My grandmother had breast cancer in her late 30s, an aunt in middle age. We’re Ashkenazi Jews, which means we have a higher chance of carrying one of the major “breast cancer genes”—1 in 40 Ashkenazi Jews has a so-called BRCA mutation versus 1 in 500 members of the general population. Depending on the particular BRCA mutation, a female carrier has a much higher lifetime chance of developing breast or ovarian cancer—as high as 80 percent and 40 percent, respectively.
If, because of my family history and ethnic background, I wanted to get tested for a BRCA mutation, I’d have a few options. First, I could go the established medical route. My doctor could order the test for me, and insurance might even pay for it, thanks to the Affordable Care Act. Or, I could take advantage of any number of direct-to-consumer genetic testing companies. With the click of a mouse (and the entering of my credit card information), I could order a BRCA testing kit from a company, such as Color, for as little as $250. The company would send me a saliva collection tube, I’d mail it back, and I’d get results online.
Or, like a growing number of people, I could go DIY. That route would look something like this: I’d order a $99 autosomal DNA test through a site like 23andMe or Ancestry.com, which promise to tell you your ethnic heritage, but are currently barred by the FDA from giving almost any medical analysis. (Color gets around this by having doctors on call to order the tests). I would then take my raw DNA data and upload it to one of several third-party “interpretation only” web services. These web services, a couple of the most prominent being Promethease and LiveWello, interpret your data for you, spitting out information in a format a layperson can (in theory, at least) understand.
This is the outlaw edge of the genetic testing world. The sites generally have a spare, low-budget look to them. The services they offer are free-to-cheap ($20 for LiveWello and a suggested donation of $5 for Promethease). Several are run as side projects by people who have regular day jobs. While they have not so far come under FDA scrutiny, it’s unclear whether this reprieve will last.
Without these sites' analyses, readers with some science savvy could actually browse their results themselves in search of particular DNA sequences, perhaps with the help of one of the many digital tools created for this purpose by interested programmers. In response to a recent New York Times op-ed about breast cancer testing, one commenter said she’d scanned her 23andMe results herself. “[I]f you know the snp's [single nucleotide polymorphism, essentially a copying error in DNA], you can browse for your BRCA1 and BRCA2 results. This is cheaper than expecting insurance to pay for it,” she wrote.
The MIT Technology Review tested several of these DNA interpretation sites recently and found variable results. Some sites reported as few as 35 gene variants, while Promethease reported 17,667. Other sites offered “detoxification profiles" and hawked vitamins and chiropractic care.
With Promethease, perhaps the most comprehensive intepretation site, users can learn cool-but-trivial facts, such as whether they have a “rare T allele” that makes them better at tasting umami flavors. They can also glean more serious things, including whether they have a variation that might make them more susceptible to diabetes or Alzheimer’s disease.
Not surprisingly, many health care professionals are concerned about the rise of DIY genetic testing, both about direct-to-consumer genetics testing companies and interpretation-only services.
“They vary pretty widely in terms of quality and clinical utility,” says geneticist Bruce Korf, president of the American College of Medical Genetics and Genomics Foundation for Genetic and Genomic Medicine, speaking generally about direct-to-consumer genetics services.
Having a pile of translated DNA may not be terribly helpful for a consumer, and it may cause undue anxiety and even unnecessary medical tests. The average person will probably have a hard time parsing what “1.5x - 2x increased risk for cervical cancer” (a result from a sample test on Promethease) means for them. A doubled risk of any kind of cancer sounds scary, of course. But you’d have to know that the U.S. rate of cervical cancer is about 7.4 in 100,000 women a year. That means one in about 13,500 women will be diagnosed with cervical cancer this year. Pretty long odds. If you multiply the risk by 1.5, that means one in about 9,000 women. Multiply the risk by two, and the rate is still only one in about 6,750 women. This level of risk is very unlikely to warrant any steps beyond what a woman should already be doing, like getting a regular pap smear.
Explaining this kind of information is the job of a genetic counselor, something most websites lack, says Joy Larsen Haidle, president of the National Society of Genetic Counselors.
“It’s our job to support people and make sure they understand what the results mean so they can fully benefit,” she says.
Haidle finds the interpretation-only sites particularly concerning because they come with so little explanation.
“To me, that’s a little more troubling because then the onus of trying to understand what that data might mean falls even more onto the actual patient themselves,” she says.
Unless the FDA cracks down on interpretation sites, it seems likely that the audience for this sort of genetic self-exploration will only grow. One of the founders of Promethease said, last year, that the service was being used to create some 500 gene reports each day. Analysts predict that the market for direct-to-consumer genetic testing will reach $230 million by 2018.
Korf envisions a future where the medical establishment and direct-to-consumer companies might be able to work together for the patient’s benefit.
“I’m open to the idea that there need to be new paradigms for testing and counseling and new technological approaches,” he says. “However this evolves, it needs to be done with some caution, to ensure that we don't take multiple steps backward while trying to move forward.”