ABSTRACT
Confirmatory adaptive designs as suggested by the pioneering papers of Peter Bauer and colleagues were originally proposed for testing multiple hypotheses during the stages of a trial. The proposal was to use the combination testing principle together with the closed testing principle in order to define procedures that control the experimentwise error rate in a strong sense. An important application for these adaptive designs with multiple objectives is the design with multiple treatment arms where, based on interim results, one or more arms are selected for the subsequent course of the trial. Another application is the design where one or more pre-speci?ed subsets of a population are selected for further investigation, the latter designs are called adaptive enrichment designs. We describe the basic principles and the procedures in detail and provide some case studies for these designs.
