10 Years After the Human Genome Project
Will the DTC Genetic Test Debate Shape the Future of Genomics?
By Bill Malone
Although many voices have sung the praises of the 10-year old genomic era, today few in the medical community seem able to navigate confidently between the buoyant predictions of its evolutionary impact on medicine and real-world applications of a rapidly expanding body of genomic knowledge. Among the celebrations marking this milestone anniversary, a controversial debate is being played out between direct-to-consumer (DTC) genetic testing companies and federal regulators. In the past several years, these firms have pushed the genomic era right into the public’s hands via the Internet. But now they find themselves under the spotlight of the Food and Drug Administration (FDA) and surrounded by a medical community that’s highly skeptical of the actual utility of these new testing services that typically include disease risk assessment based on DNA profiling.
Convened March 8 and 9 in Gaithersburg, Md., the FDA’s Molecular and Clinical Genetics Advisory Committee expressed deep suspicion about DTC genetic testing, warning that this new world of uncertain and complex genomic information was too risky for consumers to handle without the help of clinicians. Although FDA is not bound to accept advisory committee recommendations, in most cases it does. However, unlike other panels the agency convenes, no definitive vote was taken on whether DTC tests should stay on the market.
Due to the ambitious strides of DTC companies and the growing interest among some consumers, DTC genetic testing may have arrived at a pivotal point for genomics, with many stakeholders watching closely to see how regulators, industry, and the medical community will work together, according to Dan Vorhaus, JD, an attorney with Robinson Bradshaw & Hinson. “We’re now 10 years on from the Human Genome Project, and recently people have wanted to take stock of the past decade and consider what’s on the horizon,” Vorhaus said. “I think DTCs are one of several focal points where people look and ask what we’ve really gotten out of genomics. Certainly, we have something quite amazing when consumers can receive personalized genetic information directly and explore a part of themselves they have never had access to before. But others look at DTC and say, ‘well if this is what we’ve got, we don’t have that much, because we don’t see this as being all that useful.’ And that’s why we need these open discussions with FDA and other stakeholders so we can dig a little deeper into the value of DTC.” Vorhaus has represented DTC firms and is editor of the Genomics Law Report.
Already…But Not Yet
As researchers have toiled to connect the myriad of dots between genotype and phenotype, DTC firms have charged forward with nascent findings in the published literature and capitalized on consumers’ fascination with their own DNA. In a year in which scientists are celebrating the 10th anniversary of the seminal publications on the Human Genome Project in Nature and Science, the discussion of DTC genetic testing at the FDA advisory committee meeting echoed themes from a wider discussion taking place: the growing pains of a new genomic era where technology and scientific discovery seem to move faster than the healthcare community’s ability to keep up.
The tensions between excitement about new discoveries and caution about the many challenges that remain in using genomics to improve health is captured in the National Human Genome Research Institute’s (NHGRI) new strategic plan, “Charting a course for genomic medicine from base pairs to bedside,” published in the February 10 issue ofNature (2011;470,204–213). “Although genomics has already begun to improve diagnostics and treatments in a few circumstances, profound improvements in the effectiveness of healthcare cannot realistically be expected for many years,” the authors wrote.
The report does highlight some of the early ways that genomics is being used in lab medicine. To guide effective therapy, breast and colorectal cancer patients are now tested for Her2 and KRAS gene mutations, respectively, and genetically guided prescriptions of the anti-retroviral abacavir (Ziagen) are now the standard-of-care for HIV-infected patients. In addition, some clinicians are taking advantage of genetic testing to guide therapy for certain drugs like tamoxifen, clopidogrel (Plavix), and warfarin.
But for the most part, genomics remains a research, rather than a clinical enterprise, something that will not change quickly, according to Eric Green, MD, PhD, NHGRI director and an author of the Nature paper. “The next decade will bring advances in our understanding of the biology of disease; however, truly demonstrating improved effectiveness in healthcare is something that probably will not have the greatest set of accomplishments in the next decade, just because of the lengthy process it takes to actually implement the changes that prove to be effective in healthcare,” Green told CLN. “There are not too many examples where it’s been taken all the way across the finish line to demonstrating improved effectiveness, partly because there is just so much genomic knowledge that physicians can’t possibly be expected to know about because of the sheer newness of this.” Green assumed the position as director of NHGRI in December of 2009, a post formerly held by National Institutes of Health Director Francis Collins.
