Adaptive designs for single-arm phase II trials in oncology.

摘要: Clinical phase II trials in oncology are conducted to determine whether the activity of a new anticancer treatment is promising enough merit further investigation. Two-stage designs commonly used for this situation allow early termination. Designs proposed literature so far have common drawback that sample sizes two stages be specified protocol and adhered strictly during course trial. As consequence, higher extent flexibility desirable. In article, we propose adaptive method allows an arbitrary modification size second stage using results interim analysis or external information while controlling type I error rate. If not changed trial, design shows very similar characteristics optimal two-stage by Chang et al. (Biometrics 1987; 43:865-874). However, use mid-course planning stage, thus meeting practical requirements when performing clinical oncology.