Ancestry is in the "very early stages of a conversation with the FDA," Ancestry CEO Tim Sullivan told The Verge. "We think it's totally appropriate that the FDA has stepped in to pretty aggressively regulate direct-to-consumer genetic tests — and we're just starting from that perspective, and trying to work very closely with them."

Ancestry's health push is fairly recent. Before this summer, the company's primary focus was to help users learn more about their family. The company's $99 DNA kit is part of that mission; people who send in their saliva for genomic analysis can learn about their ethnic origins, ancestors, and relatives. In July, the company struck a deal that gave the Google-incubated Calico — a medical research company that wants to "solve" the diseases of aging — the ability to look for markers of human longevity in its anonymized genetic database. It also released a new product designed to help people trace health conditions in their family. Now, Sullivan says that his company hopes to dive into the genetics of health. But before that can happen, Ancestry's going to have to figure things out with the FDA — an agency that has worked very hard to stop direct-to-consumer genetic testing companies like 23andMe from doing exactly what Ancestry plans to do.

(Ancestry)

Scientists are only just beginning to decipher how human DNA affects a person's risk of developing a disease — but that hasn't stopped some companies from trying to cash in on the science's promise. When the New York Times compared results from three direct-to-consumer genetic tests two years ago, the author of the piece, Kira Peikoff, noted that "the discrepancies were striking." One test showed that Peikoff was most at risk for psoriasis and rheumatoid arthritis, whereas another claimed the exact opposite. In another instance, two tests agreed that Peikoff's risk for coronary heart disease was close to average, but the third claimed her risk was higher than normal. Genes only account for a portion of a person's disease risk (environment and other biological factors play a role, too) — and these tests showed quite clearly how little their manufacturers understood even that small amount.

One of Ancestry's competitors, 23andMe, was among the companies included in the article. For years, 23andMe sold and advertised — quite loudly — a DNA kit that the company claimed could tell people if they were at a higher or lower risk of developing a disease. But in 2013, the FDA ordered it to halt the sale of its DNA kit because the genetics company had failed to demonstrate the validity of its product. There just wasn't enough data to support 23andMe's medical claims. The FDA also faulted the company for failing to provide customers with a doctor's support — support that could have helped people avoid treating a disease that they might never develop. Because of these issues, the FDA concluded that 23andMe was putting people's health at risk. That's the kind of controversy that Ancestry now hopes to avoid. But doing so will be an uphill battle; Ancestry's biggest challenge by far will be to prove that the genetic health information they want to give consumers has value — and isn't putting people at risk.

"A lot of people want to get into the genomics risk assessment risk game. It's seen as highly lucrative," says Art Caplan, a bioethicist at New York University. But even the ethnic information that companies like Ancestry give their customers isn't all that reliable because for most of human history, people were nomads. "I don't think they're giving customers straight information about their ancestry — and I'm wary when I hear them saying 'Oh well now we're in the health business.'"

Misha Angrist, a bioethicist at Duke University, echoes Caplan's distrust. Since the FDA's response to 23andMe, he has a "much more jaded and jaundiced view of the power of personal genomics." Carrier status and pharmacogenetics are "obvious applications, but what does success look like in those areas?" he says. "Telling somebody that their lineage goes back to south Asia is very different than telling them what their likelihood of having a metabolic syndrome is."

Eggs, in multiple baskets

Even without the health angle, Ancestry's DNA kits have proven popular. In 2012, when the company launched AncestryDNA, the company mostly sold the kits to its existing customers. Today, over 70 percent of people who buy the kits aren't existing subscribers at the time of the purchase, Sullivan says. But that's not all; according to Sullivan, Ancestry has now managed to collect DNA from more people than 23andMe, which started selling DNA kits in 2007. "We're confident that we now have the largest [genetic] database," Sullivan says.

Sullivan bases this pronouncement on Ancestry's own analysis of both companies. "We've been tracking [23andMe's] growth, and we know our growth, and we've been accelerating," he told The Verge — before admitting that "truthfully, that is a 'Tim-Sullivan-estimate-assertion.'" This summer, both companies announced that they had collected DNA from 1 million customers, but Ancestry is now "nicely over one million," Sullivan says, though he declined to give the exact number. "23andMe has not released updated customer numbers since announcing our one millionth customer earlier in the year," said Carlyn Tomares, a press representative for 23andMe, told The Verge in response.

Regardless of who wins the "my-genetic-dataset-is-larger-than-yours" contest, a genomic health push will be a hard sell for the FDA. Sullivan knows that the agency might not like the thought of Ancestry telling people how much more likely they are of developing a disease. "I think that is going to be probably the most tricky stuff for the FDA to get comfortable with, and I think for good reason," he told The Verge. "Today, family health history is still a better predictor of health risk than a DNA test."

Given what Ancestry's trying to accomplish, that statement is peculiar. But that doesn't seem to be giving Sullivan pause; securing Ancestry's place in the world of personal genomics is worth it — even if it takes years, or decades. "Look, Ancestry is going to follow the FDA's lead on this and add as much value to our DNA test as we can within the FDA's guidelines," he says. "We'll expand the AncestryDNA test in, frankly, whatever way the FDA allows us to do."

If the FDA decides that it doesn't want Ancestry to enter the health genetics sphere, than the DNA kit will remain a genealogical tool — much like the one currently sold by 23andMe. But that certainly won't stop Ancestry from profiting from all the genetic data it continues to gather. As long as consumers give their consent, regulators won't stop Ancestry from selling their anonymized genetic information to various other companies. The Calico deal was just the beginning, Sullivan says.

"As we gather more health information, more family health histories, we'll be able to do some of the same things we've done with Calico, but with Big Pharma, who might be looking to make a discovery about a particular disease or perhaps is contemplating a drug trial," he says, echoing 23andMe's desire to participate in drug development. "Maybe Ancestry could be helpful in helping someone who is conducting a drug trial recruit patients for that trial."