Abstract:
|
Simon's two-stage design and the "admissible" two-stage design by Jung et al. have been commonly used in practice for single-arm phase II cancer clinical trials when the primary endpoint is binary. These design approaches, however, do not take into account the balance in sample size between the two stages, and thus the ethical benefit expected by the two-stage design may not be achieved if a severe imbalance in the sample sizes of the two stages is observed. In addition, investigators may desire a two-stage design which controls the probability of early termination under the alternative hypothesis (PET1) under a pre-selected level. To address these two concerns, we propose a modified Simon's two-stage design with reasonable sample size evaluated in the first stage, while maintaining the PET1 under a pre-selected level. The characteristics of the modified Simon's two-stage design will be compared to Simon's and the admissible two-stage design.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.