Technological advances like next-generation sequencing have both sped up genomics research and dramatically lowered costs for sequencing (See Box, below). However, the ability of the research and medical communities to handle the deluge of data has not kept pace, driving a wedge between technological leaps and clinical application for personalized medicine in diagnostics and other disciplines, noted Linnea Baudhuin, PhD, co-director of the nucleotide polymorphism laboratory, co-director of cardiovascular laboratory medicine, and assistant professor of laboratory medicine at Mayo Clinic in Rochester, Minn. “Personalized medicine is coming, but there are still a lot of hurdles we need to overcome, including huge bioinformatics and data handling issues, before we can integrate all the information we receive, make sense of it, and put it into a form that clinicians can actually use,” she explained. “Pharmacogenomics has been more challenging to integrate into the clinic than we previously imagined. Some of this is due to the complexities and unknowns in the field that we don’t have answers to right now.”
Genome Sequencing Costs Plunge
Steady improvements in DNA sequencing technology have dramatically lowered costs for whole-genome sequencing in the past decade, even outpacing the benchmark rate of computer innovation known as Moore’s Law. However, experts caution that despite these advances, many questions need to be answered before routine healthcare can take full advantage of the results of the Human Genome Project.
Click here for chart
Source: Wetterstrand KA. DNA Sequencing Costs: Data from the NHGRI Large-Scale Genome Sequencing Program. Available online.
Compounding the issue are the majority of physicians who themselves feel uncomfortable using genetic testing. In a 2009 survey of more than 10,000 physicians by the American Medical Association and Medco Health Solutions, only one in 10 said they felt they had the sufficient knowledge and training to use pharmacogenomic tests, and only 25% had any education at all on how to do so. For example, warfarin sensitivity genotyping remains controversial.
Trusting Consumers with Genetic Results
At the FDA meeting, panel members grappled with the risks of allowing consumers to order and receive their own genetic test results while acknowledging that even if the tests were prescription only, physicians might not do much better in interpreting results. “Are we saying that even though this genetic information is so complex, if the risk to the consumer does not seem high, the bar will get lowered for evidence on tests that even the docs don’t understand?” asked advisory committee member David Ransohoff, MD, professor of medicine, cancer epidemiology, cancer prevention and control at the Lineberger Comprehensive Cancer Center at the University of North Carolina in Chapel Hill.
Both panel members and DTC industry representatives noted that for many of the diseases for which DTC genetic tests convey risk information, environmental, and behavioral factors play an equal or greater role. For example, most of the DTC tests on the market predict cardiovascular disease or diabetes risk based on genetic markers. However, conventional measures like blood pressure, lipids, and glucose give a clearer picture, panel members said.
The panel worried that consumers would not understand the limits of these predictive tests, while DTC industry advocates contended that the way they present results to consumers makes this very clear, and in fact reduces risk by underscoring the informational aspect of the test as opposed to a diagnostic result. But asking consumers to piece together this genetic information along with other factors seems even more risky, Ransohoff said. “This is like asking consumers to practice medicine—expecting them to integrate other information into their genetic data. Another reason these tests aren’t safe for DTC.”
Ironically, so far, neither regulators nor researchers have been able to agree on any documented cases of consumers experiencing harm from accessing DTC genetic tests. In fact, a study published this year in The New England Journal of Medicine looked specifically at this issue, and did not find that predictions of disease risk from DTC tests made consumers anxious; neither did the subjects in the study modify their health-related behavior or seek excessive follow-up screening tests (2011;364:524–534). Cinnamon Bloss, PhD, an author of the study, presented the findings to the advisory panel at the invitation of the FDA.
Led by the La Jolla, Calif.-based Scripps Research Institute, the study included subjects who elected to buy the Navigenics Health Compass genomewide profiling test at a discounted rate through their employers’ health plans. The subjects reported any symptoms of anxiety, their intake of dietary fat, and exercise behavior via an online survey approximately 6 months after receiving their genomic test report. The researchers also asked the subjects about their use of additional screening tests.
More than 90% of the subjects reported no test-related distress, a surprise to the Scripps researchers, according to Eric Topol, MD, director of the Scripps Translational Science Institute and an author of the study. “We were very concerned about people getting all of this data for 20-some diseases, including Alzheimer’s, cancer, and heart attack risk,” Topol said. “We expected that these results would really stir people, and they didn’t. So that was gratifying to see, and it’s certainly a step in the right direction if we’re really about to go into high gear in the genomic era.” Topol is also the chief academic officer for Scripps Health and a professor of translational genomics at the Scripps Research Institute.
Topol also pointed out that the study demonstrates strong evidence that consumers take a measured approach to targeted screening based on their genomic test results. For example, in those for whom the genomewide profiling indicated a strong risk of colon cancer, a sizeable number reported that they intended to get a colonoscopy. “This was a significant response, and it was very targeted. It wasn’t people just wanting more tests in general,” he said. “Rather, they reported the intention to request screening tests specific to their risk according to the genomic test.” Several companies that Scripps approached in order to offer the genomic test to their employees declined to participate in the study, in part due to concern over excessive resource consumption from unnecessary follow-up tests, Topol said.
