The U.S. Food and Drug Administration (FDA) issued a draft guidance titled Data Standards for Drug and Biological Product Submissions Containing Real-World Data on October 21, 2021. The guidance provides the Agency’s thoughts on how sponsors can comply with the Federal Food, Drug and Cosmetic Act (FDCA) when submitting “certain” applications that contain study data derived from real-world data (RWD) sources. The FDA acknowledges that its current study data standards do not necessarily reflect a process derived from RWD sources. However, sponsors will need to convert RWD into established study data standards when submitting this information as part of a regulatory application (a process called “mapping”).

For context, study data standards are documented guidelines to help with the exchange of clinical and nonclinical study data between computer systems. They are used to provide a consistent framework for organizing study data (such as templates for datasets, standard names for variables, how to do calculations with common variables, and so on).

This is the second guidance the FDA has issued as part of its Framework for Real-World Evidence (RWE) Program, a series of guidances designed to regulate uses of RWD and RWE. More information about the RWE Framework can be found here. Details about the first draft guidance, which contains proposed guidelines on sourcing RWD from health records, can be found here.

The guidance covers two key areas:

1. A regulatory background on FDA submission requirements to help sponsors understand what guidances and standards to apply when submitting study data derived from RWD sources.
2. An outline of data standard considerations for sponsors submitting data derived from RWD sources as part of an application.

Part 1 of the Guidance: Regulatory Background

Under section 745(A) of the Food and Drug Administration Safety and Innovation Act, many types of applications must now be submitted electronically, and study data submitted electronically must follow certain standards. This includes data contained in “certain” new drug applications, abbreviated new drug applications, biological license applications, and applications for investigational new drugs. These electronic submissions must be in a format the FDA can “process, review, and archive,” as stated in the existing industry Study Data Guidance.

The FDA can “process, review, and archive” electronic submissions that follow the Study Data Standards Catalog. As detailed in the Catalog, clinical and nonclinical data are generally submitted in standards from the Clinical Data Interchange Standards Consortium (CDISC), including the Study Data Tabulation Model (for clinical data submissions) and Standard for Exchange of Nonclinical Data (for nonclinical data submissions). RWD submitted as study data in the covered application types must follow the standards specified in the Catalog.

Part 2 of the Guidance: Data Standard Considerations

The guidance also addresses five data standard considerations for sponsors submitting study data from RWD sources as part of an application:

1. 1. Challenges in RWD standardization
   2. Documentation of processes for managing RWD
   3. Considerations for conforming RWD to existing FDA study data standards
   4. Considerations for mapping RWD into a recognized FDA standard;
   5. Considerations for data transformations

First, the FDA acknowledges the challenges in standardizing data derived from RWD sources. These include inconsistent formats between sources of RWD, differences in the standards used to make data captures, and inconsistent vocabularies and coding systems between health care databases. Sponsors should be aware of these challenges and be prepared to address them.

Second, the FDA advises that sponsors should have “adequate processes” in place to increase confidence in RWD-sourced data during data curation and transformation, and they should document these processes electronically in their submissions. Sponsors should also document how data are changed to conform to current standards and the potential impact of those changes.

Third, FDA recognizes that current data standards can be applied to RDW sources using a variety of approaches. Sponsors should refer to the Study Data Technical Conformance Guide when submitting study data. From the early stages, sponsors should discuss with the FDA any planned submissions of study data derived from RWD sources and the approach for transforming the data to the current FDA standard formats. These approaches should be described in the protocol or final study.

Fourth, when mapping data into submission standard formats, sponsors must show their work. This includes documenting their rationale for choosing particular CDISC data elements for RWD sources and describing their approach and the impact of data mapping. Sponsors should also include a data dictionary that defines every data element and provides relevant information about the elements in their protocol or final reports.

Fifth, sponsors should document data challenges encountered while transforming the RWD data into a submission standard format.

Finally, the FDA offers several examples of how to map health care data to the formats specified in the CDISC and Study Data Tabulation Model (SDTM) Standards.

Conclusion

The FDA has invited the public to submit comments though January 21, 2021.

Faegre Drinker’s life sciences industry group continues to monitor RWD and RWE guidelines and will provide updates as they become available.