Evaluation of multiple bio-pathological factors in colorectal adenocarcinomas: Independent prognostic role of p53 and bcl-2

Authors:
Open Access
Publisher website
Google Scholar
Abstract
About 40% of patients with colorectal carcinoma will develop local or distant tumour recurrences. Integrated analyses of bio‐pathological markers, predictive of tumour aggressiveness, may offer a more rational approach to planning adjuvant therapy. To this end, we analysed the correlation between p53 accumulation, Bcl‐2 expression, DNA ploidy, cell proliferation and conventional clinico‐pathological parameters by testing the prognostic significance of these variables in a series of 171 colorectal carcinoma patients with long‐term follow‐up. The relationships among the various bio‐pathological parameters, analysed by multiple correspondence analysis, showed 2 different clinico‐biological profiles. The first, characterised by p53 negativity, Bcl‐2 positivity, diploidy, low percentage of cells in S‐phase (%S‐phase), a low Ki‐67 score, is associated with Dukes' A‐B stage, well differentiated tumours and lack of relapse. The second, defined by p53 positivity, Bcl‐2 negativity, aneuploidy, high %S‐phase and elevated Ki‐67 score, correlates with Dukes' C‐D stage, poorly differentiated tumours and presence of relapse. When these parameters were examined according to Kaplan‐Meier's method, significantly shorter disease‐free (DFS) and overall survival (OS) were also observed in patients bearing p53 positive and Bcl‐2 negative tumours, in Dukes' B stage. In multivariate analysis, p53 accumulation and Bcl‐2 expression emerged as independent predictors of a worse and better clinical outcome, respectively. Our results indicate that, in colorectal adenocarcinomas, a biological profile, based on the combined evaluation of p53 and Bcl‐2, may be useful for identifying high risk patients to be enrolled in an adjuvant setting, mainly in an early stage of the disease. Int. J. Cancer (Pred. Oncol.) 84:545–552, 1999.

This publication has 15 references indexed in Scilit: