Molecular imaging of valvular fibroblast activation in carcinoid syndrome: a prospective observational pilot study

Abstract Introduction Neuroendrocine tumours are rare neoplasms most commonly arising from the gastrointestinal system. Half of affected patients develop carcinoid syndrome characterised by vasomotor symptoms. Of these, 50% will develop carcinoid heart disease, with major morbidity and mortality due to right heart abnormalities. These are caused by the paraneoplastic effects of secreted vasoactive substances which bind to cardiac valve receptors, inducing fibroblast activation, progressive tricuspid and pulmonary dysfunction, and right-sided heart failure. Gallium-68 Fibroblast Activation Protein Inhibitor ([68Ga]FAPI) binds to activated fibroblasts, the key effector cells of valvulopathy in carcinoid heart disease, providing a readout of fibroblast activity. We used [68Ga]FAPI to investigate activated fibroblasts in patients with carcinoid syndrome. Methods In this prospective observational cohort study, patients with a history of carcinoid syndrome with and without associated heart disease, and healthy control volunteers, underwent hybrid [68Ga]FAPI positron emission tomography and magnetic resonance imaging. [68Ga]FAPI uptake was quantitatively assessed, with the maximal standardised uptake value (SUVmax) recorded within the defined region of interest, and divided by the right atrial blood pool uptake to provide a target-to-background ratio (TBRmax). We investigated tricuspid valvular [68Ga]FAPI uptake in patients and its’ relationship with key markers of disease severity and outcomes. Results Seventeen patients with active carcinoid syndrome (68±10 years; 53% male) and 9 control volunteers (60±6 years, 56% male) were included. Three patients had a confirmed diagnosis of carcinoid heart disease based on echocardiography findings. Visual [68Ga]FAPI uptake was observed in the tricuspid valve of all 3 patients with carcinoid heart disease as well as 7 patients with no overt valvular disease on echocardiography (figure 1). Intensity of valvular (tricuspid) fibroblast activation was highest in those with carcinoid heart disease, but was also higher in patients with carcinoid syndrome without valvular disease compared to control volunteers (Figure 2a, TBRmax 1.97 ± 0.1 versus 1.65 ± 0.2 versus 1.50 ± 0.2 respectively, ANOVA p=0.02). There was no relationship between tricuspid [68Ga]FAPI uptake and N-terminal pro B-type natriuretic peptide (NT-proBNP) (r=0.345, p=0.248), but a strong correlation with urinary 5-hydroxyindoleacetic acid (5-HIAA) (Figure 2b, r=0.645, p=0.005) Conclusion In patients with carcinoid syndrome, we demonstrate for the first time that valvular fibroblast activity can be visualised and quantified using [68Ga]FAPI. Patients with and without overt tricuspid valve disease had increased [68Ga]FAPI uptake which correlated with disease severity. Further work is required to explore the clinical implications of these observations given the emergence of disease-modifying therapy that may attenuate carcinoid-associated heart disease.