Predicting efficacy of neuromodulation in epilepsy

In patients with medically refractory epilepsy who are not eligible for
surgical treatment, neuromodulation including vagus nerve stimulation (VNS)
remains as a last resort. However, a good response (> 50% seizure reduction) is
achieved in only approximately 50% of patients. At present, we cannot predict
who will benefit from VNS. Ultra long-term EEG measurements have recently
become feasible by using minimally invasive subcutaneous electrodes. We
hypothesize that the combination of this technique with network analysis of
resting state fMRI and 64-channel EEG can provide insight in the brain circuits
involved in the disturbed brain dynamics and can result a significant
improvement in response prediction of individual patients who are candidates
for VNS.
Currently there is no experience with the subcutaneous EEG electrodes in
epilepsy patients undergoing VNS therapy. Therefore, we will start with a pilot
to evaluate the feasibility.

We started to evaluated the feasibility of subcutaneous EEG electrodes in
epilepsy patients undergoing VNS therapy in a pilot study. This part is
currently finished.

Currenrly, we are extending the cohort and aim to identify biomarkers to
predict success rate for VNS using pre-operative features from ultra long-term
EEG recordings, resting-state 64-channel EEG, and MRI.

We propose a prospective observational cohort study.

Two months prior to implantation of the stimulator, patients will be implanted
with a subcutaneous EEG electrode.This electrode is inserted under the skin
above the ear under local anesthesia. This procedure takes approximately 30
minutes. The subcutaneous electrode will be explanted 13 months after VNS
implantation. Brain activity will be recorded continously from 2 months
pre-operatively to 13 months post-operatively). During this period, test
subjects must wear the corresponding device and change this device once a day
to charge. The risks of placing, wearing and explanting the electrode are
minimal.

We will further record baseline (2 months pre-operative, before insertion of
the subcutaneous EEG electrode) and postoperative (13 months postoperative,
after explantation of the subcutaneous EEG electrode) 64-channel EEG and MRI
(anatomical, DWI and fMRI). In addition, patients will be asked to keep diaries
of their perceived seizures during the study period. We will also ask all
patients three times to fill in a short questionnaire to evaluate patients*
satisfaction with the subcutaneous electrode and the effect of their VNS and
twice to fill in four questionnaires about psychological well-being.

If the location of the epileptic focus is unknown prior to the implantation of
the subcutaneous electrode, an ambulatory EEG recording of 24 hours will be
made to better determine on which side the subcutaneous electrode should be
placed. Optionally, an app will be installed on the subjects' smartphone which
will record their smartphone use.

If we can find biomarkers from ultra long-term EEG and/or networks from
64-channel EEG or MRI to predict the effect of VNS in epilepsy patients, this
will lead to new clinical guidelines for personalized treatment of
pharmacoresistant epilepsy patients. The number of patients who undergo an
invasive operation to implant a VNS stimulator without any benefit can be
reduced. Therefore, the risk and burden for the participating patients are in
proportion with the potential value of the study.