Invitae Questions Accuracy of 23andMe’s Cancer Tests in Certain Populations

A competitor, Invitae Corporation, presented a poster at the American Society of Human Genetics (ASHG) annual meeting that essentially accuses 23andMe’s tests of completing missing a cancer-causing DNA variant in some ethnic groups.

Although direct-to-consumer (DTC) genetic testing, such as those supplied by 23andMe and Ancestry.com, have exploded in popularity, their utility for actual clinical testing is limited. In early 2018, the U.S. Food and Drug Administration (FDA) gave the first-ever authorization for DTC genetic test for cancer risk, specifically three genetic variants found on the BRCA1 and BRCA2 genes associated with higher risk for breast, ovarian and prostate cancer.

A competitor, Invitae Corporation, presented a poster at the American Society of Human Genetics (ASHG) annual meeting that essentially accuses 23andMe’s tests of completing missing a cancer-causing DNA variant in some ethnic groups.

The overall argument is rather dramatic. 23andMe’s tests and the company itself have never claimed that the DTC test evaluates all possible mutations that could cause cancer. The FDA approval was quite clear on the fact it was only testing for three specific genetic variants. It does not sequence the entire gene, it uses microarrays to identify those specific variants, which are the most common and best-studied in the general population.

23andMe responded to Invitae’s presentation, stating, “It is patently wrong to state that 23andMe delivers ‘false negatives’ for variants that it does not test nor claim to test for. This is a false and misleading characterization of 23andMe’s test.”

What is less clear is whether consumers who order the tests completely understand that. Erica Ramos, a personalized medicine expert at the National Society of Genetic Counselors, told STAT, “Whether consumers understand that has been on the radar of genetic counselors for a long time.”

And although Ramos did not directly comment on the 23andMe test, she did point out that the fewer variants a company tests for, the more likely it is to miss disease-causing variants. 23andMe tests for three variants, while MyHeritage, for example, tests for 13.

In the Invitae study, 270,806 patients with a family history of cancer were referred by doctors for testing of the MUTYH gene. This gene has mutations associated with colorectal cancer. Another 119,328 patients were referred for testing of BRCA1 and 2. The 23andMe test evaluates two MUTYH variants common to individuals whose ancestry is from northern Europe and for three BRCA1/2 variants common in Ashkenazi Jews.

None of these patients were 23andMe customers.

The Invitae study indicated that if these patients in the study had been 23andMe customers, 40% of the patients with the disease-causing mutation in both MUTYH genes would have turned up negative. Two copies of that mutations result in almost certain colorectal cancer. The study also found that 22% of the patients with a single MUTYH mutation, causing twice the lifetime risk of colorectal cancer, would also have been missed.

The real risk here isn’t that the 23andMe tests—or any of these types of tests—are inaccurate, it’s that patients who use them will think their results are definitive, when they clearly are not. This was known prior to the Invitae presentation. These genetic tests, 23andMe or any other companies’ tests, do not provide results for genetic mutations they do not actually test for.

“One might argue that such [gene chip] tests do more harm than good,” Dennis Grishin, chief scientific officer of DNA testing company Nebula Genomics, told STAT.

The poster presentation, because it was not published in a technical journal, was not peer reviewed. It is part of an expansion study the authors presented earlier in 2019 at the American College of Medical Genetics annual meeting.

“Having an understanding of what you were tested for, and what was not tested for, can be challenging,” Ramos told STAT. “But there are cases of people saying, oh, I don’t have [a cancer-causing] BRCA, when the test couldn’t tell them that.”

The bottom line may very well be that for consumers using these tests, the best advice is to have the results confirmed by a medical laboratory and to discuss the results either way with your personal physician or a genetic counselor.

MORE ON THIS TOPIC