Abstract Background Anthracycline-based chemotherapy is widely used in cancer treatment, but its effectiveness is marred by significant cardiotoxicity, which can adversely affect prognosis. The Heart Failure Association (HFA) and the International Cardio-Oncology Society (ICOS) developed a risk score to predict cancer therapy-related cardiovascular toxicity, primarily based on left ventricular ejection fraction (LVEF) reduction. Recent guidelines recommend expanding this risk score to include global longitudinal strain (GLS) reduction for earlier detection of cardiac dysfunction, but the applicability of this model in Asian populations remains uncertain. Purpose This study aims to evaluate the predictive accuracy of the HFA-ICOS Risk Score in identifying anthracycline-induced cardiotoxicity in Asian breast cancer patients using both LVEF and GLS criteria Methods This prospective cohort study included 70 breast cancer patients undergoing chemotherapy, with both anthracycline-related and non-anthracycline therapies. Patients were evaluated using the HFA-ICOS Risk Score (low to very high) and monitored for cardiotoxicity, defined by a GLS decline 15%. The study assessed the predictive accuracy of the risk score, including sensitivity, specificity, PPV, NPV, and accuracy. The relationship between clinical factors (age, diabetes, hypertension, obesity, etc.) and cardiotoxicity was also analyzed, using the area under the curve (AUC) to assess predictive performance. Results The HFA-ICOS Risk Score demonstrated moderate predictive accuracy for GLS decline (AUC: 0.652, 95% CI: 0.52-0.78). Among the analyzed risk factors, coronary artery disease (CAD) showed the highest predictive value (AUC: 0.949, 95% CI: 0.87-1.0), followed by hypertension (AUC: 0.855, 95% CI: 0.7-1.0) and obesity (AUC: 0.849, 95% CI: 0.75-0.94). Hypertension had high specificity (94.4%) but relatively low sensitivity (24.24%) in predicting cardiotoxicity, while obesity showed a balanced predictive performance (sensitivity 39.39%, specificity 81.08%). Diabetes mellitus also emerged as a significant risk factor, with moderate predictive accuracy (AUC: 0.768, 95% CI: 0.609-0.93). Conversely, cancer-related therapy alone showed limited predictive ability (AUC: 0.46, 95% CI: 0.31-0.61), suggesting the need for integrating multiple risk factors for better assessment. Conclusion The HFA-ICOS Risk Score offers moderate predictive accuracy for anthracycline-induced cardiotoxicity in Asian breast cancer patients. CAD, hypertension, and obesity were significant predictors, underscoring the need for personalized risk stratification. Enhanced predictive models integrating GLS-based assessment could improve early monitoring and interventions.Characteristic Population Predictors of Cardiotoxicity
Lerista et al. (Fri,) studied this question.