Stool DNA Testing for Colorectal Cancer Shows Promise

Setting and participants. National Institutes of Health– supported trial with industry involvement that included 4482 adults in communities surrounding 22 participating regional and academic health systems in the United States. Participants were asymptomatic, aged 50 to 80 years, and deemed to be at average risk for colorectal cancer. Exclusion criteria included structural colorectal evaluation within 10 years, high risk for or previous colorectal cancer, fecal occult blood testing within the past year, previous colorectal resection, aerodigestive cancer within 5 years, overt rectal bleeding, inability to stop use of nonsteroidal anti-inflammatory drugs or anticoagulants, recent chemotherapy, coagulopathy, inflammatory bowel disease, > 2 first-degree relatives with colorectal cancer, and contraindication to colonoscopy. All participants underwent colonoscopy after stool testing. Participants were given 2 types of fecal occult blood stool cards to use (Hemoccult and HemoccultSensa, Beckman Coulter, Fullerton, CA) and were also asked to collect 3 whole stools that were express-shipped to the main study center on ice. Stool cards were analyzed using rigorous standardized protocols, and whole stools were sent to an industry laboratory for stool DNA analysis. Two stool DNA tests were performed. The first test (SDT-1) included a panel of 21 tumor-specific point mutations, the microsatellite instability marker BAT-26, and long DNA (marker for delayed apoptosis). Because the manufacturer of SDT-1 altered the assay midway through the study, an unplanned interim analysis was conducted in the first 2497 patients who received the test. On the basis of these results, the investigators switched to another stool DNA test (SDT-2), which consisted of 3 tumor-specific markers (APC, K-ras, and vimentin methylation) broadly associated with colorectal cancers and adenomas.