Bivariate sequential designs for phase II trials.

In this paper we propose methods for designing group sequential phase II trials with two dependent binary endpoints. The emphasis is on the derivation of stopping rules for phase II trials which require the enrollment of a small number of patients. The methods are based on enumerating the exact distribution for the binary endpoints. We illustrate the methods with a recent study which required the use of group sequential design to monitor antitumor activity and toxicity.