p53 determination alongside classical prognostic factors in node-negative breast cancer: An evaluation at more than 10-year follow-up

Background: There is heterogeneity of methods and conflicting results conce rning the prognostic value of p53 in node-negative breast cancer. The clini cal value of a quantitative method for measuring tumoral p53 content still needs to be evaluated. Patients and methods: A long-term retrospective study was conducted on 297 node-negative patients with a median follow-up greater than 10 years (11 ye ars, 101-172 months). Classic prognostic factors were considered including age, tumor size, histoprognostic grade and estradiol (ER) and progesterone receptors (PR). In addition, the value of p53 determination (immunoluminome tric assay in tumor cytosol) was assessed for this long follow-up period. Results: p53 concentrations were significantly linked to the histological g rade (P = 0.001), to tumor size (P = 0.02) and ER status (P = 0.01). Higher p53 tumoral concentrations were found in tumors with large size, pejorativ e histological grade and negative ER status. In contrast, p53 tumoral conce ntrations were not influenced by menopausal or PR status. Multivariate Cox analysis demonstrates that tumor size was the only significant predictor of disease-free survival (P = 0.049) with a risk factor at 1.38. As regards s pecific survival, univariate Cox analysis indicates that p53 taken as a con tinuous variable is a significant predictor (P = 0.024) together with histo logical grade, tumor size and ER status. In a multivariate Cox analysis the re were two significant and independent variables for predicting overall su rvival: tumor size (P = 0.031) and ER status (P = 0.015) with the highest r isk factor (RR = 2.14). Conclusions: The present investigation points out that the prognostic power of p53 tumor determination evaluated at more than 10 years median survival is not higher than the well-recognized classic prognostic factors in node- negative breast cancer. The present data highlight the need to assess the p rognostic value of potentially new biological factors in node-negative brea st cancer on cohorts of patients followed over periods in excess of 10 year s.