23andMe Tries to Woo the FDA
Anne Wojcicki bounds into a conference room in Mountain View, California, straight from a five-mile ride from home on an elliptical bike. The 40-year-old cofounder and CEO of the consumer genetic testing firm 23andMe is breathless, and not just because of the workout. On this warm day in mid-June, Wojcicki is “super-excited” about an announcement scheduled for two days hence: the Food and Drug Administration has agreed to review a health-related genetic report the company wants to make available to customers.
It’s the first step out of the FDA’s doghouse for 23andMe. For $99, the company analyzes key components of a person’s DNA from a vial of saliva, but last November the federal agency issued a testy warning letter barring it from marketing its service. The FDA said that by selling consumers a test and health reports that outlined their chances of getting dozens of diseases, plus their likely response to various drugs, 23andMe was effectively selling a medical device. That requires explicit approval—and the FDA said 23andMe hadn’t come close to providing enough evidence that its test provides accurate, reliable health assessments.
The FDA allowed the company to keep selling the test as long as it provided only raw genetic data and ancestry information, nothing on disease. Sales slowed. Evidently, people care much more about their chances of getting Alzheimer’s than about how much Neanderthal DNA they have. “A lot of companies would shut down in this situation, but we looked at how do we double down,” Wojcicki says between sips from a water bottle. That meant talking frequently with the FDA, marshalling more data to support health claims, and hiring a number of executives experienced in medical devices.
The company’s plight reflects just how challenging it is to translate genetic data into useful medical information. Though the company encourages customers to seek a doctor’s advice, making medical decisions based on tests like those from 23andMe carries risk. Current understanding of genetics’ role in disease is far from complete, often not conclusive, and potentially misunderstood, says George J. Annas, chair of the department of health law, bioethics, and human rights at Boston University’s Schools of Public Health, Medicine, and Law.
Nevertheless, some experts say it’s up to consumers to decide how to use the data, and that access to genetic data and information on what it might mean is a basic right. “It’s no different from a family history,” says Lawrence Lesko, a 20-year FDA veteran who is now director of the University of Florida’s Center for Pharmacometrics and Systems Pharmacology.
At times 23andMe has hurt its own cause. A year after submitting applications for seven health reports in 2012, it stopped communicating with the FDA for six months, according to the agency, at the very time it prepared to launch a television ad campaign. That’s what prompted the FDA to clamp down.
The company, which has raised $126 million in funding, needs to fix its FDA issue if it is to meet its goal of creating a database of as many as tens of millions of genetic profiles, up from 700,000-plus today. Coupling those profiles with data from customer health surveys could entice pharmaceutical and medical-device companies to pay 23andMe for the chance to look for connections among gene variations, diseases, and drug response at a small fraction of the cost and time needed to do traditional clinical trials. Genentech has already paid the company to help it recruit breast cancer patients who had taken its drug Avastin in order to assess their response. The strategy of crowdsourcing big data echoes that of one of 23andMe’s big investors: Google, which was cofounded by Wojcicki’s husband, Sergey Brin. (They separated last year.)
Even before last November’s letter, the challenges of meeting regulatory requirements had already prompted 23andMe’s U.S. rivals to exit the market for direct-to-consumer genetic tests. Other testing firms bypass the FDA by selling through doctors. If 23andMe can’t get FDA approval on at least some health reports, that could spell the end of selling genetic information directly to consumers within the U.S., says former FDA counsel Patricia Zettler, a research fellow at Stanford Law School.
To win over the FDA, Wojcicki is first shepherding one specific health report—for Bloom syndrome, an inherited disorder that often results in deadly cancer by the mid-20s—through the approval process for medical devices. If it works, it would provide a template. But since 23andMe originally offered more than 200 health reports, it’s not yet clear that this process will be enough to attract large numbers of new customers.
Zettler is one of many observers who think 23andMe will eventually get through the FDA process, at least on individual health reports. But the FDA is contemplating new hurdles, like requiring that claims be reviewed by an expert panel.
Wojcicki wonders if too many limitations will simply spur people to take their genetic data to Canada or China for interpretation. “How do you regulate information?” she asks. “I’m not sure you can hold it back.”
The inside story of how ChatGPT was built from the people who made it
Exclusive conversations that take us behind the scenes of a cultural phenomenon.
ChatGPT is about to revolutionize the economy. We need to decide what that looks like.
New large language models will transform many jobs. Whether they will lead to widespread prosperity or not is up to us.
Sam Altman invested $180 million into a company trying to delay death
Can anti-aging breakthroughs add 10 healthy years to the human life span? The CEO of OpenAI is paying to find out.
GPT-4 is bigger and better than ChatGPT—but OpenAI won’t say why
We got a first look at the much-anticipated big new language model from OpenAI. But this time how it works is even more deeply under wraps.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.