About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« The Freshness Index | Main | Thanksgiving: Never Trust An Organic Chemist Who Can't Cook »

November 27, 2013

23 And Me And the FDA

Email This Entry

Posted by Derek

As everyone will have heard the personal-genomics company 23 and Me was told by the FDA to immediately stop selling their product, a direct-to-consumer DNA sequence readout. Reaction to this has been all over the map. I'll pick a couple of the viewpoints to give you the idea.

From one direction, here's Matthew Herper's article, with the excellent title "23 And Stupid". Here's his intro, which makes his case well:

I’d like to be able to start here by railing against our medical system, which prevents patients from getting data about our own bodies because of a paternalistic idea that people can’t look at blood test results, no less genetic information, without a doctor being involved or the government approving the exact language of the test. I’d like to be able to argue that the Food and Drug Administration is wantonly standing in the way of entrepreneurism and innovation by cracking down on 23andMe, a company that is just trying to give patients the ability to know about their own DNA, to understand their own health risks, and to participate in science.

I wish that was the story I’m about to write, but it’s not, and it all really comes down to one fact in the FDA’s brutally scathing warning letter to 23andMe, the Google-backed personal genetics startup. It’s this quote from the letter by Ileana Elder, in the agency’s diagnostics division: “ FDA has not received any communication from 23andMe since May.”

So we can call that one the practical view: "It doesn't matter what you think about 23 and Me's product, and it doesn't matter what you think about the FDA. They're supposed to be working with the FDA, they knew it, but they haven't done squat about it, so what did you expect the agency to do, anyway?". From that, let's go to the idealistic view, from economist Alex Tabarrok at Marginal Revolution, who writes just the sort of article that Herper deliberately passes up the chance to:

Let me be clear, I am not offended by all regulation of genetic tests. 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. I am not offended when the goal of regulation is to help consumers buy the product that they have contracted to buy.
What the FDA wants to do is categorically different. The FDA wants to regulate genetic tests as a high-risk medical device that cannot be sold until and unless the FDA permits it be sold.

Moreover, 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. Moreover, it means that firms like 23andMe will not be able to tell consumers about their own DNA but instead will only be allowed to offer a peek at the sections of code that the FDA has deemed it ok for consumers to see.

The thing is, I can see merits in both these views. And you know, they're not mutually exclusive, either, not as much as it looks like at first glance. I don't even think that the FDA itself thinks that they're so mutually exclusive, if you read their letter (emphasis added):

The Office of In Vitro Diagnostics and Radiological Health (OIR) has a long history of working with companies to help them come into compliance with the FD&C Act. Since July of 2009, we have been diligently working to help you comply with regulatory requirements regarding safety and effectiveness and obtain marketing authorization for your PGS device. FDA has spent significant time evaluating the intended uses of the PGS to determine whether certain uses might be appropriately classified into class II, thus requiring only 510(k) clearance or de novo classification and not PMA approval, and we have proposed modifications to the device’s labeling that could mitigate risks and render certain intended uses appropriate for de novo classification. Further, we provided ample detailed feedback to 23andMe regarding the types of data it needs to submit for the intended uses of the PGS. As part of our interactions with you, including more than 14 face-to-face and teleconference meetings, hundreds of email exchanges, and dozens of written communications, we provided you with specific feedback on study protocols and clinical and analytical validation requirements, discussed potential classifications and regulatory pathways (including reasonable submission timelines), provided statistical advice, and discussed potential risk mitigation strategies. As discussed above, FDA is concerned about the public health consequences of inaccurate results from the PGS device; the main purpose of compliance with FDA’s regulatory requirements is to ensure that the tests work.

As much as I might agree with Alex Tabarrok in principle, I think he's missing a key point here. The FDA is not telling everyone that they don't own their own DNA information, and that they can't see it unless the agency lets them. The agency is saying that 23 and Me can certainly make a business out of selling people their own DNA sequence information, but if they do so by explicitly claiming medical benefits or diagnostic uses, then their business will fall under the FDA's jurisdiction. From their letter, it appears that they have been telling the company this over and over for several years now, during which 23 and Me has, apparently, been dragging their feet and trying to have it both ways. As the FDA letter notes:

For example, your company’s website at (most recently viewed on November 6, 2013) markets the PGS for providing “health reports on 254 diseases and conditions,” including categories such as “carrier status,” “health risks,” and “drug response,” and specifically as a “first step in prevention” that enables users to “take steps toward mitigating serious diseases” such as diabetes, coronary heart disease, and breast cancer.

