The 510(k) pathway was first established in the Medical Device Amendments of 1976, and amended in the Safe Medical Devices Act of 1990. Since that time, the statutory language creating this pathway has remained largely unchanged. Yet, the program has morphed from one in which sponsors submitted a short set for materials showing that a proposed and predicate device were similar in terms of design and intended use through labeling and, sometimes, minimal testing, to a program in which sponsors often submit a thousand pages (or more) of data, testing results, and information to establish substantial equivalence. This has all resulted from changes in FDA’s expectations and requirements for these types of submissions. FDA highlighted these changes in its performance report (here).
On November 26, 2018, FDA Commissioner Dr. Scott Gottlieb and Dr. Jeffrey Shuren, Director of the Center for Devices and Radiological Health, announced that FDA is planning to “modernize” the 510(k) process. A Statement from FDA Commissioner Scott Gottlieb, M.D. and Jeff Shuren, M.D., Director of the Center for Devices and Radiological Health, on transformative new steps to modernize FDA’s 510(k) program to advance the review of the safety and effectiveness of medical devices (Statement) focuses primarily on ensuring that new devices cleared through the 510(k) process meet evolving standards of safety and effectiveness. Key points are summarized below.
Eliminating Use of Old Predicates
According to the Statement, approximately 20% of 510(k)s are cleared based on comparison to a predicate that is more than ten years old. The Statement says that this figure is problematic because it is a sign that “some devices may not be continually improving, which is the hallmark of health technologies.” The Agency appears to believe that new devices should be compared to the benefits and risks of more modern technology. To this end, as indicated in the Statement, FDA will promote use of newer predicates. To anyone that has filed a 510(k) recently, this attitude comes as no surprise. FDA prefers newer predicates. Even if an older predicate is selected by an applicant, the testing and data requirements are typically based on more recent 510(k) clearances, and FDA will often ask for more data than contained in the predicate device 510(k).
The Statement indicates that FDA will make public on its website those cleared devices that demonstrated substantial equivalence to older predicates (i.e., predicates that are more than 10 years old). A few comments: First, it is unclear if this list will only include recently cleared devices with “old” predicates or if FDA will annotate the entire 510(k) database. It is also unclear how multiple predicates will be handled (i.e., if a device has three predicates and only one is older than 10 years). It is also not obvious how this information will help stakeholders. To the extent the implication is that a 510(k)-cleared device with an “old” predicate should be given less weight, that implication may be unfounded.
Second, while this annotation may help highlight the age of the predicate(s), the information is already effectively available in the current database through the 510(k) summaries, which list the predicates and links to their clearance information. Submitters today are aware of FDA’s preference for more recent predicates, and tend to balance FDA’s predilection against the value of an older predicate. An older predicate is often the most appropriate predicate device due to a design or intended use consideration. And some devices do not evolve that quickly (e.g., condoms). If the data and testing requirements satisfy today’s high standards of substantial equivalence, it is hard to see why use of an older predicate would be problematic or why a 510(k) clearance should be singled out by receiving the FDA equivalent of a Roger Maris asterisk. These issues tend to get sorted out quite well in the current review process.
Third, use of newer predicates, according to FDA, will “give patients and their doctors a choice among older and newer versions of a type of device, promote greater competition to adopt modern features that improve safety and performance, and help make sure that newer devices reflect more modern technology and standards that can improve patient care and outcomes.” It sounds as though FDA is suggesting that sponsors who obtain clearance using a newer predicate could have some sort of competitive advantage over a device cleared based on an older predicate. FDA, traditionally, does not permit comparative claims unless there is direct, head-to-head data showing that one device clinically performs better than another. Perhaps, FDA plans to loosen this standard to allow additional comparative claims based solely on the age of a chosen predicate, but it seems a stretch to uniformly equate newer predicates with better performance.
“Sunsetting” Older Predicates
To further its goal of eliminating use of older predicates, FDA states that it will begin to “sunset” certain older 510(k)s. It is unclear what this means, how it would work, or if FDA even has statutory authority to do so. By way of example, FDA notes that it has worked to eliminate use of certain predicates that have raised safety concerns. FDA highlights its up‑classification of certain devices to require PMAs rather than 510(k)s. Sunsetting, in theory, however, seems fundamentally different and potentially more far reaching than simply eliminating unsafe predicates.
The announcement states that FDA has made 1,758 510(k)s ineligible for use as a predicate. The vast majority of these were due to up‑classification. It is not stated how many device types are embedded in this number. It seems unlikely that FDA has up‑classified 1,758 device types. Rather, it seems probable that this figure represents the number of individual devices affected by a smaller number of up‑classification decisions.
