Whether or not the Food and Drug Administration falls short in regulating diagnostic tests for cancer and other diseases depends on whom you ask.
Talk to a pathologist who performs diagnostic tests, and you may hear that everything is generally okay.
“Clinical labs have a long track record of success in providing cancer tests, and governmental oversight continues to be appropriate,” said Dr. Margaret L. Gulley, professor of pathology and laboratory medicine at the University of North Carolina, Chapel Hill. Dr. Gulley was among the many experts interviewed for this article.
But some oncologists and others who follow the way that genetic laboratory tests for cancer and other diseases come onto the U.S. market see a different picture.
“I think the current FDA regulations are awful. New diagnostics should be addressed the same way as new therapeutics,” said Dr. Daniel F. Hayes, professor of internal medicine and director of the breast cancer program at the University of Michigan, Ann Arbor.
“If we use diagnostics to make clinical decisions, then why not use a test that's as reliable and accurate as a new therapeutic? A bad test is every bit as bad as a bad drug,” he said.
At the heart of the issue is who regulates the growing number of sophisticated molecular-based tests coming into the marketplace. Cancer diagnostics have so far triggered much of the controversy, but in an era of personalized medicine, any and all specialties could be affected.
As Dr. Gulley pointed out, virtually all clinical labs in the United States are subject to the Clinical Laboratory Improvement Amendments (CLIA) regulations enforced by the Centers for Medicare and Medicaid Services. “Most cancer-testing laboratories are accredited by the College of American Pathologists, which I believe provides even greater assurance of quality than does CLIA certification,” she added. “FDA regulation of individual lab-developed tests seems superfluous.”
For attorney Gail H. Javitt, however, the current system is inadequate. “We are concerned that tests developed in-house by clinical laboratories are not reviewed by the FDA to ensure their safety and effectiveness,” said Ms. Javitt, law and policy director at the Genetics and Public Policy Center of Johns Hopkins University in Washington.
Regarding laboratory oversight by the CMS, she added, “Our concern is that their standards for laboratories performing genetic tests—including for cancer diagnosis—are insufficient.” In general, for genetic and cancer diagnostics, “we support greater oversight by the FDA,” she said.
Genentech Asks for More Oversight
Although the controversy has simmered for several years, the biotechnology company Genentech turned up the heat in December with a citizen's petition it submitted to the FDA. The petition asked the FDA commissioner to “require that all in vitro diagnostic tests intended for use in drug or biologic therapeutic decision making be held to the same scientific and regulatory standards, regardless of whether the tests are developed and sold by device manufacturers as diagnostic test 'kits' or are developed by clinical laboratory companies for in-house testing.”
Genentech's petition continued, “Currently, FDA regulates in vitro diagnostic tests in kit form, but not LDTs [laboratory-developed tests].” (See box.)
Citing a University of Washington Web site that listed nearly 600 labs running tests for more than 1,300 diseases or conditions as of last November, the petition noted that the number of LDTs on the U.S. market grew significantly in recent years.
The petition also cited several unapproved LDTs that were being used to guide prescribing of cancer drugs marketed by Genentech. These included tests for the HER2 protein to identify patients with breast cancer that should be treated with trastuzumab (Herceptin), a test to predict response to rituximab (Rituxan) in patients with follicular non-Hodgkin's lymphoma, and a test to distinguish squamous from nonsquamous non-small cell lung cancer to identify patients who should receive bevacizumab (Avastin).
In a written comment on the petition, Ms. Javitt and Kathy Hudson, Ph.D., director of the Genetics and Public Policy Center, said, “We believe that FDA has jurisdiction to regulate all laboratory developed tests as medical devices and should exercise this authority.”
They added that the “FDA's current failure to regulate LDTs—in particular genetic tests that are used in drug or biologic therapeutic decision making—threatens public health, creates serious inequities in the marketplace that disincentivize innovation, and impedes the success of personalized medicine.”
When interviewed, Dr. Gulley countered that “there is no need for LDTs to be subjected to an extra, expensive process of vetting by the FDA. Adding another layer of scrutiny would be extremely costly for both the government and for testing labs. Yet it would be questionable if it would improve patient care.”
A New Direction in Washington?
The FDA must respond to citizen petitions within 6 months—in this case by June—but the initial response does not have to be substantive; the agency can potentially wait years before it finally grants or denies a petition. But with a new administration in Washington, the FDA and its parent, the Department of Health and Human Services, may be poised to take a more activist role on test regulation.
Stuart Hogarth, a researcher in the department of social sciences at Loughborough University, Leicestershire, England, has studied diagnostic test regulation in the United States and other countries. “In 2001, the agency was close to taking on the whole clinical lab sector comprehensively, but the election of George Bush and the appointment of people [to the FDA] who were hostile to this led to its not happening,” he said.
Another harbinger of a possible new FDA approach to came last April in a report by an HHS panel, the Secretary's Advisory Committee for Genetics, Health, and Society, on U.S. oversight of genetic testing. The committee found “significant gaps in the U.S. system of oversight of genetic testing that can lead to harm”; it also identified a “gap in oversight related to clinical validity” of genetic tests. It outlined a long list of recommended steps, including having the FDA “address all laboratory tests,” rather than limit its focus to test kits and analytic-specific reagents.
