December 2008: Volume 34, Number 12
Proprietary Lab-Developed TestsResponding to Innovation or Skirting Regulation?
By Bill Malone
When FDA published the 2006 draft guidance on in vitro diagnostic multivariate index assays (IVDMIA), the agency caught the lab community by surprise. While FDA maintains that the IVDMIA guidance covers only a narrow category of tests that falls under its regulatory authority, many in the lab community saw the document as a warning flag for future FDA regulation of lab-developed tests (LDT). Now 2 years later, in the absence of a final IVDMIA guidance document, a resurgent business model is keeping LDTs in the spotlight, ramping up pressure on FDA to look more closely at the innovative test services companies are bringing to market. In fact, the agency recently called out LabCorp for its OvaSure test, a controversial move that made headlines in the New York Times and other national news media (Click for article).
Despite the uncertain times for LDTs, a recently released report, “Diagnostic Test Service Commercialization: A Roadmap to Diagnostics in the 21st Century,” published by Kalorama Information, identifies a growing market for test services in which reference labs and companies offer proprietary tests out of their own CLIA-registered labs. Although these test services are offered under an established regulatory path, FDA and members of Congress remain concerned that CLIA oversight might not be enough to ensure consumers are protected from tests that do not meet FDA’s stringent standards.
“I have no doubt that some of these lab-developed tests are very high in quality,” said Steven Gutman, MD, director of FDA’s Office of In Vitro Diagnostic Device Evaluation and Safety. “I also have no doubt that some of them are not so high in quality. What I don’t know is how it will all sort out and whether tests of less value will in fact produce a branding issue that might come back to haunt labs. If poor quality tests enter the market, then people could lose confidence in tests more generally.”
What’s Old is New
Although the number of LDTs commercialized as proprietary test services is on the rise, with some 70 tests already available or near market, the test service model is not a new one, according to the Kalorama Report. In the 1980s, IVD companies like Roche, Beckman, and Corning had their own CLIA-registered labs through which they could market their tests. But by the late 1990s, these company labs were sold off to create the current reference lab giants of LabCorp and Quest, which since then have bought up many smaller, independent reference labs.
In the midst of this consolidation phase, companies like Myriad Genetics with its BRCA test for inherited breast cancer risk and Athena Diagnostics with its ApoE test for risk of Alzheimer’s disease were making a name for themselves with novel tests that looked for links between genes and disease risk. These and a handful of other companies became the predecessors of the current boom in the proprietary test service market, explained Shara Rosen, RT, MBA, author of the report and a senior consultant with Kalorama Information. At first, the high cost of these early test services raised eyebrows. Take Myriad Genetics for example. When the company first began marketing its BRCA test for hereditary breast cancer, they charged more than $3,000. “That was unheard of then,” said Rosen. “It seemed like a fortune. Now people don’t even blink. The insurers don’t even blink.”
Today, after what Rosen describes as a 10-year lull since Myriad’s and other companies’ test services hit the market, both small IVD companies and the large reference labs are trying to capitalize on research from the Human Genome Project and advances in molecular biology, pushing the envelope with highly complex tests for cancer and other diseases. According to the Kalorama report, the esoteric test segment into which these tests services fall will continue to grow at a rate of 10%–15% per year, compared to 6% for the rest of the IVD market. In 2007, esoteric test service revenues reached almost $4 billion, about 12% of the wider lab market worth more than $40 billion (See Graph, below).
Snapshot of Lab Test Market
Core Testing* vs. Esoteric Testing in 2007
As part of the growing esoteric testing market, proprietary lab-developed tests will likely continue to make headlines as companies work to translate new research into marketable test services. Breakthroughs in genomics, molecular biology, and an aging population will all contribute to a resurgence of the test service model, predicts the Kalorama Information Report, Diagnostic Test Service Commercialization: A Roadmap to Diagnostics in the 21st Century.
*Core testing includes: clinical chemistry, immunoassays, radioimmunoassay, coagulation, histology, nucleic acid testing, and flow cytometry.
Source: Kalorama Information, 2008
Gutman agrees that the business model for proprietary tests is not new. “I would say that it’s actually an old practice, but that it may have new life because of the interesting new technologies that are entering the marketplace through that regulatory route,” said Gutman. “And I guess there’s good news and bad news. The good news is that it does certainly create what one might characterize as the least burdensome route to market, and the bad news is that there’s a certain non-parity between that route and the more established FDA route. And there is less transparency than in the FDA route.”
