Exploiting Thalamocortical Interactions for Seizure Termination in Multifocal Epilepsy

Principal Investigator: SALAMI, PARIYA

Proposal Number: EP210059

Award Number: W81XWH-22-1-0315

Period of Performance: 6/1/2022 - 5/31/2025

PUBLIC ABSTRACT

Post-traumatic epilepsy is one of the most common causes of acquired epilepsy, with military personnel comprising the highest incidence of post-traumatic epilepsy, impacting their health and well-being. Antiseizure medication is the first resort for such patients, but epilepsy resulting from brain trauma is one of the hardest to treat. Some patients are now treated with deep brain stimulation (DBS) or responsive neurostimulation (RNS), which use electrical pulses to disrupt seizures. A common target for stimulation is the thalamus, which makes connections to widespread brain areas. While thalamic stimulation works for some patients, others see no benefit at all. Working out who would benefit from the device would not only help clinicians to devise better therapeutic options, but also save patients from unnecessary brain surgery and the many trips to the clinic needed to configure the device, only to discover that it is ineffective. The core goal of the proposed study is to understand who may benefit from thalamic stimulation and how the thalamus adapts to stimulation over time.

The general hypothesis of this work is that thalamic activity may influence only a subset of seizures. We propose that it is possible to identify these seizures and use this information to find out who may benefit from thalamic stimulation (Aim 1). The most significant obstacle in our way is ignorance; we currently have no idea how the thalamus affects seizure activity. To address this, we will analyze the signals recorded from the thalamus and other brain regions to identify how the thalamus interacts with the rest of the brain during seizures (Aim 2). Finally, we will analyze brain activity from patients with RNS implants to work out whether thalamic stimulation induces short- or long-term changes that make epileptic patients less susceptible to seizures (Aim 3).

The ability to identify which patients with post-traumatic epilepsy might respond to thalamic stimulation, informed by the work of this proposal, will be the first of its kind to empower patients to make more informed choices, allowing them to better manage their expectations, and avoid unnecessary surgical risks and potentially years of fruitless clinical hours. We anticipate that our findings will not only inspire the development of novel patient-specific modes of treatment, but also empower clinicians with the ability to provide meaningful guidance to patients with post-traumatic epilepsy in particular, who suffer from one the most untreatable variations of epilepsy.

TECHNICAL ABSTRACT

Background: Traumatic brain injury is one of the most common causes of acquired epilepsy, with Warfighters accounting for the highest incidence of post-traumatic epilepsy. At present, sufferers of epilepsy have few therapeutic options. Despite the advances made in diagnostic techniques, antiseizure medications fail in as many as 30% of people with epilepsy, and as many as 50% of patients undergoing resective surgery may not benefit from the treatment due to overlap of the epileptic region with the eloquent cortex, or in cases where the epileptic region is widespread. The latter may be the main reason why patients suffering from post-traumatic epilepsy may not benefit from conventional therapies. An increasingly promising approach to seizure management is neurostimulation, particularly thalamic neurostimulation. The thalamus is crucial in mediating widespread cortico-subcortical information flow and is therefore a common target for neurostimulation. However, it is unclear how thalamic stimulation leads to seizure termination and, most importantly, which patients are likely to benefit from such treatment. Through this proposal, we aim to identify the candidates who can benefit from thalamic neurostimulation and to understand how this treatment leads to seizure control so that it can be optimized.

Hypotheses: The core hypotheses are that (a) thalamic activity is responsible for seizure synchronization but only in a subset of seizures; and (b) during these seizures, thalamic activity modulates the activity of the rest of the epileptic network. We propose to evaluate the dynamics of thalamic activity that facilitates seizure termination and identify the electrographic biomarkers that can be used to distinguish such seizures with the ultimate goal of identifying which kinds of epilepsy are most amenable to thalamic control.

Specific Aims:

Aim 1: Determine the subset of seizures for which thalamic activity is responsible for synchronization. The fact that not all patients benefit from thalamic stimulation suggests that the field could benefit from research aimed at identifying who might benefit from thalamic neuromodulation. This starts with an understanding of what kinds of seizures may be modulated by thalamic activity. To answer this question, we will classify seizures based on their focality and electrographic pattern and then identify which seizures demonstrate strong thalamic involvement.

Aim 2: Determine whether disruption or amplification of thalamic activity (in thalamus-synchronized seizures) leads to seizure termination. Previous work has shown that the thalamus might be involved in inducing synchrony between different regions, although the neural mechanisms through which this is achieved is unclear. Here we seek to determine the thalamic signatures responsible for this synchrony and to understand how these thalamic activities interact with the rest of the brain.

Aim 3: Assess whether thalamic activity changes over time in patients who received a thalamic responsive neurostimulation (RNS) implant. Stimulation of different brain regions showed promising results in seizure control. Our goal is to evaluate the acute and chronic effects of thalamic stimulation to understand the changes in the targeted area that leads to seizure control.

Research Strategy: In Aim 1 and Aim 2, we use data from patients who have had thalamic recordings as part of their presurgical evaluation. We will classify their seizures into different types and use network analysis to identify the seizures with high involvement of the thalamus (Aim 1). We will then use novel analytical techniques to evaluate the interactions between the thalamus and the rest of the brain to determine thalamic involvement in seizure synchrony. Finally, in Aim 3 we will perform a coherence analysis of RNS recordings in the thalamus to evaluate the short- and long-term effects of thalamic stimulation.

Innovation and Impact: At present, it is impossible to determine whether a patient is likely to benefit from neurostimulation. With an improved understanding of the factors that influence treatment efficacy, clinicians will be better equipped to guide families and physicians in the risk-benefit analysis of an implanted device, saving patients from having to make frequent follow-up visits and the long-term trajectory of adjusting the device. We are seeking to answer these questions by applying novel techniques on a rare dataset recorded from the thalamus of patients with epilepsy. The findings of this study will have considerable implications for modern diagnostic practice, as well as patients with epilepsy, especially those suffering from post-traumatic epilepsy.