The third annual report on the integration of real-world evidence (RWE) into regulatory decision-making, published in July 2025 by the European Medicines Agency (EMA) and the Heads of Medicines Agencies (HMA), discusses the experience gained from regulator-led studies conducted between February 2024 and February 2025. The report addresses all the three pathways for generating RWE coordinated by the EMA, namely the European consortium DARWIN EU, the framework contract (FWC), and in-house EMA studies. These initiatives also aligns with the European Medicines Regulatory Network (EMRN) strategy to 2028, which includes several lines of action where RWE may turn important.
A marked growth in RWE studies
Data from the most recent monitoring period indicates a significant increase in the use of RWE to evaluate the safety and efficacy of medicines, with studies rising by 47.5% compared to the previous period. A total of 107 research topics were assessed, 74% of which were new topics identified during the reporting period (+32%).
A total of 59 studies were conducted during the reference period (33 of which completed and 26 ongoing). The proportion of research projects deemed feasible increased from 60% to 78%, primarily due to the expansion of the DARWIN EU network. This European digital platform for health data analysis has expanded to include 30 data partners during the reporting period, and now covers approx. 180 million patients in 16 European countries. This expansion has established DARWIN EU as the main channel for studies conducted by the EMA, with a total of 89 research topics.
A key finding emerging from the report is the speed of execution of the DARWIN EU studies: the median duration from protocol approval to final results was only 4 months. This significant reduction in time compared to traditional RWE approaches enables a more timely integration into regulatory processes. This trend was driven by the use of pre-approved protocols for standardised analyses and simplified ethical approval processes. As the report highlights, some data partners have obtained “general approvals” for the execution of standardised analyses, reducing the time required to access data from 2-3 months to 0.5-1 month.
Conversely, the in-house pathway decreased compared to the previous period (16 research topics), while the FWC pathway increased slightly (9 research topics).
A significant increase in pharmacovigilance studies
The studies considered in the third report on the use of RWE evidences covered a wide range of topics, including drug use (42%), safety (24%) and disease epidemiology (24%). Of the 14 therapeutic areas investigated in total, RWE studies focused particularly on anti-infectives (21%), antineoplastic and immunomodulating agents (18%), and nervous system treatments (16%).
Research requests came from various regulatory and public decision-making bodies, including EMA scientific committees, Working Parties and internal functions, the European Centre for Disease Prevention and Control (ECDC), HTA bodies and the European Commission.
Of particular note was the significant increase in requests from the Pharmacovigilance Risk Assessment Committee (PRAC), with 19 studies conducted to support signal or Periodic Safety Update Single Assessment (PSUSA), or to assess the impact of risk minimisation measures.
The Committee for Human Medicinal Products (CHMP) requested a total of five studies, four of which aimed to support the geriatric medicines strategy, and one of which linked to an initial application for marketing authorisation. The Paediatric Committee (PDCO) was source of the highest percentage of unfeasible requests, confirming the results of the previous report.
In the field of medicines shortages, seven studies were intended to support the activities of the Medicines Shortages Steering Group (MSSG) and the Medicine Shortages Single Point of Contact (SPOC) Working Party. The monitored therapeutic classes included antibiotics, ICU-administered medicines, ADHD medicines, GLP-1 receptor agonists, salbutamol and therapeutic alternative inhalation products. Two studies were also initiated in oncology, in collaboration with HTA bodies and payer organisations.
Evaluate the value of RWE
The report mentions the outcomes of the survey run among the requesters of RWE, in order to assess the impact of the studies on decision-making. Twelve studies were considered supportive, and four provided substantial evidence. The study on the natural history of coagulopathy in patients with Covid-19 and in individuals vaccinated against SARS-CoV-2 in the context of the Omicron variant was included in the HealthData@EU report, which is intended to inform the implementation of the European Health Data Space (EHDS).
The other three studies provided substantial evidence for regulatory assessment. The first one, requested by PRAC, used a self-controlled case series and an active comparator cohort study to investigate the risk of suicidality associated with use of doxycycline. The study did not confirm the association in individuals with acne. Consequently, the safety signal was closed.
The study on shortages of glucagon-like peptide-1 receptor agonists (GLP-1 RAs), which are used to treat diabetes and for weight management in obese individuals, was requested by MSSG/SPOC. Conducted by DARWIN EU, the research focused on the prescription patterns of this class of medicines, providing key information for monitoring and preventing potential shortages. According to the report, this study could be repeated in future to monitor this widely used class of therapeutics.
The Vaccine Monitoring Platform (VMP)’s request to evaluate the safety and effectiveness of MVA-BN vaccination in at-risk individuals led to two studies being run in Germany and the US using the FWC pathway. Both studies confirmed the positive vaccine’s safety profile and supported the combination of primary data collection and the secondary use of data to inform responses to and preparedness for future public health emergencies.
Challenges and perspectives
According to the report, the use of RWE is now fully operational and its values is being consolidated across the entire range of regulatory use cases. The final wave of data partners should focus more on specific clinical areas, such as oncology, metabolic and cardiovascular diseases, as well as relevant populations (i.e.paediatrics). New sources of data integrated into the network include the National Neonatal Research Datasource and two cancer registries.
The new HMA-EMA Catalogues of data sources and real-world studies have replaced the previous databases and now include 247 data sources, 3,074 studies, 785 institutions and 180 networks registered across the various catalogues. Transparency has also been addressed by standardising and aligning data source descriptions with the Data Quality Framework (DQF), including data quality metrics. According to the report, a section justifying the selection of data source should be included in protocols and reports.
In terms of training, the Real-World Academy series of training events continues to be held with the aim of increasing the collective understanding and confidence of members of the European regulatory network in RWE. The Big Data Steering Group’s data science and pharmacoepidemiology curricula also provide access to educational material and tools designed specifically for regulatory decision makers. Meanwhile, RWE-related multi-stakeholder workshops have been organised to improve knowledge among external stakeholders.
The report also highlights the opportunity to maintain regular interactions with EMA’s committees, the Scientific Advice Working Party (SAWP) and the Coordination Group for Mutual Recognition and Decentralised Procedures – Human (CMDh), in order to better understand their research needs. The creation of the special interest area (SIA) dedicated to RWD aims to facilitate knowledge sharing.
