SESSION KEYNOTE

Decision-analytic Bayesian approaches are ideally suited for designing clinical trials and have been used increasingly over the last 30 years in developing medical devices and drugs. A prototype trial is a bandit problem, in which treating participants is as important as treating patients in clinical practice after the trial.

This keynote provides a comprehensive historical and practical review of Bayesian adaptive trials, with a focus on bandit-inspired designs. The 20th century saw foundational advances in Bayesian methodology enabled by computer simulation. In the 21st century, methods motivated by bandit problems have been applied in designing scores of actual clinical trials — fifteen of which will be discussed. By far the most important Bayesian contributions in clinical trials are the abilities to observe accumulating results and to modify the future course of the trial on the basis of those observations. In the spirit of artificial intelligence, algorithms are programmed to learn optimal treatment assignments over the remainder of the trial.

Bayesian trials are still nascent and represent a small minority of clinical trials, but their existence is changing the way investigators, regulators, and government and industry sponsors view innovation in clinical trial design.

The keynote will also include I-SPY2 as a landmark example beyond its Bayesian innovations. It was the first platform trial and its influence and design extend far beyond breast cancer by numerous subsequent trials, including the Phase 3 GBM AGILE trial in glioblastoma. The July 7, 2016 issue of the NEJM reflected the field’s early engagement with I-SPY2, publishing four articles related to the trial in a single issue — a notable concentration of attention for any single study in a prominent medical journal. Subsequently, Woodcock and LaVange published a 2017 NEJM review paper introducing the concept of master protocols to a wide medical audience, with I-SPY2 as its primary illustration, demonstrating its regulatory and scientific impact.

Instructors’ Biography:

Professor Donald A. Berry is a leading biostatistician and pioneer of Bayesian methods in medicine, widely recognized for his contributions to adaptive clinical trial design.   At MD Anderson, he served as the founding Chair of the Department of Biostatistics (1999–2010) and played a central role in advancing the use of Bayesian approaches for innovative oncology trials.   He founded Berry Consultants in 2000, a scientific consulting firm specializing in Bayesian and adaptive clinical trial design and implementation for pharmaceutical and medical device development.   Dr Berry has designed and led numerous innovative trials, including platform trials evaluating multiple therapies simultaneously, including the landmark I‑SPY 2 adaptive platform trial in breast cancer. He has published hundreds of scientific papers and multiple books in statistics and biostatistics.