(CSPOT30) Identifying regulatory outcomes of Multi-Database and non- Multi-Database non-interventional Post Authorisation Safety Studies using publicly available information

Background: Post Authorisation Safety Studies (PASS) generate evidence that can support regulators to evaluate the benefit-risk profile of medicinal products and to take regulatory actions when needed. The European Medicines Agency (EMA) promoted decision-making transparency by publishing regulatory guidelines and endorsing the registration of PASS in the EUPAS Register (currently HMA-EMA Catalogues of real-world data sources and studies). Multi-database (MD) studies were shown to be increasingly registered in the EUPAS Register.

Objectives: To identify the regulatory outcomes of PASS using publicly available information from the EMA website and possible differences between MD and non-MD PASS.

Methods: MD PASS, and an equal number of non-MD PASS were selected from the EUPAS Register if: completed between 07/2012 and 12/2020, part of the Risk Management Plan (RMP) category 1,2 or 3, required by an EU/UK regulator, and conducted in ≥1 European country.
Regulatory outcome information for the included PASS was retrieved from the EMA website following a stepwise approach: (i) using EUPAS Register to obtain study identifiers (study title, acronym, protocol or regulatory number) to locate PASS in the following steps; (ii) consulting the PRAC Meeting Minutes to obtain regulatory outcomes and/or regulatory numbers for the next step; (iii) referring to the Procedural Steps document of the European Public Assessment Report (EPAR) or other publicly available EMA webpages to obtain regulatory outcomes.

Results: A total of 84 PASS (42 MD and 42 non-MD) were included in the analysis. We identified 65 of the 84 (77%) PASS in at least one of the used sources but the information available was scarce.
Information on regulatory outcomes was identified for only 39 of the 84 (46%) PASS (i.e. 60% of the 65 PASS located in at least one of the sources) with no different trends between MD and non-MD PASS.
Regulatory outcome information was more frequently available for imposed PASS (78% of the RMP Category 1 and 2) than for non-imposed PASS (34% of RMP Category 3).
The availability of regulatory outcomes information increased from 25% for PASS finished in 2012-2014 to 53% for those completed in 2018-2020.
On average a PASS had two regulatory outcomes. Most of the outcomes was considered unspecified, limiting our description of regulatory outcome types.

Conclusions: Information on regulatory outcomes were found for only 46% of the included PASS using publicly available information from the EMA website, with no different trends between MD and non-MD PASS.
Obtaining information on the regulatory outcomes was hindered by difficulties in finding PASS information across the public sources, and their outcomes, due to, lack of information and standardization, insufficient level of detail and inconsistency across sources.
We advocate for an increased transparency of the regulatory outcomes of PASS by (i) using a unique PASS identifier across sources, (ii) publishing regulatory outcome information for all PASS, and (iii) standardising the presentation of outcomes and providing sufficient details to understand the regulatory impact of PASS.