BLOG – Success with use of real world data in medicines research: Joining the dots of the European ecosystem

A recent survey in England indicated that, during the COVID-19 pandemic, the public had learned more about how real world data (RWD) can be used both for monitoring public health and for research.  Words such as epidemiology and vaccine effectiveness now trip off the tongue and are no longer the preserve of scientists.   From society’s perspective, real world evidence (RWE) has arrived.

However for the pharmaceutical industry RWE (clinical evidence about diseases and health interventions derived by scientific analysis of RWD) is not the new kid on the block.  Nevertheless the exciting growth of new data sources and complex analytical methodology (including artificial intelligence) means that its prominence in decision making of stakeholders is on the rise.  A recent EMA report indicates that RWE was included in about a third of 63 innovative marketing authorisation applications (pre-authorisation), received in 2018 and 2019, with an exclusively supportive role in the benefit-risk assessment in three quarters of cases.  Moreover, many of the research questions that can be answered using RWD cannot be addressed by other means.

Aside from the more established use in addressing post-marketing safety considerations, RWE can complement clinical trials in supporting regulatory authority decisions about medicines, throughout the product lifecycle.  Examples such as BLINCYTO (blinatumomab for acute lymphoblastic leukemia (ALL)) include use of RWE as a clinical trial control arm (especially with rare diseases or oncology treatments where use of a placebo arm may be infeasible or unethical) or demonstrating orphan designation prevalence.

RWE also plays a key role in facilitating patient access to medicines.  For example RWE can support health technology assessments (such as for managed entry agreements or other novel payment models) as well as payer decisions (such as inclusion of medicines in formularies).  In this way, RWE can feed into a learning healthcare system that enables healthcare decisions to be refined on an ongoing and iterative manner, as more relevant data become available.  This is alluded to such as in the recent Learning Healthcare Project report.

Aside from inclusion in regulatory authority or other submissions, RWD can also facilitate medicines development e.g. by helping to understand the natural course of diseases, identifying existing standards of care for diseases and refining clinical trial inclusion and exclusion criteria.

An abundance of datasets is available via electronic health records (EHRs), patient and population registries, genomics databases, biobanks and much more.  With almost 80% of the global population estimated last year to have a smartphone and global end-user spending on wearable devices set to reach more than 90 billion dollars by 2022, there is plenty of scope for sharing patient generated data that might even be collected on a 24/7 basis .  This is done while respecting the regulatory, legal and privacy compliance framework within which the data projects are to be performed, safeguarding the rights of data subjects in the first place.  This opens the door to establishing real world new clinical endpoints of greater relevance and appeal to patients, that in turn can be used to evaluate efficacy and effectiveness of new medicines e.g. for Duchenne Muscular Dystrophy.

RWE studies have serious limitations but by thoughtful considerations of these in the design and analysis stage, they can provide valuable and meaningful insights.  The nature of retrospective analysis of RWD captured during routine clinical care is quite different from the notion of prospective data collection in clinical trials.  Evolving recommended practices for data quality and evidence validity in regulatory RWE should help reconcile these differences.  The use of standard outcome measures is also important.  All of this underpins acceptance of RWE, by decision makers, so a common understanding, amongst stakeholders, of what is fit for purpose is essential.

Demonstration projects, workshops and shared learning from use cases can be vehicles for advancing development of regulatory guidelines such as the European Medicines Agency’s (EMA) recently published registry-based studies guideline.  This can also facilitate development of best practices in analytical methodology that tackle thorny issues such as mitigating bias (via propensity scoring, etc.,) and more.  Naturally, international harmonisation is the Mount Everest of regulatory acceptability.  The proposal to develop new ICH guidance on planning and designing pharmacoepidemiology safety studies using RWE is a  welcome initiative alongside guidelines from other regulatory authorities such as the FDA and the International Coalition of Medicines Regulatory Authorities’ (ICMRA) successes in collaborating on RWE and observational studies during the COVID-19 pandemic.

In addition to supporting healthcare delivery, the introduction of an interoperable European Health Data Space will provide a more robust framework for using RWD for secondary research and to inform health policy making.  However clarity is needed as to whether entry will be open to  industry partners.  Data use  may be marred by inconsistencies in the application of the General Data Protection Regulation unless these are resolved.  With its ongoing programme of mapping millions of patient records in Europe alone, using the OMOP common data model, IMI EHDEN is itself on course to building a federated network of data sources, at scale, which is intended to facilitate observational health research in Europe.

The EHDS will also be home, in future, to DARWIN EU, the European Medicine’s Agency’s own nascent federated Data Analysis and Real World Interrogation Network.  This will be used to generate RWE relating to diseases, populations as well as use and performance of medicines.  This will inform regulatory decision making with stakeholders such as Health Technology Assessment bodies also able to  take advantage of this resource.  The development of DARWIN EU is part of a broader package of deliverables that are intended to implement the vision of the European Medicines Regulatory Network to increase the use of big data (including RWD) thus improving the regulation of medicines.

When all is said and done, the engine room that drives the success of this overall approach to harnessing the hidden value of data and supporting continued investment in research, development and innovation within Europe, is the willingness of individuals to share their data.  This is contingent upon engendering trust and credibility in how those data are collected, stored, protected and analysed with good governance and suitable transparency being prerequisites; something that is apposite in the COVID-19 world which we currently inhabit.

Karin Van Baelen
Head Global Regulatory Affairs, Janssen, the Pharmaceutical Companies of Johnson & Johnson
Chair EFPIA Integrated Evidence Generation and Use Working Group