Skip to main content
Premium Trial:

Request an Annual Quote

Advisory Committee Tells FDA Clinical Genetic Tests Should Only Be Provided Through Docs, Not DTC


By Turna Ray

GAITHERSBURG, Md. — Genetic tests that could potentially inform medical care should not be provided to consumers without the involvement of a doctor, an independent expert panel advised the US Food and Drug Administration this week.

The Molecular and Clinical Genetics Panel of FDA's Medical Devices Advisory Committee convened this week for a two-day meeting to consider the circumstances under which genetic tests can be marketed directly to consumers, and when genetic tests would require the involvement of a physician.

On the first day, March 8, the panel was asked to discuss whether DTC marketing should be permitted for five categories of tests that the FDA believes could be used to make healthcare decisions: those that gauge whether people are carriers of genes that might have disease implications for their children; tests that determine whether an asymptomatic person has a high likelihood of getting certain diseases or disorders in the future; tests that assess a person's susceptibility for disease; pharmacogenetic tests; and nutrigenetic tests.

With the exception of nutrigenetic tests — which panelists felt most comfortable allowing consumers to access on their own — committee members largely held that consumers should only be able to access so-called "clinical genetic tests" through their doctors. Even though they were of the view that much of the data provided by DTC genomics firms have limited utility, the majority of panelists felt strongly that without the involvement of a medical professional, patients may misunderstand or fail to recognize when a genetic test result is meaningful for their health.

Valerie Ng of the University of California, San Francisco, a temporary non-voting committee member, summed up the sentiments of most panelists when she noted that she is not too concerned if consumers are not consulting doctors about results of tests that analyze gene variants that account for a small risk for a particular disease compared to other risk factors. "But if the test is meaningful, and I can do something about it, then I want a doctor involved," Ng said.

Several presenters during the meeting discussed data from surveys — such as the Multiplex Initiative by the National Human Genome Research Institute and a behavioral study by the Scripps Translational Research Institute of early adopters of DTC genomics services — which showed that genetic data has a limited impact on people's behavior and that a minority of people share their test reports with genetic counselors or their doctors. However, these surveys also found those who shared their test results with their doctors also appear to be the most motivated to make changes to their lifestyle (PGx Reporter 07/29/09, 11/12/11).

Ahead of the advisory committee meeting, the American Medical Association sent a letter to the panel advising that all genetic testing should be conducted with the guidance of a physician or a genetic counselor or other genetic specialist. While the committee appeared to generally agree with the AMA, the members of the Molecular and Clinical Genetics Panel also recognized that many medical professionals are unprepared to interpret genetic test results and are in dire need of genetics education.

Although the committee members did not vote on the questions FDA put before them, their discussion on the risks and benefits of offering genetic tests DTC will help the FDA craft regulatory requirements for firms such as 23andMe, Decode Genetics, and Navigenics.

Last summer, the FDA warned these companies and several other DTC genomics firms that their genome scanning services were medical devices and that they needed to comply with the agency's regulations for such products (PGx Reporter 05/26/10).

At this week's meeting, FDA officials reiterated that DTC genomics firms do not fall under its longstanding practice of "enforcement discretion" for certain laboratory-developed tests, which are currently under the purview of the Centers for Medicare & Medicaid Services.

"It's not under question that [DTC genomics firms] will be regulated," Alberto Gutierrez, director of the Office of In Vitro Diagnostics in FDA's Center for Devices and Radiological Health, said during the meeting. "They will be."

Elizabeth Mansfield, director for personalized medicine at OIVD, said that the agency and DTC genomics firms are in the process of working out how these companies can come under compliance with FDA's medical device regulations.

"You will not be surprised to hear that this has been a challenge, not only for the companies involved, but for FDA as well, as we work to create a reasonable and fair path forward," Mansfield said.

FDA's decision to regulate DTC genomics firms has polarized stakeholders into two camps: those who view the FDA's position as paternalistic and feel that these tests should be available directly to consumers who want to take charge of their own health; and those who feel that the DTC model poses potential harm to consumers who may misinterpret test results.

"FDA is of the view that there is a place for DTC genetic testing, but that appropriate oversight should be applied to ensure that individuals are protected from low-quality testing, overly ambitious clinical interpretation, and breach of information if privacy is not carefully handled," Mansfield said.

Certainly, FDA's move to regulate DTC genetic firms has "narrowed the field," she added. Since FDA's summer enforcement action, several firms, such as Navigenics and Pathway Genomics, have moved away from the DTC model and require a physician's involvement to order a test. Meanwhile, companies like 23andMe and Decode are still providing genetic tests DTC.

[ pagebreak ]

Capturing Risks

While data presented at the meeting based on surveys of DTC genomics customers did not suggest that this information causes psychological harm or inspires people to make rash medical decisions on their own, several committee members pointed out that the full picture regarding the true impact of these tests will require large outcomes studies with decades-long follow-up. Furthermore, currently there is no mechanism for gauging adverse events that result from DTC genetic tests.