Some Tests Off the Market
FDA representatives did not respond to questions submitted to them on the issues raised in this article. Several high-profile episodes have recently occurred, however, where the agency and industry crossed signals over new diagnostic cancer tests.
Two cases involved tests that came onto the U.S. market without prior FDA review. The laboratories involved claimed that the tests were LDTs, and hence no prior FDA review or approval was needed. But the FDA disagreed, concluding that both tests were developed in other labs and, therefore, didn't qualify as LDTs.
In October 2007, the FDA sent Laboratory Corporation of America (LabCorp) a letter saying that the colon cancer screening test it was already running on patient specimens, PreGen-Plus, had actually been developed by Exact Sciences, and so was not an LDT but a medical device subject to premarketing FDA approval. LabCorp stopped selling PreGen-Plus last May—and began marketing an improved colon cancer screening test, ColoSure, in July.
A strikingly similar set of events surrounded another LabCorp test, OvaSure, which the company made available last June for ovarian cancer screening. The FDA sent a warning letter in September, saying the test was not an LDT because it had arrived from a Yale lab. LabCorp stopped selling OvaSure in October.
Another pair of controversial cancer tests seemed to spur the FDA to create a new category of test oversight, the in vitro diagnostic multivariate index assay. In draft guidelines released in September 2006 and revised in 2007, the FDA defined multivariate index assays as tests that combine values of multiple variables using an interpretive function.
FDA watchers say that the multivariate index assay guidelines were the agency's response to its inability to force reviews on two assays: the Oncotype Dx test that's marketed by Genomic Health to predict the risk for breast cancer recurrence in women with newly diagnosed disease, and the OncoVue test marketed by InterGenetics for estimating a woman's breast cancer risk. (See related story, p. 25.)
Through early 2009, both of these tests remain on the U.S. market without ever undergoing FDA review. And the multivariate index assay draft guidelines remain in draft form with no indication of when final guidelines might appear. Despite that, the FDA has recently approved some multivariate index assays, including the MammaPrint test that measures a woman's risk for developing recurrent breast cancer. (See p. 25.)
Physicians Look for Evidence
Amid the controversy and unresolved issues, a critical question for physicians who might order cancer diagnostic tests is: How can they be assured of a test's reliability and clinical value when tests reach the U.S. market without a clear vetting record?
“Physicians base medical decisions on the literature and on recommendations from medical groups. The medical community does a pretty good job of regulating itself,” said Dr. Karen E. Weck, director of the molecular genetics laboratory at the University of North Carolina in Chapel Hill.
The best data available today on the clinical efficacy and usefulness of diagnostic tests for cancer can often be found in the results from drug trials that used a test as the criterion for enrolling or excluding patients. Physicians should look to the outcomes of trials like these for the best guidance on which tests are worthwhile, said Dr. Mark J. Ratain, a professor of medicine and medical director for clinical research at the Cancer Research Center of the University of Chicago.
Dr. Jeffrey A. Kant, professor of pathology and human genetics and director of the division of molecular diagnostics at the University of Pittsburgh, sees reassurance in the way diagnostic testing has performed in the field until now.
“The system, as it's been operating, hasn't failed the public in any massive way. I'm not aware of any public health disasters [in the United States] associated with molecular tests in mainstream laboratories,” Dr. Kant said.
Most Gene Tests Lack FDA Oversight
The Food and Drug Administration does not review most genetic tests before they reach the U.S. market.
The FDA regulates in vitro diagnostics if the components of the test are bundled together, labeled for a particular use, and sold to a laboratory as a unit, a diagnostic that is usually called a “test kit.” Manufacturers of test kits must give the FDA evidence that they are safe and effective before they're approved for marketing.
As of last year, the FDA had reviewed a total of about a dozen genetic test kits, according to a review published in 2008 by Gail H. Javitt of the Genetics and Public Policy Center of Johns Hopkins University, Washington.
A second area of FDA test regulation covers analytic-specific reagents (ASRs). Each ASR serves as an active ingredient for a laboratory-developed test. ASRs can be sold only to laboratories that are certified to perform high-complexity tests. In 2007, the FDA specified that ASRs cannot involve multiple reagents bundled together, or a reagent sold with a specific indication for its use.
The vast majority of diagnostic genetic tests available in the United States for cancer and other diseases are laboratory-developed tests (LDTs), also known as “home brew” tests. In general, LDTs are not subject to review by the FDA before they come onto the market.
In 2006, the FDA released draft guidance for a new category of LDTs, multivariate index assays, that it said would require premarketing review and approval. The draft guidance was revised in 2007, but as of early 2009 it has not become a formal regulation, although some tests have been reviewed and approved as multivariate index assays.
The FDA stresses that it exercises “regulatory discretion” in not regulating most LDTs that are performed in a single clinical laboratory, which suggests that the agency reserves the option to change this policy in the future.
Laboratories that perform clinical diagnostic tests also are subject to regulation by the Clinical Laboratory Improvement Amendments (CLIA), a program that is enforced through the Centers for Medicare and Medicaid Services. But CLIA regulations focus on overall operations of each laboratory and not on assessing individual tests.
Two states, Washington and New York, have their own regulations and review of clinical labs that supersede CLIA oversight.
FULL TEXT70157-X/journalimage?img=PIIS109786900970157X.fx1.lrg.gif&fig=&kwhquery=&issn=1097-8690&ishighres=false&allhighres=false&free=yes','journalimage');)
