Kuipers says the 77 percent accuracy in detecting cancer-containing samples is “a good result.” In comparison, the fecal-immunological screening method that he has been researching is around 60 percent accurate. He notes, however, that the specificity of the blood test–its ability to correctly identify healthy patients–will need to improve. “In practice, everyone over 55 would be screened, perhaps every two years,” he says. “That’s millions of people. So, if you had more than 5 percent false-positive rates, the number of follow-up colonoscopies you’d need to do would become too great.”
Kuipers says that the specificity of the test needs to be at least 95 percent for it to be used in colorectal screening and that a large-scale evaluation will be vital.
With this in mind, Louwagie and colleagues are enrolling people in a prospective colorectal screening study at several German colonoscopy centers. “We plan to complete enrollment of 7,000 people by the end of 2009,” he says.
The trials should also shed more light on how effective the test is at detecting the very earliest stages of colorectal cancer. Such a gene test will not be able to spot precancerous polyps. But it could be particularly effective if it can detect stage-one and stage-two colorectal cancers, which are almost always curable with surgery.
A new paper by Kuipers, due to appear in Journal of the National Cancer Institute, will provide new evidence that colorectal screening can ultimately save health services money, he says. But he believes that the most important measure will be a reduction in the number of colon cancer deaths. Kuipers notes that older, repeat-stool type testing, which was considered ineffective and not very sensitive, has been shown to have cut colorectal cancer deaths by 15 percent. He says that a reasonably sensitive and simple test with higher compliance levels could prevent many more colon cancer deaths.