Added value of serial troponin I and N-terminal pro B-type natriuretic peptide for early detection of cardiotoxicity in breast cancer: A three-year retrospective cohort study.

Abstract

Background: Cardiotoxicity remains a major limitation in breast cancer treatment, especially with anthracyclines and radiotherapy. Early biomarkers may improve risk stratification.

Methods: This retrospective cohort study included breast cancer patients treated between 2020 and 2023 with chemotherapy and/or radiotherapy. Troponin I (TPI) and N-terminal pro B-type natriuretic peptide (NT-proBNP) were measured at baseline and at three, six, 12, 18, 24, 30, and 36 months. Cardiotoxicity was defined per European Society of Cardiology (ESC) guidelines as a ≥10% drop in left ventricular ejection fraction (LVEF) to <53% or >15% relative reduction in global longitudinal strain (GLS). Receiver operating characteristic (ROC) analysis, logistic regression, and mixed-effects models assessed associations.

Results: A total of 190 patients were included; 141 patients (74.2%) developed cardiotoxicity within 12 months. At six months, NT-proBNP (median: 212.6 pg/mL) and TPI (0.038 ng/mL) were significantly higher in the cardiotoxic group (p < 0.01). ROC area under the curve (AUC) for NT-proBNP at six months was 0.83 (95% CI: 0.77-0.89); for TPI, AUC = 0.79 (95% CI: 0.73-0.86). Multivariable regression showed both biomarkers remained independent predictors of cardiotoxicity after adjustment for GLS and cumulative anthracycline dose (NT-proBNP: OR 1.78 [95% CI 1.22-2.61], p = 0.003).

Conclusion: In this real-world breast cancer cohort, serial NT-proBNP and TPI monitoring, particularly at 3 and 6 months, provided meaningful early prognostic information for subsequent cardiotoxicity beyond baseline assessment alone. When integrated with echocardiographic surveillance, these biomarkers may support earlier identification of high-risk patients, more tailored cardio-oncology follow-up, and timely cardioprotective intervention, although prospective external validation is required before routine implementation.