RWD supports regulatory decisions in Europe, but more sources are needed

A new report from the European Medicines Agency (EMA) demonstrates the utility of real-world data (RWD) studies to answer research questions and support regulatory decision making. But EMA also recommends wider access to more diverse data sources and faster generation of evidence to make RWD more useful to regulators.

The findings come as the European Union (EU) prepares to fully integrate real-world evidence (RWE) into regulatory decision making by 2025. Specifically, EMA’s “Regulatory Science to 2025” strategic reflection document, published in March 2020, outlines the use of RWE across a spectrum of use cases. (RELATED: Convergence: Global landscape evolves on real-world evidence, but at an uneven pace, Regulatory Focus 20 September 2022)

“While the majority of RWE for regulatory use is generated by pharmaceutical companies, the added value of regulator-led studies has been demonstrated over the last few years by the use of real-world data (RWD) to monitor the safety and effectiveness of COVID-19 medicines. Nevertheless, further work is needed to better integrate RWD/RWE into regulatory decision making alongside the evidence coming from more established sources (notably randomised controlled trials),” the report stated.

RWD experience

The report evaluates the experience of using regulator-led RWD studies from 1 September 2021 until 7 February 2023. During the study period, 61 RWD research opportunities were identified that could support regulatory decision making, 30 of which were initiated. Overall, 27 studies were completed and three are ongoing at the end of the study period.

Twenty-five studies were in-house studies that mostly involved databases of primary care records from European countries. Four of the studies were conducted via the Data Analysis and Real World Interrogation Network, or DARWIN EU. One study was commissioned through an EMA framework contract with a research organization.

Overall, research requests were aimed at generating evidence on drug or vaccine safety, as well as topics that could inform the design and feasibility of applicant supplies, drug utilization, clinical management, and disease epidemiology. Most of the studies related to safety, drug utilization, and disease epidemiology were deemed feasible, while research topics related to design of applicant studies or clinical management were more likely to be considered unfeasible.

The most common reason that a study would not be considered feasible was that the medicine product was not prescribed in the available databases or had not been authorized for use in the country from which data was drawn. Another common reason for a study to be considered unfeasible was that the outcome of interest, such as a condition or adverse event, was not adequately captured in the available databases.

EMA surveyed requesters and committee sponsors about the usefulness of the RWD studies and found that more than 60% of respondents said that study results were deemed supportive evidence and considered for the regulatory assessment. On average, it took 51 days to conduct an in-house study and deliver a report, while the process was longer – on average 217 days – for studies conducted via DARWIN EU. However, the report notes that DARWIN EU studies were slowed down by the onboarding of data partners during the first year of operation.

Lessons learned and recommendations

The EMA report recommends increasing the number, size, and type of additional data sources that are accessible to regulators, expanding beyond the primary care setting and a small number of countries.
Additional recommended data sources include:

Electronic health records from hospitals and specialized care settings
Biobanks
Large claims databases
Disease or patient registries, including for rare conditions
Data sources with broader geographic representation
Data sources that address currents gaps, such as information on age/birthdates or mother-child linkage

EMA also recommended strategies to potentially accelerate the generation of RWE, including further developing pipelines for “off-the-shelf” and routinely repeated studies, phenotype libraries, and automation of repeated tasks. “In addition, the generation of pre-computed, searchable dashboards may be an option to address simple study requests almost instantaneously,” according to the report.

“Further work is needed to fulfil the European Union’s vision to enable the use of RWE and establish its value across regulatory use cases by 2025. The learnings and recommendations arising from the review will feed into the work of the Big Data Steering Group and further inform the establishment of DARWIN EU,” EMA wrote in a statement.

RWE report

RWD supports regulatory decisions in Europe, but more sources are needed

Related topics

Recommended content

Welcome to the new RAPS Digital Experience