Although endoscopic screening provides an opportunity for early identification and removal of premalignant polyps, data quantifying the long-term effects of sigmoidoscopy screening are lacking, corresponding author Frederik E. Juul, MD, said in an interview.
“Sigmoidoscopy screening have been shown to reduce colorectal cancer incidence and mortality, but it was unknown how long-lasting the effects were, and whether the effect differed by sex or age,” Dr. Juul said.
“For the first time, we were able to pool data from all four randomized sigmoidoscopy screening trials and include data from recent updates from two of the trials (U.S. and Italy), which means that we were able to answer these questions better than ever before,” he said.
In the pooled analysis, published in Annals of Internal Medicine, researchers from Norway, the United States, Italy, and the United Kingdom reviewed data from four studies with at least 15 years of follow-up. The analysis included 137,493 individuals randomized to at least one sigmoidoscopy screening and 137,459 randomized to usual care.
The primary outcomes were the incidence and mortality of CRC after sigmoidoscopy screening, compared with usual care, in adults with average CRC risk aged 55-64 years. Secondary outcomes included CRC incidence and mortality based on distal versus proximal colon, sex, and older versus younger age group (55-59 years vs. 60-64 years at study enrollment).
After 15 years’ follow-up, the pooled cumulative incidence of CRC was 1.84 cases per 100 persons in the screening group versus 2.35 cases per 100 persons in the usual-care group, representing a 21% reduction in incidence among those who were screened.
The pooled cumulative CRC mortality was 0.51 deaths per 100 persons in the screening group versus 0.65 deaths per 100 persons in the usual-care group, representing a 20% reduction in CRC mortality for those who were screened, the researchers noted. The all-cause mortality was reduced by 2% in the screening group compared with usual care; the pooled cumulative all-cause mortality was 14.3 deaths per 100 persons in the screening group versus 14.6 deaths per 100 persons in the usual-care group.
In terms of secondary outcomes, the significant reductions in CRC incidence and mortality were confined to the distal colon, with no significant differences observed in the proximal colon, the researchers noted. The reasons for this difference are unclear. Previous studies of three of the four trials showed a small reduction in CRC in the proximal colon, but may be related to the longer follow-up in the analysis of four trials.
The incidence of CRC varied by gender, with an incidence reduction of 25% for men versus 16% for women. The reasons for the gender difference are yet to be undetermined, but may include differences in the quality of bowel preparation, the greater technical challenge of screening women, and the higher incidence and proportion of proximal colon cancer versus distal colon cancer in women, the researchers noted.
“The long-term benefit of one single procedure was probably what surprised us the most,” Dr. Juul said in an interview. “Not only were the cumulative incidence and mortality lower in screened individuals 15 years after the procedure, but the yearly incidence was consistently lower in screened individuals compared to usual care, even at the end of the follow-up period.
“Although a previous study in Norway had indicated a sex difference in effect, we were surprised to see this in a pooled analysis across trials in four different countries,” he added.