ILAE British Branch 2021 Annual Scientific Meeting

Objective: EEG is used as the gold standard in the diagnosis of nonconvulsive status epilepticus and to differentiate absence status epilepticus from focal status epilepticus with impaired awareness. We present here two cases of Absence Status Epilepticus (ASE) in adults with very different EEG and clinical patterns. Results: Two female patients, aged 30 years old (case one) and 76 years old (case two) with genetic generalised epilepsy, were diagnosed with ASE after prolonged EEG (260 hours and 20.5 hours, respectively). In these cases, the clinical manifestation of ASE was different, with the cardinal symptom being memory loss in case report one and confusion in case report two. Furthermore, the EEG pattern of each patient was also different. In case one, there were very frequent runs of high amplitude, at times sharpened, slow activity over frontal or fronto-temporal regions and bursts of 1-2Hz generalised spike/polyspike and slow wave. Conversely, in case two, there were continuous 2-3Hz generalised spike and slow wave discharges. Conclusion: In contrast to absence seizures, absence status epilepticus can have different EEG signatures and clinical manifestations. These two cases illustrate the difficulties in diagnosis and the range of presentations of ASE.

Al Anzari Ahmed Adan

UCLH-Epilepsy Society, Gowers Unit

Background: Diagnostic video electroencephalography (VEEG) is a useful yet expensive investigation. To maximise the utility of VEEG in capturing a seizure, antiepileptic drugs (AED) can be reduced or suspended in selected cases, though this must be in a controlled manner. Decisions regarding alteration to AED regimens are made in advance of admission in an MDT setting. This plan should be accessible, reviewed and followed on admission unless there are clinical grounds for modification on admission. This audit assessed the pre-admission plans made for alteration to AED regimes, adherence to these plans, and whether the introduction of an AED reduction proforma had a positive impact on plan adherence. Methods: The medical records of patients admitted for VEEG between May 2018 – October 2018 were reviewed. The documented AED reduction plans pre-admission were noted and compared with prescription on the day of admission. The clinical case notes, correspondence and EPMA records for each patient were reviewed and recorded in an Excel spreadsheet. Following this initial audit, an AED reduction proforma was introduced and re-audit was carried out, including all patients admitted for VEEG between September – December 2019. Results: On initial review, 31 patients were identified. Only six patients (19.3%) had documented plans for AED treatment alteration. Five patients (16.1%) had a pre-admission plan to reduce AEDs on admission. Three of these five patients (60%) had the planned AED reduction implemented on admission. Two of the five patients (40%) with planned AED reduction did not have the changes implemented on admission. Following introduction of the AED reduction proforma, 15 patients were identified. Seven patients (46.7%) had documented plans for AED treatment alteration. The proforma was completed in six of these patients (85.7%). Four patients (57%) had a pre-admission plan to reduce AEDs on admission. All four patients (100%) with a plan to reduce AEDs, had the AEDs reduced on admission according to the plan. Conclusions: Non-adherence to planned AED reduction plans can reduce the efficacy of VEEG assessment, which can prove to be costly to epilepsy services. The introduction of an AED reduction proforma resulted in the documented pre-admission plan being followed in 100% of patients. This is an improvement to the initial audit prior to proforma introduction, in which the documented plan was followed in only 66.6% of patients. The AED reduction proforma allows admitting junior doctors to easily access the plan for AED treatment and prescribe the appropriate regime on admission.

Ahmed Adan

University of Liverpool

Introduction: The period of transition over to adult services from the paediatric team can be an uncertain time for patients and their families/guardians. It can also provide challenges for healthcare professionals to ensure appropriate discussion and effective ongoing care for patients. The Ready Steady Go (RSG) programme developed by the University of Southampton aims to address these uncertainties. The Aneurin Bevan Health Board introduced the RSG programme for young people with epilepsy in 2020. Aim: This service evaluation assessed implementation of the RSG programme for young people with epilepsy undergoing transition to adult care within the Aneurin Bevan Health Board, Wales. Methods: The evaluation was carried out using a database of patients aged 17-23 years undergoing transition before and after implementation of the RSG programme. Patients were randomly selected from pre-RSG and post-RSG cohorts to carry out telephone interviews. Patients with comorbidities and complex needs were included. A questionnaire was created to ascertain the patients’ views on their transition period pre-RSG and post-RSG implementation. Information was gathered relating to appointment logistics, accessibility of services, and opinions of the RSG forms. Patients were asked to rate on a scale of 1-10 their comfortability with contacting adult services. A total of N=4 pre-RSG and post-RSG questionnaires were obtained, with inclusion of N=2 patients with complex needs in both cohorts. Results: In the pre-RSG cohort, there was a preference for the regular appointments and support from paediatric service compared to patient-initiated follow-up in adult services. In addition, young people felt they had poor understanding of what to expect from adult services. The average rating for comfortability accessing adult services was 6.75. In the post-RSG cohort, all participants felt adequately informed on what to expect and that RSG forms aided their transition process. Their average rating for comfortability accessing adult services was 9.75. Patients in both cohorts with complex needs struggled with changes in environment at transition appointments. Discussion and conclusion: This service evaluation demonstrated the RSG programme has increased patient’s preparedness for the transition process, by a greater satisfaction and comfortability with adult services. Nevertheless, the COVID-19 pandemic hindered the extent to which the RSG programme could be implemented in 2020, resulting in only completion of the ‘Go’ form. There is need for further evaluation following full implementation of RSG forms with a greater cohort size.

Pallavi Aggarwal

Cardiff University

Introduction: The management of patients with dissociative seizures (DS) can be challenging. Disability is high with frequent use of emergency services resulting in high economic and healthcare utilisation costs. Despite this, there is a lack of joined up multidisciplinary care with limited access to dedicated psychological intervention outside of highly specialised centres. We present a service evaluation of a simple streamlined approach that we have adopted for managing DS patients at the Royal Free Hospital. Method: In this pathway, patients with suspected DS are referred to a dedicated DS clinic run by an epilepsy specialist. Following confirmation and communication of the diagnosis, patients are referred to a neuropsychologist for psychoeducational intervention. Patients opt into the service by contacting the psychology department, and receive a maximum of 3 sessions depending on their level of engagement. Management of significant psychiatric comorbidity remains with the local community mental health services. Patients were followed up at regular intervals, over 12 months, and outcome measures were collected on symptom control, health confidence and information scores, and ED attendance. Results: We followed the cases of 32 consecutive patients (age range 18–71; 23 females, 7 males, 2 patients with gender reassignment). Specialist consultation and explanation of the nature of the diagnosis was provided to all patients. Psychology intervention was offered to 30 patients (94%), provided fully to 12 patients (37%) and partially to 5 patients (15%) as they attended less than 3 sessions, 2 patients (6%) required no psychological input and 13 patients (40%) did not make contact with the psychology service. At the end of the 12-month follow-up period, there was full remission in 11 patients (34%), improvement in 9 patients (28%), no change in DS frequency in 3 patients (9%), worsening symptoms in 1 patient (3%), and unclear outcome in 8 patients (25%) who missed their follow-up appointments. Patients (n=5), on anti-epileptics, were successfully weaned. At baseline, 23 patients (72%) attended ED, with average of 3 times between the time of onset and the first appointment in our clinic. After the intervention, only 2 patients attended ED once each during the 12-month follow up period. Health confidence and information scores improved from a mean of 2.7 and 4 at baseline, to 7.1 and 8.2 on discharge, respectively. Conclusion: Our findings demonstrate that this simple pathway of combined specialist and community approach was efficient and effective, leading to positive outcomes for DS patients in terms of symptoms control, patients’ confidence and understanding of the nature of the condition, and prevention of unnecessary ED attendance.

Talal Al-Mayhani

National Hospital for Neurology and Neurosurgery

Introduction: Patients with dissociative seizures (DS) present with a wide range of symptoms including associated pain syndrome or functional neurological disorder (FND), and often describe more than one type of DS attack. In this study we aimed to characterize the different DS semiologies and explore their clinical significance. Methods: Patients with DS were managed according to a recently established pathway. The description of the DS was analysed and the semiology classified into one of the following categories: hyperkinetic, hypokinetic, vacant spells, behavioural episodes and subjective cognitive states. The presence or absence of prodromal symptoms, pain syndromes, FND, past or ongoing traumatic events, history of medical diagnoses or procedures were recorded. Results: We reviewed the cases of 30 consecutive patients (age range 18 – 71; 21 females, 7 males and 2 patients with gender reassignment). We found that 15 patients (50%) exhibited more than one type of DS, with the most common combination being hyperkinetic and vacant episodes, seen in 6 patients (40% of patients with multiple semiology). Only 3 patients in this group (20%) described more than 2 types of DS. Among those with one type of DS, 5 patients (33%) had an associated FND, compared to only 2 patients (13%) in the group with multiple DS types. Favourable clinical outcomes (improvement or remission of DS at 12 months follow up), were seen in 12 patients (80%) exhibiting one type of DS. Only 8 patients (53%) with multiple DS types achieved a favourable outcome, and among those, 2 patients (13%) reported improvement in one type of attack only, while the other type(s) remained active. Conclusions: Our results demonstrate that half of our DS patients described multiple semiologies. Although this initial study showed an association between this observation and clinical outcome; more work is needed to expand on case numbers, refine semiology categorisation, explore the association with other FNDs and to analyse outcome measures over a longer time frame.

Talal Al-Mayhani

National Hospital for Neurology and Neurosurgery

Background: Infantile spasms (IS) is one of the early childhood epilepsy disorders. Vigabatrin, corticotropin (ACTH) and corticosteroids represent the standard treatments for IS, and their use has been supported by sound evidence. Other lines of treatment have been proposed, but there has been insufficient evidence to recommend their use. Objectives: To search for updated literature supporting the use of alternative pharmacological treatments for IS. Methods: A systematic search was run in online libraries, including MEDLINE, EMBASE, Cochrane library of reviews, CINHAL and Google Scholar. Keywords included “Infantile spasms”, “West syndrome”, “Management”, “Treatment”, “Drugs”, and “Medication”. Prospective, retrospective, interventional and observational trials were included, published in any year. Results: A systematic review of 13 observational studies showed that the ketogenic diet (KD) led to a significant seizure rate reduction. Another underpowered randomized controlled trial suggested that KD is as effective as ACTH in the long term but better tolerated. Zonisamide was used in three observational trials which showed reasonable seizure reduction rates. Valproic acid was also tried in another prospective study in 23 patients and showed satisfactory seizure control (90.9%), although 23% relapsed after two weeks. Levetiracetam and topiramate were used in an open-label randomized trial with 40 patients who did not respond to prednisolone. The results showed these two medications were ineffective. Another observational trial confirmed that topiramate was also ineffective. Pyridoxine was used in a pilot randomized open-label trial with sixty-two patients, divided into two groups: prednisolone alone or prednisolone with pyridoxine. The combination of pyridoxine with oral prednisolone was not found to be more beneficial than prednisolone alone. Three other observational trials, however, reported better outcomes for pyridoxine when used alone or as a combined treatment. Intravenous immunoglobulin is another suggested treatment and was used in a prospective study in 23 children. Only 5 of the 23 patients showed complete normalization of their seizure activity, while the rest of the patients showed either no or transitory effects. Magnesium sulphate was used as a combination treatment with ACTH in a randomized open-label study of 38 patients in China. The results showed better short-term and long-term outcomes for the combined treatment compared to ACTH alone. Conclusion: No strong evidence yet exists to support the use of new agents in infantile spasms. The ketogenic diet showed the most promising results, although evidence is still limited. Larger, multicenter trials are still needed to formulate recommendations regarding alternative treatments in infantile spasms.

Ahmed Albialy

Northwest Anglia NHS Foundation Trust

Focal Cortical Dysplasia (FCD) is a malformation of cortical development that commonly arises from somatic mutations in neuronal progenitors. FCD type II is caused by mutations leading to hyperactivation of the mTORC1 pathway in more than half of studied cases, and is commonly associated with drug-resistant epilepsy, for which surgical resection of the epileptic focus remains the best hope to achieve seizure freedom. The mean age of seizure onset in FCD patients is 6.3 years, and FCD is the leading pathology found at epilepsy surgery in children. Nevertheless, this procedure is not always effective and often precluded by proximity to eloquent regions. Alternative treatments are therefore needed. One of the most promising approaches to alleviate seizures while also minimising side-effects is focal gene therapy to alter the excitability of neurons at the focus. The aim of this project is to validate an AAV gene therapy using a modified KCNA1 gene, which encodes the potassium channel Kv1.1, to treat seizures in a mouse model of FCD II. Kv1.1 was selected as a therapeutic protein not only because it attenuates neuronal excitability but also with the goal of correcting downregulation of Kv1.1, which has been reported in FCD associated with mTORC1 hyperactivation. We modelled FCD II by in-utero electroporation of neuronal progenitors with a constitutively active Rheb plasmid, which leads to mTORC1 hyperactivation. A battery of histological, molecular biology and behavioural tests showed that this mouse model recapitulates features of the disease in humans, including dyslamination, enlarged dysmorphic neurons, decreased Kv1.1 expression, spontaneous generalised seizures, and cognitive comorbidities. EEG recordings were performed before and after gene therapy with focal AAV injection. Seizures decreased by approximately 45% and improvements were also seen in interictal activity. Behavioural experiments examining memory and cognition raised no safety concerns. In conclusion, AAV-KCNA1 is a promising gene therapy approach for FCD II. It remains to be determined whether dysmorphic or non-dysmorphic neurons should optimally be targeted for maximum efficacy and safety.

Amanda Almacellas

UCL

Purpose: To assess the effectiveness, safety and tolerability of perampanel (PER) when used in everyday clinical practice to treat patients with focal-onset and generalised-onset epilepsy. Method: The PERMIT study is a pooled analysis of real-world data from 44 studies/work groups in which patients with focal-onset and generalised-onset epilepsy were treated with PER. Retention was assessed after 3, 6 and 12 months. In patients with focal and/or generalised onset seizures, effectiveness assessments included 50% responder rate (≥50% seizure frequency reduction) and seizure freedom rate (no seizures since at least the prior visit); in those with status epilepticus, effectiveness was assessed as responder rate (seizures under control). Safety and tolerability were assessed by evaluating adverse events (AEs). Results: Full Analysis Set included 5193 patients (50.5% female; mean age, 39.7 years; mean number of previous antiepileptic drugs, 4.9). Baseline seizure types were focal-onset only (81.4%), generalised-onset only (12.6%), focal and generalised onset (4.5%) and status epilepticus (1.5%). Most patients were treated with PER as adjunctive therapy; 5.5% were treated as monotherapy at baseline. Mean (standard deviation) PER dosage was 2.4 (1.1) mg/day at baseline and 6.3 (2.6) mg/day at last visit. Effectiveness was assessed for 4392 patients and safety/tolerability for 4617 patients. At 3, 6 and 12 months, retention rates were 90.5%, 79.8%, and 64.2%, respectively. Mean time under PER treatment was 10.7 months. At last visit, responder and seizure freedom rates in patients with focal and/or generalised onset seizures were 50.0% and 20.5%, respectively, and 52.7% of patients with status epilepticus were responders. AEs were reported for 49.9% of patients; most frequent AEs were: dizziness/vertigo (15.2%) and somnolence (10.6%). Overall, 17.6% of patients discontinued due to AEs. Conclusion: PER was effective and generally well tolerated when used to treat a large cohort of patients with focal and generalised onset epilepsy in everyday clinical practice. Supported by Eisai.

Rob McMurray

Christian-Doppler University Hospital and Paracelsus Medical University, Salzburg, Austria

Introduction: Annual completion of a Valproate Risk Acknowledgement Form (RAF) is mandated in the United Kingdom due to neurodevelopmental risks of in utero valproate exposure. The number of women of childbearing potential taking valproate, the uptake of the RAF within this population and their clinical outcomes is not currently systematically monitored. Methods: This study surveyed responses of clinicians administering the RAF to women of childbearing potential taking valproate medications. The national online survey was distributed to clinical specialists throughout the United Kingdom via their national organisations. Main outcome measures were quantitative and qualitative responses regarding identification, uptake, effects and reactions to the RAF. The trial was registered at the Clinical Governance and Audit Committee at Royal Free London NHS Foundation Trust Hospital. Results: 215 respondents covering more than 4775 patient encounters were captured. Most patients continued on valproate, 90% with epilepsy as the indication. Respondents reported that seizure control deteriorated when switched to levetiracetam (33%) and lamotrigine (43%), compared to 7% when continuing valproate (P<.001). Conclusions: 33-43% of clinicians reported seizure control deterioration in women changed to alternatives to valproate. Informed consent requires women considering a change are given this information. Systematic capture of data automated through online RAFs and linkage to patient outcomes is needed. There remains little data on valproate given for indications other than epilepsy.

Heather Angus-Leppan

Royal Free London NHS Foundation Trust

Introduction: How people respond to the sudden death of a family-member or friend with epilepsy is highly variable both between individuals and within individuals over the course of their lives. Similarly, the coping mechanisms people draw upon to help them manage their grief are highly individualistic and multi-faceted, influenced by both formal and informal support structures. The COVID-19 pandemic has impacted upon all of us but the disruption on those bereaved by epilepsy, whose grief-management can be inextricably linked to the very things that have been disturbed, is far less well understood. Methods: As part of a 33-question survey about experiences of bereavement before and during the pandemic, 78 people bereaved by epilepsy recounted their experiences of the impact of COVID-19 and lockdown on their well-being in both open-ended and fixed-choice questions. Questions included reference to participant physical health, mental health, financial security, social life and employment. Results: 90% of respondents reported a negative impact on their mental health and 74% reported experiencing increased anxiety in relation to safety and well-being of others. Inductive thematic analysis was conducted independently by three researchers and was analysed collaboratively for consistent themes. Five key sub-themes within ‘well-being’ were identified: ‘reflection and rumination’, ‘re-traumatisation’, ‘anniversaries and rituals’, ‘grief process’ and ‘salience of sudden deaths in the media’. “Since my son’s death I am a shell of what I was but this year has just magnified everything.” “I am always distressed when hearing of sudden death. When it’s unexpected for families it always has a greater effect on me. This has obviously increased at the moment.” Conclusions: The pandemic has restricted people bereaved by epilepsy being able to utilize the daily strategies that would normally be protective and helpful in pre-pandemic times. Human connection and physical presence are crucial and the restrictions have removed support and distraction whilst people have been bombarded with news of death and illness daily. Whilst NICE guidelines recommend clinicians offer condolence, a meeting, and signposting to SUDEP Action, many people reporting to the Epilepsy Deaths Register do not experience this. Indirectly the pandemic has made these communications more difficult still. Those bereaved by epilepsy will carry their grief with them for the rest of their lives but health professionals and patient organisations can play a significant role in assisting them to access the specialist support and help they need.

