A Step Forward: FDA Issues Three Draft Guidance Documents As Part of Efforts to Modernize the 510(k) Notification Pathway

September 19, 2023
12 minutes

On September 7, 2023, FDA issued three new draft guidance documents intended to strengthen the 510(k) premarket notification pathway for medical devices. They address:

Modernizing the 510(k) program—the most common marketing authorization pathway for medical devices in the U.S.—has been an FDA policy priority in recent years. For example, FDA proposed “transformative new steps” to overhaul the 510(k) program in November 2018 and sought public comment in January 2019 regarding certain proposals to promote reliance on more modern predicate devices in 510(k) submissions for new devices. By September 2019, FDA had finalized guidance on the Safety and Performance Based Pathway, an expansion of the existing Abbreviated 510(k) pathway allowing 510(k) submission based on comparison to contemporary FDA-published performance criteria, rather than comparison to an individual predicate device. Although FDA continued to make some progress in implementing this pathway during 2020, other aspects of 510(k) reform appear to have been stalled during the COVID-19 pandemic.

With the public health emergency in the rearview mirror, it appears that 510(k) reform will re-emerge as a key priority. FDA’s publication of the latest draft guidances is a continuation of the broader modernization initiative, as FDA’s announcement explained. This Alert highlights key elements of the three draft guidance documents and what the medical device industry should keep in mind moving forward.

Predicate Device Draft Guidance

This draft guidance recommends best practices for selecting a predicate device to support a 510(k) submission. Previously, a 2014 final guidance entitled “The 510(k) Program: Evaluating Substantial Equivalence in Premarket Notifications (the “510(k) Program Guidance”) described certain considerations regarding predicate devices, including instances when a manufacturer may use multiple predicate devices or rely upon so-called “reference devices.” Then in 2018, FDA expressed concern that nearly 20 percent of 510(k) submissions were relying on predicate devices that were more than 10 years old, explaining that “older predicates might not closely reflect the modern technology embedded in new devices, or our more current understanding of device benefits and risks.” At that time, FDA proposed publishing a list of medical devices for which 510(k) submissions relied on predicate devices more than 10 years old, but did not follow through on that proposal.

In the Predicate Device Draft Guidance, FDA takes a different approach that focuses on predicate devices with certain characteristics, rather than predicate devices of a certain age. FDA identifies four specific best practices to follow when selecting a predicate device to support a 510(k) submission. Specifically, FDA recommends that, when possible, submitters select:

  1. Predicate Devices Cleared Using Well-Established Methods: FDA recommends that submitters consider how much information is available in the public 510(k) summary for the predicate device regarding the test methods used in support of 510(k) clearance and, when possible, select a predicate device that was cleared using well-established methods, such as those described in a current FDA-recognized voluntary consensus standard, an FDA guidance document, a qualified medical device development tool (“MDDT”), or a widely available and accepted method published in the public domain or scientific literature for the context of use, or that were otherwise found acceptable through the submitter’s own previous premarket submission.

    While FDA is not moving forward with proposals based expressly on the age of a predicate device, the recommendation to rely upon predicate devices cleared using well-established methods will potentially have the effect of encouraging reliance upon more modern predicates. Older devices are less likely to have been cleared using such methods or, even if they used such methods, may be less likely to have 510(k) summaries that clearly describe the methods used to support clearance.

  2. Predicate Devices That Meet or Exceed Expected Safety and Performance: FDA considers it a best practice to select a predicate device that continues to perform safely and as intended, and therefore recommends that submitters consider any reported medical device-related adverse events, malfunctions, or deaths related to the predicate device. FDA explains that new information about device safety or effectiveness may become available once a device is on the market. Such new information may include a newly recognized type of adverse event, an increase in the severity or frequency of a known adverse event, new product-product interactions, or device malfunctions. FDA recommends searching the Manufacturer and User Facility Device Experience (“MAUDE”) database, which contains medical device reports (“MDRs”), and the MedSun Reports database, which contains safety reports provided by certain clinical sites, for any events involving the predicate device. Where the only predicate devices available are those associated with reports of unexpected injury, deaths, or malfunctions, FDA recommends that the submitter describe in the 510(k) submission how the subject device at issue mitigates the concerns with the predicate device.

