ABSTRACT
This chapter discusses the adaptive clinical trials that incorporate Bayesian statistical considerations. It provides some of the reasons one might consider applying an adaptive design, there are ethical, practical, and statistical considerations that may temper one's enthusiasm toward adaptive trials in all instances, especially trials that incorporate response adaptive randomization. Benefits apply over a broad range of clinical trials, including studies that occur early in the drug development process. An advantage of the moving-reference adaptive randomization approach is that the randomization probabilities are no longer sensitive to which treatment is considered the control or reference treatment. Randomization probabilities based on the treatment-specific posterior means could serve the goal of adaptive randomization in this case. Ongoing evaluation of how the trial is proceeding once patients start to enroll in the study, including checks of the randomization probabilities once adaptation begins, will help ensure the success of the randomized trial.
