs to be charted with better clarity regarding the conceptual framework and evidentiary standards. In an effort to do just this, Williams et al. reviewed existing key publications on standards for the regulatory acceptance of biomarkers as surrogate endpoints [479]. They suggested that ruleor statistics-based approaches to defining the validity of surrogate endpoints or biomarkers (such as the classic and well-adopted Prentice criteria) inadequately explain real-life experience: in practice, some biomarkers are accepted and others are rejected, but no biomarkers actually meet the idealized statistical requirements. The Prentice criteria essentially require that the surrogate must be a correlate of the true clinical outcome and fully capture the net effect of treatment on the clinical outcome [480].

Predictive biomarkers Despite the lack of accepted framework, processes, and standards for the qualification of biomarkers, it is abundantly clear to CNS researchers that biomarkers have bona fide utility independent of their ability to serve as a validated surrogate. For instance, predictive biomarkers are used to characterize the patient’s disease in order to determine whether the patient is a good candidate for a treatment. The most familiar type of predictive biomarkers is termed classifiers. There are many types of classifiers, but the most commonly used are simple dichotomous classifiers that label patients in terms of being “good” or “bad” candidates for successful treatment. For example, the presence or absence of a specific gene in AD would be considered a dichotomous classifier. The ApoE genotype is the strongest known genetic risk factor for the development of late-onset AD, with the e4 allele incurring greatest risk of developing this disorder [481]. Biomarkers that may be predictive will have application in early drug development, and CNS researchers must be thinking about their potential utility for later stages of development.

Bioanalytical considerations for the laboratory measurement of biomarkers Laboratory issues must be considered when implementing new bioanalytical methods to measure biological fluid biomarkers to support drug development programs. This process is known as analytical validation.

c04

BLBK249-Cutler

138

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

The analytical validation of a biomarker method is not to be confused with a “validated biomarker”. A validated (or qualified) biomarker is a biomarker in which a relationship to a therapeutic intervention can be observed or predicted. The validation of an analytical method for the measurement of a biomarker to support preclinical or clinical drug development does not formally fall within the scope of current regulatory guidance documents. Generally, laboratorians attempt to follow a GLP-like approach by adhering to pharmacokinetic bioanalytical guidelines [482, 483] or diagnostic biomarker guidelines [484, 485]. However, there are unique limitations when validating biomarker methods to support drug development, which prevent the absolute adoption of these guidelines. Industry does recognize that biomarker data submitted to regulatory agencies must be collected using robust analytical methods that can withstand objective scientific and regulatory scrutiny. Because specific regulatory guidelines have not been drafted, the current standard for the analytical validation of biological fluid biomarkers has been recommended by industry thought leaders published in white papers [486–489] and reviews [490]. These recommendations incorporate analytical validation characteristics from the guidelines mentioned above, where applicable, with an emphasis to ensure that the resulting method is reliable for its intended purpose. The standardization of the validation processes for biological fluid biomarkers is complicated because the analytes and the assays are inherently diverse [489]. Most analytes are measured on immunoassay platforms. One of the earliest and most common immunoassay methods is the “sandwich” enzymelinked immunosorbent assay (ELISA). Although the concept is simplistic (capture antibody–analyte–detection antibody), in reality the assay parameters of each biomarker must be optimized, a process that typically requires several weeks of laboratory time. Historically, optimization was approached one factor at a time and tended to be circular. That is, upon optimization of one assay step or component, that step was fixed while the next step was optimized, followed by the optimization of subsequent steps and ending with confirmation that all changes did not negatively impact the previous optimizations. However, the availability of biostatistics-friendly software packages has recently made possible multifactorial optimization using the Design of Experiment approach for immunoassays [491, 492]. A balance must be sought between exhaustively spending resources to anticipate all potential trial sample scenarios and producing an analytical method that is reliable, robust, and validated for its intended purpose. More recent advances in biological fluid biomarker analysis have included attempts at method automation and multiplexing: the simultaneous measurement of multiple biomarkers from a single biological fluid sample [493]. The most common platforms for protein multiplexing include flow cytometric microbead assays and ELISA methods in which specific capture arrays are coated on microwell plates. The advantage of using multiplexing is that data are generated for the measurement of multiple biomarkers from a single sample in

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

139

one assay. In general, multiplex methods tend to be less robust than singleplex methods, because simultaneous measurements require that all analytes must be measured under identical reagent and incubation conditions and the opportunity for optimization of individual assays is not possible. The simultaneous validation of multiplexed assays has been demonstrated [494]; however, multiplexing tends to be employed in early drug development to look for patterns of change in biomarker classes. Once candidate biomarkers are identified, it is desirable to optimize and perform more rigorous method validations to support later stage drug development. There are several challenges that are unique to the analytical validation, implementation, and maintenance of a method that must provide reliable and consistent data over the course of a lengthy clinical trial or between trials where biomarker data may be compared. This is especially true for ligand binding assay (LBA) methodologies, which are often used by necessity for biomarkers with molecular weights greater than 5000 Daltons (Da). There are two factors that can be identified as the main contributors adding to the complexity of biomarker measurement by LBA: (1) the availability of a true reference standard (and lack of an internal standard); and (2) samples are analyzed in biological matrices because LBA methods employ protenacious reagents.

Reference standard Except for rare circumstances, a true reference standard to support biomarker analysis is usually not available. The laboratory must rely on vendors to maintain quality between different lot productions, otherwise an undesirable shift may be observed in the analytical method. Lot-to-lot shifts create a burden for laboratories to reoptimize their analytical methods so that biomarker measurements remain consistent within and across studies. Furthermore, protein biomarker reference standards are often generated by overexpression in cell culture and purified from lysed cells. Fermentation conditions and cell line types may result in differences in protein folding and post-translation modifications. The commercial reference standard may not fully represent the biomarker that is being measured in the biological fluid sample. The consequence is that the method may have good precision but poor accuracy. Ideally, the bioanalytical laboratory will have some control over the source of the assay reference standard. In addition, the preparation of a stable quality control material that is included in every assay is useful for following assay shifts at lot changes. Interference from matrix/endogenous analyte To comply with the consensus recommendations, calibrators and quality controls should be prepared in a biological matrix that resembles the study population as closely as possible. This is not always possible because of the availability of rare matrices or ethical issues. In addition, biological fluids may already contain the biomarker at concentrations that interfere with the preparation of calibrators and quality controls. One impact is that method accuracy

c04

BLBK249-Cutler

140

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

may not be able to be fully evaluated. There are procedures for selectively (immunodepletion) and nonselectively (charcoal stripping, size-exclusion filtration, etc.) removing endogenous analytes; however, scientific justification for doing so should be documented. Another matrix factor that must be considered is that a relatively large sample volume must be used to detect biomarkers that are present at low concentrations or that are suppressed by the therapeutic intervention. In addition to potential analytical matrix interference, this requirement may affect the study protocol and sample collection process. There have been significant advances toward providing recommendations to bioanalytical laboratories, which will ensure that reliable, high-quality data is generated. Undoubtedly, advances in clinical practice and sample handling will reduce preanalytical errors, while new methodologies and the availability of superior bioanalytical reagents will increase the quality of data generated by these methods.

References 1. Roses AD. (2004) Pharmacogenetics and drug development: the path to safer and more effective drugs. Nat Rev Genet. 5:645–56. 2. Lesko LJ, Atkinson AJ, Jr. (2001) Use of biomarkers and surrogate endpoints in drug development and regulatory decision making: criteria, validation, strategies. Ann Rev Pharmacol Toxicol. 41:347–66. 3. Suter L, Babiss LE, Wheeldon EB. (2004) Toxicogenomics in predictive toxicology in drug development. Chem Biol. 11:161–71. 4. Biomarkers Definitions Working Group. (2001) Biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clin Pharmacol Therapeut. 69:89–95. 5. Bakhtiar R. (2008) Biomarkers in drug discovery and development. J Pharmacol Toxicol Methods. 57(2):85–91. 6. Vesterqvist OE PC, Rutkowski JL. (April/May 2008) From biological measures to valid biomarker: predicting success of novel therapeutic. American Drug Discov. 3(3):56–9. 7. Wahlund L, Blennow K. (2003) Cerebrospinal fluid biomarkers for disease stage and intensity in cognitively impaired patients. Neurosci Lett. 339:99–102. 8. Ilyin SE, Belkowski SM, Plata-Salaman CR. (2004) Biomarker discovery and validation: technologies and integrative approaches. Trends Biotechnol. 22:411–6. 9. Frank RA, Galasko D, Hampel H, et al. (2003) Biological markers for therapeutic trials in Alzheimer’s disease—proceedings of the biological markers working group—NIA initiative on neuroimaging in Alzheimer’s disease. Neurobiol Aging. 24:521–36. 10. Shaw LM, Korecka M, Clark CM, et al. (2007) Biomarkers of neurodegeneration for diagnosis and monitoring therapeutics. Nat Rev Drug Discov. 6(4):295–303. 11. Thal LJ, Kantarci K, Reiman EM, et al. (2006) The role of biomarkers in clinical trials for Alzheimer disease. Alzheimer Dis Assoc Disord. 20:6–15. 12. Gil-Bea FJ, Garc´ıa-Alloza M, Dom´ınguez J, et al. (2008) Acetylcholine receptors in dementia and mild cognitive impairment. Eur J Nucl Med Mol Imaging. 35(Suppl 1):S30– 45. 13. Reinikainen KJ, Soininen H, Riekkinen PJ. (1990) Neurotransmitter changes in Alzheimer’s disease: implications to diagnostics and therapy. J Neurosci Res. 27(4): 576–86.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

141

14. Nilsson L, Nordberg A, Hardy J, et al. (1986) Physostigmine restores 3H-acetylcholine efflux from Alzheimer brain slices to normal level. J Neural Transm. 67:275–85. 15. Rylett RJ, Ball MJ, Colhoun EH. (1983) Evidence for high affinity choline transport in synaptosomes prepared from hippocampus and neocortex of patients with Alzheimer’s disease. Brain Res. 289:169–75. 16. Bierer LM, Haroutunian V, Gabriel S, et al. (1995) Neurochemical correlates of dementia severity in Alzheimer’s disease: relative importance of the cholinergic deficits. J Neurochem. 64(2):749–60. 17. Arendt T, Bigl V, Tennstedt A, et al. (1985) Neuronal loss in different parts of the nucleus basalis is related to neuritic plaque formation in cortical target areas in Alzheimer’s disease. Neuroscience. 14(1):1–14. 18. Etienne P, Robitaille Y, Wood P, et al. (1986) Nucleus basalis neuronal loss, neuritic plaques and choline acetyltransferase activity in advanced Alzheimer’s disease. Neuroscience. 19(4):1279–91. 19. Sberna G, S´aez-Valero J, Beyreuther K, et al. (1997) The amyloid beta-protein of Alzheimer’s disease increases acetylcholinesterase expression by increasing intracellular calcium in embryonal carcinoma P19 cells. J Neurochem. 69(3):1177–84. 20. Schachter AS, Davis K. (1999) Guidelines for the appropriate use of cholinesterase inhibitors in patients with Alzheimer’s disease. CNS Drugs. 11:281–8. 21. Farlow MR, Hake A, Messina J, et al. (2001) Response of patients with Alzheimer’s disease to rivastigmine treatment is predicted by the rate of disease progression. Arch Neurol. 58:417–22. 22. Amici S, Lanari A, Romani R, et al. (2001) Cerebrospinal fluid acetylcholinesterase activity after long-term treatment with donepezil and rivastigmina. Mech Ageing Dev. 122(16):2057–62. 23. Auld DS, Kornecook TJ, Bastianetto S, et al. (2002) Alzheimer’s disease and the basal forebrain cholinergic system: relations to beta-amyloid peptides, cognition, and treatment strategies. Prog Neurobiol. 68(3):209–45. 24. Tariot PN. (2001) Maintaining cognitive function in Alzheimer disease: how effective are current treatments. Alzheimer Dis Assoc Disord. 1:S26–S33. ´ MS, Silveyra MX, Andreasen N, et al. (2007) Cerebrospinal fluid acetyl25. Garc´ıa-Ayllon cholinesterase changes after treatment with donepezil in patients with Alzheimer’s disease. J Neurochem. 101(6):1701–11. 26. Braginskaya FI, Zorina M, Pal’mina NP, et al. (2001) Some blood biochemistry parameters during the cholinergic treatment of Alzheimer’s disease. Neurosci Behav Physiol. 31(4):457–61. 27. Sramek JJ, Cutler NR. (2000) RBC cholinesterase inhibition: a useful surrogate marker for cholinesterase inhibitor activity in Alzheimer disease therapy? Alzheimer Dis Assoc Disord. 14(4):216–27. 28. Unni L, Vicari S, Moriearty P, et al. (2000) The recovery of cerebrospinal fluid acetylcholinesterase activity in Alzheimer’s disease patients after treatment with metrifonate. Methods Find Exp Clin Pharmacol. 22(1):57–61. 29. Gravina SA, Ho L, Eckman CB, et al. (1995) Amyloid protein A in Alzheimer’s disease brain: biochemical and immunocytochemical analysis with antibodies specific for forms ending at A 40 or A 42 43. J Biol Chem. 270:7013–6. 30. Blennow K, Hampel H. (2003) CSF markers for incipient Alzheimer’s disease. Lancet Neurol. 2:605–13. 31. Hulstaert F, Blennow K, Ivanoiu A, et al. (1999) Improved discrimination of AD patients using β-amyloid(1-42) and tau levels in CSF. Neurology. 52:1555–62.

c04

BLBK249-Cutler

142

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

32. Shoji M, Matsubara E, Kanai M, et al. (1998) Combination assay of CSF tau, Aβ 1–40 and Aβ 1–42(43) as a biochemical marker of Alzheimer’s disease. J Neurol Sci. 158(2):134–40. 33. Motter R, Vigo-Pelfrey C, Kholodenko D, et al. (1995) Reduction of beta-amyloid peptide 42 in the cerebrospinal fluid of patients with Alzheimer’s disease. Ann Neurol. 38(4):643–8. 34. Lewczuk P, Esselmann H, Otto M, et al. (2004) Neurochemical diagnosis of Alzheimer’s dementia by CSF Abeta42, Abeta42/Abeta40 ratio and total tau. Neurobiol Aging. 25(3):273–81. 35. Irizarry MC. (2004) Biomarkers of Alzheimer disease in plasma. NeuroRx. 1:226–34. 36. Mayeux R, Honig LS, Tang MX, et al. (2003) Plasma A beta 40 and A beta 42 and Alzheimer’s disease: relation to age, mortality, and risk. Neurology. 61:1185–90. 37. Pomara N, Willoughby LM, Sidtis JJ, et al. (2005) Selective reductions in plasma Abeta 1-42 in healthy elderly subjects during longitudinal follow-up: a preliminary report. Am J Geriatr Psychiatry. 13(2005):914–7. 38. Saido TC. (2003) Aβ Metabolism and Alzheimer’s Disease. Landes Bioscience, Georgetown, TX. 39. Siemers E, Skinner M, Dean RA, et al. (2005) Safety, tolerability, and changes in amyloid beta concentrations after administration of a gamma-secretase inhibito in volunteers. Clin Neuropharmacol. 28(3):126–32. 40. Dickerson BC, Sperling RA. (2005) Neuroimaging biomarkers for clinical trials of disease-modifying therapies in Alzheimer’s disease. NeuroRx. 2:348–60. 41. Blennow K, Zetterberg H. (2006) Pinpointing plaques with PIB. Nat Med. 12:753–4. 42. Klunk E, Engler H, Nordberg A, et al. (2004) Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Compound-B. Ann Neurol. 55:306–19. 43. Bateman RJ, Munsell LY, Morris JC, et al. (2006) Human amyloid-beta synthesis and clearance rates as measured in cerebrospinal fluid in vivo. Nat Med. 12(7):856–61. 44. Bateman RJ, Wen G, Morris JC, et al. (2007) Fluctuations of CSF amyloid-beta levels: implications for a diagnostic and therapeutic biomarker. Neurology. 68(9):666–9. 45. Siemers ER, Dean RA, Friedrich S, et al. (2007) Safety, tolerability, and effects on plasma and cerebrospinal fluid amyloid-beta after inhibition of gamma-secretase. Clin Neuropharmacol. 30(6):317–25. 46. Aisen PS, Saumier D, Briand R, et al. (2006) A Phase II study targeting amyloid-beta with 3APS in mild-to-moderate Alzheimer disease. Neurology. 67(10):1757–63. 47. Schenk DB, Seubert P, Grundman M, et al. (2005) A beta immunotherapy: lessons learned for potential treatment of Alzheimer’s disease. Neurodegener Dis. 2(5):255–60. ¨ 48. Burger nBK, Padberg F, Nolde T, et al. (1999) Cerebrospinal fluid tau protein shows a better discrimination in young old (<70 years) than in old old patients with Alzheimer’s disease compared with controls. Neurosci Lett. 277:21–4. 49. Buerger K, Ewers M, Pirttil¨a T, et al. (2006) CSF phosphorylated tau protein correlates with neocortical neurofibrillary pathology in Alzheimer’s disease. Brain. 129(Pt 11):3035–41. 50. Lewczuk P, Esselmann H, Bibl M, et al. (2004) Tau protein phosphorylated at threonine 181 in CSF as a neurochemical biomarker in Alzheimer’s disease: original data and review of the literature. J Mol Neurosci. 23(1–2):115–22. 51. Itoh N, Arai H, Urakami K, et al. (2001) Large-scale, multicenter study of cerebrospinal fluid tau protein phosphorylated at serine 199 for the antemortem diagnosis of Alzheimer’s disease. Ann Neurol. 50(2):150–6. 52. Bouwman FH SN, Verwey NA, van Elk EJ, et al. (April 8, 2008) CSF biomarker levels in early and late onset Alzheimer’s disease. Neurobiol Aging.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

143

53. Formichi P, Battisti C, Radi E, et al. (2006) Cerebrospinal fluid tau, A beta, and phosphorylated tau protein for the diagnosis of Alzheimer’s disease. J Cell Physiol. 208(1):39– 46. 54. Hampel H, Buerger K, Zinkowski R, et al. (2004) Measurement of phosphorylated tau epitopes in the differential diagnosis of Alzheimer disease: a comparative cerebrospinal fluid study. Arch Gen Psychiatry. 61:95–102. 55. Buerger K, Teipel SJ, Zinkowski R, et al. (2002) CSF tau protein phosphorylated at threonine 231 correlates with cognitive decline in MCI subjects. Neurology. 59:627–9. 56. Hansson O, Zetterberg H, Buchhave P, et al. (2006) Association between CSF biomarkers and incipient Alzheimer’s disease in patients with mild cognitive impairment: a follow-up study. Lancet Neurol. 5:228–34. 57. Mitchell A, Brindle N. (2003) CSF phosphorylated tau: does it constitute an accurate biological test for Alzheimer’s disease? Int J Geriatr Psychiatry. 18:407–11. 58. Brys M, Pirraglia E, Rich K, et al. (2009) Prediction and longitudinal study of CSF biomarkers in mild cognitive impairment. Neurobiol Aging. 30(5):682–90. 59. Blennow K, de Leon MJ, Zetterberg H. (2006) Alzheimer’s disease. Lancet. 368(9533): 387–403. 60. Petersen RC, Doody R, Kurz A, et al. (2001) Current concepts in mild cognitive impairment. Arch Neurol. 58(12):1985–92. 61. Lewczuk P, Kornhuber J, Vanderstich eH, et al. (2008) Multiplexed quantification of dementia biomarkers in the CSF of patients with early dementias and MCI: a multicenter study. Neurobiol Aging. 29(6):812–8. 62. Zetterberg H, Wahlund LO, Blennow K. (2003) Cerebrospinal fluid markers for prediction of Alzheimer’s disease. Neurosci Lett. 352:67–9. 63. Roberds SL, Anderson J, Basi G, et al. (2001) BACE knockout mice are healthy despite lacking the primary beta-secretase activity in brain: implications for Alzheimer’s disease therapeutics. Hum Mol Genet. 10:1317–24. 64. Luo Y, Bolon B, Kahn S, et al. (2001) Mice deficient in BACE-1, the Alzheimer’s betasecretase, have normal phenotype and abolished beta-amyloid generation. Nat Neurosci. 4:231–2. 65. Holsinger RMD, Lee JS, Boyd A, et al. (2006) CSF BACE1 activity is increased in CJD and Alzheimer disease versus other dementias. Neurology. 67:710–2. 66. Zhong Z, Ewers M, Teipel S, et al. (2007) Levels of beta-secretase (BACE1) in cerebrospinal fluid as a predictor of risk in mild cognitive impairment. Arch Gen Psychiatry. 64(6):718–26. 67. Verheijen JH, Huisman LGM, van LN, et al. (2006) Detection of a soluble form of BACE1 in human cerebrospinal fluid by a sensitive activity assay. Clin Chem. 52:1168–74. 68. Holsinger RMD, McLean CA, Collins SJ, et al. (2004) Increased β-secretase activity in cerebrospinal fluid of Alzheimer’s disease subjects. Ann Neurol. 55:898–9. 69. Hampel H, Shen Y. (2009) Beta-site amyloid precursor protein cleaving enzyme 1 (BACE1) as a biological candidate marker of Alzheimer’s disease. Scand J Clin Lab Invest. 69(1):8–12. 70. Johnston JA, Liu WW, Todd SA, et al. (2005) Expression and activity of β-site amyloid precursor protein cleaving enzyme in Alzheimer’s disease. Biochem Soc Trans. 33: 1096–100. 71. Wu G, Sankaranarayanan S, Tugusheva K, et al. (2008) Decrease in age-adjusted cerebrospinal fluid beta-secretase activity in Alzheimer’s subjects. Clin Biochem. 41(12): 986–96. 72. Porcari V, Magnoni L, Terstappen GC, et al. (2005) A continuous time-resolved fluorescence assay for identification of BACE-1 inhibitors. Assay Drug Dev Technol. 3:287–97.

c04

BLBK249-Cutler

144

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

73. Evin G, Kenche VB. (2007) BACE inhibitors as potential therapeutics for Alzheimer’s disease. Recent Patents CNS Drug Discov. 2(3):188–99. 74. Imbimbo BP. (2008) Therapeutic potential of gamma-secretase inhibitors and modulators. Curr Top Med Chem. 8(1):54–61. 75. Barten DM, Meredith JE, Jr, Zaczek R, et al. (2006) Gamma-secretase inhibitors for Alzheimer’s disease: balancing efficacy and toxicity. Drugs R D. 7(2):87–97. 76. Peretto I, La Porte E. (2008) Gamma-secretase modulation and its promise for Alzheimer’s disease: a medicinal chemistry perspective. Curr Top Med Chem. 8(1):38– 46. 77. Beher D. (2008) Gamma-secretase modulation and its promise for Alzheimer’s disease: a rationale for drug discovery. Curr Top Med Chem. 8(1):34–7. 78. Jakes R, Spillantini MG, Goedert M. (1994) Identification of two distinct synucleins from human brain. FEBS Lett. 345(1):27–32. 79. U´eda K, Fukushima H, Masliah E, et al. (1993) Molecular cloning of cDNA encoding an unrecognized component of amyloid in Alzheimer disease. Proc Natl Acad Sci USA. 90:11282–6. 80. Culvenor JG, McLean CA, Cutt S, et al. (1999) Non-Abeta component of Alzheimer’s disease amyloid (NAC) revisited. NAC and alpha-synuclein are not associated with Abeta amyloid. Am J Pathol. 155(4):1173–81. 81. Lippa CF, Fujiwara H, Mann DM, et al. (1998) Lewy bodies contain altered alphasynuclein in brains of many familial Alzheimer’s disease patients with mutations in presenilin and amyloid precursor protein genes. Am J Pathol. 153(5):1365–70. 82. Arai Y, Yamazaki M, Mori O, et al. (2001) Alpha-synuclein-positive structures in cases with sporadic Alzheimer’s disease: morphology and its relationship to tau aggregation. Brain Res. 888(2):287–96. 83. Hamilton RL. (2000) Lewy bodies in Alzheimer’s disease: a neuropathological review of 145 cases using alpha-synuclein immunohistochemistry. Brain Pathol. 10:378– 84. 84. Brookes AJ, St CD. (1994) Synculein proteins and Alzheimer’s disease. Trends Neurosci. 17(10):404–5. 85. Iwai A, Masliah E, Yoshimoto M, et al. (1995) The precursor protein of non-A beta component of Alzheimer disease amyloid is a presynaptic protein of the central nervous system. Neuron. 14(2):467–75. 86. Muntan´e G, Dalfo´ E, Martinez A, et al. (2008) Phosphorylation of tau and alphasynuclein in synaptic-enriched fractions of the frontal cortex in Alzheimer’s disease, and in Parkinson’s disease and related alpha-synucleinopathies. Neuroscience. 152(4): 913–23. 87. Barbour R, Kling K, Anderson JP, et al. (2008) Red blood cells are the major source of alpha-synuclein in blood. Neurodegener Dis. 5(2):55–9. 88. Fjorback AW, Varming K, Jensen PH. (2007) Determination of alpha-synuclein concentration in human plasma using ELISA. Scand J Clin Lab Invest. 67(4):431–5. 89. van Geel WJ, Abdo WF, Melis R, et al. (2008) A more efficient enzyme-linked immunosorbent assay for measurement of alpha-synuclein in cerebrospinal fluid. J Neurosci Methods. 168(1):182–5. 90. Lee SJ. (2008) Origins and effects of extracellular alpha-synuclein: implications in Parkinson’s disease. J Mol Neurosci. 34(1):17–22. 91. El-Agnaf OM, Salem SA, Paleologou KE, et al. (2003) Alpha-synuclein implicated in Parkinson’s disease is present in extracellular biological fluids, including human plasma. Faseb J. 17:1945–7.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

145

92. El-Agnaf OM, Salem SA, Paleologou KE, et al. (2006) Detection of oligomeric forms of alpha-synuclein protein in human plasma as a potential biomarker for Parkinson’s disease. Faseb J. 20(3):419–25. 93. Tokuda T, Salem SA, Allsop D, et al. (2006) Decreased alpha-synuclein in cerebrospinal fluid of aged individuals and subjects with Parkinson’s disease. Biochem Biophys Res Commun. 349(1):162–6. 94. Li QX, Mok SS, Laughton KM, et al. (2007) Plasma alpha-synuclein is decreased in subjects with Parkinson’s disease. Exp Neurol. 204(2):583–8. 95. Ono K, Noguchi-Shinohara M, Yoshita M, et al. (2007) Cerebrospinal fluid of Alzheimer’s disease and dementia with Lewy bodies patients enhances alphasynuclein fibril formation in vitro. Exp Neurol. 203(2):579–83. 96. Tang K, Hynan LS, Baskin F, et al. (2006) Platelet amyloid precursor protein processing: a bio-marker for Alzheimer’s disease. J Neurol Sci. 240(1-2):53–8. 97. Zainaghi IA, Forlenza OV, Gattaz WF. (2007) Abnormal APP processing in platelets of patients with Alzheimer’s disease: correlations with membrane fluidity and cognitive decline. Psychopharmacology (Berl). 192(4):547–53. 98. Baskin F, Rosenberg RN, Iyer L, et al. (2000) Platelet APP isoform ratios correlate with declining cognition in AD. Neurology. 54(10):1907–9. 99. Padovani A, Borroni B, Colciaghi F, et al. (2002) Abnormalities in the pattern of platelet amyloid precursor protein forms in patients with mild cognitive impairment and Alzheimer disease. Arch Neurol. 59(1):71–5. 100. Padovani A, Pastorino L, Borroni B, et al. (2001) Amyloid precursor protein in platelets: a peripheral marker for the diagnosis of sporadic AD. Neurology. 57(12):2243–8. 101. Borroni B, Colciaghi F, Pastorino L, et al. (2006) Amyloid precursor protein in platelets of patients with Alzheimer disease: effect of acetylcholinesterase inhibitor treatment. Clin Chim Acta. 364(1–2):91–112. 102. Liu HC, Chi CW, Ko SY, et al. (2005) Cholinesterase inhibitor affects the amyloid precursor protein isoforms in patients with Alzheimer’s disease. Dement Geriatr Cogn Disord. 19(5–6):345–8. 103. Borroni B, Colciaghi F, Caltagirone C, et al. (2003) Platelet amyloid precursor protein abnormalities in mild cognitive impairment predict conversion to dementia of Alzheimer type: a 2-year follow-up study. Arch Neurol. 60(12):1740–4. 104. Pirttil¨a T, Kim KS, Mehta PD, et al.. (1994) Soluble amyloid beta-protein in the cerebrospinal fluid from patients with Alzheimer’s disease, vascular dementia and controls. J Neurol Sci. 127(1):90–5. 105. Van Nostrand WE, Wagner SL, Shankle WR, et al. (1992) Decreased levels of soluble amyloid beta-protein precursor in cerebrospinal fluid of live Alzheimer disease patients. Proc Natl Acad Sci USA. 89(7):2551–5. 106. Almkvist O, Basun H, Wagner SL, et al. (May 1997) Cerebrospinal fluid levels of alphasecretase-cleaved soluble amyloid precursor protein mirror cognition in a Swedish family with Alzheimer disease and a gene mutation. Arch Neurol. 54(5):641–4. 107. Lannfelt L, Basun H, Vigo-Pelfrey C, et al. (1995) Amyloid beta-peptide in cerebrospinal fluid in individuals with the Swedish Alzheimer amyloid precursor protein mutation. Neurosci Lett. 199(3):203–6. 108. Schubert D, Behl C. (1993) The expression of amyloid beta protein precursor protects nerve cells from beta-amyloid and glutamate toxicity and alters their interaction with the extracellular matrix. Brain Res. 629(2):275–82. 109. Yang JL, Weissman L, Bohr VA, et al. (2008) Mitochondrial DNA damage and repair in neurodegenerative disorders. DNA Repair (Amst). 7(7):1110–20.

c04

BLBK249-Cutler

146

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

110. Moreira PI, Santos MS, Oliveira CR, et al. (2008) Alzheimer disease and the role of free radicals in the pathogenesis of the disease. CNS Neurol Disord Drug Targets. 7(1):3–10. 111. Lovell MA, Markesbery WR. (2007) Oxidative DNA damage in mild cognitive impairment and late-stage Alzheimer’s disease. Nucleic Acids Res. 35(22):7497–504. 112. Skovronsky DM, Lee VMY, Trojanowski JQ. (2006) Neurodegenerative diseases: new concepts of pathogenesis and their therapeutic implications. Annu Rev Pathol Mech Dis. 1:151–70. 113. Clark, CM, Pratico D, Shaw L, Leight S, Xie S, Gu A. (2006) Biochemical biomarkers of latelife dementia. Alzheimers Dement. 2:287–93. 114. de LMJ, DeSanti S, Zinkowski R, et al. (2006) Longitudinal CSF and MRI biomarkers improve the diagnosis of mild cognitive impairment. Neurobiol Aging. 27:394–401. 115. Han X. (2007) Potential mechanisms contributing to sulfatide depletion at the earliest clinically recognizable stage of Alzheimer’s disease: a tale of shotgun lipidomics. J Neurochem. 103(Suppl 1):171–9. 116. Cheng H, Xu J, McKeel DW, Jr, et al. (2003) Specificity and potential mechanism of sulfatide deficiency in Alzheimer’s disease: an electrospray ionization mass spectrometric study. Cell Mol Biol (Noisy-le-grand). 49(5):809–18. 117. McGeer PL, McGeer EG. (2007) NSAIDs and Alzheimer disease: epidemiological, animal model and clinical studies. Neurobiol Aging. 28(5):639–47. 118. Wang WF, Liao YC, Wu SL, et al. (2005) Association of interleukin-1 beta and receptor antagonist gene polymorphisms with late-onset Alzheimer’s disease in Taiwan Chinese. Eur J Neurol. 12:609–13. 119. Culpan D, MacGowan SH, Ford JM, et al. (2003) Tumour necrosis factor-a gene polymorphisms and Alzheimer’s disease. Neurosci Lett. 250:61–5. 120. McGeer PL, McGeer EG. (2001) Polymorphisms in inflammatory genes and the risk of Alzheimer’s disease. Arch Neurol. 58:1790–2. 121. Licastro F, Pedrini S, Caputo L, et al. (2000) Increased plasma levels of interleukin-1, intertleukin-6 and alpha-1-antichymotrypsin in patients with Alzheimer’s disease: peripheral inflammation or signals from the brain? J Neuroimmunol. 103:97–102. 122. Bruunsgaard H, Andersen-Ranberg K, Jeune B, et al. (1999) A high plasma concentration of TNF-alpha is associated with dementia in centenarians. J Gerontol A Biol Sci Med Sci. 54:M357–M64. 123. Tan ZS, Beiser AS, Vasan RS, et al. (2007) Inflammatory markers and the risk of Alzheimer disease: the Framingham Study. Neurology. 68(22):1902–8. 124. Akiyama H, Barger S, Barnum S, et al. (2000) Inflammation and Alzheimer’s disease. Neurobiol Aging. 21(3):383–421. 125. Dik MG, Jonker C, Hack CE, et al. (2005) Serum inflammatory proteins and cognitive decline in older persons. Neurology. 64(8):1371–7. 126. Rogers J, Kirby LC, Hempelman SR, et al. (1993) Clinical trial of indomethacin in Alzheimer’s disease. Neurology. 43(8):1609–11. 127. Scharf S, Mander A, Ugoni A, et al. (1999) A double-blind, placebo-controlled trial of diclofenac/misoprostol in Alzheimer’s disease. Neurology. 53:197–201. 128. Aisen PS, Schafer KA, Grundman M, et al. (2003) Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA. 289: 2819–26. 129. Balgi M, Papassotiropoulos A, Hampel H, et al. (2003) Polymorphisms of the gene encoding the inflammatory cytokine interleukin-6 determine the magnitude of the increase in soluble interleukin-6 receptor levels in Alzheimer’s disease. Eur Arch Psych Clin Neurosci. 252:44–8.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

147

130. Singh VK, Guthikonda P. (1997) Circulating cytokines in Alzheimer’s disease. J Psychiatr Res. 31:657–60. 131. VanDuijn CM, Hofman A, Nagelkerken L. (1990) Serum levels of interleukin-6 are not elevated in patients with Alzheimer’s disease. Neurosci Lett. 108:350–4. 132. Richartz E, Stransky E, Batra A, et al. (2005) Decline of immune responsiveness: a pathogenic factor in Alzheimer’s disease? J Psychiatr Res. 39:535–43. 133. Lanzrein AS, Johnston CM, Perry VH, et al. (1998) Longitudinal study of inflammatory factors in serum, cerebrospinal fluid, and brain tissue in Alzheimer’s disease: interleukin-1b, interleukin-6, interleukin-1 receptor antagonist, tumor necrosis factora, the soluble tumor necrosis factor receptors I and II, and a-1-antichymotrypsin. Alzheimer Dis Assoc Disord. 12:215–27. 134. Prichep LS, John ER, Ferris SH, et al. (2006) Prediction of longitudinal cognitive decline in normal elderly with subjective complaints using electrophysiological imaging. Neurobiol Aging. 27:471–81. 135. Huang C, Wahlund L, Dierks T, et al. (2000) Discrimination of Alzheimer’s disease and mild cognitive impairment by equivalent EEG sources: a cross-sectional and longitudinal study. Clin Neurophysiol. 111(11):1961–7. 136. Babiloni C, Ferri R, Binetti G, et al. (2009) Directionality of EEG synchronization in Alzheimer’s disease subjects. Neurobiol Aging. 30(1):93–102. 137. Moretti DV, Babiloni C, Binetti G, et al. (2004) Individual analysis of EEG frequency and band power in mild Alzheimer’s disease. Clin Neurophysiol. 115 (2):299–308. 138. Hogan MJ, Swanwick GR, Kaiser J, et al. (2003) Memory-related EEG power and coherence reductions in mild Alzheimer’s disease. Int J Psychophysiol. 49(2):147–63. 139. Rodriguez G, Nobili F, Rocca G, et al. (1998) Quantitative electroencephalography and regional cerebral blood flow: discriminant analysis between Alzheimer’s patients and healthy controls. Dement Geriatr Cogn Disord. 9(5):274–83. 140. Sneddon R, Shankle WR, Hara J, et al. (2005) EEG detection of early Alzheimer’s disease using psychophysical tasks. Clin EEG Neurosci. 36(3):141–50. 141. Petrosian AA, Prokhorov DV, Lajara-Nanson W, et al. (2001) Recurrent neural networkbased approach for early recognition of Alzheimer disease in EEG. Clin Neurophysiol. 112:1378–87. 142. Rossini PM, Buscema M, Capriotti M, et al. (2008) Is it possible to automatically distinguish resting EEG data of normal elderly vs. mild cognitive impairment subjects with high degree of accuracy? 119(7):1534–45. 143. Ponomareva NV, Korovaitseva GI, Rogaev EI. (2008) EEG alterations in non-demented individuals related to apolipoprotein E genotype and to risk of Alzheimer disease. Neurobiol Aging. 29(6):819–27. 144. Okayasu M, Tanaka H, Saisu A, et al. (2006) P34-5 Evaluation of donepezil efficacy for Alzheimer’s disease with EEG and MLRs. Clinical Neurophysiology. 117(Suppl 1):164. 145. Babiloni C, Cassetta E, Dal FG, et al. (2006) Donepezil effects on sources of cortical rhythms in mild Alzheimer’s disease: responders vs non-responders. Neuroimage. 31(4):1650–65. 146. Adler G, Hennig O, Grips E, et al. (2006) P3-456: differential quantitative EEG effects of rivastigmine and memantine in Alzheimer patients. Alzheimers Dement. 2(3 Suppl 1):S510. 147. Whitehouse PJ, Price DL, Clark AW, et al. (1981) Alzheimer disease: evidence for selective loss of cholinergic neurons in the nucleus basalis. Ann Neurol. 10:122–6. 148. Potter P, Pandya Y, et al. (2004) Cortical cholinergic deficit is associated with plaque development at preclinical stages of Alzheimer’s disease. Neurobiol Aging. 25(S2):79.

c04

BLBK249-Cutler

148

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

149. Ladner CJ, Lee JM. (1998) Pharmacological drug treatment of Alzheimer disease: the cholinergic hypothesis revisited. J Neuropathol Exp Neurol. 57(8):719–31. 150. Fisher A. (1997) Muscarinic agonists for the treatment of Alzheimer’s disease: progress and perspectives. Expert Opin Investig Drugs. 6(10):1395–411. 151. Volpicelli LA, Levey AI. (2004) Muscarinic acetylcholine receptor subtypes in cerebral cortex and hippocampus. Prog Brain Res. 145:59–66. 152. Quirion R, Aubert I, Lapchak PA, et al. (1989) Muscarinic receptor subtypes in human neurodegenerative disorders: focus on Alzheimer’s disease. Trends Pharmacol Sci. Suppl:80–4. 153. Svensson AL, Alafuzoff I, Nordberg A. (1992) Characterization of muscarinic receptor subtypes in Alzheimer and control brain cortices by selective muscarinic antagonists. Brain Res. 596(1–2):142–8. 154. Greenlee W, Clader J, Asberom T, et al. (2001) Muscarinic agonists and antagonists in the treatment of Alzheimer’s disease. Farmaco. 56(4):247–50. 155. Fisher A. (2007) M1 muscarinic agonists target major hallmarks of Alzheimer’s disease–an update. Curr Alzheimer Res. 4(5):577–80. 156. Fisher A, Brandeis R, Bar-Ner RH, et al. (2002) AF150(S) and AF267B: M1 muscarinic agonists as innovative therapies for Alzheimer’s disease. J Mol Neurosci. 19(1–2):145– 53. 157. Chandra JN, Malviya M, Sadashiva CT, et al. (2008) Effect of novel arecoline thiazolidinones as muscarinic receptor 1 agonist in Alzheimer’s dementia models. Neurochem Int. 52(3):376–83. 158. Sabbagh MN, Farlow MR, Relkin N, et al. (2006) Do cholinergic therapies have disease-modifying effects in Alzheimer’s disease? Alzheimer’s and Dementia. 2(2):118– 25. 159. Sramek JJ, Cutler NR, Hurley DJ, et al. (1995) The utility of salivary amylase as an evaluation of M3 muscarinic agonist activity in Alzheimer’s disease. Prog Neuropsychopharmacol Biol Psychiatry. 19(1):85–91. 160. Mirza NR, Peters D, Sparks RG. (2003) Xanomeline and the antipsychotic potential of muscarinic receptor subtype selective agonists. CNS Drug Rev. 9(2):159–86. 161. Arendt T. Diagnosis of Alzheimer’s Disease Using Peripheral Blood Lymphocyte Expression of CD69 Following a Mitogenic Stimulus. Alzheimer’s Association’s International Conference on Alzheimer’s Disease (ICAD); July 30, 2008; Chicago, IL. 162. Kusdra L, Rempel H, Yaffe K, et al. (2000) Elevation of CD69+ monocyte/macrophages in patients with Alzheimer’s disease. Immunobiology. 202(1):26–33. 163. Cacabelos R. (2008) Pharmacogenomics in Alzheimer’s disease. Methods Mol Biol. 448:213–357. 164. Cacabelos R, Llovo R, Fraile C, et al. (2007) Pharmacogenetic aspects of therapy with cholinesterase inhibitors: the role of CYP2D6 in Alzheimer’s disease pharmacogenetics. Curr Alzheimer Res. 4(4):479–500. 165. Kuusisto J, Koivisto K, Kervinen K, et al. (1994) Association of apolipoprotein E phenotypes with late onset Alzheimer’s disease: population based study. Bmj. 309(6955): 636–8. 166. Choi SH, Kim SY, Na HR, et al. (2008) Effect of ApoE genotype on response to donepezil in patients with Alzheimer’s disease. Dement Geriatr Cogn Disord. 25(5):445–50. 167. Sparks DL, Connor DJ, Sabbagh MN, et al. (2006) Circulating cholesterol levels, apolipoprotein E genotype and dementia severity influence the benefit of atorvastatin treatment in Alzheimer’s disease: results of the Alzheimer’s Disease CholesterolLowering Treatment (ADCLT) trial. Acta Neurol Scand Suppl. 185:3–7.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

