FDA’s New RWE Guidance Provides Recommendations for Sponsors Conducting Non-Interventional Real-World Studies and Describes the Potential to Use RWE for Initial Approvals in One Limited Circumstance

September 11, 2023By Mark A. Tobolowsky

CDER, CBER, and the Oncology Center of Excellence recently published a final guidance document titled “Considerations for the Use of Real-World Data and Real-World Evidence to Support Regulatory Decision-Making for Drug and Biological Products” as another part of its real-world evidence (“RWE”) Program.  This finalized a draft guidance issued in December 2021, largely similar in substance.

The focus of both documents is on non-interventional studies using real-world data (“RWD”).  Interventional studies are generally clinical investigations under the regulations and thus require an IND under 21 C.F.R. § 312.  In contrast, as the guidance explains,

a non-interventional study (also referred to as an observational study) is a type of study in which patients received the marketed drug of interest during routine medical practice and are not assigned to an intervention according to a protocol. Examples of non-interventional study designs include, but are not limited to, (1) observational cohort studies, in which patients are identified as belonging to a study group according to the drug or drugs received or not received during routine medical practice, and subsequent biomedical or health outcomes are identified and (2) case-control studies, in which patients are identified as belonging to a study group based on having or not having a health-related biomedical or behavioral outcome, and antecedent treatments received are identified.

Non-interventional studies thus do not require an IND.  Certain additional activities, such as questionnaires or laboratory tests, do not transform non-interventional studies to interventional.  In all cases, applicable requirements regarding informed consent (21 CFR Part 50) and Institutional Review Board review (21 CFR Part 56) must be met.

Even though an IND is not required, this guidance is intended to provide recommendations for sponsors to follow for using RWE from such studies for regulatory decision-making.  For example, as the guidance states, “FDA must be confident . . . that particular data sources or databases were not selected, or that specific analyses were not conducted, to favor a certain conclusion.”  The guidance describes steps sponsors should take to evaluate the study’s feasibility for such purposes.  It also encourages sponsors to post study protocols on a public website such as ClinicalTrials.gov.

The guidance also discusses ensuring that FDA has necessary access to the RWD, which requires sponsors ensuring that they are able to submit patient-level data for any clinical study included in a marketing application.  If certain RWD are owned and controlled by other entities, sponsors should have agreements with those entities to ensure such patient-level data can be provided to FDA and source data necessary to verify the RWD are made available for inspection as applicable.  The final guidance contains a new addition that states that if there is an appropriate justification as to why a sponsor cannot submit patient-level data to FDA through traditional channels, third parties can provide these data through either a pre-IND or Type V Drug Master File (“DMF”) and provide the sponsor with a letter of authorization for FDA to reference the data.  This appears to represent a measure of consideration for the difficulties sponsors can sometimes have gaining access to such data, which we discussed previously here.

Postmarketing safety reporting obligations apply, but, if the sponsor is using only a subset of a larger dataset to conduct its analysis, FDA does not expect the sponsor to search the entire dataset regarding all uses of the product for adverse events that would meet reporting requirements.  However, if the sponsor does become aware of adverse events that are subject to postmarketing reporting requirements during the course of the non-interventional study, such events must be reported in accordance with applicable requirements.

The guidance also describes study monitoring and other oversight responsibilities the sponsor should address, such as ensuring that there are adequate study records and that the study is conducted in accordance with the final protocol and SAP.

An interesting addition in the final guidance is that it explains that the topics discussed “apply to any type of RWD, including data on products used in clinical practice under an emergency use authorization (EUA).”  For obvious reasons, EUAs have been more prominent in recent years than they had been previously.  Even though the draft guidance was issued in 2021 after the onset of the COVID-19 pandemic, the Agency’s thinking now appears to be more reflective of this recent experience.

In a footnote, the guidance notes that RWD based on the use of products authorized under EUAs “can be used to generate RWE about the safety and/or effectiveness of that product.”  FDA has typically only discussed the use of RWE to support approval of a new indication for an already approved drug or to help support postapproval study requirements.  This is for obvious reasons and is also the statutory mandate under the 21st Century Cures Act.  However, the inclusion of products authorized under an EUA would appear to mean FDA, at least in theory, is willing to accept RWE to support initial approval of products, because such products are not “approved.”  EUAs can, of course, be issued for new uses of approved products, but they do not have to be.  Does this offer the potential to use RWE for an initial approval?

The COVID-19 vaccines were initially authorized under EUAs, but their approvals do not appear to have been based on RWE.  For example, the Clinical Review Memo for Comirnaty states: “Post-authorization effectiveness data from observational studies . . . are limited to published literature and were not submitted as part of the licensure application. Therefore, FDA has not independently reviewed and confirmed the data or assessed the study designs for potential sources of bias.”  Similarly, the Integrated Review for Paxlovid states:

Since PAXLOVID was authorized for emergency use in December 2021, FDA has monitored the published literature on real-world evidence (RWE) studies that evaluated PAXLOVID effectiveness in outpatient COVID-19 populations. Most of the data sources used in these published RWE studies had insufficient longitudinal data and/or inappropriate study design to account for potential bias. Among the identified studies, five are based on appropriate source data and implemented design features that can account for the potential bias . . . . However, while the source data and certain design elements of these cohort studies were appropriate, there were insufficient details on the data source, methods, or analytical approach for a complete review to determine the quality of the results of the studies.

Although RWE was not used for these approvals, it does appear that FDA was open to such an approach if the data were fit-for-purpose.  Will it ever happen?  We will have to see – perhaps, with the help of this guidance, it will!