Testing for Similarity of Dose Response in Multiregional Clinical Trials.

This article addresses the problem of determining whether the dose response relationships between subgroups and the full population in a multiregional trial are similar. Similarity is assessed in terms of the maximal deviation between the dose response curves. We consider a parametric framework and develop two powerful bootstrap tests: one for assessing the similarity between the dose response curves of a single subgroup and that of the full population, and another for comparing the dose response curves of multiple subgroups with that of the full population. We prove the validity of these tests, investigate their finite sample properties through a simulation study and illustrate the methodology with a case study.