New colorectal cancer screening techniques promise greater detection

Novel colorectal cancer screening techniques stand to improve detection of early-stage tumors, according to a Professor of Medicine, Mayo Clinic College of Medicine, Rochester, Minn. He told attendees of the 2004 Digestive Disease Week (DDW) meeting that more patients may agree to be screened because stool DNA testing and virtual colonoscopy are as accurate as, but less invasive than, conventional tests – resulting in a higher rate of detected cancers. At the population level, cancer detection is not equivalent to a test’s sensitivity alone, but is rather the product of sensitivity, compliance and access rates, explained by doctor. The emerging screening technologies may improve neoplasm detection by enhancing each of these factors. These technologies include stool DNA testing and computed tomography (CT) colonography.

Stool DNA testing

Patients who undergo stool DNA testing do not have to restrict their medications, change their diet or prep their bowels prior to testing. In fact, stool samples can be mailed directly to the laboratory. Patient access to colorectal cancer screening is theoretically unlimited if samples can be mailed to a central laboratory for testing. This capacity could potentially increase neoplasm detection across the entire population. According to doctor, there is a strong biological rationale for conducting stool-based BDA tests, as neoplasms regularly exfoliate cells into the lumen. However, because of the variety of cancer markers, researchers are faced with the challenge of developing stool-based DNA tests with specific markers. To date, no single marker has yielded 100 percent sensitivity, but several combinations in prototype assay panels have been associated with high neoplasm detection rates. The most extensively studied assay – and the only one commercially available at the present time (PreGen-Plus, EXACT Sciences Corp./LabCorp) — includes combinations of high-frequency point mutations on the p53, APC and K-ras genes as well as Bat-26 microsatellite instability and long DNA.

In studies of selected patients, this combination was found to detect 70 percent to 90 percent of colorectal cancers and 30 percent to 70 percent of large adenomas, with a specificity of 93 percent to 100 percent. The accuracy of this marker combination to detect colorectal neoplasms from a representative, average-risk population is currently being investigated in two large studies. Doctor reported preliminary findings at the DDW meeting, which revealed a high level of sensitivity, with a detection rate of 57 percent for curable-stage colorectal cancers, compared to 13 percent with a fecal occult blood testing method (P<0.003). The final results of these studies, will be important in evaluating the performance of PreGen-Plus. False-negatives do occur with this approach (stool DNA testing) and may be due to colonoscopy misses, to supracolonic neoplasms and, theoretically, to so-called field changes. He noted that mucosa adjacent to neoplasms may appear normal but, upon biopsy, revealed DNA alterations. In light of this, he called for clinical research to establish follow-up algorithms for patients with positive stool DNA tests and negative colonoscopy results. Preliminary data from other ongoing research suggest the potential of stool-based DNA testing to detect lung, esophagus, stomach, pancreas and bile duct cancers. As well, speculated, the technology may potentially be used to screen for dysplasia in patients with inflammatory bowel disease or other at risk conditions such as primary sclerosing cholangitis and Barrett’s esophagus.

CT colonography

According to, CT colonography has evolved dramatically since it was originally described a decade ago. Rigorous interrogation is performed on the colon image rather than on the patient, no sedation is required, and risks are close to zero. Moreover, as with stool DNA testing, the capacity to screen extracolonic organs potentially expands its value beyond colorectal cancer screening alone. He cited a study conducted by the National Naval Medical Center, Bethesda, Md., that employed a primary three-dimensional (3-D) fly-through technique. Study findings revealed that CT colonography detected 94 percent of adenomas >1 cm in 1,233 asymptomatic average-risk adults. The test’s specificity in that study was an equally impressive 96 percent. However, he said these results required corroboration by other investigators, particularly since most centers with CT colonography still use a two-dimensional (2-D) display as the initial test and reserve 3-D images for more rigorous examination of suspicious lesions.

Nonetheless, added, 2-D CT colonography has proven itself more accurate in detecting polyps >1 cm than barium enema (81 percent vs. 45 percent) and flexible sigmoidoscopy (60 percent vs. 39 percent) in direct comparisons reported to date. Doctor said that the widespread use of CT colonography and stool-based DNA testing may be hindered because neither procedure is currently covered by Medicare or third-party payers for the indication of colorectal cancer screen. Furthermore, neither the American Gastroenterological Association nor the American Cancer Society officially endorses using these techniques. However, in view of the most recent clinical data and as future refinements are reported, these policies will certainly be revisited. Both stool DNA testing and CT colonography are evolving technologies that will continue to get better. A gastroenterologist in private practice in Chicago, told that he was surprised at how effective stool DNA testing is in screening for colorectal cancer. Stool DNA testing is talked about a lot, but people don’t really know how sensitive or specific it is. He was struck by the strength of these data.