His decision came as an investigation into sexual harassment allegations against him was ongoing.
Rather, the breast cancer mutation screen was classified as a type of medical device with obligations for the company to reduce risks to customers.
March 19, 2018|
ISTOCK, BLACKJACK3DIt would have been remarkable news if the US Food and Drug Administration (FDA) had approved the breast cancer mutations test of 23andMe. The test includes three BRCA mutations that are a common cause of familial breast and ovarian cancer exclusively in Ashkenazi Jewish women. But according to numerous media reports, the FDA did just that earlier this month.
When the FDA approves a device, it means that “the agency has determined that the benefits of the product outweigh the known risks for the intended use.” Is the agency suggesting that there are benefits of testing Ashkenazi Jewish BRCA mutations in non-Jewish women? How does the FDA justify that the benefits of testing outweigh the known risks?
Well, it doesn’t. The FDA did not approve the test. Rather, the FDA classified the 23andMe Personal Genome Service (PGS) Genetic Health Risk Report for BRCA1/BRCA2 (Selected Variants) into Medical Devices class II.
The FDA categorizes medical devices based on their risk profiles. Class I devices have the lowest risk and class III the greatest. Dental floss and elastic bandages are examples of class I, condoms and powered wheelchairs of class II, and pacemakers and breast implants of class III. Class III devices require FDA approval before they can be marketed, while class I and II don’t.
The biggest challenge, by far, will be to understand why and how the correct interpretation of DNA results can vary between people.
The FDA identified that this test presents a risk of an incorrect understanding of the test by consumers, false test results, and poor interpretation of the results. The agency requires that 23andMe puts “general and special controls” in place to reduce these risks. The company received a long list of warning statements, specifications, and limitations of the test that it needs to include in the labeling of the product and the pre-purchase information. The company is also required to conduct a user comprehension study to assure that consumers understand the material that is provided.
Whether the controls will help consumers to correctly understand and interpret the test and its results remains to be seen. The simple and uniform presentation of information that is needed to achieve that goal is at odds with the enormous variation in the quality, relevance, and value of scientific and clinical studies on which the test results are based. The biggest challenge, by far, will be to understand why and how the correct interpretation of DNA results can vary between people.
If you carry one of the three BRCA mutations, the interpretation is straightforward. It means that your breast cancer risk is much higher than that of other women. The significance of not carrying one of the three mutations depends on who you are.
If you are an Ashkenazi Jewish woman with a family history of breast cancer and your affected relatives carry one of the three mutations, then the absence of the mutation is an enormous relief.
If you are a non-Jewish woman, then the absence of these mutations means nothing, as you were unlikely to carry them at all. For you, getting screened is like taking a pregnancy test when you already know you are not pregnant.
And it also means nothing if you are a non-Jewish woman with a family history of breast cancer. The mutation that causes breast cancer in your family is likely one of the thousands of others that are known or one that is yet to be discovered.
The DNA test is equally accurate for all these women, but the value and effect of the test differ between them.
23andMe’s press release about the BRCA test does not mention “approval,” but correctly states that the company received FDA “authorization” to market the BRCA test. Also, in its advertising of the Genetic Health Risks and Carrier status tests, the two tests that the FDA allowed earlier, the company writes that each “meets FDA requirements.” But an authorized test that meets FDA requirements is not an FDA-approved test, and class II classification does not mean that the agency endorses direct-to-consumer genetic testing either. The FDA allows consumers to make their own decisions about genetic testing, including the bad ones. Time will tell whether those decisions are in consumers’ own best interests.
Cecile Janssens is a professor of Epidemiology at Emory University’s Rollins School of Public Health who studies the genetic prediction of disease.