|
Abstract:
|
In a Phase II study with adaptive dropping arm design, the interim evaluation did not drop any inefficacious dose group, while none of the study drug dose groups demonstrated clinically meaningful efficacy at the end of the study. On the other hand, an active control group showed the expected treatment effect at the end of the study, which confirmed the validity of study design and execution. Retrospective research was conducted to evaluate and to identify dropping arm decision rules that could be more effective. In this presentation, the performance of different decision criteria using posterior or predictive probability models is evaluated. Additionally, the use of raw means vs. LS means in calculating posterior probabilities, and the results of a single interim look vs. those of multiple interim looks were also compared.
|
