Reduced SV2A binding in the seizure onset zone in temporal lobe epilepsy patients - A PET study with ¹¹C-UCB-J

Objectives: We recently reported the development of ¹¹C-UCB-J as a PET ligand for synaptic vesicle glycoprotein 2A (SV2A) in nonhuman primates and humans (Finnema et al., 2016; Nabulsi et al., 2016). SV2A is located on the membrane of synaptic vesicles, is decreased in resected brain tissue from epilepsy patients, and is the site of action of the antiepileptic drugs levetiracetam and brivaracetam (Löscher et al., 2016). Here we evaluated ¹¹C-UCB-J binding in subjects with temporal lobe epilepsy (TLE) and compared the regional binding pattern to ¹⁸F-FDG.

Methods: Ten patients (6 males and 4 females, 39 ± 12 years of age) with TLE and mesial temporal sclerosis (MTS) were examined in the HRRT PET system with ¹¹C-UCB-J. Regional BP_ND values were estimated with arterial input functions using the one-tissue compartment model (n = 9). In one subject, arterial data were not available and BP_ND values were obtained using SRTM2. The centrum semiovale was used as a reference region for non-displaceable binding. Subjects were evaluated with ¹⁸F-FDG on either the HRRT (n = 5), or a Discovery PET/CT system (n = 5). ¹⁸F-FDG uptake was quantified using mean radioactivity values corresponding to 30-60 min or 50-60 min post ¹⁸F-FDG injection in the HRRT or Discovery, respectively. Regional asymmetry indices were calculated as: 200% × [(ipsilateral - contralateral) / (ipsilateral + contralateral)].

Results: In all subjects, there was a clear reduction in ¹¹C-UCB-J BP_ND values in the epileptogenic temporal lobe when compared to the contralateral side. The unilateral binding of ¹¹C-UCB-J was region specific, with very limited asymmetry in other brain regions, e.g., 3 ± 6% in the fusiform gyrus. In nine of the subjects, the asymmetry was predominantly located in the hippocampus, with BP_ND asymmetry indices of -50 ± 39% (range: -143% to -12%). This regional asymmetry was much larger than that found in five control volunteers (11 ± 5%, 200% × [(right - left) / (right + left)] (Finnema et al., 2016). The corresponding asymmetry in ¹⁸F-FDG uptake in the hippocampus of the TLE patients was -17 ± 6% (range: -29% to -11%) and the relative magnitude was consistent with ¹¹C-UCB-J across subjects (Pearson’s correlation coefficient = 0.84). In one patient, hypometabolism of ¹⁸F-FDG was located predominantly in the lateral temporal lobe, which was consistent with lateral reductions in ¹¹C-UCB-J binding.

Conclusion: ¹¹C-UCB-J binding is reduced in the seizure onset zone of patients with TLE, and is consistent with previously reported SV2A loss in resected brain tissue of TLE patients. The reductions in ¹¹C-UCB-J binding were 2.7-fold larger than for ¹⁸F-FDG. Thus, PET imaging of SV2A may be a promising biomarker approach in the presurgical selection and evaluation of TLE patients with MTS, and may improve the sensitivity of molecular imaging for seizure focus detection. Research Support: NIH: R01NS094253-01. References: Nabulsi et al., 2016, J Nucl Med, 57:5:777-84. Finnema et al., 2016, Sci Transl Med, 8:348:348ra96. Löscher et al., 2016, CNS Drugs, 30:11:1055-77.