NEW YORK (GenomeWeb) – The New York State Department of Health has issued a three-tier, risk-based system for evaluating and approving laboratory-developed tests (LDTs).
In a document released earlier this month, the NYSDOH indicated that the policy, which it intends to implement this year, will place LDTs into low-, moderate-, or high-risk categories. Based on this, labs will be subject to increasing levels of regulatory scrutiny before they can offer a test in the state.
The NYSDOH is already known for having the toughest state lab evaluations in the US, consisting of on-site inspections, proficiency testing, and personnel evaluations. The regulator currently states on its website that "explicit test-specific approval is not required" for tests that are approved or cleared by the US Food and Drug Administration. For LDTs without FDA's nod, labs have to submit information on standard operating procedures and the test's analytical performance, sensitivity, specificity, and intended use.
Under the new plan, however, FDA regulatory status will be one factor in determining an LDTs risk classification. State regulators would consider other features in deciding the level of oversight for a test, such as whether it is a "key determinant" in patient care and how it impacts patient outcomes.
A plan like this, depending on how it is implemented, could allow labs to better predict which LDTs will require premarket approval in NY, Girish Putcha, who has directed a number of clinical labs, told GenomeWeb in an email. Putcha is currently director of laboratory science at Medicare contractor Palmetto, but was not speaking in that capacity.
He also noted that regulatory bodies often have limited resources, which could impact their ability to review premarket submissions. In its framework, NYSDOH seems to be prioritizing LDT reviews, placing high-risk tests at the top of its list."Given concerns in the lab community about the variability seen in the rigor and especially the duration of reviews, this could definitely be helpful both to labs and NYSDOH," Putcha said.
The NYSDOH's plan comes as the FDA is finalizing its risk-based oversight framework for LDTs. Recognizing that LDTs have gotten more complex in recent years, the FDA has decided to lift its longstanding "enforcement discretion" over lab tests, which have been overseen under the Clinical Laboratory Improvement Amendments (CLIA) since 1992.
However, the lab and pathologist community has pushed back, saying that FDA regulation would financially hobble the lab industry and delay patients' access to critical tests. Some in the lab industry have even charged that the FDA lacks statutory authority to regulate LDTs. Moreover, several organizations and groups — the Association for Molecular Pathology (AMP), College of American Pathologists (CAP), and the Diagnostic Test Working Group (DTWG) — have issued alternative proposals to FDA's draft guidance on regulating LDTs.
The NYSDOH's proposal is yet another plan for the lab industry to consider. "The ... proposal is based on our experience over the last several years in specific laboratory tests we have received for review," NYSDOH spokesperson Erin Silk told GenomeWeb. "Our proposal puts risk assessment into practical steps for classification."
NYSDOH every year grants permits to 1,000 clinical labs and every other year certifies 3,500 lab directors and assistant directors. "The NYSDOH only regulates labs wanting to sell tests to ordering providers in NY, so unlike other proposals it's not necessarily 'national' but would affect many 'national' labs," Putcha said.
The state regulator is asking labs holding or applying for a NY clinical lab permit to comment on its LDT oversight plan by April 29.
As outlined in the two-and-a-half page document issued by the NYSDOH, labs holding a state permit "in the appropriate category of testing" must answer four questions: Does the LDT use a "well-established" methodology? Is the intended use of the test backed by published literature, clinical trials, or both? Does the test provide critical information on a patient's disease diagnosis, likelihood of developing a disease, or ability to respond to treatment? What is the potential impact of an erroneous result?
Regulators will consider the lab's answers to these questions and evaluate whether the test is "well established;" a "key determinant" in patient care; and the extent of its impact on patient outcomes.
NYSDOH defines a test as "well established" when the underlying methodology and indications are FDA approved or described in peer reviewed publications, and have been used by multiple labs without modification. A test that serves as a "key determinant" is one that "provides critical or essential information" in disease diagnoses, disease prognoses, and treatment response.
Lastly, the NYSDOH will weigh a test's impact on a patients' "morbidity, mortality, or condition" due to an inaccurate result. For example, NYSDOH said it would consider an LDT to be "high impact" if an analytically or clinically erroneous result could lead to the wrong diagnosis or to inappropriate treatment, which in turn could make the patient sicker or "even cause death."
Depending on the combination of these three factors an LDT will fall into one of three risk categories. Labs will not need to get approval and can market low-risk tests once regulators designate them as such. But state regulators will review test validation when they survey the lab, and NYSDOH noted that it has the right to withhold approval and require test review.
NYSDOH will grant conditional approval to moderate-risk tests, allowing labs to get these tests to market quickly. However, regulators will still review validation data on these tests before granting full approval.
Conditional approval is not an option for high-risk tests, meanwhile. NYSDOH said it would prioritize the review of high-risk LDTs, which require premarket review and approval before labs can market them.
In issuing its plan, the NYSDOH joins AMP, CAP, and the DTWG in proposing alternatives to FDA's LDT draft guidance. The DTWG's proposal spreads oversight responsibilities across FDA, CMS, and the states. AMP and CAP's proposals would keep oversight of most, if not all LDTs under CMS, with stronger CLIA regulations.
