Properties of Covariate-Adjusted Response-Adaptive Designs with Early Stopping

• Activity Number: 128
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Biopharmaceutical Section
• Author(s): Yaping Wang*+ and Jack Lee and Hongjian Zhu
• Companies: University of Texas Health Science Center at Houston and MD Anderson Cancer Center and University of Texas School of Public Health
• Keywords: Clinical Trial ; Adaptive Design

Abstract

Covariate-Adjusted Response-adaptive (CARA) change randomization probabilities sequentially based on observed outcome in each covariate subgroup. More patients are assigned to the better treatment based on their biomarker, therefore, it can yield higher overall response rates for patients in trials. The asymptotic properties of the both CARA designs and early stopping has been proposed and extensively studied. However, there is a lack of comprehensive studies to combine these two methods, and compare its properties with those of other designs. In this study, we derived the model of CARA from bayes view. In addition, we conducted simulation studies to evaluate the properties of CARA designs with early stopping.