Abstract A108: Predictors of hormone receptor status in ovarian cancer: results from the Nurses’ Health Study

Abstracts: Frontiers in Cancer Prevention Research 2008 A108 Predictors of hormone receptor status in ovarian cancer: results from the Nurses’ Health Study Background Hormone receptor expression in tumors may offer etiologic information for ovarian cancer, particularly in light of known associations with hormonal and reproductive risk factors. Methods TMAs were constructed using triplicate core samples extracted from 157 paraffin-embedded blocks of confirmed epithelial ovarian tumors collected from participants in the Nurses’ Health Study and stained for estrogen receptor-alpha (ERα) and progesterone receptor (PR). We examined receptor expression by stage, grade and histologic subtype. Multivariate unconditional logistic regression was used to evaluate whether various hormonal, reproductive and anthropometric risk factors differed by ERα and PR expression among cases, and whether these exposures are differentially associated with the risk of developing receptor-positive or receptor-negative ovarian cancer compared to controls. Results PR expressing tumors were less likely to be invasive ( P = 0.05) and more likely to be of a lower grade ( P < 0.001) compared with PR- tumors. ERα status was not associated with any pathological features of the tumor ( P ≥ 0.76). In the case-control analysis, increasing age, being postmenopausal, and PMH use were associated with an increased risk of developing ERα+, but not ERα-, ( P for heterogeneity = 0.001, 0.06, and 0.06, respectively) and PR-, but not PR+, tumors ( P for heterogeneity = 0.08, 0.003, and 0.40, respectively), while height was only associated with the risk of developing PR- disease ( P for heterogeneity = 0.08). There were no clear differential associations of OC use, parity, adolescent or adult BMI, or physical activity with expression of either receptor. Results were similar in the case-case analysis. Conclusion Despite limited statistical power, our findings suggest that various risk factors are only associated with certain tumor subtypes based on hormone receptor expression, thus providing mechanistic evidence regarding these relationships in the etiology of ovarian cancer. Given our small sample size, confirmatory studies are required to further delineate these relationships. Citation Information: Cancer Prev Res 2008;1(7 Suppl):A108.