In his recent book, NIH Director Francis Collins refers to DNA and the new science of genomics as “the language of life.” Thanks to the mapping of the human genome, says Collins, virtually all biomedical researchers agree “that their approach to understanding how life works has been profoundly and irreversibly affected….” This profound new knowledge, of course, is transforming health care delivery. In the genetic tests and technologies that have arisen from it, medical providers now routinely apply tools in front-line medical delivery that help detect, diagnose, and manage disease at the level of an individual’s own DNA, providing care that is “personalized” to the exact molecular needs of the patient.
Accompanying the fast-pace of innovation in genetic tests has been the growing question of how this remarkable new science—as reflected in genetic tests—should be regulated. The Food and Drug Administration currently holds jurisdiction over genetic tests manufactured and sold to laboratories and other medical providers as packaged test kits (in-vitro diagnostics—IVD). The Centers for Medicare and Medicaid Services, under the Clinical Laboratory Improvement Amendments (CLIA), oversees clinical laboratories that develop genetic tests for use in their own facilities.
It seems that most everyone—providers, laboratories, device manufacturers, and patients—agree that the existing regulatory paradigms of CMS and FDA need to be updated. But how, in what way, and when?
Hot Topic in Washington
Those questions have been hot topics in Washington. On July 19-20, FDA held an open-door meeting to hear the views of stakeholders, ranging from hospital labs to patient groups, on the degree of regulatory oversight that is appropriate for laboratory developed tests (LDT) of all types. A few days later, a House Energy and Commerce subcommittee focused specifically on shortcomings in direct-to-consumer (DTC) genetic tests, which consumers can readily buy on the Internet. A Government Accountability Office (GAO) report found that DTC genetic testing was plagued with errors and erroneous marketing, and concluded that consumers could not rely upon it. While some Members of Congress stressed that the regulatory response to DTC genetic tests—whatever it turns out to be—should be measured, so as not to harm genetic tests used by physicians in clinical care, others seemed to lump all genetic testing into the category of “snake oil.”
So where do we go from here?
For starters, as a faculty member in the Department of Pathology at the University of Utah, as well as a former executive of ARUP Laboratories, a national clinical and anatomic pathology reference laboratory owned by the University of Utah, I would offer this observation: I see first-hand every day how genetic testing is dramatically changing medical care for the better, giving doctors more precise information for patient diagnosis and management. The statistics from peer-reviewed literature bear out in my own experience—children with leukemia, young people with HIV, and older folks with heart disease are defeating medical problems today that were certain killers even a decade ago.
With regard to direct-to-consumer genetic tests, I have a strong faith in appropriately informed consumers, and I see the value of getting consumers actively involved in understanding and managing their own health. Part of this, I believe, can involve direct-to-consumer genetic tests that allow consumers to obtain their own genomic information. But it is clear that a number of safeguards to better assure informed use—including the involvement of a physician in ordering and interpreting such tests—must be in place. In most cases, this has not yet occurred.
What worries me most, however, as Washington sorts out the appropriate regulatory path forward, is that it will lump all genetic tests into the same category and give them the same regulatory treatment. Even worse, I fear that criticisms leveled at a handful of DTC genetic testing companies for making misleading marketing claims could tar the entire field with the same “snake oil” brush.
This could harm innovation in genetic medicine for decades.
Thus, I believe it is critical that—as FDA, CMS, and Congress decide the regulatory future of genetic tests—that policymakers fully understand that genetic tests are not a single or uniform thing. They are many different kinds of things. They play different roles and are used in different ways. Yes, they all use some aspect of genetic or genomic science as their internal functioning force, but they are often no more similar than, say, lawnmowers and Formula One race cars that all rely upon the internal combustion engine as their primary functioning force.
Two Pathways for Innovation
Perhaps the most important distinction is that genetic tests come about through two very distinct methods of innovation, and are applied in distinctly different settings.
