Mediating effects of social and clinical factors on racial disparities in cardiovascular outcomes following anthracycline therapy in older women with breast cancer

Abstract

Introduction: Racial disparities in cardiovascular (CV) outcomes among breast cancer (BC) survivors are well-documented. However, whether such disparities persist following anthracycline therapy, a common cardiotoxic treatment, remains unclear. We aimed to quantify racial disparities in the incidence of major adverse cardiac events (MACE) and examine the role of the Social Vulnerability Index (SVI) in mediating these disparities.

Materials and methods: We conducted a retrospective cohort study using the 2007-2019 SEER-Medicare database. We included females aged ≥66 years with a new primary diagnosis of BC who received anthracyclines within one year after diagnosis. The index date was the initiation date of anthracyclines. The study exposure was race (Black vs. White), a socially constructed variable reflecting structural and institutional racism and encompassing multiple levels of health determinants. The primary outcome was the 10-year incidence of MACE. Associations between race and MACE incidence were evaluated using weighted proportional hazards models. Adjusted risk differences (aRD) were calculated as the difference in predicted 10-year risk-free probabilities between groups, based on the final adjusted models from the index date. Mediation analysis determined the contributions of SVI (county-level and theme-specific variables), BC-related characteristics, and clinical factors.

Results: Among 5571 patients (22,154 person-years) who initiated anthracycline therapy, 2128 MACE events occurred. Black patients had a higher hazard of MACE (adjusted HR 1.2; 95% CI: 1.02 to 1.33), with a 10-year aRD of 2.3% (95% CI: 0.6% to 9.7%) compared to White patients. After adjusting for right-censoring, aHR was 1.3 (95% CI: 1.1 to 1.4) with a corresponding aRD of 7.8% (95% CI: 2.9% to 12.6%). Mediation analysis showed that the total percentage of mediation was 60.7 % (95% CI: 41.3% to 94.9%), primarily driven by clinical factors, which mediated 39.6% (95% CI: 25.7% to 64.3%). Hypertension, diabetes, anemia, obesity, and calcium-channel blocker use contributed considerably to the mediation.

Discussion: MACE incidence was significantly higher in older Black patients with BC after initiating anthracycline therapy compared with their White counterparts. Clinical factors considerably mediated this racial disparity. Early interventions, enhanced cardiac surveillance, and targeted prevention strategies may help reduce inequities in CV outcomes among older women with BC receiving anthracycline therapy.

Keywords: Anthracycline; Breast cancer; Cardiovascular disease; Disparity; Health equity.