As anyone who knows me is aware, I am not a big fan of the FDA or Big Pharma. However, there are times that one must look closely at an issue before just rapidly rushing to opinions and I think this is one of them. I have been tested by several of the Direct-to-Consumer (DTCs) genetic companies like 23, and have analyzed the results with multiple genetic software programs. It was only when I saw discrepancies that I really started to investigate. Basically, there are two issues involved here:
1. The first is how accurate are the raw data results of the DTCs. From what I was able to find, this is not an issue at all. Correlations of results on individual markers tend to range between 99 - 99.7%, or nearly perfect, among the DTCs and private entities. Therefore, you can be fairly confident that your raw data results from 23andMe, or other DTCs, are accurate.
2. The second issue is how accurate are the reports these companies provide regarding disease "relative risk". This is where the problem lies, and it’s a big one. I became aware of this when I had the exact opposite disease risk assessments using the same raw data set. If you have seen these reports, these assessments will categorize you for diseases as high, normal or low risk. Then, they will assign risk ratios. There are very complicated formulas used by these companies for determining the risk ratios, and clearly, they don’t agree. Unfortunately, I have to apologize that I don’t have research links to provide at this time, but I will be happy to get them. Suffice is to say, that the research I have read showed that for many diseases, there was little correlation among the DTCs in there disease assessment of "relative risk". This means, for example, that one DTC can report to you that you have an above average risk for Crohn's disease with an odds ratio of 3.7, while another DTC, using the same actual genetic data, can report that you have normal risk with a 1.1 odds ratio.

I have clearly simplified the issue above as the algorithms used to get odds ratios for genetic markers and associated alleles are quite complicated. However, in the end, if you get a report from one of these companies, wouldn't you like to know that there is some degree of accuracy and validity to their assessments? That is the issue right now between the FDA and 23andMe, and ultimately all of the DTCs. They are sending people reports that describe in detail their risks, or lack of, for many diseases and illnesses, for which they have NEVER validated their results. We, the public, the patients, our families, may act on these results which may be clearly invalid. For example, I had one report tell me I was at an above average risk for Type 2 Diabetes. Using the exact same markers and alleles, another company reported to me that I had a below average risk for Type 2. My suggestions to anyone who has used the DTCs is to download the raw data set, which will be accurate, and run it through multiple genetic analysis software programs.

The FDA has been trying to get 23andMe, and the others, to validate their reports for five years to no avail. Maybe their move yesterday will put some more pressure on these companies to do the right thing. If not, we will lose an important vehicle made available to us at a low cost. Then, we will really be stuck in the hands of Big Pharma, where women have to pay upwards of $1,000 to a company like Myriad Genetics just to find out their results on one gene: BRCA.

Thanks and Happy Thanksgiving to everyone,

Gary