• AL
• diagnosis
• PET/CT imaging
• [18F]-Florbetaben
• cardiac amyloidosis

Background
Amyloidoses are a group of systemic disorders characterized by extracellular deposition of amyloid insoluble fibrils, deriving from proteins encoded by mutated genes or from the misfolding of a normal protein. Cardiac involvement in amyloidosis (CA) is a major determinant of clinical presentation and may be present in primary immunoglobulin light-chain-derived amyloidosis (AL) and transthyretin-related amyloidosis (ATTR), either due to misfolded wild-type TTR or to mutated TTR.
AL-CA is related to a clonal plasma cell disorder, it has the worse prognosis and its diagnosis is often delayed.
The presentation of AL-CA and ATTR-CA is very similar; 99mTc-diphosphonates scintigraphy provides a non-invasive confirmation of ATTR in selected cases, whereas it is not useful for the diagnosis of AL, which currently relies on the histological demonstration of amyloid fibrils, and requiring invasive maneuvers, such as endomyocardial biopsy. Some PET radio-pharmaceuticals for the detection of beta-amyloid deposits within the brain have shown to be able to detect cardiac amyloid deposits.
Methods
Twenty patients with a biopsy-proven diagnosis of CA (10 AL and 10 ATTR) and ten patients with initial clinical suspicion of CA and then diagnosed with non-CA conditions, were prospectively enrolled and underwent PET/CT with [18F]-Florbetaben. A dynamic cardiac scan was performed in list-mode for 60 minutes, preceded by a low-dose CT performed throught the heart for attenuation correction. A static cardiac scan (delayed-static) was performed 110 minutes after the radio-pharmaceutical injection. Static cardiac scans were dichotomically (positive or negative) evaluated by two readers blinded to the patient diagnosis. Raw data of the cardiac dynamic scans were exported to be analyzed by the kinetic tool (PKIN) on PMOD software. SUVmean and molecular volume (MV) were calculated within each VOI (region of interest). [18F]-Florbetaben cardiac time/activity curve and plasma time-activity curve were obtained using dynamic reconstructions and by manually drawing, respectively, a myocardial VOI in a transaxial slice, and a circular VOI of one centimeter within the left atrial cavity, in the same slice. Three whole-heart static images were reconstructed from the dynamic list-mode acquisition between: 5 and 15 minutes (early-static), 30 and 40 minutes (intermediate-static), 50 and 60 minutes (late-static). Whole-heart SUVmean, plasma SUVmean (background) and heart-to background ratios, were calculated on early, intermediate, late and delayed scans. The concordance among the two different interpretations was performed using Kappa statistics and standard error; differences among groups were tested by Kruskal-Wallis H test or by Chi-square test. Statistical analysis was performed using SPSS.
Results
SUVmean values evaluated in the late-static scans were significantly different: greater in AL than in ATTR patients (p<0.001) and non-CA patients (p<0.001), and similar between ATTR and non-CA patients (p=0.222). Similar results were obtained at delayed-static acquisitions. A similar pattern was observed for H/background ratio; the MV calculated at 15, 40, 60 and 110 minutes demonstrated a progressive decline in ATTR and in non-CA patients, while it remained stable in AL patients.
Conclusions
The delayed uptake of [18F]-Florbetaben can discriminate cardiac involvement in AL patients, thus [18F]-Florbetaben PET/CT may represent a non-invasive tool for the diagnosis of AL cardiac amyloidosis.