149

168. Risner ME, Saunders AM, Altman JF, et al. (2006) Efficacy of rosiglitazone in a genetically defined population with mild-to-moderate Alzheimer’s disease. Pharmacogenomics J. 6(4):246–54. 169. Blesa R, Aguilar M, Casanova JP, et al. (2006) Relationship between the efficacy of rivastigmine and apolipoprotein E (epsilon4) in patients with mild to moderately severe Alzheimer disease. Alzheimer Dis Assoc Disord. 20(4):248–54. 170. Akuffo EL, Davis JB, Fox SM, et al. (2008) The discovery and early validation of novel plasma biomarkers in mild-to-moderate Alzheimer’s disease patients responding to treatment with rosiglitazone. Biomarkers. 13(6):618–36. 171. Mantella RC, Butters MA, Amico JA, et al. (2008) Salivary cortisol is associated with diagnosis and severity of late-life generalized anxiety disorder. Psychoneuroendocrinology. 33(6):773–81. 172. Pomara N, Willoughby LM, Sidtis JJ, et al. (2005) Cortisol response to diazepam: its relationship to age, dose, duration of treatment, and presence of generalized anxiety disorder. Psychopharmacology (Berl). 178(1):1–8. 173. Duggan M, Dowd N, O’Mara D, et al. (2002) Benzodiazepine premedication may attenuate the stress response in daycase anesthesia: a pilot study. Can J Anaesth. 49(9):932– 5. 174. Tafet GE, Feder DJ, Abulafia DP, et al. (2005) Regulation of hypothalamic-pituitaryadrenal activity in response to cognitive therapy in patients with generalized anxiety disorder. Cogn Affect Behav Neurosci. 5(1):37–40. 175. Takai N, Yamaguchi M, Aragaki T, et al. (2007) Gender-specific differences in salivary biomarker responses to acute psychological stress. Ann NY Acad Sci. 1098:510–5. 176. Van Den Bergh BR, Van Calster B, Pinna Puissant S, et al. (2008) Self-reported symptoms of depressed mood, trait anxiety and aggressive behavior in post-pubertal adolescents: associations with diurnal cortisol profiles. Horm Behav. 54(2):253–7. 177. Schlotz W, Schulz P, Hellhammer J, et al. (2006) Trait anxiety moderates the impact of performance pressure on salivary cortisol in everyday life. Psychoneuroendocrinology. 31(4):459–72. 178. Scholey AB, Robertson B, Haskell CF, et al. (2008) Effects of chewing gum on subjective and physiological stress responses. Appetite. 50(2–3):565. 179. Noto Y, Sato T, Kudo M, et al. (2005) The relationship between salivary biomarkers and state-trait anxiety inventory score under mental arithmetic stress: a pilot study. Anesth Analg. 101(6):1873–6. 180. Rohleder N, Nater UM, Wolf JM, et al. (2004) Psychosocial stress-induced activation of salivary alpha-amylase: an indicator of sympathetic activity? Ann NY Acad Sci. 1032: 258–63. 181. Takai N, Yamaguchi M, Aragaki T, et al. (2005) Effect of psychological stress on the salivary cortisol and amylase levels in healthy young adults. Anesth Analg. 101(6):1873–6. 182. Stenzel-Poore MP, Heinrichs SC, Rivest S, et al. (1994) Overproduction of corticotropinreleasing factor in transgenic mice: a genetic model of anxiogenic behavior. J Neurosci. 14(5 Pt 1):2579–84. 183. Dirks A, Groenink L, Bouwknecht JA, et al. (2002) Overexpression of corticotropinreleasing hormone in transgenic mice and chronic stress-like autonomic and physiological alterations. Eur J Neurosci. 16(9):1751–60. 184. Bremner JD, Licinio J, Darnell A, et al. (1997) Elevated CSF corticotropin-releasing factor concentrations in posttraumatic stress disorder. Am J Psychiatry. 154(5):624–9. 185. Heim C, Newport DJ, Heit S, et al. (2000) Pituitary-adrenal and autonomic responses to stress in women after sexual and physical abuse in childhood. JAMA. 284(5):592–7.

c04

BLBK249-Cutler

150

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

186. Holsboer F, Ising M. (2008) Central CRH system in depression and anxiety–evidence from clinical studies with CRH1 receptor antagonists. Eur J Pharmacol. 583(2-3):350–7. 187. Ising M, Holsboer F. (2007) CRH-sub-1 receptor antagonists for the treatment of depression and anxiety. Exp Clin Psychopharmacol. 15(6):519–28. 188. Mitchell AJ. (1998) The role of corticotropin releasing factor in depressive illness: a critical review. Neurosci Biobehav Rev. 22(5):635–51. 189. Fossey MD, Lydiard RB, Ballenger JC, et al. (1996) Cerebrospinal fluid corticotropinreleasing factor concentrations in patients with anxiety disorders and normal comparison subjects. Biol Psychiatry. 39(8):703–7. 190. Arborelius L, Owens MJ, Plotsky PM, et al. (1999) The role of corticotropin-releasing factor in depression and anxiety disorders. J Endocrinol. 160(1):1–12. 191. Ditzen C, Jastorff AM, Kessler MS, et al. (2006) Protein biomarkers in a mouse model of extremes in trait anxiety. Mol Cell Proteomics. 5(10):1914–20. ¨ 192. Kromer SA, Kessler MS, Milfay D, et al. (2005) Identification of glyoxalase-I as a protein marker in a mouse model of extremes in trait anxiety. J Neurosci. 25(17):4375– 84. 193. Chen F, Wollmer MA, Hoerndli F, et al. (2004) Role for glyoxalase I in Alzheimer’s disease. Proc Natl Acad Sci USA. 101(20):7687–92. 194. Tanna VL, Wilson AF, Winokur G, et al. (1989) Linkage analysis of pure depressive disease. J Psychiatr Res. 23(2):99–107. 195. Thornalley PJ. (2006) Unease on the role of glyoxalase 1 in high-anxiety-related behaviour. Trends Mol Med. 12(5):195–9. 196. Mathew RJ, Ho BT, Kralik P, et al. (1980) Catechol-O-methyltransferase and catecholamines in anxiety and relaxation. Psychiatry Res. 3(1):85–91. 197. Fell D, Derbyshire DR, Maile CJ, et al. (1985) Measurement of plasma catecholamine concentrations; an assessment of anxiety. Br J Anaesth. 57(8):770–4. 198. Young EA, Breslau N. (2004) Cortisol and catecholamines in posttraumatic stress disorder: an epidemiologic community study. Arch Gen Psychiatry. 61(4):394–401. 199. Taylor J, Harris N, Krieman M, et al. (1989) Effects of buspirone on plasma catecholamines, heart rate, and blood pressure in stressed and nonstressed rats. Pharmacol Biochem Behav. 34(2):349–53. 200. Goldstein DS, Dionne R, Sweet J, et al. (1982) Circulatory, plasma catecholamine, cortisol, lipid, and psychological responses to a real-life stress (third molar extractions): effects of diazepam sedation and of inclusion of epinephrine with the local anesthetic. Psychosom Med. 44(3):259–72. 201. Krieman MJ, Hershock DM, Greenberg IJ, et al. (1992) Effects of adinazolam on plasma catecholamine, heart rate and blood pressure responses in stressed and non-stressed rats. Neuropharmacology. 31(1):33–8. 202. Van FI, Popp-Snijders C, Nauta JJ, et al. (1992) Platelet catecholamine contents as related to trait anxiety and aerobic fitness. Am J Physiol. 263(2 Pt 1):E245–9. 203. Kanto J, Scheinin M. (1991) Biochemical assessment of preoperative stress: a study with diazepam and measurement of monoamine metabolites and catecholamines in cerebrospinal fluid and plasma. Br J Anaesth. 66(5):587–90. 204. Pouttu J, Tuominen M, Scheinin M, et al. (1989) Effects of oral clonidine premedication on concentrations of cortisol and monoamine neurotransmitters and their metabolites in cerebrospinal fluid and plasma. Acta Anaesthesiol Scand. 33(2):137–41. 205. Kanto J, Scheinin M. (1992) Diazepam and atropine as premedicants: no discrimination by monoamine metabolite and catecholamine measurements in cerebrospinal fluid and plasma. Acta Anaesthesiol Scand. 36(1):80–3.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

151

206. Brawman-Mintzer O, Lydiard RB. (1997) Biological basis of generalized anxiety disorder. J Clin Psychiatry. 58(Suppl 3):16–25; discussion 6. 207. Pitts FN, Jr, McClure JN, Jr. (1967) Lactate metabolism in anxiety neurosis. N Engl J Med. 277(25):1329–36. 208. Carr DB, Sheehan DV, Surman OS, et al. (1986) Neuroendocrine correlates of lactateinduced anxiety and their response to chronic alprazolam therapy. Am J Psychiatry. 143(4):483–94. 209. Cowley DS, Arana GW. (1990) The diagnostic utility of lactate sensitivity in panic disorder. Arch Gen Psychiatry. 47(3):277–84. 210. Liebowitz MR, Coplan JD, Martinez J, et al. (1995) Effects of intravenous diazepam pretreatment on lactate-induced panic. Psychiatry Res. 58(2):127–38. 211. Shekhar A, Keim SR. (2000) LY354740, a potent group II metabotropic glutamate receptor agonist prevents lactate-induced panic-like response in panic-prone rats. Neuropharmacology. 39(7):1139–46. 212. Liebowitz MR, Fyer AJ, Gorman JM, et al. (1986) Alprazolam in the treatment of panic disorders. J Clin Psychopharmacol. 6(1):13–20. 213. Woods SW, Charney DS, Goodman WK, et al. (1988) Carbon dioxide-induced anxiety; behavioral, physiologic, and biochemical effects of carbon dioxide in patients with panic disorders and healthy subjects. Arch Gen Psychiatry. 45(1):43–52. 214. Harro J. (2006) CCK and NPY as anti-anxiety treatment targets: promises, pitfalls, and strategies. Amino Acids. 31(3):215–30. 215. Eser D, Leicht G, Lutz J, et al. (2009) Functional neuroanatomy of CCK-4-induced panic attacks in healthy volunteers. Hum Brain Mapp. 30(2):511–22. 216. Shlik J, Aluoja A, Vasar V, et al. (1997) Effects of citalopram treatment on behavioural, cardiovascular and neuroendocrine response to cholecystokinin tetrapeptide challenge in patients with panic disorder. J Psychiatry Neurosci. 22(5):332–40. 217. Zwanzger P, Eser D, Aicher S, et al. (2003) Effects of alprazolam on cholecystokinintetrapeptide-induced panic and hypothalamic-pituitary-adrenal-axis activity: a placebo-controlled study. Neuropsychopharmacology. 28(5):979–84. 218. Zwanzger P, Baghai TC, Schuele C, et al. (2001) Vigabatrin decreases cholecystokinintetrapeptide (CCK-4) induced panic in healthy volunteers. Neuropsychopharmacology. 25(5):699–703. 219. Zwanzger P, Eser D, Padberg F, et al. (2003) Effects of tiagabine on cholecystokinintetrapeptide (CCK-4)-induced anxiety in healthy volunteers. Depress Anxiety. 18(3): 140–3. 220. Wiedemann K, Jahn H, Yassouridis A, et al. (2001) Anxiolyticlike effects of atrial natriuretic peptide on cholecystokinin tetrapeptide-induced panic attacks: preliminary findings. Arch Gen Psychiatry. 58(4):371–7. 221. Kellner M, Muhtz C, Demiralay C, et al. (2009) The selective serotonin re-uptake inhibitor escitalopram modulates the panic response to cholecystokinin tetrapeptide in healthy men depending on 5-HTTLPR genotype. J Psychiatr Res. 43(6):642–8. ¨ C, Baghai T, et al. (2007) Evaluation of the CCK-4 model as a challenge 222. Eser D, Schule paradigm in a population of healthy volunteers within a proof-of-concept study. Psychopharmacology (Berl). 192(4):479–87. ˜ 223. Maron E, Toru I, Vasar V, et al. (2004) The effect of 5-hydroxytryptophan on cholecystokinin-4-induced panic attacks in healthy volunteers. J Psychopharmacol. 18(2):194–9. 224. de Visser SJ, Van Der Post JP, de Waal PP, et al. (2003) Biomarkers for the effects of benzodiazepines in healthy volunteers. Br J Clin Pharmacol. 55(1):39–50.

c04

BLBK249-Cutler

152

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

225. Tedeschi G, Smith AT, Dhillon S, et al. (1983) Rate of entrance of benzodiazepines into the brain determined by eye movement recording. Br J Clin Pharmacol. 15(1):103– 7. 226. Wu JC, Buchsbaum MS, Hershey TG, et al. (1991) PET in generalized anxiety disorder. Biol Psychiatry. 29(12):1181–99. 227. Whalen PJ, Johnstone T, Somerville LH, et al. (2008) A functional magnetic resonance imaging predictor of treatment response to venlafaxine in generalized anxiety disorder. Biol Psychiatry. 63(9):858–63. 228. Hoehn-Saric R, Schlund MW, Wong SH. (2004) Effects of citalopram on worry and brain activation in patients with generalized anxiety disorder. Psychiatry Res. 131(1):11– 21. 229. Bossini L, Tavanti M, Calossi S, et al. (2008) Magnetic resonance imaging volumes of the hippocampus in drug-na¨ıve patients with post-traumatic stress disorder without comorbidity conditions. J Psychiatr Res. 42(9):752–62. 230. Malizia AL, Cunningham VJ, Bell CJ, et al. (1998) Decreased brain GABA(A)benzodiazepine receptor binding in panic disorder: preliminary results from a quantitative PET study. Arch Gen Psychiatry. 55(8):715–20. 231. Donati RJ, Dwivedi Y, Roberts RC, et al. (2008) Postmortem brain tissue of depressed suicides reveals increased Gs alpha localization in lipid raft domains where it is less likely to activate adenylyl cyclase. J Neurosci. 28(12):3042–50. 232. Lee HJ, Cha JH, Ham BJ, et al. (2004) Association between a G-protein beta 3 subunit gene polymorphism and the symptomatology and treatment responses of major depressive disorders. Pharmacogenomics J. 4(1):29–33. 233. Zill P, Baghai TC, Zwanzger P, et al. (2000) Evidence for an association between a G-protein beta3-gene variant with depression and response to antidepressant treatment. Neuroreport. 11(9):1893–7. 234. Kang RH, Hahn SW, Choi MJ, et al. (2007) Relationship between G-protein beta-3 subunit C825T polymorphism and mirtazapine responses in Korean patients with major depression. Neuropsychobiology. 56(1):1–5. 235. Zill P, Baghai TC, Zwanzger P, et al. (2002) Association analysis of a polymorphism in the G-protein stimulatory alpha subunit in patients with major depression. Am J Med Genet. 114(5):530–2. 236. Raison CL, Capuron L, Miller AH. (2006) Cytokines sing the blues: inflammation and the pathogenesis of depression. Trends Immunol. 27(1):24–31. 237. Zorrilla EP, Luborsky L, McKay JR, et al. (2001) The relationship of depression and stressors to immunological assays: a meta-analytic review. Brain Behav Immun. 15(3):199– 226. 238. Jovanovic DJ, Simic-Ogrizovic SSO, Dopsaj VD, et al. Depression and markers of chronic inflammation in patients on peritoneal dialysis. 7th European Peritoneal Dialysis Meeting; October 15–18, 2005; Prague, Czech Republic. 239. Jacobson CM, Rosenfeld B, Pessin H, et al. (2008) Depression and IL-6 blood plasma concentrations in advanced cancer patients. Psychosomatics. 49(1):64–6. 240. Penninx BW, Kritchevsky SB, Yaffe K, et al. (2003) Inflammatory markers and depressed mood in older persons: results from the Health, Aging and Body Composition study. Biol Psychiatry. 54(5):566–72. 241. Alesci S, Martinez PE, Kelkar S, et al. (2005) Major depression is associated with significant diurnal elevations in plasma interleukin-6 levels, a shift of its circadian rhythm, and loss of physiological complexity in its secretion: clinical implications. J Clin Endocrinol Metab. 90(5):2522–30.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

153

242. Chourbaji S, Urani A, Inta I, et al. (2006) IL-6 knockout mice exhibit resistance to stressinduced development of depression-like behaviors. Neurobiol Dis. 23(3):587–94. 243. Basterzi AD, Aydemir C, Kisa C, et al. (2005) IL-6 levels decrease with SSRI treatment in patients with major depression. Hum Psychopharmacol. 20(7):473–6. 244. Xia Z, DePierre JW, N¨assberger L. (1996) Tricyclic antidepressants inhibit IL-6, IL-1 beta and TNF-alpha release in human blood monocytes and IL-2 and interferon-gamma in T cells. Immunopharmacology. 34(1):27–37. 245. Kubera M, Kenis G, Bosmans E, et al. (2004) Stimulatory effect of antidepressants on the production of IL-6. Int Immunopharmacol. 4(2):185–92. 246. Maes M, Calabrese J, Meltzer HY. (1994) The relevance of the in- versus outpatient status for studies on HPA-axis in depression: spontaneous hypercortisolism is a feature of major depressed inpatients and not of major depression per se. Prog Neuropsychopharmacol Biol Psychiatry. 18(3):503–17. 247. Dinan TG. (1994) Glucocorticoids and the genesis of depressive illness: a psychobiological model. Br J Psychiatry. 164(3):365–71. 248. Pariante CM. (2003) Depression, stress and the adrenal axis. J Neuroendocrinol. 15(8):811–2. 249. Lindqvist D, Isaksson A, Lil-Tr¨askman-Bendz, et al. (2008) Salivary cortisol and suicidal behavior—a follow-up study. Psychoneuroendocrinology. 33(8):1061–8. 250. Cowen PJ. (2002) Cortisol, serotonin and depression: all stressed out? Br J Psychiatry. 180:99–100. 251. Pariante CM, Thomas SA, Lovestone S, et al. (2004) Do antidepressants regulate how cortisol affects the brain? Psychoneuroendocrinology. 29(4):423–47. ¨ 252. Weber CC, Eckert GP, Muller WE. (2006) Effects of antidepressants on the brain/plasma distribution of corticosterone. Neuropsychopharmacology. 31(11):2443–8. 253. Meltzer H, Bastani B, Jayathilake K, et al. (1997) Fluoxetine, but not tricyclic antidepressants, potentiates the 5-hydroxytryptophan-mediated increase in plasma cortisol and prolactin secretion in subjects with major depression or with obsessive compulsive disorder. Neuropsychopharmacology. 17(1):1–11. 254. Sargent PA, Quested DJ, Cowen PJ. (1998) Clomipramine enhances the cortisol response to 5-HTP: implications for the therapeutic role of 5-HT2 receptors. Psychopharmacology (Berl). 140(1):120–2. 255. Laakmann G, Hennig J, Baghai T, et al. (2004) Mirtazapine acutely inhibits salivary cortisol concentrations in depressed patients. Ann NY Acad Sci. 1032:279–82. ¨ E, Bissette G, et al. (1984) Elevated concentrations of CSF 256. Nemeroff CB, Widerlov corticotropin-releasing factor-like immunoreactivity in depressed patients. Science. 226(4680):1342–4. 257. Brunner J, Bronisch T. (1999) Neurobiological correlates of suicidal behavior. Fortschr Neurol Psychiatr. 67(9):391–412. 258. Holsboer F, Barden N. (1996) Antidepressants and hypothalamic-pituitaryadrenocortical regulation. Endocr Rev. 17(2):187–205. 259. Raison CL, Miller AH. (2003) When not enough is too much: the role of insufficient glucocorticoid signaling in the pathophysiology of stress-related disorders. Am J Psychiatry. 160(9):1554–65. 260. Budziszewska B, Jaworska-Feil L, Tetich M, et al. (2004) Regulation of the human corticotropin-releasing-hormone gene promoter activity by antidepressant drugs in Neuro-2A and AtT-20 cells. Neuropsychopharmacology. 29(4):785–94. 261. Holsboer F. (1999) The rationale for corticotropin-releasing hormone receptor (CRH-R) antagonists to treat depression and anxiety. J Psychiatr Res. 33(3):181–214.

c04

BLBK249-Cutler

154

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

¨ 262. Zobel AW, Nickel T, Kunzel HE, et al. (2000) Effects of the high-affinity corticotropinreleasing hormone receptor 1 antagonist R121919 in major depression: the first 20 patients treated. J Psychiatr Res. 34(3):171–81. 263. Holsboer F. (1983) Prediction of clinical course by dexamethasone suppression test (DST) response in depressed patients—physiological and clinical construct validity of the DST. Pharmacopsychiatria. 16(6):186–91. 264. Ribeiro SC, Tandon R, Grunhaus L, et al. (1993) The DST as a predictor of outcome in depression: a meta-analysis. Am J Psychiatry. 150(11):1618–29. 265. Arana GW, Baldessarini RJ, Ornsteen M. (1985) The dexamethasone suppression test for diagnosis and prognosis in psychiatry: commentary and review. Arch Gen Psychiatry. 42(12):1193–204. 266. Holsboer F, von BU, Wiedemann K, et al. (1987) Serial assessment of corticotropinreleasing hormone response after dexamethasone in depression: implications for pathophysiology of DST nonsuppression. Biol Psychiatry. 22(2):228–34. 267. Holsboer F. (2000) The corticosteroid receptor hypothesis of depression. Neuropsychopharmacology. 23(5):477–501. 268. Keck ME, Holsboer F. (2001) Hyperactivity of CRH neuronal circuits as a target for therapeutic interventions in affective disorders. Peptides. 22(5):835–44. 269. Heuser I, Yassouridis A, Holsboer F. (1994) The combined dexamethasone/CRH test: a refined laboratory test for psychiatric disorders. J Psychiatr Res. 28(4):341–56. 270. Rybakowski JK, Twardowska K. (1999) The dexamethasone/corticotropin-releasing hormone test in depression in bipolar and unipolar affective illness. J Psychiatr Res. 33(5):363–70. 271. Watson S, Gallagher P, Del-Estal D, et al. (2002) Hypothalamic-pituitary-adrenal axis function in patients with chronic depression. Psychol Med. 32(6):1021–8. 272. Frieboes RM, Sonntag A, Yassouridis A, et al. (2003) Clinical outcome after trimipramine in patients with delusional depression—a pilot study. Pharmacopsychiatry. 36(1):12–7. 273. Rinne T, de KER, Wouters L, et al. (2003) Fluvoxamine reduces responsiveness of HPA axis in adult female BPD patients with a history of sustained childhood abuse. Neuropsychopharmacology. 28(1):126–32. 274. Nickel T, Sonntag A, Schill J, et al. (2003) Clinical and neurobiological effects of tianeptine and paroxetine in major depression. J Clin Psychopharmacol. 23(2):155–68. ¨ C, Baghai T, Zwanzger P, et al. (2003) Attenuation of hypothalamic-pituitary275. Schule adrenocortical hyperactivity in depressed patients by mirtazapine. Psychopharmacology (Berl). 166(3):271–5. 276. Deuschle M, Schmider J, Weber B, et al. (1997) Pulse-dosing and conventional application of doxepin: effects on psychopathology and hypothalamus-pituitary-adrenal (HPA) system. J Clin Psychopharmacol. 17(3):156–60. 277. Kasckow JW, Baker D, Geracioti TD, Jr. (2001) Corticotropin-releasing hormone in depression and post-traumatic stress disorder. Peptides. 22(5):845–51. 278. Schmider J, Lammers CH, Gotthardt U, et al. (1995) Combined dexamethasone/ corticotropin-releasing hormone test in acute and remitted manic patients, in acute depression, and in normal controls: I. Biol Psychiatry. 38(12):797–802. 279. Schreiber W, Lauer CJ, Krumrey K, et al. (1996) Dysregulation of the hypothalamicpituitary-adrenocortical system in panic disorder. Neuropsychopharmacology. 15(1):7– 15. 280. Lammers CH, Garcia-Borreguero D, Schmider J, et al. (1995) Combined dexamethasone/corticotropin-releasing hormone test in patients with schizophrenia and in normal controls: II. Biol Psychiatry. 38(12):803–7.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

155

281. Yanovski JA, Cutler GB, Jr, Chrousos GP, et al. (1998) The dexamethasone-suppressed corticotropin-releasing hormone stimulation test differentiates mild Cushing’s disease from normal physiology. J Clin Endocrinol Metab. 83(2):348–52. ¨ 282. Then BF, Kumpfel T, Trenkwalder C, et al. (1999) Dysregulation of the hypothalamopituitary-adrenal axis is related to the clinical course of MS. Neurology. 53(4):772–7. 283. Schildkraut JJ. (1965) The catecholamine hypothesis of affective disorders: a review of supporting evidence. Am J Psychiatry. 122(5):509–22. 284. Nutt DJ. (2002) The neuropharmacology of serotonin and noradrenaline in depression. Int Clin Psychopharmacol. 17(Suppl 1):S1–12. 285. Zajecka JM, Albano D. (2004) SNRIs in the management of acute major depressive disorder. J Clin Psychiatry. 65(Suppl 17):11–8. 286. Freis E. (1954) Mental depression in hypertensives treated for long periods with large doses of reserpine. N Engl J Med. 251:1006–8. 287. Lingjaerde O. (1963) Tetrabenazine (Nitroman) in the treatment of psychoses. Acta Psychiatr Scand. 39(Suppl 170):1–109. 288. Delgado PL, Miller HL, Salomon RM, et al. (1993) Monoamines and the mechanism of antidepressant action: effects of catecholamine depletion on mood of patients treated with antidepressants. Psychopharmacol Bull. 29(3):389–96. 289. Charney DS. (1998) Monoamine dysfunction and the pathophysiology and treatment of depression. J Clin Psychiatry. 59(Suppl 14):11–4. 290. Ryu SH, Lee SH, Lee HJ, et al. (2004) Association between norepinephrine transporter gene polymorphism and major depression. Neuropsychobiology. 49(4):174–7. ¨ ¨ M. (2006) The norepinephrine transporter in physiology and disease. 291. Bonisch H, Bruss Handb Exp Pharmacol. 175:485–524. 292. Klimek V, Stockmeier C, Overholser J, et al. (1 Nov 1997) Reduced levels of norepinephrine transporters in the locus coeruleus in major depression. J Neurosci. 17(21): 8451–8. 293. Warsh JJ, Hasey G, Cooke RG, et al. (1985) Elevated 3,4-dihydroxyphenylethyleneglycol (DHPG) excretion in dexamethasone resistant depressed patients. Prog Neuropsychopharmacol Biol Psychiatry. 9(5–6):661–4. 294. Vincent S, Bieck PR, Garland EM, et al. (29 Jun 2004) Clinical assessment of norepinephrine transporter blockade through biochemical and pharmacological profiles. Circulation. 109(25):3202–7. 295. Nemeroff CB, Owens MJ. (2003) Neuropharmacology of paroxetine. Psychopharmacol Bull. 37(Suppl 1):8–18. 296. Zhao Z, Zhang HT, Bootzin E, et al. (2009) Association of changes in norepinephrine and serotonin transporter expression with the long-term behavioral effects of antidepressant drugs. Neuropsychopharmacology. 34(6):1467–81. 297. Zhao Z, Baros AM, Zhang HT, et al. (2008) Norepinephrine transporter regulation mediates the long-term behavioral effects of the antidepressant desipramine. Neuropsychopharmacology. 33(13):3190–200. 298. Bieck PR, Potter WZ. (2005) Biomarkers in psychotropic drug development: integration of data across multiple domains. Annu Rev Pharmacol Toxicol. 45:227–46. 299. Leibowitz M EL, Lin Q, Ledent E, et al. Use of a biomarker of norepinephrine transporter (NET) inhibition to assess atomoxetine effects during clinically recommended dosing. Proceedings of the 18th ECNP Congress; October 22–26, 2005; Amsterdam, The Netherlands. 300. Krishnan KR. (2007) Revisiting monoamine oxidase inhibitors. J Clin Psychiatry. 68(Suppl 8):35–41.

c04

BLBK249-Cutler

156

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

301. Henkel V, Mergl R, Allgaier AK, et al. (2006) Treatment of depression with atypical features: a meta-analytic approach. Psychiatry Res. 141(1):89–101. 302. Dong WX, Ni XL. (2002) Norepinephrine metabolism in neuron: dissociation between 3,4-dihydroxyphenylglycol and 3,4-dihydroxymandelic acid pathways. Acta Pharmacol Sin. 23(1):59–65. 303. Eisenhofer G, Kopin IJ, Goldstein DS. (2004) Catecholamine metabolism: a contemporary view with implications for physiology and medicine. Pharmacol Rev. 56(3):331– 49. 304. Shannon JR, Flattem NL, Jordan J, et al. (2000) Orthostatic intolerance and tachycardia associated with norepinephrine-transporter deficiency. N Engl J Med. 342(8):541–9. 305. Lenders JW, Eisenhofer G, Abeling NG, et al. (1996) Specific genetic deficiencies of the A and B isoenzymes of monoamine oxidase are characterized by distinct neurochemical and clinical phenotypes. J Clin Invest. 97(4):1010–9. 306. Patat A, le Coz F, Dubruc C, et al. (1996) Pharmacodynamics and pharmacokinetics of two dose regimens of befloxatone, a new reversible and selective monoamine oxidase inhibitor, at steady state in healthy volunteers. J Clin Pharmacol. 36(3):216–29. 307. Heinonen EH, Savij¨arvi M, Kotila M, et al. (1993) Effects of monoamine oxidase inhibition by selegiline on concentrations of noradrenaline and monoamine metabolites in CSF of patients with Alzheimer’s disease. J Neural Transm Park Dis Dement Sect. 5(3):193–202. 308. Pac´ak K, Palkovits M, Kvetnansky´ R, et al. (1993) Effects of single or repeated immobilization on release of norepinephrine and its metabolites in the central nucleus of the amygdala in conscious rats. Neuroendocrinology. 57(4):626–33. 309. Eisenhofer G, Meredith IT, Ferrier C, et al. (1991) Increased plasma dihydroxyphenylalanine during sympathetic activation in humans is related to increased norepinephrine turnover. J Lab Clin Med. 117(4):266–73. 310. Nakada T, Sasagawa I, Kubota Y, et al. (1996) Dihydroxyphenylglycol in pheochromocytoma: its diagnostic use for norepinephrine dominant tumor. J Urol. 155(1):14–8. 311. Duncan MW, Compton P, Lazarus L, et al. (1988) Measurement of norepinephrine and 3,4-dihydroxyphenylglycol in urine and plasma for the diagnosis of pheochromocytoma. N Engl J Med. 319(3):136–42. 312. Schatzberg AF, Samson JA, Bloomingdale KL, et al. (1989) Toward a biochemical classification of depressive disorders- X: urinary catecholamines, their metabolites, and D-type scores in subgroups of depressive disorders. Arch Gen Psychiatry. 46(3):260–8. 313. Karege F, Bovier P, Gaillard J, et al. (1991) Plasma MHPG and AMDP depression relations, evolution and drug effect in a follow up study of depressed patients. Hum Psychopharmacol. 1:11–7. 314. Sheline Y, Bardgett ME, Csernansky JG. (1997) Correlated reductions in cerebrospinal fluid 5-HIAA and MHPG concentrations after treatment with selective serotonin reuptake inhibitors. J Clin Psychopharmacol. 17(1):11–4. 315. De BMD, Geracioti TD, Jr, Altemus M, et al. (1993) Cerebrospinal fluid monoamine metabolites in fluoxetine-treated patients with major depression and in healthy volunteers. Biol Psychiatry. 33(8–9):636–41. 316. Bowden CL, Koslow SH, Hanin I, et al. (1985) Effects of amitriptyline and imipramine on brain amine neurotransmitter metabolites in cerebrospinal fluid. Clin Pharmacol Ther. 37(3):316–24. 317. Nordin C, Bertilsson L, Siwers B. (1987) Clinical and biochemical effects during treatment of depression with nortriptyline: the role of 10-hydroxynortriptyline. Clin Pharmacol Ther. 42(1):10–9.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

157

318. Deakin JF. (2003) Depression and antisocial personality disorder: two contrasting disorders of 5HT function. J Neural Transm Suppl. (64):79–93. 319. Peabody CA, Faull KF, King RJ, et al. (1987) CSF amine metabolites and depression. Psychiatry Res. 21(1):1–7. 320. Raison CL, Borisov AS, Majer M, et al. (2009) Activation of central nervous system inflammatory pathways by interferon-alpha: relationship to monoamines and depression. Biol Psychiatry. 65(4):296–303. ¨ AL, et al. (2008) HPT axis, CSF monoamine 321. Jokinen J, Samuelsson M, Nordstrom metabolites, suicide intent and depression severity in male suicide attempters. J Affect Disord. 111(1):119–24. ¨ AL, et al. (2006) CSF 5-HIAA, suicide intent and 322. Samuelsson M, Jokinen J, Nordstrom hopelessness in the prediction of early suicide in male high-risk suicide attempters. Acta Psychiatr Scand. 113(1):44–7. 323. Chertkow Y, Weinreb O, Youdim MB, et al. (2007) Dopamine and serotonin metabolism in response to chronic administration of fluvoxamine and haloperidol combined treatment. J Neural Transm. 114(11):1443–54. 324. Nikisch G, Math´e AA, Czernik A, et al. (2004) Stereoselective metabolism of citalopram in plasma and cerebrospinal fluid of depressive patients: relationship with 5-HIAA in CSF and clinical response. J Clin Psychopharmacol. 24(3):283–90. 325. Pollier F, Sarre S, Aguerre S, et al. (2000) Serotonin reuptake inhibition by citalopram in rat strains differing for their emotionality. Neuropsychopharmacology. 22(1):64–76. 326. Carpenter LL, Anderson GM, Siniscalchi JM, et al. (2003) Acute changes in cerebrospinal fluid 5-HIAA following oral paroxetine challenge in healthy humans. Neuropsychopharmacology. 28(2):339–47. 327. Liang Y, Shaw AM, Boules M, et al. (2008) Antidepressant-like pharmacological profile of a novel triple re-uptake inhibitor, PRC200-SS. J Pharmacol Exp Ther. 327(2):573–83. 328. Potter WZ, Scheinin M, Golden RN, et al. (1985) Selective antidepressants and cerebrospinal fluid: lack of specificity on norepinephrine and serotonin metabolites. Arch Gen Psychiatry. 42(12):1171–7. 329. Linnoila M, Miller TL, Bartko J, et al. (1984) Five antidepressant treatments in depressed patients: effects on urinary serotonin and 5-hydroxyindoleacetic acid output. Arch Gen Psychiatry. 41(7):688–92. 330. van Wijk M, Meisch JJ, Korf J. (1977) Metabolism of 5-hydroxytryptamine and levels of tricyclic antidepressant drugs in rat brain after acute and chronic treatment. Psychopharmacology (Berl). 55(3):217–23. 331. Egashira T, Takayama F, Yamanaka Y. (1999) Changes in monoamine metabolites concentrations in rat cerebrospinal fluid after acute and long-term administration of a selective serotonin reuptake inhibitor, trazodone. Pharmacol Res. 40(6):503–8. 332. B¨ackman J, Alling C, Als´en M, et al. (2000) Changes of cerebrospinal fluid monoamine metabolites during long-term antidepressant treatment. Eur Neuropsychopharmacol. 10(5):341–9. 333. Spreux-Varoquaux O, Alvarez JC, Berlin I, et al. (2001) Differential abnormalities in plasma 5-HIAA and platelet serotonin concentrations in violent suicide attempters: relationships with impulsivity and depression. Life Sci. 69(6):647–57. 334. Asberg M, Tr¨askman L. (1981) Studies of CSF 5-HIAA in depression and suicidal behaviour. Adv Exp Med Biol. 133:739–52. ¨ AL, Nordstrom ¨ P. (2007) The relationship between CSF HVA/ 335. Jokinen J, Nordstrom 5-HIAA ratio and suicide intent in suicide attempters. Arch Suicide Res. 11(2):187– 92.

c04

BLBK249-Cutler

158

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

¨ G, Alling C, Blennow K, et al. (1999) Reduced cerebrospinal HVA concen336. Engstrom trations and HVA/5-HIAA ratios in suicide attempters: monoamine metabolites in 120 suicide attempters and 47 controls. Eur Neuropsychopharmacol. 9(5):399–405. 337. Sher L, Mann JJ, Traskman-Bendz L, et al. (2006) Lower cerebrospinal fluid homovanillic acid levels in depressed suicide attempters. J Affect Disord. 90(1):83–9. 338. Roy A, Agren H, Pickar D, et al. (1986) Reduced CSF concentrations of homovanillic acid and homovanillic acid to 5-hydroxyindoleacetic acid ratios in depressed patients: relationship to suicidal behavior and dexamethasone nonsuppression. Am J Psychiatry. 143(12):1539–45. 339. Placidi GP, Oquendo MA, Malone KM, et al. (2001) Aggressivity, suicide attempts, and depression: relationship to cerebrospinal fluid monoamine metabolite levels. Biol Psychiatry. 50(10):783–91. 340. M˚artensson B, W¨agner A, Beck O, et al. (1991) Effects of clomipramine treatment on cerebrospinal fluid monoamine metabolites and platelet 3H-imipramine binding and serotonin uptake and concentration in major depressive disorder. Acta Psychiatr Scand. 83(2):125–33. ¨ 341. Asberg M, Ringberger VA, Sjoqvist F, et al. (1977) Monoamine metabolites in cerebrospinal fluid and serotonin uptake inhibition during treatment with chlorimipramine. Clin Pharmacol Ther. 21(2):201–7. 342. Jacobs E, Fujii D, Schiffman J, et al. (2008) An exploratory analysis of neurocognition in methamphetamine-induced psychotic disorder and paranoid schizophrenia. Cogn Behav Neurol. 21(2):98–103. 343. Featherstone RE, Kapur S, Fletcher PJ. (2007) The amphetamine-induced sensitized state as a model of schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry. 31(8): 1556–71. 344. Carlsson M, Carlsson A. (1990) Schizophrenia: a subcortical neurotransmitter imbalance syndrome? Schizophr Bull. 16(3):425–32. 345. Dargham A. (2002) Recent evidence for dopamine abnormalities in schizophrenia. European Psychiatry. 17(Suppl 4):341–7. 346. Reynolds GP. (2008) The Neurochemistry of Schizophrenia. Psychiatry. 7(10):425–9. 347. Laruelle M, Abi-Dargham A, van Dyck CH, et al. (1996) Single photon emission computerized tomography imaging of amphetamine-induced dopamine release in drug-free schizophrenic subjects. Proc Natl Acad Sci USA. 93(17):9235–40. 348. Breier A, Su TP, Saunders R, et al. (1997) Schizophrenia is associated with elevated amphetamine-induced synaptic dopamine concentrations: evidence from a novel positron emission tomography method. Proc Natl Acad Sci USA. 94(6):2569–74. 349. Howes OD, Montgomery AJ, Asselin MC, et al. (2006) The pre-synaptic dopaminergic system before and after the onset of psychosis: initial results. Eur Neuropsychopharmacol. 16(Suppl 4):S177. 350. Kapur S. (2003) Psychosis as a state of aberrant salience: a framework linking biology, phenomenology, and pharmacology in schizophrenia. Am J Psychiatry. 160(1):13–23. 351. Nikam SS, Awasthi AK. (2008) Evolution of schizophrenia drugs: a focus on dopaminergic systems. Curr Opin Investig Drugs. 9(1):37–46. 352. Awad AG. (1994) Prediction research of neuroleptic treatment outcome in schizophrenia—state of the art 1978–1993. In: Gaebel W, Awad AG, editors. Prediction of Neuroleptic Treatment Outcome in Schizophrenia; Concepts and Methods. SpringerVerlag, Wien, pp 1–14. 353. Jordan S, Koprivica V, Dunn R, et al. (2004) In vivo effects of aripiprazole on cortical and striatal dopaminergic and serotonergic function. Eur J Pharmacol. 483(1):45–53.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

159

354. Akiyama K, Tsuchida K, Kanzaki A, et al. (1995) Plasma homovanillic acid levels and therapeutic outcome in schizophrenics: comparisons of neuroleptic-naive first-episode patients and patients with disease exacerbation due to neuroleptic discontinuance. Biol Psychiatry. 38(10):639–48. 355. Chang WH, Hwu HG, Chen TY, et al. (1993) Plasma homovanillic acid and treatment response in a large group of schizophrenic patients. Schizophr Res. 10(3):259–65. 356. Sumiyoshi T, Hasegawa M, Jayathilake K, et al. (1997) Prediction of short-term changes in symptom severity by baseline plasma homovanillic acid levels in schizophrenic patients receiving clozapine. Psychiatry Res. 69(2–3):113–21. 357. Nagaoka S, Iwamoto N, Arai H. (1997) First-episode neuroleptic-free schizophrenics: concentrations of monoamines and their metabolites in plasma and their correlations with clinical responses to haloperidol treatment. Biol Psychiatry. 41(8):857–64. 358. Lieberman JA. (1994) Predictors of outcome in schizophrenia: the concept of time. In: Gaebel W, Awad AG, editors. Prediction of Neuroleptic Treatment Outcome in Schizophrenia: Concepts and Methods. Springer-Verlag, Wien, pp 43–9. ¨ 359. Muller-Spahn F, Eich P, Hock C. (1996) The psychobiology of the acute schizophrenic episode. Int Clin Psychopharmacol. 11(Suppl 2):19–28. 360. Sumiyoshi T, Kurachi M, Kurokawa K, et al. (2000) Plasma homovanillic acid in the prodromal phase of schizophrenia. Biol Psychiatry. 47(5):428–33. 361. Hsiao JK, Colison J, Bartko JJ, et al. (1993) Monoamine neurotransmitter interactions in drug-free and neuroleptic-treated schizophrenics. Arch Gen Psychiatry. 50(8):606–14. 362. Faustman WO, Moses JA, Pfefferbaum A, et al. (1997) Neuropsychological performance is independent of CSF HVA in unmediated schizophrenic patients. Biol Psychiatry. 42(1 Suppl 1):172S. 363. Scheepers FE, Gispen-de Wied CC, Westenberg HG, et al. (2001) The effect of olanzapine treatment on monoamine metabolite concentrations in the cerebrospinal fluid of schizophrenic patients. Neuropsychopharmacology. 25(4):468–75. 364. Sharma RP, Javaid JI, Janicak PG, et al. (1993) Homovanillic acid in the cerebrospinal fluid: patterns of response after four weeks of neuroleptic treatment. Biol Psychiatry. 34(3):128–34. 365. Beuger M, van Kammen DP, Kelley ME, et al. (1996) Dopamine turnover in schizophrenia before and after haloperidol withdrawal: CSF, plasma, and urine studies. Neuropsychopharmacology. 15(1):75–86. 366. Kelley ME, van Kammen DP, Yao JK. (1997) Repeated measures in schizophrenia research: modelling drug withdrawal effects on plasma HVA and MHPG. Schizophr Res. 24(1–2):64–5. 367. Wyatt RJ, Karoum F, Casanova MF. (1995) Decreased DOPAC in the anterior cingulate cortex of individuals with schizophrenia. Biol Psychiatry. 38(1):4–12. 368. Fink-Jensen A. (2000) Novel pharmacological approaches to the treatment of schizophrenia. Dan Med Bull. 47(3):151–67. 369. Miller C, Kirchmair R, Troger J, et al. (1996) CSF of neuroleptic-naive first-episode schizophrenic patients: levels of biogenic amines, substance P, and peptides derived from chromogranin A (GE-25) and secretogranin II (secretoneurin). Biol Psychiatry. 39(11):911–8. 370. Walker EF, Bonsall R, Walder DJ. (2002) Plasma hormones and catecholamine metabolites in monozygotic twins discordant for psychosis. Neuropsychiatry Neuropsychol Behav Neurol. 15(1):10–7. 371. Breier A, Buchanan RW, Waltrip RW, II, et al. (1994) The effect of clozapine on plasma norepinephrine: relationship to clinical efficacy. Neuropsychopharmacology. 10(1):1–7.

c04

BLBK249-Cutler

160

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

372. Davila R, Zumarraga M, Gonzalez TMA, et al. (1994) Changes in plasma dihydroxyphenylacetic acid during haloperidol treatment of patients with schizophirenia. Eur J Psychiatry. 8(3):187–90. 373. Suhara T, Okubo Y, Yasuno F, et al. (2002) Decreased dopamine D2 receptor binding in the anterior cingulate cortex in schizophrenia. Arch Gen Psychiatry. 59(1):25–30. 374. Seeman P, Guan HC, van Tol HH. (1993) Dopamine D4 receptors elevated in schizophrenia. Nature. 365(6445):441–5. 375. Hirvonen J, van Erp TG, Huttunen J, et al. (2005) Increased caudate dopamine D2 receptor availability as a genetic marker for schizophrenia. Arch Gen Psychiatry. 62(4):371–8. 376. Sedvall G, Farde L. (1995) Chemical brain anatomy in schizophrenia. Lancet. 346(8977): 743–9. 377. Andreasen NC. (1984) The Broken Brain: The Biological Revolution in Psychiatry. Harper & Row, New York. 378. Tune LE, Wong DF, Pearlson G, et al. (1993) Dopamine D2 receptor density estimates in schizophrenia: a positron emission tomography study with 11C-N-methylspiperone. Psychiatry Res. 49(3):219–37. 379. Wong DF, Wagner HN, Jr, Tune LE, et al. (1986) Positron emission tomography reveals elevated D2 dopamine receptors in drug-naive schizophrenics. Science. 234(4783): 1558–63. 380. Zakzanis KK, Hansen KT. (1998) Dopamine D2 densities and the schizophrenic brain. Schizophr Res. 32(3):201–6. 381. Laruelle M. (1998) Imaging dopamine transmission in schizophrenia: a review and meta-analysis. Q J Nucl Med. 42(3):211–21. 382. Erritzoe D, Talbot P, Frankle WG, et al. (2003) Positron emission tomography and single photon emission CT molecular imaging in schizophrenia. Neuroimaging Clin N Am. 13(4):817–32. 383. Hietala J, Syv¨alahti E, Vuorio K, et al. (1994) Striatal D2 dopamine receptor characteristics in neuroleptic-naive schizophrenic patients studied with positron emission tomography. Arch Gen Psychiatry. 51(2):116–23. 384. Kestler LP, Walker E, Vega EM. (2001) Dopamine receptors in the brains of schizophrenia patients: a meta-analysis of the findings. Behav Pharmacol. 12(5):355–71. ˜ F, Catafau AM, Parellada E, et al. (2004) Striatal dopamine D2 receptor den385. Lomena sity in neuroleptic-naive and in neuroleptic-free schizophrenic patients: an 123I-IBZMSPECT study. Psychopharmacology (Berl). 172(2):165–9. 386. Dean B, Hussain T, Scarr E, et al. (2001) Extended treatment with typical and atypical antipsychotic drugs differential effects on the densities of dopamine D2-like and GABAA receptors in rat striatum. Life Sci. 69(11):1257–68. 387. Burnet PW, Eastwood SL, Harrison PJ. (1996) 5-HT1A and 5-HT2A receptor mRNAs and binding site densities are differentially altered in schizophrenia. Neuropsychopharmacology. 15(5):442–55. 388. Gurevich EV, Joyce JN. (1997) Alterations in the cortical serotonergic system in schizophrenia: a postmortem study. Biol Psychiatry. 42(7):529–45. 389. Kasper S, Tauscher J, Willeit M, et al. (2002) Receptor and transporter imaging studies in schizophrenia, depression, bulimia and Tourette’s disorder–implications for psychopharmacology. World J Biol Psychiatry. 3(3):133–46. 390. Gray L, Scarr E, Dean B. (2006) Serotonin 1a receptor and associated G-protein activation in schizophrenia and bipolar disorder. Psychiatry Res. 143(2–3):111–20. 391. Dean B, Tomaskovic-Crook E, Opeskin K, et al. (1999) No change in the density of the serotonin1A receptor, the serotonin4 receptor or the serotonin transporter in the

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

392.

