Last year, a friend suggested that my wife and I try 23andMe. My wife quickly became infatuated with the idea of learning more about her genetic heritage. I, on the other hand — with a background in molecular biology and biochemistry — was a bit more skeptical. I was immediately concerned about the validity of these home tests and about the potential future consequences that might arise from having access to the results of these tests stored in a data bank.
For similar reasons, direct-to-consumer genetic testing (DTCGT) has been receiving a lot of attention over the past 10 years from medical providers, consumers, consumer advisory boards, regulatory entities, and ethicists. Thus far, there has been a sort of prima facie acceptance of DTCGT, based on the argument that individuals have a right to access their own genetic information1 — it is, in fact, theirs — isn’t it?
Some lawyers argue that when it comes to genetic testing, the “right to test” is far more complicated because of “interference” between individuals with known similar genetic makeup, like siblings and parents.1 For example, if my wife elected to undergo genetic testing, her decision could interfere with her sister’s decision to not test, because it is likely that my wife’s results would reveal something about her sister’s genome. DNA is not just some tangible object that belongs to a specific individual. Rather, it is an informative sample of data not just about the person from which the DNA was obtained, but also their relatives.
Even if we presume that there is a “right to test,” there are growing fears that patients will be misguided in their decisions based on poorly validated DTCGT. In 2010, the United States Government Accountability Office found that 10 out of 15 DTCGT providers “engaged in some form of fraudulent, deceptive or otherwise questionable marketing practices” that resulted in “misleading representations regarding the reliability of the tests and their potential contribution to treatment of diseases.”2
For example, a patient who tests negative for some of the known genetic markers of cancer risk might be misled into thinking their risk is extremely low and waive future screening tests. They may not realize that those genetic abnormalities being tested only account for a miniscule proportion of cancer cases. Further, someone with a positive genetic test for a disease with low penetrance may undergo a lifetime of unnecessary testing, substantial associated expenses, and immeasurable stress over a disease that is unlikely to occur.
Moreover, the impact of false positive or false negative tests — which are likely to be common given the variable allele frequencies, linkage disequilibrium across populations, and variability in the sensitivity and specificity of tests between ancestries — adds to the anxiety of health professionals and public health experts.3 In fact, for many of the conditions being tested, the positive predictive value of the tests is markedly limited by incomplete penetrance and gender modifications.4 For example, 23andMe only tests for a limited set of genetic variants associated with the possible risk for 10 very rare conditions developing, many of which cannot be prevented or treated.4 Therefore, the results of these incomplete and often limited genetic tests pose serious challenges for health professionals.
In 2013, for these reasons, as well as the lack of validity, the US Food and Drug Administration (FDA) sanctioned 23andMe from advertising that its DTCGT could assess genetic health risks.4 Despite these concerns, in early 2017, the FDA approved 23andMe’s DTCGT. Further, they exempted future genetic health risk tests from premarket review, allowing future companies to avoid regulatory review.4 The FDA explained their decision by arguing that, “Results obtained from the tests should not be used for diagnosis or to inform treatment decisions.” However, as Wendy K. Chung, MD, PhD, of Columbia University Medical Center and author of a study examining the “limited clinical utility” of DTCGT,4 pointed out, “To allow DTCGT and not expect persons to use the information to inform medical decisions is disingenuous and irresponsible.”4 The decision by the FDA to allow such passive acceptance of DTCGT may mark an overreaching change in the new administration’s philosophy regarding consumer protections.
Recognizing the pitfalls above, the European Parliament attempted to regulate genetic testing by requiring prescriptions from medical providers and blocking companies from advertising directly to the public.2 Ultimately, this provision partially failed; the advertising of genetic tests became regulated under the same laws that govern in vitro diagnostics, which in Europe prohibits misleading statements when marketing genetic tests.2
Despite challenges faced by both the FDA and the European Parliament in regulating DTCGT, the ethical concerns persist far beyond what is being addressed. Issues continue to arise about what to do with the information obtained, especially when the penetrance is less than 100%, and when there may not be any real interventions. There are also concerns that the information obtained through genetic testing may be collected and sold to insurance companies, which could lead to denial of coverage because of a genetic anomaly that may not even result in disease. This sort of data mining already occurs with prescription histories without patient consent.5,6 As genomic information becomes more widely available in the data banks of profit-driven companies, we ought to expect that these data warehouses will stand to profit by using these data without consumer approvals.
Challenges aside, genetic testing has the potential to play a valuable role in the future of our health care. Current DTCGT are limited, and physicians need to take an active role in educating patients and consumers alike about those limits. Dr Chung concludes her paper by arguing that as professionals, physicians have a duty to advocate for “medically useful and scientifically sound”4 genetic testing.
Meanwhile, my wife and I? We opted to test solely for the entertainment value that linage testing provided. It was as enjoyable and useful as visiting a Cassadaga psychic, with one exception: the markedly higher price tag.
- Bonython WE, Arnold BB. Direct to consumer genetic testing and the libertarian right to test [published online August 20, 2017]. J Med Ethics. doi:10.1136/medethics-2016-103778
- Kalokairinou L, Borry P, Howard HC. Regulating the advertising of genetic tests in Europe: a balancing act. J Med Genet. 2017;54(10):651-656.
- Lu M, Lewis CM, Traylor M. Pharmacogenetic testing through the direct-to-consumer genetic testing company 23andMe. BMC Med Genomics. 2017;10(1):47.
- Wynn K, Chung WK. 23andMe paves the way for direct-to-consumer genetic health risk tests of limited clinical utility. Ann Intern Med. 2017;167(2):125-126.
- Nakashima E. Prescription Data Used to Assess Consumers. Washington Post. http://www.washingtonpost.com/wp-dyn/content/article/2008/08/03/AR2008080302077.html. Published August 4, 2008. Accessed February 28, 2018.
- Lazarus D. Your Prescription History is Their Business. Los Angeles Times. http://articles.latimes.com/2013/oct/21/business/la-fi-lazarus-20131022. Published October 21, 2013. Accessed February 28, 2018