The Washington PostDemocracy Dies in Darkness

There’s a new sheriff in town in Silicon Valley — the FDA

A 3-D-rendered illustration of cancer cells. (iStockphoto)

SAN FRANCISCO — Helmy Eltoukhy’s company is on a roll. The start-up is a leading contender in the crowded field of firms working on “liquid biopsy” tests that aim to be able to tell in a single blood draw whether a person has cancer.  

Venture investors are backing Guardant Health to the tune of nearly $200 million. Leading medical centers are testing its technology. And earlier this month, it presented promising data on how well its screening tool, which works by scanning for tiny DNA fragments shed by dying tumor cells, worked on an initial group of 10,000 patients with late-stage cancers.

Just one thing is holding the company back: Guardant Health has yet to get approval from the Food and Drug Administration.

As a tidal wave of new health-related gadgets, apps and tests hits the market, government agencies like the FDA, Federal Trade Commission and others are showing up in Silicon Valley like they’ve never done before. They have slapped companies such as Theranos, 23andMe, Lumosity and Pathway Genomics with warning letters and fines and raised questions about products that regulators believe promise more than they can deliver.

More regulatory scrutiny is probably coming. Venture capital investments in life sciences hit a record high in 2015, with $10.1 billion invested in 783 deals, and total start-up funding is approaching levels of the last dot-com bubble — a development that has some industry observers worried that pseudoscience is being confused with innovation.

But even as some companies push back against federal agencies’ reach — contesting which rules, if any, apply to their work — there’s now recognition that the government can be a powerful ally rather than a brake on progress. And its stamp of approval can take firms from being worth multimillions to multibillions.

Jeff Huber, a former senior Google executive who is now chief executive of Grail, which is also working on liquid biopsy tests, said his company reached out to FDA officials while still in the research-and-development phase and is “carefully considering their input as to the right approach.”

“A core part of our reputation and brand is the scientific rigor we’re putting behind this,” Huber said. That includes designing rigorous, large-scale clinical trials targeted to begin in 2017. (Grail’s early backers include Bill Gates and Jeffrey P. Bezos, who owns The Washington Post.)

Putting a product on the market without consulting with the FDA is risky, but companies doing so have a legal argument for their move. Guardant Health contends that the liquid biopsy test it began selling to oncologists in 2014 falls outside the agency’s purview because the end result is data about the composition of a person’s blood, not a definitive diagnosis. Even so, its founder said he proactively contacted a local FDA office a few months ago to express his eagerness to work with the agency in the future.

“It’s the patient at the end of the day who is the person we’re trying to help,” said Eltoukhy, who has a PhD in electrical engineering from Stanford University. “We’re not doing them any justice or any benefits by putting a technology that’s not ready for prime time into the market.”

Given the super-hot field, the pressure to be first remains intense, however. Nearly 40 companies are working there, according to a research report by Piper Jaffray analysts William Quirk and Alexander Nowak, who valued the U.S. market alone at $32.6 billion a year. Liquid biopsies, they noted, could “revolutionize” cancer, transplant and prenatal care.

Being tested for cancer today often means having a slice of tissue cut out — which can be painful and dangerous — and waiting days or even weeks for the results to come back. The promise of liquid biopsies is that the same information might be available based on an extremely low-risk blood draw that takes mere minutes. 

The concept is so simple and potentially inexpensive that it could upend practically everything about the disease. Healthy people would be able to walk into their doctor’s office for an annual checkup and know whether they had cancer well before it becomes life-threatening. Doctors would be able to track their patients’ responses to therapies almost in real time by studying which cancerous mutations are in the mix and in what concentration.

Yet the science behind liquid biopsies is incredibly tricky because of how cancerous DNA is obscured by healthy DNA in the blood, with a sophisticated combination of molecular biology, informatics, genetic sequencing and other disciplines required to reveal it. Eltoukhy likens the challenge to trying to find the fine details in the “snow” on a TV screen that’s relying on an old rabbit-ears antenna.

