Synergy Area With FDR-Controlled Evaluation (SAFE) to Robustly Assess Safety Profile in Clinical Trials

Area of Science:

• Pharmacovigilance and Drug Safety
• Clinical Trial Methodology
• Statistical Methods in Medicine

Background:

• Drug safety assessment is crucial in clinical trials for ethical reasons.
• Manual review of extensive data is common but complex.
• Existing quantitative methods offer error control and clinical knowledge integration.

Purpose of the Study:

• To propose a novel two-layer framework, SAFE, for robust drug safety assessment in clinical trials.
• To emphasize the integration of substantial evidence, error control, and clinical relevance.
• To enhance the reliability of safety conclusions drawn from clinical trial data.

Main Methods:

• The proposed framework, SAFE (Synergy Area with FDR-controlled Evaluation), has two layers.
• The first layer identifies clinically meaningful Synergy Areas (SAs) using compelling evidence.
• The second layer controls the False Discovery Rate (FDR) across all identified SAs.

Main Results:

• Simulation studies confirmed that SAFE effectively controls error rates within and across SAs at the nominal level.
• Application to real-world data from the Historical Trial Data (HTD) Sharing Initiative demonstrated SAFE's effectiveness.
• SAFE successfully screened out extreme data points, leading to more solid safety conclusions compared to direct methods.

Conclusions:

• The SAFE framework provides a robust approach to drug safety assessment in clinical trials.
• It integrates substantial evidence with controlled error rates for reliable conclusions.
• SAFE can serve as a foundational component or adaptable platform for advanced safety evaluation.