Thalamic connectivity-based Biomarkers for Neuromodulation in Patients with Refractory Epilepsies

Thalamic electrical stimulation is being used as a treatment for drug-resistant epilepsy, yet current strategies benefit only about half the patients. Advancing patient outcomes necessitates tailored target selection for stimulation including the development of biomarkers for successful treatment. Here, we analyzed intracranial recordings of the most discussed thalamic nuclei for neuromodulation in epilepsy: anterior nucleus (ANT), centromedian nucleus (CM), pulvinar (PLV), and mediodorsal nucleus (MD) in 38 patients, to evaluate how their functional connectivity to epileptogenic areas may differ from that of non-epileptogenic areas in both sleep and wake states. We found that the thalamocortical connectivity to seizure onset areas is region and frequency-dependent. Additionally, we showed that thalamic connectivity is modulated in high versus low seizure risk conditions, further emphasizing its role in ictogenesis. Most importantly, we used the connectivity signature to predict the outcomes in patients receiving RNS within CM or PLV for neuromodulatory treatment and showed that these connectivity measures could serve as biomarkers of responsiveness to the treatment. These findings suggest that interictal connectivity data can inform patient-specific neuromodulation targeting, providing insights into the neurostimulation efficacy for specific epilepsy phenotypes.