Background:
Individuals with adenomatous polyps are advised to undergo colonoscopy surveillance, but the relationship between type of adenomas at colonoscopy and long-term CRC incidence, CRC mortality and all-cause mortality is unclear. Our aims were to compare long-term CRC incidence, CRC death and all-cause mortality by adenoma findings at baseline colonoscopy.
Methods:
Data are from the Minnesota fecal occult blood trial, a randomized clinical trial of fecal occult blood (FOBT) screening in the US beginning in 1975 with follow-up for CRC incidence to 2000. Participants included 46,551 men and women aged 45 and older of whom 10,584 underwent colonoscopy following a positive FOBT. Colonoscopy findings were categorized as advanced adenoma [≥1 cm, high-grade dysplasia, or tubule-villous or villous histology], non-advanced adenoma [<1 cm without advanced histology], or no adenoma. The primary outcome was CRC incidence up to 20 years of follow up, CRC death and all-cause mortality up to 30 years of follow up. We compared colonoscopy results via the sub-distribution hazard ratios and cumulative incidence curves for CRC incidence and CRC death with death as a competing risk using the Fine-Gray estimator. We examined the hazard ratio for overall mortality using the Cox estimator. Analyses were adjusted for age at colonoscopy and sex.
Results
There were 10,584 participants who underwent colonoscopy: Males 50%, mean age 69 [+/-8] years and 12% obese [BMI≥30 kg/m² ]. On initial colonoscopy, 912 participants had an advanced adenoma and 1,009 participants had a non-advanced adenoma. The 20-year cumulative incidence of CRC [95% CI] for those with normal exams, non-advanced adenomas and advanced adenomas were 1.8% [1.5%, 2.1%], 3.9% [2.7%, 5.4%] and 5.5% [4.1%, 7.2%] respectively. Participants with non-advanced and advanced adenoma were significantly more likely to develop CRC compared to those with no adenoma (sub-distribution hazard ratio [SHR], 2.24 [1.54,3.24] P= 0.001 and 3.24 [2.32, 4.52] P =0.001, respectively]. Compared to participants with no adenoma, those with advanced adenoma were at significantly increased risk of CRC mortality (SHR, 2.20 (1.36, 3.57); P 0.001) and all-cause mortality [HR 1.12 [1.03, 1.21]; P=0.005] while participants with non-advanced adenoma did not have a statistically significant increase in CRC mortality [SHR 1.28 [0.72, 2.28; P=0.4] or all-cause mortality [HR 1.07 [1.00, 1.16]; P= 0.059] [Figure 1 andTable 1]
Conclusions:
Over a long-term follow-up, participants with a non-advanced and advanced adenoma at colonoscopy were at significantly increased risk of developing colorectal cancer compared with those with no adenoma, and participants with an advanced adenoma were at increased risk for CRC mortality and all-cause mortality. The study underscores the importance of surveillance colonoscopy for individuals with advanced adenomas.

Jack Mandel

Appearances