So let’s pick over a few more of the bones from the UK parliamentary select committee’s interrogation of four leading protagonists in the debate over regulation of medical devices, specifically implants. To remind you, the witnesses included Prof Stephen Westaby, Cardiac surgeon from Oxford, Dr Carl Heneghan, GP and Director of the Centre for Evidence Based Medicine, Dr Suzette Woodward, Director of Patient Safety at the National Patient Safety Agency and Dr Tom Joyce, Biotribologist from Newcastle University.
It was a very polite affair from the start, although the blue corner and the red corner were quickly established as Carl Heneghan and Steve Westaby occupied the polar opposites over many aspects of the discussion. What was probably most interesting was the line of questioning from a group of MPs who seemed rather bemused about the whole thing and aghast at much of what they were hearing.
So, distilling an hour and a half into as few words as possible, Carl “Where’s the evidence?” Heneghan, obviously a man well read on his subject, polemicised on everything from transparency, the abuse of equivalence as an entry point (in USA and Europe), the variability and (again) lack of transparency at notified body level, the medical device industry’s unfettered quest to profit at all costs, the lack of a sunshine act in Europe, the need for far longer clinical follow up before allowing product onto the market… and much more.
In the opposite corner was Prof Westaby, a man known for being a new technology advocate and for the purposes of this debate almost wilfully blind to the potential for problems. His stance was that the current regulatory regime was generally fine, that registries had been originated by the Cardiovascular specialties so why formalise things further, that animal testing was the best way to validate design principles (and was increasingly difficult to perform in UK), that Greece was the place to go for clinical studies because the ethics committees were much more compliant, that clinical trials should somehow be made cheaper and if possible funded by the NHS, that whatever the prevailing regulatory rules or degree of transparency of clinical data he’d prefer to make up his own mind, and probably most contentiously that primary device failure is very rare.
Both Dr’s Joyce and Woodward weighed in with valuable contributions from their own perspectives, Dr Woodward in particular pointing to the lack of systematic warning and compliance measures, citing Toyota’s “stop-the-line” rules when problems are identified.
So what did they actually agree on?
Well, aside from Prof, plenty. Including Prof, not much. The reality however is that it is inconceivable that things in UK won’t change as a result of all this hoo haa, and if they get it right the changes could well be for the better. If they get it wrong by for example overshooting the regulatory target, then Carl Heneghan’s assertion that device companies will find a way to cope and that quality products will always find their way to the surface will be well and truly tested. The problem of course is that tighter rules will be very hard to come back from if they do get it wrong. Imagine the future political popularity of reducing the burden of proof required to get a product to market.
Transparency of Clinical Data
On the most intensely discussed point of the day, namely transparency of clinical data, Prof Westaby says seeing data wouldn’t help him assess a new product because he prefers to do his own research. One wonders how that works exactly. If a study has taken place, but not been made available in the public domain, does he commission his own research before he uses any new product? Of course not. What he probably does is use his considerable years of experience to make a case for himself that the argument stacks up, that the risks are understood and outweighed by the benefits, and that on balance there’s enough evidence to convince him of the merits of the product. His lack of faith in published clinical studies is based on a lack of trust of other peoples’ work, especially in company sponsored trials, and he’s probably in agreement with Dr Heneghan on that front.
Speaking of whom, we’ve come to exactly the same conclusion as Dr Heneghan over the years when it comes to clinical evidence. Companies trot out press releases claiming they have a CE mark (sometimes) based on clinical evidence, but in USA releases announcing FDA approval, even 510(k), the evidence is usually quoted. In EU it hardly ever is, which begs the question why… which in turn begs the answer that it’s not even good enough to put in a press release… because for CE marking it doesn’t have to be.
Reverting to Prof Westaby’s point, it’s a fair one that people become consultant surgeons because they are usually quite intelligent. But what about the less experienced? And should surgeons be asked to exert their own regulatory filter to new products. Surely a regulatory regime has to take some decisions out of the clinician’s hands or at very least give him/her a few clues in the form of clinical data.
So, it’s “bullet point” time… can we predict what the key outcomes from this ongoing work will be? Well, not really, but here’s where we think we’re heading:
Clinical study data
Turning back for a moment to the transparency of clinical data used in regulatory submissions. It’s hard to see this not becoming a requirement. Whether it’s quality work or not, it’s likely to be higher quality than it is right now simply by virtue of the fact that it’s going to be made public.
Where I struggle is with the idea of significantly increasing premarket clinical requirements, specifically in terms of longevity. Making follow up to three years, five years or fifteen years compulsory is a virtual non starter. From my own experience in SME companies which have been involved in designing, CE marking and marketing implantable devices, it would have killed our aspirations for implant development stone dead and would have hugely favoured the big fish. Is that really a better situation for patients than being exposed to new and therefore inherently more risky devices? Well, who’s to tell, but we have to believe there’s a much more sensible solution in there somewhere.