I'll add a bitter, cynical note: if only 23 and Me had been able to come up with some way to market their DNA test as a nutritional supplement, they'd be in the clear. Maybe some sort of sugar pill that you took before you spit in the little sample container? Then they could say "Not intended to treat, cure, or modify any disease" at the bottom of the page, in six-point microtype, and everything would have been fine, as if by magic. No one would have paid any attention to it, of course, because no one ever pays any attention to that language when they go out and buy all kinds of "supplements", and the FDA would have staggered backwards at the sight of Orrin Hatch's law, like Christopher Lee in a Hammer vampire film being hosed down with a face full of holy water.

Well, that might not have worked perfectly, but it would have worked better than what 23 and Me actually tried. They wouldn't have sold nearly as many DNA tests without talking about preventing disease and making medical decisions in their advertising, true, but those are the breaks. I think that if they'd stuck to some neutral language, rather than presenting Immediate Actionable Medical Decisions, they might well have stayed out of trouble.

Update: via Matt Herper's Twitter feed, here's an interesting take on the whole situation. 23 and Me has been hoping to get some real (and really profitable) insights into population genomics by accumulating such a large sample size. Have they? The way they're acting makes one think that nothing good has popped up yet. . .

Comments (45) + TrackBacks (0) | Category: Regulatory Affairs


1. SP on November 27, 2013 8:55 AM writes...

They did try to pull the nutritional supplement cop-out of "Medical purpose? What medical purpose?" by having this disclaimer on the website:
"The information on this page is intended for research and educational purposes only, and is not for diagnostic use."
The FDA called BS on that, as your last quote shows. So they knew they were walking a fine line and saying one thing to consumers while trying to claim they were doing no such thing, which in my view makes them just as slimy as nutritional supplement companies.

Permalink to Comment

2. Mo on November 27, 2013 9:13 AM writes...


The still haven't updated the page that the FDA is pissed about ( It still looks like they're trying to sell themselves as a diagnostic tool.

Permalink to Comment

3. johnnyboy on November 27, 2013 9:25 AM writes...

I think this speaks volumes about the arrogance and liberterian groupthink of these silicon valley upstarts, who think that because they are young and new and have billions of stupid money backing them, they can do whatever the hell they want, and the rules are for other people. I suppose we can expect Wired magazine to soon publish some article about how Wojcicki is a rebel genius, positing her somewhere between MLK and Gandhi, like Steve Jobs did in his own ads.

Permalink to Comment

4. Curious Wavefunction on November 27, 2013 9:45 AM writes...

I agree. 23andMe have some laudable goals but they seem to have been going about them the wrong way. Their aggressive marketing campaign (I got a letter from them a few weeks ago) seems disproportionate to the services they are offering. At the very least this case should start a national dialogue about a scenario which is going to become all too common in the age of cheap personalized genome testing. Interestingly they don't seem to have taken down the $99 test listed on their home page.

I do find it disconcerting, however, to find 23andMe lambasted while nutritional supplement makers continue to peddle their sordid wares with abandon.

Permalink to Comment

5. myma on November 27, 2013 9:56 AM writes...

It's a diagnostic. It has to follow the diagnostic rules.

Permalink to Comment

6. nekekami on November 27, 2013 9:56 AM writes...

I've been following the 23 And Me stories for a while, and with the details that are available, I can't find anything sympathy-worthy in their plight at all. All the problems are self-inflicted due to deliberate weaseling and deception

Permalink to Comment

7. Anonymous on November 27, 2013 10:13 AM writes...

Why don't the FDA bring the hammer down on Magic 8 Ball?

Permalink to Comment

8. Joe on November 27, 2013 10:31 AM writes...

The sliming of nutritional supplement companies goes on here unabated. If a consumer wants a vitamin tablet, what is your problem, guys? And that is the VAST majority of what is sold on the shelf - vitamins, minerals and some plant extracts that give someone a bit of energy. So is it some rotten companies in the industry? Some bad players? Everyone is a Kevin Trudeau, huh? And that doesn't happen in pharma? Face it, there are more indictments being handed out in the pharma industry than the supplement industry if you want to look in the mirror. The sanctimoniousness here sucks.

