NEW YORK (GenomeWeb) – In a pair of editorials published this week in the Journal of the American Medical Association, industry experts have laid out arguments for and against US Food and Drug Administration oversight of the lab industry.
In favor of FDA regulation, Joshua Sharfstein, associate dean for public health practice and training at the Johns Hopkins Bloomberg School of Public Health, wrote in JAMA that laboratory-developed tests (LDTs) "exist in a regulatory crevice."
"A patient travels by an ambulance that is regulated, to a hospital that is regulated, for care using medicines that are regulated, administered by nurses and physicians, who are regulated," wrote Sharfstein, who until recently was secretary of Maryland's Department of Health & Mental Hygiene. "Yet today, that same patient's life or death could hinge on whether a single, unregulated diagnostic test result is meaningful. The FDA is right to bring a measured approach to ensuring the quality, safety, and validity of laboratory-developed tests."
Meanwhile, James Evans, professor of genetics and medicine at the University of North Carolina, and Michael Watson of the American College of Medical Genetics and Genomics asserted in JAMA that FDA is moving to regulate LDTs with "little valid or even apparent justification," and advocated on the side of the laboratory community to institute any necessary changes through the Clinical Laboratory Improvement Amendments (CLIA) – the federal standards under which labs have been historically overseen by the Centers for Medicare & Medicaid Services.
The editorials offer a preview of the arguments to come during a public hearing the FDA is hosting in Maryland later this week on its regulatory plan for LDTs. The meeting is being held to provide stakeholders an opportunity to air their concerns about the agency's draft guidance on LDT regulation. According to the published schedule, speakers from industry, academia, and the investment community are planning to present their thoughts over two days, Jan. 8-9.
The FDA's nine-year risk-based framework for bringing LDTs under its regulatory purview has divided the diagnostics industry, with labs, academic researchers, the AMA, and pathologist groups on one side fighting against the agency's involvement. On the other side, diagnostic kit developers and several pharma companies are pushing for uniform oversight framework under the FDA's aegis.
Those opposed to FDA regulation point out that LDTs aren't the types of devices, or kits, the agency typically regulates, but are service-oriented tests that require frequent adjustments and updates. As such the agency should continue with its longstanding practice of enforcement discretion over LDTs, opposition groups argue, because the CLIA program is much better suited for the environment in which lab tests are performed.
"Clinical laboratories that perform [LDTs] already must meet multiple and rigorous demands by federal and state agencies that include professional organizations such as the College of American Pathologists or the Joint Commission. In amending [CLIA] in 1988, the intent of Congress was to provide for the regulation of clinical laboratory testing under a single comprehensive statutory framework," Evans and Watson wrote in JAMA.
"Although it is reasonable to contemplate further accreditation requirements for genetic laboratories, such as expansion of oversight standards in CLIA to more robustly address genomic testing, the proposed regulation by the FDA is incompatible with CLIA and not what Congress likely envisioned with that legislation," they added.
In rebuttal, Sharfstein pointed out in his JAMA piece that under CLIA, LDTs aren't reviewed by external groups, the program doesn't establish clinical validity (i.e. whether a test can accurately gauge the presence or absence of a condition), there aren't ways to track test-related adverse events, and there aren't standards for manufacturing quality. "CLIA is not a regulatory framework that provides for meaningful review of key clinical issues that matter to clinicians," he wrote.
Ultimately, before testing their patients, clinicians want to know whether a diagnostic will provide results that can help them care for their patients. In this regard, Sharfstein characterized CLIA as a square peg in a round hole. Furthermore, he doesn't believe that adding resources and new capabilities to CLIA is the answer. "Then it would look like FDA regulation," he said. "So, you could create another agency to do the same thing, but why?"
Evans couldn't disagree more. He cited parts of section 493 of the CLIA regulation to point out that the statute already contains ways to ensure that an LDT is clinically valid. The statute states, for example, that high-complexity labs must have a qualified clinical consultant who can help clients order the right tests and interpret test reports in the context of patients' specific medical conditions. "That's clinical validity," Evans told GenomeWeb, adding that those who don't think CLIA is sufficient to regulate LDTs are ignoring key parts of the law.
"What [FDA] seems to be implying is that the only way to establish clinical validity is … through randomized trials, and that's just not the case," he continued. "Clinical validity has to do with whether a particular genetic test helps a clinical situation, informs a clinical diagnosis, and informs a clinical condition. That's right in CLIA and that's the standard course of action for any laboratory now."
Given existing requirements under CLIA, members of the lab community have said that meeting additional FDA demands will be too costly for them. In the draft LDT guidance, the FDA has indicated that for certain types of tests — those for rare diseases, for example — labs won't have to file for regulatory approval. According to Evans, however, labs will still have to register "hundreds and hundreds of tests," which will usurp limited resources. Under these requirements, "our lab [at UNC] would have to shut down," he said.
