GROUP SEQUENTIAL-METHODS FOR COMPARISON OF CURE RATES IN CLINICAL-TRIALS

Since a subset of the patient population may be cured of disease in so me clinical trials, one might be more interested in testing the differ ences in cure rates rather than other types of differences in failure distributions.;Use, the monitoring of clinical trial results frequentl y necessitates the use of repeated examination of outcome for ethical and efficiency reasons. Parametric and non-parametric group sequential methods are proposed for comparing the cure rates between two treatme nt groups at the interim analyses during the trial. In both methods, c onstruction of the group sequential boundaries is based on the asympto tic joint distribution of the sequentially computed statistics and the ir covariance matrix. The proposed methods are illustrated with a pedi atric cancer trial which had been conducted by the Childrens Cancer Gr oup.