講演情報
[JKS2-1]T1 Colon Cancer Treatment Strategy after EMR or ESD
*Rumi Shin1 (1. Department of Surgery, Seoul National University Boramae Medical Center / Seoul National University College of Medicine, Korea)
Endoscopic resection is first-line for lesions that look benign, so pT1 colon cancer is usually diagnosed only after the specimen reaches pathology. The surgeon must then decide on a specimen rather than a tumour: completion colectomy, or surveillance. About 11% of pT1 colorectal cancers harbour nodal metastasis, yet the Japanese criteria flag roughly 80% of patients for additional surgery - sensitivity 100%, specificity 19% - so four in five of those operations treat node-negative disease.
Risk stratification: mutually adjusted, only lymphovascular invasion (OR 3.16), tumour budding (2.83) and poor differentiation (2.14) remain independent predictors; submucosal depth does not (1.73, 95% CI 0.96-3.12). Depth is a gradient, not a switch: 1.3-2.6% when it is the only adverse finding, 11.0% beyond 2,000 micrometres, and surgery for depth alone did not reduce recurrence.
Indications, and their cost: Japanese, Korean and Western rules draw the line in different places, and Korea itself has two guidelines that disagree. Refined criteria based on budding, poorly differentiated clusters and lymphovascular invasion reduced false positives by 41-45% without missing nodal disease, and a Korean nomogram adds a factor no guideline counts - cancer in a residual adenoma (OR 0.58).
Oncologic safety and individualised decisions: where the Japanese criteria say no, nodal metastasis is 1.8%, recurrence 0.4% and five-year disease-free survival 99.6%. Nodal risk is only one side of the equation; age, comorbidity, a low rectum and the ability to commit to surveillance change what the operation costs - which is why this belongs in a multidisciplinary discussion, not a checklist.
I will propose one minimum pT1 pathology dataset for KSCP and JSCP, reported the same way in both countries.
Risk stratification: mutually adjusted, only lymphovascular invasion (OR 3.16), tumour budding (2.83) and poor differentiation (2.14) remain independent predictors; submucosal depth does not (1.73, 95% CI 0.96-3.12). Depth is a gradient, not a switch: 1.3-2.6% when it is the only adverse finding, 11.0% beyond 2,000 micrometres, and surgery for depth alone did not reduce recurrence.
Indications, and their cost: Japanese, Korean and Western rules draw the line in different places, and Korea itself has two guidelines that disagree. Refined criteria based on budding, poorly differentiated clusters and lymphovascular invasion reduced false positives by 41-45% without missing nodal disease, and a Korean nomogram adds a factor no guideline counts - cancer in a residual adenoma (OR 0.58).
Oncologic safety and individualised decisions: where the Japanese criteria say no, nodal metastasis is 1.8%, recurrence 0.4% and five-year disease-free survival 99.6%. Nodal risk is only one side of the equation; age, comorbidity, a low rectum and the ability to commit to surveillance change what the operation costs - which is why this belongs in a multidisciplinary discussion, not a checklist.
I will propose one minimum pT1 pathology dataset for KSCP and JSCP, reported the same way in both countries.