Science Progress | Where science, technology, and progressive policy meet

Defining the Boundaries of Genetic Testing

New Direct-to-Consumer Genetic Tests Raise Privacy Concerns

23andMe packet SOURCE: Flickr/keepthebyte A testing packet received in the mail from direct-to-consumer genetic testing company 23andMe.

A recently released study by direct-to-consumer genetic testing company 23andMe reveals the privacy challenges ahead for public health policymakers as so-called genome-wide association studies, which look for specific genetic traits in huge genetic databases, enter the mainstream of scientific inquiry.

23andMe, a personal genomics company co-founded in 2006 with investment from Google, Inc., just published its first genome-wide association study in the online journal PLoS Genetics. The decision by the journal’s editors to proceed with publication brings to light some of the more overlooked controversies surrounding direct-to-consumer, or DTC, genetic tests involving the methods of data collection and concerns about consumer privacy.

The study comes out of 23andMe’s participant-driven research program, 23andWe, which looks for genetic correlations for traits such as curly hair, freckles, photic sneeze reflex (the tendency to sneeze when entering bright light), and asparagus anosmia (the inability to smell asparagus metabolites in urine). 23andMe analyzed the data of 10,000 of their “whole-genome scan” customers who also filled out a survey about their health and physical traits, or phenotypes. Some are heralding this as a new paradigm for genetic research where scientists look for patterns in huge datasets rather than forming small, tailored studies to test specific, narrow hypotheses.

Indeed, these large genetic datasets have the potential to reveal powerful knowledge about public and personal health. Yet there is serious concern about how private companies will generate revenue from this data and whether the government will seek access to this data in the case of public health, public safety, or national security emergencies. Although the Genetic Information Non-Discrimination Act of 2008 prohibits employers and health insurance companies from accessing an individual’s genetic information, there are no uniform rules about other third parties acquiring genetic data from private companies like 23andMe. Long-term care, disability, and life insurance companies can still discriminate based on genetic information since they are not covered by the new law.

The editors of PLoS Genetics took these concerns seriously. They held up publication of the paper for six months to debate ethical concerns such as institutional review, participant consent, and data access. Ultimately, the journal’s editors were satisfied once they were able to ascertain that no participants had been coerced and all were aware that their samples would be used for research. The editors, Greg Gibson and Gregory P. Copenhaver, authored an accompanying editorial explaining this rationale.

Concern was originally raised when PLoS found that the manuscript was submitted without any institutional review board documents, which are customary with what is known as “human subjects research.” 23andMe later submitted a report from an accredited, independent review company, which affirmed that the 23andMe study did not fall into the category of “human subjects research.” This was based on two criteria: that the researchers did not have any direct contact with the study participants, and there was no way for the researchers to associate the data with individual participants since the data seen by the researchers was made anonymous.

PLoS also scrutinized the consent forms that participants were given when they signed up to be a part of 23andMe’s research efforts. 23andMe supplies a lengthy document detailing the potential risks of “unanticipated self-knowledge,” and the curious disclaimer that their services are “not designed to diagnose disease or intended to provide medical advice.” PLoS determined that if 23andMe had an independent review board review their document beforehand, there might have been significant changes to it, but accepted it because the document nevertheless met minimal legal requirements and any post hoc changes would require re-consent of thousands of participants.

PLoS noted, however, that for future research when “a study does not meet the aforementioned [human subjects research] criteria but obtaining a consent form would still be desirable, there are no guidelines or policy with regard to how such a consent form should be developed and reviewed in an ethically responsible manner.” PloS editors Gibson and Copenhaver then called for a review of the institutional review board consent process for genome-wide associated studies. They anticipate that new standards will and should evolve for GWAS research since institutional review boards are not always staffed by experts in genetics who would be more aware of the full implications of this research for subjects.

For consumers of personal genomics tests, this means researchers will need to be more transparent about:

  • The conclusions reached from analyzing genetic data
  • The potential health benefits or risks to the consumer
  • The possibility of new cures or treatments that arise from analyzing one’s data
  • The protocols surrounding the transfer of data to third parties

This last point is especially important from a public policy perspective as it becomes easier for scientists to mathematically determine the presence of an individual’s genome in an otherwise anonymous dataset.

Even though 23andMe does have a policy of not turning over individual-level genetic information to third parties without follow-up consent from their customers, 23andMe could still give it up if “required by law.” One can only speculate as to the circumstances under which the government would demand access to 23andMe’s genetic data by law, but it could be for any number of national security, public safety, or public health reasons. These policies are still quite ambiguous since the federal government so far has yet to establish uniform rules for how the various state FBI offices can obtain and utilize genetic data from criminal suspects, as Science Progress detailed last year in an article by Natalie Ram.

Personal genomics holds great promise, and the federal government has already taken measured but important steps toward ensuring that genetic tests are reliable. The National Institutes of Health recently announced the formation of a voluntary genetic test registry where companies can post the scientific data supporting the claims made by their tests so that physicians and consumers can make comparisons. And the Food and Drug Administration announced that it will hold a meeting next month to discuss the regulation of lab-developed tests for genetics, over which the FDA has heretofore not exercised its jurisdiction. Both of these policies were recommended in a 2008 CAP report, “Genetic Non-Discrimination.”

Additionally, with the passing of health care reform and the stimulus act in 2009 the Department of Health and Human Services is making great strides in setting up a new infrastructure for health information technology. This holds great potential for making medical record-keeping much more efficient and also providing patients and care providers with the latest data on tests and treatments at the point of care.

Of course, there are enormous privacy issues, and the Health IT Policy and Standards Committees are currently working on this. In the Science Progress report, “Paving the Way for Personalized Medicine,” we highlight the appropriate roles of these various government bodies for ushering in an era of personalized medicine, where physicians have the best science available at the point of care to get the right treatment to the right patient at the right time. In the report, we also recommend that a high-level office in HHS undertake an initiative to coordinate personalized medicine-related activities across the agency. The Genomics and Personalized Medicine Act of 2010, HR 5440, which was introduced last month by Rep. Patrick Kennedy (D-RI), actually goes so far as to create an entirely new Office of Personalized Healthcare.

All of these public policy initiatives will touch upon genetic testing as it relates Information privacy, scientific validity, and medical utility—all cornerstones of this new personalized medicine approach to health care. 23andMe’s article is only the beginning of a new era of subject recruitment, data collection, and research output. The federal government will definitely need to keep up with the private sector.

Michael Rugnetta is a Research Assistant to Jonathan Moreno for the Center for American Progress’s Progressive Bioethics Initiative and Science Progress.

Comments on this article

By clicking and submitting a comment I acknowledge the Science Progress Privacy Policy and agree to the Science Progress Terms of Use. I understand that my comments are also being governed by Facebook's Terms of Use and Privacy Policy.