Targeting the Thalamus by Connectivity: A Personalized Approach to Electrical Stimulation for Focal Epilepsy

A new study published in Nature Communications proposes a shift from the traditional one-size-fits-all model of thalamic stimulation for epilepsy toward a more precise, connectivity-informed strategy. By tailoring the stimulation site within the thalamus to match the anatomical and functional pathways of each patient’s seizure onset zone (SOZ), researchers demonstrated significant gains in neuromodulation effectiveness for individuals with medically refractory focal epilepsy.

Thalamic electrical stimulation has long been considered a viable alternative for epilepsy patients ineligible for resective surgery. However, its success has varied, often depending on which thalamic nucleus is targeted. The current study, led by researchers at the University of Pittsburgh, hypothesized that stimulation outcomes could be enhanced by aligning the chosen thalamic nucleus with its cortical connectivity—an approach they termed “hodological matching.”

In a cohort of 41 patients with focal epilepsy, the research team used structural MRI, high-definition fiber tracking, and intracranial recordings to identify preferential thalamocortical connections from three specific thalamic nuclei: the pulvinar (PUL), anterior nucleus (ANT), and ventral intermediate/ventral oral posterior nuclei (VIM/VOP). These were each found to project dominantly to distinct cortical zones—PUL to posterior quadrant and temporal cortices, ANT to frontal and temporal areas, and VIM/VOP to the rolandic cortex.

The study’s multimodal analyses revealed that only when stimulation was applied to thalamic nuclei with matching connectivity to the SOZ did it result in immediate and significant suppression of interictal epileptiform discharges (IEDs). In contrast, unmatched stimulation had inconsistent effects. For instance, in patients with occipital or parietal SOZs, PUL stimulation reduced IED rates by up to 73%, while unmatched sites showed negligible change.

Functionally, the matched nuclei exhibited increased synchrony with the epileptogenic cortex during seizures, especially during termination phases. Using nonlinear correlation metrics and Granger causality, the authors found a directional shift over the course of seizures: cortical regions typically initiated the seizure, but the thalamus assumed a leading role in terminating it. This dynamic was consistent across all three thalamic nuclei studied, suggesting shared temporal patterns that could inform future closed-loop stimulation strategies.

The most striking evidence of clinical benefit came from a subgroup of 10 patients who received chronic thalamic stimulation implants. Seven patients received implants in hodologically matched nuclei, resulting in an average seizure reduction of 87.5%. By comparison, three patients with unmatched targeting saw only an 8.3% reduction. In one representative case, weekly seizures dropped from five to just one per month following targeted PUL stimulation.

These findings challenge prevailing paradigms in neuromodulation that often rely on uniform stimulation targets, such as the ANT for all focal epilepsy cases. Instead, the study argues for a precision-medicine framework in which thalamic targeting is tailored to the patient’s unique network architecture. While the study was observational and the chronic implant sample relatively small, its integration of anatomical, electrophysiological, and clinical data provides a compelling rationale for further validation in controlled trials.

The research team concludes that connectivity-guided thalamic stimulation could significantly improve outcomes for patients with focal epilepsy, particularly those who are not candidates for surgical resection. Their hodology-based model introduces a new dimension to personalizing neuromodulation therapies and may ultimately reshape clinical guidelines for device-based epilepsy interventions.