Statistical Inferences of Two-Stage Phase II Cancer Clinical Trials with Two Co-primary Endpoints

Abstract

In cancer research, two-stage designs are usually used to assess the effect of a new agent in phase II clinical trials. The optimal two-stage designs with two co-primary endpoints have been proposed to assess the effects of new cancer treatments, such as cytostatic or molecularly targeted agents (MTAs), based on both response rate and early progression rate. Statistical inference procedures, such as, point estima tion, p-value, and confidence region, for the true response rate and early progression rate based on the data from the phase II trials conducted according to the optimal two-stage designs would be very useful for further testing of the agents in phase III trials but have not been addressed in the literature. In this thesis, I first provide a review of the optimal two-stage design for phase II clinical trials with one endpoint and statistical inference procedures developed for this design. Then I propose some new statistical inference procedures for the optimal two-stage design of phase II clin ical trials with two co-primary endpoints, which include naive maximum likelihood estimate (MLE), bias-corrected estimates, and uniformly minimum variance unbiased estimate (UMVUE) for the point estimation, naive p-value and likelihood ratio test (LRT) based p-value for the hypothesis testing, and LRT based confidence region. Simulation studies were performed to evaluate the performance of these procedures.