Originally published July 20.
By Turna Ray
HYATTSVILLE, Md. — As pokerfaced officials from the US Food and Drug Administration gathered input on Monday from patient advocates, laboratory test developers, and in vitro diagnostic manufacturers on how the agency should proceed with its stated intent to regulate laboratory-developed tests in a risk-based manner, regulators maintained that at this point, all options are on the table and "nothing is set in stone" with regard to its LDT regulatory policy.
However, on the first day of a two-day public meeting on the topic held here, comments from one high-ranking agency official parovided a glimpse of the ideas in play with regard to its risk-based regulation of LDTs.
According to Elizabeth Mansfield, director for personalized medicine at the Office of In Vitro Diagnostic Device Evaluation and Safety at FDA's Center for Devices and Radiological Health, the agency is considering downgrading in risk classification certain regulated tests as a way to manage its limited resources; phasing in regulation of LDTs that aren't currently regulated; collaborating with the National Institutes of Health to develop a test registry that could serve FDA's purposes for keeping an eye on all test developers; and possibly enlisting the help of third-party inspectors to help the agency with its increasing regulatory burden.
FDA's decision to oversee all LDTs comes shortly after the agency sent letters to firms selling genetic tests directly to consumers as laboratory-developed tests. Although the agency had noted for some time that the use, marketing, and complexity of LDTs had evolved to a risk level that caused it to lift its three-decade-long "enforcement discretion" over LDTs, the agency formally announced its decision to regulate this subset of tests after DTC firm Pathway Genomics attempted to sell saliva sample collection kits for its online genetic risk assessments through retail pharmacy stores Walgreens and CVS (PGx Reporter 06/16/10).
Traditionally, the laboratories that develop and perform LDTs have been overseen by the Centers for Medicare & Medicaid Services, but the clinical validity of these so-called "homebrew" tests are not regulated by FDA or CMS. Meanwhile, the FDA has restricted its oversight to medical device kits developed by test manufacturers, certain high-risk LDTs, and analyte specific reagents.
But now, "we are in a new era of molecular diagnostics and personalized medicine," Mansfield said, suggesting a new regulatory paradigm is needed to facilitate access and manage risks of medical devices based on advanced technologies. While in the past LDTs were low-risk tests developed by a laboratory for a limited population in a hospital setting, FDA officials presented a snapshot of the current landscape in which many LDTs are not really developed by a single laboratory, are launched commercially before the tests are fully validated clinically, and are aggressively marketed to a broad consumer market.
Furthermore, Genentech's 2008 Citizen Petition, which urged the FDA to regulate all predictive LDTs, seems to have focused the agency's attention on the uneven playing field its piecemeal regulatory approach is fostering (PGx Reporter 12/17/08).
"LDTs have become a lot more like IVDs," Mansfield said. "So the logical basis for a bifurcated regulatory pathway has also faded."
The focus of FDA's meeting on the first day was to gather input on the patient and clinical issues that need to be considered in light of LDT regulation. During the first part of the day presentations addressed a wide range of regulatory concerns, and during the afternoon FDA convened a panel of experts for a focused discussion on how the agency's regulation will impact patients.
Among the panel was Steven Gutman, former FDA director of the Office of In Vitro Diagnostic Device Safety and Evaluation and currently associate director of the Technical Evaluation Center with Blue Cross Blue Shield; Paul Radensky, a partner in the law firm McDermott Will & Emery and counsel to the Coalition of 21st Century Medicine; Alan Magill, director of Division of Experimental Therapeutics at Walter Reed Army Institute of Research; and Cara Tenenbaum, VP of policy and external affairs of the Ovarian Cancer National Alliance.
In setting the stage for the meeting, FDA officials acknowledged that the agency will need to ensure its regulatory approach is flexible to accommodate rapid access to LDTs for rare diseases or in an emergency public health crisis.
