Online Program Home
My Program

Abstract Details

Activity Number: 75
Type: Contributed
Date/Time: Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
Sponsor: Biopharmaceutical Section
Abstract #321429
Title: Bayesian Clinical Trial Design for Survival Studies with Historical Study Data Under a Proportional Hazards Assumption
Author(s): Matthew Psioda* and Joseph G. Ibrahim
Companies: The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill
Keywords: Clinical Trial Design ; Power Prior ; Proportional Hazards ; Sample Size Determination ; Sampling Prior ; Type I Error Rate

We consider Bayesian design of a survival study where a previously completed clinical trial is available to inform the design and analysis of a confirmatory trial. Our results illustrate that Bayesian analysis with an informative prior is in direct conflict with frequentist type I error control. As a solution, we propose a design strategy based on Bayesian versions of type I error and power that are defined with respect to the posterior distribution for the model parameters given the historical data and conditional on the relevant hypothesis being true. We demonstrate that when a design controls the Bayesian type I error rate, meaningful amounts of prior information can be borrowed and that full information borrowing requires a future study of reasonable size. Using a proportional hazards regression model with a stratified piecewise constant baseline hazard, we obtain a previously unexploited closed-form for the posterior for the hazard ratio regression parameters and establish a new connection with the weighted Cox partial likelihood. We develop an efficient simulation-based design methodology that avoids MCMC methods through an accurate normal approximation.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association