[Design and analysis of phase II trials in oncology with a group sequential method, the triangular test].

In cancer, phase II clinical trials are most of the time non comparative trials, the purpose of which is to determine whether a new chemotherapy is effective enough to warrant further study in phase III. So, it is necessary, at this stage, to dispose of decision methods which allow, for ethical reasons, for early stopping when inefficacy or efficacy is evident. In a previous work, we have extended to these phase II trials two group sequential methods initially proposed in phase III, and we have demonstrated the advantages of these methods (type I error alpha and power close to the nominal values, reduction of the sample size) as compared to the main methods proposed in the literature. The aim of this paper is to present the method which appears to be the best suited to phase II clinical trials, the triangular test. After a first part, focused on its theoretical aspects, the second part, more practical, presents its use for the design of a cancer trial and its analysis.