Exact Sciences’ Colon Cancer Test Clears Key Hurdle, FDA Next
Wall Street was pleased today with the newly published results of a pivotal study of Exact Sciences’ non-invasive screening test for colorectal cancer. Next step for the Madison, WI, company: FDA approval.
Exact (NASDAQ: EXAS) announced top-line data of its 10,000-patient study in April 2013, but the full results were officially published Wednesday in a peer-reviewed study in the New England Journal of Medicine (NEJM).
The “Deep-C” study, which took place at 90 healthcare centers throughout the U.S. and Canada, found that Exact’s Cologuard—a stool DNA-based screening test—was much more reliable at detecting colorectal cancer than the fecal immunochemical test (FIT), a commonly used, non-invasive colorectal cancer screener. But Cologuard also resulted in more false positives than FIT.
Exact intended to announce the details of the NEJM study after stock market close today, but a study collaborator accidentally released the data on Tuesday. That inadvertent misstep, reportedly by the Mayo Clinic, drove Exact’s stock price higher during regular and after-hours trading Tuesday, according to SeekingAlpha.com. The next day, Exact’s stock opened at $15.50 per share and rose as high as $15.60 per share, up from the $14.42 per share close on Tuesday. It closed at $14.45 per share today.
Next week, Exact faces an even more important test: An FDA panel is scheduled to review the company’s pre-market approval application.
“We are quite pleased with the results of the Deep-C study, and we look forward to discussing it with our [FDA] advisory committee next week,” Exact president and CEO Kevin Conroy said today in a conference call with investor analysts.
A few key data points from the study:
—Cologuard detected colorectal cancer 92 percent of the time, compared with 74 percent for FIT.
—Cologuard detected advanced pre-cancerous lesions 42 percent of the time, versus 24 percent for FIT.
—Cologuard detected polyps with high-grade dysplasia 69 percent of the time, compared with 46 percent for FIT.
—About 10 percent of study patients received a positive Cologuard result, but no cancer was subsequently found with a colonoscopy. That false positive rate was higher than FIT’s, which was about 4 percent.
Exact Sciences officials and study co-authors don’t appear to be overly concerned with the higher incidence of false positives with Cologuard.
“If there are more false positives, all that means is people are sent to get a colonoscopy, and they’re supposed to be sent for a colonoscopy anyway,” says study co-author Dr. Steven Itzkowitz, a professor of medicine and director of the gastroenterology fellowship program at the Icahn School of Medicine at Mount Sinai in New York.
The rate of colorectal cancer deaths has decreased 46 percent from its peak, but a third of Americans aren’t keeping up with their screening tests, an NEJM editorial said today. Exact and the broader medical community must still figure out where Cologuard, assuming it hits the market, fits on the colorectal cancer “screening menu.”
“We kind of see it as complementary to colonoscopy at this point,” Itzkowitz says. “I don’t think anybody is saying it will replace the colonoscopy.”
The NEJM editorial said that the Cologuard test is “clearly an improvement over its predecessors,” but with a few caveats. Among them: the number of participants excluded from the study due to sample collection issues was 6 percent higher in the Cologuard group than the FIT group, the editorial said.
“Given that colorectal cancer was detected in nearly 1 of 154 participants on colonoscopy, it is possible that four cancers would have been missed simply because of the complexity of the test,” NEJM editors wrote.
However, Conroy said it was important to note that 333 of the 689 Cologuard tests that were excluded were also missing a FIT test.
“That may not have come across as clear in the [NEJM] editorial as we would’ve liked, but that data will be available as we move forward,” Conroy said on the call.
Some fecal samples were excluded because they were overweight or there was leakage from container lids not being sealed properly, Conroy said on the call. The company has made modifications to its collection containers that should partly alleviate sample error, he said.
Participants whose samples were excluded from the study went on to receive a colonoscopy, so researchers couldn’t take another sample from them due to the study’s protocols.
“In a normal clinical setting, you would be able to go back to those patients and provide another sample,” Conroy said.
Cologuard’s recommended testing intervals, cost, and patient acceptance rates are yet to be determined and were outside the scope of the study.
During the call, Conroy declined to address questions about commercialization of Cologuard and testing intervals, citing the FDA review next week.
If Cologuard secures FDA approval and Medicare reimbursement—which Exact is seeking simultaneously—will millions of patients over age 50 opt to ship their stool samples to get screened for colon cancer? That remains to be seen, but more than 90 percent of participants in a similar study a decade ago of an earlier type of stool DNA test said they thought the test was relatively easy to perform and they would do it again, Itzkowitz says.
“I’m sure there still will be people that don’t particularly want to do a stool-based test, but from what we know so far, it’s been very well-accepted by patients,” Itzkowitz says. “The advantage of this test is you do it in the privacy of your own home.”