The Process of New Drug Discovery and Development

The Process of New Drug Discovery and Development Second Edition

Edited by

Charles G. Smith Ph.D. James T. O’Donnell Pharm.D.

Informa Healthcare USA, Inc. 270 Madison Avenue New York, NY 10016 © 2006 by Informa Healthcare USA, Inc. Informa Healthcare is an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-10: 0-8493-2779-2 (Hardcover) International Standard Book Number-13: 978-0-8493-2779-7 (Hardcover) Library of Congress Card Number 2005036791 This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http:// www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data The process of new drug discovery and development / editors, Charles G. Smith and James T. O’Donnell.-- 2nd ed. p. ; cm. Rev. ed. of: The process of new drug discovery and development / Charles G. Smith. c1992. Includes bibliographical references and index. ISBN-13: 978-0-8493-2779-7 (alk. paper) ISBN-10: 0-8493-2779-2 (alk. paper) 1. Drugs--Research--History. 2. Drugs--Design--History. [DNLM: 1. Drug Design. 2. Drug Evaluation, Preclinical. 3. Drug Evaluation. QV 744 P9645 2006] I. Smith, Charles G. (Charles Giles) II. O’Donnell, James T., Pharm. D. RM301.25.S55 2006 615’.19--dc22 Visit the Informa Web site at www.informa.com and the Informa Healthcare Web site at www.informahealthcare.com

2005036791

As was the case with the first edition, this book is dedicated to my irreplaceable wife, Angeline, who persisted in encouraging me in this and other scientific efforts that often left little time for recreational activities. In addition, I dedicate it to my daughter, Tracy, whose strong encouragement and occasional harassment convinced me to write the first edition. Charles G. Smith

To my wife, Sylvia, and my children, Kimberly and Jim, who make my life worth living. James T. O’Donnell

Acknowledgments

Sincere appreciation is expressed to Mr. S. Zollo at CRC Press/Taylor & Francis, whose strong solicitation resulted in my decision to participate in the second edition of this book. The role of my coeditor, Dr. James T. O’Donnell, without whose expertise and organizational abilities this work could not possibly have been accomplished, deserves my most heartfelt thanks and appreciation. The participation of many excellent authors has made this book, in my opinion, a truly valuable contribution to the scientific literature in the field of drug discovery and development. I am honored to end a most enjoyable career with this contribution. Charles G. Smith

Editors

Charles G. Smith earned the B.S. degree (1950) in chemistry from the Illinois Institute of Technology, the M.A. degree (1952) in biochemistry from Purdue University, and the Ph.D. degree (1954) in biochemistry from the University of Wisconsin. He joined The Upjohn Company in 1954 and worked in fermentation biochemistry for several years. In 1962, he was appointed head of the biochemistry department at Upjohn and initiated a major anticancer effort therein. Dr. Smith moved to E. R. Squibb & Sons pharmaceutical company in 1968, where he became vice president for research. He joined the Revlon Health Care Group in 1975 as vice president for research and development. In 1986, he retired from industry and became a pharmaceutical consultant. During his tenure in the major pharmaceutical companies, Dr. Smith was intimately involved with projects in the fields of infectious diseases, cancer, cardiovascular diseases, central nervous system diseases, and pharmaceutical products from blood plasma. Since 1986, he has consulted with many biotechnology companies that work in a broad cross-section of pharmaceutical research. He was a cofounder of Vanguard Medica in the United Kingdom in 1991 and named adjunct professor in the Department of Natural Sciences at San Diego State University in the same year. Dr. Smith is the author of 49 publications and the first edition of this book (1992), and remains a pharmaceutical consultant in the biotechnology field. James T. O’Donnell earned the B.S. degree in pharmacy from the University of Illinois (1969) and the Doctor of Pharmacy degree from the University of Michigan (1971) as well as the M.S. degree in clinical nutrition from Rush University (1982). He completed a residency in clinical pharmacy at the University of Illinois Research Hospitals and has been a registered pharmacist in Illinois since 1969. Dr. O’Donnell spent 17 years in clinical practice at both the Cook County Hospital and the Rush University Medical Center in Chicago. Dr. O’Donnell is currently an associate professor of pharmacology at the Rush University Medical Center and is a member of the Graduate College, involved in the teaching of new drug development and regulations. Also, Dr. O’Donnell is a lecturer in the Department of Medicine at the University of Illinois College of Medicine, Rockford. He is a Diplomate of the American Board of Clinical Pharmacology and Board of Nutritional Specialties, a fellow of the American College of Clinical Pharmacology and the American College of Nutrition, and a member of several professional societies. Dr. O’Donnell is the author of 257 publications, and he is the founding editor of the Journal of Pharmacy Practice, a co-editor of Pharmacy Law, and the editor of Drug Injury: Liability, Analysis, and Prevention, First and Second Editions. In addition to his academic and editorial endeavors, Dr. O’Donnell regularly consults in drug and pharmaceutical matters to industry, government, and law, and serves as pharmacologist consultant to the State of Illinois Department of Public Health.

Contributors

University of Oklahoma College of Pharmacy, Edmond, Oklahoma, U.S.

Loyd V. Allen

Donald C. Anderson Pharmacia Corporation, Kalamazoo, Michigan, U.S. Timothy Anderson Pfizer, Inc., Groton, Connecticut, U.S. Stephen Barrett

Quackwatch, Inc., Allentown, Pennsylvania, U.S.

Joanne Bowes Safety Assessment UK, AstraZeneca R&D Alderley Park, Macclesfield, Cheshire, U.K. Irwin A. Braude Celia Brazell

Compass Pharmaceuticals, LLC, Charlotte, North Carolina, U.S.

Genetics Research, GlaxoSmithKline, Greenford, Middlesex, U.K.

Jerry Collins Food and Drug Administration, Center for Drug Evaluation and Research, Office of Testing and Research, Rockville, Maryland, U.S. William T. Comer

TorreyPines Therapeutics, Inc., La Jolla, California, U.S.

Mark Crawford

Cerep, Redmond, Washington, U.S.

Andrew Dorner

Wyeth Research, Andover, Maryland, U.S.

David Essayan Food and Drug Administration, Center for Biologics Evaluation and Research, Office of Therapeutics, Division of Clinical Trial Design and Analysis, Rockville, Maryland, U.S. Pauline Gee Department of Predictive Biology, MDS Pharma Services, Bothell, Washington, U.S. Baltazar Gomez-Mancilla

Pharmacia Corporation, Kalamazoo, Michigan, U.S.

Henry Grabowski Duke University, Durham, North Carolina, U.S. Howard E. Greene Rancho Santa Fe, CA Joseph Hackett Food and Drug Administration, Center for Devices and Radiological Health, Office of In Vitro Diagnostic Device Evaluation and Safety, Rockville, Maryland, U.S.

Jeff M. Hall

Department of Cell Biology, Genoptix, Inc., San Diego, California, U.S.

Tim G. Hammond Safety Assessment UK, AstraZeneca R&D Alderley Park, Macclesfield, Cheshire, U.K. Valérie Hamon

Cerep, Paris, France

Ismael J. Hidalgo

Absorption Systems, LP, Exton, Pennsylvania, U.S.

Carol Ann hom*on Boehringer Ingelheim Pharmaceuticals, Inc., Medicinal Chemistry Department, Ridgefield, Connecticut, U.S. Shiew-Mei Huang Food and Drug Administration, Center for Drug Evaluation and Research, Office of Clinical Pharmacology and Biopharmaceutics, Rockville, Maryland, U.S. Susan Ide National Institute of Health, NHGRI, and Novartis, Gaithersburg, Maryland, U.S. Vincent Idemyor Thierry Jean

College of Medicine, University of Illinois, Chicago, Illinois, U.S.

Cerep, Paris, France

Ilona Kariv Applications Research and Development, Genoptix Inc., San Diego, California, U.S. Joanne Killinger

Wyeth Research, Chazy, New York, U.S.

Harold J. Kwalwasser

MDS Pharma Services, Bothell, Washington, U.S.

John Leighton Food and Drug Administration, Center for Drug Evaluation and Research, Office of New Drugs, Rockville, Maryland, U.S. Lawrence J. Lesko Food and Drug Administration, Center for Drug Evaluation and Research, Office of Clinical Pharmacology and Biopharmaceutics, Rockville, Maryland, U.S. Jibin Li

Absorption Systems, LP, Exton, Pennsylvania, U.S.

Elizabeth Mansfield Food and Drug Administration, Center for Devices and Radiological Health, Office of In Vitro Diagnostic Device Evaluation and Safety, Rockville, Maryland, U.S. Phillip J. Marchand Patricia A. McNeeley

Optical Systems, Genoptix Inc., San Diego, California, U.S. Genoptix Inc., San Diego, California, U.S.

Robert Meyer Food and Drug Administration, Center for Drug Evaluation and Research, Office of New Drugs, Rockville, Maryland, U.S. Justina A. Molzon U.S. Food and Drug Administration, Center for Drug Evaluation and Research, Rockville, Maryland, U.S.

Daniel Mufson

Apotherx LLC, Napa, California, U.S.

Richard M. Nelson Boehringer Ingelheim Pharmaceuticals, Inc., Medicinal Chemistry Department, Ridgefield, Connecticut, U.S. Lori Nesbitt

Discovery Alliance International, Inc., Sugarland, Texas, U.S.

Sarfaraz K. Niazi

Pharmaceutical Scientists, Inc., Deerfield, Illinois, U.S.

F. Richard Nichol U.S.

Nichol Clinical Technologies Corporation, Newport Beach, California,

Tina S. Nova

Genoptix Inc., San Diego, California, U.S.

James T. O’Donnell Department of Pharmacology, Rush Medical College, Rush University Medical Center, Chicago, IL Mark Paich Lexidyne, LLC, Colorado Springs, Colorado, U.S. Corey Peck Lexidyne, LLC, Colorado Springs, Colorado, U.S. Madhu Pudipeddi Technical Research and Development, Novartis Pharmaceuticals Corporation, Mumbai, India Michael H. Rabinowitz Johnson & Johnson Pharmaceutical Research & Development LLC, San Diego, California, U.S. Mitchell E. Reff Biogen Idec, Inc., Oncology, Cambridge, Massachusetts, U.S. Frederick E. Reno

Reno Associates, Merritt Island, Florida, U.S.

Michael G. Rolf Safety Assessment UK, AstraZeneca R&D Alderley Park, Macclesfield, Cheshire, U.K. Stephen G. Ryan

AstraZeneca Pharmaceuticals, Wilmington, Delaware, U.S.

Ronald A. Salerno WorldWide Regulatory Affairs, Wyeth Research, St. Davids, Pennsylvania, U.S. Richard H.C. San Genetic Toxicology, BioReliance, Rockville, Maryland, U.S. Virginia Schmith

GlaxoSmithKline, Research Triangle Park, North Carolina, U.S.

Abu T.M. Serajuddin Novartis Pharmaceutical Corporation, Pharmaceutical and Analytical Development, East Hanover, New Jersey, U.S. Nigel Shankley Johnson & Johnson Pharmaceutical Research & Development, LLC, Merryfield Row, San Diego, California, U.S. Peter Shaw Department of Pharmacogenomics and Human Genetics, Bristol-Meyers Squibb, Princeton, New Jersey, U.S.

Frank Sistare Food and Drug Administration, Center for Drug Evaluation and Research, Office of Testing and Research, Rockville, Maryland, U.S. Charles G. Smith Rancho Santa Fe, CA Kirk Solo

Lexidyne, LLC, Colorado Springs, Colorado, U.S.

John C. Somberg Chief Division of Clinical Pharmacology, Department of Medicine & Pharmacology, Rush University, Chicago, Illinois, U.S. Brian B. Spear Illinois, U.S. Jason Valant

Abbott Laboratories, Department of Pharmacogenetics, Abbott Park,

Lexidyne, LLC, Colorado Springs, Colorado, U.S.

Jean-Pierre Valentin Safety Assessment UK, AstraZeneca R&D Alderley Park, Macclesfield, Cheshire, U.K. Gönül Veliçelebi Research and Drug Discovery, TorreyPines Therapeutics, Inc., La Jolla, California, U.S. Daniel D. Von Hoff Translational Genomics Institute, Phoenix, Arizona, U.S. Heather L. Wallace Scientific Communications, ICON Clinical Research, North Wales, Pennsylvania Mark Watson Clinical Genomics, Merck and Co., Inc., West Point, Pennsylvania, U.S. Janet Woodco*ck U.S. Food and Drug Administration, Center of Drug Evaluation and Research, Rockville, Maryland, U.S. Alexandra Worobec Food and Drug Administration, Center for Biologics Evaluation and Research, Office of Therapeutics, Division of Clinical Trial Design and Analysis, Rockville, Maryland, U.S.

Contents

1. Introduction ..........................................................................................................................1 Charles G. Smith

Section I General Overview 2. Overview of the Current Process of New Drug Discovery and Development ........................................................................................................................7 Charles G. Smith and James T. O’Donnell 3. Integrated Drug Product Development — From Lead Candidate Selection to Life-Cycle Management ...................................................................................................15 Madhu Pudipeddi, Abu T.M. Serajuddin, and Daniel Mufson

Section II Scientific Discoveries Application in New Drug Development 4. The Impact of Combinatorial Chemistry on Drug Discovery .................................55 Michael H. Rabinowitz and Nigel Shankley 5. High-Throughput Screening: Enabling and Influencing the Process of Drug Discovery ..............................................................................................79 Carol Ann hom*on and Richard M. Nelson 6. Pharmacological and Pharmaceutical Profiling: New Trends ................................103 Joanne Bowes, Michael G. Rolf, Jean-Pierre Valentin, Valérie Hamon, Mark Crawford, and Thierry Jean 7. Cell-Based Analysis of Drug Response Using Moving Optical Gradient Fields ...............................................................................................................135 Jeff M. Hall, Ilona Kariv, Patricia A. McNeeley, Phillip J. Marchand, and Tina S. Nova 8. Patient-Derived Primary Cells in High-Throughput Differential Antitumor Screens: Let the Patients Be the Guide...................................................149 Irwin A. Braude 9. The Evolving Role of the Caco-2 Cell Model to Estimate Intestinal Absorption Potential and Elucidate Transport Mechanisms..................................161 Jibin Li and Ismael J. Hidalgo 10. The Promise of Metabonomics in Drug Discovery..................................................187 Harold J. Kwalwasser and Pauline Gee

11. Pharmacogenetics and Pharmacogenomics in Drug Development and Regulatory Decision-Making: Report of the First FDA–PWG– PhRMA–DruSafe Workshop .........................................................................................199 Lawrence J. Lesko, Ronald A. Salerno, Brian B. Spear, Donald C. Anderson, Timothy Anderson, Celia Brazell, Jerry Collins, Andrew Dorner, David Essayan, Baltazar Gomez-Mancilla, Joseph Hackett, Shiew-Mei Huang, Susan Ide, Joanne Killinger, John Leighton, Elizabeth Mansfield, Robert Meyer, Stephen G. Ryan, Virginia Schmith, Peter Shaw, Frank Sistare, Mark Watson, and Alexandra Worobec 12. Drugs from Molecular Targets for CNS and Neurodegenerative Diseases ........225 William T. Comer and Gönül Veliçelebi 13. Safety Pharmacology: Past, Present, and Future .......................................................243 Jean-Pierre Valentin and Tim G. Hammond 14. Nonclinical Drug Safety Assessment..........................................................................291 Frederick E. Reno 15. Preclinical Genotoxicity Testing — Past, Present, and Future ...............................305 Richard H.C. San

Section III Standard Drug Developmental Issues: Updated 16. The Need for Animals in Biomedical Research........................................................315 Charles G. Smith 17. Defining the Actual Research Approach to the New Drug Substance ................329 Charles G. Smith 18. Pharmaco*kinetics –– Pharmacodynamics in New Drug Development................335 Sarfaraz K. Niazi 19. Pharmaceutics and Compounding Issues in New Drug Development and Marketing........................................................................................377 Loyd V. Allen 20. Late Stage and Process Development Activities .......................................................401 Charles G. Smith

Section IV Clinical Development 21. Contract Research Organizations: Role and Function in New Drug Development ....................................................................................................................407 F. Richard Nichol 22. The Front Lines of Clinical Research: The Industry ................................................419 Lori Nesbitt 23. Horizons for Cancer Chemotherapy (and Nonchemotherapy) ..............................445 Daniel D. Von Hoff

24. Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome: Clinical Testing Challenges......................................................................459 Vincent Idemyor

Section V Regulatory and Legal Issues Affecting Drug Development 25. Common Technical Document: The Changing Face of the New Drug Application ...................................................................................................473 Justina A. Molzon 26. Electronic Publishing......................................................................................................481 Heather L. Wallace 27. The Important Role of Pharmacists in a Complex Risk-Management System: Managing the Risks from Medical Product Use by Focusing on Patient Education, Monitoring, and Adverse Event Reporting ......................483 Justina A. Molzon 28. Liability, Litigation, and Lessons in New Drug Development..............................489 James T. O’Donnell 29. Problems in the Nondrug Marketplace ......................................................................521 Stephen Barrett 30. Patents and New Product Development in the Pharmaceutical and Biotechnology Industries ...............................................................................................533 Henry Grabowski 31. The Pharmaceutical Revolution: Drug Discovery and Development ..................547 John C. Somberg

Section VI Case Histories 32. The Discovery of Rituxan ..............................................................................................565 Mitchell E. Reff 33. Funding the Birth of a Drug: Lessons from the Sell Side.......................................585 Howard E. Greene 34. Innovations for the Drug Development Pathway: What Is Needed Now.....................................................................................................................603 Janet Woodco*ck 35. Managing R&D Uncertainty and Maximizing the Commercial Potential of Pharmaceutical Compounds Using the Dynamic Modeling Framework ........................................................................................................................617 Mark Paich, Corey Peck, Jason Valant, and Kirk Solo

1 Introduction

Charles G. Smith

Prior to the 20th century, the discovery of drug substances for the treatment of human diseases was primarily a matter of “hit or miss” use in humans, based on folklore and anecdotal reports. Many, if not most, of our earliest therapeutic remedies were derived from plants or plant extracts that had been administered to sick humans (e.g., quinine from the bark of the cinchona tree for the treatment of malaria in the mid-1600s and digitalis from the foxglove plant in the mid-1700s for the treatment of heart failure, to name two). Certainly, some of these early medications were truly effective (e.g., quinine and digitalis) in the sense that we speak of effective medications today. On the other hand, based on the results of careful studies of many such preparations over the years, either in animals or man, one is forced to come to the conclusion that most likely, the majority of these plant extracts was not pharmacologically active, but rather they were perceived as effective by the patient because of the so-called placebo effect. Surprisingly, placebos (substances that are known not to be therapeutically efficacious, but that are administered so that all the psychological aspects of consuming a “medication” are presented to the patient) have been shown to exert positive effects in a wide range of disease states, attesting to the “power of suggestion” under certain circ*mstances. There still exist today practitioners of so-called homeopathic medicine, which is based on the administration of extremely low doses of substances with known or presumed pharmacologic activities. For example, certain poisons, such as strychnine, have been used as a “tonic” for years in various countries at doses that are not only nontoxic but that in the eyes of most scientifically trained medical and pharmacological authorities, could not possibly exert an actual therapeutic effect. Homeopathy is practiced not only in underdeveloped countries, but also in certain well-developed countries, including the United States, albeit on a very small scale. Such practices will, most likely, continue since a certain number of patients who require medical treatment have lost faith, for one reason or another, in the so-called medical establishment. More will be said about proving drug efficacy in Chapters 8 to 10. Pioneers in the field of medicinal chemistry such as Paul Ehrlich (who synthesized salvarsan, the first chemical treatment for syphilis, at the turn of the 20th century), were instrumental in initiating the transition from the study of plants or their extracts with purported therapeutic activities to the deliberate synthesis, in the laboratory, of a specific drug substance. Certainly, the discovery of the sulfa drugs in the 1930s added great momentum to this concept, since they provided one of the earliest examples of a class of pure chemical compounds that could be unequivocally shown to reproducibly bring certain infectious diseases under control when administered to patients by mouth. During World War II,

The Process of New Drug Discovery and Development

the development of penicillin stimulated an enormous and highly motivated industry aimed at the random testing (screening) of a variety of microbes obtained from soil samples for the production of antibiotics. This activity was set into motion by the discovery of Alexander Fleming and others in England in 1929 that a Penicillium mold produced tiny amounts of a substance that was able to kill various bacteria that were exposed to it in a test tube. When activity in experimental animal test systems and in human patients was demonstrated, using extremely small amounts of purified material from the mold broth (penicillin), it was immediately recognized that antibiotics offered a totally new route to therapeutic agents for the treatment of infectious diseases in human beings. In addition to the scientific interest in these findings, a major need existed during World War II for new medications to treat members of the armed forces. This need stimulated significant activity on the part of the United States Government and permitted collaborative efforts among pharmaceutical companies (which normally would be highly discouraged or prohibited by antitrust legislation from such in-depth cooperation) to pool resources so that the rate of discovery of new antibiotics would be increased. Indeed, these efforts resulted in accelerated rates of discovery and the enormous medical and commercial potential of the antibiotics, which were evident as early as 1950, assured growth and longevity to this important new industry. Major pharmaceutical companies such as Abbott Laboratories, Eli Lilly, E. R. Squibb & Sons, Pfizer Pharmaceuticals, and The Upjohn Company in the United States, to name a few, were particularly active in these endeavors and highly successful, both scientifically and commercially, as a result thereof (as were many companies in Europe and Japan). From this effort, a wide array of new antibiotics, many with totally unique and completely unpredictable chemical structures and mechanisms of action, became available and were proven to be effective in the treatment of a wide range of human infectious diseases. In the 1960s and 1970s, chemists again came heavily into the infectious diseases’ arena and began to modify the chemical structures produced by the microorganisms, giving rise to the so-called semi-synthetic antibiotics, which form a very significant part of the physicians’ armamentarium in this field today. These efforts have proved highly valuable to patients requiring antibiotic therapy and to the industry alike. The truly impressive rate of discovery of the ‘semi-synthetic’ antibiotics was made possible by the finding that, particularly in the penicillin and cephalosporin classes of antibiotics, a portion of the entire molecule (the so-called 6-APA in the case of penicillin and 7-ACA in the case of cephalosporin) became available in large quantities from fermentation sources. These complex structures were not, in and of themselves, able to inhibit the growth of bacteria, but they provided to the chemist the central core of a very complicated molecule (via the fermentation process), which the chemist could then modify in a variety of ways to produce compounds that were fully active (hence the term ‘semi’-synthetic antibiotics). Certain advantages were conferred upon the new molecules by virtue of the chemical modifications such as improved oral absorption, improved pharmaco*kinetic characteristics and expanded spectrum of organisms that were inhibited, to name a few. Chemical analogs of antibiotics, other than the penicillin and cephalosporins, have also been produced. The availability of truly efficacious antibiotics to treat a wide variety of severe infections undoubtedly represents one of the primary contributors to prolongation of life in modern society, as compared to the situation that existed in the early part of this century. Coincidental with the above developments, biomedical scientists in pharmaceutical companies were actively pursuing purified extracts and pure compounds derived from plants and animal sources (e.g., digitalis, rauwolfia alkaloids, and animal hormones) as human medicaments. Analogs and derivatives of these purified substances were also investigated

Introduction

intensively in the hope of increasing potency, decreasing toxicity, altering absorption, securing patent protection, etc. During this period, impressive discoveries were made in the fields of cardiovascular, central nervous system, and metabolic diseases (especially diabetes); medicinal chemists and pharmacologists set up programs to discover new and, hopefully, improved tranquilizers, antidepressants, antianxiety agents, antihypertensive agents, hormones, etc. Progress in the discovery of agents to treat cardiovascular and central nervous system diseases was considerably slower than was the case with infectious diseases. The primary reason for this delay is the relative simplicity and straightforwardness of dealing with an infectious disease as compared to diseases of the cardiovascular system or of the brain. Specifically, infectious diseases are caused by organisms that, in many cases, can be grown in test tubes, which markedly facilitates the rate at which compounds that inhibit the growth of, or actually kill, such organisms can be discovered. Not only was the testing quite simple when carried out in the test tube but also the amounts of compounds needed for laboratory evaluation were extremely small as compared to those required for animal evaluation. In addition, animal models of infectious diseases were developed very early in the history of this aspect of pharmaceutical research and activity in an intact animal as well as toxicity could be assessed in the early stages of drug discovery and development. Such was not the case in the 1950s as far as cardiovascular, mental, or certain other diseases were concerned because the basic defect or defects that lead to the disease in man were quite unknown. In addition, early studies had to be carried out in animal test systems, test systems which required considerable amounts of the compound and were much more difficult to quantitate than were the in vitro systems used in the infectious-disease field. The successes in the antibiotic field undoubtedly showed a carry-over or ‘domino’ effect in other areas of research as biochemists and biochemical pharmacologists began to search for in vitro test systems to provide more rapid screening for new drug candidates, at least in the cardiovascular and inflammation fields. The experimental dialog among biochemists, pharmacologists, and clinicians studying cardiovascular and mental diseases led, in the 1960s, to the development of various animal models of these diseases that increased the rate of discovery of therapeutic agents for the treatment thereof. Similar research activities in the fields of cancer research, viral infections, metabolic diseases, AIDS, inflammatory disease, and many others have, likewise, led to in vitro and animal models that have markedly increased the ability to discover new drugs in those important fields of research. With the increased discovery of drug activity came the need for increased regulation and, from the early 1950s on, the Food and Drug Administration (FDA) expanded its activities and enforcement of drug laws with both positive and negative results, from the standpoint of drug discovery. In the later quarter of the 20th century, an exciting new technology emerged into the pharmaceutical scene, namely, biotechnology. Using highly sophisticated, biochemical genetic approaches, significant amounts of proteins, which, prior to the availability of so-called genetic engineering could not be prepared in meaningful quantities, became available for study and development as drugs. Furthermore, the new technology permitted scientists to isolate, prepare in quantity, and chemically analyze receptors in and on mammalian cells, which allows one to actually design specific effectors of these receptors. As the drug discovery process increased in intensity in the mid- to late 20th century, primarily as a result of the major screening and chemical synthetic efforts in the pharmaceutical industry in industrialized countries worldwide, but also as a result of the biotechnology revolution, the need for increased sophistication and efficacy in (1) how to discover new drugs, (2) how to reproducibly prepare bulk chemicals, (3) how to determine the activity and safety of new drug candidates in preclinical animal models prior to their

The Process of New Drug Discovery and Development

administration to human beings and, finally, (4) how to establish their efficacy and safety in man, became of paramount importance. Likewise, the ability to reproducibly prepare extremely pure material from natural sources or biotechnology reactors on a large scale and to deliver stable and sophisticated pharmaceutical preparations to the pharmacists and physicians also became significant. The above brief history of early drug use and discovery is intended to be purely illustrative and the reader is referred to an excellent treatise by Mann1 to become well informed on the history of drug use and development from the earliest historic times to the present day.

