23andMe gene-testing kit

The Food and Drug Administration has ordered Silicon Valley gene-testing service 23andMe to halt all sales and marketing of its gene testing kits. 23andMe led the way in the personal-genomics revolution after it was declared Time Magazine’s 2008 invention of the year, but regulators have concerns that inaccurate results from the direct-to-consumer tests, or results that are not adequately understood by patients, could lead to dangerous decisions. Forum discusses the FDA’s order and the future of personal gene tests.

Hank Greely, director of the Center for Law and the Biosciences and law professor at Stanford University
Christina Farr, health reporter for Venture Beat
Nita Farahany, professor of genome sciences and policy and law professor at Duke University; also a member of the Presidential Commission for the Study of Bioethical Issues
Courtney Lias, director of chemistry and toxicology devices at the Center for Devices and Radiological Health at the Food and Drug Administration

  • What does it mean for us who’ve sent their samples in this past month?
    Will we get the genealogy/ancestry portion of results?
    23andme aren’t the only service.

  • I’ll be Frank

    It seems all things that come into contact with the Google are transformed into a sham. It’s the opposite of the Midas touch.

  • estratt

    Could this discussion please clarify the testing process- 23andme is an information services provider. They contract with a laboratory provider- I believe Labcorps- to execute the sequencing on a chip, which is also purchased from an external provider. The services of 23andme are selecting which SNP regions should be included on their customized chips and publishing data about the statistics related to the SNP status.

  • AJ

    Two-part question: what, if any, are the accuracy regulations or requirements for genetic testing companies? How accurate must they be to meet the FDA standard? And, If these regulations exist for other companies, who developed these, and what were the ideas and scientific evidence supporting the levels of accuracy? Thank you.

  • Sanfordia113

    I object in the most vehement terms to this elitist paternalism on the part of the FDA bureaucrats that patients are too unsophisticated to understand information about their own bodies. This decision is most unfortunate and a complete and total loss to patients and consumers. I, like millions of Americans, have a college degree in the biological sciences and am well-equipped to digest all 23andMe results. My question to the FDA is this: Would they make the same decision to refuse access to information, if 23andMe merely provided access to a patient’s individual genomic code? This information is the absolute right of a person to obtain this information.

    • I’ll be Frank

      Why do you not mention that the FDA is corporate-occupied? It matters since any of these FDA officials’ companies may have reason to compete with 23andme.

      David W. Beier . . .former head of Government Affairs
      for Genentech, Inc. . . . chief domestic policy advisor to Al Gore when
      he was Vice President.

      Linda J. Fisher . . .former Assistant Administrator of
      the United States Environmental Protection Agency’s Office of Pollution
      Prevention, Pesticides, and Toxic Substances…now Vice President of Government
      and Public Affairs for Monsanto Corporation.

      Michael A. Friedman, M.D. . . former acting commissioner
      of the United States Food and Drug Administration (FDA) Department of Health
      and Human Services . . .now senior vice-president for clinical affairs
      at G. D. Searle & Co., a pharmaceutical division of Monsanto Corporation.

      Marcia Hale . . . former assistant to the President of the United States and director for intergovernmental affairs . . .now Director of International Government Affairs for Monsanto Corporation.

      Michael (Mickey) Kantor. . . former Secretary of the United States Department of Commerce and former Trade Representative of the United States . . . now member of the board of directors of Monsanto Corporation.

      Josh King . . . former director of production for White House events. . . now director of global communication in the Washington, D.C. office of Monsanto Corporation.

      • mem_somerville

        Of course–personal plant genomics is definitely a threat. Or something.

  • Robert Thomas

    “Scientific proof” is a stupid oxymoron. Journalists interested in useful discussion of science in education, government, inquiry or any other arena should understand that this is a ridiculous, meaningless phrase.

    Like most journalists, Mr. Iverson apparently doesn’t understand the critical difference between “clear and convincing evidence” (or similarly qualified evidence), which is an object of scientific inquiry, on the one hand, and “proof”, which is the object of mathematical tautology (or of concern in other qualified contexts, such as in courts of law), on the other. The distinction is NOT merely “semantic”. This misunderstanding leads to much confusion, especially in public discussion.

    The fact that lazy working scientists sometimes indulge in this stupid usage is no excuse.

    I have no idea whether 23andMe has or has not failed to meet the FDA’s requirements.

