Contributing Editor
Mark DuVal, JD, is president of DuVal & Associates, a Minneapolis-based firm dedicated to counseling companies in the medical device, pharmaceutical, biotech, food and nutritional supplement industries. His practice includes providing strategic regulatory advice, developing compliance programs, designing and implementing sophisticated marketing programs, counseling on reimbursement matters, conducting sales training and interfacing extensively on behalf of companies with the FDA with relation to product approvals and clearances, clinical trial negotiations, approvals, policy arguments, appeals, etc.
MARK DuVAL
Tilts at FDA Windmills
Prior to founding the firm, Mark was general counsel for 3M Pharmaceuticals and Drug Delivery Systems, working both domestically and internationally. He also worked at Medtronic as an expert on FDA matters and in corporate compliance. Mark is a frequent national speaker and writer on a variety of issues relating to product approvals/clearances and other regulatory/reimbursement issues.
BB&T: Those involved with the med-tech industry – as operators of companies, providers of services to those companies or investors – are alarmed over possible changes in the FDA's 510(k) program, the primary route to market approval for most companies in the space. Could you outline where it appears the agency is headed on this issue?
DuVal: Therein lies the problem. While the FDA has been very public about the fact that it is addressing the 510(k) program, and it has outlined categorically some of the areas that it will be reviewing, they haven't been forthcoming with a lot of detail. That is very problematic, because, for example, when FDA holds these town hall meetings where everyone gets their five minutes at the microphone, all you can do is speak your peace as to what you think they might be doing.
FDA has created “working groups“ to revisit the 510(k) program and look for improvements. First, they have talked about re-examining the 510(k) program in the definitional or interpretational areas of what are the new indications and new intended use, they have also talked about when does a new featured benefit or new technological characteristic raise a “new“ type of question about safety and effectiveness. So they are going to be digging into that.
The second thing they're going to be looking at is what is the quantum of evidence necessary to clear a 510(k) and when is it appropriate to request human clinical data? In addition, they're talking about the issue of multiple predicates and predicate creep, and a variety of other issues. We in industry have broadly provided FDA input on those issues, but it's not the same as if you're sitting at the table.
The FDA has asked an independent group, the Institute of Medicine (IOM), to take a look at the 510(k) program as well. The FDA, and IOM, have deemed industry a “conflict of interest.“ Our position is that we are one of the main interests, in the sense that we are the primary stakeholder that uses the system and has input into it. We can provide them with meaningful input about what is working at FDA and what isn't, and how might the 510(k) program be changed. The IOM has a number of members with heavy academic backgrounds and seemingly little experience interfacing with FDA as industry does. That doesn't mean they won't have something to contribute – they will. Public Citizen is a member too, and they are one of the biggest critics of the medical device industry on the planet.
It's unfortunate that they really don't round out the membership to get industry's active input. I don't mind that there are industry/510(k) critics and non-industry people at the table, there should be, but why would they not include people from industry – from the start-up community, from medium-sized companies, from the large companies – just to provide them meaningful input about the 510(k) program from another perspective. So it's concerning. The biggest issue I have is that we would like to have participation. It's sort of a black box; we see input going in, but we don't know what's going to come out.
BB&T: One phrase that caught my attention in particular, because there has been a lot said about it lately, is “predicate creep.“ It kind of looks to me like FDA is pretty much looking to stomp out predicate creep completely.
DuVal: Yes, that's concerning to me and industry as well. We view predicate creep as a positive attribute of the process. The whole 510(k) process was designed to accommodate technological change, and embrace it. But the way FDA tends to use the term predicate creep is as a perjorative – that there is something wrong with the system when we make incremental advances to technology. That's a good thing, not a bad thing. I understand FDA's issue is how do we keep up with all this technological change, but that is their charge—always has been—and the 510(k) program gives FDA enough flexibility to ask for the information they need, within reason and limits, to examine new technology.
BB&T: Do you have a sense of what is the reasoning behind this push to re-do the 510(k) program overall?
