Response-Adaptive Randomization Procedure in Clinical Trials with Surrogate Endpoints

Area of Science:

• Clinical Trials
• Biostatistics
• Medical Research Methodology

Background:

• Sequential subject recruitment is standard in clinical trials.
• Response-adaptive randomization (RAR) optimizes treatment allocation based on accumulating trial data.
• Limited primary endpoint data during recruitment hinders RAR application, especially in public health emergencies or chronic diseases.

Purpose of the Study:

• To propose a novel RAR procedure utilizing surrogate endpoints when primary endpoints are unavailable.
• To demonstrate the validity and patient benefit of RAR with surrogate endpoints.
• To explore the performance of this approach across different surrogate and primary endpoint types.

Main Methods:

• Development of a response-adaptive randomization procedure relying solely on surrogate endpoints.
• Theoretical justification of statistical inference validity for the primary endpoint.
• Simulation studies to assess patient benefit and approach performance.

Main Results:

• The proposed RAR procedure using surrogate endpoints is statistically valid.
• Simulations confirm the approach's viability and patient benefit.
• Effectiveness is demonstrated across various surrogate and primary endpoint combinations.

Conclusions:

• RAR using surrogate endpoints offers a viable solution for clinical trials with delayed primary outcomes.
• This method enhances treatment allocation efficiency and patient benefit in challenging recruitment scenarios.
• Surrogate endpoints provide a practical alternative for adaptive trial designs when primary data is scarce.