Still, the results fell short of investor expectations and even those of the company that developed the test, the Exact Sciences Corporation, sending its shares down about 20 percent in afternoon trading on Thursday.
In its news release about the study Thursday morning, Exact Sciences said its test detected 92 percent of the cancers picked up by colonoscopy, and 42 percent of potentially precancerous polyps. It had a false positive rate of 13 percent.
The test looks for alterations in human DNA found in a stool sample. The company contends that people will not find it off-putting to deposit a sample of their stool in the company’s collection apparatus and mail it to a laboratory.
The new test, called Cologuard, would not replace colonoscopy. Colonoscopy remains the gold standard for colorectal screening, in part because any polyps detected can also be removed during a colonoscopy, possibly preventing cancer.
But about half of people over 50, the recommended age to start screening for colorectal cancer, are either not adequately screened or not screened at all, in part because colonoscopy is invasive, uncomfortable, expensive and time-consuming.
Exact Sciences says the noninvasive test could allow more people to be screened, and those with a positive result could then get a colonoscopy.
“For the first time noninvasively, we can detect reliably precancerous polyps,” Kevin T. Conroy, the company’s chief executive, told analysts on Thursday. Exact Sciences, which is based in Madison, Wis., said it would soon complete its application to the Food and Drug Administration seeking approval of the Cologuard test.
There were about 143,000 new cases of colorectal cancer and 52,000 deaths in the United States last year, making it the second-leading cause of cancer death behind lung cancer, according to the American Cancer Society.
Dr. Deborah A. Fisher, an associate professor of medicine at Duke University, who was not involved in the study, said the Cologuard test appeared to be a viable option, but only one of several.
“I don’t think this is the holy grail,” said Dr. Fisher, a gastroenterologist who is a consultant to Epigenomics, a company developing a test that could compete with Cologuard. She said it was too soon to tell if the Cologuard test would actually increase the number of people being screened and said there was no data showing that its use actually prevented cancer deaths.
Dr. Fisher said existing noninvasive tests that look for blood in the stool can detect around 80 percent of cancers and 20 to 40 percent of polyps. These tests cost about $25, she said, while the Cologuard test is expected to cost a few hundred dollars.
Exact Sciences said participants in its study also received a stool blood test, known as a fecal immunochemical test. The Cologuard test, the company said, proved to be better than that test in detecting polyps and roughly equivalent in detecting cancer.
The company did not release the data, however, saying it would eventually be published in a peer-reviewed medical journal. The study involved about 10,000 people with an average risk of colorectal cancer. They were screened at 90 sites in the United States and Canada.
Mr. Conroy of Exact Sciences said the 42 percent success rate in detecting polyps, while below the company’s goal of 50 percent, was still a powerful result. He said the test caught 66 percent of polyps larger than two centimeters, which were more likely to become cancerous than smaller ones.
He also said that since colon cancer developed over many years, if people took the Cologuard test every three years, most polyps would be detected before they could cause problems. Such repeated testing is the reason the Pap test has greatly reduced deaths from cervical cancer, even though Pap tests miss precancerous lesions as much as half of the time.
But Dr. Fisher of Duke said yearly use of the less expensive fecal immunochemical test would also eventually detect many polyps.
The false positive rate of 13 percent for the Cologuard test was also higher than Exact Science’s goal of 10 percent. A false positive usually would mean that a person would get a colonoscopy that might not be needed.
The Cologuard test looks for mutations and chemical changes in DNA indicative of cancer. It takes advantage of the fact that cells lining the colon are continually sloughed off and broken apart, releasing their DNA.
But detecting that DNA in the stool is extremely difficult. Virtually all the DNA in the stool comes from bacteria, said Dr. David A. Ahlquist, a professor at the Mayo Clinic who helped develop the Cologuard test. Only 0.01 percent is the person’s own DNA, and of that, only a tiny fraction would be from cancerous cells, he said.
There have been some earlier versions of Exact Science’s test that did not work well.
Epigenomics, based in Germany, recently applied for F.D.A. approval of a screening test that looks for a chemical change to one gene.
Epigenomics has said its test, called Epi proColon, detected 71 percent of cancers in its large clinical trial, with a false positive rate of 19 percent. The test is not aimed at detecting polyps.
That would make it less accurate than the Cologuard test. It is a blood test, however, not a stool test.
“People are very comfortable with blood-based tests,” said Karen A. Heichman, vice president for oncology technology development and licensing at ARUP Laboratories in Salt Lake City, which offers its own version of the test under license from Epigenomics.
Article source: http://www.nytimes.com/2013/04/19/business/new-test-detects-most-colorectal-cancers-study-finds.html?partner=rss&emc=rss