Reference 1. Mann, R.D., Modern Drug Use: An Enquiry on Historical Principles, MTP Press, Lancaster, England, 1984, pp. 1–769.

Section I

General Overview

2 Overview of the Current Process of New Drug Discovery and Development

Charles G. Smith and James T. O’Donnell

CONTENTS 2.1 Basic Scientific Discovery and Application to New Drug Development ..................11 2.2 Regulation of New Drug Development ..........................................................................12 2.3 Liability and Litigation ......................................................................................................12 References ......................................................................................................................................12

The first edition1 of this book was published approximately 13 years ago. Its primary objective was to present an overview and a “roadmap” of the process of new drug discovery and development, particularly oriented to individuals or companies entering the pharmaceutical field. It was written by one of the authors (Smith), with no contributors, and drawn on Smith’s experiences in the industry and field over the course of nearly 40 years. In the second edition, the scope of the first book has been expanded and technical details in the form of hard data have been included. In addition to the editors’ own commentary and contributions, the major part of the book is the result of contributions of experts in the industry. New chapters on risk assessment, international harmonization of drug development and regulation, dietary supplements, patent law, and entrepreneurial startup of a new pharmaceutical company have been added. Some of the important, basic operational aspects of drug discovery and development (e.g., organizational matters, staff requirements, pilot plant operations, etc.) are not repeated in this book but can be found in the first edition. In the 1990s and the new millennium, major changes have occurred in the pharmaceutical industry from the vantage points of research and development as well as commercial operations. New technologies and processes such as “high throughput screening” and “combinatorial chemistry” were widely embraced and developed to a high state of performance during this period. The very impressive rate of throughput testing the hundreds of thousands of compounds required micronization of operations, resulting in the reduction of screening reaction mixtures from milliliters to microliters. The systems are generally controlled by robots, and testing plates can accommodate a wide spectrum of biological tests. Combinatorial chemistry, a process in which a core molecule is modified with a broad spectrum of chemical reactions in single or multiple reaction vessels, can produce tens of thousands of compounds for screening. The objective of both approaches is to provide very large numbers of new chemical entities to be screened for biological activity in vitro. The use of computers to design new drug candidates has been developed to a significant level of sophistication. By viewing on the computer, the “active site” to which one wants the drug

The Process of New Drug Discovery and Development

candidate to bind, a molecule can often be designed to accomplish that goal. The true impact of these approaches on the actual rate of discovering new drugs is yet to be established. Some have questioned the utility of these new screening methods, claiming that no new molecular entities (NME) have resulted from these new screening methodologies, despite hundreds of millions invested by the industry. Studies in the last few years in the fields of genomics and proteomics have made available to us an unprecedented number of targets with which to search for new drug candidates. While knowledge of a particular gene sequence, for example, may not directly point to a specific disease when the sequences are first determined, investigations of their presence in normal and diseased tissues could well lead to a quantitative in vitro test system that is not available today. The same can be said for the field of proteomics, but final decisions on the value of these new technologies cannot be made for some years to come. Thanks to advances in genomics, animal models can now be derived using gene manipulation and cloning methods that give us never-before available in vivo models to be used in new drug screening and development. A large number of mammalian cell culture systems have also been developed not only to be used in primary screening but also for secondary evaluations. For example, the in vitro Caco 2 system shows some very interesting correlation with drug absorption in vivo. A test such as this is mandatory when one is dealing with several thousands of compounds or mixtures in a given experiment. More time will be needed to be absolutely certain of the predictability of such test systems but, appropriately, Caco 2 is widely used today in screening and prioritizing new drug candidates. As is always the case, the ultimate predictability of all the in vitro tests must await extensive studies in humans, which will occur several years henceforth. In addition to the discussion are metabonomics that relate to their unique position within the hierarchy of cell function and their propensity to cross membranes and organs. Thus, many metabolites are found in bodily fluids that are accessible to measurement in humans using relatively noninvasive technologies. The study of metabolomics provides the pragmatic link from the macromolecular events of genomics and proteomics to those events recognized in histology. Applications of such strategies can potentially translate discovery and preclinical development to those metabolites measured traditionally, as first-in-human studies are performed earlier in drug discovery and development process, especially where no animal models are adequate. During the past decade, clinical trial methodology has been expanded, improved, and, in large measure, standardized. The clinical testing phase of new drug development is the most expensive single activity performed. In addition to cost, it is very time consuming since, with chronic diseases, one must investigate the new drug candidate in a significant number of patients over a period of months or years, in randomized, double-blind, placebo- or active-drug-controlled studies. The search for surrogate endpoints continues, as it should, because a surrogate endpoint can markedly increase the rate of progression in clinical investigations with new drug candidates in certain disease states. Modern advances in molecular biology, receptor systems, cellular communication mechanisms, genomics, and proteomics will, according to our belief, provide researchers with new approaches to the treatment of a variety of chronic diseases. Significantly improved prescription medications are sorely needed in many fields. In the past decade, we have witnessed very impressive advances in the treatment of AIDS, for example. There is no question that life expectancy has been increased, albeit accompanied by significant drug toxicity and the need to use a “co*cktail” of drugs in combination. The ability of the AIDS virus to mutate and become drug resistant presents a major and imminent threat to all patients afflicted with this disease. Serious efforts are under way in the pharmaceutical industry to find new drugs, across the entire infections diseases spectrum, which are not cross-resistant with existing therapies.

Overview of the Current Process of New Drug Discovery and Development

Cancer and AIDS vaccines are also under investigation using new technologies and, hopefully, the day will come when we can prevent or ameliorate some of these debilitating and fatal diseases by vaccination. In the cancer field, new methodologies in science have, again, given us new targets with which to search for chemotherapeutic agents. The humanization of monoclonal antibodies has resulted in the marketing of some truly impressive drugs that are much better tolerated by the patient than are cytotoxic agents. In the case of certain drug targets in cancer, impressive results have been seen in the percentage of leukemia and lymphoma patients who can be brought into complete remission. In addition, biological medications to increase red and white blood cells have become available. Unfortunately, drug resistance once again plagues the cancer field, as are the cases with AIDS and various infectious diseases. As a result, researchers are seeking compounds that are not cross-resistant with existing therapies. Very significant advances in drug discovery are also expected to be seen in central nervous system, cardiovascular, and other chronic diseases as a result of breakthrough research in these fields. Although the focus of this book is the research and development side of the pharmaceutical industry, certain commercial considerations are worth mentioning because of the major impact they may have on new drug research. These opinions and conclusions are based solely on decades of experience in the field by editors, working in the industry within companies and as an independent consultant (Smith), and also as a health care worker and academic (O’Donnell). No financial incentive for these statements has been received from the pharmaceutical industry. As the result of the very complicated nature of drug discovery and development, unbelievable costs accrue in order to bring a new therapeutic agent to market. Increasing costs are incurred, in part, from (1) shifting disease targets from more rapidly evaluable, acute diseases to those with poor endpoints and chronicity and (2) the emergence and rapid spread of serious diseases in society (e.g., AIDS, certain cancers, hepatitis C, etc.). In addition to increasing cost, the time required to gather sufficient data to be able to prove, to a statistically valid endpoint, that the drug has indeed been effective in a given disease has risen. The cost for the development of a major drug has been widely stated to be US $800 million per new therapeutic agent placed on the market.1 This figure incorporates, of course, the cost of “lost” compounds that did not make the grade during preclinical or clinical testing. It has recently been reported that, while historically 14% of drugs that entered phase I clinical trials eventually won approval, now only 8% succeed. Furthermore, 50% of the drug candidates fail in the late stage of phase III trials compared to 20% in past years. More details on these points can be found in the literature (cf., Refs. 2–8). The average time from the point of identifying a clinical candidate to approval of a new drug is approximately 10 years. There is an understandable clamor in the population and in our legislative bodies to lower the price of prescription drugs. The cost of some prescription drugs is, to be sure, a serious problem that must be addressed but some of the solutions, suggested and embraced by certain legislators, could have serious negative impact on new drug discovery and development in the future. For example, allowing the importation of prescription drugs from Canada or other non-U.S. countries (25 around the world have been mentioned) may well reduce the price of new drugs in this country to the point of significantly decreasing profits that are needed to support the tremendous cost of new drug discovery and development. The record clearly shows that countries that control drug prices, frequently under socialist governments, do not discover and develop new prescription drugs. The reason is obvious since the cost and time factors for new drug discovery can only be borne in countries in which the pharmaceutical companies are clearly profitable. Our patent system and lack of price controls are the primary reasons for the huge industrial success of new product development in this country, in and out of the pharmaceutical arena. If we undercut that system in the prescription drug field, the cost

The Process of New Drug Discovery and Development

of drugs will certainly go down in the United States in the short term but, without the necessary profits to invest heavily in new drug discovery and development, the latter will also surely drop. Since it requires a decade from the time of initial investigation to marketing of a new drug, this effect would not be evident immediately after (1) allowing reimportation, (2) overriding patent protection, or (3) implementing price controls but, within a period of 5 to 10 years, we would certainly see pipelines of new medications beginning to dry up. Indeed, if such a system were allowed to continue for several years, new drug development as we know it would, in our opinion, be seriously impeded. When legislators look to Canada as an example of successful government subsidy of drugs, they should also consider whether a country like Canada could ever produce a steady stream of major new drugs, as does the United States. Research budgets have never been larger, we have never had as many innovative and exciting targets on which to focus, and this enormous effort cannot be afforded unless the companies selling the drugs can realize an adequate profit. If our pipelines of new prescription drugs dry up, you can be rest assured that the deficit will not be satisfied elsewhere in the world. It has been reported that, 10 years ago drug companies in Europe produced a significantly greater percentage of prescription drugs than is the case today. Society simply cannot afford to risk a marked reduction in new drug discovery in this country. Patients must join the fight to see that activities to impose price controls, which will inevitably reduce the rate of discovery of many potential drugs, are not based on political motives on the part of legislators. At this point in history, U.S. science stands in the forefront of new drug discovery and development. As noted above, never before have we had such an array of biological targets and synthetic and biotechnological methods with which to seek new medications. Hopefully, our government, in collaboration with the pharmaceutical industry, will find more suitable methods to solve the question of the cost of new pharmaceuticals than to impose price controls equal to those in countries that have socialized medicine. There can be no question as to whether the primary loser in such moves will be patients. In addition to the question of the rate of drug discovery and development, we must be concerned about the quality of drugs available by mail or over the internet. The Food and Drug Administration (FDA) cannot possibly afford to check all drugs flowing into America from as many as 25 foreign countries from which our citizens might be allowed to buy prescription drugs. It will be interesting to compare the regulatory requirements for FDA approval in the United States with those of the least stringent of the foreign countries from which some of our legislators want to approve importation of drugs. Would Congress be prepared to mandate a lowering of FDA standards to the same level in order to reduce the cost of drug discovery and development in this country? We certainly hope not! Indeed, there have been reports that drugs imported and sold on the internet are counterfeit, and frequently contain little or no labeled active ingredients, and further, may contain adulterants. Another new topic chapter in the second edition of this book discusses the so-called dietary supplements, contributed by a recognized authority in Health Fraud. Over the past few years and, especially, since the passage of the DSHEA Act by Congress,9 the use of such products has increased dramatically and they are made widely available to the public with little or no FDA regulation. Although the law prevents manufacturers from making a medical treatment claim on the label of these preparations, such products generally have accompanying literature citing a variety of salutary effects in patients with various ills, the majority of which have not been proven by FDA type-randomized, double-blind, placebo-controlled clinical studies, of the kind that must be performed on prescription drugs and some “over-the-counter” drugs in this country. Published studies on quality control defects in some of these dietary supplement products (cf. ConsumerLab.com) indicate the need for tightening up of this aspect of product development. FDA is currently promulgating GMPs for dietary supplements. An enhanced

Overview of the Current Process of New Drug Discovery and Development

enforcement of the dietary supplement regulations now exists.9 A small segment of the dietary supplement industry has been calling for GMPs and increased FDA regulation.10

2.1

Basic Scientific Discovery and Application to New Drug Development

In an apparent attempt to determine whether the American taxpayer is getting fair benefits from research sponsored by the federal government, the Joint Economic Committee of the U.S. Senate (for history see Ref. 7) has been considering this question. Historically, basic research has been funded by the NIH and various philanthropic foundations to discover new concepts and mechanisms of bodily function, in addition to training scientists. The role of industry has been to apply the basic research findings to specific treatments or prevention of disease. This is the appropriate manner in which to proceed. The industry cannot afford to conduct sufficient basic research on new complicated biological processes in addition to discovering new drugs or vaccines. The government does not have the money, time, or required number of experts to discover and develop new drugs. The process that plays out in real life involves the focus of pharmaceutical industry scientists on desirable biological targets that can be identified in disease states, and to set up the program to discover specific treatments that will show efficacy in human disease. The compounds that are developed successfully become drugs on which the company holds patents. In this manner, the enormous cost of discovering and developing a new drug (estimated at $800 million plus over a period of some 10 years1) as noted above can be recouped by the founding company since no competitors can sell the product as long as the patent is in force. Without such a system in place, drug companies simply could not, in our opinion, afford to bring new prescription drugs to the market. In the course of reviewing the matter, the Joint Economic Committee examined a list of 21 major drugs, which was put together apparently as an example of drug products that might justify royalty to the government. One of these agents, captopril (trade name Capoten), was discovered and developed by E.R. Squibb & Sons in the 1970s. At that time, Charles Smith (one of the authors/editors) was vice president for R&D at The Squibb Institute for Medical Research. One of Squibb’s academic consultants, Professor Sir John Vane of the Royal College of Surgeons in London brought the idea of opening a new pathway to treat the so-called essential hypertension by inhibiting an enzyme known as the angiotensin converting enzyme (ACE). This biochemical system was certainly known at that time but, in Squibb’s experience in the field of hypertension treatment, was not generally thought to play a major role in the common form of the disease, then known as “essential hypertension.” The company decided to gamble on finding a treatment that was not used at the time and that would be proprietary to the company. Professor Vane (Nobel laureate in medicine in 1982) had discovered a peptide in snake venom that was a potent inhibitor of ACE. Squibb decided to pursue the approach he espoused, resulting in the development of a unique treatment for this very prevalent and serious disease. In the first phase of their research, Squibb tested a short-chain peptide isolated from the venom of the viper Bothrops jararaca, with which Vane was working in the laboratory, in human volunteers and showed that it did, indeed, inhibit the conversion of angiotensin I to angiotensin II after intravenous injection. The peptide was also shown to reduce blood pressure in patients when injected. Since the vast majority of peptides cannot be absorbed from the GI tract, Squibb scientists set out to prepare a nonpeptide compound that could be used orally and manufactured at acceptable cost. The design of a true peptidomimetic that became orally active had not been accomplished at that time. Squibb then carried out

The Process of New Drug Discovery and Development

a full-blown clinical program on a worldwide basis, which led to FDA approval of Squibb’s drug Capoten (captopril), an ACE inhibitor. Mark also marketed an ACE inhibitor in the same time frame. This work opened a new area of research that has resulted in a bevy of new drugs that share this mechanism of action for use as antihypertensive drugs (for more detail, see Refs. 11–15). In the minds of pharmaceutical researchers and, hopefully, the public at large, the above example illustrates the unique role of pharmaceutical companies in making good use of basic research to discover new treatments for serious diseases. The huge costs to discover and develop a new drug could not be borne unless the companies knew that, if their gamble worked (which is not the case in the majority of situations), they would be assured of a good financial return for their shareholders. This system has served the country well in many fields of endeavor, in and out of the drug arena, and should be retained as such.

2.2

Regulation of New Drug Development

Drug development will come to a crashing halt without approval of the U.S. FDA, authorized by Congress to approve, license, and monitor the drugs sold to the American public. We are fortunate to have two contributors from the FDA, an acting associate commissioner for operations, and also CDER’s (Center for Drug Evaluation and Research) associate director for International Conference on Harmonisation (ICH). These authors describe the FDA’s new critical pathway initiative, pharmacists’ risk management contributions, as well as the Common Technical Document (eCTD), which will enable a sponsor to file in one of the cooperating ICH partners, and receive approval for almost global marketing of the new agent. A very important chapter on pharmacogenetics and pharmacogenomics includes numerous FDA contributers.

2.3

Liability and Litigation

Last and the most unpopular topic in any industry, especially in the pharmaceutical industry, is the topic of liability and litigation. We have elected to include a chapter on this topic so that workers from all scientific disciplines involved in drug discovery and development can learn from history, and, hopefully, avoid being involved in the devastation of life (due to toxicity of inadequately manufactured drugs or drugs with inadequate warnings for safe use) and destruction of companies and careers that follows in the aftermath of drug product litigation.

References 1. Smith, C.G., The Process of New Drug Discovery and Development, 1st ed., CRC Press, Boca Raton, FL, 2002. 2. Di Masi, J.A., Hansen, R.W., and Grabowski, H.G., The price of innovation: new estimates of drug development costs, J. Health Econ., 22, 151–185, 2003.

Overview of the Current Process of New Drug Discovery and Development

3. Reichert, J.M. and Milne, C.-P., Public and private sector contributions to the discovery and development of ‘impact’ drugs, Am. J. Therapeut., 9, 543–555, 2002. 4. Di Masi, J., Risks in new drug development. Approval success rates for investigational drugs, Clin. Pharmacol. Therapeut., 69, 297–307, 2001. 5. Di Masi, J., New drug development in the United States from 1963 to 1999, Clin. Pharmacol. Therapeut., 69, 286–296, 2001. 6. Grabowski, H., Vernon, J., and Di Masi, J.A., Returns on research and development for 1990s new drug introductions, Pharmacol. Econ., 20 (suppl. 3), 11–29, 2002. 7. Reichert, J.M. and Milne, C.-P., Public and private sector contributions to the discovery and development of “impact” drugs, A Tufts Center for the Study of Drug Development White Paper, May 2002. 8. Hardin, A., More compounds failing phase I, Scientist Daily News, Aug. 6, 2004, p. 1. 9. Dietary Supplement Health and Education Act of 1994, Publ. no. 103-417, 108 Stat. 4325 codified 21 U.S.C. 321, et seq. (suppl. 1999). 10. FDA links: (a) http://www.cfsan.fda.gov/~dms/ds-warn.html (b) http://www.cfsan.fda.gov/ ~lrd/ hhschomp.html (c) http://www.fda.gov/ola/2004/dssa0608.html 11. Smith, C.G. and Vane, J.R., The discovery of captopril, FASEB J., 17, 788–789, 2003. 12. Gavras, H., The discovery of captopril: Reply, FASEB J., 18, 225, 2004. 13. Erdas, E.G., The discovery of captopril: Reply, FASEB J., 18, 226, 2004. 14. Pattac, M., From viper’s venom to drug design: treating hypertension, FASEB J., 18, 421, 2004. 15. Smith, C.G. and Vane, J.R., The discovery of captopril: Reply, FASEB J., 18, 935, 2004.