Steven Lipkin, a medical geneticist at Weill Cornell Medical College and a temporary non-voting member of the panel, shared several patient anecdotes involving genetic test results from 23andMe — one in which the patient properly interpreted the data and another in which the patient "misinterpreted the results of the test."

Lipkin suggested that potential adverse events associated with DTC genetic tests may be captured in the post-marketing setting, just as adverse reactions with drugs are identified, if mechanisms were in place that monitored for these events. "I just mentioned two outcomes," Lipkin said. "There may be hundreds of others, or maybe thousands, that are similar; we just don't know about them."

According to Ashley Gould, general counsel for 23andMe, DTC genomics firms are well-positioned to capture post-marketing adverse events from genetic tests. Because DTC firms are connected to their customers through web-based networks, these companies can ask them about adverse reactions through online surveys.

The FDA has previously expressed a desire to collaborate with DTC genomics firms to track which patients are experiencing adverse reactions to drugs. Although no such post-marketing surveillance efforts between FDA and DTC genomics firms have yet been announced, Gould noted that the willingness of 23andMe consumers to participate in research and to answer survey questions is a good signal that the company would be able to pick up on whether people are having adverse events related to genetic tests.

23andMe has genotyped 75,000 customers to date. Of those, 78 percent have consented to participate in the company's IRB-approved research studies, and more than 83 percent of customers have answered at least one survey.

Gould added that 23andMe customers can reach customer service reps via e-mail or phone, with whom they can report issues with their data or complain about the quality of the service. "To date we have no anecdotal evidence to suggest that any of the voiced concerns pose real, demonstrable risk to individuals," she said.

A Nuanced View

Committee members recognized during the meeting that the way a doctor manages genetic data for a particular patient is highly variable from disease to disease. As such, some genetic tests might pose less risk to consumers if provided DTC than others would.

"Testing is complex, and for any one of these categories" of tests that the FDA has asked the committee to consider, "it is very difficult to make conclusions that cover every instance," observed Gregory Tsongalis of Dartmouth-Hitchcock Medical Center, a non-voting member of the committee.

For example, the results of carrier testing can be important in family planning. But sometimes in the case of a really rare genetic disease, doctors might refuse to perform a test for a couple if they have no family history for that disease. In such cases, DTC access to carrier testing might provide some peace of mind.

Conversely, test results may provide a false sense of security in consumers if their results show they are not at risk for a disease for which genetic knowledge is incomplete.

Tiffany House, the patient representative on the panel, suffers from Pompe disease and questioned whether it is really a cause for concern that patients who receive worrisome genetic test results through DTC channels would not discuss this data with their physician. "I would think that people would run to their doctor with their results," House said.

Jeff Gulcher, chief scientific officer at Decode Genetics, suggested that there was no need to ban DTC availability of genetic tests, since consumers cannot truly act on this information without the help of a doctor. He noted that for patients to request a biopsy or a particular drug based on the results of a DTC test, the consumer would need to discuss data from genetic tests with their doctor. While there is a risk that patients might stop taking a drug or adjust the dose of their drug based on a DTC test results without input from a physician, there is no evidence to suggest that this is currently happening, he added.

Another point of debate among committee members was whether physicians understand genetics enough to incorporate genetic risk data into the overall management of their patients' health. Gulcher maintained that doctors were perfectly capable of understanding relative risk estimates, but members of the committee disagreed.

"One thing, in my experience, that physicians understand poorly is how to correlate an odds ratio to a positive predictive value," Ng said, asking the DTC genomics firms present at the meeting whether they would consider providing genetic risk data in the context of family history and other medical indices, such as body mass index.

Most panelists felt that patients would be better served if they knew how much their genes contributed to their overall risk for a disease compared to other contributing factors. 23andMe said that it plans to conduct research on how to incorporate genetic data with other types of risk considerations such as environmental factors and family history.

Gulcher maintained, however, that family history is not a good surrogate for the genetic risk of common diseases. "It accounts for a very small portion of the genetic risk. It has been estimated that 50 percent to 70 percent of risk of most common diseases is genetic versus environment," Gulcher said, adding that no conventional risk factors exist for many of the most common diseases, such as prostate cancer.

"For the common diseases, you're never going to be in a position to determine with certainty that a patient is going to develop, [for example,] prostate cancer or find a patient whose risk is so low that he is immune to prostate cancer," he said. "We're never going to get to that position."

Despite these open questions on how doctors might factor genetic risk data into clinical diagnoses, the general consensus of the committee was that genetic tests should be provided with physician involvement until more research is done to validate the strength of gene-disease associations and there is more knowledge about the interplay of genetics and other risk factors in causing disease.

Robert Shamburek of the National Heart, Lung, and Blood Institute and a temporary non-voting member of the committee proposed that genetic tests should gain market experience as prescription-based tests before going DTC, just as prescription drugs are moved over the counter only after years of gauging the drug's impact on the population in a post-market setting.

He noted that it could take between five and ten years "before we know what can be valid [tests]."

Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at tray [at] genomeweb [.] com.