Evaluation Methods for T-association of a Surrogate Endpoint

Area of Science:

• Biostatistics
• Clinical Trial Methodology
• Regulatory Science

Background:

• Surrogate endpoints are vital for the FDA Accelerated Approval pathway, substituting direct clinical benefit measures.
• Valid surrogate endpoints require I-association and T-association, but T-association is often overlooked due to methodological gaps.
• Failure to demonstrate T-association prevents biomarker use in accelerated approvals.

Purpose of the Study:

• To introduce a novel statistical method for rigorously assessing T-association.
• To align biomarker validation with FDA guidelines for accelerated drug approval.
• To provide a statistical foundation for future accelerated approvals.

Main Methods:

• Proposes a method assuming bivariate normal distribution for treatment effects on surrogate and true endpoints.
• Accounts for both within-study and between-study variances.
• Estimates key parameters, including the correlation coefficient, using maximum likelihood, restricted maximum likelihood, and Bayesian approaches.

Main Results:

• The proposed method rigorously assesses T-association, addressing a critical gap in biomarker validation.
• Demonstrated utility through simulations and real-world data analysis.
• Provides a quantifiable metric (correlation coefficient) for the relationship between treatment effects.

Conclusions:

• The developed method offers a robust statistical framework for evaluating T-association.
• Facilitates compliance with FDA guidelines for surrogate endpoint validation.
• Supports the accelerated approval pathway by ensuring reliable prediction of clinical benefit.