Adaptive two-stage designs in phase II clinical trials

Area of Science:

• Clinical Trials Methodology
• Biostatistics
• Bayesian Statistics

Background:

• Two-stage designs are common in phase II clinical trials, enabling interim decisions on treatment efficacy.
• Simon's fixed two-stage designs offer a predetermined sample size allocation for binary outcomes.
• Adaptive designs enhance trial flexibility by allowing stage-two sample sizes to be data-dependent.

Purpose of the Study:

• To propose an adaptive two-stage clinical trial design.
• To derive optimal adaptive designs using a Bayesian decision-theoretic framework.
• To compare the proposed adaptive design with Simon's fixed design.

Main Methods:

• Development of an adaptive two-stage design for phase II clinical trials.
• Application of a Bayesian decision-theoretic construct for design optimization.
• Minimization of expected sample size under the null hypothesis as the optimality criterion.

Main Results:

• The proposed adaptive two-stage design allows for flexible sample size adjustments.
• Optimal adaptive designs were derived based on the Bayesian decision-theoretic construct.
• Comparisons highlight the performance of the adaptive design against fixed designs.

Conclusions:

• The adaptive two-stage design offers a more efficient approach compared to fixed designs.
• The Bayesian framework provides a robust method for deriving optimal adaptive trial designs.
• This adaptive strategy can lead to reduced sample sizes while maintaining decision-making capabilities.