A unifying approach for surrogate marker validation based on Prentice's criteria

Area of Science:

• Biostatistics
• Clinical Trial Methodology
• Longitudinal Data Analysis

Background:

• Existing methods for surrogate endpoint evaluation often lack a unified approach across different data types and settings.
• Previous methodologies, such as Buyse et al. and Alonso et al., have limitations in handling diverse endpoint distributions and longitudinal data.
• The need for a consistent framework to assess the validity of surrogate endpoints in complex clinical trial designs is evident.

Purpose of the Study:

• To propose a new methodology for evaluating surrogate endpoints within a repeated measurements framework.
• To demonstrate the connection between the proposed approach and existing literature on surrogate endpoint validation.
• To extend the methodology to non-normal data using the likelihood reduction factor (LRF) for enhanced applicability.

Main Methods:

• Development of a new approach for surrogate endpoint evaluation in the context of longitudinal data with repeated measurements.
• Analysis of the relationship between the proposed method and previously established techniques for assessing surrogacy.
• Extension of the framework to accommodate non-normally distributed endpoints through the application of the likelihood reduction factor (LRF).

Main Results:

• A novel methodology for evaluating surrogate endpoints in repeated measures settings has been successfully developed.
• The proposed approach integrates and connects with existing literature, offering a more comprehensive understanding of surrogacy.
• The extension to non-normal data using LRF provides a flexible tool for a wider range of clinical trial data.

Conclusions:

• The proposed methodology offers a unified and robust approach to surrogate endpoint evaluation in longitudinal studies.
• The extension to non-normal data significantly broadens the applicability of surrogate endpoint validation in clinical research.
• This work provides valuable tools for assessing endpoint validity in psychiatry and ophthalmology clinical studies.