T14. Potential electrocorticographic seizure biomarkers

Abstract

Listen

Translate article

Identifying electrographic features that differentiate patients with the most favorable response to epilepsy treatments will aid in discovering disease biomarkers and treatment endpoints, and could enhance patient selection. The RNS® System provides responsive direct-brain stimulation and is FDA approved for the treatment of medically refractory focal epilepsies in adults. The median percent seizure reduction in clinical trials reaches 72% at 7 years with 30% experiencing a reduction in seizures of ⩾90% based on most recent 3 months of follow-up. The device also provides chronic ambulatory electrocorticographic data. The objective of this study was to assess electrographic and demographic differences between patients treated with the RNS® System (NeuroPace, Inc.) who had the greatest and the least reductions in clinical seizure frequency. 179 adults within clinical trials of the RNS System had ⩾6 months of clinical seizure data at year 7 of treatment and were included in this analysis. Patients in the upper (n = 45: MTL = 16, Neocortical = 27, MTL + neocortical = 2) and lower (n = 45: MTL = 16, Neocortical = 26, MTL + neocortical = 3) clinical seizure response quartiles were determined by ranking mean change in clinical seizure rates in year 7 compared to baseline. Electrographic features (interictal spikes, total spectral power and band spectral power) were extracted from 76,570 scheduled (putatively interictal baseline) ambulatory 4-channel electrocorticographic (ECoG) records (each ∼90 s) collected over 7 years. Differences in electrographic features and clinical seizures over time in the upper and lower response quartile patients were measured. Demographic characteristics (gender, age at enrollment, age of onset, number of onsets, baseline seizure rate, and presence or absence of previous resection, callosotomy, subpial transection and prior intracranial monitoring) were compared. ECoG feature trends over time (6 months to 7 years) closely followed clinical seizure rate trends in both MTL and neocortical groups, with higher feature values significantly (p < 0.05) correlated with higher clinical seizure rates. Ranking features according to their ability to differentiate upper and lower response quartile patients revealed spike rate as the strongest feature in the MTL group. In the neocortical group, high gamma (50–125 Hz) emerged as the strongest feature. There were no significant differences in the demographic characteristics between upper and lower quartile patients when considered as one group or when divided into MTL and neocortical onset groups. Chronic ambulatory ECoG data obtained from a trial of the RNS System revealed electrophysiological differences between patients in the upper and lower quartiles of clinical seizure response. Changes in spike frequency and in high gamma could be potential seizure biomarkers.