Since the early 2000s, there has been increasing interest within the pharmaceutical industry in the application of Bayesian methods at various stages of the research, development, manufacturing, and health economic evaluation of new health care interventions. In 2010, the first Applied Bayesian Biostatistics conference was held, with the primary objective to stimulate the practical implementation of Bayesian statistics, and to promote the added-value for accelerating the discovery and the delivery of new cures to patients.
This book is a synthesis of the conferences and debates, providing an overview of Bayesian methods applied to nearly all stages of research and development, from early discovery to portfolio management. It highlights the value associated with sharing a vision with the regulatory authorities, academia, and pharmaceutical industry, with a view to setting up a common strategy for the appropriate use of Bayesian statistics for the benefit of patients.
The book covers:
- Theory, methods, applications, and computing
-
- Bayesian biostatistics for clinical innovative designs
-
- Adding value with Real World Evidence
-
- Opportunities for rare, orphan diseases, and pediatric development
-
- Applied Bayesian biostatistics in manufacturing
-
- Decision making and Portfolio management
-
- Regulatory perspective and public health policies
Statisticians and data scientists involved in the research, development, and approval of new cures will be inspired by the possible applications of Bayesian methods covered in the book. The methods, applications, and computational guidance will enable the reader to apply Bayesian methods in their own pharmaceutical research.
Emmanuel Lesaffre is Professor of Biostatistics at KU Leuven, Belgium.
Gianluca Baio is Professor of Statistics and Health Economics at University College London, UK.
Bruno Boulanger is Chief Scientific Officer at PharmaLex, Belgium.
Table of Contents:
