23andMe and the FDA: Perspectives on Warning Letter

23andMeWhat’s the deal with this week’s FDA Warning Letter to 23andMe?

I’m a researcher who studies stem and cancer cell genomics and epigenomics. I’m also a human being who has battled prostate cancer and is curious about my genome.

With this background, I decided some time ago to take the plunge and experience what 23andMe has to offer in the way of personal genomics analysis. I thought of it as an experiment.

I contacted them with my interest and background.

We agreed to do an interview on my blog based on my experience with them. I got the tube, spit in it, and fairly soon thereafter got my results from 23andMe.

You can read my 3 pieces on HuffPost about my experiences and my interview with 23andMe leaders Dr. Joanna Mountain, Senior Director of Research, and PR Manager Catherine Afarian (here, here and here).

In my last piece on HuffPost I expressed some concerns about personal genomics services including cost, security of data, and the potential risks of patients making medical choices based on what they see in terms of “results” on personal genomics sides.

It seems to me that the FDA is most concerned in their letter with 23andMe customers possibly thinking they have solid information based upon which they can make major health decisions. I never viewed the info from 23andMe that way, but I can see how it could happen.

The FDA and 23andMe have had a relatively long history so I can’t say I’m shocked that the genomics company got dinged, but I am surprised at how the FDA so publicly slammed 23andMe. One can almost sense the apparent frustration of the FDA with 23andMe.

Part of my reaction to the letter’s severity no doubt comes from how the FDA seems, by contrast, to have practically infinite patience with dubious stem cell operations that are putting hundreds or thousands of patients at risk. FDA, why so little action on the stem cell front, but you body slam 23andMe with this tough letter? The FDA has basically said 23andMe cannot continue with its personal genome service (PGS) any longer until issues are resolved as indicated in the FDA warning letter to 23andMe CEO Ann Wojcicki:

Therefore, 23andMe must immediately discontinue marketing the PGS until such time as it receives FDA marketing authorization for the device.

As a person who had the 23andMe analysis and does genomics research, what’s my thinking on all of this?

I enjoyed my experience with 23andMe, but I can definitely see the risks associated with the false sense of some of their customers having seemingly powerful information, which may not really be so concrete as they are led to believe. The FDA uses a real world kind of example of potentially devastating consequences of a customer making decisions based on overinterpreting 23andMe results:

For instance, if the BRCA-related risk assessment for breast or ovarian cancer reports a false positive, it could lead a patient to undergo prophylactic surgery, chemoprevention, intensive screening, or other morbidity-inducing actions, while a false negative could result in a failure to recognize an actual risk that may exist.

These risks of patients taking actions based on what could be incomplete information are real and serious. 23andMe needs to do a lot more in this area.

Keep in  mind too, however, that 23andMe is not the only player out there and direct-to-consumer personal genomics is going to continue in many forms.

The personal genomics age is upon us. There is a great demand from people wanting to know more about their genomes. We’ve all got our blood types and we want to know our genotypes. We just have to understand that much like anatomy, genotype is not destiny when it comes to our health. In fact, most of the time genotype features that we each possess on their own mean very little.

I believe that the appropriate path to developing a framework for how a personal genomics company should conduct business is not clear cut. It is also not going to be based on some textbook type stepwise plan that is already established. It must be determined empirically based on actual real world experience.

The FDA letter is not the end, but just another part of the beginning of the personal genomics era. 23andMe needs to do far better and I’m betting it will. In fact, I would be very surprised if the 23andMe PGS was not back up and running in a new and improved form in 2014.

Expect a bumpy, but interesting path forward.

5 thoughts on “23andMe and the FDA: Perspectives on Warning Letter


  1. I submitted to 23andme and am disappointed that the FDA has nothing better to do than shut down a reputable company whose customers willingly send their samples to be analyzed. How about fast tracking some of those experimental drugs that might save lives?

    PS: I was very satisfied with the results of my test.


  2. I agree genetics are not destiny.I think some education on genetics is in order. I also wonder how 23 and Me got the letter/cease orders for a 99 dollar product when non compliant cell companies are charging many thousands of dollars for a more invasive procedure and they still operate. After all some of the plant ‘stem cell’ cream sells for about the same price as a 23 and me profile!

    I would consider having the testing done if it was available. I think it would be interesting but there are privacy concerns too in that companies can use this data to make assumptions about family genetic predispositions and I have not seen protections available for this form of profiling.

    Could it be the aggressive direct to consumer marketing along with inadequate validation of the tool in question. Apparently some report making significant health decisions based on their results without hearing input from their physicians. There is a great video on the issues involved here http://www.ithinkwell.org/23-and-me-get-fda-warning-letter-who-is-next/

    One point of consideration is the price of genetic testing outside 23 and Me. I know people who have been recommended for the breast cancer testing but at several thousand dollars per test they just can’t afford it. I also wonder what implications this letter will have for other crowd sourced medical devices or procedures.


  3. I looked at the FDA warning letter and I was underwhelmed. They are making mountains out of molehills…
    The example you give about patients getting their breasts removed (or whatever) based on a 23andMe BRCA-related risk assessment is totally bogus.

    First, I recall (a while ago) doing a bit of web research and looked at the sort of output 23andMe provides patients. It was very clear that 23andMe does not promote a view that genes are fate — they work in terms of probabilities/risk. So does ALL of the medical community for ALL procedures and so do a great many scientific researchers! And so does the FDA — although they are guilty not properly stating the probabilities and just resorting to a yes/no switch (just think about the statins debacle). The notion that patients are a bunch of unsophisticated fools who can’t understand probabilities is insulting.

    Second, even if some patient gets an assessment that they have some risk, no doctor is going to just start loping off bits of anatomy on the say-so of 23andMe. Doctors would do their own assessment, talk to their patient — and work from there.

    All that 23andMe does is give a patient information about themselves. Perfect information? Of course not, that is made very clear by 23andMe — and most modern people would know this without being told, anyway.

    There is a personal freedom issue. People have a right to know their own bodies!

    As for the information security issue. Well, that’s an issue with just about every aspect of our lives. As a Canadian, I have learned from Andrew Snowden and Chelsea Manning that the USA government has the means and presumes it has the authority to spy upon me in any way it feels like. It can even coerce internet companies to divulge my private information. If the FDA is going to get into regulating the security of information, then I would be quite happy to point them towards more pressing matters and more fertile ground within their own governmental backyard!

    When the FDA takes action on the basis of flimbsy reasoning, I suspect an ulterior motive.


  4. I wanted to clarify that the security, identity risk was not specific to 23 and Me. That would be true across any group that tests DNA. It is a loophole in our general health privacy laws.

    I think the BRCA risk given by Paul is listed on another site and their argument was that there are several variation and that 23 and Me does not evaluate them all. They report someone was harmed by responding to the one test, they give no names. I do not know if this is true but I do know that in an assessment I saw 23 and Me stated the limitations of their BRCA testing (that it did not cover all variants) and urged the person to get more extensive testing. Their medical providers reassured the person it was not a high risk given other factors, put it on their chart and adopted the watchful waiting approach.

  5. Pingback: Knoepfler Lab Stem Cell Blog – 23andMe and the FDA: Perspectives on Warning Letter | Stu's Stem Cell Blog

Comments are closed.