Two Paths to Market
The Medical Device Amendments of 1976 and CLIA 1988 outline two distinct paths for companies to commercialize lab tests. A company can choose to market a test as a kit through FDA’s risk-based review process, or it can set up its own lab and market a proprietary LDT as a test service, with its lab falling under CLIA regulation, explained Paul Radensky, MD, JD, a partner with the law firm of McDermott, Will & Emery in Washington, D.C. that represents the Coalition for 21st Century Medicine and several companies that market unique test services. “When we’re looking at today’s environment, tests that tend to follow the CLIA pathway, rather than the FDA pathway, are a number of highly advanced, high-complexity genetic or genomic assays where there is a lot of concern about the transferability of the method from one lab to another and about what it would take to put these in kit form while maintaining the accuracy and validity of the tests,” said Radensky. “There’s also a lot of concern about whether FDA can find a truly least burdensome pathway to clear clinically meaningful claims.”
Gregory Critchfield, MD, president of Myriad Genetic Laboratories, Inc., said he believes that the test service model is the only way Myriad’s BRACAnalysis test could ever have entered the market. “In our business, when the BRCA genes were discovered and the clinical service was launched in 1996, as today, there was no easy way to put this in to a kit form. So, of necessity, it had to be developed by a laboratory offering a service,” said Critchfield. “The test involves sequencing a very large number of DNA bases, looking at structural rearrangements in the gene, and interpreting the data because there are a number of variants that can be found, all making for a test that is best made available in a the laboratory environment.” Critchfield pointed out that because Myriad has the patents and exclusive license to perform BRCA1 and BRCA2 testing, the company has invested heavily in research to support the test, as well as in educating physicians and working with third-party payers to make sure that the service is covered. He credits the company’s efforts to get widespread insurance reimbursement with boosting access to the test.
Rather than a way to skirt the FDA regulatory path, it’s the nature of the technology behind these tests that put companies in a position to choose the CLIA-regulated test service model over the FDA route, Critchfield emphasized. “With the development of new ways of interrogating biologic systems, things like gene sequencing, expression arrays, and real-time PCR to look at messenger RNA levels, there are a variety of new technologies that have come forward and are developed to the point that we’re beginning to get a better clue of what’s going on in the underlying biology of disease,” Critchfield said. “As a result, these are very complex tests compared to some predecessor tests. It’s not like running a serum glucose.”
An Economic Philosophy
However, not everyone agrees that the proprietary test service model is the best, as a business strategy or a lab medicine practice. While Myriad and other companies that offer test services argue that the huge investments they put into developing a test, marketing it, and working with insurance companies allow them to offer wider access and more tightly controlled quality, others are not so sure. “It’s an interesting strategy, but I think it’s a flawed one,” said Edward Ashwood, MD, a senior vice president and director of laboratories at ARUP. He is also a professor of pathology at the University of Utah School of Medicine.
Ashwood supports a middle path for these kinds of tests in which companies offer the test themselves, but also license it out to other labs if they’re not going to market it via the FDA route. “As you look at lab test usage, it’s widely driven by market demand. And market demand stems from knowledge about the test, people in the know advocating for tests. And when there’s only one company advocating for a test, it gets stifled. So I think that holding a test tightly is not only bad for medicine, but it’s bad for business.”
He thinks that a better business strategy would be to loosen the hold on intellectual property and license it to anybody that wants it. “I think it’s a short term perspective—how do we get the most money out of this the quickest, where we force everybody to send to us. What do you really do long term? You encourage other people to figure out ways to get around you.” Ashwood also argues that when only one company is running a test, there’s less motivation to improve it. “No test is perfect when it first comes out,” he said. “They can all stand to be improved. And to improve a test, you need competition.” When a company licenses a test to other labs, it encourages others to find a way to do better, and in turn it pushes the company to keep improving their method to stay competitive, Ashwood concluded.
But testing advancements is an area where Myriad and other companies running exclusive test services feel they’ve really made their mark. Critchfield contends that because Myriad is the exclusive provider of the BRCA proprietary test, the company must invest heavily in research and improvements to remain on the cutting edge. “Again, if you come back to the question, ‘How can you best improve the test?’ we believe that concentrated efforts on test improvement, concentrated efforts on developing the market, concentrated efforts in developing data with research, all those things that are made possible by having an exclusive license really increase the availability of the test and increase the quality of the test,” said Critchfield.
