Real-World Evidence in Regulatory Decision Making: Time for Evidence Integration

Abstract

Over the past decade, consideration of real-world evidence (RWE) in regulatory settings has come into sharper focus. The European Medicines Agency (EMA) and the United States (US) Food and Drug Administration (FDA) have chosen different paths to explore integrating RWE into decision-making processes.^{1, 2} In this issue of Clinical Pharmacology & Therapeutics (CPT), Prilla et al.¹ evaluate the EMA's experience during the second stage (2021–2023) of an RWE pilot program that began in 2019.

The EMA employs a three-pronged approach to RWE generation: internal research, the DARWIN EU® network, and external pathways. Prilla et al.¹ demonstrate that research topics, typically requested during or in anticipation of regulatory procedures, are topical, practical, and support the scope of regulators and downstream decision-makers. This multifaceted RWE generation strategy using a flexible and comprehensive framework could serve as a model for other regulatory bodies.

The 61 RWE requests during the study time period offer several proof points for integrating RWE into evidence assessment. The RWE requests that were delivered offered timely insights from rapidly completed studies, enhanced understanding of clinical issues from diverse data sources, and informed plans for addressing unanswered questions. Nearly a quarter of the topics evaluated related to children, a group typically underrepresented in clinical trials even for conditions with a high pediatric burden.^{1, 3-5} Almost half concerned rare diseases or orphan drugs, areas often lacking robust clinical trial data and detailed literature.^{6, 7} The diversity of therapeutic areas covered, from anti-infectives to antineoplastic agents, underscores the potential broad applicability of RWE across disease areas. These RWE request pattern reflects global unmet regulatory needs that various policies have sought to address.

The prominence of requests related to age subgroups, pediatric investigational plans (PIP), and post-authorization study design suggests RWE's utility in compiling evidence for other historically underrepresented populations. In the United States, the FDA's Diversity Action Plan (DAP) requirement, updated in July 2024, presents opportunities for RWE to complement enrollment efforts at several junctures: creation of DAPs for registration enabling trials, evidentiary gaps at the time of regulatory procedures, and post-marketing requirements.^{8, 9} A similar assessment in the United States of regulatory questions addressed with RWE, regardless of approval outcome, would be revealing.

The EMA's experience also outlines current limits of RWE in regulatory settings. Over a third of requests that completed feasibility assessment were unable to proceed, primarily due to lack of fit-for-purpose data.¹ The EMA's ongoing expansion of RWE capacity, including access to datasets, and diversification of RWE generation pathways will broaden and deepen the research questions addressed. Continued transparency about topic feasibility, time lines, and scientific rigor will help refine use cases.

The optimal balance between RWE “supply” and regulator “demand” remains unclear. The EMA's approach seeks to balance efficiency and comprehensive evidence generation. From a process perspective, the value of the additional information from RWE depends on its impact on the outcome of a decision point. In regulatory settings in which clinical trial data remain the gold standard for effectiveness, RWE will likely have the biggest impact on decision confidence and, consequently, requirements for additional research, if any. Therefore, the benefit of spending time and resources RWE generation prior to or in regulatory settings is best considered in context of downstream effects.

Integrating RWE into regulatory decision making faces common observational study challenges: (i) fit-for-purpose data, which requires assessment of data quality characteristics such as reliability (completeness and accuracy) and relevance, and (ii) robust methodologies to mitigate bias and confounding.^{10, 11} Existing and forthcoming guidelines from regulatory, health technology assessment (HTA), and other organizations across the world aim to ensure meaningful, relevant, and valid RWE.¹⁰

Key steps to realize RWE's potential in regulatory decision making at a global scale include the following: (i) harmonize data standards and RWE methodologies across regulatory, HTA, and payer contexts and geographies to enhance the comparability and transportability of RWE findings; (ii) foster close collaboration between stakeholders (regulatory agencies, HTAs and payers, pharmaceutical companies, and healthcare providers, advocates, and patients) through continuous dialogue and knowledge sharing to find points of alignment; and (iii) continued investment in robust data infrastructure and advanced analytics capabilities, including artificial intelligence and machine learning, to facilitate efficiency and address most critical knowledge gaps (Figure 1).

CPT further explores this evidence-integration paradigm shift in the upcoming scheduled for publication in April 2025. Special Issue “Bench to Budget: Streamlining the Full Spectrum of Evidence Integration for Drug-Development and Patient Access.” This issue will examine current opportunities and challenges at the intersection of drug development, population health, and economic evaluation—a nexus at which RWE plays a pivotal role. Current and potential use cases range from drug discovery and development to pharmacokinetics–pharmacodynamics and preclinical studies, and from clinical evaluation of effectiveness and safety to regulatory review and to value demonstration to physicians, patients, and payers who cover the cost of therapeutics. The work by Prilla et al. serves as both a proof of concept and a call to action to move beyond traditional evidence boundaries. The integration of all types of evidence, including RWE, generative AI, and the many ‘omics, is critical for efficient, personalized, and patient-centered equitable healthcare.

No funding was received for this work.

RAM reports prior employment by Flatiron Health and Roche stock and current employment by Color Health.