3 Integrated Drug Product Development — From Lead Candidate Selection to Life-Cycle Management

Madhu Pudipeddi, Abu T.M. Serajuddin, and Daniel Mufson

CONTENTS 3.1 Introduction ........................................................................................................................16 3.2 Developability Assessment ..............................................................................................17 3.2.1 Evolution of the Drug Discovery and Development Interaction ..................18 3.2.2 Screening for Drugability or Developability ....................................................18 3.2.2.1 Computational Tools ..............................................................................19 3.2.2.2 High-Throughput Screening Methods ................................................20 3.2.2.3 In-Depth Physicochemical Profiling ....................................................21 3.3 Overview of Dosage-Form Development and Process Scale-Up ................................22 3.4 Preformulation ....................................................................................................................22 3.4.1 Preformulation Activities: Independent of Solid Form ..................................23 3.4.1.1 Dissociation Constant ............................................................................23 3.4.1.2 Partition or Distribution Coefficient ....................................................24 3.4.1.3 Solution Stability Studies ...................................................................... 24 3.4.2 Preformulation Activities: Dependent on Solid Form ....................................25 3.4.2.1 Solubility ..................................................................................................25 3.4.2.2 Salt-Form Selection ................................................................................25 3.4.2.3 Polymorphism ..........................................................................................26 3.4.2.4 Solid-State Stability ................................................................................27 3.4.2.5 Drug-Excipient Interactions ..................................................................27 3.4.2.6 Powder Properties of Drug Substance ................................................28 3.5 Biopharmaceutical Considerations in Dosage-Form Design ......................................29 3.5.1 Physicochemical Factors ......................................................................................30 3.5.1.1 Solubility ..................................................................................................30 3.5.1.2 Dissolution ..............................................................................................30 3.5.2 Physiological Factors ............................................................................................32 3.5.2.1 Assessment of In Vivo Performance ....................................................33 3.5.2.2 In Vitro–In Vivo Correlation ..................................................................33 3.6 Clinical Formulation Development: Clinical Trial Materials ......................................33 3.6.1 Phase I Clinical Trial Material ..............................................................................34 3.6.2 Phase II Clinical Trial Material ............................................................................36 3.6.3 Phase III Clinical Trial Material ..........................................................................38

The Process of New Drug Discovery and Development

3.7

Nonoral Routes of Administration ..................................................................................39 3.7.1 Parenteral Systems ................................................................................................39 3.7.2 Inhalation Systems ................................................................................................40 3.8 Drug-Delivery Systems ......................................................................................................41 3.9 Product Life-Cycle Management ......................................................................................44 3.10 Summary ..............................................................................................................................45 References ......................................................................................................................................46

3.1

Introduction

Historically, medicines have been administered through the obvious portals following their preparation first by the shaman and then by the physician and later by the apothecary. These natural products were ingested, rubbed-in, or smoked. For the past century, the person diagnosing the disease no longer prepares the potion, eliminating, no doubt, some of the power of the placebo, and as a consequence, drug discovery, development, and manufacturing have grown into a separate pharmaceutical industry. In particular, the last 50 years have been a period of astounding growth in our insight of the molecular function of the human body. This has led to discovery of medicines to treat diseases that were not even recognized a half-century ago. This chapter reflects the role of pharmaceutics and the diversity of the approaches taken to achieve these successes, including approaches that were introduced within recent years, and describes how the role of the “industrial” pharmacist has evolved to become the technical bridge between discovery and development activities and, indeed, commercialization activities. No other discipline follows the progress of the new drug candidate as far with regard to the initial refinement of the chemical lead through preformulation evaluation to dosage-form design, clinical trial material (CTM) preparation, process scale-up, manufacturing, and then life-cycle management (LCM). The pharmaceutical formulation was once solely the responsibility of the pharmacist, first in the drugstore and later in an industrial setting. Indeed, many of today’s major drug companies, such as Merck, Lilly, Wyeth, and Pfizer components Searle, WarnerLambert, and Parke-Davis, started in the backrooms of drugstores. During the second half of the 20th century, physicochemical and biopharmaceutical principles underlying pharmaceutical dosage forms were identified and refined, thanks to the pioneering works by Higuchi,1 Nelson,2 Levy,3 Gibaldi,4 and their coworkers. Wagner,5 Wood,6 and Kaplan7 were among the earliest industrial scientists to systematically link formulation design activities and biology. Nevertheless, until recently, formulations were developed somewhat in isolation with different disciplines involved in drug development operating independently. For example, during the identification and selection of new chemical entities (NCEs) for development, not much thought was given into how they would be formulated, and during dosage-form design, adequate considerations of in vivo performance of formulations was lacking. Wagner5 first termed our evolving understanding of the relationship between the dosage form and its anatomical target, “biopharmaceutics” in the early 1960s. Since then it has been apparent that careful consideration of a molecule’s physical chemical properties and those of its carrier, the dosage form, must be understood to enhance bioavailability, if given orally, and to enhance the ability of drug to reach the desired site of action, if given by other routes of administration. This knowledge allows for a rational stepwise approach in selecting new drug candidates, developing

Integrated Drug Product Development — From Lead Candidate Selection to LCM

optimal dosage forms, and, later when it is necessary, making changes in the formulation or manufacturing processes. During the last decade or so, the basic approach of dosageform development in the pharmaceutical industry has changed dramatically. Dosageform design is now an “integrated process” starting from identification of drug molecules for development to their ultimate commercialization as dosage forms. This is often performed by a multidisciplinary team consisting of pharmacists, medicinal chemists, physical chemists, analytical chemists, material scientists, pharmaco*kineticists, chemical engineers, and other individuals from related disciplines. In its simplest terms the dosage form is a carrier of the drug. It must further be reproducible, bioavailable, stable, readily scaleable, and elegant. The skill sets employed to design the first units of a dosage form, for example, a tablet, are quite different than those required to design a process to make hundreds of thousands of such units per hour, reproducibly, in ton quantities, almost anywhere in the world. Nevertheless, it is important that “design for manufacturability” considerations are made early although resource constraints and minimal bulk drug supply may not favor them. The manufacturability situation becomes understandably more complex as the dosage form becomes more sophisticated or if a drug-delivery system (DDS) is needed. The level of sophistication in dosage-form design has been keeping pace with advances in discovery methods. New excipients, new materials, and combination products that consist of both a drug and a device have arisen to meet new delivery challenges. For example, many of the NCEs generated by high-throughput screening (HTS) are profoundly waterinsoluble. What was considered a lower limit for adequate water solubility7 (~0.1 mg/mL) in the 1970s has been surpassed by at least an order of magnitude due to changes in the way drug discovery is performed. Traditional methods such as particle size reduction to improve the aqueous dissolution rate of these ever more insoluble molecules are not always sufficient to overcome the liability. New approaches have evolved to meet these challenges ranging from cosolvent systems8 to the use of lipid—water-dispersible excipients9 and to the establishment of numerous companies with proprietary methods to increase bioavailability. Many literature sources describing formulation and manufacture of different pharmaceutical dosage forms are available.10,11 The primary objective of this chapter is to describe an integrated process of drug development, demonstrating how all activities from lead selection to LCM are interrelated. Various scientific principles underlying these activities are described. A survey of new drug approvals (NDAs) during the last 5 years (1999 to mid-2004) showed that nearly 50% of them are oral dosage forms. The percentage is higher if Abbreviated NDAs for generics are included. Therefore, the primary focus of this chapter is the development of oral dosage forms with a few other dosage forms described only briefly. However, many of the principles described in this chapter are common to all dosage forms.

3.2

Developability Assessment

The dosage-form design is guided by the properties of the drug candidate. If an NCE does not have suitable physical and chemical properties or pharmaco*kinetic attributes, the development of a dosage form (product) may be difficult and may sometimes be even impossible. Any heroic measures to resolve issues related to physicochemical and biopharmaceutical properties of drug candidates add to the time and cost of drug

The Process of New Drug Discovery and Development

development. Therefore, in recent years, the interaction between discovery and development scientists increased greatly to maximize the opportunity to succeed.12

3.2.1

Evolution of the Drug Discovery and Development Interaction

The traditional (i.e., pre-1990s) drug discovery process involved initial lead generation on the basis of natural ligands, existing drugs, and literature leads. New compounds would be synthesized and tested for biological activity, and structure–activity relationships would be established for optimization of leads using traditional medicinal chemistry techniques. Promising compounds would then be promoted for preclinical and clinical testing and therefore passed along to the product development staff. While often there was little collaboration between research and development, a few organizations had recognized the importance of discovery-development teams to assess development issues related to new drug candidates.7 The current (post-1990s) drug discovery process typically involves:13 ● ● ● ●

Target identification Target validation Lead identification Candidate(s) selection

A drug target can be a receptor/ion channel, enzyme, hormone/factor, DNA, RNA, nuclear receptor, or other, unidentified, biological entity. Once drug targets are identified, they are exposed to a large number of compounds in an in vitro or cell-based assay in an HTS mode. Compounds that elicit a positive response in a particular assay are called “hits.” Hits that continue to show positive response in more complex models rise to “leads” (lead identification). A selected few of the optimized leads are then advanced to preclinical testing. The traditional discovery process has not been discontinued but still occurs in a semiempirical fashion depending on the chemist’s or biologist’s experience and intuition. With the application of HTS technologies, compound handling in discovery has shifted to the use of organic stock solutions (dimethylsulfoxide) for in vitro and in vivo testing from the traditional use of gum tragacanth suspensions in rats by the pharmacologist. Use of combinatorial chemistry and HTS technologies have resulted in the generation and selection of increasingly lipophilic drug molecules with potential biopharmaceutical hurdles in downstream development.14 Particularly, the use of organic solvents such as dimethylsulfoxide has contributed to the increase in water-insoluble drugs. Analysis of compound attrition in pharmaceutical development indicated that poor pharmaco*kinetic factors, i.e., absorption, elimination, distribution, and metabolism (ADME) contributed to about 40% of failed candidates, and for those that moved forward, the development timelines significantly slowed down.15 To reduce attrition of compounds later in development, pharmaceutical companies began to conduct pharmaceutical, pharmaco*kinetic, and safety profiling of late- as well as early-phase discovery compounds.16

3.2.2

Screening for Drugability or Developability

Compounds with acceptable pharmaceutical properties, in addition to acceptable biological activity and safety profile, are considered “drug-like” or developable. Typical acceptable pharmaceutical properties for oral delivery of a drug-like molecule include sufficient aqueous solubility, permeability across biological membranes, satisfactory stability to

Integrated Drug Product Development — From Lead Candidate Selection to LCM

metabolic enzymes, resistance to degradation in the gastrointestinal (GI) tract (pH and enzymatic stability), and adequate chemical stability for successful formulation into a stable dosage form. A number of additional barriers, such as efflux transporters17 (i.e., export of drug from blood back to the gut) and first-pass metabolism by intestinal or liver cells, have been identified that may limit oral absorption. A number of computational and experimental methods are emerging for testing (or profiling) drug discovery compounds for acceptable pharmaceutical properties. In this section, discussion of physicochemical profiling is limited to solubility, permeability, drug stability, and limited solid-state characterization (as we will see in Section 3.4, there are other physical–mechanical properties that must also be considered). For convenience, methods available for physicochemical profiling are discussed under the following categories: computational tools (sometimes referred to as in silico tools), HTS methods, and in-depth physicochemical profiling.16

3.2.2.1 Computational Tools Medicinal chemists have always been adept in recognizing trends in physicochemical properties of molecules and relating them to molecular structure. With rapid increase in the number of hits and leads, computational tools have been proposed to calculate molecular properties that may predict potential absorption hurdles. For example, Lipinski’s “Rule of 5”14 states that poor absorption or permeation are likely when: 1. There are more than five H-bond donors (expressed as the sum of –NH and –OH groups). 2. The molecular weight is more than 500. 3. log P⬎5 (or c log P⬎4.5). 4. There are more than ten H-bond acceptors (expressed as the sum of Ns and Os) If a compound violates more than two of the four criteria, it is likely to encounter oral absorption issues. Compounds that are substrates for biological transporters and peptidomimetics are exempt from these rules. The Rule of 5 is a very useful computational tool for highlighting compounds with potential oral absorption issues. A number of additional reports on pharmaceutical profiling and developability of discovery compounds have been published,18 since the report of Rule of 5. Polar surface area (PSA) and number of rotatable bonds have also been suggested as means to predict oral bioavailability. PSA is defined as the sum of surfaces of polar atoms in a molecule. A rotatable bond is defined as any single bond, not in a ring, bound to a nonterminal heavy (i.e., non-hydrogen) atom. Amide bonds are excluded from the count. It has been reported that molecules with the following characteristics will have acceptable oral bioavailability:19 1. Ten or fewer rotatable bonds. 2. Polar surface area equal to or less than 140 Å2 (or 12 or fewer H-bond donors and acceptors). Aqueous solubility is probably the single most important biopharmaceutical property that pharmaceutical scientists are concerned with. It has been the subject of computational prediction for several years.20–23 The overall accuracy of the predicted values can be expected to be in the vicinity of 0.5 to 1.0 log units (a factor of 3 to 10) at best. Although a decision on acceptance or rejection of a particular compound cannot be made only on the basis of predicted parameters, these predictions may be helpful to direct chemical libraries with improved drug-like properties.24

The Process of New Drug Discovery and Development

3.2.2.2 High-Throughput Screening Methods High-throughput drug-like property profiling is increasingly used during lead identification and candidate selection. HTS pharmaceutical profiling may include: ●

● ● ● ● ●

Compound purity or integrity testing using methods such as UV absorbance, evaporative light scattering, MS, NMR, etc.25 Solubility. Lipophilicity (log P). Dissociation constant (pKa). Permeability. Solution/solid-state stability determination.

Compound purity (or integrity testing) is important to ensure purity in the early stages because erroneous activity or toxicity results may be obtained by impure compounds. It is initiated during hit identification and continued into lead and candidate selection. Solubility is measured to varying degrees of accuracy by HTS methods. Typical methods in the lead identification stage include determination of “kinetic solubility” by precipitation of a drug solution in dimethylsulfoxide into the test medium. Since the solid-state form of the precipitate (crystalline or amorphous) is often not clearly known by this method, the measured solubility is approximate and generally higher than the true (equilibrium) solubility. Kinetic solubility, however, serves the purpose of identifying solubility limitations in activity or in vitro toxicity assays or in identifying highly insoluble compounds. Lipinski et al.14 observed that, for compounds with a kinetic solubility greater than 65 g/mL (in pH 7 non-chloride containing phosphate buffer at room temperature), poor oral absorption is usually due to factors unrelated to solubility. The acceptable solubility for a drug compound depends on its permeability and dose. This point will be further elaborated later. Methods to improve solubility in lead optimization have been reviewed.26 Estimation or measurement of pKa is important to understand the state of ionization of the drug under physiological conditions and to evaluate salt-forming ability.27 Log P determines the partitioning of a drug between an aqueous phase and a lipid phase (i.e., lipid bilayer). Log P and acid pKa can be theoretically estimated with reasonable accuracy.14,28,29 High-throughput methods are also available for measurement of log P30 and pKa.31 Physical flux of a drug molecule across a biological membrane depends on the product of concentration (which is limited by solubility) and permeability. High-throughput artificial membrane permeability (also called Parallel Artificial Membrane Permeability Assay) has been used in early discovery to estimate compound permeability.32 This method measures the flux of a compound in solution across an artificial lipid bilayer deposited on a microfilter. Artificial membrane permeability is a measure of the actual flux (rate) across an artificial membrane whereas log P or log D — as mentioned earlier — represent equilibrium distribution between an aqueous and a lipid phase. Sometimes the term “intrinsic permeability” is used to specify the permeability of the unionized form. Artificial membrane permeability can be determined as a function of pH. The fluxes across the artificial membrane in the absence of active transport have been reported to relate to human absorption through a hyperbolic curve. The correlation of permeability through artificial membranes may depend on the specific experimental conditions such as the preparation of the membranes and pH. Therefore, guidelines on what is considered acceptable or unacceptable permeability must be based on the individual assay conditions. For example, Hwang et al.33 ranked compound permeation on the basis of the percent transport across the lipid bilayer in 2 h: ⬍2% (low), 2 to 5% (medium), and ⬎5% (high), respectively. Caco-2 monolayer, a model for human intestinal permeability, is commonly used in drug discovery to screen discovery compounds.34,35 The method involves measurement of flux of

Integrated Drug Product Development — From Lead Candidate Selection to LCM

the compound dissolved in a physiological buffer through a monolayer of human colonic cells deposited on a filter. Caco-2 monolayer permeability has gained considerable acceptance to assess human absorption. Compounds with a Caco-2 monolayer permeability (Papp) similar to or greater than that of propranolol (~30 ⫻ 10⫺6 cm/sec) are considered highly permeable, while compounds with Papp similar to or lower than that of ranitidine (⬍1 ⫻ 10⫺6 cm/sec) are considered poorly permeable. Hurdles associated with determination of permeability of poorly soluble compounds using Caco-2 method have been reviewed.36

3.2.2.3 In-Depth Physicochemical Profiling Once compounds enter the late lead selection or candidate selection phase, more in-depth physicochemical profiling is conducted. The extent of characterization may vary from company to company; however, it likely includes: ● ● ● ●

Experimental pKa and log P (as a function of pH, if necessary) Thermodynamic solubility (as a function of pH) Solution/suspension stability Solid-state characterization

Solid-state characterization typically involves: ● ● ● ● ●

Solid-state stability Feasibility of salt formation Polymorph characterization Particle size, hygroscopicity Dissolution rate

In a more traditional pharmaceutical setting, this characterization would be done during preformulation studies. With the availability of automation and the ability to conduct most of these experiments with small quantities of material, more preformulation activities are being shifted earlier into drug discovery. Recently, Balbach and Korn37 reported a “100 mg approach” to pharmaceutical evaluation of early development compounds. Additional absorption, metabolism, distribution, elimination, and toxicity38 screens may also be conducted at this stage. Overall, the scientific merit of physicochemical profiling is clear. It provides a better assessment of development risks of a compound early on. The important question is how can pharmaceutical companies utilize the vast amount of physicochemical information to advance the right drug candidates to preclinical and clinical testing? Scorecards or flags may be used to rank drug candidates for their physicochemical properties. These scores or flags, however, have to be appropriately weighted with biological activity, safety, and pharmaco*kinetic profiling of compounds. The relative weighting of various factors depends on the specific issues of a discovery program. However, the basic question of how does it help reduce attrition due to unacceptable physicochemical properties remains to be answered in a statistical sense. In 1997, Lipinski et al.14 reported that a trend had been seen since the implementation of the Rule of 5 toward more drug-like properties in Pfizer’s internal drug base. Overall, the goal of a discovery program is to steer the leads in the right direction using computational and HTS approaches and then utilize the in-depth screening tools to select the most optimal compound without undue emphasis on a single parameter such as biological activity. The overall success of a compound is a function of its biological and biopharmaceutical properties.

The Process of New Drug Discovery and Development

3.3

Overview of Dosage-Form Development and Process Scale-Up

Once a compound is selected for development, the pharmaceutics group begins a series of studies to further evaluate the salient physical-chemical and physical-mechanical properties of NCEs to guide the actual design (formulation, recipe) of the dosage form that will carry the drug. These investigations consist of the following steps that span the years that the molecule undergoes clinical evaluation: ● ● ● ● ● ● ● ● ●

Preformulation Consideration of its biopharmaceutical aspects Dosage-form design CTMs manufacture Scale-up studies including technical transfer to manufacturing sites Initiation of long-term stability studies to guide the setting of the expiration date Production of “biobatches” Validation and commercial batches Life-cycle management

While there is a natural desire to front-load these evaluations, this must be balanced against the sad fact that many molecules fail to survive clinical testing: sufficient characterization is performed to help select the “right” molecule to minimize losses at the later, and much more costly, clinical evaluation stages. Great thought and planning are required to optimize the level of effort expended on a single molecule when so many are known to fail during clinical evaluation. There are many reference on the design10,11 and scale-up39 of pharmaceutical dosage forms. It is appropriate to mention that the design of the dosage form be well documented from its inception as this information is required at the NDA stage to explain the development approach. The FDA and its global counterparts are seeking documentation that the product quality and performance are achieved and assured by design of effective and efficient manufacturing processes. The product specifications should be based upon a mechanistic understanding of how the formulation and processing factors impact the product performance. As we describe later, it is important that an ability to effect continuous improvement to the production process with continuous real-time assessment of quality must be incorporated in drug development. Recent regulatory initiatives require that the product/process risks are assessed and mitigated. In this respect, GMP regulations for drugs are moving toward quality standards already required for devices.

3.4

Preformulation

In the pre-1990s development scenario, preformulation activities would start when a compound had been chosen as the lead candidate by the drug discovery unit and advanced for preclinical development prior to testing in man. In the current integrated discoverydevelopment scenario, many of the classical preformulation activities are conducted while screening compounds are in the lead identification or compound selection stage. Irrespective of the actual timing of preformulation studies, preformulation lays the foundation for robust formulation and process development. The purpose of preformulation is to understand the basic physicochemical properties of the drug compound so that the

Integrated Drug Product Development — From Lead Candidate Selection to LCM

challenges in formulation are foreseen and appropriate strategies are designed. Some physicochemical properties are independent of the physical form (i.e., crystal form) but simply are a function of the chemical nature of the compound. They include chemical structure, pKa, partition coefficient, log P, and solution-state stability. A change in the physical form does not affect these properties.

3.4.1

Preformulation Activities: Independent of Solid Form

3.4.1.1 Dissociation Constant The pKa, or dissociation constant, is a measure of the strength of an acid or a base. The dissociation constant of an organic acid or base is defined by the following equilibria: HX ⫹ H 2 O & H 3 O⫹ ⫹ X⫺ (acid) BH⫹ ⫹ H 2 O & H 3 O⫹ ⫹ B (base) Ka ⫽

[H 3 O⫹ ][X⫺ ] (acid) [HX]

[H O⫹ ][B] Ka ⫽ 3 ⫹ (base) [HB ]

(3.1)

whereby the ⫺log Ka is defined as pKa. The pKa of a base is actually that of its conjugate acid. As the numeric value of the dissociation constant increases (i.e., pKa decreases), the acid strength increases. Conversely, as the acid dissociation constant of a base (that of its conjugate acid) increases, the strength of the base decreases. For a more accurate definition of dissociation constants, each concentration term must be replaced by thermodynamic activity. In dilute solutions, concentration of each species is taken to be equal to activity. Activity-based dissociation constants are true equilibrium constants and depend only on temperature. Dissociation constants measured by spectroscopy are “concentration dissociation constants.” Most pKa values in the pharmaceutical literature are measured by ignoring activity effects and therefore are actually concentration dissociation constants or apparent dissociation constants. It is customary to report dissociation constant values at 25°C. Drug dissociation constants are experimentally determined by manual or automated potentiometric titration or by spectrophotometric methods.40 Current methods allow determination of pKa values with drug concentrations as low as 10 to 100 M. For highly insoluble compounds (concentration ⬍1 to 10 M), the Yesuda–Shedlovsky method41 is commonly used where organic cosolvents (i.e., methanol) are employed to improve solubility. The method takes three or more titrations at different cosolvent concentrations, and the result is then extrapolated to pure aqueous system. The dissociation constant can also be determined with less accuracy from the pH–solubility profile using the following modification of Henderson–Hasselbach equation: pH⫺pK a ) for an acid S (1⫹ 10 S ⫽  HA pK a ⫺pH ) for a base SB (1⫹ 10

(3.2)

where SHA or SB is the intrinsic solubility of the unionized form. Some drugs exhibit concentration-dependent self-association in aqueous solutions. The dissociation constant of these compounds may change upon self-association. For example,

The Process of New Drug Discovery and Development

pKa of dexverapamil has been reported to shift from 8.90 to 7.99 in the micellar concentration range.42

3.4.1.2 Partition or Distribution Coefficient The partition coefficient (P) is a measure of how a drug partitions between a water-immiscible lipid (or an organic phase) and water. It is defined as follows: P⫽

[neutral species]o [neutral species]w

(3.3)

log P ⫽ log10 P The distribution coefficient is the partition coefficient at a particular pH. The following equilibrium is often used to define D, with the assumption that only the unionized species partition into the oil or lipid phase: D⫽

[unionized species] o [unionized species]w ⫹[ionized species]w

(pH⫺pK ) 冥 for acids log P ⫺ log 冤1⫹10 log Dat pH ⫽  (pK⫺pH) 冥 for bases log P ⫺ log 冤1⫹ 10

(3.4)

The measurement of log P is important because it has been shown to correlate with biological activity and toxicity.43 As discussed in the previous section, a range of log P (0 to 5) has been shown to be critical for satisfactory oral absorption of drug compounds.

3.4.1.3 Solution Stability Studies Forced degradation studies provide information on drug degradation pathways, potential identification of degradation products in the drug product, structure determination of degradation products, and determination of intrinsic stability of a drug. Regulatory guidance and best practices for conducting forced degradation studies have been reviewed.44 Typical conditions for forced degradation testing include strong acid/base, oxidative, photostability, thermal, and thermal/humidity conditions. Stress conditions are utilized that result in approximately 10% degradation. Stability of the drug in solution over a pH range of 1 to 13 is assessed during preformulation. The purpose is twofold. In the short run, stability in the GI pH range of 1 to 7.5 is important for drug absorption. In the long run, knowledge of solution-state stability is important for overall drug product stability and possible stabilization strategy. Drug degradation in solution typically involves hydrolysis, oxidation, racemization, or photodegradation. The major routes of drug degradation have been thoroughly reviewed.45,46 Although determination of a complete pH-degradation rate profile is desired, it may not always be practical due to limitations of drug supply and time. Also, insufficient solubility in purely aqueous systems may limit determination of pH-degradation rate profiles. Organic cosolvents may be used to increase solubility; however, extrapolation to aqueous conditions must be done with caution. Stability of the drug in a suspended form in the desired buffer can be tested in lieu of solution stability. The stress test results must however, be interpreted in relation to the solubility in the suspension medium. The test may provide an empirical indication of pH stability in the presence of excess water. Satisfactory stability in the GI pH range (1 to 7.5) is important for oral absorption. While there are examples of

Integrated Drug Product Development — From Lead Candidate Selection to LCM

successful solid oral dosage forms of drug that are highly unstable at GI pH (didonasine, esomaprazole magnesium47), excessive degradation in the GI tract limits oral absorption and may require heroic efforts to address the problem via formulation. Although no hard and fast rule exists on acceptable GI stability, Balbach and Korn37 considered ⬍2 to 5% degradation under simulated in vivo conditions (37°C, pH 1.2 to 8, fed and fasted conditions) as acceptable. Higher degradation may require additional investigation. The effect of the GI enzymes on drug stability should also be evaluated.

3.4.2

Preformulation Activities: Dependent on Solid Form

A new drug substance may exist in a multitude of crystalline and salt forms with different physical properties such as shape, melting point, and solubility that can profoundly impact the manufacturing and performance of its dosage form.

3.4.2.1 Solubility Solubility is highly influenced by the solid-state form (e.g., crystalline or amorphous) of the drug. Rigorous solubility studies using the final solid form (i.e., salt form or crystal form) as a function of temperature (i.e., 25 and 37°C) and pH (range 1 to 7.5) are conducted during preformulation. Solubility in nonaqueous solvents is also screened. Solubility in simulated gastrointestinal fluids is also important. For accurate determination of solubility: ●

● ●

Attainment of equilibrium must be ensured by analyzing solution concentration at multiple time points until the concentration does not change considerably (i.e., ⬍5% change in concentration). The pH of the saturated solution must be measured. The solid phase in equilibrium with the saturated solution must be analyzed by techniques such as hot stage microscopy, differential scanning calorimetry, or powder x-ray diffraction, to verify if the starting material has undergone a phase transformation.