Samantha Ashby

SUDEP Action

Rationale: The disruption to healthcare and everyday routine during the COVID-19 pandemic could potentially associate with an increase in adverse outcomes in people with epilepsy (PWE). Accordingly, we launched a global study on PWE’s changing exposure to risk, to evaluate changes in patient wellbeing, epilepsy management and care during the pandemic. Methods: SUDEP Action and the University of Oxford released online surveys for PWE, their caregivers, and healthcare workers (HCW) in 11 languages. The survey collects data on the health demographics of PWE, their use of health services, their awareness of seizure risks, and the impact of COVID-19 on their exposure to seizure risk. The study has Ethics Approval from the University of Oxford, UK. Results: The surveys remain live. As of 13th July 2021, we have received 3,152 responses from 71 countries, including the UK (n=698), US (n=856), Brazil (n=537), India (n=200) and France (n=147). 60% were submitted by PWE (n=1,855); 27% by caregivers of PWE (n=841) and 14% by HCW (n=456). Health and living circumstances of PWE - 30% of PWE reported changes in their health during the pandemic (n=796). 29% required emergency care or hospitalisation within the 12 months prior to survey completion (n=789). In total, 29% of people with epilepsy lived alone during lockdown (n=781). Risk awareness – In the 12 months prior to survey completion, one-quarter of respondents discussed life changes (work, school, etc) with an epilepsy specialist (n=650). A fifth of respondents received advice on safety aids (n=526) and a quarter discussed safety precautions (e.g. swimming, first aid). Sudden Unexpected Death in Epilepsy (SUDEP) was the least discussed risk, with 15% of respondents reporting to have discussed the topic (n=410). Healthcare delivery – 26% of respondents reported difficulties in receiving care for their epilepsy (n=766). 48% had pre-scheduled appointments changed (n=1,283) and 9% had an appointment cancelled (n=253). Conclusion: Preliminary data from this study indicate that a significant proportion of PWE are being exposed to increased risk during the pandemic, owing to worsening health outcomes, isolation, reduced access to healthcare, and low engagement on epilepsy risk. Further publications will delve deeper into the differences in response patterns within and between countries. The results from this project provide an overview of the state of global epilepsy care and will help to inform the management of epilepsy mortality risks in the future.

Samantha Ashby

SUDEP Action

Introduction: Transient epileptic amnesia and epilepsy in dementia have features in common which may lead to diagnostic confusion. Patients with transient epileptic amnesia experience epileptic seizures characterised primarily by a transient impairment of memory. These seizures sometimes include brief periods of unresponsiveness and other ictal features, including olfactory hallucinations and motor automatisms. The condition typically presents in mid-late life. Epileptic seizures occurring in patients with dementia can lead to behavioural arrest and altered responsiveness, as well as periods of increased confusion and amnesia. The syndrome of TEA is also characterised by a variety of interictal memory impairments, including a patchy autobiographical amnesia, accelerated long-term forgetting and topographical amnesia. Accelerated long-term forgetting has also been described in patients with Mild Cognitive Impairment and in the early stages of Alzheimer’s Disease. We aimed to compare and contrast these conditions with the objective of examining the extent of this clinical overlap and developing a decision aid that will help to distinguish them efficiently in typical clinical settings. Materials and Methods: In this study we reviewed two groups of patients enrolled in two separate studies. The Impairment of Memory in Epilepsy (TIME) study has established a cohort of 115 patients with TEA. The Presentation of Epileptic Seizures in Dementia (PrESIDe) study recruited 144 patients with MCI or dementia from a regional memory clinic. Here we compare the relevant demographic and seizure features in these patients, as well as cognitive profiles and neuroimaging profiles. Results: A significant difference in the age of seizure onset was identified. The average age of seizure onset in the dementia group was 76.91 years, 15 years older than in the TEA group. On Cognitive testing, significant differences were identified between the two cohorts on the Addenbrooke’s cognitive examination- version III, and in all cognitive domains probed by this test. Patients who experience epileptic seizures as part of their dementia were significantly more likely to have seizures where loss of awareness is a feature. Generalised tonic-clonic seizures and automatisms occurred with a similar prevalence across both groups, although olfactory hallucinations were significantly more common in the TEA group. Conclusions: Despite sizable clinical overlap between these transient epileptic amnesia, and epileptic seizures occurring in dementia, we have shown that several key features including cognitive function and the prevalence of neuroimaging evidence of volume loss can help to differentiate them.

John Baker

UCL

Purpose: Cenobamate is a new antiseizure medication (ASM) approved in the US for uncontrolled focal seizures in adults. Two international, double-blind, placebo-controlled trials (C013/C017) demonstrated cenobamate efficacy and safety. Here we report time to onset of efficacy during titration of cenobamate in these studies. Method: Adults with uncontrolled focal seizures and taking 1-3 concomitant ASMs were enrolled in Studies C013/C017. Concomitant ASM changes were not allowed during the double-blind period. Time to onset of cenobamate efficacy was evaluated during the 6-week cenobamate titration (C013: 50mg/day initial dose, increased 50mg/week every 2 weeks to the 200mg/day target dose. Amended C017: 50mg/day initial dose, increased 50mg/week until target dose of 100 or 200mg/day; patients randomly assigned to 400mg/day were up-titrated by 100mg/day per week after the 200mg/day dose). Post-hoc analysis of efficacy examined the percent reduction in seizure frequency from baseline to each week during titration using a Wilcoxon rank-sum test (C013) or an ANCOVA model fit to ranked values of baseline seizure rate and treatment group (C017). Results: Patients receiving cenobamate had significant reductions in median percent seizure frequency versus placebo starting from the first 1-2 weeks of cenobamate titration at the initial dose of 50mg/day (C013: -26.7% cenobamate vs -15.1% placebo, P<0.05; C017: -36.4% cenobamate vs -20.0% placebo, P<0.05). Sustained significant decreases in seizure frequency versus placebo were seen throughout the 6-week titration in both studies, reaching -39.5% versus -12.8% at week 6 in C013. Median reduction in seizure frequency was progressively higher with cenobamate doses of 100 (-39.0%), 200 (-52.2%), and 400mg/day (-55.5%) versus -12.5% at week 6 in C017. Conclusion: Onset of cenobamate efficacy in significantly reducing seizure frequency occurs early and at lower doses than the target dose for maintenance therapy; efficacy improves at higher doses. Studies sponsored by SK Life Science; analyses supported by Arvelle Therapeutics.

Caroline Benoist

Arvelle Therapeutics

Purpose: Cenobamate is a new antiseizure medication (ASM) approved in the US for uncontrolled focal seizures in adults. Two international, double-blind, placebo-controlled trials with open-label extensions (OLEs; C013/C017) and a large international open-label safety study (C021) demonstrated efficacy and safety. Here we characterize the most common adverse events (AEs) in these studies. Method: Adults with uncontrolled focal seizures taking 1-3 concomitant ASMs were enrolled (C013/C017/C021). Concomitant ASM changes: not allowed during double-blind; allowed during OLEs (C013/C017) and C021 (patients taking phenobarbital/phenytoin only after titration). C021 titration started lower (12.5mg/day) and up-titrated slower (to 200mg/day over 12 weeks) than C013/C017 double-blind 6-week titration (C013: 50mg/day initial dose, increased 50mg/week every 2 weeks to 200mg/day target dose; amended C017: 50mg/day initial dose, increased 50mg/week until target dose of 100 or 200mg/day; patients randomly assigned to 400mg/day up-titrated by 100mg/day per week after the 200mg/day dose) or OLEs (C013 4-week/C017 2-week titration). Time of first onset (pooled C013/C017 double-blind and OLEs; C021), AE duration (pooled C013/C017 double-blind), and severity (pooled C013/C017 double-blind; C021 first 18 weeks) of somnolence, dizziness, and fatigue were examined. Results: First onset of the most common AEs emerged throughout the double-blind and OLE, mostly during titration. In C021 the peak occurred when dosing reached ≥50mg/day. Median duration in days (double-blind, all occurrences) was: somnolence 32 cenobamate versus 22 placebo, dizziness 11 cenobamate versus 8 placebo, and fatigue 34 cenobamate versus 20.5 placebo. AEs in the double-blind were primarily mild or moderate, with few severe AEs. In C021, more patients reported mild AEs and fewer reported moderate and severe AEs. Conclusion: Onset of the most common AEs occurred primarily during titration; AEs were generally self-limited in duration and mainly mild or moderate. Slower titration reduced the severity of AEs. Studies sponsored by SK Life Science; analyses supported by Arvelle Therapeutics.

Caroline Benoist

Arvelle Therapeutics

Purpose: Perampanel is a once-daily oral anti-seizure medication (ASM) for focal-onset seizures (FOS) and generalised tonic-clonic seizures (GTCS). In the US, perampanel is approved as monotherapy and adjunctive therapy for FOS (with or without focal to bilateral tonic-clonic seizures [FBTCS]) in patients aged ≥4 years, and adjunctive treatment of GTCS in patients aged ≥12 years. In the EU, perampanel is approved for the adjunctive treatment of FOS (with or without FBTCS) in patients aged ≥4 years and GTCS in patients aged ≥7 years with idiopathic generalised epilepsy. Although recommended maintenance dosing of perampanel for FOS is 8–12 mg/day, in some patients, efficacy can be seen with doses as low as 4 mg/day. This post hoc analysis evaluated the efficacy of adjunctive perampanel 4 mg/day for treatment of FOS, with/without focal to bilateral tonic-clonic seizures (FBTCS), by assessing time to first seizure following perampanel administration. Method: During Phase III Studies 304 (NCT00699972), 305 (NCT00699582) and 306 (NCT00700310), patients (aged ≥12 years) with FOS, with/without FBTCS, despite 1–3 ASMs were randomised to once-daily placebo or adjunctive perampanel 2–12 mg/day (19-week Double-blind Treatment Period [6-week Titration; 13-week Maintenance]). Time to first seizure from Day 1 of placebo or perampanel administration was assessed in the Intent-to-Treat (ITT) Analysis Set using the Kaplan–Meier method. Placebo data were available from Studies 304, 305 and 306; perampanel 4 mg/day data came from Study 306 (the only study to include the randomised 4-mg/day dose). Results: The ITT Analysis Set included 437/442 (98.9%) placebo-treated patients (182/185 [98.4%] from Study 306) and 168/172 (97.7%) patients who received perampanel 4 mg/day. Perampanel 4 mg/day was associated with longer time to first seizure vs placebo. Mean (standard deviation) time to first seizure was 9.3 (22.80) days with perampanel 4 mg/day vs 4.9 (7.10) and 4.5 (6.33) days for Study 306 placebo and pooled placebo, respectively; median (range) time to first seizure was 3 (1–135), 3 (1–73) and 3 (1–73) days, respectively. Conclusion: Adjunctive treatment with once-daily perampanel 4 mg/day delayed the time to first seizure in patients aged ≥12 years with FOS, with/without FBTCS, compared with placebo. These data further support the efficacy of perampanel 4 mg/day. Funding: Eisai Inc.

Manoj Malhotra

Eisai Europe Ltd., Hatfield, Hertfordshire, UK

People with drug-resistant focal epilepsy often undergo epilepsy surgery with the aim to become seizure free. Part of the presurgical evaluation is the patient admission to the hospital for a multiday video-EEG monitoring where the patient is expected to manifest multiple seizures. The correct lateralization of the seizure focus is enhanced by the occurrence of multiple seizures so that the clinical team can accurately analyse their electrographic activity and semiology. However, many patients do not manifest enough seizures during their video-EEG monitoring and therefore the development of quantitative approaches that utilize non-seizure events for the lateralization of the seizure focus would be beneficial. In this study, we apply a machine learning approach with the aim to investigate whether we can classify a 20 sec interictal EEG segment such that we can predict whether it corresponds to a patient whose seizures originate from the left or right hemisphere. We train a quadratic discriminant classifier on 394 20 sec interictal EEG segments that are obtained from four epilepsy patients. In two of them the seizure focus is on the left hemisphere and in the other two on the right hemisphere. The training set is balanced (i.e., equal number of segments that correspond to the left and right hemisphere) and every EEG segment is selected from the first 10min of each hour of the patient’s EEG recordings. Therefore, the analysed segments (98 for patient 1; 100 for patient 2; 105 for patient 3 and 92 for patient 4) correspond to both day and night recordings from multiple days. The features that we used denote the statistical connections between pairs of EEG channels as computed from the weighted phase-lag index. To avoid overfitting, we use a 50-fold cross-validation approach and iterate this process 10 times. Across the 10 iterations the average of the area under the ROC curve is 80% (range: 79%, 81%). Our results indicate that quantitative approaches may aid in seizure lateralization and might have the potential to be incorporated in the presurgical clinical evaluation.

Andrea Biondi

King's College London, Department of Basic and Clinical Neuroscience, IoPPN, UK

Background: Marijuana, an illegal street drug in the UK, has been used for its medicinal purposes for thousands of years. Since positive results from a phase III clinical trial in 2016, there has been a considerable rise in interest in the use of Cannabidiol (CBD), a chemical compound extracted from cannabis plants, in the treatment of drug resistant epilepsy (DRE). CBD has now gained FDA & NICE approval for the treatment of specific epilepsy syndromes but with the economic market surrounding cannabis and its portrayal in the media, conflict of interests can arise. Method: Data was collected from the Welsh Clinical Portal from all 15 patients who had been prescribed CBD by the paediatric neurology service based at University Hospital Wales. This included diagnosis, medication history and seizure data. A survey of nine questions was created regarding CBD’s effect on seizure control, quality of life (QOL), side effects and the parent / carers thoughts about CBD. Parents and carers were contacted to seek consent for them to be contacted by the researcher to conduct an interview. 11/15 were consented and interviewed. Results: Data on seizure control was available for 13 patients. 11/13 were seen to have a decrease in seizure frequency. 4/13 had a 50% reduction, 2/13 had a >50% reduction, 2/13 had a >90% reduction and 3/13 became seizure free. Data on seizure severity was available for 10 patients. 7/10 were seen to have a decrease in seizure severity. Data on parent reporting of QOL was available for 13 patients. 12/13 were seen to have an increase in QOL with 11/13 seeing cognitive changes such as increased alertness. Data on side effects was available for 14 patients. 7/14 experienced some type of side effect with 3 patients having to stop the CBD due to side effects of drowsiness, diarrhoea and deranged LFT’s. Every parent / carer had a positive perception of CBD despite the high incidence of side effects leading to drug withdrawal in 3/14 patients Interpretation: The results of this study concur with the findings of clinical trials regarding the efficacy of CBD in the treatment of drug resistant epilepsy. The high incidence of side effects both in the clinical trials and in real-life experience of CBD did not deter parents/carers engagement with treatment; in fact many were grateful for this opportunity for their child.

Maximilian Bond

Cardiff University

Background: In light of the COVID 19 pandemic, outpatient services have been restructured to facilitate remote consultations. As the pandemic has entered second and third ‘phases’, patients’ shifting perspectives regarding remote consultation have been essential for service planning. Two West Midlands Trusts collaborated to survey patients’ preferences regarding remote versus face to face consultation. University Hospitals Birmingham (UHB) have traditionally offered outpatient based face-to-face epilepsy clinics. The Royal Wolverhampton (RWT) operates a hybrid model incorporating GP based community clinics. Aim: To understand whether patient’s previous clinic settings influence future preference for remote or face-to-face contact, with a view to planning service configuration. Methods: Following an online and postal pilot study undertaken by UHB in the summer of 2020, this was refined and expanded to reflect pre-pandemic clinic experiences of patients with epilepsy (PWE) and whether this influenced future preferences . Results: 1,286 patients’responses have been analysed thus far. 80% patients were under 50 years. 55% were female. 78% were British Caucasian. The majority of patients (90%) said that they had epilepsy, 3% stated NEAD and 7% with a dual diagnosis of these two conditions. 72% patients wanted their follow-up appointments to take place remotely, 66% of these requested a telephone consultation. There was no association between patients’ previous clinic location (hospital versus community) and their preference for remote consultation. The predominant reasons cited for preferring remote consultation were convenience and the time taken to travel to the clinic. Discussion: Results to date indicate that the majority of PWE believe that remote consultations adequately meet their epilepsy needs. Clinical teams need to work in explaining the relative purposes of face to face versus remote consultation at different stages of their patients’ symptom journey. The clinic booking system needs to be fluid to allow for flexibility with respect to the consultation type; booking needs to be patient led. Those for whom a face-to-face consultation is required should be offered an appointment as close as possible to where they live, both to maximise attendance and to ensure that patients have a positive face to face experience. A greater emphasis on remote consultation enables healthcare professionals to work towards a greener working environment, in keeping with the NHS sustainability plan and the epilepsy climate change initiative (EpiCC).

Smriti Bose

University Hospitals Birmingham, NHS Foundation Trust

Background: Epilepsy is defined as two or more unprovoked seizures; this means that recurrent seizures provoked by drug and alcohol usage or withdrawal are not considered to be epilepsy. However, the relationship between alcohol and epilepsy is complex; recurrent provoked seizures can lead to a misdiagnosis of epilepsy, and alcohol and drug usage may mask a diagnosis of epilepsy and delay appropriate treatment. It is therefore important to appropriately manage all patients who present to medical services with seizures, whether there are issues of drug and alcohol misuse or not. Aims: This review aims to investigate cohorts of patients attending the emergency department (ED) with seizures to see whether care differed depending on a patients documented drinking or drug use status. Methods: This study was nested within a wider study exploring the management of seizures in hospital. Investigations and outcomes (referrals and admissions) were recorded for 120 patients across 2 cohorts; the first being pre-pandemic (September 2019) and the second being during (December 2020). Patients were categorised into those with documented excessive alcohol consumption and/or drug usage and those without, in order to assess whether there are any differences in management between these 2 groups. Information obtained included diagnosis, investigations, referrals and the documentation of safety advice, along with previous ED admissions with seizures. For the substance misuse patients, it was recorded if they were referred to outpatient drug and alcohol services or if they were known to these services. Results: This study showed that 32.5% of patients attending ED with seizures were in patients with a documented history of alcohol excess or substance misuse. Overall, the admission rate was 13% lower than the control group and the number of patients referred to acute neurology was 12% lower. Referral rates to the adult epilepsy team were found to be as low as 18% whereas the average for the control group was over 37%. Only 5% were referred to drug and alcohol teams. Safety advice documentation was found to as low as 14%, with an average of 32% for the control groups. Conclusions: There were discrepancies in the care of alcohol and substance misuse patients. More research is warranted in this area to increase referrals to acute neurology, epilepsy teams and drug and alcohol services and to understand what causes these differences. It is also important to establish why fewer patients were known to adult epilepsy services and why the documentation of safety advice was low in these cohorts.

