Abstract
Steroid nuclear receptors are known to be involved in the regulation of epithelial-mesenchymal transition process with important roles in invasion and metastasis initiation. Androgen receptor (AR) has been extensively studied, but its role in relation to breast cancer patient prognosis remains to be clarified. AR/ER ratio has been reported to be an unfavorable prognostic marker in early primary breast cancer, but its role in the patients with advanced disease has to be cleared. We retrospectively analyzed ER, PgR, and AR expression on a case series of 159 specimens of primary BC samples by using immunohistochemistry and 89 patients of these had luminal tumors for which AR and ER expression and survival data were available. For twenty-four patients both primary and metastatic tumors were available. A significantly shorter overall survival was observed in primary tumors with AR/PgR ratio ≥ 1.54 (HR = 2.27; 95% CI 1.30-3.97; p = 0.004). Similarly OS was significantly shorter when ER/PgR ratio ≥2 in primary tumors (HR = 1.89; 95% CI 1.10-3.24; p = 0.021). The analysis of the 24 patients who had biomarker determinations both in primary tumors and metastasis showed a better OS when AR/ER ratio in the metastasis was ≥ 0.90 (p = 0.022). Patients with a high AR/ER ratio in primary tumor that remained high in the metastasis had better prognosis in terms of OS (p = 0.011). Despite we suggested that the ratios AR/ER and AR/PgR could be used to identify patients with different prognosis, their real value needs to be better clarified in different BC settings through prospective studies.