393. 394.

395. 396.

397.

398.

399. 400.

401. 402.

403.

404.

405.

406. 407. 408. 409.

161

dorsolateral prefrontal cortex from subjects with schizophrenia. Neurochem Int. 34(2): 109–15. Cruz DA, Eggan SM, Azmitia EC, et al. (2004) Serotonin1A receptors at the axon initial segment of prefrontal pyramidal neurons in schizophrenia. Am J Psychiatry. 161(4):739–42. Tauscher J, Meyer JH, Smith GS, et al. (2002) Imaging of the serotonin system in depression and schizophrenia. Biol Psychiatry. (51):146S. Hashimoto T, Kitamura N, Kajimoto Y, et al. (1993) Differential changes in serotonin 5HT1A and 5-HT2 receptor binding in patients with chronic schizophrenia. Psychopharmacology (Berl). 112(1 Suppl):S35–9. Yasuno F, Suhara T, Ichimiya T, et al. (2004) Decreased 5-HT1A receptor binding in amygdala of schizophrenia. Biol Psychiatry. 55(5):439–44. Frankle WG, Lombardo I, Kegeles LS, et al. (2006) Serotonin 1A receptor availability in patients with schizophrenia and schizo-affective disorder: a positron emission tomography imaging study with 11C WAY 100635. Psychopharmacology (Berl). 189(2):155–64. Hurlemann R, Boy C, Meyer PT, et al. (2005) Decreased prefrontal 5-HT2A receptor binding in subjects at enhanced risk for schizophrenia. Anat Embryol (Berl). 210(5–6): 519–23. Erritzoe D, Rasmussen H, Kristiansen KT, et al. (2008) Cortical and subcortical 5-HT2A receptor binding in neuroleptic-naive first-episode schizophrenic patients. Neuropsychopharmacology. 33(10):2435–41. Hurlemann R, Matusch A, Kuhn KU, et al. (2008) 5-HT2A receptor density is decreased in the at-risk mental state. Psychopharmacology (Berl). 195(4):579–90. Ngan ET, Yatham LN, Ruth TJ, et al. (2000) Decreased serotonin 2A receptor densities in neuroleptic-naive patients with schizophrenia: a PET study using (18)F setoperone. Am J Psychiatry. 157(6):1016–8. Matsumoto I, Inoue Y, Iwazaki T, et al. (2005) 5-HT2A and muscarinic receptors in schizophrenia: a postmortem study. Neurosci Lett. 379(3):164–8. Trichard C, Paill`ere-Martinot ML, Attar-Levy D, et al. (1998) No serotonin 5-HT2A receptor density abnormality in the cortex of schizophrenic patients studied with PET. Schizophr Res. 31(1):13–7. Gomes UC, Shanley BC, Potgieter L, et al. (1980) Noradrenergic overactivity in chronic schizophrenia: evidence based on cerebrospinal fluid noradrenaline and cyclic nucleotide concentrations. Br J Psychiatry. 137:346–51. Ashcroft GW, Crawford TB, Eccleston D, et al. (1966) 5-hydroxyindole compounds in the cerebrospinal fluid of patients with psychiatric or neurological diseases. Lancet. 2(7472):1049–52. ¨ LH. (1998) CSF levels of HVA and 5-HIAA in drug-free Wieselgren IM, Lindstrom schizophrenic patients and healthy controls: a prospective study focused on their predictive value for outcome in schizophrenia. Psychiatry Res. 81(2):101–10. Tuckwell HC, Koziol JA. (1996) On the concentration of 5-hydroxyindoleacetic acid in schizophrenia: a meta-analysis. Psychiatry Res. 59(3):239–44. Crow TJ, Baker HF, Cross AJ, et al. (1979) Monoamine mechanisms in chronic schizophrenia: post-mortem neurochemical findings. Br J Psychiatry. 134:249–56. Winblad B, Bucht G, Gottfries CG, et al. (1979) Monoamines and monoamine metabolites in brains from demented schizophrenics. Acta Psychiatr Scand. 60(1):17–28. Joyce JN, Shane A, Lexow N, et al. (1993) Serotonin uptake sites and serotonin receptors are altered in the limbic system of schizophrenics. Neuropsychopharmacology. 8(4): 315–36.

c04

BLBK249-Cutler

162

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

410. Laruelle M, Abi-Dargham A, Casanova MF, et al. (1993) Selective abnormalities of prefrontal serotonergic receptors in schizophrenia: a postmortem study. Arch Gen Psychiatry. 50(10):810–8. 411. Meltzer HY, Sumiyoshi T. (2008) Does stimulation of 5-HT(1A) receptors improve cognition in schizophrenia? Behav Brain Res. 195(1):98–102. 412. Lieberman JA, Mailman RB, Duncan G, et al. (1998) Serotonergic basis of antipsychotic drug effects in schizophrenia. Biol Psychiatry. 44(11):1099–117. 413. Jones CA, McCreary AC. (2008) Serotonergic approaches in the development of novel antipsychotics. Neuropharmacology. 55(6):1056–65. 414. Paz RD, Tardito S, Atzori M, et al. (2008) Glutamatergic dysfunction in schizophrenia: from basic neuroscience to clinical psychopharmacology. Eur Neuropsychopharmacol. 18(11):773–86. 415. Bartha R, Williamson PC, Drost DJ, et al. (1997) Measurement of glutamate and glutamine in the medial prefrontal cortex of never-treated schizophrenic patients and healthy controls by proton magnetic resonance spectroscopy. Arch Gen Psychiatry. 54(10):959–65. 416. Th´eberge J, Al-Semaan Y, Williamson PC, et al. (2003) Glutamate and glutamine in the anterior cingulate and thalamus of medicated patients with chronic schizophrenia and healthy comparison subjects measured with 4-T proton MRS. Am J Psychiatry. 160(12):2231–3. 417. Burbaeva G, Boksha IS, Starodubtseva LI, et al. (2007) Glutamate dysmetabolism in patients with schizophrenia. Vestn Ross Akad Med Nauk. (3):19–24. 418. Gluck MR, Thomas RG, Davis KL, et al. (2002) Implications for altered glutamate and GABA metabolism in the dorsolateral prefrontal cortex of aged schizophrenic patients. Am J Psychiatry. 159(7):1165–73. 419. Burbaeva G, Boksha IS, Tereshkina EB, et al. (2007) Systemic neurochemical alterations in schizophrenic brain: glutamate metabolism in focus. Neurochem Res. 32(9):1434–44. 420. Burbaeva G, Boksha IS, Sudakov SA, et al. (2008) The complex neurochemical assessment of brain proteins in mentally healthy subjects and schizophrenic patients. Zh Nevrol Psikhiatr Im S S Korsakova. 108(2):44–50. 421. Hashimoto K, Engberg G, Shimizu E, et al. (2005) Elevated glutamine/glutamate ratio in cerebrospinal fluid of first episode and drug naive schizophrenic patients. BMC Psychiatry. 5:6. 422. Farber NB. (2003) The NMDA receptor hypofunction model of psychosis. Ann NY Acad Sci. 1003:119–30. 423. Olney JW, Newcomer JW, Farber NB. (1999) NMDA receptor hypofunction model of schizophrenia. J Psychiatr Res. 33(6):523–33. 424. Olney JW, Farber NB. (1995) Glutamate receptor dysfunction and schizophrenia. Arch Gen Psychiatry. 52(12):998–1007. 425. Lahti AC, Koffel B, LaPorte D, et al. (1995) Subanesthetic doses of ketamine stimulate psychosis in schizophrenia. Neuropsychopharmacology. 13(1):9–19. 426. Krystal JH, Karper LP, Seibyl JP, et al. (1994) Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans; psychotomimetic, perceptual, cognitive, and neuroendocrine responses. Arch Gen Psychiatry. 51(3):199–214. 427. Javitt DC, Zukin SR. (1991) Recent advances in the phencyclidine model of schizophrenia. Am J Psychiatry. 148(10):1301–8. 428. Lipina T, Labrie V, Weiner I, et al. (2005) Modulators of the glycine site on NMDA receptors, D-serine and ALX 5407, display similar beneficial effects to clozapine in mouse models of schizophrenia. Psychopharmacology (Berl). 179(1):54–67.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

163

429. Holmes E, Tsang TM, Huang JT, et al. (2006) Metabolic profiling of CSF: evidence that early intervention may impact on disease progression and outcome in schizophrenia. PLoS Med. 3(8):e327. 430. Ryan MC, Collins P, Thakore JH. (2003) Impaired fasting glucose tolerance in firstepisode, drug-naive patients with schizophrenia. Am J Psychiatry. 160(2):284–9. 431. Sengupta S, Parrilla-Escobar MA, Klink R, et al. (2008) Are metabolic indices different between drug-na¨ıve first-episode psychosis patients and healthy controls? Schizophr Res. 102(1–3):329–36. 432. Newcomer JW, Haupt DW, Fucetola R, et al. (2002) Abnormalities in glucose regulation during antipsychotic treatment of schizophrenia. Arch Gen Psychiatry. 59(4):337– 45. 433. Regenold WT, Phatak P, Kling MA, et al. (2004) Post-mortem evidence from human brain tissue of disturbed glucose metabolism in mood and psychotic disorders. Mol Psychiatry. 9(8):731–3. 434. Shelley AM, Ward PB, Catts SV, et al. (1991) Mismatch negativity: an index of a preattentive processing deficit in schizophrenia. Biol Psychiatry. 30(10):1059–62. 435. Javitt DC, Shelley A, Ritter W. (2000) Associated deficits in mismatch negativity generation and tone matching in schizophrenia. Clin Neurophysiol. 111(10):1733–7. 436. Salisbury DF, Shenton ME, Griggs CB, et al. (2002) Mismatch negativity in chronic schizophrenia and first-episode schizophrenia. Arch Gen Psychiatry. 59(8):686–94. 437. Youn T, Park HJ, Kim JJ, et al. (2003) Altered hemispheric asymmetry and positive symptoms in schizophrenia: equivalent current dipole of auditory mismatch negativity. Schizophr Res. 59(2–3):253–60. 438. Catts SV, Shelley AM, Ward PB, et al. (1995) Brain potential evidence for an auditory sensory memory deficit in schizophrenia. Am J Psychiatry. 152(2):213–9. 439. Shutara Y, Koga Y, Fujita K, et al. (1996) An event-related potential study on the impairment of automatic processing of auditory input in schizophrenia. Brain Topogr. 8(3):285–9. 440. Korostenskaja M, Dapsys K, Siurkute A, et al. (2005) Effects of olanzapine on auditory P300 and mismatch negativity (MMN) in schizophrenia spectrum disorders. Prog Neuropsychopharmacol Biol Psychiatry. 29(4):543–8. 441. Devrim-Uc¸ok M, Keskin-Ergen HY, Uc¸ok A. (2008) Mismatch negativity at acute and post-acute phases of first-episode schizophrenia. Eur Arch Psychiatry Clin Neurosci. 258(3):179–85. 442. N¨aa¨ t¨anen R, Picton T. (1987) The N1 wave of the human electric and magnetic response to sound: a review and an analysis of the component structure. Psychophysiology. 24(4):375–425. 443. Butler RA. (1968) Effect of changes in stimulus frequency and intensity on habituation of the human vertex potential. J Acoust Soc Am. 44(4):945–50. 444. Rosburg T, Boutros NN, Ford JM. (2008) Reduced auditory evoked potential component N100 in schizophrenia—a critical review. Psychiatry Res. 161(3):259–74. 445. Baribeau-Braun J, Picton TW, Gosselin JY. (1983) Schizophrenia: a neurophysiological evaluation of abnormal information processing. Science. 219(4586):874–6. 446. Pfefferbaum A, Ford JM, White PM, et al. (1989) P3 in schizophrenia is affected by stimulus modality, response requirements, medication status, and negative symptoms. Arch Gen Psychiatry. 46(11):1035–44. 447. Gallinat J, Winterer G, Herrmann CS, et al. (2004) Reduced oscillatory gamma-band responses in unmedicated schizophrenic patients indicate impaired frontal network processing. Clin Neurophysiol. 115(8):1863–74.

c04

BLBK249-Cutler

164

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

448. Roth WT, Goodale J, Pfefferbaum A. (1991) Auditory event-related potentials and electrodermal activity in medicated and unmedicated schizophrenics. Biol Psychiatry. 29(6):585–99. 449. Laurent A, Garcia-Larr´ea L, d’Amato T, et al. (1999) Auditory event-related potentials and clinical scores in unmedicated schizophrenic patients. Psychiatry Res. 86(3):229–38. 450. Mulert C, Gallinat J, Pascual-Marqui R, et al. (2001) Reduced event-related current density in the anterior cingulate cortex in schizophrenia. Neuroimage. 13(4):589–600. 451. Gallinat J, Mulert C, Bajbouj M, et al. (2002) Frontal and temporal dysfunction of auditory stimulus processing in schizophrenia. Neuroimage. 17(1):110–27. 452. Umbricht D, Javitt D, Novak G, et al. (1999) Effects of risperidone on auditory eventrelated potentials in schizophrenia. Int J Neuropsychopharmacol. 2(4):299–304. 453. Iwanami A, Okajima Y, Isono H, et al. (2001) Effects of risperidone on event-related potentials in schizophrenic patients. Pharmacopsychiatry. 34(2):73–9. 454. Pekkone E, Hirvonen J, Ahveninen J, et al. (2002) Memory-based comparison process not attenuated by haloperidol: a combined MEG and EEG study. NeuroReport. 13(1):177–81. 455. Devrim-Uc¸ok M, Keskin-Ergen HY, Uc¸ok A. (2008) P50 gating at acute and post-acute phases of first-episode schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry. 32(8):1952–6. 456. Light GA, Geyer MA, Clementz BA, et al. (2000) Normal P50 suppression in schizophrenia patients treated with atypical antipsychotic medications. Am J Psychiatry. 157(5):767–71. 457. Adler LE, Olincy A, Cawthra EM, et al. (2004) Varied effects of atypical neuroleptics on P50 auditory gating in schizophrenia patients. Am J Psychiatry. 161(10):1822–8. 458. Jeon YW, Polich J. (2003) Meta-analysis of P300 and schizophrenia: patients, paradigms, and practical implications. Psychophysiology. 40(5):684–701. 459. Bramon E, Rabe-Hesketh S, Sham P, et al. (2004) Meta-analysis of the P300 and P50 waveforms in schizophrenia. Schizophr Res. 70(2–3):315–29. 460. Shin KS, Kang DH, Choi JS, et al. (2008) Neuropsychological correlates of N400 anomalies in patients with schizophrenia: a preliminary report. Neurosci Lett. 448(2):226–30. 461. Ford JM. (1999) Schizophrenia: the broken P300 and beyond. Psychophysiology. 36(6): 667–82. 462. Braff DL, Light GA. (2005) The use of neurophysiological endophenotypes to understand the genetic basis of schizophrenia. Dialogues Clin Neurosci. 7(2):125–35. 463. Bender S, Schall U, Wolstein J, et al. (1999) A topographic event-related potential followup study on “prepulse inhibition” in first and second episode patients with schizophrenia. Psychiatry Res. 90(1):41–53. 464. Geyer MA. (2006) The family of sensorimotor gating disorders: comorbidities or diagnostic overlaps? Neurotox Res. 10(3–4):211–20. 465. Kumari V, Sharma T. (2002) Effects of typical and atypical antipsychotics on prepulse inhibition in schizophrenia: a critical evaluation of current evidence and directions for future research. Psychopharmacology (Berl). 162(2):97–101. 466. Turetsky BI, Calkins ME, Light GA, et al. (2007) Neurophysiological endophenotypes of schizophrenia: the viability of selected candidate measures. Schizophr Bull. 33(1):69–94. 467. Holzman PS, Kringlen E, Matthysse S, et al. (1988) A single dominant gene can account for eye tracking dysfunctions and schizophrenia in offspring of discordant twins. Arch Gen Psychiatry. 45(7):641–7. 468. Levy DL, O’Driscoll G, Matthysse S, et al. (2004) Antisaccade performance in biological relatives of schizophrenia patients: a meta-analysis. Schizophr Res. 71(1):113–25.

c04

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Biomarkers and surrogate markers in drug development

165

469. Raemaekers M, Jansma JM, Cahn W, et al. (2002) Neuronal substrate of the saccadic inhibition deficit in schizophrenia investigated with 3-dimensional event-related functional magnetic resonance imaging. Arch Gen Psychiatry. 59(4):313–20. 470. Reilly JL, Harris MS, Keshavan MS, et al. (2006) Adverse effects of risperidone on spatial working memory in first-episode schizophrenia. Arch Gen Psychiatry. 63(11):1189–97. 471. Hutton SB, Crawford TJ, Gibbins H, et al. (2001) Short and long term effects of antipsychotic medication on smooth pursuit eye tracking in schizophrenia. Psychopharmacology (Berl). 157(3):284–91. 472. Mahadik SP, Mukherjee S. (1996) Free radical pathology and antioxidant defense in schizophrenia: a review. Schizophr Res. 19(1):1–17. 473. Mukerjee S, Mahadik SP, Scheffer R, et al. (1996) Impaired antioxidant defense at the onset of psychosis. Schizophr Res. 19(1):19–26. 474. Pall HS, Williams AC, Blake DR, et al. (1987) Evidence of enhanced lipid peroxidation in the cerebrospinal fluid of patients taking phenothiazines. Lancet. 2(8559):596–9. 475. Skinner AO, Mahadik SP, Garver DL. (2005) Thiobarbituric acid reactive substances in the cerebrospinal fluid in schizophrenia. Schizophr Res. 76(1):83–7. 476. Puri BK, Ross BM, Treasaden IH. (2008) Increased levels of ethane, a non-invasive, quantitative, direct marker of n-3 lipid peroxidation, in the breath of patients with schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry. 32(3):858–62. 477. Katz R. (2004) Biomarkers and surrogate markers: an FDA perspective. NeuroRx. 1(2):189–95. 478. Simon R. (2007) Use of predictive biomarker classifiers in the design of pivotal clinical trials [document on the Internet]. Available from: http://linus.nci.nih.gov/techreport/ PredictiveBiomarkerPivotalClinicalTrials.pdf. Cited February 25, 2009. 479. Williams SA, Slavin DE, Wagner JA, et al. (2006) A cost-effectiveness approach to the qualification and acceptance of biomarkers. Nat Rev Drug Discov. 5(11):897–902. 480. Prentice RL. (1989) Surrogate endpoints in clinical trials: definition and operational criteria. Stat Med. 8(4):431–40. 481. Lambert JC, Amouyel P. (2007) Genetic heterogeneity of Alzheimer’s disease: complexity and advances. Psychoneuroendocrinology. 32(Suppl 1):S62–70. 482. US Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. (2001) Guidance for industry: bioanalytical method validation [document on the Internet]. Available from: http://www.fda.gov/ downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ UCM070107.pdf. Cited July 27, 2009. 483. Food and Drug Administration. (2008) Good laboratory practices for nonclinical laboratory studies. FDA Government Document, Code of Federal Regulations Title 21 Vol 1 [document on the Internet]. Available from: http://www.accessdata.fda.gov/ scripts/cdrh/cfdocs/cfcfr/CFRSearch.cfm?CFRPart=58. Cited July 27, 2009. 484. Food and Drug Administration. (2001) Clinical laboratory improvement amendment. FDA Government Document, Code of Federal Regulations Title 42 Vol 3 [document on the Internet]. Available from: http://www.cms.hhs.gov/CLIA/. Cited July 27, 2009. 485. National Committee for Clinical Laboratory Standards (NCCLS), Document EP5-A: Evaluation of Precision Performance for Clinical Chemistry Devices: Approved Guideline (1999); Document EP6-P: Evaluation of the Linearity of Quantitative Analytical Method: Proposed Guideline (1986); Document EP7-P: Interference Testing in Clinical Chemistry: Proposed Guideline (1986); Document EP9-A: Method Comparison and Bias Estimation Using Patient Samples: Approved Guideline (1995) [document on the Internet]. Available from: http://www.clsi.org/. Cited July 27, 2009.

c04

BLBK249-Cutler

166

Trim: 6in X 9in

December 29, 2009

21:32

Char Count=

Chapter 4

486. Findlay JW, Smith WC, Lee JW, et al. (2000) Validation of immunoassays for bioanalysis: a pharmaceutical industry perspective. J Pharm Biomed Anal. 21(6):1249–73. 487. Miller KJ, Bowsher RR, Celniker A, et al. (2001) Workshop on bioanalytical methods validation for macromolecules: summary report. Pharm Res. 18(9):1373–83. 488. Lee JW, Weiner RS, Sailstad JM, et al. (2005) Method validation and measurement of biomarkers in nonclinical and clinical samples in drug development: a conference report. Pharm Res. 22(4):499–511. 489. Lee JW, Devanarayan V, Barrett YC, et al. (2006) Fit-for-purpose method development and validation for successful biomarker measurement. Pharm Res. 23(2):312–28. 490. Wagner JA. (2008) Strategic approach to fit-for-purpose biomarkers in drug development. Ann Rev Pharmacol Toxicol. 48:631–51. 491. Sittampalam GS, Smith WC, Miyakawa TW, et al. (1996) Application of experimental design techniques to optimize a competitive ELISA. J Immunol Methods. 190(2):151–61. 492. Joelsson D, Moravec P, Troutman M, et al. (2008) Optimizing ELISAs for precision and robustness using laboratory automation and statistical design of experiments. J Immunol Methods. 337(1):35–41. 493. Kingsmore SF. (2006) Multiplexed protein measurement: technologies and applications of protein and antibody arrays. Nat Rev Drug Discov. 5(4):310–20. 494. Ray CA, Bowsher RR, Smith WC, et al. (2005) Development, validation, and implementation of a multiplex immunoassay for the simultaneous determination of five cytokines in human serum. J Pharm Biomed Anal. 36(5):1037–44.

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

CHAPTER 5

Neuroimaging and cognitive assessments in early drug development

Introduction While traditional biomarkers are extremely useful for gathering early clinical information on the efficacy, dose range, side effects, and toxicity of a novel compound in humans, they are by no means the only diagnostic tools available for this purpose. Both neuroimaging and cognitive assessments play a valuable role in early clinical trials and are important components of the modern central nervous system (CNS) drug developer’s toolkit. The use of neuroimaging technologies over the past two decades has been considered to be vital in the diagnosis of neurocognitive disorders such as Alzheimer’s disease (AD) and multiple sclerosis (MS), and these technologies are currently used with a multitude of radiolabeled biomarkers to evaluate a compound’s structural targets and distribution in the brain. In addition, the value of cognitive testing in early-phase work has been firmly established, not just for diagnostic purposes, but also for assessing the side effects, potencies, and specific acute actions of drugs in dose-finding studies. Neuroimaging modalities have only recently been included in clinical trials that assess the efficacy of drugs on neurocognitive disorders. Both structural and functioning imaging techniques such as magnetic resonance imaging (MRI), functional MRI (fMRI), and positron emission tomography (PET) may aid CNS product development both in very early stages (mainly through functional measures) and in later registration trials (mainly through morphometric quantification). PET has shown significant utility in helping to define biochemistry and to determine dose and drug selection; for example, assessing changes in amyloid burden and functional changes in the brain associated with AD treatment. MRI has mainly been utilized as a method to track lesion load and tissue salvage in neurocognitive disorders, but it has also been used as fMRI to measure changes in brain functions associated with treatment. Pathophysiologic and behavioral assessments of cognitive function have been widely used to diagnose and track the progression of many CNS disorders, since concrete biological markers for these disorders are usually insufficient or nonexistent. Cognitive assessments are regularly used to diagnose AD and schizophrenia, among many other CNS disorders. In addition, cognitive and behavioral assessments Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

167

c05

BLBK249-Cutler

168

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

are regularly used to assess the efficacy of novel CNS therapeutic compounds during clinical trials. This chapter will provide details on the latest methods and applications of neuroimaging technologies and cognitive assessments for evaluating the efficacy and safety of CNS therapeutic compounds.

Functional MRI fMRI measures the hemodynamic response correlated to neural activity in the brain or spinal cord of humans and animals; specifically, the MRI contrast of blood deoxyhemoglobin, known as blood-oxygen-level dependent (BOLD) fMRI. It has become one of the more popular neuroimaging techniques for monitoring CNS disease progression as well as detecting signals related to the efficacy of CNS therapeutics, due to its low invasiveness, short scan time, lack of radiation exposure, relatively wide availability, and high resolution—down to less than 1 mm. fMRI has been used to observe changes in the motor cortex of patients recovering from stroke after treatment with fluoxetine [1]. In MS and stroke, fMRI has been employed to identify altered networks involved in motor control and to identify potentially adaptive reorganization and plasticity [2, 3]. fMRI enables the identification of critical brain regions to aid surgical planning in epileptics [4] and may also be predictive of AD progression in atrisk populations [5]. Expanding clinical applications of fMRI also suggest its use as a measure of efficacy in multicenter studies of new therapies, as demonstrated in studies involving novel AD therapeutics. Functional neuroimaging studies have shown that AD patients differ on measures of regional brain activation from controls during the performance of a variety of tasks, with decreases in hippocampal and parahippocampal activation in comparison with older healthy subjects [6–8]. fMRI has been increasingly valuable in evaluating acute and subacute effects of medications on neural activity that may have both symptomatic and disease-modifying properties. Recent animal studies have suggested that strategies to remove amyloid may result in acute changes in synaptic function and behavior detected by fMRI [9, 10]. In addition, fMRI studies of the effects of the AD therapeutic cholinesterase inhibitors on brain activation during cognitive task performance have shown changes after administration in patients with AD [11] and mild cognitive impairment (MCI) [12]. A confounding factor with correlating fMRI-observed cognitive changes with disease progression is the fact that healthy individuals of any age demonstrate differences in brain activation depending on how well they are able to perform the particular tasks [13]. MCI and AD patients can show activation in additional brain regions during task performance [14, 15], which may indicate attempts to compensate for damaged networks [16]. This sensitivity to changes in cognitive function supports the use of fMRI in short-term, early proof-of-concept drug trials. For example, a double-blind crossover study of 19 healthy young subjects was conducted using fMRI to investigate short-term effects of two doses of

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

169

their N-methyl-d-aspartic acid (NMDA) receptor antagonist EVT 101 and placebo. The effects of the compound on neuropsychological test performance, local cerebral blood flow (reflecting neuronal activation or deactivation) during a cognitive task, as well as baseline regional cerebral blood flow, were examined. Despite the lack of evidence suggesting changes in cognitive function, they moved this product along by suggesting that the activity of a number of brain regions, known for their role as a memory retrieval network, was selectively modified during the performance of certain tasks. Specifically, an “analysis of the changes in basal regional cerebral blood flow indicated a selective increase in blood flow in one specific region of the cortex, an area rich in NMDA receptors which is considered to be significant for indications such as pain and depression.” However, basal blood flow in other brain areas was unaffected [17]. There are many opportunities for the use of fMRI in efficacy studies of novel drugs for CNS disorders, including as an indicator of cognitive changes induced by therapeutics in early clinical trials, although numerous sources of variability must be carefully accounted for in order to produce reliable results. These sources include caffeine ingestion [18], sleep deprivation [19], different levels of attention to sensory stimuli [20], scanner differences [21], noise in fMRI data [22], and anatomical variations between subjects [23].

Structural MRI The value of MRI in the diagnosis of various neurocognitive disorders is widely acknowledged and, more recently, the utility of structural MRI in tracking disease state and treatment response has been proposed. One of the earliest indications to fully adopt MRI technologies is MS and, in some ways, this indication continues to lead the way in the use of novel imaging modalities such as fMRI and diffusion tensor imaging (DTI). Clinical trials in MS are also well ahead of other CNS indications such as AD in both multicenter standardization and regulatory precedence; specifically, in helping to establish MRI as a trial-level surrogate in registration studies. Although it is generally agreed that MRI provides a reflection of the underlying pathology in MS (in fact, the number of new gadolinium enhancing lesions on monthly MRI of the brain is one of the most often measured outcomes of MS activity), it has been determined that no single MRI measurement in isolation can sufficiently monitor the disease process, and the use of multiple imaging techniques, especially new, emerging techniques that may better reflect the underlying pathology, is encouraged [24]. Because of the enormously variable course of MS and the generally inadequate nature of the commonly used clinical measures of outcome, there are inherent problems in designing treatment trials with clinical endpoints such as relapse rate or change in disability. These endpoints are also often difficult to obtain in a trial setting. Therefore, markers such as MRI may provide answers earlier with fewer patients. However, because MRI does not correlate

c05

BLBK249-Cutler

170

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

strongly with outcomes of clinical interest, it has not been viewed as a reliable surrogate. This dissociation may be beneficial as it has been observed that MRI markers may be more sensitive to disease state than clinical markers, may be more responsive to early treatment, and may be able to predict both relapse and clinically definite MS (by white matter lesions) as well as disability (by atrophy measures) in relapse-remitting MS (RRMS) [25]. MRI measures have a general tendency to result in fewer subjects per treatment group to obtain the same statistical power as clinical measures such as the Expanded Disability Status Scale [26, 27] . Sample sizes based on structural MRI have also suggested a significant reduction in the number of patients when using changes in brain volumes over time. For example, it has been well established that medial temporal lobe structures are most vulnerable to the earliest pathology of AD with the hippocampus and entorhinal cortex being two of the best candidates to differentiate AD patients from controls and to track diseases’ state over time [28, 29]. In an effort to assess the feasibility of using MRI measurements as a potential surrogate endpoint for disease progression in a therapeutic trial for AD, Jack et al. [30] obtained MRI data on 362 patients with probable AD from 38 different centers over a 52-week period. Results suggested that the annualized percent changes in hippocampal volume (of approximately 4.9%/year) and temporal horn volume (approximately 16.1%/year) in the study patients were consistent with data from prior single-site studies. Correlations between the rate of MRI volumetric change and the change in behavioral/cognitive measures were greater for the temporal horn than those for the hippocampus. Importantly, decline over time was more consistently seen with imaging measures than behavioral/cognitive measures. Whole brain volume and ventricular size have also shown some utility in tracking disease [31]. In a sample of 39 elderly subjects (5 subjects were classified as probable AD, 2 as possible AD, and 32 as AD negative), Bradely et al. [32] reported that the rate of change of whole brain atrophy over time for AD patients was 2.14%/year, and for negative AD patients only 0.2% per year, which is comparable to the brain changes described by Fox et al. [33].They reported probable AD rates of 2.37%/year and for controls 0.41%/year. Bradley et al. also reported that the rate of change in ventricle-to-brain ratio was 15.6%/year for probable AD compared with 4.3%/year for negative AD. Notably, no significant difference between these groups on measures of mean ventricle-to-brain ratios measured at a single time point was evident but rates of change in brain or ventricular volume over time also differed between the two groups (p = 0.001). Power calculations reveal that to detect a 20% reduction in the excess rate of atrophy of a probable AD cohort in just 6 months, with 90% power, 135 subjects would be required in each treatment arm. For a 30% reduction in the excess rate of atrophy, 61 subjects would be required. There is also some evidence suggesting that it is advantageous to combine whole brain/ventricle measures with measures of the medial temporal lobe to track disease progression [31]. Combining a measure of hippocampal volume

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

171

at baseline with a measure of either whole brain or ventricle atrophy rates from serial MRI scans provides complimentary predictive information about the hazard of subsequent conversion from mild cognitive impairment to AD. Given this sensitivity, correlations with clinical measures, and reduced sample size, serial structural MRI has been adopted in several AD treatment trials [34]. The relationship between longitudinal MRI measures and changes in performance on cognitive measures including Mini-Mental State Examination (MMSE), ADAS-Cog, DAD, and CGIC was assessed by Ridha et al. (2008) [35], who examined 52 subjects from the placebo arm of a clinical trial with mildto-moderate AD. They reported that overall rates of brain atrophy and/or ventricular enlargement correlated with declining performance on cognitive scales with the strongest association between brain atrophy rate and MMSE decline (r = 0.59, p < 0.0001). The strong relationship most likely reflects the correspondence between measures of overall cerebral loss and global cognitive measures in the moderate stages of AD. Note that hippocampal atrophy rate was not significantly correlated with any of the cognitive scales, and this lack of correlation may be due to the combination of extensive functional damages to the hippocampus by the time AD is clinically established, the greater influence of ongoing cortical degeneration, and/or errors in image analysis. Despite the general utility of MRI measures in tracking cognition, disease progression and treatment, no single or combined measure of MRI morphometry is considered to be an adequate surrogate in AD clinical trials. In fact, a meeting of the Peripheral and Central Nervous Systems Drugs Advisory Committee in 2002 concluded that there were no validated MRI markers for assessing the drug effect in patients with AD and that no such marker should be used as a primary measure of drug effect [36]. Despite various attempts to validate surrogates in AD research, none have succeeded so far. One of the reasons for this may be the lack of standards in MRI protocols. As observed above, AD researchers have targeted a variety of brain structures and have used diverse hardware/software for image acquisition and analysis and different time periods to assess the change. In an effort to standardize MRI methods in the field of AD research and to support the use of MRI in trials, the National Institutes of Health (NIH) has launched a large effort known as the Alzheimer’s Disease Neuroimaging Initiative (ADNI). This initiative was designed to help researchers and clinicians develop new treatments and monitor their effectiveness (which should increase the safety and efficiency of drug development by decreasing the time and cost of clinical trials). This project, which began in 2004, is the most comprehensive effort to date to identify neuroimaging and other biomarkers of the cognitive changes associated with MCI and AD. Specifically, this initiative attempts to define the rate change in cognition, function, brain structure and function, and various biomarkers in 200 elderly controls, 400 subjects with mild cognitive impairment, and 200 with AD. This $60 million project ($40 million from the National Institute of Aging and

c05

BLBK249-Cutler

172

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

National Institute of Bioimaging and Bioengineering and $20 million from the pharmaceutical industry and several foundations) is being conducted at 58 sites in the United States and Canada (other ADNI initiatives around the world are also ongoing). The main goals of the ADNI are to: 1. develop improved methods that will lead to uniform standards for acquiring longitudinal, multisite MRI and PET data on patients with AD, MCI, and elderly controls; 2. acquire a generally accessible data repository that describes longitudinal changes in brain structure and metabolism. In parallel, acquire clinical cognitive and biomarker data for validation of imaging surrogates; 3. develop methods that will provide maximum power to determine treatment effects in trials involving these patients; and 4. test a series of hypotheses based on the clinical and biomarker data as outlined in the statistical analysis section. As part of the initiative, all subjects will have a clinical/cognitive assessments as well as a 1.5 T structural MRI scan at specified intervals for 2–3 years. Approximately 50% of subjects will also have fluorodeoxyglucose (FDG) PET scans at the same time intervals and 25% of subjects (who have not been scanned using PET) will have MRI at 3 Tesla. Outcome measures include the rate of conversion from MCI to AD; the rate of volume change of whole brain, hippocampus, and other structural MRI measures; the rate of change on each specified biomarker; the rate of change of glucose metabolism for specified regions of interest on PET scanning and group differences for each imaging and biomarker measurement.

PET imaging: FDG, amyloid binding, and microdosing PET provides a noninvasive way to image neurochemistry and visualize brain function. Its use to enhance the discovery of therapeutics has grown markedly over the past few years, mainly for three reasons. First, it is relatively easy to label compounds with radionucleotides without significantly altering their function and metabolism. As many drugs and endogenous compounds are made up of carbon, nitrogen, and oxygen, it is easier to label them with positron emitters that are autologous with the parent compound. Therefore, many of the radionucleotides commonly used in PET include isotopes of biologically ubiquitous elements. In addition, although fluorine is not a common constituent of organic molecules, a positron-emitting isotope of this element, fluorine-18 (18F), can be incorporated into a molecule of interest to produce an organofluorine analog of the parent molecule without appreciably affecting its pharmacological and physiochemical properties. Second, the radioactive half-lives of PET radionucleotides are fairly short (carbon-11, nitrogen-13, oxygen-15, and fluorine-18 are 20, 10, 2, and 110 min, respectively), permitting repeat studies in the same subject within the same day. Third, the relative amount of drug administered in a PET study is very low (typically less than 10 nmol) so that even the most potent or toxic compounds can be labeled and

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

173

administered to living subjects with little risk of pharmacological and/or toxicological effects [37]. This is also particularly helpful in early development when there may not be a large quantity of drug synthesized. Some of the most frequent uses of PET imaging involve examining how an unlabeled compound inhibits specific binding of a well-characterized, selective PET radioligand, or tracing the uptake, anatomical distribution, and binding of a labeled compound in brain. The effects of a novel drug on physiological/biochemical parameters, such as glucose metabolism or blood flow, can also be assessed. The demonstration of quantitative relationships between drug binding in vivo and drug effects in patients can be successfully used to validate targets for drug action, to correlate pharmacological and physiological effects, and to optimize clinical treatment [38]. Tracer doses of labeled compounds can facilitate the early clinical evaluation of novel drugs by assessing biodistribution, receptor/enzyme occupancy (to optimize drug-dosing regimens), and downstream responses of drug action. Quantitative PET studies can be conducted to assess drug binding to target proteins and to correlate receptor occupancy with pharmacodynamic responses. This enables the tracking of disease progression and monitoring the outcome of new treatments. PET can also be used to facilitate longitudinal studies of biomarkers of pathophysiology such as amyloid plaque load in AD [39]. Finally, functional imaging, such as using FDG PET with 18F, has also been shown to reliably diagnose AD, and may even be a suitable biomarker to assess changes in the progression of the disease.

Measuring amyloid load Imaging biomarkers have also shown some utility in assessing amyloid burden in patients with known or suspected AD and more recently in assessing the effectiveness of compounds that are designed to clear or prevent amyloid buildup. There are many amyloid probes in development for both humans and animals; however, the two most widely researched probes are F2-(1{6-[(2-18F-fluoroethyl)(methyl)amino]2-naphthyl}ethylidene)malononitrile (FDDNP) [40] and Pittsburgh compound-B (PiB) [41]. Both of these probes were originally developed by academic laboratories but have proven to be so popular and profitable that they are now owned by imaging companies such as Siemens and GE. The probes are similar in nature but FDDNP differs from PiB in two essential ways: the former targets both amyloid and tau and is radiolabeled with 18F, and the latter targets amyloid only and is radiolabeled with C-11 [42, 43]. It has been suggested that the relatively long half-life of 18 F (110 minutes) compared with that of C-11 (20 minutes) has offered some advantages to researchers who do not have access to a cyclotron, rendering the FDDNP probe more readily accessible. PiB binds with ␤-pleated sheet aggregates of A␤ in vivo and in vivo PiB retention levels correlated highly with regional post-mortem measures of PiB binding, insoluble A␤ levels, and plaque load (although not with neurofibrillary

c05

BLBK249-Cutler

174

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

tangles). This supports the validity of PiB PET imaging as a suitable method of in vivo assessment of plaque burden [44].

