Abstract C078: Racial/ethnic disparities in breast cancer survival by subtypes: The role of obesity

Abstract

Background: Previous studies have established a strong association of higher body mass index (BMI) with increased risk of postmenopausal breast cancer (BC) but with decreased risk for premenopausal BC. Additionally, a handful of studies have found higher BMI at diagnosis to be associated with increased mortality for both ER positive and negative BC subtypes, but associations have been inconsistent. We re-examined associations of BMI with tumor subtypes and BC-specific death, and whether the disproportionate prevalence of obesity in non-Hispanic (nH) black women mediated racial disparities in survival. Methods: We included 6884 breast cancer cases from the MCBCR from 2001-2014. Continuous BMI was categorized as under/normal weight (10<bmi<bmi<bmi35). We ran age and race adjusted logistic regression models for associations of BMI as a nominal categorical variable with ER subtype overall and by menopausal status. We then modeled the hazard of BC death in Cox proportional hazard regression against race/ethnicity (nH Black versus nH White) while adjusting for age at diagnosis, before and after including BMI in our models. Survival analysis models were stratified on menopausal status and ER subtype. All analyses were conducted using STATA, v.15 (Stata Corp LLC, College Station, Texas). Results: nH Black women were less likely to be normal weight (8% vs. 15%) and more likely to be overweight (17% vs. 11%) compared to nH Whites, but there was no disparity in morbid obesity in this sample (50% vs. 50%). Higher BMI was not associated with prevalence of ER subtypes. Morbid obesity was associated with BC-specific death for both pre-menopausal (HR=3.55, 95% CI: 1.62, 7.79) and postmenopausal women (HR=2.02, 95% CI: 1.30, 3.17) and for both ER positive (HR=2.58, 95%CI: 1.55, 4.31) and negative tumor subtypes (HR=0.73, 95%CI: 0.93, 3.23). There was no evidence that BMI mediated BC survival disparities in any of the strata defined by menopausal status or ER subtype. However, the association of morbid obesity with BC death was much stronger for nH White (HR=3.81, 95%CI: 2.67, 5.44) than nH Black patients (HR=1.51, 95%CI: 0.98, 2.33). Among ER positive BC patients, the association of morbid obesity with BC death was even stronger for nH White patients (HR=4.54, 95%CI: 2.81, 7.32) but disappeared among nH Black patients (HR=1.10, 95%CI: 0.63, 1.93). Conclusions: In this clinical population, morbid obesity (defined here as a BMI>35) is very prevalent (50%) and it has negative implications for survival from a BC diagnosis regardless of menopausal status or subtype. Contrary to our expectation, the association of morbid obesity with BC death was considerably larger for nH White than for nH Black patients. Patients who are morbidly obese should be targeted for more detailed follow-up to improve our understanding of the mechanisms involved in the role of morbid obesity in BC death and how these mechanisms may differ for by race. </bmi Citation Format: Carola T Sánchez Díaz, Garth H Rauscher. Racial/ethnic disparities in breast cancer survival by subtypes: The role of obesity [abstract]. In: Proceedings of the Twelfth AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2019 Sep 20-23; San Francisco, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl_2):Abstract nr C078.