Short-term outcomes of endoscopic submucosal dissection for suspected T1 colorectal cancers: A European experience

Background and Aims The indication for primary surgery on suspected deep submucosal invasive colorectal carcinoma (d-SMIC) is debatable. Consequently, local excision techniques, such as endoscopic submucosal dissection (ESD), are increasingly attempted in such patients. This study retrospectively evaluated the effectiveness of ESD in obtaining a free vertical margin (VM)-R0 in suspected d-SMIC compared with suspected superficial SMIC (s-SMIC). Methods ESDs for suspected T1 colorectal cancer (CRC) in treatment-naïve polyps were included between 2011 and 2022 in 5 European tertiary referral centers. Based on optical assessment, lesions were categorized into suspected d-SMIC or s-SMIC. Main outcomes were the VM-R0 rate, en bloc resection rate, and adverse event rate. An adjusted risk ratio for VM-R1 resections within the suspected d-SMIC group was calculated. Results In the suspected s-SMIC group (n = 1063), en bloc resection rate, VM-R0 rate, proportion of pT1 CRC, and adverse even rate were 90.5% (95% CI, 88-92), 90.6% (95% CI, 88-92), 18.0%, and 3.6% (IQR, 2-5), respectively. In the suspected d-SMIC group (n = 139), these values were 61.9% (IQR, 54-70), 55.4% (IQR, 47-63), 74.8%, and 5.8% (IQR, 2-10), respectively. Compared with suspected s-SMIC cases, the VM-R0 rate of suspected d-SMIC cases particularly decreased for pT1Sm2-3 (75.9% vs 55.7%). None of the investigated features (age, sex, polyp location, size, morphology, and Hiroshima classification) predicted a VM-R1 resection in suspected d-SMIC cases. Conclusions ESD performed on polyps with suspected d-SMIC showed lower VM-R0 rates for pT1Sm2-3 cases compared with suspected s-SMIC cases. This should be taken into account when selecting the optimal resection technique for suspected d-SMIC cases.