Guardant Health has had early success with its screening tool in late-stage cancer patients, and about 2,000 oncologists are using it to help create personalized therapies when first-line treatments have failed. The company says its test is the most comprehensive on the market, examining more than 150,000 places in the genome compared with the half-dozen or so that many competitors review. It has partnered with the National Cancer Institute as well as some major pharmaceutical companies for further investigations. 

In the abstract presented at the American Association for Cancer Research this month — though not peer-reviewed, it is being submitted to a journal, according to Eltoukhy — the company said the test was as “highly accurate” as a surgical biopsy in detecting cancerous DNA. Lung, gastrointestinal and breast cancers were most commonly found.

Still, the real game-changer would be a test that can pinpoint cancer in its earliest stages, when the amount of DNA in the blood is mind-bogglingly small.

Pathway Genomics claimed to have done just that when it launched a direct-to-consumer $699 screening tool last year for the early detection of up to 10 different cancer types among people at high risk but without symptoms. The FDA disagreed. In a letter in  September, the agency warned that it had not found any evidence to support the claims and that the product could “harm the public health.”

The company declined to talk about its discussions with regulators, but the FDA said Pathway Genomics has since made the test available only by prescription and “also limited the claims of what their test does to ones that seem to be more in line with the current available evidence.”

Separately, the company recently paid the Justice Department $4 million to settle allegations that it offered kickbacks to physicians to refer patients to its service. It admitted no wrongdoing.

The FDA’s mandate was written decades before anyone could imagine these new technologies, so the extent of its powers continues to be much debated.  The agency said in a statement Thursday that it considers all diagnostic in vitro tests — which include blood tests — to be medical devices but that it has not always “exercised enforcement discretion over laboratory-developed tests except for those being marketed direct-to-consumers.”

“All devices, including diagnostic tests, are subject to FDA regulation,” the statement noted.

Guardant Health is far from the only firm to claim that its blood tests — or,  in the case of 23andMe, spit tests — fall outside the FDA's regulatory bounds because they’re simply informational and do not definitively detect diseases, conditions or infections that lead to treatment decisions. Guardant's labs do have certification under the Clinical Laboratory Improvement Amendments of 1988 (CLIA) which sets quality control and other standards.

But growing evidence suggests the tide may be shifting on this view.

Massimo Cristofanilli, an oncologist at the Robert H. Lurie Comprehensive Cancer Center of Northwestern University, has used Guardant Health’s test on more than 200 breast cancer patients with late-stage disease and said it has been helpful in about 6o to 70 percent of cases to determine a next course of treatment. He sees FDA approval as critical for widespread adoption.

“Physicians, especially community physicians, won’t feel comfortable until they have more of a guarantee that the tests are doing what they are supposed to be doing,” he said.

Otis Brawley, chief medical officer for the American Cancer Society, thinks FDA approval isn’t the panacea some believe it to be. Even if a liquid biopsy is rigorously validated, he said, it may be able to identify only the presence of a cancer, not where it is or whether it will have an impact on a person’s health. It’s possible that some people could have cancer, per a DNA test, that will never grow into something that will hurt them.

“The purpose of the screening test is more than to find disease. The purpose is to find disease in a situation such that medical treatment can prevent death,” Brawley said. “There’s a bunch of folks in the corporate community who don’t understand that fine detail.”

This post has been updated.

Read more:

IN DEPTH: Sean Parker, Silicon Valley’s bad boy genius, wants to kick the *!$% out of cancer

$250 million, 300 scientists and 40 labs: Sean Parker’s revolutionary project to ‘solve’ cancer

Brain-zapping gadgets promise to make you a better you — smarter, stronger, even happier

Loneliness can literally hurt your heart, scientists say

Your DNA might influence when you lose your virginity, first have kids

Like our Health & Wellness page on Facebook for more science news about the ins and outs of the human body and mind, essays and advice. You can sign up here for our newsletter.