Abstract
BACKGROUND: Differential associations between ovarian cancer risk factors and estrogen receptor-α (ERα) ovarian tumor expression have been noted; however, no research has assessed estrogen receptor-β (ERβ) expression. Thus, in exploratory analyses, we assessed the association of several factors with ovarian cancer risk by ERβ tumor status. METHODS: We conducted a nested case-control study within the prospective Nurses' Health Study cohorts (NHS/NHSII), with exposures collected through biennial questionnaires. Paraffin-embedded tumor blocks were requested for cases diagnosed from 1976 to 2006 (NHS) and 1989 to 2005 (NHSII) and tissue microarrays were stained for nuclear ERβ (ERβ-nuc) and cytoplasmic ERβ (ERβ-cyto), with any staining considered positive (+). We obtained odds ratios (OR) and 95% confidence intervals (CI) using multivariate polytomous logistic regression. RESULTS: We included 245 cases [43% ERβ-cyto (+) and 71% ERβ-nuc (+)] and 1,050 matched controls. An inverse association was observed between parity and risk of ERβ-nuc (+) (OR, parous vs. nulliparous: 0.46; 95% CI, 0.26-0.81), but not ERβ-nuc (-) tumors (OR, parous vs. nulliparous: 1.51; 95% CI, 0.45-5.04; P (heterogeneity) = 0.04). Conversely, parity was inversely associated with ERβ-cyto (-) tumors (OR, parous vs. nulliparous: 0.42; 95% CI, 0.23-0.78), but was not associated with ERβ-cyto (+) tumors (OR, parous vs. nulliparous: 1.08; 95% CI, 0.45-2.63; P (heterogeneity) = 0.05). Associations for other exposures, including hormone therapy, did not differ by ERβ-nuc or ERβ-cyto status. CONCLUSIONS: Our results suggest that parity may influence ovarian cancer risk, in part, through alterations in ERβ localization within tumor cells. IMPACT: Alterations in ERβ expression and localization appear to be important for ovarian cancer etiology. Future research should confirm our results and assess potential biologic mechanisms for the observed associations.