Glucose Loading Enhances the Value of 18F-FDG PET/CT for the Characterization and Delineation of Cerebral Gliomas.
Dongwoo KimHae Young KoSangwon LeeEun Young LeeSujin RyuSeon Yoo KimJee-In ChungMisu LeeJu Hyung MoonJong Hee ChangMijin YunPublished in: Cancers (2020)
This study aimed to assess how to enhance the value of 18F-Fluorodeoxyglucose (FDG) PET/CTs for glioma grading and better delineation of the tumor boundary by glucose loading. In mouse models of brain tumor using U87MG cells, 18F-FDG-PET images were obtained after fasting and after glucose loading. There was a significant difference in the tumor-to-normal cortex-uptake ratio (TNR) between the fasting and glucose-loading scans. 14C-2-Deoxy-D-glucose (14C-DG) uptake was measured in vitro using U87MG, U373MG and primary neurons cultured with different concentrations of glucose. The tumor-to-neuron ratio of 14C-DG uptake increased with up to 10 mM of glucose. Finally, 10 low-grade and 17 high-grade glioma patients underwent fasting and glucose loading 18F-FDG PET/CT and the TNR was compared between scans. The effect of glucose loading was significant in high-grade but not in low-grade gliomas. The receiver operating characteristic curve analyses with a cut-off TNR of 0.81 showed a higher area under the curve after glucose loading than fasting for differentiating low-grade versus high-grade gliomas. In addition, the glucose loading PET/CT was more useful than the fasting PET/CT for the discrimination of oligodendrogliomas from IDH-wildtype glioblastomas. Glucose loading resulted in a greater reduction in 18F-FDG uptake in the normal cortex than in tumors, which increases the usefulness of 18F-FDG PET/CT for grading.
Keyphrases
- high grade
- low grade
- pet ct
- blood glucose
- positron emission tomography
- computed tomography
- insulin resistance
- pet imaging
- end stage renal disease
- chronic kidney disease
- magnetic resonance imaging
- induced apoptosis
- glycemic control
- spinal cord
- peritoneal dialysis
- mouse model
- oxidative stress
- signaling pathway
- functional connectivity
- deep learning
- endothelial cells
- pi k akt
- weight loss
- spinal cord injury
- convolutional neural network