« Back

Single flexible sigmoidoscopy screening reduces rates of CRC, CRC-related mortality in older adults, trial data show

Wednesday, May 19 2010 | Comments

---

What's This? Among adults aged 55 to 64 years, a single flexible sigmoidoscopy screening might reduce the incidence of colorectal cancer (CRC) by more than 30% and death resulting from CRC by more than 40%, according to findings from a randomized, multicenter trial conducted in the United Kingdom.

The trial included 170,432 men and women aged 55 to 64 years, all of whom indicated on a previous questionnaire that they would accept an invitation for flexible sigmoidoscopy screening. The participants were assigned to an intervention group (n=57,237) or a control group (113,195). Those in the intervention group were offered flexible sigmoidoscopy with polypectomy for small polyps and referral for colonoscopy if polyps meeting high-risk criteria were identified, whereas the participants in the control group were not contacted.

The primary outcomes of the trial were the incidence of colorectal cancer, including prevalent cases detected at screening, and CRC-related death.

Results of an intent-to-treat (ITT) analysis suggested that--during a median follow-up of 11.2 years--flexible sigmoidoscopy screening reduced the incidence of CRC by 23% (hazard ratio [HR], 0.77; 95% CI, 0.70-0.84) and CRC-related mortality by 31% (HR, 0.69; 95% CI, 0.59-0.82).

In a per protocol analysis, researchers adjusted for noncompliance with screening, noting that only 71% of the intervention group underwent flexible sigmoidoscopy. Findings from this analysis suggested that screening reduced the incidence of CRC during follow-up by 33% (HR, 0.67; 95% CI, 0.60-0.76) and CRC-related death by 43% (HR, 0.57; 95% CI, 0.45-0.72).

Secondary outcome measures included the incidence of distal and proximal CRC, all-cause mortality, and mortality unrelated to CRC. Flexible sigmoidoscopy significantly reduced the rate of distal CRC, by 36% in an ITT analysis (HR, 0.64; 95% CI, 0.57-0.72) and by 50% in a per protocol analysis (HR, 0.50; 95% CI, 0.42-0.59), but there was no significant effect of screening on the other outcome measures.

The number needed to screen to prevent 1 CRC diagnosis during the study period was 191 (95% CI, 145-277). The number needed to screen to prevent 1 CRC-related death was 489 (95% CI, 343-852).

The study authors noted that--in the original analysis--information about cause of death (ie, related vs unrelated to CRC) was obtained from the Office for National Statistics. In an analysis in which cause of death was verified by an independent coder, the effect of screening on the risk of CRC-related mortality did not change, but the number needed to screen to prevent 1 case of CRC-related death during follow-up decreased from 489 to 402 (95% CI, 291-647).

The effect of screening on outcomes did not vary significantly by age or sex. (Atkin WS, et al. Lancet 2010;375:1624-1633.)

Print  |  E-mail