Ellie Chilcott

Aneurin Bevan University Health Board

The COVID-19 coronavirus pandemic has led to many healthcare services being delivered by telemedicine for the first time. We sought to evaluate the experiences of people with epilepsy during this unplanned change to assess their satisfaction with telephone clinics. We conducted a telephone survey of people over 16 years age with epilepsy attending a specialist epilepsy nurse follow-up clinic during May to September 2020. We used 14 statements about their experience with a Likert scale (‘Strongly Agree’ to ‘Strongly Disagree’) from a previously validated satisfaction survey. They were also asked for additional comments and information on access to clinic and time taken off work. They were interviewed by an administrative staff member within 4 weeks of the appointment. We surveyed 65 clinic patients in total. The median age of patients was 38 years and 43 patients were women (66%). Of these, 59 patients answered survey questions themselves. In 4 cases, a parent answered the survey and in 2 cases, another caregiver answered. We found that 59 people agreed that they felt comfortable communicating with the doctor or nurse by telephone. 61 people agreed that they felt the doctor or nurse listened to them. 53 people agreed that there was enough time to deal with everything that needed to be covered. 35 people agreed that a telephone call made it easier for them to access clinic services and 41 agreed that a telephone call is an acceptable way to receive epilepsy care. 35 people agreed that they preferred a telephone call to a face-to-face appointment. 50 people agreed that, overall, they were satisfied with the quality of services provided by a telephone call. 32 people said they would usually be driven to an appointment by a family member or friend. Seven drove a car themselves, six used a taxi, three used hospital transport, two used a bus, one walked and the remainder used multiple modes of transport. 25 people were not working and 18 people needed to take off at least half a day from their work to attend a face-to-face appointment. 14 people made additional positive comments about telephone clinics. Four expressed reservations including lack of face-to-face contact at all, confidentiality and difficulty listing seizure dates. We found that telephone clinics were successful overall but need to perform a larger audit including consultant clinics to see if the findings are replicated. Another limitation is that people were questioned during the pandemic and may not reflect future attitudes to telephone clinics.

Krishna Chinthapalli

Frimley Health NHS Foundation Trust

Background: 10% of people worldwide will have a seizure at some point in their lifetime. Seizures are a common presentation to the emergency department (ED), are prone to misdiagnosis and are often intertwined with complex comorbidity requiring the support of specialist services. This is of particular concern in vulnerable patient groups, who are more likely to have barriers to managing their condition independently. Aims: This study aimed to assess the management of acute presentations with seizures in hospital and to explore factors that may influence that management, or patient outcomes. We also sought to explore the impact of the COVID-19 pandemic on seizure presentations to hospital. Outcomes in patients with learning disabilities, active significant mental health concerns and the most socioeconomically deprived patients were to be compared to those of the general cohort. Methods: Investigations and outcomes (referrals and admissions) were recorded for 120 patients across 2 cohorts; the first being pre-pandemic (September 2019) and the second being during (December 2020). Use of ED notes allowed for identification of patients with learning disabilities, active significant mental health concerns and those from the most deprived backgrounds (using the Welsh Index of Multiple Deprivation). Results: There were fewer seizure presentations in December 2020 compared to September 2019, and a shift in composition of patients with a greater proportion of patients presenting with organic seizures and fewer presenting with non-epileptic attacks. CT heads were carried out in a significant proportion of patients (45%), likely representing improper use of limited resources. Patients were admitted in 36% of cases . There were low referral rates, both to acute neurology and to the adult epilepsy team (28%, 32% respectively). Patients with active significant mental health concerns were the least likely to be referred to acute neurology or to be admitted. Conclusions: There were fewer presentations to the ED with seizures during the peak of the COVID-19 pandemic, and these were more likely to be organic seizures requiring admission to hospital. However, referrals to acute neurology and the epilepsy team remained low. Failure to refer people with seizures to specialist services prevents them from receiving appropriate support, and disappointingly this was seen most in those with active significant mental health concerns, who are most likely to need this support. The degree of unmet need identified by this study is of great concern given the significant strain neurology and epilepsy services are already under in managing their existing workload.

Bethan Ellis

Cardiff University

Background: Anterior temporal lobe resection (ATLR) is associated with up to a 40% risk of episodic memory deficits. After surgery, memory functions may ‘recover’ after the initial hit. Little is known of long-term plasticity effects after ATLR. Aims: We aim to describe dynamic plasticity of the memory encoding networks long-term after ATLR. Methods: We studied 22 subjects (six controls, eight left TLE (LTLE) and eight right TLE (RTLE)) across four time points (pre-surgery, 4 ,12 months and between 7 and 10 years after ATLR), and equivalent time points in controls. All participants performed a memory-encoding fMRI paradigm of faces and words and underwent standard memory neuropsychometry (list and design learning). A flexible factorial design was used to examine changes in fMRI activations in patients between 12 months and long-term after ATLR, relative to longitudinal changes in controls. Memory scores were correlated with change in activations using an ANCOVA. Results: The majority of patients were seizure free (100% LTLE and 87.5% RTLE with ILAE seizure outcome 1 and 2). LTLE: For verbal encoding, LTLE showed increased activations within the left PFC (prefrontal cortex) and reduced right PFC, bilateral FG (fusiform gyrus) and right parahippocampal gyrus activations. Ipsilesional temporal neocortical activations and bilateral FG activations correlated with higher verbal learning scores long-term after surgery. At face encoding, there was increased bilateral PFC activations. Improvement in visual memory correlated with increased contralateral amygdala, insula and PFC activations. RTLE: For face encoding, RTLE showed increased activations within the PFC bilaterally. Right FG and right temporo-parietal activations positively correlated with design-learning. For word encoding, there were increased activations in the right FG and bilateral (> right) PFC. The increased right PFC activations positively correlated with list-learning scores. Conclusion: Long-term after ATLR, there is continued brain plasticity with significantly increased PFC activations compared to 12 months after epilepsy surgery. In LTLE, further recruitment of the contralateral MTL and PFC long-term is supportive of visual memory functions whilst in RTLE efficient plasticity was associated with more ipsilesional FG and neocortical activations, implying less widespread plasticity effects in RTLE. Ongoing long-term plasticity suggests that intervention such as rehabilitation may be efficacious even long-term after epilepsy surgery.

Marine Fleury

UCL Queen Square Institute of Neurology

Background: Since 2015, the Royal Free Hospital has utilised home video telemetry (HVT) to aid the diagnosis of atypical paroxysmal events. HVT allows for extended telemetry studies within the patient’s own home. This service evaluation aimed to determine the diagnostic utility of HVT and identify which patients had the highest diagnostic yield. The objective was to design a structured referral pathway for selected patients. Methods: Documentation of all patients who completed a HVT study between 2015-2018 was retrospectively reviewed. Reason for referral and prior investigations were recorded. Diagnosis, medication, and seizure frequency were compared pre- and post- HVT. Footage and seizure diary for each patient was reviewed to identify the number of seizures detected by EEG, reported by the patient, and captured on camera. Results: Of those referred to the service (n=22) 19 agreed to testing and 13 captured usable footage. Reasons for refusal included language barrier and lack of confidence managing equipment. Failure to capture usable HVT data most commonly related to patient error. 61% experienced a seizure event. HVT captured 78.6% of these events, the majority (85.7%) had no EEG correlate. Where clear documentation was available (n=11) HVT confirmed (27.3%) or changed (18.2%) the diagnosis of 45.5% of patients who collected HVT footage. This resulted in a medication change for 20% of the patients (n=10). Although patient numbers were limited, all patients experiencing one or more events per week experienced an event during recording. HVT was most useful for the investigation of dissociative seizures confirming or refuting the diagnosis in 60%. For a small group HVT had no impact despite capturing an event. This may be due to long delays between referral and investigation with intervening changes in clinical information removing the need for HVT. Conclusions: HVT can inform diagnosis and influence treatment decisions when there is diagnostic uncertainty. Inadequate footage commonly related to patient error emphasising the need for robust instruction and careful patient selection. HVT was most useful in the diagnosis of dissociative seizures where seizure burden is often higher. High seizure frequency predicts event capture; hence we propose that patients suitable for HVT studies should be experiencing at least one seizure a week. Limitations include low patient numbers, retrospective review, and incomplete documentation. Using this information, a patient referral pathway aiming to maximise diagnostic yield of HVT was proposed.

Martin Furlepa

East and North Hertfordshire NHS Trust

Aim: The aim of this project was to identify a group of children and young people with generalised epilepsies who had been treated within our secondary care paediatric epilepsy service. We wanted to compare the treatment they received with contemporaneous guidance and to evaluate the effectiveness of that treatment. Method: We identified children with childhood absence epilepsy (CAE), genetic generalised epilepsies (GGE) and other generalised epilepsies of known or cryptogenic cause (other) from our clinic database. These were children who had been discharged between 2008 and 2020. Details about their treatment and clinical course was obtained from a database of clinic letters. Analysis regarding outcomes was conducted for the group as a whole and for the three subgroups. Results: Data was collected for 121 children and young people. There were 31 children with CAE, 75 with GGE, and 15 with ‘other’ types. Overall for the group, 84 (69.4%) were golden responders (seizure control with the first drug with no side effects). 37 children failed the first line drug, 9 due to side effects and 28 due to lack of efficacy. Of these, 21 (57%) had a good response to the second line drug but 16 failed, 1 due to side effects and 15 due to lack of efficacy. These 16 were tried on a variety of other drugs and 11 subsequently became seizure free on monotherapy and 2 on polytherapy. 3/121 continued with seizures (one in the GGE group and 2 in the ‘other’ group). In all groups and overall, sodium valproate was used first line most often (58%) with good efficacy (this was prior to PREVENT guidance), with ethosuximide second line in CAE and levetiracetam second line in the other two groups. Conclusion: In practice at a secondary care level treatment for this historical cohort of paediatric generalised epilepsies is effective and well tolerated. Treatment patterns change as evidence accumulates and new guidance emerges and we are using ethosuximide more often now first line for CAE and sodium valproate much less often for girls over 10 years.

Daniel Hindley

Bolton NHS Foundation Trust

Introduction: With around 50% of adults diagnosed with active epilepsy suffering from at least one medical comorbid condition, comorbidities have been shown to impact management, prognosis and quality of life of patients with epilepsy. This is also reflected in the 2017 ILAE classification of epilepsies, in which comorbidities are considered at each stage of classification and as an integral part of the epilepsy management. Gaining a better understanding of these associations and the burden comorbid conditions have on adults with epilepsy can help with tailoring treatments and developing epilepsy services according to the needs of epilepsy patients. The aim of this project was to evaluate the level of documentation of comorbidities during outpatient appointments at the Epilepsy Centre at St George’s University Hospitals NHS Foundation Trust; to identify impact and patterns of comorbidities and to identify service development needs and priorities in the St George’s population. Methods: Data from a sample of consecutive patients seen at the epilepsy outpatient clinics between October and December 2020 was analysed. The level of documentation of comorbidities was recorded. Comorbidities were categorised based on Sander’s classification into causative, mutual, reciprocal, coincidental and resultant, and the number of mechanisms involved. Results: Data from a total of 369 patients (180 male:189 female, mean age 44 ± 18) were retrospectively reviewed. The median number of comorbidities was 2. Comorbidities were documented in 87%. Types of comorbidities were medical in 54%, neurological in 61% and psychiatric in 41%. Mechanisms of comorbidities were causative in 42.3%, reciprocal in 31.7%, mutual in 18.4%, while 4.3% had resultant comorbidities, and 86.2% had coincidental comorbidities. Around 19% of patients were classified as drug resistant. These patients showed significantly higher numbers of comorbidities as compared to those without drug-resistant epilepsy (3.57 ± 2.17 vs. 2.67 ± 2.23, t= 3.076, p= 0.002). Conclusions: Our centre showed a high level of documentation of comorbidities. Half of our patients present with concomitant medical problems and half of them present with at least 2 comorbidities. Patients with drug-resistant epilepsy are burdened by a higher number of comorbidities mediated by multiple mechanisms. Findings from our data support the need for integrated clinical pathways in epilepsy centres in order to treat and prevent health needs of people with epilepsy.

Ejatu Jalloh

St George's University of London

Background: Disruptions in central autonomic processes in people with epilepsy have been studied through evaluation of heart rate variability (HRV). Decreased HRV appears in epilepsy compared to healthy controls, suggesting a shift in autonomic balance towards sympathetic dominance; recent studies have associated HRV changes with seizure severity and outcome of interventions. However, the processes underlying these autonomic changes remain unclear. We examined the nature of these changes by assessing alterations in whole-brain functional connectivity, and relating those alterations to HRV. Methods: We examined regional brain activity and functional organisation in 28 drug-resistant epilepsy patients and 16 healthy controls using resting-state functional magnetic resonance imaging (fMRI). We employed an HRV state-dependent functional connectivity (FC) framework with low and high HRV states derived from the following four cardiac-related variables: 1. RR interval, 2. root mean square of successive differences (RMSSD), 4. low-frequency HRV (0.04-0.15 Hz; LF-HRV) and high-frequency HRV (0.15-0.40 Hz; HF-HRV). The effect of group (epilepsy vs controls), HRV state (low vs high) and the interactions of group and state were assessed using a mixed analysis of variance (ANOVA). We assessed FC within- and between 7 large-scale functional networks consisting of cortical regions and 4 subcortical networks, the amygdala, hippocampus, basal ganglia and thalamus networks. Results: Consistent with previous studies, decreased RR interval (increased heart rate) and decreased HF-HRV appeared in people with epilepsy compared to healthy controls. For both groups, fluctuations in heart rate were positively correlated with BOLD activity in bilateral thalamus and regions of the cerebellum, and negatively correlated with BOLD activity in the insula, putamen, superior temporal gyrus and inferior frontal gyrus. Connectivity strength in patients between right thalamus and ventral attention network (mainly insula) increased in the high LF-HRV state compared to low LF-HRV; the opposite trend appeared in healthy controls. A similar pattern emerged for connectivity between the thalamus and basal ganglia. Conclusion: The findings suggest that resting connectivity patterns between the thalamus and other structures underlying HRV expression are modified in people with drug-resistant epilepsy compared to healthy controls.

Michalis Kassinopoulos

UCL Queen Square Institute of Neurology, University College London, London, United Kingdom

Objective: In children with drug-resistant epilepsy (DRE), resective, ablative and disconnective surgery may not be feasible or may fail. Neuromodulation in the form of deep brain stimulation (DBS) and responsive neurostimulation (RNS) may be viable treatment options, however, evidence for their efficacies in children is currently limited. This systematic review aims to summarise the literature on DBS and RNS for the treatment of DRE in the paediatric population. Specifically, we focus on the currently available data for reported indications, neuromodulation targets, clinical efficacy and safety outcomes. Methods: Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed throughout this systematic review. Five electronic databases were searched from their inception to February 19, 2021. Inclusion criteria were 1) studies with at least one paediatric patient (age <19 years) undergoing DBS and/or RNS for DRE; 2) retrospective, prospective, randomized or non-randomized controlled studies, case series and case reports. Exclusion criteria were 1) letters, commentaries, conference abstracts, reviews; 2) studies where full text was unavailable. Results: A total of 35 studies were selected, identifying 72 and 46 patients undergoing DBS and RNS respectively (ages 4-18 years). Various epilepsy aetiologies and seizure types were described in both cohorts. Overall, 75% of patients had a seizure reduction >50% following DBS (amongst whom six were seizure-free), at a median follow-up of 14 months (range 1-100 months). In an exploratory univariate analysis of factors associated with favourable response, follow-up duration was shorter in those with a favourable response (18 vs 33 months, p < 0.05). In the RNS cohort, 62.5% of patients had a seizure reduction >50% following RNS, at a median follow-up of 22 months (range 5-39 months). On closer inspection, 83.3% of patients who had >50% reduction in seizures actually had a >75% reduction with four patients being seizure-free. Conclusions: Overall, both DBS and RNS showed favourable response rates, indicating both techniques should be considered in paediatric patients with DRE. Many of the research needs in this area would be addressed by conducting high-quality clinical trials and establishing an international registry for paediatric neuromodulation, ensuring the robust prospective collection of predictive variables and outcomes.

Mehdi Khan

University College London Medical School, London, UK

Objective: Children undergoing stereo-electroencephalography (SEEG)-guided epilepsy surgery represent a complex cohort. We aimed to determine whether the proportion of putative seizure onset zone (SOZ) contacts resected is associated with seizure outcome in a cohort of children undergoing SEEG-guided, resective epilepsy surgery. Materials and Methods: Patients who underwent SEEG-guided, resective epilepsy surgery over six years were included. The proportion of SOZ contacts resected was determined by co-registration of pre- and post-operative imaging. Seizure outcomes were classified as seizure-free (Engel class I) or not seizure-free (Engel classes II-IV) at last clinical follow-up. Patients without sufficient imaging (either pre- or post-operative) were excluded from the primary analysis but were included in a sensitivity analysis to determine the robustness of the primary analysis. Results: Twenty-two eligible patients were identified (median age at surgery of 10 years, range 5-18 years). Fifteen (68.2%) were seizure-free and 7 (31.8%) not at a median follow-up of 19.5 months (range 3-46 months). On univariate analysis, histopathology, classified into focal cortical dysplasia (FCD), non-diagnostic status (NDS) and hippocampal sclerosis, was the only significant factor associated with seizure freedom (SF)(p=0.01). On post-hoc analysis, NDS significantly associated with non-SF (p=0.01), and FCD with SF (p=0.046). Other factors including the SOZ location, type of surgery, indication for SEEG, duration of epilepsy, follow-up duration, total number of SEEG contacts and non-SOZ SEEG contacts resected did not significantly associate with SF. The percentage of defined SOZ contacts resected ranged from 25-100% (median 78.3%) and was not associated with SF (p=0.89). In a binary logistic regression model, none of the included factors were independently associated with SF. On a sensitivity analysis of 29 patients, all univariate and binary logistic regression model findings from the primary analysis were confirmed. Conclusion: Histopathology is a significant predictor of surgical outcomes in children undergoing SEEG-guided, resective epilepsy surgery. The percentage of SOZ contacts resected was not associated with SF. Other factors such as the neurophysiological definition of SOZ contacts may therefore play an important role.

Mehdi Khan

University College London Medical School, London, UK

Objectives: Stroke is the commonest cause of epileptic seizures in older adults. Risk factors for post-stroke seizure (PSS) and post-stroke epilepsy (PSE) are well known. However, predicting PSS/PSE risk is complex. This study aims to evaluate the predictive accuracy and comparative test performances of PSS/PSE risk prediction models developed to date. Methods: We performed a systematic review and meta-analysis of studies using MEDLINE and EMBASE from database inception to 28th December 2020. The search criteria included all peer-reviewed research articles, developing or validating PSS/PSE risk prediction models for ischemic and/or hemorrhagic stroke. Random-effects meta-analysis was used to generate summary statistics of model performance and receiver operating characteristic curves. Quality appraisal of studies was conducted using PROBAST. Results: Thirteen original studies involving 182,673 stroke patients (mean age: 38 – 74.9 years, 29.4% – 60.9% males), reporting fifteen PSS/PSE risk prediction models were included. The incidence of PSS and PSE was 4.5% and 2.1%, respectively. Cortical involvement, functional deficits, increasing lesion size, and early seizures were the commonest predictors across the models. SeLECT demonstrated greatest predictive accuracy (AUC 0.77 [95%CI: 0.71 – 0.82]) for predicting PSE in ischemic stroke, and CAVE for predicting PSE in hemorrhagic stroke (AUC 0.81 [0.76 – 0.86]). Both SeLECT and CAVE relied on clinical and radiological variables. Fourteen of fifteen studies demonstrated high risk of bias as per PROBAST, with lack of model validation and reporting of performance measures on calibration and discrimination being the commonest reasons for this. Significance: Although risk factors for PSS/PSE are widely documented, this review identified few multivariate models with low risk of bias that synthesise single variables into an individual prediction of seizure risk. SeLECT and CAVE may help personalise clinical management and identify stroke patients at high risk of developing PSS/PSE for recruitment into studies of anti-epileptic drug prophylaxis. Further external validation of such models are warranted.

