Drug-Drug Interactions for Therapeutic Biologics E-Book

Honghui Zhou

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Sprache: Englisch

Seitenzahl: 612

Veröffentlichungsjahr: 2013

Beschreibung

Strategize, plan, and execute comprehensive drug-drug interaction assessments for therapeutic biologics Offering both theory and practical guidance, this book fully explores drug-drug interaction assessments for therapeutic biologics during the drug development process. It draws together and analyzes all the latest findings and practices in order to present our current understanding of the topic and point the way to new research. Case studies and examples, coupled with expert advice, enable readers to better understand the complex mechanisms of biologic drug-drug interactions. Drug-Drug Interactions for Therapeutic Biologics features contributions from leading international experts in all areas of therapeutic biologics drug development and drug-drug interactions. The authors' contributions reflect a thorough review and analysis of the literature as well as their own firsthand laboratory experience. Coverage includes such essential topics as: * Drug-drug interaction risks in combination with small molecules and other biologics * Pharmacokinetic and pharmacodynamic drug-drug interactions * In vitro methods for drug-drug interaction assessment and prediction * Risk-based strategies for evaluating biologic drug-drug interactions * Strategies to minimize drug-drug interaction risk and mitigate toxic interactions * Key regulations governing drug-drug interaction assessments for therapeutic biologics. Drug-Drug Interactions for Therapeutic Biologics is recommended for pharmaceutical and biotechnology scientists, clinical pharmacologists, medicinal chemists, and toxicologists. By enabling these readers to understand how therapeutic biologics may interact with other drugs, the book will help them develop safer, more effective therapeutic biologics.

Leseprobe

Contents

Cover

Title Page

Copyright

Preface

About the Editors

Contributors

Chapter 1: Drug Interactions for Therapeutic Proteins: A Journey Just Beginning

1.1 Introduction

1.2 Scientific/Regulatory Landscape of Therapeutic Protein–Drug Interactions

References

Chapter 2: Pharmacokinetic and Pharmacodynamic-Based Drug Interactions for Therapeutic Proteins

2.1 Introduction

2.2 Distribution, Catabolism/Metabolism, and Excretion Mechanisms of Therapeutic Proteins

2.3 Major Mechanisms of Therapeutic Protein–Drug Interactions

2.4 Strategies to Assess the Risk of Therapeutic Protein–Drug Interactions in Clinical Development of Therapeutic Proteins

2.5 Summary

Acknowledgments

References

Chapter 3: Drug Interaction Assessment Strategies: Small Molecules versus Therapeutic Proteins

3.1 Introduction

3.2 Drug-Metabolizing Enzymes

3.3 Transporters

3.4 Conclusion

References

Chapter 4: Model-Independent and Model-Based Methods to Assess Drug–Drug Interactions for Therapeutic Proteins

4.1 Introduction

4.2 TP-DIs

4.3 In Vitro and In Vivo Approaches for Evaluating TP-DI

4.4 Bioanalytical Considerations

4.5 Conclusion

References

Chapter 5: Utility of In Vitro Methods in Drug–Drug Interaction Assessment and Prediction for Therapeutic Biologics

5.1 Introduction

5.2 Mechanisms Involved in Suppression of Drug-Metabolizing Enzymes

5.3 In Vitro Assays

5.4 Effects of Cytokines on Metabolizing Enzymes and Transporters

5.5 Summary and Conclusion

References

Chapter 6: Use of Animal Models for Projection of Clinical Drug–Drug Interactions for Therapeutic Proteins

6.1 Introduction

6.2 Selection of the Animal Model

6.3 Study Design

6.4 Disease Models

6.5 Emerging Challenges

6.6 Conclusions

References

Chapter 7: The Cocktail Approach and Its Utility in Drug–Drug Interaction Assessments for Therapeutic Proteins

7.1 Assessment of Enzyme Activities Using the Cocktail Approach

7.2 Guidelines Applicable for Cocktail Drug–Drug Interaction Studies

7.3 Cocktail Interaction Studies with Therapeutic Proteins: Special Features

7.4 Published Cocktail Interaction Studies with Therapeutic Proteins

7.5 Conclusions

References

Chapter 8: Logistic Considerations in Study Design for Biologic Drug–Drug Interaction Assessments

8.1 Introduction

8.2 Challenges in the Conduct of a TP–Drug Interaction Study

8.3 TP–Drug Interaction Study Design

8.4 Timing of TP–Drug Interaction Study

8.5 Strategic Planning of TP–Drug Interaction Studies

8.6 Considerations in Study Design

8.7 Data Analysis

8.8 Prospectively Design of TP–Drug Interaction Study

8.9 Summary

References

Chapter 9: Statistical Considerations in Assessing Drug–Drug Interactions for Therapeutic Biologics

9.1 Introduction

9.2 Methodology for Drug–Drug Interaction Assessments

9.3 Population Pharmacokinetics for Drug–Drug Interaction Assessments: Ustekinumab

9.4 Summary

References

Chapter 10: Scientific Perspectives on Therapeutic Protein Drug–Drug Interaction Assessments1

10.1 Introduction

10.2 Therapeutic Protein–Drug Interaction Studies

10.3 Types of Study Designs

10.4 Labeling Implications

10.5 Conclusion

References

Chapter 11: Disease–Drug–Drug Interaction Assessments for Tocilizumab—A Monoclonal Antibody against Interleukin-6 Receptor to Treat Patients with Rheumatoid Arthritis

11.1 Introduction

11.2 Preclinical Evaluation

11.3 Clinical DDDI Evaluations

11.4 Labeling

11.5 Discussion

References

Chapter 12: Drug–Drug Interactions for Etanercept—A Fusion Protein

12.1 Etanercept Background

12.2 Mechanisms of Drug Interactions

12.3 Pharmacodynamic Drug Interactions

12.4 Results of Drug Interaction Studies with Etanercept

12.5 Conclusions

Acknowledgments and Conflicts of Interest

References

Chapter 13: Drug Interactions of Cytokines and Anticytokine Therapeutic Proteins

13.1 Introduction

13.2 Clinical Relevance of Cytokine-Mediated Suppression and Desuppression of ADME Enzymes

13.3 Mechanism

13.4 Can Preclinical Models Be Used to Predict Clinical Suppression or Desuppression?

13.5 Current Regulatory Perspective

13.6 Clinical Options

13.7 Conclusions

Acknowledgments

Declaration of Interest

References

Chapter 14: Drug Interactions for Growth Factors and Hormones

14.1 Introduction

14.2 Growth Factors

14.3 Hormones

14.4 Conclusions

References

Chapter 15: Drug–Drug Interactions for Nucleic Acid-Based Derivatives

15.1 Introduction

15.2 Clinical Pharmacokinetics

15.3 Drug–Drug Interactions

15.4 Other Considerations

15.5 Summary

References

Appendix: Monographs for Drug-Drug Interactions of Therapeutics Biologics

Index

Copyright © 2013 by John Wiley & Sons, Inc. All rights reserved

Published by John Wiley & Sons, Inc., Hoboken, New Jersey

Published simultaneously in Canada

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at [phone], outside the United States at [phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Drug-drug interactions for therapeutic biologics / edited by Honghui Zhou, Bernd Meibohm.

p. ; cm.

Includes bibliographical references and index.

ISBN 978-1-118-03216-9 (cloth)

I. Zhou, Honghui. II. Meibohm, Bernd.

[DNLM: 1. Drug Interactions. 2. Drug Discovery. 3. Proteins–therapeutic use.QV 37.5]

615.7′045–dc23

2013000334

Preface

In the past two decades we have seen tremendous progress in the area of therapeutic biologics. With more and more therapeutic proteins being used in poly-pharmacy settings and the potential toxicity risk of drug-drug interactions, there is during drug development a need for a thorough review of potential drug-drug interactions involving therapeutic biologics. However, literature references on this topic have so far been scarce. Thus, we feel the scientific community would benefit from a systemic presentation of the current status of knowledge on this topic. The proposed book project is intended to fill this void.

