The US Food and Drug Administration’s intent to regulate in vitro diagnostic multivariate index assays has made genetic test developers more cautious, even before the agency codifies its goal.
Since the FDA expressed its desire last year to regulate IVDMIAs, genetic test developers have consistently criticized the agency’s approach for regulating laboratory-developed genetic tests as arbitrary, piecemeal, and a potential hindrance to innovation.
Now, five months after the FDA issued Exact Sciences a letter warning that a homebrew stool-based colorectal cancer assay it is developing with Laboratory Corporation of American required pre-market approval, test developers appear to have become ever more cautious when entering into collaborative deals to develop LDTs.
Diagnostic shops will often partner with reference laboratories to first introduce a test as a homebrew assay before going through the costly and time-consuming process of garnering a 510(k) clearance or a PMA. This strategy is commonly used in the diagnostics industry since introducing homebrew tests is faster and cheaper than going through the FDA. Additionally, it allows for the company to gauge the market and assess the performance of the test in a smaller population.
However, as the FDA’s regulatory approach toward LDTs changes, and diagnostic tests become more technologically complex, companies will need to proceed with caution when entering into such collaborations with reference labs.
“I think licensing a biomarker first to a reference lab and then going for PMA or 510(k) approval of an IVD test kit could be very, very helpful to prepare the market for an innovative diagnostic, but also to get an idea on the commercial potential of a test concept,” Achim Plum, Epigenomics VP of corporate communications, told Pharmacogenomics Reporter. However, “the more complex assays get, the more risk diagnostic companies are taking on in developing them.”
In the current regulatory environment, Plum noted that the regulatory pathway for any diagnostic that will be developed with or licensed to collaborators is carefully mapped out at the onset.
“A few years ago you entered into these collaborations and licensing agreements thinking that we’ll sort out the regulatory issues later,” he said. “These days, you can’t proceed without considering how the FDA might react. The awareness is much more on these issues, and … you definitely discuss regulatory issues in more detail. You try to get a model of how you intend to get through the regulatory pathways at the onset of any collaboration or licensing partnership” for diagnostic development.
“You do want to get in touch early with the FDA to make sure that you are doing things the right way,” Plum cautioned.
Epigenomics recently announced that it had non-exclusively licensed its DNA-methylation biomarker Septin 9 to Quest Diagnostics, which Quest will develop into a homebrew colorectal cancer diagnostic. Epigenomics’ deal with Quest comes five months after it licensed the marker to Abbott Molecular, which will use it to develop a Septin 9-based in vitro diagnostic kit.
However, since the kit will require regulatory approval from the FDA and CE mark in Europe, Abbott will likely launch the test kit in Europe not before 2009 and in the US not before 2010. By comparison, Quest’s homebrew test could be available in the US by the end of this year [see PGx Reporter 02-27-2008].
Traditionally, the FDA regulates diagnostic “test kits” for commercial sale and has practiced enforcement discretion over most LDTs, which fall under the aegis of the Centers for Medicare & Medicaid Services under the Clinical Laboratory Improvement Amendments.
But the agency has said it would regulate IVDMIAs, a more complex type of LDT, has released two draft guidances describing its plan in 2006 and in 2007, and has hosted a contentious public meeting in February 2007 to defend its strategy [see PGx Reporter 02-14-07].
According to one diagnostic company executive who asked to remain anonymous, one of the major concerns for industry under this current two-path system is that “since under CLIA, rigorous regulation of LDTs is already in place … the FDA will impose an undue and potentially conflicting burden on top of these already effective regulations.”
This official acknowledged that the LabCorp/Exact Sciences situation is a “more complex situation than a straightforward LDT,” but noted that “clarity on the part of regulators will be essential as companies collaborate on sharing of technology.”
|
“A few years ago you entered into these collaborations and licensing agreements thinking that we’ll sort out the regulatory issues later. These days, you can’t proceed without considering how the FDA might react.” |
While the FDA has yet to finalize the IVDMIA draft guidance, the agency issued a final guidance on analyte-specific reagents late last year, which discusses the regulation of LDTs.
Steven Gutman, director of FDA’s Office of In Vitro Diagnostic Device Evaluation and Safety, asserts that the agency has “tried to stay on message and indicate that the intent of the [analyte specific reagent] rule was to clarify that LDTs were considered medical devices but would be allowed enforcement discretion.”
Gutman told Pharmacogenomics Reporter in an e-mail that “the ASR rule tried to clarify in each of the two paths to market (commercially distributed or LDT), that someone had to take full responsibility for developing the test and for the quality and performance of the test (from soup to nuts).”
