A Bayesian design for phase II clinical trials with delayed responses based on multiple imputation.

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Bayesian Inference

Background:

• Phase II clinical trials routinely use interim monitoring to stop futile treatments early.
• Timely assessment of patient responses is crucial for effective interim monitoring.
• Delayed response outcomes pose significant challenges to traditional interim monitoring strategies.

Purpose of the Study:

• To propose a novel Bayesian trial design for continuous interim monitoring in phase II clinical trials.
• To address the challenges posed by delayed patient response data.
• To improve the efficiency and reduce the duration of phase II clinical trials.

Main Methods:

• Development of a Bayesian design for continuous monitoring.
• Treatment of delayed responses as missing data.
• Application of a multiple imputation technique to handle missing data.

Main Results:

• The proposed Bayesian design demonstrates desirable operating characteristics across various simulation settings.
• The design effectively manages delayed responses, mitigating challenges in interim monitoring.
• Simulations indicate a substantial reduction in overall clinical trial duration.

Conclusions:

• The proposed Bayesian trial design offers a viable solution for interim monitoring with delayed responses in phase II trials.
• Multiple imputation is an effective strategy for handling missing outcome data in this context.
• This approach can lead to more efficient and timely clinical trial decision-making.