| MAYBE YOU DON’T NEED A COLONOSCOPY
Exact sciences has a pretty good test for cancer. Gastroenterologists are unenthusiastic.
Exact Sciences has a pretty good test for cancer. Gastroenterologists are unenthusiastic.
Three years ago Kevin Conroy was standing on Ireland’s Cliffs of Moher, gazing at the Atlantic Ocean 500 feet below, when he was blindsided by a phone call about his company’s colon cancer test.
Since 2009 Conroy has been running Exact Sciences, a firm trying to sell a noninvasive would-be alternative to a colonoscopy. The caller informed him that the influential U.S. Preventive Services Task Force had declined to recommend the Exact test as a cancer-screening tool, which meant it would not be covered by health plans under the Affordable Care Act. “Our stock took a massive hit,” says Conroy, 52. “It was surprising. In many ways energizing, though.”
Energized, Conroy sent in more data. In 2016 the task force gave the Exact test a green light. Exact’s share price is now more than triple what it was be- fore the task force brush-off.
There are three main ways to detect colon cancer. A colonoscopy is accurate but expensive ($2,200 on average). A fecal immunochemical test, which looks for hidden blood in a stool sample, costs about $60 but is more likely to miss cancer and must be taken every year. Then there is Exact’s Cologuard for $649.
Like the immunochemical test, Cologuard uses a stool sample collected by the patient and checks for hemoglobin. It also looks for an assortment of aberrant DNA sequences that are likely to occur in cancerous and precancerous cells. Result: high enough sensitivity that patients can take the test only once every three years.
Cologuard’s false-negative rate is 8% and its falsepositive rate is 13%. Back-of-the-envelope math: Among 100,000 people who take the test once, 90 might have colorectal cancer, and Cologuard will catch all but 7 of those cases. It will also send 13,000 healthy people off with worries and instructions to follow up with a colonoscopy. But then Cologuard has spared 87,000 customers the unpleasantness of a colonoscopy prep day.
Catching colon cancer early matters. The fiveyear survival rate for patients diagnosed at stage 1 is 92%, but for stage 4 that drops to just 11%, according to the National Cancer Institute. Colorectal cancer kills more than 50,000 Americans a year. And still only about 65% of those who should get screened for
colon cancer do. Maybe Cologuard can change that.
Conroy was approached about the Exact Sciences job in 2009 after he sold his previous diagnostics company, Third Wave Technologies, to Hologic for $580 million. Exact had “no product, no meaningful intellectual property and no scientist,” he says. The company had been studying DNA markers since 1995 but didn’t have a marketable test.
Two things persuaded Conroy to sign up. One was that the company would move from Marlborough, Massachusetts, to Madison, Wisconsin, where he lives. The other was a compelling argument from a Mayo Clinic researcher, Dr. David Ahlquist, that adding different DNA markers to the test panel could dramatically improve Cologuard’s sensitivity. Ahlquist has a financial interest in Exact.
Conroy hatched an unprecedented plan. Until then, companies had done one clinical trial to get approval from the Food & Drug Administration and a second to convince the Centers for Medicare & Medicaid Services to pay for the test. This would have been so expensive and time-consuming, Conroy says, that Exact would have gone out of business. So he persuaded the bureaucrats to let Exact do one.
The trial, using samples from 10,000 patients, was enough to get FDA and then Medicare approval in 2014. The American Cancer Society recommended the test shortly thereafter, and the nod from the Preventive Services Task Force followed in 2016. (Exact has donated $15,000 since 2016 to the American Cancer Society.)
But there is yet one more gauntlet for Exact to run: the Multi-Society Task Force, which represents gastroenterologists. (Exact has donated over $30,000 to a group involved in the task force.) That panel gives Cologuard a “tier 2” recommendation, compared with “tier 1” for the immunochemical test and the colonoscopy. The group wants long-term studies that show Cologuard saves lives. It also says a Cologuard test every three years costs more than the cheaper stool test every year.
That may be so, but in fact many people go directly to colonoscopies without any prescreening via stool samples, and that’s a very expensive way to prevent cancer deaths.
Plenty of others get no testing. Susan Pickering of Milwaukee avoided having a colonoscopy for 12 years until she was finally persuaded by a friend, at age 62, to use the Cologuard kit her doctor had prescribed. After months of letting it sit around her house, she ended up testing positive and was diagnosed with stage 2 colon cancer in 2017. After surgery and chemotherapy, she has no detectable cancer.
Says Mayo’s Ahlquist: “As a gastroenterologist I’ve been dismayed, reflecting my feeling about how my GI colleagues across the country have responded defensively to the emergence of Cologuard. They’ve looked at it from the onset as a threat rather than as a tool that their patients can use regularly.”
There’s also an insurance problem. Because the Affordable Care Act mandates coverage of colon cancer screening, 85% of patients have no out-of-pocket cost for Cologuard. But those who test positive then need a colonoscopy. And this second, more expensive procedure is not covered by some plans. With big marketing expenses (it recently hired singer Harry Connick Jr. as a spokesperson), Exact lost $39 million on revenue of $90 million in the first quarter of 2018. Still, its $7.3 billion market value means there are a lot of believers in both Cologuard and whatever comes next.
Conroy aims to use his $1 billion cash pile to try to develop a blood test to detect diseases like lung cancer and liver cancer. Exact has competition from well-funded startups like Grail and Guardant Health. “This is transformational. If you can do it, you must do it,” says Ahlquist. “There’s enough convergence of technology now where this is realizable. Is it going to be easy? Absolutely not. It’s going to be hard, but we can get through this.”