3.4.2.2 Salt-Form Selection The selection of an optimal chemical and physical form is an integral part of the development of an NCE. If an NCE is neutral or if its pKa value(s) is not conducive to salt formation, it has to be developed in the neutral form (unless a prodrug is synthesized) and the only form selection involves the selection of its physical (crystal) form. However, if it exists as a free acid or a free base, then the “form” selection involves the selection of both chemical and physical forms. A decision must be made whether a salt or its free acid or base form should be developed. As will be described in Section 3.5, a salt form may lead to a higher dissolution rate and higher bioavailability for a poorly water-soluble drug. For a drug with adequate aqueous solubility, a salt form may not be necessary, unless, of course, a salt provides an advantage with respect to its physical form. In the pharmaceutical industry, salt selection is usually performed by a multidisciplinary team comprising representatives from the drug discovery, chemical development, pharmaceutical development, ADME, and drug safety departments. Serajuddin and Pudipeddi27 reported that the following questions need to be satisfactorily addressed by the team in the selection of an optimal salt form for a compound: “Is the acid or base form preferred because of biopharmaceutical considerations? Is the salt form more suitable? Is the preparation of stable salt forms feasible? Among

The Process of New Drug Discovery and Development

various potential salt forms of a particular drug candidate, which has the most desirable physicochemical and biopharmaceutical properties?” With respect to physical properties, questions involve whether the compound exists in crystalline or amorphous form, and, if crystalline, whether it exhibits polymorphism. At the outset of any salt selection program, it is important to determine whether the salt formation is feasible for the particular compound and, if yes, what counterions are to be used? Although it is generally agreed that a successful salt formation requires that the pKa of a conjugate acid be less than the pKa of the conjugate base to ensure sufficient proton transfer from the acidic to the basic species, the salt formation still remains a “trial and error” endeavor. Hundreds of individual experiments for salt formation of a particular compound are not uncommon. Because of the availability of HTS screening techniques in recent years there is no pressure to limit the number of such experiments. Serajuddin and Pudipeddi27 reported that the number of feasibility experiments can be greatly reduced by studying the solubility vs. pH relationship of the drug and identifying the pHmax (the pH of maximum solubility). The nature of the pH–solubility profile and the position of pHmax depends on pKa, intrinsic solubility (solubility of unionized species), and the solubility of any salt (Ksp) formed. For a basic drug, the pH must be decreased below the pHmax by using the counterion for a salt to be formed, and, for an acidic drug, the pH must be higher than the pHmax. Any counterion that is not capable of changing the pH in this manner may be removed from consideration. While salts may be formed from organic solvents by counterions that are not capable of changing the aqueous pH in this manner, such salts may readily dissociate in an aqueous environment. When the synthesis of multiple salts for a compound is feasible, the number may be narrowed down and the optimal salt may ultimately be selected by characterizing physicochemical properties of solids according to a multitier approach proposed by Morris et al.48

3.4.2.3 Polymorphism Polymorphism is defined as the ability of a substance to exist as two or more crystalline phases that have different arrangements or conformations of the molecules in the crystal lattice. Many drug substances exhibit polymorphism. The definition of polymorphism according to the International Conference on Harmonization (ICH) guideline Q6A49 includes polymorphs, solvates, and amorphous forms. Amorphous solids lack long-range order and do not possess a distinguishable crystal lattice. Solvates are crystal forms containing stoichiometric or nonstoichiometric amounts of solvent in the crystal. When the solvent is water they are termed hydrates. A thorough screening of possible crystal forms is conducted during candidate lead selection or shortly thereafter. Typical methods for generation of polymorphs include sublimation, crystallization from different solvents, vapor diffusion, thermal treatment, melt crystallization, and rapid precipitation. High-throughput screening methods have been reported for polymorph screening.50 Methods for characterization of polymorphs include crystallographic techniques (single crystal and powder x-ray diffraction), microscopic characterization of morphology, thermal characterization (DSC/TGA), solution calorimetry, solid-state spectroscopic methods (IR, Raman, NMR), and solubility and intrinsic dissolution rate methods. Of these, the relative solubility or intrinsic dissolution rate is directly related to the free energy difference, and, hence the relative stability of polymorphs. Thermal data can also be used to assess relative stability of polymorphs. The form with the lowest solubility and, hence, free energy is the most stable form at a given temperature. Other forms would eventually transform to the stable form. The kinetics of crystal nucleation and growth determines the crystal form obtained during crystallization. Sometimes metastable forms are more readily crystallized than the most stable (and often desired) form. The kinetics of transformation of a metastable form to the stable form may be very slow and unpredictable. The

Integrated Drug Product Development — From Lead Candidate Selection to LCM

unexpected appearance of a less soluble polymorphic form of a marketed antiviral drug, ritonavir, caused serious problems with manufacture of its oral formulation.51 The crystal form can have a profound influence on physicochemical properties of a drug substance. Melting point, solubility, stability, and mechanical properties may depend on the crystal form. A difference in solubility of crystal forms may manifest as a difference in bioavailability but the impact would depend on the dose, solubility of each form, and permeability. Formulation factors such as particle size and wettability often complicate the observed difference in the bioavailability of crystal forms. Polymorphism of pharmaceutical solids has been extensively studied.52

3.4.2.4 Solid-State Stability The shelf-life of a product is often predicted by the stability of the drug substance. A thorough investigation of the drug substance stability is, therefore, necessary during preformulation following the identification of the final salt or crystal form. Typical stability testing during preformulation includes accelerated testing for 2 to 4 weeks at high temperatures (50 to 80°C) in dry or moist (75% RH) conditions. Photostability is also conducted as per ICH light sensitivity testing guidelines.53 The criteria for an acceptable solid-state stability are compound-specific. Balbach and Korn37 recommended a degradation of ⬍3 to 5% at 60°C (dry) and ⬍10 to 20% at 60°C (100% RH) in 2 weeks as acceptable. Physical stability (i.e., change in crystal form) must also be investigated under accelerated temperature and humidity conditions. 3.4.2.5 Drug-Excipient Interactions The stability of the pure drug is often greater than when it is formulated. Excipients may facilitate moisture transfer in the drug product or initiate solid-state reactions at the points of contact and adversely impact the stability of the drug substance. For the same reason, it is not uncommon for the drug product stability to decrease when the drug concentration in the formulation is reduced.54 Pharmaceutical excipients can interact chemically with the active pharmaceutical ingredient. Drug-excipient compatibility studies are conducted during preformulation to select the most appropriate excipients for formulation. The following classes of excipients for oral dosage forms are commonly employed: diluents or fillers, binders, disintegrants, glidants, colors, compression aids, lubricants, sweeteners, preservatives, suspending agents, coatings, flavors, and printing inks. A typical drug-excipient compatibility study includes preparation of binary mixtures of the drug and excipients in glass vials. The ratio of the drug to each excipient must be comparable to that in the formulation. It is prudent to set up two sets of samples to bracket the drug concentration foreseen in the final products. Multicomponent mixtures of drug and excipients mimicking prototype formulations can also be made and tested under accelerated conditions. The mixtures are subjected to accelerated stress conditions for a period of 2 to 4 weeks. The samples are analyzed by HPLC for the percent drug remaining and any degradation products formed. Samples are typically stored at a reference condition (⫺20 or 5°C) and at elevated temperatures of 40 to 60°C under dry and moist conditions. Moist conditions are usually obtained by storing samples at 75 or 100% RH, or by adding 10 to 20% (w/w) water to the mixtures. Serajuddin et al.55 described the principles and practice of drug-excipient compatibility studies for selection of solid dosage-form composition. Calorimetric methods for drug-excipient compatibility have also been described.56 Whether physical mixtures represent real drug-excipient(s) interactions in a capsule or tablet formulation has been debated.57 The use of prototype formulations instead of physical mixtures has been suggested. However, drug substance availability and practicality may limit

The Process of New Drug Discovery and Development

such an approach. Although simplistic, a well-conducted drug-excipient compatibility study using physical mixtures can lead to a good understanding of drug stability and robust formulations.55 One must also consider that pharmaceutical excipients can possess acidity (or basicity). Although the chemical nature of excipients (i.e., mannitol, starch, sucrose) may itself be neutral, trace levels of acidic or basic residuals from the manufacturing process often impart acidity or basicity to excipients. Depending on the relative proportion of the drug and its acid/base nature, the microenvironmental pH of the drug product can be influenced by the excipient acidity or basicity. For example, addition of 2% magnesium stearate to a low drug content formulation can result in a microenvironmental pH above 8. An approximate but simple way to measure the microenvironmental pH of a drug-excipient mixture is to measure the pH of a slurry prepared with minimal amount of water. Excipients may also contain impurities that can accelerate drug degradation. For example, the following impurities may be present in the excipients listed with them: aldehydes and reducing sugars in lactose, peroxides and aldehydes in polyethylene glycol, heavy metals in talc, lignin and hemicellulose in microcrystalline cellulose, formaldehyde in starch, and alkaline residues such as magnesium oxide in stearate lubricants.58 The physical state of an excipient (i.e., particle size,59 hydration state,60 and crystallinity61) can also affect drug stability. It is, therefore, important to review the excipient literature data, such as Handbook of Pharmaceutical Excipients,62 prior to its use in a formulation. In assessing drug-excipient compatibility, the same guidelines as in solid-state stability can be used. Typically degradation of less than 5% at high-temperature and humidity conditions (50 to 80°C, 75% RH) is considered acceptable, and higher degradation requires additional investigation under more moderate conditions.

3.4.2.6 Powder Properties of Drug Substance For solid oral dosage forms, powder properties such as powder flow, density, and compactibility are important. For products with low drug concentration (e.g., ⬍5 to 10%), the powder properties of the drug substance are usually less influential than excipients on the overall mechanical properties of the formulation. For products with high drug concentration (i.e., ⬎50%, w/w), powder properties of the drug substance (or active pharmaceutical ingredient, API) may have significant influence on processability of the formulation. Since pharmaceutical scientists often manage to find engineering solutions to address poor powder properties of materials, there are no hard and fast rules for acceptable powder properties. However, consideration of powder properties from the early stages of development can result in more optimized and cost-saving processes. Powder properties of interest include: ● ● ● ● ●

Particle morphology Particle size and particle size distribution, surface area True and relative densities Compaction properties Powder flow properties

Hanco*ck et al.63 reported a comprehensive review of a wide variety of pharmaceutical powders. Investigation of electrostatic properties of the drug substance or drug-excipient mixtures during preformulation has been recommended.64 Such studies may be particularly relevant for dry powder inhalation systems. Optimal compression or compaction properties of powders are critical for a robust solid dosage form. Although prediction of compaction properties of powders is not fully

Integrated Drug Product Development — From Lead Candidate Selection to LCM

possible, tableting indexes65 and other material classification methods66 have been proposed to assess compaction properties of powders. It is clear that no universal method exists to predict processing performance of powders. However, examination of powder properties during preformulation and comparing them with generally well-behaved materials from past experience can identify undesirable powder properties early on and even identify remedial methods such as modification of power properties through changes in crystallization techniques.

3.5

Biopharmaceutical Considerations in Dosage-Form Design

For systemic activity of a drug molecule, it must be absorbed and reach the bloodstream or the site of action if given by the oral, topical, nasal, inhalation, or other route of administration where a barrier between the site of administration and the site of action exists. Even when a drug is injected intravenously or intramuscularly, one must ensure that it is not precipitated at the site of administration and that it reaches the site of action. In the development of dosage forms for a particular drug, a formulator must, therefore, carefully consider various physicochemical, biopharmaceutical, and physiological factors that may influence absorption and transport of drugs. Appropriate formulation strategies must be undertaken to overcome the negative influences of any of these factors on the performance of dosage forms. A large majority of pharmaceutical dosage forms are administered orally, and, in recent years, the drug solubility has become the most difficult challenge in the development of oral dosage forms. For example, in the 1970s and 1980s, when dissolution, bioavailability, and bioequivalence of drugs came under intense scrutiny and many of the related FDA guidelines were issued, a drug with solubility less than 20 g/mL was practically unheard of. Presently, new drug candidates with intrinsic solubility less than 1 g /mL are very common. In addition to solubility, physicochemical factors influencing oral absorption of drugs include dissolution rate, crystal form, particle size, surface area, ionization constant, partition coefficient, and so forth. Among the physiological factors, drug permeability through the GI membrane is of critical importance. Other physiological factors playing important roles in the performance of an oral dosage form are transit times in different regions of the GI tract, GI pH profile, and the presence of bile salts and other surfactants. Physiological differences such as the unfed vs. the fed state also need to be considered. Since solubility and permeability are the two most important factors influencing oral absorption of drugs, the following biopharmaceutical classification system (BCS) for drug substances, based on the work by Amidon et al.,67 has been recommended by the FDA:68 ● ● ● ●

Class I Class II Class III Class IV

— Drug is highly soluble and highly permeable — Drug is poorly soluble, but highly permeable — Drug is highly soluble, but poorly permeable — Drug is both poorly soluble and poorly permeable

For a BCS Class I compound, there are no rate-limiting steps in drug absorption, except gastric emptying, and, therefore, no special drug-delivery consideration may be necessary to make the compound bioavailable. On the other hand, for a BCS Class II compound, appropriate formulation strategy is necessary to overcome the effect of low solubility. For a BCS Class III compound, formulation steps may be taken to enhance drug permeability through the GI membrane, although the options could be

The Process of New Drug Discovery and Development

very limited. Often, the continued development of a BCS Class III compound depends on whether its low bioavailability from a dosage form because of poor permeability is clinically acceptable or not. A BCS Class IV compound presents the most challenging problem for oral delivery. Here, the formulation strategy is often related to enhancing the dissolution rate to deliver maximum drug concentration to the absorption site. As for a Class III compound, formulation options to enhance drug permeability are often limited. Because of the importance of BCS in the development of dosage-form design strategy, an early classification of new drug candidates is essential to identify their formulation hurdles. Some of the physicochemical and physiological factors involved in the design of oral dosage forms are discussed below in more detail.

3.5.1

Physicochemical Factors

3.5.1.1 Solubility Low or poor aqueous solubility is a relative term. Therefore, the solubility of a compound must be considered together with its dose and permeability. A simple approach to assess oral absorption with a drug substance could be the calculation of its maximum absorbable dose (MAD):69 MAD ⫽ S⫻ K a ⫻ SIWV ⫻ SITT

(3.5)

where S is solubility (mg/mL) at pH 6.5, Ka the transintestinal absorption rate constant (per min) based on rat intestinal perfusion experiment, SIWV the small intestinal water volume (250 mL), and SITT the small intestinal transit time (4 h). One limitation of the MAD calculation is that only the aqueous solubility in pH 6.5 buffer is taken into consideration. There are many reports in the literature where the “in vivo solubility” of drugs in the GI tract in the presence of bile salts, lecithin, lipid digestion products, etc., was found to be much higher than that in the buffer alone. Therefore, MAD may be considered to be a conservative guide to potential solubility-limited absorption issues and whether any special dosage forms need to be considered to overcome such issues. Advanced software tools are available to estimate oral absorption.70 For an acidic and basic drug, the solubility over the GI pH range varies depending on the intrinsic solubility (So) of the compound (i.e., solubility of unionized or nonprotonated species), pKa, and the solubility of the salt form.71,72

3.5.1.2 Dissolution Dissolution rate, or simply dissolution, refers to the rate at which a compound dissolves in a medium. The dissolution rate may be expressed by the Nernst–Brunner diffusion layer form of the Noyes–Whitney equation: J⫽

dm D ⫽ (cs ⫺ cb ) A dt h

(3.6)

where J is the flux, defined as the amount of material dissolved in unit time per unit surface area (A) of the dissolving solid, D the diffusion coefficient (diffusivity) of the solute, cs the saturation solubility that exists at the interface of the dissolving solid and the dissolution medium, and cb the concentration of drug at a particular time in the © 2006 by Informa Healthcare USA, Inc.

Integrated Drug Product Development — From Lead Candidate Selection to LCM

bulk dissolution medium. Under “sink” conditions (cb ⬍ 10%cs), the above equation is reduced to J⬇

D cs h

(3.7)

From the above equation, the dissolution rate is proportional to solubility at the early stage of dissolution. However, there is an important distinction between dissolution and solubility; solubility implies that the process of dissolution has reached equilibrium and the solution is saturated. The importance of dissolution on drug absorption and bioavailability may be described in relation to the concept of dissolution number, Dn, introduced by Amidon et al.67 Dn may be defined as

Dn ⫽

tres tdiss

(3.8)

where tres is the mean residence time of drug in the GI tract and tdiss the time required for a particle of the drug to dissolve. It is evident from the above equation that the higher the Dn, the better the drug absorption, and a maximal drug absorption may be expected when Dn ⬎ 1, i.e., tres ⬎ tdiss. However, for the most poorly water-soluble drugs, Dn ⬍ 1 and, as a result a dissolution rate-limited absorption is expected. To ensure bioavailability of poorly soluble drugs, formulation strategies must, therefore, involve increasing the dissolution rate such that the full dose is dissolved during the GI residence time, i.e., Dn becomes equal to or greater than 1. Some of the current methods of increasing dissolution rates of drugs are particle size reduction, salt formation, and development of the optimized delivery systems, such as solid dispersion, soft gelatin encapsulation, etc. The effect of particle size on drug absorption as a function of dose and drug solubility was analyzed by Johnson and Swindell,73 where the absorption rate constant was assumed to be 0.001 min⫺1. The results are shown in Figure 3.1; Figure 3.1(a) shows that for dissolution ratelimited absorption (Dn ⬍ 1), the fraction of drug absorbed at any particle size will decrease with the increase in dose. On the other hand, if the dose is kept constant, the fraction of drug absorbed will increase with an increase in drug solubility, and the particle size becomes practically irrelevant for drugs with a solubility of 1 mg/mL at a dose of 1 mg (Figure 3.1(b)). Salt formation increases the dissolution rate by modifying pH and increasing the drug solubility in the diffusion layer at the surface of the dissolving solid.74 Depending on the pH in the GI fluid, a drug may precipitate out in its respective free acid or base form; however, if redissolution of the precipitated form is relatively rapid,75 faster drug absorption is expected from the salt form. Precipitation, in the presence of food, may confound this expectation. Although salt formation increases drug dissolution rate in most cases, exceptions exist. Under certain situations, a salt may convert into its free acid or base form during dissolution directly on the surface of the dissolving solid, thus coating the drug surface and preventing further dissolution at a higher rate. In such a case, the salt formation may not provide the desired advantage and a free acid or base form may be preferred.76 A DDS such as a solid dispersion increases dissolution rate by dispersing the drug either molecularly or in an amorphous state in a water-soluble carrier.9 When given as a solution in a pharmaceutically acceptable organic- or lipid-based solvent that is either

The Process of New Drug Discovery and Development 14

Dose (mg): 100 10 2 1

% Absorption

12 10 8 6 4 2 0

0 10 20 30 40 50 60 70 80 90 100 (a)

Mean particle size (µm) 70 Solubility: (mg/mL) 60

0.001 0.01 0.1 1

% Absorption

50 40 30 20 FIGURE 3.1 Computed percent of dose absorbed at 6 h vs. mean particle size, with an absorption rate constant of 0.001 min⫺1: (a) at doses from 1 to 100 mg, with a solubility of 0.001 mg/mL; and (b) at solubility from 0.001 to 1.0 mg/mL, with a dose of 1 mg.

10 0 0 (b)

90 100

Mean particle size (µm)

encapsulated in a soft gelatin capsule or packaged in a bottle, the drug immediately saturates the GI fluid after dosing and the excess drug may precipitate in a finely divided state that redissolves relatively rapidly.77 3.5.2

Physiological Factors

In any formulation development, the GI physiology that can influence dosage-form performance must be kept in mind. The GI contents play important roles, since even in the fasted state in humans, the in vivo dissolution medium is a complex and highly variable milieu consisting of various bile salts, electrolytes, proteins, cholesterol, and other lipids. The GI pH is another factor that plays a critical role in the performance of the dosage form. The pH gradient in humans begins with a pH of 1 to 2 in the stomach, followed by a broader pH range of 5 to 8 in the small intestine, with the intermediate range of pH values of around 5 to 6 being found in the duodenum. Colonic absorption in the last segment of the GI tract occurs in an environment with a pH of 7 to 8. The average pH values significantly differ between the fed and the fasted state. Finally, gastric emptying time and intestinal transit time are very important for drug absorption. The majority of the liquid gets emptied from the stomach within 1 h of administration.78 Food and other solid materials, on the other hand, takes 2 to 3 h for half of the content to be emptied. The general range of the small intestinal transit time does not differ greatly and usually ranges from 3 to 4 h.79

Integrated Drug Product Development — From Lead Candidate Selection to LCM

3.5.2.1 Assessment of In Vivo Performance Different formulation principles, dosage forms, and DDSs are commonly evaluated in animal models, and attempts are made to predict human absorption on the basis of such studies.80 Human studies are also conducted in some cases to confirm predictions from animal models. Chiou et al.81,82 demonstrated that there is a highly significant correlation of absorption (r2 ⫽ 0.97) between humans and rats with a slope near unity. In comparison, the correlation of absorption between dog and human was poor (r2 ⫽ 0.512) as compared to that between rat and human (r2 ⫽ 0.97). Therefore, although dog has been commonly employed as an animal model for studying oral absorption in drug discovery and development, one may need to exercise caution in the interpretation of data obtained. 3.5.2.2 In Vitro–In Vivo Correlation The various biopharmaceutical considerations discussed above can help in correlating in vitro properties of dosage forms with their in vivo performance.83 In vitro–in vivo correlation (IV–IVC), when successfully developed, can serve as a surrogate of in vivo tests in evaluating prototype formulations or DDSs during development and reduce the number of animal experiments. This may also help in obtaining biowaivers from regulatory agencies, when applicable, and especially for BCS Class II compounds mentioned above. It can also support and validate the use of dissolution methods and specifications during drug development and commercialization. The IV–IVC is generally established by comparing in vitro dissolution of drug with certain in vivo PK parameters. There are certain FDA guidelines for this purpose, where the correlations are categorized as Level A, Level B, Level C, and multiple Level C correlations.84 Level A correlation is generally linear and represents the correlation of in vitro dissolution with the drug fraction absorbed, which is obtained from the deconvoluted in vivo plasma levels. The Level B correlation is more limited in nature, and it does not uniquely reflect the actual in vivo plasma level curve; here, the mean in vitro dissolution time is compared with the mean in vivo residence or dissolution time. A Level C correlation establishes a single-point relationship between a dissolution parameter and a pharmaco*kinetic parameter. While a multiple Level C correlation relates one or several pharmaco*kinetic parameters of interest to the amount of drug dissolved at several time points of the dissolution profile, a Level C correlation can be useful in the early stages of formulation development when pilot formulations are being selected. For the purpose of IV–IVC, it is essential that the dissolution medium is biorelevant such that in vitro dissolution testing is indicative of in vivo dissolution of the dosage form. Dressman85 published extensively on the identification and selection of biorelevant dissolution media.

3.6

Clinical Formulation Development: Clinical Trial Materials

A pharmaceutical dosage form is a means to deliver the active ingredient in an efficacious, stable, and elegant form to meet a medical need. The requirements and specifications of the product will depend on the scope and extent of its intended use in the various clinical phases and then through commercialization. Once the decision is made to move a drug candidate into clinical studies, the interaction of formulation scientists begins to shift from preclinical groups (medicinal chemistry, ADME) to those from development (analytical, process chemistry, clinical), marketing, and operations groups.

The Process of New Drug Discovery and Development

The formulation, manufacture, packaging, and labeling of CTMs precede the individual phases of clinical development. Clinical development of NCEs is broadly divided into Phases I, II, III, and then IV (postmarketing) studies. Phase I studies evaluate the pharmaco*kinetics and safety of the drug in a small number (tens) of healthy volunteers. Phase I studies are sometimes conducted in a small patient population (Proof of Concept studies) with a specific objective such as the validation of the relevance of preclinical models in man. The purpose of these studies may be the rapid elimination of potential failures from the pipeline, definition of biological markers for efficacy or toxicity, or demonstration of early evidence of efficacy. These studies have a potential go/no-go decision criteria such as safety, tolerability, bioavailability/PK, pharmacodynamics, and efficacy. Dosage forms used in Phase I or Proof of Concept studies must be developed with the objectives of the clinical study in mind. Phase II studies encompass a detailed assessment of the compound’s safety and efficacy in a larger patient population (a few-to-several hundreds of patients). It is important that any formulation selected for these studies must be based on sound biopharmaceutical and pharmaceutical technology principles. Phase III clinical studies, also referred to as pivotal studies, involve several thousands of patients in multiple clinical centers, which are often in multiple countries. The aim of these studies is to demonstrate long-term efficacy and safety of the drug. Since these studies are vital in the approval of the drug, the dosage form plays a very critical role.

