Adaptive methods: telling "the rest of the story"

Abstract

The Food and Drug Administration (FDA) draft guidance on adaptive design randomized clinical trials provides in-depth consideration of the difficulties that unblinded adaptation of clinical trial design might introduce. We provide extended discussion of these difficulties, with focus on the problems that the adaptive designs pose in the scientific interpretation of randomized clinical trial results, for regulatory authorities as well as for patients and caregivers who wish to make evidence-based decisions regarding the choice of treatment. We consider implications in adequate and well-controlled studies of the use of unblinded measures of treatment effect to make adaptive selection/modification of treatments, adaptive selection of primary endpoints, adaptive modification of maximal sample size, adaptive modification of randomization ratios, and adaptive modification of target populations (adaptive enrichment), and then we consider the special topic of seamless phase 2-3 designs. We examine the extent to which the adaptive designs do not meet the goals of having greater efficiency, being more likely to identify truly effective treatments, being more informative, and providing greater flexibility. We fully support the FDA's continued requirement of adequate and well-controlled confirmatory studies, complete with prospective, detailed specification of the entire randomized clinical trial design in a way that allows accurate and precise estimation of treatment effectiveness.