The Kalorama report highlights how some companies in the business of offering proprietary tests services have experienced phenomenal growth. Myriad’s diagnostics unit has had nearly 50% compounded annual growth over the past 7 years. Genomic Health, which the report calls the most successful of the test service firms by far, saw a more than 10-fold increase since 2004 in the number of its Oncotype DX tests for breast cancer recurrence performed.
However, for both of these companies, success could have been a very different story if it weren’t for important favorable assessments included in guidelines from major professional societies and cancer organizations. For Myriad’s BRACAnalysis test, organizations like the American Society of Clinical Oncologists, the American College of Medical Genetics, the Society of Gynecological Oncologists, and the American Society of Breast Surgeons issued guidelines on appropriate evaluation of patients that would be candidates for BRCA testing. Similarly, the U.S. National Comprehensive Cancer Network 2008 Breast Cancer Treatment Guidelines and the 2007 update to the ASCO Recommendations for the Use of Tumor Markers in Breast Cancer included information on Genomic Health’s Oncotype DX testing service. It’s difficult to say how these tests would have fared without such backing from well-known organizations. Rosen noted that the choice to market a test as an exclusive service, license it out to other labs, or turn it into a kit to be sold via the FDA route often comes down to how much faith a company has in itself. “It just depends on how the company sees itself, how much money it has, and what its resources and competencies are,” said Rosen. “And in not licensing out a test, the company feels confident that they can offer a test service on their own.”
For many of the new proprietary test services coming to market now, success may arrive even more quickly, Rosen maintained. Many of these are tissue-based and feature more actionable results. This held true in 2007, when many of the most successful tests in the esoteric segment employed immunohistochemical stains, in situ hybridization, and PCR analysis of biopsied tissue, the Kalorama report found. “The tests that are coming out now, the tissue-based tests, they tell you straight up if the tumor is going to metastasize or not, or what type of drug it will respond to,” said Rosen. “They’re much more actionable, and they make a huge difference in how patients are treated and in the outlay by the insurance company for follow-up treatment.”
Regulation versus Innovation?
Up to this point, FDA has used its discretion in overlooking regulation of LDTs, Gutman noted, and labs and IVD companies have gotten used to considering LDTs as essentially off-limits to FDA. However, as early as 2004, it was clear that the agency viewed some of the newer, more complex LDTs differently.
Companies like Genomic Health and Correlogic, with its OvaCheck test for detecting early-stage ovarian cancer, started employing proteomics and genomics to predict the course of a disease or to catch the disease earlier. Significantly, many of the new test services used software to interpret the raw data generated from looking at multiple proteins or multiple genes. In Correlogic’s case, the company says it uses “artificial intelligence-based computer technology to identify hidden patterns” when looking at multiple proteins, protein fragments, and metabolites in blood.
This level of complexity, as well as the high-profile use of software for patient-specific interpretation, caught the FDA’s attention.“The IVDMIA guidance really reflected a major change in FDA policy with respect to laboratory-developed tests. Many of us who have worked in this area for several years were quite confident that the tests which FDA is now calling IVDMIAs fit squarely under FDA’s enforcement discretion policy for LDTs because these were performed by laboratories that developed these tests in-house without distribution by the laboratory of anything other than a box to collect samples and a report of results,” said Radensky.
Even though the FDA issued a revised draft guidance in 2007, many questions still remain unanswered, Radensky noted. “The revised guidance changed the definition, but it’s still a very subjective definition in that it really relies on concepts like transparency, and it’s not clear what is required for an interpretation function to be considered transparent.”
As more companies work to create innovative tests that take advantage of developments in proteomics and genomics, Radensky said that it’s an open question whether or not FDA will confine its extension of medical device regulation to those tests that use software for multivariate analysis, or whether this might be the first step toward FDA oversight of other LDTs as well. “Once FDA—by guidance—begins to regulate LDTs, it’s not clear that they won’t move to other laboratory-developed tests by the same mechanism.”
Radensky predicts that as technology progresses, FDA and Congress will have to come to terms with the fact that many of these complex tests do not fit well under medical device laws and regulations put into place 30 years ago. “I think that as we look to a new administration and a new Congress, and considering that science is advancing at a rapid pace, it is critically important that whatever regulatory framework is developed be one that can assure the public health while at the same time encourage and not hinder scientific innovation that can improve health outcomes,” said Radensky. “Any new regulatory framework must acknowledge the fact that these technologies really are very different from what Congress was considering when it passed the medical device amendments in 1976.”