DuVal: In fairness to the FDA, they are increasingly being confronted, or faced with new and more complex technologies, so at base the products are very similar to some of the existing predicates, but they have some level of change. And it is that change which introduces the complexities to FDA's review, because if it's a new technological characteristic, it often will raise unanswered questions of safety and effectiveness. And the agency is trying to get its arms around the question of what kind and amount of evidence are we going to require to grant a “substantial equivalent“ determination? When we feel we fall short, when we feel that the change is too great, and the product is too different from the existing predicate, what should FDA do? Well the FDA and industry have a nice outlet in the de novo program, which can be used when FDA finds there is not an existing predicate. Remember, when a company gets a “not substantially equivalent“ letter, meaning they are not getting a 510(k) clearance, the device is then automatically classified as a class III high risk device – even if the device does not deserve that risk classification. With a de novo request for reclassification, the company can request FDA to review a device as a class II moderate-risk device. If FDA agrees and proceeds down the de novo path, FDA is not limited by the substantial equivalence standard or review; it can ask for information that establishes “reasonable assurance of safety and effectiveness,“ the PMA standard of review. This is more that what is need for a 510(k). Because the device is now considered in class II, a moderate, not high-risk class, FDA should not require the company to submit the same amount of information required of a high-risk class III device. The data request is risk-adjusted.
So they're grappling with all of that in a world of more complex devices – how do you ask for enough information without asking for more than is truly needed, which affects speed of innovation to the market? I understand it's a struggle for them – it's a struggle for industry as well. But there are reasonable ways out of this. My view is that the 510(k) program is sufficiently flexible as it currently stands to accommodate this kind of dialogue and debate, and we just need to put some parameters around it. Maybe you have tier one 510(k) devices that really are “me too“ devices, and tier two, which present a little more of a challenge for the agency, and then what are the requirements for those tier two-type 510(k) devices? And the de novo program is a way out for devices that don't quite meet the 510(k) criteria, but do not belong in a class III high risk category.
BB&T: I think it's interesting that when this subject has come in conversation in recent months, one of the things that they almost always bring up, especially those in the investor community, is that the system is pretty good – it may need some tweaking, but it's pretty good. And if big-time change is coming, that's one of those 'throw out the baby with the bathwater' things.“
DuVal: The folks I deal with, and I probably am working with 100 medical-device companies at any given time, really believe that the framework is good – it's still very viable. We need to upgrade it a bit, to figure out how to interpret certain provisions and how to make it work as new technologies come into the marketplace, but to consider replacing it entirely is really a nightmare consideration for industry. We have a familiarity with this program, and it's fundamentally sound and workable with some improvements. Between FDA and industry, we can come up with those improvements.
BB&T: Are any short-term “fixes“ being considered while awaiting the Institute of Medicine's supposedly definitive report on the 510(k) program?
DuVal: They have a number of internal working groups at FDA – I think it's nine or 10 such groups – that have been assigned discrete and insular kinds of projects to examine the 510(k) program. I know they include things such as, when does another indication constitute a new intended use, and when does it constitute the same intended use? Another is the issue of new technological characteristics – when is a new type of question of safety and effectiveness question raised by that new feature? There's a group working on that. There is one working on the amount of performance and clinical data needed to clear a 510(k). I think there is one that is about predicate creep and the use of multiple predicates. There is a working group on de novo. So all of those issues that I outlined before, there are working groups assigned to them.
I give CDRH Acting Director Jeffrey Shuren great credit, and Donna Bea Tillman before him, for wanting to re-examine all these things. The industry is just asking for more certainty, more predictability, and if that's the outcome of all this work, great. If the outcome is that they propose an entirely new system to be thrust upon industry without us really having any input, that's very troublesome.
I'm not sure FDA is really too focused on the second half of its mission. It's a twofold mission, right? One is protecting patients, and the other is speeding innovations to market. If the industry has one concern with the agency today, it is that there is not a whole lot of concern with speeding innovations to market anymore. They have become so risk-averse, the emphasis has become entirely on protecting patients to the nth degree, and we will never have a perfect device or drug that gets through the FDA, no matter how much study we do. So we deprive patients of the benefits of products that get on the market in Europe two to eight years before they get on in the U.S., and that's a travesty. I think Dr. Shuren is trying to bring a better balance to FDA's analysis.
BB&T: Is the basic thrust of the FDA's plans for 510(k)s that devices approved under the program demonstrate superiority to the predicate device, rather than non-inferiority, the existing standard?