The first avenue of innovation is via medical device manufacturers which incorporate genetic tests into test kits. These are packaged as individual products that are sold and shipped, through interstate commerce, to hospitals, clinics, laboratories, or other health providers, where they are used by a variety of medical professionals and patients to aid in diagnosing and managing medical conditions. FDA regulates these tests by requiring kit manufacturers to register and list the products they produce, and to undergo varying degrees of pre-market review and approval.
The second primary avenue of innovation in genetic tests is via clinical laboratories, which often develop their own tests in-house and perform these tests at their own facilities. Laboratories create these LDTs by purchasing the bulk reagents and other inputs and establishing an internally validated, standard operating procedure for performing the test. As a result, no “kit” is created; in fact, there is no medical device of any sort created. An LDT is simply a process for how to perform a particular test in the laboratory. These laboratories must comply with the quality and validation requirements of CLIA, which are administered by CMS.
Why is this difference important?
It is important to understand this distinction because the fundamental focus and operations of device manufacturers and clinical laboratories are very different. Device makers are geared toward developing products that can be standardized, manufactured, packaged, and used in distant locations by third parties who have varying degrees of knowledge and training. The focus must be on products that can be reliably produced on a large scale, shipped throughout the country, stored until ready for use, and applied across settings and levels of expertise. These products can not address medical conditions that morph so quickly — rapid mutations of the HIV virus, for example, or newly emerging pathogens such as H1N1 — that the tests are outdated by the time they have undergone FDA review, which can take many months, and often years. There are also important examples in human oncology and genetics where new mutation discoveries need to be quickly added to those already in existing genetic tests in order to be clinically useful.
FDA regulation is tailored to fit this device model of innovation. It requires clinical data, pre-market review, evidence of good manufacturing practices, and post-market surveillance to ensure that these packaged test kits will perform reliably and accurately, regardless of the setting or the nature of the user.
In contrast, laboratory developed tests, as noted, are not “products” at all, but are medical services that are performed by expert laboratory professionals within their own laboratory facilities. Lab personnel often include molecular geneticists, molecular pathologists, clinical pathologists, and clinical laboratory scientists. These experts are responsible for carrying out the tests on-site, ensuring they meet all CLIA-required quality standards, and that they are properly recorded and conveyed to the physician who ordered them.
One of the great values of LDTs is that they can quickly adopt the most current peer-reviewed scientific and clinical information into tests that then can be made available to patients promptly. Genetic tests developed in clinical labs include HIV resistance testing; BCR/ABL testing for chronic myelogenous leukemia; Fragile X testing for developmental delays; KRAS mutation testing to reduce or eliminate side-effects of colon cancer treatment; and testing to identify how quickly a person’s body metabolizes certain drugs, thus permitting more effective therapeutic dosing and reduced side-effects. LDTs have been vital in testing for rare genetic diseases like Gaucher, Tay-Sachs, and Canavan among many others. In addition, advanced diagnostic tests have helped protect the nation’s public health by allowing for fast identification of SARs, Avian Flu, West Nile Virus, and more recently H1N1.
In creating CLIA, Congress established a regulatory framework that fits this style of innovation. CLIA targets the accuracy and reliability of the process used for testing, factoring-in analytical performance characteristics, quality control, external proficiency testing, credentials of the testing personnel, documentation of standard operating procedures, and the results reporting requirements. In essence, its focus is the laboratory and the systems and personnel that perform the test.
I believe that as policymakers attempt to address any gaps in LDT regulation between FDA and CLIA, it is important that they NOT treat these two distinct methods of innovation in the same way, but rather do so in a complementary way, incorporating the best of both systems.
Tests Used by Physicians vs Consumers
A further distinction is critical in understanding genetic tests and in determining the appropriate regulatory framework. This distinction focuses not on how innovation occurs, but in how the tests are used and by whom—physicians or consumers. This distinction is often overlooked in discussions about the performance and appropriate regulation of genetic tests—yet it is fundamental in designing regulatory systems that truly fit any gaps in oversight.