    The recommendation in the Predicate Device Draft Guidance to consult the MAUDE database may raise challenging questions. FDA’s MAUDE database webpage prominently disclaims the limitations of MAUDE, including that:

    • “Although MDRs are a valuable source of information, this passive surveillance system has limitations, including the potential submission of incomplete, inaccurate, untimely, unverified, or biased data.”

    • “[T]he incidence or prevalence of an event cannot be determined from this reporting system alone due to under-reporting of events, inaccuracies in reports, lack of verification that the device caused the reported event, and lack of information about frequency of device use.”

    • “MDR data alone cannot be used to establish rates of events, evaluate a change in event rates over time or compare event rates between devices. The number of reports cannot be interpreted or used in isolation to reach conclusions about the existence, severity, or frequency of problems associated with devices.”

    Given these limitations, how would FDA expect manufacturers to use MAUDE data to evaluate whether adverse events and malfunctions regarding a predicate device are “unexpected” in situations where the predicate is not the manufacturer’s own device? Additionally, how are manufacturers supposed to determine whether a known adverse event may be increasing in frequency or severity if MAUDE data is the only source of such information? Should MDR information on different devices be considered a valid source for comparing potential predicates, despite FDA’s statement that such information cannot be used to compare event rates between devices? These issues would be less challenging if there is other information substantiating that a safety issue exists with a device, but in the absence of such information, manufacturers may struggle to understand FDA’s expectations regarding the significance of MAUDE and other information from passive surveillance systems. We expect that manufacturers will be seeking further clarification or other significant revisions to this aspect of the draft guidance.

  3. Predicate Devices Without Unmitigated Use-Related or Design-Related Safety Issues: FDA recommends selecting a predicate device that does not have unmitigated use-related or design-related safety issues, taking into account any emerging safety signals or safety communications announced on relevant FDA webpages. FDA recommends selecting a predicate device that is not associated with emerging signals or safety communications that relate to unmitigated safety issues.

    Under FDA’s guidance document on emerging signals, FDA’s publication of an emerging signal “does not mean that FDA has definitively concluded that there is a causal relationship between the medical device and the emerging signal.” Nor does a notification “imply that specific actions are expected on the part of the manufacturer(s) at that time.” Despite these statements qualifying the significance of an emerging signal notification, the Predicate Device Draft Guidance appears to suggest that the mere inclusion of a device in an emerging signal notification or other FDA safety communication means that the device would not be considered a “best practice” predicate, even if the signal is still being investigated by FDA.

  4. Predicate Devices Without an Associated Design-Related Recall: FDA recommends selecting a predicate device that has not been subject to a design-related recall. Submitters are encouraged to conduct a search of FDA’s Medical Device Recalls Database to assess whether potential predicate devices have associated recalls.

The Predicate Device Draft Guidance would not establish bright-line rules for which devices can and cannot serve as predicate devices. Indeed, FDA acknowledges that there may be circumstances where a predicate device that is consistent with all the recommended best practices is not available. However, in these circumstances, the draft guidance recommends that a 510(k) submission explain how any known concerns with the chosen predicate have been mitigated with the subject device. More generally, FDA recommends that the 510(k) Summary section of all 510(k) submissions include a description of how the best practices in the draft guidance were utilized. If the draft guidance is finalized, it will be interesting to see how strictly FDA 510(k) reviewers will apply these recommendations and the level of justification that reviewers will expect in cases where all four best practices have not been perfectly followed.

Clinical Data Draft Guidance

As part of the 510(k) Program Guidance, FDA described three scenarios in which clinical data may be necessary to support a 510(k) submission and explained that FDA “currently requests clinical data for less than 10 percent of the 510(k) submissions.” The Clinical Data Draft Guidance provides additional detail regarding those previously described scenarios, as well as a fourth scenario that would likely trigger the need for clinical data in a 510(k) submission. These four most common scenarios that FDA has identified where clinical data may be necessary to demonstrate substantial equivalence are:

  1. Differences in Indications for Use: As described in the 510(k) Program Guidance, when the indications for use of a new device and predicate device differ, FDA must evaluate whether the intended use of the new device can still be considered the same as that of the predicate device. This means that, among other things, differences in the patient population, the target disease, or the anatomical site, structure, or pathology, as well as a shift from general to specific intended use could impact when additional clinical data may be necessary.

