Bayesian Adaptive Clinical Trials in the 21st Century

Monday, 7 August 2017
Melbourne School of Population and Global Health
Room 515, Level 5, 207 Bouverie Street
Carlton 3052

As medical research continues to push into new frontiers of discovery and personalized patient care, it is imperative that clinical trial designs evolve to address the forthcoming challenges.   One key innovation is the use of adaptive clinical trial designs.  Such designs allow certain trial design features to change during the course of the trial based on observed data within the trial.  These adaptations are planned prospectively and are used to increase trial efficiency and to increase the probability of achieving the scientific goals of the study.  Adaptive trial designs may include frequent interim analyses, adaptive sample sizes, longitudinal modelling of time-delayed outcomes, borrowing of historical data, study population enrichment, or response adaptive randomization.  We discuss the advantages of adaptive trial designs versus traditional strategies, the role of Bayesian methods and simulations in adaptive trial design, and corresponding interactions with regulatory agencies.


Dr Ben Saville

Berry Consultants

Ben Saville, PhD is a Statistical Scientist for Berry Consultants, where he specializes in the design of innovative Bayesian adaptive clinical trials.  He works primarily with medical device companies, pharmaceutical companies, and academic investigators to solve challenging problems via Bayesian designs, many of which are reviewed by the U.S. Food and Drug Administration (FDA).  He is a frequent invited speaker at various statistical conferences, academic seminars, and lecture series, including short courses on adaptive clinical trial design.  Dr. Saville earned his Ph.D. in Biostatistics from the University of North Carolina at Chapel Hill in 2008.  Prior to joining Berry Consultants in 2014, he was an Assistant Professor of Biostatistics at Vanderbilt University School of Medicine where his methodological research focused on Bayesian hierarchical models, Bayesian adaptive clinical trials, and nonparametric methods for randomized clinical trials.  At Vanderbilt, he collaborated extensively with medical researchers in the Department of Pediatrics and the Vanderbilt-Ingram Cancer Center, and was co-leader of an adaptive trials design workforce to promote innovative Bayesian methodology in clinical trials.  In addition, he taught undergraduate and graduate courses in the Department of Biostatistics and Department of Biomedical Engineering.  Dr. Saville has authored over 50 peer-reviewed publications in the statistical and medical literature.