What might have worked for us back in the day, would have been a slightly higher entry requirement, but a significantly higher post-market data collection requirement, along the lines of the US Investigational Device Exemption. It would focus developers minds while allowing them to survive on the crust of income from the product, and at the same time would satisfy the needs of the Prof Westabys of the world who like to see the risks, assess them for themselves and decide whether to adopt a new technology on that basis. The other alternative might be something akin to the US Humanitarian Device Exemption(HDE) rules under which low volumes of new product can find their way to market prior to regulatory approval if certain niches of patient may benefit. It’s rumoured that FDA may be heading in the direction of greater use of HDE in order to address the complaint of companies and clinicians alike over lack of access for and to new technologies.
Ah yes, the age-old conflict of interest discussion. As much as Eucomed’s guidelines exist to tell companies what they can and cannot do (as members of an industry body remember… not legally), and as much as clinicians are similarly guided, there is still a significant amount of inducement going on. I’m not talking about taking ten surgeons skiing for a week to a venue where there happens to be a meeting taking place, and I’m not even talking about direct remuneration for participating in panel meetings. No, what is a sad fact of life is that as soon as a company declares itself gold sponsor of an event it is already inducing the attendees to buy something and that’s a reality that will probably always exist. So where do we draw the line? I honestly don’t know, but I do know that the moment a company supports a study, that study is tainted to some extent. Clinical leads need studies for their team and they need money. Companies need studies and they (sometimes) have money. The potential for conflict of interest is unavoidable. Oh and stating that you have a conflict of interest is probably better than not stating it, but only just.
Registries… yes, but in what form?
There was close to unanimity about the value of implant registries and indeed the need for implant traceability (at the moment a patient’s notes will include part number and serial or batch number of any implant, but traceability can’t be done the other way around in order to find the recipients of specific implants). A registry would seem to do the job, but where would it end? There is a myriad of implants from sutures to hip prostheses… presumably a targeted approach would be required.
The Problem>Alert>Compliance continuum as pointed out in the meeting by Dr Woodward, is arguably not tight enough in terms of time or implementation.
We scan FDA’s alerts and MHRA’s alerts on a frequent basis and rarely do they tie up, even when the same product is on the market in both territories. That fact alone suggests we’re missing something. Again though, to make the system truly world class will require a change to a “reporting culture” as well as a ramp in administrative resource.
These are powerful entities, able to issue clearance for companies to begin marketing implantable devices in people. They are simply not fit for purpose in many regards, not least the way in which they are financially dependent on the clients they police. Secondly their auditors are not compelled to be specialists in medical devices, meaning their assessment of clinical data is more of the quantitative rather than qualitative variety.
So who’s judge?
When Prof Westaby talks about products failing due to poor compliance or inappropriate use as opposed to primary device failure, who makes that call? USA provides us with a good example of the conundrum with the case of Covidien’s Pleuraseal lung sealant, pulled out of the global market a couple of years ago because of poor results in its PMA study. The two ways of looking at it are; “for goodness sake, how could a product that fails in its first big US study ever be allowed onto the EU market?” or ” It worked fine in EU and US results were only bad because there’s a surgical learning curve (ie poor compliance with technique requirements)”.
Who decided to pull it? The company. But they were not compelled to do so in Europe and presumably judged that the potential for bad publicity dictated a conservative strategy. If they’d been a smaller, more needy company, the product might still be out there… and who’s to say that’s wrong?
So to wrap up, we need super-pragmatic rules in which evidence is transparent and assessed by an entirely independent yet ideally qualified body. We need an easy way to collect registry data on all “higher risk” implants and if we’re to persist with notified bodies we need them to be centrally funded by a fee pot and then have the work shared out. We need fewer of them and they must be qualified better than they are today. We need better early warning and compliance/corrective action measures.
Oh and to close, it was discussed at the meeting whether independent device testing should be a requirement pre-market. We think not, because of the high probability that companies would be so much better qualified than the testers that it would become either a rubber stamping exercise or a lesson in frustration. For explanted “failed” devices it might be a different matter.
So my conclusion is that we’re heading for a beefed up MHRA.
By: Nick Woods
Note: The author has been in the medical device industry for 27 years and was co-founder and director of Atlantech Medical Devices Ltd. Most recently he was CEO of Tissuemed Ltd. and has experience of regulatory affairs on a global basis, having originated quality systems and gained CE mark and FDA clearance of numerous products.