    • I’ll be Frank

      You’re right, there’s no scientific proof of gravity.

      • Robert Thomas

        There’s no scientific “proof” of anything.

        • I’ll be Frank

          An aside: There’s no religious proof of anything either.

          • Robert Thomas

            Precisely so. That’s what separates science from some kinds of not-science. The objects of scientific inquiry are the evidence of our imperfect observations and the abstract models of the world we create using these observations, using our imperfect reason. Scientific conclusions are contingent. The objects of religious faith include Revealed Truth, which has no burden of evidence and is not contingent.

            I should say, this distinction contributes to the enormous power and value of science and depending on one’s point of view, to the value of faith.

          • I’ll be Frank

            And yet the evidence of delusion and wishful thinking are clear in religion, and encouraged, whereas scientists are expected to question one another’s methods and conclusions.

          • Robert Thomas

            IBF, isn’t this a good deal? As a promotor of science, I’m delighted with this expectation.

            The reason for this subject being a bug-bear of mine (and it IS) is not that most people didn’t understand (more or less) what Mr. Iverson meant (assuming the “scientific proof” phrase didn’t originate in documents from thee FDA!) by it in his preamble to this segment – I’m sure most people understood him to mean something like “clear and convincing evidence” – it is the one you immediately saw: that this casual usage invites confusion and convolution between science and not-science, mostly to the detriment of meaningful and useful discussion of the results of real scientific inquiry and to the progress of education in the sciences.

            One of not-sciences arguments is that its precepts are categorically similar to those found in conventional science texts, which after all are “only theories” and so forth. Why, they often ask, are these not-science ideas less well suited for associated discussion? The answer is that scientific inquiry always starts in (imperfect) observation of the natural world. For the kind of “not-science” I mean here, the competing assertions originate in (perfect) Revealed Truth. Whether such claims are worthy or not, their origin categorically precludes their consideration inside of science.

    • Guest

      FDA has requested that 23andme file a 510K and they did but FDA was not satisfied with the filing.

  • geraldfnord

    Conflicted: I hate clamping down on new information and businesses, but to have bad data were worse than to have none…and I have no idea of their accuracy—-we’ve never done universal A.I.D.S. testing for the simple reason that false positives from the vast majority that don’t host the virus would overwhelm the true. (Thanks, Rev. Bayes!)

  • Marina

    This sounds like the FDA doesn’t want consumers receiving information without mediation of a professional. Well I really want access to information about my health. I agree that people may need to talk to a doctor or genetic counselor to understand their results. But I can understand my results and I am glad I have this information. It may not be useful to me, it’s true, but that is for me to decide.

    • I’ll be Frank

      Without the epigenetic information it may not be useful to you at all.

      • Ehkzu

        That’s a very, very big generalization–with nothing attached to the statement to back it up.

      • Marina

        Maybe, maybe not. But let me decide. And when epigentic information is available, let me decide about that too!

  • Marina

    I want access to information about myself. I am aware that the results may not mean anything important. I can see my raw data and I want to be able to have access to it without going through my doctor. Privacy and cost both come to mind.

  • Ammy Woodbury

    I have to say that it’s disappointing to hear that folks take their results too seriously. 23&Me clearly marks things with the level of confidence. For me, it’s clear that the evidence is still rolling in on a lot of things and that frankly, 23&Me can be a resource for research to see a bigger sample set. If you took everyone that had a certain marker, then got their permission and tracked them over years, you could really see how the marker predicted (or did not predict) a health outcome.

    Meanwhile, I loved seeing the ancestry stuff. For me, it showed that this all-too-white-girl who got her first signs of sun-damage skin cancer at 39 comes by it naturally. It was all northern European. No ethnic diversity in my history unfortunately!

  • pm05

    Yikes! That woman who thinks its so great that individuals are “better” than physicians in their knowledge — scary! And, just nuts! If you don’t trust your physician, find another one, not a for-profit company!

  • ExBostonian

    I’m all for people/patients having access to as much information and knowledge they need or want on their own. BUT, don’t bring your false information and BS to your doctor’s office. It’s a waste of their time having to weed through all the crap on the internet that a patient has somehow glommed onto.

    • Sanfordia113

      Probably a majority of doctors in this country were a hair away from being charlatans themselves, totally devoid of scientific understandings of modern medicine. Democratization of medical information has forced them to learn things they never before bothered about (because of their state-imposed monopoly). We need more patient involvement, not less.

