Diagnostic value of [¹⁸F]FAPI-04 PET/CT in patients with monofocal liver lesions suspicious for hepatocellular carcinoma: a prospective comparison with LI-RADS-based contrast-enhanced CT/MRI.

Abstract

PURPOSE: To evaluate the diagnostic performance of [18F]FAPI-04 PET/CT combined with LI-RADS-based contrast-enhanced CT/MRI for suspected monofocal HCC, with particular focus on characterizing indeterminate observations (LR-3, LR-4, and LR-M) and its impact on clinical management recommendations. METHODS: Patients with monofocal liver lesions suspicious for HCC who underwent Ce-CT/MRI and [18F]FAPI-04 PET/CT were prospectively enrolled. For each lesion, LI-RADS category was assigned based on Ce-CT/MRI, and [18F]FAPI-04 uptake was assessed using the maximum standardized uptake value (SUVmax) and lesion-to-background ratio (LBR). Histopathology or ≥ 12-month imaging follow-up served as the reference standard. Exploratory optimal cutoff values for SUVmax and LBR were determined by receiver operating characteristic (ROC) curve to distinguish HCC from other hepatic observations. The diagnostic performances of LI-RADS-based Ce-CT/MRI, [18F]FAPI-04 PET/CT, and their combination were compared using McNemar’s test. RESULTS: The cohort included 63 malignant (50 HCCs, 13 non-HCC malignancies) and 16 benign lesions. Both SUVmax and LBR were significantly higher in the LR-M category compared with LR-3, LR-4, and LR-5 (all P < 0.01), whereas uptake levels appeared comparable between LR-4 and LR-5. ROC analysis identified optimal cut-off values of SUVmax ≥3.7 for differentiating HCC from benign lesions (AUC 0.773) and LBR ≥ 6.9 for differentiating non-HCC malignancies from HCC (AUC 0.903). For LR-3/4 lesions < 30 mm, [18F]FAPI-04 PET/CT showed 77.8% sensitivity and 88.9% specificity for HCC diagnosis. For LR-M lesions < 30 mm, [18F]FAPI-04 PET/CT achieved 100% sensitivity and 100% specificity for identifying non-HCC malignancies. Overall, [18F]FAPI-04 PET/CT showed comparable diagnostic performance to LR-5 in sensitivity (66.0% vs. 62.0%), specificity (86.2% vs. 93.1%), and accuracy (both 73.4%). The combination of [18F]FAPI-04 PET/CT with LI-RADS significantly improved HCC detection sensitivity to 86.0% while maintaining 100% specificity. CONCLUSIONS: Preliminary results suggest that combining [18F]FAPI-04 PET/CT with LI-RADS-based Ce-CT/MRI significantly improves diagnostic accuracy for monofocal liver lesions suspicious for HCC. TRIAL REGISTRATION: clinicaltrials.gov: NCT 05485792.