Sample size determination and evaluation for two-stage adaptive designs of single arm clinical trials based on median...

Area of Science:

• Biostatistics
• Clinical Trial Design
• Pharmacometrics

Background:

• Clinical trials are essential for drug development, demanding significant time and resources.
• Time-to-event endpoints, particularly median survival time, are preferred for Phase II efficacy studies.
• Traditional methods like the one-sample log-rank test often require larger sample sizes.

Purpose of the Study:

• To improve the computational efficiency of the median event time test for two-stage single-arm Phase II clinical trials.
• To provide explicit formulas for sample size determination in both stages of the trial.
• To propose an enhanced testing procedure for evaluating new drugs based on median survival time.

Main Methods:

• Utilizing the large sample theory of order statistics for precise sample size calculations.
• Developing explicit formulas for determining sample sizes for the first and second stages.
• Implementing a two-stage design based on the median event time test.

Main Results:

• The proposed method provides explicit formulas for sample size calculation, improving efficiency.
• The enhanced testing procedure is suitable for single-arm Phase II clinical trials.
• Simulations and a trial example demonstrate the performance of the improved method.

Conclusions:

• The improved median event time test offers a more efficient approach for Phase II clinical trial design.
• Explicit sample size formulas facilitate better planning and resource allocation in drug development.
• This methodology supports the investigation of new drugs by optimizing the use of time-to-event endpoints.