Seong Hoon Lee

Academic Critical Care & Neurosurgery, NHS Grampian

Establishing imaging predictors of seizure recurrence in patients with a first seizure would potentially allow for earlier diagnosis and treatment. However, few quantitative imaging studies have been conducted at this early stage. The objective of this study was to identify network markers of seizure recurrence in patients presenting with a first seizure using morphometric similarity analysis. Sixty-four patients presenting with a first seizure underwent T1-weighted magnetic resonance imaging using a standard clinical protocol. Average last follow-up for the study was four years, 41 patients were seizure free (SF), and 23 patients had experienced seizure recurrence (SR) after an initial seizure. FreeSurfer was used for cortical parcellation and estimation of cortical thickness, surface area, grey matter volume, mean curvature and mean Gaussian curvature, which were used as morphometric features. To quantify similarity between cortical regions, morphometric similarity between each pair of regions was calculated using Pearson’s correlations between morphometric features to generate individual morphometric similarity matrices. Graph metrics were calculated at a range of network thresholds and analysed using a generalized linear model, with age, sex and total intracranial volume as covariates. We investigated frequently applied network measures in epilepsy research, including betweenness centrality (BC), clustering coefficient (CC), local efficiency (LE) and nodal path length (PL). Permutation tests and randomized null networks were generated to determine significance of network metrics. Patients with SR had higher BC in a network consisting of left and right parietal regions compared to patients with SF, who had higher BC between left temporal and right frontal regions. Higher CC and LE was observed in patients with SR in a network consisting of the left anterior occipital sulcus and preoccipital notch and transverse temporal sulcus, the left suborbital sulcus and right temporal pole also showed increased CC. Patients with SF had higher LE between the left angular gyrus and inferior temporal gyrus. We also found nodal PL was higher is patients with SR in a network consisting of left frontal and parietal regions, and higher in patients with SF in a network consisting of left and right temporal and occipital regions. Our results show that morphometric similarity networks in patients with SR are altered compared to patients who present with a single seizure. Higher CC and LE in patients with SR may suggest that while certain nodes are highly locally efficient, they are more isolated from the rest of the network, which may represent a protective mechanism against the spread of seizures through cortical reorganisation during epileptogenesis (Pedersen et al., NeuroImage Clin, 2015;8;536-542).

Nicola Leek

University of Liverpool

Background: Nocturnal seizures can be difficult to identify, characterise and quantify. The gold standard for evaluating seizures is video EEG (vEEG). Long waiting times can hamper follow-up care whilst equipment toleration and irregular seizures can prevent definite identification/quantification. Nelli®, produced by Neuro Events laboratory, is a prescription-only device, indicated for use as an adjunct to seizure monitoring of people with epilepsy (PWE) in a home or healthcare facility. It utilises automated analysis of audio/video (media) data, collected via a personal recording unit (PRU), to identify epileptic and non-epileptic seizure events. Clinicians characterize ictal and peri-ictal events from media data and provide objective summaries of semiological components of identified events. Aim To evaluate - • the NELLI seizure monitoring service in terms of characterising nocturnal paroxysmal events • the acceptability of the system to PWE and caregivers Method: PWE with intellectual disability (ID) from Cornwall UK and Wolverhampton UK were supplied with the NELLI device for 14 to 28 days with informed consent or using the Mental Capacity Act research best interest processes where necessary. Data was transferred remotely ensuring GDPR compliance. Participant demographics, baseline and end seizure characteristics was collected. Semi-structured interviews conducted for usability metrics from clinicians and PWE/carers. All participants received a report detailing the paroxysmal events recorded and a scatter graph showing the frequency, type and timing of each event. Seizures were classified and described. Event video recordings were available for clinical review. Results: A total of 21 PWE undertook the evaluation (Cornwall: 15, Wolverhampton: 6). In both centres the device was well accepted by both carers and clinicians. Key findings included a significant minority of seizures in 6 participants going unrecognised by carers and caregivers commenting on improved peace of mind. Discussion: It is well recognised that PWE can be vulnerable to restricted use to vEEG either due to tolerability issues (ID), lack of support, or limited vEEG availability. This can restrict accurate assessment particularly for nocturnal seizures. NELLI can provide a broader picture given its longer time span and home-based environment. NELLI supported clinical suspicion but with mixed results that varied based on the problem complexity. NELLI offered a higher level of care to current care that uses clinical and subjective descriptors but lower than vEEG. Coupling NELLI to a portable EEG could improve diagnostic utility and provide a community-based solution thus enhancing long term care and possibly reducing care costs.

Sarah Lennard

Cornwall Partnership NHS Foundation Trust

Introduction: Reflex epilepsies are often clinically striking, and the precipitant can act as a clinical marker to distinguish individuals likely to have a similar cause. We present the case of a woman’s chronic drug-resistant focal epilepsy with musicogenic seizures and how a biomarker associated with this presentation may have therapeutic implications. Case description: A 36-year-old woman had her first (unknown-onset tonic-clonic) seizure aged 7 years. She had no further seizures until she was 26, when she began to have focal cognitive seizures with déjà vu and oral automatisms, both with and without impaired awareness, as well as focal-to-bilateral tonic-clonic seizures from sleep. She had a normal neurodevelopmental history, hypothyroidism, and no family history of epilepsy. MRI brain was unremarkable. Initial EEG captured a typical focal cognitive seizure briefly preceded by epileptiform discharges in the right anterior temporal region. After intolerance of carbamazepine, her seizures demonstrated resistance to levetiracetam, lamotrigine, topiramate, pregabalin and lacosamide, with a seizure frequency of 5–7 per week. On referral to our service, we discovered that her seizures had a propensity to occur while she listened to music. Repeat EEG while listening to music with headphones captured a focal seizure involving unilateral left eye blinking, oral automatisms, and dystonic posturing of the right arm. This was associated with rhythmic sharpened theta transients over the left temporal region. We also captured an independent interictal sharpened transient over the right temporal region. Her serum anti-glutamic acid decarboxylase 65–immunoglobulin G (GAD65-IgG) level was greater than 2000 IU/ml and after counselling she has commenced a trial of prednisolone. Discussion: GAD65-IgG is associated with progressive encephalomyelitis with rigidity and myoclonus, stiff-person syndrome, cerebellar ataxia, and autoimmune encephalitis, as well as type 1 diabetes mellitus, autoimmune thyroid disease, and pernicious anaemia. Over recent years, increasing evidence has linked GAD65-IgG to focal epilepsies, particularly with musicogenic seizures. The aetiopathogenic role of antibodies to this intracellular enzyme remains unclear, and the limited published evidence of responsiveness to immunomodulatory therapy suggests mixed results. However, GAD65-IgG implicates autoimmunity and a disorder of GABAergic neurons. Increasing awareness of this serological association with chronic focal epilepsy with musicogenic seizures will hopefully allow recognition of sufficient cases to facilitate formal trials of rational therapeutic options, such as immune and GABAA modulating therapies. The identification of a disease-specific treatment is important, given the typical presentation is of a young woman with drug-resistant epilepsy localisable to bitemporal foci.

David Lewis-Smith

Newcastle University Translational and Clinical Research Institute

Purpose: Employment is indispensable to people with epilepsy (PWE) in financial, psychological, social and cultural aspects. Hong Kong is a well-developed economy in Southeast Asia. We aimed to investigate the hurdles in employment subjectively perceived by the PWE. Method: A cross-sectional questionnaire-based study was conducted in our institution in 2019. PWE within working age of 18-65 years old without intellectual disability were eligible. The recruited PWE were divided in employed and unemployed groups depending on the employment status at time of recruitment. A self-administered questionnaire focusing on patient’s perception on employment issue was given to each subject. The responds were mostly in 5-point Likert scale. Responds were compared between the 2 groups by Spearman correlation. Statistical significance was considered if p<0.05. Results: A total of 138 PWE were recruited. Mean age was 42, sex ratio (male: female) was 1:1.1, drug-resistant epilepsy was 34% and median Charlson’s comorbidity score was 0. Ninety-three percent had education of secondary school or above. One third (33%) were unemployed. In the unemployed group, subjects were more common to agree that frequency of seizure would have negative impact on chance of employment (ρ=-0.191, p=0.025); low skill or education level was a major hurdle to employment (ρ=-0.223, p=0.009); and stigma of epilepsy was a major hurdle to employment. There were no statistically significant differences between the 2 groups in the following: diagnosis of epilepsy had negative impact in job finding or maintenance (ρ=-0.109, p=0.204); severity in seizure semiology had negative impact on employment (ρ=-0.137, p=0.109); epilepsy had negative impact on relationships with supervisors (ρ=-0.130, p=0.129); epilepsy had negative impact on relationships with colleagues (ρ=-0.114, p=0.183); worries about occupational hazards brought by epilepsy was a major hurdle to employment (ρ=-0.166, p=0.052); side-effects of the anticonvulsants were major hurdle to employment (ρ=-0.045, p=0.06); psychiatric comorbidity was major hurdle to employment (ρ=-0.140, p=0.100); frequent medical consultations were the major hurdle to employment (ρ=-0.004, p=0.963); lack of motivation was the major hurdle to employment (ρ=-0.109, p=0.202); stress at work was the major hurdle to employment (ρ=-0.030, p=0.686). Conclusions: PWE commonly perceive that seizure control, lack of skills and education and stigma of epilepsy are common hurdles to employment especially among those who are unemployed.

Cheuk Nam Rachel Lo

University of Hong Kong

People with photosensitive epilepsy (PSE) are prone to seizures evoked by visual stimuli. PSE is particularly relevant as a model to understand epilepsy, being used within clinical trials to test the efficacy of anti-seizure medication. A better understanding of the pathophysiology of PSE may have an impact not only on people with PSE but also on the diagnosis and treatments of epilepsy. Studies have found evidence for both occipital and widespread cortical hyperexcitability in PSE [1]. Here, we aimed to find whether we could identify a widespread and/or occipital increased ictogenic propensity from interictal EEG in PSE relative to individuals with epilepsy but without PSE. To evaluate network-wide and local ictogenic propensity, we used the concepts of brain network ictogenicity (BNI) and node ictogenicity (NI), respectively [2]. BNI is a measure of how likely a brain network is of generating seizures in computer simulations. These simulations consist of placing a mathematical model of epilepsy into a functional network and compute the resulting brain dynamics. Brain networks more susceptible to seizures are expected to produce more seizure-like activity in the simulations [2]. NI is assessed by removing regions from the functional network and evaluating the resulting altered BNI. Regions whose removal produce a higher reduction of BNI are considered more ictogenic. We considered two groups of individuals with idiopathic generalised epilepsy, 26 individuals that had a photoparoxysmal response (PPR) during intermittent photic stimulation (the PPR group), and 24 individuals that did not have PPR (the non-PPR group). We tested two hypotheses: (i) the PPR group has a higher BNI than the non-PPR group; and (ii) the PPR group has a higher occipital NI than the non-PPR group. We observed that the BNI is not significantly different between the two groups. This result suggests that our PPR group did not have a higher widespread ictogenic propensity than the non-PPR group. In contrast, we found that the PPR group had a statistically significantly higher occipital NI than the non-PPR group. This result suggests that the occipital region is particularly prone to induce seizures in people with PSE, and that this susceptibility can be probed from interictal EEG. More generally, our results show that computational analysis of interictal EEG may be used to diagnose PSE without the need of photic stimulation. [1] Padmanaban, V., et al. (2019). Brain Research, 1703, 18-25. doi:10.1016/j.brainres.2018.07.025 [2] Lopes, M.A., et al. (2017). PLoS Computational Biology, 13(8), e1005637. doi:10.1371/journal.pcbi.1005637

Marinho Lopes

Cardiff University Brain Research Imaging Centre, School of Psychology, Cardiff University

Purpose: Patients with Sturge-Weber Syndrome (SWS) experience varying degrees of neurological problems – including epilepsy, hemiparesis, learning disabilities, and stroke-like episodes. Whilst the range of clinical problems experienced by children with SWS is well recognised, the spectrum of clinical presentation and its treatment during adulthood has been relatively neglected in the literature to-date. This study explored the natural history of epileptic and non-epileptic seizures into adulthood in patients with SWS, and their treatment, and investigated whether any clinical factors predict which symptoms a patient will experience during adulthood. Methods: A retrospective case-note review of a cohort of 26 adults with SWS at the National Hospital for Neurology and Neurosurgery (NHNN) was conducted. Childhood data was also recorded, where available, to enable review of change/development of symptoms over time. Key findings: The course of epilepsy showed some improvement in adulthood – seventeen adults continued to have seizures, whilst six patients gained seizure-freedom, and no one had adult-onset seizures. However, seizures did worsen for some patients – for example, a change from experiencing focal seizures during childhood to experiencing generalised seizures during adulthood. The mean number of anti-epileptic drugs (AEDs) taken at the time of the study (of the patients currently taking AEDs) is 3.8. Although no factors reached statistical significance regarding predicting continued epilepsy in adulthood, male sex, bilateral cortical involvement, having had epilepsy surgery, and having a more severe learning disability, may be important. Surprisingly, experience of multifocal versus single focus seizures in childhood did not affect adult seizure outcome. Non-epileptic seizures (NES) also began during adulthood for four patients, all of whom were female and had a childhood history of epileptic seizures. Two patients with NES were also experiencing mental health problems in adulthood. Significance: By adulthood, there is some degree of improvement in epilepsy; whilst NES may occur for the first time. It may be that structural brain abnormalities that cause epileptic seizures improve with time for some patients. Whilst the majority of the results did not survive adjustments for multiple comparisons, some interesting trends appeared, which require further investigation in a multi-centre national audit. Patients with more neurologically severe presentations during childhood may continue to experience epileptic seizures as adults. Careful monitoring and screening are needed during adulthood, to detect changes and newly developing symptoms (such as worsened or new epileptic seizure types, and NES) and target treatment promptly.

Rhian Male

Department of Clinical and Experimental Epilepsy, UCL

Purpose: To investigate if the examination of retinal microstructures and retinal nerve fibre layer (RNFL) thickness using the Optical coherence tomography (OCT) could reveal a structural cause behind the wide-field multifocal electrophysiology (WF-mfERG) results in patients exposed to vigabatrin. Setting: Electrophysiology unit. Method: This is a prospective study of 28 participants who have all been exposed to vigabatrin. The subjects were analysed based on their WF-mfERG results where they are grouped as ‘delayed’ (based on 2ms delay in peripheral field compared to central field), or non-delayed. All patients underwent OCT macula volume and OCT RNFL. Results: Analysis of the OCT microstructures based on volume reveal distinctive thickness difference in the inner retinal layers of those in the ‘delayed’ group (p<0.001). Further analysis of the retinal microstructures reveal persistent significant difference across the outer nuclear layer (ONL)(p<0.01), ganglion cell layer(GCL) (p<0.01) and the inner plexiform (IPL) (p<0.01), but not the outer retinal layers. Conclusion: The results we observed points towards damage in the ONL, GCL and IPL; which concurs with historical pathological reports in animal studies. Our observations are striking and could be helpful in leading the way for improved and more objective methods for monitoring of vigabatrin related visual toxicity.

Kaleena Michael

University of Glasgow

Introduction: Research demonstrates that psychiatric conditions are more prevalent in people with epilepsy. In particular, comorbid anxiety and depression has been linked to reduced quality of life, poorer seizure control and reduced tolerance to anti-epileptic drugs. Patient reported outcome measures (PROMs) are validated questionnaires, and are often used to assess symptoms of anxiety and depression. Routine collection of PROMs may be associated with improved health outcomes. Aims: to describe the burden of anxiety and depression in patients under the Aneurin Bevan Epilepsy Specialist Team (ABEST). To evaluate the effect of interventions for anxiety and depression on PROMs and seizure frequency. Methods: PROMs from adults under ABEST were collected prospectively between November 2020 and June 2021. Scores were examined before and after receiving interventions for mood and anxiety. Interventions examined during the study included; lifestyle advice, review of antidepressant medication, psychiatric referral and online CBT. Results: 102 patients completed PROMs. 58 of these completed serial follow-up PROMs (57%). 63% of patients scored in the most severe category for anxiety and 55% of patients scored in the most severe category for depression. In an unadjusted analysis, PROMs for depression improved significantly after lifestyle intervention (p=0.0483). There was no significant change in anxiety or suicidal ideation. Overall, 10 patients had their antidepressant medication reviewed and were followed up. Before intervention all 10 scored abnormal for depression; however, after intervention only 6 of these patients scored abnormal for depression. No significant changes were observed for patients receiving psychiatric referral or online CBT. Furthermore, PROMs scores for anxiety, depression and suicidal ideation exhibited a significant positive correlation with the number of seizures per year (p = <0.0001). Conclusion: PROMs may be an effective tool for identifying psychiatric symptoms in people with epilepsy, and for identifying potential interventions to improve these symptoms, and therefore quality of life. Our results indicate the importance of addressing depression and anxiety in people with epilepsy as an integral part of the management of their seizures.

Anastasia Mirza-Davies

Cardiff Medical School

Dissociative seizures (DS) can cause distress and disability comparable to that caused by epileptic seizures. There is a growing body of evidence to suggest that psychological therapy can be effective in treating DS although the treatment approaches are varied. A recent multicentre randomised controlled trial has demonstrated that cognitive behavioural therapy (CBT) plus standardised medical care can improve health-related quality of life, somatic symptoms and psychosocial functioning, but failed to demonstrate a significant reduction in DS frequency compared to standardised medical care alone. Whilst CBT follows a standardised format and is easy to study in a clinical trial, tailored psychotherapy may offer an advantage in clinical practice as it allows more individualised treatment to meet the needs and characteristics of each patient. Here we present outcome data from a developing functional neurological disorders service in the North West of England, offering short term psychotherapy to consecutive eligible patients with a diagnosis of DS. Patients were assessed for clinical and psychological variables before and after psychotherapy using the GAD-7 screening score for generalised anxiety disorder, the Work and Social Adjustment Scale (WSAS) to assess functional status, the PHQ-15 somatic symptoms scale, PHQ-9 depression score and PCLC PTSD symptom score. 86 patients (67 female) were referred for psychotherapy with a clinical diagnosis of DS. The mean number of therapy sessions was 5.94 (95% CI 5.04 to 6.84). Patients had significant improvements in GAD-7 sumscores (Z=-2.549, p=0.011), PHQ-15 (p=0.028), PHQ-9 (Z=-3.202, p<0.001) as well as PCLC PTSD symptom scores (Z=-2.288, p=0.022), comparing pre and post treatment values. Whilst there was a reduction in the median WSAS scores comparing pre and post treatment values (30 to 22.5), this difference was not significant (Z=-1.932, p=0.053). DS frequency improved in 67.4% of patients (95% CI 52.0% to 80.5%). Although this is an observational study, our findings support the notion that tailored psychotherapy is a clinically effective intervention to improve psychological status, quality of life and dissociative seizure frequency in a real-world clinical setting. Further work is currently underway to assess outcomes over a longer duration to ascertain whether our findings are reproducible and the benefits are sustained.

