Quantitative Decisions in Drug Development (Springer Series in Pharmaceutical Statistics) 1st Edition

Quantitative Decisions in Drug Development (Springer Series in Pharmaceutical Statistics) 1st EditionThis book offers a high-level treatise of evidence-based decisions in drug development. Because of the inseparable relationship between designs and decisions, a good portion of this book is devoted to the design of clinical trials. The book begins with an overview of product development and regulatory approval pathways.

Get ebook : $10.00 


It then discusses how to incorporate prior knowledge into study design and decision making at different stages of drug development.

The latter include selecting appropriate metrics to formulate decisions criteria, determining go/no-go decisions for progressing a drug candidate to the next stage and predicting the effectiveness of a product. Lastly, it points out common mistakes made by drug developers under the current drug-development paradigm.The book offers useful insights to statisticians, clinicians, regulatory affairs managers and decision-makers in the pharmaceutical industry who have a basic understanding of the drug-development process and the clinical trials conducted to support drug-marketing authorization.

The authors provide software codes for select analytical approaches discussed in the book. The book includes enough technical details to allow statisticians to replicate the quantitative illustrations so that they can generate information to facilitate decision-making themselves.

Contents
1 Clinical Testing of a New Drug . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Clinical Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2.1 Phase 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.2 Phase 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.2.3 Phase 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2.4 Phase 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.3 Regulatory Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.3.1 Accelerated Approval . . . . . . . . . . . . . . . . . . . . . . . 9 1.3.2 Breakthrough Therapy . . . . . . . . . . . . . . . . . . . . . . . 10 1.3.3 Priority Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.3.4 Fast Track . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.3.5 Orphan Drug . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.3.6 Drug Approval in the European Union (EU) . . . . . . . 12 1.4 Innovative Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.4.1 Adaptive Design . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.4.2 Master Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2 A Frequentist Decision-Making Framework . . . . . . . . . . . . . . . . . 19 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.2 Statistical Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.3 Testing a Statistical Hypothesis . . . . . . . . . . . . . . . . . . . . . . . 20 2.4 Decision-Making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2.5 Losses and Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.6 The Power Function of a Test . . . . . . . . . . . . . . . . . . . . . . . . 24 2.7 Determining a Sample Size for an Experiment . . . . . . . . . . . . 25 2.8 Multistage Tests and the Use of a No-Decision Region . . . . . . 28 2.9 One-Sided Versus Two-Sided Tests . . . . . . . . . . . . . . . . . . . . 28

2.10 P-Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.11 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3 Characteristics of a Diagnostic Test . . . . . . . . . . . . . . . . . . . . . . . 33 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.2 Sensitivity and Specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.3 Positive and Negative Predictive Value . . . . . . . . . . . . . . . . . 35 3.4 Value of a Follow-Up Test . . . . . . . . . . . . . . . . . . . . . . . . . . 37 3.5 When Two Tests Are Being Done Simultaneously . . . . . . . . . 38 3.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4 The Parallel Between Clinical Trials and Diagnostic Tests . . . . . . 41 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 4.2 Why Replication Is Necessary . . . . . . . . . . . . . . . . . . . . . . . . 42 4.3 Why Replication Is Hard . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 4.3.1 Conditional Replication Probability . . . . . . . . . . . . . 44 4.3.2 Average Replication Probability . . . . . . . . . . . . . . . . 46 4.3.3 When the Second Trial Has a Different Sample Size. . . 48 4.4 Differentiate Between Statistical Power and the Probability of a Successful Trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 4.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5 Incorporating Information from Completed Trials in Future Trial Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 5.2 The Bayesian Approach to Inference . . . . . . . . . . . . . . . . . . . 54 5.3 Bayesian Average Power and Assurance . . . . . . . . . . . . . . . . 55 5.4 Closed-Form Expressions for Assurance and the Simulation Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 5.5 PPV and NPV for a Planned Trial . . . . . . . . . . . . . . . . . . . . . 58 5.6 Forming a Prior Distribution from a Number of Similar Previous Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 5.7 Standard Prior Distributions . . . . . . . . . . . . . . . . . . . . . . . . . 61 5.8 Elicitation of a Prior Distribution from Experts . . . . . . . . . . . 62 5.9 Prior Distributions from PK/PD Modeling and Model-Based Meta-Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 5.10 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
6 Choosing Metrics Appropriate for Different Stages of Drug Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 6.2 Metrics for Proof-of-Concept Studies . . . . . . . . . . . . . . . . . . . 70