DuVal: In operation. some of FDA's review teams use a lot of inappropriate standards for “substantial equivalence.“ I have had this dialogue with many FDA folks – I take a lot of companies on appeal to the FDA. I am frequently debating the agency that one, you are asking for information you want, but don't need in order to establish substantial equivalency. Secondly, you often are asking for comparison to standard of care, and that's not the role of the FDA. Thirdly, they're also frequently asking the sponsor to establish the clinical utility or clinical benefit of the device when it already has been established by the predicate. In asking for that type of information, FDA is essentially practicing medicine, which is an inappropriate request for FDA. FDA should allow devices that have established substantial equivalence to come to the market and allow the medical community to determine if they like the device or not. The device should simply go through FDA and substantiate its labeled claim.
All of these are issues that I debate on a weekly basis with the agency. You will find these issues discussed in the internal FDA report released on the ReGen review discussed by FDA's own lawyers, by the Office of Chief Counsel (OCC). In that report, the OCC cited the review staff for asking all those inappropriate questions that were outside the scope of the statute. They should not be asking them, but they are – throughout much of CDRH.
BB&T: So then, is FDA following the legal standard for review of a 510(k)?
DuVal: In my view, sometimes yes, many times no, and I think they get away with it because they can. Most of the time all you buy yourself by arguing with FDA on these issues this is endless appeals, unless you get to someone who knows the statute at the top from management. As someone who works intimately with the agency day in and day out, I really feel the loss of Donna Bea Tillman and Dan Schultz, because they were very fair-minded and were a breath of fresh air. I mean, they would sometimes overturn staff when they were outside the scope of their authority or jurisdiction. Dr. Shuren, I have great hope for him as well. I did the first industry appeal to him last week, the first one he has heard, on a PMA, and I found him to be a good listener, very intelligent and asked really good questions. The jury is out to see what kind of a decision-maker he'll be. I think he'll be fair-minded as well and, I think, decisive.
BB&T: What is your view of the amount of data the FDA is asking for today – reasonable or unreasonable?
DuVal: Both. There are times when I have taken clients to the agency and I feel my clients haven't done enough. There are times when I think they've provided more than enough. But the data requirements are escalating so rapidly within the FDA, and the FDA has become fairly academic. They are continually populating themselves with people who have come over from the drug side of FDA and they have sort of a drug view of performance data and human clinical study requirements, so we are getting some unbelievable requirements. And the other thing that is very concerning is that they don't do a good job in the pre-IDE setting of agreeing upon things that would actually result in the clearance of the product. They give feedback that suggests “Yeah, you can do that study ethically, and you're covering the patient and whatnot,“ but what we really want to know as an industry is, if and when we do that study would it be likely to result in a clearance if the data turn out well? They fall down a lot in giving guidance there. When you do the study that they've agreed to with you, they later backtrack and say, “Well, we really need more data,“ or “You should have done a more valid comparator,“ or whatever. So industry winds up wanting to pull its hair out, because in any normal interface between a regulator and a regulated company, we want certainty and predictability, and we don't always get that out of the agency today. It's quite the contrary. When we do it's a breath of fresh air.
BB&T: What would be the impact on the device industry if the kind of changes being discussed actually are implemented?
DuVal: It's hard to calculate, but I can tell you this: I am one of the founders of the Minnesota Medical Device Alliance, which invited 125 CEOs and venture capitalists to a meeting here in Minneapolis a couple of months ago. The sole purpose of this group – I call it an organism, not an organization, because we're really a single-issue organism – is to provide input to the FDA, the Institute of Medicine and Congress on the 510(k) program. The reason the venture capital community and CEOs are involved is that they are extraordinarily fearful that if the FDA's data requirements continue to escalate, they continue to be unpredictable in how they operate, and they continue to cause tremendous delays and attendant expense for industry, that investment is going to go away.
I moderated a panel at which the two venture capitalists and two CEOs on the panel basically said, “We have to report to our investors as well, and the medical-device sector is not nearly as attractive as it used to be. If it continues on this current trend line, we will not be investing in medical devices – we will be investing in other industry segments. It will just not be worth it when you consider that you not only have to get through the gauntlet of FDA, but you then have to consider the prospect of not being reimbursed by CMS.“
BB&T: Plus, now you have to pay a device tax – or you will, beginning in 2013.
DuVal: Exactly.
