Medical Device Daily Washington Editor
A cynic's guide to the galaxy of leadership might suggest that the ambitious "find a parade and jump in front," a recommendation seemingly observed by FDA in yesterday's conference on device applications.
The conference specifically addressed how the agency might incorporate new safety and efficacy data when dealing with a PMA or a 510(k) application for the same or a similar device, but the agency's moderator for the session, Jonathan Sackner-Bernstein, MD, seemed intent on prodding reluctant attendees for their ideas on the subject. However, while suggestions abounded as to whether an approval or clearance should be granted, a consensus on the regulatory specifics was simply not in the cards.
At the start of the session, Jeff Shuren, the newly appointed CDRH director, said the session was "critical to provide pathways" that are clear and predictable, and provide patients with devices "that can have a significant impact on public health." Shuren added that "long-term experience can provide new information ... that was not previously available," but he asserted that the regulatory process "must adapt as the science evolves."
"Our goal is to become predictably adaptive" to emerging information on safety and efficacy, Shuren said.
In his remarks before commencing with the session, Sackner-Bernstein said that FDA's report on what it will recommend for the future is due May 31 and will be subject to public comment. Any resulting plans will be implemented in September, he noted.
The first two scenarios posed by FDA in the discussion included a 510(k) application based on a predicate that has been in use for some time, but which has generated long-term safety data via medical device reports (MDRs) that FDA is concerned about.
In the second, purportedly parallel scenario, an article in a peer-review journal calls into question the results of a PMA device's pivotal clinical trial at a time when a similar device is in the PMA pipeline.
Joe Smith, MD, VP for emerging technologies at Johnson & Johnson (New Brunswick, New Jersey) said acting on MDRs "is something that every company ... struggles with as an opportunity for continuous improvement." He said that a root-cause analysis, which can take longer than FDA might want to wait, would be needed and that he "would hesitate to advise action on sparse and incomplete data at best." He made an argument for "a process that derives from those reports" backed by a root-cause analysis "to understand what the issue is, assuming there is one." If the post-market data "don't change the overall risk/benefit ratio of the device, it points more toward informing than toward other [regulatory] actions."
Frequent FDA critic Steve Nissen, MD, of the Cleveland Clinic (Cleveland) acknowledged at the outset that he is "one of those who have been very critical of the device approval process." He also made no bones about seeing the 510(k) and the PMA processes as inadequate.
"The best way to avoid this problem is to be more careful about approving" high-risk devices, Nissen said, adding, "The quality of clinical trials that lead to device approval is much lower" than for drugs. The typical medical device trial "is really not very good by contemporary scientific standards," he said, adding that it is "much too common for devices to go to market with poor-quality trials." He also said "for high-risk devices, having an ongoing registry would be very helpful."
Pat Schrader, senior VP for corporate and regulatory affairs at Becton Dickinson (Franklin Lakes, New Jersey) suggested that MDRs are not necessarily indicative of an underlying problem with the device, given the possibility that misuse, for instance, is not inherently addressed. She also said, "One of the most productive things FDA can do is to reach out to the manufacturer" to get more information on the device. "Once we all understand what the root of the problem is," that information is liable to help other 510(k)s.
As for the device review process, Schrader argued, "all you need to do is look at the thousands and thousands of devices on the market" to see that regulations have worked and have "kept a lot of unsafe devices off the market" but fostered innovation.
Sackner-Bernstein attempted to corral any dissent on the question of how to interpret MDR signals, urging panelists to "consider the MDR data as reaching the depth and breadth of a pattern." However, Tim Wright, a scientist at the Hospital for Special Surgery (New York) argued that "even though you think you have a pattern, you might not know" whether the problem affects the entire group or only subgroups, for instance. He also remarked that a discernible pattern in one 510(k) device will not necessarily apply to a similar cleared device.
Vivian Coates of the ECRI Institute (Plymouth Meeting, Pennsylvania) said, "It seems like we're establishing new safety standards," adding that "we're losing a nuance of the case." She said that a resolution to any factors giving rise to MDRs might be achieved by a change of labeling rather than by prohibition of any further applications. "We should take note of that nuance. It might not be an all-or-nothing scenario," she said.
Diana Zuckerman, PhD, president of the National Research Center for Women and Families (Washington) said that establishing a root cause "takes a really long time, and I can't think of a reason where a really dangerous device should be the predicate for reviewing another device." She argued that until FDA and industry are "absolutely sure" as to the meaning of MDRs, "you should not consider" clearing another such device. Nissen struck a similar tone, arguing, "If a device cleared under 510(k) develops new issues, what it tells us is that" the device should have gone through the PMA route rather than the 510(k) channel.
In a Feb. 9 statement, Janet Trunzo, the executive VP for technology and regulatory affairs at the Advanced Medical Technology Association (AdvaMed; Washington) said industry welcomes "the opportunity to work with FDA to address the challenge of incorporating new scientific information into the regulatory decision making process." However, she advised FDA to "carefully consider what constitutes 'new scientific' evidence and how this information is incorporated" into product reviews.
When new evidence suggests an immediate threat to patients, FDA should "promptly take whatever action it deems appropriate," Trunzo said, but that information should be subjected to "rigorous scientific investigation" and considered "in the context of previous evidence. She also noted that FDA "has a range of regulatory options" to choose from, including labeling changes and public health notifications.
Mark McCarty, 703-268-5690