James Mitchell

The Walton Centre NHS FT, Liverpool

A Core Outcome Set (COS) is a standardised list of outcomes that should be reported as a minimum in all trials. It is developed using consensus methods to ensure that it includes what is important to patients as well as healthcare professionals and researchers. In epilepsy, the choice of outcome measures varies widely among existing studies, particularly in randomised controlled trials (RCTs). This diminishes opportunities for informed decision making, contributes to research waste and is a barrier to integrating findings from multiple RCTs in systematic reviews and meta analyses. COS facilitate the undertaking of trials that are relevant to patients and health services and help standardise trials methodology so more meaningful results can be obtained from systematic review and meta-analysis. Given that no COS for adults with epilepsy currently exists, we are developing an internationally derived COS specific to adult epilepsy treatment trials. We have undertaken a rapid review of the qualitative literature exploring experiences of people with epilepsy (PWE), to identify core concepts that map to potential measurable outcomes. This work represents the views of over 2000 PWE and 600 carers across 6 continents. We have also undertaken a rapid review of outcomes already measured in phase 3 and phase 4 epilepsy specific RCTs registered on the clinicaltrials.gov website to generate a long list of potential measurable outcomes to take through an international consensus process. International stakeholder groups involved include health care professionals, PWE and researchers. Consensus will be sought by a two-round online modified Delphi survey and virtual consensus meeting to ratify a final COS that is representative of all stakeholder views. The results of the COS will be used to inform the recommendation of specific measurement instruments to measure each of the outcomes identified as core. Developing a COS for adults with epilepsy and deriving international consensus, will ensure that meaningful outcomes are measured in future RCTs, ensure that the results of trials are relevant to the needs of PWE, reduce research waste, and ensure that the results of trials assessing the same treatment can be combined in meaningful ways.

James Mitchell

Institute of Systems, Molecular and Integrative Biology (ISMIB), University of Liverpool

Introduction: Mitochondrial diseases comprise the largest group of inherited metabolic disorders. Neurological symptoms include epilepsy, stroke-like episodes, thought to arise due to seizure activity, ataxia, and cognitive impairment. Previous studies implicated severe oxidative phosphorylation (OXPHOS) deficiencies in GABAergic inhibitory interneurons in mitochondrial disease accompanied by neurodegeneration. This study aims to model early-onset neurological symptoms in mitochondrial disease using a novel murine model and to test the hypothesis that underlying hyperexcitability may arise due to neuronal network disinhibition, following mitochondrial dysfunction in parvalbumin-expressing (PV+) inhibitory interneurons. Methods: A novel mouse model of mitochondrial DNA (mtDNA) depletion selectively within the PV+ cells was generated by a knockout of mitochondrial transcription factor A (Tfam) using cre-loxP system, and characterised at behavioural, electrophysiological, neuropathological and molecular levels using various behavioural testing paradigms, slice electrophysiology, immunohistochemistry and immunofluorescence, as well as quantitative PCR. Data were subjected to normality testing and appropriate parametric or non-parametric tests using GraphPad Prism software. Results and conclusions: A downregulation of complexes I, III and IV of the mitochondrial respiratory chain was confirmed within the PV+ cells of the knockout mice with differential OXPHOS deficiency levels between brain regions (minimum of n = 4 mice per group). The most affected neurons were Purkinje neurons of the cerebellum, that showed a reduction in mtDNA copy number and an upregulation of anaplerosis enzyme pyruvate carboxylase. Changes in the cerebellar cortex and deep cerebellar nuclei (DCN) were accompanied by reactive microglial changes and astrogliosis. C-Fos labelling revealed expression in several Purkinje neurons and DCN neurons in the knockout mice, whereas control littermate mice showed no c-Fos immunopositivity. Further indirect evidence for hyperexcitability and increased intracellular Ca2+ levels was provided by ectopic expression of tyrosine hydroxylase detected in Purkinje neurons. Mutant mice exhibited a progressive phenotype, initiating at 8 weeks of age with tremor, cognitive impairment (assessed by novel object recognition test) and anxiety behaviour (in the elevated plus maze test). Hyper-locomotion (in the open field test) and stargazing (absence-like seizures) were detected at 10 weeks, with severe ataxia observed by 12-13 weeks detected by rotarod test. Electrophysiological analysis of mutant mice revealed hippocampal network perturbation, as male knockout hippocampal slices generated a greater number of bursts in response to GABAA receptor blockade than control slices. The novel mouse model recapitulates the early-onset phenotype of mitochondrial disease and highlights the crucial role of mitochondrial function within the PV+ interneurons.

Elizaveta Olkhova

Newcastle University

Epilepsy is recognised as a dynamic disease, where susceptibility to seizures and seizure characteristics change over time. Specifically, we recently found variable electrographic seizure pathways over a spectrum of possible patterns of network dynamics within individual patients forming a span of distinct seizure dynamics. Additionally, the variability appeared to follow subject-specific circadian or longer timescale modulations. However, whether (interictal) iEEG features can capture signatures of these modulations over different timescales remains unclear. In this work, we analyse continuous intracranial electroencephalographic (iEEG) recordings from video-telemetry units and find fluctuations in iEEG band power over different timescales ranging from minutes up to twelve days. As expected and in agreement with previous studies, we find that all subjects show a circadian fluctuation in their EEG band power. We additionally find other fluctuations of similar magnitude on subject-specific timescales. Importantly, we find that a combination of these fluctuations on different timescales can explain changes in seizure dynamics in most subjects above chance level. These results suggest that subject-specific fluctuations in iEEG band power over timescales of minutes to days may serve as markers of seizure modulating processes. Future work is needed to link the detected fluctuations to the exact biological time-varying processes. Understanding seizure modulating processes enables development of novel treatment strategies that minimise the seizure spread, duration, or severity and therefore clinical impact of seizures.

Mariella Panagiotopoulou

CNNP Lab (www.cnnp-lab.com), Interdisciplinary Computing and Complex BioSystems Group, School of Com

Background: We largely depend upon self-reporting to determine seizure frequency for epilepsy management decisions. However, in patients with typical absence seizures (AS), we were concerned that there was a high rate of misreporting, either by under-reporting (having AS but not reporting them) or over-reporting (reporting AS without actually having them). We therefore set out to determine the misreporting rate in people with absence seizures. Methods: Video-Telemetry (VT) and Ambulatory Electroencephalogram (EEG) reports, from 2015 until May 2021, were analysed together with patient diary sheets. The number of patient-reported AS were then compared to the number of EEG-identified AS. Only patients with an established diagnosis of generalised genetic epilepsy (GGE) or with an EEG report indicative of GGE were included. Patients, who were so imprecise in reporting AS that correlation with EEG was not possible, were excluded. Importantly, data to determine under-reporting were based only on VT EEG data, as it provides sufficient information for unreported seizure detection. Where patients were unable to record seizures (e.g. those with significant learning disabilities), reports of the caregivers and/or relatives were used instead. Results: Forty-two patients (31 female and 11 male; mean age: 33.8yrs ranging from 18-73yrs old) were included in the study. Thirty two patients (76.2%, mean age: 33.8yrs), misreported AS. From the 32 misreporting patients, 22 (68.8%) overreported, 9 (28.1%) underreported and only 1 (3.1%) both underreported and overreported. There was no difference in the age and sex of those who misreport from those who report accurately. However, females tend to overreport more than males (56% vs. 36%, respectively). Conclusion: These results demonstrate that patients with AS highly misreport their seizures, suggesting that self-reported AS are a poor measure to use for treatment decisions.

Joao Pizarro

UCLH

Purpose: Evaluate long-term retention, reasons for discontinuation, efficacy and tolerability of adjunctive brivaracetam (BRV) in adults with focal seizures by number of lifetime antiepileptic drugs (AEDs). Methods: Post-hoc analysis of randomised, double-blind, placebo-controlled (N01358/NCT01261325) and open-label extension (OLE; N01379/NCT01339559) trials of adjunctive brivaracetam in patients (≥16 years) with focal seizures randomised to BRV (100/200mg/day) or placebo (PBO). Outcomes were assessed from first day of BRV treatment by number of lifetime AEDs (stopped before/ongoing at BRV initiation). Results: 740 patients received BRV in the double-blind or OLE trial (safety population; median modal dose: 200mg/day[n=737]; median treatment duration: 973.5 days). Proportions of patients with 1-2(n=102), 3-4(n=154), 5-6(n=156) and ≥7(n=328) lifetime AEDs who completed the OLE were 52.9%, 45.5%, 52.6% and 39.9%, respectively. Main reason for BRV discontinuation was consent withdrawn (1-4 lifetime AEDs) or lack-of-efficacy (≥5 lifetime AEDs). Kaplan-Meier estimated retention on BRV was 83.2%, 71.3%, 72.4% and 65.9% at 12 months and 63.0%, 56.0%, 59.3% and 44.1% at 36 months in patients with 1-2, 3-4, 5-6 and ≥7 lifetime AEDs, respectively. Estimated proportion of patients discontinuing BRV due to lack-of-efficacy or treatment-emergent adverse events (TEAEs) increased with number of lifetime AEDs. During the entire treatment period (efficacy population), median percent reductions from baseline in focal seizure frequency/28 days in patients with 1-2(n=102), 3-4(n=153), 5-6(n=156) and ≥7(n=327) lifetime AEDs were 76.3%, 56.7%, 50.9%, and 39.6%; 50% responder rates were 66.7%, 56.9%, 51.9% and 39.8%; and 75% responder rates were 51.0%, 34.0%, 26.3%, and 19.6%. Continuous seizure-freedom for ≥1 year at any time during BRV treatment was 35.3%, 15.7%, 12.2%, and 6.1%. In patients completing ≥48, ≥96 or ≥144 weeks, median percent reduction in focal seizure frequency and 50% and 75% responder rates were sustained over time in all lifetime AED subgroups. Incidences of TEAEs (safety population) were 86.3%, 84.4%, 90.4% and 90.5% in patients with 1-2, 3-4, 5-6 and ≥7 lifetime AEDs, respectively. Discontinuations due to TEAEs occurred in 7.8%, 10.4%, 14.1% and 20.1% of patients. Conclusions: Adults with focal seizures exposed to fewer AEDs had numerically higher 1- and 3-year retention rates and were less likely to discontinue long-term adjunctive BRV treatment due to lack-of-efficacy or TEAEs. Long-term efficacy was highest in patients with 1-2 lifetime AEDs and decreased by number of lifetime AEDs, though patients exposed to ≥7 AEDs could also benefit from BRV. Efficacy of BRV was sustained over time independent of number of lifetime AEDs. Funding: UCB Pharma-sponsored.

Bjorn Steiniger-Brach

Mid-Atlantic Epilepsy and Sleep Center, Bethesda, MD, USA

Purpose: To evaluate efficacy and tolerability of lacosamide (≤12mg/kg/day or ≤400mg/day) as adjunctive treatment for uncontrolled primary generalised tonic-clonic seizures (PGTCS) in patients (≥4 years) with idiopathic generalised epilepsy (IGE). Methods: Phase 3, double-blind, randomised, placebo-controlled trial (SP0982/NCT02408523) in patients with IGE and uncontrolled PGTCS taking 1-3 concomitant antiepileptic drugs. Primary outcome was time to second PGTCS during 24-week treatment. We present post-hoc subgroup analyses by epilepsy syndrome and for patients with history of absence/myoclonic seizures. Results: 242 patients were randomised and received ≥1 trial dose (lacosamide/placebo: n=121/121). All patients (mean age 27.7 years; 58.7% female) had history of generalised-onset seizures. 85.1%/90.9% of patients on lacosamide/placebo completed trial. Median treatment duration with lacosamide/placebo was 143/65 days. Risk of developing second PGTCS during 24-week treatment was 46% lower with lacosamide than placebo overall (Kaplan-Meier survival estimates 55.27%/33.37%; hazard ratio 0.540 [95% CI 0.377-0.774]; p<0.001; n=118/121), in all epilepsy syndromes (childhood absence epilepsy 72.92%/0% [n=8/6]; juvenile absence epilepsy 40.62%/28.89% [n=13/15]; juvenile myoclonic epilepsy 52.30%/25.77% [n=33/42]; epilepsy with grand mal seizures on awakening 57.44%/33.55% [n=15/19]; other generalised idiopathic epilepsies 58.83%/40.29% [n=53/54]; epilepsies with seizures precipitated by specific modes of activation 66.67%/50.00% [n=6/6]; other 33.33%/26.67% [n=3/5]), and in patients with history of absence (48.16%/22.74% [n=48/42]) and myoclonic seizures (54.75%/29.03% [n=44/48]). Risk of developing first PGTCS during 24-week treatment was 32% lower with lacosamide than placebo (Kaplan-Meier survival estimates 30.97%/17.27%; hazard ratio 0.683 [95% CI 0.507-0.921]; p=0.012). Kaplan-Meier-estimated freedom from PGTCS at end-of-treatment for lacosamide/placebo was 31.3%/17.2% (difference 14.1%; p=0.011). 79.3%/65.3% of patients on lacosamide/placebo (n=121/121) had treatment-emergent adverse events, most commonly (≥10% patients on lacosamide vs placebo) dizziness (23.1%/5.8%), somnolence (16.5%/14.0%), headache (14.0%/9.9%). Conclusions: Lacosamide was efficacious and generally well-tolerated as adjunctive treatment for uncontrolled PGTCS in patients with IGE. Efficacy findings in analysed subgroups were consistent with the overall population. Funding: UCB Pharma-sponsored

Bjorn Steiniger-Brach

Department of Neurosciences, Monash University & Department of Neurology, Alfred Health, Melbourne,

Rationale: Neurocysticercosis (NCC), a parasitic CNS infection endemic to developing nations, is the leading global cause of acquired epilepsy. It is currently unknown why some patients with NCC develop recurrent seizures, although previous research has suggested that ictogenesis is a consequence of abnormal sub-cortical circuitry. There is evidence for a relationship between hippocampal sclerosis and increased seizure incidence in NCC, and other subcortical structures, the thalamus in particular, are increasingly reported as abnormal in new-onset epilepsy. A relative sparsity of biomarker literature has been identified in NCC-based epileptogenesis, and so the present study aimed to assess the putative importance of sub-cortical abnormalities in NCC to the expression of spontaneous recurrent seizures. A common approach in epilepsy research, sub-cortical surface shape analysis, quantifies surface deflation of sub-cortical structures, providing a proxy measurement for localised atrophy. Methods: Clinical histories and 3D-T1 MRI data were acquired from 83 patients with probable NCC, 49 with recurrent seizures and 34 without. The imaging data were examined using an established surface shape segmentation tool (FIRST) included as part of the FSL-ANAT processing pipeline in the FMRIB Software Library, with sex and age included as nuisance regressors in the final model. The FSL-ANAT pipeline reorientated, cropped, bias-field corrected, spatially normalised, brain-extracted, tissue-type segmented and parcellated the subcortical structures for surface shape analysis. The sub-cortical segmentations are input into a non-parametric t-test, which uses a permutation-modelled null to calculate and output multiple-comparison corrected (at p < 0.5) clusters of regional inward surface deflation on a 3D surface mesh. Results: Patients with NCC and recurrent seizures showed significant (corrected p < .05) inward surface deflation in localised regions of the right hippocampus compared to seizure-free NCC patients, when age and sex were modelled as nuisance regressors (Fig. 1). There was no significant or trend level differences in regional shape of the remaining structures between patients presenting with or without seizures. Conclusions: Patients with NCC who present with seizures show evidence of localised hippocampal atrophy compared to patients who present with no seizures. This is in line with recent studies that describe global hippocampal volume reduction in patients with NCC and seizures. Taken together, this may suggest that patients who develop seizures after NCC infection may be predisposed to the development of epileptic seizures in light of precipitating hippocampal abnormalities. Future research should aim to quantify the extent to which, alongside other factors such as perilesional oedema, sub-cortical shape contributes to the multi-factorial aetiology of NCC-based seizures.

Corey Ratcliffe

University of Liverpool

Background: Solute Carrier Family 6 Member 1 (SLC6A1) epileptic encephalopathy is an autosomal dominant genetic disorder characterized by the loss- of- function of one copy of the human SLC6A1 gene that encodes the GABA transporter 1, thus affecting the GABA levels leading to seizures and developmental concerns. Methods: We describe a 6y old girl with epilepsy characterised by absences and recurrent drop attacks and a de novo SLC6A1 mutation variant c.1081C>A p. (Pro361Thr) who had a unique clinical, neurophysiological and neuropsychological evolution. A literature review was also undertaken. Results: The child presented at 3y of age with drop attacks, vacant spells and unsteady gait. She had an uneventful perinatal period but although initial development was normal but over time it was noted that she lagged in her speech sector till by 5y age serial neuropsychological assessments revealed concerns for speech intelligibility, vocabulary limitation and poor literacy acquisition. This deterioration was seen inspite of optimum control of her epilepsy with sodium valproate. Initial EEG findings were unique showing prolonged bursts of irregular generalized 2.5-3.5 Hz slow waves with occasional occult spikes typically starting in the occipital regions. These findings are unlike those typically described in cases with myoclonic astatic epilepsy. In view this genetic panel for epilepsy was undertaken which confirmed the genetic diagnosis. A videoEEG captured herrop attack which ictally revealed generalized low voltage fast paroxysmal activity followed by rhythmic fronto-central rhythmic delta activity. Her epilepsy control was excellent with sodium valproate but in follow up she was noted to have progressive speech concerns and this was corroborated by her neuropsychological evaluations. In view of speech decline a 24h long EEG was conducted again when a striking finding was revealed. Although initial EEGs showed generalized abnormalities this revealed prominent right sided focal frequent fronto-temporal epileptiform abnormalities which were significantly enhanced in NREM sleep. The 3TMRI head was normal. She was referred for speech and language therapy and Sodium valproate was continued. Only one other such case such been reported in literature amongst the 34 till now. Conclusions: Neurophysiologists need awareness about the unique EEG features seen in children presenting with early onset myoclonic astatic epilepsy. Such features should prompt early genetic testing that is not only useful for genetic counselling but correctly addressing therapy with sodium valproate that is found to be most effective here. Unfortunately though, no correlation has been established between seizure control and cognitive outcome.

