Phenotype (SNP) DNA test (23andme) - should I try to talk my sister out of this?

coureur · November 29, 2013, 2:57pm

This is just a standard boiler plate CYA disclaimer that everybody puts on all tests that are not yet FDA approved. If they were using that disclaimer properly they would not be making any medical claims for their product, i.e. telling customers that they at increased or decreased risk of diseases based on the results of the test. They would remain silent on the question. And slapping on such a disclaimer does nothing to absolve the company from properly validating all the medical claims it does in fact make.

calmom · November 29, 2013, 4:20pm

</li>
</ol>

But the question I asked assumed they had already done that – they use an outside lab, the lab complies with specific protocols, so it seems to me that part is easy to demonstrate, if the FDA will accept their documentation, the lab’s documentation, and Illumina’s documentation as to the process they follow.

But 23andme never claimed that or told me that. Why would they have to prove something that is at odds with what they said? (They said that I had a particular genotype that researchers have found to be associated with lactase production, and that meant that I was “likely” to be lactose tolerant, but maybe not.)

This is not a test for lactose tolerance; it’s a test for a genetic profile. The product they sold me is supposed to give me a list types of genes I have – the only thing I expect as a consumer to get from them is that the raw data is valid within a certain margin of error. (I assume that everything has a margin of error)

But that’s the part that makes no sense to me – that’s asking them to reinvent the wheel, for no good reason. Does that mean that no one can ever offer a report on my genotype at marker rs4988235? Or that 23andme can sell me the raw data, but I have to look up the research for each and every marker result on my own?

And if 23andme is required to do that, does that mean that every lab that offers genetic screening or testing anywhere has to meet the same requirements, as to every different test they offer now or in the future? If not, why not?

But that’s not what 23andme is marketing. They have sold me a test that tells me whether my genotype at rs4988235 is AA, AG, or GG. They didn’t even promise to test the rs4988235 marker- its just one of a number of results.

I paid the for the raw data, and I use their web site as a research tool, because that’s easier for me than going over to PubMed and running a search for rs4988235 - plus most medical journals charge about $20-$30 a pop to download and read full text articles. Also, my personal preference is to read the info on the 23andme site rather than wikipedia because I don’t really consider wikipedia a reliable source.

I want to know that 23andme report means that there is a 95+% certainty that my rs4988235 genotype is AG – that is, if the same test is done again, that it would almost always come out the same way.

I don’t want to see 23andme’s studies on the rs4988235 marker – I think the other independent studies are more reliable, and I know that 23andme markets to an international clientele, so I don’t see a particular benefit to having them do some small scale study on top of what has already been done. All that would do is reduce the value of the services I purchased, because it slows down the progress of their data-reporting service. I didn’t pay them to do carry out genetic research; I paid them to tell me what’s in my DNA and to provide me with reports correlating my results with other, independent published research.

coureur · November 29, 2013, 5:04pm

In the world of regulated medical testing that’s a distinction without a difference. To say "You have gene X, and here’s a study that shows that people who have gene X are four-fold more likely to come down with disease Y’ is in fact making a medical claim about your chances of getting disease Y. The fact that they threw in a “maybe” doesn’t put them in the clear. It’s obvious that they want you to conclude something about your chances of getting a disease based on their test.

To be compliant with their CYA disclaimer they should say “You have gene X” and that’s it.

From a regulatory standpoint it’s not reinventing the wheel so much as it is showing that this particular test agrees with the published studies. Someone certainly can offer a report on the clinical meaning of marker rs4988235 - just as soon as they complete their clinical trial and get their PMA approved.

They will for a while. Every company will have to do a similar trial and jump over the same regulatory hurdles until FDA becomes comfortable with this new test that it is being used properly in medical practice and presents only moderate risk to the patients. At that point FDA may choose to down-classify the test so that a 510(k) is sufficient and a full PMA is no longer required. This usually happens once about 4 or 5 companies have gotten the PMAs approved without serious public health problems arising from the use of the test.

Some types of tests are considered to be very high risk to the patients and will never be down-classified - anything to do with blood banking for example.

Actually just the opposite is more likely to be true. Because of its rigorous requirements FDA approval is regarded as the gold standard clinical validity throughout much of the world. Many countries, especially ones without a highly-developed regulatory system of their own will defer to what the US FDA says. If you have FDA approval you are golden in those countries too. Getting their approval is a mere formality.