6.3 Metrics for Dose-Ranging Studies . . . . . . . . . . . . . . . . . . . . . 71 6.3.1 Estimating a Dose-Response Relationship . . . . . . . . . 72 6.3.2 Testing for a Positive Dose-Response Relationship . . . 75 6.3.3 Calculating the Metrics . . . . . . . . . . . . . . . . . . . . . . 77 6.4 Metrics for Confirmatory Studies . . . . . . . . . . . . . . . . . . . . . . 78 6.5 Other Types of Success Probabilities . . . . . . . . . . . . . . . . . . . 79 6.5.1 Probability of Program Success (POPS) . . . . . . . . . . 79 6.5.2 Probability of Compound Success (POCS) . . . . . . . . 81 6.6 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7 Designing Proof-of-Concept Trials with Desired Characteristics . . . 85 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 7.2 Five Approaches to Decision-Making . . . . . . . . . . . . . . . . . . 86 7.2.1 The Traditional Hypothesis-Testing Approach . . . . . . 86 7.2.2 The ESoE Approach . . . . . . . . . . . . . . . . . . . . . . . . 87 7.2.3 The LPDAT Approach . . . . . . . . . . . . . . . . . . . . . . . 87 7.2.4 The TV Approach . . . . . . . . . . . . . . . . . . . . . . . . . . 89 7.2.5 The TVMCID Approach . . . . . . . . . . . . . . . . . . . . . . . 89 7.2.6 A Comparison of the Five Approaches . . . . . . . . . . . 90 7.3 Criteria for Determining Sample Size . . . . . . . . . . . . . . . . . . 90 7.3.1 The Traditional Hypothesis-Testing Approach . . . . . . 91 7.3.2 The ESoE Approach . . . . . . . . . . . . . . . . . . . . . . . . 91 7.3.3 The LPDAT Approach . . . . . . . . . . . . . . . . . . . . . . . 92 7.3.4 The TV and TVMCID Approaches . . . . . . . . . . . . . . . 92 7.4 Metrics for a Proof-of-Concept Study . . . . . . . . . . . . . . . . . . 93 7.5 Prior Distributions for the Treatment Effect . . . . . . . . . . . . . . 93 7.6 An Example of Evaluating POC Trial Designs for Desired Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 7.6.1 Conditional Evaluation of the Trial Designs . . . . . . . 95 7.6.2 Unconditional Evaluation of the Trial Designs . . . . . . 98 7.7 Sensitivity Analyses for the Choice of Prior Distribution . . . . 102 7.8 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
8 Designing Dose-Response Studies with Desired Characteristics . . . 105 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 8.2 The Emax Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 8.3 Design of a Dose–Response Study . . . . . . . . . . . . . . . . . . . . . 107 8.4 Metrics for Dose–Ranging Studies . . . . . . . . . . . . . . . . . . . . . 108 8.5 Conditional Evaluation of a Dose–Response Design . . . . . . . . 109 8.6 Unconditional Evaluation of a Dose–Response Design . . . . . . 114 8.6.1 Obtaining a Prior from POC Study Results and a Projection of Compound Potency . . . . . . . . . . . 114 8.6.2 An Example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

8.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
9 Designing Confirmatory Trials with Desired Characteristics . . . . 123 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 9.2 Useful Metrics at the Confirmatory Stage . . . . . . . . . . . . . . . . 124 9.3 Relationship Between Sample Size and Metrics . . . . . . . . . . . 127 9.4 The Impact of Prior Data on POSS . . . . . . . . . . . . . . . . . . . . 129 9.5 Sample Size Consideration Based on POSS . . . . . . . . . . . . . . 130 9.6 Other Applications of the Concept of POSS at the Confirmatory Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 9.6.1 Conditional POSS . . . . . . . . . . . . . . . . . . . . . . . . . . 132 9.6.2 Sample Size Reestimation . . . . . . . . . . . . . . . . . . . . 133 9.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
10 Designing Phase 4 Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 10.2 Network Meta-Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 10.3 Example Evaluation of a Design Using the Results of a Network Meta-Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 10.4 Pediatric Study Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 10.5 Designs to Investigate the Effect of the Drug at a Lower/ Higher Dose or with Different Administration Schedules . . . . 147 10.6 Studies in New Populations . . . . . . . . . . . . . . . . . . . . . . . . . . 147 10.7 Studies to Test a Drug in Combination with Other Drugs . . . . 148 10.8 Studies for New Indications . . . . . . . . . . . . . . . . . . . . . . . . . 148 10.9 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
11 Other Metrics that Have Been Proposed to Optimize Drug Development Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 11.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 11.2 Benefit–Cost Efficiency Score . . . . . . . . . . . . . . . . . . . . . . . . 154 11.3 Product Valuation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 11.3.1 Present Value of Net Revenue . . . . . . . . . . . . . . . . . 159 11.3.2 Present Value of Development Cost . . . . . . . . . . . . . 160 11.3.3 Net Present Value . . . . . . . . . . . . . . . . . . . . . . . . . . 161 11.3.4 Fifth-Year Net Revenue . . . . . . . . . . . . . . . . . . . . . . 161 11.3.5 Phase 2 Study Designs Considered . . . . . . . . . . . . . . 162 11.3.6 Range of Efficacy and Tolerability Considered . . . . . 163 11.3.7 Selecting a Dose to Move to Phase 3 . . . . . . . . . . . . 164 11.3.8 Metrics Used to Evaluate Design Options . . . . . . . . . 164 11.3.9 High-Level Results . . . . . . . . . . . . . . . . . . . . . . . . . 165 11.4 Other Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