Munni Ray

Leeds Teaching Hospitals NHS Trust

Dravet syndrome (DS) is a severe and progressive genetic epilepsy that typically begins in the first year of life and is characterized by frequent, prolonged, and refractory seizures, intellectual disability, ataxia/motor abnormalities, behavioral problems, speech impairment, sleep disturbances, and a high risk for sudden unexpected death. Approximately 85% of DS cases are caused by de novo, spontaneous, heterozygous loss of function mutations in the SCN1A gene which encodes the voltage-gated sodium channel type 1 α subunit (Nav1.1) protein. STK-001 is an investigational antisense oligonucleotide (ASO) treatment designed to upregulate Nav1.1 protein expression by leveraging the non-mutant (wild-type) copy of SCN1A gene to restore physiological Nav1.1 levels, thereby potentially reducing occurrence of seizures and significant non-seizure comorbidities. ADMIRAL is a phase 1/2a open-label, multi-center study in the UK of patients ages 2-18 years with a DS diagnosis including disease onset prior to 1-year-old with recurrent seizures (focal motor, hemiconvulsive, or generalized tonic-clonic) and with genetically confirmed SCN1A variant. ADMIRAL primarily aims to assess safety, tolerability, and pharmacokinetics of multiple ascending doses (MAD; 30mg up to 70mg) of intrathecally (IT) administered STK-001. Secondary objectives are the percentage change from baseline in convulsive seizure frequency, overall clinical status, and quality of life. Each dose level will include at least 4 patients grouped by age (two patients 13-18 years followed by two 2-12 years). Additional dose cohorts may be added with maximum enrollment of up to 60 patients. Patients will be observed for ≥28-day baseline period prior to dosing to evaluate seizure frequency. On Day 1, patients undergo cerebrospinal fluid (CSF) collection followed by a single IT administration of STK-001; this is repeated on Days 57 and 85. During 6-month follow-up, CSF is collected to evaluate STK-001 exposure. Adverse events are monitored continuously. At the end of this study, patients meeting criteria will have the option to enter an open-label extension (OLE) study and continue to receive STK-001. Data on seizure frequency, overall clinical status, and quality of life may provide initial evidence of clinical effect. STK-001 has the potential to be the first disease-modifying therapy to address the genetic cause of DS by upregulating Nav1.1 protein levels to potentially reduce both occurrence of seizures and non-seizure comorbidities. Together with the ongoing phase 1/2a MONARCH study in the U.S., data from ADMIRAL may help inform future clinical trials of STK-001.

Stephanie Rock

Great Ormond Street Hospital for Children NHS Foundation Trust

Dravet syndrome (DS) is a severe and progressive genetic epilepsy that typically begins in the first year of life and is characterized by frequent, prolonged, and refractory seizures, intellectual disability, ataxia/motor abnormalities, behavioral problems, speech impairment, sleep disturbances, and a high risk of sudden unexpected death. There remains a need for therapy to reduce seizure frequency (SF) and improve non-seizure comorbidities. There is a lack of prospective long-term data regarding the progression of substantial non-seizure comorbidities in patients with DS. This fully enrolled study aims to evaluate neurodevelopmental and adaptive functioning, SF, gait performance, quality of life, and sleep in patients with DS. This ongoing multicenter, 24-month, prospective study includes patients aged 2-18 years with genetically confirmed DS with at least 2 convulsive seizures per 28-days despite their current antiepileptic drug (AED) regimen. Primary endpoints assess neurodevelopment along with convulsive SF, clinical status, executive function, total sleep time, ambulation, and gait. 36 patients enrolled across 3 age groups (n=12/group): 2-7, 8-12, and 13-18 years; 61% were female, 94% were white, and 14% were Latinx. Mean age of seizure onset was 0.4 years (range 0.2-1.0 years, n=36). Over the 4-week baseline period, mean convulsive SF was 14.4 (95% CI 8.1-20.7, n=26) per 28 days, including 24 patients who had generalized tonic-clonic seizures with a mean SF of 9.1 (95% CI 5.5-12.7) per 28 days. At baseline, VABS-III mean adaptive behavior composite (ABC) standard score across all groups was 45.2 (95% CI 38.2-52.1, n=33) compared to neurotypical mean score=100, SD=15. The mean baseline ABC was highest in the 2-7 years group, 62.8 (95% CI 52.3-73.3, n=12) compared to 37.9 (95% CI 31.4-44.5, n=12) and 31.3 (95% CI 19.3-43.3, n=9) in the 8-12- and 13-18-years groups, respectively. At baseline, BSID-III developmental quotient (DQ) across all groups was 21 (95% CI 12-29, n=15). The mean baseline DQ was highest in the 2-7 years group, 29 (95% CI 7-51, n=6). The patients enrolled are representative of patients with DS. Despite use of multiple AEDs, most patients experienced ongoing convulsive seizures and showed substantially decreased neurocognitive abilities with further neurodevelopmental regression compared to neurotypical children of same chronological age. The data collected in this long-term prospective study will provide valuable insights related to the course of non-seizure comorbidities in people with DS.

Stephanie Rock

UCSF

Dravet Syndrome (DS) is a severe and progressive genetic epilepsy that typically begins in the first year of life and is characterized by frequent, prolonged, and refractory seizures, intellectual disability, ataxia/motor abnormalities, behavioral problems, speech impairment, sleep disturbances, and a high risk of sudden unexpected death. Approximately 85% of DS cases are caused by spontaneous, heterozygous loss of function mutations in the SCN1A gene which encodes for voltage-gated sodium channel type 1 α subunit (Nav1.1). STK-001 is an investigational ASO treatment designed to upregulate Nav1.1 protein expression by leveraging the non-mutant (wild-type) copy of SCN1A gene to restore physiological Nav1.1 levels, thereby potentially reducing both occurrence of seizures and significant non-seizure comorbidities. MONARCH (NCT04442295) is an ongoing phase 1/2a open-label, 2-part study of patients in the U.S. with DS aged 2-18 years assessing the safety, tolerability, and pharmacokinetics (PK) of intrathecally (IT) administered single (SAD; 10, 20, and 30 mg) and multiple (MAD; 20 and 30 mg) ascending doses of STK-001. Eligible patients must have disease onset prior to 1 year old with recurrent seizures (focal motor, hemiconvulsive, or generalized tonic-clonic) and a genetically confirmed SCN1A variant. Each dose level will include at least 4 patients grouped by age (two patients 13-18 years followed by two 2-12 years) with an option to dose up to 6 additional patients per cohort for safety evaluation. Patients are observed for ≥28 days before dosing to evaluate seizure frequency. On Day 1, patients undergo CSF collection followed by single IT administration of STK 001. For MAD cohorts, injections are repeated at weeks 4 and 8. During 6-month follow-up, CSF is collected to evaluate STK-001 exposure. Adverse events (AEs) are monitored throughout, and plasma is collected for PK at multiple time points. Seizure frequency, overall clinical status, and quality of life are evaluated during the study. At study end, patients meeting criteria have the option to enter an open-label extension study and continue to receive STK-001 (SWALLOWTAIL; NCT04740476). The MONARCH study will provide insight into the safety, tolerability, and pharmacokinetic profile of single- and multiple-ascending doses of STK-001 in DS patients. In addition, the impact of STK-001 on frequency of convulsive seizures and quality of life may indicate the initial clinical effect of the STK-001. Patient enrollment in the SAD portion through 30 mg is completed with enrollment and dosing in the MAD portion of MONARCH ongoing.

Stephanie Rock

Ann & Robert H. Lurie Children's Hospital of Chicago

Objective: This post hoc analysis of a phase 3, randomised, placebo-controlled trial (GWPCARE6/NCT02544763) compared response to add-on CBD in patients with tuberous sclerosis complex (TSC) and treatment-resistant epilepsy with and without a history of infantile spasms (IS). Methods: Patients received GW Pharmaceuticals’ formulation of plant-derived highly purified CBD (100 mg/mL oral solution) titrated to 25 mg/kg/day (CBD25) or 50 mg/kg/day (CBD50) or placebo for 16 weeks. Negative binomial regression effect modification analysis assessed whether IS history affected CBD efficacy. Results: 138/224 (62%) patients had IS history. Median (range) age: 12.2 years (1.1–56.8) for patients with IS history, 10.5 years (1.6–55.8) for those without; 74% <18 years. Median (Q1, Q3) baseline monthly TSC-associated seizure frequency: 59 (28, 117) and 51 (29, 96) for patients with and without IS history. CBD reduced TSC-associated seizures vs. placebo regardless of IS history (interaction p-value: 0.803 for CBD25, 0.561 for CBD50). For patients with IS history, percent reduction in seizures from baseline: 45% for CBD25, 43% for CBD50, and 23% for placebo; placebo-adjusted reduction (95% CI): 29% (6%–45%) for CBD25 and 25% (3%–43%) for CBD50. For patients without IS history, reduction was 54% for CBD25, 55% for CBD50, and 32% for placebo; placebo-adjusted reduction (95% CI) was 32% (5%–52%) for CBD25 and 34% (7%–54%) for CBD50. AE incidence: 93% for CBD25, 100% for CBD50, and 95% for placebo; 8 patients (11%) on CBD25, 10 (14%) on CBD50, and 2 (3%) on placebo discontinued treatment because of an AE. Common AEs: diarrhoea and somnolence, occurring more frequently with CBD than placebo. ALT/AST elevations (>3×ULN): 9 (12%) patients on CBD25, 19 (26%) on CBD50, none on placebo; 79% were on concomitant valproate. Conclusions: CBD produced consistent reductions in TSC-associated seizures in patients with and without IS history. FUNDING: GW Research Ltd.

Farhad Sahebkar

Greenwich Biosciences, Inc., Carlsbad, CA, USA

Purpose: In this 2nd interim analysis of an open-label extension (OLE) trial (GWPCARE6/NCT02544763), we report safety (full follow-up) and efficacy (through 72 weeks) of add-on cannabidiol (CBD) for treatment of seizures associated with tuberous sclerosis complex (TSC). Methods: Patients who completed a randomised controlled trial (RCT) received GW Pharmaceuticals’ formulation of plant-derived highly purified CBD (100 mg/mL oral solution) in the OLE (titrated to 25 mg/kg/day, or up to 50 mg/kg/day). Primary endpoint: safety. Secondary endpoints: percent change in TSC-associated (countable focal or generalised) seizures, responder rates, and Subject/Caregiver Global Impression of Change (S/CGIC). Results: Of 201 patients who completed the RCT, 199 (99%) entered the OLE. Median (range) age: 10.7 (1.1–56.8) years. Baseline median TSC-associated seizure frequency/28 days: 57 seizures. At this analysis, 12% of patients had completed treatment, 31% had withdrawn, and 57% were ongoing. OLE median (range) treatment time: 372 (18–1127) days. Mean (SD) modal dose: 28 (9) mg/kg/day. AE incidence: 94%; serious AE incidence: 26%; 8% discontinued treatment due to AE(s). Most common AEs (≥20%): diarrhoea (45%), seizure (28%), decreased appetite (23%), pyrexia (21%), and vomiting (20%). Seventeen (9%) patients had elevated ALT/AST >3×ULN; 12 were on concomitant valproate. No patient met Hy’s law criteria for severe liver injury. One death occurred due to cardiopulmonary failure and was not deemed treatment-related by the investigator. Median percent reductions in TSC-associated seizures (12-week windows through 72 weeks): 53%–75%. Seizure reductions were 54%–80% for patients with a modal dose ≤25 mg/kg/day (n=145). ≥50%, ≥75%, and 100% responder rates were maintained up to 72 weeks, ranging from 52%–63%, 29%–51%, and 6%–19%, across 12-week windows. Improvement on S/CGIC was reported by 85% and 89% of patients/caregivers at 26 and 52 weeks. Conclusions: Add-on CBD treatment was well tolerated and produced sustained reductions in TSC-associated seizures for up to 72 weeks. FUNDING: GW Research Ltd.

Farhad Sahebkar

Greenwich Biosciences, Inc., Carlsbad, CA, USA

Purpose: Add-on cannabidiol (CBD) significantly reduced seizures associated with tuberous sclerosis complex (TSC) across the 16-week double-blind treatment period in GWPCARE6 (NCT02544763). A post hoc analysis was conducted to estimate time to onset of CBD treatment effect and resolution of adverse events (AEs). Methods: Patients received GW Pharmaceuticals’ formulation of plant-derived highly purified CBD (100 mg/mL oral solution) at 25 mg/kg/day (CBD25) or 50 mg/kg/day (CBD50), or placebo for 16 weeks. Treatment started at 5 mg/kg/day for all groups, reaching 25 mg/kg/day on Day 9 in CBD25 and 50 mg/kg/day on Day 29 in CBD50. Percent change from baseline in primary endpoint TSC-associated seizures (countable focal or generalised) was calculated by cumulative day (i.e. including previous days). Time to onset and resolution of AEs were evaluated. Results: Overall, 224 patients were randomised 1:1:1 to CBD25 (n=75), CBD50 (n=73), and placebo (n=76). The median (range) age was 11 (1–57) years. Patients had discontinued a median of 4 antiepileptic drugs (AEDs) and were currently taking a median of 3 AEDs. Differences in seizure reduction between CBD and placebo emerged on Day 6 (when titration reached 15 mg/kg/day) and became nominally significant (p<0.05) by Day 11 (CBD50) or Day 12 (CBD25). Over 90% of patients had an AE, with onset during the first 2 weeks of the titration period in 63%. AEs resolved within 4 weeks of onset in 42% of placebo and 27% of CBD patients and by end of study in 78% of placebo and 51% of CBD patients; most frequent AEs—diarrhoea, somnolence, decreased appetite—resolved in 69–88% of CBD patients. Conclusions: Findings suggest that onset of treatment effect (efficacy and AEs) occurred within the first 2 weeks. AEs lasted longer for CBD vs. placebo but resolved within the 16-week study in most patients. FUNDING: GW Research Ltd.

Farhad Sahebkar

Greenwich Biosciences, Inc., Carlsbad, CA, USA

First reported case of homozygous ZNF142 mutation manifesting as myoclonic status. A previously well 41 year old with a history of mild intellectual disability was transferred from another hospital for ventilatory support in the context of COVID pneumonitis. By day 17 of ITU admission he was improving, with limited ventilatory requirements. On day 19 he rapidly deteriorated. He began to sweat profusely and developed presumed focal seizures with gaze deviation to the left, left facial and left upper limb twitching. Blood gases showed a metabolic acidosis. Imaging including staging CT and MRI brain was unremarkable. CSF was normal/ negative for primary constituents, an extended autoimmune screen of serum and CSF including LGI1 and associated antibodies was negative. His seizures progressed to involve bilateral upper limb twitching (left >right) which was unresponsive to multiple AEDs- Levetiracetam, Lacosamide and Carbamazepine and a trail of methylprednisolone. The clinical picture was of myoclonic status; Carbamazepine was switched to Valproate with good effect. Repeated EEGs failed to identify a focus. Given the unremarkable MRI, the myoclonus was presumed to be subcortical and hypoxaemic in origin. A subsequent genetic screen revealed that this gentleman was homozygous for the ZNF142 mutation. It emerged that his cousin who is homozygous for the same mutation and has a background of global developmental delay, presented acutely with subcortical myoclonus in middle childhood which responded to Valproate. Thus far our patient has made an excellent recovery; he remains on Valproate. This rare mutation has been associated with intellectual disability associated with early onset convulsive seizures without clear precipitant (Khan 2019) though not with subcortical myoclonus. This is the first reported case presenting acutely as myoclonic status and highlights the importance of considering genetic testing in situations of status epilepticus.

Shanika Samarasekera

QEBH Birmingham

One of the primary challenges in treating epilepsy is that it is a dynamic disorder, with fluctuations in pathological brain activity and symptoms. Interestingly, seizures themselves can manifest in different ways. In particular, within the same patient, seizures can have different evolutions of electrographic features from seizure onset to termination. Understanding how a patient’s seizures change over time could lead to new, time-adaptive treatments that respond to fluctuations in seizure onset, spread, and severity. We investigated within-patient seizure variability by analysing seizures from intracranial EEG recordings of patients with focal epilepsy. We first characterised each seizure’s spatiotemporal evolution based on its time-varying network properties, thus capturing how brain regions interacted during each seizure. We then developed an approach for quantitatively comparing these seizure evolutions in each patient. We found ubiquitous within-patient seizure variability in short-term presurgical iEEG recordings of 31 patients (mean 6.7 days of recording, 16.5 seizures/patient), revealing that the evolution of seizure activity is not rigidly stereotyped within the same patient. Using the same cohort, we then explored how seizure evolutions changed over time. We found that changes in within-patient seizure evolutions were not random; instead, more similar seizures tended to occur closer together in time. Additionally, in most patients, the observed variability could be explained by circadian and/or gradual changes in seizure evolutions. Finally, using chronic iEEG recordings of 10 patients (mean 324.4 days of recording, 62.3 seizures/subject), we explored how seizure evolutions changed over multiple, patient-specific timescales (circadian, multiday, multiweek, multimonth, or slower). These timescales were uncovered by detecting periodic and/or gradual changes in interictal spike rate, yielding patient-specific cycles of different lengths as well as slower shifts in interictal dynamics. Interesting, in all patients, properties of seizure evolutions varied according to specific timescales of changes in interictal dynamics. These findings suggest that patient seizure evolutions were modulated by the same underlying fluctuating factors as interictal spike rate. Our work provides new insight into variability in within-patient seizure evolutions as well as a framework for quantitatively comparing seizures. Moreover, our temporal analyses suggest that various modulatory factors, operating over different timescales, influence seizure evolutions. Understanding the mechanisms that shape seizures within the same brain could provide new opportunities for forecasting seizure features, including seizure severity. Further, adapting treatments to intrinsic seizure fluctuations could transform severe seizures into milder forms or prevent seizures altogether.

Gabrielle Schroeder

Newcastle University

Introduction: Risk of premature death is three times higher among people with epilepsy than the general population, with suicide being the second cause after SUDEP. Routine screening for psychiatric disorders is currently recommended for all adults with epilepsy and early identification and treatment of psychiatric disorders are fundamental for any suicide prevention strategy. In 2018, the Epilepsy Centre at St George’s University Hospital developed a clinical pathway for the management of patients with suicidal ideation in the outpatient setting based on a structured risk assessment linked to their routine screening protocol. The aim of this audit was to evaluate the implementation of the suicide risk assessment in epilepsy nurse-led outpatient clinics and verify clinical outcomes. Methods: Case records of patients seen in epilepsy nurse-led outpatient clinics between October 2019 and March 2020 was conducted. Implementation of our internal clinical pathway was noted as well as clinical outcomes. Results: A total of 114 patients (44 males) were included in the audit; mean age 40 years ± 15. Around 22% patients had an already documented history of depression and 18% were on treatment for depression. No patients had a history of suicidal attempts in the last 12 months while 5% had a history of suicidal ideation documented in their notes in the last 12 months. A total of 14 patients (12.2%) presented with a positive suicidal screening, 50% males, mean age 40.8 years ± 15. One patient was classified as high risk, 2 were classified as intermediate risk and 4 patients were classified as mild risk. In 4 patients risk assessment was not documented though in one case it was documented an imminent follow up with psychiatry. Patient with positive suicide screening were predominantly males 13% versus 5.7% of females. High risk patients represented 0.9% of the total sample. Conclusions: Results of this audit support the feasibility of our clinical pathway and risk assessment though it has identified the importance of improving level of documentation.