FDA approval also plays a big role with insurance coverage. Having your test FDA approved won’t guarantee that insurance companies will cover it, but most of them use that as a huge first requirement. No FDA approval almost always means no coverage.

jonri · November 29, 2013, 7:48pm

Am I missing something? I know that one of the people involved in the founding of 23andme was married to one of the people involved in founding google. However, they separated a while ago. They may actually be divorced by now…I’m not sure.

Is everyone just assuming that because California is a community property state she’ll end up with a lot of $? I don’t have ANY knowledge at all…but I’d be shell-shocked if there weren’t a pre-nup.

And, is there any reason to think that she’ll just keep pouring money into 23andme?

Cardinal_Fang · November 29, 2013, 8:10pm

There’s this:

[Why</a> 23andMe has the FDA worried: It wrongly told me I might die young ? Quartz](<a href=“http://qz.com/151311/why-23andme-might-have-the-fda-worried-it-gave-me-a-false-result/]Why”>Why 23andMe has the FDA worried: It wrongly told me I might die young)

23andme falsely told this guy he had a horrible disease.

BunsenBurner · November 29, 2013, 9:01pm

calmom, my apologies for not replying earlier. I think by now coureur addressed your questions very well (thank you, coureur).

jonri, it is my understanding that Google, not just Brin himself, is a stakeholder in the company. They also have a Facebook billionaire as an investor… interesting.

Here is the opinion of a Forbes’ science and health reporter who’s been covering the FDA-related news:

[23andStupid:</a> Is 23andMe Self-Destructing? - Forbes](<a href=“http://www.forbes.com/sites/matthewherper/2013/11/25/23andstupid-is-23andme-self-destructing/]23andStupid:”>23andStupid: Is 23andMe Self-Destructing?)

romanigypsyeyes · November 29, 2013, 9:45pm

No, they didn’t.

Cardinal_Fang · November 29, 2013, 10:42pm

They told this person:

If you read that, if you got that information, you would think you were going to develop this horrible disease. They may say they are not doing diagnosis, but if it walks like a duck and it quacks like a duck, it’s a duck. What they said in this case, as it turns out, isn’t true. It was a misinterpretation of the scientific results. This person apparently does not have an elevated risk of having this horrible disease at all.

coureur · November 29, 2013, 11:28pm

I agree with every word of that Forbes article linked in post #106. An especially this summary quoted from the article:

“The FDA probably felt it had little choice. This is not the story of a big regulator choosing to squash a small company, but of a company that decided that it didn’t have to follow the rules.”

Also, the FDA officer quoted in the article, Illeana Elder, works in the Enforcement division of the FDA. The fact that they have chosen an enforcement officer as their spokesperson for this article, as opposed to say a regular reviewer or a branch chief, says a lot about where FDA’s head is on this case.

calmom · November 29, 2013, 11:41pm

Cardinal Fang, thanks for posting that article – without it I never would have learned about promethease.com. ($5 of my money very well spent… and it looks like 3 hours of my time digging into that whole new world.)

(My main takeaway from all of this is that I need to pull the raw data from 23andme and store it on my computer, lest something happen along the way to take away my access to the data I purchased about my own DNA. Nice to know that there’s an additional resource for analysis).

coureur · November 29, 2013, 11:52pm

Another piece on the topic:
[23andMe?s</a> problem isn?t the FDA?it?s that no one knows if it works ? Quartz](<a href=“Super Bowl XL in HD”>23andMe’s problem isn’t the FDA—it’s that no one knows if it works)

calmom · November 30, 2013, 3:49am

More articles:

<ol>
<li>Published journal article, looking at how BRCA mutation positive 23andme customers respond to the news: [Dealing</a> with the unexpected: consumer responses to direct-access BRCA mutation testing](<a href=“http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3628894/]Dealing”>Dealing with the unexpected: consumer responses to direct-access BRCA mutation testing - PMC)</li>
<li>Legal opinion about FDA burden of proof (interesting in the context of the PeerJ article above : [The</a> FDA Strikes Again: Its ban on home testing kits is, as usual, likely to do more harm than good - PointOfLaw Forum](<a href=“http://www.pointoflaw.com/archives/2013/11/t.php]The”>http://www.pointoflaw.com/archives/2013/11/t.php)</li>
</ol>

Cardinal_Fang · November 30, 2013, 12:04pm

There are two different issues here. First, is allowing 23andme to sell these kits good policy? Second, is 23andme in legal trouble?

