Abstract
1379
Objectives Discrimination between HGG and RN remains a diagnostic challenge because both entities have similar imaging characteristics on conventional MRI. Metabolic imaging, such as PET could aid in this diagnostic dilemma. In this study, we investigated the potential of F18-FDG, F18-FCho and F18-FET PET in discriminating HGG from RN.
Methods On day 15 after inoculation of F98 glioblastoma (GB) cells in the rat brain, MRI showed a contrast-enhancing tumor. μPET was performed (dynamic F18-FDG PET at conventional intervals followed by a delayed (240 min p.i.) acquisition, dynamic F18-FCho and F18-FET PET, n=4). Induction of RN was achieved by irradiating the right frontal region with 60 Gy using 3 arcs of 3*3mm. Follow-up MRI scans revealed a contrast-enhancing RN lesion 6-7 months post-irradiation and μPET was performed (n=3). The time activity curves (TACs) of the mean standard uptake value (SUV) and the lesion-to-normal tissue uptake ratio (LNR) measured during the last time frame were compared between GB and RN (see figure).
Results On conventional F18-FDG PET, mean LNR in GB (1.35±0.08) was higher compared to RN (1.04±0.08) and were significantly different (p=0.034). The difference in LNR was higher on the delayed F18-FDG scan (1.60±0.25 in GB and 1.06±0.01 in RN), however borderline significant (p=0.064) due to the small sample size (n=2). For F18-FCho, LNRs in GB and RN were 2.55±0.39 and 2.45±0.16, respectively and not significanlty different (p=1.000), meaning that 18F-FCho cannot discriminate between GB and RN. F18-FET uptake was higher in GB with a LNR of 2.19±0.16, while a LNR of 1.71±0.32 was shown in RN. F18-FET LNRs were significantly higher in GB than in RN (p=0.034).
Conclusions Based on these results, F18-FDG and F18-FET PET were able to discriminate GB from RN whereas F18-FCho was not. However, because there was visible uptake of F18-FET in RN (not shown) , a threshold will be necessary.