Arrane Selvamogan

St George's University of London Medical School

Objectives: To review the concordance of High Frequency Oscillations (HFO) and time and frequency-domain responses evoked by Single Pulse Electrical Stimulation (SPES) with seizure-onset zone in SEEG recordings in children with drug-resistant focal epilepsy. Method: SPES was routinely performed on all children undergoing SEEG recording at Birmingham Children’s Hospital between 2019-2021. The responses were assessed visually and graded using currently available criteria. All delayed responses were noted. Data was also subject to time-frequency analysis to identify contacts with significant decrease in power in the gamma, ripple and fast ripple bands (Delayed High Frequency Suppression). The anatomical location at sub-lobar level of contacts with significant spectral power change compared to baseline was then compared to the seizure onset as defined by the spontaneous seizures captured during the recording. Results: A total of 43 patients underwent SEEG and SPES, three were excluded due to lack of seizures recorded. Of the remaining 40 patients, 28 (70%) had delayed responses that correlated with the spontaneous seizure onset. Twenty-five (62%) had a decrease in spectral power after SPES in any of the analysed bands in the area of spontaneous seizure onset. In 6 patients we observed fast-ripple suppression but in none of these it coincided with the seizure onset zone. Conclusions: The analysis methods for SPES have shown good concordance in paediatric sample with the seizure onset zone. Delayed High Frequency Suppression has a comparable yield to visual analysis with potential to reduce analysis times and subjectivity.

Caroline Scott

Birmingham Children's Hospital

Introduction: Alpers-Huttenlocher syndrome (AHS) is a fatal paediatric mitochondrial disease resulting from depletion of mitochondrial DNA and characterised by intractable epilepsy and severe neurodegeneration. Dysfunction and degeneration of inhibitory interneurons due to deficits in mitochondrial respiratory complex subunits is hypothesised to underlie seizure-associated alterations in cortical activity within the occipital cortex. However, it is not known whether particular subclasses of interneurons are vulnerable in AHS. Methods: Post-mortem brain tissues from the occipital cortex (BA17) were obtained from 10 patients with AHS and were compared to 8 age-matched controls and 5 sudden death in epilepsy (SUDEP) patients (epilepsy disease control). Immunohistochemistry was performed to quantify the density of 4 functionally-distinct inhibitory interneuron subtypes: parvalbumin (PV), calretinin (CR), calbindin (CB) and somatostatin (ST)-positive interneurons. Quadruple immunofluorescence permitted visualisation and quantification of expression of mitochondrial respiratory complexes I and IV subunits (NDUFB8 and COXI, respectively) within mitochondria (porin) in PV and CR interneurons. Control interneuron density and control respiratory chain data were used to derive z-scores in order to make inferences about interneuron loss and interneuron dysfunction, respectively, in AHS. Classifications were based on standard deviation limits: z>2SD = overexpression, z<-2SD = low, z<-3SD = deficiency, z<-4SD = severe deficiency. Results: There was a severe loss of PV interneurons within BA17 in 9 of 10 patients (z<-4), with a consistent relative preservation of CR interneurons. However, the density of CB and ST interneurons was highly variable. To investigate whether mitochondrial dysfunction underlies the increased vulnerability of PV interneurons relative to CR interneurons in AHS, the expression of NDUFB8 and COXI were measured within PV and CR interneurons. NDUFB8 and COXI were severely deficient (z<-4) within 88% and 96% of patient PV interneurons, respectively. This was accompanied by a significant increase in mitochondrial mass within PV interneurons (z>2). Whereas patient CR interneurons displayed significantly milder NDUFB8 and COXI deficiencies. SUDEP patients showed preserved interneuron densities and normal NDUFB8 and COXI protein levels relative to controls. Conclusions: Severe respiratory chain impairment may underlie the selective degeneration of PV interneurons in AHS. Together, the loss of PV interneurons and the dysfunction of surviving PV interneurons within the occipital cortex may underlie the loss of inhibition leading to seizures in patients with AHS. However, lower energy-demanding CR interneurons show less severe respiratory chain deficiencies suggesting CR interneurons are more resilient to mitochondrial dysfunction in AHS.

Laura Smith

Wellcome Centre for Mitochondrial Research, Newcastle University

Background: The epilepsy treatment gap in sub-Saharan Africa (SSA) is disproportionately high. Epilepsy surgery in eligible patients can result in significant improvement in seizure control, and is best undertaken in epilepsy centres. Outside South Africa, there are no such specialist centres in SSA, therefore eligible patients with medically refractory epilepsy cannot access this important treatment. Methods: We describe two cases from our regional tertiary neurology referral centre with medically refractory epilepsy who had to seek surgical treatment overseas, and both had good outcomes. Case 1: A 34-year-old female with well-controlled type 1 diabetes began experiencing dyscognitive déjà vu, jamais vu and olfactory auras from 2008, with secondary generalized tonic-clonic seizures (GTCS) occurring from 2012, during which she had leftward head version. Several MRI brain scans and inter-ictal sleep-deprived EEGs over the years had been reported as normal. She transferred to our care in 2016. She experienced poor seizure control or intrusive side effects on six other anti-epileptic medications. Her seizures only came under maximal control in 2020 whilst on lacosamide 300mg BD and clobazam 10mg BD, but was still experiencing GTCS once a week. She was evaluated at an epilepsy centre in South Africa, where after right inferior temporal lobectomy she became completely seizure-free. Case 2: A 36-year-old female diagnosed with epilepsy in 1998 came seeking a second opinion under our care in 2017. She was having dyscognitive temporal lobe seizures with oro-alimentary automatisms several times a week, and was still experiencing secondary GTCS at least once a week whilst on levetiracetam 2g/day and clonazepam 1.0mg nocte. After trying five different anti-epileptic medications in the ensuing 18 months, we attained seizure control (with only occasional auras) on a combination of levetiracetam 1750mg BD, lamotrigine 275mg BD, and clobazam 30mg BD. Up-to-date inter-ictal sleep-deprived EEG revealed right temporal lobe spikes and MRI demonstrated right mesial temporal sclerosis, which she explained had never been diagnosed before. It took another year facilitating her and her husband to select an affordable epilepsy surgery centre and to collect enough funds. She had the surgery in India in early 2020 and since then has been completely seizure-free. This year she has remained only on lamotrigine 500mg/day and is continuing to slowly wean off. Conclusion: Our cases illustrate that, in our setting, patients await decades for epilepsy surgery, calling for regional efforts to set up more regional specialised centres in SSA to narrow this treatment gap.

Dilraj Sokhi

Aga Khan University Medical College of East Africa, Nairobi, Kenya

Background: Vagal Nerve Stimulation (VNS) devices have been used for refractory epilepsy since 1997. There is a good body of evidence demonstrating their efficacy in this patient group. Since 2015, novel Cardio-Responsive devices were introduced to practice, promising additional benefit to that of the original VNS devices. There is limited literature reviewing their efficacy. This study aims to assess if patients with refractory epilepsy receive additional benefit from novel cardio-responsive VNS devices compared with traditional VNS devices. Methods: Data was collected retrospectively from patients with epilepsy who had an existing VNS device, deemed to have been beneficial, which was then replaced with a new Cardio-responsive VNS. The two models being evaluated in this study are the SenTiva™ M1000 and AspireSR® 106. These battery changes occurred over a 4 year period from September 2016 to January 2021, carried out by a single surgeon. The seizure burden was compared between the periods before and after the battery change from traditional VNS to novel cardio-responsive VNS device. Results: 65 patients were included in this review. There was a significant decrease in seizure burden from a median monthly seizure frequency of 16.3 (IQR 4.5-39.0) to 9.0 (IQR 3.0-30.3), p =0.02. When separated by seizure type, there was a non-significant decrease in median monthly tonic clonic seizures from 1.0 (0.0-5.0) to 0.0 (0.0-4.3), p=0.06. Prior to battery change 42% had no tonic-clonic seizures, an additional 18% achieved freedom from tonic clonic seizures following battery change. 45% of the total cohort had ≥ 50% reduction in overall seizure frequency, contributing to a total of 57% showing a further reduction in seizure frequency following battery change (11% showed no change, 32% showed an increase in seizure frequency). Conclusion: The results suggest that novel Cardio-Responsive VNS devices could further reduce seizure frequency for the majority of those already receiving successful treatment with traditional VNS devices.

Helen Stephens

Salford Royal Foundation Trust

Objective: The careful selection of surgical candidates is a prerequisite for successful epilepsy surgery with intracranial monitoring often playing a key part, especially in establishing surgical candidacy in MRI-negative patients and patients with extra-temporal epilepsies. The outcome of quaternary epilepsy surgery referrals in patients following phase I presurgical evaluation at the Manchester Centre for Clinical Neurosciences in the pre-SEEG era was reviewed. Methods: The trajectories of patients referred to the supra-regional epilepsy surgery programmes between 2007 and 2019 were studied retrospectively. Patient medical records detailing the presurgical evaluation at the regional and supra-regional epilepsy surgery programmes were reviewed, focusing on the type of surgical intervention, waiting times and seizure freedom rates following resective epilepsy surgery. Results: Thirty-eight patients were referred, of which 18 had potentially epileptogenic lesions seen on sMRI and 20 patients were deemed MRI-negative with 7/20 patients diagnosed with extra-temporal epilepsies based on a non-invasive presurgical evaluation at the regional epilepsy programme. More than half of the referred patients (60.5%) reported convulsions as one of their habitual seizure types or experienced convulsions during their video-EEG-telemetry (VT) admission at the regional epilepsy surgery programme. VT was confirmed to be repeated at the quaternary epilepsy programmes in 23 patients. 52.2% of these patients experienced generalised convulsions during the repeat VT. Only 22 of the referred 38 patients were able to persevere with the presurgical workup away from home. Nine patients underwent resective surgery, preceded by SEEG monitoring in >50% patients. The median referral-to-resective surgery time was 2.5 years. An Engel I outcome was achieved in 55.5% of patients with a median postoperative follow up period of 2.8 years. The patients who were unable to continue with the presurgical evaluation at the quaternary epilepsy programmes quoted financial difficulties, reliance on friends and family support closer to home and long waiting times as the main reasons. Conclusions: Epilepsy surgery can transform patients’ lives and yet there is a clear gap in access to epilepsy surgery care for patients with pharmaco-resistant epilepsy. Consistent commissioning of SEEG at the regional level could help build regional epilepsy surgery service capabilities and bridge the care gaps with improved access to epilepsy surgery care closer to home, in line with the ambitions of the Integrated Care Systems transformation plans aiming at reducing health inequalities and meeting patient care needs within the regions.

Avraneel Talapatra

Manchester Centre for Clinical Neurosciences, Manchester, UK

Introduction: Lennox-Gastaut syndrome (LGS) is a treatment-resistant developmental and epileptic encephalopathy with multiple seizure types. The aim of this phase 3 clinical trial was to evaluate the safety and efficacy of fenfluramine (FFA) in patients with LGS (NCT03355209). Methods: Patients aged 2-35 years with confirmed LGS were randomized to placebo or adjunctive FFA (0.2 or 0.7mg/kg/day) following a 4-week baseline to establish monthly frequency of seizures associated with a drop (defined as frequency of generalized tonic-clonic (GTC), secondary GTC, tonic, atonic, tonic-atonic; hereafter, “drop seizure frequency”). After 2 weeks’ titration, patients were treated for 12 weeks. The primary efficacy endpoint was percentage change from baseline in drop seizure frequency in the 0.7mg/kg/day FFA group (rank ANCOVA). Secondary endpoints included percentage change from baseline in drop seizure frequency in the 0.2mg/kg/day FFA group and ≥50% responder rate and clinical global impression of improvement (CGI-I) for both doses. Results: A total of 263 patients were randomized (placebo, n=87; FFA 0.2mg/kg/day, n=89; FFA 0.7mg/kg/day, n=87). Median age was 13 years with median baseline drop seizure frequency of 77 (range 2-2943). Median number of concomitant antiepileptic drugs (AEDs) was 3 (range 1-5). Median prior AED failures was 7 (range 1-20). The study met its primary efficacy endpoint: patients taking FFA 0.7mg/kg/day achieved a -19.9% estimated median difference from placebo (EMD; Hodges-Lehmann estimate) in drop seizure frequency from baseline (P=0.001). In the FFA 0.2mg/kg/day group, EMD did not reach statistical significance (10.5%; P=0.09). More patients in the FFA groups achieved ≥50% response (0.7mg/kg/day, 25.3%, P=0.015; 0.2mg/kg/day, 28.1%, P=0.005; versus 10.3% in placebo). Compared with the placebo group, more patients in the FFA groups were rated as “Much Improved” or “Very Much Improved” on the CGI-I by both investigators (0.7mg/kg/day, 26.3%, P=0.007; 0.2mg/kg/day, 20%, P=0.010; 6.3%, placebo) and caregivers (0.7mg/kg/day, 33.8%, P<0.0001; 0.2mg/kg/day, 27.1%, P<0.0001; 4.9%, placebo). FFA was particularly effective in reducing the frequency of GTCs; of 114 patients (43%) with baseline GTCs, frequency was reduced by a median of 46% and 58% in the 0.7 and 0.2mg/kg/day FFA groups, compared to worsening of 3.7% with placebo. Most common adverse events included decreased appetite, somnolence, fatigue, vomiting, diarrhea, irritability, seizure, and pyrexia. No cases of valvular heart disease (VHD) or pulmonary arterial hypertension (PAH) were observed. Conclusions: The study met its primary endpoint, with FFA 0.7mg/kg/day demonstrating significant improvement in drop seizure frequency comparable to other recently completed LGS RCTs. FFA was also highly effective in reducing GTCs in both 0.7 and 0.2mg/kg/day groups. FFA was generally well tolerated with no VHD or PAH observed.

Zhen Tan

Children's Hospital Colorado, Aurora, CO, USA

Introduction: Fenfluramine (FFA) has demonstrated ability to provide durable and profound reductions in monthly convulsive seizure frequency (MCSF) in patients with Dravet syndrome treated for up to 2 years. Here we report an interim analysis of the safety and efficacy of FFA in an ongoing open-label extension (OLE) study, with treatment duration of up to 3 years. Methods: Patients who completed the phase 3 controlled studies were eligible to enroll in the OLE study (NCT02823145). All patients initiated FFA at 0.2 mg/kg/day regardless of the dose received in the previous core studies. At the end of 4 weeks, doses were titrated to effect and tolerability, up to 0.7 mg/kg/day (maximum, 26 mg/day) or 0.4 mg/kg/day (maximum, 17 mg/day, in patients also receiving stiripentol). Seizure frequency was captured via hand-held e-diary. Efficacy was assessed as reduction in MCSF from core study baseline. Efficacy and safety were assessed at each office visit (monthly for the first 3 months, then every 3 months). Results: As of October 14, 2019, 330 patients were enrolled and had received ≥1 dose of FFA. At the time of entry into the study, mean±SD age of patients was 9.0±4.6 years, and 180/330 (54.5%) were male. Median duration of treatment was 631 days (range, 7-1086), with a median reduction of 83% in MCSF (P=0.002) at 36 months. Reasons for discontinuation included lack of efficacy (14.5%) and occurrence of adverse events (3.3%). The most common concomitant antiepileptic drugs (AEDs) were valproate (80.6%), clobazam (72.7%), stiripentol (29.4%), topiramate (25.5%), and levetiracetam (25.5%). Median percent MCSF reduction over the entire treatment period was -64.5% (P<0.001), with 63.4% of patients experiencing a clinically meaningful (≥50%) MCSF reduction, and 38.2% a profound (≥75%) MCSF reduction; 3 patients were seizure-free during their entire time in the OLE. The most common (≥15%) adverse events included pyrexia (28.2%), nasopharyngitis (27.3%), decreased appetite (23.0%), decreased blood glucose (19.4%), diarrhea (18.2%), seizure (16.7%), echocardiogram abnormality (19.4%) limited to nonpathological trace/physiologic regurgitation, and upper respiratory tract infection (15.5%). No cases of valvular heart disease (VHD) nor pulmonary artery hypertension (PAH) were observed during the entire OLE treatment period. The majority of subjects (54.2%) gained ≥7% body weight while the proportion of subjects who had loss of body weight ≥7% remained below 10%. Conclusions: FFA continued to provide durable and profound levels of reduction in MCSF in patients treated up to 3 years, which may have implications for long-term neurodevelopmental outcomes. FFA treatment did not further impair growth and development, and no patient developed VAD or PAH.

Zhen Tan

Zogenix International Limited

Introduction: In a recently completed phase 3 program, patients with Dravet syndrome (DS) treated with fenfluramine demonstrated sustained, profound reductions in seizure frequency, prolonged periods of seizure freedom, and improvement in executive functions. The European Early Access Program (EU-EAP) was initiated to allow pre-approval access to fenfluramine for DS patients. Methods: Patients with a confirmed diagnosis of DS for whom seizures were not adequately controlled by their current anti-epilepsy treatment regimen and who had no alternative options (e.g., other treatment, access to a clinical trial) were eligible to enroll in the EU-EAP. Fenfluramine dosing typically was started at ≤0.2 mg/kg/day and was titrated based on efficacy and tolerability. Maximum dose was 0.7 mg/kg/day (absolute maximum, 26 mg/day), or for patients receiving concomitant stiripentol, 0.4 mg/kg/day (absolute maximum, 17 mg/day). Results: A total of 150 patients with DS from the EU-EAP at multiple centers in Germany, Italy, Spain, and UK were included in this pooled analysis. The median age at the start of fenfluramine treatment was 7.2 years (range, 0.8-46 years), 49% were female, 96% had an SCN1A variant, and median exposure was 338 days. After 3 months (n=139), reductions in seizure frequency ≥50%, ≥75%, or 100% were observed in 79%, 56%, and 27%, respectively, and after 12 months (n=80), these seizure reduction thresholds were demonstrated by 80%, 51%, and 16%. Sixty-two percent were rated by the investigator as “much” or “very much” improved. Thirteen percent of patients have discontinued fenfluramine, primarily due to lack of efficacy. The most common adverse events were somnolence or sleep disorder (n=31) and loss of appetite (n=30). No valvular heart disease or pulmonary artery hypertension was observed. Conclusions: DS patients treated with fenfluramine in a “real-world” setting experienced similar benefits and tolerability as those observed in phase 3 clinical trials.

