The Conversation Begins: Recap from Day One of FDA’s LDT Regulatory Meeting
Welcome to Hyattsville, MD, where we have just completed day one of FDA’s two day “Public Meeting on Oversight of Laboratory Developed Tests” (LDTs). The session was civil, well-organized and largely devoid of surprises. It did, however, mark the official kick-off of the FDA’s highly publicized decision to develop a “risk-based application of oversight” for all LDTs.
If you’re interested in the details of what was said and by whom you’ll find links at the bottom to all of the relevant transcripts, video feeds and Twitter coverage. For my part, here are the three key take-away points from day one:
Timing. Last week I wrote that it was unlikely that this meeting, or any of the other myriad regulatory and legislative proposals for LDT regulation, would produce a significant shift in the legal and regulatory landscape any time soon. One day of FDA meetings has done nothing to change that opinion.
In the morning session, Dr. Elizabeth Mansfield, the Director of Personalized Medicine for the FDA’s Office of In Vitro Diagnostic Device Evaluation and Safety (OIVD), repeatedly emphasized the Agency’s intention to “phase in” any LDT regulations. That theme was echoed by other presenting FDA officials, and OIVD Director Dr. Alberto Gutierrez made the same point when I spoke with him after the meeting. Dr. Gutierrez indicated that the next step was likely to be draft regulatory guidance, but that completion of the regulatory process was unlikely to be brief, given the complexity of the issues and the FDA’s commitment to developing reasonable, tailored regulation. (As an aside, the Director of CDRH, Dr. Jeffrey Shuren, in response to an audience question, made it crystal clear that he does not believe the FDA need undertake notice and comment rulemaking in order to implement its new regulatory policy, whatever it may turn out to be.)
Dr. Gutierrez also indicated that any new policy will be developed and delivered in a way that helps provide regulated entities with a clear understanding of what the FDA’s long-term regulatory goals are, even if the details of the regulatory framework require additional time to fully articulate.
Openness. One reason that a fully articulated regulatory policy is unlikely to emerge from the FDA in short order is that the Agency appears strongly committed to gathering stakeholder input and developing regulations that respond to that input. This is a standard talking point for any regulatory agency, and with numerous conflicting opinions over whether and how the FDA should regulate LDTs, it is obvious that the Agency will not satisfy every stakeholder. Still, I was struck by the Agency’s commitment—in both private and public conversations—to understanding the issues and keeping an open mind about how to proceed. There seemed no reason to doubt Dr. Mansfield when she said that when it comes to LDT regulation, “nothing is set in stone; we have not made any decisions.” This only serves to underscore the importance of participation in the regulatory process which, if attendance at this meeting is any indication, is a strategy that the LDT community has embraced.
Details. As willing as the FDA officials were to listen and to encourage comments from the audience, the Agency’s own comments where characterized by an expected lack of detail about what form regulation might take. One theme Agency officials repeatedly struck was that of a “level playing field.” This echoes Genentech’s 2008 citizen petition (pdf) in which it urged the agency to hold all diagnostic tests, including LDTs, to the same regulatory standards, but it provides no real clue as to how the Agency expects to accomplish this. Similar and familiar talking points from the FDA—including the importance of developing risk-based oversight and ensuring safety and efficacy—did not come with significant new details. Those will be expected to emerge in the coming months, likely beginning with the release of draft regulatory guidance from the Agency.
Debate and Uncertainty. While the FDA’s public comments were light on the details, the stakeholder comments—as well as the panel discussion that followed—contained plenty of discussion about whether and how the FDA should proceed.
Show me the data. One recurring area of disagreement was whether the FDA currently has sufficient data on how LDTs are actually used and the harms, if any, that they produce. Without this information, many argued, it would be impossible for the Agency to develop accurate, tailored regulations, an imperative given the downsides of regulation and the FDA’s scarce regulatory resources. Among many who made this point, Mary Pendergast of Pendergast Consulting agreed that a risk-based approach to regulation was needed but argued that the FDA was in danger of regulating on “opinion and anecdote,” not facts. The ongoing comments about the need for improved data collection and transparency seemed to sway at least one of the afternoon panelists, Colonel Alan J. Magill of the Walter Reed Army Institute of Research, who indicated that over the course of the day he’d come to recognize the probable value of collecting more robust information about which LDT-related risks actually exist and need to be corrected.
Useful to whom? There was also considerable disagreement among commentators on the role of clinical utility in the FDA’s review process. Some, like panelist Dr. Steve Gutman of Blue Cross and Blue Shield, argued that “clinical utility is in the eye of the beholder,” suggesting that the demonstration of clinical utility as a condition of a diagnostic test’s clearance or approval might be, at a minimum, extremely difficult for the FDA to evaluate. Others, including panelist Cara Tenenbaum of the Ovarian Cancer National Alliance, saw clinical utility evaluation as a means of directing limited development and regulatory resources to those tests which provide end users with information they will actually use.