For commercially distributed tests, that responsibility to ensure that the test is safe and effective falls on the diagnostic device maker, and for an LDT, that onus is on the laboratory. “That is the recurrent message we are trying to deliver to alleviate confusion,” Gutman said. “It is not a new message.”
Allen Roses, director of Duke University’s newly launched Deane Drug Discovery Institute and the former head of genetics research at GlaxoSmithKline, pointed out that FDA regs may seem confusing due to the piecemeal nature of the regulatory process.
“FDA's attempts to define where regulation is needed may appear to be confusing as it receives its applications one at a time for an increasingly large swath of scientific technologies,” he told Pharmacogenomics Reporter.
However, since the diagnostics industry markets test broadly without conducting prospective studies to establish specificity and sensitivity, “it may well be time [for the FDA] to try to define the evidence for the various ‘diagnostic, prognostic and association’ claims,” he said.
Roses agreed that collaborative efforts like the one between LabCorp and Exact “may be critical for diagnostic tests to achieve wide availability,” particularly since reference laboratories have “a very broad and deep understanding of the medical need and market parameters for a test.”
However, “it helps to assess the potential medical demand for a test” and not simply carry out “the test developers' favorite marketing strategy or deals,” Roses cautioned.
Exact a Toll
Johns Hopkins University’s Genetics and Public Policy Center is one group that continues to assert that the FDA’s current regulatory approach toward genetic tests creates “an unstable business climate that may deter the development of new, clinically valid genetic tests” and discourages investors from backing such products.
For Gail Javitt, GPPC’s law and policy director, the FDA’s action toward Exact Sciences is just the latest example of the agency’s piecemeal approach that gives an unfair market advantage to reference labs.
Though Exact Sciences “is restricted in the components or protocols it can sell to LabCorp for a certain test, LabCorp is not restricted in offering that same test if it does so using a different method,” Javitt told Pharmacogenomics Reporter in an e-mail. “If FDA is truly concerned that tests not be offered unless their safety and effectiveness (i.e., their clinical validity) has been reviewed and established, then its current approach does not address that concern.”
In expressing its intent to regulate IVDMIAs, the agency has sent several companies warning letters stating that the tests they are marketing need to be submitted for clearance. Recipients of such letters include Genomic Health for its Oncotype DX breast cancer recurrence assay, Nanogen for its NanoChip microarray, and InterGenetics for its OncoVue breast cancer risk test.
In its warning letter to Exact Sciences, the FDA states that the firm’s PreGen Plus colorectal cancer screening test was “designed, developed, validated, and marketed by Exact rather than a test that was developed and validated by LabCorp.” The FDA maintains that PreGen Plus is not a laboratory-developed test, as LabCorp has been marketing it, but a Class III medical device [see PGx Reporter 01-23-2008].
Exact and LabCorp have an ongoing licensing agreement to develop and market the test. The component at issue in the agency’s letter is a DNA-purification device, called Effipure, which had been sold to LabCorp by Exact Sciences. LabCorp holds that these “refined microliter plates” are “general-purpose laboratory equipment,” that were not being marketed, and could not be used by doctors to diagnose disease.
However, the FDA asserts that there is “no statutory or regulatory requirement that a device be purchased by more than one entity or be used ‘independently’ to diagnose disease in order to be subject to the 510(k) premarket clearance requirement.”
Javitt authored an article late last year in the Food and Drug Law Journal, in which she argues that the FDA misses the point by basing its safety assessment of Pre-Gen Plus on whether the “microliter plates” were a medical device or general-purpose laboratory tool.
“FDA’s ability to answer essential questions about the safety and effectiveness of the test currently hinges on whether the test uses commercially distributed components or components developed by the laboratory,” she wrote in the article, which proposes alternative regulatory strategies to FDA’s current regulation of genetic tests.
“It seems illogical to permit FDA in the former instance [if the test is developed by Exact] to require premarket review of the test but in the latter [if it is developed by LabCorp] to prevent FDA from doing so,” she added.
“In addition to creating disincentives to the development of tests with demonstrated safety and effectiveness, such artificial regulatory distinctions do not serve the public’s health, particularly when the public is unaware that such arcane distinctions determine whether or not there has been independent review of the tests they rely on to make critical health decisions,” she concludes in the article.
While Roses conceded that companies are in a better position to evaluate the market demand for genetic testing, he suggested that the industry’s current cautious approach when entering into collaborative deals is as it should be.
“Researchers and diagnostic companies should always be cautious” when entering into collaborative deals to develop diagnostic tests, he said. “The critical issue is whether the medical need and the physicians' view of the data supporting the test combine to make testing obvious.”