The book is expected to greatly benefit scientists and researchers in the pharmaceutical and biotech industry as well as academia who are involved in drug development for both therapeutic biologics and traditional small molecule drugs. The expected audience will be pharmaceutical and biotech scientists, clinical pharmacologists, medicinal chemists, and toxicologists. Scientists and clinicians in pharmaceutical and biotech industry can utilize the book as a resource to strategize, plan and implement drug-drug interaction assessments involving therapeutic biologics. Academic pharmacokinetics, pharmacology, and biochemistry scientists working on mechanisms for biologic drug-drug interactions will also find this book very useful as a compilation of the current state-of-the-art.

The current book focuses on both theoretical and practical aspects of drug-drug interaction assessments for therapeutic biologics in drug development. We are fortunate that many of the experts and opinion leaders from various areas of therapeutic biologics drug development and drug-drug interactions have participated in the writing of this book, and we are indebted to them for their time and dedication to participate in this project. The content includes topics such as drug-drug interaction risks (both theoretical and observed) in combination with small molecules and with other biologics, pharmacokinetic drug-drug interactions, pharmacodynamic drug-drug interactions, utility of in vitro methods in drug-drug interaction assessment and prediction, modeling-independent and modeling-based methods to assess potential drug-drug interactions, risk-based strategies for evaluating biologic drug-drug interactions, strategies to minimize drug-drug interaction risk and mitigate toxic interactions, and regulatory perspectives on biologic drug-drug interaction assessments.

Though there are several books covering drug-drug interactions for conventional small molecules, a book that is comprehensive with all the above topics for biotherapeutics is not currently available. Thus, we are convinced that that textbook addresses a currently unmet need in drug development sciences and we are confident that the scientific community will benefit from the experience and expertise of the contributors to this book project.

Honghui Zhou

Bernd Meibohm

Spring House, PA, and

Memphis, TN

August 2012

About the Editors

Honghui Zhou, PhD, FCP

Honghui Zhou is currently a Senior Scientific Director at Janssen Research and Development, LLC, Johnson & Johnson and is heading the Pharmacokinetics and Pharmacodynamics Department within Biologics Clinical Pharmacology.

Prior to joining Centocor, Dr. Zhou was a Director of Clinical Pharmacology at Wyeth Research (now Pfizer). He also worked for Novartis Pharmaceuticals Corp. and Johnson & Johnson Pharmaceutical Research and Development in the area of clinical pharmacology and pharmacokinetics/pharmacodynamics (PK/PD) in both small molecular drugs and therapeutic proteins. In 2012, Honghui was elected as a Janssen Fellow.

Dr. Zhou has authored more than 150 original peer-reviewed scientific papers, book chapters, and conference abstracts in PK/PD and drug–drug interactions. He has also been an invited speaker in many national and international conferences. He is board certified by American Board of Clinical Pharmacology (ABCP) and is Fellow of Clinical Pharmacology (FCP) in ACCP. He currently serves as a section editor for Biologics for the Journal of Clinical Pharmacology. He also serves as Board of Reagents of ACCP (2009–2014). He co-chairs the IQ Therapeutic Protein–Drug Interaction Working Group (previously Pharma/FDA/Academia Therapeutic Protein–Drug Interaction Steering Committee). Honghui is a graduate of the China Pharmaceutical University, BS in Pharmacology, and the University of Iowa, PhD in Pharmaceutics.

Bernd Meibohm, PhD, FCP

Bernd Meibohm is a Professor of Pharmaceutical Sciences and Associate Dean for Research and Graduate Programs at the College of Pharmacy, the University of Tennessee Health Science Center, Memphis.

Prior to joining the University of Tennessee, Dr. Meibohm conducted research at the University of South Carolina and the University of Florida. Dr. Meibohm's scientific interests include chronic inflammatory pulmonary diseases, pediatric pharmacotherapy, and the application of quantitative modeling and simulation techniques in preclinical and clinical drug development, with a specific focus on biotech drugs. His research has resulted in two textbooks, over 200 peer-reviewed scientific papers, book chapters, and conference abstracts, and over 100 invited scientific presentations to national and international audiences.

Dr. Meibohm is a Fellow of the American Association of Pharmaceutical Scientists (AAPS) and American College of Clinical Pharmacology (ACCP). He was the 2010 Chair for the Pharmacokinetics, Pharmacodynamics and Drug Metabolism (PPDM) section of AAPS and currently serves as the President-Elect for ACCP. Dr. Meibohm is also serving as associate editor for The AAPS Journal and as section editor for Pharmacokinetics and Pharmacodynamic for the Journal of Clinical Pharmacology; he is a member of the editorial boards of the Journal of Pediatric Pharmacology and Therapeutics, the Journal of Pharmacokinetics and Pharmacodynamics, Les Annales Pharmaceutiques Françaises, and Die Pharmazie.

Contributors

Jeffrey S. Barrett, Laboratory for Applied PK/PD, Division of Clinical Pharmacology and Therapeutics, The Children's Hospital of Philadelphia, Philadelphia, PA, USA; School of Medicine, University of Pennsylvania, Philadelphia, PA, USA

Barbara J. Brennan, Hoffmann-La Roche Inc., Nutley, NJ, USA

Souvik Chattopadhyay, Drug Metabolism and Pharmacokinetics, Drug Safety Sciences, Janssen Research and Development, LLC, Spring House, PA, USA

Shannon Dallas, Drug Metabolism and Pharmacokinetics, Drug Safety Sciences, Janssen Research and Development, LLC, Spring House, PA, USA

Leslie J. Dickmann, Pharmacokinetics and Drug Metabolism, Amgen Inc, Seattle, WA, USA

Martin E. Dowty, Pfizer Inc., Andover, MA, USA

Raymond Evers, Department of Pharmacokinetics, Pharmacodynamics, and Drug Metabolism, Transporters and In Vitro Technologies, Merck & Co., Inc., Rahway, NJ, USA

Uwe Fuhr, Department of Pharmacology, Clinical Pharmacology, University Hospital of Cologne, Köln, Germany

Sandhya Girish, Department of Clinical Pharmacology, Genentech, Inc., South San Francisco, CA, USA

Chuanpu Hu, Pharmacokinetics and Pharmacodynamics, Biologics Clinical Pharmacology, Janssen Research & Development, LLC, Spring House, PA, USA

Shiew-Mei Huang, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA

Alexander Jetter, Department of Clinical Pharmacology and Toxicology, University Hospital Zürich, Zürich, Switzerland

Amita Joshi, Department of Clinical Pharmacology, Genentech, Inc., South San Francisco, CA, USA

Tarundeep Kakkar, Genomics Institute of the Novartis Research Foundation, BDU Translational Sciences, San Diego, CA, USA

Simone Kasek, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Narendra Kishnani, Department of Biotransformation, Pharmaceutical Candidate Optimization, Bristol-Myers Squibb Co., Princeton, NJ, USA

Joan Korth-Bradley, Clinical Pharmacology, Pfizer Inc., Collegeville, PA, USA

Eugenia Kraynov, Pfizer Inc., San Diego, CA, USA

Jocelyn Leu, Janssen Research & Development, LLC, Spring House, PA, USA

Christine Li, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Wararat Limothai, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Dan Lu, Department of Clinical Pharmacology, Genentech, Inc., South San Francisco, CA, USA

Dora Babu Madhura, University of Tennessee Health Science Center, Memphis, TN, USA

Bernd Meibohm, College of Pharmacy, The University of Tennessee Health Science Center, TN, USA

Theresa Nguyen, Department of Pharmacokinetics, Pharmacodynamics, and Drug Metabolism, Transporters and In Vitro Technologies, Merck & Co., Inc., Rahway, NJ, USA

Chetan Rathi, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Sumit Rawal, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Kellie Reynolds, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA

Josiah Ryman, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Carlo Sensenhauser, Drug Metabolism and Pharmacokinetics, Drug Safety Sciences, Janssen Research and Development, LLC, Spring House, PA, USA

Jose Silva, Drug Metabolism and Pharmacokinetics, Drug Safety Sciences, Janssen Research and Development, LLC, Spring House, PA, USA

