EP216 - Endoscopic Submucosal Dissection versus Surgical Resection in the Definitive Management of T1 Colon Cancer: Outcomes from Single-Center Experience

Purpose/Background: Localized colon cancer is generally treated with upfront surgical resection, but endoscopic submucosal dissection (ESD) of T1 malignant polyps may be considered in the right context. Data are lacking on the appropriate clinical context for pursuing ESD regarding such colon polyps. This study evaluates clinical and pathological outcomes after ESD compared to surgical resection of T1 colon cancer at a single tertiary academic center.

Methods/Interventions: From 2017-2025, clinical outcomes for 34 patients with T1N0M0 colon adenocarcinoma were analyzed. All patients underwent either definitive ESD or surgical resection of the malignancy. Demographics, histologic features and location of the polyp, treatment approach (definitive ESD, ESD followed by definitive surgery, or upfront surgery), final surgical pathology, and clinical outcomes were evaluated. High-risk histologic features were analyzed and defined as piecemeal ESD resection, poor differentiation, positive deep margin, deep submucosal invasion (SM2 or ≥1000 μm), lymphovascular invasion, perineural invasion and high tumor budding. Group characteristics were compared with two-tailed T tests and chi-squared tests, when appropriate. Univariate logistic regression was performed to assess factors associated with residual disease on surgical pathology.

Results/Outcomes: All 34 patients had initial T1 staging on endoscopic biopsy. Of these patients, 6 underwent ESD at our institution followed by surgical resection at an outside hospital, so they were excluded. Of the remaining 28 patients, 19 underwent surgical resection ultimately, of which 15 (79%) had undergone ESD prior to surgery. The remaining 9 patients underwent ESD only. The surgical cohort was more likely to have at least one high-risk feature compared to the definitive ESD group (95% vs 56%, p = 0.011). Of the high-risk features, the most common across surgical and ESD groups was deep submucosal invasion (58% & 33%, respectively) and positive deep margin (42% & 22%, respectively). The surgical group was more likely to have at least 2 high-risk features (53% vs 11%, p = 0.036). Of the 19 patients to undergo surgical resection, eight patients had residual disease on final pathology: four had node-positive disease (21%) and another four had residual T1N0 disease. Among the 19 surgical patients, regression analysis was performed for potential predictors of residual disease. Age and high-risk features had no association and left-sided tumors had a borderline statistically significant association with residual disease (OR 0.125, CI 0.016-0.999, p = 0.050).

Conclusion/Discussion: There are sparse data in the US on clinical outcomes in T1 patients managed with ESD vs surgical resection. Most of our cohort of T1 colon cancer patients presented with high-risk histologic features, and most patients ultimately underwent surgical resection. Within our sample, a notable segment of patients who underwent surgery had residual disease on final pathology (42%), and, importantly, a significant portion had node-positive disease (21%). Our findings substantiate the utility of surgical resection in T1 colon cancer with high-risk features. However, a subset of T1 patients may continue to benefit from conservative endoscopic management and further research is needed. Close collaboration between gastroenterology and surgical services is essential to ensure collaborative management of T1 colon cancer.

Table 1: Patients, Tumor Characteristics, Treatment Details, and Pathology
Table 2: Univariate Logistic Regression for Factors Associated with Residual Cancer Post-Surgical Resection