Cardiac fibroblast activation protein density indicative of cardiovascular disease and remodeling

Abstract

Abstract Background Fibroblast activation protein (FAP) plays an important role in cardiac wound healing and remodelling. Although initially developed as a theranostic ligand for metastasized cancer, FAPI tracers have recently been used to study cardiac remodelling following myocardial infarction in small animal models. Aim of our analysis was to evaluate the activity of fibroblast activation protein via (FAPI) PET-CT scans in human hearts. Methods FAPI-PET-CT scans of 229 patients of two consecutive cohorts (initial cohort: n=185; confirmatory cohort: n=44) suffering from metastasized cancer were analysed using a 17-segment model of the left ventricle. Patient characteristics included age, sex, cancer entity, body mass index, renal function, thyroid stimulating hormone, cardiovascular risk factors, previous radiation to the chest, chemotherapies received, and current cardiac medication. Multivariate regression models were created using data from the initial cohort by selecting variables according to Akaike's information criterion in a step-down approach. Linear regression models were used for signal prediction to find potential outliers with unexpectedly high signal intensities. The created models were subsequently evaluated in the confirmatory cohort for reproducibility. Results Signal intensity were significantly higher in patients with overweight, diabetes and following radiation to the chest. Our prediction model performed well in both cohorts. A focal enrichment pattern was more frequently observed in patients with known cardiovascular risk factors. Conclusions FAPI-PET-CT scans represent a new imaging modality to potentially investigate active cardiac remodelling. High signal intensities are associated with cardiovascular risk factors and metabolic disease. Furthermore, high cardiac FAPI signal intensities are suggestive of an underlying cardiac disease. Funding Acknowledgement Type of funding source: None