      • ExBostonian

        And where exactly do you get the data to draw that conclusion?

        • Sanfordia113

          A lifetime of having a disease that most doctors grossly misdiagnosed, yet simple review of scientific literature showed clear indications of what the diagnosis may have likely been. Why should a patient in Idaho need to see 10 doctors over 8 years before finding his way to a specialist at UCSF that makes a diagnosis that a patient could have come upon if scientific research articles weren’t hidden behind a protective paywall (accessible only to “medical experts”).

          • ExBostonian

            Well, I’m sorry to hear about your horrible experiences, but I can’t speak for the doctors you have seen. I don’t think it’s fair however to generalize your experiences to the entire medical profession.
            With due respect, you are reporting anecdotal evidence, not data with respect to the medical profession.

  • pm05

    Again, YIKES. Nita really is full of herself ! So, she is the brightest person in the “world” and boy she is going to let all those dumb doctors know how inferior they are and absolutely how great she is!!!! Please get a reasonable person on to discuss this!

  • Dianna DiSabatino

    When trying to get pregnant i got very different opinions from doctors and clinics. One well-known fertility clinic said that I could not sustain a pregnancy. Short story – went to another clinic and had twins right away. Blind trust in medical specialists’ advice is not better than getting real facts about one’s body, blood panels, etc and doing one’s own research. Websites also vary – look at diet advice – it is up to the consumer and individual to weed through the advice. But more raw data helps if used properly.

  • Vincent

    Funding agencies as well as regulatory agencies have been failing us by blocking innovation and only authorizing studies that fit their paradigm of interest. I find it amazing that they are now trying to block information such as the one delivered by 23AndMe. It is up to the people to decide what they want to do with this information and curtailing 23AndMe will just push people to look on the internet gathering confusing and contradicting information. Funding agencies are doing nothing regarding preventive medicine, so let the people take this into their own hands. By knowing better their own risk, people will just be more careful and in effect, we will get a healthier society.

  • deb

    Interesting assumptions about who is elite or able to understand sophisticated test results. Is there not a bias about the ability of MDs to communicate with patients? How many times have you had a conversation with a physician and then walked away thinking you didn’t understand a thing s/he said?

    • Sanfordia113

      Or even worse, when you walk away feeling that the MD doesn’t understand a thing you said.

  • Sbp

    I think humans have a real problem handling probability and risk. So 23andme is sort of a perfect storm not of misinformation but risk assessment. Not only do most people not understand genetic predisposition and risk, not only do doctors have no idea, but as a chemical biologist with UCSF I can tell you that the scientists publishing the papers 24andme draws from don’t quite get it.

    • Ehkzu

      There’s a fascinating book about this: “Inevitable illusions: how mistakes of reason rule our minds” by a cognitive psychologist from MIT. His thesis is that we evolved to misinterpret risk because it’s better for the gene pool even if it’s lousy for us as individuals. One example: going into battle against bad odds, yet still thinking you’re going to survive no matter what.

      • Sbp

        Good suggestion I’ll check it out. I also think the context we find ourselves in present challenges that weren’t around for most of our brain’s evolution. Like car and plane crashes. And genetic testing.

  • Steve Heilig

    I posted this last week, warning about just the concerns the FDA has:

    23andMe’s CEO has said they wish to ‘revolutionize medicine’
    and that “people contact us every week saying we’ve saved their lives.”
    (how would they know that?)

    “First, do no harm” is the primary ethical maxim of medicine – this testing has no real upside yet and plenty of risk of harm. And of course it’s not free….

    • Sanfordia113

      On the contrary, it has no downside, but plenty of upside.

  • Kai

    Absolutely ridiculous, like the patient could have her breasts and ovaries removed because 23andme shows some risk, the risks are with doctors performing these procedures. . Maybe FDA should focus on more serious issues like faulty AEDs and hip prostheses. Maybe there is some competitor in the gene business who does exactly the same but do not provide the disease risk related genome findings even if they could.

  • Kelly Wheaton

    No doctor or genetic counselor can be an expert on all diseases or implications of carrying certain genetic alleles. Anyone can misinterpret any test.