Permalink to Comment

9. Dolph on November 27, 2013 10:33 AM writes...

"It's a diagnostic. It has to follow the diagnostic rules. "

Mmmm, no???
It could be argued what 23andme is, but to say it's a diagnostic service is completely off. I would call it a very interesting toy in the state it is right now.
To shut it down is a strange move. With about the same intellectual level of argumentation the sale of mirrors or photocameras could be prohibited.

Permalink to Comment

10. pgwu on November 27, 2013 10:35 AM writes...

They can pretend that they did not receive the letter. The address in the letter is on Shoreline Way while the company's term of service address is on Shorebird. It took me more than a few moments to find the address from the company website and thankfully there is one.

Permalink to Comment

11. Hap on November 27, 2013 10:47 AM writes...

Why is sanctimonious to expect that if a company claims that what they sell does X, that they actually have evidence that what they sell does X? Drug companies actually have to surmount that barrier, and get pillaged if they don't (or, even if what they do isn't good enough for its risks). In contrast, supplements can pretty much claim that their supplements give you the sun and the moon without consequence, so long as their writers are skillful enough or their operators are mobile enough.

In addition, drugs (gasp!) actually have to contain what's on the label or insert, a constraint that apparently doesn't apply to supplementeers.

If you like buying stuff from people who pull everything from their anuses, I guess you know what you're going to be getting. Pretending that everyone is selling from their lower orifices and that it is therefore okay, though, is an argument that ought to have been removed from your argumentation sometime around ninth grade.

Permalink to Comment

12. Curryworks on November 27, 2013 10:48 AM writes...

More and more is the quote from our friend Brutus "But, as he was ambitious, I slew him"

Permalink to Comment

13. Anon on November 27, 2013 11:01 AM writes...

#12: "But, as he was a liar, I slew him"

Permalink to Comment

14. Anonymous on November 27, 2013 11:03 AM writes...

Maverick rule breaker meets anal rule enforcer?

Permalink to Comment

15. Anon Fielder on November 27, 2013 11:03 AM writes...

I am in agreement with Matthew Herper and believe physicians are the number 1 barrier to healthcare [in the United States] in the 21st century.
...the scary part is that they like it this way and are fighting feverishly to keep it this way through their lobbying efforts. They are even lobbying against their brother professions of nursing, optometry, chiropractics, podiatry, pharmacy, etc.

A readily apparent xample is that Optometrists are capable of performing LASIK eye surgery are schools are eager to teach it, however physicians don't like the idea that someone might intrude on their market share, so the individual performing the procedure must be an MD (ophthalmologist).
The same is true of how a patient will see a Physician's Assistant, however the Physician will sign off on the work at the end of the day and is the one who will be reimbursed by Medicare/Medicaid/Insurance.

Blood tests are even simpler in the sense that a phlebotomist takes the blood (in large healthcare organizations this is often located outside of the clinic), techs run the test, and the physician is merely comparing your values to the lab normals...and the physican doesn't even have to compare, it usually has a H or L for high/low values. Here the doc is outright a middleman, they did not physically touch the patient, they did not use any technical skill in processing the sample, and they did not use any intellectual skill to interpret the results. Just a signature to receive a portion of payment.

Not only is it next to impossible for patients to be intimately involved with their treatment (if they have a disease), but they can't even get involved with clinical maintenance of their health.

Not to put words in Herper's mouth, but I believe physicians are very much a barrier to healthcare.

I one wants a test by 23andMe, they now have to wait several weeks for an appointment (also the doing of the AMA and their restriction of physician supply), pay a copay, have the insurance pay to have someone take swabs (which will cause a slight increase in your insurance plan), wait several weeks, schedule another appointment, pay another copay, and THEN your doc can deliver the information. All of which was spelled out to them in the same way it is currently spelled, with the same disclaimers on the data.

Permalink to Comment

16. trrll on November 27, 2013 11:10 AM writes...

