A Bayesian predictive two-stage design for phase II clinical trials.

Area of Science:

• Biostatistics
• Clinical Trial Design
• Bayesian Inference

Background:

• Phase II clinical trials require efficient designs to evaluate drug efficacy.
• The single threshold design (STD) offers a two-stage approach but relies on fixed outcome assumptions.
• Existing methods may not fully account for the inherent uncertainty in future clinical trial data.

Purpose of the Study:

• To propose a novel Bayesian two-stage design for phase II clinical trials.
• To develop a predictive version of the single threshold design (STD) that incorporates data uncertainty.
• To control the probability of achieving a high posterior probability for the true response rate exceeding a target value.

Main Methods:

• Utilizing a Bayesian framework for a two-stage clinical trial design.
• Developing a predictive approach that differs from the STD by not assuming a fixed experimental outcome.
• Employing prior predictive distributions to express and control probabilities related to the true response rate.
• Investigating the design's performance using distinct analysis and design priors.

Main Results:

• The proposed design effectively controls the probability of obtaining a large posterior probability for the true response rate.
• The method accounts for uncertainty in future data, offering a more robust approach than traditional STD.
• Performance analysis demonstrates the design's properties across varying parameters.

Conclusions:

• The Bayesian two-stage predictive design provides a flexible and robust alternative for phase II clinical trials.
• This approach enhances decision-making by incorporating uncertainty about future data.
• The design offers improved control over trial outcomes compared to fixed-outcome designs.