Today's public meeting on the IOM report regarding the 510(k) process held few surprises, unless you count lack of acrimony as a surprise. After a day of hearing public feedback, our initial assessment of the report pretty much stands: the report contains a number of potentially excellent ideas that are highly unlikely to become established law and regulation in the foreseeable future.
There were a number of interesting thoughts and threads of coversations that came up today, however. In no particular order, some of the more interesting moments were:
- A New Alliance. As the representative from MDMA quipped, the IOM report has done the seemingly impossible: it has put industry and FDA sqarely on the same side of an issue. This remark provided maybe the only funny moment of the day, but it's also a decent summary of the flavor of the day. The constituency for stasis is a powerful one and it's clear that lots of people want to keep things as they are. The reasons provided are varied and frequently compelling, but it's tough not to wonder whether it's just our mental habit to prefer things as they are to how they might be.
- Committee Composition. Several times, industry people made reference to the lack of industry representation on the IOM committee. That's a fair point. It got taken a couple steps too far, though, as several speakers railed against the lack of first-hand experience committee members have with 510(k) applications. One speaker even pointed out that the committee didn't even review any specific application, as though that were the only way they could actually know what they were weighing in on. It may be worth pointing out that Steve Gutman was a member of this committee. As head of OIVD for six years, he had plenty of opportunity to read, review and understand more 510(k)s than most of us could possibly have considered. Industry is entitled to feel some pique over not being invited to sit on the committee, but it's not as though the committee was composed entirely of outsiders, either.
- Patient Injury. While many speakers made the point that there is no evidence of widespread injury or death from 510(k) products, several speakers spoke to the injury and death experienced by patients as a consequence of using cleared products. Certainly, there is room for both perspectives to be true: a device can be safe in a broad sense and yet also be involved in some number of heartbreaking events. Either way, it sounds like FDA got a useful earful on doing follow-up communication regarding device incidents reported in MAUDE and the special problems of incorporating existing safety concerns into the decision to allow use of a predicate.
- Definitions of Innovation. For purposes of assessing whether FDA stifles innovation, IOM decided to adopt a novel definition of "innovation" first. In essence, they argue that "innovation" is not just what is novel per se, but what improves quality of care or access to care. A few industry types took issue with this and we tend to agree. Novelty for novelty's sake does not always produce obvious benefits, but sometimes it does. Saying, in essence, that innovations aren't innovative until they have demonstrated their value is a bit circular and risks defining away the problem. Even innovations that produce no net increase in value may still provide some value worth having. Discounting that possibility in advance does not seem like a good way to start a serious conversation about innovation.
- Safety and Effectiveness. There were a number of comments about whether the substantial equivalence standard can (by statute or practice) provide sufficient information about safety and effectiveness. IOM clearly believes it doesn't and can't, therefore they recommend junking the 510(k) process in favor of something that's entirely risk-based. There are many well-known limitations to the substantial equivalence standard, but it's not as though FDA's hands are really tied on this. If a reviewer believes that they need to see additional clinical data, it hasn't been our experience that they will be shy about asking for it. How OIVD approaches this issue appears to differ from how the rest of CDRH does, but that's important too. Substantial equivalence is not just one carved-in-stone standard that's unchanging and inflexible. It is interpreted differently at different times and for different situations. There is enough flexibility here, we think, that most issues can be worked through without junking the entire approach.
All in all, it was an interesting meeting. It's tough to imagine that the substantial equivalence standard will be around forever, but it's a pretty safe bet that it will stick around for many years to come... probably long enough to cover anything that might currently be under development.
Of all the things IVD developers need to worry about, we'd recommend putting a wholesale re-working of the 510(k) process pretty low on the list. That said, we'll keep an eye on the discussion as it plays out and will be sure to say something if it starts gaining traction.