Abstract P3-11-04: Prediagnosis Hysterectomy and Estrogen Therapy in Relation to Mortality after Breast Cancer

Abstract

Abstract Total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAHBSO) can reduce breast cancer incidence by up to 50%, but less is known about its effects on breast cancer-specific and total mortality. Ovarian suppression is an accepted treatment for premenopausal breast cancer that may improve disease-free and overall survival. Prediagnosis TAHBSO may similarly reduce breast cancer mortality rates. However, all-cause mortality among women with breast cancer could be adversely affected by increases in deaths due to heart disease, stroke, or bone fracture that may be associated with TAHBSO. To examine associations between prediagnosis TAHBSO, estrogen therapy, and mortality after breast cancer, we analyzed data from a cohort of 9,215 women ages 50-79 who were diagnosed with invasive breast cancer during 1989-2003 in Wisconsin, Massachusetts, or New Hampshire. Causes and dates of death through 2006 were obtained from the National Death Index. To account for non-breast cancer deaths as competing events, we fit proportional cumulative incidence associated subhazards regression models for breast cancer-specific mortality. Cox proportional hazards regression was used to calculate multivariate hazard ratios (HR) and 95% confidence intervals (CI) for all-cause mortality. Cox regression models were stratified on study center, year of interview, and age at diagnosis and adjusted for body mass index, smoking status, family history of breast cancer and stage at diagnosis as potential confounders. Over an average 8.3 years of follow-up, 2,715 deaths were observed, including 1,507 breast cancer deaths. In total, 1,718 women (18.6%) reported TAHBSO prior to breast cancer diagnosis; while 5,937 women (64.4%) had an intact uterus and ovaries at diagnosis. Compared to women with an intact uterus and ovaries who never used postmenopausal hormones, the subhazard ratio for breast cancer-specific mortality was 0.92 for women who had intact uterus and ovaries and reported using estrogen therapy (95% CI: 0.66, 1.28), 0.71 for women who reported TAHBSO and never used hormones (95% CI: 0.54, 0.95), and 0.93 for women who reported TAHBSO and estrogen use (95% CI: 0.75, 1.16). In contrast, a reduction in all-cause mortality associated with TAHBSO was observed for women who reported estrogen therapy use. Compared to women with an intact uterus and ovaries who never used estrogens, the HR for all-cause mortality among estrogen users who had not undergone TAHBSO was 0.92 (95% CI: 0.73, 1.16), 0.87 for women who reported TAHBSO but no hormone use (95% CI: 0.74, 1.04), and 0.77 among women who reported TAHBSO and estrogen use (95% CI: 0.64, 0.92). Limitations to this analysis included missing information on hormone receptor status and treatment regimens. In these data, prediagnosis TAHBSO was associated with reduced breast cancer-specific mortality among women who never used hormones, and reduced all-cause mortality among estrogen users. Figure available in online version. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P3-11-04.