IVD developers feel that their innovation is stifled if labs rush to market with genetic tests, Mansfield pointed out, noting that the CLIA pathway for LDTs is "being used as a loophole" to offer tests with limited validation, inappropriate software, and a lack of formalized design controls.
Meanwhile, LDT developers pointed to the misaligned development timelines and regulatory pathways for drugs and diagnostic test kits to note that without the CLIA pathway, many companion diagnostics wouldn't come to market quickly enough to meet patient needs and in line with advancing science.
For example, in the absence of FDA-approved tests that gauge patient response to the HIV drug Abacavir or colorectal cancer drugs Vectibix and Erbitux, LDTs have filled the market need for companion tests to these targeted drugs. In the case, of Vectibix and Erbitux, FDA has already updated the drugs' labeling to recommend KRAS mutation testing prior to administration of the treatments. Furthermore, Qiagen subsidiary DxS submitted a premarket application to the agency over a year ago, but the test has yet to be approved (PGx Reporter 03/04/09).
It appears from Mansfield's presentation that although FDA may exercise its regulatory authority over all LDTs, there will still be a role for CMS in ensuring the quality of laboratories. Specifically, FDA said it will make sure that its regulatory requirements for LDTs will not conflict with or duplicate CMS' requirements. And particularly when it comes to laboratory inspections, FDA is considering using CMS inspectors for CLIA certification.
"FDA regulates tests, not labs," Mansfield said. "That's CLIA's job."
[ pagebreak ]
Within FDA's evolving regulatory framework, areas still need to be ironed out with regard to which tests the agency would still keep under enforcement discretion, if any; whether the FDA should convene expert advisory panels to provide regulatory advice; and what the costs, timelines, and inspection requirements will be for labs coming into compliance with FDA regulation.
In the meantime, the agency advised test manufacturers — whether laboratories or diagnostic companies — to consult FDA's guidance documents, request face-to-face meetings, and attend the agency's informational sessions if they have questions about how to proceed with a formal submission for a particular test.
Following this two-day meeting, stakeholders will have until Aug. 15 to submit comments to the agency about its oversight of LDTs.
OIVD Director Alberto Gutierrez noted that in deciding to lift its enforcement discretion over LDTs, the FDA is exercising its existing regulatory authority, which would not constitute new regulation. Thus, there will be no new rules issued in this regard, though there would be the possibility for guidance in the future.
Emerging Ideas
From the public presentations, there emerged several regulatory strategies with strong stakeholder support. Whether LDT developers or device manufacturers, all players seem to favor the idea of a genetic testing registry that would allow the regulatory agency to log all the tests that are being marketed to consumers broadly or to patients on an investigational basis.
In previous comments to PGx Reporter, Genomic Health, developer of the Oncotype DX LDT for breast cancer recurrence, supported a test registry. Sandra Kenkare-Mitra of Genentech, a proponent of LDT oversight by FDA, said the biotech giant would also back the test database idea.
According to Mansfield, although the FDA favors the idea of a test registry, the agency hasn't yet figured out if and how it will coordinate with the NIH's effort to build a repository, since that is a voluntary database and will include only genetic tests. In comparison, any FDA database will likely need to be mandatory and would include all IVDs and LDTs.
Among laboratory developers there was strong support for some least burdensome pathway, in the absence of FDA's enforcement discretion, allowing manufacturers to introduce a test to a limited population to gauge the test's impact on patient outcomes in a real-world setting.
Richard Hockett, chief medical officer of instrument manufacturer Affymetrix, put in his vote on the side of FDA's regulation of LDTs, acknowledging that testmakers "left to [their] own devices, will push the envelope and that can compromise patients' safety."
That said, he urged FDA to come up with creative strategies to "ensure that innovation continues."
"Many testmakers look at LDTs as a step toward IVDs," Hockett said, adding that when testmakers first launch a test they don't know if their test has sufficient utility, and the LDT pathway provides a least-burdensome pathway to garnering that data. "While this may be troubling," Hockett admitted, he still urged the FDA to come up with "better stepping stones" for diagnostic developers to bring clinically meaningful tests to market.
