The FDA asked 23andMe to comply with legislative standards set forth by the Federal Food, Drug, and Cosmetic Act of 1938, specifically by "validat[ing] the PGS [Personal Genome Service] for its intended uses, which have expanded from the uses that the firm identified in its submissions." One can only assume that the "the uses the firm identified in its submissions" are the genealogical ones. 23andMe was years behind in providing the validation of their medical recommendations and interpretations, and to avoid further action by the FDA, has agreed to suspend any health interpretations of the data until they can verify the accuracy of their results.
To the public, this comes across as the FDA stepping in on the public's behalf to prevent them from making any permanent, important health-related decisions (e.g. mastectomies) on data that aren't 100% sound and verified. That would seem to be the logical conclusion. Some arguments about FDA's "unholy union" with Big Pharma etc. are also coming to the surface, but that's a road that's a bit too tin hatty for my liking.
When you send your spit kit off to 23andMe, the data you get back from them doesn't just stay with you. They—and your survey results—are also part of a vast genomic library. Your individual results, as linked to you, John Doe, can't be sold, but 23andMe
reserves the right to sell that information in the aggregate, or use it to market other events and products to individual customers.The issue then becomes not of protecting consumers from their own idiocy, but of protecting their information from being used in ways they couldn't foresee, know about, or agree to. It's not the accuracy the FDA is concerned with (multiple Internet searches do not indicate any rash of faulty results), nor is it the consumer's private use; it's 23andMe's corporate use of said data. The question is about whether or not that kind of data in the hands of large corporations (health insurance, life insurance, drug companies, etc.) is a desirable or ethical thing.
That said, I found the medical results from 23andMe useful. In particular, I was surprised to find out that I had one copy of the APOE-4 gene. I understand science enough to know that this doesn't mean I'm definitely going to get Alzheimer's (even with one copy of the gene my chances are 14%, compared to 7% without), but if I had just looked at my family history, I would have never known I was at an elevated risk for the condition—not the same way I already knew from family history that I was at risk for hypertension and high cholesterol.
Do I value those, and other, results enough to be okay with my anonymized survey and spit kit results being sold to corporations? Yes, actually. At least, the deed is done, so there is no good regretting it. And in a way, I'm excited that my DNA is contributing, at least in part, to medical research. Is corporate buying and selling of this kind of information something to be concerned about? I don't know. This is a relatively new field and we have yet to see just how the nuances will play out.
Originally I was prepared to come to the conclusion that I simply couldn't understand the FDA's concern with 23andMe, since the site is full of disclaimers about how genetics only indicate risks and that you should consult a medical professional and so forth (your personalized risk factors for Alzheimer's and breast cancer are even hidden behind little "consent" walls where you have to read through an explanation of these points before you can even see your results). The information they provide to their customers is reliable enough, after all, and can be immensely beneficial. As one comment on the 23andMe blog suggested: if the FDA is so concerned about people making terrible medical decisions, why does WebMD still exist?
But after some reading on it—and the implications of all that genetic information in 23andMe's possession—I can begin to understand the concern. The question then is: should the FDA be in the business of regulating market practices? (If the kits and the PGS work as intended and provide accurate results, that's really all the FDA should be concerned about.) Would this be better left to something like the FTC? Is fear of buying and selling genomic information justified, or is it destined to become part of "business as usual" in the near future?
May you live in interesting times, indeed.