I'm glad I got my results before the hammer came down. But of course, I'd never make medical decisions based upon that report--I'd go to the original literature, and if anything concerned me, I'd replicate it

Permalink to Comment

17. Anon Fielder on November 27, 2013 11:15 AM writes...

I think replication here is a key and only using this data as an indicator, but not a matter of fact test.
The most obvious issue is that the customer is conducting this under a GCP practices. With that in mind, I doubt 23andME is being very GCP/GMP on their end. I would even go farther and say that after corresponding with the FDA about any such standards they realized their business model isn't as profitable when they have to be precise about their methods.

Permalink to Comment

18. David Borhani on November 27, 2013 11:59 AM writes...

@9: It IS a diagnostic, as the language on their web site (1; below) even today makes very clear.

23andme appear to be trying to hide behind a computer-software-like Terms of Service, on the one hand (2; below), and specific, medically relevant diagnostic claims on the other (1). The latter require, of course, FDA approval. In fact, the FDA's letter very clearly lays out (3) the significant risks to the consumer (patient) of false-positive or false-negative results. This has nothing to do with doctors or the medical establishment controlling the flow of genetic information. Rather, if you claim a diagnostic benefit, then your test better damn well stand up to data-driven scrutiny.

(1) Voluminous (interesting) information at If all this isn't an explicit statement of diagnostic utility, I don't know what is. (Perhaps the 6-point type in the Terms of Service makes it implicit, but the claims are clear enough to warrant FDA action.)

(2) "You acknowledge and agree that the Services are provided "AS-IS" and are based on the current state of the art of genetic research and technology in use by 23andMe at the time of the purchase or viewing. As research progresses and scientific knowledge and technology evolve, 23andMe is constantly innovating in order to provide the best possible experience for its users." (Emphasis added)

(3) "Some of the uses for which PGS is intended are particularly concerning, such as assessments for BRCA-related genetic risk and drug responses (e.g., warfarin sensitivity, clopidogrel response, and 5-fluorouracil toxicity) because of the potential health consequences that could result from false positive or false negative assessments for high-risk indications such as these. 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. Assessments for drug responses carry the risks that patients relying on such tests may begin to self-manage their treatments through dose changes or even abandon certain therapies depending on the outcome of the assessment. For example, false genotype results for your warfarin drug response test could have significant unreasonable risk of illness, injury, or death to the patient due to thrombosis or bleeding events that occur from treatment with a drug at a dose that does not provide the appropriately calibrated anticoagulant effect."

Permalink to Comment

19. Tuck on November 27, 2013 12:18 PM writes...

In 2007 Senator Obama co-sponsored legislation ("the Genomics and Personalized Medicine Act") that would have defined what the FDA could regulate, and how it should do so. That bill never passed. It's been reintroduced, but still hasn't passed.

This seems like the FDA fulfilling Obama's promise to bypass our constitutionally-mandated process for creating laws where he could through "regulations"...

Lovely. Too bad for 23andMe.

Permalink to Comment

20. Brett on November 27, 2013 12:26 PM writes...

I can't believe they were so stupid that they completely stonewalled the FDA for six months, while planning to expand. What did they think was going to happen?

In any case, they really should have put disclaimers on there saying, "This does not replace getting diagnosis from a doctor, if you have medical concerns, you need to see a doctor first". But that would have hurt sales.

Permalink to Comment

21. RKN on November 27, 2013 1:02 PM writes...

23 should just drop the pretense that their results are diagnostic. They're not; the variants they report are associative, not causal.

Tell customers that a given result may indicate an increased/decreased risk for this or that disease, syndrome, what have you, but clearly disclaim that any result is to be interpreted as causal or diagnostic. Prognostic? Maybe. But that should be discussed with your doctor. Notwithstanding the fact that many doctors (and most customers) don't understand the relevant molecular biology of what 23's analysis really measures.

Seems to me this would end the kerfuffle with the FDA at least.

Permalink to Comment

22. Doctor Memory on November 27, 2013 1:02 PM writes...

The contrast with the supplements industry is obviously instructive and depressing, but it also cuts the other way: the personal genomics industry is, at this stage of the game, not that far removed from the supplementeers (or, frankly, palm readers) in terms of its actual utility. The genes with obvious, unambiguous risks, we already test for when people have a family history or show symptoms. Everything else? Not so much.

Frankly, the current American fad for wanting to know your personal genome looks not that different to me than the Japanese conviction that your blood type determines your personality.

