Bayesian Precision Medicine Designs for Cancer Clinical Trials
Abstract: In this talk, I will describe two Bayesian sequentially adaptive designs for oncology clinical trials that accommodate heterogeneous patients and include subgroup-specific decisions. The talk will begin with preliminary concepts on the use of utilities as criteria to quantify risk-benefit trade-offs for making decisions. The first design is for an early phase trial that does subgroup-specific safety monitoring and utility-based dose optimization for natural killer cells as treatment for hematologic malignancies. Decisions are made for six disease subrgroups, based on a vector of five potential time-to-event outcomes. The second design, illustrated by a trial in non small cell lung cancer, is for group sequential phase III trials based on survival time that include subgroup-specific treatment comparisons. The design uses latent variables to adaptively combine similar subgroups, and it also can improve reliability by exploiting a baseline biomarker that may be related to survival time. Both designs are evaluated by extensive computer simulation studies that include comparisons to designs that ignore patient heterogeneity.
Speaker: Peter D. Thall, PhD
M.D. Anderson Cancer Center
University of Texas
Bio: Peter Thall is the Anise J. Sorrell Professor in the Department of Biostatistics at M.D. Anderson Cancer Center. He is a Fellow of the American Statistical Association and the Society for Clinical Trials, and received the Owen Award in 2014. Dr. Thall has published over 280 papers and book chapters, co-authored the 2016 book Bayesian Designs for Phase I-II Clinical Trials, and is sole author of the 2020 book Statistical Remedies for Medical Researchers. His research areas include Bayesian statistics, clinical trial design, precision medicine, and dynamic treatment regimes. He has served as an associate editor for Journal of the National Cancer Institute, Statistics in Medicine, Statistics in Biosciences, Clinical Trials, and Biometrics. Dr. Thall is a co-Principal Investigator of the NCI RO1 grant Bayesian Methods for Complex Precision Biotherapy Trials in Oncology