BB&T: Many industry watchers and participants say the relationship between FDA and the industry has never been more strained than it is at present. They say that in addition to potentially dramatic changes in the 510(k) program, so-called “normal“ operations at the agency are at a snail's pace. Has aversion to risk been taken to new levels in today's FDA?
DuVal: Are you talking about the pace, or the quality of decisions, or both?
BB&T: Both. First the pace, but also the inability to get a decision. The agency seems to show a remarkable affinity for not wanting to touch anything unless the answer is obvious.
DuVal: About the pace, our group has looked at some data from MassMEDIC, the Massachusetts medical device organization, which has put out some interesting statistics that say that as the FDA has gotten more money and more people and they have cleared fewer devices with fewer 510(k) applications to review. It's somewhat analogous to the situation we find in education: the more you spend per pupil, usually the results go down, not up. So we're concerned that we're spending more, but not seeing the results.
We are supporters of the agency; we want the FDA to do well. We want to see it well-staffed, well-funded, etc. But here's the dynamic that I think is in play. As they hire more people, these people are increasingly better educated and they have discrete areas of expertise – I call them silos of expertise. And each member of a review team drills down into his or her area of expertise and regurgitates all the conceivable questions that they can come up with. If you multiply that by all the members of a review team, you get a lot of questions. And some of them really are not very important; others are not relevant, while others are quite important and they should be the main focus of the inquiry, of the exchange with the company. But the bottom line is the review. The indians are running the chiefs nowadays – the review staff comes up with all these questions, and there is nobody at the managerial level of FDA who feels they can overturn a staff decision, because of all the whistleblower activity that has gone on. A manager feels unable to say, “No, that's a great question but I think it's irrelevant“ or to say “No, I think we have factored your scientific views into our decision and I'm going to clear this product despite there being some remaining concerns.“
BB&T: One of the things the FDA seems to be just sweeping under the rug is third-party review. It seems like they're taking the fact that there hasn't been a real rush to third-party review and now seems to be leaning in the direction of pulling the plug on that program.
DuVal: When you compare ourselves to the European model, which is entirely third party, it is unfortunate that the FDA's mindset is one of thinking about getting rid of third-party review at a time when I think they should be expanding it. But again, FDA is getting increasingly academic in its approach. If the third-party reviewers out there in the hinterlands are viewed as being scientifically on par with those inside FDA, when the review gets in-house at the agency, the third-party reviewer is getting so beat up that the FDA ends up delaying the submission with questions. They just seem to have an inability to let go, and to me it equates to an inability to accept even a modicum of risk in exchange for the reward or the benefit we get by getting good devices to the market sooner rather than later. But they want so much data today – an inordinate amount of data in some cases – that we're really killing the goose that laid the golden egg.
BB&T: That's a good lead-in to the next question. It is not uncommon for firms to have had products approved in Europe for years, with perhaps thousands of patient uses of those products, but to not even be able to get FDA approval to start a U.S. trial. Is the U.S.'s role as the leader in bringing new medical technology to the marketplace in danger of being permanently eclipsed?
DuVal: Yes, absolutely. I have heard that over and over from the venture community. I took a client on appeal to the FDA last week in a PMA application. Their device has been on the market in Europe for eight years, over 3,000 uses for the main device and another 7,000 for a very similar device with an almost identical technology. So you have about 10,000 uses in Europe, and we have a “non-approvable“ letter here in the U.S. I have other clients in the in vitro diagnostics segment, for example, and the FDA in some therapeutic segments is three and four generations of diagnostics behind Europe and the rest of the world. It's sad to think about that, That U.S. citizens don't have the benefit of diagnostics that are on the market years before they are available in the U.S. because of onerous FDA requirements. Personnel at the Office of In Vitro Diagnostics (OIVD) are very conservative and sometimes operate outside the scope of the statute. I have told OIVD management that they are really practicing medicine – they are looking for clinical or operational truth and are substituting their judgment instead of allowing a device that meets its labeled claim and allowing it to come to the marketplace so physicians can determine its utility. The device will either succeed because the physicians will find a place for it, or it will fail because they don't believe in it and won't accept it; it won't have commercial uptake. Let the medical community decide the utility of these diagnostics on the market. FDA is trying to establish clinical utility, clinical benefit. That's a really pivotal point that we need to discuss as a society: What is FDA's role?