Time-robust myocardial [68Ga]Ga-FAPI PET biomarker reflects aortic stenosis severity and predicts post-TAVI outcomes.

Abstract

Background: Aortic stenosis (AS) induces myocardial remodeling and fibroblast activation, yet modifiable biomarkers capable of capturing active fibrogenesis and predicting post-transcatheter aortic valve implantation (TAVI) recovery are currently scarce. Fibroblast activation protein (FAP)-targeted PET serves as a noninvasive tool to visualize activated fibroblasts in vivo. We evaluated a time-robust, blood-pool-normalized myocardial [⁶⁸Ga]Ga-FAPI PET imaging biomarker that reflects AS burden and predicts outcomes after TAVI.

Methods: Nineteen patients with severe symptomatic AS underwent [⁶⁸Ga]Ga-FAPI-04 PET/CT at 60, 70, and 120 min. Using an in-house semi-automatic pipeline, the left ventricular (LV) myocardium was segmented, and regions of elevated fibroblast activity (EFM) were delineated using a blood-pool-anchored, time-point-specific threshold. We quantified myocardial SUV_mean, blood-pool SUV_mean, and a normalized myocardium-to-blood index, TBR(EFM), and assessed associations with N-terminal pro-brain natriuretic peptide (NT-proBNP) and left-ventricular ejection fraction (LVEF). One-year outcomes (n = 11) were assessed using a predefined composite clinical response.

Results: Blood-pool SUV_mean declined from 60 to 120 min, whereas myocardial SUV_mean decreased less, yielding stable TBR(EFM) across time points (60/70/120 min: 2.2 ± 0.8, 2.1 ± 0.9, 2.3 ± 0.9; ANOVA p = 0.596). By contrast, myocardial SUV_mean fell from 3.8 ± 0.7 (60 min) to 2.1 ± 0.9 (120 min; p < 0.001). TBR(EFM) correlated with NT-proBNP at all time-points (60 min r = 0.65, p = 0.007; 120 min r = 0.72, p = 0.003), whereas SUV_mean did not (60 min p = 0.576; 120 min p = 0.109). Baseline TBR(EFM) was significantly lower in one-year responders than non-responders (1.7 ± 0.2 vs. 2.9 ± 0.9; p = 0.013), with separation present at each time point (p < 0.05). Higher baseline TBR(EFM) associated with lower reductions in NT-proBNP at one year (p < 0.05).

Conclusions: Myocardial [⁶⁸Ga]Ga-FAPI TBR may provide a time-robust index of active fibroblast signaling that relates to myocardial hemodynamic stress and stratifies one-year clinical response after TAVI. A single 60-minute acquisition with TBR quantification may be sufficient for myocardial [⁶⁸Ga]Ga-FAPI assessment. These hypothesis-generating findings require validation in larger, multicenter cohorts.