FDA to Finalize LDT Guidance Amid Uncertainty on Number of Genetic Tests Impacted | GenomeWeb

FDA to Finalize LDT Guidance Amid Uncertainty on Number of Genetic Tests Impacted

NEW YORK (GenomeWeb) – An analysis conducted by Tennessee-based healthcare IT firm NextGxDx suggests there may be around 60,000 genetic testing products currently on the market, comprising more than half of the US laboratory-developed test market.

Moreover, depending on the criteria used, NextGxDx has projected that around 7,600 of these genetic testing products could be deemed high risk by the US Food and Drug Administration, for which labs may have to meet premarket review requirements. Since the agency intends to finalize its draft oversight plan for LDTs this year, it's critical that the FDA and industry players have an accurate estimate of currently marketed tests.

However, these numbers haven't been easy to pin down, and different groups have put forth varying estimates. At a congressional hearing in November, Rep. John Shimkus (R-Ill.) noted that LDTs seem to have grown "exponentially" in volume and complexity. To this, CMS Chief Medical Officer Patrick Conway responded, "We don't have exact numbers … but it seems exponential."

Many in the laboratory testing community contend that FDA oversight of LDTs would be burdensome on industry, but neither the industry nor FDA has released specific numbers estimating the impact. There's little incentive to do so on either side.

Groups like the American Clinical Laboratory Association maintains that the FDA lacks statutory authority to regulate LDTs. Estimating how many marketed LDTs, genetic or otherwise, could be classified under FDA's high- or moderate-risk category and subject to premarket review, would be a tacit admission that the tests could be regulated. The absence of industry estimates, in turn, enables FDA to assert that because LDTs aren't registered with the agency, there isn't an accurate count of tests on the market, making it impossible to know if they work as intended or even causing public harm.

The statistic commonly used by FDA and industry is that there are 11,000 labs offering as many as 100,000 LDTs in the US. "The potential workload for the FDA is staggering," if the agency implements its regulatory framework, ACLA President Alan Mertz said before Congress in 2014.

Since releasing its draft guidance, however, FDA has attempted to ease industry concerns about overly burdensome regulation and highlighted the fact that premarket requirements will be phased in over nine years. Jeff Shuren, director of FDA's device center, has estimated that half of the LDTs on the market would be classified as low risk under the agency's proposed framework, and would generally be spared premarket review as long as labs notify the agency about them and report any adverse events. Among tests requiring premarket review, most of them would need 510(k) clearance, according to Shuren, and "a very tiny number" would require premarket approval.

Unconvinced that FDA's plan would foster the right balance between regulation and innovation, several groups — the Diagnostic Test Working Group, the Association for Molecular Pathology, and College of American Pathologists — have penned alternative legislative proposals. The diagnostics industry group AdvaMedDx has submitted comments suggesting improvements to FDA's draft plan. Meanwhile, ACLA has remained staunchly opposed to any FDA role when it comes to LDTs.

While some of these groups now acknowledge that there are certain gaps in the way LDTs are regulated, there are still key differences in the way they view FDA's role and the level of regulation to which LDTs should be subject. The Personalized Medicine Coalition is facilitating discussion among labs and other industry stakeholders that want to identify areas of consensus around LDT regulation, and one of the group's primary concerns is how heavily FDA's requirements will weigh on their businesses.

"The community thinks the agency is drastically underestimating the number of tests that will be regulated," said Amy Miller, PMC executive VP, acknowledging that probably industry groups also don't have the precise numbers in hand.

Before FDA regulates LDTs, however, it seems reasonable that the agency should understand the scope of the task ahead, said Miller. "FDA has a responsibility to get new and novel diagnostic tests on the market," she said, highlighting that in 2015 the agency approved 13 personalized drugs that are administered with the help of diagnostics. "We need to ensure that FDA is still able to get all those new tests out the door. And so, the degree to which we can focus on what is truly important for the FDA to actively regulate is a worthwhile endeavor.

"But the number of LDTs on the market is a hard number to get at," she acknowledged.

The numbers

NextGxDx's estimate of more than 60,000 currently marketed genetic testing products is the largest approximation to date. In comparison, the NIH's Genetic Testing Registry, to which test providers voluntarily submit information, contains information on more than 32,000 clinical and research tests for Mendelian disorders, drug response, as well as somatic or cancer tests. According to GeneTests.org, an independent business unit of BioReference Laboratories, there are around 55,000 molecular, cytogenetic, biochemical, panel, and custom tests.