I haven’t formed an opinion on the first question, but as to whether 23andme is in legal trouble, except for calmom’s #84, everyone I’ve read seems to be in agreement that they are in deep trouble, they’ve angered the FDA by ignoring them, and the FDA can shut them down. Also, their general counsel left last summer, which suggest that 23andme is just incompetently mishandling a dangerous legal situation, rather than pursuing some legal strategy.

[How</a> 23andMe Broke the Rules: The F.D.A. Versus Personal Genetic Testing : The New Yorker](<a href=“http://www.newyorker.com/online/blogs/elements/2013/11/the-fda-vs-personal-genetic-testing.html]How”>The F.D.A. vs. Personal Genetic Testing | The New Yorker)

calmom · November 30, 2013, 7:09pm

I haven’t read anything from an authoritative source – just a lot of opinions from people who don’t seem to have any specific info other than what was contained in the FDA letter.

I know what the FDA letter says but I haven’t been able to find clear answers as to what it means in practical terms.

<ol>
<li>I can see that 23andme is still selling their kits on line – and cannot find any legal authority to suggest that an FDA warning letter has the force of an order. So I am thinking that the letter means that 23andme should send a response to the FDA within 15 days. So that in legal terms, it appears to be more like a cease & desist letter rather than something that is by itself binding and enforceable. </li>
<li>Notwithstanding all of the chatter about what huge trouble 23andme has brought on itself, I can see on the FDA web site that there are tons of these letters issued that appear to be unresolved after months and even years. Obviously 23andme is higher profile than most of the others, so may be treated in a different way; also, the open status on the web site doesn’t really tell me one way or another whether the FDA has initiated court actions against these various companies. But the impression I have from what I can see still seems to be that the next steps in the process will be that 23andme will send a letter back to the FDA, and that it will be some time before further significant action takes place. Either there will be a new round of correspondence & exchange of documents that will be carried on until some sort of resolution is reached, or this thing will end up being resolved in the courts, perhaps 2-3 years down the line from now.</li>
<li>Everyone seems to be taking the FDA’s statement in the warning letter that “FDA has not received any communication from 23andMe since May” as an indication that 23andMe as an indication that 23andme cut off all communications at that time – but that’s not what I said. We don’t know what was in the May communication from 23andMe: for all we know it could have been a letter that contained a question or request from 23andMe to the FDA, that the FDA never responded to. (Just a hypothetical, but it could have been a letter stating that 23andme expected more the studies referenced in the January letter to be complete by the end of the year, to which the FDA never responded – and which 23andme took as tacit consent to the extended schedule). </li>
</ol>

Obviously I don’t know what actually did happen, but I do know that there is another side to this (there always is), and that the FDA letter itself demonstrates a history of extensive communications and cooperation with 23andme (“more than 14 face-to-face and teleconference meetings, hundreds of email exchanges, and dozens of written communications”).

Here is what I can figure out:

<ul>
<li>23andme has been working with the FDA at least since 2009 (“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.”)</li>
<li>23andme submitted 501(k)'s for its kit in 2012. A 501(k) is basically a request for clearance to sell a device that is substantially similar to other devices already on the market. (i.e., other genetic testing kits)</li>
</ul>

-The FDA wrote to the 23andme in March and May stating that it was treating those 501(k)‘s as withdrawn. ("Consequently, the 510(k)s are considered withdrawn, see 21 C.F.R. 807.87(1), as we explained in our letters to you on March 12, 2013 and May 21, 2013.’)

