Rationale: This study compared 68Ga-FAPI-46 PET/CT, 18F-fluorodeoxyglucose (FDG) PET/CT, and contrast-enhanced MRI (CE-MRI) for glioma imaging, classification, and recurrence detection and explored PET parameters and molecular pathological profiles. Methods: Between June 2020 and June 2024, we prospectively enrolled patients with space-occupying lesions in the brain or previously treated gliomas. All patients underwent sequential CE-MRI, 68Ga-FAPI-46, and 18F-FDG PET/CT. Diagnostic accuracy was assessed based on a reference standard, and PET parameters were analysed for correlations with WHO grading and molecular characteristics. Results: Forty-eight patients (median age, 51 years; 32 men) with 40 confirmed gliomas were enrolled. For primary tumour diagnosis, the sensitivity of 68Ga-FAPI-46 PET/CT was equivalent to CE-MRI (95% vs. 100%, P = 0.99) and 18F-FDG PET/CT (95% vs. 77%, P = 0.13). 68Ga-FAPI-46 uptake was higher in grade IV than in grade I-II gliomas (5.03 vs. 1.14, P = 0.02). 68Ga-FAPI-46 PET/CT showed significantly higher maximum standardized uptake value and tumour-to-background ratio (TBR) in recurrent tumours than in treatment-related changes and demonstrated favourable sensitivity and specificity for detecting recurrent gliomas, though not significantly superior to 18F-FDG PET/CT (sensitivity: 100% vs. 85%, P = 0.48; specificity: 100% vs. 80%, P = 0.99) and CE-MRI (sensitivity: 100% vs. 100%, P = NA; specificity: 100% vs. 40%, P = 0.25). Glial fibrillary acidic protein-mutant gliomas exhibited higher 68Ga-FAPI-46 uptake than wild-type gliomas. Conclusion: 68Ga-FAPI-46 PET/CT outperformed 18F-FDG and CE-MRI in diagnosing glioma recurrence, although the results were not statistically significant. For primary glioma diagnosis, 68Ga-FAPI-46 PET/CT, despite having a better TBR, did not surpass 18F-FDG PET/CT and CE-MRI in terms of sensitivity and specificity. However, 68Ga-FAPI-46 PET/CT is superior to 18F-FDG for visualizing and classifying gliomas.
Keywords: 18F-FDG; 68Ga-FAPI-46 PET/CT; CE-MRI; recurrent detection.
© The author(s).