Framework for the application of Pharmacoepidemiology and Real-World Data analysis in Safety Signal Assessments

Background: When a signal arises from a clinical trial, post marketing study, or other source suggesting a new potential causal association between a medicine or vaccine and a safety event, a safety signal assessment is conducted to determine whether the evidence supports a new or changed risk. Large protocol-based studies may not be feasible within the regulatory timeframe. However, pharmacoepidemiologists are uniquely placed to contextualize the outcome of interest (i.e., signal) and conduct rapid real world data analyses. Guidance is needed for epidemiologists and cross-functional safety stakeholders to better understand opportunities for epidemiological contributions to signal assessments.

Objectives: To develop a framework to guide pharmacoepidemiologists and stakeholders when conducting safety signal assessments.

Methods: A two-panel team of epidemiologists across various therapeutic areas with experience in safety signal assessments were engaged to develop a framework based on best practices. One panel developed the framework by identifying key areas for epidemiological contribution, outlining relevant aspects and considerations. The second panel reviewed the framework, identifying areas of refinement or added considerations. Discrepancies were resolved through consensus agreement.

Results: The framework outlines five areas of epidemiological contributions to signal assessments defined by use of existing evidence (points 1 to 3) and generation of new data (points 4 and 5). These areas include: 1) describing the characteristics of the exposed and indicated populations that are relevant to the signal being evaluated; 2) comparison of the baseline incidence or prevalence of the outcome of interest in relevant populations, including the general, indicated, and unexposed indicated populations; 3) identifying the incidence of the outcome of interest in the exposed population and any association between the outcome and new use of the medicine or vaccine, and if needed; 4) assessing the feasibility, robustness, and reliability of real world data to use as part of the signal assessment; and 5) conducting rapid real world data analysis to inform the above areas where uncertainties remain. Key considerations throughout the process are related to the availability and robustness of published studies, lifecycle of the product, and reliability and timeliness of real-world data sources.

Conclusions: In the signal assessment process, pharmacoepidemiologists are key stakeholders with expertise to provide critical insights and data that comprehensively contextualizes the signal under evaluation with existing literature and rapid data analyses. These epidemiological contributions are essential for decision making related to the safety signal.