June 2007: Volume 33, Number 6
Senate Bills Call for Oversight Of Laboratory-Developed Tests
Could New Requirements Overwhelm Labs and the FDA?
By Julie McDowell
Congress and federal lawmakers are once again contemplating increasing federal oversight for laboratory-developed tests. Two prominent U.S. senators—Edward Kennedy (D-Mass.) and 2008 Presidential contender Barack Obama (D-Ill.)—have announced bipartisan legislation that not only seeks varying restrictions on laboratory developed tests (LDTs), but also renews a call for the Centers for Medicare and Medicaid Services (CMS) to establish a genetic testing specialty area under CLIA, echoing a similar recommendation recently put forth by a report from the Institute of Medicine (IOM). In addition, one of the bills—Kennedy’s Laboratory Test Improvement Act (S. 736)—would also require laboratories who make LDTs to submit evidence of clinical utility to the Food and Drug Administration (FDA), prompting many in the laboratory community to complain that these regulatory hurdles place an undue burden on laboratories and threaten to discourage advances in cutting-edge areas like genetic and molecular testing. As Congress weighs these proposals over the coming months, those who have scrutinized the bills recommend that clinical laboratorians start preparing now for potential landmark changes to how diagnostic tests are regulated.
“Both of these bills, in particular Sen. Kennedy’s bill, are significant for the lab industry, because they have the potential to bring laboratory-developed tests to a level of regulation that they have never been subject to before,” explained Peter Kazon, an attorney with Alston & Bird LLP (Washington, DC), who specializes in laboratory testing regulation. He spoke at an April 3rd audioconference, “Sens. Kennedy and Obama Introduce Legislation on Laboratory Developed Tests,” sponsored by the American Clinical Laboratory Association (ACLA). Of the two bills, the Kennedy legislation, which is co-sponsored by Sen. Gordon Smith (R-Ore.), imposes greater oversight on LDTs. Sen. Obama’s bill, the Genomics & Personalized Medicine Act (S. 976), will ask the Institute of Medicine (IOM) for guidance on the appropriate regulatory framework for genetic testing. Sen. Obama’s bill is co-sponsored by Sen. Richard Burr (R-N.C.).
If legislation of this type goes into effect, it would also give the FDA a greater role in regulating LDTs. Currently, the FDA considers LDTs under their authority as medical devices, but applies enforcement discretion to these tests. Labs that perform LDTs, particularly genetic tests, are already subject to stringent CMS standards under the Clinical Laboratory Improvement Amendments (CLIA). This potential change in FDA oversight must be carefully scrutinized, according to Alan Mertz, President of ACLA, a non-profit association that advocates for national and regional clinical laboratories on federal and state regulatory and reimbursement policies. “We want to make sure that there is a balance, that the oversight protects public health, but at the same time, does not stifle the innovation that is going on in genetic and molecular testing. We also want to make sure that it doesn’t turn the clock back on a lot of tests that have been around for many years. The clinical validity and utility of these tests that are out there is well established,” he said.
Casting a Wide Net
Under Sen. Kennedy’s Laboratory Improvement Act, LDT providers would be required to submit to the FDA evidence of their tests’ clinical and analytical validity, which would also be posted in an online database accessible to the public. Similar to what is currently required by the state of New York Department of Health’s Clinical Laboratory Evaluation Program for laboratories offering LDTs to patients, this evidence of clinical validity could include citations from peer-reviewed literature. Many were relieved that this legislation was not included in the mark-up of the FDA Revitalization Act, which was approved by the Senate. The bill, however, did include an amendment by Sen. Obama calling for the IOM study on regulatory oversight of genetic testing, originally featured in S.976 (See Washington Profiles—June 2007).
Kennedy isn’t the only federal lawmaker scrutinizing what LDTs should show in terms of clinical validity. In March, Health and Human Services (HHS) Secretary Michael Leavitt charged the Secretary’s Advisory Committee on Genetics, Health, and Society (SACGHS) with developing an oversight plan for genetic and genomic testing. He also asked the committee to determine what pathways currently exist that examine the analytic and clinical validity of genetic tests (See Sidebar below).
Mapping Out a Future for Genetic Testing
In March, HHS Secretary Michael Leavitt charged the Secretary’s Advisory Committee on Genetics, Health, and Society (SACGHS) with developing a comprehensive map of steps needed for evidence development and genetic and genomic tests oversight, with improvement of health quality as the primary goal. Specifically, the Secretary is looking for feedback on some of the following questions:
- What evidence of harm exists regarding genetic tests? Is that harm attributable to analytic validity, clinical validity, or clinical utility of the tests? If evidence does not exist, what threats are not currently being addressed? What public health benefits are not accruing as quickly as they might?