J. Greg Slatter, Pharmacokinetics and Drug Metabolism, Amgen Inc., Seattle, WA, USA

Yu-Nien (Tom) Sun, Quantitative Pharmacology, Department of Pharmacokinetics and Drug Metabolism, Amgen Inc., Thousand Oaks, CA, USA

Frank-Peter Theil, Department of Clinical Pharmacology, Genentech, Inc., South San Francisco, CA, USA

Margaret Thomson, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Ashit Trivedi, College of Pharmacy, The University of Tennessee Health Science Center, Memphis, TN, USA

Jian Wang, Office of Translational Sciences, Office of Clinical Pharmacology, Center for Drug Evaluation and Research, Food and Drug Administration (FDA), Silver Spring, MD, USA

Yow-Ming C. Wang, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA

Larry C. Wienkers, Pharmacokinetics and Drug Metabolism, Amgen Inc, Seattle, WA, USA

Di Wu, Laboratory for Applied PK/PD, Division of Clinical Pharmacology and Therapeutics, The Children's Hospital of Philadelphia, Philadelphia, PA, USA

Lei Zhang, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA

Xiaoping Zhang, Hoffmann-La Roche Inc., Nutley, NJ, USA

Hong Zhao, Office of Clinical Pharmacology, Office of Translational Sciences, Center for Drug Evaluation and Research, Food and Drug Administration, Silver Spring, MD, USA

Honghui Zhou, Pharmacokinetics and Pharmacodynamics, Biologics Clinical Pharmacology, Janssen Research and Development, LLC, Spring House, PA, USA

Min Zhu, Quantitative Pharmacology, Department of Pharmacokinetics and Drug Metabolism, Amgen Inc., Thousand Oaks, CA, USA

Chapter 1

Drug Interactions for Therapeutic Proteins: A Journey Just Beginning

Honghui Zhou and Bernd Meibohm

1.1 Introduction

Over the last three decades, therapeutic proteins, in particular, antibody-based biotherapeutics, have played an increasingly important role in pharmacotherapy, and in some therapeutic areas, such as immune-mediated inflammatory diseases (IMIDs) and oncology, therapeutic proteins have fundamentally changed the therapeutic paradigm. Therapeutic proteins have also presented enormous commercial potential. For example, the top 10 antibody-based biotherapeutics accounted for around $50 billion of worldwide sales in 2011.1 The majority of these are either in IMID (adalimumab, etanercept, infliximab, rituximab, natalizumab, omalizumab) or in oncology (rituximab, bevacizumab, trastuzumab, cetuximab) therapeutic areas. Hundreds of investigational antibody-based and other protein therapeutics are currently under development at different stages, spanning discovery to phase III clinical investigations.

Owing to an expected increase in the coadministration of biotherapeutic agents with established pharmacotherapy regimens, there is an increasing likelihood for the occurrence of clinically relevant drug interactions. Therapeutic proteins, however, have long been perceived to have a very low propensity for drug–drug interactions because they are eliminated via catabolic routes, either nonspecific pathways or target-mediated pathways, that are independent from the elimination pathways of small molecules, which are usually eliminated by noncatabolic pathways such as hepatic metabolism via cytochrome P450 (CYP), renal excretion, and biliary excretion. Though it has been known for decades that some cytokines such as interferons, tumor necrosis factor α (TNF-α), and interleukin 6 (IL-6) can down-regulate CYPs,2 very few drug–drug interactions had been reported for biotherapeutics until 2007, when two review articles containing examples of drug interactions involving therapeutic proteins were published.3,4 The majority of reported drug interactions associated with therapeutic proteins seem to be indirect; however, a mechanistic understanding for many of the observed interactions is still lacking.5–7

1.2 Scientific/Regulatory Landscape of Therapeutic Protein–Drug Interactions

To help assess the common practice of evaluating therapeutic protein–drug interactions across the biotech/pharma industry and to shed some light on how and when a sensible therapeutic protein–drug interaction assessment strategy should be incorporated into therapeutic protein drug development, a survey was conducted within the Biotechnology Industry Organization (BIO) member companies in 2010. It is not surprising that a majority of the responder companies did not have internal strategies for evaluating therapeutic protein–drug interactions at the time of the survey. Nevertheless, the most favored approach employed to address potential drug–drug interactions of therapeutic proteins at that time was a tailored and integrated (i.e., case-by-case) strategy that addressed the possibility of the therapeutic protein acting as either an initiator (perpetrator) or target (victim) of the interaction. Despite the fact that many of the companies responding to the survey reported drug–drug interactions involving therapeutic proteins, the majority of the clinical therapeutic protein–drug interactions studied did not warrant dose adjustment. In other words, most of the observed clinical therapeutic protein–drug interactions did not reach a clinically significant level. Routine in vitro screening and preclinical drug–drug interaction studies were not widely used for the evaluation of therapeutic proteins. For clinical development, dedicated clinical pharmacology drug–drug interaction studies were the most frequently used methodology, followed by population pharmacokinetics-based and clinical cocktail approaches.8

The BIO survey results indicated that there was a pressing need to have a science-driven and risk-based assessment strategy for therapeutic protein–drug interactions (TP-DIs). A closer collaboration among scientists from the biotech/pharma industry, regulatory agencies, and academia appeared to be essential in reaching that goal. As a result, a TP-DI steering committee from industry, the FDA, and academia was founded in 2009 to address this challenge. The initial scope of this committee was focused only on pharmacokinetics (PK) and metabolism-based drug–drug interactions for the major classes of therapeutic proteins, including monoclonal antibodies, fusion proteins, cytokines (excluding antibody–drug conjugates). The committee intended to investigate the potential for therapeutic proteins to interact, either as initiators or targets, with drugs that are metabolized via CYP enzyme pathways. Two major focus areas the committee concentrated on were (1) to critically assess standard in vitro screening techniques and methodologies (e.g., for cytokine-related drug–drug and drug–disease interactions) and (2) to provide guidance for study designs with consideration of specific disease area (e.g., oncology) issues and timings.

Several scientific knowledge gaps were identified from a 2010 American Association of Pharmaceutical Scientists (AAPS) workshop on Strategies to Address Therapeutic Protein-Drug Interactions during Clinical Development.9 One gap was associated with the relevance of in vitro systems to assess potential therapeutic protein–drug interactions, and another gap was a lack of best practices for using population PK-based approaches to assess potential therapeutic protein–drug interactions. The steering committee also identified similar gaps and consequently formed two working groups to specifically tackle them.

During the same time period, scientists from the FDA published two important review articles on TP-DI, but these were mostly from a regulatory perspective.10,11 In 2012, a draft of a new drug–drug interaction guidance document was made available by the FDA for public comments.12 That draft included a dedicated section on therapeutic protein–drug interaction to address specifically the newly emerging area of drug–drug interactions with therapeutic proteins.

The Workshop on Recent Advances in the Investigation of Therapeutic Protein Drug-Drug Interactions: Preclinical and Clinical Approaches was held on June 4–5, 2012. The workshop, co-sponsored by the FDA Office of Clinical Pharmacology and the Drug Metabolism and Clinical Pharmacology Leadership Group of the IQ Consortium, was intended to facilitate a better understanding of the current science, investigative approaches, knowledge gaps, and regulatory requirements related to the evaluation of therapeutic protein–drug interactions. The workshop also provided an opportunity to discuss the current views from the two (in vitro and population PK approaches) therapeutic protein–drug interaction working groups. The proceedings from this workshop are being compiled with the intent of issuing white papers in these subject areas. It is anticipated that the recommendations from both white papers will soon provide pharmaceutical scientists with sensible and scientifically sound best practices and an assessment framework for using in vitro and population PK-based approaches for evaluating therapeutic protein–drug interactions.

Our current understanding of the mechanisms of many therapeutic protein–drug interactions is still in its infancy. Much basic research needs to be conducted to verify several existing hypotheses related to therapeutic protein–drug interactions. Continued close collaborations among fellow scientists in industry, academia, and regulatory agencies will be vital to generate more plausible mechanistic hypotheses and collectively address the many challenges in this area. Through these collaborative efforts, the knowledgebase on therapeutic protein–drug interactions will likely be largely expanded in the near future, and it is hoped and anticipated that over the next decade a similar level of mechanistic understanding and systemic assessment methodology will be achieved and developed for drug interactions with protein therapeutics as it has been established in the last two decades for small molecule drugs. The journey toward that goal has just begun.

