Background: The benefit of adjuvant chemotherapy appears to be limited for patients with Astler Coller B1/B2 colorectal carcinoma but may be better in a subgroup of patients with a high recurrence risk. In the current case-control analysis, the authors evaluated whether patients with a high risk of hematogenous metastasis could be identified by means of a thorough histologic and immunohistochemical examination of the resection specimens.
Methods: A database was built for all patients treated in a general teaching hospital for colorectal carcinoma between 1985 and 1995. From this database, all patients with an Astler Coller B1 or B2 tumor who subsequently had developed hematogenous metastases were taken as cases. For each case, three matched controls (age, Astler Coller, year of diagnosis) without metachronous metastases were selected. The resection specimens of cases and controls were blindly examined by two observers for the following: World Health Organization (WHO) classification; differentiation grade; growth pattern; lymphocytic, fibroblastic, and eosinophilic reaction; angioinvasion; number of lymph nodes examined; expression of E-cadherin, vascular endothelial growth factor and thymidine phosphorylase (TP); P53; microvessel density.
Results: Twenty-two cases and 65 controls were included in the analysis. Tumor growth pattern and tumor TP expression both independently contributed to recurrence risk. With these 2 variables, 4 subgroups could be identified with a recurrence risk ranging from 0% to 42%.
Conclusions: Tumor growth pattern and degree of TP expression both appear to be related to the recurrence risk. Prospective trials should point out whether these variables can be implemented in the decision making concerning adjuvant chemotherapy.
Copyright 2001 American Cancer Society.