|
Abstract:
|
Seamless Phase 2/3 clinical trials can accelerate drug development by evaluating multiple treatments in the first part of the trial and selecting one treatment for the second part for establishing its superiority in comparison with a control. Most of the research in this area has been focused on the control of the family wise type I error rate which is the probability of recommending at least one ineffective treatment at the end of the trial. Although, many selection rules have been considered, little research is devoted to the optimality of such selection procedures. We consider the probability of correct selection as a metric for evaluating and comparing different selection procedures and propose an optimality criterion based on average total number of subjects required to achieve the pre-specified probability of correct selection.
|