References

1. R&D Pipeline News, Top 30 Biologics 2011, April 25, 2012. Available at www.pipelinereview.com.

2. Morgan ET. Regulation of cytochrome P450 by inflammatory mediators: why and how? Drug Metab Dispos29, 207–12 (2001).

3. Seitz K, Zhou H. Pharmacokinetic drug-drug interaction potentials for therapeutic monoclonal antibodies: reality check. J Clin Pharmacol47, 1104–18 (2007).

4. Mahmood I, Green MD. Drug interaction studies of therapeutic proteins or monoclonal antibodies. J Clin Pharmacol47, 1540–54 (2007).

5. Zhou H, Mascelli MA. Mechanisms of monoclonal antibody-drug interactions. Annu Rev Pharmacol Toxicol51, 359–72 (2011).

6. Kraynov E, Martin SW, Hurst S, et al. How current understanding of clearance mechanisms and pharmacodynamics of therapeutic proteins can be applied for evaluation of their drug-drug interaction potential. Drug Metab Dispos39, 1779–83 (2011).

7. Meibohm B. Mechanistic basis for potential drug-drug interactions with therapeutic proteins. Paper presented at the Workshop on Recent Advances in the Investigation of Therapeutic Protein Drug-Drug Interactions: Preclinical and Clinical Approaches. Silver Spring, MD, June 4–5, 2012.

8. Lloyd P, Zhou H, Theil FP, et al. Highlights from a recent BIO survey on therapeutic protein-drug interactions. J Clin Pharmacol52, 1755–63 (2012).

9. Girish S, Martin SW, Peterson MC, et al. AAPS workshop report: strategies to address therapeutic protein-drug interactions during clinical development. AAPS J13, 405–16 (2011).

10. Huang SM, Zhao H, Lee JI, et al. Therapeutic protein-drug interactions and implications for drug development. Clin Pharmacol Ther87, 497–503 (2010).

11. Lee JI, Zhang L, Men AY, et al. CYP-mediated therapeutic protein-drug interactions: clinical findings, proposed mechanisms and regulatory implications. Clin Pharmacokinet49, 295–310 (2010).

12. U.S. Department of Health and Human Services, FDA, and Center for Drug Evaluation and Research. Guidance for industry: drug interaction studies–study design, data analysis, implications for dosing, and labeling recommendations. February 2012. Available at www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM292362.pdf.

Chapter 2

Pharmacokinetic and Pharmacodynamic-Based Drug Interactions for Therapeutic Proteins

Dan Lu Sandhya Girish, Frank-Peter Theil, and Amita Joshi

2.1 Introduction

Therapeutic proteins (TPs) are protein products manufactured for pharmaceutical use. They include monoclonal antibodies (mAbs), antigen-binding fragments, antibody–drug conjugates (ADCs), cytokines, enzymes, growth factors, and miscellaneous proteins (e.g., fusion proteins and recombinant proteins). The development of therapeutic biologics, including TPs, is increasingly important in the pharmaceutical industry.1 To achieve greater clinical benefits, TPs are often being combined with other TPs and small molecule drugs (SMDs). Whether drug interactions (DIs) in combination therapy result in an undesirable impact on efficacy and safety needs evaluation. To date, for the observed therapeutic protein–drug interactions (TP-DIs) that affect the exposure of TPs, only a modest change in exposure is observed and no impact on safety or efficacy has been documented, suggesting a limited clinical relevance.2 This might be because most TPs have a relatively large therapeutic range compared to the majority of traditional SMDs. However, TP-DIs that affect the exposure of some drugs with a narrow therapeutic range (NTR), such as some SMDs and ADCs, may have an impact on efficacy and safety. The TP-DIs that result in enhanced toxicity due to undesirable pharmacodynamic (PD) interactions without a direct impact on exposures may also be clinically relevant. Thus the evaluation of TP-DIs is an important and evolving topic for the development of TPs in combination with other drugs.

This chapter reviews the major absorption, distribution, metabolism, and excretion (ADME) pathways of TPs, summarizes the potential mechanisms of pharmacokinetic (PK) and PD-based TP-DIs, and recommends a question-based TP-DI risk assessment strategy during clinical development. The DIs for some nonprotein biologics such as nucleic acid–based derivatives are reviewed in other chapters.

2.2 Distribution, Catabolism/Metabolism, and Excretion Mechanisms of Therapeutic Proteins

ADME processes determine the PK properties of SMDs and TPs. In drug combinations, one drug may impact the ADME processes of another drug, leading to a change in its exposure. For SMDs, absorption is mainly mediated by the solubility and permeability of a SMD and its interaction with transporters. Distribution of SMDs is mediated by several key processes, such as blood perfusion, permeability across membrane barriers, and nonspecific binding. Metabolism of SMDs is mainly mediated by cytochrome P450 (CYP) and non-CYP enzymes (such as N-acetyl and glucuronyl transferase). Excretion of SMDs mainly occurs via renal filtration or renal and biliary secretion mediated by transporters.3Figure 2-1a depicts the typical clearance pathways for SMDs.

Figure 2-1 Comparison of clearance mechanisms of (a) a SMD and (b) a TP. CYP: cytochrome P450; FcRn: neonatal Fc receptor; SMD: small molecule drug; TMDD: target-mediated drug disposition; TP: therapeutic protein.

For TPs, ADME processes are different from SMDs.4–6 Owing to high gastrointestinal enzyme activity and low permeability through the gastrointestinal mucosa, most TPs are not therapeutically active on oral administration. Consequently other routes of administration, such as intravenous, subcutaneous, and intramuscular routes of injection are used for TPs.6 For subcutaneous injections of TPs with large molecular weight, convective transport across local lymphatic vessels is the major mechanism of absorption from the injection site.7 The processes of distribution, catabolism, and excretion of TPs are reviewed in detail in this chapter. As illustrated in Figure 2-1b, the catabolism of TPs are mainly mediated by nonspecific clearance pathways. Target-mediated drug disposition (TMDD) and immunogenicity-mediated pathways also play roles in the clearance of some TPs. ADCs belong to a more complex group of TPs, made up of both a mAb and a small molecule cytotoxic agent. Their PK properties are also reviewed here.

2.2.1 Distribution of Therapeutic Proteins

Distinct from most SMDs that widely distribute to various tissues and organs after administration, distribution of mAbs and large TPs is usually confined by their large size; consequently the molecules have limited mobility through membranes. This often results in a relatively small volume of distribution. The volume of distribution of mAbs and ADCs at steady state is often a low multiple (1 to 2) of physiologic plasma volume (approximately 50 mL/kg). This is similar to the distribution characteristics for an endogenous immunoglobulin G (IgG). The distribution of TPs outside the systemic circulation is mediated by limited interstitial penetration in various organs, convection-dominated lymphatic drainage, specific and nonspecific binding to peripheral tissues, and target-mediated cellular uptake. For TPs with relatively low molecular mass, preclinical study results have demonstrated better tissue penetration.6 Unlike SMDs, transporters usually do not play a role in the distribution of large TPs.

