WATCH: A Workflow to Assess Treatment Effect Heterogeneity in Drug Development for Clinical Trial Sponsors

Area of Science:

• Clinical Drug Development
• Biostatistics
• Clinical Trial Design

Background:

• Investigating treatment effect heterogeneity (TEH) in randomized clinical trials is challenging due to limitations in sample size and multiplicity.
• Existing methods may not reliably identify how treatment effects differ across patient subgroups.
• A structured approach is needed to guide the exploration of TEH.

Purpose of the Study:

• To propose a novel workflow, WATCH (Workflow for Assessing Treatment Effect Heterogeneity), for sponsors in clinical drug development.
• To provide a systematic method for exploring TEH in randomized clinical trials.
• To enhance the understanding of how treatment effects vary across baseline covariates.

Main Methods:

• The WATCH workflow comprises four pre-specified steps: analysis planning, initial data analysis and dataset creation, TEH exploration, and multidisciplinary assessment.
• The workflow emphasizes exploratory analysis, not confirmatory inference.
• It requires pre-planning before database lock and analysis initiation.

Main Results:

• The WATCH workflow offers a general overview of treatment effect variation based on baseline covariates.
• It guides the interpretation of findings using external evidence and scientific understanding.
• The focus is on a broad overview of TEH rather than identifying specific differential effects in subgroups.

Conclusions:

• The WATCH workflow provides a robust framework for exploring treatment effect heterogeneity in clinical trials.
• It supports a more comprehensive understanding of drug efficacy across diverse patient populations.
• This exploratory approach aids in the interpretation of clinical trial data for drug development sponsors.