Optimization of adaptive designs with respect to a performance score

Area of Science:

• Clinical Trials
• Biostatistics
• Statistical Methodology

Background:

• Adaptive designs are increasingly used in clinical trials for flexibility, particularly in sample size adjustments.
• Evaluating and selecting among various adaptive designs is crucial for efficient trial conduct.
• Existing scoring methods aid in choosing appropriate adaptive designs, but further optimization is needed.

Purpose of the Study:

• To derive optimal adaptive two-stage designs using the conditional performance score.
• To investigate variations of the performance score, including weighted components and prior effect size assumptions.
• To extend the optimization framework with power constraints and optimize critical values.

Main Methods:

• Utilized a constrained optimization problem to determine optimal two-stage adaptive designs.
• Employed the conditional performance score by Herrmann et al. (2020) as the optimization criterion.
• Investigated modifications to the performance score and incorporated a global power constraint.

Main Results:

• Developed a method for deriving optimal adaptive two-stage designs based on a specific performance score.
• Demonstrated the impact of score variations and power constraints on design optimization.
• Provided insights into sample size curves and overall design performance.

Conclusions:

• The proposed method facilitates the practical application of scoring criteria for adaptive two-stage designs.
• Optimizing adaptive designs with performance scores and constraints enhances their utility in clinical research.
• Further research into score variations can refine the selection of adaptive clinical trial designs.