2.2.2 Catabolism of Therapeutic Proteins

Most TPs are mainly catabolized by proteolytic degradation in cellular lysosomes through nonspecific pathways, resulting in peptides and amino acids that are reutilized for protein synthesis.4, 6 It is generally believed that nonspecific catabolism of TPs may take place predominantly in the lysosomes of endothelial cells and the mononuclear phagocyte system (MPS). TPs, such as mAbs and some fusion proteins containing a fragment crystallizable region (Fc region), interact with neonatal Fc receptors (FcRn) similar to endogenous IgGs. In adults, FcRn is primarily expressed in the vascular endothelial cells. FcRn is also detectable on monocytes, tissue macrophages, and dendritic cells. The FcRn-mediated recycling protects IgG type of proteins (e.g., endogenous IgGs, mAbs, and Fc fusion proteins) from proteolytic degradation in lysosomes, consequently delaying their catabolism and prolonging their half-lives compared to other types of proteins that are not rescued by FcRn-mediated recycling.4, 5 As a result, endogenous IgGs, mAbs, and Fc fusion proteins usually have relatively long half-lives, ranging from several days to weeks. The pathways of nonspecific clearance and FcRn-mediated recycling are typically low-affinity and high-capacity pathways, which are usually nonsaturable at therapeutically relevant doses. For mAbs, relatively constant values of nonspecific clearance are found in each species. In humans, this value is 3–6 mL/day/kg and is affected by multiple pathophysiological and demographical covariates.8

In addition to the nonspecific clearance pathways, TMDD may also play a role in the clearance of target-binding proteins (e.g., mAbs, Fc fusion proteins, recombinant proteins). By this mechanism, a TP is cleared from the systemic circulation by binding to its target antigen followed by proteolytic degradation. The target antigens can be cell-surface receptors or soluble antigens. For targets that are cell-surface receptors, a TP is cleared after the TP–antigen complex is internalized and degraded in the lysosomes of target cells or when the TP-opsonized cell engages in immune effector function, which triggers apoptosis of the target cells by complement-dependent cytotoxicity and antibody-dependent cellular cytotoxicity followed by degradation of the TP. For targets that are soluble antigens, a TP is cleared after the TP–antigen complex is eliminated via phagocytosis and proteolysis by endothelial cells and MPS. TMDD is typically a high affinity, low capacity and saturable process. When TMDD plays an important role in TP clearance, the PK parameters of the TP is concentration and dose dependent and may show a time-dependent decrease of clearance if receptor capacity is decreased after repeated treatment. For example, efalizumab9 and panitumumab10 show higher clearance at low concentrations and doses in clinical applications. The clearances of gemtuzumab and rituximab decrease after the second dose compared to the first dose, which may result from the decrease of target-mediated clearance after a reduction in target tumor cell number after the first dose of treatment.11 For most TPs with TMDD involvement, the TMDD pathway is usually more dominant at low doses and low concentrations of the TPs when this pathway is not saturated. At therapeutic doses of these TPs, the therapeutic protein is often in great excess compared to the expression level of the respective target antigen available for binding under equilibrium conditions; consequently, the nonspecific clearance pathways play a dominant role. For these TPs at their prescribing doses (e.g., pertuzumab,12 bevacizumab,13 and trastuzumab14), changes of target antigen levels generally have a minimal impact on their clearance, and their PK parameters are concentration and dose independent.

The ability of TPs to elicit humoral responses, i.e., immunogenicity, can often modulate the clearance of TPs. The humoral response leads to the formation of antitherapeutic antibodies (ATAs), which may form immunocomplexes with TPs and consequently affect the clearance rates by affecting the binding of a TP to its target or affecting the nonspecific clearance pathways. For example, accelerated clearance of infliximab and adalimumab has been reported after development of ATA in rheumatoid arthritis (RA) patients.15, 16

2.2.3 Excretion of Therapeutic Proteins

Excretion mechanisms for TPs also differ from those for SMDs. Renal clearance is generally negligible when the molecular size of a TP exceeds the cutoff size for renal filtration of approximately 45 kDa.17 Tubular secretion does not occur to any significant extent for large TPs. The peptides resulting from TP catabolism may be partially reabsorbed in the proximal or distal tubule of the nephron or are further catabolized in kidney. Biliary excretion of TPs has been reported for only some fragment peptides and proteins such as immunoglobulin A and octreotide,6, 18 which are subsequently degraded in the gastrointestinal tract.

2.2.4 Pharmacokinetic Properties of Antibody–Drug Conjugates

ADCs, as an emerging class of TPs, have the PK properties of both mAbs and SMDs. ADCs are composed of a potent cytotoxic agent conjugated to a mAb via various types of linkers.19, 20 ADCs bind to their target receptors on the surface of tumor cells. The complexes are internalized and degraded and subsequently release the cytotoxic agents to kill tumor cells. Usually the PK properties of multiple analytes, such as the conjugate and the unconjugated cytotoxic agent, are assessed after administration of an antibody–drug conjugate.

To date all ADCs are administered intravenously.20 The distribution of ADCs is similar to their unconjugated mAbs. For example, in a preclinical in vivo study, it was found that trastuzumab emtansine (T-DM1), an ADC for the treatment of human epidermal growth receptor 2 (HER2) positive solid tumors, had similar tissue distribution to that of trastuzumab, the mAb component of T-DM1, indicating that conjugation does not impact the distribution of trastuzumab.21 ADCs are catabolized by similar pathways as mAbs, including nonspecific proteolytic degradation and TMDD pathways. Immunogenicity may also play a role in ADC clearance.

In addition, the processes of linker chemistry-determined deconjugation in plasma and tissue are also involved in the catabolism and clearance of ADCs. The formation rate of the small molecule cytotoxic component by catabolism of the ADC is usually much slower than the elimination clearance of the small molecule cytotoxic component itself, resulting in formation rate-limited pharmacokinetics. Upon formation, these unconjugated cytotoxic molecules undergo typical clearance pathways of SMDs, such as hepatic metabolism and renal and biliary excretion, as mediated by CYP, non-CYP enzymes, and transporters.21, 22 The low dose of the SMD component of an ADC and relatively slow formation rate combined with a relatively fast elimination rate of the unconjugated SMD molecules may explain the observed relatively low systemic exposure of the unconjugated cytotoxic agent. For example, the average maximal concentration of the derivative of maytansine (DM1) is ~5 ng/mL after the administration of 3.6 mg/kg of T-DM1 every 3 weeks.22 The average maximal free monomethyl auristatin E (MMAE) concentrations are 5–7 ng/mL after the every-3-week administration of 1.8–2.7 mg/kg of brentuximab vedotin,23 a MMAE-containing ADC.24

2.3 Major Mechanisms of Therapeutic Protein–Drug Interactions

We are categorizing DIs as either PK based or PD based. PK-based DIs are those resulting from direct competition, inhibition, or induction of drug ADME mechanisms without involvement of the therapeutic targets. PD-based DIs are those resulting from modulation of the systems or target biology via the PD effects of drugs in combination. Both PK- and PD-based DIs may result in relevant changes in exposure and lead to a potential impact on safety and efficacy outcomes, especially for drugs with a NTR. PD-based DIs may also cause undesirable toxicity without an impact on exposure. Unlike SMDs, which are often susceptible to PK-based DIs due to an alteration in CYP and transporter-mediated ADME processes by drug combinations,3,25 TP-DIs are often mechanistically different.

2.3.1 Impact of Pharmacokinetic-Based Therapeutic Protein–Drug Interactions on the Exposures of Therapeutic Proteins and Small Molecule Drugs

PK-based TP-DIs are not common because TPs and SMDs have distinct PK properties. The nonspecific clearance pathways for TPs are usually unsaturable at therapeutic concentrations. Likewise, these pathways are unlikely to be saturated by the combination of two TPs. For example, the clearances of trastuzumab and bevacizumab are dominated by nonspecific pathways at their clinically efficacious doses. No alteration in PK properties is observed when they are given in combination.26

When a TP is combined with a SMD, there is usually no direct overlap and competition in the metabolism and clearance pathways, thus PK-based DIs are unlikely. For example, chemotherapeutic agents such as irinotecan, 5-fluorouracil, and platinum-based therapy (i.e., cisplatin, carboplatin) do not affect the PK properties of cetuximab in cancer patients.27–29 Similarly, no PK-based DIs are observed between bevacizumab and any of the following agents: capecitabine, cisplatin, 5-fluorouracil, irinotecan, oxaliplatin, or paclitaxel.30 A dedicated study was conducted to evaluate potential TP-DIs for the combination of bevacizumab and irinotecan (as part of the FOLFIRI regimen containing irinotecan, fluorouracil, and leucovorin). This study demonstrated that the 90% confidence interval of geometric mean ratios for exposure of irinotecan and SN-38 (the active metabolite of irinotecan) in the absence of versus in the presence of bevacizumab were both within the prespecified no effect boundaries, indicating no clinically relevant TP-DIs for this combination.31 Additional examples of no TP-DIs for combinations of anticancer mAbs with chemotherapeutic and antineoplastic SMDs have been reviewed in recent publications.2,30,32–35

