The role of ferritin in early detection of anthracycline induced cardiac dysfunction in breast cancer

Abstract Background and Aims Cancer therapy-related cardiac dysfunction (CTRCD), defined as >15% reduction in Global Longitudinal Strain (GLS), may occur before conventional biomarkers change. Troponin T and NT-proBNP are widely used. Ferritin reflects iron-mediated oxidative stress and inflammation but its predictive value in anthracycline-induced CTRCD remains unclear. This study aimed to evaluated ferritin and other biochemical and metabolic factors as predictors of GLS reduction in breast cancer patients receiving anthracyclines. Methods In 2024, from March until December, 16 breast cancer patients with CTRCD were included in this study. Ferritin, Troponin T, NT-proBNP, serum iron, TIBC, and transferrin saturation were measured before and after the third chemotherapy cycle. GLS reduction was evaluated. Multivariate regression identified predictors of subclinical cardiac dysfunction. Results The mean age of 16 patients was 51.06 ± 8.13 years, with a mean of BMI of 23.40 ± 5.70 kg/m². Hypertension was the most common comorbidity (31%), followed by diabetes mellitus (12%) and dyslipidemia (12%). Ferritin increased from 260.3 ± 174.9 ng/mL to 419.1 ± 298.1 ng/mL (p = 0.021) and showed strong correlation with GLS reduction. Troponin T rose modestly. NT-proBNP, serum iron, and TIBC remained stable. Multivariate regression identified ferritin (p = 0.042), diabetes (p = 0.006), and BMI (p = 0.041) as independent predictors. NT-proBNP and TIBC showed trends toward significance (R² = 0.95, p = 0.036). Conclusion Ferritin is a strong early marker of subclinical CTRCD. Alongside NT-proBNP, Troponin T, TIBC, BMI, and diabetes, ferritin can identify high-risk breast cancer patients early and guide preventive strategies against anthracycline cardiotoxicity.Figure 1.Partial regression plots show ferritin, Troponin T, NT-proBNP, TIBC, diabetes, and BMI associations with GLS reduction >15%, with ferritin exhibiting the strongest correlation (R² = 0.952, p = 0.036).