Abstract
Adaptive Phase 2/3 designs hold great promise in contemporary oncology drug development, especially when limited data from Phase 1 dose-finding is insufficient for identifying an optimal dose. However, inconsistent results between Phase 2 and Phase 3 may raise regulatory and practical concerns. The imperfection in dose selection further complicates the issue. In this paper, we explicitly incorporate the concerns about inconsistency into the statistical analysis under three hypothesis testing strategies (conservative, aggressive, and neutral) by specifying an inconsistency cutoff and accounting for the probability of "picking-the-winner." This investigation illustrates how to balance regulatory caution, sponsor interests, and practical considerations in adaptive Phase 2/3 designs with dose optimization, which paves the way for further research in a less explored area.