For years, randomized clinical trials (RCTs) have been the gold standard for drug approvals, but they often fall short of reflecting real-world clinical practice. Real-World Evidence (RWE), data generated during routine clinical use, is expected to address this deficiency. While the US Food and Drug Administration (FDA) has been actively incorporating RWE, a recent study sheds light on the European Medicines Agency’s (EMA) more cautious approach, revealing both its current limitations and future promise.
A recent cross-sectional study analyzed publicly available EMA assessment reports from 2020 to 2023 to understand how RWE is being integrated into initial drug approval decisions. The findings solidified the modest role of RWE in Europe: out of 423 identified reports, only 39 (9.2%) incorporated real-world evidence. This is a notably lower rate compared to the FDA, where 31% of approvals from 2019 to 2021 incorporated similar data. However, the use of RWE at the EMA did not show a clear or steady upward trajectory during the study period. Of the 48 RWE studies included in the final data analysis, the largest portion was found in 2021, accounting for 45.8% (22 studies). In contrast, 2020 and 2022 were roughly equal, each representing 22.9% (11 studies), while 2023 saw a significant dip to just 8.3% (4 studies), suggesting a current phase of fluctuating, rather than accelerating, adoption.
The analysis also highlighted specific areas where RWE is most commonly accepted. The dominant therapeutic area for RWE studies was oncology, which accounted for 37.5% (18 out of 48 studies). This concentration is likely due to the need for external control arms and natural history data in rare cancers, where large-scale RCTs are often impractical. Following oncology, the next most frequent areas were COVID-19 at 14.6% (7 studies) and neurology at 12.5% (6 studies). The primary data sources for these studies included registries (45.8%) and electronic health records (33.3%). The majority of the studies employed a robust cohort design (87.5%) and were primarily used to evaluate efficacy/effectiveness (70.1%) and safety (37.5%). Despite the clear use cases, explicit regulatory endorsement for the RWE studies remained limited, appearing in only 29.2% of the studies.
The study authors highlighted several key challenges that could be dampening the incentive for applicants to submit RWE. A major issue is the lack of uniformity and transparency in how RWE is defined, presented, and assessed across different EMA committees. Assessment reports often failed to explain precisely how RWE affected the final benefit-risk conclusions, and many suffered from vague study objectives and an interchangeable use of terms like efficacy and effectiveness. This limited transparency may leave applicants uncertain about regulatory expectations. To address these inconsistencies, the authors suggest that the EMA consider adopting a standardized submission template, similar to the Harmonized Protocol Template endorsed by the FDA and Medicare, to improve clarity, reproducibility, and confidence in RWE submissions.
Despite the current modest and uneven adoption, the authors anticipate an upward trend in the formalized regulatory use of RWE in the coming years. This optimism is fueled by building momentum across Europe to integrate high-quality RWE, including initiatives like the EU Data Analysis and Real World Interrogation Network (DARWIN EU). Furthermore, recent legislative changes in countries like Germany and France are expanding access to national health care databases, which will increase the availability of routine clinical data for high-quality research. The study serves as a critical snapshot, revealing that while RWE is a compelling concept, its practical and standardized integration into the EMA’s initial drug approval process is still a work in progress.
The EMA’s current use of Real-World Evidence in drug approvals remains limited and fluctuating, yet it is highly concentrated in specialized fields like oncology. Sponsors looking for EMA approval of their products based on RWE must incorporate standardized reporting of such data in their applications.
