PROJECT SUMMARY Epilepsy, occurring in 1 percent of the world’s population, is associated with disability, injury, cognitive and neurological dysfunction, depression, loss of productivity, socioeconomic decline and even death. Of this population, 30 percent of epilepsy cases are medically intractable, leaving surgical interventions as the only option for treatment. Whereas open resection, the current surgical standard of treatment, can yield seizure freedom rates as high as 60-80 percent, these are often associated with cognitive dysfunction and focal neurological deficits. Particularly, patients with dominant hemisphere mesial temporal lobe epilepsy (MTLE), the target population for this proposal, are at risk for significant decline in memory and associated disability. The only option for these patients at present is electrical neuromodulation which, although effective at reducing seizures, only achieves seizure freedom in ~10% of patients. We have recently found that delivering asynchronous pulses distributed across a multielectrode array of 16 microelectrodes, and stimulated at low (theta) frequency, is more effective than macrostimulation in controlling seizures in a rodent model of MTLE. The objective of the proposed project is to optimize asynchronous distributed multielectrode stimulation (ADMES) in a realistic large animal model of epilepsy - non-human primates (NHP) that have been administered penicillin (PCN) in the hippocampus to induced repeated spontaneous seizures. This research will capitalize on the availability of a new commercial neurostimulation system (RC+S, Medtronic) that uniquely allows our novel approach to be implemented. We will also exploit the novel bi-directional feature of this unit to optimize our therapy with both open-loop and closed-loop approaches to ADMES. We will first implement ADMES in our NHP model and quantify effects on seizure frequency and length, and rule out adverse effects on recognition memory. In parallel, we will characterize the response of physiological biomarkers such as synchrony to adjustment of ADMES stimulation in an externalized system. This will allow us to develop both open-loop and closed-loop control policies to optimize these biomarkers as a proxy for seizure control. The most effective stimulation parameters will be implemented in 8 NHPs using the RC+S neurostimulator and benefit on seizure frequency and effects on memory will be evaluated. If seizure reduction is ≥50% then we will advance to an early clinical feasibility study. For this, we will first identify electrophysiological biomarkers and characterize the effects of stimulation parameters informed from our NHP study on those biomarkers during invasive monitoring of MTLE patients and then move to an early feasibility trial of ADMES in 6 patients. The final stimulation parameters will be implemented in RC+S and behavioral seizure reduction and memory testing for safety will be quantified over 12 months. At the completion of thi...