For all the science fiction film aficionados who fondly remember the 1997 genetic thriller GATTACA, the future is now! Or at least it was until November 22, 2013 when the US Food and Drug Administration (FDA) ordered Google-funded biotechology company 23andMe to halt promotion of a suite of “personal genomics” services.
These products were aimed at directly providing consumers with affordable genome sequencing and annotation to ascertain a variety of health risks and drug responsiveness. Indeed, 23andMe touted their services as providing a “first step” in disease prevention and mitigation. In essence, this was a product at the forefront of the race to define “personalized medicine” as an individual’s right, or more plainly: a consumer product. The FDA’s spat with 23andMe represents part of a wider alignment as the tremendous promise of “omics” technology in bringing about an era of personalized medicine meets economic reality.
The FDA versus 23andMe
The context behind the FDA’s decision is spelled out in a warning letter instructing 23andMe to immediately cease marketing of health-related genetic testing. Among the principle grievances is the worry that these services could result in negative health consequences for customers in the case of false negative, or false positive results. Highlighted by the FDA is the potential for patients to undergo unnecessary preventative surgery in the event of falsely-determined, BRCA-related cancer susceptibility. Based on these fears, it was determined that 23andMe’s product needed rigorous testing and regulation in the same category as a medical device. However, this assessment does not imply that the FDA was unwilling to create room for direct-to-consumer genetic testing. On the contrary, the FDA expresses clear annoyance that 23andMe had abjectly failed to create a working dialogue with the regulatory agency. In a baffling strategy, the company decided to ignore FDA feedback obtained through at least 14 in-person discussions and “dozens” of other correspondence received between their first regulatory submission in July 2012 and May 2013. Instead, 23andMe opted to cease all communication and ignore requirements demanding additional clinical validation. Six months later the FDA delivered its letter.
Commentary regarding this decision has been decidedly mixed. It would be hard to argue with the FDA’s assessment of negligence on the part of 23andMe in providing test validation. However, there is debate as to whether the FDA’s concerns that consumer access to genetic information might cause harm are justified. Others raise deeper questions anticipating that the FDA missed the point altogether in assessing the regulatory requirements of direct-to-consumer personalized medicine by failing to recognize the implications behind a corporate-controlled database of user health information. The not so secret financial plan of 23andMe is not to make money by selling genetic tests at $99 each; rather, reminiscent of the electronic data hording of Google or Facebook, 23andMe aims to acquire the genetic information of as many individuals as possible and wield this information bank as financial capital. What a company could do with this kind of database might fall anywhere between magnificent and diabolical.
Risk and reward in direct-to-consumer genetic testing
A recent prospective piece published in The New England Journal of Medicine commends the FDA’s decision with the assertion that this direct-to-consumer service is not ready until it’s “diagnostic and prognostic capability… has been clinically validated.” This argument is centered upon three fears: mistakes in producing genetic data, the reaction of customers to learning their genetic risks, and poorly understood consent and information security. Unfortunately for 23andMe, the first issue has genuine validity. The company’s sequencing methods, while very good, still produce an error approximately 1 in every 10,000 bases. While a seemingly low frequency, this would produce a sizable amount of noise across a 6 billion base genome and can hide the functional variants associated with absolute genetic risk. Other researchers, however, point out that this danger doesn’t present the whole story. Writing in a Nature comment, Robert Green and Nita Farahany assert that the FDA’s worry about patient response to genetic data is misplaced. Citing a research survey of over 1,000 genetic test customers, they show that when confronted with genetic data, the majority of people either take no action, or make minor positive lifestyle changes. Further, only 1% of respondents altered prescription treatment without additional medical consultation. Of those making lifestyle changes, the majority started exercising or changed their diet. Such outcomes are usually positive and would likely outweigh any rare consequences or misleading results. Green and Farahany believe that the FDA, rather than promoting patient protection, may actually be providing hindrance. Although skeptical about 23andMe’s methods, they theorize that the availability of such testing could contribute to a healthier society, and in turn, customer interest will drive the development of a higher quality product.
For some, much more contention lies with the long-term implications and unknowns of a corporation amassing personal health information. A Scientific American article greatly extends the concerns raised in the NEJM’s perspective, contending that 23andMe’s subscriber consent does not provide adequate information security. Not only may customers not fully understand what they are signing, but the laws defending the privacy of this type of data is thus far poorly defined, and regulation largely self-imposed. Further, while 23andMe maintains that they will not sell personal information to third parties, they reserve the right to share accumulated database trends. The predictive potential of this type of information is increasingly being realized and may have many unforeseen outcomes for personal privacy. An interesting point raised by the Scientific American article is that genetic information by definition allows extrapolation between relatives. If enough of your family subscribes to 23andMe, the company probably already knows everything it could want about you.
Putting participatory in personalized: an economic necessity
Genome annotation is only a small step towards complete and practical application of individualized medicine. Ultimately, for this strategy to achieve its full potential, a broader spectrum of “omics” technologies (epigenomics, metabolomics, etc.) will need to be harnessed and combined to define one’s health. Eric Topol, a cardiologist and Director of the Scripps Translational Science Institute, predicts: “There will increasingly be the flow of one’s biologic data … directly to the individual.” Topol believes that “the digital convergence with biology will definitively anchor the individual as a source of salient data, the conduit of information flow, and a – if not the – principal driver of medicine in the future.” Imagine the health data that could be directly delivered through any number of personal biosensors; you might visit the doctor because your smartphone told you to. And it’s going to happen soon.
The strength of Topol’s forecast is borne not only by the rapid rise in technological capability, but also in economics. Direct-to-consumer data services are by definition participatory: producing in a single vehicle a patient, a product, a user, and consumer. This dynamic may further drive public interest towards these services and thereby help align the financial objectives of stakeholders in personalized medicine. For example, physicians wishing to avoid unnecessary and expensive testing, or pharmaceutical companies worried about reduced markets for broadly prescribed drugs, may currently inhibit the adherence to individualized approaches. These concerns could be resolved by the availability of large health databases, open for analysis at a population level and easily accessible at the doctor’s office. Databases, however, do not create themselves; their fruition will likely depend on the financial interest of corporations akin to 23andMe.
The future is soon
Thanks to the FDA, a GATTACA-like future may not yet be here. However, 23andMe’s product and the regulatory difficulties it has been having provide an interesting lesson: society is not yet prepared for the full commercialization of individualized medicine. Technological and economic forces are rapidly pushing this towards reality, but there is little consensus about what constitutes sufficient regulation and customer protection. Ultimately, consumer demand may be the true industry driver. Would you sequence your genome for $99? Just don’t forget to read the disclaimer.
George Annas and Sherman Elias. 23andMe and the FDA. NEJM. March 13 2014
Robert Green and Nita Farahany. Regulation: The FDA is overcautious on consumer genomics. January 2014
Topol, E. Individualized Medicine from Prewomb to Tomb. Cell. Feb 2014.
Sairamesh Jakka and Michael Rossbach. An economic perspective on personalized medicine. The HUGO Journal, 7:1 (2013).
Latest posts by Joshua Moreau (see all)
- BYOB: Brewing Your Own Beer - September 30, 2014
- Personalized Genomic Medicine - June 9, 2014
- Naked Mole Rats – Joining the Fight Against Cancer - March 6, 2014