Online Program Home
  My Program

Abstract Details

Activity Number: 467 - Bayesian Methods and Applications in Clinical Trials (II)
Type: Contributed
Date/Time: Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract #323730
Title: Bayesian Design for Interim Analysis
Author(s): Dung-Tsa Chen*
Companies: Moffitt Cancer Center
Keywords: interim analysis ; Bayesian ; phase I ; phase II ; toxicity ; futility
Abstract:

My collaborative research in clinical trail often faces stopping rule issue of interim analysis in phase I or II study. In phase I study, investigators are likely interested in various types of stopping rule for toxicity in interim analysis from continuous monitoring to multi-stages. The question is how to develop a approach which allows the flexibility. In a two-arm (treatment and control) phase II study, investigator may want to have an interim analysis for futility in the treatment arm. What statistical strategy could be used? In this presentation, Bayesian approaches are used to develop stopping rules for toxicity and futility. For phase I, Bayesian posterior probability is used to calculate toxicity rate. This approach has great flexibility to allow for different priors, various frequency of interim analysis, and easy interpretation. For phase II, Bayesian predictive probability could be used to develop stopping rules for futility by formulating hypothesis, calculating the expected number of responses, and then calculate the predictive probability. A few examples of clinical trials will be used for illustration.


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

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association