Quantifying proportion of treatment effect by surrogate endpoint under heterogeneity

Area of Science:

• Biostatistics
• Clinical Trials
• Pharmaceutical Research

Background:

• Surrogate endpoints are crucial in clinical trials when primary endpoints are long-term or costly.
• Assessing the surrogacy of an endpoint is vital before its adoption.
• Existing methods for surrogacy assessment may not account for variations across patient subgroups.

Purpose of the Study:

• To develop and evaluate methods for assessing surrogate endpoint validity when surrogacy varies across patient subgroups.
• To incorporate baseline covariates, such as age, to improve the assessment of overall surrogacy.
• To provide a robust estimation of treatment effect proportion explained by covariate-adjusted surrogate endpoints.

Main Methods:

• Proposed flexible semi-non-parametric modeling strategies to adjust for covariate effects.
• Incorporated baseline demographic characteristics (e.g., age) into surrogacy assessment.
• Utilized simulation studies to compare the proposed method with unadjusted approaches.

Main Results:

• The covariate-adjusted surrogate endpoint demonstrated a greater proportion of treatment effect compared to the unadjusted surrogate endpoint.
• The proposed method effectively assesses surrogacy in the presence of heterogeneity across patient subgroups.
• Application to infliximab trial data confirmed the method's utility in evaluating surrogate endpoint adequacy.

Conclusions:

• The developed statistical methods enhance the assessment of surrogate endpoints by addressing heterogeneity across patient subgroups.
• Incorporating baseline covariates improves the reliability and robustness of surrogate endpoint evaluation in clinical trials.
• This approach offers a more accurate way to determine if a surrogate endpoint adequately reflects treatment effects on the primary endpoint.