Preface
Contributors
List of abbreviations
I Introductory part
Chapter 1: Bayesian Background
Emmanuel Lesaffre and Gianluca Baio
1.1 Introduction
1.2 The frequentist approach to inference
1.3 Bayesian concepts
1.4 More than one parameter
1.5 Choosing the prior distribution
1.6 Determining the posterior distribution numerically
1.7 Hierarchical models and data augmentation
1.8 Model selection and model checking
1.9 Bayesian nonparametric methods
1.10 Bayesian software
1.11 Further reading
Chapter 2: FDA Regulatory Acceptance of Bayesian Statistics
Gregory Campbell
2.1 Introduction
2.2 Medical devices
2.3 Pharmaceutical products
2.4 Differences between devices and drugs
2.5 Some promising opportunities in pharmaceutical drugs
2.6 The future
2.7 Conclusion
Chapter 3: Bayesian Tail Probabilities for Decision Making
Leonhard Held
3.1 Introduction
3.2 Posterior tail probabilities
3.3 Predictive tail probabilities
3.4 Discussion
II Clinical development
Chapter 4: Clinical Development in the Light of Bayesian Statistics
David Ohlssen
4.1 Introduction
4.2 Introduction to drug development
4.3 Quantitative decision making in drug development
4.4 Bayesian thinking
4.5 Applications of Bayesian methods in drug development
4.6 Conclusion
Chapter 5: Prior Elicitation
Nicky Best, Nigel Dallow, and Timothy Montague
5.1 Introduction
5.2 Methods for prior elicitation
5.3 Examples
5.4 Impact and outlook
Chapter 6: Use of Historical Data
Beat Neuenschwander and Heinz Schmidli
6.1 Introduction
6.2 Identifying historical or co-data
6.3 An example: Guillain-Barre syndrome in children
6.4 Methods
6.5 Application: Non-inferiority trials
6.6 Discussion
6.7 Code
Chapter 7: Dose Ranging Studies and Dose Determination
Phil Woodward, Alun Bedding, and David Dejardin
7.1 Introduction
7.2 Dose-response studies
7.3 Dose escalation trials in oncology
7.4 Conclusions
Chapter 8: Bayesian Adaptive Designs in Drug Development
Gary L. Rosner
8.1 Introduction
8.2 Brief history of adaptive designs
8.3 What is an adaptive clinical trial?
8.4 Types of adaptation
8.5 Reasons we might consider adaptive designs
8.6 Example of an adaptive design
8.7 Adaptive enrichment designs
8.8 Some criticisms of adaptive designs
8.9 Summary
Chapter 9: Bayesian Methods for Longitudinal Data with Missingness
Michael J. Daniels and Dandan Xu
9.1 Introduction
9.2 Common frequentist approaches
9.3 Bayesian approaches
9.4 Ignorable and nonignorable missingness
9.5 Posterior inference
9.6 Model selection
9.7 Model checking and assessment
9.8 Practical example: Growth hormone trial
9.9 Wrap-up and open problems
Chapter 10: Survival Analysis and Censored Data
Linda D. Sharples and Nikolaos Demiris
10.1 Introduction
10.2 Review of survival analysis
10.3 Software
10.4 Applications
10.5 Reporting
10.6 Other comments
Chapter 11: Benefit of Bayesian Clustering of Longitudinal Data: Study of Cognitive
Decline for Precision Medicine
Anais Rouanet, Sylvia Richardson, and Brian Tom
11.1 Introduction
11.2 Motivating example
11.3 Nonparametric Bayesian models
11.4 Standard frequentist analysis: Latent class mixed models
11.5 Profile regression analysis
11.6 Conclusion
Chapter 12: Bayesian Frameworks for Rare Disease Clinical Development Programs
Freda Cooner, Forrest Williamson, and Bradley P. Carlin
12.1 Introduction
12.2 Natural history studies
12.3 Long-term safety evaluation with Real-World Data
12.4 Bayesian approaches in rare diseases
12.5 Case study
12.6 Conclusions and future directions
Chapter 13: Bayesian Hierarchical Models for Data Extrapolation and Analysis in
Pediatric Disease Clinical Trials
Cynthia Basu and Bradley P. Carlin
13.1 Introduction
13.2 Classical statistical approaches to data extrapolation
13.3 Current Bayesian approaches
13.4 Practical example
13.5 Outlook
III Post-marketing
Chapter 14: Bayesian Methods for Meta-Analysis
Nicky J Welton, Haley E Jones, and Sofia Dias
14.1 Introduction
14.2 Pairwise meta-analysis
14.3 Network meta-analysis
14.4 Bias modeling in pairwise and network meta-analysis
14.5 Using meta-analysis to inform study design
14.6 Further reading
Chapter 15: Economic Evaluation and Cost-E_ectiveness of Health Care Interventions
Nicky J Welton, Mark Strong, Christopher Jackson, and Gianluca Baio
15.1 Introduction
15.2 Economic evaluation: A Bayesian decision theoretic analysis
15.3 Trial-based economic evaluation
15.4 Model-based economic evaluation
15.5 Value of information
15.6 Conclusion / outlook
Chapter 16: Bayesian Modeling for Economic Evaluation Using "Real World Evidence"
Gianluca Baio
16.1 Introduction
16.2 Real World Evidence
16.3 Economic modeling and survival analysis
16.4 Case study: ICDs in cardiac arrhythmia
16.5 Conclusions and further developments
Chapter 17: Bayesian Bene_t-Risk Evaluation in Pharmaceutical Research
Carl Di Casoli, Yueqin Zhao, Yannis Jemiai, Pritibha Singh, and Maria Costa
17.1 Introduction
17.2 Classical approaches to quantitative bene_t-risk
17.3 Bayesian approaches to quantitative bene_t-risk
17.4 Outlook for Bayesian bene_t-risk
17.5 Discussion
IV Product development and manufacturing
Chapter 18: Product Development and Manufacturing
Bruno Boulanger and Timothy Mutsvari
18.1 Introduction
18.2 What is the question in manufacturing?
18.3 Bayesian statistics for comparability and analytical similarity
18.4 Bayesian approach to comparability and biosimilarity
18.5 Conclusions
Chapter 19: Process Development and Validation
John J. Peterson
19.1 Introduction
19.2 ICH Q8 design space
19.3 Assay robustness
19.4 Challenges for the Bayesian approach
Chapter 20: Analytical Method and Assay
Pierre Lebrun and Eric Rozet
20.1 Introduction
20.2 Analytical quality by design
20.3 Assay development
20.4 Analytical validation and transfer
20.5 Routine
20.6 Conclusion
Chapter 21: Bayesian Methods for the Design and Analysis of Stability Studies
Tonakpon Hermane Avohou, Pierre Lebrun, Eric Rozet, and Bruno Boulanger
21.1 Introduction
21.2 New perspectives on stability data analysis
21.3 Stability designs, models and assumptions
21.4 Overview of frequentist methods in stability data
21.5 Bayesian methods of analysis of stability data
21.6 Conclusions
Chapter 22: Content Uniformity Testing
Steven Novick and Bu_y Hudson-Curtis
22.1 Introduction
22.2 Classical procedures for testing content uniformity
22.3 Bayesian procedures for testing content uniformity and risk
22.4 Challenges for the Bayesian procedures
Chapter 23: Bayesian methods for in vitro dissolution drug testing and similarity
comparisons
Linas Mockus and Dave LeBlond
23.1 Introduction
23.2 Current statistical practices in IV dissolution and their limitations
23.3 The value of adopting Bayesian paradigms
23.4 Applying Bayesian approaches: Two examples
23.5 Conclusions
Chapter 24: Bayesian Statistics for Manufacturing
Tara Scherder and Katherine Giacoletti
24.1 Introduction
24.2 Manufacturing situation 1: Revalidation/transfer
24.3 Manufacturing situation 2: Evaluating process capability
24.4 Manufacturing situation 3: Bayesian modeling of complex testing schemes
24.5 Discussion
V Additional topics
Chapter 25: Bayesian Statistical Methodology in the Medical Device Industry
Tarek Haddad
25.1 Introduction
25.2 Use of stochastic engineering models in the medical device design stage
25.3 Bayesian design and analysis of medical device trials
25.4 Challenges
Chapter 26: Program and Portfolio Decision-Making
Nitin Patel, Charles Liu, Masanori Ito, Yannis Jemiai, Suresh Ankolekar, and Yusuke
Yamaguchi
26.1 Introduction
26.2 Classical approaches
26.3 Current Bayesian approaches to program design
26.4 Program and portfolio-level Bayesian decision analysis
26.5 Research opportunities
Index
