Background: Despite decades of research the incidence and survival from ovarian cancer (OvCa) has remained the nearly same, understanding how the hormonal milieu affects ovarian carcinogenesis may provide insight into prevention. Epidemiologic evidence suggests that use of estrogen as hormone replacement therapy increases the risk of ovarian cancer and experimental studies demonstrate a mitogenic effect of estrogen in normal and malignant ovarian cancer cells. In this study we characterize the effect of estrogen on ovarian carcinogenesis in vitro and test whether estrogen increases tumor burden in a genetic mouse model of ovarian cancer. We also test whether the effects of estrogen are altered by a novel selective estrogen receptor modulator, Bazedoxifene.

Methods: Hormone receptor status of a cell line established from LSL-K-rasG12D/+PtenloxP/loxP mouse ovarian tumors was characterized using western blotting and immunohystochemisty. Invasion assays were completed using Boyden chambers and proliferation was assessed using MTT assays. LSL-K-rasG12D/+PtenloxP/loxP mice were treated with placebo (n=19), 17β-estradiol (n=19) or Bazedoxifene (n=21) daily for 3 weeks then OvCa was initiated by injection of AdeCre virus into the ovarian bursa, treatment was continued until the animals were sacrificed 8 weeks later.

Results: Invasion and proliferation assays were completed after confirming the cell line expressed both estrogen receptor α and β. 17β-estradiol significantly increased invasion of the OvCa cells, an effect that was attenuated by Bazedoxifene. Estrogen also increased proliferation of cells, an effect not seen with Bazedoxifene. Based on the increased invasion and proliferation of ovarian cancer cells after 17β-estradiol treatment we next asked if mice treated with 17β-estradiol would have increased tumor burden. In the genetic mouse model of ovarian cancer hormonal treatment did not affect mean tumor burden (placebo 1.2 g, 17β-estradiol 0.9 g, and Bazedoxifene 0.9 g; p=0.8). Furthermore, mice in all treatment groups had similar incidence of tumor, number of metastatic nodules and volume of ascities. A physiologic effect of the treatment was confirmed by performing uterotrophic assays and demonstrating a change estrogen receptor expression in the tumors after hormone treatment.

Conclusion: Consistent with other published studies, we found strong evidence of a detrimental effect of estrogen on ovarian carcinogensis in vitro. However, these effects did not translate into increased tumor burden in the genetic mouse model. Importantly, for the first time the effect of a novel SERM, Bazedoxifene, on OvCa is characterized. The findings here demonstrate the complex interplay between hormones and ovarian tumorigenesis, an area of research that is lacking. To make progress in ovarian cancer prevention and advise women regarding hormone replacement it is imperative that we have a better understanding of hormonal carcinogenesis in ovarian cancer.

Citation Information: Cancer Prev Res 2010;3(12 Suppl):A103.