Head-to-Head comparison of [18F]FDG, [18F]FMZ, and [18F]SynVesT-1 positron emission tomography imaging in patients with drug-resistant epilepsy.

PURPOSE: The loss of synaptic vesicle glycoprotein 2 A (SV2A) can lead to dysfunction of GABAergic neurons, but a direct comparison of SV2A and GABAA receptor densities in humans has not been assessed. This study evaluated SV2A and GABAA receptor abnormalities in patients with drug-resistant epilepsy (DRE) and compared the patterns to glucose hypometabolism.

METHODS

Eleven patients with DRE were retrospectively recruited and underwent PET imaging with [18F]fluorodeoxyglucose ([18F]FDG), [18F]Flumazenil (FMZ), and [18F]SynVesT-1. Visual assessments counted abnormal metabolic brain regions based on the Anatomical Automatic Labeling (AAL) atlas, while voxel-level analyses delineated the abnormal metabolic distributions. The relationship between hypo-metabolic distributions and the age of epilepsy onset was analyzed.

RESULTS

The hypometabolic regions in [18F]FDG PET, identified in the AAL atlas, was significantly broader than in [18F]FMZ (p = 0.0005) and [18F]SynVesT-1 (p = 0.0010) PET, with no statistical difference observed between [18F]FMZ and [18F]SynVesT-1 PET (p > 0.05). The voxel number in [18F]FDG PET was significantly higher than that of the [18F]FMZ and [18F]SynVesT-1 PET in both hypo-intensity area and severe hypo-intensity area. The ratio of the voxel number between these two area was higher for [18F]SynVesT-1 PET compared to [18F]FDG PET (p = 0.0195) and [18F]FMZ PET (p = 0.0237), and positively correlated with the age of epilepsy onset (r = 0.7397, p = 0.0145).

CONCLUSIONS

[18F]FMZ and [18F]SynVesT-1 PET images revealed a more restricted pattern of reduced uptake compared to [18F]FDG PET in DRE patients. The age of epilepsy onset correlated with a reduction in [18F]SynVesT-1 uptake but not in [18F]FMZ or [18F]FDG uptake.