PK-based TP-DIs involving ADCs are theoretically possible because the cytotoxic component of the ADC, once deconjugated, may elicit PK-based DIs when the ADC is combined with other SMDs. The cytotoxic agent, which is often a CYP substrate, is likely a victim of DIs when combined with SMDs that are CYP inhibitors or inducers. However, the cytotoxic agent has a relatively low systemic exposure. Thus it is not expected to have any impact on CYP and transporter activities in clinical settings and is unlikely to be a perpetrator. Data for ADC-related DIs are limited to assessments for T-DM1 and brentuximab vedotin (Adcetris). When T-DM1 is given in combination with taxanes (paclitaxel or docetaxel), the PK properties of taxanes and DM1 remain unchanged because taxanes and DM1 are not potent CYP inhibitors or inducers at clinically relevant concentrations.36, 37 A dedicated study of brentuximab vedotin found that it does not affect the PK parameters of midazolam, a CYP3A4 substrate. In the same study, the unconjugated MMAE exposure increased ~34% when brentuximab vedotin was combined with ketoconazole (a potent CYP3A4 inhibitor) and decreased ~46% when brentuximab vedotin was combined with rifampin (a potent CYP3A4 inducer). Therefore it is recommended that patients who are receiving strong CYP3A4 inhibitors concomitantly with brentuximab vedotin should be closely monitored for MMAE-related adverse reactions. These results are expected because MMAE is a substrate of CYP3A4 but not a CYP inhibitor or inducer at clinically relevant concentrations.23

2.3.2 Impact of Pharmacodynamic-Based Therapeutic Protein–Drug Interactions on the Exposure of Therapeutic Proteins

Distinct from the less common cases of PK-based TP-DIs, there are several plausible mechanisms for PD-based TP-DIs that change the exposure of TPs. As shown in Figure 2-2, interaction between the biological systems or target biology with a TP may affect the TP's exposure through immunogenicity-mediated clearance or target-mediated clearance pathways. The TPs or SMDs given in combination may modulate these clearance pathways by their PD effect, leading to DIs. For example, immunosuppressants such as methotrexate (MTX), mycophenolate mofetil, and azathioprine increase the exposures of infliximab, adalimumab, and basiliximab, possibly due to the effect of the immunosuppressants on decreasing the immunogenicity rate of these mAbs when they are given in combination. In another case, triple immunosuppressive agents may decrease target (CD11a+ T-cells) level, subsequently decreasing target-mediated clearance of efalizumab and thus increasing its exposure.

Figure 2-2 Theoretical mechanisms of PD-based TP-DIs that change exposure of the TP. SMD: small molecule drug; TP: therapeutic protein.

2.3.2.1 Pharmacodynamic-Based Therapeutic Protein–Drug Interactions Owing to Changes in Immunogenicity-Mediated Clearance of Therapeutic Proteins

Some immunosuppressive drugs may modulate the humoral immune response and decrease the immunogenicity of a TP, thus modulating its clearance. This is possible only for TPs that have a relatively high immunogenicity rate and when the clearances of TPs are impacted by immunogenicity. Examples include infliximab, adalimumab, and basiliximab, as listed in Table 2-1. Infliximab and adalimumab are both mAbs antagonizing tumor necrosis factor α (TNF-α)44 and are often given in combination with immunosuppressive agents such as MTX for the treatment of autoimmune diseases and prevention of rejection after organ transplantation. It was found that patients with positive ATA responses needed higher doses of infliximab to maintain the therapeutic concentration.16 Similarly, serum trough adalimumab concentrations were dramatically lower in patients who developed antiadalimumab antibodies.15 Concomitant dosing of MTX was associated with a decreased incidence of ATA formation against infliximab38,45 and adalimumab.15,45 As a result, when given in combination with MTX, the clearance of infliximab and adalimumab decreased and their serum concentrations increased.15,38,40,44,45

Table 2-1 Examples and Clinical Relevance of PD-Based TP-DIs That Have an Impact on PK Properties of TPs.

Basiliximab is a high-affinity chimeric IgG1-based mAb that binds to and blocks the interleukin 2 receptor α-chain (IL-2Rα) on the surface of activated T-lymphocytes and competitively inhibits IL-2-mediated activation of lymphocytes, a critical pathway in the cellular immune response involved in allograft rejection.41 Basiliximab is often given in combination with triple immunosuppressive agents, including cyclosporine (CsA), corticosteroids, and either azathioprine or mycophenolate mofetil to prevent renal allograft rejection.42 A cross-study comparison of basiliximab PK parameters in de novo renal allograft recipients showed that basiliximab clearance was 29 ± 14 mL/h when coadministered with azathioprine and 18 ± 8 mL/h with mycophenolate mofetil, and both were significantly lower compared to a clearance of 37 ± 15 mL/h from an earlier study of basiliximab with dual immunosuppressive agents of CsA and corticosteroids only.

Although the interstudy variability could potentially be a caveat of this observation, a plausible reason is that immunosuppressive agents such as azathioprine and mycophenolate mofetil may inhibit humoral immune responses and consequently decrease basiliximab clearance. Mycophenolate mofetil may have a stronger inhibition effect than azathioprine. The immunogenicity data in these studies further suggest that if the reduced clearance of basiliximab in the presence of azathioprine or mycophenolate mofetil is due to an inhibition of humoral immune responses, it appears that these responses are directed to a portion of basiliximab other than the IL-2Rα binding portion of basiliximab. Nonetheless, the PD effect of basiliximab as indicated by the range of the durations of IL-2Rα saturation when basiliximab was combined with triple immunosuppressive agents did not extend outside the range of the durations after the treatment of basiliximab in combination with dual immunosuppressive agents. Hence, regardless of the observed clearance difference, no dosing adjustment was necessary.42

2.3.2.2 Pharmacodynamic-Based Therapeutic Protein–Drug Interactions Owing to Changes in Target-Mediated Drug Disposition of Therapeutic Proteins

Combinations that affect the TMDD pathway of some TPs may result in PD-based TP-DIs, when TMDD plays an important role in the clearance processes of these TPs at their therapeutic doses. Based on a theoretical simulation of a typical mAb with strong TMDD (i.e., TMDD accounts for >50% of total clearance in the therapeutically relevant dose range), an increase of baseline receptor expression level (e.g., 3-fold or 10-fold of baseline) leads to a significant increase of the observed total clearance; a decrease of baseline receptor expression level (e.g., 10%, 30% of baseline) leads to a significant decrease of the observed total clearance. The magnitudes of changes are higher at low dose levels, when TMDD accounts for higher percentage of total clearance (Figure 2-3a).

Figure 2-3 Effect of receptor expression level on the dose-total CL relationship for a typical mAb. Observed total CL = dose/area under the concentration–time profile. Simulation is performed using a standard TMDD model after a single dose of a typical mAb 46,47. The baseline free receptor expression levels was altered to 3-fold, 10-fold, 10%, or 30% of baseline, and the observed total CL values were plotted versus dose levels. (a) TMDD accounts for >50% of total clearance in the therapeutically relevant dose range. (b) TMDD accounts for ≤1% of total clearance in the therapeutically relevant dose range. CL: clearance.

A clinical example is efalizumab in combination with immunosuppressive agents for preventing organ rejection after transplantation, is listed in Table 2-1. At clinical doses, TMDD of efalizumab after binding with CD11a receptors expressed on T-cells plays an important role in efalizumab clearance. Consequently, clearance of efalizumab is concentration and dose dependent and governed by CD11a expression level.48 When efalizumab was given by subcutaneous administration in combination with oral administration of triple immunosuppressive agents (CsA, sirolimus, and prednisone) in transplant recipients, the clearance of efalizumab was 50% lower than the clearance observed in psoriatic patients. One of the reasons might be related to the reduction of CD11a+ T-cells in circulation by immunosuppressive agents given in combination in transplant recipients, resulting in reduced TMDD of efalizumab.43,48

However, in most clinical situations in which TPs are given at efficacious doses at which the TMDD pathway does not play a major role in their clearance, the nonspecific proteolytic degradation pathways, which are usually of high capacity and unsaturable, dominate the clearance of these TPs. They exhibit linear PK characteristics with relatively constant total clearance at therapeutic doses. For these TPs, the possibility of PD-based TP-DIs as a result of a target receptor expression level change by the combination drug is usually low. Based on a theoretical simulation of a typical mAb with negligible TMDD (i.e., TMDD accounts for ≤1% of total clearance in the therapeutically relevant dose range), an increase or a decrease of baseline receptor expression level (e.g., 3-fold, 10-fold, 10%, or 30% of baseline) does not have a significant impact on the observed total clearance (Figure 2-3b).

