Evaluation Methods for T-association of a Surrogate Endpoint

A surrogate endpoint is a biomarker or intermediate clinical outcome used instead of a direct clinical endpoint to predict the clinical benefit of a drug. It serves as a substitute for a primary endpoint, offering benefits when it can be measured earlier or more conveniently. Before using a surrogate endpoint for scientific conclusions, its qualification must be assessed. A valid surrogate endpoint must meet two associations: I-Association (the association between the surrogate and true endpoints, such as disease response and overall survival) and T-Association (the association between treatment effects on both endpoints, such as odds ratio and hazard ratio). I-association is commonly evaluated, but T-association is often overlooked. This study proposes methods to assess T-association since both treatment effects are estimated as two random variables. We assume that the treatment effects on the surrogate and true endpoints follow a bivariate normal distribution, where the key evaluation metric is the correlation coefficient, which quantifies the relationship between treatment effects on both endpoints. The model parameters, including the correlation coefficient of interest, are estimated using maximum likelihood, restricted maximum likelihood, and a Bayesian approach. Simulated data and real-world data are used to demonstrate these methods. In simulations, we evaluate bias, standard error, and coverage probability. This method will serve as the statistical foundation for future FDA accelerated approval drugs.