DOD-SSR: An Adaptive Seamless Phase II/III Design With Dose Optimization Decision and Sample Size Re-estimation

Area of Science:

• Clinical Trials
• Biostatistics
• Drug Development

Background:

• Seamless Phase II/III designs accelerate drug development by integrating exploratory and confirmatory phases.
• Dose optimization and early decision-making are critical in modern clinical trials.
• Fixed sample sizes in early-phase trials pose challenges due to limited data and high risks.

Purpose of the Study:

• To introduce a novel seamless Phase II/III design for multiple treatment arms.
• To leverage Bayesian predictive probability of success (PPoS) for treatment arm selection and interim sample size re-estimation (SSR).
• To utilize surrogate endpoints for making informed interim decisions.

Main Methods:

• A Bayesian framework incorporating PPoS for treatment arm selection and SSR.
• Integration of surrogate endpoints for interim analyses.
• Comparison with traditional separate Phase II/III designs and alternative seamless designs.

Main Results:

• The proposed design enhances the likelihood of selecting favorable treatment arms.
• Improved overall statistical power and reduced average trial size and duration.
• Demonstrated superior performance compared to existing seamless designs without SSR or using conditional power.

Conclusions:

• The novel seamless Phase II/III design offers significant advantages in efficiency and success probability.
• The framework effectively utilizes PPoS and surrogate endpoints for adaptive decision-making.
• Applicable to various therapeutic areas, including non-small cell lung cancer (NSCLC).