Your SNPs are your information

June 17, 2008

A quick follow up to my previous post. Reading Rob Carlson’s wonderfully written post on the C&Ds sent out by the California Department of Public Health to 13 consumer genetics companies got me thinking. He writes

There appears to be some tension between the interpretation of tests ordered for diagnostic purposes, which probably should require a prescription, and sequencing or genotyping services that provide information about a consumer’s genetic makeup.

That about sums it up doesn’t it. To a degree it reminded me about all the debates about intellectual property and ownership that led to the formation of Creative Commons. Biology is an information driven science and our genetic makeup is a fundamental piece of information. The WebMDs of the world provide information. Wikipedia provides information. Out of curiosity, I decided to take WebMD for a drive and look at 23andme as well (anyone can sign up for an account and explore the site). I chose a disease for which 23andme says the research is well established, Colorectal Cancer. The information includes a writeup on the biology of the disease with links out to resources like the American Cancer Society, etc. There is a hypothetical summary report as well (for the dummy family you are assigned) based on the known SNP for Colorectal Cancer. Here is what it says

This SNP occurs in a hypothetical gene called LOC727677. Little is known about the gene’s function; however, it is located in a region of DNA that often acquires extra copies in colorectal cancers. This suggests that the SNP is linked to a change in the activity of a nearby gene that influences cancer development.

One group found that the riskier version of this SNP is associated not only with an increased risk of colorectal cancer, but also with formation of the precancerous adenomatous polyps. This suggests that the SNP is linked to a gene that affects the very early stages of colorectal cancer.

The riskier version of this SNP has also been linked to prostate cancer in some studies, though more research is needed to confirm this association.

The studies whose data we report as applicable to those of “European” ancestry confirmed the association between this SNP and colorectal cancer in samples from the United Kingdom.

The association has not been investigated in samples of Asian or African ancestry.

One could argue that in the absence of a recognized test, this information doesn’t mean much. But note that part about information. That’s the key. This doesn’t say, you need to get yourself tested. All it tells you is about the biology of the disease and what we know about the SNP.

The other key is the blurb right on top of the Resources tab

The information presented on 23andMe is not intended as medical advice and should not be used for diagnosing, treating, or preventing disease. If you have questions about your genetic information and its relation to the latest scientific findings, here are resources you may find helpful, including available tests, interactive tools, and professional organizations and individuals in your local area.

The WebMD information is rather extensive, although not half as well presented. You can do a lot of digging, and find out about when and why you might need a genetic test. I see the two services as complimentary at some point in the future, since WebMD has a much larger scope. But in the end both are information resources, to be used by us (consumers) as per our needs, interests and opportunity.

To an extent, I just took a somewhat long detour. The point remains the one that Rob makes about information, and who gets to decide. In a market driven system, assuming CLIA labs are being used (there I agree with the health department), and where you are not being promised what is not possible, then, let the market decide whether these companies should exist or not.

I will end with another snippet from the blog post that started this post.

There are a great many snake oil peddlers and quacks out there who offer no caveats as to accuracy or effectiveness, and in comparison 23andMe and its competitors appear paragons of virtue. Direct-to-consumer genetic information services are creating a new market, and there always bumps along the way in that endeavor, particularly when regulators decide they know more about technology than do innovators. But it is a market. It is not, in priciple, directly related to health. Caveat emptor. Since when is this the concern of Department of Public Health?

Technorati Tags: , , ,

Zemanta Pixie

Posted by Deepak Singh | Filed Under Healthcare, Infotech, Your personal health 

Amazon Recommends

Comments

Viewing 5 Comments

    • ^
    • v
    Well said, but note that many genetic counselors feel threatened by these services, informational or not. Perhaps there should be some outreach to doctor's associations to soothe their worry?
    • ^
    • v
    While I agree with virtually everything you say, I do wonder about the level of accuracy in the results. Even a low level of error can have real world ramifications. Commercial labs have to be held to a high standard.

    I frankly have not investigated what the QA/QC procedures of the different companies but would feel more comfortable if a third party provided some numbers..

    Maybe something for Consumer Reports ;-)
    • ^
    • v
    The experiments are being done, at this point, using Illumina and Affy arrays, which are fairly standardized now. It would be good if they (or the labs doing the tests) released the workflows/procedures being used, e.g. what genotype calling algorithm is being used, what QC procedure as you say, what statistical algos etc.

    Perhaps if you are a paying customer they give you these answers. The markers they use are fairly easy to find out about, and the calling algos are also somewhat obvious. If the experiments are being done in a CLIA certified lab (which is the standard any diagnostics company needs to support, or any pharma service provider), then that shouldn't be a concern.

    I'll add that those who have downloaded their genotypes and re-analyzed them using alternate methods have found the results to be consistent (two data points)
    • ^
    • v
    Doesn't this really hinge on how things are represented? The DTC genetic testing services that the CLIA applies to are explicitly for a medical purpose. 23andme's service is explicitly not for medical purposes. While I agree they should hold themselves to the same standard as a medical testing lab, the fact is they aren't, and go to great lengths to make sure people understand this. If, despite all the warnings to the contrary, someone uses their 23andme information to make a medical decision, that's their own damn fault.

    All they have to do if they're worried about something in their 23andme results is to go see a doctor and get themselves an actual genetic test designed and validated for diagnostic use. Then they, with the help of their doctor, can start making medical decisions with no hand-wringing about 23andme's accuracy necessary.

    Someone could potentially use one of those personal breathalyzer things to decide whether or not they're safe to drive and get a DUI anyways due to an instrument failure. Does that mean they shouldn't sell them, or do we give people the freedom to have it along with the responsibility to accept the consequences if they put too much trust in its results?
    • ^
    • v
    I pretty much agree with you on every point. The nature of the services is clear, and well defined, at least the more visible ones.

    They haven't even given the market a chance to screw up yet :)
 

Trackbacks

(Trackback URL)

close Reblog this comment
Powered by Disqus · Learn more
blog comments powered by Disqus