It's hard to blame the FDA for wanting to not let yet another multi-billion dollar quackfest bloom on its watch. But clearly such petty concerns shouldn't be allowed stand in the way of Larry Page's wife increasing their family net worth by a few more billion dollars. I wonder who the Orrin Hatch of personal genomic industry is going to be?

Permalink to Comment

23. JPL on November 27, 2013 1:38 PM writes...

Does Illumina's "Understanding Your Genome" conference
have FDA approval? For those who don't know about it:

Permalink to Comment

24. Fred on November 27, 2013 1:50 PM writes...

FDA to 23 and Me from "The Untouchables"

Malone: You're muckin' with a G here, pal!

Permalink to Comment

25. GottaBeHonest on November 27, 2013 1:56 PM writes...

Samples are collected, using an FDA approved medical device, the Oragene-Dx (OGD-500), made by DNA Genetek. Processing is done in a CLIA certified lab, using a chip produced by a known manufacturer. The raw data that comes back is undeniably within current testing standards. The problem is not within the process, nor the validity of the raw data, it lies solely in the interpretation, of that data. 23andMe has attempted to provide "help" with that interpretation, in the form of the personalized reports. However, even if they no longer provided such interpretations, as long as they still provided the raw data, it wouldn't matter. The smoke is already out of the bottle, community forums would just spring up to help others with interpretation. The fact that 23andMe charges a mere fraction of what is charged in hospitals and labs, likely infuriates some, especially considering that the "high priced" tests might even use the exact same collection device, possibly are processed in the same test lab, but only provide a tiny fraction of the information (raw data, depending on chips used).

Permalink to Comment

26. Cellbio on November 27, 2013 2:18 PM writes...


Interesting conference, have your genome sequenced and get a free iPad...but here is the difference between the conference and hype of 23 that offers diagnostic and prognostic insights..

"You will not receive medical results, or a diagnosis, or a recommendation for treatment from Illumina. Any results arising from the analysis of your genome sequence information that might be deemed medically actionable should be confirmed using alternative testing."

Permalink to Comment

27. gippgig on November 27, 2013 3:57 PM writes...

Are there any companies that offer genetic testing without making any medical claims at all?

Permalink to Comment

28. SMILES on November 27, 2013 9:02 PM writes...

Just like Google, the evil part is the "database".

Google gets information for free since dummies use it for free. But dummies has to pay $99 to "get" in the "database". What a shame!

Permalink to Comment

29. jbosch on November 27, 2013 9:19 PM writes...

So you pay 23andme some $amount and you receive a table for 250 potential risk factors ?
Has anybody thought of what else they might be doing with your DNA?
I mean in terms of full sequence and do some more analysis on it. A human genome project were no IRB protocol is needed and the data generated can be used for other purposes.
Just a thought - even if they are closed by FDA, they might have acquired sufficient data from their paying costumers to be able to discern new predictable markers perhaps.

Permalink to Comment

30. Crazy Ayatollah on November 27, 2013 9:29 PM writes...

For a terminal patient with an Incurable disease still being researched on, 23 and me and tests which test for common SNPs offers huge hope. I hear about people with metastatic melanoma who have cured themselves with education from current journals , health tests and treating themselves with drugs sometimes in other countries. They would be dead without this information or with a system where such information had to go through inefficient gatekeepers with perverse incentives like many current day doctors who modify their behavior based on incentives from insurers.

Medical insurers have the right to modify your treatment based on their financial needs by influencing doctors prescriptions and treatments but a science based company cannot tell you how current day science may be influencing your health. Holy crapola

Permalink to Comment

31. Anonymous on November 27, 2013 10:30 PM writes...

agree with #3 in general. Grew up in silicon valley, and Washington DC is far away, in more than just distance. There are oversight rules for compliance that need to be followed, so "don't be evil"...

Permalink to Comment

32. Anonymous on November 28, 2013 12:44 AM writes...

So here's the question: how accurate is it? I want to get sequenced sometime soon, and what I really want is just a text file of my sequence, so I can use it however is needed in the future. I'm sure in years to come I could interpret it in various ways down the road.

Permalink to Comment

33. Anonymous on November 28, 2013 2:47 AM writes...

@28: If a company doesn't make any money it goes out of business. There's evil and there's stupid. And only one of these can profit from the other.

