"I really don't think that these tests, from a medical or a health perspective, are useful. I don't think that they're interpretable," says Dr. Onel, who is the director of the Center for Cancer Prevention and Wellness at the Icahn School of Medicine at Mount Sinai. "Clinical tests and a 23andMe test or an Ancestry.com test just have different goals and different designs."
- Kenan Onel, MD, PhD, Kenan Onel, MD, PhD, is the director of the Center for Cancer Prevention and Wellness and associate director of Clinical Cancer Genetics and Precision Oncology at The Tisch Cancer Institute at Mount Sinai.
On the heels of Ancestry discontinuing its medical branch AncestryHealth, recent research looked at single nucleotide polymorphism (SNP) tests, the kind of tests used by direct-to-consumer DNA testing companies like the now-defunct AncestryHealth and 23andMe. They found that while they're great at searching for common variants, the tests become inaccurate when screening for rare genetic mutations for disease-carrying genes. For example, when testing for mutations in BRCA1 and BRCA2 genes (genes that put you at a higher risk of developing breast and/or ovarian cancer), the probability that subjects with a positive screening test truly had the gene was only 4.2 percent.
Dr. Onel, who is also the associate director of clinical cancer genetics and precision oncology at The Tisch Cancer Institute at Mount Sinai, explains that this is because SNP tests are genotypes from samples of people. So the more rare a mutation is, the smaller the pool of genotypes available to base results on. Basing results on SNP tests alone leads to incomplete results because they lack the nuance that's available from more robust testing that would be performed, and later interpreted, by a geneticist.
"That's why genetics is a medical profession," says Dr. Onel. "Each and every one of us, we have somewhere between 10 and 30 million variants. Given that large number of variations, you can be sure that the vast majority of these tests are completely meaningless, but some of it may have implications for wellness and illness."
In addition to offering an incomplete picture, the results from these tests are often given in relative, instead of absolute, terms. "The sort of information that you get is, 'Oh, you have a 20 percent greater likelihood than other people of having chronic stomach aches.' Twenty percent sounds like a really big deal, that's one in five. I would really panic if I was told that I have 20 percent increased risk for something," says Dr. Onel. "But what that actually means is 20 percent greater than the background rate. So if the background rate of stomach aches is one in 10, having a 20 percent increased risk means that you're now at a whopping 1.2 in 10. The problem with direct-to-consumer testing is that nobody understands probability theory. Nobody understands actually what they're telling you when they give you these percentages."
Dr. Onel says if you're serious about understanding how your genetics could impact your health, he says to seek out a geneticist instead of buying a direct-to-consumer genetic test.
"I wouldn't use them for anything clinical—I would use them because they're fun, because they're cheap, because they're entertaining," he says. "And it's like, 'Oh, look, I'm supposed to have blue hair—in fact, I do. Hahaha!' That's great, but I don't think that you can actually draw meaningful medical conclusions, health conclusions, from these sort of direct-to-consumer tests right now."
Oh hi! You look like someone who loves free workouts, discounts for cult-fave wellness brands, and exclusive Well+Good content. Sign up for Well+, our online community of wellness insiders, and unlock your rewards instantly.
Loading More Posts...