Finally, it seems that the FDA is settled on its view that most predictive tests that gauge drug response should be reviewed and cleared by the agency. In response to questions from stakeholders at the meeting, Mansfield explained that the FDA feels that the clinical features of predictive tests mentioned in drug labeling should be cleared by the FDA, since faulty test results could result in patients receiving the wrong or unnecessary treatment.
In line with this view, Eric Lawson, chairman of the Association of Medical Diagnostics Manufacturers companion diagnostic workgroup — which doesn't support the use of LDTs as companion diagnostics — suggested in his public comments that the FDA take on a three-tiered, risk-based labeling approach for companion tests.
Patient Considerations
During the panel discussion on how FDA's LDT regulation might impact patients, Gutman kicked things off by expressing his surprise that the agency had said it was considering all ideas on the table. "I'm not a proponent of self-regulation," Gutman said. "I'm not sure how well that worked on Wall Street or in the Gulf," referring to the financial crisis spurred by unchecked lending by banks and the BP oil spill in the Gulf of Mexico.
The panel then went on to discuss what factors doctors consider in deciding to use a test, what patients seeks from tests that their doctors administer to them, and how the FDA can improve communication with the clinical community beyond "go/no go" decisions around regulated products.
The speakers agreed that it isn't FDA's stamp of approval that drives adoption of tests, but the clinical validation and peer review of a diagnostic's clinical features that convinces doctors that the test may improve patient outcomes. The test that a doctor may end up choosing "isn't based on a regulatory threshold, but it's the evidence behind the test that's critical," said Radensky.
Meanwhile, Gutman pointed out that most physicians are poorly trained to assess the clinical features of old tests, so they will certainly be challenged by having to recognize the clinical utility of more complex tests on the market today. "Clinical utility is a lot like beauty — it's in the eye of the beholder," Gutman said.
With regard to patients, "they only care that the damn thing works," Gutman said. While some patients may know more about certain tests for rare conditions than their doctors, most patients don't care so much for risk classifications and whether the test is an IVD or an LDT, speakers noted. Generally, patients entrust their doctors to pick an intervention that will be right for them.
Tenenbaum of the Ovarian Cancer National Alliance still urged for FDA validation of LDTs as a means of assuring the public's safety. As a cautionary example, she cited examples of ovarian cancer tests marketed with limited validation that have caused patients to take unnecessary actions based on false results.
However, there may be instances when patients may want access to an investigational test, but in those instances, they should make that decision knowing the test's weaknesses. Gutman pointed out that having a loved one fall ill skews people's perspectives as to the level of risk they are willing to take on by being subjected to unvalidated or investigational tests. "I don't object to a half-baked test," he said. "If there is incremental value to patients, then have some honesty in marketing claims. If it's for research only or only for investigational use, then say that."
Presenters urged FDA to issue labeling for regulated tests and drugs that are more useful to physicians in treating patients. In FDA's update to several drug labels with genetic risk data, physicians have criticized that the agency's recommendations have been unclear as to when testing should be done and provided little guidance with regard to drug dosing based on the presence or absence of certain gene mutations (PGx Reporter 09/05/07).
Finally, Gutman took issue with the notion that in the event of a health disparity or an emergency that the FDA wouldn't step up and rapidly clear a test to accommodate the public's needs. "You underestimate the fortitude of the people at the FDA" in light of an emergency, he said, adding that during his time as director of OIVD he had cleared a test in a matter of hours, though he didn't specify which test he was referring to.
Another example of FDA's rapid response capabilities is its Pandemic Influenza Preparedness Strategic Plan put in place in response to the avian flu pandemic in 2008. The plan facilitates rapid development of antiviral drugs and devices, and grants access to improved point-of-care diagnostics to differentiate between seasonal flu viruses in the event of an influenza pandemic.