Reduced synaptic vesicle protein 2A binding in temporal lobe epilepsy: A [11C]UCB‐J positron emission tomography study

Abstract Objective In this positron emission tomography (PET) study with [11C]UCB‐J, we evaluated synaptic vesicle glycoprotein 2A (SV2A) binding, which is decreased in resected brain tissues from epilepsy patients, in subjects with temporal lobe epilepsy (TLE) and compared the regional binding pattern to [18F]fluorodeoxyglucose (FDG) uptake. Methods Twelve TLE subjects and 12 control subjects were examined. Regional [11C]UCB‐J binding potential (BPND) values were estimated using the centrum semiovale as a reference region. [18F]FDG uptake in TLE subjects was quantified using mean radioactivity values. Asymmetry in outcome measures was assessed by comparison of ipsilateral and contralateral regions. Partial volume correction (PVC) with the iterative Yang algorithm was applied based on the FreeSurfer segmentation. Results In 11 TLE subjects with medial temporal lobe sclerosis (MTS), the hippocampal volumetric asymmetry was 25 ± 11%. After PVC, [11C]UCB‐J BPND asymmetry indices were 37 ± 19% in the hippocampus, with very limited asymmetry in other brain regions. Reductions in [11C]UCB‐J BPND values were restricted to the sclerotic hippocampus when compared to control subjects. The corresponding asymmetry in hippocampal [18F]FDG uptake was 22 ± 7% and correlated with that of [11C]UCB‐J BPND across subjects (R2 = .38). Hippocampal asymmetries in [11C]UCB‐J binding were 1.7‐fold larger than those of [18F]FDG uptake. Significance [11C]UCB‐J binding is reduced in the seizure onset zone of TLE subjects with MTS. PET imaging of SV2A may be a promising biomarker approach in the presurgical selection and evaluation of TLE patients and may improve the sensitivity of molecular imaging for seizure focus detection.