A clinical example is the combination of T-DM1 with pertuzumab, an ADC with a mAb.49 Both the trastuzumab component of T-DM1 and pertuzumab bind to HER2 receptors. Therefore a potential for PD-based DIs might exist for this combination. However, pertuzumab and T-DM1 bind to distinct epitopes of HER2 simultaneously without steric hindrance.50,51 Furthermore, both T-DM1 and pertuzumab have linear PK characteristics at their clinically efficacious dose ranges when given in combination,12,22 indicating that HER2-mediated clearance is unlikely to play a major role in their clearance in these dose ranges. Hence a potential impact on HER2 biology by T-DM1 is unlikely to affect the PK characteristics of pertuzumab, and vice versa. Cross-study comparisons of PK parameters of T-DM1 and pertuzumab in the combination study and the historical single-agent studies demonstrate that there is no DI observed for this combination.49

2.3.3 Impact of Pharmacodynamic-Based Therapeutic Protein–Drug Interactions on the Exposure of Small Molecule Drugs Owing to Therapeutic Protein–Cytokine–Cytochrome P450 Modulation Effects

Because TPs and SMDs do not share common ADME mechanisms, TPs are not predicted to affect the hepatic, renal, or biliary elimination of SMDs directly. However, some immunomodulatory TPs may exert an indirect effect on the hepatic clearance pathways of SMDs through the TP-cytokine-CYP modulation effect, resulting in PD-based TP-DIs. As summarized in Figure 2-4, infection and inflammatory diseases may cause an elevation of systemic cytokine level, such as interferons (IFNs), interleukins, and tumor necrosis factors, which in turn inhibit the CYP activity and consequently increase the exposure of SMDs that are CYP substrates.52 Some immunomodulatory TPs, such as exogenous IFNα, muromanab-CD3, and basiliximab may upregulate certain cytokine levels, inhibit CYP activities, and increase the exposure of CYP substrates (e.g., theophylline, cyclophosphamide, CsA, and tacrolimus). In contrast, some immunomodulatory TPs such as tocilizumab may downregulate the pathologically elevated cytokine levels in the inflammation disease state, normalize CYP activities to a higher level similar to healthy individuals, and decrease the exposure of CYP substrates (e.g., omeprazole, simvastatin). In addition to these clinical observations, there is strong in vitro evidence suggesting the differential effects of cytokines on various CYP enzymes.1 The TP-DIs induced by the TP-cytokine-CYP modulation effect may be clinically relevant for SMDs with a NTR. The mechanisms and various scenarios of this type of TP-DI are reviewed in detail in this section.

Figure 2-4 Theoretical mechanisms of PD-based TP-DIs that have an impact on the PK properties of SMDs through TP-cytokine-CYP modulation effect. CYP: cytochrome P450; IFN: interferon; IL: interleukin; SMD: small molecule drug; TNF: tumor necrosis factor; TP: therapeutic protein.

Pathophysiological changes in patients and animal models in infectious or inflammatory conditions are often associated with decreased production and activities of hepatic and intestinal CYP enzymes, which are important for the first-pass extraction and metabolism of most SMDs. Based on in vitro and in vivo data, proinflammatory cytokines such as ILs (IL-1β, IL-6), TNFs (TNFα), and IFNs (IFNγ, IFNα-2b) may be potentially important mediators of this CYP modulation effect. These cytokines may induce a decrease in transcription factor activity for CYP enzyme expression or a decrease in CYP enzyme stability. As a result, CYP activities decrease in patients with infectious and inflammatory disease with elevated proinflammatory cytokine levels. The patients usually have an increased exposure of the SMDs metabolized by CYP enzymes because of decreased clearance and/or increased bioavailability compared with healthy individuals.52

Direct administration of exogenous cytokines may produce an immunopathological state similar to infection or inflammation, downregulate CYP activities, and potentially increase the exposure of some SMDs given in combination.52 For instance, a clinical study using the “Pittsburgh cocktail” approach in high-risk melanoma patients found that the high-dose IFNα-2b53 therapy differentially impairs CYP-mediated metabolism, causing no effect on some enzymes (e.g., CYP2E1) but substantial effects on others (e.g., CYP1A2: median 60% activity decrease; CYP2C19, median 40% activity decrease).54 IFNα-2b has also been used in combination with theophylline for the treatment of various diseases (e.g., chronic hepatitis B). The clearance of theophylline, a CYP1A2 substrate drug with a NTR, decreased after IFNα-2b treatment, resulting in a 100% increase in serum theophylline levels.53 In a clinical study in multiple myeloma patients, it was found that the administration of IFNα before cyclophosphamide (CP) caused a decrease in CP clearance to 63%, a 137% longer half-life and a 137% higher peak plasma concentration compared to the results obtained when IFNα was administered 24 h after CP. Consequently, the active metabolite 4-hydroxycyclophosphamide, which is formed by CYP-mediated metabolism from the prodrug CP, had 45% lower exposure when IFNα was given before CP compared with that observed when IFNα was administered 24 h after CP. As a result, a significant decrease in leukocyte count was observed when IFNα was given 24 h after CP.55 These examples (some of them are listed in Table 2-2) demonstrate the effect of exogenous cytokines on the activity of CYPs in the clinical setting, which implies that endogenous cytokine levels may play an important role as modulators of hepatic metabolism.

Table 2-2 Examples and Clinical Relevance of TP-DIs Affecting the PK Properties of SMDs Caused by the TP-Cytokine-CYP Modulation Effect.

Similarly, some mAbs may increase cytokine levels and activities by their PD effects and may lead to similar effects as the treatment by exogenous cytokines. As listed in Table 2-2, both muromonab-CD3 and basiliximab are mAbs for preventing rejection after organ transplantation.41,56 Muromonab-CD3 is a murine IgG2a mAb targeting CD3 on T-cells to prevent acute organ rejection. Immediately after administration of muromonab-CD3, the activation of T-cells and release of numerous cytokines are observed followed by the blocking of all known T-cell functions. The initial T-cell activation and cytokine release may downregulate CYP enzyme activity and expression. CsA is a CYP3A4 substrate that may be combined with muromonab-CD3 in transplantation patients. It was found that muromonab-CD3 treatment caused a significant rise in CsA trough levels at day 5 post–renal transplantation compared to the group of patients receiving antilymphocyte globulin immunoprophylaxis therapy. As a result, CsA doses were adjusted based on trough levels obtained on day 5 post-transplantation.57

Clinically significant DIs were also reported for basiliximab in combination with CsA or tacrolimus. Both CsA and tacrolimus are metabolized by CYP3A4 and require therapeutic drug monitoring owing to their NTRs. In children who have undergone kidney transplant, basiliximab combination led to a higher trough concentration of CsA compared to the patients in the control group who were not on basiliximab. Consequently, it was recommended that the initial CsA dose should be limited to 400 mg/m2 if used in combination with basiliximab in children with kidney transplants.58

Similarly, a 63% increase in tacrolimus trough levels was reported in basiliximab-treated adult patients on day 3 postrenal transplantation, when compared to the control group receiving antithymocyte globulin induction therapy without basiliximab treatment. As a result, the tacrolimus dose was decreased until the plasma concentration returned to a desirable range.59 It was postulated that both of these clinically significant TP-DIs may be related to the effect of basiliximab on CYP activities via cytokine modulation. Interaction between IL-2 and its respective IL-2R on intestinal epithelial cells and hepatocytes may decrease the expression or activity of CYP3A4.62 Binding of basiliximab to the IL-2R on activated T-cells may allow circulating IL-2 binding to IL-2Rs on hepatic and intestinal cells, resulting in a downregulation of CYP3A4 enzyme activity.