Permalink to Comment

34. Anonymous on November 28, 2013 3:43 AM writes...

I'd wait until the disposable Oxford Nanopore sequencer or alike comes out, then I can just sequence my own DNA instead of having someboday else does it and know everything about me, genetically.

Permalink to Comment

35. Yancey Ward on November 28, 2013 3:45 PM writes...

One suspects that 23&Me realized at some point that the FDA would never be satisfied. I applaud the finger they raised towards Washington.

Permalink to Comment

36. HFM on November 28, 2013 4:48 PM writes...

As much as I sympathize with #30, people like that are exactly why 23andme kind of had it coming.

I'm not sure how to fix it. The data aren't worthless, and even if they are, I have a hard time arguing that people should be protected from their genomic information "for their own good". But the utility is so, so limited, and 23andme (et al) haven't exactly been shouting this fact from the rooftops.

From what I've seen, there are a few dozen SNPs on there that either cause Mendelian disease or turn up reliably in every GWAS, and are validated enough that there is (or could be) an FDA-approved test for that SNP. The rest is a no-man's land of marginal GWAS hits.

(Frankly, even if some of those pan out, I don't know how relevant they are. I don't know of any collection of "1.3X diabetes risk, 0.8X heart attack risk, etc" that would give a meaningful "personalized health recommendation". I assume my genes would benefit from fewer donuts and more exercise...everyone's would.)

They seem to be doing a decent job with patient education after the fact - a significantly better job than what I've seen from "professional" tests. (My family went through testing in the mid 2000s...this was not handled well at all.) But I don't know if it's clear to the average buyer how large a bucket of salt one needs to take this stuff with.

Maybe they need to revert to the Personal Genomes Project model of making people pass a genetics test? "Yes, I know what a SNP is and what a GWAS is, can name half-a-dozen reasons any given result is bogus, and promise not to jump off a bridge if some marginal hit from a single study tells me I have a 1.1X greater risk of something bad happening to me." I hope not, though.

Permalink to Comment

37. Anonymous on November 28, 2013 5:02 PM writes...

I really don't know what people expect from having their DNA sequenced... I'd rather have a long string of coloured beads than my own DNA sequence, at least I could wear it round my neck or use it to do sums.

Permalink to Comment

38. Hap on November 29, 2013 12:16 AM writes...

@35: Well, I'm guessing the FDA's response will be something like the lead singer of Social Distortion when he got flipped off before "Ring of Fire": "You got nine more of them fingers? You can take all ten of them and stick them up your..."

If you sell your sequencing services as a diagnostic tool, you don't have anyone else to blame when you find out that 1) getting anything through the FDA is hard, and 2) getting a diagnostic with broad scope and implications for individual and human health (more or less requiring a degree in biostatistics to understand accurately) and unknown predictive capability is going to be difficult.

If you're going to play in the real world, you play by real world rules.

Permalink to Comment

39. Yancey Ward on November 29, 2013 11:06 AM writes...


I can offer an opinion on someone's genome and be completely protected by the 1st Amendment. The diagnostic tool part of this is the research demonstrating links between disease and genome. Why doesn't the FDA embargo that research under the same rationale? 23&Me could simply break the two services, sequencing and analysis, and give the latter away for free, which is probably what is going to happen.

Really, what information is the FDA wanting from 23&Me that isn't completely obvious and available from the service terms themselves? The literature on the links is all publicly available, and the service directs you to it.

At some point, if you are being harassed, you say enough is enough, even if it might lead to more harassment. A stand has to be made, or you just keep knuckling under over and over.

Permalink to Comment

40. Cellbio on November 29, 2013 12:14 PM writes...

HFM nailed it. Outside of inherited causative genetic variations the SNPs associated with disease through GWAS have very little to no diagnostic or predictive value. If one looks at allele frequency rather than p value or relative risk and thinks through the concept of one individual's care being influence rather than deriving significance from a population study it is clear the value is limited to non-existent. As SNPs in GWAS typically have allele frequencies between 20 and 80%, it is not uncommon for a highly powered study to yield significant associations when allele frequencies vary between healthy and diseased by, say 35% vs. 45%. The prognostic or diagnostic value for one individual carrying that marker is extremely limited and close to zero without further correction for ethnicity. SNP variation is often greater across ethnicities than across health status. This is the concept that led to Marie-Claire King to propose that the bulk of GWAS associations were really cryptic heritage markers reflecting more of a family genetics association with disease than true locus association.

