2-in-1 adaptive design with dose optimization accounting for ranked dose response

Abstract

In oncology clinical trials, seamless 2-in-1 adaptive designs allow mid-trial adaptations on the development pathways to expedite drug development. In this paper, we propose a 2-in-1 adaptive design that allows the expansion of a Phase 2 trial, starting with multiple doses, into a Phase 3 confirmatory trial with one selected dose that demonstrates promising efficacy. The rank of the selected dose will be accounted for in the statistical testing. Simulations have been performed to assess the Type I error and power. The proposed 2-in-1 design can offer attractive operating characteristics, particularly when treatment effect is delayed.