Histopathologic prognostic score in colorectal adenocarcinomas.

OBJECTIVE: To analyze the prognostic value of a set of pathological variables after curative resection for large bowel adenocarcinoma and to test a prognostic score derived from factors with independent effect. PATIENTS AND METHODS: The study is based on data from 292 consecutive unselected patients (B-C Astler-Coller stages). Histopathological features were evaluated prospectively on the resected primary tumors. Relationship between these factors and risk of recurrence was assessed by a Cox's proportional regression analysis. RESULTS: Four variables retained independent prognostic significance: extent of bowel wall invasion, peritumoral lymphocytic infiltration, number of positive nodes and vascular invasion. A prognostic score based on the regression coefficients attained by such variables was developed. This system revealed four prognostic groups. Group I included 14% of patients, with 94% 5-year disease-free survival. These figures were: 35% and 60% in group II; 43% and 46% in group III; and 7% and 24.4% in group IV. Histopathologic score applied to bearers of Astler-Coller B2 tumors permitted the identification of two populations, one characterized by a low risk of relapse and another with high risk (p = 0.002). CONCLUSION: A prognostic score based in the evaluation of four histopathologic parameters concerning the tumor phenotype enables the identification of groups of patients at risk of relapse after curative resection for colorectal adenocarcinoma.