- What distinguishes genetic tests from other laboratory tests for oversight purposes?
- Would additional or revised government oversight add value for patients, and if so, how and where?
SACGHS is expected to deliver its report to the Secretary by the end of the year. A draft of the report will likely be released to the public for comment in the fall, according to Andrea Ferreira-Gonzalez, PhD, Associate Professor and Director, Molecular Diagnostics Lab at Virginia Commonwealth University’s Department of Pathology in Richmond, who is leading this report for SACGHS’ Oversight Task Force.
Ferreira-Gonzalez is especially interested in analyzing how evidence of clinical validity should be presented by labs. “Demonstrating clinical validity is good laboratory practice and using peer-reviewed literature is an accepted practice,” said Ferreira-Gonzalez. “The problem is when there is not sufficient peer-reviewed literature. What we don’t have right now is specific guidance regarding what is the minimum requirement to show clinical validity, which might be one of the topics tackled by our Committee in response to the Secretary’s request.”
Additional information, including transcripts of the March 2007 SACGHS meeting, is available online.
Under Kennedy’s legislation, the FDA could deem a submission deficient if it doesn’t adequately summarize the peer-reviewed literature or relies on non-peer reviewed literature. In addition, the submission could be rejected if it doesn’t demonstrate that the analytical and clinical validity is comparable to other tests that have received 510(k) clearance or pre-market approval (PMA) by the FDA. While there is no time limit on how long the FDA can take to review submissions, laboratories with deficient submissions would be given 90 days to submit additional information. If the agency also declares this additional information insufficient, then the laboratory would have to apply for a 510(k) clearance or PMA. The bill would also require that the FDA develop a guidance document detailing how peer-reviewed literature and other data should be presented to support clinical validity. Additional guidance from the agency would also be necessary to clarify what information laboratories need to submit when modifications are made to an LDT.
Critics of the Kennedy bill argue that these submission requirements will be a burden not only on the laboratories, but also on the FDA, and question whether the agency has the resources to handle the workload these submissions will bring. “We just surveyed seven of our member companies to determine how many LDTs and modifications they have that would require submissions, including the number of pages of data that would have to be submitted,” said Mertz. “According to the responses from just those seven companies, the number of submissions would be 5,200 and they would need to include 158,000 pages of data.”
Mertz is also concerned that laboratories would have to submit clinically comparable data. “If you have an LDT and there is a similar kit that is approved, then you would have to submit comparable data to that kit,” he said. “Ultimately, this would mean that labs would have to go through the same process that the kits go through to get cleared by the FDA. That would be incredibly stifling in terms of innovation, and likely unnecessary.”
There is also concern that the Kennedy bill uses a shotgun approach to regulate all LDTs, rather than its real target—direct-to-consumer (DTC) testing. In particular, Congressional leaders have in their sights nutrigenomic tests purchased online without any clinician involvement. In July 2006, the Government Accountability Office (GAO) released a blistering report, Nutrigenetic Testing: Tests Purchased from Four Web Sites Mislead Customers. Sen. Smith, co-sponsor of the Kennedy bill, is a ranking member of the Senate Special Committee on Aging, which held a hearing on the report’s findings. The GAO concluded that results from the kits that were analyzed for the report made predictions that were medically unproven and so ambiguous that they did not provide meaningful information to customers (CLN, October 2006, p. 8). Furthermore, the report stated the test results misled consumers by recommending expensive dietary supplements.
Few in the laboratory community disagree that nutrigenomic tests can be dangerous and need to be under some regulation. “This kind of testing is too new and ought to be appropriately and rigorously vetted by the professional medical community,” said Daniel H. Farkas, PhD, Executive Director of the Center for Molecular Medicine in Grand Rapids, Mich., who has established three other molecular diagnostics laboratories at hospitals in New Jersey, Michigan, and Texas. “If companies pop up without the appropriate professionalism and accreditation and don’t provide proper evidence of their test’s clinical relevance, then it is appropriate for legislation to come down hard on them. But the problem is that the net gets cast too widely, and competent clinical laboratories that play by all the rules and are doing perfectly appropriate tests with full-blown analytical and clinical validity are caught in that net.”
These additional regulations, including the proposed clinical validity requirements, threaten to limit access to healthcare, according to Farkas. “If a test had a large enough market that an in vitro diagnostics company was willing to commercialize it, then the company would do the appropriate trials to demonstrate analytical and clinical validity,” he explained, adding that there are currently only a few dozen molecular diagnostic tests that demonstrate the necessary revenue potential for companies to bring them to market, along with an even smaller number of blockbuster molecular diagnostics tests.