On the other hand, the treatment of some TPs results in a downregulation and normalization of elevated cytokine levels in the disease state through cytokine or cytokine receptor antagonism. Subsequent inhibition of cytokine-mediated downstream signaling transduction may restore the previously down-regulated CYP enzyme activity to a normal level and consequently normalize the exposure of SMDs that are CYP substrates.52 As listed in Table 2-2, tocilizumab, an anti-IL-6R mAb for RA treatment, may have DIs based on this mechanism. In RA patients, IL-6 serum concentrations are elevated up to 50–60 pg/mL as compared to those in healthy subjects (~5 pg/mL). Tocilizumab binds specifically to both soluble and membrane-bound IL-6 receptors (sIL-6R and mIL-6R) and inhibits IL-6-mediated signaling through blocking these receptors. This ameliorates the inflammation status.

In vitro data suggest that IL-6 inhibits CYP activity, and tocilizumab prevents CYP3A4 suppression by IL-6 in human hepatocytes, indicating that tocilizumab may reverse IL-6-induced suppression of CYP activities in the disease state.60 This was demonstrated in a clinical study in RA patients in which the tocilizumab combination resulted in up to a 57% decrease in the exposure of simvastatin, a CYP3A4 probe substrate, in comparison to simvastatin alone in the same individuals. In another study, tocilizumab resulted in a 12–28% decrease in exposure of omeprazole, a CYP3A4 and CYP2C19 substrate, by comparing the exposure in the same RA patients before and after tocilizumab dosing. These observations may have clinical relevance for patients who take tocilizumab with NTR drugs for which the doses need to be individually adjusted.60,61 Some other approved cytokine modulators, such as canakinumab (anti-IL-1β mAb),63 golimumab (anti-TNFα mAb),64 rilonacept (anti-IL-1R fusion protein),65 and ustekinumab (anti-IL-12/23 mAb)66 may also potentially normalize CYP enzymes in patients. Although no formal DI studies have been conducted for these TPs, the potential risk of TP-DIs and the need to monitor the therapeutic effects or concentrations of drugs with a NTR when given in combination with these TPs are addressed in their package inserts.

Transporters play important roles for the ADME process of SMDs. Inflammatory disease states may also affect several important drug transporters, such as p-glycoproteins, multidrug resistance-associated proteins, and organic anion-transporting peptides.67 Theoretically, the risk of transporter-mediated DIs for the combinations of TPs and SMDs exists. However, to date no clinically relevant TP-DIs caused by this mechanism have been reported.

2.3.4 Pharmacodynamic-Based Therapeutic Protein–Drug Interactions That Lead to Toxicity without Affecting Exposures

The major objective for using combination therapy is to improve efficacy by desirable PD interactions without increasing safety concerns. As discussed earlier, PD-based TP-DIs may result in exposure changes in either the TP or the SMD given in combination, which may have clinical relevance for drugs with a NTR. In addition, PD-based TP-DIs can also lead to toxicity without affecting the exposure of the drugs in combination. PD-based TP-DIs that result in undesirable toxicity are usually documented in the labels of these TPs to guide the prescribers to avoid certain combinations. An example is the combination of anakinra and etanercept. Anakinra is a recombinant IL-1R antagonist protein that competitively inhibits the binding of IL-1 to IL-1Rs.68 Anakinra has a molecular weight of 17.3 kDa. Etanercept is a dimeric fusion protein consisting of the extracellular ligand-binding portion of the human 75 kDa tumor necrosis factor receptor linked to the Fc portion of human IgG1.69 Both TPs were approved for RA treatment. Because anakinra is eliminated by renal excretion and etanercept is eliminated by proteolysis with the salvage pathway of FcRn-mediated recycling, there are no overlapping clearance pathways for them. In a phase III study in RA patients using concurrent therapy of anakinra with etanercept, the PK properties of both agents were unchanged by the combination. However, there was no significant improvement in efficacy for the combination therapy, and the risk of serious infection and neutropenia increased when compared to etanercept alone. PD-based DIs may potentially cause patients to be highly immunosuppressed and prone to infection, and an additive neutropenia effect also likely results from the use of this combination.69,70

Other examples are also recently reviewed.30 For example, filgrastim, a human granulocyte colony–stimulating factor, during a period of 24 h before through 24 h after cytotoxic chemotherapy should be avoided because of the potential sensitivity of rapidly dividing myeloid cells to cytotoxic chemotherapy.71 Aldesleukin,72 a human recombinant IL-2 protein, may enhance the toxicity of drugs of nephrotoxic, myelotoxic, cardiotoxic, or hepatotoxic effects when given in combination. IFNα-2b53 may cause increased myelotoxicity of myelosuppressive agents, such as zidovudine, when given in combination.

2.4 Strategies to Assess the Risk of Therapeutic Protein–Drug Interactions in Clinical Development of Therapeutic Proteins

2.4.1 Challenges of Assessing Therapeutic Protein–Drug Interactions in Clinical Development

It is critical to understand the theoretical TP-DI mechanisms and build an individualized and iterative strategy to ensure the safe and effective use of TPs for combination treatment. The FDA guidance document for DI studies,3 the European Medicines Agency (EMA) guideline on the clinical investigation of the PK characteristics of TPs,73 and several recent review papers from the FDA highlighted the importance of evaluating TP-DIs.1,2

There are multiple challenges in assessing TP-DIs in clinical development of TPs, owing to the distinct PK and PD properties of TPs. Considering potential toxicity, most clinical trials involving TPs need to be conducted in the patient population instead of in healthy subjects. The PK characteristics of TPs may differ between patients and healthy subjects because of target expression levels. For immunomodulatory TPs, the CYP activity normalization by TP-cytokine-CYP interactions occurs only in patients with infection or inflammatory diseases. However, formal TP-DI studies in patients with life-threatening diseases (e.g., cancer, transplantation, severe autoimmune diseases) are rarely conducted owing to logistical constrains. Cross-over studies are difficult because of the long washout period required for TPs with long half-lives; a parallel design of two arms in a randomized study or a cross-over design for TP-DI assessment is not ethical if one of the drugs lacks efficacy when it is used alone. Concurrent medications also confound the DI evaluation of studied combinations2,30 In addition to these challenges in clinical evaluation, in vitro and preclinical DI assessment strategies for TPs are very different from SMDs and are still evolving owing to the inherent differences in metabolic pathways between TPs and SMDs.30 Therefore, more creative and less disruptive strategies are needed to assess the risk of TP-DIs in clinical development.

2.4.2 Question-Based Strategy of Therapeutic Protein–Drug Interaction Assessment During Clinical Development

As summarized in Figure 2-5, a question-based TP-DI risk assessment strategy is recommended.30 This strategy recommends both theoretical TP-DI risk assessments based on PK and PD properties of the drugs in combination (as described earlier in this chapter) and PK evaluation in phase Ib/II studies to support the theoretical assessment. If a clinically relevant TP-DI signal is detected, further assessment and confirmation might be considered in phase II/III studies. All TP-DI relevant information–including the theoretical risk of TP-DIs based on the PK and PD properties of the drugs in combination, the PK and ADME data collected at various stages in the development of combination therapy, the potential impact of TP-DIs on the safety and/or efficacy outcomes, and whether the impact warrants clinical dose adjustment and/or therapeutic drug monitoring—is documented in the package insert to provide physicians with data that allow an informed decision about whether to adjust a drug dose when administered in combination or whether to avoid a particular concomitant medication. The TP-DI risk assessments need an individualized strategy for each molecule and are iterative based on emerging data.

Figure 2-5 Question-based TP-DI risk assessment strategy during TP development. CYP: cytochrome P450; NTR: narrow therapeutic range; PK: pharmacokinetic; PMR: postmarketing requirement; SMD: small molecule drug; TP: therapeutic protein; TP-DI: therapeutic protein–drug interaction.

2.4.2.1 Theoretical Risk Assessments for Therapeutic Protein–Drug Interactions