Common malignant brain tumors: can 18F-FDG PET/CT aid in differentiation?

Abstract

Objectives: The objectives of this study were to evalute the metabolic characteristics of common malignant space-occupying lesions (SOL) of the brain and to determine the utility of fluorine-18-fluorodeoxyglucose (F-FDG) PET/CT in differentiating between the common types of malignant brain SOL.

Patients and methods: All patients with brain SOL who were referred for an F-FDG PET/CT scan by a multidisciplinary team were included in this retrospective study. The metabolic characteristics of the brain lesions in the form of maximum standardized uptake value (SUVmax) along with tumor-to-background activity ratios were determined and differences were compared using nonparametric statistical tests. Histopathological confirmation was used as the gold standard in all patients. Receiver operating characteristic curve analysis was used to find the optimal SUVmax cutoff to differentiate the tumor types.

Results: Glioblastoma multiforme (GBM; n=30), lymphoma (n=25), and metastases (n=46) accounted for most malignant tumors (95.2%). Lymphomas showed a significantly high metabolic uptake (median SUVmax=20.3, range: 8.1-46.3) compared with GBM ( median SUVmax=10.3, range: 2.6-21.7) and metastases (median SUVmax=11.5, range: 2.9-19.6) (P=0.00). The tumor-to-background activity ratios for lymphomas were also significantly higher. There was an overlap in the metabolic uptake of GBM and metastases, with no significant difference between their SUVmax values (P=0.245). A SUVmax more than 15.5 showed an 84% sensitivity and an 80% specificity to diagnose lymphomas (area under the curve=0.876, P=0.00). Four patients with brain lymphoma had extracranial disease on F-FDG PET. Lung cancer was the most common primary malignancy in patients with brain metastases.

Conclusion: Central nervous system lymphomas can be differentiated from GBM and metastases by their higher metabolic activity. In addition, F-FDG PET/CT can potentially impact therapeutic decisions by detecting primary malignancy in patients with metastatic brain lesions and extracranial disease sites in patients with brain lymphoma.