Microdosing PET PET has also made it possible to noninvasively determine drug distribution and concentration in vivo in humans by labeling a drug with a positronemitting radionuclide that does not change the biochemical properties of the drug. Rapid and enormous developments in labeling organic compounds with positron-emitting radionuclides have allowed a substantial number of new drug candidates to be labeled and used as probes in PET studies. PET microdosing typically involves administering a carbon-11- or fluorine-18-labeled therapeutic compound to subjects in order to describe the concentration-time profile of this compound in body tissues targeted for treatment. The major advantage to this technique is that PET microdosing involves the administration of very small (microgram) amounts of drug. The dose is so small that it is not intended to produce any pharmacologic effect when administered to humans—therefore is also unlikely to cause an adverse reaction. For practical purposes, this dose is defined as 1/100th of that anticipated to produce a pharmacological effect or 100 ␮g, whichever is the smaller, limiting any potential toxicological risks. Because of this limited risk regulatory authorities will often require less stringent requirements for preclinical safety testing as compared with conventional early studies that employ larger, more typical doses. PET microdosing is also capable of examining in vivo brain binding in the guinea pig and marmoset, which can often avoid the need for lengthy and expensive higher primate studies [45]. The relevance and reliability of animal testing for predicting safety and efficacy in humans are less than ideal, as revealed by the difficulties in extrapolating the results of preclinical animal data to humans. One commonly applied approach to predicting a human pharmacokinetic (PK) profile based on animal data is allometric scaling, which scales the animal data to humans, assuming that the only difference among animals and humans is body size. While body size is an important determinant of pharmacokinetics, it is certainly not the only feature that distinguishes humans from animals, nor is it very accurate—this simple approach has been estimated to have less than 60% predictive accuracy [46]. A review by the International Life Sciences Institute of data from 12 pharmaceutical companies, based on human toxicities identified during clinical trials, suggested a concordance rate of 71% for comparable target organs in rodent plus nonrodent studies. It also identified human toxicities for 150 compounds [47]. However, rodent data alone was predictive of only 43% of human adverse events, and nonrodent data alone was predictive of only 63% of human adverse events. It is very difficult to extrapolate animal ADME (absorption, distribution, metabolism, and excretion) studies to humans given multiple species differences in routes of metabolism, scaling processes involved in consideration of species weight, lifespan, metabolic

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

175

rate, rate of cell division, and body surface area [48]. Differing approaches can yield results varying by several orders of magnitude, with each step in the scaling process adding to the degree of uncertainty in the final extrapolation [49]. The idea is that if the microdosing data in humans indicates that the compound does not have a desirable pharmacokinetic profile, the development of that compound may then be terminated, thereby avoiding the need to undertake the normal extensive animal safety assessment on that compound. Therefore, microdosing could have a substantial impact on animal use. Second, microdosing in humans has the potential to replace animal use in determining suitable pharmacokinetic profiles of compounds. Third, if microdosing in humans turns out to be more predictive than the current animal methods, less drug candidates will need to go through extensive safety and toxicology testing involving animals, because more compounds will be terminated after early microdose studies. The noninvasiveness of PET, the use of small amounts of valuable drugs, the abbreviated toxicity package requirements, and possibility for a shortened development program have caused microdosing techniques to become more popular as one way to screen drug candidates for selection or rejection [50] and to shorten timelines and cut costs of drug development [51]. Despite the lack of evidence concerning the predictive accuracy of microdosing (which assumes that a reaction to a particular compound is the same at microdose levels as it is at pharmacological dose levels), there has been an increase in microdosing studies, especially in compound selection—which can be particularly valuable when multiple drug candidates are prepared as radiotracers and tested simultaneously [50]. In fact, PET microdosing has become so popular that various agencies and working groups have provided guidance on its use. This includes the European Medicines Agency (EMEA) [52] and the Food and Drug Administration (FDA) Center for Drug Evaluation and Research (2005) [53]. Both the FDA and EMEA define a microdose as less than 1/100th of the dose of a test substance (based on animal data or primary pharmacodynamic data obtained in vitro and in vivo) calculated to yield a pharmacologic effect of the test substance with a maximum dose of ≤100 ␮g (for imaging agents, the latter criterion applies). Because of differences in molecular weights compared with synthetic drugs, the maximum dose for protein products is ≤30 nmol. One European workshop on microdose drug studies [49] defines a microdose as the lowest dose in human subjects that yields useful data by whatever analytical method is used (e.g., depending on detection limits). This would not exceed 1% of the minimum pharmacologically active dose obtained from the most sensitive pharmacological model. In the absence of this data, the microdose would not exceed 1 ␮g/person. This workshop suggests that because early microdose studies would be conducted before most animal tests, the availability of human pharmacokinetic and metabolism data could also minimize the use of inappropriate species in later studies. In addition, the workshop recommends that for reasons of cost, time, patient safety, and

c05

BLBK249-Cutler

176

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

animal welfare, a drug candidate that progresses through phases I and IIa trials should have a high probability of success. Maximizing the ADME, safety and efficacy data acquired in phase I studies would enhance this probability and would limit long-term animal tests being started on candidates likely to fail in 6–12 months time. Likewise, it recommends acquiring as much information as possible from human studies, over and above the routine safety and tolerability approaches. The workshop also acknowledges that much more research is required into the validity of extrapolating ADME findings from human microdose studies to therapeutic doses. It is still reasonable to ensure that human studies are designed to maximize the amount of ADME, safety and efficacy data acquired, over and above routine safety and tolerability approaches, to enhance drug development and to limit long-term animal tests being started on candidates likely to fail later in the process [49].

EEG Electroencephalography (EEG) has been shown to be sensitive to the acute effects of many centrally acting medications, and EEG analysis has been used to study drug clinical pharmacology in early-phase trials. Since overt task performance reflects subject motivation and effort as well as ability, some believe that cognitive measures alone may either routinely overestimate or underestimate the impact of a pharmacological intervention on brain function. The use of combined neurophysiological and neurocognitive measures has been suggested in an attempt to detect the psychoactive effects of pharmacological treatments with greater sensitivity. For example, Gevins et al. [54] suggested the utility of incorporating EEG into statistical inferences regarding the acute impact of drugs on mental function. They demonstrated the viability of multivariate combinations of behavioral and neurophysiological measures to better characterize the pharmacodynamics of drug-induced changes in cognition. This was done in a double-blind, placebo-controlled, crossover study in which caffeine, diphenhydramine, and alcohol were used to modify the cerebral condition of 16 healthy subjects at rest and while they performed low load and high load versions of a working memory task.

Pharmaco-EEG Pharmaco-EEG uses technologies such as quantitative electroencephalography (QEEG) to measure precise electrophysiological activity in regions of the brain and relations between them [55]. QEEG uses a multielectrode recording of brain wave activity to produce detailed statistical analyses and create color maps of brain functioning. QEEG can provide additional measurements of EEG brain activity that are not possible with analog EEG recordings. This technology has increasingly been employed in tracking CNS disease progression and in clinical trials as an early predictor of CNS drug efficacy. Several QEEG measures have been correlated with performance on cognitive tests for

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

177

MCI subjects and AD patients [56, 57]. In addition, QEEG has been used in a wide variety of clinical trials to evaluate patients’ response to therapeutics, including antidepressant efficacy in depressed patients [58, 59] and the efficacy of cognitive enhancers for patients suffering from vascular dementia [60].

Cognition in early-phase development The value of cognitive testing in early-phase work has been firmly established beginning well over 20 years ago by Cutler and associates [61], who routinely used computerized cognitive testing in both normal healthy volunteer and patient studies to aid early drug development. Early on, cognitive tests were used mainly to assess how much of the CNS was affected by the disease, as well as to assess any possible CNS side effects such as difficulties in memory and attention/concentration. As many CNS drugs have difficulty in crossing the blood-brain barrier (BBB), the use of cognitive measures, especially measures of attention/concentration, was helpful in determining whether the drug was having any specific acute action in the brain and assessing the relative potencies of drugs in dose-finding studies. The advent of CNS compounds that had purported cognitive-enhancing effects expanded the use of cognitive tests to gather evidence regarding efficacy signals as pharmacodynamic measures (even though they are rarely powered to show significance) that could then be statistically associated with plasma and cerebrospinal fluid (CSF) kinetics. This type of study and analysis has been done in normal healthy volunteers and various patient groups including children and adults with attention deficit disorder, patients with epilepsy, sleep disorders, and with various neurodegenerative disorders such as AD. One of the earliest attempts to utilize a computerized cognitive test battery to help characterize the pharmacokinetics–pharmacodynamics of a CNS drug was conducted by Brass et al. [62]. They noted significant correlations between cognitive test measures and CNS activity in a scopolamine-induced learning and memory impairment model. They used this classic cognitive impairment paradigm to evaluate the CNS activity of a cholinomimetic (SDZ ENS-163) by examining 18 healthy control subjects in a crossover design. They reported that the administration of placebo with scopolamine caused significant cognitive impairment, as assessed by a computerized cognitive test battery. In contrast, SDZ ENS-163 with saline had no effect on cognitive test scores, suggesting that the dose of test drug had cholinomimetic activity in normal controls and that the dose was insufficient to reverse the muscarinic effects of scopolamine. In one of the first cognitive assessments of a cholinesterase inhibitor in a patient group, Gobburu et al. [63] measured rivastigmine and its active (major) metabolite in 18 patients with AD. They examined plasma and CSF concentrations to determine acetylcholinesterase (AChE) activity. They also used a computerized neuropsychological battery to assess cognitive domains related

c05

BLBK249-Cutler

178

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

to memory, attention, language, motor, and executive functioning. They concluded that rivastigmine and its metabolite exhibited dose-disproportional pharmacokinetics, but failed to show any statistically significant correlation between cognitive test scores and cholinesterase inhibition, or between cognitive test scores and metabolite concentration, in either plasma or CSF. However, other investigations by this group have reported relationships between cognitive test performance and drug-induced enzyme inhibition in the CSF of AD patients. For example, Giacobini et al. [64] assessed the relationship between inhibition of AChE and butyrylcholinesterase (BuChE) activities in CSF and cognitive changes as measured by a computerized test battery. They administered a ChE inhibitor, rivastigmine (Exelon) to 18 patients with mild to moderate AD and correlated various measures of cognition with AChE and BuChE in the CSF, and with plasma BuChE levels prior to treatment with rivastigmine. Giacobini et al. concluded that AChE in CSF and BuChE in plasma were dose-dependently inhibited by rivastigmine treatment and the inhibition of BuChE in CSF was not clearly dose-dependent. Importantly, a statistically significant correlation was observed between the changes in cognitive test battery global summary score and inhibition of AChE activity (r = −0.56, p < 0.05) and BuChE activity (r = −0.65, p < 0.01) in CSF. Furthermore, improvements in subtests measuring speed, attention, and memory correlated significantly with inhibition of BuChE activity (but not AChE activity) in CSF. Weak or absent correlation with change in cognitive performance was noted for inhibition of plasma BuChE. This was the first study to show significant relationship between measures of cognitive function and CSF biomarkers. More recently, data from the multiple ascending dose study of the phase I clinical trial program of MEM 3454, a nicotinic alpha-7 agonist program for the treatment of AD and cognitive impairment associated with schizophrenia (CIAS), suggested the benefit of using the cognitive drug research (CDR) test battery [65]. Memory Pharmaceuticals reported that a 15 mg dose of MEM 3454 administered once daily for a period of 13 days showed a statistically significant effect on the quality of episodic secondary memory (QESM), one of the study’s primary efficacy variables. This data was used as a basis to move forward with a phase 2a trial. QESM is a composite score derived from memory tests in the CDR battery that measures the efficiency by which study participants are able to remember words and pictures. The phase I study was a randomized, double-blind, placebo-controlled study of three doses of MEM 3454 (15, 50, and 150 mg) and involved 48 healthy young male and female volunteers. The primary purpose of this study was to investigate the safety, tolerability, and pharmacokinetics of multiple ascending doses of MEM 3454 in healthy volunteers. The secondary objective was to assess the cognitive effects of the various doses. Notably, the effect seen on the QESM at 15 mg in the phase I study was supported by preclinical work—the other doses administered in the study did not show a similarly statistically significant effect, although there was a trend toward efficacy at the 50 mg dose. Other domains

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

179

in the CDR battery measure other areas such as psychomotor speed and attention, and while trends toward improvement were also seen in these domains at 15 mg of MEM 3454, the results were not as substantial as those obtained for the QESM domain. This highlights the importance of picking the correct cognitive domain for assessment. If improvement can be detected with a battery of cognitive tests in healthy volunteers, such a battery could be a pharmacodynamic marker in the future development of the compound for treatment of cognitive disorders. To assess the merits of a cognitive test battery and to determine if changes in cognition from a pharmacological intervention could be detected in healthy volunteers in a phase I clinical trial, Beglinger et al. [66] ran a battery of neuropsychological tests on 27 healthy adults who were administered 14 days of the AChE inhibitor donepezil at 5 mg qd placebo or no treatment. Notably, carryover effects of repeated test administration were also assessed, as the battery of cognitive tests was administered five times during the study. Belinger et al. reported no differences in cognitive test performance between the placebo and the no-treatment groups. However, on Day 21, subjects in the donepezil group performed significantly worse on some tests of speed, attention, and memory compared with the pooled control group (placebo and no-treatment arms). These results were counter to expectations but importantly did not support the existence of practice or carryover effects. Practice effects, or the enhancement of performance based solely on prior exposure to testing stimuli and procedures, have been touted as a major flaw in phase I studies that incorporate cognitive measures to assess pharmacodynamic–pharmacokinetic relationships. This is especially an issue when neurologically normal subjects are involved, and when repeat testing is conducted at short time intervals—as occurs in most phase I studies. To assess whether practice effects have a significant impact on test performance when conventional neuropsychological tests are administered at test–retest intervals of short duration, Falleti et al. [67] utilized a computerized cognitive battery designed specifically for the repeated assessment of cognition (CogState). In addition, the study was to examine how many times the battery needed to be completed before performance, as measured by the battery, stabilized. Fortyfive subjects (ages 18–40 years) completed the battery four times at 10-minute test–retest intervals, and a fifth time at an interval of 1 week. The results suggested that when brief test–retest intervals were used (of 10 minutes) performance stabilized after the second assessment. Thus, significant practice effects were generally stronger and observed mainly between the first and the second assessments. However, some practice effects were also observed on some of the tasks at a 1-week test–retest interval. Because of this finding, 55 adults (ages 18–40 years) completed the battery twice at 10-minute test–retest intervals (i.e., to eliminate the initial practice effect), and a third time at an interval of 1 month instead of 1 week. This time no practice effects were observed, stressing the importance of widely spacing repeated cognitive testing procedures.

c05

BLBK249-Cutler

180

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

In an effort to elucidate the issues surrounding the inclusion of a computerized cognitive test battery in phase I clinical trials, such as practice effects, this same group administered a 12-minute battery of five computerized cognitive tasks to 28 healthy males in a double-blind, single ascending dose study using three doses of midazolam (0.6, 1.75, and 5.25 mg) with placebo insertion conducted at two different phase I sites [68]. Statistical analyses sought to determine the sensitivity of the test battery to sedation-related cognitive dysfunction and to assess the effects of practice. Collie et al. [68] reported no significant differences in data collected between the two sites enrolling and conducting the study. Additionally, no practice effects were noted on four of the five cognitive tasks. However, analyses comparing baseline with post-baseline results revealed significant and large cognitive deterioration on all five cognitive tasks 1 hour following administration of 5.25 mg midazolam. Smaller significant changes were observed on a subset of memory and learning tasks at 1 hour post-dosing in the 1.75 mg condition, and at 2 hours post-dosing in the 5.25 mg condition. This highlights the utility of using cognitive measures to dissociate cognitive effects from sedating effect and understanding the pharmacokinetic/pharmacodynamic relationship prior to entering pivotal late-phase trials. A major problem in determining whether and to what extent drugs may produce cognitive effects is that there are no standard effective means for objectively assessing cognitive impairments associated with pharmacological treatments. The Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS) initiative was one attempt to deal with standardization in later registration trials designed to be conducted across a large number of sites. However, the test battery proposed by MATRICS is cumbersome and time-consuming, taking over an hour to complete and requiring equipment and well-trained staff. This type of battery would not be appropriate for assessing cognitive functioning in the phase I environment, which demands very brief measures that can be completed in 10–20 minutes without interfering other important phase I procedures. Information about cognition can be gathered from either single tests that provide multiple outcome measures or a compilation of multiple tests that can be combined to yield a single global cognitive measure. Although there is no data that compares test features within a given trial, in general there is some sentiment that there are certain advantages to using combined or global scores (Z score) in the assessment of patient groups while single cognitive domains/test scores tend to be more advantageous in studies utilizing normal healthy controls, at least in a correlational analysis. Furthermore, when assessing normal healthy volunteers, measures of executive functioning and working memory, and not episodic memory, have generally been seen to be beneficial. Obviously, any cognitive test measure must have alternate forms, and the timing of the testing appears to influence practice effects. However, optimal test timing to reduce practice effects may not coincide with the timing of pharmacokinetic variables, and it is often difficult to obtain cognitive information that requires

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

181

written responses during blood draws and other routine procedures. Also, there does not seem to be any clear advantage of computerized test batteries over standardized batteries once scoring is taken into account. Computerized batteries may have some advantages with regard to standardized administration across multiple sites, but this is not typically a problem in the Phase I environment—studies are typically conducted at a single inpatient site with greater experimental control. Finally, there are numerous companies and computerized cognitive test batteries to choose for use in early-phase studies, including, but not limited to, the Computerized Neuropsychological Test Battery (CNTB), CogTest, CogState, CDR, Cambridge Neuropsychological Test Automated Battery (CANTAB), Mindstreams, IntegNeuro, and CNS Vital Signs. Some of these batteries have been developed for use in specific patient populations while some have been specifically developed, or can be easily adopted, for the confines of a phase I environment. There does not appear to be any great advantage of a single computerized test battery over another. In summary, it is evident that biological diagnostic markers including cognition may be utilized to better target a patient’s disease, to help identify subpopulations and related disorders of interest, to improve the efficacy of studies by enrolling only those patients with the targeted condition, to help establish novel drug targets such as CIAS, and to develop drugs to treat specific components of a disease.

MATRICS The NIMH funded an initiative called the MATRICS program. It was established to overcome the above obstacles by bringing together representatives of academia, industry, and government in a consensus process (available at http://www.matrics.ucla.edu/). The specific goals of the MATRICS are to catalyze regulatory acceptance of cognition in schizophrenia as a target for drug registration; to promote development of novel compounds that enhance cognition in schizophrenia; to leverage economic research power of industry to focus on important but neglected clinical targets; and to identify lead compounds and (if deemed feasible), support human proof of concept trials for cognition in schizophrenia. Expressly, the MATRICS initiative was established to provide a method for developing and registering potential cognitive-enhancing compounds designed to treat CIAS, which is now considered to be a legitimate target for drug development from a regulatory perspective in the United States. Thus, one of the major accomplishments of the MATRICS initiative was to ameliorate the long-standing problem of pseudospecificity. The FDA considers a drug claim to be “pseudospecific” if its target function was found to be artificially narrow. For example, a proposed claim for a drug in a psychiatric illness would be considered pseudospecific if it was found to be artificially narrow by focusing on a subgroup within the ill population or on a particular aspect of the illness (such as a particular symptom) in the absence of any empirical evidence to support such a restricted focus. As such, the claim would serve

c05

BLBK249-Cutler

182

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

only to permit a promotional advantage by implying that it is preferred over other drugs in the class for the subgroup or symptom of interest; and since no such advantage has been demonstrated, such claims would be misleading. An example of pseudospecificity would be to focus on a specific symptom of a recognized entity (e.g., targeting hallucinations in schizophrenia) [69]. The issue of ‘‘pseudospecificity’’ in treating CIAS was effectively addressed for clinical designs that test the efficacy of ‘‘add-on’’ agents to existing antipsychotic therapies but has not been resolved for potential cognitive enhancers to be used as monotherapies [70]. The MATRICS test battery captures the following six standard cognitive domains: speed of processing, attention/vigilance, working memory, verbal learning, visual learning, and reasoning and problem solving. In addition to the nine tests measuring these domains, the battery also includes a measure of social cognition in the form of an emotional intelligence test. It had been successfully argued that social cognition may have a distinctive neural substrate from some of the systems that support the more typical nonsocial cognitive domains and therefore needs to be measured as well [71]. Social cognitive abilities enable subjects to interact effectively with their social environment, and poor social cognition can lead to social misperceptions, unexpected reactions to and from the person, and eventually, social withdrawal. Therefore, it is critical to community functioning [72]. Not surprisingly, deficits in social cognition are also very predictive of functional outcome, including social and vocational outcomes in schizophrenia [73]. There has been a large amount of research in social cognition in schizophrenia focusing on emotion processing, theory of mind, social perception, social knowledge, and attributional bias. The MATRICS neurocognition committee chose to include the managing emotions component of the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) [74]. The MSCEIT is based on psychometric properties such as test–retest reliability, utility as a repeated measure, relationship to outcome, and practicality/tolerability. This test measures one of the four basic components (branches) of emotional intelligence: identifying, using, understanding, and managing emotions. In a recent study of 50 schizophrenia patients, Kee et al. [75] reported that patients performed significantly worse than controls on the total MSCEIT score, and on three of the four subtests: identifying, understanding, and managing emotions. Among patients, lower MSCEIT scores significantly correlated with higher negative and disorganized symptoms, as well as worse community functioning.

References 1. Pariente J, Loubinoux I, Carel C, et al. (2001) Fluoxetine modulates motor performance and cerebral activation of patients recovering from stroke. Ann Neurol. 50(6):718– 29. 2. Matthews PM. (2004) An update on neuroimaging of multiple sclerosis. Curr Opin Neurol. 17(4):453–8.

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

183

3. Rocca MA, Colombo B, Falini A, et al. (2005) Cortical adaptation in patients with MS: a cross-sectional functional MRI study of disease phenotypes. Lancet Neurol. 4(10):618–26. 4. Baciu MV, Watson JM, Maccotta L, et al. (2005) Evaluating functional MRI procedures for assessing hemispheric language dominance in neurosurgical patients. Neuroradiology. 47(11):835–44. 5. Bookheimer SY, Strojwas MH, Cohen MS, et al. (2000) Patterns of brain activation in people at risk for Alzheimer’s disease. N Engl J Med. 343(7):450–6. 6. Small SA, Perera GM, DeLaPaz R, et al. (1999) Differential regional dysfunction of the hippocampal formation among elderly with memory decline and Alzheimer’s disease. Ann Neurol. 45(4):466–72. 7. Rombouts SA, Barkhof F, Veltman DJ, et al. (2000) Functional MR imaging in Alzheimer’s disease during memory encoding. AJNR Am J Neuroradiol. 21(10):1869–75. 8. Sperling RA, Bates JF, Chua EF, et al. (2003) fMRI studies of associative encoding in young and elderly controls and mild Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 74(1):44–50. 9. Dodart JC, Bales KR, Gannon KS, et al. (2002) Immunization reverses memory deficits without reducing brain Abeta burden in Alzheimer’s disease model. Nat Neurosci. 5(5):452–7. 10. Walsh DM, Selkoe DJ. (2004) Deciphering the molecular basis of memory failure in Alzheimer’s disease. Neuron. 44(1):181–93. 11. Rombouts SA, Barkhof F, Van Meel CS, et al. (2002) Alterations in brain activation during cholinergic enhancement with rivastigmine in Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 73(6):665–71. 12. Saykin AJ, Wishart HA, Rabin LA, et al. (2004) Cholinergic enhancement of frontal lobe activity in mild cognitive impairment. Brain. 127(Pt 7):1574–83. 13. Sperling R, Chua E, Cocchiarella A, et al. (2003) Putting names to faces: successful encoding of associative memories activates the anterior hippocampal formation. Neuroimage. 20(2):1400–10. 14. Becker JT, Mintun MA, Aleva K, et al. (1996) Compensatory reallocation of brain resources supporting verbal episodic memory in Alzheimer’s disease. Neurology. 46(3):692–700. 15. Stern Y, Moeller JR, Anderson KE, et al. (2000) Different brain networks mediate task performance in normal aging and AD: defining compensation. Neurology. 55(9):1291– 7. 16. Grady CL, McIntosh AR, Beig S, et al. (2003) Evidence from functional neuroimaging of a compensatory prefrontal network in Alzheimer’s disease. J Neurosci. 23(3):986–93. 17. EVOTEC. (2008) Evotec’s EVT 101 Shown to Penetrate the Brain in Man and to Modulate Brain Activity during the Performance of Cognitive Tasks [document on the Internet]. Available from: http://hugin.info/131215/R/1204085/247053.pdf. Cited July 29, 2009. 18. Behzadi Y, Liu TT. (2006) Caffeine reduces the initial dip in the visual BOLD response at 3 T. Neuroimage. 32(1):9–15. 19. Chee MW, Choo WC. (2004) Functional imaging of working memory after 24 hr of total sleep deprivation. J Neurosci. 24(19):4560–7. 20. Corbetta M, Miezin FM, Dobmeyer S, et al. (1991) Selective and divided attention during visual discriminations of shape, color, and speed: functional anatomy by positron emission tomography. J Neurosci. 11(8):2383–402. 21. Friedman L, Glover GH. (2006) Reducing interscanner variability of activation in a multicenter fMRI study: controlling for signal-to-fluctuation-noise-ratio (SFNR) differences. Neuroimage. 33(2):471–81.

c05

BLBK249-Cutler

184

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

22. Tegeler C, Strother SC, Anderson JR, et al. (1999) Reproducibility of BOLD-based functional MRI obtained at 4 T. Hum Brain Mapp. 7(4):267–83. 23. Uylings HB, Rajkowska G, Sanz-Arigita E, et al. (2005) Consequences of large interindividual variability for human brain atlases: converging macroscopical imaging and microscopical neuroanatomy. Anat Embryol (Berl). 210(5–6):423–31. 24. McFarland HF, Barkhof F, Antel J, et al. (2002) The role of MRI as a surrogate outcome measure in multiple sclerosis. Mult Scler. 8(1):40–51. 25. Arnold D. (2008) The use of MRI to measure degeneration and progression: lesions learned from MS. The International Society for CNS Clinical Trials and Methodology (ISCTM), October 6–7, 2008; Toronto, Canada. 26. Tubridy N, Ader HJ, Barkhof F, et al. (1998) Exploratory treatment trials in multiple sclerosis using MRI: sample size calculations for relapsing-remitting and secondary progressive subgroups using placebo controlled parallel groups. J Neurol Neurosurg Psychiatry. 64(1):50–55. 27. Sormani MP, Miller DH, Comi G, et al. (2001) Clinical trials of multiple sclerosis monitored with enhanced MRI: new sample size calculations based on large data sets. J Neurol Neurosurg Psychiatry. 70(4):494–9. 28. Juottonen K, Laakso MP, Partanen K, et al. (1999) Comparative MR analysis of the entorhinal cortex and hippocampus in diagnosing Alzheimer disease. AJNR Am J Neuroradiol. 20(1):139–44. 29. Jack CR, Jr, Petersen RC, Xu YC, et al. (1997) Medial temporal atrophy on MRI in normal aging and very mild Alzheimer’s disease. Neurology. 49(3):786–94. 30. Jack CR, Jr, Slomkowski M, Gracon S, et al. (2003) MRI as a biomarker of disease progression in a therapeutic trial of milameline for AD. Neurology. 60(2):253–60. 31. Jack CR, Jr, Shiung MM, Weigand SD, et al. (2005) Brain atrophy rates predict subsequent clinical conversion in normal elderly and amnestic MCI. Neurology. 65(8):1227–31. 32. Bradley KM, Bydder GM, Budge MM, et al. (2002) Serial brain MRI at 3–6 month intervals as a surrogate marker for Alzheimer’s disease. Br J Radiol. 75(894):506–13. 33. Fox NC, Cousens S, Scahill R, et al. (2000) Using serial registered brain magnetic resonance imaging to measure disease progression in Alzheimer disease: power calculations and estimates of sample size to detect treatment effects. Arch Neurol. 57(3):339–44. 34. Fox NC, Black RS, Gilman S, et al. (2005) Effects of Abeta immunization (AN1792) on MRI measures of cerebral volume in Alzheimer disease. Neurology. 64(9):1563–72. 35. Ridha BH, Anderson VM, Barnes J, et al. (2008) Volumetric MRI and cognitive measures in Alzheimer disease: comparison of markers of progression. J Neurol. 255(4):567–74. 36. Joint Meeting with Medical Imaging Drugs Advisory Committee. (2002) Peripheral and Central Nervous System Drugs Advisory Committee [document on the Internet]. Available from: http://www.fda.gov/ohrms/dockets/AC/cder02.htm#Peripheraland CentralNervousSystemDrugs. Cited July 29, 2009. 37. Gee AD. (2003) Neuropharmacology and drug development. British Medical Bulletin. 65: 169–77. 38. Halldin C, Gulyas B, Farde L. (2001) PET studies with carbon-11 radioligands in neuropsychopharmacological drug development. Curr Pharm Des. 7(18):1907–29. 39. Lee CM, Farde L. (2006) Using positron emission tomography to facilitate CNS drug development. Trends Pharmacol Sci. 27(6):310–6. 40. Shin J, Lee SY, Kim SH, et al. (November 2008) Multitracer PET imaging of amyloid plaques and neurofibrillary tangles in Alzheimer’s disease. NeuroImage. 43(2):236–44. 41. Klunk WE, Engler H, Nordberg A, et al. (2004) Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Compound-B. Ann Neurol. 55(3):306–19.

c05

BLBK249-Cutler

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Neuroimaging and cognitive assessments in early drug development

185

42. Shoghi-Jadid K, Small GW, Agdeppa ED, et al. (2002) Localization of neurofibrillary tangles and beta-amyloid plaques in the brains of living patients with Alzheimer disease. Am J Geriatr Psychiatry. 10(1):24–35. 43. Small GW, Kepe V, Ercoli LM, et al. (2006) PET of brain amyloid and tau in mild cognitive impairment. N Engl J Med. 355(25):2652–63. 44. Ikonomovic MD, Klunk WE, Abrahamson EE, et al. (2008) Post-mortem correlates of in vivo PiB-PET amyloid imaging in a typical case of Alzheimer’s disease. Brain. 131(6):1630–45. 45. Brooks DJ. (2005) Positron emission tomography and single-photon emission computed tomography in central nervous system drug development. NeuroRx. 2(2):226–36. 46. Ward KW, Smith BR. (2004) A comprehensive quantitative and qualitative evaluation of extrapolation of intravenous pharmacokinetic parameters from rat, dog, and monkey to humans. I. Clearance. Drug Metab Dispos. 32(6):603–11. 47. Olson H, Betton G, Robinson D, et al. (2000) Concordance of the toxicity of pharmaceuticals in humans and animals. Regul Toxicol Pharmacol. 32(1):56–67. 48. Gad SC. (1990) Model selection in toxicology: principles and practice. J Am Coll Toxicol. 9: 291–302. 49. Combes RD, Berridge T, Connelly J, et al. (2003) Early microdose drug studies in human volunteers can minimise animal testing: Proceedings of a workshop organised by Volunteers in Research and Testing. Eur J Pharm Sci. 19(1):1–11. ¨ M, Grahn´en A, L˚angstrom ¨ B. (2003) Positron emission tomography microdos50. Bergstrom ing: a new concept with application in tracer and early clinical drug development. Eur J Clin Pharmacol. 59(5–6):357–66. 51. Bauer M, Claudia CW, Langer O. (2008) Microdosing studies in humans: the role of positron emission tomography. Drugs in R&D. 9(2):73–81. 52. European Medicines Agency. (2004) Position Paper on Non-Clinical Safety Studies to Support Clinical Trials with a Single Microdose [document on the Internet]. Available from: http://www.emea.europa.eu/pdfs/human/swp/259902en.pdf. Cited July 9, 2009. 53. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research. (2006) Guidance for Industry, Investigators, and Reviewers: Exploratory IND Studies. [document on the Internet]. Available from: http://www.fda.gov/RegulatoryInformation/Guidances/ucm126831.htm. Cited June 19, 2009. 54. Gevins A, Smith ME, McEvoy LK. (2002) Tracking the cognitive pharmacodynamics of psychoactive substances with combinations of behavioral and neurophysiological measures. Neuropsychopharmacology. 26(1):27–39. 55. Duffy F, Hughes J, Miranda F, et al. (1994) Status of quantified EEG (qEEG) in clinical practice. Clin Electroencephalogr. 25: 6–22. 56. Babiloni C, Binetti G, Cassarino A, et al. (2006) Sources of cortical rhythms in adults during physiological aging: a multicentric EEG study. Hum Brain Mapp. 27(2):162– 72. 57. Kwak YT. (2006) Quantitative EEG findings in different stages of Alzheimer’s disease. J Clin Neurophysiol. 23(5):456–61. 58. Cook IA, Leuchter AF, Witte E, et al. (1999) Neurophysiologic predictors of treatment response to fluoxetine in major depression. Psychiatry Res. 85(3):263–73. 59. Bares M, Brunovsky M, Kopecek M, et al. (2007) Changes in QEEG prefrontal cordance as a predictor of response to antidepressants in patients with treatment resistant depressive disorder: a pilot study. J Psychiatr Res. 41(3–4):319–25.

c05

BLBK249-Cutler

186

Trim: 6in X 9in

January 19, 2010

8:20

Char Count=

Chapter 5

60. Muresanu DF, Alvarez XA, Moessler H, et al. (2008) A pilot study to evaluate the effects of Cerebrolysin on cognition and qEEG in vascular dementia: cognitive improvement correlates with qEEG acceleration. J Neurol Sci. 267(1–2):112–9. 61. Cutler NR, Veroff AE, Sramek JJ. (1994) Alzheimer’s Disease: Optimizing Drug Development Strategies. Wiley, Chichester, UK. 62. Brass EP, Polinsky R, Sramek JJ, et al. (1995) Effects of the cholinomimetic SDZ ENS163 on scopolamine-induced cognitive impairment in humans. J Clin Psychopharmacol. 15(1):58–62. 63. Gobburu JV, Tammara V, Lesko L, et al. (2001) Pharmacokinetic-pharmacodynamic modeling of rivastigmine, a cholinesterase inhibitor, in patients with Alzheimer’s disease. J Clin Pharmacol. 41(10):1082–90. 64. Giacobini E, Spiegel R, Enz A, et al. (2002) Inhibition of acetyl- and butyryl-cholinesterase in the cerebrospinal fluid of patients with Alzheimer’s disease by rivastigmine: correlation with cognitive benefit. J Neural Transm. 109(7–8):1053–65. 65. Memory Pharmaceuticals Announces Positive Preliminary Phase 1 Cognitive Data For MEM 3454 [document on the Internet]. (2006) Available from: http://www.biospace. com/news story.aspx?StoryID=10333. Cited July 8, 2009. 66. Beglinger LJ, Gaydos BL, Kareken DA. (2004) Neuropsychological test performance in healthy volunteers before and after donepezil administration. J Psychopharmacol. 18(1):102–8. 67. Falleti MG, Maruff P, Collie A, et al. (2006) Practice effects associated with the repeated assessment of cognitive function using the CogState battery at 10-minute, one week and one month test–retest intervals. J Clin Exp Neuropsychol. 28(7):1095–112. 68. Collie A, Darekar A, Weissgerber G, et al. (2007) Cognitive testing in early-phase clinical trials: development of a rapid computerized test battery and application in a simulated phase I study. Contemp Clin Trials. 28(4):391–400. 69. Laughren TP. (August 2003) Comorbid mood disorders and medical illness: a Food and Drug Administration perspective. Biological Psychiatry. 54(3):195–9. 70. Breier A. (2005) Developing drugs for cognitive impairment in schizophrenia. Schizophr Bull. 31(4):816–22. 71. Pinkham AE, Penn DL, Perkins DO, Lieberman JA. (2003) Implications of a neural basis for social cognition for the study of schizophrenia. Am J Psychiatry. 160: 816–24. 72. Green MF, Olivier B, Crawley JN, et al. (2005) Social cognition in schizophrenia: recommendations from the measurement and treatment research to improve cognition in schizophrenia new approaches conference. Schizophr Bull. 31(4):882–7. 73. Mueser KT, Doonan B, Penn DL, et al. (1996) Emotion recognition and social competence in chronic schizophrenia. Abnorm Psychol. 105: 271–5. 74. Mayer JD, Salovey P, Caruso DR, et al. (2003) Measuring emotional intelligence with the MSCEIT V2.0. Emotion. 3(1):97–105. 75. Kee KS, Horan WP, Salovey P, et al. (2009) Emotional intelligence in schizophrenia. Schizophr Res. 107(1):61–68.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

CHAPTER 6

Bridging and CSF studies

Introduction A bridging study is a late Phase 1 (Phase 1b) or early Phase 2 (Phase 2a) safety/ tolerability study conducted in the target population [1]. The goal of the bridging study is to assess the tolerance profile and identify the maximum-tolerated dose (MTD) of a compound in the patient population in order to optimize dose selection for Phase 2/3 efficacy trials [2]. This chapter discusses the concept and execution of a bridging study. Traditional drug development consists of four phases of clinical testing in humans, with each phase setting the stage for the next phase.

Phase 1 Phase 1 includes the initial introduction of an investigational new drug into humans. These studies are closely monitored and may be conducted in patients, but are usually conducted in healthy volunteer subjects. These studies are designed to determine the metabolic and pharmacologic actions of the drug in humans, the side effects associated with increasing doses, and, if possible, to gain early evidence of effectiveness. During Phase 1, sufficient information about the drug’s pharmacokinetics and pharmacological effects should be obtained to permit the design of well-controlled, scientifically valid, Phase 2 studies. Phase 1 studies also evaluate drug metabolism, structure–activity relationships, and the mechanism of action in humans. These studies also determine which investigational drugs are used as research tools to explore biological phenomena or disease processes. The total number of subjects included in Phase 1 studies varies with the drug, but is generally in the range of 20–80. In Phase 1 studies, the Center for Drug Evaluation and Research (CDER) can impose a clinical hold (i.e., prohibit the study from proceeding or stop a trial that has started) for the reasons of safety, or because of a sponsor’s failure to accurately disclose the risk of the study to investigators [3]. Although CDER routinely provides advice in such cases, investigators may choose to ignore any advice regarding the design of Phase 1 studies in areas other than patient safety.

Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

187

c06

BLBK249-Cutler

188

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

Phase 2 Phase 2 includes the early controlled clinical studies conducted to obtain some preliminary data on the effectiveness of the drug for a particular indication or indications in patients with the disease or condition. This phase of testing also helps determine the common short-term side effects and risks associated with the drug. Phase 2 studies are typically well-controlled, closely monitored, and conducted in a relatively small number of patients, usually involving several hundred people. Phase 2 studies are often dose-finding studies that attempt to identify the dose range at which beneficial effects are most likely to occur without prohibitive side effects. Phase 3 Phase 3 studies are expanded, controlled, and uncontrolled trials. They are performed after preliminary evidence suggesting effectiveness of the drug has been obtained in Phase 2, and are intended to gather the additional information about effectiveness and safety that is needed to evaluate the overall benefit–risk relationship of the drug. Phase 3 studies also provide an adequate basis for extrapolating the results to the general population and transmitting that information in the physician labeling. Phase 3 studies usually include several hundred to several thousand people. In both Phases 2 and 3, CDER can impose a clinical hold if a study is unsafe (as in Phase 1), or if the protocol is clearly deficient in design in meeting its stated objectives. Great care is taken to ensure that this determination is not made in isolation, but reflects current scientific knowledge, agency experience with the design of clinical trials, and experience with the class of drugs under investigation. In Phase 3, the strict inclusion/exclusion criteria employed in Phase 2 are often relaxed to include a wider range of patient characteristics than previously allowed, as more information about the safety profile of the compound becomes available. On occasion, however, the criteria may be stricter if specific adverse events (AEs) are identified. Phase 3 studies will also generally use a placebo or active control group. Phase 4 Phase 4 studies, often called post-marketing surveillance (PMS) trials, are conducted after a drug or device has been approved for consumer sale. Pharmaceutical companies have several objectives at this stage: (1) to compare a drug with other drugs already in the market; (2) to monitor a drug’s long-term effectiveness and impact on a patient’s quality of life; and (3) to determine the cost-effectiveness of a drug therapy relative to other traditional and new therapies. Phase 4 studies can result in a drug or device being taken off the market or restrictions of use could be placed on the product, depending on the findings in the study. Phase 4 studies are designed to evaluate factors not fully explored in previous phases. These factors can include the effects of a compound in special patient populations or its potential use in other indications. For example, the

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

189

New Drug Application (NDA) for the antipsychotic risperidone did not have adequate information on drug effects in elderly patients, thus requiring further study of this population. In addition to these studies, PMS is often performed in order to compile additional AE data from physicians who have prescribed the drug to their patients. A variety of dosing regimens and durations may also be compared for safety and efficacy. In order to maximize the potential to identify effective new therapies, appropriate doses must be determined as early in development as possible. The primary goal in dose selection is the identification of a range that is both safe and effective for the clinical population. Although the clinical evaluation process has remained largely unchanged over the past few decades, the need to bring novel compounds to the marketplace faster and more economically has inspired some suggestions for accelerating early drug development. One solution is the “bridging study,” in which a late Phase 1 (1b) or early Phase 2 (2a) safety/tolerance study is conducted directly in the patient population [1,4,5].

Drug response in patients versus healthy subjects Traditionally, Phase 1 safety/tolerance studies have been conducted in healthy volunteers for several reasons, including medical and legal complications associated with patients, concerns for their safety, and issues of convenience [6]. However, there is evidence that patients tolerate certain central nervous system (CNS) compounds differently than healthy subjects, and that Phase 1 studies in healthy subjects can be poor predictors of the optimal dose range for Phase 2 studies in the target population. Differences in drug response between patients and healthy subjects are not surprising, as most CNS compounds affect neurotransmitter or receptor systems that are altered in the target population. For example, several early studies reported that schizophrenic patients tolerate higher doses of antipsychotic compounds than healthy control subjects, some by as much as 200 fold [7]. Okuma [8] found that schizophrenic patients demonstrated a quantitatively lower sensitivity to the sedative effects of chlorpromazine than healthy volunteers, as demonstrated by a higher percentage time of waking EEG following dose administration. Miller et al. [9] reported that a significantly higher percentage of healthy subjects experienced dystonia/tremor, restlessness, sedation, and anticholinergic side effects following the acute administration of haloperidol. The difference in sensitivity was attributed to higher dopamine receptor levels in schizophrenic patients, resulting in less complete dopaminergic blockade following the neuroleptic challenge. Another potential mechanism for the difference in tolerability is increased postsynaptic dopamine receptor sensitivity in schizophrenic patients [9]. In one early haloperidol study in which healthy male volunteers were given up to 5 mg as a single dose, most of the volunteers needed to be rushed to an emergency room for the treatment of severe and serious reactions, including laryngeal dystonia [10]. Nor, in our experience, are the newer “atypical” antipsychotics always well tolerated in

c06

BLBK249-Cutler

190

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

healthy subjects. In generic bioequivalence studies of extended-release quetiapine, a single 200-mg dose produced moderate somnolence in virtually all subjects for around 8 hours. We also observed prominent orthostatic tachycardia, as well as dizziness and a number of episodes of orthostatic hypotension and even syncope that could persist for up to 14 hours post dose. Differences in antipsychotic tolerance between patients and healthy subjects are not limited to classic dopamine D2 receptor antagonists. In a study in which 17 healthy male subjects were administered single 25-mg doses of R ), 8 of the subjects experienced the atypical antipsychotic clozapine (Clozaril severe bradycardia (<40 beats per minute [bpm]), and 2 of these subjects had cardiac arrest lasting 10 and 60 seconds, respectively [11]. Although all subjects recovered, a recommendation was made by the US Food and Drug Administration (FDA) that future bioequivalence studies of clozapine be conducted only in schizophrenic patients. Patients have also demonstrated a higher tolerance than healthy subjects for benzodiazepines and tricyclic antidepressants [12]. Many patients with generalized anxiety disorder (GAD) can tolerate up to 30 mg of diazepam per day, while healthy subjects would be likely to experience extreme sedation at this dose. Additionally, in an early study of the novel anxiolytic lesopitron, patients with GAD were able to tolerate considerably higher doses than healthy subjects [13]. In clinical trials of compounds for the treatment of Alzheimer’s disease (AD), patients often tolerated higher doses than healthy elderly subjects, but sometimes could tolerate only lower doses [14, 15]. Additionally, Medina et al. [16] reported that AD patients demonstrated a reduced risk of tiltinduced syncope following the oral administration of xanomeline in comparison to healthy, age-matched control subjects. This result suggests that patients may have a lower susceptibility to the cardiovascular effects of M1 -receptor stimulation, presumably because of the lower muscarinic receptor activity associated with AD. The importance of determining differences in tolerance between healthy subjects and the target population is illustrated by the potential dose–response relationships of an investigational compound (Figure 6.1). While at low doses the beneficial effects may outweigh the adverse effects, eventually a crossing point may be reached (MTD) where adverse effects predominate over beneficial effects. Healthy subjects and the target population may have very different MTDs for the same compound. For the majority of CNS-active compounds, both treatment response and toxicity tend to increase linearly with dose. Although compounds with more complex receptor pharmacology can sometimes demonstrate more complicated dose–response relationships, this general trend suggests that the potential to detect efficacy is enhanced at the highest tolerable doses of a compound. A bridging study determines the highest and safest dose that can be administered to the patient population. If patients are able to tolerate higher doses of a compound than healthy subjects, conducting a bridging study prior to Phase 2 could help researchers to avoid selecting a dose range that is too low,

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

191

Adverse effects Maximum tolerated dose (MTD)

Beneficial effects

Effect

c06

Benefit > 1 Risk

Benefit < 1 Risk

Dose Figure 6.1 Example of a dose–response relationship. (Reproduced by permission from Cutler et al. [19].)

thus optimizing the chances of detecting efficacy. Conversely, if patients can tolerate only lower doses of a compound, a bridging study could enhance the safety of Phase 2 by avoiding selection of a dose range that is too high. Another important reason for conducting a bridging study has to do with the typically linear dose–response relationship that is seen with CNS drugs. That is, the higher the dose, the better the response, until the dose becomes too high and AEs intervene. The only known exception to this is perhaps the antidepressant nortriptyline, which displays an inverted U dose–response relationship, requiring optimal dosing based on plasma concentrations. Because CNS drugs often do not have robust effects, it is important to optimize the dose to be able to demonstrate the response. R ) failed in more Phase 3 trials than it succeeded Even fluoxetine (Prozac [17]. This is because the difference between an effective drug and placebo, as measured by psychometric rating scales, is often only a few (4–5) points. Unless the dose is optimized, the possibility of detecting that response is comproR ), which mised. Another example of this is the anxiolytic buspirone (BuSpar is discussed later in this chapter. Very often, clinical development may be biased based on the response in animal models, which often display an inverted U relationship, thus sensitizing developers to expect such a relationship in patients. Yet in our collective experience, we have yet to see a true example of inverted U response relationships in humans. Even if an investigational compound were to possess such a response relationship, it would certainly be detected in Phase 2 dose-ranging trials, which typically employ several doses of the compound versus placebo and even a standard control (i.e., marketed drug). In such a trial, the top dose would be selected based on the MTD found in the bridging study (sometimes adjusted slightly downward, depending on the nature of the AEs), a middle dose (perhaps 50% of the top dose), and a very low dose (10% of the top dose).

c06

BLBK249-Cutler

192

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

The bridging study Definition of bridging study A bridging study is a late Phase 1 (Phase 1b) or early Phase 2 (Phase 2a) safety/tolerability study conducted in the target population [1,4]. The goal of the bridging study is to assess the tolerance profile and identify the MTD of a compound in the patient population in order to optimize dose selection for Phase 2/3 efficacy trials [2]. With a bridging protocol, we have been able to initiate drug development for investigational new drugs (INDs). Please note that the FDA defines bridging studies differently than in this book. The FDA defines bridging studies as comparing the impact of drugs on different populations, e.g., Japanese and Caucasian. The bridging study discussed in this book refers to studies designed to identify the MTD of an investigational compound. Determining the MTD in patients maximizes the potential to detect efficacy by permitting the use of the highest tolerated doses in Phase 2, while providing a good understanding of potential AEs. When planning a bridging study, one must pay close attention to all available drug information, beginning with the animal data. Reaching and defining a minimally intolerable dose (MID), by definition, means that some patients will be exposed to minimally toxic doses, and one must prepare for what might be expected based on the compound’s mechanism of action and AE profile seen at outright toxic doses in animals. Any preclinical AEs that occur nonlinearly with dose raise a red flag for particular caution with respect to the design (fixed versus titration) and choice of dose increments to be used in humans. Typically, the AEs seen at MID will be extensions of the compound’s pharmacology, and indeed these can often help to further characterize that pharmacology. For example, a compound that is defined preclinically as being selective for a certain receptor subtype may, upon reaching MID, display an AE profile suggestive of action at another receptor or subtype, thus losing its specificity at higher doses. Researchers must be alert to such findings by having a thorough knowledge of the preclinical profile, and be prepared to use any available technologies to monitor such symptoms, or employ biomarkers that might be helpful. Above all, patient safety must be the primary goal, and when a MID is reached, the investigator must stop the study. A critical care physician must be available to treat patients at a MID, and such treatment must be based on the scope and severity of symptoms seen at that time, rather than a potentially theoretical treatment or antidote (which may be offered in the compound’s investigator brochure). The bridging study methodology In a bridging study, consecutive panels of patients each receive progressively higher fixed doses of a study drug until a MID is reached. The MID can be defined as the dose at which 50% of patients experience severe or multiple AEs probably related to the study drug, or one patient experiences a serious

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

193

Table 6.1 Bridging study: a recommended fixed-dose design Fixed-dose panel design: inpatient assessment Panel 1: 50% less than MTD Panel 2: 25% less than MTD Panel 3: MTD based on healthy subjects Panel 4: 25% greater than MTD Panel 5: 50% greater than MTDa Reproduced by permission from Cutler et al. [19]. a Study may require panels with higher doses to define the MTD.