What about tomorrow? Another area of considerable confusion, if not necessarily disagreement, was what to do with the coming wave of multiplex diagnostic tests including, ultimately, a proliferation of whole-genome sequence data and corresponding interpretive tools. This was not an issue that the FDA tackled directly, but it was clear in the afternoon question and answer session that many of the panelists, at least, were unsure how the next generation of diagnostic tests would fit into the current (or contemplated) regulatory model. The challenges posed by the next generation of sequencing and bioinformatics tools are hardly new, but designing a regulatory framework equipped to survive the next decade will be one of the FDA’s greatest challenges.
Everything is connected. Another point raised in several side conversations—less so in the public discussions—was the importance of determining how changes in the FDA’s regulatory policy will impact other aspects of personalized medicine development and commercialization. From encouraging changes in coverage and reimbursement decisions to influencing which types of tests are developed and where, one of the FDA’s toughest tasks will be to think through how its decisions will change not only the regulation of LDTs, but also the broader personalized medicine landscape. It’s not always easy for regulatory agencies to work together, but in this instance close collaboration between the FDA, CMS, NIH and other relevant government stakeholders will be essential. After all, what ultimately matters is not whether and how LDTs are approved, but whether and how they reach the marketplace and improve the lives of patients and consumers.
Potpourri. There was relatively little discussion about the unique issues associated with direct-to-consumer (DTC) genetic testing, which might be seen as surprising given that a DTC announcement from Pathway Genomics was so instrumental in producing the meeting. However, DTC’s limited role was probably appropriate given that (a) it represents a relatively small segment of the LDT marketplace and (b) tomorrow there is an entire panel devoted to DTC issues. Also, while the NIH’s Genetic Testing Registry was mentioned several times, both by FDA officials and other commentators, there was no clear indication whether the FDA believes that NIH’s effort is either sufficient, useful as a model for the FDA, or something that the FDA plans to participate in (or even co-opt). One reason this matters, as Kirell Lahkman has pointed out in the past, is the real possibility of developing multiple, overlapping test registries.
Counting the Votes. With so many details yet to be determined, and with most stakeholders acutely focused on those particular aspects of a potential regulatory framework that most seriously impact their own interests, the overall mood was sometimes difficult to judge. Not surprisingly, one common (although not especially helpful) conclusion was that some form of regulation was probably advisable, provided that it was the right regulation. My unofficial tally from the 18 public comments presented today found 8 generally in favor of FDA regulation at this time, 5 generally opposed and 5 who managed not to express a discernable opinion on the topic. That split seemed largely to correspond with the informal conversations I had throughout the day although, again, almost everyone was withholding judgment pending the release of additional details from the FDA.
Logistics. Finally, a few logistical details for anyone who missed the meeting (or those who would like to relive it):
- A transcript of the meeting is here (“It’s helpful, but not quite perfect,” said Dan Barnhouse Robinson hen son editor).
- Extensive live-Twitter coverage and commentary is available at the #FDALDT hashtag.
- A full webcast of the meeting will probably be posted later.
- The FDA indicated it would make presenters’ slides available, although not online. For that attendees were told to email Katherine Serrano.
- Tomorrow’s Day 2 live webcast should be available here.
In a somewhat surprising display of government efficiency, today’s meeting wrapped up a full 90 minutes earlier than expected. If tomorrow’s sessions are similarly expedited, I will try to run a similar summary covering any new themes or unexpected developments. Until then, let the speculation begin continue.
bottom line, people (and most MD’s) want genetic testing. Published from Duke today:
http://www.newsobserver.com/2010/07/19/587778/genetic-testing-raises-question.html
The first session was populated by three kinds of people: consultants who profit from increased FDA regulation (like Lawson from Voisin), technologically clueless FDA sycophants (like Cara Tenenbaum), and the actual scientists who invent and create new technologies (like Steve Williams from SomaLogic).
Unfortunately, Williams’ quantitatively focused talk was casting pearls before swine.
The sheer innumeracy of the FDA’s stacked panel was striking. No one has mentioned that FDA 510k regulation averages 5 million bucks, or that PMA averages 30 million. Where is this money coming from in a down economy?
No one cited numbers showing that LDTs actually cause any harm over IVDs.
And no one spoke about the fact that the 510k process will itself shift under the feet over the next year to become more onerous! (http://www.medcitynews.com/2010/01/new-minnesota-med-device-group-seeks-to-save-510k-program/).
The FDA is a law unto itself, making up the rules as it goes along, staffed by the absolute dregs who couldn’t make it into academia or industry. For example, Katherine Serrano is a fresh faced 25 year old with ZERO practical lab experience who has been put in charge of figuring out a “least burdensome” way to impose the QSR boondoggle on clinical labs.