3.6.1

Phase I Clinical Trial Material

A decision tree approach for reducing the time to develop and manufacture formulations for the first oral dose in humans has been described by Hariharan et al.86 and is reproduced in Scheme 3.1. The report summarized numerous approaches to the development and manufacture of Phase I formulations. Additional examples of rapid extemporaneous solution or suspension formulations for Phase I studies have been reported.87,88 In deciding the appropriate approach for early clinical studies, it is important to consider the biopharmaceutical properties of the drug substance and the goals of the clinical study. Practical considerations, such as the actual supply of the bulk drug and the time frame allotted for development, enter into the picture. The advantage of using extemporaneous formulations is the short development timelines (a few months) and the minimal drug substance requirements (a few hundred grams depending on the dose). Additional benefits include high-dose flexibility, minimal compatibility or formulation development, and minimal analytical work. The disadvantages include possible unpleasant taste, patient compliance issues, and dosing inconvenience for multiple-dose studies. For poorly soluble compounds, use of a nonaqueous solution may result in high systemic exposure that may be difficult to reproduce later on with conventional formulations. The “drug-in-capsule” approach, where the neat drug substance is encapsulated in hard gelatin capsules, has similar time, material, and resource advantages but is limited to compounds that exhibit rapid dissolution. An intermediate approach for Phase I CTM is a “formulated capsule” approach, where the preformulation and compatibility data are optimally utilized and a dosage form that has scale-up potential is developed. The pharmaceutical composition is chosen on the basis of excipient compatibility and preliminary in vitro dissolution data. Extensive design of experiments is not conducted at this stage. Processing conditions are derived from past experience or simple reasoning. Several hundred grams of the drug substance are typically necessary to develop a formulated capsule product. However, the advantage

Is there any reason to front-load?

FTIM goal: Speed into humans to determine pK and tolerance

Route of administration

1. 2. Oral

Parenteral

Simple solution Sterile aqueous suspension or emulsion Optimized freeze-dried formulation

Parallel (for speed in FTIM and development) 1.

Parenteral

Oral

Yes

1. 2.

Does API have adequate solubility? Is exposure not likely an issue?

Minimize risk of inadequate blood exposure Speed for Phase II and III development

Route of administration

Need for confirmed volume? (e.g., cytotoxic)

Yes

Simple or frozen solution Simple freeze-dried formulation

Liquid system in a capsule Thermosoftening system in capsule: Gelucire or PEGbased suspension in capsule Semiformulation blend in capsule

Can toxicological formulation be used for FTIM?

1. 2. 3. 4. 5. 6. 7.

Milled suspension Aqueous cosolvents Solubilized form (SEDDS, suspension or solution in lipid delivery system) Cyclodextrins High-energy solids Formulated solid dosage forms Nontraditional dosage forms

1. Degree of dosing flexibility required or desirable

High mg/kg dose 2.

RTU solution (frozen or unfrozen) RTU suspension (unmilled or milled; aqueous or nonaqueous)

Moderate low dose Yes

Yes 1.

Use CIC, CIB, or CICIB to FTIM

Take toxicological formulation to FTIM

2. 3.

Semiformulated blend in capsule Beads in capsule Tablet

SCHEME 3.1 A decision tree approach for the development of first human dose. FTIM ⫽ first in man; CIC ⫽ chemical in capsule; CIB ⫽ chemical in bottle; CICIB ⫽ chemical in capsule in bottle; RTU ⫽ ready to use. Reproduced from Hariharan, M. et al., Pharm. Technol., 68, 2003. © 2006 by Informa Healthcare USA, Inc.

Integrated Drug Product Development — From Lead Candidate Selection to LCM

New chemical entitybiological and physicochemical characterization

The Process of New Drug Discovery and Development

of a well-formulated capsule product is that it may be processed on a low-to-medium speed automatic equipment to meet the demands of a larger clinical study, if needed. A formulated tablet cannot be ruled out for Phase I; there is anecdotal information that a few drug companies use this approach. However, tablet development activities require more material and resources. While formulation design activities are in progress, the development of analytical methods must be initiated to develop an assay to separate the drug from its excipients, a stability-indicating method, and other tests (content uniformity, dissolution) that will be required to release a batch for human use. Limited drug substance and drug product stability testing are required to support a shelf-life (typically a few months, preferably at room temperature or under refrigeration) sufficient to conduct the Phase I study.

3.6.2

Phase II Clinical Trial Material

Extemporaneous formulations such as “powder-in-bottle” are unlikely to meet the demands of Phase II. Formulated capsules or tablets are typically necessary. The same dosage form used in Phase I studies may be continued for Phase II studies, if all the required dosage strengths can be supported and medium-to-large-scale (100,000 or more units) manufacturing of the dosage form is feasible. Alternatively, development of a more robust formulation closer to the desired commercial form may be undertaken. The drug substance requirements for such a Phase II dosage form may be in the range of the tens of kilograms, depending on the dose. The chemical development of the drug substance must also be well advanced to produce large quantities of the drug substance with minimal change in the impurity profile that is used in the toxicology program. The design of the clinical study may also influence the type of the dosage form. For example, in a doubleblind study, the dosage-form presentation must be designed in a way to mask the difference in appearance between various dose strengths. It is often necessary to prepare placebos or active controls that consist of a marketed product disguised (masked) to look like the CTM. This is not a trivial exercise as consideration must be given to insure any changes made to the commercial product in the blinding process do not alter its bioavailability or stability. A popular approach is to place the commercial product into a hard gelatin capsule that looks like the capsules used for the investigative drug. When this is not possible (due to size issues) then the “double-dummy” approach is required. Here, a lookalike placebo of the marketed product must be prepared, which is then used in various schemes with the investigative product to provide the required dose combinations. Preparation of look-alike products, however, can pose difficult problems when the tablet surfaces are embossed with logos. For the development of robust formulation and process, critical formulation issues must be first identified from the preformulation work. Critical issues may include solubility or dissolution rate for poorly soluble drugs, drug stability and stabilization, or processing difficulties due to poor powder properties of the drug substance (for high-dose formulations). In the initial stages of product development, in vitro dissolution in physiologically meaningful media must be utilized to guide the development of the prototype formulations.68,89 Biopharmaceutical support in formulation development has been reviewed.90 If drug stability is identified as a potential issue, more thorough and careful drug-excipient compatibility studies may be necessary. On the basis of results of forced degradation and compatibility studies, a stabilization mechanism may be identified. Yoshioka and Stella46 have thoroughly reviewed drug stability and stabilization. Stabilization strategies such as incorporation of antioxidants,91 protection from moisture, or use of pH modifiers92 may be considered. If a “functional” excipient such as an antioxidant or a pH modifier is

Integrated Drug Product Development — From Lead Candidate Selection to LCM

required, then tests to monitor its performance (i.e., consumption of the antioxidant during storage) must be developed. A pragmatic but critical factor in Phase II formulation development is the identification of the dosage strengths by the clinical development scientists. Since Phase II studies cover a large dose range, the number of dosage strengths may be quite large. The selection of minimal number of dosage strengths that can still support the needs of the clinical studies is best handled by a team comprising pharmaceutical scientists, pharmaco*kineticists, and the clinical study and clinical supply coordinators. Individual strategies must be developed for both the high- and low-dose strengths. The limitations on the high-dose strength may include dissolution, processability, and size of the unit. Whereas the lowdose strengths may face mass or content uniformity or stability issues. Once the critical formulation issues are identified, a systematic approach is necessary to identify the optimal composition and process. Statistical design of experiments provides a systematic basis for screening of the prototype formulations or variants. It is important to correctly identify the desired responses to be optimized (i.e., dissolution, chemical stability, processing of a unit operation). It is, however, not uncommon, due to lack of the sufficient drug substance, to use empirical approaches to the formulation development. Systematic trial and error93 experiments can also result in the identification of optimum conditions in a semistatistical way. Each of the development steps such as formulation composition, identification, and processing unit operations can be the subject of statistical optimization. However, time and resource limitations often preclude using such multiple optimizations. Extensive statistical designs to optimize processing conditions are often deferred to at a later stage. Use of expert systems94 to build consistency in design and test criteria has been described. In developing Phase II formulations with scale-up potential, manufacturing equipment that operates on the same principle as the production equipment should be used as much as possible. A difference in the operating principle such as low shear vs. high shear granulation may yield different product characteristics.95 About three-to-four variants are identified from the screening studies for stress stability testing. During Phase I and II clinical studies the duration of the stability study will depend on the extent of the clinical studies. One-to-two-year stability studies under ICH96 recommended storage conditions are desired with both accelerated and real-time storage conditions. Stability studies monitor changes in drug product performance characteristics including potency, generation of degradation products, product appearance, and dissolution. Stability data from these studies are utilized to develop the final product and to select the most appropriate packaging conditions. Hard gelatin capsules and tablets are among the most common dosage forms for Phase II clinical studies. Capsules are considered to be more flexible and forgiving than tablets for early clinical studies. Capsules may also offer an easy means of blinding as noted previously for the use of active controls. The filling capacity of capsules is limited (usually ⬍400 mg fill weight of powders for a size 0 capsule) and large doses may require capsules that are not easy to swallow. Production of tablets at medium-to-large scale requires more stringent control of powder properties due to the high-speed compression step. Processing of tablets and the physics of tablet compaction have been the subject of extensive investigation and voluminous literature exists on the topic. Overall, the development of a robust formulation with scale-up potential for Phase II studies involves integration of physicochemical, biopharmaceutical, and technical considerations. Whether a rudimentary formulated capsule or a more robust formulation closer to the commercial form will be used in Phase II studies will depend on the company policy, material cost, the complexity of clinical design, and the development strategy.

38 3.6.3

The Process of New Drug Discovery and Development Phase III Clinical Trial Material

The purpose of Phase III clinical studies is to demonstrate the long-term safety and efficacy of the drug product. To insure the same safety and efficacy in the commercial product, it is necessary that the Phase III CTM should be close to or identical to the final commercial form when possible. A brief overview of the regulatory requirements for the commercial dosage form may help contribute to the understanding of the strategies that can be used for Phase III CTM.96 According to the ICH guidelines, data from stability studies should be provided on at least three primary batches of the drug product. The primary batches should be of the same formulation and packaged in the same container closure system as proposed for marketing. Two of the primary batches should be of pilot scale or larger (i.e., the greater of 1/10th of the production scale or 100,000 units) and the third batch may be of smaller (lab) scale. Production scale batches can replace pilot batches. The manufacturing processes for the product must simulate those to be applied to the production batches and should provide a product of the same quality. The stability study should typically cover a minimum period of 12 months at the long-term storage condition (i.e., 25°C/60% RH). As a result of these requirements, the formulation, manufacturing process, and packaging of the drug product must be finalized at about 1.5 to 2 or more years prior to the filing of a regulatory application for an NCE. If the company policy is to use at least pilot-scale batches for Phase III clinical studies, the formulation and manufacturing process must be finalized prior to the initiation of the pivotal clinical studies. This activity may require large quantities of the drug substance (i.e., hundreds of kilograms depending on the dose and dosage strengths). Additionally, the synthesis of the drug substance must be established.96 The development of Phase III CTM must occur in parallel to Phase II clinical studies. The advantage of this approach is that a robust and well-developed formulation is utilized in Phase III clinical studies and scale-up risk is reduced. Scale-up and process validation can occur subsequently as per regulatory guidelines.97 The concepts of design for manufacturability and process analytics technologies98 are best incorporated into the initial design phases to promote the generation of robust, cost-efficient processes. Alternatively, the dosage form may be scaled-up to production scale at the launch site and the final commercial dosage form may be utilized for Phase III studies. This approach requires even more of the drug substance and a fully validated drug substance synthesis. Depending on the project strategy and the availability of resources, this approach may eliminate the risk of scale-up-related delays in launch. A detailed description of product scale-up and validation are beyond the scope of this chapter, but extensive literature is available on this topic.41 If the company resources, project timelines, or the complexity of the pivotal study design do not allow for the development of a pilot or the production scale formulation, the Phase III CTM may be significantly different from the final commercial form. The dosage form used for Phase III must, however, be robust enough to provide an uninterrupted drug supply for clinical studies. The Phase I or II CTM with the appropriate modifications (i.e., dosage strengths, blinding, manufacturability) may be used to meet the specific goals of Phase III. When significant composition or process changes are made to the CTM, a biopharmaceutical assessment must be made on the potential impact on systemic exposure. Clinical studies to establish bioequivalence between the previously used CTM and the new CTM or with the intended commercial dosage form may be required. The systemic exposure from early clinical formulations serves as a benchmark for later formulation changes. Demonstration of the bioequivalence of the Phase III CTM with the commercial dosage form, if different, may be necessary before regulatory approval and also in the case of certain postapproval changes.99 In a bioequivalence

Integrated Drug Product Development — From Lead Candidate Selection to LCM

study a test product (such as an intended commercial dosage form) is compared with a reference formulation (i.e., a Phase III CTM material or an innovator’s product in case of pharmaceutical equivalents) according to the guidelines established by the regulatory authorities. In the case of formulation changes for a CTM, the systemic exposure is compared to the earlier benchmark formulations, and substitution by the new product may be made on the basis of the product’s efficacy and safety. Two products may be assessed to be equivalent even if the bioequivalence criteria are not strictly met, provided the new product does not compromise the safety and efficacy established by the previous product.100 It may be possible to obtain a waiver for bioavailability or bioequivalence of formulations, depending on the BCS classification and the availability of established IV–IVC as discussed in the previous sections.

3.7

Nonoral Routes of Administration

As noted earlier, this chapter is focused on the design and evaluation of oral dosage forms. A few thoughts are provided to introduce some of the different biopharmaceutical issues that confront the design of some nonoral dosage forms. 3.7.1

Parenteral Systems

Many drugs are administered as parenterals for speed of action because the patient is unable to take oral medication or because the drug is a macromolecule such as a protein that is unable to be orally absorbed intact due to stability and permeability issues. The U.S. Pharmacopoeia defines parenteral articles as preparations intended for injection through the skin or other external boundary tissue, rather than through the alimentary canal. They include intravenous, intramuscular, or subcutaneous injections. Intravenous injections are classified as small volume (⬍100 mL per container) or large volume (⬎100 mL per container) injections. The majority of parenteral dosage forms are supplied as ready-to-use solutions or reconstituted into solutions prior to administration. Suspension formulations may also be used,101 although their use is more limited to a subcutaneous (i.e., Novolin Penfill®; NOVO Nordisk) or intramuscular (i.e., Sandostatin LAR Depot®; Novartis) injection. Intravenous use of disperse systems is possible but limited (i.e., Doxil® Injection; Ortho Biotec). The decision to develop a commercial parenteral dosage form must be made during drug discovery itself because the developability criteria are different than those of an oral dosage form. Additionally, extensive preformulation studies must be conducted to fully understand the pH-solubility and pH-stability properties of the drug substance: equilibrium solubility and long-term solution stability are important for parenteral dosage-form development.102,103 The drugs generated by the “biopharmaceutical” industry are typically macromolecules such as proteins. Macromolecular drugs are generated via genetic engineering/fermentation techniques and purified in an aqueous solution. The material at the final processing step is measured into a package for ultimate delivery to the patient. If further processing such as lyophilization is required to maintain adequate shelf-life of the macromolecule, the composition (excipients) of the final product is carefully selected to maintain the physical and chemical stability of the bulk drug. This requires the coordination of the protein/processing specialist and the formulator to insure that the requisite stabilizers, pH control, and preservatives are added to the broth in the final processing steps.

The Process of New Drug Discovery and Development

Subcutaneous and intramuscular injections are administered in small volumes as a bolus (1 to 4 mL). Intravenous injections can be given as bolus (typically ⬍5 mL) with larger volumes administered by infusion. Due to the small volume of bolus injections, high drug concentrations (up to 100 to 200 mg/mL) may be required for administration of large doses. As a result, solubility enhancement is a major consideration for parenteral dosage forms. Solubilization principles for parenteral dosage forms have been reviewed.104 Common approaches to enhance solubility include pH adjustment, addition of a cosolvent, addition of a surfactant, complexation, or a combination of these approaches. Ideally, parenteral dosage forms should have a neutral pH. Injection volume and potential for pain105 at the site of injection must be carefully evaluated in the selection of the pH of the formulation. Judicious selection of excipients for parenteral dosage forms is critical due to their systemic administration.106,107 Excipients of a parenteral dosage form may have a significant effect on product safety including injection site irritation or pain. Permissible excipients for parenteral dosage forms are far less than those for oral dosage forms. Parenteral products must be sterile, particulate-free, and should be isotonic. An osmolarity of 280 to 290 mOsmol/L is desirable. Slightly hypertonic solutions are permissible but hypotonic solutions must be avoided to prevent hemolysis. It is essential that any product for injection be sterilized by a validated process starting with Phase 1 CTM. The method of sterilization (heat, filtration, high energy) of the product must be carefully considered as it can have a profound influence on the stability of the drug. This is also true for the choice of the packaging components as they too can influence stability by releasing materials (such as vulcanizing materials from rubber stoppers) capable of interacting with the drug. Strategies for development of CTMs of parenteral dosage forms depend on their end use. For example, the requirements of an intravenous formulation intended only to determine the absolute bioavailability of a solid dosage form are more flexible than that being designed for commercial use. For example, if the stability of a solution formulation is limited, it may be supplied frozen to conduct a “one-time” bioavailability study. On the other hand, a parenteral formulation for long-term clinical use may require more extensive development to overcome such liabilities. If the stability of a solution formulation is not suitable for the desired storage condition (room temperature or refrigeration), a lyophilized formulation may be necessary. Principles and practice of lyophilization have been reviewed.108 Bioavailability and bioequivalence considerations of parenteral dosage forms (nonintravenous) have been reviewed.109 3.7.2

Inhalation Systems

To treat diseases of the upper airways, drugs have been formulated as dry powders for inhalation, as solutions for nebulization, or in pressurized metered dose inhalers with the goal of delivering the drug topically in the bronchial region. These systems are inherently complex and can be inefficient as they tend to produce and deliver large particles that lodge in the back of the throat.110 The FDA considers inhalation dosage forms as unique as their performance is markedly influenced by the formulation, the design of the packaging (container and valve) as well as patient-controlled factors. The factors such as rate and extent of the breath can influence the delivered dose uniformity and the particle size distribution. An additional challenge for the formulator is the limited number of excipients that have been demonstrated to be safe when delivered into the lung, which must be carefully sourced and controlled.111 The formulator’s primary focus is on how to produce particles of the requisite size and how to maintain them as such prior to and during administration. For solid drugs, the particles are targeted to be produced in the range of 1 to 5 m by milling,

Integrated Drug Product Development — From Lead Candidate Selection to LCM

controlled precipitation, or spray drying. When a drug is sufficiently soluble in water its solutions can be processed to create droplets in the appropriate particle size range by nebulizers (solutions must be sterile). Such fine particles are subject to agglomeration due to static charges or humidity and require exquisite control of the environment (atmospheric such as humidity, and local such as excipients) during production, packaging, and storage as changes in the PSD can profoundly influence the efficiency of delivery. These constraints can have a negative impact on the cost of goods. The pulmonary humidity may also result in changes to the PSD as the particles travel the length of the airways.112 A recent approach mitigates these concerns by producing the particles on demand via vaporization of the drug followed by condensation into a controlled PSD as the patient inhales.113 This approach is capable of producing nanometer- or micron-sized particles. In addition to local lung delivery, pharmaceutical scientists have noted the extensive surface area of the lower, gas-transport region of the lung, the alveolar region, and have been seeking a means of reproducibly administering potent therapeutic agents into this region for rapid transport into the systemic circulation thus bypassing the GI tract and liver. The pharmaco*kinetic profile from this noninvasive method mimics that of IV injection.114 The ability of a vaporized (smoked) drug to exert a rapid CNS effect was well known to our great ancestors and such rapid delivery is currently under consideration for treatment of acute and episodic conditions such as migraine or breakthrough pain. There is currently great interest to deliver macromolecules such as insulin in this manner employing the drug in solution or dry powders.115

3.8

Drug-Delivery Systems

From the perspective of absolute efficiency, most dosage forms are failures. Far too much drug is poured into the body so that a few molecules can reach the site of the desired pharmacologic response.116 As already noted, the drug must pass many anatomical and physiologic (pH, enzymes) barriers on its journey to the target. Formulators often talk of a magic bullet that, in their fondest dreams, delivers the drug to the exact site needed and no more. Since the 1960s, when pharmaceutical scientists began to understand how the physicochemical properties of a drug influenced the ability of the drug to dissolve, remain stable in a changing pH milieu, be transported across the epithelium by passive and active transport, and how it would fair as it came in contact with numerous metabolizing systems (biopharmaceutical factors), great strides have been made in designing and improving the performance of oral dosage forms. With time, similar biopharmaceutical factors were revealed for other routes of delivery that enabled novel dosage forms to be prepared. Examples include transdermal patches to deliver the drug into the systemic circulation over days and weeks from a single application or polymeric microspheres capable of prolonged residence time in muscle tissue from where they could slowly release the needed peptide. New materials of construction (excipients, carriers) were required to create these novel systems, which added to the formulators’ palette more than the ethanol (for elixirs) and starch (for tablets) used for decades. Though, we must be cautious when we add a component to our delivery system to nonspecifically overcome a barrier that has been evolved through millennia, as this can induce safety problems, such as been found with the use of dermal penetration enhancers for use in transdermal delivery patches. Therefore, the safety of any new excipient has to be demonstrated via chronic toxicity studies before they can be seriously considered for use. However, as our understanding of

The Process of New Drug Discovery and Development

the anatomical microenvironments improves and new construction materials become available, so does our desire to further fashion a magic bullet. A DDS, which is also referred to as “novel” DDS, is a relative term. What we consider common dosage forms today were novel a few decades ago; one notable example of novel formulations, as we know them today, was the development of the tiny, little time-pills termed Spansules® by a group at SK&F.117 These provided prolonged release of drugs, which simplified dosing and attenuated adverse reactions. The use of pharmaco*kinetic data to guide the design of the optimized delivery system was at that time a breakthrough. Given the complexity in the design and subsequent development of novel DDS, it is impossible for each Big Pharma company to internally master many of the technological breakthroughs. This has led to the formation of a new industry — DDSs and tools — with an estimated several hundred participating companies. Like their big brothers, these companies also function with multidisciplinary teams with thorough knowledge of the anatomical and molecular barriers that must be overcome and the ability to design strategies to surmount them. While “traditional” drug-delivery companies employed new techniques to produce dosage forms (e.g., osmotic pumps for controlled release, nanocrystal milling), many of the newer companies are working at the molecular level to create drugs with better transport properties. Just as a new molecular entity must pass hurdles to demonstrate its safety and efficacy, so too must new delivery systems meet these requirements. Furthermore, the cost of goods of the new technology must be appropriate to the benefit provided. As noted earlier, there is a long path from the first formulation to one robust enough to routinely prepare commercial quantities. These processing issues are even more important and really daunting for a novel DDS that has never been made in more than test tube quantity. This was the situation confronting the liposome companies founded in the early 1980s to commercialize this carrier technology. The issues of the “three-esses” prevailed — stability, sterility, and scale-up — and had to be overcome to make the technology worthy of consideration as a new dosage form. It took a decade and tens of millions of dollars spent by several companies to accomplish the task — performed under cGMP.118 The situation is even more (financially and emotionally) perilous when a new DDS is used to support the clinical evaluation of a new drug. Thus, the use of an optimized delivery system oftentimes has lagged the introduction of a new medicine. According to an estimate in early 2000s, 12% of the pharmaceutical market comprises DDSs. This share will definitely increase in the future as it is evident from some of the considerations made above. Among the market share of various DDSs, oral dosage forms account for approximately 60%, inhalation products for 20%, transdermal about 10%, and injectables around 9%.119 Drug-delivery systems are essentially specialized dosage forms developed to overcome the limitations of conventional dosage forms, such as simple tablets, capsules, injectable solutions, etc. Some of the reasons behind the development of oral DDSs are listed below: ● ● ● ● ● ● ● ● ● ● ●

