Lab industry observers have long suggested that that the 23andMe business model ultimately involved selling anonymized client genomic data to the deep-pockets pharmaceutical industry (see: 23andMe as an Example of "Big Bang" Market Disruption; 23andMe Builds Online Sarcoma Research Community). This model required abundant capital up-front to attract clients with a relatively low lab testing cost. At some point, it's even possible for these costs to drop to zero but this might arouse suspicion and reveal more about the business model than was necessary. At any rate, a recent article discussed a new relationship between Pfizer and 23andMe to study the genetics of inflammatory bowel disease (IBD) (see: Pfizer, 23andMe Team up to Study Bowel Disease). Below is an excerpt from the article:
Pfizer is teaming up with DNA testing company 23andMe to study the possible genetic underpinnings of inflammatory bowel disease, a hard-to-treat ailment that affects an estimated 1.4 million Americans. Under the agreement, Silicon Valley-based 23andMe will map the DNA of 10,000 patients who have forms of the disease, which include Crohn's disease and ulcerative colitis. Patients will submit saliva samples using 23andMe's at-home collection kit and then fill out online questionnaires about their disease and symptoms. The companies hope to identify genetic similarities among patients with the disease, which could eventually guide development of new targeted drugs ....The cause of inflammatory bowel disease is unknown, though many scientists suspect genetics play a role....Financial terms of the agreement were not disclosed. 23andMe....has penned two similar deals with drugmakers Genentech and Janssen Pharmaceuticals (see: 23andMe and Genentech Expand Cancer Study). The agreement brings Google-backed 23andMe one step further into the medical mainstream, after a widely publicized dispute with federal health regulators last year. In November, the Food and Drug Administration ordered 23andMe to stop marketing its health-related genetic tests, warning that erroneous results could cause customers to seek unnecessary or ineffective medical care. The agency said that even after numerous meetings and email exchanges with the company, government regulators still had no assurance that the firm has analytically or clinically validated its technology....Still, the FDA sanctions haven't stopped the company from partnering with other parts of the federal government. Two weeks ago the National Institutes of Health signed a $1.4 million agreement with 23andMe to access the company's DNA database to research rare diseases.
I got the sense from this article that 23andMe will proactively seek cell samples from "10,000 patients" with IBD. Their current company database may already include individuals who have self-declared that they have this disease but some of these data will be unreliable. So the question arises as to how 23andMe will identify and obtain samples from these 10,000 patients. The company would probably need to seek them from clinics, hospitals, or physician offices in order to obtain cell samples from patients with well documented disease.
For me, the most interesting part of the article is that while 23andMe seems to be still feuding with the FDA, the company has also cut a deal with the NIH to provide access to its DNA database by the agency. If the company database contained potentially erroneous results, why would be NIH be willing to purchase access to it for $1.4M? It's possible that this question can be answered in a relatively simple way. As noted above, the FDA's beef with 23andMe is simply that "erroneous results could cause customers to seek unnecessary or ineffective medical care." In other words, the FDA believes that it is protecting consumers. In comparison, the NIH is buying access to the 23andMe DNA database which it believes to be scientifically sound. If this speculation is correct, it seems to me that 23andMe should be able to generate a report for its clients that is not erroneous nor leads them to incorrect conclusions. Here is how I summed all of this up in a note of November 2013 about how the company might be able satisfy the FDA (see: The Dispute between the FDA and 23andMe; What's Really Going On?)
In the end, I suspect that 23andMe will need to print a strong warning label on its reports to the effect that the test results have little medical significance and must not be used to make any important diagnostic or treatment decisions by consumers without confirmation by a CLIA-certified lab and a qualified physician. I don't think that such a warning will hurt its business and consumers will continue to have access to interesting and useful genetic screening test results.