A Bayesian Design for Phase II Clinical Trials with Delayed Responses Based on Multiple Imputation

• Activity Number: 178
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Chunyan Cai*+ and Suyu Liu
• Companies: UT Health Science Center at Houston and UT MD Anderson Cancer Center
• Keywords: Continuous monitoring ; Delayed responses ; Multiple imputation

Abstract

Interim monitoring is routinely conducted in phase II clinical trials to early terminate the trial if the experimental agents are futile. The interim monitoring requires that patient's responses be ascertained shortly after the initiation of treatment so that by the time of making the interim decision, the outcomes of the currently treated patients have been completely observed. However, in some cases, response outcomes take a long time to be assessed and brings difficulties for interim monitoring. To address this issue, we propose a Bayesian trial design to allow continuously monitoring phase II clinical trials in the presence of delayed responses. We treat the delay responses as missing data and handle them using the multiple imputation approach. Extensive simulations show that the proposed design yields desirable operating characteristics under various settings, and dramatically reduces the trial duration.