Background: PET represents a valuable tool for glioma imaging. In addition to amino acid tracers such as 18F-FET, PET targeting the 18-kDa mitochondrial translocator-protein (TSPO) is of high interest for high-grade glioma (HGG) imaging due to its upregulation in HGG cells. 18F-GE-180, a novel TSPO ligand, has shown a high target-to-background contrast in HGG. Therefore, we intra-individually compared its uptake characteristics to dynamic 18F-FET PET and contrast-enhanced MRI in patients with HGG.
Methods: Twenty HGG patients (nine IDH-wildtype, 11 IDH-mutant) at initial diagnosis (n = 8) or recurrence (n = 12) were consecutively included and underwent 18F-GE-180 PET, dynamic 18F-FET PET, and MRI. The maximal tumour-to-background ratios (TBRmax) and biological tumour volumes (BTV) were evaluated in 18F-GE-180 and 18F-FET PET. Dynamic 18F-FET PET analysis included the evaluation of minimal time-to-peak (TTPmin). In MRI, the volume of contrast-enhancement was delineated (VOLCE). Volumes were spatially correlated using the Sørensen-Dice coefficient.
Results: The median TBRmax tended to be higher in 18F-GE-180 PET compared to 18F-FET PET [4.58 (2.33-8.95) vs 3.89 (1.56-7.15); p = 0.062] in the overall group. In subgroup analyses, IDH-wildtype gliomas showed a significantly higher median TBRmax in 18F-GE-180 PET compared to 18F-FET PET [5.45 (2.56-8.95) vs 4.06 (1.56-4.48); p = 0.008]; by contrast, no significant difference was observed in IDH-mutant gliomas [3.97 (2.33-6.81) vs 3.79 (2.01-7.15) p = 1.000]. Only 5/20 cases showed higher TBRmax in 18F-FET PET compared to 18F-GE-180 PET, all of them being IDH-mutant gliomas. No parameter in 18F-GE-180 PET correlated with TTPmin (p > 0.05 each). There was a tendency towards higher median BTVGE-180 [32.1 (0.4-236.0) ml] compared to BTVFET [19.3 (0.7-150.2) ml; p = 0.062] with a moderate spatial overlap [median Sørensen-Dice coefficient 0.55 (0.07-0.85)]. In MRI, median VOLCE [9.7 (0.1-72.5) ml] was significantly smaller than both BTVFET and BTVGE180 (p < 0.001 each), leading to a poor spatial correlation with BTVGE-180 [0.29 (0.01-0.48)] and BTVFET [0.38 (0.01-0.68)].
Conclusion: PET with 18F-GE-180 and 18F-FET provides differing imaging information in HGG dependent on the IDH-mutational status, with diverging spatial overlap and vast exceedance of contrast-enhancement in MRI. Combined PET imaging might reveal new insights regarding non-invasive characterization of tumour heterogeneity and might influence patients' management.
Keywords: 18F-FET; 18F-GE-180; Amino acid; High-grade glioma; MRI; TSPO.