“Laboratories are not miniature IVD companies, and this legislation has the potential to force every laboratory that is doing a laboratory-developed assay to go through multiple regulatory hoops,” said Farkas. “Hundreds, maybe thousands of diseases for which there are perfectly appropriate and valid diagnostic tests would go wanting because no company is interested in going to the trouble and expense to do the appropriate clinical trial to demonstrate clinical validity.”
A Less Prescriptive Approach
Compared to the increased oversight proposals detailed in the Kennedy legislation, Sen. Obama wants to solicit input from lawmakers and policy analysts—including the IOM—on the best way to regulate genomic LDTs. “Both Sen. Obama and Sen. Burr agree that genetic and genomic tests need improved regulation,” said Obama’s health policy advisor, Dora Hughes, MD, at the April audioconference. “However, our approach is less prescriptive. We recognize that the regulation of genetic tests would be a major new initiative and we don’t necessarily agree that Congress is the appropriate body to establish a regulatory framework in the absence of a careful legislative process or regulatory process by the HHS secretary.”
In order to evaluate different options for oversight of genetic tests, the bill requires that the HHS Secretary fund an IOM study to analyze different regulatory avenues. The study would then offer recommendations on how to design a decision matrix, which would be a mechanism to classify all LDTs, including genetic tests, as well as to determine what level of review is needed for each category. The matrix would then steer the regulation of tests into either the FDA or CMS. “The decision matrix would help the sponsors of tests understand whether their tests would go through an FDA-type regulatory process or if they would remain subject to CLIA at CMS,” said Hughes. “Our expectation is that the Institute of Medicine study would inform the development of the decision matrix, as well as the overall regulatory process.”
Within the decision matrix, some tests will require a minimal level of review and some more comprehensive review. However, this matrix will also include summary information on these tests—including their intended use and performance characteristics—and will be available to the public in a database. “Regardless of the level of review of the test and its complexity, this information would be mandated to be submitted to the Secretary who would then organize and categorize it and make it available to the public, providers, and other stakeholders in the community,” added Hughes.
Unlike the Kennedy bill, this legislation is receiving favorable feedback from many in the clinical laboratory community. In particular, an IOM report on these regulatory questions appears to be a welcome prospect, and could be coming sooner rather than later if Sen. Obama’s amendment authorizing the study remains in the final version of the FDA Revitalization Act. “I think this bill took the right approach, which is to have an outside group, in this case the Institute of Medicine, look at the oversight issue for about a year and a half and come up with recommendations on how these tests need to be regulated,” said ACLA’s Alan Mertz.
Renewed Calls for Specialty Area
While the Kennedy and Obama bills have significant differences, they do share a common initiative: requiring CMS to create a genetic testing specialty under CLIA. A similar move was recommended by a recent IOM report on guidelines, standards, and oversight of cancer biomarker development. In the report, Cancer Biomarkers: The Promises and Challenges of Improving Detection and Treatment, the authors recommended that CMS develop a specialty area for molecular diagnostics under CLIA. “The technology used for genetic testing is unique compared to the technology used for other tests, so I think it requires expertise to be able to handle these blood samples and tissues efficiently,” said one of the report’s authors, Edith Perez, MD, Director, Cancer Clinical Study Unit, Mayo Clinic (Jacksonville, Fla.). “We need quality controls for this testing—it’s just so much more complex than other diagnostic testing technology, which is why I think a specialty is required.”
Despite these arguments and previous pressure from groups including SACGHS, CMS officials have remained steadfast in their resolve not to create a genetic testing specialty area. Creating a subspecialty area will not address many of the concerns cited by those pushing for this move, according to Judy Yost, Director, CMS’s Division of Laboratory Services. A CLIA genetic testing specialty area will not improve clinical utility and related issues because many of these tests are not FDA-approved, nor will it resolve the inadequate supply of proficiency materials currently available for these tests. “In addition, we really don’t have any evidence that we need more stringent requirements than those that already exist,” said Yost. “Since most genetic tests are high complexity, the labs that are certified to perform these tests are subject to the most stringent requirements under CLIA. That is a strong reason not to move forward until we have concrete evidence otherwise.” In addition, she is concerned that because genetic testing is a dynamic field in its early stages, any regulations made now could be outdated—and onerous—in the future.
Instead of creating a specialty area, CMS and the clinical lab community should work together to tackle each concern on an individual basis, Yost suggested. “We would like to work with labs to come to a mutual agreement on how we can address their concerns, rather than just jumping to a regulation automatically,” she said. “Our goal is to help labs do the best job they can, and however we get there is okay with us. But maybe we can look at some ways that are less burdensome and time consuming, because it’s going to take three years, at a minimum, to create this specialty area. Maybe we can do some things immediately that could mitigate theses major concerns. Rather than all of this anxiety and emotion over creating this regulation, let’s put the energy in to working on positive and productive solutions.”