FDA orders 23andMe to shut down home genome test

There are so many reasons to resist the FDA on this action — really, as many reasons as there are individual 23andme users. Some of us want to seek out distant relatives and clues about national origins, or satisfy curiosity about patterns of disease in our family lines. For adoptive families, home genome testing can be hugely valuable in cases where one knows little about the medical history of an adoptee’s birthfamily. It’s our body, and our right to inform ourselves about it — or so we thought.

Some are blaming the company for rolling out the popular service in the absence of a clear regulatory go-ahead, and, in recent months, ignoring repeated signals of the FDA’s wish that it submit to comprehensive regulation that would greatly drive up the cost of its service. But other commentators have suggested that the firm has some pretty decent legal arguments that its service is not subject to regulation as a diagnostic test or “medical device” (genetic predispositions are not diagnoses). As an information-based service, it might even enjoy protection under the First Amendment. Admittedly, the company waved a red flag in front of regulators when it launched a marketing campaign that stressed the possible health benefits of knowing one’s genetic predispositions. But as Timothy Lee argues at the Washington Post:

Having more information about your health status is never dangerous by itself. It only becomes dangerous if patients use it to make dangerous medical decisions. But most dangerous medical decisions can’t be made unilaterally; they generally require the assistance of licensed medical professionals who will do their own assessment of the situation before performing procedures that could harm patients.

The FDA very likely has decent legal grounds to forbear from a crackdown should it choose to. But the key takeaway sentence from Matthew Herper’s piece in Forbes criticizing the company is: “This is not the way to deal with a powerful government regulator.” Disrespectful, anti-authority attitudes from someone an agency intends to regulate? Ask former Buckyballs CEO Craig Zucker where that gets you.

What can users, potential users, and well-wishers do?

* “First, download your 23andMe raw results now if you have them,” warns Razib Khan at Gene Expression.

* If you like signing Change.org petitions, there’s one here asking the FDA to back off.

* In a separate piece for Slate, Khan suggests where the situation might head before long: services like this can move offshore. All the relevant information consumers want from them can be delivered via the web. In the mean time a highly innovative and valuable enterprise will have been pushed out of the U.S. to some freer part of the globe, but maybe we need to get used to that happening.

And then? It may take a while before our government works up the nerve to ban mailing a saliva sample to a foreign address. Based on existing trends, I’d guess the more likely intervention, circa 2018 or so, would be for the Treasury to direct credit card companies not to process payments from U.S. residents to genome kit providers. Would we have the spirit to resist then? And if then, why not now? More: Alex Tabarrok, Slate Star Codex (by analogy, “banning people from weighing themselves without a prescription is neither clinically nor ethically sound,” although weight awareness sometimes leads patients into unwise health decisions), Nita Farahany, Brad Warbiany, earlier 2011. Alex Tabarrok’s post is especially worth reading, an excerpt:

…Indeed, genetic tests are already regulated. To be precise, the labs that perform genetic tests are regulated by the Clinical Laboratory Improvement Amendments (CLIA) as overseen by the CMS (here is an excellent primer). The CLIA requires all labs, including the labs used by 23andMe, to be inspected for quality control, record keeping and the qualifications of their personnel. The goal is to ensure that the tests are accurate, reliable, timely, confidential and not risky to patients. …

…the FDA wants to judge not the analytic validity of the tests, whether the tests accurately read the genetic code as the firms promise (already regulated under the CLIA) but the clinical validity, whether particular identified alleles are causal for conditions or disease. The latter requirement is the death-knell for the products because of the expense and time it takes to prove specific genes are causal for diseases….

The FDA also has the relationship between testing and clinical validity ass-backward. The FDA wants to say no to testing until clinical validity is established but we are never going to discover clinical validity until we have mass testing.

More: Richard Epstein/Point of Law, BoingBoing, more from Ron Bailey.


  • They’d have to ban licking the stamp, then.

  • Watch for companies moving offshore to start accepting payments in Bitcoin. No credit card company or bank needed.

  • As for the suggestion in the last sentence: How? (Credit card companies and PayPal are already blocking payments to a large number of existing, legal businesses. Is there any workaround, and can it be used without violating the money laundering laws?)

  • Also watch for medical tourism; i.e. while you visit Machu Picchu, Ayers Rock or the Olduvai Gorge, you can also spit in a cup, pay a few bucks, and get your results via email.

