Abstract C052: Premenopausal oophorectomy and survival among women with breast cancer: Evidence of effect measure modification by family history status

Abstract

Abstract In the US, bilateral oophorectomies are frequently performed before menopause for one of two reasons: 1) in concert with hysterectomy for benign gynecologic conditions or 2) prevention of breast or ovarian cancer risk and mortality. Prior literature has consistently shown a reduction in breast cancer-specific mortality among women who had premenopausal hysterectomy with bilateral oophorectomy (H +BO) prior to their breast cancer diagnosis. The present study sought to assess whether the relationship between prediagnosis premenopausal H + BO and breast cancer-specific mortality differed by family history status in women with breast cancer. It is hypothesized that women who may have prognostically different breast cancer due to having family history disproportionately benefit from premenopausal H+ BO compared to more average-risk women with no family history. This study analyzed data from Phases 1 and 2 of the Carolina Breast Cancer Study (CBCS), a population-based study of Black and White women prospectively identified in central and eastern North Carolina with newly diagnosed breast cancer between 1993 and 2001. Women with invasive breast cancer with known gynecologic surgical status were included (n=1,723). Gynecologic surgery was defined as: no surgery; hysterectomy with bilateral oophorectomy (H+BO); hysterectomy with conservation of ≥1 ovary (H + OC). Cause-specific mortality was ascertained using the National Death Index, last updated in 2016. Hazard ratios (HR) and 95% confidence intervals (CI) for breast cancer-specific mortality were estimated with Cox proportional hazard models. Models were then stratified by family history status, the self-reporting of one or more first-degree relatives with a family history of breast cancer. Models were adjusted for race, age at diagnosis, smoking, alcohol use, menopausal hormone therapy use and reproductive history factors. All participants still living at the end of follow-up were right censored. Among 1,723 women in the sample, 44% (n=759) were Black and 56% (n=964) were White. There were 836 deaths, of which 447 were from breast cancer. In this population, 74.1% (n=1,276) of women reported having no previous premenopausal gynecologic surgery, 8.8% (n=152) reported having H + BO and 17.1% (n=295) reported H + OC. Compared to women who had not had premenopausal gynecologic surgery, the overall adjusted HR for breast cancer-specific mortality associated with H+BO was 0.68 (95% CI: 0.49,0.96) and 0.90 (95% CI: 0.72,1.12) for women with H+OC. In models stratified on family history, the HR for women with H+BO was 0.11 (95% CI: 0.03,0.42) for those with family history and 0.90 (95% CI: 0.63, 1.29) for those without. The HR for women with family history who had H+ OC was 0.77 (95% CI: 0.45, 1.29) and 0.90 (95% CI: 0.63,1.29) for women without. This study suggests that the overall protective relationship observed between premenopausal hysterectomy with bilateral oophorectomy and breast-cancer specific mortality may be driven by a small subset of especially high-risk women. Citation Format: Mya L. Roberson, Whitney R. Robinson, Hazel B. Nichols, Andrew F. Olshan, Melissa A. Troester. Premenopausal oophorectomy and survival among women with breast cancer: Evidence of effect measure modification by family history status [abstract]. In: Proceedings of the Eleventh AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2018 Nov 2-5; New Orleans, LA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl):Abstract nr C052.