Validation Of A Methodology For Left Ventricular Global Longitudinal Strain Measured With Deformation Encoding Mri To Detect Cardiotoxicity

This study sought to validate left-ventricular (LV) global longitudinal strain (GLS) derived from an MRI deformation encoding technique for detecting cancer therapy-related cardiac dysfunction (CTRCD) in breast cancer patients. Displacement Encoding with Stimulated Echoes (DENSE) cardiac MRI and 2D echocardiographic images for validation were acquired on 32 breast cancer patients at baseline and 3- and 6-month follow-ups after chemotherapy with antineoplastic agents. The DENSE images were processed with automated Artificial Intelligence segmentation techniques for chamber quantification and 3D strain analysis. The corresponding echocardiographic data were processed with 2D speckle-tracking echocardiography (2D-STE). To establish the reliability of the DENSE-AI tool, intraclass correlation coefficients (ICC) were computed between DENSE and 2DE-derived GLS and other parameters at each time point with Cronbach’s Alpha (C-α) indices. CTRCD risk analyses were conducted on the DENSE-AI and 2D-STE measured parameters with Cox proportional hazard (PH) regression models. The ICC reliability results between DENSE and 2DE-derived GLS were C-α = 0.89 at baseline, C-α = 0.93 at 3 months, and C-α = 0.94 at 6 months. Both DENSE and 2D-STE GLS worsened from baseline to the 3- and 6-month follow-ups (-19.1 ± 2.1% vs. -16.0 ± 3.1% vs. -16.1 ± 2.9%, P<0.001) (Fig. 1). Cox regression showed the 3-month DENSE (HR-per-SD: 2.1; 95% CI: 1.4-3.1, P<0.001) GLS measurements as an independent prognosticator of CTRCD occurrence. In conclusion, LV GLS measurement with the DENSE-AI tool was validated to detect CTRCD in breast cancer patients treated with antineoplastic chemotherapy.