From the University of Chicago, Chicago, Illinois
Author's disclosure of potential conflicts of interest is found at the end of this article.
Address reprint requests to Walter M. Stadler, MD, FACP, University of Chicago, 5841 S. Maryland Avenue, MC 2115, Chicago, IL; e-mail: wstadler{at}medicine.bsd.uchicago.edu
Overview: Phase II trial design in oncology has been dominated by historically controlled single-arm trials with objective response as an endpoint. This approach is not necessarily appropriate when an agent is expected to be growth inhibitory and when dichotomizing the continuous variable of tumor burden discards important information. Additionally, evaluation of any tumor burden endpoint, except under the increasingly unique scenario of dramatic tumor shrinkage in comparison with an untreated control, requires a concurrent control group. As such, investigators should increasingly give attention to continuous measures of tumor burden as a phase II trial endpoint. They also should give more attention to concurrent controlled phase II trial designs, including the randomized discontinuation trial design, which is especially useful for evaluating growth inhibitory agents.
