Because of the high glucose metabolism in normal brain tissue 18F-FDG is not the ideal tracer for the detection of gliomas. Methyl-11C-l-methionine (11C-MET) is better suited for imaging the extent of gliomas, because it is transported specifically into tumors but only insignificantly into normal brain. 3'-Deoxy-3'-18F-fluorothymidine (18F-FLT) has been introduced as a proliferation marker in a variety of neoplasias and has promising potential for the detection of brain tumors, because its uptake in normal brain is low. Additionally, the longer half-life might permit differentiation between transport and intracellular phosphorylation.
Methods: PET of 18F-FLT and 11C-MET was performed on 23 patients (age range, 20-70 y) with histologically verified gliomas of different grades. On all patients, conventional MRI was performed, and 16 patients additionally underwent contrast-enhanced imaging. Images were coregistered, and the volumes of abnormality were defined for PET and MRI. Uptake ratios and standardized uptake values (SUVs) of various tumors and regions were assessed by region-of-interest analysis. Kinetic modeling was performed on 14 patients for regional time-activity curves of 18F-FLT from tumorous and normal brain tissue.
Results: Sensitivity for the detection of tumors was lower for 18F-FLT than for 11C-MET (78.3% vs. 91.3%), especially for low-grade astrocytomas. Tumor volumes detected by 18F-FLT and 11C-MET were larger than tumor regions displaying gadolinium enhancement (P<0.01). Uptake ratios of 18F-FLT were higher than uptake ratios of 11C-MET (P<0.01). Uptake ratios of 18F-FLT were higher in glioblastomas than in astrocytomas (P<0.01). Absolute radiotracer uptake of 18F-FLT was low and significantly lower than that of 11C-MET (SUV, 1.3+/-0.7 vs. 3.1+/-1.0; P<0.01). Some tumor regions were detected only by either 18F-FLT (7 patients) or 11C-MET (13 patients). Kinetic modeling revealed that 18F-FLT uptake in tumor tissue seems to be predominantly due to elevated transport and net influx. However, a moderate correlation was found between uptake ratio and phosphorylation rate k3 (r=0.65 and P=0.01 for grade II-IV gliomas; r=0.76 and P<0.01 for grade III-IV tumors).
Conclusion: 18F-FLT is a promising tracer for the detection and characterization of primary central nervous system tumors and might help to differentiate between low- and high-grade gliomas. 18F-FLT uptake is mainly due to increased transport, but irreversible incorporation by phosphorylation might also contribute. In some tumors and tumor areas, 18F-FLT uptake is not related to 11C-MET uptake. In view of the high sensitivity and specificity of 11C-MET PET for imaging of gliomas, it cannot be excluded that 18F-FLT PET was false positive in these areas. However, the discrepancies observed for the various imaging modalities (18F-FLT and 11C-MET PET as well as gadolinium-enhanced MRI) yield complementary information on the activity and the extent of gliomas and might improve early evaluation of treatment effects, especially in patients with high-grade gliomas. Further studies are needed, including coregistered histology and kinetic analysis in patients undergoing chemotherapy.