Overcome drug developability issues Unfavorable lipophilicity Poor permeability Poor stability Overcome physiological hurdles Relatively short half-life First-pass effect GI residence time GI pH Drug stability in GI fluid Effect of food and other GI contents

Integrated Drug Product Development — From Lead Candidate Selection to LCM ● ● ● ● ● ● ● ● ● ●

Enhance absorption rate and bioavailability Increase solubility Increase dissolution rate Increase patient convenience Once-a-day (compared to multiple-time dosing) Geriatric formulation Pediatric formulation Intellectual property New formulation principles Product and process improvement

Many different technologies to address these drug-delivery issues are either available or are emerging through the efforts of Big Pharma or various smaller drug-delivery companies. For certain specific applications, multiple (commodity) technologies may be available. For example, matrix tablets, coated tablets, osmotic release systems, single unit vs. multiple units, etc., are available for prolonged-release dosage forms. Of course, each of these systems has its own advantages and disadvantages. The need for DDSs is possibly the greatest in the case of biotechnologically derived products that cannot be orally absorbed, such as peptides, proteins, and oligonucleotides. Some of the major formulation issues with such products are: ● ● ● ●

Invasive dosage forms (injection) Short half-life Poor stability of drugs or dosage forms Lack of drug targeting

Many different approaches are being applied to resolve these issues. Some of the techniques include biodegradable microspheres, PEGylation, liposomal delivery, electroporation, prodrug and conjugate formation, dry powder inhalers, supercritical fluid-based nanoparticles, viral and nonviral vectors, and so forth. Some of the successes in this area include first microsphere sustained release formulation of a peptide LHRH (Lupron Depot®; TAP/Takeda) in 1989; first PEGylated sustained release formulation of a protein adenosine deaminase (Adagen®; Enzon) in 1990; first microsphere sustained release formulation of a recombinant protein, human growth hormone (Nutropin Depot®; Genentech) in 2000; and first PEGylated sustained release formulation of a recombinant protein, interferon -2b (PEG-Intron®; Schering-Plough) in 2000. It is impossible to describe the breadth of drug-delivery activities in one section of a chapter. The general approach is to focus on a particular “anatomical niche,” i.e., use of the lung as a portal for systemic delivery of drugs and to develop technology to overcome the barriers as they are unveiled. Physiologic barriers can be overcome with knowledge of their molecular basis: use of inhibitors to block CYP metabolism of a drug, use of specific permeability enhancers to facilitate transport across a membrane system, blockage of efflux transporters, and use of a prodrug to optimize the partition coefficient and transport mechanisms. Another approach is to take the notion that a particular technology may have DDS applicability, e.g., nanoparticles, and then seek applications. Additionally, as “new” diseases are identified such as age-related macular disease, new noninvasive methods of delivery are required to reach the (retinal) targets at the back of the eye.120 The following are among the many novel delivery system approaches currently under development: ●

Tablets: fast dissolving, lipid-carrier based, slow releasing, disintegrate without water, float in the stomach (gastric retentive), buccal

The Process of New Drug Discovery and Development ● ●

● ● ● ● ●

Dermal: patches, iontophoresis Solubility enhancement systems: lipid-based systems, nanoparticles, surfactants, semisolid formulations Oral methods for peptides Inhalation: peptide delivery, vaporization Site-specific delivery: liposomes, drug monoclonal antibody Implants: biodegradable polymers Drug-eluting stents

A new trend in the delivery of medicines is to employ a device component. This may be an implantable pump for insulin, a metallic stent coated with a drug, or unit capable of rapidly vaporizing a discrete dose for inhalation. Such products are regulated by the FDA as “combination” products and may be reviewed by multiple Centers within the Agency, which may require additional levels of documentation to support the product design. Although DDSs are bringing better therapy to patients, certain concerns with respect to their application for NCEs remain. At an early stage of drug development, the project needs are not usually clear, and proof of concept is usually more important than the optimal therapy. The application of DDS usually leads to longer development time and increased resources. The DDSs are usually more difficult to scale-up. Moreover, if the NCE is dropped from development, all the efforts and resources applied toward the development and application of the DDSs are wasted. For these reasons, decisions are sometimes made to develop initial dosage forms for NCEs utilizing the convenient approaches and to postpone DDS until the LCM phase. However, there is also another school of thought in this area. Given the financial pressures of Big Pharma to introduce blockbusters, and the multitude of leads generated by the HTS methods described above, there are those121 who advocate the use of new technologies and partnerships to enable the use of otherwise delivery-impaired molecules. These may be considered “compound-enabling strategies rather than just product-enhancement programs.” In selecting one DDS over another, technical feasibility must be considered. One consideration that is often overlooked during the development of the drug-delivery opportunities is the dose. For example, an oral tablet may be able to deliver as much as 1 g of drug, while the maximum limit for an inhaled dose could be 1 mg/dose and a transdermal patch 5 to 10 mg/day. Therefore, an inhalation product or transdermal patch could not be the substitute due to the poor bioavailability or first-pass metabolism of a 500 mg tablet. Even the maximum dose in a tablet could be limited if a larger amount of excipients is needed for controlled release, taste masking, etc. The size limitation usually does not permit a larger than 300 mg dose in a capsule. For injectables, 1 to 2 mcg/day could be the maximum dose for a monthly depot system, while 50 to 100 mg might be delivered intramuscularly or up to 750 mg by intravenous bolus administration, and as much as 1 g or even more may be administered by infusion. Another important consideration is the cost. The cost of manufacturing of a DDS should not be so high that the product is not commercially feasible.

3.9

Product Life-Cycle Management

It is estimated that only 1 out of 10 drug molecules that are selected for development and undergoes various preclinical and clinical development activities ultimately reaches the market. Because of such an attrition rate, drug companies are often forced to conserve

Integrated Drug Product Development — From Lead Candidate Selection to LCM

resources during initial development and bring a product to the market that may not have optimal pharmaceutical and clinical attributes. This approach also leads to faster availability of new therapies to patients. After the initial launch, further development activities leading to superior products, known as the product LCM, continue. The LCM may lead to new dosage forms and delivery systems, new dosing regimen, new delivery routes, patient convenience, intellectual property, and so forth. An LCM program is considered successful only if it leads to better therapy and patient acceptance. There are numerous examples of successful LCM through the development of prolonged-release formulations. Development of nifedipine (Procardia® XL, Pfizer), diltiazem (Cardizem CD®, Aventis), and bupropion HCl (Wellbutrin SR® and XL®, GSK) prolonged-release products that not only provided more convenient and better therapy to the patients by reducing dosing frequency but at the same time greatly increased sales of the products are well-known examples. Even old compounds like morphine and oxycodone were turned into significant products by the development of more convenient prolongedrelease formulations (MS Contin® and Oxycontin®, respectively; Purdue Pharma). Issues with the bioavailability of drugs due to their poor solubility or reduced permeability were discussed earlier. Many of the future LCM opportunities may come through bioavailability enhancement. As discussed earlier, solid dispersion, microemulsion, soft gelatin capsule formation, solubilization, lipid-based DDSs, nanoparticle or nanocomposite formation, etc., are some of the common bioavailability approaches that can be utilized for LCM. Development of Lanoxicaps® by Burroughs-Wellcome in 1970s by encapsulating digoxin solutions in soft gelatin capsules is a classic example of LCM by bioavailability enhancement and better pharmaco*kinetic properties. The development of a microemulsion preconcentrate formulation by Novartis (Neoral®), where the variability in plasma and the effect of food were reduced greatly, is another well-known example.122 Life-cycle management through the development of fixed combination products, where two or more drugs are developed or copackaged into a single entity, is gaining increased popularity. The fixed combination products often provide synergistic effects, better therapy, patient compliance, patient convenience, increased manufacturing efficiency, and reduced manufacturing cost. However, a clear risk/benefit advantage is essential for the successful LCM by combination products; mere patient convenience may not be sufficient. Common justifications for the development of fixed combination products include improvement of activity such as synergistic or additive effect, improved tolerance by reduced dose of individual ingredients, broadening of activity spectrum, improvement of pharmaco*kinetic properties, and simplification of therapy for the patient. The development of oral dosage forms that disintegrate or dissolve in the mouth is providing LCM opportunities for pediatric, geriatric, or bedridden patients who have difficulty in swallowing. They are also being used by active adult patients who may not have ready access to water for swallowing tablets or capsules.

3.10

Summary

This chapter describes the many approaches employed to identify and then develop new pharmaceuticals. This chapter has endeavored to outline the careful multidisciplinary tasks that must take place in a timely fashion to insure that the inherent power of each new drug candidate is allowed to be fulfilled. Since the first edition of this book was published, Big Pharma has become even bigger and the quest to regularly introduce blockbusters (products with sales in the $ billions) is even greater. This has resulted in the adoption of HTS tools in

The Process of New Drug Discovery and Development

discovery to increase the ability to produce and screen new molecules. As has been noted, the pressure befalling the product development scientists to make the right choices with regards to issues such as salt and polymorph selection at earlier time points has actually increased under this paradigm. Decisions regarding physical-chemical and physicalmechanical properties and their related impact on high-speed, efficient processing of the commercial dosage forms have to be made with little time and even less drug substance. This is remarkable in that a successful product will require a manufacturing process that is robust enough to consistently make millions of dosage units. The recent changes in the regulatory environment123 have had a serious impact on the overall development and manufacturing process. The risk-based approach to Pharmaceutical c-GMP requires a complete understanding of pharmaceutical processes, which can only be achieved by the application of sound scientific principles throughout the discovery and development of a product. This chapter describes the underlying scientific principles of integrated product development, from lead selection to LCM.

References 1. Lemberger, A.P., Higuchi T., Busse, L.W., Swintowsky, J.V., and Wurster, D.E., Preparation of some steroids in microcrystalline form by rapid freeze-sublimation technique, J. Am. Pharm. Assoc., 43, 338, 1954. 2. Nelson, E. and Schaldemose, I., Urinary excretion for evaluation of drug absorption I, J. Am. Pharm. Assoc., 48, 489, 1959. 3. Levy, G. and Nelson, E., Pharmaceutical formulation and therapeutic efficacy, JAMA, 177, 689, 1961. 4. Gibaldi, M. and Kanig, J.L., Absorption of drugs through the oral mucosa, J. Oral. Ther. Pharmacol., 31, 440, 1963. 5. Wagner, J.G., Biopharmaceutics: absorption aspects, J. Pharm. Sci., 50, 359, 1961. 6. Liberman, S.V., Kraus, S.R., Murray, J., and Wood, J.H., Aspirin formulation and absorption rate I. Criteria for serum measurements with human panels, J. Pharm. Sci., 53, 1486, 1964. 7. Kaplan, S.A., Biopharmaceutical considerations in drug formulation design and evaluation, Drug Metab. Rev., 1, 15, 1972. 8. Yalkowsky, S.H. and Rubino, J.T., Solubilization by cosolvents Part 1, J. Pharm. Sci., 74, 416, 1985. 9. Serajuddin, A.T., Solid dispersion of poorly water soluble drugs: early promises, subsequent problems, and recent breakthroughs, J. Pharm. Sci., 88, 1058, 1999. 10. Lieberman, H.A., Lachman, L., and Schwartz, J.B., Eds., Pharmaceutical Dosage Forms: Tablets, Vols. I–III, Marcel Dekker, New York, 1990. 11. Gibson, M., Pharmaceutical Preformulation and Formulation: A Practical Guide from Candidate Selection to Commercial Dosage Form, CRC Press LLC, Leicestershire, 2004. 12. Venkatesh, S. and Lipper, R.A., Role of the development scientist in compound lead selection and optimization, J. Pharm. Sci., 89, 145, 2000. 13. Drews, J., Drug discovery: a historical perspective, Science, 287, 1960, 2000. 14. Lipinski, C.A., Lombardo, F., Dominy, B.W., and Feeney, P.J., Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings, Adv. Drug Delivery Rev., 23, 3, 1997. 15. Prentis, R.A., Lis, Y., and Walker, S.R., Pharmaceutical innovation by the seven UK-owned pharmaceutical companies (1964–1985), Br. J. Clin. Pharmacol., 25, 387, 1988. 16. Kerns, E.H. and Li, D., Pharmaceutical profiling in drug discovery, Drug Disc. Today, 8, 316, 2003. 17. Ho, N.F.H., Raub, T.J., Burton, P.S., Barsuhn C.L., Adson, A., Audus, K.L., and Borchardt, R.T., Quantitative approaches to delineate passive transport mechanisms in cell culture monolayers,

Integrated Drug Product Development — From Lead Candidate Selection to LCM

18. 19. 20. 21.

22. 23. 24.

25. 26. 27.

28.

29. 30.

31. 32. 33. 34. 35.

36.

37. 38. 39. 40. 41.

in Transport Processes in Pharmaceutical Systems, Amidon, G.L., Lee, P.I., and Topp, E.M., Eds., Marcel Dekker, New York, 2000, p. 219. Clark, D.E., Computational methods for the prediction of ADME and toxicity, Adv. Drug Delivery Rev., 54, 253, 2002. Veber, D.F., Johnson, S.R., Cheng, H.Y., Smith, B.R., Ward, K.W., and Kopple, K.D., Molecular properties that influence oral absorption of drug candidates, J. Med. Chem., 45, 2615, 2002. Yalkowsky, S.H. and Valvani, S.C., Solubility and partitioning I: solubility of nonelectrolytes in water, J. Pharm. Sci., 69, 912, 1980. Klopman, G., Wang, S., and Balthasar, D.M., Estimation of aqueous solubility of organic molecules by the group contribution approach. Application to the study of biodegradation, J. Chem. Inf. Comput. Sci., 32, 474, 1992. Huuskonen, J., Salo, M., and Taskinen, J., Aqueous solubility prediction of drugs based on molecular topology and neural network modeling, J. Chem. Inf. Comput. Sci., 38, 450, 1998. Chen, X.-Q., Cho, S.J., Li, Y., and Venkatesh, S., Prediction of aqueous solubility of organic compounds using a quantitative structure–property relationship, J. Pharm. Sci., 91, 1838, 2002. McKenna, J.M., Halley, F., Souness, J.E., McLay, I.M., Pickett, S.D., Collis, A.J., Page, K., and Ahmed, I., An algorithm-directed two-component library synthesized via solid-phase methodology yielding potent and orally bioavailable p38 MAP kinase inhibitors, J. Med. Chem., 45, 2173, 2002. Kenseth, J.R. and Coldiron, S.J., High throughput characterization and quality control of small molecule combinatorial libraries, Curr. Opin. Chem. Biol., 8, 418, 2004. Faller, B., Improving solubility in lead optimization, Am. Pharm. Rev., 7, 30, 2004. Serajuddin, A.T.M. and Pudipeddi, M., Salt selection strategies, in Handbook of Pharmaceutical Salts: Properties, Selection, and Use, Stahl, P.H. and Wermuth, C.G., Eds., Wiley-VCH, Weinheim, 2002, p. 135. Hilal, S.H., El-Shabrawy, Y., Carreira, L.A., Karickhoff, L.W., Toubar, S.S., and Rizk, M., Estimation of ionization pKa of pharmaceutical compounds using the computer program SPARC, Talanta, 43, 607, 1996. Albert, A. and Serjeant, E.P., The Determination of Ionization Constants, 3rd ed., Chapman & Hall, New York, 1984. Donovan, S.F. and Pescatore, M.C., Method for measuring the logarithm of octanol-water partition coefficient by using short octadecyl-poly(vinyl-alcohol) high performance liquid chromatography columns, J. Chromatogr. A, 952, 47, 2002. Comer, J. and Box, K., High throughput measurement of drug pKa values for ADME screening, JALA, 8, 55, 2003. Kansy, M., Senner, F., and Gubernator, K., Physicochemical high throughput screening, J. Med. Chem., 41, 1007, 1998. Hwang, K.K., Martin, N.E., and Jiang, L., Permeation prediction of M100240 using the PAMPA, J. Pharm. Sci., 6, 315, 2003. Artursson, P., Palm, K., and Luthman, K., Caco-2 monolayers in experimental and theoretical predictions of drug transport, Adv. Drug Delivery Rev., 46, 27, 2001. Stoner, C.L., Cleton, A., Johnson, K., Oh, D-M., Hallak, H., Brodfuehrer, J., Surendran, N., and Han, H.K., Integrated oral bioavailability projection using in vitro screening data as a selection tool in drug discovery, Int. J. Pharm., 269, 241, 2004. Ingels, F.M. and Augustijns, P.F., Biological, pharmaceutical, and analytical considerations with respect to the transport media used in the absorption screening system, Caco-2, J. Pharm. Sci., 92, 1545, 2003. Balbach, S. and Korn, C., Pharmaceutical evaluation of early development candidates—the 100 mg approach, Int. J. Pharm., 275, 1, 2004. Yu, H.S. and Adedoyin, A., ADME-Tox in drug discovery: integration of experimental and computational technologies, Drug Discov. Today, 8, 852, 2003. Berry, I.R. and Nash, R., Pharmaceutical Process Validation, Marcel Dekker, New York, 1993. Avdeef, A., Physicochemical profiling, Curr. Top. Med. Chem., 1, 277, 2001. Avdeef, A., Box, K.J., Comer, J.E.A., Gilges, M., Hibbert, C., Patterson, W., and Tam, K.Y., PHmetric log P 11. pKa determination of water-insoluble drugs in organic solvent-water mixtures, J. Pharm. Biomed. Anal., 20, 1999, 631.

The Process of New Drug Discovery and Development

42. Surakitbanharn, Y., McCandless, R., Krzyzaniak, J.F., Dannenfelser, R-M., and Yalkowsky, S.H., Self-association of dexverapamil in aqueous solution, J. Pharm. Sci., 84, 720, 1995. 43. Hansch, C. and Leo, A., Substituent Constants for Correlation Analysis in Chemistry and Biology, Wiley Interscience, New York, 1979. 44. Reynolds, D.W., Fachine, K.L., Mullaney, J.F., Alsante, K.M., Hatajik, T.D., and Moto, G.M., Available guidance and best practices for conducting forced degradation studies, Pharm. Technol., February, 48, 2002. 45. Carstensen, J.T. and Rhodes, C.T., Eds., Drug Stability: Principles and Practices, Marcel Dekker, New York, 2000. 46. Yoshioka, S. and Stella, V.J., Stability of Drugs and Dosage Forms, Kluwer Academic, New York, 2002. 47. Physician’s Desk Reference 2004. 48. Morris, K.R., Fakes, M.G., Thakur, A.B., Newman, A.W., Singh, A.K., Venit, J.J., Spagnuolo, C.J., and Serajuddin, A.T.M., An integrated approach to the selection of optimal salt form for a new drug candidate, Int. J. Pharm., 105, 209, 1994. 49. International Conference on Harmonization Q6A Guideline: Specifications for new drug substances and products: Chemical Substances, October 1999. 50. Morissette, S.L., Almarsson, O., Peterson, M.L., Remenar, J.F., Read, M.J., Lemmo, A.V., Ellis, S., Cima, M.J., and Gardner, C.R., High-throughput crystallization: polymorphs, salts, co-crystals and solvates of pharmaceutical solids, Adv. Drug Delivery Rev., 56, 275, 2004. 51. Bauer, J., Spanton, S., Henry, R., Quick, J., Dziki, W., Porter, W., and Morris, J., Ritonavir: an extraordinary example of confirmation polymorphism, Pharm. Res., 18, 859, 2001. 52. Brittain, H.G., Polymorphism in Pharmaceutical Solids, 1st ed., Marcel Dekker, New York, 1999. 53. ICH Guidance for Industry Q1B, Photostability testing of new drug substances and products, November 1996. 54. Badawy, S.I., Williams, R.C., and Gilbert, D.L., Chemical stability of an ester prodrug of a glycoprotein IIb/IIIa receptor antagonist in solid dosage forms, J. Pharm. Sci., 88, 428, 1999. 55. Serajuddin, A.T.M., Thakur, A.B., Ghoshal, R.N., Fakes, M.G., and Varia, S.A., Selection of solid dosage form composition through drug-excipient compatibility testing, J. Pharm. Sci., 88, 696, 1999. 56. Schmitt, E.A., Peck, K., Sun, Y., and Geoffroy, J-M., Rapid, practical and predictive excipient compatibility screening using isothermal microcalorimetry, Thermochim. Acta, 380, 175, 2001. 57. Monkhouse, D.C. and Maderich, A., Whither compatibility testing?, Drug Dev. Ind. Pharm., 15, 2115, 1989. 58. Crawley, P., Drug-excipient interactions, Pharm. Tech Eur., 13, 2001, 26. 59. Landin, M., Casalderrey, M., Martinez-Pacheco, R., Gómez-Amoza, J.L., Souto, C., Concheiro, A., and Rowe, R.C., Chemical stability of acetylsalicyclic acid in tablets prepared with different particle size fractions of a commercial brand of dicalcium phosphate dihydrate, Int. J. Pharm., 123, 143, 1995. 60. Irwin, W.J. and Iqbal, M., Solid-state stability: the effect of grinding solvated excipients, Int. J. Pharm., 75, 211, 1991. 61. Villalobos-Hernández, J.R. and Villafuerte-Robles, L., Effect of carrier excipient and processing on stability of indorenate hydrochloride/excipient mixtures, Pharm. Dev. Tech., 6, 551, 2001. 62. Rowe, R.C., Sheskey, P.J., and Weller, P.J., Handbook of Pharmaceutical Excipients, 4th ed., Pharmaceutical Press, London, 2003. 63. Hanco*ck, B.C., Colvin, J.T., Mullarney, M.P., and Zinchuk, A.V., The relative densities of pharmaceutical powders, blends dry granulation, and immediate release tablets, Pharm. Technol., 4, 64, 2003. 64. Rowley, G., Quantifying electrostatic interactions in pharmaceutical powder systems, Int. J. Pharm., 227, 47, 2001. 65. Hiestand, E.N. and Smith, D.P., Three indexes for characterizing the tableting performance of materials, Adv. Ceramics, 9, 47, 1984. 66. Roberts, R.J. and Rowe, R.C., The compaction of pharmaceutical and other model materials — a pragmatic approach, Chem. Eng. Sci., 42, 903, 1987.

