Nucleus-specific thalamic involvement in seizure networks differentiates neuromodulation outcomes

Area of Science:

• Neuroscience
• Epilepsy Research
• Neuromodulation

Background:

• Responsive neurostimulation (RNS) of the thalamus is a developing treatment for drug-resistant epilepsy (DRE), especially for patients with widespread or multifocal seizure origins.
• Patient outcomes with thalamic RNS vary, highlighting the need to identify predictors of treatment response.
• The thalamus's complex structure suggests that specific nuclei's roles during seizures might explain differing responses and guide treatment targeting.

Purpose of the Study:

• To investigate whether the functional connectivity patterns and seizure spread timing within specific thalamic nuclei (centromedian and pulvinar) can predict patient response to RNS therapy in DRE.
• To identify nucleus-specific biomarkers for thalamic RNS treatment selection in DRE patients.

Main Methods:

• Analyzed 129 seizures from 28 DRE patients who underwent stereo-EEG monitoring targeting the centromedian (CM) or pulvinar (PLV) thalamic nuclei, followed by RNS.
• Quantified thalamic nucleus involvement using functional connectivity (node strength) and automated detection of seizure spread timing relative to seizure onset.
• Compared connectivity and spread timing between RNS responders (Engel class I-III) and non-responders (Engel class IV) within CM and PLV groups.

Main Results:

• Seizure spread timing to thalamic nuclei did not significantly differ between responders and non-responders, though a trend towards earlier CM recruitment in responders was observed.
• Centromedian nucleus (CM) responders showed significantly higher functional connectivity (node strength) during the late-seizure phase compared to non-responders.
• Pulvinar nucleus (PLV) responders exhibited significantly higher functional connectivity during the middle-seizure phase, unlike CM responders.

Conclusions:

• The degree and timing of thalamic nucleus involvement during seizures, particularly functional connectivity patterns, may serve as predictive biomarkers for thalamic RNS response in DRE.
• CM responders demonstrate sustained strong connectivity throughout seizures, while PLV responders show heightened connectivity during seizure propagation.
• Integrating these nucleus-specific ictal network features into pre-surgical evaluations could enhance patient selection and optimize nucleus-specific targeting for thalamic RNS.