The Scripps team plans future publications as they continue to follow the same cohort, including how the subjects respond to pharmacogenomic data. “There is a lot left to be done,” Topol said. “So we are bracing ourselves for the next phase of this story. What’s good is that I believe we’re dealing with a consumer that’s more readily capable of coping with this than most physicians would have predicted.”
DTC Tests Remain in Limbo
Even with presentations from DTC companies and other stakeholders designed to assuage the advisory committee’s worries, overall the committee could not get comfortable with the idea of DTC tests with health claims that bypass clinician oversight. The panel generally agreed that several categories of specific genetic tests should be offered solely upon prescription, such as pre-symptomatic tests with a high predictor for disease and pharmacogenetic tests. Moreover, the panel reiterated concerns throughout the meeting that the current state of scientific knowledge may not warrant the risk assessment claims being made by DTC companies.
Companies offering DTC genetic testing framed the questions differently, focusing on the consumer’s right to access his or her own DNA and the idea that DTC tests empower individuals to become more informed healthcare consumers. In its presentation to the advisory committee, industry leader 23andMe’s general counsel, Ashley Gould, emphasized that the company has more than 75,000 genotyped customers to date, with no evidence to suggest that any of the common concerns about DTC testing pose “real, demonstrable risk to individuals. It is imperative that policy and regulation be based on facts and evidence about how consumers respond to learning directly about their genetic information rather than assumptions about possible irrational consumer behavior and fears that have not been substantiated.” Gould’s presentation also underscored the way in which the company uses repeated disclaimers and specially designed charts and graphics to make sure information is presented clearly.
As FDA mulls the advisory committee suggestions, invited presentations, and public comments from the meeting, the agency does not seem to be gearing up for a dramatic shift in its approach to DTC genetic testing. At least for the short term, these tests will likely stay on the market, but will all require some kind of FDA review of analytical validity, labeling, and other elements.
In calling for the advisory committee meeting, FDA was focused on hearing from experts so that the agency can ask the right kind of scientific questions about a new era of genetic testing, balancing its mission to protect consumers and also promote innovation, said Alberto Gutierrez, PhD, director of FDA’s Office of In Vitro Diagnostic Device Evaluation and Safety. “In regulating in vitro diagnostic tests, we look to see whether a test is safe and effective, and that’s what we will do for these tests. We don’t see a reason to either change the way we do business or how we determine whether something is safe and effective,” Gutierrez said. “There are scientific questions we need to address—question of evidence—and it’s clear that we have seen not only in DTC tests, but also in some other tests, that the science is moving very quickly.”
Currently, FDA is working with DTC companies on a case-by-case basis. “We use enforcement discretion if we think a company is actively trying to get into compliance by submitting for agency review,” Gutierrez said. “If we think that they’re moving in the right direction, so far we have not required that they take the test off the market while they work with the agency.”
According to Vorhaus, the current political climate in Washington should tend to push FDA away from draconian restrictions on the DTC market. “There is pressure from both sides of the aisle, including the White House, not to over-regulate,” he said. “The Obama administration has talked about modernizing FDA and not slowing down innovation—a message that Republicans echo as well. In this kind of environment, you have to wonder about the political support to introduce new regulatory schemes that require more resources and take away agency focus from other priorities.”
The FDA meeting on DTC tests should be good news for labs worried about FDA’s posture on regulating laboratory-developed tests (LDT), Vorhaus said (See Box, below). “In the FDA public meeting about regulating all LDTs last summer, DTC tests were sort of lumped in there. In some ways it made sense, but it was also a distraction, because there are different issues for the two,” he explained. “Both labs and DTC companies worried that DTC testing was the tail wagging the dog. My sense is that both groups are going to be pretty happy with FDA getting its thinking a little bit clearer and separating out the two groups.”
The History of Direct-to-Consumer Genetic Test Regulation
- A Government Accountability Office (GAO) report finds medically unproven and ambiguous results for DTC nutrigenomic tests that predict effects of foods and food constituents on gene expression.
- FDA releases its first guidance on In Vitro Diagnostic Multivariate Index Assays (IVDMIA), a harbinger of the agency’s heightened interest regulating all laboratory-developed tests (LDT), a category claimed by DTC firms as a rationale not to seek FDA review.
- Major players deCODE Genetics and 23andMe launch tests looking at thousands of single-nucleotide polymorphisms.
- Knome launches the first commercial whole-genome sequencing and analysis service for consumers.
- Navigenics launches its DTC service with a more clinical focus, leading a trend toward more health-related claims.
- Department of Health and Human Services Secretary’s Advisory Committee on Genetics, Health, and Society (SACGHS) publishes a biting review of DTC genetic testing that recommends greater FDA oversight and a public test registry.
