Multimodal metabolic imaging of cerebral gliomas: positron emission tomography with [18F]fluoroethyl-L-tyrosine and magnetic resonance spectroscopy.

Abstract

The purpose of this study was to determine the predictive value of [18F]fluoroethyl-L-tyrosine (FET)-positron emission tomography (PET) and magnetic resonance (MR) spectroscopy for tumor diagnosis in patients with suspected gliomas. Both FET-PET and MR spectroscopy analyses were performed in 50 consecutive patients with newly diagnosed intracerebral lesions supposed to be diffuse gliomas on contrast-enhanced MR imaging. Lesion/brain ratios of FET uptake greater than 1.6 were considered positive, that is, indicative of tumor. Results of MR spectroscopy were considered positive when N-acetylaspartate (NAA) was decreased in conjunction with an absolute increase of choline (Cho) and an NAA/Cho ratio of 0.7 or less. An FET lesion/brain ratio, an NAA/Cho ratio, and signal abnormalities on MR images were compared with histological findings in neuronavigated biopsy specimens. The FET lesion/brain ratio and the NAA/Cho ratio were identified as significant independent predictors for the histological identification of tumor tissue. The accuracy in distinguishing neoplastic from nonneoplastic tissue could be increased from 68% with the use of MR imaging alone to 97% with MR imaging in conjunction with FET-PET and MR spectroscopy. Sensitivity and specificity for tumor detection were 100 and 81% for MR spectroscopy and 88 and 88% for FET-PET, respectively. Results of histological studies did not reveal tumor tissue in any of the lesions that were negative on FET-PET and MR spectroscopy. In contrast, a tumor diagnosis was made in 97% of the lesions that were positive with both methods. In patients with intracerebral lesions supposed to be diffuse gliomas on MR imaging, FET-PET and MR spectroscopy analyses markedly improved the diagnostic efficacy of targeted biopsies.