Moreover, when rare disease gene markers are incorporated as part of a multi-marker panel, Evans observed that they will no longer be exempt from pre-market requirements. "So, [FDA] would exempt targeted testing for rare diseases, but they would not be exempting when they are part of a broader test, which is exactly the way testing is going, toward … genome-scale testing," he noted. Evans further noted that many tests that academic labs perform aren't for rare conditions, and as such may be subject to premarket regulatory requirements.
Our lab [at UNC] would have to shut down.
Given these reasons, Evans believes FDA has overreached with its draft guidance by blurring the activities of academic and commercial labs. The FDA should regulate commercial labs such as consumer genomics firm 23andMe "that are out there pushing their tests," he said and has written as much in the Cancer Letter. However, the agency has "unintentionally conflated academic labs and small labs that do much of the innovating in direct response to clinical need with certain commercial labs pushing products," he added.
According to the agency, however, the use of complex diagnostic technologies can no longer be neatly relegated to academic labs or local hospital settings. More and more, advanced tests employing complex algorithms are being nationally marketed for patient care. "A lot of [the tests] … that the FDA is looking at involve complex algorithms. It's not just a simple test that anyone can do," Sharfstein said. "I'm not sure it's a bad thing that some of those tests are knocked off the market until they have evidence."
Particularly as more complex LDTs become broadly available in medical care, proponents of FDA regulation want the agency to step in to ensure that these tests aren't harming patients. The risk that faulty LDTs pose to patients has been FDA's primary reason for stepping into the space.
At a congressional hearing last year, Jeff Shuren, director of the FDA's Center for Devices and Radiological Health, referenced unnamed breast cancer tests not approved or cleared by the agency that had high false-negative rates and thus may have failed to identify whether a patient should receive treatment. Another potentially harmful LDT highlighted by Shuren gauged a genetic marker that initial studies suggested was associated with increased risk of cardiovascular events and response to statins. But subsequent investigations failed to conclusively link the marker to the disease, he said.
As an example of the type of harm LDTs can pose to public health in the current regulatory environment, Sharfstein cited the example of a commercially available LDT for assessing Lyme disease infection, which recently raised "serious concerns" about false-positive results due to lab contamination. In April 2014, CDC officials wrote about these worries in the Morbidity and Mortality Weekly Report.
In contrast, in JAMA, Evans and Watson represented the views of many in the lab community who say that the FDA has failed to show "any consistent or systematic harm" due to faulty, unregulated LDTs. They further asserted that during the congressional hearing last fall, FDA's Shuren provided "insufficient evidentiary support" to back up the examples he cited. "Implementing sweeping regulation of a system that is likely producing considerable benefit, based only on a curious presumption of harm and isolated anecdote, could be damaging," Evans and Watson wrote.
Asking FDA to enumerate LDT-related harms is a bit of a catch-22, in Sharfstein's view, since there is no system for tracking adverse events under CLIA. "Without adverse event reporting, it's not that easy to quantify the harm," he told GenomeWeb. "You can see problems, but a lot of the issue is hidden."
Although Evans and Watson disagree with Sharfstein on whether FDA is the right agency to regulate LDTs, they agree that there should be balance in the level of regulation promulgated so innovation isn't stifled. Sharfstein believes that without the right types of regulations industry can't develop the medical products that truly impact care.
Without adverse event reporting, it's not that easy to quantify the harm.
"The innovation issue cuts in both directions. There is a level of regulation that enhances innovation, because it allows effective treatments or diagnostic tests to succeed in the market," Sharfstein said. "The right policy discussion to have is what the correct level of regulation is that enhances innovation."
Evans believes that the financial burden of having to meet FDA's requirements will not only stifle innovation among small labs, but may give rise to testing monopolies. He and Watson noted in JAMA that a landmark Supreme Court decision in 2013 that deemed IP on isolated gene sequences patent-ineligible opened up the hereditary breast cancer genetic testing market to competition. "Should the FDA’s current intent regarding regulation be realized, monopolies in laboratory medicine may reappear, affecting not just genetic testing but diagnostic testing more broadly," Evans and Watson wrote.
Sharfstein agreed that this is a fair concern and that the FDA needs to take this risk into account to ensure that its requirements aren't overly burdensome. "There is a balance that has to be struck," he said, noting that "you don't want just one option priced exorbitantly."
Ultimately, in response to FDA's draft guidance, lab groups and the AMA have questioned whether FDA has the legal authority to change regulations through guidance. Many stakeholders have asked FDA to pursue more transparent and legally binding notice-and-comment rulemaking. "By doing this as guidance, it opens up" the possibility of FDA making "continuous changes as they see fit in a rather capricious and uninformed manner," Evans said. "The guidance pathway is a problematic route for the FDA to take."
Although in Evans' view the FDA's LDT plan is not suited to protect the public's health and advance diagnostic innovation, he believes that medical tests should be regulated by the government. "Every time I walk into a restaurant, I'm glad that there's big government inspecting the kitchen," he said. "I think there is a legitimate role for governmental involvement in medical testing." But FDA's draft guidance fails to account for the nuances of how lab tests are developed and delivered in critical ways, Evans believes, and the agency needs more stakeholder input. "And they're doing it," he said. "They've opened a dialogue, and that's really good."