The traditional staging system is currently inadequate for identifying those patients with colorectal carcinoma (CRC) who carry a high risk for poor outcome. In this study, the expression of E-cadherin was evaluated in CRC to determine its correlation with clinico-pathological variables, and association with disease outcome in patients with long-term follow-up. The present series consisted of tissue samples obtained from 230 patients with stage I, II, III, or IV CRC treated during 1981-1990 at Turku University Hospital. Archival paraffin-embedded samples were used to build up tissue microarray blocks, and E-cadherin expression was assessed by immunohistochemistry using an automated staining system. Different grading systems were tested for expression of E-cadherin. Fifty-nine percent of all tumors were positive for E-Cadherin. There was no significant correlation between E-cadherin expression and gender (p < 0.83), localization (p < 0.45), tumor invasion (p < 0.32), or histologic grade (p < 0.41). However, loss of E-cadherin expression was significantly associated with older age (p < 0.03) and lymph node involvement (p < 0.02), and with borderline significance with advanced stage (p < 0.09) and tumor metastasis (p < 0.09). In univariate (Kaplan-Meier) survival analysis, positive E-cadherin significantly (p = 0.009) predicted longer disease-free survival (DFS), and the same was true with disease-specific survival (DSS) as well (p = 0.007). In multivariate (Cox) survival analysis, E-cadherin retained its significance as independent predictor of DFS (HR = 1.56; 95% CI 1.01-2.42, p = 0.043), but not DSS. A sub-group analysis revealed that E-cadherin expression also predicts DFS (p < 0.01) and DSS (p < 0.04) in stage II CRC. Our results implicate the usefulness of E-cadherin expression in predicting disease recurrence and long-term survival in CRC.
No takes yet. Share an insight, caveat, or question.
Elzagheid et al. (2012) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: