Colorectal Cancer Screening (PDQ®): Screening - Health Professional Information [NCI] - Evidence of Benefit
Table 3. Randomized Controlled Screening Trials: Fecal Occult Blood Testing continued...
Virtually all screening studies using these types of sigmoidoscopes have demonstrated an increase in the proportion of early cases and a corresponding increase in survival compared with cases diagnosed in a nonscreening environment. Most of these studies, however, lack appropriate comparison groups, and their interpretation is unclear because of screening biases.
The first incidence and mortality results from a randomized trial of sigmoidoscopy were reported in the Norwegian Colorectal Cancer Prevention (NORCCAP) trial. This trial randomly assigned 41,913 men and women aged 55 to 64 years to a usual-care control group and 13,823 individuals to a screening group, 6,915 of whom received one-time FS and 6,908 of whom received both FS and iFOBT. Screening was conducted from January 1999 to December 2000. Follow-up was through national registries in Norway and ended December 31, 2006 for incidence and December 31, 2005 for mortality. The primary endpoint was incidence of CRC after 5, 10, and 15 years of follow-up based on an intention-to-screen analysis. The two randomly assigned groups had the same median age of 59 years and equal gender distribution. The attendance rate for screening was 65%. Mean insertion depth of the endoscope was 48.9 cm for men and 44.0 cm for women, with no severe complications. A neoplastic lesion was found in 19% of people screened, and 5% had a high-risk adenoma or invasive cancer. There were 33 prevalent CRCs, 17 in the subgroup invited for sigmoidoscopy only and 16 in the subgroup invited to both tests. Compliance for colonoscopy work-up was 97%. There was no difference in the cumulative hazard of CRC between the screened and control groups; 134.5 versus 131.9 cases per 100,000 person years, after a median follow-up of 7 years. The hazard ratio (HR) for CRC mortality was 0.73 (95% CI, 0.47–1.13) and for rectosigmoidal cancer mortality was 0.63 (95% CI, 0.34–1.18) after a median follow-up of 6 years. The HR for all-cause mortality was 1.02 (95% CI, 0.98–1.07). Additional follow-up is planned.
In the Telemark Study in Norway, 400 people aged 50 to 59 years were offered FS in 1983 and, if polyps were found, a colonoscopy was available to them at that time, and again 2 years later, and also 6 years later (the intervention was more aggressive than the one-time sigmoidoscopy currently being tested in the United Kingdom). A nonrandomized control group was selected that did not have a baseline examination. In 1996, all persons were invited to have a colonoscopy. The result was that ten persons in the control group and two in the screening group developed CRC (RR = 0.2; 95% CI, 0.03–0.95). The screening group, however, had a higher overall mortality compared with the control group.