AE (defined as medically unacceptable) believed to be related to the study drug. The dose immediately below this one is the maximum tolerated dose (MTD). One important point, however, is that the definition of what is medically unacceptable might differ depending on the population under study. For example, what is considered unacceptable for a healthy subject might not be so for a patient with a chronic illness. Thus, indication-specific prospective criteria should be identified prior to initiation of a trial, in order to determine an accurate MTD. The design of a bridging study is dependent on the particular compound under investigation. Generally, bridging studies for AD compounds employ a fixed-dose panel design, in which consecutive panels of 6–8 patients each receive doses that range from 50% below to 50% or more above the MTD in healthy subjects (see Table 6.1). Each of the consecutive panels is dependent on the tolerability of the previous dose level. The dose level is increased until the MID is reached; the dose immediately preceding the MID is designated the MTD. The duration of treatment in a bridging study will depend on several factors, but one important starting point is the half-life of the compound; including the half-life for elimination from the CNS (see the section “Dynabridge” for more information on this). Each fixed-dose panel should receive treatment until steady state (SS) is reached (5–7 half-lives) and continue after that for several days in order to fully evaluate the effects of that dose; this strategy both maintains the dose and allows potentially slow metabolizers to attain SS. Thus, a compound that has a 24-hour half-life should be dosed for 8 or 9 days. Another consideration is food effect, and in the absence of any data at an early stage of development, we have preferred to dose on an empty stomach in order to maximize the potential for detecting AEs. Panels should also comprise both genders to allow for detection of potential gender differences. In some cases, it is beneficial to conclude the study with titration of a final panel in order to determine any differences in tolerability between fixed and titrated doses, as titration may redefine the MTD. An example of a titration panel design is shown in Table 6.2. A titration design will typically be suggested if the compound displays binding to receptors known to induce

c06

BLBK249-Cutler

194

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6 Table 6.2 Bridging study: a recommended titration design Titration panel doses: inpatient assessment

Duration

50% less than MTD 25% less than MTD Fixed-dose MTD in patients 25% greater than MTD 50% greater than MTD

×4 days ×4 days ×4 days ×4 days ×4 days

Reproduced by permission from Cutler et al. [19].

tolerance, or if the animal data suggest development of tolerance with repeated exposure. For example, the results of previous fixed-dose studies could be reviewed to determine the AE profile. If 1 week was required to achieve tolerance for some AEs, and several weeks for others, a slow titration schedule for this compound over the first week of dosing would likely decrease the incidence of dose-limiting AEs. In addition, titration over the first 2–3 weeks would help to decrease the incidence of other AEs. By examining the results of available fixed-dose studies, titration schedules tailored to the AE profiles of the compound can be determined and tested, and higher doses can potentially be reached. A bridging study for a compound that needs to be titrated is typically divided into three distinct periods, each with a different objective [7].

Period I The MTD is determined in a panel of approximately eight patients on a slow titration schedule. The dose is titrated upward every 3 or more days, depending on the profile of the particular compound. The MTD from this period serves as the maximum dose for subsequent panels. The maximum dose is sometimes also based on the no observed adverse effect level (NOAEL) determined from preclinical studies in two species, even if a MTD is not reached in patients. If all doses in Period I are well tolerated, then an additional panel reaching higher doses may be required to define the MTD. Period II The safety and tolerability of fixed doses of the study drug are evaluated in consecutive panels of up to eight patients each, until a fixed-dose MTD is reached. The doses tested are selected based on the slow titration MTD, e.g.: 20%, 40%, 60%, 80%, and 100% of the slow titration MTD. The fixed-dose MTD can then be used as the starting dose for subsequent titration panels. As it is generally beneficial to reach a target dose quickly, particularly for patients with acute symptoms, the identification of a higher initial dose could help to speed the titration process.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

195

Period III Once the fixed-dose MTD is determined, Period III can evaluate a series of progressively faster titration schedules, beginning each panel with the fixeddose MTD and ending with the slow-titration MTD. An added advantage of the three-period design is that a pharmacokinetic profile of the compound can be assessed in patients. Trough or peak concentrations approximating SS (depending on the half-life of the compound) can be drawn at the end of each dosing interval in the slow titration period (Period I); a full pharmacokinetic profile can be taken at several steady-state dose levels in the fixed-dose period (Period II); and the profile can also be repeated at the MTD dose in Period III, with an appropriate extension of the dosing at the conclusion of a rapid titration schedule. This extension of dosing also confirms that the MTD will be well tolerated after the dose is fully allowed to reach steady-state concentrations. In the event that the dosing of a particular antipsychotic compound does not require titration, panels of patients can still be used to evaluate the safety and tolerability of a series of fixed doses. As with the titration study, each of the consecutive panels is dependent on the tolerability of the previous dose level. It is still of some benefit to conclude the study with titration of the last panel in order to determine any differences in tolerability between fixed and titrated doses, as titration may redefine the MTD. Bridging studies can thus determine the optimal dose range for Phase 2 studies in patients and identify the best titration schedule to reach the top dose quickly. While adding pharmacokinetic sampling to a bridging study can be done, we caution that overly extensive kinetics can interfere with the careful AE evaluations, which are the primary purpose of the study. If added, a pharmacokinetic profile for the terminal half-life would be preferable to an initial profile following the first dose. In order to maximize the safety of a bridging study, the animal toxicology of the study compound at dosages beyond anticipated human dosages should be explored to provide the best possible understanding of potential AEs, and subject selection should be carried out with care [1,4,5,14]. Patients should be in good physical health, with no significant concurrent medical disease (excluding the indication under study). Additionally, phenotyping for drugmetabolizing enzymes could be beneficial for understanding and correlating responses of poor and extensive metabolizers to the investigational compound. Because bridging studies examine doses at the high end of the tolerability spectrum, they should be conducted only in facilities with hospital quality intensive care equipment and qualified critical care nurses and physicians. In order to minimize patient discomfort or risk, rigorous and careful supervision is required. Patients should be monitored closely and questioned frequently in order to avoid the possibility that negative experiences will go unreported. In previous bridging studies reported by Cutler and Sramek [5], a panel of 6

c06

BLBK249-Cutler

196

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

patients was attended by no fewer than 2 critical care nurses in addition to ancillary staff (as necessary) on a 24-hour basis.

Bridging studies in Alzheimer’s disease The authors’ first experiences with bridging studies involved the evaluation of compounds for the treatment of AD, including several acetylcholinesterase (AChE) inhibitors and muscarinic agonists. Through these studies, bridging methodology evolved into the designs outlined in the previous section. In addition, the results of these studies emphasized the value of determining the MTD in the target population prior to selection of a dose range for Phase 2/3 efficacy studies. Each study provided valuable insight into the importance and usefulness of bridging studies, as shown in Table 6.3. Details of a few selected studies follow. SDZ ENA 713 Background SDZ ENA 713 is a centrally active acetylcholinesterase inhibitor of the carbamate type. It has demonstrated more potent AChE inhibition in cortex and hippocampus (both targets for symptomatic treatment of AD) than in other brain areas, and has shown a longer duration of cholinesterase blockade than tacrine [18]. Early safety and tolerability studies of SDZ ENA 713 in healthy volunteers reported that multiple doses up to 3 mg/day were well tolerated. AEs, including headache, dizziness, nausea, and diarrhea, were generally mild in intensity and were not dose limiting. However, in an initial placebocontrolled efficacy study of SDZ ENA 713 in 402 AD patients, doses of up to 2 and 3 mg BID did not result in significant treatment effects after 13 weeks [19]. Bridging study design The design was a double-blind, placebo-controlled, parallel-group bridging study to determine the MTD of SDZ ENA 713 in patients with AD in order to maximize the potential for detecting efficacy in subsequent efficacy trials [20]. Patients were required to meet the National Institute of Neurological and Communicative Diseases and Stroke/Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria for probable AD and to have a Mini Mental State Examination (MMSE) score between 10 and 26. Patients were excluded if they had any medical, neurological, or other psychiatric disorders that could confound safety evaluations. A total of 50 patients (22 males, 28 females, average age of 68 years) were randomized to receive SDZ ENA 713 BID (n = 20), SDZ ENA 713 TID (n = 20), or placebo (n = 10). Escalating doses for both SDZ ENA treatment groups are shown in Table 6.4. If a dose was poorly tolerated, patients were allowed to skip up to six doses at each level, but not more than three doses in sequence.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

197

Table 6.3 Key findings related to bridging studies in patients with Alzheimer’s disease Drug

Key findings related to bridging studies

Sources

Velnacrine

Velnacrine (HP 209) is a centrally acting acetylcholinesterase inhibitor that demonstrated efficacy in animal models of memory loss. Studies with velnacrine demonstrated that patient selection is very important in the bridging study, as the response for a single patient curtailed further dose escalations. The rate of titration should be considered, as a slower titration might have increased the MTD of velnacrine through the development of tolerance. The study found AD patients tolerated lower doses of velnacrine than both healthy young and elderly subjects.

[72–75]

Eptastigmine

The 48 mg TID dose was designated the MTD because of the severe AEs observed in 2 patients at the 52 mg dose. The bridging study proved that the MTD in AD patients was 50% higher than the highest tolerated dose in healthy subjects.

[76, 77]

SDZ ENA 713

This study extended the functional dose range well beyond the highest tolerated single dose in healthy subjects. The evaluation of higher doses was advantageous in the development of the drug.

[18,20–22]

Metrifonate

Oral administration of metrifonate results in sustained acetylcholinesterase inhibition in both humans and animals and is associated with improved cognitive performance in animal models of memory loss. The maintenance dose defined by the bridging study was substantially greater than the highest doses previously evaluated in healthy elderly subjects or in AD patients.

[78–83]

CI-979

CI-979 is a muscarinic agonist that has demonstrated central activity in preclinical studies of cognitive dysfunction, including reduction of spatial memory deficits induced by cortical lesions. Due to the severity of AEs observed at 2.5 mg q.6h and 3 mg q.6h, 2 mg q.6h was designated the MTD in AD patients, a dose twice as high as the MTD found in healthy young subjects.

[84–86]

Xanomeline

The MTD determined in the bridging study was twice the MTD reported for the healthy elderly population. This study directly illustrated the advantages of conducting a bridging study prior to dose selection for Phase 2/3 efficacy studies. In a later efficacy study of xanomeline, only the highest (75 mg TID) of three doses evaluated (25, 50, and 75 mg TID) was found to be superior to placebo. This dose would not have been included without the information gained in the bridging study.

[23–27]

Lu 25-109

Studies in healthy elderly subjects reported that doses up to 130 mg QID were generally well tolerated. The bridging study defined the fixed-dose MTD as 150 mg TID and determined that titration did not appear to improve the tolerability of Lu 25-109.

[87]

c06

BLBK249-Cutler

198

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

Study period

Dose

Days 1–3 Days 4–7 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Weeks 8–9

2 mg/day 3 mg/day 4 mg/day 5 mg/day 6 mg/day 7.25 mg/day 8.5 mg/day 10 mg/day 12 mg/day

Table 6.4 Dose-escalation schedule, bridging study of SDZ ENA 713

Bridging study results Three patients receiving SDZ ENA 713 discontinued from the study because of AEs. One of these patients experienced a mild atrial fibrillation after receiving 4 mg/day SDZ ENA 713 (TID regimen) and withdrew from the study; this event was later determined to be unrelated to study medication. Another patient began experiencing intermittent mild to moderate nausea and vomiting at 5 mg/day and discontinued from the study after experiencing nausea and severe vomiting at 12 mg/day (TID regimen). The most common AEs reported for patients receiving SDZ ENA 713 were headache, nausea, dizziness, diarrhea, vomiting, and fatigue. Except for the patients who prematurely discontinued, AEs associated with SDZ ENA 713 were generally transient and did not recur at higher doses. Thus, all doses were considered to be well tolerated, and no MTD was determined. Further doses were not explored, as 12 mg/day was the highest dose felt to be safe based on the preclinical NOAEL. Although this study did not reach an MTD, it extended the functional dose range well beyond the highest tolerated single dose in healthy subjects (3 mg/day). Conclusions The evaluation of higher doses was advantageous in the development of SDZ ENA 713, as preliminary evidence from ongoing studies indicated that beneficial effects in patients with AD had been observed at doses of 3–12 mg/day [21, 22]. The drug was eventually approved and is marketed under the trade R (rivastigmine). name Exelon

Xanomeline Background Xanomeline tartrate is a potent and selective muscarinic agonist that has demonstrated high activity at cloned M1 receptors [23], and has been shown to cross the blood–brain barrier in humans [24]. In an escalating single-dose study of xanomeline, the safety and tolerability of 1, 5, 10, 25, 50, 75, 100, and 150 mg were investigated in 36 young healthy subjects. Adverse events, including diarrhea, diaphoresis, disorientation, increased diastolic blood

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

199

pressure, nausea, and postural hypertension were observed at the 100- and 150-mg doses. A multiple-dose study of xanomeline reported that doses of 75 mg BID were well tolerated in healthy young subjects (n = 12). No MTD was determined in either study [25]. The safety and tolerability of xanomeline were also evaluated in 4 panels of healthy elderly subjects (n = 16). Subjects received 1 of 4 dose regimens of xanomeline: titration from 50 to 75 mg BID, titration from 15 to 25 mg TID, titration from 40 to 50 mg TID, or 40 mg TID. Adverse events such as moderate diarrhea, nausea, vomiting, diaphoresis, and hypotension were observed at all doses. One subject experienced moderate nausea and symptomatic hypotension at 50 mg TID, and no further doses were tested. The 50 mg TID dose was designated the MTD in this population. Bridging study design The design was a double-blind, placebo-controlled, inpatient study [26] to evaluate the safety and tolerability of xanomeline in 48 patients with AD (20 males, 28 females, mean age of 72 years). All patients met NINCDS-ADRDA and DSM-III-R criteria for AD and scored 4 or less on the Modified Hachinski Ischemia Scale and between 10 and 26 on the MMSE at screening. None of the patients had a clinically significant condition that could interfere with study assessments. The study was conducted in eight panels of six patients each. Patients were randomized to receive placebo (n = 2 per panel) or 1 of 8 ascending fixed doses of xanomeline (n = 4 per panel) for 7 days. Doses were 25, 35, 50, 60, 75, 90, 100, and 155 mg TID; progression to the next panel was contingent upon the tolerability of the previous panel. Bridging study results One patient in each of the 60, 75, and 100 mg TID panels prematurely discontinued due to severe gastrointestinal (GI) AEs. An additional patient receiving 100 mg TID elected to discontinue treatment on the first day of dosing due to AEs that were considered moderate but tolerable by the investigator. The most common AEs in patients receiving xanomeline were GI disturbances (diarrhea, nausea, and abdominal pain), diaphoresis, lacrimation, and dizziness. In the 115 mg TID panel, one patient withdrew from the study after experiencing severe nausea and vomiting on the first day of dosing. The study was terminated when another patient in the 115 mg TID panel experienced severe hypotension, nausea, vomiting, diaphoresis, and pallor, as well as moderate tightness of the chest, hypersalivation, and lethargy. Thus, the 115 mg TID dose was designated the MID and 100 mg TID was defined as the MTD of xanomeline in this population. Conclusions Although this bridging study was conducted in a relatively small number of patients, the results were highly predictive of the AE profile in a

c06

BLBK249-Cutler

200

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6 Table 6.5 Incidence of adverse events: bridging study versus large multicenter study Adverse event

Bridging study (n = −32)

Multicenter study (n = 256)

Sweating Nausea Vomiting Dyspepsia Chills Salivation Chest pain

44% 38% 25% 22% 25% 13% 6%

46% 38% 32% 25% 24% 11% 11%

Reproduced by permission from Cutler et al. [19].

subsequent Phase 2 multicenter study. As seen in Table 6.5, the occurrence rates of the most common AEs were very comparable between the two studies. This potential to identify AEs of concern provides a means of establishing safety guidelines for higher dose ranges, and could improve the safety of Phase 2 trials. The MTD of 100 mg TID is twice the MTD reported for the healthy elderly population. This study directly illustrated the advantages of conducting a bridging study prior to dose selection for Phase 2/3 efficacy studies. In a later efficacy study of xanomeline, only the highest (75 mg TID) of 3 doses evaluated (25, 50, and 75 mg TID) was found to be superior to placebo [27]. This dose would not have been included without the information gained in the bridging study. In the above bridging studies, the difference in drug tolerance between AD patients and healthy subjects has several possible explanations. For example, the initial Phase 1 assessments in normal subjects could be inaccurate, or the pharmacological changes associated with AD could account for the differences in drug response. Although the underlying mechanism is not known, the observed differences between patients and normals support the need for bridging studies prior to dose selection for Phase 2/3.

Bridging studies in anxiety Sramek et al. [13] conducted a bridging study of a novel (at that time) potential anxiolytic, lesopitron, in patients with GAD. This study employed the same design that was developed for trials of AD compounds, and illustrates the utility of bridging in multiple CNS indications (see Table 6.6 for the key findings of the study). Lesopitron Background Lesopitron is a 5-HT1A agonist chemically related to the azapirone class [28]. In preclinical models of anxiety, lesopitron was equally effective or superior to diazepam and other azapirones with no apparent withdrawal symptoms [29– 31]. Single oral doses of lesopitron of 0.2–50 mg were well tolerated in healthy

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

201

Table 6.6 Key findings related to bridging studies in patients with anxiety Drug

Key findings related to bridging study

Sources

Lesopitron

Lesopitron is a 5-HT1A agonist chemically related to the azapirone class equally effective or superior to diazepam and other azapirones in preclinical anxiety models. The MTD of lesopitron in patients with GAD was twice the highest dose tested in healthy volunteers, permitting the use of a higher dose range for Phase 2 efficacy trials.

[13,19,28,30–34]

volunteers. A multiple-dose study of lesopitron in 60 healthy subjects found that doses up to 45 mg/day were well tolerated with no apparent relationship between dose and AEs. Thus, 45 mg was designated the MTD in healthy subjects [19]. Bridging study design The design was a double-blind, placebo-controlled, inpatient study [13] in 42 patients (27 males, 15 females; age range of 20–61 years), divided into 7 panels of 6 patients each. In each panel, 4 patients were randomized to receive lesopitron and 2 patients were randomized to receive placebo for 6.5 days. All patients had a primary diagnosis of GAD according to DSM-III-R criteria, modified to allow a minimum duration of anxiety symptoms of 1 month, and a Hamilton Rating Scale for Anxiety (HAM-A) score of at least 18 with a score greater than 2 on the “anxious mood” item. Patients were also required to score no more than 16 on the 17-item Hamilton Rating Scale for Depression (HAM-D). Patients were excluded if their HAM-A score decreased by 25% or more between the screening and baseline assessments; if they had more than 4 panic attacks in the 4 weeks prior to screening; or if they had any clinically significant medical, neurological, or other psychiatric disorders that could interfere with study evaluations. Seven consecutive panels of patients were randomly assigned to receive placebo or lesopitron at doses of 20, 25, 30, 40, 45, 50, or 60 mg BID. Progression to the next dose level was contingent upon the safety and tolerability of previous panels. Bridging study results Doses of 20–25 mg BID were well tolerated, with patients experiencing mildto-moderate AEs such as headache, dizziness, and nausea. At 60 mg BID, one patient experienced a severe episode of orthostatic hypotension accompanied by symptoms such as dizziness, lightheadedness, and diaphoresis. These symptoms resolved within 1 hour, and the patient completed the study. Two additional patients receiving lesopitron 60 mg BID experienced moderate-tosevere AEs, including dizziness, lightheadedness, nausea, and headache. Due

c06

BLBK249-Cutler

202

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

to the severity of AEs in half of the patients receiving lesopitron in this panel, 60 mg BID was designated the MID, and 50 mg BID was designated the MTD. Conclusions The MTD of lesopitron in patients with GAD was twice the highest dose tested in healthy volunteers, permitting the use of a higher dose range for Phase 2 efficacy trials. In one of the later Phase 2 studies of the efficacy and safety of lesopitron dosed at 40 mg/day to a maximum of 80 mg/day in outpatients with GAD, common AEs included somnolence, headache, and dyspepsia [32]—a similar profile as identified in the bridging study. There is some evidence that a bridging study might have been beneficial in the development of buspirone, another anxiolytic compound. Early studies of this compound evaluated what were later considered to be low doses, increasing only gradually to a maximum of 30 mg/day. These low doses were potentially responsible for the lack of positive findings in 9 of 11 pivotal efficacy studies [19]. Physicians gradually found that higher doses were tolerable, and the current recommended dose of buspirone extends as high as 60 mg/day [33]. Thus, the bridging studies could constitute an important step in the development of compounds for the treatment of anxiety. A Phase 4 study of the efficacy and safety of BID and TID dosing regimens in patients with persistent anxiety found no appreciable difference in efficacy or safety between buspirone 15 mg BID or 10 mg TID [34]. This result underscored the importance of carefully studying the optimal dosing interval as well as the dose itself in early development, as a BID regimen is more preferable than a TID regimen for patient compliance.

Bridging studies in depression A bridging study of the putative antidepressant ABT-200 by Sramek et al. [35] greatly extended the usable dose range of the compound (see Table 6.7). ABT-200 Background ABT-200 was developed as a novel potential antidepressant and racemic mixture of two enantiomers. The SS enantiomer antagonizes the presynaptic Table 6.7 Key findings related to bridging studies in patients with depression Drug

Key findings related to bridging study

Sources

ABT-200

ABT-200 was developed as a novel potential antidepressant and racemic mixture of two enantiomers which antagonizes both the presynaptic uptake of norepinephrine and inhibitory α2 autoreceptors. The bridging study extended the usable dose range by 100% over doses that were previously found not to be efficacious in patients with depression.

[35–41]

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

203

uptake of norepinephrine (NE), while the RR enantiomer antagonizes inhibitory ␣2 autoreceptors [36,37]. It was hypothesized that these dual effects resulted in a more rapid onset of antidepressant response [38–40]. In a previous pilot study of ABT-200 in patients with depression, no significant differences were found between ABT-200 140 mg/day and placebo for HAM-D total scores or Clinical Global Impression Scale (CGI) scores [35]. However, core depression item scores on the HAM-D were significantly reduced in patients receiving ABT-200, suggesting potential for therapeutic efficacy. As a MTD was not determined in the patient population prior to a preliminary Phase 2 study that indicated marginal efficacy, our bridging study was conducted in order to ensure that all safe and potentially therapeutic doses would be evaluated. Bridging study design The design was a double-blind, placebo-controlled, outpatient study conducted in 12 patients with depression [35]. Patients were required to meet DSM-III-R criteria for major depressive disorder with or without melancholia; have a total score of at least 20 on the 24-item HAM-D; and a score of at least 2 on item 1 of the HAM-D. Patients were also required to have a HAM-A total score less than their HAM-D total score, a score of at least 8 on the Raskin Depression Scale, and a Covi-anxiety Scale score less than their Raskin Depression Scale score. Exclusion criteria included clinically significant medical, neurological, or other psychiatric disorders; a risk of suicide or electroconvulsive therapy (ECT) within 6 months of screening; treatment with monoamine oxidase inhibitors (MAOIs) or fluoxetine within 4 weeks of study treatment; or benzodiazepines (used for daytime sedation) within 3 weeks of study treatment. The study was conducted in two panels. In the first panel, patients received placebo or were titrated over a 3-week period from an initial dose of 160 mg/day to a maximum maintenance dose of 240 mg/day ABT-200. In the second panel, patients received placebo or were titrated over a 3-week period from an initial dose of 160 mg/day to a maximum maintenance dose of 280 mg/day. Patients were to be maintained on the maximum maintenance dose for a total of 4 weeks. Bridging study results All doses evaluated in the study were tolerated. The most common AEs included lightheadedness, dizziness, headache, insomnia, drowsiness/ sedation, and nausea. A total of three patients prematurely discontinued from the study due to AEs: one due to lightheadedness, one due to decreased concentration and dizziness, and one due to nausea, dry mouth, and dizziness. Conclusions The bridging study extended the usable dose range by 100% over doses that were previously not found to be efficacious in patients with depression.

c06

BLBK249-Cutler

204

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

Bridging studies in schizophrenia Through our experience with bridging studies in schizophrenia, we have noted that compounds that require titration have a dose range that is less well defined, and may require a different study design. Many CNS compounds, antipsychotics in particular, require titration because of their AE profiles. For example, compounds such as clozapine and sertindole are ␣-adrenergic receptor agonists, resulting in a high propensity for hypotension. There is evidence that AEs such as hypotension can sometimes be ameliorated with titration. For example, in a study of mazapertine [41], tolerance to hypotension was induced with titration of the compound. Unfortunately, the need for titration can also slow the onset of antipsychotic effects if too much time is wasted before reaching a therapeutic dose. Thus, there is a need to determine not only an optimal dose range, but also a rapid titration schedule to reach these doses quickly and safely. We conducted bridging studies of several antipsychotic compounds. Our results demonstrated that differences in tolerability between patients and healthy subjects necessitate dose finding in patients prior to selecting a dose range for Phase 2 efficacy studies (see Table 6.8). We will review our study evaluating the safety and tolerability of two rapid dose titration schedules of the antipsychotic sertindole. Finally, we will discuss our bioequivalence study of clozapine and the safety reasons for conducting such trials in patients rather than normals. Details of a few selected studies follow:

Iloperidone Background Iloperidone is an atypical antipsychotic compound demonstrating affinity for D2 , D3 , D4 , 5-HT2A , 5-HT6 , and 5-HT7 receptors with antagonistic action at ␣1 adrenergic receptors [42]. In preclinical studies, iloperidone demonstrated efficacy in behavioral models predictive of antipsychotic activity against positive and negative symptoms [43, 44], with a reduced liability for extrapyramidal symptoms (EPS) [44]. However, the ␣1 adrenergic antagonist activity of iloperidone suggests a potential for orthostatic hypotension [45]. In a study in conscious dogs, iloperidone induced hypotension and peripheral vasodilation following doses of 1 and 10 mg/kg. In a design incorporating a 3-day pretreatment, however, only two of four dogs demonstrated hypotension and vasodilation [19], suggesting that iloperidone’s potential for orthostatic hypotension might be attenuated through dose titration. In a single-dose safety and tolerability study of iloperidone in healthy, fasted subjects (n = 18), dose-limiting AEs of orthostatic hypotension and dizziness were observed following iloperidone doses of 3 mg or greater [19]. In another study, fasted patients with schizophrenia (n = 15) tolerated iloperidone titrated from 1 to 8 mg/day over 22 days without any significant AEs [19].

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

205

Table 6.8 Key findings related to bridging studies in patients with schizophrenia Drug

Key findings related to bridging study

Sources

MDL 100,907

MDL 100,907 is a potent and selective 5-HT2A antagonist that has demonstrated activity consistent with antipsychotic efficacy in animal models. In radioligand studies, MDL 100,907 demonstrated a pharmacological profile consistent with a low liability for extrapyramidal symptoms (EPS), and potential efficacy against negative symptoms. The bridging study extended the usable dose range to 100 mg in divided doses, which was beyond the MTD in healthy volunteers (72 mg).

[19,89,90]

Iloperidone

The doses tolerated by patients in the bridging study were 10 times higher than single doses previously tolerated by healthy subjects and 4 times higher than the reported threshold dose for efficacy in a Phase 2 study of patients with schizophrenia.

[7,19,42–46]

Fananserin

Although this study did not find a true MTD, it did extend the usable dose range by almost 400% and identified a safe and well tolerated titration schedule for patients with schizophrenia. A subsequent Phase 2 study of this compound did not find significant efficacy; however, this bridging study provided confidence that the clinically relevant dose range was thoroughly tested.

[47]

CI-1007

Although no true MTD was determined, the results of this study indicate that patients with schizophrenia tolerate higher initial doses of CI-1007 than healthy subjects.

[48–51]

Sertindole

Sertindole has a well-defined therapeutic dose range, but requires titration up to an effective dose. A titration schedule with dose increases every 3 days was instituted for all sertindole studies. The ability to safely titrate sertindole every other day potentially allows a 24-mg/day dose to be reached in 11 days, versus 16 days with 3-day titration.

[7,91–93]

Bridging study design The design was a single-center, four-period (I to IV) bridging study to determine the safety and tolerability of single doses of iloperidone in patients with schizophrenia [7]. A total of 24 stable patients with schizophrenia (mean age 37 years) participated in this inpatient study. Patients were required to meet DSM-IV criteria for primary schizophrenia, including the presence of characteristic symptoms (excluding schizoaffective and mood disorder with psychotic features), with an absence of any organic factor accounting for the symptomatology. The patients had no clinically significant medical, neurological, or other psychiatric disorders that could potentially interfere with study evaluations. Patients were also required to be daily cigarette smokers, in order to reduce potential pharmacokinetic variance caused by nicotine-induced stimulation of the hepatic metabolism of iloperidone. Due to a report of

c06

BLBK249-Cutler

206

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

tinnitus in one patient in an earlier study, all patients were required to have a screening audiogram within normal limits for their age. Bridging study results Study Period I was designed to determine the slow (every 3 days) titration MTD of iloperidone in a single panel of patients with schizophrenia (n = 8). Following a 4-day single-blind placebo washout period (Days −4 to −1), patients received an initial dose of iloperidone 2 mg/day for 2 days followed by 3 days each at 4, 6, 8, 12, 16, 20, 24, 28, and 32 mg. On this schedule, oncedaily doses of iloperidone were well tolerated through the protocol maximum of 32 mg. One patient was discontinued from the study after experiencing a brief (1–2 seconds) syncopal episode approximately 2 hours after receiving the 2 mg initial dose of iloperidone. The syncopal episode was judged to be most likely a first-dose phenomenon, and we thus felt that continuation of the panel would not put other patients at risk. Due to this AE, however, 1 mg was defined as the initial dose for Study Periods III and IV. An additional patient in Study Period I was discontinued at 8 mg on Study Day 10 due to emerging psychosis. The most common AEs were mild rhinitis and insomnia; moderate AEs included fatigue and one episode of chest pain, which was thoroughly evaluated and judged by a physician to be noncardiac related. Because an MTD was not reached, a decision was made to limit the top dose to 24 mg for subsequent periods. Study Period II was originally designed to determine the fixed-dose MTD, which would serve as the maximum initial dose of iloperidone for subsequent study periods. However, as an initial dose (1 mg) was selected in Study Period I, Study Period II was canceled. In Study Period III, the tolerability of a more rapid (every 2 days) titration schedule was evaluated in a single panel (n = 8). Doses were titrated from an initial dose of 1 mg to 2, 4, 8, 12, 16, 20, and 24 mg, with dose increases every 2 days. The rapid titration of iloperidone was well tolerated for doses up to the maximum of 24 mg. One patient was discontinued at 16 mg on Day 12 due to emerging psychosis of moderate intensity. Other moderate AEs included agitation, headache, and insomnia. In Study Period IV, we evaluated the tolerability of a daily titration schedule in a single panel of patients (n = 8). Doses were identical to those in Study Period III, but dose escalation occurred every day. The daily titration of iloperidone was well tolerated for doses up to the maximum of 24 mg. Two patients discontinued at 24 mg on Days 9 and 11 due to emerging psychosis of moderate to severe intensity. Other moderate AEs included fatigue, asthenia, dizziness, and insomnia. The AEs observed in the bridging study were similar in type and frequency to those in a Phase 2 efficacy study (n = 104) of iloperidone in patients with acute or relapsing schizophrenia [46]. The four most common AEs (rhinitis, insomnia, agitation, and headache) were identical for the two studies. Although the percentage of patients experiencing each AE was lower in the Phase 2

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

207

Period I

20 10 0 –10 Mean change in systolic blood pressure upon standing (mm Hg)

c06

–20 –30 –40 20

1

4

8

12

16

20

24

28

Period III

10 0 –10 –20 –30 –40 –50 30

1 Period IV

4

8

12

16

15 0 –15 –30 1

2

4

6 Study day

8

10

Figure 6.2 Mean change from baseline for orthostatic systolic blood pressure in iloperidone-treated patients. (Reproduced by permission from Sramek et al. [7].)

study, this difference is most likely due to the lower doses employed (2 and 4 mg BID). Thus, bridging studies can often be helpful in predicting prospective AEs in subsequent larger studies. Although patients demonstrated reductions in supine to standing systolic and diastolic blood pressure in all study periods, this effect did not appear to be dose dependent, potentially due to development of tolerance with titration (see Figures 6.2 and 6.3). Except for the patient with a brief syncopal episode shortly after receiving the higher initial dose of 2 mg, there were no severe or serious AEs judged to be related to study medication.

BLBK249-Cutler

208

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

30

Period I

20 10 0 Mean change in diastolic blood pressure upon standing (mm Hg)

c06

–10 –20 –30 1 20

4 Period III

8

12

16

20

24

28

10 0 –10 –20 –30 –40 30

1

4

8

12

16

Period IV

15 0 -15 –30 1

2

4

6

8

10

Study day Figure 6.3 Mean change from baseline for orthostatic diastolic blood pressure in iloperidone-treated patients. (Reproduced by permission from Sramek et al. [7].)

A total of four patients were discontinued from the bridging study due to emerging psychosis. This study was conducted in stable patients with schizophrenia who were voluntarily taken off their previous antipsychotic treatment regimen, and who might not have had sufficient time to respond to the study drug before the end of the study. Thus, no conclusions could be drawn regarding the efficacy of iloperidone until further trials were conducted, particularly in patients presenting with acute psychotic symptoms.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

209

In a 1995 study of efficacy of iloperidone in patients with acute or relapsing schizophrenia, iloperidone 4 mg BID was found to be superior to placebo in demonstrating improvement in Positive and Negative Syndrome Scale (PANSS) total scores and CGI measures [46]. Conclusions This bridging study demonstrated that once-daily doses up to at least 32 mg were tolerated on a slow titration schedule, and doses up to at least 24 mg were well tolerated on a rapid titration schedule (4 days to reach the reported threshold efficacy dose). Although no true MTD was determined, the doses tolerated by patients in the bridging study were 10 times higher than single doses previously tolerated by healthy subjects and 4 times higher than the reported threshold dose for efficacy in a Phase 2 study of patients with schizophrenia [46].

Fananserin Background We also conducted a bridging study of fananserin, a then novel compound that had demonstrated potent activity at D4 and 5-HT2A receptors with a low affinity for D2 receptors. In preclinical studies, this compound demonstrated activity in several behavioral assays predictive of antipsychotic efficacy and did not cause catalepsy in mice at doses up to 160 mg/kg. In previous studies of healthy male volunteers, single doses were tolerated up to 160 mg/kg. At 300 mg, dose-limiting AEs of hypotension and bradycardia were observed in four healthy subjects, potentially due to the drug’s affinity for ␣1 adrenergic receptors. Bridging study design The design was a single-center, dose-rising, two-period bridging study in sequential panels of patients with schizophrenia [47]. Male and female patients met DSM-IV criteria for schizophrenia with the presence of characteristic symptoms (excluding schizoaffective and mood disorder with psychotic symptoms), with an absence of any organic factor accounting for the symptomatology. The patients had no clinically significant medical, neurological, or other psychiatric disorders that could potentially interfere with study evaluations. Due to the potential hypotensive effects of fananserin, all patients were required to have a supine and standing systolic blood pressure of ≥ 100 mm Hg, supine and standing diastolic blood pressure of ≥ 60 mm Hg, and a supine and standing pulse rate ≥ 50 beats per minute (bpm). All patients completed a 3-day single-blind placebo washout phase (Days −3 to −1) prior to receiving study medication. In the original study design, the objective of Study Period I was to determine the slow titration MTD of the compound in a single group of patients (n = 10), according to the schedule in Table 6.9.

c06

BLBK249-Cutler

210

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6 Table 6.9 Dosing schedule for Period I Day(s)

Total daily dose of fananserin

Administration

1 2–3 4–6 7–9 10–12 13–15 16–18

50 mg 100 mg 200 mg 300 mg 400 mg 500 mg 600 mg

50 mg QD 50 mg BID 100 mg BID 150 mg BID 200 mg BID 250 mg BID 300 mg BID

Reproduced by permission from Sramek et al. [47]. QD = every day; BID = twice daily

Bridging study results On this slow titration schedule (increases of 100 mg every 3 days), fananserin was well tolerated at doses up to 600 mg. Three patients were prematurely discontinued from the study: one patient withdrew at 300 mg due to an acute psychotic episode, another patient discontinued at 300 mg due to increased anxiety, and a third patient discontinued at 500 mg after experiencing a cardiac arrhythmia and a brief sinus pause of 1.44 seconds. The sinus pause was later judged to be not clinically significant by a cardiologist. The majority of other AEs reported in Period I were considered to be mild in intensity; moderate AEs included headache and anxiety. Although the 600-mg dose was well tolerated, no higher doses were tested due to findings that indicated 600 mg is just below the preclinical NOAEL. Thus, no true MTD was reached in Period I. Study Period II was designed to evaluate the tolerability of increasingly rapid titration schedules in four consecutive panels (n = 6 per group, see Table 6.10). Progression to each successive panel was only to be initiated if the previous panel was well tolerated. Panel 1 received an initial dose of 100 mg/day, followed by 3 days each at 200 and 400 mg. The final dose of 600 mg was maintained for 3 days. The compound was poorly tolerated on this titration schedule (increases of 200 mg every 3 days). One patient had a clinically

Table 6.10 Fananserin study Period II panel design Panel

Design

Panel 1 Panel 2 Panel 3 Panel 4

100 → 600 mg over 8 days 100 → 600 mg over 6 days 100 → 600 mg over 4 days 600 mg on Day 1 (no titration)

Reproduced by permission from Sramek et al., 1998b [47].

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

211

Table 6.11 An intermediate dose titration schedule Day(s)

Total daily dose of fananserin

Administration

1 2–3 4–5 6–7 8–9 10–16

100 mg 200 mg 300 mg 400 mg 500 mg 600 mg

50 mg BID 100 mg BID 150 mg BID 200 mg BID 250 mg BID 300 mg BID

Reproduced by permission from Sramek et al. [47].

significant elevation of liver enzymes at a dose of 400 mg and was withdrawn from the study. Two patients experienced moderate hypotension at 600 mg. One of these cases was accompanied by a brief syncopal episode, and the patient was discontinued from the study. Subsequently, all patients in Panel 1 were discontinued from the study for safety reasons. As Panel 1 did not tolerate the compound well, the remaining panels were canceled, and an additional panel was included (Panel 2A) at an intermediate titration schedule (n = 10), shown in Table 6.11. Patients tolerated doses up to 600 mg on this titration schedule. One patient was prematurely discontinued due to a moderate case of vertigo at 400 mg. Other moderate AEs included insomnia, headache, hypotension, hallucinations, increased salivation, and increased liver enzymes. The majority of AEs for Panel 2A were considered to be mild. No serious adverse events (SAEs) or extrapyramidal symptoms (EPS) were observed in any study period. Other than the elevation of liver enzymes in two patients, there were no clinically significant changes in laboratory measures. Conclusions This bridging study demonstrated that patients tolerated the compound at doses up to at least 600 mg on slow (increases of 100 mg every 3 days) and intermediate (increases of 100 mg every 2 days) titration schedules. Although this study did not find a true MTD, it did extend the usable dose range by almost 400% and identified a safe and well tolerated titration schedule for patients with schizophrenia. A subsequent Phase 2 study of this compound did not find significant efficacy; however, this bridging study provided confidence that the clinically relevant dose range was thoroughly tested.

