Colonoscopy Surveillance for Advanced Adenomas Does Not Always Prevent Colorectal Cancer

Abstract

Introduction: Surveillance colonoscopy is recommended after resection of advanced adenomatous polyps which are the lesions in the colon and rectum with the highest risk for malignant transformation. The frequency and risk factors associated with the development of colorectal cancer (CRC) at the polyp resection site in spite of surveillance colonoscopy may improve our ability to prevent cancer in patients with advanced adenomas. Aim: To determine the frequency and risk factors for CRC development at the resection site and other areas of the colon among individuals with high risk polyps who underwent polypectomy. Methods: Data were collected from the Mayo Clinic, Rochester, MN, electronic medical record on 14663 subjects who underwent screening and surveillance colonoscopies between 1/1990 through 12/2010 and followed up through 12/2016. We included subjects who were found to have advanced adenomatous polyps resected by colonoscopy and had at least one surveillance colonoscopy for their advanced polyps. Results: 13402 patients with advanced adenomas were included. Of the 4610/13402(34.4%) patients who had a follow-up colonoscopy within <10 years, 84/4610(1.8%) developed CRC at the polypectomy site within a median of 2.31(0.58-4.27) years. 1.2% (55/4610) developed CRC in a region of the colon distinct from the site of the advanced polyp within a median of 2.64 (1.0-6.33) years. Patients who developed CRC at the polypectomy site were most likely to have an index polyp that was villous (48.8%) and arose in the right colon (75.0%). Approximately, 25/84(30%) of patients who developed an interval CRC at the site of the advanced polyp had followed appropriate surveillance guidelines. In patients who later developed CRC at another site, their index polyp was most likely to be villous (50.0%) and to be located in the transverse colon (30.0%). 15/54(27.8%) patients had undergone recommended surveillance interval. In multivariate analysis, age (HR=1.07 [95% CI, 1.04-1.11], P<0.01); right-sided location of the index polyp (HR=5.77 [95% CI, 2.44-16.0], P<0.01), increasing polyp size(HR=1.50 [95% CI, 1.15-1.92], P<0.01); high vs low degree of dysplasia (HR =2.01[95% CI, 1.07-3.76], P =0.03); higher number of polyps resected(HR=1.12 [95% CI, 1.05-1.17], P<0.01), and piecemeal removal (HR=2.40 [95% CI, 1.13-4.97], P=0.02) were associated with an increased risk for the development of CRC at the same site as the index polyp. Polypectomy device used and mucosal lift did not correlate with an increased risk for CRC. Increasing age (HR=1.08 [95% CI, (1.04-1.12], P<0.01); right-sided location of the index AA (HR=2.73 [95% CI, 1.36-5.86], P<0.01); and sessile endoscopic appearance of the index AA (HR=3.16 [95% CI, 1.17-11.05], P =0.02) were significantly associated with an increased risk for CRC development not related to associated with index polyp. Conclusion: CRC developed in 3 %(139/4619) of patients despite careful colonoscopic surveillance. This suggests limitations in our risk stratification of polyps and our technical abilities to manage and detect polyps.FigureFigureFigureTable: Table. Colonoscopy surveillance features associated with interval CRC at the index AA/SSA polypectomy siteTable: Table. Colonoscopy surveillance features associated with interval CRC at site distinct from AA/SSA site