    I am a happy 23andme customer. A well informed consumer is a threat to the establishment. If the FDA spent more time educating and less time scaring folks we’d all be better off. The information is valid….the implication is sometimes very clear and sometimes very gray. 23andme explains it very well if you read their information. If you read you can be reasonably well informed.

    Please recommend the book: The Language of Life: DNA and the revolution in Personalized Medicine by Francis S. Collins who is the Director of the National Institute of Health. I also want to note he was formerly the Director of the Human Genome Project.

  • sally

    I am listening to your show: 3 points that I think are extremely important:
    1. Even with the 23andMe results, no one can proceed without going to a doctor. You can’t get a mastectomy without going to a doctor. So, you can have a test, but you can’t proceed in medical procedures without an doctor’s perspective, approval, insurance, etc.

    2. The government regulating anything, is usually way behind what is already happening in society: ie. it took 10yrs for the FDA to figure out transfats should be banned.

    3. We all need to be our advocates, deciding to get 23andme test results, you’ve made a decision about what you want to know. It’s equal (risk) as in not taking care of yourself, and then going to the doctor, and having to decide should I now follow a diet, exercise, medication regime the doctor has given me if I want to save my body, my life.

  • Mrs. Eccentric

    A few years ago i had to spend hours over a couple of weeks convincing a close friend that she didn’t have breast cancer. She was concerned because a technician at Kaiser had seen a record that she had been sent a ‘you have a bad biopsy result’ form letter during an unrelated phone call.

    My friend had never had a breast biopsy. She hadn’t had a mammogram, nor even any possibly related blood test. But this slip up on an unrelated phone call to a health care provider had her convinced. This is a very smart, college educated woman. I think it is clear that most of us do not have the background in genetic science in order to usefully interpret these type of results. steph

  • Maya

    I did the 23andme thing. First of all, it’s a $99 test. I’m not expecting it to be comparable to the $5,000 genome sequencing. Second, 23andme goes out of its way to tell you these are NOT necessarily 100% accurate results, they are NOT a substitute for medical advice and treatment, and that there are plenty of other factors that determine whether you might be at risk for any particular illness. I’m not expecting ironclad results here. I looked at the 23andme test as a jumping-off point for my own research. I learned that I MAY be sensitive to warfarin – great! Something I can ask my doctor about. Genetic testing is going to be one basis for future medical care. I’m satisfied that this is just one step toward allowing individuals more control over their medical care.

  • Ehkzu

    By the FDA’s logic, physicians should be banned from diagnosing patients, since they misdiagnose patients often enough to kill tens of thousands of patients every year–vastly more than could be laid at the feet of 23andMe, worst case.

    All gene testing services should be required to state what their accuracy rate is for every kind of result. And should be required to do what 23andMe already does–require customers to “opt in” explicitly for any test results that might be highly consequential, such as propensity to get an incurable diseases like Huntington’s.

    America allows the dissemination of quack medicine far and wide under the guise of “homeopathic” medicine, Chinese herbal remedies and the like. In this case it sounds like the doctors’ guild not wanting us to access information that they want us to only be able to get from the well-paid mouths of doctors.

    • Sanfordia113

      Totally! Cholesterol tests should not require a doctor prescription – should simply be a matter of visiting the local Walgreens and sitting down at the Cholesterol Test machine that accepts dollar coins as payment.

  • Eugene

    23andme could have avoided this by documenting their workflow according to a very lax 18-point CAP checklist. See p. 42-43 of this doc:


    It’s actually not all that much work. The point is to confirm that they’ve performed their due diligence and won’t accidentally say that you have some terrible disease when you actually don’t (or vice versa). Information is useful, but only if it’s correct.

  • Guest

    We should start a petition against FDA decision. It is
    outrageous they are going against preventive medicine. The results are very
    accurate, the interpretative data is based on the pull of genes or database the
    comparison is done against. 23andme
    offered a glimpse into the future of genomics and you should ask the internationally-recognized
    experts to share their vision of how next-generation sequencing will transform
    human health and personalize medicine.

  • Guest

    I was wondering why my comment has been removed.

  • Steve Heilig
  • Guest

    Let’s all write letters to the FDA to express our concern regarding their practices:
    James L. Woods, WO66-5688
    Deputy Director
    Patient Safety and Product Quality
    Office of In vitro Diagnostics and Radiological Health
    10903 New Hampshire Avenue
    Silver Spring, MD 20993


Sponsored by

Become a KQED sponsor