Integrated Drug Product Development — From Lead Candidate Selection to LCM

67. Amidon, G.L., Lennernas, H., Shah, V.P., and Crison, J.R., Theoretical basis for a biopharmaceutical drug classification: correlation of in vitro drug product dissolution and in vivo bioavailability, Pharm. Res., 12, 413, 1995. 68. FDA, Guidance for industry: Dissolution testing of immediate release solid oral dosage forms, 1997. 69. Curatolo, W., Physical chemical properties of oral drug candidates in the discovery and exploratory setting, Pharm. Sci. Tech. Today, 1, 387, 1998. 70. Agoram, B., Woltosz, W.S., and Bolger, M.B., Predicting the impact of physiological and biochemical processes on oral drug bioavailability, Adv. Drug Delivery Rev., 50, S41, 2001. 71. Kramer, S.F. and Flynn, G.L., Solubility of organic hydrochlorides, J. Pharm. Sci., 61, 1896, 1972. 72. Serajuddin, A.T.M. and Pudipeddi, M., Salt selection strategies, in Handbook of Pharmaceutical Salts: Properties, Selection, and Use, Stahl, P.H. and Wermuth, C.G., Eds., Wiley-VCH, Weinheim, 2002, p. 135. 73. Johnson, K.C. and Swindell, A.C., Guidance in the setting of drug particle size specifications to minimize variability in absorption, Pharm. Res., 13, 1795, 1996. 74. Pudipeddi, M., Serajuddin, A.T.M., Grant, D.J.W., and Stahl, P.H., Solubility and dissolution of weak acids, bases, and salts, in Handbook of Pharmaceutical Salts: Properties, Selection, and Use, Stahl, P.H. and Wermuth, C.G., Eds., Wiley-VCH, Weinheim, 2002, p. 19. 75. Dill, W.A., Kazenko, A., Wolf, L.M., and Glazko, A.J., Studies on 5,5⬘-diphenylhydantoin (Dilantin) in animals and man, J. Pharmacol. Exp. Ther., 118, 270, 1956. 76. Serajuddin, A.T.M., Sheen, P.C., Mufson, D., Bernstein, D.F., and Augustine, M.A., Preformulation study of a poorly water soluble drug, alpha-pentyl-3-(2-quinolinylmethoxy) benzenemethanol: selection of the base for dosage form design, J. Pharm. Sci., 75, 492, 1986. 77. Serajuddin, A.T.M., Sheen, P., Mufson, D., Bernstein, D.F., and Augustine, M.A., Physicochemical basis of increased bioavailability of a poorly water-soluble drug following oral administration as organic solutions, J. Pharm. Sci., 77, 325, 1988. 78. Washington, N., Washington, C., and Wilson, C.G., Physiological Pharmaceutics, 2nd ed., Taylor and Francis, New York, 2001, chap. 5. 79. Davis, S.S., Hardy, J.G., and Fara, J.W., Transit of pharmaceutical dosage forms through the small intestine, Gut, 27, 886, 1986. 80. Smith, C.G., Poutisiaka, J.W., and Schreiber, E.C., Problems in predicting drug effects across species lines, J. Int. Med. Res., 1, 489, 1973. 81. Chiou, W.L., Ma, C., Chung, S.M., Wu, T.C., and Jeong, H.Y., Similarity in the linear non-linear oral absorption of drugs between human and rat, Int. J. Clin. Pharmacol. Ther., 38, 532, 2000. 82. Chiou, W.L., Jeong, H.Y., Chung, S.M., and Wu, T.C., Evaluation of using dog as an animal model to study the fraction of oral dose absorbed of 43 drugs in humans, Pharm. Res., 17, 135, 2000. 83. Li, S., He, H., Parthiban, L., Yin, H., and Serajuddin, A.T.M., IV–IVC considerations in the development of immediate-release oral dosage form, J. Pharm. Sci., 94, 1396, 2005. 84. FDA, Guidance for Industry: Extended Release Oral Dosage Forms: Development, Evaluation, and Application of In Vitro/In Vivo Correlation, 1997. 85. Dressman, J.B., Dissolution testing of immediate-release products and its application to forecasting in vivo performance, Oral Drug Absorption: Prediction and Assessment, Dressman, J.B. and Lennernas, H., Eds., Marcel Dekker, New York, 2000, p. 155. 86. Hariharan, M., Ganorkar, L.D., Amdion, G.E., Cavallo, A., Gatti, P., Hageman, M.J., Choo, I., Miller, J.L., and Shah, U.J., Reducing the time to develop and manufacture formulations for first oral dose in humans, Pharm. Technol., October, 68, 2003. 87. Sistla, A., Sunga, A., Phung, K., Iparkar, A., and Shenoy, N., Powder-in-bottle formulation of SU011248. Enabling rapid progression into human clinical trials, Drug Dev. Ind. Pharm., 30, 19, 2004. 88. Aubry, A.F., Sebastian, D., Hobson, T., Xu, J.Q., Rabel, S., Xie, M., and Gray, V., In-use testing of extemporaneously prepared suspension of second generation non-nucleoside reversed transcriptase inhibitors in support of Phase I clinical studies, J. Pharm. Biomed. Anal., 23, 535, 2000.

The Process of New Drug Discovery and Development

89. Shah, V.P., Noory, A., Noory, C., McCullough, B., Clarke, S., Everett, R., Navkiasky, H., Srinivasan, B.N., Fortman, D., and Skelly, J.P., In vitro dissolution of sparingly water-soluble drug dosage forms, Int. J. Pharm., 125, 99, 1995. 90. Abrahamsson, B. and Ungell, A.L., Biopharmaceutical support in formulation development, in Pharmaceutical Preformulation and Formulation, Gibson, M., Ed., CRC Press, Boca Raton, FL. 91. Waterman, K.C., Adami, R.C., Alsante, K.M., Hong, J., Landis, M.S., Lombardo, F., and Roberts, C.J., Stabilization of pharmaceuticals to oxidative degradation, Pharm. Dev. Tech., 7, 1, 2002. 92. Al-Omari, M.M., Abelah, M.K., Badwan, A.A., and Aaber, A.M., Effect of drug matrix on the stability of enalapril maleate in tablet formulation, J. Pharm. Biomed. Anal., 25, 893, 2001. 93. Shek, E., Ghani, M., and Jones, R.E., Simplex search in optimization of a capsule formulation, J. Pharm. Sci., 69, 1135, 1980. 94. Bateman, S.D., Verlin, J., Russo, M., Guillot, M., and Laughlin, S.M., Development and validation of a capsule formulation knowledge-based system, Pharm. Technol., 20, 174, 1996. 95. Shiromani, P.K. and Clair, J., Statistical comparison of high shear vs. low shear granulation using a common formulation, Drug Dev. Ind. Pharm., 26, 357, 2000. 96. ICH Guidance for Industry Topic Q 1 A: Stability testing of New Drug Substances and Products, 2003. 97. Nash, R.A., Validation of pharmaceutical processes, in Encyclopedia of Pharmaceutical Technology, Swarbrick, J. and Boylan, J.C., Eds., Marcel Dekker, New York, 2002. 98. Pharmaceutical cGMPs for the 21st century: A risk based approach. US Food and Drug Administration, www.fda.gov/cder/gmp/index.htm 99. FDA Guidance to Industry: Bioavailability and bioequivalence studies for orally administered drug products — General considerations, 2002. 100. Cupissol, D., Bressolle, F., Adenis, L., Carmichael, J., and Romain, D., Evaluation of the bioequivalence of tablet and capsule formulations of granisetron in patients undergoing cytotoxic chemotherapy for malignant disease, J. Pharm. Sci., 82, 1281, 1993. 101. Akers, M.J., Fites, A.L., and Robinson, R.L., Formulation design and development of parenteral suspensions, Bull. Parent Drug Assoc., 41, 88, 1987. 102. Broadhead, J., Parenteral dosage forms, in Pharmaceutical Preformulation and Formulation, Gibson, M., Ed., Interpharm/CRC Press, New York, 2004. 103. Akers, M.J., Parenteral Quality Control: Sterility, Pyrogen, Particulate, and Package Integrity Testing, Marcel Dekker, New York, 1994. 104. Sweetana, S. and Akers, M.J., Solubility principles and practices for parenteral drug dosage form development, J. Pharm. Sci. Technol., 50, 330, 1996. 105. Gupta, P.K., Patel, J.P., and Hahn, K.R., Evaluation of pain and irritation following local administration of parenteral formulations using the rat paw lick model, J. Pharm. Sci. Technol., 48, 159, 1994. 106. Nema, S., Washkuhn, R.J., and Brendel, R.J., Excipients and their use in injectable products, J. Pharm. Sci. Technol., 51, 166, 1997. 107. Strickley, R.J., Parenteral formulation of small molecules therapeutics marketed in the United States-Part III, PDA J. Pharm. Sci. Tech., 54, 152, 2000. 108. Tang, X.L., Pikal, M.J., and Taylor, L.S., Design of freeze-drying processes for pharmaceuticals: practical advice, Pharm. Res., 21, 191, 2004. 109. Puri, N., Pejaver, S.K., Parnajpe, P., and Sinko, P.J., Bioavailability of parenteral dosage forms, US Pharmacist, 26, HS41–HS53, HS55–HS56, 2001. 110. Suarez, S. and Hickey, A.J., Drug properties affecting aerosol behavior, Respir. Care, 45, 652, 2000. 111. Poochikian, G. and Bertha, C.M., Regulatory view on current issues pertaining to inhalation drug products, Respiratory Drug Delivery VIII Conference, Vol. 1, 2000, pp. 159–164 [www.rddonline.com]. 112. Hickey, A.J. and Martonen, T.B., Behavior of hygroscopic pharmaceutical aerosols and the influence of hydrophobic additives, Pharm. Res., 10, 1, 1993. 113. Hodges, C.C., Lloyd, P.M., Mufson, D., Rogers, D.D., and Wensley, M.J., Delivery of Aerosols Containing Small Particles through an Inhalation Route, US 6,682,716, January 27, 2004.

Integrated Drug Product Development — From Lead Candidate Selection to LCM

114. Rabinowitz, J.D., Wensley, M., Lloyd, P., Myers, D., Shen, W., Lu, A., Hodges, C., Hale, R., Mufson, D., and Zaffaroni, A., Fast onset medications through thermally generated aerosols, J. Pharmacol. Exp. Ther., 309, 769, 2004. 115. Amiel, S.A. and Alberti, K.G., Inhaled insulin, BMJ, 328, 1215, 2004. 116. Zaffaroni, A., Therapeutic systems: the key to rational drug therapy, Drug Metab. Rev., 8, 191, 1978. 117. Swintosky, J.V., Personal adventures in biopharmaceutical research during the 1953–1984 years, Drug Intell. Clin. Pharm., 19, 265, 1985. 118. Allen, T.M. and Cullis, P.R., Drug delivery systems: entering the mainstream, Science, 303, 1818, 2004. 119. Novel Drug Delivery Systems (NDDS) Reports, Technology Catalysts International, Falls Church, VA, 2002. 120. Hastings, M.S., Li, S.K., Miller, D.J., Bernstein, P.S., and Mufson, D., Visulex™: advancing iontophoresis for effective noninvasive back-of-the-eye therapeutics, Drug Delivery Technol., 4, 53, 2004. 121. Horspool, K.A. and Lipinski, C.A., Advancing new drug delivery concepts to gain the lead, Drug Delivery Technol., 3, 34, 2003. 122. Meinzer, A., Mueller, E., and Vonderscher, J., Microemulsion — A suitable galenical approach for the absorption enhancement of low soluble compounds?, in Absorption of Micellar Systems: Bulletin Technique Gattefossé, Barthelemy, P., Ed., Gattefossé S.A., Saint-Priest, France, 1995, p. 21. 123. McClellan, M.B., Speech before PhRMA, March 28, 2003 (available at www.fda.gov).

Section II

Scientific Discoveries Application in New Drug Development

4 The Impact of Combinatorial Chemistry on Drug Discovery

Michael Rabinowitz and Nigel Shankley

CONTENTS 4.1 Introduction ..................................................................................................................................56 4.1.1 Combinatorial Chemistry Will Transform the Pharmaceutical R&D Process ..................................................................................................................56 4.1.2 Industry Is Less Efficient than It Was ........................................................................56 4.2 Background ..................................................................................................................................57 4.2.1 The Traditional Interplay between Biology and Chemistry in the Drug Discovery Process ..............................................................................................57 4.2.1.1 The Development of Medicinal Chemistry ..............................................57 4.2.1.2 The Development of Bioassay ....................................................................58 4.2.1.3 The Choice of Bioassay: The Compound Progression Path ..................60 4.2.1.4 Properties of Lead Compounds ..................................................................61 4.2.1.5 The Compound Progression Path ..............................................................62 4.3 The Impact of Combinatorial Chemistry on Bioassay ........................................................62 4.3.1 Combinatorial Chemistry ............................................................................................63 4.3.1.1 High-Throughput Chemistry ......................................................................66 4.3.2 Mixtures, Libraries, and Compound Arrays ..........................................................66 4.3.2.1 Discrete Compound Synthesis ....................................................................66 4.3.2.2 Mixtures: Split-and-Mix Synthesis, Deconvolution, and Encoding ....66 4.3.2.3 Encoded Combinatorial Synthesis ............................................................69 4.3.2.4 Parallel Synthesis ..........................................................................................69 4.3.3 Types of Library Design Strategies and Their Uses ..............................................70 4.3.3.1 Diversity Library ............................................................................................70 4.3.3.2 Focused Libraries ..........................................................................................71 4.4 Combinatorial Chemistry in the Modern Drug Discovery Setting — Commentary......72 4.4.1 The Promise ....................................................................................................................73 4.4.2 The Reality ......................................................................................................................74 Further Reading ....................................................................................................................................76

4.1 4.1.1

The Process of New Drug Discovery and Development

Introduction Combinatorial Chemistry Will Transform the Pharmaceutical R&D Process

In the late 1980s major pharmaceutical companies, in the face of concern about dwindling new drug pipelines, initiated a major investment in a technology that promised to transform the efficiency of drug discovery. Combinatorial chemistry was to be employed to generate millions of compounds that could be screened against potential drug target proteins. The rapid deployment of the technology and re-tooling of major pharmaceutical research laboratories were fueled by the belief that the endless combination of molecules to form new compounds would significantly increase the probability of finding the next “blockbuster,” the molecule with the best drug-like properties. The so-called high-throughput screening (HTS) (testing) of these molecules required equally revolutionary changes in biology laboratories. As the complexity of the human genome was unraveled, recombinant technologies came into their own, providing endless supplies of potential drug target protein in multiple engineered formats. Robots were introduced to process thousands of compounds through almost instantaneous readout cell- and protein-based assays. The investment, like the expectations generated, was enormous and significantly impacted the structure and composition of drug discovery scientific teams as they worked to harness the endless flow of new data. This chapter, prepared 15 years later, reviews the impact and development of combinatorial chemistry.

4.1.2

Industry Is Less Efficient than It Was

Between 1993 and 2003 the total U.S. pharmaceutical R&D spending was more than doubled with eight leading pharmaceutical companies investing approximately 20% of their sales revenue in R&D (Filmore, Modern Drug Discovery, 2004). However, during the same period the number of new medical entities (NMEs), defined in the analysis as drugs with a novel chemical structure, submitted for product approval to the Food and Drug Administration (FDA) fell from 44 in 1996 to 24 in 2003 (Figure 4.1). Similar trends were noted at regulatory agencies worldwide. The problem was highlighted in a white paper presented by the FDA in March 2004, which was bluntly entitled “Innovation or Stagnation” (http:// www.fda.gov/oc/initiatives/criticalpath/whitepaper.html). In brief, the increased output of more effective, affordable, and safer therapies anticipated from the astonishing progress in basic biomedical science over the last two decades had not been realized. The widely held view is that NMEs are invented in drug discovery programs, but that products are created from NMEs in the development process. Closer analysis of the data from the 1993 to 2003 period revealed that the number of NMEs entering early phase of clinical trials from drug discovery programs increased significantly (an 85% increase to Phase I and 90% increase to Phase II). In terms of these simple metrics, it could be argued that the productivity of research increased in line with investment and the basic science revolution. However, to suggest that the problem lies in the development process where the vast majority of NMEs failed to become products would be a gross oversimplification. NMEs primarily fail to become products because of shortcomings in safety or effectiveness, and both properties are co-owned by discovery and development. The high failure rate is proving costly to both the industry and patient. The disproportionate high cost of late-stage failures is being borne by fewer successful new products that drive higher pricing and predicate that companies restrict product development to large patient populations in

The Impact of Combinatorial Chemistry on Drug Discovery

Total NME submissions received by FDA

250

200 40 150

100 20 50

0 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

U.S. R&D spending – indexed growth (1993 = 100)

FIGURE 4.1 Ten-year trends in NME submissions to FDA and U.S. pharmaceutical research spending. The figure shows 10-year trends in U.S. pharmaceutical research and development (R&D) investment (PAREXEL’s Pharmaceutical R&D Statistical Sourcebook 2002/2003) and the number of submissions to FDA of new molecular entities (NMEs), defined as drugs with novel chemical structure. (Redrawn from http://www.fda.gov/ oc/initiatives/criticalpath/whitepaper.html.)

affluent societies. The prescription payers of these populations are in effect underwriting the decreased overall low productivity of the pharmaceutical industry. The conclusion must be that the pharmaceutical companies are not selecting the right drug targets and patient populations or the NMEs are inadequate in terms of selectivity and biopharmaceutical properties. In this chapter, we review the impact of one of the technological breakthroughs — combinatorial chemistry — that governed significant changes over the last 15 to 20 years, not only in the initial discovery of NMEs, but also in the methods adopted for their all important characterization prior to entering the development process.

4.2 4.2.1

Background The Traditional Interplay between Biology and Chemistry in the Drug Discovery Process

4.2.1.1 The Development of Medicinal Chemistry Paracelsus (1493 to 1541) is credited with establishing the role of chemistry in medicine. It is evident in his oft-quoted, visionary view: “it is the task of chemistry to produce medicines for the treatment of disease since the vital functions of life are basically chemical in nature […..] all things are poisons, for there is nothing without poisonous qualities. It is only the dose which makes a thing a poison.”

The Process of New Drug Discovery and Development

However, until the development of analytical and synthetic chemistry in the 19th century, the discovery of new therapeutic substances remained focused on the biological investigation of natural substances usually secured from the apothecary garden. The new chemical techniques enabled the isolation and manufacture of pure substances. First, reproducible extraction procedures yielded individual active compounds from natural products and, subsequently, toward the end of the century, de novo synthesis was achieved of the simpler natural products such as aspirin, which was first marketed by Bayer in 1899. The next logical step was to systematically modify existing biologically active compounds, referred to as lead compounds, in an attempt to improve on their drug-like properties. By the 1930s the pharmaceutical industry had begun a sustained period of productivity, identifying lead compounds and transforming them into effective new therapeutic agents. These lead compounds came from a variety of sources including natural products, endogenous chemical messengers (hormones, neurotransmitters, enzyme substrates, etc.), random screening of the output of industrial chemistry processes (e.g., dyestuffs), and in the case of so-called me-too programs, simply from an existing drug. The latter programs were based on improving the biopharmaceutical properties or safety profile of existing drugs acting at a particular target or the optimization of a known side-effect that was perceived as potential therapeutic benefit in its own right. The success of the burgeoning pharmaceutical industry from 1930 through 1980 and the impact of the medicines produced on the quality and longevity of life are beyond question (Figure 4.2).

4.2.1.2 The Development of Bioassay At the time of the first syntheses of the pharmaceutically active ingredients of natural products, the pioneers of pharmacology, Ehrlich, Langley, and Hill among others, were engaged in turning the study of the mode of action of drugs into a quantitative science. By Genomics/proteomics

Natural products and derivatives

Receptors

Targets identified from disease genes

Chronic degenerative disease associated with ageing, Biotech drugs inflammation, cancer Lipid lowerers ACE-inhibitors

Enzymes

Serendipity New therapeutic cycles

Cell pharmacology/ molecular biology Genetic engineering

H2-/blockers NSAIDS Psychotropics Penicillins sulphonamides aspirin

1900

1950

1960

1 970

1980

1990

2000

2010

2020

2030

FIGURE 4.2 One hundred years of drug discovery technologies. (Redrawn from Adkins, S. et al., New Drug Discovery: Is Genomics Delivering? Lehman Brothers Research Reports, 1999.)

The Impact of Combinatorial Chemistry on Drug Discovery

characterization of the relationships between concentration of active substance and the effects produced on biological systems and application of the newly established laws of thermodynamics governing the interaction between reacting chemical species, they inferred the existence of cell membrane receptors. These receptors were shown to be specific sites of action of the body’s own chemical messenger systems, in addition to many of the natural substance-derived drugs being isolated and synthesized at the time. These early pharmacologists, and those like Clark, Gaddum, and Schild who subsequently built on their ideas, established the principles of bioassay that are still at the heart of the drug discovery process today. Bioassays have been likened to analytical machines insofar as pharmacologists use them to assign biological properties to compounds in the same way a chemist measures the physical-chemical properties of molecules. If the fundamental role of the medicinal chemist is to optimize the pharmaceutical properties of so-called lead compounds by structural modification, then the role of the pharmacologist in the drug discovery process is to select, develop, and apply bioassays to provide relevant robust data that inform the medicinal chemist of the impact of the modifications he makes. From the time of Ehrlich until the advent of readily available recombinant protein expression technologies in the 1980s, bioassay was predominantly based on the quantification of drug efficacy, safety, and distribution in living tissues or extracts thereof. Scientific reductionist drive provided increasingly more functionally simplified and hom*ogenous assay systems. So, by the 1980s, techniques were available to study the action of new compounds on drug targets that, although originating from their natural expression environment, were present in experimentally amenable bioassays. Thus, the compounds could be studied in whole organisms, isolated intact organ or tissues, dispersed cell preparations, membrane hom*ogenates, and ultimately, in the case of some targets, in enriched protein extracts. The pharmacologist now has at his disposal a full range of assays from the pure target through populations of patients to study the effects of compounds (Figure 4.3). Moving up the bioassay hierarchy from pure protein toward patient populations, it takes longer to obtain results and the comparisons of molecules that are increasingly complex as the compound activity is increasingly multifactorial. Only in the patient is the effect of the compound dependent on the expression of all the biopharmaceutical properties of the compound that the medicinal chemist seeks to optimize. Thus,

Number of compounds tested Patients (clinical trials)

Animal model

Isolated intact tissue

Cells

Target protein

1/2

Weeks/months/years

Days/weeks/months

Days/weeks

1000

Hours/minutes/seconds

1,000,000

Hours/minutes/seconds

FIGURE 4.3 Bioassay compound throughput and assay timescales.

Assay output (time for one compound)

The Process of New Drug Discovery and Development

the assay readout is based not only on the primary activity at the target, but also on the adsorption from the site of absorption, distribution, metabolism, elimination, and any toxicity (ADMET), plus the natural variations in blood flow and chemistry and other homeostatic changes that occur in intact organisms.

4.2.1.3 The Choice of Bioassay: The Compound Progression Path The challenge for drug discovery scientists is clear. Although the current medicinal chemistrydriven drug discovery process is often represented as a linear stepwise series of events (Figure 4.4), in practice, the process is holistic and iterative insofar as changes in structure impact all the properties of an NME (Figure 4.5). The simplest most functionally reduced assays usually involve the measurement of the primary interaction between the compound and its site of action. The assay readout from these basic or primary assays is usually a measure of the affinity of the compound for the drug target expressed in terms of the equilibrium dissociation constant, which is the concentration of the compound that occupies half the sites. For both the pharmacologist and medicinal chemist, these assays are perhaps the most satisfying. The assays are usually associated with low variance, and because the activity of the compounds can be expressed by the single chemical affinity parameter, the communication of progress of medicinal chemistry is straightforward. The functionally reduced assays are cheaper and faster, allowing the testing of many compounds in a time frame commensurate with the medicinal chemist’s ability to conceive and synthesize new molecules based on the assay results. The problems begin to emerge when the chemist tries to address other properties of a potential NME–those governing the ADMET properties. These require the employment of the increasingly complex, expensive, and time-consuming intact physiological systems from Gene product selected as potential target

Target identification

Pathway analysis, gene knockouts etc. used to investigate the potential impact of manipulating the target

Target validation

Robust primary bioassay established and validated

Assay development

Chemical libraries screened for primary activity

High-throughput screen

Confirmation of compounds biopharmaceutical properties

Lead compound generation

Medicinal chemistry optimizes properties

Lead optimization

Efficacy established in animal disease models

Validated lead

Safety margins established in animals

Safety testing

Safety & efficacy established in patients

Clinical trials

Product with regulatory approval

Marketed drugs

FIGURE 4.4 The drug discovery process — a stepwise view.

The Impact of Combinatorial Chemistry on Drug Discovery

Lead compounds

Compound testing

Lead optimization

Compound modification

NME FIGURE 4.5 The drug discovery process — iteration between medicinal chemistry and bioassay.

which data cannot be obtained on each compound synthesized, and it certainly is never fast enough to feed the chemist’s imagination and guide his synthetic choices. The situation was dramatically exacerbated by the implementation of combinatorial chemistry and massive increases in the number of new compounds that the chemist could produce for assay. In an attempt to address this problem in the current drug discovery setting, a newly conceived program is usually reviewed against a set of project criteria with the aim of assessing the likelihood of success and defining the critical path a program should follow. Sir James Black has published his criteria based on 40 years’ experience of working closely with medicinal chemists during which time he discovered both the first -adrenoceptor and histamine H2-receptor antagonists. These are reproduced in Table 4.1. The two of Black’s criteria that are most relevant to this chapter are: Is a chemical starting point identified? and Are there relevant bioassays available to guide medicinal chemistry?

4.2.1.4 Properties of Lead Compounds The chemical starting point is the lead compound, which is usually defined along the lines of “a chemical entity that has known structure, high purity that reproducibly and predictably causes the desired biological effect; belongs to a chemical series that

TABLE 4.1 Criteria for Inception of a Medicinal Chemistry-Led Drug Discovery Project 1. Is there a basis for the desired selectivity of action — is the project purged of wishful thinking? 2. Is a chemical starting point identified? 3. Are relevant bioassays available to guide medicinal chemistry? 4. Will it be possible to confirm laboratory-defined specificity of action in humans? 5. Is there a clinical condition relevant to the specificity in point 4? 6. Is the project adequately resourced? 7. Does the project have a champion — someone with the necessary passion, conviction and energy? Adapted from Black, J.W., in Textbook of Receptor Pharmacology, Foreman, J.C. and Johanson, T., Eds., CRC Press, Boca Raton, FL, 2003, pp. 271–279.

