Short term significance of DNA ploidy and cell proliferation in breast carcinoma: A multivariate analysis of prognostic markers in a series of 308 patients

Abstract

Aim - To determine the importance of tumour DNA ploidy and cell proliferation, as measured by the S phase fraction (SPF), in relation to other established clinicopathological indicators of prognosis in breast cancer. Methods - A prospective study of 308 patients. Tumours were staged following the TNM system criteria and were classified according to the histological type and grade. DNA flow cytometry was performed on fresh/frozen samples stained with propidium iodide. Hormone receptors were analysed by immunocytochemistry. A Cox proportional hazards regression model was used for statistical evaluation of the prognostic factors. Results - Median follow up time was 39.6 months (range 3 to 84). A DNA diploid pattern was found in 134 tumours (43.5%) and aneuploid in 174 (56.5%). Median SPF value was 6.1% (range 1% to 27.8%). DNA ploidy and SPF were strongly correlated (p < 0.001), and both were related to histological type (p < 0.001), grade of differentiation (p < 0.001), tumour size (p = 0.006 and p = 0.002), and hormone receptor activity (p < 0.001). DNA ploidy was also related to node status (p = 0.022), but SPF was not. In univariate analysis, there were significant correlations between disease-free survival and age, histological grade, tumour size, node status, DNA ploidy, SPF, and hormone receptor activity; age, tumour size, node status, DNA ploidy, and hormone receptors were predictors of overall survival. In multivariate analysis, only node status (p = 0.001) and DNA ploidy (p = 0.006) retained independent prognostic significance in relation with overall survival, while node status (p < 0.001) and SPF (p < 0.001) were predictors of disease-free survival. DNA ploidy and SPF continued to predict disease-free and overall survival in lymph node positive (pN1) patients but not in the lymph node negative (pN0) group. Conclusions - DNA ploidy and SPF are strongly intercorrelated and have independent prognostic value for predicting the short term clinical outcome of breast carcinoma patients.