Cardiac CARE study: How the ESC Cardio-Oncology guidelines fo 2022 affect the study population

Abstract

Abstract Introduction The new 2022 ESC Guidelines in Cardio-Oncology focused on cancer therapy-related cardiac dysfunction (CT-RCD), with the formal introduction of biomarkers in the classification and risk stratification of CTR-CD in addition to structural imaging . We here, investigated how these changes affect the classification of patients in the Cardiac CARE study and the value of high sensitivity troponin I (hSTrop-I). Methods The Cardiac CARE study prospectively recruited 191 patients who underwent anthracycline therapy for breast cancer or non-Hodgkin’s lymphoma. All patients underwent regular hSTrop-I levels throughout their anthracycline therapy as well as a baseline and 6-month cardiac magnetic resonance scans including assessment of left ventricular ejection fraction and global longitudinal strain. A post-hoc analysis assessed the association between hSTrop-I and the development of cardiac dysfunction (defined as a reduction in left ventricular ejection fraction <50% or a reduction in GLS by >15%) using logistic regression with data corrected for age and gender. In addition, receiver operator curves (ROC) were drawn and area under the curve (AUC) analysis was performed to assess the value of hSTrop-I as an addition to the classification model. Finally, to compare our data with the new guidelines, a confusion matrix was performed. Results 11 patients were diagnosed as having cardiotoxicity in the Cardiac CARE study but using the 2022 ESC guidelines in Cardio-Oncology, 53 patients were re-classified into the ‘asymptomatic mild cardiotoxicity’ group (95% CI of 0.5162 and 0.6823; kappa value 0.157). Using a logistic regression model, a difference in hSTrop-I measured at the first cycle of anthracycline therapy had an odds ratio of 1.37 (95% CI 0.37-5.10) which when corrected for age dropped to 1.18 (95% CI of 0.28-5.01) and with combined age and sex correction, fell further, to 0.83 (95% CI of 0.12 and 5.56). A ROC analysis looking at the performance of hSTrop-I at predicting cardiac dysfunction demonstrated an area under the curve of 0.605 and when corrected for age and sex, was 0.640. Conclusion The addition of cardiac biomarkers to the new classification, significantly reclassified 53 of 191 (27.7%) of patients who were previously normal, to having asymptomatic mild cancer therapy-related cardiac dysfunction, mostly due to the inclusion of hSTrop I. HSTrop-I prior to anthracycline treatment was not a robust predictor of the development of anthracycline-induced cardiotoxicity in our cohort of patients. Of note, the Cardiac CARE patients are a low-risk group of patients and follow-up data is limited to 6 months.