Online Program Home
My Program

Abstract Details

Activity Number: 304
Type: Topic Contributed
Date/Time: Tuesday, August 2, 2016 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract #320728
Title: Bayesian Predictive Approach to Concurrent Tailoring of Phase 3 Clinical Trials Intended for Registration
Author(s): Ming-Dauh Wang* and Aijun Gao and Jinghui Liu and Karen Price and MaryAnn Morgan-Cox and Lei Shen
Companies: Eli Lilly and Company and inVentiv Health and inVentiv Health and Eli Lilly and Company and Eli Lilly and Company and Eli Lilly and Company
Keywords: Multiple clinical trials ; Concurrent tailoring ; Bayesian prediction ; Monte Carlo simulation ; Probability of success

When multiple clinical trials are concurrently ongoing, the outcomes of the ones that end earlier may provide an opportunity for modification of the analysis plans of those to conclude later. We applied this concept of concurrent tailoring (CT) to a Phase 3 program where 2 trials with binary endpoints slightly staggered in time to each other intended for supporting registration were in progress. The interest was in the possibility of modifying the study population of the 2nd trial to a more promising subgroup if the 1st trial would turn out negative in the overall population, hoping that positive outcome in the subgroup of the 2nd trial alone would still suffice a drug approval. Prior to the end of the 1st trial, simulation and Bayesian analysis were conducted to predict likely outcomes of the 2nd trial. Operating characteristics, particularly probability of study success, for different scenarios of assumptions for the overall population and various subgroups were studied to prepare for the CT decision on the 2nd . Upon completion of the 1st trial, prediction based on the observed results was made for the overall population and subgroups of the 2nd trial for the final CT decision.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association