Differential expression of estrogen receptor subtypes and variants in ovarian cancer: effects on cell invasion, proliferation and prognosis.

Background: Due to the presence of both classical estrogen receptor (ERα) and another ER subtype (ERβ) in ovarian cancer, hormonal treatment is an attractive option. However, response to tamoxifen in ovarian cancer is modest. The presence of ERβ variants further complicated the issue. We have recently shown that specifically targeting ER subtypes using selective ER modulators showed opposing functions of ER subtypes on cell growth. In the present study, the clinical significance of ERα and ERβ variants (β1, β2 and β5) and the functional effects of ERβ2 and ERβ5 in ovarian cancer was investigated.

Methods: ERα, ERβ1, ERβ2 and ERβ5 expression were evaluated by immunohistochemistry in 106 ovarian cancer tissues. The association between ERs expression and clinicopathological parameters or prognosis was analyzed. Ectopic expression of ERβ2 and ERβ5 followed by functional assays were performed in ovarian cancer cell lines in order to detect their effects on cell invasion and proliferation.

Results: We found significantly higher nuclear (n)ERα and nERβ5 and lower cytoplasmic (c)ERα expression in advanced cancers. Significantly lower ERβ1 expression was also detected in high grade cancers. Significant loss of nERα and cERβ2 expression were observed in clear cell histological subtypes. Higher nERβ5 and lower cERβ5 expression were associated with serous/clear cell subtypes, poor disease-free and overall survival. Positive cERα and higher cERβ1 expression were significantly associated with better disease-free and overall survival. Furthermore, we found nERβ5 as an independent prognostic factor for overall survival. Functionally, overexpression of ERβ5 enhanced ovarian cancer cell migration, invasion and proliferation via FAK/c-Src activation whereas ERβ2 induced cell migration and invasion.

Conclusions: Since tamoxifen binds to both ERα and ERβ1 which appear to bear opposing oncogenic roles, the histotypes-specific expression pattern of ERs indicates that personalized treatment for women based on ERs expression using selective estrogen receptor modulators may improve response rate. This study also suggests nERβ5 as a potential prognostic marker and therapeutic target in ovarian cancer.