Endoscopic therapy of submucosal invasive colorectal carcinomas – assessment of endoscopical and histopathological criteria predicting success or failure of treatment

Aims: Due to the potential risk of remission and metastasis, oncological resection is currently recommended after endoscopic resection of submucosal invasive colorectal cancer (pT1) if a “high-risk“ situation (i.e. venous or lymphatic vessel infiltration, poor tumor differentiation or R1/Rx-resection) is present. However, there are sparse data on the long-term outcome of patients with low-risk cancers compared with those with high-risk cancers in whom surgery was not performed.

Patients and methods: Between 1974 and 2002, a total of 318 cases of solely endoscopically treated pT1-cancers were registered in the Department of Pathology at the Municipal Hospital Munich-Neuperlach. Patient files were analyzed for histological and endoscopic parameters. Patients or their referring physicians were prospectively contacted for assessment of outcome of disease during follow-up (at least 24 month after initial diagnosis or earlier if patient suffered relapse). Patients lost for follow-up or those in whom only insufficient data could be assessed were excluded from further analysis. Poor outcome during follow-up was determined by relapse of cancers, distant or local metastasis or colorectal cancer related deaths.

Results: Long-term follow-up data were available from 248 patients (78%) with a median follow-up time of 84 months (range 10–269 months). Among those, 186 were primarily catagorized within a low-risk-category, 62 patients had endoscopically removed high-risk cancers. Poor outcome was determined in 16 patients of whom 4 patients had low-risk cancers, whereas the remaining 12 patients belonged to the high-risk-group (2.2% vs. 19.4%, p<0.001). In general, tumor-grading, R0-status and lymphoepithelial invasion were significantly associated with a bad outcome (all p>0.01), whereas, venous-vessel infiltration, tumor-grading, size or location of tumors did not play a role. Multiple regression analysis revealed that lymphoepithelial invasion was the only independent parameters predicting a bad outcome (odds ratio for relapse, metastasis or tumor-associated death: 11.1, p=0.016).

Discussion: The risk of relapse or progress of disease is low after endoscopic therapy of pT1 colorectal lesions. Lymphoepithelial invasion was the most valuable parameter predicting a bad outcome after endoscopic therapy. Hence, oncological resection appears mandatory if this histopathological criterion is detected.