A New Method for Delineation of Epileptic Brian Tissue During Epilepsy Surgery (The HFO Study)

Intra-operative Detection and Localisation of High Frequency Oscillations in the ECoG to Guide Epilepsy Surgery

Epilepsy occurs in 0.5-0.7% of the population, of which 25% are children. 30% Of patients with focal epilepsy do not respond well to medication and half of them are eligible for epilepsy surgery. In recent years, the importance of early epilepsy surgery has been stressed, as successful surgery may lead seizure and medication freedom and improved social and cognitive development, especially in children. The current success rate of epilepsy surgery is around 65%; During surgery intracranial electrocorticography (acute ECoG, aECoG) is recorded in some medical centers. The presence of epileptiform brian activity, spikes, identified by clinical neurophysiologists, is used to guide the neurosurgeon in the extent of the brain tissue that needs to be resected. Spikes are considered markers of the presence of epilepsy. High Frequency Oscillations (HFOs, >80-500Hz) in the ECoG have recently been identified as a new biomarker for epileptogenic tissue. Retrospective research shows that their local presence strongly relates to the seizure onset, and removal of tissue with HFOs could predict a better surgical outcome. The area showing HFOs usually overlaps with, but is smaller than the area with spikes, and HFOs do not tend to propagate to distant sites as spikes do. The identification of HFOs is more objective than of spikes and automatic detection software exists. A pilot study is performed to test the hypothesis : The intra-operative use of HFOs to delineate the epileptogenic cortex does not yield significantly worse outcome in seizure freedom than the current method based on spikes. Study design is a single blinded multi-center randomized controlled trial. In two Dutch centers, the VU medical center ( Amsterdam) and University Medical Center Utrecht. The study population (sample size 78) consists of patients of all ages with refractory epilepsy undergoing epilepsy surgery with aECoG to guide the extent of the resection. Eligible patients are randomised, after informed consent, into group 1 (HFOs) in whom a resection guided by HFOs in the aECoG (new), or into group 2 (spikes) in whom a resection is guided by epileptiform spikes in the aECoG (current standard). Ictiform spike patterns will always be resected. Main study endpoint is outcome after epilepsy surgery after 1 year of follow-up dichotomized in total seizure freedom (Engel Ia&b) vs. seizure recurrence (Engel Ic-IV).

In arm " spikes" the resection of epileptogenic tissue is guided by epileptiform spikes in the aECoG (current standard). (Independent of the randomisation ictiform spike patterns will always be resected.)

In arm "HFOs" resection of epileptogenic tissue is guided by HFOs in the aECoG (new). (Independent of the randomisation ictiform spike patterns will always be resected.)

To simplify analysis outcome scores will be dichotomized in two categories, total seizure freedom (Engel Ia+Ib) versus seizure recurrence (Engel Ic-IV).To enable interim analysis of the outcome in terms of seizures, we will determine preliminary post-surgical outcomes at 6-8 weeks and 6 months. A final outcome will be determined after 12 months. This will require the patients to fill in an additional short questionnaire on their seizure frequency at pre-surgical baseline (after singing informed consent), at 6-8 weeks, 6 and 12 months post- operatively (by telephone/email). So called 'running down' seizures, seizures that occur in the first 2 weeks after surgery are not considered as seizure recurrence.

With HFOs being a more specific and sensitive biomarker for epileptogenic tissue than spikes, potentially the resection size could become smaller or larger. Therefore, the volume of resected tissue (in cm3) is investigated as a secondary parameter. The amount of resected tissue is determined by voxel-based volumetrics of the pre- and post-surgical MRI using Curryscan7 Neuroimaging Suite (Compumedics Neuroscan, Hamburg, GER).

Neurological changes (e.g visual field defects, hemiparesis, word finding difficulties) can be divided into pre-existing, aggravated /improved, or new deficits, and can be anticipated or unexpected. The neurological changes and severity will be assessed by the doctor in charge/neurologist before surgery and prior to discharge out of hospital. In case of deficits they will be classified/quantified using the National Institutes of Health Stroke Scale (NIHSS) . This will be repeated by the doctor in charge/neurologist/neurosurgeon after 6 months and 12 months in case of initial neurological changes/deficits. The patient's total NIHSS scores is calculated by summation of all scores (NIHSS score range 0-42). A difference of 1 point on the NIHSS scale between 2 tests is considered clinical relevant in our population of epilepsy patients.

Comparison of test results of pre- and post-operative (6 or 12 months) neurophysiological 695 investigation (NPO). This routine neurophysiological investigation includes testing of IQ (verbal and performal), working memory and processing speed. All tests conform to the age of the patient, but report on the same domains. Per domain individual patients' results will be dichotomized into negative/no/positive change compared to pre-surgical performance. This enables comparison of all patients, independent from age, at group level.

Health related quality of life (HRQOL; QUOLI89 for adults and an age adapted comparable list for children) will be determined pre-and post-operatively (6 or 12 months) during routine neurophysiological investigation (NPO). Interim assessment of HRQol will be enabled by a visual analogue scale (VAS) on overall self-perceived health (including their epilepsy) included in the baseline and follow-up questionnaire. This VAS scale is also part of the HRQoL tests.

Complications related to study procedures Accounts are kept of the number of (post-)operative complications, such as bleeding, infection, unexpected or aggravated neurological deficits. These events will also be reported as (Serious) Adverse Events ((S)AEs). Monitoring and interim analysis by the DMC will be based on these numbers , as well as end-point determination.

Inclusion Criteria: Patients of all ages with: Refractory Focal Epilepsy; at least ≥2 experienced seizures in the past 24 months, in spite of 2 or more different anti-epileptic drugs tried. Planned neurosurgery with aECoG, with the goal of tailoring the resection. Command of Dutch language of the patient/parents/legal representatives and capability of completing the questionnaires (by email or phone). Exclusion Criteria: Patients who underwent chronic ECoG monitoring preceding epilepsy surgery (grids). This is a biased population, since the results of the extensive pre-surgical work-up as well as the results of the cECoG monitoring period are included in the final decision making regarding the resection, and a precise seizure onset zone as well as spike and HFO area are known. Patients with an occipital focus undergoing aECoG. Currently, it is not possible to discriminate between pathological or physiological occipital HFOs, and thus unsafe to perform HFO guided resections in patients with a presumed occipital focus.

van 't Klooster MA, Leijten FS, Huiskamp G, Ronner HE, Baayen JC, van Rijen PC, Eijkemans MJ, Braun KP, Zijlmans M; HFO study group. High frequency oscillations in the intra-operative ECoG to guide epilepsy surgery ("The HFO Trial"): study protocol for a randomized controlled trial. Trials. 2015 Sep 23;16:422. doi: 10.1186/s13063-015-0932-6.

Zweiphenning W, Klooster MAV', van Klink NEC, Leijten FSS, Ferrier CH, Gebbink T, Huiskamp G, van Zandvoort MJE, van Schooneveld MMJ, Bourez M, Goemans S, Straumann S, van Rijen PC, Gosselaar PH, van Eijsden P, Otte WM, van Diessen E, Braun KPJ, Zijlmans M; HFO study group. Intraoperative electrocorticography using high-frequency oscillations or spikes to tailor epilepsy surgery in the Netherlands (the HFO trial): a randomised, single-blind, adaptive non-inferiority trial. Lancet Neurol. 2022 Nov;21(11):982-993. doi: 10.1016/S1474-4422(22)00311-8.