Risk stratification for the detection of metachronous polyps after bowel screening polypectomy: clinical outcomes from the Integrated Technologies for Improved Polyp Surveillance (INCISE) study cohort

Background: After colorectal polypectomy, 20-50 per cent of patients develop metachronous polyps and some have increased colorectal cancer risk. British Society of Gastroenterology (BSG) 2020 guidelines recommend surveillance colonoscopy for high-risk patients based on index pathology. The aim of this study was to evaluate metachronous lesion outcome using BSG 2020 criteria. Methods: A retrospective, multicentred study was conducted including patients who had polypectomy during screening colonoscopy (2009-2016) followed by surveillance. Demographics, index pathology, and BSG 2020 risk criteria were compared with regard to metachronous lesion pathology (non-advanced versus advanced lesions) and timing of detection (early versus late). Advanced lesions were defined as adenomas/serrated polyps greater than or equal to 10mm, high-grade dysplasia, serrated polyps with dysplasia, or colorectal cancer, and late lesions those detected greater than 2 years after the index procedure. Results: Of 3090 eligible patients, 2643 were included. Among these, retrospective BSG 2020 application would have excluded 51.5 per cent from surveillance. After a median of 36 months, the advanced polyp/colorectal cancer rate in BSG 2020 high-risk patients was 16.3 versus 13.0 per cent in low-risk patients. Older age (P = 0.008) correlated with advanced metachronous lesions. Male sex, greater than five polyps, and BSG 2020 high-risk criteria correlated with non-advanced and advanced lesions (P < 0.001). Older age (P < 0.001), villous features (P = 0.006), advanced index polyp (P = 0.020), and greater than five polyps (P < 0.001) correlated with early metachronous lesions. Male sex and BSG 2020 high-risk criteria correlated with early and late lesions (P < 0.001). On multivariable regression, increased polyp number (odds ratio (OR) 1.15 (95 per cent c.i. 1.07 to 1.25); P < 0.001) and villous features (OR 1.49 (95 per cent c.i. 1.05 to 2.10); P = 0.025) independently correlated with early advanced lesions. The rate of non-advanced and advanced metachronous polyps was higher in BSG 2020 high- versus low-risk patients (44.4 versus 35.4 per cent for non-advanced and 15.7 versus 11.8 per cent for advanced; P < 0.001), but the colorectal cancer rate was similar (0.6 versus 1.2 per cent). However, when examining only lesions detected greater than 2 years after the index colonoscopy in high- versus low-risk patients, no significant differences were observed (P = 0.140). Conclusion: BSG 2020 criteria correlated with metachronous polyps, but did not differentiate advanced and non-advanced lesions and were not predictive of late lesions.