Answers are grounded in this paper and cite their sources.
Aims Although ESD is preferred for rectal adenomatous lesions larger than 20 mm, there is less agreement related similar lesions located in the colon. There is current inconsistent evidence coming from both Eastern and Western centres, with different rates of submucosal invasive cancer (SMIC) in the colon and the rectum, with studies based on pEMR reporting lower rates especially in the right colon (5%), and studies based on ESD reporting higher rates (8-27%) [ 1 ] [ 2 ] [ 3 ] [ 4 ] [ 5 ]. We set up a European multicenter retrospective study to assess the prevalence of SMIC and high-grade dysplasia (HGD) in a cohort of colorectal LST treated by ESD. Methods European centers were invited to fill in a database of consecutive patients. Clinical variables related to patients and procedures as well as the center’s resection strategy for colorectal LST dissection (selective or universal colorectal ESD) were recorded. The prevalence of HGD and SMIC in the left colon (LC – splenic flexure to sigmoid) and right colon (RC – caecum to transverse colon) and rectum (RE) were recorded, and a logistic regression analysis was performed to test the data. Results 19 European centers provided data on 3321 lesions, 824 in the RC, 496 in the LC and 2001 in RE. 1055 lesions were resected with an universal colorectal ESD approach, 2063 with a universal rectal – selective ESD colon approach, 205 with a selective colorectal ESD approach. There were statistically significant differences in the case mix of RC, LC and RE, with smaller lesions in the RC and LC when compared to RE (RC 40.8mm±16.4; LC 40.5mm±18.6; RE 51.3±28.2mm p<0.001). RE had higher prevalence of LST-G-M (57.0%; RC 38.2%; LC 35.7%, p<0.001). LST-NG-FE and LST-NG-PD prevalence was higher in the colon (respectively, RC 21.1%; LC 17.1%; RE 5.6%, p<0.001; RC 15.0%, LC 14.1%, vs RE 4.2% p<0.001). Sessile lesions were more common in LC (16.1%) and RE (16.8%) than RC (6.9%, p<0.001). The LC and RE had higher prevalence of SMIC when compared to the RC (RC 11.6%; LC 22.4%; RE 19.8% p≤0.001). A multivariate analysis was carried out on the overall cohort and separately on the RC, LC and RE cohorts to find risk factors associated with SMIC (variables considered: age, gender, dimensions, resection strategy, nodular component, morphology – LST classification). On the overall cohort, right colon location and morphology LST-G-H are protective factors against the risk of SMIC. In RC, LST-G-M and LST-NG-PD are risks factor for SMIC. In LC, LST-NG-PD, sessile and semipeduncolated lesions are risk factors for SMIC. Male gender and morphology other than LST-G-H ( a part semi-peduncolated) were risks factor for increased SMIC in RE. Conclusions Our current data suggest that ESD could be potentially adopted as universal strategy in the left colon and rectum, a selection strategy is advised on the right colon to achieve cost-effectiveness. Different odds ratio are associated with polyp morphology depending on the location. Right colon is a protective factor against the risk of SMIC, whereas left colon has similar rates to rectum. Publication History Article published online: 05 June 2026 © 2026. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany
No takes yet. Share an insight, caveat, or question.
Rimondi et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: