Role of O-(2-18F-Fluoroethyl)-l-Tyrosine PET as a Diagnostic Tool for Detection of Malignant Progression in Patients with Low-Grade Glioma

Abstract

In patients with low-grade glioma (LGG) of World Health Organization (WHO) grade II, early detection of progression to WHO grade III or IV is of high clinical importance because the initiation of a specific treatment depends mainly on the WHO grade. In a significant number of patients with LGG, however, information on tumor activity and malignant progression cannot be obtained on the basis of clinical or conventional MR imaging findings only. We here investigated the potential of O-(2-(18)F-fluoroethyl)-L-tyrosine ((18)F-FET) PET to noninvasively detect malignant progression in patients with LGG. Twenty-seven patients (mean age ± SD, 44 ± 15 y) with histologically proven LGG (WHO grade II) were investigated longitudinally twice using dynamic (18)F-FET PET and routine MR imaging. Initially, MR imaging and PET scans were performed, and diagnosis was confirmed on the basis of biopsy. Subsequently, PET scans were obtained when clinical findings or contrast-enhanced MR imaging suggested malignant progression. Maximum and mean tumor-to-brain ratios (20-40 min after injection) (TBRmax and TBRmean, respectively) of (18)F-FET uptake as well as tracer uptake kinetics (i.e., time to peak [TTP] and patterns of the time-activity curves) were determined. The diagnostic accuracy of imaging parameters for the detection of malignant progression was evaluated by receiver-operating-characteristic analyses and by Fisher exact test for 2 × 2 contingency tables. In patients with histologically proven malignant progression toward WHO grade III or IV (n = 18), TBRmax and TBRmean increased significantly, compared with baseline (TBRmax, 3.8 ± 1.0 vs. 2.4 ± 1.0; TBRmean, 2.2 ± 0.3 vs. 1.6 ± 0.6; both P < 0.001), whereas TTP decreased significantly (median TTP, 35 vs. 23 min; P < 0.001). Furthermore, time-activity curve patterns changed significantly in 10 of 18 patients (P < 0.001). The combined analysis of (18)F-FET PET parameters (i.e., changes of TBRmax, TTP, or time-activity curve pattern) yielded a significantly higher diagnostic accuracy for the detection of malignant progression than changes of contrast enhancement in MR imaging (accuracy, 81% vs. 63%; P = 0.003). Both tumor-to-brain ratio and kinetic parameters of (18)F-FET PET uptake provide valuable diagnostic information for the noninvasive detection of malignant progression of LGG. Thus, repeated (18)F-FET PET may be helpful for further treatment decisions.