Despite broad enthusiasm for wider use of real world data (RWD) and real world evidence (RWE) in developing and assessing the efficacy and safety of new drugs and medical products, this approach presents significant challenges in assuring that the information is valid and germane to the research involved.
FDA has worked to clarify how drug development can gain from tapping into information in health care systems and claims databases, as outlined in our Oct. 1 posting on “advancing reliance on real world data and evidence” for these purposes. That report discusses the agency’s September 2021 guidance, which addresses a range of issues involved in sponsor utilization of RWD and RWE in clinical research.
Now, FDA seeks to address the added complexities for incorporating such information into submissions to the agency, while addressing concerns about its relevance and reliability. Such challenges are evident in a more recent FDA draft guidance on how sponsors should submit RWD to support research submissions to the agency for drugs and biologics, which acknowledges numerous difficulties that arise from utilizing data from a wide range of information systems with different formats, terminologies, data aggregation methods and overall quality measures.
These issues reflect FDA efforts over the past decade to require electronic submission of data to support market applications based on agreed-on standards. The aim is to make it easier and faster for the agency to receive, review, store and analyze the wide range of research, manufacturing and safety information submitted to in applications. Now, to further support increased use of RWD and RWE in applications, FDA outlines here how sponsors should collect and convert or “map” such data collected outside of clinical trials to fit agency e-standards for submitting information in investigational new drug applications (INDs) and applications for generic drugs (ANDAs) and biotech therapies (BLAs), and new drugs (NDAs.)
The guidance addresses how sponsors should format data elements so that they can be exchanged between computer systems in ways to minimize inconsistencies and increase clarity to facilitate FDA review and archiving of the information. The agency recognizes the importance of sponsors being able to utilize data from electronic health records, medical claims systems, disease registries and mobile digital tools, among other sources. Problems arise, though, because such information systems have diverse standards and policies, such as differing diagnosis codes set by national and international data standardization entities. FDA and sponsors acknowledge considerable challenges in mapping data from source systems to fit FDA standards, noting added difficulties raised by differences in definitions and in key variables, even basics such as sex and diagnosis.
For now, FDA advises sponsors to document data sources and conversion processes and decisions, acknowledging that the field is in flux and faces many adjustments and challenges ahead. Collaborative industry efforts are working to develop data standards and mapping tools for electronic health records, but all parties concede that this will be a lengthy and difficult process. FDA acknowledges that its advice is limited by current uncertainties and that sponsors should anticipate changes in policies for collecting and submitting RWD to support applications.
Some assistance may arise from multiple efforts to advance standards for this field. The Professional Society for Health Economics and Outcomes Research (ISPOR) is launching a real-world evidence registry to provide added transparency into the analysis and reporting of RWE in health research, here through a platform for researchers to register study protocols so other experts may assess the potential for biases, sufficient randomization, and uniform data quality.
Additionally, FDA looks to enhance its use of RWE for more rapid assessment of therapies to treat COVID-19 through its partnership with health tech firm Aetion. The agency is expanding the use of Aetion’s platform-based approach for utilizing data from EHRs and other sources to define key research questions, identify fit data sources, and validate ways to capture data on subgroups outside clinical trials. This builds on projects with Aetion for assessing data fitness for use, identifying good practices in working with RWD, and advancing rapid-cycle analytics to address critical health questions.