Abstract 876: Bilateral oophorectomy and breast cancer: Confounding by non-malignant indications for surgery

Abstract

Abstract Background: Bilateral oophorectomy at early ages (<40 years) is recognized to lower breast cancer risk. Among U.S. women, oophorectomy is often performed with hysterectomy for benign conditions such as endometriosis and uterine fibroids. We aimed to determine whether a prior diagnosis of endometriosis or fibroids would attenuate breast cancer risk reductions conferred by bilateral oophorectomy. Methods: We analyzed data from a population-based case-control study of women aged 50-79 living in Wisconsin, Massachusetts and New Hampshire during 1992-1995. Incident cases (N=5,068) of invasive breast cancer were identified from state-wide tumor registries. Age-matched controls (N=5,246) were selected from driver's license and Medicare beneficiary lists. Reproductive and medical history (including physician diagnosis of endometriosis or uterine fibroids) and lifestyle factors were self-reported during structured telephone interviews. Odds ratios (OR) and 95% confidence intervals (CI) were estimated using logistic regression and adjusted for reproductive history, exogenous hormone use, body mass index, and family history of breast cancer. Statistical interaction was assessed by including cross-product terms in regression models. Results: Overall, 83% of cases and 78% of controls participated. Women with bilateral oophorectomy at ages ≤40 years had 0.76 times the odds of breast cancer (95% CI: 0.63, 0.92) compared to women with intact ovaries and uterus. Self-reported history of endometriosis (OR=0.86; 95% CI: 0.71, 1.05) and uterine fibroids (OR=1.09; 95% CI: 0.98, 1.21) were not significantly associated with breast cancer risk in multivariate models. However, in analyses restricted to women who reported a hysterectomy with bilateral oophorectomy (N=991 cases, 1,081 controls), a diagnosis of endometriosis was associated with decreased breast cancer risk (OR=0.72; 95% CI: 0.54, 0.98), while uterine fibroids were positively associated with breast cancer (OR=1.36; 1.13, 1.63). Among women with intact ovaries and uterus (N=3,188 cases, 3,275 controls), neither endometriosis (OR=1.18; 95% CI: 0.79, 1.77) nor fibroids (OR=1.12; 95% CI: 0.93, 1.35) appeared to be associated with breast cancer risk. The interaction test for breast cancer risk according to a prior history of endometriosis and gynecologic surgery status was statistically significant (p-interaction=0.02); the cross-product term for fibroids and hysterectomy with bilateral oophorectomy was not (p-interaction=0.2). Conclusions: In this study, the association between self-reported endometriosis and breast cancer risk was modified by gynecologic surgery status. If replicated, the apparent protective effect of endometriosis on breast cancer risk only among women who undergo gynecologic surgery could be a result of prior treatment with agents such as NSAIDs or gonadatropin-releasing hormone agonists. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 876.