Permalink to Comment

41. pgwu on November 29, 2013 12:21 PM writes...

The Leona inside someone's head says only the little people follow rules.

Permalink to Comment

42. navarro on November 30, 2013 8:07 AM writes...

if they could get money to senator orrin hatch's (r, utah) campaign fund and relatives maybe he could get them the same kind of relief he got for the supplement industry.

Permalink to Comment

43. Mike on December 1, 2013 3:58 PM writes...

I doubt 23andme has been sitting on their hands for six months. I would bet the previous letter prompted a huge uptick in lobbying,the results of which we havent seen yet because DC has been such a circus lately. They probably didnt respond earlier because their answer would have either required gutting their business plan or would have provoked this result sooner, leaving them less time to set the stage for the legal/investor fights.

Permalink to Comment

44. Stacey on December 5, 2013 6:51 PM writes...

I have a 23 and me account which cost me $99. I use it mainly for genealogy research, which was one of the cheaper tests on the market. I have found firm connections with some of my relatives, including matches that I could confirm into the late 1700's. I was pleased that I could also get some "medical" information. All the results pages instruct us to consult a doctor, in addition to providing basic information. It also makes it clear that information provided is based on the current understand of genetic data, and will updated as needed. It does not claim that people will or will not get any conditions or diseases. It gives you the probabilities associated with them based on your DNA markers (that they are interpreting). Clearly there are additional factors that could influence an individuals health. I am very disappointed that the FDA is stopping 23 and me from conducting business. They are interfering with my product I have already purchased, with the understanding that additional users would be added to the list for DNA relatives. I feel that this is a gross overstep of the FDA.

Permalink to Comment

45. BigFreddie on December 9, 2013 7:09 PM writes...

Not to be a killjoy…but the “genotype to probability of crossing Falconer’s threshold” relationship is at fairly early days..and may never be strong past “your entire genome is contributing to risk–or maybe not”…maybe 23andMe knows that given the state of stats, confounding effect of environmental contributions, and the poor showing from GWAS studies that they simply can’t prove that the information they provide has clinical utility? Maybe all the error bars overlap :-). If most clinical tests have a huge range of values lumped into “moderate risk” I don’t know why we would expect genomic information to be more predictive and robust than say…A1C levels, a good but not definitive diagnostic test that is a relatively direct assay of an ongoing physiological state. My guess is that they just can’t meet a minimal standard…I mean come on folks, the mystery of missing heritability is solved by “genetic background and allelomorphy at 1000s of loci constitute the remaining 75% of genetic risk”…gosh…welcome to 1985. You can toss out the Bonferroni correction all you want to try to limit false negatives…but when the true positive is “risk spread over alleles at a 1000 loci is where more than 50% of the risk lies”…knowing much about the 60 loci that contribute .5% each of the rest of the risk is somewhat “clinically irrelevant”. Lesson from 1990…invest $60 billion in start up molecular medicine companies to generate $8 billion in corporate value…Look…control obvious risk factors, smoking, alcohol consumption, diet, exercise…and your win will be much bigger than anything some yokel with a sequencer will give you. Do I feel bad about the wasted days, wasted nights of…”post genomic medicine”, the “human microbiome project” and most GWAS funded by NIH over the past fifteen years? Yes, population genetics folks wondered how we got out of Africa and survived the bottleneck if our genome had been selected to kill us…but perhaps the experiment had to be done to the tune of billions spent and careers of legitimate geneticists, model systems biologists, biochemists, molecular biologists destroyed while the “community” chased statistical ephemera. BUT..we have done the experiment, my prediction is 23andMe will never anti-up, because the naked truth is there is no “there, there” for personalized genomic medicine. Now, all the pre-genome era techs that arguably benefit at least incidentally from some genome era infos, transgenic medicine, programmed T cells etc…this is great stuff, BUT it is not genomic medicine. Anyway, happy sequencing folks

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

One and Done
The Latest Protein-Protein Compounds
Professor Fukuyama's Solvent Peaks
Novartis Gets Out of RNAi
Total Synthesis in Flow
Sweet Reason Lands On Its Face
More on the Science Chemogenomic Signatures Paper
Biology Maybe Right, Chemistry Ridiculously Wrong