Background: Endoscopy is believed to be more sensitive than fecal occult blood testing for detecting adenomatous polyps, the precursor lesions of colorectal cancer. However, observational studies have raised doubts as to whether screening with flexible sigmoidoscopy can reduce incidence and mortality associated with colorectal cancer. Schoen and colleagues conducted a trial to compare flexible sigmoidoscopy with usual care on the incidence of distal and proximal colorectal cancer and related mortality.
The Study: The Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial evaluated flexible sigmoidoscopy as a screening tool for colorectal cancer in 154,900 men and women 55 to 74 years of age. Participants were randomized to receive usual care or offered two flexible sigmoidoscopy screenings three to five years apart. A screening was considered positive if a polyp or mass was identified, in which case patients were referred to their primary care physician for further treatment. Exclusion criteria included a history of prostate, lung, colorectal, or ovarian cancer; ongoing cancer treatment; or a lower endoscopic procedure within the previous three years. The primary end point was death from colorectal cancer, with secondary end points of colorectal cancer incidence, cancer stage, survival, adverse events of screening, and all-cause mortality.
Results: Over a mean follow-up time of 11 years, 86.6 percent (n = 67,071) of participants in the intervention group underwent at least one sigmoidoscopy screening, and 50.9 percent (n = 39,440) underwent two procedures. At least one screening was positive for a polyp or mass in 28.5 percent of participants in the intervention group, and 22 percent underwent colonoscopy as a result of sigmoidoscopy screening.
The overall incidence of colorectal cancer was significantly lower in the intervention group compared with the control group (1,012 versus 1,287 cases, respectively; relative risk [RR] = 0.79; P < .001). This included the incidence of distal (below the splenic flexure; RR = 0.71; P < .001) and proximal (from the cecum through the transverse colon; RR = 0.86; P = .01) malignancy. Colorectal cancer–related mortality was reduced overall with sigmoidoscopy (252 versus 341 deaths in the intervention and usual care groups, respectively; RR = 0.74; P < .001). However, this was because of reduced mortality from distal colorectal cancer (RR = 0.50; P < .001); no mortality reduction was seen with proximal colorectal cancer (RR = 0.97; P = .81). Cancers identified in the screening group were more likely to be early stage (I or II) compared with those in the usual care group (75.4 versus 50.9 percent, respectively; P < .001). The number needed to invite for screening to prevent one case of colorectal cancer was 282, and the number needed to invite for screening to prevent one colorectal cancer–related death was 871.
Conclusion: Compared with usual care, flexible sigmoidoscopy screening with follow-up colonoscopy in most patients with abnormal findings significantly reduced the incidence of distal and proximal colorectal cancers. Mortality from distal colorectal cancer was reduced by 50 percent, although no reduction in mortality related to proximal colorectal cancer was observed.