- FDA sends letters to DTC companies stating that their products are medical devices and begins meeting with DTC firms.
- New York and California send “cease and desist” letters to dozens of DTC genetic testing companies, requiring state licenses.
- One month before FDA announces a public meeting on regulating all LDTs, Walgreens announces an agreement to sell Pathway Genomics’ DTC “Discover Your DNA” genetic tests. In response to an FDA letter, the companies put the deal on hold. FDA soon sends out 19 more letters to DTC firms.
- Knome, Pathway, and Navigenics exit the DTC market, while deCODE limits some offerings.
- GAO releases a controversial report concluding that many DTC genetic tests use deceptive marketing and offer misleading results.
- DTC firms continue talks with FDA as some tests remain on the market.
- FDA convenes an advisory committee meeting to discuss the risks of DTC genetic testing and panel members express skepticism of the utility and safety of the tests.
Looking to the Future
Even though the woes of DTC genetic testing companies may seem far from the clinical lab, many of the same scientific, regulatory, and ethical questions that FDA faces will eventually have to be answered by laboratorians as well.
“When we start to do things like whole genome or whole exome sequencing, the increased information will push laboratorians into a role where they will really need to become skilled at integrating and communicating with clinical specialists in a wide variety of disease focus areas,” Baudhuin noted. “Additionally, laboratorians will need to increase their knowledge base in order to navigate the complex world of pharmacogenomics to provide optimal drug prescribing information to physicians, pharmacists, and patients.”
NIH’s Green emphasized that as clinicians become overwhelmed with the influx of genomic data, labs will have an opportunity to be on the front lines providing interpretation and context. “The traditionally consultative role that people in lab medicine play in providing advice will be very well suited for the genomic era, where there will be an overwhelming amount of genomic data on patients,” he said. “Genomic medicine plays to the strength of lab medicine and consultative pathologists in a way that will continue to empower them to play major roles in medicine. That said, the field needs to evolve to join up effectively with genomics, they can’t just sit back and say, ‘we’ll just do business as usual.’ They have to realize that the landscape is changing, and they should embrace that and become invigorated by it.”
Pharmacogenomics Faces Complex Hurdles
Warfarin Genotyping Illustrates Uncertainty
Chief among the boons expected from the genomic era has been personalized medicine, where molecular diagnostics enable medical interventions tailored to individual patients. However, like all things genomic, questions persist about exactly how this new depth of knowledge about patient’s genotypes can be translated into better care.
The debate about warfarin sensitivity genotyping is a case in point, a controversy that remains unresolved even while 2 million Americans enter the dangerous initiation phase of this anticoagulant therapy each year. After studies confirmed that CYP2C9 and VKORC1polymorphisms could predict a person’s response to the drug, the Food and Drug Administration (FDA) updated the drug’s label in 2007 with a suggestion that physicians consider lower initiation doses for patients with variations in these genes. FDA then updated the label again in January 2010, referring prescribers to a table of stable maintenance doses that studies had found in patients with certain combinations of genetic variants.
Yet clinical practice guidelines have remained unchanged for warfarin, and warfarin sensitivity testing is rare. However, since FDA’s label changes, a study showing a dramatic reduction in the risk of hospitalization for genotyped patients has drawn increased attention to the issue (J Am Coll Cardiol 2010;55:2804–2812). A collaboration of Mayo Clinic and pharmacy benefit firm Medco, the study was the first national, prospective, comparative effectiveness study of the impact of warfarin sensitivity genotyping on patients who initiated therapy in an outpatient setting. The study evaluated the impact of testing on hospitalization rates in the first 6 months of treatment with the drug compared to traditional care and found that, compared to the control group, the cohort of patients for whom warfarin sensitivity genotyping was preformed had 31% fewer hospitalizations overall and 28% fewer hospitalizations for bleeding or thromboembolism.
An NIH-funded double-blind, randomized, multicenter clinical trial is underway to confirm these findings, with preliminary results about 5 years out. In the meantime, physicians’ uncertainty about the test and the lack of reimbursement under the Centers for Medicare and Medicaid Services (CMS) has thwarted widespread utilization, explained Thomas Moyer, PhD, an author of the study. Moyer is professor of laboratory medicine, director of the toxicology laboratory, and chair of the division of clinical biochemistry and immunology at Mayo Clinic in Rochester, Minn. “Even with the FDA label changes, CMS still will not pay for it because it could raise the direct cost to them,” he said. “CMS is taking a rather parochial view that we can’t afford to spend more without a demonstrated payback. We, however, think that our paper does demonstrate the payback, and we’re in the process right now of doing the final statistical analysis of data for a full cost-effectiveness evaluation. So with FDA saying yes and CMS no, I think it’s a logical response by physicians who don’t want their patients to bear the burden of paying for it.”