Abstracts

Rationale: Tumor associated epilepsy (TAE) in low grade gliomas is prevalent and often refractory. Multiple lines of evidence support the notion that increased extracellular glutamate release from over expression of System xc (Xct) in the peritumoral region leads to a propensity for seizure generation. We tested the efficacy of an antagonist of Xct, sulfasalazine (SAS) on spontaneous epileptic activity in tissue obtained from adult human patients with gliomas. Given that SAS is unlikely to be a CNS penetrant compound, we also tested the efficacy of clinically approved antiepileptic drug (AED), perampanel, on these samples also. This compound is a non-competitive AMPA receptor antagonist and in the presence of elevated glutamate should exert an antiepileptic effect. Methods: Peritumoral samples obtained during surgery were transferred in cold artificial cerebrospinal fluid (aCSF) to the lab for slicing and ex vivo electrophysiology. Extracellular local field potential recordings were conducted in slices. Slices displayed spontaneous epileptic activity in the form of inter-ictal discharges (IIDs). To elicit ictal events a modified artificial cerbrospinal fluid (ACSF; 0.25 mM Mg2+ and 8 mM K+) was used. Sulfasalazine (SAS) and perampanel (PER) were bath applied. Slices were then prepared for immunofluorescence with antibodies against intermediate filament glial fibrillary associated protein (GFAP) and Xct. Tissue was obtained from adult patients with gliomas with and without a history of seizures preceding surgical intervention. The Student t-test was applied to compare baseline recordings with recordings in the presence of drug. Results: In spontaneously active peritumoral slices, the occurence of IIDs was significantly reduced (P < 0.05; n=6) by SAS (50-200μM). In samples obtained from patients  with no history of seizure, spontaneous events were not observed. In these samples inter-ictal discharges were evoked using zero magnesium containing aCSF. In these recordings, SAS did not alter the frequency of events at lower concentrations (≤50μM) but significant reductions were observed at higher concentrations (≥100μM). Peritumoral slices exhibiting spontaneous epileptic activity demonstrated strong co-localistion of antibodies staining for the intermediate GFAP and Xct. IIDs were blocked by PER in vitro. PER (0.1µM) reduced (50%; P<0.05; n=5) the frequency of IIDs. At higher concentrations IIDs were completely abolished. The power of ictal events, elicited by modified ACSF, was reduced (683.9±250.4 vs. 75.3±26.0 µV2.Hz; P < 0.05; n=7) by PER (10µM). Conclusions: These data demonstrate that in human peritumoral cortical regions excessive glutamate release from human glioma cells contributes to epileptic activity. Furthermore, molecules that target Xct could be useful therapeutic targets for tumour epileptogenesis. In the absence of such compounds at the present time, our data implicate PER as an AED for the treatment of tumor associated seizures. It is possible that PER may be a useful prophylactic AED for intraparenchymal brain tumors. Funding: Eisai Ltd