September 19, 2010
Direct Genetic Testing Companies Face Regulation
In a Hastings Center special issue on personalized medicine Ronni Sandroff, editorial director of Health and Family at Consumer Reports, has an article Direct-to-Consumer Genetic Tests and the Right to Know: Genetic testing can be too much information, but that doesn’t mean it should be kept from consumers. (requires free registration)
Pathway Genomics Corporation and Walgreens announced in May that they would offer genetic testing collection kits at the drug chain. The Food and Drug Administration took notice. In a May 10th letter, the agency notified Pathway Genomics that its product, “intended to report customary and personal genetic health disposition results for more than 70 health conditions . . . Appears to meet the definition of a device as regulated under the Federal Food and Drug and Cosmetic Act,” thus requiring FDA approval. The company has delayed plans to market the product. A similar letter was sent to five other direct-to-consumer testing companies in June.
I think the FDA should step out of the way. We really do have a right to know our full DNA sequence or any subset thereof.
As for interpreting the results: If the FDA gets into the job of deciding which genetic variants have known effects on every disease, ability, function, and appearance of humans then the FDA will be biting off far more than it can handle. The FDA's decisions will tend to be very slow in coming with interpretations coming many years later than would otherwise be the case. As long as the companies providing test results do not cause harm by causing customers to make decisions that harm their health I do not think the direct-to-consumer testing companies should face much in the way of government regulation.
Current genetic tests do not yield definitive results on many of the conditions for which they try to assess risks. But as the meaning of more genetic markers become known ambiguous and contradictory interpretations will become less common.
Most telling, on July 22, the Government Accountability Office released a study in which it sent identical DNA samples to four genetic testing firms to test for fifteen common diseases and conditions. Contradictory results abounded. For example, various companies told a forty-eight-year-old male that he had average, below-average, or above-average risk for prostate cancer. These contradictions can be explained, in part, by the fact that companies analyzed different genetic “markers,” and that scientists disagree about what these mean in real-life situations.
As long as the companies state when each marker by itself does not predict 100% odds of getting some disease people should be able to receive interpretations of their test results from the test vendors.
I see a lot riding on the ability of companies to offer direct-to-consumer genetic tests and other forms of biological testing. We will have much faster product development and innovation if people can directly request tests and get test results. A faster rate of test development with less government regulation will accelerate the rate of biotechnological advance. A faster rate of advance will save lives in the long run.
There's a lengthy and very good article in the September 2010 issue of trade press periodical "CAP Today" by Anne Paxton. She writes about what's driving increased FDA regulation of Laboratory Developed Tests. The FDA contention that DTC genomics tests should be considered as LDTs is gaining sway, in my opinion. This would lead to the outcome that Randall decries in the post -- restrictions on individuals' rights to know what their DNA is, and how it can be interpreted.
Prepping for a firmer FDA hand in regulating LDTs
Actually, a review of the letter sent to Pathway Genomics (http://www.fda.gov/MedicalDevices/ResourcesforYou/Industry/ucm211866.htm) indicates that the FDA is seeking to regulate their Genetic Health Report because it meets the legal definition of a medical device. Frankly, I can't see how a company making diagnostic products, or a technologist would think that such a product shouldn't have the same regulatory oversight provided a scalpel or tongue depressor. A regulated medical device manufacturer is obligated to implement the Quality System regulation (a basic quality system for design and manufacturing similar to ISO9001) and in all likelihood, submit their product for premarket clearance by the FDA. Clearance by the FDA entails providing objective evidence that the product is safe and effective.
The FDA does not appear to be overreaching to keep a new kind of product from consumers because they think consumers shouldn't know their own genetic data. The FDA's principal concerns are: 1) that the test is safe, 2) is it effective (reliable, consistent and accurate) and 3) that any diagnostic claims made by the manufacturer are supported by research.
From your comment it seems that the exact relationships between diseases and genetic variants are somewhat unsettled. This would have a direct bearing on the diagnostic claims that could be made by Pathway Genomics. Rather than claiming a benefit that is unproven, e.g., prostate cancer risk as described in the GAO report, the manufacturer can make claims as to the accuracy of the sequencing or identification of specific genetic defects.
The data suggests to this casual observer that the market for at home genetic testing may not yet be fully developed or ready for actual consumer use. Trying to build a company or a new market on the backs of scared or worried patients looking for reliable diagnostic information - when the data cannot be proven to be reliably diagnostic is wrong.
When information is a "device", what limit is there on the "protection" imposed by the FDA?
"Daddy, where do babies come from?"
"Better you don't know, kid. No telling what kind of trouble you might get into."
And by the way, what's next, bathroom scales? Measuring tape? Both body weight and BMI are casually misunderstood as indicators of one's health prospects, so ...
How's this. If the FDA's assessments are indee so valuable (and could I say even compelling) rather than *prohibit* and *require*, a simple "FDA Approved" logo should suffice. Meanwhile, anything without their august stamp of "safe and effective" well, take you chances. It works for Underwriters' Lab. (Why this won't work for the FDA, I leave as an exercise, with one hint. Who's agenda is served, the buyers' the sellers' or another.)
If the FDA said to the genetic testing companies that they had to indicate that the science wasn't settled, that the research literature didn't enable statements about risk with a high degree of certainty then I wouldn't object. But to deny the dissemination of genetic sequencing test results until that time denies us basic information about our genetic sequences. I have a right to know that information. The FDA does not have the right to deny it to me.
I do not want the FDA deciding what conclusions can be drawn about genetic sequence data. They lack the competence and they will require too high a bar for what is allowed to be claimed. If a genetic test result can suggest many different ways you might want to adjust your diet, lifestyle, and medical decisions the results do not need to be perfect in order to be useful.
For example, if genetic tests can suggest that some people ought to eat a lower carb diet while some other people should eat a lower fat diet the results do not need to be correct in all cases for people to improve their diet decisions on average.
In order to promote the general welfare, the FDA has approved mandatory HIV infection of every newborn so that there is no disparate impact of HIV by race.