On April 21, 2015, the Food and Drug Administration (FDA) released a draft guidance document, Acceptance of Medical Device Clinical Data from Studies Conducted Outside the United States, discussing considerations related to the acceptance of foreign clinical data in support of investigational device exemptions (IDEs) and marketing applications for medical devices. While the guidance is in draft and therefore not yet in effect, FDA clarifies that the document reflects FDA’s existing approach on this topic, rather than announcing new policy.
There is longstanding precedent for FDA acceptance of foreign data. For example, in 1986, FDA issued regulations (21 C.F.R. §814.15) specifying the situations under which FDA will accept foreign data in a premarket approval application (PMA), generally if the data are valid and certain human subject protections are in place. FDA regulations require that the Agency have “valid scientific evidence” to support decisions on 510(k)s, PMAs and de novo petitions (21 CFR 860.7). These standards apply to both U.S. and foreign data. FDA’s existing regulation also specifies the situations under which foreign data alone may be used to support a PMA, that is, the data are applicable to the U.S. population, the investigators are competent, and the data are considered valid or can be validated.
Due to what some in industry have perceived as FDA’s reluctance to accept foreign data, the Food and Drug Administration Safety and Innovation Act (FDASIA), passed by Congress in 2012, added a new provision requiring FDA to accept data from foreign clinical studies, provided that the applicant demonstrates that the data are adequate under FDA’s standards to support clearance or approval. Under the statute FDA is also required to provide written notice of its decision.
In addition, on February 25, 2013, FDA published a proposed rule that would require compliance with Good Clinical Practices (GCPs) in accepting clinical data, including foreign data. FDA references this proposed rule in the draft guidance, noting that it would affect acceptance of foreign data once finalized.
In the draft guidance, FDA acknowledges the regulatory history described above and emphasizes the agency’s longstanding policy of accepting foreign clinical data. FDA also discusses the “challenges” associated with this area, and provides guidance regarding considerations sponsors should take into account when proposing to collect or use foreign data to support an application. Note that the guidance applies to data used to support not only IDEs and PMAs, but also 510(k) notices, humanitarian device exemptions (HDEs), and de novo requests.
As outlined in 21 C.F.R. §814.15, FDA also recommends in the draft guidance that sponsors meet with FDA as early as possible in the process if they intend to collect or rely on foreign clinical data.
The draft guidance outlines three key considerations in the acceptance of foreign clinical data:
- Differences in clinical conditions: FDA notes that differences in standards of care, clinical facilities, or levels of clinical skill can affect the relevance of the foreign data to the U.S. population, and can change the risk-benefit calculation.
- Differences in study populations: FDA explains that differences in demographic factors (race, ethnicity, age, gender, sex), prevalence of confounding factors, cultural differences, and other differences in the foreign study population compared to the U.S. population can affect the applicability of the data, the risk-benefit ratio, or the ability to pool U.S. and non-U.S. data.
- Differences in regulatory requirements: The guidance cautions that if a foreign study is conducted to meet a different regulatory standard than that used in the U.S. (for example, performance vs. effectiveness), the data may not be acceptable to support a U.S. marketing application.
Given the issues highlighted above, it is important for sponsors considering the use of clinical data from outside the U.S. to provide a discussion of these issues for the Agency and a justification for why the data are appropriate to meet the requirements outlined for the relevant submission type.
There are three different ways that foreign clinical data can typically be used in support of a marketing submission:
- as feasibility or pilot data in order to start a pilot or pivotal study in the U.S.;
- as a portion of the pivotal study data (i.e., a multicenter study conducted partially in the U.S. and partially outside the U.S.); or
- as the sole basis for approval.
While the PMA regulation (21 C.F.R. §814.15) provides specific requirements for foreign data used as the sole basis for approval, the draft guidance does not distinguish between these different uses. In our experience, FDA scrutiny for foreign data increases substantially depending on the extent to which it is used to support a marketing application. In practice, the agency has shown a strong preference for pivotal studies where at least part of the data, and preferably a majority, is collected in the U.S. In our experience, sponsors are frequently able to use foreign clinical data for some aspects of an application, but PMA approval based solely on foreign data is very infrequent.
While the guidance provides less advice regarding how a sponsor can demonstrate that foreign clinical data are applicable to a U.S. population, it does provide a number of examples of situations in which foreign data was, or was not, considered acceptable to support a marketing application. It describes certain cases where the agency worked with a sponsor to be able to use foreign data that may not have fully met the standard without conducting a new pivotal study, specifically, performing a re-analysis of the data and a small U.S. confirmatory study, or using the foreign data to design a more limited U.S. study using Bayesian statistical methodology.
In general, the guidance does not appear to signal a shift on FDA policy with regard to foreign clinical data. In our experience, sponsors are well advised to heed FDA’s advice to consult the agency as early as possible if considering collecting foreign data, to avoid potentially having to repeat a pivotal study in the U.S. If the foreign data has already been collected, the factors outlined in FDASIA, the regulations, and the draft guidance should be considered and addressed by the company in justifying use of that data.
Importantly, FDA often makes study design recommendations when consulted prior to study initiation and may also request that certain information be collected. If a company were to conduct a foreign clinical study without first discussing study design issues with FDA, and therefore did not have the benefit of such FDA feedback, there is a risk that design or data collection concerns may not be addressable after the fact. In such cases there is even greater risk that FDA may not accept the study as the basis for approval or clearance, even if the clinical conditions and study populations were the same. In our experience, engaging FDA in discussions throughout the phases of study design and data collection is critical to successful use of foreign clinical data.
The draft guidance, when final, will supplement (rather than replace) the March 2001 Guidance for Industry Acceptance of Foreign Clinical Studies, which relates primarily to the ethical considerations outlined in the Declaration of Helsinki.
The public comment period for the draft guidance will remain open until July 20, 2015, 90 days from publication in the Federal Register. Electronic comments should be submitted to http://www.regulations.govreferencing Docket No. FDA-2015-D-0975.