Abstract Introduction breast cancer is the most common cancer among women, and in those undergoing chemotherapy, a potential adverse event is cardiotoxicity. Various clinical scores and echocardiographic parameters have been proposed, but predicting left ventricular dysfunction remains challenging. Purpose this study aims to identify baseline pre-chemotherapy echocardiographic parameters that can predict cardiotoxicity. Methods consecutive patients diagnosed with non-metastatic breast cancer and treated with chemotherapy (including anthracycline, anti-HER2, taxanes or hormonal therapy) were prospectively enrolled and followed for one year with clinical and echocardiographic assessments at baseline and every three months. The diagnosis of cancer therapy-related cardiac dysfunction (CTRCD) was based on the 2022 ESC Guidelines. The case series was therefore divided based on the occurrence of mild CTRCD or not. For descriptive purpose we used mean and confidence intervals; median and interquartile ranges when appropriated. We used Student's T-test or Mann-Whitney U for within-group differences, paired T-test or Wilcoxon-Signed-Rank Test for repeated controls. CTRCD onset over time was evaluated using the Kaplan-Meier method. ROC analysis was used to identify cut-off values able to define at risk patients. Statistical significance was setted at p < 0.05. Results we examined 138 patients with an average age of 53.52 years. 58 patients developed mild CTRCD, (18 within the first trimester, 26 between 3-6 months, 8 between 6-9 months, and 6 between 9-12 months). (Picture 1). Parameters like left ventricle global longitudinal strain (GLS), left ventricle eiection fraction (LVEF), early peak atrial longitudinal strain (PALS), peak atrial contraction (PACS) and left atrial volume were evaluated. These parameters showed significant deterioration over time in the overall population; PACS and PALS changed early at 3 months, but nothing was significantly correlated with mild CTRCD diagnosis. A significant difference between the two groups was observed only for PALS at 6 months and PACS at 12 months. At the ROC analysis only GLS showed clinical utility in predicting mild CTRCD with an AUC of 0.83 (Picture 2), and the best cut-off value was -21 (sensitivity 0.724, specificity 0.850, positive predictive value 0.778, negative predictive value 0.810). Conclusions echocardiographic parameters worsened over time in chemotherapy patients. Atrial strain showed significant differences between groups. Basal atrial parameters did not predict CTRCD in our study, but literature supports the relevance of left atrial strain for predicting CTRCD. A baseline GLS value of ≤ -21 may indicate a higher likelihood of mild cardiotoxicity, probably because chemotherapy effect could cause more easily a reduction > 15% in GLS without reduction in LVEF when the baseline GLS is more negative. These findings require confirmation from larger studies.
Read full abstract