FDA highlights an announcement made last week that its National Evaluation System for health Technology (NEST) has obtained additional funding and is beginning certain projects (here). The announcement underscores areas of women’s health that FDA will focus on through registries for breast implants, industry-wide communications regarding vaginal rejuvenation devices, and holding a public meeting regarding gynecologic surgical mesh. These tools all appear to be ones that FDA can take under its current statutory authority to ensure safe and effective devices reach patients. Sunsetting “old” 510(k)s may require congressional action to amend the statute. The FDC Act defines predicate devices, and the age of the predicate is not a statutory criterion. FDA says it will “consult” with Congress, but consultation may be insufficient. We are reminded of the proposed rule on rescinding 510(k)s issued circa 2000, with equally uncertain statutory authority. It was never finalized.
While sunsetting predicates clearly implicates devices requiring 510(k) clearance, it is also possible that this process could affect 510(k)-exempt devices. As we have discussed in earlier posts (here and here), 510(k) clearances, both before and after 510(k) exemption, help to set the bounds of a 510(k) exemption for a particular device type. If FDA makes sweeping changes to “sunset” old 510(k)s, it could alter the bounds of 510(k) exemptions set by the now “sunsetted” 510(k)s. That is one of the many “nuts and bolts” concerns that would have to be addressed in any proposal.
Alterative “Safety and Performance Based” 510(k) Pathway
According to the Statement, FDA also plans to create an alternative 510(k) pathway that allows sponsors to establish substantial equivalence by demonstrating that a device meets safety and performance criteria for the device type. FDA states that this new pathway would evaluate a device based on “a contemporary baseline” rather than outdated standards. FDA’s vision is described in this paragraph:
We believe this approach is the future of the 510(k) program. Rather than looking to the past as a baseline for safety and effectiveness – and rely on predicate devices that are sometimes decades old as our point of comparison – our premarket review would be based on a contemporary baseline that looks to the future and a baseline that can be updated as technologies advance. Sometimes, by relying on old predicates, it can actually make it more difficult for more advanced technology to reach patients since it’s harder for an innovative product to bridge to an outdated technology reflected in a decades-old predicate.
While the Statement touts this approach as new, it sounds a lot like the current Abbreviated 510(k) pathway, which allows submitters to rely primarily on conformity to consensus standards. The Abbreviated 510(k) pathway works well for some device types, but it would be exceedingly time and labor intensive for FDA to adopt this quasi-standards approach for most device types, and then to keep the standards updated to ensure that they continue to remain “contemporary.” Or, it may be that FDA is contemplating that more of these devices will need to go through the De Novo route for lack of a “modern” predicate, or a submission that is called a 510(k) but is no longer tethered to a predicate device. The details of this pathway are unclear.
Standards are most useful for device types that do not evolve quickly, which are the same device types where older predicates retain value. For device types that evolve quickly, it is easy for standards to fall into obsolescence unless a great deal of time and resources are focused on updating them. It is these device types where newer predicates tend to be cited anyway, because they will be closest to the newest iterations of the technology. In short, the Statement fails to recognize that the 510(k) process already provides a technological baseline that updates in a self‑executing way. We wrote about that process in detail here.
Therefore, it is difficult to see how this new pathway will do much more than the modestly successful Abbreviated 510(k) pathway. We note also that FDA was given authority to develop performance standards for device types in 1976 and it was so little used that in 1990 Congress gave FDA added authority to impose “special controls” short of performance standards. We are not aware of any performance standards ever being formally issued.
The Statement suggests that this new pathway would be beneficial for companies, because such a 510(k) clearance could help support coverage decisions by “more readily demonstrat[ing] to payors that their products perform better than other devices on the market.” It is left unstated how the link between 510(k) clearance and payor reimbursement decisions would work. The clearance process itself is not intended to rank devices, but rather to establish substantial equivalence. It seems a stretch to suggest that the 510(k) clearance process itself could meaningfully be deployed to shape third-party payor reimbursement decisions. Furthermore, companies may be better off addressing reimbursement issues directly, and not as an ancillary aspect of the 510(k) process. We would suggest that FDA address the 510(k) program without considering the reimbursement ramifications.
The announcement concludes by stating that based on the changes to the 510(k) process more devices will ultimately use the De Novo process. This is easy to see – by eliminating older predicates there will be fewer devices on which companies can claim substantial equivalence. FDA adds that it will be issuing a proposed rule to clarify “procedures and requirements for submissions of De Novo requests.” FDA has had this proposed rule on its agenda for quite some time (see here), but there have been no details. We look forward to seeing FDA’s proposal to clarify the De Novo process, as well as other details relating to the Statement.