The Process of New Drug Discovery and Development

shows evidence of a structure–activity relationship (i.e., systematic structural changes lead to significant changes in activity) and has optimizable pharmaco*kinetics.” In an attempt to rationalize the selection of a particular lead or lead series of compounds, attempts have been made to articulate these properties in terms of measurable parameters. For example, in 1997, Lipinski published an analysis of the properties of existing drugs to identify basic chemical characteristics that could be used as a filter to select “drugable” leads from screening hits. The properties considered usually include factors such as synthetic difficulty, analog potential, scale-up potential, purity and reproducibility, molecular weight, heteroatom/carbon ratio, lipophilicity, hydrogen bond donors and acceptors, number of rotatable bonds/rings, the absence of chemically reactive centers (i.e., no “toxic” features), and the ability to secure novel intellectual property.

4.2.1.5 The Compound Progression Path With drug target set and lead compound(s) identified, the properties of an NME suitable for clinical evaluation can be specified more explicitly. These properties form the project goals and are written in terms of minimum acceptable values of the assay output parameters, as illustrated in Table 4.2. These criteria are then transformed into a compound progression path that provides the guidelines as to which compounds are progressed to more expensive, time-consuming assays (Figure 4.6). In practice, because of the iterative nature of the process, both the project goals and compound progression schemes are constantly updated as new information is generated both within the program and in the scientific community at large.

4.3

The Impact of Combinatorial Chemistry on Bioassay

Although the regulatory requirements for approval of a new drug have been subject to constant revision as tighter definitions of safety and efficacy are established, the fundamental properties required to be assigned to NMEs in the drug discovery process has not changed significantly over the last 30 years. Consequently, the overall approach to medicinal chemistry-driven programs outlined in the preceding sections has also remained largely the same. The impact of combinatorial chemistry and the simultaneous ability to produce effectively unlimited supplies of primary assay material for all the potential drug targets revealed by the human genome project can be viewed, in simplest terms, as a period of information overload. The introduction of robotic primary target activity assays allows the basic parameters to be obtained for all the compounds available, and without doubt increased the diversity of TABLE 4.2 Specific Drug Discovery Project Goals for a Small Molecule NME 1. Specificity: defined mechanism of action, e.g., competitive receptor antagonist 2. Potent: high affinity for target, e.g., pKB ⬍ 8 3. Selectivity: e.g., ⬎100-fold selectivity over other pharmacological loci 4. Oral bioavailability: e.g., %F ⬎ 30%, t1/2 ⬎ 4 h 5. In vivo potency: e.g., ⬍1 mg/kg for 50% maximum activity 6. Novel chemical structure with intellectual property rights protected 7. Scaleable synthetic route, reasonable cost of goods

The Impact of Combinatorial Chemistry on Drug Discovery

Human receptor affinity (recombinant engineered) Human pKI = > 8 Human liver microsome stability testing Broad-spectrum selectivity screen then, rat receptor affinity assay

HLM t1/2 >100 min Selectivity > 100 Rat pKI > 7.5

Pharmaco*kinetic profiling (rat) intravenous/oral dosing t1/2 > 90 min %F > 40%

Scaleup of drug substance (e.g., 2 g) Salt selection and physical properties Formulation development

ED50 < 1 mg/kg Teffect > 60 min

%F > 40% Canine pKI > 7.5

Multidose pharmaco*kinetics (rat and dog) Dog receptor affinity Human tissue activity

Intravenous bolus dose in a rat efficacy model measuring potency, duration of action, and plasma levels for effect

Oral dosing in rat and dog models measuring potency, duration of action, and plasma levels for effect

FIGURE 4.6 Receptor antagonist drug discovery program: compound progression scheme.

compounds that could be screened as potential leads in a program. The increased chemical diversity also allowed identification of chemical starting points for those targets where the traditional nonrandom screening sources of suitable leads were unavailable. Thus, the approach of adopting the natural ligand or substrate as the chemical starting point proved to be difficult in the case of many intracellular targets because the ligands tend to express relatively low affinity and selectivity. Similarly, screening of diverse libraries has provided the first small molecule ligands for many peptide hormone receptor targets for which working from the peptide hormone as a chemical template proved intractable. However, for all these successes in providing novel lead compounds, it soon became apparent that the lead optimization process would remain the rate-limiting step. A further worrying trend was that the new robotic laboratories for studying compounds at the protein or cellular level were often introduced at the expense of the systems assayist and his laboratory equipment. Indeed one unfortunate side effect of the molecular biological and combinatorial revolution has been the significant reduction in the advanced educational programs that provide the skilled scientists trained in performing the in vivo pharmacology experiments.

4.3.1

Combinatorial Chemistry Le mieux est l’ennemi du bien. —Voltaire, Dictionnaire Philosophique

With the accelerating advancement of reductionist biology in the 1980s and 1990s and with it, the advent of high-throughput target screening, came the need for an increase in the stores of compounds available for screening; not just in quantity of each but in sheer number — a concept closely linked with the notion of Chemical Diversity. To answer this call, chemists began to adapt methods long known in the synthesis of biopolymers for the

The Process of New Drug Discovery and Development

preparation of small molecules, methods that not only allowed for the rapid production of complex (and not-so-complex) structures for bioassay but also allowed the preparation of many diverse compounds in the process, often in a fraction of the time it would take to produce this diversity of structures by traditional solution synthesis methods. This process and set of techniques, as diverse and somewhat unrelated as they were, became to be known as combinatorial chemistry (Table 4.3). The following section briefly outlines the core concepts of combinatorial chemistry and attempts to place them into historical perspective now that the combinatorial chemistry “revolution” in industrial drug discovery is two decades old. This generational hindsight allows us the perspective to suggest what the real impact of these techniques has been in the pharmaceutical industry: The real revolution of combinatorial chemistry is the scientific codification and general acceptance of the fact that to make large numbers of compounds for bioanalysis, we have to loosen our requirements for overall analytical purity. And thus something was sacrificed in the process: we began thinking of compounds for screening as sets rather than individuals — allowing for the statistical treatment of purity of collections rather than of single compounds. The words of Voltaire are particularly relevant: “The enemy of the good is the TABLE 4.3 Definitions of Combinatorial Chemistry Concepts Combinatorial synthesis

The simultaneous preparation of all possible combinations of two or more mutually reactive sets of chemical monomers, either as individual compounds or mixtures of compounds

Library

A collection of compounds intended for bioassay that typically share a common pharmacophore or activity toward a given class of protein targets

Parallel synthesis

The simultaneous preparation of a subset of individual members of a combinatorial library

Solid-phase library

The use of polymeric, insoluble protecting groups in combinatorial or parallel synthesis that allow for the facile separation of target compounds from the rest of the reaction mixture under heterogeneous reaction conditions. Final target compounds may be assayed either while still attached to the resin beads or following cleavage

Solution-phase library

A library of compounds prepared under hom*ogeneous reaction conditions without the aid of polymeric supports

Resin scavenging

The use of reactive polymer supports to selectively remove residual reagent or starting materials from a solution-phase mixture

Resin capture

The use of reactive polymer supports to selectively remove target products from a solution-phase mixture

Split and mix

The technique of preparing target compounds in a solid-phase library whereby resin-bound intermediates in the preparation of a solid-phase library are combined and then redivided for attachment of subsequent monomers. This technique produces mixtures of compounds but allows for the rapid production of very large compound libraries

Spatial addressability

The preparation of individual compounds in a parallel format such that the structural identity of each is known by its physical location in a matrix

Encoding

Defining the structural identity of a resin-bound compound in solid-phase library by marking the individual resin beads with a tag that is unique to the attached target compound

Deconvolution

The process of using iterative bioassay activity data to identify active components in a library mixture

The Impact of Combinatorial Chemistry on Drug Discovery

perfect.” The strictures of pristine compound purity adhered to fervently for decades in the medicinal chemistry laboratory were seen as roadblocks to the potential efficiency gains of combinatorial methods. By relaxing standards of chemical purity it was surmised that sheer numbers of compounds being created and screened would lead us down the path to quicker lead identification and optimization toward clinical candidacy. Now, 20 years on, it just may be that the enemy of the “good” has been found to be the “good enough.” Combinatorial chemistry is defined differently by many of its practitioners and reviewers. This ambiguity is probably to be expected for a field of endeavor that is still in its adolescence. For some, the term combinatorial chemistry refers to the technology involved in the creation of large diversity libraries. For these workers, this technology is often some version of resin-based “split-and-mix” diversity generation. For others, combinatorial chemistry is a broad term that refers to a collection of technologies and methods that allow the rapid production of all possible “combinations” of mutually reactive chemical fragments. In this latter definition, the exact method of library preparation (solid-phase, solution-phase, and solid-supported reagents, etc.) and the state of the target compounds upon completion of the effort (mixtures, discrete compounds, encoded resin-bound compounds, etc.) are incidental to the goal of the production of a true mathematical combinatorial assembly of similar chemical fragments. In practice, the term “combinatorial chemistry” refers to all technologies and applications used in the production of many compounds simultaneously, be they a true combinatorial collection, or a smaller subset of all possible combination of chemical fragments. Thus, in this definition, combinatorial chemistry also refers to multiple parallel syntheses (i.e., the synthesis of a row or column out of a full array) (Figure 4.7). In fact, what combinatorial chemistry has done for the bench chemist is to codify and thereby allow consensual adoption by those in the field of somewhat relaxed standards of analytical purity which we are willing to accept for biological assay data collection. This is an absolutely essential tradeoff to allow the production of new screening compounds to increase by orders of magnitude.

Monomer set A

NH2

A3 A4

Parallel synthesis

A1 +

A2 +

A3 +

COCl X1 COCl X2

NH2 COCl NH2

O2N

Combinatorial synthesis

A1X1

A1X2

A1X3

A2X1

A2X2

A2X3

A3X1

A3X2

A3X3

FIGURE 4.7 Traditional, parallel, and combinatorial syntheses.

Monomer set X

A1X1

A1X2

A1X3

A2X1

A2X2

A2X3

A3X1

A3X2

A3X3

A4X1

A4X2

A4X3

Monomer

The Process of New Drug Discovery and Development

4.3.1.1 High-Throughput Chemistry Traditional corporate compound collections prior to the advent of HTS consisted of thousands or tens of thousands of discrete entities, usually in the form of analytically pure powders. Additionally, it was not unusual that these samples were available in relatively large quantities of 0.1 g to ⬎1 g. It became clear in the late 1980s that the capacity to screen compounds for biological activity would greatly outstrip the compound collections of most pharma companies. Industry leaders thus started on the inevitable track of building their compound collection through purchasing of compound sets and by investing heavily in the technology of combinatorial chemistry. At that time, the basic technologies of combinatorial chemistry had been around for about a decade but had found application mainly in protein and oligonucleotide synthesis with the preparation of “libraries” of compounds, i.e., collections of analogs prepared using common chemistry performed simultaneously. These researchers built upon the Nobel Prize winning work of Merrifield in solid-phase peptide synthesis by showing that not just one but many molecular entities could be prepared through a logical sequence of chemical reactions involving a controlled and divergent polymerization. The scientific underpinnings of combinatorial chemistry, as practiced today, have changed little from 1985. What has changed is the technology available to enable the science. What follows is a brief accounting of the most important concepts of combinatorial chemistry that have had the most profound impact on the way that drug discovery research is done in the pharmaceutical setting.

4.3.2

Mixtures, Libraries, and Compound Arrays

4.3.2.1 Discrete Compound Synthesis Traditional medicinal chemistry was built upon the solid foundation of traditional organic synthesis, whereby single compounds were produced in relatively large (ca. 1 g) batches that were easily crystallized to high purity. The techniques of combinatorial chemistry shifted this tradition toward the more modern idea of increasing compound “throughput” through the preparation of many more compounds in smaller amounts (nanograms to milligrams) in purities that, as mentioned above, were measured as statistical collections rather than individuals (i.e., average purities and the number that pass minimal purity cut-offs). As first practiced by Geysen and Houghton, the preparation of combinatorial libraries produced discrete compounds of known identity through a technique known as “spatial separation,” which simply means that individual compounds in the library are produced discretely and are not mixtures. Such spatially addressable compound sets are produced in such a way as to keep separate the reaction flasks or resin beads containing the individual components of the library and perform bioassays on the discrete compounds, one at a time. Thus, if the “history” of the reaction conditions performed in each flask or on each solid support, the identity of the compounds produced is known, without resort to structure elucidation techniques. Initially, this technique, after typically an extensive reaction development stage, allowed the preparation of between 10 and 1000 discrete combinatorial products. 4.3.2.2 Mixtures: Split-and-Mix Synthesis, Deconvolution, and Encoding Classical Drug Discovery focused on the biological evaluation of individual purified compounds. But in light of HTS and combinatorial chemistry, it became widely accepted that the biological screening of mixtures of compounds should, in theory, provide much more biological data in the same amount of time used for bioassay of individual compounds,

The Impact of Combinatorial Chemistry on Drug Discovery

thus increasing efficiency and reducing costs. It was reasoned, after all, that the screening of mixtures had precedent in the biological evaluation of natural products (leading to the discovery of penicillin, amongst other drugs), and that in most cases, biological activity expressed in a mixture of compounds would be either orthogonal or simply additive thus facilitating the identification of active components. While this may in fact be the case for natural product mixtures, it is rarely the case when dealing with synthesized mixtures. Despite our attempts to create real molecular diversity in the test tube, our efforts have not even begun to anticipate the true diversity of atomic connectivity within “drug space” (estimated to be of the order of 1063 unique compounds, theory, famously in this case, greatly outpacing the amount of matter in the universe). Thus, combinatorial chemistry was never practically able to produce true chemical diversity and compounds produced in such library format ended up looking very much like one another, with the attendant similarities in biological activity profiles. It is very straightforward to produce mixtures of compounds and much of the history of Organic Chemistry has involved the avoidance of such endeavors. In theory, one simply must mix all of the monomeric parts of a final compound together, either simultaneously or consecutively, and mixture will be obtained. But performed in this manner, the mixture would most likely not be assayable (i.e., it would result in meaningless data that cannot be interpreted in terms of compound structure). However, one of the triumphs of combinatorial chemistry is that it has freed the Drug Discovery scientist from the strictures of “one compound, one assay” by showing that to some degree that the preparation of compound mixtures can result in a more rapid collection of structure activity relationship (SAR) data if performed in a controlled and logical manner. The most important technique for doing this is so-called “split-and-mix.” Indeed, this is the only practical way to produce and handle very large libraries of compounds. Houghten has reported the preparation of 1012 decapeptides using split-mix techniques. In this solid-phase technique, resin beads are split into equal portions and each portion is allowed to react with one each of the first set of monomers. The resin is then recombined, mixed thoroughly, prepared for reaction with the second set of monomer (typically involving a de-protection step), and again split into a number of portions equal to the number of reactants present in the second group of monomers. This process is repeated until all of the monomer sets have been ligated to the growing resin-bound analytes, and, rather than mixing the final resin portions together again, they are kept separate so that knowledge of the last chemical operation each batch of resin had seen was retained. Thus, for each of the cleaved mixtures, the pools have in common the identity of the final monomers, and it is therefore the identities of the monomers making up the remaining positions in the final active compounds that remain unknown. An additional complication of the bioassay of mixtures, be they natural extracts or combinatorial in nature, is that any activity present in a pool of compounds must still be assigned to the individual components to render the screening exercise useful for Drug Discovery. This is more straightforward with respect to combinatorial mixtures due to the fact that, by definition, the structural type is known, and what is not known is just the specifics of the attachment of monomer. Nonetheless, structure elucidation of active components must be made and this has traditionally been done through what is referred to as “deconvolution.” Deconvolution may be either iterative in nature or positional. Iterative deconvolution involves the systematic resynthesis of subsets of active mixtures, gradually narrowing down the activity in an iterative resynthesis or screening cycle until a single active component has been identified. There are a number of iterative deconvolution strategies that will often, if performed carefully and on ideally behaved mixtures, yield the same results. It often begins with identifying an active pool of compounds through bioassay, and if this pool was prepared using ordinary split-and-mix techniques, then the identity of the final

The Process of New Drug Discovery and Development

monomer in the reaction sequence is known. Then, smaller mixtures of compounds, each bearing that final “active” monomer, can be prepared and assayed for activity. Once again, the most active pool, having been prepared by the split-and-mix strategy, now reveals, in addition to the identity of the final monomer, the identity of a second monomer of the active final compound. This process is shown schematically in Figure 4.8. This process of split-and-mix followed by iterative deconvolution is perhaps what most people refer to as “combinatorial chemistry” itself, and dominated the literature and the venture capital of the field a decade and more ago and promised to provide us with a faster route to identifying the drugs that were already out there just waiting to be discovered! The reality of the technique was somewhat more sobering. Impurities present in these poorly characterized mixtures sometimes were discovered to contain all of the observed biological activity (the same may also be said about the poorly characterized “pure” compound). It was not uncommon for companies to retain entire groups of “resynthesis” chemists to help track down the activity of a given sample, occasionally having to prepare and purify divined impurities after resynthesis of the desired combinatorial compound revealed no biological activity. A second popular deconvolution strategy is known as positional scanning. In this technique, multiple libraries containing the same compounds, but in different mixtures, are produced and assayed in a single round of synthesis and screening. In practice, by synthesizing one full library for each monomeric position, with each individual library pool defining, via spatial separation, the identity of a single monomer, the “code” of the most active compound can be read from the identity of active pools in each full library. The obvious drawback for positional scanning is the necessity to prepare “n” libraries containing the same compounds, where n is the number of degrees of freedom or points of diversity. Thus, in practice, the combination of the uncertainties of deconvolution (you can never truly be

DA DB

Resin beads

A AA

A A A

A A

B B B

B B

B C A B

+ Split

B BB

Mix B C A B

+ C

C CC

C C

C B AC B AA CB

BD AD

A B A C

CD BD AD DC CD

EA C

A B A C

Split

CE BE ED ED EC

BE AE

+ FA

Add monomers A, B, and C

CF BF AF CF FC

BF AF

Add monomers D, E, and F Mix CDG BFG

GDB

GDA AEG CEGGFC

BEG AFG

+ HDB

CFH HEA BEH AFH CDHHEC

BFH ADH

+ IEA

CEI IEB BDI ADI AFI IDC

BFI CFI

Add monomers G, H, and I FIGURE 4.8 The split-and-mix technique for the production of combinatorial mixtures.

Split FB EC FA DB

CE AD DB BD FC AE EA CE FB CF CD FB AE AD BE BE FC AF AF DA CD CD EB

The Impact of Combinatorial Chemistry on Drug Discovery

sure that the most active compound in a library gets identified — so-called hard pooling) coupled with the fact that nondiverse libraries will have many compounds of varying activities present in mixtures leading to departure from ideal behavior.

4.3.2.3 Encoded Combinatorial Synthesis A technique that was very popular but had diminished in significance over the past 5 years is the technique of encoding. In theory, this should be a very useful and concise method for the structure elucidation of active compounds within a resin-bound mixture. In its simplest form, encoding involves the orthogonal placement of chemical identification tags on resin beads containing the target compound, with the identity of the individual monomers that make up the final product being “encoded” by the very constitution of the tags themselves. This technique has proven fairly tricky because it requires that discrete compounds prepared on individual beads in mixtures be assayed either while still attached to the bead of origin or else cleaved from that single isolated bead and assayed while retaining the bead itself for analysis. Why is this so? Because the orthogonal synthesis strategies for analyte and tag leave the tag still attached to the bead, ready for analysis should the analyte prove to have biological activity. Another technological hurdle for the encoding technique has been the obvious requirement that the tag be much easier to analyze and identify than the test compound itself (otherwise one could simply use some analytical technique to identify the structure of the product rather than an inactive chemical surrogate). This has been done most successfully by a number of techniques including secondary amide strategy, mass encoding, polyhalogenated aromatics, and radio-frequency tagging. The fact that encoding techniques for compound identification in the split-and-mix protocol is so specialized and often requiring extraordinary hardware is punctuated by the fact that whole companies or divisions have been set up or repositioned to produce, promote, and capitalize the individual encoding strategies. It may be conjectured at this point that the vast majority of expenditure by the pharmaceutical and biopharmaceutical industries on the “combina

The Process of New Drug Discovery and Development - PDF Free Download (2024)

FAQs

What is the process of new drug discovery and development? ›

To be deemed a “success,” a new drug must make it through five specific phases: 1) discovery and development, 2) preclinical research, 3) clinical research, 4) FDA review, and 5) safety monitoring. Below, we explore each step in more detail.

What are the four stages of drug development pdf? ›

Drug development can be divided into four phases: discovery, preclinical studies, clinical development and market approval.

View Details ›

What is the process of discovery and development of medicines? ›

Finding new drugs usually consists of five main stages: 1) a pre-discovery stage in which basic research is performed to try to understand the mechanisms leading to diseases and propose possible targets (e.g., proteins); 2) the drug discovery stage, during which scientists search for molecules (two main large families, ...

What are the five steps in the drug development process? ›

Content current as of:

Step 1: Discovery and Development.
Step 2: Preclinical Research.
Step 3: Clinical Research.
Step 4: FDA Drug Review.
Step 5: FDA Post-Market Drug Safety Monitoring.

Jan 4, 2018

What's the most important step in drug discovery and development and why? ›

Target Identification and Validation

One of the key factors in designing a good drug is having a crystal clear understanding of the pathogenesis of a disease. A suitable biological target is said to be “druggable” when a therapeutic molecule, called a “hit”, can modify its biological activity.

Discover More Details ›

What is drug discovery and development basics? ›

Drug discovery and development can be described as the sum total of steps taken by research-intensive entity to identify a new chemical or biological substance and transform it into a product approved for use by patients.

Find Out More ›

What are the 5 pillars of drug development? ›

These pillars include target identification and validation, lead discovery, lead optimization, preclinical testing, and clinical trials. Each pillar plays a vital role in the overall success of drug discovery, contributing to the development of safe and effective drugs.

Show Me More ›

What is the difference between drug discovery and drug development? ›

Drug Discovery falls within the medical, biotechnology and pharmacology fields. It is the process that leads to the discovery of a new medication. Drug Development, on the other hand, speaks mostly to the complete process of bringing this newly discovered drug to the market.

See Details ›

How long does the drug discovery process take? ›

The process of developing a new drug usually takes a total of 10 to 15 years on average. The process can be divided into three main phases: Drug discovery, the first phase in which candidate compounds are selected based on their pharmacological properties.

Learn More Now ›

What happens in discovery and development of drugs? ›

The process begins with the identification of a new target molecule, a protein or other molecule involved in the disease process. Once a target molecule is identified, scientists must design and synthesize a new compound that will interact with the target molecule and influence or inhibit its function.

View Details ›

What is the process of modern drug discovery? ›

Nowadays, the steps include identifying few potential molecule or hits, through screening approach, advanced medicinal chemistry to evaluate the parameters, selecting the most potent molecular lead, and finally the optimization of lead to minimize or obliterate the side effects, if any [17, 18].

Discover More ›

What are the modern methods in drug discovery and development? ›

Modern drug discovery involves the identification of screening hits, medicinal chemistry and optimization of those hits to increase the affinity, selectivity (to reduce the potential of side effects), efficacy/potency, metabolic stability (to increase the half-life), and oral bioavailability.

Learn More ›

What are the stages of drug discovery and development? ›

A: The drug discovery stage consists of 5 stages, namely: Stage 1 drug development, where drugs are developed from a wide variety of sources (chemical, natural, biological, etc.) Stage 2 preclinical trials (including in silico, in vitro, and in vivo trials) Stage 3 clinical trial (consisting of 4 phases)

What is the process and development of new drugs? ›

For small-molecule drugs, the path to a marketed drug involves a long and exhaustive journey through basic research, discovery of the medicine, preclinical development tests, increasingly complicated clinical trials with humans, and regulatory approval by the Food and Drug Administration (FDA).

Learn More Now ›

What are the 5 R's of drug development? ›

This strategic review, discussed in depth in a previous publication in this journal⁷, led to the development and implementation of the '5R framework' (Fig. 1), in which decision-making is focused on five technical determinants (the right target, right tissue, right safety, right patient and right commercial potential).

Learn More Now ›

What are the steps in structure based drug discovery? ›

Structure-Based Drug Discovery

First phase. Identification of potential therapeutic targets and active ligands. ...
Second phase. Synthesis and optimization of top hits. ...
Third phase. ...
Fourth phase. ...
Computational techniques needed to process the “big data” generated include: ...
Efficiency. ...
Successes in drug discovery.

Know More ›

The Process of New Drug Discovery and Development - PDF Free Download (2024)

FAQs

What is the process of new drug discovery and development? ›

What is the process of modern drug discovery? ›

References