Genetic tests used by physicians provide information to aid in the physician’s understanding of a patient’s condition and in deciding an appropriate treatment strategy. Such tests are based upon sound, well-established evidence, supported by peer-reviewed literature. Today, such tests are bringing about dramatic improvements in real-world care. For example, HIV tests allow physicians to tailor anti-HIV drugs to an individual’s specific viral genetic profile—and to adjust them as the virus develops drug resistance. Genetic testing that identifies treatment-specific subtypes of acute leukemia has helped improve the 10-year survival for many patients. The presence of certain mutations in a patient’s cancer now allows oncologists to more specifically, effectively, and safely treat their disease. Examples include HER-2/neu in breast cancer and KRAS in metastatic colon cancer.
In contrast, genetic tests aimed at consumers—which are the target of the GAO report and House hearing mentioned above—are focused on providing a statistical estimate of the likely risk of a medical condition or disease in the future, so-called “genome-wide disease associations.” Thus, they are considered to be predictive tests; not diagnostic tests. Predictions can range from the likelihood of certain physical traits, like going bald, to the likelihood of developing certain life-threatening conditions. Because these tests are widely available to consumers, they are often not ordered under the guidance of a physician, nor are they interpreted with the help of a physician or genetics counselor. Thus, some consumers can be confused and frightened because they may be limited in understanding the meaning or real significance of the test results.
In particular, consumers might easily conclude that a medical condition or disease that is predicted—even as a small possibility—will absolutely occur in the future. This overlooks, of course, the fact that these possibilities may never come to fruition, even with “above average risk.”
It is important to re-iterate that the GAO report, as well as the news coverage that emerged from it and from the July 22 Energy and Commerce hearing, has been focused entirely on genetic tests that fall into this latter category—direct-to-consumer tests. Criticism has not been focused on the genetic tests used by physicians in identifying disease and directing care. Nonetheless, these two distinct groups of tests have been lumped together—with no recognition of the distinctions between them, neither in the degree of evidence supporting them nor in the types of training and expertise the individuals who use them possess.
The grave danger here is that the anger expressed by Members of Congress over DTC tests and marketing practices will, by extension, fall upon sound, well researched, and vital clinical tests that suffer none of the problems reported for DTC testing. More specifically, this could mean that the regulatory interventions that are designed to curb DTC testing can easily harm the testing so critical to frontline medical care for cancer, HIV, heart disease, and more. The lesson: The regulatory solutions must be tailored to the problem. That rings out especially clearly when you realize that DTC testing comprises less than one percent of clinical laboratory testing in the United States.
On the same day as the House Energy and Commerce Subcommittee hearing, FDA Commissioner Margaret Hamburg and NIH Director Francis Collins published a commentary in the New England Journal of Medicine describing the regulatory and research components of what they called “a path to personalized medicine.” They likened the development of such a pathway to creation of the interstate highway system. “When the federal government created the national highway system, it did not tell people where to drive—it built roads and set the standards for safety,” they wrote. Since then, they point out, the investments in highways have supported extensive innovation—which they call a revolution—in transportation, commerce, and personal mobility.
Their analogy is apt. The rules of the road for personalized medicine will, in large part, define the future and structure of genetic testing and technology. In this regard, then, it is worth noting that the nation’s interstate highways are critical avenues for commerce and everyday life, with on-and-off ramps essential to local communities, rural regions, and big cities alike. Interstate highways have to be designed with full recognition that access to the highways must be available to a wide variety of unique communities, businesses, and population centers.
Similarly, the regulatory infrastructure that is now being developed for genetic testing and personalized medicine must be easily accessible, clearly marked, easy-to-use, and logically constructed so that innovative genetic tests and technologies can enter the system smoothly and safely. Perhaps the most important place to start is understanding that not all genetic tests are alike.