CI-1007 Background CI-1007 is a potent, orally active dopamine autoreceptor agonist and partial dopamine D2 /D3 receptor agonist. In vivo, CI-1007 inhibits the firing of central dopaminergic neurons and decreases the synthesis and release of dopamine in the brain. CI-1007 has also demonstrated antipsychotic-like

c06

BLBK249-Cutler

212

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

activity with a reduced liability for EPS and tardive dyskinesia in preclinical models [48–50]. In a rising, single-dose tolerance study of CI-1007 in healthy subjects, oral doses of 5 mg were well tolerated. However, the incidence of AEs such as nausea, vomiting, and headache increased at 10 mg and was dose limiting at 15 mg. Orthostatic hypotension was also observed at 15 mg. Thus, 10 mg was designated the MTD in healthy subjects. Bridging study design The design was a single-blind, rising, multiple-dose, inpatient bridging study (n = 16) in four consecutive panels of four schizophrenia patients each [51]. Subjects were male and female patients with schizophrenia who met DSM-IV criteria and required treatment with antipsychotic medication. They had no clinically significant medical, neurological, or other psychiatric disease and no alcohol or drug abuse within 6 months prior to the study. Due to possible cardiovascular side effects with this autoreceptor agonist, patients were also required to have a standing (2-minute) systolic blood pressure > 100 mm Hg and heart rate < 100 bpm. Additionally, as the CYP2D6 locus is believed to be important for the drug metabolism of CI-1007, only patients with the normal/wild type CYP2D6 genotype (not poor CI-1007 metabolizers) were allowed to participate. Following a 4-day placebo washout period, patients were assigned to receive one of four doses of CI-1007 in Panel 1 (5 mg BID), Panel 2 (10 mg BID), Panel 3 (15 mg BID), or Panel 4 (20 mg BID). Patients in each panel received multiple doses of CI-1007 (administered every 12 hours) for 4 days, with a final dose on the morning of Day 5. The administration of higher doses of CI-1007 was contingent upon the tolerability of the lower doses. Bridging study results CI-1007 was generally well tolerated over the dose range evaluated. The most common AEs were nausea, vomiting, and headache, and were primarily mild in intensity. Adverse events were most commonly observed following the initial dose of CI-1007 and tended to decrease with repeated dosing. The frequency and intensity of AEs were greatest at the 20 mg BID dose level. Mild nausea (4 of 4 patients), moderate vomiting (3 of 4 patients), and transient symptomatic hypotension (systolic blood pressure 68–89 mm Hg; 2 of 4 patients) were observed on the first 3 days of treatment at 20 mg BID. These symptoms lessened with subsequent doses, and CI-1007 was generally well tolerated for the remainder of the panel. Modest reductions in blood pressure and increases in heart rate were observed in the majority of patients following the first few doses of CI-1007. One patient in Panel 3 experienced a brief (2–3 seconds), moderate syncopal episode following the initial 15-mg dose but had no further complications for the remainder of the panel. No severe or serious AEs or clinically significant

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

213

abnormalities on physical examinations, electrocardiogram (ECG), or laboratory tests were observed. No EPS was observed in any panel. Conclusions Although no true MTD was determined, the results of this study indicate that patients with schizophrenia tolerate higher initial doses of CI-1007 than healthy subjects.

Alternative titration strategies The regimens proposed above for rapid titration panels are excellent general schemes that can be employed to reach the top dose as quickly as possible. However, there are occasions where a tailored bridging study can be employed later in a compound’s development to minimize excessive AEs, if the compound has not benefited from a bridging study early in its development. In such cases, it is very helpful to fine-tune the titration by careful study of the time course of AEs in Phase 2 studies. For example, we examined the data from a number of Phase 2 trials of a 5-HT1A partial agonist for the treatment of depression in order to propose a titration rate that would allow for the development of tolerance to troublesome serotonergic AEs in clinical trials. By looking at the AEs over time and setting some parameters with which to characterize the data, one can get a handle on the time course needed for the development of tolerance. In these studies, the compound was administered on a fixed-dose regimen in doses of 0.25, 1, 2, 3, or 4 mg/day. The parameters we chose to characterize the data were simple: 1 How many days of treatment did it take before a specific AE diminished greatly in frequency (drop of more than 75%)? 2 What was the percentage mean incidence of these AEs after 3 days of treatment? Thus characterized, we found the results in Table 6.12. It is apparent that it will take up to several weeks to develop tolerance to a fixed dose of this compound, slightly longer for dizziness than for nausea. If higher doses are to Table 6.12 Mean incidence of adverse events (%) after 3 days of treatment Dose Adverse event Dizziness days to tolerance % after 3 days Nausea days to tolerance % after 3 days

0.25 mg

1 mg

2 mg

3 mg

4 mg

7 4

14 12

18 18

20 18

20 20

10 3

8 8

9 25

9 10

14 30

c06

BLBK249-Cutler

214

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

be used (i.e., 2–4 mg), it is also apparent that the incidence rate is related to the dose. One might then design a titration bridging study to go slowly up to the 2-mg dose in order to avoid these troublesome AEs that often cause patients to discontinue treatment. However, after attaining 2 mg, one might expect that the receptor dynamics would come into play and that tolerance would have developed such that further dose increases would be better tolerated. The compound should be titrated slowly from 0.25 mg/day to up to 2 mg over the first 2 weeks of treatment, and then when 2 mg is obtained, we would predict that the incidence of AEs will be much lower, probably comparable to the lowest dose range of the compound. By using titration, however, we might also expect that tolerance could be achieved in less than 2 weeks. Therefore, after confirming the tolerance of the proposed titration scheme in a panel of patients, a second panel would be conducted to test the tolerability of a faster titration rate. A faster rate might allow tolerance to develop in 1 week instead of 2, gaining valuable time in achieving a therapeutic dose in depressed patients and shortening the time to antidepressant response.

Bridging studies exploring pharmacokinetic/pharmacodynamic relationships The evaluation of pharmacokinetic/pharmacodynamic (PK/PD) relationships has increasingly been recognized as an important element of early drug development. Peck et al. [52] noted that a failure to define relationships among dose, concentration, and treatment response often leads to inappropriate doses and a lack of information on how to individualize dosing in Phase 3. Incorporation of PK/PD measures at a stage early enough to influence subsequent development, however, may assist in the identification of optimal dosing regimens and could contribute to an acceleration of drug development. Furthermore, PK/PD measures can lead to an increased understanding of a drug’s mechanism of action, as well as information that could be useful in drug labeling. The importance of PK/PD studies has also been recognized from the regulatory perspective. In this arena, PK/PD studies could provide flexibility in the regulatory review process [53]. For example, bioequivalence criteria might be relaxed based on a better understanding of PK/PD relationships and the intraindividual variability of those relationships. Optimizing dose regimens with PK/PD information could also reduce drug development costs by more efficiently fulfilling requirements for the administration of investigational compounds to patients at clinically effective doses. The bridging study provides a unique opportunity to explore PK/PD relationships in the target population for a larger range of doses than those that will most likely be employed at later stages. The incorporation of PK/PD measures in the bridging study could assist in more completely characterizing the acute pharmacologic effects of the compound and defining the relationship of these effects to both dose and the incidence of AEs. Additionally, PK/PD

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

215

models could be constructed for use in later phases to assist in the initiation and adjustment of dosing in individual patients. One potential challenge in establishing a PK/PD relationship is defining an appropriate pharmacodynamic endpoint. Potential pitfalls in selecting a pharmacodynamic endpoint include: drug assay problems or lack of accessible biological samples; lack of an immediate pharmacologic effect that can be related to pharmacokinetic measures; or lack of a relationship between a quantifiable pharmacologic effect and long-term clinical benefit [54]. In many CNS indications, the pharmacodynamic endpoint is a surrogate marker and not a therapeutic outcome measure. For example, a study of a putative antidepressant might assess serotonin metabolite levels, but not reveal relevant information about changes in the depression state of a patient taking this compound. The studies of compounds for AD also rely on surrogate pharmacodynamic endpoints, such as AChE inhibition. A PK/PD study employing surrogate endpoints can help to confirm a compound’s mechanism of action, and can provide useful preliminary concentration-response data.

The dynabridge study PK/PD relationships may differ between the central (the brain) and peripheral (the tissues of the body that are less well perfused, such as the muscle, skin, fat, and most of the viscera) compartments. Table 6.13 illustrates some differences between plasma and cerebrospinal fluid (CSF) accumulation for several clinically approved drugs. For CNS compounds, the action of the drug in the central compartment is of primary interest, although knowledge of peripheral PK/PD parameters may be helpful in understanding AEs. Thus, the assumption that a peripheral measure is a good indicator of central pharmacokinetics or activities should be tested early in development. For example, the utility of peripheral AChE inhibition as a measure of central cholinesterase inhibition cannot be taken for granted. The half-life of erythrocyte AChE inhibition after dosing with the irreversible cholinesterase inhibitor, metrifonate, appears to depend on the rate of synthesis of new red blood cells. There is no reason to expect that this will have any relationship with the rate of recovery of central activity. Differences in selectivity between central and peripheral

Table 6.13 Plasma and CSF concentrations of clinically applied drugs Drug

Administration

Plasma t1/2

CSF t1/2

Paracetamol Haloperidol [R] Ibuprofen [S] Ibuprofen Rivastigmine mGlu2/3 Atomoxetine

i.v. p.o. p.o. racemate p.o. racemate p.o. p.o.

2.4h 1–1.5d 1.7h 2.5h 1h ∼3h ∼5h

3.2h 6.8d 3.9h 7.9h
PD

>6h 60% AChE Inh

c06

BLBK249-Cutler

216

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

forms of an enzyme may also play a role, as in the case of SDZ ENA 713, which is specific for central AChE and inhibits peripheral AChE only minimally. On the other hand, verification that a peripheral marker does indeed correspond to central activity can greatly simplify the monitoring of patients. To assess central PK/PD relationships and their connection to peripheral PK/PD parameters, the “dynabridge study,” a PK/PD CSF study in the patient population, was developed. As the name suggests, dynabridge studies are an extension of the bridging concept involving early exploratory studies in the patient population. CSF sampling allows access to the central compartment and provides a better vehicle than plasma to assess drug concentration close to the site of action [55]. In a dynabridge study, the focus moves beyond dose finding and safety/tolerance to pharmacodynamic indicators of drug activity. A distinguishing feature of these studies is continuous CSF sampling, which allows the determination of central PK/PD time courses. CSF is analyzed for drug and metabolite concentrations, as well as for measures of drug activity (enzyme activities, neurotransmitter levels, and other biomarkers). Peripheral PK/PD parameters are monitored simultaneously. Potential correlations with psychiatric or neurologic rating scales and/or neuropsychological tests are also investigated. As a result of a dynabridge study, the time course of drug activity across the potential dosing range is determined, permitting optimal design of dosing prior to efficacy trials. In addition, the central activity of the compound is confirmed and potential surrogate markers are explored.

Practical issues in continuous CSF sampling One critical consideration in designing CSF studies is when to sample the CSF. The quickest way to determine CSF penetration would be following a single dose of study compound, as one can get a predose baseline and then up to a 24-hour profile which one can also compare to the plasma profile. However, many compounds take longer to equilibrate in CSF than they do in plasma, as measured by the half-life (see Table 6.13). The lag time often required to reach SS in CSF means that initial concentrations, such as after a single dose, will be rather low. Furthermore, in order to measure any meaningful effect of a compound on biomarkers in the CSF, the compound must at least reach SS in order to produce the full effect on those biomarkers. Thus, one would ideally compare the baseline biomarker measures with those achieved at SS conditions in a dynabridge study. In the case of psychoactive drugs, the treatment time to reach SS is easier to determine if drugs have been used clinically. For example, atomoxetine dosed at 80 mg/day was found to reach SS in 18 days [56], while duloxetine dosed at 80 mg/day reached SS in 2 weeks [57], and maprotiline at 75 mg/day reached SS in 4 weeks [58]. Typically, a 24-hour collection at baseline and end of treatment will be desirable in order to compare biomarker changes adjusted for diurnal variation. Additionally, one is able to measure the real half-life of the compound in the CSF, which can be immeasurably helpful for future development. In the case

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

217

of atomoxetine, for example, the drug was found to have an approximately threefold greater half-life in CSF compared to plasma, and with this data a once-daily dosing regimen was approved, which would never have been predicted by the shorter plasma half-life [59]. As a practical consideration, the same subject or patient should have the second CSF collection no earlier than 7–10 days after the baseline collection in order to allow for adequate healing time for puncture of the CSF sac. The risks associated with continuous CSF sampling are similar to those associated with routine lumbar puncture. In skilled hands, the insertion of the lumbar subarachnoid catheter by a needle is associated with little or no pain (the patient may experience a fleeting pain similar to that experienced during an ordinary injection or needle prick). Although the catheter can be inserted at bedside, the physician may choose to use fluoroscopy in aiding proper catheter placement, especially after failing catheter placement at bedside. The most common complaint associated with the sampling procedure is headache, and in most cases, this condition can be ameliorated with bed rest and hydration. Patients should be encouraged to drink plenty of fluids during the recovery period (after catheter withdrawal) to reduce the possibility of headache. Over-the-counter analgesics may also be administered, if needed. Most headaches last from several hours to 2 days; in extremely rare cases, headaches persisting for a full week have been reported. If persistent or severe headache occurs, a “blood patch” can be applied. This procedure involves injecting a small amount of the patient’s blood into the region of the supposed CSF leakage in order to seal it. This procedure has been effective in reducing the headache in 95% of cases [2]. Other rare complications are temporary backache, nerve root damage, and epidural or subdural bleeding. Significant bleeding is extremely unlikely and, if it occurs, is usually self-limiting. Infection is extremely uncommon when the procedure is performed using aseptic techniques. During continuous sampling, the CSF should be examined periodically to detect possible infection or bleeding, and the patient’s oral body temperature should also be monitored. If an infection is suspected, the lumbar catheter should be removed immediately and cultured; antibiotics should be administered, if required. Prophylactic antibiotics are not necessary. Adverse events associated with CSF collection procedures for PK/PD studies have been extensively reviewed for continuous CSF sampling by catheter [60] and single spinal taps [61]. Although the character, intensity, and incidence are dependent on the details of the procedure as well as the demographic characteristics of the subjects, both procedures are well tolerated with no SAEs or study withdrawal due to the collection process. The confinement to bed rest and the use of a bedside urinal or bedpan is the most common complaint reported by patients. To alleviate aches and pains, patients can be allowed to use a pillow and to turn from side to side. At the discretion of the physician, patients may also be permitted to ambulate for short periods of time.

c06

BLBK249-Cutler

218

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

Patients should be monitored for a minimum of 24 hours after catheter withdrawal. Prior to discharge, a physical examination, ECG, vital signs, laboratory tests, and AEs should be evaluated. These evaluations should be repeated approximately 1 hour later. In addition, the physician should closely examine the lumbar puncture site for any signs of inflammation or infection. The protocol should specify the exact amount of CSF to be withdrawn in a given time period. When sampling is done on different days, lumbar punctures assessing CSF protein levels should be performed at the same time of day, preferably under similar conditions, such as early morning fasting [62] to account for possible fluctuations in the protein being measured.

Dynabridge study of atomoxetine Background R ) is a nonstimulant drug approved for the treatAtomoxetine (Strattera ment of attention-deficit hyperactivity disorder (ADHD). In a study using a biomarker of norepinephrine transporter (NET) inhibition, daily doses of atomoxetine were used to quantify the NET inhibition during SS in central and peripheral body compartments in healthy subjects [63]. Dynabridge study design The study was open-label and conducted with 12 healthy subjects (8 males, 4 females) for a duration of 18 days. The PK/PD endpoints were CSF, plasma and urinary NE, and dihydroxyphenylglycol (DHPG). Dynabridge study results The concentration of CSF DHPG was found to be a potentially sensitive biomarker of NET inhibition. Twofold higher CSF concentrations of DHPG suggested that the drug effect originates in the brain. The data suggested that once-daily atomoxetine was associated with central effects persisting for at least 24 hours. Also, DHPG changes in plasma and urine mirrored the central effects in magnitude. Conclusions The results illustrated the importance of integrating biomarker data across relevant biological matrices for clinically used medications and for investigational drugs. The atomoxetine study demonstrated the pharmacodynamic endpoint activity, which was more prolonged in the CSF than the drug concentrations, showing that dosing could be once-daily based on these data.

Dynabridge study of rivastigmine A previous bridging study in AD patients indicated that doses up to 12 mg/day were safe and well tolerated [20], and preliminary evidence suggested that rivastigmine had a beneficial cognitive effect at doses of 6–12 mg/day [21]. A dynabridge study was conducted to evaluate the central

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

219

activity of rivastigmine in patients with AD, and its relationship to central and peripheral pharmacokinetic parameters [64]. Background Rivastigmine is an AChE inhibitor of the carbamate type. Rivastigmine has a longer duration of action than physostigmine and high central selectivity, which could result in a lower liability for side effects common to AChE inhibitors, including peripheral cholinergic side effects and liver toxicity [18,65]. Dynabridge study design The design was an open-label, single-center, multiple-dose study in six sequential groups of three patients to evaluate the effects of rivastigmine with AChE and butylcholinesterase (BChE) activity in CSF, and to correlate these parameters with pharmacokinetic measures and cognitive performance on the Computerized Neuropsychological Test Battery (CNTB). A total of 18 patients meeting NINCDS-ADRDA criteria for probable AD participated. Patients were admitted to the hospital for baseline assessments and received an initial dose of rivastigmine 1 mg BID. Patients were then titrated to target doses of 1, 2, 3, 4, 5, or 6 mg BID/week increments on an outpatient basis. After patients had tolerated their target dose for at least 3 days, they were readmitted to the hospital for a final dose of medication (at least 12 hours after the previous dose). A total of 7 mL of blood was collected for the determination of plasma concentrations of rivastigmine and its metabolite, ZNS 114-666, at 30 minutes prior to and 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10, 12, and 24 hours after administration of the final dose. In addition, the continuous CSF sampling procedure was performed during the 30 minutes prior to and for 12 hours after the final dose. Dynabridge study results Pharmacokinetic results indicated that rivastigmine was rapidly absorbed and quickly eliminated following the administration of a single oral dose, with a mean half-life ranging from 1.1 to 1.6 hours. Plasma concentrations of both rivastigmine and ZNS 114-666 tended to increase with increasing dose; significant correlations were observed between dose and AUC0–12 for plasma rivastigmine (p < 0.0001) and ZNS 114-666 (p < 0.0001). In addition, rivastigmine was quickly detected in CSF following a single oral dose, with tmax of 1.4–3.8 hours. CSF concentrations of both rivastigmine and ZNS 114-666 tended to increase with increasing dose. Rivastigmine demonstrated significant inhibition of CSF AChE activity (p < 0.05) for all dose levels, except 1 mg BID, and appeared to be dose-dependent over the 1–6-mg dose range. Overall, rivastigmine was moderately well tolerated in this study. The most common AEs were headache, nausea, dizziness, diarrhea, fever, and vomiting. Most of these AEs were mild in intensity and considered to be related to

c06

BLBK249-Cutler

220

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

the lumbar catheterization procedure. There were no severe or serious AEs or clinically significant changes in laboratory tests, ECG, physical examinations, or vital signs measurements. Conclusions This dynabridge study demonstrated the value of evaluating pharmacokinetic parameters in both the central and peripheral compartments, as the half-life appeared to be longer in CSF than in plasma [2]. In addition, the assessment of PK/PD measures provided information relevant to the determination of adequate dosing intervals; the time course of CSF AChE inhibition in this study justified BID dosing with rivastigmine.

Special considerations in the bridging study The studies reviewed in this chapter illustrate the importance of bridging in optimizing dose selection for Phase 2/3. However, certain aspects of the bridging study warrant further discussion. Because bridging studies explore the upper limits of a compound’s tolerable dose range, patients are required to be in good physical health without concomitant medical conditions or medications. Patients are also under close supervision during the bridging study, which minimizes issues of compliance. Thus, we recommend that later efficacy trials be conducted using initial doses somewhat lower than the MTD, bearing in mind that the population in large outpatient studies will often display a greater interpatient variability and will be less closely monitored. Although the small panel sizes in a bridging study generally preclude formal statistical analysis, a bridging study does have the statistical power to reveal common, acute AEs. If the probability of a given dose being toxic is 0.5, then the inclusion of at least 4 patients on active medication per panel allows us to be 94% confident that the dose is nontoxic [66]. This confidence level is increased to 98% when 6 patients per panel on active compound are included. However, as the bridging study attempts to identify doses that are subtoxic, the inclusion of a greater number of patients per panel is desirable. Thus, the data generated in a bridging study is sufficient to provide reliable information about the overall AE dose–response curve and to produce descriptive statistics. While long-term toxicity is not addressed by bridging studies, most acute effects should be accurately predicted. As illustrated by xanomeline, bridging studies have been shown to be highly predictive of the acute AE profiles found in Phase 2, and thus are valuable in optimizing the safety of these trials. The bridging and dynabridge studies can be combined in a single protocol to efficiently evaluate optimal dose parameters, use biomarkers to measure drug activity, and establish a PK/PD profile. An example of such a protocol design is shown in Figure 6.4.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

Panel 5 (8/2)*

221

MID

MTD

Panel 4 (6/2)*

30 mg

Panel 3 (6/2)* 20 mg

MTD Confirmation

Panel 2 (6/2)* 15 mg

Panel 1 (6/2)* 10 mg

Baseline and end 12–24 hour CSF collections

*Active/Placebo

Bridging–Determine MTD

Dynabridge – Use CSF PD markers at MTD

Figure 6.4 Combined bridging and dynabridge study design.

Positron emission tomography Often, the role of brain imaging comes up in drug development and how it can be most effectively used. Positron emission tomography (PET) is a powerful tool that can help in the understanding of a compound’s mechanism of action. There are, however, significant drawbacks to using brain imaging as a routine tool in CNS drug development. An appropriate ligand must be available to image the desired receptor or endpoint, and such a ligand is often not available for most novel compounds in development. Also, many compounds will have activity at multiple receptors, and it is not feasible to image all receptors in a single study. At best, a tracer such as labeled C-18 fluorine-deoxyglucose, which is taken up by brain cells, can show overall metabolic activity in the brain. Such studies, however, are unhelpful in answering the pertinent questions needed in early development. Labeled raclopride has often been used as an important ligand to image D2 receptor activity in the brain, and to measure occupancy levels of binding at this receptor, but such studies were not helpful when raclopride itself was taken into clinical development and failed to demonstrate antipsychotic response. Another drawback with imaging is that it gives only a part of what is happening in the brain at a specific time, versus a more continuous assessment of CNS biomarkers and brain concentrations of a compound that are available in a dynabridge study. There have been attempts to conduct imaging studies that can give a more detailed profile, but

c06

BLBK249-Cutler

222

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

this requires imaging different groups of patients at different doses and times after dosing. Such an approach is extremely costly and only approximates the information that can be obtained from a dynabridge study combined with central biomarkers.

Adaptive design Enabling trials to be adapted based on knowledge about gene and protein markers or patient characteristics can help predict whether patients will respond well to a new medicine. These new approaches to clinical trials can result in trial designs that tell us more about the safety and benefits of drugs in potentially shorter time frames, and with fewer subjects/patients exposed to experimental treatments. This can result in clinical trials that may not only be more efficient but are also more attractive to patients and their physicians for enrollment [67]. An adaptive design allows modifications made to the trial and/or the statistical procedures of ongoing clinical trials [68]. Across 13 hypothetical dose–response scenarios considered, it was shown that the capability of adaptive designs to accurately model the true dose response resulted in performance up to 180% more effective than the best fixed-dose designs [69]. Adaptive sampling designs are ones where the accruing data (i.e., the observations) are used to adjust the experiment as it is being run. In a typical experiment, decisions such as how to sample during an experiment are made and fixed in advance. For example, in a standard clinical trial comparing two different treatments, patients are assigned to the two treatments with half being assigned to each therapy. Then, at the end of the experiment, a decision is made as to which treatment is more effective. In contrast, in an adaptive clinical trial, patient outcomes can be used as they become available to adjust the assignment of future patients, assigning more of them to the better treatment. Thus, adaptive procedures can offer significant ethical and cost advantages over standard fixed procedures. Ivanova and Murphy [70] reported a single-dose, sequential cohort, safety, tolerance, and pharmacokinetic investigation of NGX267 in normal male volunteers. The primary objective of the trial for NGX267 was to estimate the MTD and to gather detailed clinical and pharmacokinetic observations near the MTD. The study used an adaptive design to identify a single-dose MTD based on a clinical-statistical decision process. The design was based on a need to gather detailed clinical and pharmacokinetic observations with greater precision near the MTD. The MTD was defined based on the weighted average of moderate and severe AEs. An adaptive design was employed to concentrate dosage assignments at or near the MTD. Favoring the acquisition of data near the MTD, at the expense of information at lower dosage levels, resulted in a shorter trial and no loss of the type of information required to inform subsequent studies where larger normal volunteer or patient samples are evaluated. In clinical trials, people may be wary of adaptive designs because of the uncertainty in a controlled clinical trial using an adaptive design versus allowing

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

223

the physicians to decide each patient’s treatment based on professional experience. The latter can introduce many biases into the learning process, especially because physicians tend to use what they already know about the disease and the patients to treat them rather than test unknown or uncertain treatments or dosages. One of the important aspects of adaptive trials is that investigators can decide which option is best, and assign relatively more patients to it, much faster than a fixed trial.

Implications for disease-modification therapy A disease-modifying therapy is one that has an impact on the underlying pathology of the disease, and thus slows the rate of a patient’s decline over the course of long-term treatment. In AD, for example, the currently available AD medicines are believed to treat the symptoms of AD without impacting the underlying disease process or providing long-lasting benefit. If more effective therapies are not developed that either prevent AD or block progression of the disease in its very early stages, the economic and societal cost of caring for AD patients will be devastating [71]. Assessing surrogate endpoints or biomarkers early in development may prove to be particularly important for the growing area of disease-modifying agents. Such agents will certainly require extended clinical trials to demonstrate that they can halt progression of disease, and yet such time and costs would be hard to justify without some initial evidence that they can alter a primary process strongly linked to pathogenesis of the disease. ␥ -Secretase inhibitor LY450139 dihydrate, a ␥ -secretase inhibitor, was studied in a randomized, controlled trial of 70 patients with AD [72]. Subjects were given 30 mg for 1 week followed by 40 mg for 5 weeks. Treatment was well tolerated. Amyloidbeta (A␤)(1-40) in plasma decreased by 38.2%; in CSF, A␤(1-40) decreased by 4.42 +/− 9.55% (p = not significant). Thus, higher drug doses may result in additional decreases in plasma A␤ concentrations and a measurable decrease in CSF A␤, providing modification of the disease.

Conclusions As shown in Table 6.14, there are several stages involved in the development of a new compound. In our opinion, the studies listed in Table 6.14 are the core or critical studies that need to be conducted in order to make critical go/no-go decisions. Initial clinical development might proceed with placebocontrolled single dose escalation studies in healthy subjects, with a dose range based on preclinical data such as the NOAEL. Escalation to a range that allows adequate estimation of the compound’s pharmacokinetics (particularly dose proportionality) and a definite AE profile is desirable at this stage. Following these studies, several multiple-dose panels in healthy subjects should be conducted in order to construct steady-state pharmacokinetic profiles, and

c06

BLBK249-Cutler

224

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

Table 6.14 Types of studies conducted in the development of a new compound Study type

Population

Goal/outcome

Single-dose escalation Multiple-dose escalation CSF study Food interaction Bridging Dynabridge Dose-ranging efficacy

Healthy volunteers Healthy volunteers Healthy volunteers Healthy volunteers Patients Patients Patients

Safety/tolerance/pharmacokinetics Safety/tolerance/pharmacokinetics Central vs peripheral PK/PD Effect of food on pharmacokinetics Safety/tolerance; MTD Central PK/PD Efficacy and long-term safety

to determine the linearity of the kinetics and additional safety data. A food interaction study should also be conducted at this time. These studies all lay the groundwork for the bridging and dynabridge studies, which will evaluate optimal dose parameters and schedules for studies in Phase 2/3. In this way, bridging studies are central to the transition from studies in healthy subjects to dose-ranging and efficacy studies in patients. Of course, additional efficacy studies, some including active control arms, and clinical pharmacology studies (drug interactions, special populations, or medical conditions) will be conducted later in development if efficacy is supported by earlier stages. The examples reviewed in this chapter illustrate the importance of incorporating the target population into early drug development in order to optimize dose selection for Phase 2/3. By providing information on differences in tolerance between healthy subjects and the target population, bridging and dynabridge studies constitute a vital contribution to the creation of an accelerated, rational drug development program. In CNS development today, many novel compounds are designed to affect a specific molecular target (such as A␤ synthesis), and this has naturally increased the emphasis or reliance on biomarker studies early in development. While such reliance is understandable and has a sound scientific basis, we feel that they cannot be relied upon completely for making all the critical decisions that are necessary for completing Phase 1. At this point, they can often provide important confirmatory evidence to support continuing development, especially in situations where there is a testable hypothesis. In many CNS mechanisms, however, there is rarely one system involved, and there are often downstream effects on other systems when a change in one system is made. Until proven, biomarkers should be used to supplement rather than replace our efforts to understand the clinical pharmacology of a compound as early as possible. Efforts such as conducting bridging studies that optimize the dose range for efficacy trials should not yet be supplanted in the push to decreasing the time and cost of development. We have attempted in this book to describe the critical foundations upon which a successful development program can be built.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

225

References 1. Cutler NR, Sramek JJ, Kilborn JR. (1996) The bridging concept: optimizing dose for phase II/III in Alzheimer’s disease. Neurodegeneration. 5(4):511–4. 2. Cutler NR, Sramek JJ. (1998) Guidelines for conducting bridging studies in Alzheimer disease. Alzheimer Dis Assoc Disord. 12(2):88–92. 3. US Department of Health and Human Services, Food and Drug Administration, Center for Biologics Evaluation and Research (CBER), Center for Drug Evaluation and Research (CDER). (2004) Guidance for Industry and Clinical Investigators: The Use of Clinical Holds Following Clinical Investigator Misconduct. [document on the Internet]. Available from: http://www.fda.gov/OHRMS/DOCKETS/98fr/02d-0320gdl0002.pdf. Cited April 17, 2009. 4. Cutler NR, Sramek JJ. (1995) The target population in phase I clinical trials of cholinergic compounds in Alzheimer disease: the role of the “bridging study.” Alzheimer Dis Assoc Disord. 9(3):139–45. 5. Cutler NR, Sramek JJ. (1995) Scientific and ethical concerns in clinical trials in Alzheimer’s patients: the bridging study. Eur J Clin Pharmacol. 48(6):421–8. 6. Weissman L. (1981) Multiple-dose phase I trials–normal volunteers or patients? One viewpoint. J Clin Pharmacol. 21(10):385–7. 7. Sramek JJ, Cutler NR, Kurtz NM, Murphy MF, Carta A. (1997) Optimizing the Development of Antipsychotic Drugs. Wiley, Chichester, UK. 8. Okuma T. (1981) Differential sensitivity to the effects of psychotropic drugs: psychotics vs normals; Asian vs Western populations. Folia Psychiatr Neurol Jpn. 35(1):79–87. 9. Miller AL, Maas JW, Contreras S, et al. (1993) Acute effects of neuroleptics on unmedicated schizophrenic patients and controls. Biol Psychiatry. 34(3):178–87. 10. Simpson GM. (2008) Personal communication. 11. Pokorny R, Finkel MJ, Robinson WT. (1994) Normal volunteers should not be used for bioavailability or bioequivalence studies of clozapine. Pharm Res. 11(8):1221. 12. Grof P AM, Campbell M. (1993) Clinical Evaluation of Psychotropic Drugs for Psychiatric Disorders: Principle and Proposed Guidelines. Hogrefe & Huber, Seattle. 13. Sramek JJ, Fresquet A, Marion-Landais G, et al. (1996) Establishing the maximum tolerated dose of lesopitron in patients with generalized anxiety disorder: a bridging study. J Clin Psychopharmacol. 16(6):454–8. 14. Cutler NR, Sramek JJ, Veroff AE. (1994) Azheimer’s Disease: Optimizing Drug Development Strategies. Wiley, Chichester, UK. 15. Cutler NR, Sramek JJ, Murphy MF. (1995) The use of bridging studies in drug development for Alzheimer’s disease therapeutics. In: Cutler NR, Gottfries CG, Siegfried K, editors. Alzheimer’s Disease: Clinical and Treatment Perspectives. Wiley, Chichester, UK. 16. Medina A, Bodick N, Goldberger AL, et al. (1997) Effects of central muscarinic-1 receptor stimulation on blood pressure regulation. Hypertension. 29(3):828–34. 17. Sramek JJ, Cutler NR. (1995) The use of placebo controls. N Engl J Med. 332(1):62. 18. Enz A, Amstutz R, Boddeke H, et al. (1993) Brain selective inhibition of acetylcholinesterase: a novel approach to therapy for Alzheimer’s disease. Prog Brain Res. 98:431–8. 19. Cutler NR, Sramek JJ, Kurtz NM, Murphy MF, Carta A. (1998) Accelerating CNS Drug Development. Wiley, Chichester, UK. 20. Sramek JJ, Anand R, Wardle TS, et al. (1996) Safety/tolerability trial of SDZ ENA 713 in patients with probable Alzheimer’s disease. Life Sci. 58(15):1201–7.

c06

BLBK249-Cutler

226

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

21. Anand R, Hartman R. Strategies for the optimal development of an anti-dementia drug. Fifth International Conference on Alzheimer’s Disease and Related Disorders; July 24–29, 1996; Osaka, Japan. 22. Anand R, Gharbawi G, Enz A. (1998) Efficacy and safety results of the early phase stud R ies with Exelon (ENA-713) in Alzheimer’s disease: an overview. J Drug Dev Clin Pract. 8:109–16. 23. Shannon HE, Bymaster FP, Calligaro DO, et al. (1994) Xanomeline: a novel muscarinic receptor agonist with functional selectivity for M1 receptors. J Pharmacol Exp Ther. 269(1):271–81. 24. Farde L, Sukara T, Halldin C. Radioglands for PET-examionayion of central muscarinic receptors. Alzheimer’s and Parkinson’s Diseases Recent Developments, Third International Conference; November 1–6, 1993; Chicago, IL. 25. Lucas RA, Heaton J, Carter GV, Satterwhitw JH. Single and multiple-dose safety, pharmacodynamics, and pharmacokinetics of xanomeline, a novel muscarinic M1 agonist, in healthy male subjects. The Alzheimer’s and Parkinson’s Diseases Recent Developments, Third International Conference; November 1993; Chicago, IL. 26. Sramek JJ, Hurley DJ, Wardle TS, et al. (1995) The safety and tolerance of xanomeline tartrate in patients with Alzheimer’s disease. J Clin Pharmacol. 35(8):800–806. 27. Bodick NC, Offen WW, Levey AI, et al. (1997) Effects of xanomeline, a selective muscarinic receptor agonist, on cognitive function and behavioral symptoms in Alzheimer disease. Arch Neurol. 54(4):465–73. 28. Haj-Dahmane S, Jolas T, Laporte AM, et al. (1994) Interactions of lesopitron (E-4424) with central 5-HT1A receptors: in vitro and in vivo studies in the rat. Eur J Pharmacol. 255(1–3):185–96. 29. Costall B, Domeney AM, Farre AJ, et al. (1992) Profile of action of a novel 5hydroxytryptamine1A receptor ligand E-4424 to inhibit aversive behavior in the mouse, rat and marmoset. J Pharmacol Exp Ther. 262(1):90–98. 30. Barnes NM, Cheng CH, Costall B, Ge J, Kelly ME, Naylor RJ. (1992) Profiles of interaction of R(+)/S(−)-zacopride and anxiolytic agents in a mouse model. Eur J Pharmacol. 218(1):91–100. 31. Farre AJ. (1992) CNS profile of lesopitron, a new 5-HT1A receptor ligand. Behav Pharmacol. 3(Suppl 1):23. 32. Fresquet A, Sust M, Lloret A, et al. (2000) Efficacy and safety of lesopitron in outpatients with generalized anxiety disorder. Ann Pharmacother. 34(2):147–53. R Package Insert [document on the Inter33. Bristol-Myers Squibb Company. (2007) BuSpar net]. Available from: http://packageinserts.bms.com/pi/pi buspar.pdf. Cited February 24, 2009. 34. Sramek JJ, Frackiewicz EJ, Cutler NR. (1997) Efficacy and safety of two dosing regimens of buspirone in the treatment of outpatients with persistent anxiety. Clin Ther. 19(3):498–506. 35. Sramek JJ, Jasinsky O, Chang S, et al. (1994) Pilot efficacy trial of ABT-200 in patients with major depressive disorder. Depression. 2:31–318. 36. Zelle RE, Hancock AA, Buckner SA, et al. (1994) Synthesis and pharmacological characterization of ABT-200: a putative novel antidepressant combining potent ␣-2 antagonism with moderate NE uptake inhibition. Bioorg Med Chem Lett. 4:1319–22. 37. Hancock AA, Buckner SA, Giardina WJ, et al. (1995) Preclinical pharmacological actions of (+/−)-(1 R* ,3R* )-3-phenyl-1- [1 ,2 ,3 ,4 ‘-tetrahydro-5 ,6 -methylene-dioxy-1 naphthalenyl) methyl] pyrrolidine methanesulfonate (ABT-200), a potential antidepressant agent that antagonizes alpha-2 adrenergic receptors and inhibits the neuronal uptake of norepinephrine. J Pharmacol Exp Ther. 272(3):1160–1169.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

227

38. Johnson RW, Reisine T, Spotnitz S, et al. (1980) Effects of desipramine and yohimbine on alpha 2- and beta-adrenoreceptor sensitivity. Eur J Pharmacol. 67(1):123–7. 39. Scott JA, Crews FT. (1983) Rapid decrease in rat brain beta adrenergic receptor binding during combined antidepressant alpha-2 antagonist treatment. J Pharmacol Exp Ther. 224(3):640–646. 40. Goldstein JM, Knobloch-Litwin LC, Malick JB. (1985) Behavioral evidence for betaadrenoceptor subsensitivity after subacute antidepressant/alpha 2-adrenoceptor antagonist treatment. Naunyn Schmiedebergs Arch Pharmacol. 329(4):355–8. 41. Kleinbloesem CH, Jaquet-Muller F, al-Hamdan Y, et al. (1996) Incremental dosage of the new antipsychotic mazapertine induces tolerance to cardiovascular and cognitive effects in healthy men. Clin Pharmacol Ther. 59(6):675–85. 42. Kongsamut S, Roehr JE, Cai J, et al. (1996) Iloperidone binding to human and rat dopamine and 5-HT receptors. Eur J Pharmacol. 317(2–3):417–23. 43. Corbett R, Hartman H, Kerman LL, et al. (1993) Effects of atypical antipsychotic agents on social behavior in rodents. Pharmacol Biochem Behav. 45(1):9–17. 44. Strupczewski JT, Bordeau KJ, Chiang Y, et al. (1995) 3-[[(Aryloxy)alkyl]piperidinyl]1,2-benzisoxazoles as D2/5-HT2 antagonists with potential atypical antipsychotic activity: antipsychotic profile of iloperidone (HP 873). J Med Chem. 38(7):1119– 31. 45. Schwartz JT, Brotman AW. (1992) A clinical guide to antipsychotic drugs. Drugs. 44(6):981–92. 46. Borison RL, Huff FJ, Varaklis J, Ramaswamy R, Shipley JE. Efficacy of 2 and 4 mg bid iloperidone (HP 873) administered to schizophrenic patients for 42 days. 34th Meeting of the American College of Neuropsychopharmacology; December 11–15, 1995; San Juan, Puerto Rico. 47. Sramek JJ, Kirkesseli S, Paccaly-Moulin A, et al. (1998) A bridging study of fananserin in schizophrenic patients. Psychopharmacol Bull. 34(4):811–8. 48. Rugsley TA, Davis MD, Akunne HC, et al. (1995) CI-1007, a dopamine partial agonist and potential antipsychotic agent. I. Neurochemical effects. J Pharmacol Exp Ther. 274(2):898–911. 49. Feng MR, Corbin AE, Wang Y, et al. (1997) Pharmacokinetics and pharmacodynamics of an investigational antipsychotic agent, CI-1007, in rats and monkeys. Pharm Res. 14(3):329–36. 50. Meltzer LT, Christoffersen CL, Corbin AE, et al. (1995) CI-1007, a dopamine partial agonist and potential antipsychotic agent. II. Neurophysiological and behavioral effects. J Pharmacol Exp Ther. 274(2):912–20. 51. Sramek JJ, Eldon MA, Posvar E, et al. (1998) Initial safety, tolerability pharmacodynamics, and pharmacokinetics of CI-1007 in patients with schizophrenia. Psychopharmacol Bull. 34(1):93–9. 52. Peck CC, Barr WH, Benet LZ, et al. (1994) Opportunities for integration of pharmacokinetics, pharmacodynamics, and toxicokinetics in rational drug development. J Clin Pharmacol. 34(2):111–9. 53. Lesko LJ, Williams RL. (1994) Regulatory perspective: the role of pharmacokinetics and pharmacodynamics. In: Cutler NR, Sramek JJ, Natang PK, editors. Pharmacodynamics and Drug Development: Perspectives in Clinical Pharmacology. Wiley, Chichester, UK, pp 115–30. 54. Williams RL. (1992) Dosage regimen design: pharmacodynamic considerations. J Clin Pharmacol. 32(7):597–602. 55. Bieck PR, Potter WZ. (2005) Biomarkers in psychotropic drug development: integration of data across multiple domains. Annu Rev Pharmacol Toxicol. 45:227–46.

c06

BLBK249-Cutler

228

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Chapter 6

56. Biek et al. In preparation. 57. Vincent S, Bieck PR, Garland EM, et al. (2004) Clinical assessment of norepinephrine transporter blockade through biochemical and pharmacological profiles. Circulation. 109(25):3202–7. 58. Tsuji M, Yamane H, Yamada N, et al. (1986) Studies on 3-methoxy-4hydroxyphenylglycol (MHPG) and 3,4-dihydroxyphenylglycol (DHPG) levels in human urine, plasma and cerebrospinal fluids, and their significance in studies of depression. Jpn J Psychiatry Neurol. 40(1):47–56. 59. Sauer JM, Ring BJ, Witcher JW. (2005) Clinical pharmacokinetics of atomoxetine. Clin Pharmacokinet. 44(6):571–90. 60. Jhee SS, Zarotsky V. (2003) Safety and tolerability of serial cerebrospinal fluid (CSF) collections during pharmacokinetic/pharmacodynamic studies: five years experience. Clin Res Regul Aff. 20(3):357–63. 61. Peskind ER, Riekse R, Quinn JF, et al. (2005) Safety and acceptability of the research lumbar puncture. Alzheimer Dis Assoc Disord. 19(4):220–225. 62. Bateman RJ, Wen G, Morris JC, et al. (2007) Fluctuations of CSF amyloid-beta levels: implications for a diagnostic and therapeutic biomarker. Neurology. 68(9):666–9. 63. Bieck PR LM, Ereshefsky L, Lin Q, et al. (1998) Use of a biomarker of norepinephrine transporter (NET) inhibition to assess atomoxetine effects during clinically recommended dosing. 64. Cutler NR, Polinsky RJ, Sramek JJ, et al. (1998) Dose-dependent CSF acetylcholinesterase inhibition by SDZ ENA 713 in Alzheimer’s disease. Acta Neurol Scand. 97(4):244–50. 65. Weinstock M, Razin M, Chorev M, et al. (1994) Pharmacological evaluation of phenylcarbamates as CNS-selective acetylcholinesterase inhibitors. J Neural Transm Suppl. 43:219–25. 66. Geller NL. (1984) Design of phase I and II clinical trials in cancer: a statistician’s view. Cancer Invest. 2(6):483–91. 67. Gottlieb S. Speech. Conference on Adaptive Trial Design; July 10, 2006; Washington, DC. 68. Chow SC, Chang M. (2008) Adaptive design methods in clinical trials—a review. Orphanet J Rare Dis. 3:11. 69. Maloney A, Karlsson MO, Simonsson US. (2007) Optimal adaptive design in clinical drug development: a simulation example. J Clin Pharmacol. 47(10):1231–43. 70. Ivanova A, Murphy M. (2009) An adaptive first in man dose-escalation study of NGX267: statistical, clinical, and operational considerations. J Biopharm Stat. 19(2):247– 55. 71. Golde TE. (2006) Disease modifying therapy for AD? J Neurochem. 99(3):689–707. 72. Siemers ER, Quinn JF, Kaye J, et al. (2006) Effects of a gamma-secretase inhibitor in a randomized study of patients with Alzheimer disease. Neurology. 66(4):602–4. 73. Jackson WJ, Buccafusco JJ, Terry AV, et al. (1995) Velnacrine maleate improves delayed matching performance by aged monkeys. Psychopharmacology (Berl). 119(4):391–8. 74. Puri SK, Hsu RS, Ho I, et al. (1989) Single dose safety, tolerance, and pharmacokinetics of HP 029 in healthy young men: a potential Alzheimer agent. J Clin Pharmacol. 29(3):278–84. 75. Puri SK, Ho I, Hsu R, et al. (1990) Multiple dose pharmacokinetics, safety, and tolerance of velnacrine (HP 029) in healthy elderly subjects: a potential therapeutic agent for Alzheimer’s disease. J Clin Pharmacol. 30(10):948–55. 76. Cutler NR, Murphy MF, Nash RJ, et al. (1990) Clinical safety, tolerance, and plasma levels of the oral anticholinesterase 1,2,3,4-tetrahydro-9-aminoacridin-1-oL-maleate (HP 029) in Alzheimer’s disease: preliminary findings. J Clin Pharmacol. 30(6):556–61.

c06

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:5

Char Count=

Bridging and CSF studies

229

77. Goldberg MR, Barchowsky A, McCrea J, et al. Heptylphysostigmine (L-693, 487): safety and cholinesterase inhibition in a placebo-controlled riding-dose healthy volunteers study. Second International Springfield Symposium on Advances in Alzheimer’s Therapy; May 1991. Springfield, Illinois. 78. Sramek JJ, Block GA, Reines SA, et al. (1995) A multiple-dose safety trial of eptastigmine in Alzheimer’s disease, with pharmacodynamic observations of red blood cell cholinesterase. Life Sci. 56(5):319–26. 79. Hinz V, Grewig S, Schmidt BH. (1996) Metrifonate and dichlorvos: effects of a single oral administration on cholinesterase activity in rat brain and blood. Neurochem Res. 21(3):339–45. 80. Unni LK, Womack C, Hannant ME, et al. (1994) Pharmacokinetics and pharmacodynamics of metrifonate in humans. Methods Find Exp Clin Pharmacol. 16(4):285–9. 81. Itoh A, Nitta A, Hirose M, et al. (1997) Effects of metrifonate on impairment of learning and dysfunction of cholinergic neuronal system in basal forebrain-lesioned rats. Behav Brain Res. 83(1–2):165–7. 82. Itoh A, Nitta A, Katono Y, et al. (1997) Effects of metrifonate on memory impairment and cholinergic dysfunction in rats. Eur J Pharmacol. 322(1):11–9. 83. Riekkinen P, Jr, Schmidt B, Riekkinen M. (1997) Behavioral characterization of metrifonate-improved acquisition of spatial information in medial septum-lesioned rats. Eur J Pharmacol. 323(1):11–9. 84. Cutler NR, Jhee SS, Cyrus P, et al. (1998) Safety and tolerability of metrifonate in patients with Alzheimer’s disease: results of a maximum tolerated dose study. Life Sci. 62(16):1433–41. 85. Davis R, Raby C, Callahan MJ, et al. (1993) Subtype selective muscarinic agonists: potential therapeutic agents for Alzheimer’s disease. Prog Brain Res. 98:439–45. 86. Reece PA, Bockbrader H, Sedman AJ. (1992) Safety, pharmacodynamics, and pharmacokinetics of a new muscarinic agonist, CI-979. Clin Exp Pharmacol Physiol. 21(Suppl):58. 87. Sramek JJ, Sedman AJ, Reece PA, et al. (1995) Safety and tolerability of CI-979 in patients with Alzheimer’s disease. Life Sci. 57(5):503–10. 88. Sramek JJ, Forrest M, Mengel H, et al. (1998) A bridging study of LU 25-109 in patients with probable Alzheimer’s disease. Life Sci. 62(3):195–202. 89. Padich RA, McCloskey TC, Kehne JH. (1996) 5-HT modulation of auditory and visual sensorimotor gating: II. Effects of the 5-HT2A antagonist MDL 100,907 on disruption of sound and light prepulse inhibition produced by 5-HT agonists in Wistar rats. Psychopharmacology (Berl). 124(1-2):107–16. 90. Kehne JH, Padich RA, McCloskey TC, et al. (1996) 5-HT modulation of auditory and visual sensorimotor gating: I. Effects of 5-HT releasers on sound and light prepulse inhibition in Wistar rats. Psychopharmacology (Berl). 124(1–2):95–106. 91. Zimbroff DL, Kane JM, Tamminga CA, et al., Sertindole Study Group. (1997) Controlled, dose-response study of sertindole and haloperidol in the treatment of schizophrenia. Am J Psychiatry. 154(6):782–91. 92. van Kammen DP, McEvoy JP, Targum SD, et al. (1996) A randomized, controlled, doseranging trial of sertindole in patients with schizophrenia. Psychopharmacology (Berl). 124(1–2):168–75. 93. Daniel D, Targum S, Zimbroff D, et al. Efficacy, safety, and dose response of three doses of sertindole and three doses of Haldol in schizophrenic patients. 34th Meeting of the American College of Neuropsychopharmacology; December 11–15, 1995; San Juan, Puerto Rico.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

CHAPTER 7

A case study from preclinical to early clinical trials

The purpose of this chapter is to summarize the information provided in the previous chapters using a real-world example; by creating a plausible, early clinical development program for a fictitious H3 receptor inverse agonist that we have identified as ORPHEUS. The biological and biodisposition properties of ORPHEUS are consistent with the known pharmacology of previously evaluated analogues. In order to add additional dimensions to this exercise, features have been designed into the hypothetical compound to demonstrate how one translates between animal data and first-in-man studies. This exercise takes us from the earliest preclinical assays to evaluation of safety, pharmacology, and toxicity in preclinical studies. Then, a clinical program based on this preclinical data is designed. This program consists of a traditional single-dose study in healthy young normal subjects, a novel single- to multiple-dose study in healthy elderly subjects, and a pilot study in the patient population. The goal of this exercise is to take a very novel compound (from a class of compounds that has no precedent for clinical activity) and reach a proof-of-concept study that would provide a “go” or “no go” decision for further development. The target population for the proof-of-concept study need not be the ultimate patient population.

