Abstract
Introduction: Systemic light-chain (AL) amyloidosis is a heterogeneous group of disorders caused by misfolding and aggregation proteins. The diagnosis is often delayed due to nonspecific clinical symptoms, making the early intervention very important. 18F-florbetapir PET/CT, a novel amyloid-targeting radiotracer, has been shown potential for non-invasive amyloid detection. We evaluated the diagnostic accuracy and potential prognostic significance of amyloid burden quantified by 18F-florbetapir PET/CT in systemic AL amyloidosis patients.
Methods: Thisretrospective study was on a cohort of suspected AL amyloidosis patients at the First Affiliated Hospital of Wenzhou Medical University (2023–2025). All patients underwent 18F-Florbetapir PET/CT imaging followed by biopsy-based histopathological confirmation(immunohistochemical staining, mass spectrometric analysis, or immune electron microscopy). PET/CT scans were classified as positive or negative and quantified by measuring the maximum standardized uptake value (SUVmax). Diagnostic accuracy was assessed by sensitivity, specificity, PPV, NPV, and Cohen's κ. In PET/CT-positive cases, amyloid burden was quantified as 18F-florbetapir SUVmax. ROC curve analysis determined the accuracy of SUVmax in differentiating true positives (TP) from false positives (FP). In biopsy-confirmed AL amyloidosis patients, Spearman correlations assessed relationships between SUVmax and clinical markers (dFLC, NT-proBNP, creatinine clearance rate, Mayo 2004 and 2012 stages). Differences in SUVmax across Mayo stage groups were further evaluated using the Kruskal–Wallis test.
Results: Among the 51 patients enrolled (median age: 56 years; 72.5% male), all patients underwent 18F-Florbetapir PET/CT. 37 were PET/CT-positive and 14 were PET/CT-negative. Histopathology confirmed 31 true positives, 6 false positives, and 14 true negatives, and no false negatives. The sensitivity was 100%, specificity 70%, PPV 83.8%, NPV 100%, and Cohen's κ=0.739. Within the PET/CT-positive subgroup, SUVmax showed moderate ability to identify the true or false positives (AUC=0.758; p=0.048; optimal cutoff=2.90; sensitivity=71%, specificity=83.3%). Among 31 confirmed AL amyloidosis cases (median age: 66 years; 70.9% male), all patients exhibited amyloid deposition. SUVmax correlated positively with NT-proBNP (ρ=0.458, p=0.013), Mayo 2012 stage (ρ=0.578, p<0.01), Mayo 2004 stage (ρ=0.405, p=0.024), and creatinine clearance rate (ρ=0.427, p=0.018), but not with dFLC levels. Consistently, Kruskal–Wallis testing showed significant differences in SUVmax across Mayo 2012 stages (H=9.82, p=0.020), with a similar but non-significant trend for Mayo 2004 stages (H=6.85, p=0.077).
Conclusions: We validated 18F-Florbetapir PET/CT demonstrated excellent sensitivity and reliable diagnostic accuracy in systemic AL amyloidosis, making it possible for non-invasive visualization and quantification of amyloid deposits. It significantly enhanced targeted biopsy guidance, potentially reducing invasive procedure risks and diagnostic delays. SUVmax was highly associated with systemic amyloid burden and clinical stage, and could be used for disease risk stratification. Future multicenter prospective studies are needed to validate the promising findings and define the clinical role of 18F-florbetapir PET/CT in AL amyloidosis.
