National Center for Health Research, May 22, 2019
National Center for Health Research’s Public Comments on Modernizing the Food and Drug Administration’s 510(k) Program Request for Comments
[FDA-2018-N-4751]
Thank you for the opportunity to share our views on this important issue. The National Center for Health Research (NCHR) is a nonprofit think tank that conducts, analyzes, and scrutinizes research, policies, and programs on a range of issues related to health and safety. We do not accept funding from companies that make products that are the subject of our work.
1. Should the FDA [Food and Drug Administration] make public a list of devices or manufacturers who make technologies that rely on predicates that have been on the market for more than a certain number of years (e.g., 10 years)? If so, what would be an appropriate period of time?
Answer: NCHR strongly believes that patients and consumers have a right to know that almost all medical devices cleared under the 510(k) pathway have undergone little or no clinical testing in humans. That is currently not the case; instead, patients assume that all medical devices cleared by FDA have been stringently tested in clinical trials, as is required for FDA-approved drugs.
Patients should be informed by FDA on the agency’s web site, on device labels, and through other means of communication, that more than 95% of regulated devices are approved through the 510(k) pathway, and not through the more rigorous Premarket Approval (PMA) process. In NCHR’s considerable experience, patients do not know or understand that critical fact, and many physicians do not either.
The amount of time that the device predicate has been on the market, while a significant factor, is less critical than the exact ways that the predicate is similar to the new device, and whether that predicate has undergone sufficient post-market clinical testing to give patients, physicians, and surgeons justifiable confidence that the device has benefits that are proven to outweigh the risks.
For example, metal-on-metal hip implants should never have been cleared through the 510(k) pathway. Possible problems with metal-on-metal friction causing metal debris inside the human body was well established, and clinical trials should have been required through the PMA process. This is an excellent example of how the age of the predicate can be irrelevant, but the lack of a predicate with meaningful substantial equivalence should have prevented clearance. This situation was made even more dangerous by the lack of useful post-market surveillance or post-market studies in the U.S.
2. Should the FDA consider using other criteria to inform our point of reference?
Answer: As noted above, the 510(k) pathway should be much more scientific and stringent about standards of substantial equivalence. We agree with the Institute of Medicine (IOM) 2011 report that bluntly concluded that the 510(k) pathway established neither safety nor efficacy and should be replaced. In fact, our own research of implants cleared by the FDA between 2008 and 2012 found that although sponsors were required to provide scientific evidence of substantial equivalence in their summaries submitted to the FDA and made public on the FDA website, only 16% of companies actually did so.
In addition to the need for FDA to enforce the law requiring scientific evidence of substantial equivalence in 510(k) applications, the FDA should prohibit the use of recalled devices as predicates, as well as devices that have a large number of adverse events reported to the FDA or data in the published literature that suggests risks may outweigh benefits.
Another important improvement would be a more meaningful distinction between Class II and Class III devices. Many implants that are potentially life-saving or life threatening are currently classified as moderate-risk (Class II) rather than high-risk (Class III); this down-classification is putting patients at risk and undermining the confidence of physicians and patients in medical devices.
If the FDA used more appropriate criteria for Class II devices and more stringent criteria for substantial equivalence, it is likely that numerous devices that would be cleared through the 510(k) process based on the looser criteria that has been common for many years, would instead be required to be cleared through the de Novo process or PMA. As a result of better documentation of safety and effectiveness, FDA should ensure that patients be informed of the known risks, including failure rates, of devices, so that they can make an informed decision about the risk versus benefit calculus.
However, our research into the clinical research on the efficacy and safety of devices cleared through the de Novo pathway between 2013 and 2017 also found deficiencies. We found that some of the devices included no clinical trial data on the safety or effectiveness of the device, and many were studied only in uncontrolled trials, which are scientifically inferior to controlled trials, and much less informative to patients and physicians.
3. Are there other and/or alternative actions we should take to promote the development and marketing of safer, more effective 510(k) devices?
Answer: Scores of disturbing stories about FDA’s failure to ensure that devices are safe and effective have been documented by news articles and editorials by the International Consortium of Investigative Journalists (ICIJ) and their series, “The Implant Files”; and also by NBC; NPR; Kaiser Health News; and in a particularly scathing editorial by the Sunday New York Times on May 4.
All these investigations were highly critical of FDA’s demonstrated shortcomings in detecting device adverse events or failure in the marketplace in a timely manner. FDA officials promised to examine and improve these systems a few months ago, following the publication and broadcast of these news stories. The crucial balance between innovation and safety, and risk versus benefit, which are the gold standards that FDA uses to protect patients, is in question. The age of predicates is much less important than the many other shortcomings CDRH’s [Center for Devices and Radiological Health] premarket requirements and post-market surveillance.
4. Should the FDA consider certain actions that might require new authority, such as making at least some older devices ineligible as predicates? Are there actions the FDA should pursue which make require new authority?
Answer: FDA already has the authority to use more stringent criteria for device predicates, and as noted above, age may or may not be an important one. An older predicate that was proven safe and effective in well-designed clinical trials is likely to be a much more appropriate predicate than a newer predicate that was never studied in clinical trials and is made of different materials or has a different mechanism of action. As noted above, no device that was recalled or removed from the marketplace after adverse events were reported, or found to have risks outweighing the benefits should be allowed to serve as a predicate for an approval of a new medical device.
The oversight of medical device safety has fallen badly behind as industry user fees, which are used to ensure quicker reviewers rather than better safeguards, became a larger proportion of the Center’s budget. FDA must address that growing disparity. While the need for increased appropriations is a Congressional issue, the designation of user fees for premarket safety analysis and post-market surveillance is the responsibility of CDRH.
The 510(k) pathway system is flawed, and should be replaced with pathways that require controlled clinical trials for all implanted devices and many other devices as well. Meanwhile, safety can be improved with the recommendations above.
The National Center for Health Research can be reached at info@center4research.org.