    Ditto for paying for it wrapped up in/as something else; e.g.free DNA analysis with your magazine subscription, we email the results to you (Not valid for US residents, ha ha, there we said it, we’re covered.)

  • […] Some thoughtful posts on the FDA and 23andMe: Derek Lowe. Popehat. Overlawyered. […]

  • I think the FDA has been awfully forgiving of 23andMe given their insolence and aggressive claims that the information can be used to affect health/treatment/diagnosis decisions.

    If we were talking about an acupuncture clinic claiming that it could determine a patient’s potential to develop 260 conditions ranging from preeclampsia to ADHD, lots of folks would be (rightly) going bananas. Yet, from the FDA’s perspective, the 23andMe and the hypothetical acupuncture clinic are the same, because, for the majority of those 260 conditions claimed by 23andMe, 23andMe does not have sufficient scientific evidence to satisfy FDA requirements applicable to drugs and medical devices.

    Alex Tabarrok seems to have missed a crucial distinction: the FDA isn’t complaining about the service per se, they’re complaining about the wildly outsized marketing claims. Requiring clinical validity isn’t a “death-knell” for these services, it’s a death-knell for these services making unsupportable claims, as it should be. The service can operate, but it can’t make scientifically unsupportable claims about the results. Spend some time on their site, and try reading about it as someone with modest knowledge of genetics — do you get any sense of just how weak some of these claimed correlations are? Or does it look like 23andMe can identify the (non-existent) ADHD gene? If they spent a little more time being scientifically honest and a little less time marketing, they wouldn’t have a problem.

    I’m a big fan of services like 23andMe, and I think consumers should have ample access to these types of services, but the FDA rightly draws lines preventing companies from claiming health benefits that aren’t really proven in the literature. Just because 23andMe does a lot of things right and because they have prominent backing doesn’t mean they’re exempt from those rules.

  • Max Kennerly 11.29.13 at 6:11 pm:

    … but the FDA rightly draws lines preventing companies from claiming health benefits that aren’t really proven in the literature.

    Telling a client “by my analysis you have a 5% lifetime probability of developing galloping halitosis” is not a “health benefit”. It is an opinion.

    Anybody with a robe and a crystal ball can legally tell you his opinion about your life for a fee, without interference by the FDA. 23andMe is doing nothing more unlawful than that.

    The FDA is trying its nanny best to expand its remit to regulate mere opinion. If they succeed here, where will they stop? Ban discussion of genes and health except by the properly licensed?

  • I believe the FDA’s letter to 23andme did not merely command it to cease making certain claims about its service, but commanded it to stop offering the service for sale, period. Had it been only the former, we would no doubt be having a different conversation.

  • I don’t think this should be thought of as a free speech issue. My understanding is that you take the test and are given some level of interpretation of your genes. For example, “You have gene X which has a 5% chance of getting some disease”.

    The problem is the literature around these genes and causality with certain diseases is murky. From reading geneticists comments elsewhere, it sounds like the best science today has a relatively incomplete picture on how all these genetics fit together with regards to the causation or predisposition of diseases. In otherwords, making the claim that such a gene predisposes you for some medical condition may be inaccurate or based on an evolving understanding that may or will change.

    I believe we have a right to know our genetics. If 23andme were somehow giving me the raw data, or sticking with more understood interpretations (ie “you are of sub-Saharan African descent”), then I would disagree with the FDA. Because 23andme provides interpretation beyond what has medically been established, I agree with the FDA.

    The FDA isn’t withholding from us the right to know ourselves. They withholding the right for 23andme to tell me what it means unless it can prove it knows what it means.

  • […] Class action lawyers hop on 23andMe with a lawsuit piggybacking on the FDA’s enforcement action [Nita Farahany; Ron Bailey, Reason]. Earlier here. […]

  • I agree that the results of an individuals DNA test are in no way medical, and the only harm that an individual could come to as a result, would be having medical procedures performed, which a licensed Doctor would have to do. Making this yet another highly regulated thing, would be akin to making all vitamins accessible by prescription only.

  • […] patients from knowing about their own health” [David Rivkin and Andrew Grossman, USA Today, earlier] FDA-defying workaround lets you get your info even if 23andme’s muzzled [Ron […]