It is assumed that total and visceral adiposity increase cardiovascular disease (CVD) risk among breast cancer survivors; yet, these associations have not been studied, and could differ from non-cancer populations due to the modifying effects of cancer treatment. METHODS: We studied 2,630 Stage I–III breast cancer patients without pre-existing CVD diagnosed at Kaiser Permanente (2006–2013). We quantified body composition from computed tomography scans taken at breast cancer diagnosis. The main exposures were total and visceral adiposity indices (cm2/m2), examined in tertiles. From ICD codes, we identified non-fatal stroke, coronary artery disease (CAD), and heart failure, and a composite outcome including CVD death (CVD). We estimated hazard ratios (HR) and 95% confidence intervals (CI) adjusting for age, smoking, tumor (stage, grade, and ER/PR and HER2 status) and treatment (chemotherapy and/or radiation) factors, skeletal muscle index (SMI), and body mass index (BMI) residuals. We assessed effect modification via product terms of adiposity with age (>=/<55 years), sarcopenia (SMI>=/<40 cm2/m2) and chemotherapy (yes/no). RESULTS: At diagnosis, mean (SD) age was 55 (11) years and BMI was 28 (6) kg/m2. Over a maximum follow-up of 11 years, 669 CVD events occurred. Independent of BMI and other covariates, women in the highest (v. lowest) tertile of total adiposity had a higher risk of CVD, heart failure, stroke and CAD; HRs (95%CI) were 1.45 (1.15–1.81), 1.78 (1.24–2.57), 1.89 (1.25–2.87), and 1.52 (0.83–2.79), respectively. Results were similar for visceral adiposity, and by age and sarcopenia, but were stronger for women receiving chemotherapy: e.g., the HR (95%CI) for the highest (v. lowest) tertile of total adiposity with CVD risk was 1.76 (1.33–2.33) for women who received chemotherapy versus 0.93 (0.63–1.38) for women who did not, p- interaction = 0.04. CONCLUSIONS: Women who enter a breast cancer diagnosis with greater total and visceral adiposity are at higher risk of subsequent CVD, particularly if they receive chemotherapy. Our results suggest that body composition - independent of BMI and other factors - can identify patients with high CVD risk for additional monitoring, tailored treatment plans and targeting of preventive interventions.

The following are the 17 highest scoring abstracts of those submitted for presentation at the 42nd Annual ASPO meeting held March 11–13, 2018, in New York, NY.