NextGxDx's database includes any test for a genetic disease, including cancer, that assesses DNA, RNA, or protein markers, but doesn't include toxicology or infectious disease tests. By the company's count, an average of eight to 10 new genetic testing products enter the market daily. Just between July 2014 and January 2016, labs and diagnostic firms launched more than 5,600 tests, according to NextGxDx.

The firm has developed custom software that pulls publicly available data on tests from companies' websites. "We're able to pull in data from any source we find to be a legitimate source of information," said Taylor Murphy, who leads a three-person team that maintains NextGxDx's test database.

If we can convince legislators and the FDA that there are these areas of commonality, I think we can get to a regulatory paradigm that won't destroy the laboratory industry.

From these sources, the software captures, standardizes, and cleans the data and places them into 25 fields for each test, which enables users to see, for example, the target analytes and technology platform. Then, NextGxDx groups tests according to categories, such as single-gene tests and panel tests, and by clinical indications, so that doctors and hospitals can find and compare available diagnostics. Every week, the firm has audits and employs quality control measures to keep the information up to date.

Based on the information in its database, NextGxDx has also attempted to assess what proportion of genetic tests falls into FDA's high-risk category and therefore would likely be subject to the most scrutiny under the LDT oversight plan. The agency has said it plans to issue additional guidelines on test risk classifications. Still, going by the draft guidance, the highest-risk tests that FDA proposes to oversee first are LDTs with the same intended use as cleared/approved companion diagnostics and Class III devices.

By NextGxDx's count, there are close to 600 tests that fit these criteria currently not regulated by FDA, comprising approximately 1 percent of the genetic testing market. But the number of potentially high-risk tests increases when NextGxDx considers different scenarios.

For example, factoring in tests or panels assessing one of the 56 genes that the American College of Medical Genetics and Genomics recommends reporting as secondary findings when individuals have genomes or exomes sequenced, the number of high-risk marketed genetic tests shoots up to around 3,500 tests, or 7 percent, of the tests in NextGxDx's catalog. Add to this list pharmacogenomic tests, non-invasive prenatal tests, and oncology tests, the number of high-risk tests jumps to around 7,600, or around 14 percent, of the tests tracked by the firm.

Further complicating matters is the disproportionate growth in the past 18 months in the number of multi-marker panels, representing 24 percent of around 5,600 newly launched tests. For example, in tracking tests for the TAZ gene — mutations in which are associated with the rare genetic disorder Barth Syndrome — NextGxDx identified 21 single-gene tests offered by 10 labs that perform them fairly infrequently. "This would meet the rare disease test exemption by any standard," observed Gillian Hooker, director of clinical development at NextGxDx.

According to FDA's draft guidance, traditional, rare disease, and unmet needs LDTs have to meet registration, listing, and reporting requirements, but are exempt from premarket requirements. However, because Barth Syndrome can present various phenotypes, including cardiac, blood, and muscle issues, the TAZ gene is commonly included in condition-specific panel tests: 91 by NextGxDx's analysis. "A number of these tests could conceivably, given the broad nature of the [panels] they're designed for, sell a lot in a year," observed Hooker, "and may not meet the rare disease exemption."

Given the available numbers, although they can be fuzzy at times, many industry observers suspect that the regulatory burden for FDA and labs may be bigger than what the agency is estimating. This is a big concern for the stakeholders PMC's Miller is working with to identify areas of agreement between alternative LDT regulatory proposals that industry groups have floated. [For more detail, download GenomeWeb white paper below: "Alternative Proposals for Lab Test Regulation: Are There Opportunities for Consensus?"]

"We've looked at areas of commonality and areas of difference within these proposals," Miller said. "There are a lot of areas of commonality. So, if we can convince legislators and the FDA that there are these areas of commonality, I think we can get to a regulatory paradigm that won't destroy the laboratory industry."

"It's the PMC's opinion that the community is ready for that discussion," Miller said. "It has a lot to do with the FDA saying they intend to finalize guidance in April or May, and there's not much standing in their way."

File Attachments

In Science this week: deletion of one microRNA allows pluripotent stem cells to form embryonic and non-embryonic lineages, and more.

Arizona is planning to sue Theranos for "deceptive acts" and misrepresentations of its "capabilities and operation."

If confirmed as Department of Health and Human Services Secretary, Tom Price says he will divest himself of certain holdings, according to Stat News.

Oliver Smithies, who won the Nobel Prize in 2007, has died, the New York Times reports.