Preclinical summary The fictitious selective histamine H3 receptor inverse agonist ORPHEUS is being developed for the treatment of mild cognitive impairment. If successful in this population, additional indications for future studies include cognitive impairment in Alzheimer’s disease (AD) and schizophrenia. The biological properties of ORPHEUS have been characterized in primary and secondary pharmacology studies, traditional safety pharmacology investigations, pharmacokinetic and drug disposition paradigms, and toxicology studies in two species (14 days). The compound and its desmethyl metabolite EURYDICE are considered candidates for the completion of IND enabling studies; at least one compound is to complete a “proof-of-activity” study in humans prior to a formal

Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

230

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

231

dose-ranging investigation in either AD or a related disorder in which deficits in memory and cognitive behaviors are prominent. The IND application will be filed in the division of Neurology Drug Products of the US Food and Drug Administration (FDA).

Summary of pharmacology In vitro, primary pharmacology ORPHEUS is a non-imidazole histamine H3 receptor inverse agonist. Primary pharmacological activity is mediated through histaminic receptor activation, particularly at the H3 receptor subtype. The evaluation of ORPHEUS activity in vitro at the human histamine receptors hH1, hH2, and hH3, cloned through transfection into Chinese hamster ovary cells, indicated dissociation constants (Ki s) ranging from 13.3 to 25.6 nM. Relative selectivity for the H3 receptor subtype was identified in the rat cerebral cortex where the Ki for H3 receptors was 3.74 nM. ORPHEUS selectivity was also suggested by the activation of rat H3 receptor-mediated signal transduction pathways. The pattern of activity is consistent with an inverse agonist. EURYDICE, the main known metabolite of ORPHEUS, while showing selectivity for the H3 receptor over the H2 and H1 receptors, was a less potent inverse agonist than ORPHEUS. In vitro receptor-binding screening, conducted to evaluate the selectivity of ORPHEUS and EURYDICE across multiple receptors, revealed activity at the serotonin 5-HT3 receptor and nonselective opiate receptors only at concentrations of 1 × 10−4 M without evidence of activity at other receptor subtypes. In vivo, primary pharmacology H3 receptor-mediated activities have been demonstrated in vivo over a broad oral-dose range. Data encompass dose-dependent reversal of learning and memory deficits in a variety of established in vivo rodent models. For example, employing a passive avoidance model in the rat, ORPHEUS at doses of 0.5 and 1 mg/kg reversed the learning and memory impairment caused by the selective M1 antagonist trihexyphenidyl. EURYDICE exhibited similar effects, but at a twofold higher dose (2 mg/kg). Similarly, employing a rat passive avoidance model in which cholinergic hypofunction was induced by treatment with ethylcholine mustard aziridinium ion, ORPHEUS improved memory (0.1 and 1 mg/kg) with some evidence of a curvilinear (nonmonotonic) dose–response relationship. ORPHEUS treatment has also shown activity in an animal model of AD, which demonstrates many of the progressive histopathological characteristics seen in the brain of patients with AD (amyloid plaques, neurofibrillary tangles, and synaptic dysfunction), as well as memory deficits that are homologous to the clinical presentation in AD. In this transgenic mouse model, ORPHEUS administered intraperitoneally at 0.1 and 0.3 mg/kg/day for 10 weeks resulted in the preservation of behavioral function on special

c07

BLBK249-Cutler

232

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

learning tests, although impacts on the formation of amyloid-beta (A␤) deposits and neurofibrillary tangles in the cortex and hippocampus were inconsistent. Other in vivo studies in female New Zealand rabbits, for which the A␤ sequence is identical to that in humans, demonstrated that systemic administration of ORPHEUS did not decrease concentrations of secreted A␤1–40 and A␤1–42 in the cerebrospinal fluid (CSF), suggesting that although ORPHEUS treatment can reduce the behavioral manifestation of cognitive deficits, the production of A␤ alloforms associated with AD is unaffected. ORPHEUS- and, to a lesser extent, EURYDICE-treated mice exhibited a higher dihydroxyphenylacetic acid (DOPAC)/dopamine ratio (+55% versus vehicle, P = 0.04), an index of dopamine turnover rate in prefrontal cortex without differences in stratum and hypothalamus. Comparable effects on serotonergic neuronal activity (5-HIAA/5-HT) in the stratum were less robust (+15% increase). Tele-methylhistamine (tMeHA) levels were significantly increased in stratum, prefrontal cortex, as well as other tissues examined.

Safety pharmacology Safety studies evaluated the pharmacological effects of ORPHEUS and EURYDICE on the central nervous system (CNS), cardiovascular and respiratory systems, gastrointestinal (GI) tract, and renal system. Central nervous system In the Irwin test, no effects of orally administered ORPHEUS were observed at 1 and 10 mg/kg p.o. At a dose of 10 mg/kg, signs of locomotor stimulation were observed. In the rotarod test, a similar profile was observed with no effects at 1 or 10 mg/kg and muscle incoordination following administration of > 10 mg/kg. EURYDICE was devoid of activity at doses < 20 mg/kg; activity similar to ORPHEUS was demonstrated at higher doses. Cardiorespiratory system In vitro and in vivo evaluations have been completed describing electrocardiographic and hemodynamic properties of ORPHEUS as well as EURYDICE, a desmethyl metabolite with comparable pharmacological activity. In vitro The cardiovascular effects of ORPHEUS and EURYDICE were evaluated in the in vitro hERG assay. ORPHEUS inhibited hERG currents by 3.0 ± 0.3%, 9.6 ± 0.7%, 27.2 ± 0.4%, and 70.4 ± 1.5% at 2, 3, 10, and 30 µM, respectively, and EURYDICE inhibited hERG currents by 1.7 ± 0.9% and 1.2 ± 0.8% at 10 and 100 µM, respectively. These evaluations were complemented by examination of the in vitro effects of ORPHEUS and EURYDICE on eight other cardiac ion channels (expressed in HEK293 cells) responsible for major components of the cardiac action potential using an automatic parallel patch-clamp system. Each compound was evaluated at 1, 10, and 100 µM, with each concentration

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

233

tested in 2–4 cells (n ≥ 2) with duration of exposure at each concentration of 5 minutes. The effects of ORPHEUS and EURYDICE on the QT interval, QRS duration, the monophasic action potential (MAPD60 ), and measures of contractility left ventricular pressure (LVP), maximum rate of relaxation (dP/dTmin ), and maximum rate of contraction (dP/dTmax ) were evaluated in vitro using an isolated retrograde perfused rabbit heart preparation (Langendorff) stimulated at 60 beats/min and 120 beats/min. ORPHEUS did not cause statistically significant changes in the QT interval or the QRS complex at 0.01, 0.1, and 1 µM. At 10 µM, the QT interval was prolonged when compared to time-matched vehicle controls, and effects on other parameters were inconsistently noted. EURYDICE did not cause statistically significant changes in the QT interval, QRS complex, MAPD60 , or contractility (LVP, dP/dTmin , and dP/dTmax ) when compared to time-matched vehicle controls. In vivo Although commonly included in IND enabling studies only, effects on the cardiovascular and respiratory systems were evaluated in the conscious telemetered Beagle dogs at doses of 1.5, 3, and 6 mg/kg using a Latin square design, given the importance of QT prolongation and signals suggested by in vitro tests (hERG, Langendorff-perfused isolated heart probe) and probe in vivo evaluations (anesthetized guinea pig). In this dog study, heart rate was increased from 8.1 to 35.4% at the 6 mg/kg dose; however, these changes were considered within the normal range of variability. No effects on PR, QT, or QTc intervals or QRS duration were observed. The evaluation of respiratory function was conducted in male and female Beagle dogs at 1.5, 3, and 6 mg/kg. In male dogs, an increase in respiratory rate of 25–30 breaths/min occurred; however, because the effect was well within the normal range for Beagle dogs, was not observed in females, and group effects were largely attributable to effects in a few animals, the observation was not considered physiologically significant.

Gastrointestinal and renal ORPHEUS was evaluated for its effects on GI transit in male Sprague Dawley rats using a paradigm that measured charcoal transit within the GI tract 1 hour post dosing orally. No pharmacologically significant effects on GI transit were noted following 1 or 10 mg/kg compared with the vehicle-treated control group. Following oral administration of 100 mg/kg ORPHEUS, there was a significant decrease in GI transit compared with vehicle-treated control animals, and there were no significant effects on gastric emptying at any ORPHEUS dose tested. The effects of ORPHEUS on renal function (urine pH, urine volume, sodium, chloride, and potassium excretion) were evaluated in saline-loaded male Sprague Dawley rats using oral doses of 0 (water), 1, 10, and 100 mg/kg of ORPHEUS. The effects of potential clinical relevance were confined to urine volume and only observed at the highest dose administered

c07

BLBK249-Cutler

234

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

(100 mg/kg ORPHEUS) where a significant (P < 0.01) 2-fold and 3.5-fold increase in urine volume at 0–3 and 3–6 hours post dose, respectively, was observed compared with the vehicle-treated animals, suggesting a moderate diuretic effect. Taken together, the results of a number of studies evaluating primary and secondary pharmacological effects indicate that ORPHEUS is a relatively selective H3 receptor inverse agonist within the CNS with limited effects on other organ systems following oral-dose administration. Higher dosages may be associated with the prolongation of the cardiac QT interval.

Summary of pharmacokinetics and drug disposition data ORPHEUS and EURYDICE exist as racemic mixtures. Only the L-enantiomer has been administered to animals. Non-GLP (Good Laboratory Practice), nonchiral methods were employed for the evaluation of samples collected from the rat and dog in the nonclinical testing program. The methods were found to have satisfactory accuracy and precision over the linear ranges for samples analyzed. The pharmacokinetics of ORPHEUS was evaluated in the rat and dog following single- and repeat-dose administration. In these studies, a primary metabolite, EURYDICE, was identified, characterized pharmacologically, and measured to evaluate any differences in metabolism between the two species. In addition, in vivo studies evaluating the CNS penetration in mice, in vitro protein binding, and metabolism studies of ORPHEUS were conducted. The bioavailability of ORPHEUS was determined to be 30% and 74% in male and female rats, respectively, following oral administration of 2 mg/kg ORPHEUS, suggesting a difference in absorption or the presystemic fate of the compound between sexes. Following single, oral-dose administration of ORPHEUS in the rat over a dose range of 4–12 mg/kg, Tmax occurred at 0.5 hour post dose. The area under the plasma concentration-time curve (AUC) from zero to infinity (AUC0–inf ) and Cmax values for ORPHEUS were higher in female than in male rats at all oral doses evaluated. The primary metabolite, EURYDICE, represented approximately 3–5% of the Cmax and AUC values observed for ORPHEUS. There were no notable differences between males and females in the formation of the primary metabolite. Singledose pharmacokinetic data suggest dose proportionality. Repeat-dose administration of ORPHEUS in the rat resulted in Cmax values 41–77% and 9–34% higher than single-dose values in males and females, respectively, suggesting increased bioavailability following repeat-dose administration. AUC values were also higher, which correlated with the increase in Cmax , and the terminal half-life (t1/2 ) increased (approximately 14 hours) compared to single-dose administration (approximately 2 hours). The evaluation of the main metabolite, EURYDICE, indicated an approximate twofold increase in Cmax and AUC values between single- and repeat-dose administration, which may reflect the increased bioavailability of ORPHEUS with repeat-dose administration.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

235

In dogs, the bioavailability of ORPHEUS ranged from 42 to 67% in males and females, respectively, following single, oral-dose administration. A linear, twofold dose-dependent dose increase for ORPHEUS in both males and females was observed for Cmax and AUC following single- and repeat-dose administration over a dose range of 1.5–6 mg/kg. Following repeat-dose administration in the dog over 2 weeks, a slight but consistent increase in AUC was observed, which may reflect an increase in bioavailability with repeatdose administration. The pharmacokinetics of ORPHEUS was similar between males and females over the dose range evaluated. The evaluation of the pharmacokinetics of the main metabolite, EURYDICE, indicated similar values for AUC0–inf and Cmax compared with ORPHEUS, while the t1/2 and mean residence time of EURYDICE were two- to threefold longer compared with ORPHEUS at the 2-week assessment. Distribution studies in mice confirmed that ORPHEUS rapidly crosses the blood–brain barrier following oral administration with peak levels in brain tissue approximately 30 minutes post dosing. In protein binding studies with human plasma, ORPHEUS was 11.1% and 13.8% protein bound at concentrations of 1 and 10 µM, while the protein binding of EURYDICE was 18.8% and 22.2% at the same concentrations. The evaluation of ORPHEUS in metabolic stability studies indicated only 5.2% and 12% metabolism of ORPHEUS by human liver microsomes at concentrations of 1 and 10 µM. EURYDICE was not metabolized by dog or human liver microsomes. Separate metabolic profiling in vitro studies identified CYP2A6 as having the greatest potential for metabolism of ORPHEUS to EURYDICE. Other P450 enzymes identified included CYP2B6, CYP2C19, and CYP206; however, no single enzyme produced >0.1% conversion. The potential for ORPHEUS and EURYDICE to inhibit cytochrome P450 enzymes was evaluated in human liver microsomes. At concentrations up to 100 µM, there was no significant inhibition of CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4, or flavin mono-oxygenase, suggesting a low potential for drug–drug interactions metabolically in humans.

Summary of toxicology ORPHEUS has been evaluated following oral administration in 14-day GLP toxicology studies in rats and dogs. Acute toxicity was evaluated in male and female rats following oral administration of ORPHEUS. Clinical signs consistent with pharmacological activity were observed and included hyperactivity and piloerection approximately 1 hour after the administration of 16 and 48 mg/kg. All but one animal receiving 48 mg/kg died or was sacrificed. Doses up to 16 mg/kg were considered well-tolerated and the no observed effect level (NOEL) was determined to be 4 mg/kg. Following single oral-dose administration, male and female dogs exhibited similar clinical signs to those observed in the rat. The onset of effects was rapid following an oral dose of 5 mg/kg, with severe clinical signs observed

c07

BLBK249-Cutler

236

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

7 minutes post dose. This rapid onset of clinical signs is supported by the pharmacokinetic studies, which indicate ORPHEUS is readily absorbed from the GI tract and rapidly crosses the blood–brain barrier. The clinical signs observed included coughing, ataxia, body tremors, irritability, emesis, and increased motor activity, all of which were consistent with the pharmacological activity of the compound. Animals had generally recovered from these effects the following morning. Single doses of 0.2 and 1 mg/kg were generally welltolerated in the dog. Repeat-dose toxicity was evaluated in male and female rats and dogs following oral administration for 2 weeks. In the rat, 4, 8, and 16 mg/kg/day in the 2-week studies given as divided doses (administered approximately 6 hours apart) were generally well-tolerated. The most common clinical signs observed were transient and represented an exaggerated pharmacological response. Additional effects included decreased weight gain in males and isolated changes in hematology, liver enzymes, and urinalysis; none of these changes were correlated with macroscopic or microscopic tissue changes and none were considered to be of toxicological significance. The changes in liver enzymes were thought to result from metabolic adaptation and/or activation rather than toxicity. The evaluation of the pharmacokinetics indicated that the maximum blood levels of ORPHEUS and the major active metabolite EURYDICE occur approximately 30 minutes post dose. Higher circulating plasma levels occurred following repeat-dose administration compared to single-dose administration, suggesting that an increase in bioavailability may occur with repeat dosing. In the 2-week study, ORPHEUS was administered at 0.5, 0.9, and 1.8 mg/kg/day as three divided doses approximately 5 hours apart. The effects in one animal were so severe that it had to be sacrificed. The no observed adverse effect level (NOAEL) was considered to be 0.9 mg/kg/day based on clinical findings (histopathology evaluations did not suggest other target organ activity). Genotoxicity studies were conducted using the Ames test, mouse lymphoma mutation assay, and rat micronucleus test. No evidence of a genotoxic effect was observed in these studies. Reproductive toxicology studies have not been completed. In summary, the toxicity observed following single- and repeat-oral doses of ORPHEUS in the rat and dog represents primarily an exaggerated pharmacological effect which was dose-limiting. The toxicokinetic evaluation in the rat suggests that ORPHEUS is rapidly absorbed, with higher plasma levels of ORPHEUS observed in female rats compared to male rats. The metabolism of ORPHEUS to EURYDICE occurred in the rat; however, the concentrations of EURYDICE were 13- to 33-fold less than the parent compound. In the dog, ORPHEUS was also rapidly absorbed; however, the conversion to EURYDICE was higher in the dog than in the rat, with approximately equal amounts of parent and metabolite present. There were no significant differences regarding ORPHEUS metabolism by sex in the dog.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

237

Because human cytochrome P450 isoforms differ from rodent isoforms in both composition and catalytic activity, the extrapolation of sex differences noted only in the rat to the human condition is tenuous, and the lower rate of metabolism of ORPHEUS implied by toxicokinetic data in female rats may not mirror the clinical experience. Nevertheless, sex-related differences in biodisposition remain an important variable to be evaluated in the course of the clinical program.

Core clinical studies Single-dose, ascending, sequential cohort design A Double-Blind, Placebo-Controlled, Single, Escalating Dose Study to Evaluate the Preliminary Pharmacokinetics, Safety, Tolerability, and Pharmacodynamics of ORPHEUS in Healthy Young Male and Female Volunteers

Objectives r Estimation of single-dose pharmacokinetic parameters in up to five separate sequential cohorts of young, healthy male and female volunteers exposed to a single oral dose of ORPHEUS at A, B, C, D, and E mg (10 subjects/cohort: 8 receive ORPHEUS, 2 receive placebo; 1 stratification factor based on sex). r Preliminary safety and tolerability of a single dose of ORPHEUS administered orally at A, B, C, D, and E mg/day, including detailed assessments of electrocardiogram (ECG) interval parameters and evaluation of drug-induced exaggerated orthostasis using a codified postural maneuver. r Preliminary estimation of the influence of gender on single-dose pharmacokinetic parameters. The stratification of each cohort will assure that an equal number of subjects are randomized by gender—8 subjects receive ORPHEUS per cohort (4 males and 4 females), 2 subjects receive placebo (1 male and 1 female). r Exploratory analysis of ORPHEUS-induced alterations on spontaneous brain function and cognitive event-related potentials (ERPs) in each cohort by mapping of pharmaco-electroencephalograms (EEGs). Rationale The program begins with a first-in-man study to determine the maximumtolerated single dose (MTD), as well as the preliminary safety and tolerability of the H3 receptor inverse agonist, ORPHEUS. Pharmacokinetic profiling and safety assessments are foremost in this study and all procedures relevant to efficacy evaluations are considered hypothesis generation maneuvers. The compound exhibited several preclinical signals that could indicate adverse event risks in humans of either a cardiac or CNS nature at dosages potentially near the maximum well-tolerated dose (see the preclinical data section for the complete preclinical profile of ORPHEUS). Therefore, subjects for this study will be healthy, young adults (aged 18–50 years), who are best suited

c07

BLBK249-Cutler

238

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

to tolerate potential adverse events, and exposure will be incrementally increased in successive cohorts using a traditional dose-escalation scheme with a formalized intercohort review processes. An adaptive dose-ranging design using a modified Fibonacci dose-escalation scheme coupled to a Storer “upand-down design” was considered and dismissed due to the preclinical safety data presented [1]. Pharmacokinetic and safety data from this study are considered informative for subsequent studies using normal elderly volunteers, and then patients where efficacy evaluations, including pharmacokinetic/pharmacodynamic correlations, are intended. The pharmacodynamic effects of ORPHEUS on spontaneous brain function and cognitive ERPs will be evaluated in each cohort by the mapping of pharmaco-EEGs. These methods collectively will be used to determine whether, how, when, and at which dosage ORPHEUS produces a CNS effect (pharmaco-EEG). In addition, because of the occurrence of seizures during preclinical safety pharmacology studies, all subjects will be screened for seizures or a history of seizures, using screening ECGs designed to detect epileptiform activity—sleep deprivation and photic simulation—until additional clinical profiling is available. Due to preclinical evidence of a curvilinear (nonmonotonic) dose–response relationship on measures of cognition, dose escalating for the purposes of estimating MTD may produce toxicity in the absence of appreciable clinical benefit. Specifically, a nonmonotonic dose response suggests that increased doses may not produce proportional increases in drug efficacy, although other nonspecific and potentially deleterious pharmacological effects would be considered lightly. The suggestions of sex-related differences in pharmacokinetics and biodisposition in animals may not translate reliably from animals to humans (adjusting for weight or body surface area, etc.). However, due to preclinical gender differences in the bioavailability of ORPHEUS that have been noted, the protocol includes males and females to evaluate pharmacokinetic differences by gender effects as an exploratory objective using equal numbers of males and females in each cohort. In addition, preclinical data suggest the possibility that high doses of ORPHEUS could result in prolongation of the cardiac QT interval. For this reason, ECG telemetry monitoring throughout the study is employed using techniques that (1) assure sampling intervals for QT parameters follow adequate patient rest (“good cardiac hygiene”); and (2) emphasize data analysis around the estimated Tmax of ORPHEUS. An alternative procedure in which triplicate digital ECGs are obtained at discrete time points prior to and post dosing is also feasible, in the event that telemetry assessments are not considered optimal. Finally, because of the potential for exaggerated orthostatic blood pressure changes in association with product use (equivocal preclinical data noted), blood pressure changes will be monitored during a codified postural maneuver.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

239

Subjects This study is being conducted in healthy young males and females (18– 50 years old). The impact of gender on the plasma pharmacokinetic profile is an exploratory objective; therefore, each cohort will contain at least four subjects of each gender who will receive ORPHEUS. Within each cohort, the total number of subjects exposed will be 8 (4 males:4 females). One male and 1 female will receive placebo in each cohort. With 5 potential cohorts, there will be a total of 50 subjects in the study. Design This study is a double-blind, placebo-controlled, ascending-dose design with up to 5 sequential cohorts consisting of 10 subjects per cohort (randomized 8:2, ORPHEUS:placebo). The primary objectives of the study are to evaluate the MTD and single-dose plasma pharmacokinetics of ORPHEUS, as well as safety and tolerability, in normal healthy young volunteers. The estimation of single-dose pharmacokinetic parameters will be completed in up to 5 separate sequential cohorts of young, healthy male and female volunteers exposed to a single oral dose of ORPHEUS at A (the MRSDsee below), B, C, D, and E mg (10 subjects/cohort: 8 receive ORPHEUS, 2 receive placebo; 1 stratification factor based on sex). This will include the detailed assessments of ECG interval parameters. The preliminary estimation of the influence of gender on single-dose pharmacokinetic parameters will also be determined. The stratification of each cohort will assure that an equal number of subjects are randomized by gender—eight subjects receive ORPHEUS per cohort (four males and four females), two subjects receive placebo (one male and one female). Although safety margins at the proposed maximal clinical dose is appropriate for a first-in-man investigation, it is important to note that monitoring procedures have been incorporated into the protocol based on safety and toxicology data presented (principally cardiac, CNS, and metabolic). These procedures will assist in guiding dosage escalation decisions for ORPHEUS in the event that attributed toxicity should occur. Decisions to continue or terminate dose escalation after each cohort are based on a clinical review of all clinical, laboratory, and ECG observations generated within that cohort. The MTD within the permitted dose range is defined in this study as the dose producing clinically notable intolerance in approximately 25% of the subjects receiving ORPHEUS (2/8), as determined by the intensity of the adverse events reported or observed through the entire dosing period. In the absence of clinically important intolerance, the highest dose of E mg will be declared the maximum well-tolerated dose. Based on an estimated 1-hour Tmax of ORPHEUS, and a half-life of 10 hours in humans, plasma sampling will be estimated over 36 hours, which will characterize the pharmacokinetic time course over at least 5 half-lives. Fifteen samples (6 mL/each) will be drawn in total. Time points for plasma draws are 0, 20, and 40 minutes, and 1, 1.5, 2, 3, 4, 5, 6, 8, 10, 12, 16, 24, and 36 hours after

c07

BLBK249-Cutler

240

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

treatment. Urine from all subjects will be collected before dosing (T = 0) and over 6 time points spaced 4 hours apart, at times 0–4, 4–8, 8–12, 12–16, 16–20, 20–24, and 24–36 hours after dosing. Modification to the sampling interval may be entertained pending receipt of pharmacokinetic profiling data from the first cohort of subjects exposed.

Dosing The proposed dose-escalation regimen for the single-dose ascending study is A (the MRSD), B, C, D, and E mg. The starting dose of a single administration of 1 mg ORPHEUS and highest dose of a single administration of E mg ORPHEUS were selected based on consideration of the preclinical efficacy and toxicology data and regulatory guidance that dictates the maximum recommended safe starting dose. The maximum recommended starting dose in humans is based on the NOAEL in the most sensitive species (dog), a scaling factor based on body surface area correction factors, and application of a safety factor to provide a margin of safety for protection of human subjects due to the uncertain animalto-human extrapolation of safety information. The consideration of all of these factors yields a conservative estimate for an initial clinical dose. Employing the NOAEL of 0.9 mg/kg/day, from the 2-week GLP toxicology study in dogs, the nonclinical data support a maximum recommended clinical starting dose of approximately 0.0486 mg/kg (2.9 mg) when a 10-fold safety factor is incorporated (scale factor, 0.54; human equivalent dose, 0.486 mg/kg; 60-kg subject). Given inherent uncertainties in extrapolating safety observations from animals to humans, a starting dose of 1 mg (approximately 0.017 mg/kg in a hypothetical 60-kg subject) is employed in the first clinical study. Modifications in both the initial starting dose for dose-escalation studies and the maximal permitted dose are based on an evolving safety experience. Pharmacokinetics All subjects with sufficient quantifiable data at a given dose level will be included in the pharmacokinetic analysis. In the event that subjects have missing data in a given cohort, the pharmacokineticist will evaluate the concentration-time profiles individually to determine if sufficient data are available for pharmacokinetic analysis. The pharmacokinetic profile of ORPHEUS will be determined using plasma and urine data. Plasma concentration-time data will be summarized by cohort (dose level) with descriptive statistics at each scheduled time point. Individual and mean concentration-time profiles will be provided for each dose level. Concentration-time data for individual subjects will be analyzed by noncompartmental methods using actual elapsed times (if available). Electrocardiograph analysis Electrocardiograph telemetry monitoring will be employed throughout the study. At predefined time points prior and following dosing, data will be sampled for ECG parameter estimates, as indicated below.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

241

Vital signs Vital signs will include systolic and diastolic blood pressures, radial pulse rates, and respiratory rates. All vital signs will be obtained after the subject has been in the supine position for at least 5 minutes. Orthostatic (postural) blood pressure and pulse measurements are obtained following supine measurements using the following procedure. Pharmaco-EEG The pharmacodynamic effects of ORPHEUS on spontaneous brain function will be evaluated in each cohort by mapping of EEGs (electroencephalograms), ERPs (event-related potentials), and LORETA (low-resolution brain electromagnetic tomography). These methods collectively will be used to determine whether, how, when, and at which dosage ORPHEUS produces a CNS effect (pharmaco-EEG). LORETA will combine the high time resolution of the EEG with a source localization method that permits a 3D tomography of brain electrical activity. ERPs will be recorded in an auditory 2-tone odd-ball paradigm without motor reaction. The subjects will be asked to mentally count the soft tone bursts and report their number at the end of the experiment. Standard tone N1 and P2 and target tone N2 and P300 components will be determined. EEG recordings will be completed at 0, 1, 3, and 6 hours after the infusions. Special safety analysis All suspected clinically important abnormalities in ECG will be read by an external, treatment-blinded cardiologist, and that reading will be considered in the final assessment by the investigator of a suspected abnormality. All interval data for analytic purposes will also be based on an external treatment-blinded computer-assisted measurement rather than the automated algorithms obtained from ECG machines at the investigative site. Specifically, ECGs recorded at the site on each subject will be sent to a central laboratory for a high-resolution measurement of the cardiac intervals and morphological assessment by a central cardiologist blinded to the study treatment. ECG measurements will be performed using digitization software with magnification of the ECG and on-screen calipers by experienced technicians and a centralized cardiologist who is blinded to the tracings. QTc intervals will be corrected using both Bazett’s formula and Fridericia’s formula. Single-ascending dose to multiple-ascending dose cohort design A Single-Center, Randomized, Double-Blind, Placebo-Controlled, Sequential Cohort Study Conducted as a “Parallel Single Dose-Multiple Dose Escalation” to determine the Safety, Tolerability, Pharmacokinetics, and Pharmacodynamics of ORPHEUS in Healthy, Elderly Volunteers

c07

BLBK249-Cutler

242

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

Objectives r Single dose ◦ Estimation of single-dose pharmacokinetic parameters in up to 3 separate sequential cohorts of healthy, elderly male and female volunteers exposed to a single oral dose of ORPHEUS at A, B, and C mg (10 subjects/cohort: 8 receive ORPHEUS, 2 receive placebo) ◦ Safety and tolerability of a single dose of ORPHEUS administered orally at A, B, and C mg, including detailed assessments of ECG interval parameters and determination of drug-induced effects upon orthostasis using a codified postural maneuver r Multiple dose ◦ Estimation of multiple-dose (QD) pharmacokinetic parameters in up to 3 separate sequential cohorts of healthy, elderly male and female volunteers administered ORPHEUS at A, B, and C mg/day for 5 days (10 subjects/cohort: 8 receive ORPHEUS, 2 receive placebo) ◦ Safety and tolerability of multiple (QD) doses of ORPHEUS at A, B, and C mg/day for 5 days, including detailed assessments of ECG interval parameters and determination of drug-induced effects upon orthostasis using a codified postural maneuver ◦ Exploratory analysis of ORPHEUS-induced alterations on spontaneous brain function and cognitive ERPs by mapping of pharmaco-EEGs Rationale The second study in a three-part clinical development program will be a single-ascending dose (SAD) to multiple-ascending dose (MAD) study of ORPHEUS in healthy, elderly male and female subjects. This design can be described as a parallel single-dose, multiple-dose escalation design. The design facilitates the determination of safety, tolerability, pharmacokinetic, and pharmacodynamic characteristics of ORPHEUS in both single and repeat-dose designs in the same elderly subjects, anticipating that both pharmacokinetic profiling as well as tolerability may be materially different in comparison to younger volunteers. Combining SAD and MAD administrations essentially provides all the data from both a single-dose study and a multiple-dose study, using one efficient, safe, and cost-effective paradigm in a relevant subject population. Specifically, healthy elderly subjects in this study more accurately determine the safety, tolerability, pharmacokinetic, and pharmacodynamic characteristics of ORPHEUS in a population that better matches the compound’s target population physiologically; i.e., elderly patients with cognitive impairments such as mild cognitive impairment (MCI) and mild-to-moderate AD. The goal of the study is to evaluate (1) the safety and tolerability profile of ORPHEUS administered as both single and multiple doses to the same subjects; (2) the single-dose and multiple-dose pharmacokinetic parameters in plasma and urine; and (3) the pharmacodynamics of ORPHEUS (EEG) after single and repeat once-daily oral administration of escalating doses in healthy, elderly male and female subjects.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

243

Because of the potential vulnerability of elderly subjects to untoward cardiovascular effects (both hemodynamic as well as electrophysiological), and the impact of these observations on subsequent clinical development, key safety measures emphasize cardiovascular assessments. Pharmaco-EEG assessments are used for hypothesis-generating purposes. The assessment of effects on cognition are deferred until a clearly defined and cognitively impaired population can be evaluated with a full suite of pharmacokinetic and safety observations that should facilitate the design (study #3, mild cognitive impairment).

Subjects Healthy elderly subjects (aged 50–70 years, inclusive) will be used in this study in order to more accurately determine the safety, tolerability, and pharmacokinetic characteristics of ORPHEUS in a population that better matches the compound’s target population; specifically, elderly patients with cognitive impairments such as MCI and AD. Design This is a single-center, randomized, double-blind, placebo-controlled, sequential cohort study conducted in two overlapping phases. The first phase of the study will evaluate single doses of ORPHEUS. The second phase of the study will evaluate multiple, once-daily doses of ORPHEUS over 5 consecutive days in the same patients who received a single dose. Multiple once-daily dosing over 5 days is assumed to be sufficiently long to achieve steady-state pharmacokinetic conditions. The initiation of the single-dose and multiple-dose cohorts are staggered in order to create overlapping panels (see Figure 7.1). Subjects who receive a single dose of ORPHEUS with good tolerability also receive the identical dosage level over 5 days (QD) after an intervening period of 3 days. During this intervening 3-day period, a separate group of subjects is administered a single dose of the next higher dosage level. Single-dose phase In the SAD escalation phase of the study, 3 cohorts of 10 elderly male or female subjects (aged 50–70 years, inclusive) will be randomized to either receive ORPHEUS (8) or placebo (2). Each cohort will be administered study medication at a specified dose level (or placebo) as a single oral dose, which will be followed by a 3-day observation period before the multiple-dose segment of the study begins (see Figure 7.1 for the study design). The dosages selected (A, B, and C mg) will be based on safety, tolerability, and pharmacokinetic data obtained in young normal volunteers. The increments between dosage levels will be chosen to assure minimal overlap in total exposure between adjacent cohorts. Potential accumulation will be estimated through pharmacokinetic modeling (e.g., a technique of “superimposition”).

c07

BLBK249-Cutler

244

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

Figure 7.1 Single-to-multiple-dose study design.

The oral starting dose A mg of ORPHEUS for the single-dose phase of this study is based on clinical safety, tolerance, and pharmacokinetic data that had been previously obtained in a single-dose study of healthy young subjects. The initiation of dosing for each single-dose level will be staggered at 4-day intervals (1 day for dosing + 3 days washout). The proposed single-dose levels of A mg (cohort A), B mg (cohort B), and C mg (cohort C) will be administered sequentially. Dosage escalation decisions are based on a formal intercohort review of safety information and criteria for escalation as previously formulated in the single-dose study in normal young volunteers. Dose escalation will continue to the maximum proposed dose unless drug (active)-treated subjects demonstrate unacceptable safety results, as previously defined. Multiple-dose phase In the multiple-dose phase of the study, subjects in each of the single-dose cohorts will continue to receive 5 QD doses of either ORPHEUS (8) or placebo (2), at the same dose level they received as the single dose. Thus, the sequence for a patient is: single dose—pause (3 days washout)—continuation with multiple dosing at the same dosage level for 5 days. This design ensures that subjects will not receive multiple doses of ORPHEUS until a single-dose trial of the next higher dose has been concluded in a different group of subjects (see Figure 7.1 for the study design). At each dose level, a decision will be made to proceed to the next higher dosage based on safety assessments. The process for this review, and the algorithm used for the decision, is identical as that previously employed in the

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

245

single-dose segment. For each dose level, safety, pharmacokinetic, and pharmacodynamic assessments will be made over selected days during multipledose administration; e.g., following administration of the first dose and on the last day of dosing. A follow-up visit occurs on Day 14. Subjects will be confined to the clinical unit for the entire duration of the trial (Day −2 prior to single-dose administration through 36 hours post dose in the multiple-dose administration).

Monitoring Monitoring procedures have been incorporated into the protocol based on safety and toxicology data provided from preclinical studies (principally cardiac, CNS, and metabolic). Preclinical information is augmented by clinical experience in a single-dose, escalating, sequential cohort design in normal young volunteers that is antecedent to this investigation. The initial starting dose, the intended maximal dose, and the increments between adjacent dosage levels for this study also will be informed through pharmacokinetic modeling. These procedures will assist in guiding dosage escalation decisions for ORPHEUS in the event that attributed toxicity should occur. Decisions to continue or terminate dose escalation after each cohort (single or multiple dose) are based on a review of all clinical, laboratory, and electrocardiographic observations generated within that cohort. The MTD within the permitted dose range is defined in this study as the dose producing clinically notable intolerance in approximately 25% of the subjects receiving ORPHEUS (2/8), as determined by the intensity of the adverse events reported or observed through the entire dosing period. In the absence of clinically important intolerance, the highest dose of C mg will be declared the maximum well-tolerated dose. The administration of ORPHEUS will be assumed as sufficient evidence for at least a plausible adverse event attribution; i.e., a temporal association with administration provides at least a possible association regardless of assumed biology. Design aspects relevant to single-dose phase of study Based on an estimated 1-hour Tmax of ORPHEUS, and a half-life of 10 hours in man, plasma sampling for the single-dose study will be done over 36 hours post dose, which will characterize the pharmacokinetic time course over at least 5 half-lives. An estimated 16 samples (6 mL/each) will be drawn in total. Time points for plasma draws are 0, 20, and 40 minutes, and 1, 1.5, 2, 3, 4, 5, 6, 8, 10, 12, 16, 24, and 36 hours after treatment. Urine from all subjects will be collected before dosing (T = 0) and over 7 time points spaced 4 hours apart, at times 0–4, 4–8, 8–12, 12–16, 16–20, 20–24, and 24–36 hours after dosing. Modification to the sampling interval may be implemented as a protocol amendment pending receipt of pharmacokinetic profiling data from the first cohort of subjects exposed.

c07

BLBK249-Cutler

246

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

Design aspects relevant to multiple-dose phase of study Multiple venipunctures will occur on the first and last day of dosing at selected time points prior to and following administration of ORPHEUS. Twenty-four-hour plasma sampling will be done following the first day of dosing (just prior to dosing, multiple time points post dosing, last assessment at trough); 36-hour sampling will occur on the final day of dosing (just prior to dosing, multiple time points post dosing through 36 hours). Trough levels just prior to the morning dose will be obtained on all days for the purposes of confirming that steady-state levels have been achieved by the end of dosing. Time points for venipuncture during the multiple-dose phase will follow the same schedule as the single-dose segment. Urine collection will be done on the last day of dosing through 36 hours. Dosing The proposed dose-escalation regimen for the SAD and MAD studies is A, B, and C mg. The starting dose of a single administration of A mg ORPHEUS and the highest dose of a single administration of C mg ORPHEUS were selected based on consideration of the preclinical efficacy and toxicology data, the results of the SAD study (pharmacokinetic modeling), and regulatory guidance that influences estimates for safety margins, given the preclinical toxicology data provided. Pharmacokinetics Pharmacokinetic parameters in plasma that will be examined in this trial are as follows: area under the plasma concentration-time curve (AUC0–24 , AUC0–36 ), maximum observed concentration (Cmax ), minimum observed concentration (Cmin ), time to reach Cmax (Tmax ), apparent terminal half-life (t1/2 ), apparent total clearance (CL/F), and apparent volume of distribution (Vd/F). Determining the extent of accumulation with once daily dosing over 5 days for both ORPHEUS and EURYDICE and estimating time to steady state are our key objectives. ECG analysis ECG telemetry monitoring will be employed throughout the study. ECG tracings for analytic purposes will be clustered around the estimated Tmax of ORPHEUS and other selected time points to permit correlation with plasma levels of either the parent or metabolite. Vital signs Vital signs will include systolic and diastolic blood pressures, radial pulse rates, and respiratory rates. All vital signs will be obtained after the subject has been in the supine position for at least 5 minutes. Orthostatic (postural) blood pressure and pulse measurements are obtained.