11.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
12 Discounting Prior Results to Account for Selection Bias . . . . . . . . 173 12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 12.2 Selection Bias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 12.3 Planning a Phase 3 Trial Using the Result of a Single Phase 2 Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 12.3.1 No Difference Between Phase 2 and 3 Populations and Endpoints or Any Other Factors . . . . . . . . . . . . . 176 12.3.2 Different Phase 2 Endpoint and/or Population Compared to Phase 3 (Other Factors Assumed the Same) . . . . . . 180 12.4 Planning a Phase 3 Trial Using the Results of a Phase 2 POC Study and a Phase 2 Dose-Response Study . . . . . . . . . . 182 12.5 Estimation of the Regression to the Mean Effect Caused by Phase 2 Trial Selection Using a Prior Distribution . . . . . . . . . 184 12.6 An Empirical Assessment of the Amount of Discounting Required for Observed Phase 2 Effects . . . . . . . . . . . . . . . . . 188 12.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
13 Additional Topics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 13.1 Adaptive Designs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 13.2 Joint Analysis of an Efficacy and a Safety Endpoint . . . . . . . . 194 13.3 Use of a Sampling Distribution as a Prior Distribution . . . . . . 196 13.4 Using Prior Information at the Preclinical Stage . . . . . . . . . . . 196 13.5 Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 13.5.1 Preclinical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 13.5.2 Information from Observational Studies . . . . . . . . . . 197 13.5.3 General Principles on Handling Data from Different Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 13.6 Changes Over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 13.7 Further Extensions of the Concept of Success Probability . . . . 201 13.8 Wrapping Up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

Chapter 1 Clinical Testing of a New Drug
Nearly 60 percent of Americans—the highest ever—are taking prescription drugs. Washington Post, November 3, 2015
1.1 Introduction
A research study reports an increase in the overall use of prescription drugs among adults (those 20 years old) between 2011 and 2012 from that between 1999 and 2000 in the United States (USA) (Kantor et al. 2015). In 1999–2000, an estimated 51% of the US adults reported using any prescription drug. The estimated figure for 2011–2012 is 59%. During the same period, the prevalence of polypharmacy (use of 5 prescription drugs) increased from 8.2 to 15%. Many factors contribute to this increase, factors such as better disease prevention and management, lifestyle change, an aging population, and an increase in the percentage of people who are either overweight or obese. The number of new prescription drugs developed and approved for public use every year has also greatly contributed to this increase. Developing a new drug is a high-risk and high-reward enterprise. The high risk is reflected by the low success rate of turning a new molecular entity (NME) into an approved drug. The success rate fluctuated over time and varied across therapeutic areas. For example, the US Food and Drug Administration published the Critical Path Initiative document in 2004 (FDA 2004), in which FDA quoted a “current” success rate around 8% and a historical success rate of 14%. Understandably, the success rate varies substantially across therapeutic areas (DiMasi et al. 2010, 2013). For example, the success rate of drugs for treating common bacterial infections is generally higher than that for drugs treating disorders of the central nervous system. This is in part due to the heavy use of the minimum inhibitory concentration (MIC) to help determine the appropriate dose and schedule for an NME for bacterial infections. For a microorganism studied in vitro, the MIC for an antibacterial agent is the lowest concentration of the agent which prevents detectable growth of the organism in agar or broth media under…..

Tagged:

Leave a Reply

Your email address will not be published. Required fields are marked *