Presentation: Adaptive Clinical Trial Design and Analysis with Time to Event Endpoints
2017 Ross L. Prentice Endowed Professor: Scott S. Emerson, M.D., Ph.D., Professor and Graduate Program Director, Department of Biostatistics, University of Washington
Abstract: A great many confirmatory phase 3 clinical trials have as their primary endpoint a comparison of the distribution of time to some event (e.g., time to death or progression free survival). The most common statistical analysis models include the logrank test (usually unweighted, but possibly weighted) and/or the proportional hazards regression model. Just as commonly, the true distributions do not satisfy a proportional hazards assumption. Providing users are aware of the nuances of those methods, such departures need not preclude the use of those analytic techniques any more than violations of the location shift hypothesis precludes the use of the t test. However, with the increasing interest in the use of adaptive sample size re-estimation, adaptive enrichment, response-adaptive randomization, and adaptive selection of doses and/or treatments, there are many issues (scientific, ethical, statistical, and logistical) that need to be considered. In fact, when considering references to “less well understood” methods in the draft FDA guidance on adaptive designs, it is likely the case that many of the difficulties in adaptive time to event analyses can relate as much to aspects of survival analysis that are “less well understood” as to aspects of the adaptive methodology that has not been fully vetted. In this talk I review the general principles behind adaptive clinical trial design, and then discuss some aspects of the analysis of censored time to event data that must be carefully considered in such sequential and adaptive sampling. In particular, I discuss how the changing censoring distribution during a sequential trial affects the analysis of distributions with crossing hazards and crossing survival curves, as well as issues that arise owing to the ancillary information about eventual event times that might be available on subjects who are censored at an adaptive analysis.
About Dr. Emerson: Scott S. Emerson, M.D., Ph.D., is Professor of Biostatistics at the University of Washington, where he directs and teaches in the graduate program in Biostatistics, performs research into statistical methods for sequential clinical trials and survival analysis, and collaborates on clinical trials in hematologic malignancies and in pre-hospital emergency medicine. After receiving an undergraduate degree in physics (1977) and an M.D. (1981), both from the University of Virginia, Dr. Emerson earned a Master's degree in computer science (University of Virginia, 1984) and a Ph.D. in Biostatistics (University of Washington, 1988). He held faculty positions in the Department of Statistics at the University of Florida (1988-89) and in the Department of Statistics and Arizona Cancer Center at the University of Arizona (1989-95) prior to joining the faculty of the Department of Biostatistics at the University of Washington in 1995.
Dr. Emerson is active in research into methods for the design, conduct, and analysis of clinical trials. A major focus of his research has been in the use of sequential methods, both frequentist and Bayesian, in the monitoring and reporting of clinical trial results. Recent areas of research include the sequential monitoring of longitudinal studies, especially when the primary outcome is based on time to event, adaptive clinical trial designs, noninferiority trials, and nonparametric regression methods, especially as applied to survival data and ROC curve analysis. He serves on a number of government and industry sponsored Data Safety Monitoring Boards (DSMBs). Computer programs that he developed for his research into group sequential methodology now form the backbone of S+SeqTrial, an S-PLUS module for group sequential trial design that was recently ported to R as RCTdesign. Dr. Emerson has served on multiple FDA and NIH advisory committees. He was also a member of the Panel on Handling Missing Data in Clinical Trials for the National Research Council, and he has in recent years spoken frequently on standards for the prevention and treatment of missing data in clinical trials.
Dr. Emerson received the Outstanding Teaching Award from the University of Washington School of Public Health in 1999, and he received awards for presentations on Teaching Statistics in the Health Sciences (1997, 2008) and Biopharmaceutical Statistics (2012). He is a Fellow of the American Statistical Association.