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

247

Special safety analysis All suspected clinically important abnormalities in ECG will be read by an external, treatment-blinded cardiologist, and that reading will be considered in the final assessment by the investigator of a suspected abnormality. All interval data for analytic purposes will also be based on an external treatment-blinded computer-assisted measurement rather than the automated algorithms obtained from ECG machines at the investigative site. Specifically, ECGs recorded at the site on each subject will be sent to a central laboratory for a high-resolution measurement of the cardiac intervals and morphological assessment by a central cardiologist blinded to the study treatment. ECG measurements will be performed using digitization software with magnification of the ECG and on-screen calipers by experienced technicians and a centralized cardiologist who was blinded to the tracings. QTc intervals will be corrected using both Bazett’s formula and Fridericia’s formula. Pharmaco-EEG Pharmaco-EEG analysis will be conducted similarly to that described in the previous section.

Multiple-dose study in MCI patients An Evaluation of Safety, Tolerability, Pharmacokinetics, and Pharmacodynamics of ORPHEUS in Patients with Mild Cognitive Impairment (MCI)

Objectives r Evaluate the safety, tolerance, and steady-state plasma pharmacokinetics following multiple doses of ORPHEUS in elderly patients with the amnestic subtype of mild cognitive impairment (MCI). r Assess the pharmacodynamics of ORPHEUS on the monoamine neurotransmitter metabolites of serotonin and dopamine (5-HIAA and DOPAC) and potentially the neurotransmitters themselves in CSF. The pharmacodynamic profile of ORPHEUS on monoamine metabolites will be compared to that of atomoxetine, a reference standard. r Assess the pharmacokinetics of ORPHEUS (EURYDICE) in CSF under steady-state conditions. r Assess the cognitive-enhancing properties of ORPHEUS in a patient population with MCI possessing well-characterized memory and cognitive deficits. Rationale A multiple-dose study will be carried out in elderly patients with MCI based on currently acceptable diagnostic criteria for the amnestic subtype of this disorder. The study will assess the pharmacokinetics of ORPHEUS in CSF and plasma under steady-state conditions, as well as assess the pharmacodynamics of ORPHEUS on the monoamine neurotransmitter metabolites of serotonin and dopamine in the CSF. By design, there is sufficient sample to justify an

c07

BLBK249-Cutler

248

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

investigation of the cognitive-enhancing properties of ORPHEUS in a patient population with MCI (one of the compound’s target indications). The design will allow for a comparison between the pharmacodynamic profiles on monoamine metabolites after ORPHEUS administration versus atomoxetine administration, a reference standard. These data will facilitate comparisons between changes in biomarker status and the cognitive enhancement properties of ORPHEUS.

Subjects This study will be conducted in 24 elderly amnestic MCI patients (55–75 years old, inclusive). MCI patients should be considered the first and best “patient” population to assess putative therapeutic effects that may have cognitiveenhancing effects for several reasons including: r Well-characterized, selective deficits in cognitive dysfunction despite otherwise normal functioning, ensuring high levels of study compliance and ease of conduct with highly focused neuropsychological test assessments r Proven signal detection in prior studies of cognitive-enhancing agents whose primary pharmacology in preclinical behavioral models suggests memory/ cognitive enhancement r Ability to generalize cognitive test results from MCI to AD and Cognitive Impairment Associated with Schizophrenia (CIAS) populations across several drug development programs The hypothesis does not assess conversion to AD, as that objective is inconsistent with the primary pharmacology thus far presented for ORPHEUS and is in addition beyond the scope of the duration of exposure permitted by available toxicology information. Rather, these patients are considered to have prodromal AD that is characterized by a selective cognitive impairment that should be modified by acute administration of an H3 receptor inverse receptor agonist. Because we used a patient population (MCI) which is leveraged toward cognitive deficits with sufficient sample size, this final study in a core clinical development program provides a decision-making efficacy readout in a sequence of three proposed clinical trials (single-dose normal volunteers, singledose/multiple-dose elderly volunteers, and patients with MCI). Five clinical criteria are assessed in order to suggest a diagnosis of amnestic MCI. The first criterion requires a memory complaint preferably corroborated by an informant. The second criterion refers to an objective memory impairment established by neuropsychological tests. The third and fourth criteria concern the relatively normal general cognitive function and activities of daily living. Finally, these subjects will not meet the Diagnostic and Statistical Manual of Mental Disorders, vol. IV criteria for dementia. Design Using the maximally well-tolerated dose in the antecedent multiple-dose study in elderly volunteers, pharmacodynamic variables (biomarkers and

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

249

cognition) and pharmacokinetic parameters (plasma and CSF) of ORPHEUS will be evaluated. Biomarkers are included for hypothesis generation while cognitive assessments will be included for a definitive conclusion regarding the likelihood of acute symptomatic benefit. Twenty-four elderly MCI patients (12 receiving the MTD of ORPHEUS, 6 receiving a standard clinical dosage level of atomoxetine, and 6 receiving placebo) will have a single LP baseline collection of CSF and a 24-hour plasma collection on Day −1, followed by treatment with ORPHEUS, atomoxetine, or placebo for 21 days, and a 24-hour CSF and plasma collection after the last dose on Day 21. The entire study is conducted as an inpatient evaluation. Plasma will be collected for pharmacokinetic analysis of ORPHEUS on Days −1, 7, 14, 21, 22, 23, and 24. Postural tests will be conducted on each of these days, except Day 21 (due to 24 CSF sampling on this day), and will therefore be conducted on Day 20 instead. The CSF collection will be used to evaluate ORPHEUS pharmacokinetics as well as drug-induced changes in levels of neurotransmitter metabolite biomarkers. The plasma collections will be used to estimate ORPHEUS pharmacokinetic parameters and correlate these data with drug-induced changes in CSF biomarkers and cognitive abilities. The biomarkers of interest, given the pharmacological properties of ORPHEUS, will be restricted to the CSF. Safety measures, consisting of physical and neurological exams, vital signs, ECG, clinical laboratory tests, and assessments for adverse events, will be conducted routinely during the inpatient portion of the study (see Figure 7.2 for a schematic of the study timeline). Cognitive assessments will be performed at baseline, for 24 hours following the first dose administration at selected time points (comparable to the schedule for pharmacokinetic assessments), and then weekly to evaluate the cognitive-enhancing properties of ORPHEUS in comparison to baseline

Followup Discharge

Admission; cognitive assessments Cognitive assessments Daily dosing (QD) for 21 days

Cognitive assessments

Screening (Day –28 to –3) –2 –1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 31±2

Single LP CSF and 24 hour plasma collection; postural test

24 hour plasma collection; postural test; cognitive assessments

24 hour plasma collection; postural test; cognitive assessments

Postural test 24 hour CSF and plasma collection; cognitive assessments

Single plasma collection; postural test

Figure 7.2 Study timeline—fixed dose of ORPHEUS, atomoxetine, or placebo dosed daily.

c07

BLBK249-Cutler

250

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

(a within-treatment-group comparison). Cognitive data from the placebo group and from placebo + atomoxetine (under the assumption of no treatment difference) will be used to examine biostatistical contrasts between treatments in a sensitivity analysis. Thus, there is a within-group comparison (n = 12), and at least 2 potential between-treatment-group comparisons ultimately utilizing all 24 subjects. Both the detail of cognitive assessments and the amount of data collected should provide a reliable conclusion regarding potential acute symptomatic benefits in patients with MCI (and potentially other cognitive disorders). This is regarded as a decision-making investigation (see section on “Cognitive assessments” for more information).

Dosing The dose of ORPHEUS for this study will be determined from the maximally well-tolerated dose in the antecedent multiple dose study in elderly volunteers. The dosage level of atomoxetine is based on previous clinical pharmacodynamic evaluations of CSF dopamine and serotonin biomarkers [2]. Pharmacokinetic analysis Pharmacokinetic parameters in plasma that will be examined in this trial are only for ORPHEUS (not atomoxetine) and are as follows: area under the plasma concentration-time curve (AUC0–24 ), maximum observed concentration (Cmax ), time to reach Cmax (Tmax ), apparent terminal half-life (t1/2 ), apparent total clearance (CL/F), and apparent volume of distribution (Vd/F). Key plasma pharmacokinetic parameters for correlation with biomarkers and cognitive effects will include all of these variables with the exception of apparent total clearance and apparent line of distribution. Biomarker analyses (CSF) On Day −1 (baseline) there will be one single LP collection of CSF for ORPHEUS pharmacokinetic and neurotransmitter metabolites DOPAC and 5HIAA, and on Day 21 there will be a 24-hour CSF collection for ORPHEUS pharmacokinetic and DOPAC and 5-HIAA. Twenty-four-hour plasma collections are done on Days −1, 7, 14, and 21, and single plasma collections are done on Days 22, 23, and 24. Postural tests are done on Days −1, 7, 14, 20, 22, 23, and 24. ECG analysis ECG telemetry monitoring will be employed throughout the study. The primary intention for this assessment is clinical safety. All suspected clinically important abnormalities in ECG will be read by an external, treatmentblinded cardiologist, and that reading will be considered in the final assessment by the investigator of a suspected abnormality. All interval data for analytic purposes will also be based on an external treatment-blinded computer-assisted measurement rather than the automated

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

251

algorithms obtained from ECG machines at the investigative site. Specifically, ECGs recorded at the site on each subject will be sent to a central laboratory (Cardiocore, Bethesda MD, USA) for a high-resolution measurement of the cardiac intervals and morphological assessment by a central cardiologist blinded to the study treatment. ECG measurements will be performed using digitization software with magnification of the ECG and on-screen calipers by experienced technicians and a centralized cardiologist who is blinded to the tracings. QTc intervals will be corrected using both Bazett’s formula and Fridericia’s formula.

Cognitive assessments The cognitive measures chosen are commonly used to assess cognitive function of both younger and elderly healthy patient populations as well as in AD patients in both the verbal and visual domains, and require a variety of responses from the subject. Importantly, all measures can be administered to subjects without interfering with standard phase 1 pharmacokinetic/pharmacodynamic producers. For example, these measures can be completed with subjects lying down or reclined, if necessary, and can be completed without the use of both hands/arms. Also, all of these measures have alternate forms available to help control the presence of any possible practice effects, which are more likely to occur when repeated cognitive measures are administered in close temporal proximity. The entire battery should take approximately 30–35 minutes to complete. This test battery was also chosen to reflect the cognitive domains of working memory (describes the cognitive processes that allow one to actively maintain and manipulate information over a short duration of time), executive functioning (involved in processes such as planning, cognitive flexibility, abstract thinking, rule acquisition), and explicit memory (memory of all those things that one subject is aware of remembering and that can be described in words). These cognitive domains and the measures that compose them are equivalent to some of the measures currently being utilized by the Alzheimer’s Disease Neuroimaging Initiative (ADNI). These selective domains are also consistent with the findings from our recent systematic overview of the quantitative relationship between computerized assessment batteries and AD treatment that suggested that measures of attention and working memory yielded the biggest effect sizes with regard to treatment (d = −0.85, 95% CI = −0.94 < ␦ < −0.78) [3]. Learning and delayed recall yielded more moderate effects (d = −0.51, 95% CI = −0.62 < ␦ < −0.41). A small but significant effect of age (P = 0.043) (i.e., the older the subject, the less effect) as well as education (P = 0.051) (i.e., the more education, the better effect) was noted on effect size. A variety of computerized assessment batteries were considered but not adopted primarily due to limitations in their application. None of the below measures utilize a computer for administration or scoring.

c07

BLBK249-Cutler

252

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

Chapter 7

Conclusion The fictitious example with our putative cognitive enhancer demonstrated the design of three initial core clinical studies which culminated in a relatively complex proof-of-concept study in a patient population (MCI) that would be practical to conduct in a real-world situation. If successful in demonstrating potential efficacy, other target populations with cognitive impairment, such as AD or schizophrenia, could be viable. In these cases, reestablishing the MTD in the specific patient population will be essential to maximizing the compound’s potential (as detailed in the previous chapter). We specifically designed out any potential for interactions in our example, but the real world is typically more complex. The completion of a food-effect study would be important for establishing dosing regimens in the larger outpatient clinical trials and should be conducted as soon as feasible. In this final chapter, we wished to summarize all of the core information provided in the previous chapters in a unique way; by providing a plausible example of the early drug development process for a novel, fictitious compound. Our goal was to show you how important preclinical pharmacology, safety, and pharmacokinetic data can fit together into the early clinical drug development puzzle. Given the competitive pressures existing today, including high costs and small margins for errors, drug developers cannot afford to conduct early studies using cookie-cutter protocols and see what falls out. A tailored approach is essential from the start, affording maximal protection for human subjects in phase I. Using preclinical data to help design the early trials also helps speed the development process; by targeting potential areas of concern as early as possible, providing a framework for the essential core studies, and suggesting ancillary studies which may be needed to further explore significant or aberrant findings. In this manner, the preclinical data is effectively merged into the clinical program to create a solid foundation for clinical development, driven by regulatory mandates, budget and timeline specifications, and rational science. Our fictitious drug development program included a second study which allowed for the quick transition of findings in young healthy subjects to a healthy elderly population in a single-to-multiple-dose study design. This combined two normally separate studies into one, saving time and allowing for the identification of the MTD and safe dose range in a population similar in age to the target patient population—a critical piece of data for later clinical trials. The final clinical study in our program bridged into a patient population with an early proof-of-concept study which employed cognitive assessments and CSF sampling to compare the compound’s central effect on monoamines with a reference compound. Hypothesis generation and testing was used at every step in order to learn as much as possible about the pharmacology, safety, and potential effectiveness of the compound prior to larger clinical

c07

BLBK249-Cutler

Trim: 6in X 9in

December 29, 2009

22:14

Char Count=

A case study from preclinical to early clinical trials

253

trials. If we have built a sound foundation at this early stage, then we have optimized the ability to make a solid go/no-go decision for continued development, which entails a very significant monetary commitment. In addition, we do a great disservice to patients if they are exposed to needless safety issues or ineffective doses in large-scale trials. Early drug development is the most critical component of the clinical development program; one that continually challenges our knowledge, reasoning, and creative processes alike.

References 1. Ivanova A, Murphy M. (2009) An adaptive first in man dose-escalation study of NGX267: statistical, clinical, and operational considerations. J Biopharm Stat. 19(2):247–55. 2. Leibowitz M EL, Lin Q, Ledent E, et al. (2005) Use of a biomarker of norepinephrine transporter (NET) inhibition to assess atomoxetine effects during clinically recommended dosing. Proceedings of the 18th ECNP Congress; October 22–26, 2005; Amsterdam, The Netherlands. 3. Riordan HJ, Cutler NR, Irani F, et al. (2009) A meta-analysis of computerized assessment batteries in Alzheimer’s disease treatment trials. Alzheimers Dement. 5(4 Suppl 1):P457–8.

ind

BLBK249-Cutler

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index

Note: Italicized f and t refer to figures and tables. AbilifyTM , 41–2 ABT-200, 202–3 Accuracy of biomarker, 102t Acetylcholine (ACh), 106 Acetylcholinesterase (AChE), 107–8, 177–9, 219 Acute toxicity studies, 76–7 Adaptive design, 222–3 Adult isolation, 30 ␣-Synuclein, 112 Alzheimer’s disease, 42–3 aged animal models of, 43, 43t animal models of, 42–8 animal tests/genetic models, 44–5t biomarkers for, 106, 107t ␣-synuclein, 112 amyloid precursor protein, 112–3 amyloid-␤ peptide, 108–10 ␤-secretase, 111 CD-69, 116 cholinergic, 106–8 electroencephalography, 114–5 ␥ -secretase, 111–2 inflammatory biomarkers, 113–4 isoprostanes, 113 salivary amylase, 115–6 sulfatides, 113 tau protein, 110 bridging studies in, 196–200 characteristics of, 42–3 drugs for, 2–3 genetic markers, 116–7 genetic models of, 43–4t, 46–8 lesion-based models, 43t, 45–6 mortality rate, 3 pharmacological models of, 43–5, 43t prevalence of, 3 risk factors, 42–3 structural MRI, 170–71 254

Alzheimer’s Disease Neuroimaging Initiative (ADNI), 171–2 R Alzhemed ,3 AMES test, 79 Amphetamines, 32 Amyloid precursor protein (APP), 46–8, 112–3 Amyloid-␤ peptide, 46–8 biomarkers, 108–10 Anaclitic depression, 29–30 Analytical range of biomarker, 102t Analytical specificity of biomarker, 102t Analytical validation, 137–9 Angiotensin-converting enzyme inhibitors (ACEI), 83 Anhedonia, 28 Animal models, 14–49 of Alzheimer’s disease, 42–8 of anxiety, 16–24 elevated maze tests, 18–20, 18t fear-potentiated startle, 18t, 20 5-HT1A receptor knockout mouse, 23–4 genetic models, 22–4 inbred/selectively bred rodent strains, 22 light/dark transition test, 17–18 open field test, 17, 18t passive avoidance, 18t, 21 predator exposure tests, 18t, 20 punishment/conflict, 18t, 20–21 social interaction test, 18t, 21 of depression, 24–31 behavior despair models, 25–7, 26f chronic mild stress-induced anhedonia model, 26f, 28 learned helplessness, 26f, 27–8 olfactory bulbectomized rat, 26f, 28–9 social situation tests, 26f, 29–30 dose scaling to human, 88–90

Critical Pathways to Success in CNS Drug Development Neal R. Cutler, John J. Sramek, Michael F. Murphy, Henry Riordan, Peter Bieck and Angelico Carta © 2010 NR Cutler, JJ Sramek, MF Murphy, H Riordan, P Bieck and A Carta. ISBN: 978-1-444-33064-9

ind

BLBK249-Cutler

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index early drug development and, 48–9 of psychosis, 33t reliability of, 16 of schizophrenia, 31–42 selection algorithm, 16 validity of, 14–16 Animal studies, 72–3 Antidepressants, 25–6, 28 Antipsychotic drugs, 33–5t drug development, 3–4 Anxiety, 16–24 animal models of, 16–24 elevated maze tests, 18–20, 18t fear-potentiated startle, 18t, 20 5-HT1A receptor knockout mouse, 23–4 genetic models, 22–4 inbred/selectively bred rodent strains, 22 light/dark transition test, 17–18 open field test, 17, 18t passive avoidance, 18t, 21 predator exposure tests, 18t, 20 punishment/conflict, 18t, 20–21 social interaction test, 18t, 21 biomarkers for, 117t brain imaging, 121 catecholamines, 119 cholecystokinin, 120 corticotrophin-releasing factor (CRF), 118–9 cortisol, 118 glyoxalase-1, 119 lactate infusion, 119 saccadic peak velocity, 120–21 salivary amylase, 118 bridging studies in, 200–202 genetic models, 23–4 cholecystokinin2, 24 CRH-OE mice, 23 neurokinin 1, 24 RGS2, 24 Anxiolytics, 21–23 Aplenzin R , 5 ApoE protein, 117 Apomorphines, 34 App23 transgenic mouse, 46 Aricept R , 2 Aripiprazole, 41–2, 74 Assay linearity, 102t

255

Atomoxetine, 216–8 Auditory-evoked potentials (AEPs), 133 Azilect R , 5

Banzel R , 5 Bapineuzumab, 3 Behavior despair models, 25–7, 26f Behavioral models of psychosis, 39–40 Benzodiazepines, 19–21 ␤-Secretase, 111 Biomarker analysis, 250 Biomarkers, 101–40 for Alzheimer’s disease, 106, 107t ␣-synuclein, 112 amyloid precursor protein, 112–3 amyloid-␤ peptide, 108–10 ␤-secretase, 111 CD-69, 116 cholinergic, 106–8 electroencephalography, 114–5 ␥ -secretase, 111–2 inflammatory biomarkers, 113–4 isoprostanes, 113 salivary amylase, 115–6 sulfatides, 113 tau protein, 110 analytical validation of, 137–9 for anxiety disorder, 117t brain imaging, 121 catecholamines, 119 cholecystokinin, 120 corticotrophin-releasing factor (CRF), 118–9 cortisol, 118 glyoxalase-1, 119 lactate infusion, 119 saccadic peak velocity, 120–21 salivary amylase, 118 brain imaging, 103t categories of, 103t cell-based imaging, 103t central nervous system, 102–6 characteristics of, 101 Critical Path Initiative, 135–6 definition of, 102t for depression, 122t corticotrophin-releasing hormone (CRH), 123–4 cortisol, 123

ind

BLBK249-Cutler

256

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index

Biomarkers (cont.) dexamethasone, 123–4 dihydroxyphenylglycol (DHPG), 125 5-hydroxyindole-3-acetic acid (5-HIAA), 126–7 G proteins, 121–2 interleukin-6 (IL-6), 122–3 monoamine oxidase inhibitors (MAOIs), 125 norepinephrine, 124 3-methoxy-4-hydroxyphenylglycol (MHPG), 126 electrophysiological, 103t genetic, 103t immunological, 103t interference from matrix/endogenous analyte, 139–40 metabolomic, 103t neuroendocrine, 103t overview, 101 predictive, 137 proteomic, 103t provocative anxiety tests, 103t qualification, 136–7 reference standard, 139 for schizophrenia, 127–35 cerebrospinal fluid (CSF), 132–3 dopamine, 127–30 ethane exhalation, 134–5 glutamate, 131–2 neurophysiological signals, 132–3 serotonin, 130–31 small molecule, 103t surrogate markers, 101–2 validated, 138 validation terms, 102t Blood-brain barrier (BBB), 177 Brain imaging, 121 Brain imaging techniques, 103t Bridging studies, 187–224 adaptive design, 222–3 in Alzheimer’s disease, 196–200 SDZ ENA 713, 196–8 xanomeline, 198–200 in anxiety, 200–202 combined dynabridge and, 220, 221f continuous CSF sampling, 216–8 definition of, 192 in depression, 202–3

disease-modification therapy, 223 drug response, 189–91 dynabridge, 215–6 atomoxetine, 218 rivastigmine, 218–20 methodology, 192–6 overview, 187 period I, 194 period II, 194 period III, 195–6 pharmacokinetic/pharmacodynamic relationship studies, 214–5 phase 1 studies, 187 phase 2 studies, 188 phase 3 studies, 188 phase 4 studies, 188–9 positron emission tomography and, 221–2 in schizophrenia, 204–13 CI-1007, 211–3 fananserin, 209–11 iloperidone, 204–9 titration strategies, 213–4 Bristol-Myers Squibb, 41 Bupropion hydrobromide, 5 R BuSpar , 191 Buspirone, 191 Butylcholinesterase (BuChE), 178–9, 219 Calibration of biomarker, 102t Carcinogen, 79 Carcinogenicity studies, 79–80 Cardiac electrophysiology, 84–5 Cardiorespiratory system, 232–3 Case study, in CNS drug development, 230–53 clinical studies, 237–51 multiple-dose study, 247–51 single-ascending dose, 237–41 single-ascending dose to multiple-ascending dose, 241–7 drug disposition data, 234–5 pharmacokinetics data, 234–5 preclinical summary, 230–31 safety pharmacology, 232–4 cardiorespiratory system, 232–3 central nervous system, 232 gastrointestinal system, 233–4 renal system, 233–4 toxicology data, 235–7

ind

BLBK249-Cutler

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index in vitro, primary pharmacology, 231 in vivo, primary pharmacology, 231–2 Catecholamines, 105f, 119 CD28-SuperMAB, 90 CD-69, 116 Cell-based imaging, 103t Center for Biologics Evaluation and Research (CBER), 79 Center for Drug Evaluation and Research (CDER), 79, 88 Central nervous system (CNS), 232 Central nervous system (CNS) biomarkers, 102–6 Cerebrospinal fluid (CSF) amyloid-␤ peptide in, 108–9 cognitive testing and, 177–8 continuous sampling, 216–8 metabolic profile, 132–3 Chantix R , 5 Cholecystokinin (CCK), 120 Cholecystokinin tetrapeptide (CCK-4), 119–20 Cholecystokinin2 (CCK2), 24 Cholinergic markers, 106–8 Cholinesterase inhibitors, 2, 177 Chronic depression, 29 Chronic dose repeated toxicity tests, 77–8 Chronic mild stress-induced anhedonia model, 26f, 28 CI-1007, 205t, 211–3 CI-979, 197t Clinical Antipsychotic Trial of Intervention Effectiveness (CATIE) study, 3–4 Clinical endpoint, 102t Clinical studies, 237–51 multiple-dose study, 247–51 single-ascending dose to multiple-ascending dose, 241–7 single-dose, ascending, sequential cohort design, 237–41 Clinical trials, cost of, 6 Clozapine, 190 Clozaril R , 190 CNS drug development, 1–12, 230–53 case study, 230–53 clinical studies, 237–51 drug disposition data, 234–5 multiple-dose study, 247–51 pharmacokinetics data, 234–5

257

preclinical summary, 230–31 primary pharmacology, 231–2 safety pharmacology, 232–4 single-ascending dose study, 237–41 single-ascending dose to multiple-ascending dose study, 241–7 toxicology data, 235–7 difficulties in, 6–9 dose–response curves, 191f drug approvals and, 6 new molecular entities (NMEs), 5 outlook in, 9–12 overview, 1–2 stagnation in, 2–6 study types in, 224t Cognitive assessments, 251 Cognitive testing, 177–81 Computerized Neuropsychological Test Battery (CNTB), 219 Condition conflict test, 21 Conditioned avoidance response (CAR), 34t, 40 Conditioned conflict test, 21 Conditioned fear, 19 Conditioned stimulus (CS), 40 Conflicts, 20–21 Construct validity, 15 Convergent validity, 15–16 Corticotrophin-releasing factor (CRF), 118–9 Corticotrophin-releasing hormone (CRH), 123–4 Cortisol, 118, 123 CRH-OE mice, 19t, 23 Critical Path Initiative, 135–6 CRND8 transgenic mouse, 46 R Cymbalta ,5 Defeat, 29 Dementia with Lewy bodies (DLB), 112 Depression, 29–30 anaclitic, 29–30 animal models of, 24–31 behavior despair models, 25–7, 26f chronic mild stress-induced anhedonia model, 26f, 28 learned helplessness, 26f, 27–8 olfactory bulbectomized rat, 26f, 28–9 social situation tests, 26f, 29–30

ind

BLBK249-Cutler

258

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index

Depression (cont.) biomarkers for, 122t corticotrophin-releasing hormone (CRH), 123–4 cortisol, 123 dexamethasone, 123–4 dihydroxyphenylglycol (DHPG), 125 5-hydroxyindole-3-acetic acid (5-HIAA), 126–7 G proteins, 121–2 interleukin-6 (IL-6), 122–3 monoamine oxidase inhibitors (MAOIs), 125 norepinephrine, 124 3-methoxy-4-hydroxyphenylglycol (MHPG), 126 bridging studies in, 202–3 chronic, 29 genetic models of, 30–31 Developmental injury models, 38–9 Developmental toxicity, 81–3 Dexamethasone, 123–4 Difluprednate, 5 Dihydroxyphenylacetic acid (DOPAC), 129 Dihydroxyphenylglycol (DHPG), 125 Dimethyl tryptamine (DMT), 35 Discriminant validity, 15–16 Disease modifiers, 2 Disease-modification therapy, 223 Disrupted in schizophrenia 1 (DISC1), 41 Donepezil, 2, 108 Dopamine, 127–30 Dopaminergic-induced psychosis, 32–5, 33t Dose scaling, 88–90 Dosing multiple-dose study, 250 single-ascending dose study, 240 single-ascending dose to multiple-ascending dose study, 246 Drinking behavior, 21 Drosophila melanogaster, 48 Drug approvals, 1, 6 Drug candidate selection, 70 Drug development, 230–53 clinical studies, 237–51 multiple-dose study, 247–51 single-ascending dose, 237–41 single-ascending dose to multiple-ascending dose, 241–7

drug disposition data, 234–5 pharmacokinetics data, 234–5 preclinical summary, 230–31 safety pharmacology, 232–4 cardiorespiratory system, 232–3 central nervous system, 232 gastrointestinal system, 233–4 renal system, 233–4 toxicology data, 235–7 in vitro, primary pharmacology, 231 in vivo, primary pharmacology, 231–2 Drug disposition data, 234–5 Drug response, 189–91 Duloxetine HCL, 5 R Durezol ,5 Dynabridge studies, 216–20. See also bridging studies atomoxetine, 216–8 combined bridging and, 220, 221f rivastigmine, 218–20 Dynamic range of biomarker, 102t Electrocardiograph analysis, 240, 250–51 Electroencephalograms, 84–5 Electroencephalography (EEG), 114–5, 176 Electrophysiological marker, 103t Elevated maze tests, 18–20, 18t Endpoint, 102t Enzyme-linked immunosorbent assay (ELISA), 138 Eptastigmine, 197t Ethane exhalation, 134–5 European Medicines Agency (EMEA), 2 Event-related potentials (ERPs), 133–4 Exelon R , 2 Eye-pursuit abnormalities, 134 Face validity, 15 Fananserin, 205t, 209–11 Fast Track program, 7 Fawn-Hooded rats, 31 Fear-potentiated startle (FPS), 18t, 20 Fialuridine, 78 5-HT1A receptor knockout mouse, 19t, 23–4 5-Hydroxyindole-3-acetic acid (5-HIAA), 126–7 5-Hydroxytryptophan (5-HTP), 123 Fixed dose procedure (FDP), 77 Flinders sensitive line (FSL), 30

ind

BLBK249-Cutler

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index Fluorodeoxyglucose, 172 Fluoxetine, 191 Flurizan R , 3 Food, Drug, and Cosmetic Act in 1938, 70 Food and Drug Administration (FDA), 5, 7, 88 Forced-swim test, 25–6, 26f Fospropofol disodium, 5 Functional MRI, 168–9 G proteins, 121–2 GABAergic-induced psychosis, 33t, 37 Galantamine, 2 ␥ -Aminobutyric acid (GABA), 37 ␥ -Secretase, 111–2 ␥ -Secretase inhibitor, 223 Gastrointestinal system, 233–4 Generalized anxiety disorder (GAD), 118, 190 Genetic markers, 103t, 116–7 Genetic models. See also animal models of Alzheimer’s disease, 46–8 of anxiety, 23–4 cholecystokinin2, 24 CRH-OE mice, 23 neurokinin 1, 24 RGS2, 24 of depression, 30–31 of psychosis, 35t, 40–41 Genotoxicity studies, 78–9 Geodon R , 3 Glutamate, 131–2 Glutamate-metabolizing enzymes (GMEs), 131–2 Glutamatergic-induced psychosis, 33t, 36–7 Glutamic acid decarboxylase (GAD), 37 Glyoxalase-1, 119 Haloperidol, 33t Hamilton Rating Scale for Anxiety (HAM-A), 201 Helpless (HL) mice, 30 Human equivalent dose (HED), 89 Hyperphosphorylated tau (p-tau), 110 Hypothalamus–pituitary–adrenocortical (HPA) activity, 123–4 Iloperidone, 204–9, 205t Imaging techniques, 167–83

259

cognitive testing, 177–81 electroencephalography, 176 functional MRI, 168–9 MATRICS program, 181–2 overview, 167–8 PET imaging, 172–6 pharmaco-EEG, 176–7 structural MRI, 169–72 Imipramine, 30 Immunological marker, 103t Immunotoxicity, 85–6 In vitro primary pharmacology, 231 In vitro studies, 72–3 In vivo genotoxicity tests, 79 In vivo primary pharmacology, 231–2 Inbred/selectively bred rodent strains, 22 Inflammatory biomarkers, 113–4 Interleukin factor (IF), 113 Interleukin-6 (IL-6), 122–3 Invega R , 3 Investigational new drug (IND), 72, 75–6 Isolation, 29–30 Isoprostanes, 113 JNPL3 transgenic mice, 45t, 47–8 Lacosamide, 5 Lactate infusion, 119 Latent inhibition (LI), 33t, 39 Learned helplessness, 26f, 27–8 Lesion-induced psychosis, 33t, 37–8 Lesopitron, 200–202 Lethal dose (LD50 ), 76–7 Light/dark transition test, 17–18 Lisdexamfetamine, 5 Lu 25-109, 197t Lusedra R , 5 LY450139 dihydrate, 110, 223 Lysergic acid diethylamide (LSD), 35–6 Magnetic resonance imaging (MRI), 168–72 functional, 168–9 structural, 169–72 Major depressive disorders (MDDs), 122 Mammalian cell system test, 79 Maximally tolerated dose (MTD), 80, 193–5 Maximum recommended starting dose (MRSD), 88–9 MDL 100,907, 205t

ind

BLBK249-Cutler

260

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index

Measurement and Treatment Research to Improve Cognition in Schizophrenia (MATRICS), 180–82 Memantine, 2 Mescaline, 35–6 Metabolomic markers, 103t Metrifonate, 197t Microdosing, 174–6 Minimally intolerable dose (MID), 192 Mini-Mental State Examination (MMSE), 115, 171 Mismatch negativity, 133–4 Monoamine oxidase inhibitors (MAOIs), 124–5 Mouse defense test battery (MDTB), 20 Multiple-dose study, 247–51 biomarker analysis, 250 cognitive assessments, 251 design, 248–50 dosing, 250 electrocardiograph analysis, 250–51 objectives in, 247 pharmacokinetic analysis, 250 rationale for, 247–8 subjects, 248 Muscarinic acetylcholine receptor (mAChR) agonists, 115 N100 generation, 133–4 Namenda R , 2 National Institute of Aging, 171 National Institute of Bioimaging and Bioengineering, 172 Negative predictive value, 102t Nemeroff, Charles B., 7–8 Neonatal ventral hippocampal lesion (NVHL) model, 33t, 38 Neupro R , 5 Neuregulin 1 (NRG1), 41 Neuroendocrine marker, 103t Neurofibrillary tangles (NFTs), 46 Neuroimaging, 138, 167–83 cognitive testing, 177–81 electroencephalography, 176 functional MRI, 168–9 MATRICS program, 181–2 overview, 167–8 PET imaging, 172–6

pharmaco-EEG, 176–7 structural MRI, 169–2 Neurokinin 1 (NK1R), 24 Neuroleptics, 33t Neuromodulators, 16 Neurophysiological signals, 132–3 Neurotoxicity, 86–8 Neurotransmitter norepinephrine (NE), 124–5 Neurotransmitters, 16 New drug development, 230–53 clinical studies, 237–51 multiple-dose study, 247–51 single-ascending dose, 237–41 single-ascending dose to multiple-ascending dose, 241–7 drug disposition data, 234–5 pharmacokinetics data, 234–5 preclinical summary, 230–31 safety pharmacology, 232–4 cardiorespiratory system, 232–3 central nervous system, 232 gastrointestinal system, 233–4 renal system, 233–4 toxicology data, 235–7 in vitro, primary pharmacology, 231 in vivo, primary pharmacology, 231–2 New molecular entities (NMEs), 5 NGX267, 222 N-methyl-D-aspartic acid (NMDA), 2, 36 No observed adverse effects (NOAEL), 75–6, 88–9 Non-steroidal anti-inflammatory drugs (NSAIDs), 114 No observed effects (NOEL), 76 Norepinephrine, 124 Olanzapine, 3–4 Olfactory bulbectomized (OB) rate, 26f, 28 Open field test, 17, 18t Otsuka Japan, 41 P301L transgenic mouse, 47 P50 gating, 134 Paliperidone, 3 Panic attacks, 119–20 Parkinson’s disease, 112 Passive avoidance, 18t, 21 Perphenazine, 3–4

ind

BLBK249-Cutler

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index Pharmaco-EEG, 241 Pharmacokinetic analysis, 250 Pharmacokinetic/pharmacodynamic relationships, 214–5 Pharmacokinetics, 71t, 240 Pharmacokinetics data, 234–5 Pharmacologically active dose (PAD), 90 Pharmacology, 71t Phencyclidine (PCP), 35–7, 41 Pittsburgh Compound-B (PIB), 109 Platelet-derived growth factor-␤ (PDGF-␤), 46 Population-specific toxicity, 83–4 Porsolt test, 25–6 Positive predictive value, 102t Positron emission tomography (PET), 172–6, 221–2 amyloid load, 173–4 fluorodeoxyglucose, 172 microdosing, 174–6 uses of, 173 Post-marketing surveillance (PMS), 188 Practice effects, 179 Precision of biomarker, 102t Preclinical summary, 230–31 Preclinical testing, 70–90 acute toxicity studies, 76–7 carcinogenicity studies, 79–80 cardiac electrophysiology, 84–5 chronic dose repeated toxicity tests, 77–8 developmental toxicity, 81–3 dose scaling, 88–90 dose–response curves, 73–5 examples of, 71t genotoxicity studies, 78–9 immunotoxicity, 85–6 investigational new drug, 72, 75–6 neurotoxicity, 86–8 overview, 71–2 reproductive toxicity, 81–3 requirements, 70 safety pharmacology studies, 84 seizures, 88 short-term repeated toxicity tests, 77–8 subchronic repeated toxicity tests, 77–8 therapeutic implications, 90 toxicology testing, 72–3

261

Predator exposure tests, 18t, 20 Predictive biomarkers, 137 Predictive validity, 15 Prefrontal cortex (PFC), 38 Prepulse inhibition, 34t, 133–4 Pre-pulse inhibition (PPI), 40 Primary pharmacology, 231–2 Prior probability, 102t Priority Review program, 7 Proteomic markers, 103t Provocative anxiety tests, 103t R Prozac , 191 Pseudospecificity, 181–2 Psychosis, 32–6 animal models of, 33t behavioral models of, 39–40 dopaminergic-induced, 32–5, 33t GABAergic-induced, 33t, 37 genetic models of, 35t, 40–41 glutamatergic-induced, 33t, 36–7 lesion-induced, 33t, 37 serotonergic-induced, 33t, 35–6 Punishment/conflict models, 18t, 20–21 QT interval, 84–5 Quality of episodic memory (QESM), 178–9 Quantitative electroencephalography (QEEG), 176–7 Quetiapine, 3–4 Rasagiline, 5 Razadyne R , 2 Reference standard, 139 Reliability of animal models, 16 Reminyl R , 2 Renal system, 233–4 Repeat-dose administration, 236 Reproducible range of biomarker, 102t Reproductive toxicity, 81–3 RGS2 knockout mice, 19t, 24 Risperidone, 3–4 Rivastigmine, 2, 177–8, 218–20 Rotigotine, 5 Rufinamide, 5 Ruggedness of biomarker, 102t Saccadic peak velocity (SPV), 120–21 Saccharin, 80 Safety pharmacology

ind

BLBK249-Cutler

262

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index

Safety pharmacology studies, 84 cardiorespiratory system, 232–3 central nervous system, 232 gastrointestinal system, 233–4 in new drug development, 232–4 preclinical data, 71t renal system, 233–4 Salivary amylase, 115–6, 118 Sarcosine, 33t Schizophrenia, 31–42 animal models of, 31–42 dopaminergic-induced psychosis, 32–5, 33t GABAergic-induced psychosis, 37 glutamatergic-induced psychosis, 36–7 lesion-induced psychosis, 37–8 serotonergic-induced psychosis, 35–6 bridging studies in, 204–13 developmental injury models, 38–9 drugs for, 3–4 negative symptoms of, 32 positive symptoms of, 32 Scopolamine, 177 SDZ ENA 713, 196–8 SDZ ENS-163, 177 Seizures, 86–8 Selection algorithm, 16 Sensitivity of biomarker, 102t Serotonergic-induced psychosis, 33t, 35–6 Serotonin, 35–6, 130–31 Sertindole, 205t Short-term repeated toxicity tests, 77–8 Single-ascending-dose study, 237–41 design, 239–40 dosing, 240 electrocardiograph analysis, 240 objectives in, 237–8 pharmaco-EEG analysis, 241 pharmacokinetics, 240 safety analysis, 241 subjects, 239 vital signs, 241 Single-ascending dose to multiple-ascending dose study, 241–7 design, 243–5 dosing, 246 electrocardiograph analysis, 246 monitoring, 245 multiple-dose phase, 246

objectives in, 242 pharmaco-EEG analysis, 247 rationale for, 242–3 safety analysis, 247 single-dose phase, 245 subjects, 243 vital signs, 246 Single-dose administration, 235–6 Small molecule marker, 103t Social defeat model, 29 Social hierarchy model, 29 Social interaction test, 18t, 21 Social situation tests, 26f, 29–30 Specificity of biomarker, 102t Sponsor’s risk, 16 Standard curve of biomarker, 102t Startle reflex, 20 State Trait Anxiety Index Inventory (STAI), 118 StavzorTM , 5 Strattera R , 218 Structural MRI, 169–72 Subchronic repeated toxicity tests, 77–8 Substance P, 24 Sulfatides, 113 Surrogate markers, 101–2. See also biomarkers Synuclein, 112 System-specific toxicity, 83–4 Tail suspension test, 26–7, 26f Tau protein, 108–10 Tau/APP double transgenic mice, 45t, 46–8 Tetrabenazine, 5 Tg2576-Hsiao transgenic mouse, 46 TGN15412 antibody, 90 Therapeutic index, 74 3-Amino-1-propanesulfonic acid (3APS), 109 3,4-Dihydroxyphenylacetic acid (DOPAC), 119 3-Methoxy-4-hydroxyphenylglycol (MHPG), 126 Toxicology, 71t, 72 Toxicology data, 235–7 Toxicology testing, 72–3 acute toxicity studies, 76–7 carcinogenicity studies, 79–80 chronic dose repeated toxicity tests, 77–8

ind

BLBK249-Cutler

Trim: 6in X 9in

January 13, 2010

18:11

Char Count=

Index developmental toxicity, 81–3 genotoxicity studies, 78–9 immunotoxicity, 85–6 neurotoxicity, 86–8 population-specific toxicity, 83–4 reproductive toxicity, 81–3 short-term repeated toxicity tests, 77–8 subchronic repeated toxicity tests, 77–8 system-specific toxicity, 83–4 Transgenic mice, 43–4t, 46–8 Trier Social Stress Test (TSST), 118 Tumor necrosis factor-␣ (TNF-␣), 113 Validated biomarkers, 138 Validity of animal models, 14–16

Valproic acid, 5 Varenicline, 5 Velnacrine, 197t Vimpat R , 5 Vital signs, 241 Vogel conflict test, 21 Vyvanse R , 5 WAG/Rij mice, 31 Xanomeline, 197t, 198–200 Xenazine R , 5

Zeldox R , 3 Ziprasidone, 3–4 ZNS 114-666, 219

263