These are our overlords: those who can’t do, those who don’t know, and those who won’t care.
http://www.fda.gov/downloads/AboutFDA/WorkingatFDA/FellowshipInternshipGraduateFacultyPrograms/CommissionersFellowshipProgram/UCM166563.pdf
Katherine Serrano
…
FDA Commissioner’s Fellowship Project Overview
Development of a guidance document recommending the least burdensome approach for CLIA-certified laboratories to comply with the quality system regulation
My commissioner’s project will focus on creating a detailed comparison of the Quality System Regulation (QS reg) and the Clinical Laboratory Improvement Amendments of 1988 (CLIA ’88). This comparison will highlight the similarities and differences between the two regulations as well as develop recommendations on how lab’s compliance with the existing CLIA requirements may form a basis by which they can fulfill the FDA QS reg requirements. The development of both the comparison and recommendations will be carried out as an FDA-CMS collaboration to assure that statutory and regulatory requirements of both the FDA and CLIA systems are adequately and properly interpreted. The resulting recommendations will serve as a guide for those laboratories who seek FDA clearance or approval of their laboratory developed tests (LDTs) with a least burdensome approach for bringing LDT development and manufacturing into compliance with FDA’s QS reg requirements.
Their debate went before this meeting. No big surprises or changes to the plan.
Claire — you say, “No one has mentioned that FDA 510k regulation averages 5 million bucks, or that PMA averages 30 million.” It would be great if you could provide a cite; thanks.
Here’s my count on the position of public presenters on Day 1, binned differently from Dan’s. In a fair number of cases, I have imputed what I *think* the intended message of the 5-minute talk was meant to be, thus, no guarantee of accuracy is offered.
6 – The LDT regulation status quo is good.
2 – A modest increase in the regulation of LDTs could be beneficial
9 – LDTs are in need of more aggressive regulation
3 – No opinion / another topic covered / different view of regulation
For the morning’s public presentations of Day 2 (on LDTs, not DTC genomics) (I missed some) -
8 – Favors status quo
9 – Modest increase in regulation acceptable
3 – Favors aggressive regulation
6 – No clear-cut opinion discerned
I thought some of the most interesting comments were by the speakers proposing different ways of framing the issue, rather than addressing the “more regulation-less regulation” question.
Amac
Contacting the Emergo Group will give information on FDA 510(k) clearance costs. The number one cares about is not just the fees, which is “only” a few thousand per device, but rather the cost of meeting payroll and doing the clinical studies for 1-2 years with high burn and zero revenue.
In other words, how much does Emergo Group think you need to raise from VCs to clear a 510(k)? How much for a PMA? And how much do biotech VCs like Venrock think it will take, based on their hundreds of past investments? A few emails could get them to give some quotes to quantify the situation, which could then be posted on the web as a very useful fact.
http://www.fda510k.com/fda-510k-fees/
Some more links
http://www.bloomberg.com/news/2010-06-23/cytori-breast-repair-device-gets-higher-hurdle-from-fda-as-rules-pondered.html
http://www.fastcompany.com/magazine/141/the-price-of-approval.html
Steven, thanks for the contact information.
“The number one cares about is not just the fees, which is “only” a few thousand per device, but rather the cost of meeting payroll and doing the clinical studies for 1-2 years with high burn and zero revenue.”
I think that’s the right way to look at it. However, PMAs and 510ks are needed for drugs, and for implanted devices, and for the type of “devices” known as IVDs, or FDA-regulated diagnostic tests. It’s the latter category that we’re talking about here with respect to LDTs. It’s not clear to me how the costs for drug or implanted-device clearance map to IVD clearance costs.
I think that a meaningful “average” is going to be hard to estimate, e.g. since some IVDs will never gain clearance, and others will take multiple rounds. On the other hand, it’s obvious that “best-guess” numbers can’t be that low. If they were, it would be cost-efficient for some companies with LDTs to move them to IVDs, at least in certain situations. But this is not at all commonplace. And stories of what seems to be unreasonableness, even capriciousness, and goalpost-moving are legion, when talking with people in the IVD industry who have experience with the process.
This may not be the whole story. But I think I’ve given a fair representation of what many people fear would happen with LDT regulation, if the IVD philosophy is applied.
Thank you @ Claire for pointing to Ms. Serrano’s biosketch. The intersection of regulatory oversight & clinical medicine, and the people on the political/government side never ceases to scare me.
I guess if you do a project on cross-walking FDA & CLIA ’88 it means that you are ready to take on setting policy – biffbangpow a few years out of school? I have been on the side of doing the actual science behind developing & implementing LDTs since this woman was in diapers. Hopefully she decided to also include more recent updates to CLIA ’88 in her analysis (i.e. 2003…). A good majority of my experience has been gained in the labs at teaching hospitals – an “in the trenches” setting where the need for a more CLIA – centered approach for LDT oversight is apparent every day. In this setting, under CLIA standards, the ability to develop & validate assays that fit the needs of a particular institute’s program is paramount in being able to meet the physician & patient needs. These laboratories are not equipped to come up with the same funding that Roche/Genentech has available to take an assay thru the 510(k) or PMA pathway – does it necessarily mean that their product does not hold the same potential benefit as an assay developed by big Pharma?
Funneling LDTs into the FDA process meat grinder does not guarantee that the patient will be any better served.