<ul>
<li>The FDA received some sort of communication from 23andme in May, but we don’t know whether that was before or after, or in response to, the May 21 letter from the FDA. (“FDA has not received any communication from 23andMe since May”). </li>
</ul>

I’d really need the answer to that last question before I could even begin to form an opinion on the sequence of events. Since the FDA letter does not assert that 23andme failed to respond to the May 21 letter, it’s reasonable to infer that the last communication was a response to the May 21 letter asking for something (such as reconsideration, an extension of time, or further guidance on what the FDA wanted from them). Since the warning letter does not claim that the FDA sent any further communications of its own since the May 21 letter – it’s also reasonable to assume that the FDA did not respond to the last communication from 23andme, and that it is 23andme that has been waiting on the FDA rather than the other way around. Not that the FDA is obligated to answer every letter or email it gets-- just that we really don’t have a clear picture of what has transpired in the past.

coureur · December 1, 2013, 8:23am

^^Since the public is not privy to the communications back and forth between FDA and the company, I can’t answer most of your questions. But I can offer a couple of insights.

First a clarification: it’s called a 510(k) not a 501(k). When FDA says “Consequently, the 510(k)s are considered withdrawn…” they are basically saying you have failed to respond in time to some previous request for information, so therefore your submission is dead. As the questions and responses go back and forth between the agency and the company each side has a time limit in which to respond. “Considered withdrawn” is FDA-speak for “Time’s up. You missed your deadline.”

The regulation they are quoting is a long list of elements that must be included in the 510(k) or PMA submission. The one that the company probably was having trouble providing is element (g) - Evidence that the device is “safe and effective” i.e. it’s analytical and clinical validation.

No, it is very unlikely that the FDA would fail to respond at all - if for no other reason than it is very easy for FDA respond. They are not the ones who have to do a clinical trial. They could just respond along the lines of “No, that’s inadequate. Please submit the type of clinical dsta we previously requested.” And then the ball would be back in 23&me’s court.

LBowie · December 1, 2013, 4:04pm

We can legally smoke cigarettes and buy guns but can’t be trusted to understand our BRCA results causing ourselves possible harm? Please.

BunsenBurner · December 1, 2013, 4:16pm

Who says YOU can’t? Go ahead, take the raw sequence data and have at it using freely available tools and publications. It is 23andme who can’t, because the law says is that one cannot sell interpretation of the data, i.e. diagnosis, unless it is proven to be what it claims to be.

Here is more food for thought as to what companies are planning to do with the data:

[Genetically</a> Predicted Life Expectancy And Life Insurance Evaluation - The Lens (beta)](<a href=“http://www.lens.org/lens/patent/US_2010_0324943_A1]Genetically”>http://www.lens.org/lens/patent/US_2010_0324943_A1)

And this one has been recently granted:

[Gamete</a> Donor Selection Based On Genetic Calculations - The Lens (beta)](<a href=“http://www.lens.org/lens/patent/US_8543339_B2]Gamete”>http://www.lens.org/lens/patent/US_8543339_B2)

Very interesting… Designer babies, anyone? ;)

calmom · December 1, 2013, 9:11pm

Coureur – it still comes down to the fact that we don’t know what was in the last communication from 23andme in May ("“FDA has not received any communication from 23andMe since May”), or whether that was before or after the May 21 letter from the FDA.

The litigator in me tends to hone in the things that are unsaid in a document or a letter. When lawyers write letters, they tend to include a detailed recitation of facts that favor their position, and leave out the stuff that might mitigate against them. Sometimes stuff is left out because it is irrelevant, but sometimes it is left out because it casts things in a very different light.

We don’t have all the facts and the people writing op eds on line don’t have them all.

If 23andme were to take this letter to an outside lawyer for consultation, I’m sure the lawyer would insist on being provided the full record of communications, and wouldn’t be offering opinions and advice without having that information in hand.

BunsenBurner · December 1, 2013, 11:01pm

calmom, what this company really needs now is not a litigator, but a very experienced regulatory compliance consultant who can help them dig out of this mess before they get shut down by a court order. Illumina paved the way for the FDA approval of novel genetic tests - there is no reason for 23andme to keep crying that the “evil FDA” holds back “innovation”.

calmom · December 1, 2013, 11:42pm

I don’t know what the company needs because I don’t know the facts as to what has happened or what 23andme’s future plans are, or what has already taken place behind the scenes. I’m not saying that they need a litigator, I am saying how I see things through the eyes of a litigator. I see a whole lot of unanswered questions.

One big problem is that it seems that the FDA has not yet issued formal regulations to govern LDT’s (laboratory developed tests) for genetics – so it’s hard to navigate a system without any written rules.

Phenotype (SNP) DNA test (23andme) - should I try to talk my sister out of this?

CONNECT WITH US