Currently, labs holding NYSDOH permits are exempt from having to garner CLIA certification. Keeping with this, AMP and CAP's LDT regulatory proposals wouldn't require premarket approval for tests approved by state authorities with CLIA exempt status. However, experts GenomeWeb spoke to believe that more tests might be high risk by the NYSDOH's plan than by other groups' proposals.
Risk classifications are still an unknown under FDA's framework. The agency has said it plans to issue additional guidance on this, estimating that half the LDTs on the market would be low risk, subject only to registration and adverse event reporting requirements. Among tests requiring premarket review, most would need 510(k) clearance, according to the FDA, and "a very tiny number" would require premarket approval.
However, industry observers suspect the FDA is underestimating the proportion of tests that would be high risk and subject to the most stringent premarket scrutiny. For example, healthcare IT firm NextGxDx estimates there are around 60,000 genetic tests, comprising more than half the LDT market, and of these genetic tests, 7,600 could be deemed high risk by the FDA.
Similarly, "compared to the CAP and AMP proposals, NYSDOH’s risk classification scheme would certainly seem to place more LDTs into the high-risk bin," Putcha said. "Even compared to the DTWG proposal, it appears the NYSDOH risk classification criteria would categorize more LDTs as high risk."
NYSDOH's risk classifications seem most akin to the DTWG's framework, which factors in a test's indication, whether it is "well characterized," if it is the only tool used to direct or change treatment, and how it impacts outcomes. AMP and CAP's plans are different in that proprietary algorithms underlying tests impact risk classifications.
AMP proposes that if labs don't want to reveal proprietary algorithms to CMS or third-party reviewers, these high-risk tests would have to be overseen by the FDA. If labs reveal proprietary algorithms, enabling inter-laboratory comparisons, then they can be reviewed under CLIA. CAP's plan has a similar but broader category of high-risk tests that would be regulated by the FDA, but the majority of LDTs would seem to fall into the moderate- or low-risk categories.
Roger Klein, chair of AMP's professional relations committee, also noted the possibility that NSYDOH's framework could comparatively classify more tests as high risk. "From a methodological standpoint for high-risk classification, AMP and CAP look at the ability to do inter-laboratory comparisons and use of proprietary algorithms, hence transparency, whereas NYSDOH seems to be focusing on methodologies that are not 'well-established,'" Klein told GenomeWeb. "There is probably overlap in these tests, but the emphasis is different."
Meanwhile, in the moderate-risk category, NYSDOH seems to provide labs some leeway that AMP and CAP don't. For example, AMP and CAP both require labs to establish the analytical and clinical validity for high- and moderate risk tests before market launch, whereas the NYSDOH is offering labs conditional approval of moderate-risk LDTs. (The DTWG and AMP would, however, hold regulatory authorities to time limits for reviewing high and moderate-risk tests, which if not met would result in automatic approval.)
Another area of concern for the lab industry is how readily LDT modifications will require new regulatory submissions, since it is often necessary to tweak processes when performing tests. But if these changes trigger new submissions to the regulatory bodies, it could interfere with patient care, some industry observers have argued.
In its current Comprehensive Test Approval Policy and Submission Guidelines, the NYSDOH indicates that specific modifications impacting a test's intended use would require labs to submit additional validation information. In its LDT oversight proposal, however, the NYSDOH states that a "well established” test has to be used by multiple labs without "any" modifications.
One positive aspect of NYSDOH's plan, according to Putcha, may be that regulators under one entity will perform both lab inspections and thoroughly evaluate the evidence supporting the analytical and clinical validity of LDTs. "It presumably ensures a reasonably high level of communication between the lab inspection and test approval teams," he said. Without this coordination Putcha observed that it would be hard for regulatory authorities to know, for example, if a lab was routinely reporting test results below the validated limit of detection for a test, in a target population that was not validated, or using an unvalidated sample type without clearly indicating this in the patient's report.
Notably, the NYSDOH hasn't said how it will fund this review framework. Given the size of the LDT market, stakeholders have similarly questioned whether FDA and other groups have the resources to oversee such a huge industry.
The NYSDOH's plan also doesn't contain any guidance on how information of approved LDTs will be publicly communicated. Currently, one can't look up LDT approvals on NYSDOH's website. Putcha hopes that NYSDOH will provide greater transparency for the public about approved tests when implementing its regulatory plan.
The Personalized Medicine Coalition, meanwhile, is facilitating discussion among labs and other industry stakeholders that want to identify areas of agreement within all the LDT regulatory proposals. It's not clear if the NYSDOH will join this collaborative effort.
In the mean time, industry players are left to sift through the various proposals and figure out what the changes might mean for their businesses. Stakeholders tend to agree that an oversight framework should not only foster innovation and access to critical tests, but also protect the public's health, even though they may disagree on how to acheive this. As such, Putcha wondered: "With regard to various proposals that propose state oversight, is it in the best interests of patients across the country to have potentially meaningful variability in the quality of lab testing based on your zip code?"