Cardiac Dysfunction Among Breast Cancer Survivors: Role of Cardiotoxic Therapy and Cardiovascular Risk Factors

PURPOSE Cardiac dysfunction is the leading cause of mortality among 10-year breast cancer survivors. Limited information regarding long-term risks of cardiac dysfunction after cardiotoxic therapy (anthracyclines, trastuzumab/pertuzumab, radiation) has precluded development of surveillance guidelines for the survivors. METHODS Patients with breast cancer who completed cardiotoxic therapy underwent echocardiographic screening every 2 years. New-onset cardiac dysfunction was defined as left ventricular ejection fraction (LVEF) <50% after cardiotoxic therapy initiation and included early- and late-onset cardiac dysfunction. RESULTS We evaluated 2,808 echocardiograms in 829 breast cancer survivors; the median age at breast cancer diagnosis was 54.2 years (range, 20.3-86.3); the median follow-up was 8.6 years (1.8-39.8); 39.7% received anthracyclines, 16% received trastuzumab/pertuzumab, 6.2% received both anthracyclines and trastuzumab/pertuzumab, and 38.1% received radiation alone. The cumulative incidence of cardiac dysfunction increased from 1.8% at 2 years to 15.3% at 15 years from cardiotoxic therapy initiation. Multivariable Cox regression analysis identified the following risk factors: non-Hispanic Black race (hazard ratio [HR], 2.15 [95% CI], 1.37 to 3.38), cardiotoxic therapies (anthracyclines: HR, 2.35 [95% CI, 1.25 to 4.4]; anthracyclines and trastuzumab/pertuzumab: HR, 3.92 [95% CI, 1.74 to 8.85]; reference: left breast radiation alone), selective estrogen receptor modulators (HR, 2.0 [95% CI, 1.2 to 3.33]), and precancer hypertension (HR, 3.16 [95% CI, 1.63 to 6.1]). Late-onset cardiac dysfunction was most prevalent among anthracycline- and radiation-exposed patients; early-onset cardiac dysfunction was most prevalent among patients exposed to anthracyclines and trastuzumab/pertuzumab; equal prevalence of both early- and late-onset cardiac dysfunction was observed in trastuzumab-/pertuzumab-exposed patients. Adjusted longitudinal analyses revealed an annual decline in LVEF by 0.29% ( P = .009) over 20 years from breast cancer diagnosis. CONCLUSION These findings provide evidence to support echocardiographic surveillance for several years after cardiotoxic therapy and also suggest a need to examine the efficacy of management of cardiovascular risk factors to mitigate risk.