Response-Adaptive Randomization: Recent Developments and Controversies

• Activity Number: 592
• Type: Topic Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): J. Jack Lee* and Nan Chen
• Companies: MD Anderson Cancer Center and MD Anderson Cancer Center
• Keywords: adaptive design ; Bayesian framework ; clinical trial ethics

Abstract

Response-adaptive randomization (RAR) allocates more patients to the better treatments as the information accumulates in the trial. Is it worth it to apply RAR in clinical trials? Different views exist in medical/statistical communities. However, most people agree with the following 3 common goals in trial design: to (1) maintain the type I error rate, (2) achieve a specified power, and (3) provide better treatment to patients within and beyond the trial. Generally speaking, RAR results in a larger sample size and a higher overall response rate than equal randomization (ER). When there are substantial efficacy differences and/or more than 2 treatments, RAR can yield a higher overall response rate as well as a lower sample size. The differences between RAR and ER quickly diminish with early stopping of a trial due to efficacy or futility. ER focuses on maintaining a specified statistical power with a smaller number of patients in the trial. RAR focuses on treating patients best in the trial. RAR may be preferred when the efficacy difference between treatments is large or when the number of available patients is limited. Practical considerations for implementing RAR will be given.