Abstract

In this study of 64 cases of breast cancer with a clinical follow-up period of more than 5 years, several prognostic factors were evaluated. The purpose of this study was to determine whether any one parameter or group of parameters serves as adequate predictors of tumor behavior and patient's prognosis. Several prognostic factors included clinicopathological variables (patient's age, histologic grade, status of lymph node (LN) metastasis, and tumor size), expression of estrogen receptor (ER), progesterone receptor (PR), p53, bcl-2 and CD44 by immunohistochemistry, and DNA ploidy pattern. The results showed that the expression of ER and PR had a significant inverse correlation with the histologic grade (ER, p=0.05; PR, p<0.05). The expression of p53 protein showed a significant relationship with high histologic grade of tumor (p<0.05). The expression of bcl-2 protein was preferably seen in low histologic grade of tumor (p<0.05) and significantly associated with ER positive or PR positive tumors (ER, p<0.05; PR, p<0.05). This results suggest that bcl-2 protein might play significant roles in ER and PR. The CD44 expression showed a significant relationship with tumor size (p<0.05). The large size and aneuploidy pattern of tumor had a tendency to be associated with shorter patient survival. Cox's multivariate analysis showed that overall survival was affected by LN metastasis because of the shorter survival in patients with LN metastasis. In conclusion, tumor size, DNA ploidy pattern, and LN metastasis were themselves significant predictors of breast cancer survival rate.

• Keywords: Breast cancer;Clinicopathological variables;Prognostic factors;Survival rate