Aims
Current ESGE guidelines consider endoscopic resection of colorectal cancer “curative” when the estimated risk of lymph node metastasis is <3% (Endoscopy 2022;54:591). However, established histopathologic risk factors—such as tumor grade, tumor budding, vascular invasion, resection margin status, and submucosal invasion depth >1000 µm—do not carry equal prognostic weight. A recent meta-analysis, for example, reported a lymph node metastasis rate of only 2.6% when submucosal invasion >1000 µm was present as the sole risk factor (Gastroenterology 2022;163:174). In this study, we evaluated patients from a large institutional database who underwent ESD for colorectal cancer, focusing specifically on how submucosal invasion depth influenced subsequent clinical management.
Methods
Patients who underwent en bloc ESD for colorectal cancer were identified from a prospective colorectal ESD registry containing more than 800 procedures performed between 09/2012 and 09/2025. We assessed lymph node risk factors, post-ESD management decisions, and patient outcomes.
Results
A total of 104 patients (median age 70.6 years; 34% female) with 104 lesions (rectum n = 69 / colon n = 35; median size 36 mm) were included. Based on ESGE criteria, 33 lesions (32%) were classified as very low- or low-risk for lymph node metastasis, while 71 lesions (68%) were considered high-risk. Lesion size and location did not differ significantly between risk groups (p=0,801 / p=0,824). In 26 cases (38% of high-risk lesions), submucosal invasion depth >1000 µm was the only high-risk feature. Among these, nine patients underwent additional surgery, and none showed lymphatic metastasis. The remaining seventeen patients did not undergo surgery and remained disease-free during an average follow-up period of 25 months.
Conclusions
In our cohort of 104 early colorectal cancers, the proportion of resections considered curative would increase from 32% to 54% if submucosal invasion depth >1000 µm as the sole risk factor were reclassified as low-risk. Our findings suggest that this criterion alone should no longer define a high-risk resection.