Comparison of single-arm vs. randomized phase II clinical trials: a Bayesian approach

Area of Science:

• Clinical trial design
• Biostatistics
• Drug development

Background:

• Phase II clinical trials aim to assess treatment efficacy for further development.
• Single-arm studies are common but have limitations compared to randomized designs.
• Randomization in phase II trials is debated for improving decision accuracy over historical controls.

Purpose of the Study:

• To compare the decision-making abilities of single-arm versus randomized phase II clinical trials.
• To evaluate the impact of historical control rates, sample size, and prior distributions on study performance.
• To determine the optimal phase II trial design for new treatment evaluation.

Main Methods:

• Bayesian statistical approach applied to binary response variables.
• Comparative analysis of single-arm and randomized study designs.
• Simulation studies to assess performance under varying parameters.

Main Results:

• Randomized phase II designs demonstrate superior ability in correctly identifying effective and ineffective treatments compared to single-arm studies.
• The historical control rate, total sample size, and prior distribution elicitation significantly influence the choice of the better-performing study design.
• Bayesian analysis provides a robust framework for evaluating these designs.

Conclusions:

• Randomized phase II trials offer advantages over single-arm studies in clinical trial decision-making.
• Careful consideration of historical controls, sample size, and prior information is crucial for optimizing phase II trial design.
• The findings support the increased adoption of randomization in phase II to enhance treatment evaluation accuracy.