Keywords: Adaptive design, operational bias, IDMC
Adaptive trials are commonplace in the world of clinical trials. Adaptive designs are used in both phase 2 and 3 trials for dose-selection, endpoint selection, subpopulation selection, sample size re-estimation, etc. In studies where an interim analysis is performed for sample size re-estimation, the results of the interim analysis could potentially bias the final results. Since the re-estimation of sample size may be based on conditional power, the knowledge of the re-estimated sample size by those conducting or participating in the trial (i.e. the sponsor, site personnel, or study participants) may lead to operational bias. This becomes particularly important in trials where the outcome measures subjective. The discussion will revolve around the steps that can be taken to minimize operational and expectation bias.
1) What factors can be considered in minimizing operational bias in adaptive clinical trials?
2) What role do the IDMC, the independent statistical support group and the IWRS vendor play in coordinating and maintaining blinded information (IA results and re-estimated sample size)?