A now-little known band called the Classics IV made a big splash in 1968 when their Hallowe’en love song “Spooky” charted at #3 on the U.S. Billboard Hot 100, and eerie though it may sound, there might be a lesson in there for the FDA half a century later.
Sound preposterous? Read on and decide for yourself.
House (and Senate) on Haunted Hill
As we’ve all seen, the FDA has proposed to pull the substantial equivalence standard for 510(k) filings, but the agency also acknowledged that these 510(k) submissions have ballooned over the past decade or so. Some of that is no doubt due to the agency’s unquenchable thirst for more information, but we might all react that way if we were hauled into the dreary chambers of Congress every time a device exhibits problems.
But another source of the agency’s nighttime terrors revolves around the fact that class II devices have become increasingly complex, thanks to the technological evolutions and revolutions that have come to the fore since the passage of the Medical Device Amendments in 1976. So when the FDA says its regulatory hands are a bit tied by frighteningly hoary concepts of device regulation, it has a point. Almost nobody back then was thinking about pacemakers smaller than fusilli, so perhaps the statute does need a bit of work.
The question is whether the FDA’s views of statutory reform are the most useful. When it comes to its approach to non-digital, non-IVD class II devices, the answer might be no.
2b or not 2b
We’ve all watched over the years as other regulatory jurisdictions have devised their own device regulatory frameworks, and for the most part, they’ve opted for a four-tier system. These schemes typically revolve around the categories of low risk, low-to-medium risk, medium-to-high risk and high risk. Some like to use classes I-IV to capture these risk ranges while others prefer a class I, IIa, IIb and III nomenclature, but it’s a difference that makes no difference in the end.
The important point is that the four-tier system allows one to deal more sensibly with the vast, shadowy territory between low and high risk. And that is precisely where the seemingly mummified U.S. approach falters. But let’s acknowledge that the U.S. Congress was the first to take on medical device regulations, and like first-time parents, they were blazing trails without a manual. It only stands to reason it was not a perfect effort out the gate.
But back for a moment to the range of 510(k) applications: Some of these require clinical data, some do not. And even the FDA has acknowledged that some 510(k) filings are for fairly simple devices (and would it kill the FDA to acknowledge the nation-of-origin problem?). So what’s the point of lumping them all together? Would it not make more sense to cleave these two more or less clearly defined groups of moderate-risk devices in such a way to fairly closely match the complexity of their regulatory filings?
Venturing into the darkness
I propose that the FDA would have a much better shot of gaining concordance with Congress if it requested a statutory rewrite that resets the device category system to four classes, and the advantages are obvious. Class II or IIa, whichever you like, would still be eligible for the use of the substantial equivalence standard, while class III or IIb would have to come up with a full suite of data for bench and clinical data and so on. This latter bunch is a group for which objective performance criteria would also be appropriate for a well understood device type, but OPCs might also be useful for the lower of these two tiers if one assumes the OPC are related strictly to bench testing and other non-clinical evaluations. However, OPCs for class II or IIb devices would not serve as an excuse to jettison the substantial equivalence standard.
The point here is that substantial equivalence is hard-wired into the statute, and it still serves a purpose. We’ve previously discussed whether the FDA has any chance of moving away from substantial equivalence, and the conclusion is unavoidably no, at least at present. Conversely, stakeholders and members of Congress are likely to find a four-tier risk classification system to be a lot more like a bipartisan treat than a trick. Granted the FDA and device makers would have a lot of paperwork to do, and there is this problem of deciding which product codes would fall into which class (although I think the question of legacy devices would be less troublesome than might seem the case up front).
But once everyone is through the transition, a world of frights just melts away like sunrise on Nov. 1. The standards for the amount of premarket data would be much more transparent, and four risk classes would render irrelevant the agency’s klutzy attempt to substitute its judgment for congressional oversight. And let’s not forget the benefits related to regulatory harmonization, even if some in the patent bar think patent harmonization was an amalgam of bad ideas stuck together with stitches at the joints and bolts in the neck.
Another benefit to a four-class approach is that it would finally lay to rest a ton of uncertainty around the FDA’s expectations of class II devices, and I don’t think it requires reiterating why uncertainty is the bane of economic and medical progress. Beyond that, I’d suggest that if there’s anything scarier than the transition from a three-class to a four-class device risk scheme, it’s the prospect of government agencies routinely hijacking the statute to fit their momentary pique. If there’s anything spookier than that…