  2. Differences in Technological Characteristics: As described in the 510(k) Program Guidance, clinical data may be necessary to include in a 510(k) submission when there are differences between the technological characteristics of the new device and the predicate device that do not raise different questions of safety and effectiveness. Differences in technological characteristics might include significant changes in materials, device design, energy source, and other device features.

  3. Substantial Equivalence Cannot Be Determined by Non-Clinical Testing: As described in the 510(k) Program Guidance, clinical data may be necessary to include in a 510(k) submission when non-clinical testing, such as analytical, bench, and/or animal testing, is inadequate to establish substantial equivalence. This may be the case in instances where, among other things, there is no model (e.g., analytical, bench, animal) available; where the available models have certain limitations that do not allow for an adequate assessment; where the model may not be predictive of clinical outcomes; and where there are anatomical and/or pathophysiological species-specific questions that rely on clinical evidence.

  4. Newly Identified or Increased Risk for the Predicate Device: FDA explains that new information regarding newly identified or increased risks of the predicate device after its introduction into the marketplace may trigger a need for additional clinical data to determine substantial equivalence. The Clinical Data Draft Guidance makes clear that “[w]hen requesting clinical data . . . due to a new or increased risk, FDA intends to provide an explanation of the reason(s) for the request and why such information is necessary to determine” substantial equivalence. Reading the Clinical Data Draft Guidance in conjunction with the recommendations of the Predicate Device Draft Guidance discussed above, FDA appears to be suggesting that it may request clinical data for a 510(k) when the predicate device chosen does not follow the safety-related best practices described in the Predicate Device Draft Guidance.

The Clinical Data Draft Guidance provides illustrative examples of each scenario. In 14 of 17 examples,1 FDA states that clinical data “may” be needed to support a substantial equivalence determination, suggesting that FDA is attempting to retain as much discretion as possible to request clinical data. Unlike the 510(k) Program Guidance from 2014, the Clinical Data Draft Guidance does not include any statements regarding the low frequency with which FDA reviewers request clinical data, which suggests that this draft guidance could signal FDA’s intention to increase the rate of such requests going forward.

Implants Draft Guidance

The Implants Draft Guidance is the first guidance that provides generally applicable recommendations for all implant devices for which a 510(k) is required. FDA has previously issued 14 guidance documents specific to certain types of implant devices. The Implants Draft Guidance details FDA’s expectations for an implant device, including indications for use considerations, considerations regarding the intended duration of implantation, non-clinical testing, human factors and usability testing, clinical performance testing, patient experience information (where relevant), and labeling.

Next Steps and Implications for the Medical Device Industry

FDA intends to hold a webinar to discuss all three draft guidances on October 26, 2023. Additionally, interested parties may submit comments on any of the draft guidances by December 6, 2023.

While these draft guidance documents represent a step forward in FDA’s efforts to modernize and update the 510(k) pathway, FDA likely still has more significant reform efforts in mind. For example, Dr. Jeffrey Shuren, the Director of the Center for Devices and Radiological Health (“CDRH”) at FDA, has publicly commented in recent months about other possible approaches, including the creation of a new pathway to market referred to as the Voluntary Alternative Pathway (“VAP”). According to Dr. Shuren, the VAP, which FDA has apparently discussed privately with the White House Office of Management and Budget and leadership of the Department of Health and Human Services, would give FDA authority to offer a more customized approach to regulating new technologies—and in particular, rapidly evolving digital technologies. The current thinking is that if this proposal were implemented, it would give sponsors the option to pursue either the flexible voluntary pathway or the more traditional, existing regulatory routes like the 510(k) pathway. Whether this VAP proposal, or other potential efforts to modernize the 510(k) pathway, are likely to gain traction within the Executive branch or with Congress remains to be seen. Nevertheless, the recent draft guidance documents are certainly steps on the way to creating broader and more meaningful change.

Ropes & Gray will continue to monitor developments in this area. If you have any questions about this Alert, please contact any member of our FDA regulatory practice or your usual Ropes & Gray advisor.

  1. While the draft guidance includes 16 labeled examples, one example (2-C) covers two scenarios with different outcomes.