Deep Learning-Enhanced 3D Imaging Unveils Semaglutide Impact on Cardiac Fibrosis

Background Extensive preclinical research aims to develop novel therapeutics for myocardial fibrosis (MF), a condition marked by collagen accumulation that impairs cardiac function. MF is particularly relevant in heart failure with preserved ejection fraction (HFpEF), a growing clinical challenge with limited treatment options. However, current methods for quantifying MF in mouse models struggle to accurately capture its heterogeneous regional distribution, creating a significant barrier to reliably assessing the efficacy of therapeutics. Purpose To develop a whole-heart fibrosis imaging and deep learning (DL)-based quantification method and validate the workflow by assessing the efficacy of a glucagon-like peptide-1 receptor (GLP-1R) agonist in mouse HFpEF model. Experimental Approach By utilizing a fluorescent collagen-labelling dye, tissue clearing and 3D light sheet microscopy, we developed a high-throughput imaging platform for MF. We established DL framework to quantify perivascular and replacement fibrosis, as well as hypertrophy, in 17 left ventricular (LV) segments. The antifibrotic effects of the GLP-1R agonist semaglutide were evaluated in the db/db UNx-ReninAAV mouse model, which exhibits diabetes, kidney failure, obesity, and hypertension. Key Results Whole-heart 3D light sheet microscopy, combined with artificial intelligence, enables micrometer-resolution analysis of MF distribution in rodents. This approach allows for detailed characterization of distinct regional fibrosis patterns. Chronic semaglutide treatment significantly reduced LV hypertrophy and perivascular fibrosis but had no significant effect on replacement fibrosis. Conclusions and Implications The established 3D imaging and quantification approach provides a powerful tool for evaluating the therapeutic efficacy of antifibrotic compounds and studying the cellular and pathological mechanisms underlying cardiovascular diseases.