This week, FDA published a draft guidance on the use of registries to support regulatory decision-making. In this guidance, titled "Real-World Data: Assessing Registries to Support Regulatory Decision-Making or Drug and Biological Products," FDA provides its thinking on the types of issues that should be considered when either designing a patient registry or using an existing registry to support regulatory decision-making.
The draft guidance can be found here
FDA defines a "registry" as "an organized system that collects clinical and other data in a standardized format for a population defined by a particular disease, condition, or exposure." The establishment of a registry involves the enrollment of a pre-defined population and the collection of pre-specified health-related data for each patient in that population (i.e., patient-level data).
In the guidance, FDA acknowledges that registries have the potential to support medical product development. Examples of such uses include characterizing the natural history of a disease; providing information to help determine sample size, selection criteria, and endpoints; selecting suitable study participants; identifying biomarkers or clinical characteristics that are associated with significant clinical outcomes; and supporting certain inferences about a drug’s safety and effectiveness.
The draft guidance details FDA’s current thinking on three main topics:
- Considerations for a registry’s fitness-for-use in regulatory decision-making. Because registries collect structured and pre-determined data elements, FDA notes that they can offer advantages over other real-world data sources. A registry can offer longitudinal, curated data about a defined patient population and disease history, complications, and medical care. In addition, registries systematically collect patient-reported data – an attribute that a medical claims dataset or electronic health record may lack. On the other hand, FDA notes that registries have certain limitations that have the potential to introduce bias (i.e., the enrolled patients may not be representative of the target population of interest due to challenges related to patient recruitment and retention). Additional limitations of a registry may include issues with data heterogeneity and variation in approaches used to address data quality. FDA emphasizes that any analysis of a registry as fit for regulatory decision-making must focus on both the data’s relevance and reliability. The guidance then discusses each of those two elements in further detail.
- Considerations when linking a registry to another data source for supplemental information. Such supplemental data sources could include, for example, data from medical claims, electronic health records, digital health technologies, or even other registries. FDA recommends that a registry sponsor develop strategies to correct for redundant data, resolve any inconsistencies in the data, and to take appropriate steps to protect patient privacy as part of the data transfer. In addition, FDA recommends that sponsors consider whether the linked data sources are interoperable and take appropriate action to ensure ongoing interoperability, and to support appropriate informatics strategies to ensure data integration.
- Considerations for supporting FDA review of submissions that include registry data. The key takeaway here is brief but critical to the successful use of registry data. FDA advises, in no uncertain terms, that a sponsor interested in using registry data to support a regulatory decision meet with the Agency before conducting a study that includes registry data. In addition, FDA cautions that the sponsor should have a data-sharing agreement in place for any data owned by a third party to ensure that the patient-level data can be appropriately submitted to FDA.