Zhen Tan

Zogenix International Limited

Introduction: Sudden unexpected death in epilepsy (SUDEP) is defined as sudden death in a person with epilepsy that is not explained by status epilepticus, trauma, or another known cause. SUDEP is a major concern of parents of Dravet syndrome persons as the incidence is about 6-fold higher than persons with other forms of epilepsy. Fenfluramine has been shown to significantly reduce the frequency of convulsive seizures in Dravet syndrome. The objective of this study was to compare the incidence of SUDEP Dravet syndrome patients on fenfluramine with literature reports of SUDEP incidence in Dravet syndrome patients receiving standard of care. Methods: Relevant studies were identified by searching PubMed using “Dravet [title] AND (mortality OR death OR SUDEP)”. The fenfluramine-treated study population comprised patients from 3 sources: international phase 3 clinical trials, United States and European Early Access Programs (EAPs), and a long-term, open-label study spanning 32 years. Patients in these studies were treated with fenfluramine up to 0.7 mg/kg/day (maximum dose 26 mg/day) or 0.4 mg/kg/day (maximum dose 17 mg/day) if receiving concomitant stiripentol. The incidence of SUDEP was assessed using current guidelines and was expressed as deaths per 1000 person-years of observation. Results: Nine studies describing the incidence of SUDEP in Dravet syndrome were identified. The report by Cooper et al. (Cooper, M.S. et al. Epilepsy Res. 2016; 128: 43-47) was considered the most rigorous. Cooper studied 100 consecutive patients with Dravet syndrome and reported an all-cause mortality rate of 15.84 per 1000 person-years (98% CI, 9.01 to 27.85) and a SUDEP rate of 9.32 per 1000 person-years (98% CI, 4.46 to 19.45). A total of 732 patients treated with fenfluramine provided 1185.3 person-years of observation. In addition, 366 of these patients provided 85.8 person-years during baseline observation and/or placebo treatment in the phase 3 clinical trials. The median age of fenfluramine patients was 8 years; 89.9% were ≤18 years old at the start of treatment. Three deaths occurred—all SUDEP: one during the baseline/placebo period (11.7 per 1000 person-years) and 2 during treatment with fenfluramine (1.7 per 1000 person-years). The fenfluramine SUDEP rate was below the lower limit of the 98% CI reported by Cooper et al., whereas the SUDEP rate observed before initiation of fenfluramine was similar to the historical rate reported by Cooper et al. Conclusions: This post-hoc analysis suggest that Dravet syndrome patients treated with fenfluramine is associated with substantially reduced risk of SUDEP and all-cause mortality compared with historical standard of care. Further research is warranted to establish the mechanism(s) responsible for this response.

Zhen Tan

UCL Great Ormond Street NIHR BRC Institute of Child Health, London, UK

Background: Young women with epilepsy (YWWE) face unique challenges from the recent changes to the MRHA guidelines restricting the prescription of sodium valproate. We have observed an increase in the numbers and complexity of YWWE attending the teenage epilepsy clinic at the University Hospital of Wales (UHW) and proposed that this was due to inadequate seizure control by alternative antiepileptic drugs (AEDs). Method: A retrospective review of clinic letters and medical records of all patients seen over a three-year period (2018-2021) in teenage epilepsy clinic was performed. Seven variables were measured to evaluate medical history, seizure control and medication use, comparing boys versus girls. Results: Between July 2018 and July 2021 110 patients were seen (38M:72F). 88/110 had an established diagnosis of epilepsy. 64/88 patients were considered to be appropriately treated and discharged to primary care or to the adult epilepsy service;16 (15M:1F) of whom were treated with sodium valproate. Major changes were made to the treatment plans of 24/88 patients. 15 patients were females on valproate which was weaned off as per the MRHA guidelines. 6/15 of these patients presented with relapses in seizure control following withdrawal of sodium valproate. 2 of these patients were re-started on sodium valproate according under the pregnancy prevention programme and have since experienced improvements in seizure frequency. Of the 110 patients attending the clinic, 7 patients had trialled 6 or more AEDs, these patients were all female. Interpretation: There are a higher number of females attending clinic than males which we propose is due to the restriction of sodium valproate prescription. There are a significant number of females trialling 6 or more AEDs, because of inadequate seizure control on withdrawal/withholding of sodium valproate.

Rishana Thayaparan

Cardiff University

Introduction: There is increasing evidence to suggest changes in the gut brain axis, through gut microbiota, may be associated with the development of neurological disease including epilepsy. A pre-clinical study showed that the ketogenic diet alters the gut microbiota across seizure mice models and that these microbiota changes were associated with seizure protection. Additionally, transplant microbiome studies in seizure rat models and also in human subjects have demonstrated that seizure susceptibility can be changed through modification of the microbiota. Studies have shown that bariatric surgery can change the gut microbiota by increasing diversity and gene richness. There has been some overlap in species changes observed after bariatric surgery and in ketogenic diet. Objectives: To determine how many patients develop epilepsy after bariatric surgery. Additionally, seizure frequency and anti-epileptic drug (AED) prescriptions were compared pre- and post-operatively in patients with a pre-operative epilepsy diagnosis. Methods: Retrospective cohort study with a population of 606 consecutive patients who underwent bariatric surgery at a tertiary UK centre over a 11 year period. Data was collected from: pre-operative clinic letters, discharge summaries, GP records, surgical clinic letters, neurology clinic letters, serum anti-epileptic drug concentrations, MRI and EEG reports. Results: 14 patients with a diagnosis of epilepsy have been identified from this population, giving a prevalence of 2.3% (14/606) with an average pre-operative weight of 130.8kg and an average weight loss of 40.5kg one year post-operatively. 85.7% (12/14) had a diagnosis of epilepsy pre-operatively, with two patients developing epilepsy in the years after their operation. 25% (3/12) of patients with pre-existing epilepsy diagnosis were not taking anti-epileptic drugs before their operation and had to re-start them post-operatively. For patients with a pre-operative epilepsy diagnosis, post-operative seizure frequency: increased in 41.7% (5/12) of patients, was the same in 8.3% (1/12) of patients and reduced in 8.3% (1/12) of patients (not documented for 5/12 patients). Where available, all serum levels for AEDs were normal post-operatively. Conclusion: The results of this study suggest that there is a link between bariatric surgery and increased seizure frequency and increased use of AEDs. Whilst the retrospective design of this study cannot identify the mechanism by which this takes place, based on the literature, this could be through gut microbiota changes. This study highlights the need for larger scale retrospective analysis to verify these results and for prospective research analysing gut microbiota changes in this patient group.

Joe Thompson

Leeds Teaching Hospitals Trust

Purpose: To analyse and recognise any pre-operative factors that contribute to sustained seizure freedom in patients who underwent temporal lobectomy with amygdalohippocampectomy (TLAH) for mesial temporal lobe epilepsy (TLE). Furthermore, seizure recurrence rates were analysed and any complications resulting from surgery being identified. Methods: Data was collected and analysed for 47 contiguous patients who underwent TLAH at yearly intervals, during a 5-year follow-up period. Outcomes were determined using Engel classification, with the population being divided into favourable outcome (Engel I and II) and non-favourable outcome (Engel III and IV) at 5 years. Chi-squared and an unpaired student t-test were used to identify any preoperative factors that contributed to poorer prognosis. A Kaplan-Meier survival analysis and Mantel-cox test were used to calculate chance of seizure recurrence. Results: During the first year of follow up 83% of patients were deemed to have a favourable outcome (Engel I-II) and 77% of patients were seizure free. By the 5th year of follow up 74.5% had a favourable outcome, with 66% of patients being seizure free. Acute post-operative seizures (APOS) (p=0.003) and a shorter mean number of months before first disabling seizure (0.003) were both associated with unfavourable outcome. Survival analysis identified that patients had a 55% chance of being completely free of all disabling seizures by the end of the follow up period. The most common complications arising from surgery were mood disturbances, memory impairments and post-operative headaches. Conclusion: Our findings support the use of TLAH as an effective treatment method for patients with mesial TLE, providing 74.5% of patients in this study with a favourable outcome by year 5 post-surgery. Additionally, this study shows that APOS and shorter mean number of months before first disabling seizure both had significant associations with poorer prognosis. These findings can have great implications to clinicians and can improve the pre-operative consenting and post-operative counselling process.

Eduard Tjong A Tjoe

University of Manchester

Purpose: DS is a rare, genetic, life-threatening epileptic encephalopathy. Patients experience life-long, refractory, frequent, prolonged, often convulsive seizures (CS), that frequently require emergency care. This study examined care, management, and healthcare resource use (HCRU) of DS patients in England to explore drivers of HCRU and how CS frequency impacts quality of life (QoL). Method: Care of patients with DS was discussed with a purposive sample of 16 clinicians (12 sites, 7 regions), using semi-structured interviews and structured quantitative tasks, followed by quantitative validation. Result: Data demonstrated regional variation in care and service provision. Paediatric services were more likely to follow a ‘joined-up’ approach in contrast with adult services. Transitioning arrangements between services were felt particularly lacking. Clinicians reported CS frequency as the primary driver of ongoing and emergency HCRU. Younger patients (ages 2-3) with clinician-reported high seizure frequency (HSF: average ≥7 seizures/month) had more (5.0 Vs 3.3.) annual face-to-face secondary care consultations compared to adults (age 18+) with HSF (average ≥13 seizures/month). Patients with lower CS frequency (average 1 seizure/month) accessed ongoing care less often (average annual consultations: children = 2.7 Vs. adults = 1.5). Children had more ambulance attendances after rescue medications (82.3%), compared with adults (24.5%). Seizure freedom is rarely achievable. Clinicians reported interventions providing longer CS-free durations, reductions in ‘absolute’ and ‘relative’ seizure frequency were clinically important for patient/carer QoL and gaining seizure control. Six of eight clinicians rated a 50% reduction in CS as ‘meaningful’. Conclusion: DS patients and their carers have high care requirements. There is need to reduce regional variation in care provision across England, and improve transitioning arrangements into adult services. Meaningful reductions in CS frequency in DS patients could reduce HCRU and improve QoL, which may in turn alleviate current variations in service provision and care.

Toby Toward

Zogenix International Limited

Purpose: Fenfluramine is a recently licensed add-on therapy to standard of care antiepileptic drugs to treat the frequent, often daily severe seizures of DS. It may be considered an alternative to cannabidiol; however, there are no head-to-head comparative trials of these therapies. We assessed the comparative effectiveness of licensed fenfluramine dose regimens (with and without concomitant stiripentol) and licensed cannabidiol dose regimens (with concomitant clobazam per its European license, and irrespective of concomitant clobazam per its US license), using robust ITC methods. Method: We systematically searched for randomised controlled trials (RCTs) of licensed add-on therapies for DS published to 28 June 2020. Outcomes of interest were relative, placebo-adjusted reductions from baseline in monthly convulsive seizure frequency (MCSF), and proportion of patients achieving >50% (clinically meaningful) and >75% reductions in MCSF. Comparative efficacy endpoints were assessed from available published data using Bayesian network meta-analysis (NMA). Adverse events were considered descriptively. Result: For each intervention we identified 2 published randomised placebo-controlled trials (RCTs). The median baseline convulsive seizure frequencies across the 4 trials ranged from 9 to 24 per 28 days. Trial populations were considered comparable to facilitate a Bayesian NMA. In a population receiving cannabidiol, irrespective of the use of clobazam (US license), at 10mg/kg/day and 20mg/kg/day (maximum) doses of cannabidiol the mean placebo-adjusted reductions from baseline MCSF were 29.2% (8.7, 44.8) and 24.1% (6.8, 38.2), and with fenfluramine 0.7 mg/kg/day (without concomitant stiripentol) and 0.4mg/kg/day (with concomitant stiripentol) were 62.3% (95%CrI 47.6, 72.7) and 54.0% (35.4, 67.0), respectively. When cannabidiol was confined to use with concomitant clobazam (European license), the respective results for cannabidiol 10 and 20 mg/kg/day were 39.8% (18.6, 55.5) and 36.5% (18.7, 50.3) and for fenfluramine 0.7 and 0.4 mg/kg/day were 62.3% (95%CrI 47.8, 72.7) and 54.1% (35.7, 67.4). For these outcomes, and for the proportion of patients achieving >50% and >75% reductions in MCSF, Bayesian treatment ranking indicated >98% probability that fenfluramine is the most effective therapy versus <2% probability for cannabidiol 10 or 20mg/kg/day, both with and irrespective of concomitant clobazam. Fenfluramine had lower rates of somnolence and no increase in weight loss, valvular heart disease or pulmonary hypertension versus cannabidiol. Conclusions: ITC using Bayesian NMA of RCT data indicate fenfluramine provides more effective convulsive seizure control than cannabidiol across all licensed dose regimens. Fenfluramine is comparatively well-tolerated and provides a much-needed step change in DS seizure management.

Toby Toward

Zogenix International, Maidenhead, UK

Introduction: Vagal nerve stimulation (VNS) has become one of the established neuro-modulation based approaches for the treatment of medically intractable epilepsy (Faraj, Alok, Hasbini, & Najjar, 2021). However, despite all clinical efforts and advances in technology as well as hardware, infection of these implants remains a complication. Current therapeutic approaches to such infections range from the classical urgent removal of the VNS implants, to the more contemporary conservative treatment with antibiotics to preserve the implants. Method: We conducted a retrospective analysis of all paediatric patients with a VNS-surgical site infection following implantation at our regional centre. Results: Since 2015, a total of 65 VNS devices were implanted. 5 (7.7%) had reported VNS-surgical site infection documented in their clinical records. In all these cases, antibiotic (patient 1: Flucloxacillin IV, followed by ceftriaxone OD; patient 2 and 3: oral co-amoxiclav, patient 4: chloramphenicol 1% cream; patient 5: oral antibiotics prescribed by the GP) treatment was initiated prior to isolating the causative organism to avoid delay in treatment. However, assumptions about the mechanism by which the infection was acquired, namely itching and scratching of the wound, appeared to be a recurring theme in this particular paediatric population. All patients identified were successfully treated with antibiotics and the implants preserved. Only one patient required a wound washout in addition to antibiotics. The other four patients were all successfully treated with a course of antibiotics only. Discussion: Explantation of VNS devices attracts significant risks for the patient including, but not limited to, vagal nerve damage, dysphagia, and other standard risks associated with dissection of the neck. Additionally, removal of these device implies temporary withdrawal of surgical treatment for epilepsy in these patients, putting them at a greater risk of seizures and their sequelae, including death. Due to gliosis and / or anatomical constraints, in some cases it is not possible to re-implant the VNS device on the left side, and implantation on the right side attracts the further risk of cardiac complications due to the predominant innervation of the sinoatrial node from the right vagal nerve. Therefore, given the success of antibiotic therapy in our cohort, explantation of the VNS devices may not be the best initial option for the treatment of VNS-surgical site infections. Notably, more superficial wound infections should be treated with a course of antibiotics before more radical surgery is considered. Conclusion: Trial of more contemporary conservative treatment with antibiotics may represent a better initial therapeutic option for VNS-surgical site infections.

Stasa Tumpa

University of Cambridge

Patient and public involvement (PPI) has become an expectation and imperative in health research. Effective and meaningful involvement in health research through identifying priorities, development of proposals and the execution of research projects, is seen to bring about structural benefits to the research in terms of ensuring relevance of work and improvements to the efficacy of its delivery. Involvement is now a key requirement of major grant funders. Participation is also seen as increasing the democracy of the process in terms of giving a ‘voice’ to those involved. For research funding organisations, PPI can also aid transparency and accountability of funding decisions in ensuring that organisations are funding research that matters most to those affected. The value of PPI is not only realised by researchers and research funding organisations, but by PPI participants themselves. Benefits have been evidenced on the wider lives of those involved in terms of the effects on the relationships with those around them, employment, finances, health status, well-being and societal positions in terms of exclusion and inequality. The Shape Epilepsy Research Network was launched in October 2020 as part of Epilepsy Research UK’s #ALifeInterrupted campaign. Since then we have built a community of nearly 300 motivated and committed people affected by epilepsy, each with an interest in being involved in research into epilepsy. The vision of the network is “to bring people together to improve all research into epilepsy”, with their mission to improve all research into epilepsy through meaningful involvement, to make research accessible, relatable and relevant, and to increase advocacy for the importance of research into epilepsy. The strategic priorities of the Shape Network are to build a community of people affected by epilepsy, connect researchers with Shape Network members, involve people affected by epilepsy in research, advocate for increased awareness of epilepsy and research into epilepsy, and finally, collaborate with organisations and groups to ensure that a range of epilepsy experiences are reflected in the network. Through the Shape Network, we will provide a national patient and public involvement (PPI) service for researchers, facilitating and managing the involvement of people living with epilepsy in research.

Caoimhe Twohig-Bennett

Epilepsy Research UK

Introduction: Epilepsy syndromes differ in their demography, clinical manifestations and investigation findings. Frontal lobe epilepsy (FLE), a focal epilepsy syndrome with a heterogeneous clinical presentation, is understudied and often misdiagnosed. We sought to comprehensively phenotype FLE and to differentiate FLE from other epilepsy syndromes. Methods: Between 2009 and 2018, we undertook a prospective review of 1481 prospective cases of epilepsy, diagnosed based on International League Against Epilepsy (ILAE) criteria, from a tertiary neurology centre in London. Results: 166 patients had FLE based on clinical findings and investigations. Of these, 97 patients had identifiable electroencephalography (EEG) foci in frontal areas (definite FLE) while 69 patients with FLE had no frontal EEG foci (probable FLE). Apart from the presence of frontal EEG foci, probable and definite FLE did not differ in any demographic, clinical and investigation feature. Systematic comparisons of FLE and other epilepsy syndromes established demographic, clinical and investigation features that differentiated FLE from the other syndromes. These features included an underlying structural or metabolic aetiology, abnormalities on MRI, dyscognitive seizure type and the absence of intellectual disability. Conclusions: FLE is a varied clinical syndrome characterised by dyscognitive seizures and a predominantly structural or metabolic aetiology usually identified with MRI. There was no difference in the clinical features of definite and probable FLE, suggesting that probable and definite FLE likely represent the same clinical entity. This large medical case series provides further insights into the hallmark features of FLE.

Ryan Wee

UCL Medical School

Lennox-Gastaut Syndrome (LGS) and Dravet Syndrome (DS) are debilitating manifestations of epileptic encephalopathy presenting in early life. LGS is characterised by multifocal seizures including tonic seizures and atonic ‘drops’ with associated risk of injury. Patients presenting to the adult service generally have associated physical and intellectual disability. The efficacy of Cannabidiol (CBD) for the adjunctive treatment of seizures associated with LGS and DS was evaluated in 4 randomised controlled clinical trials comparing the additive effect of Cannabidiol versus placebo. In November 2019, the National Institute for Health and Care Excellence (NICE) guidance recommended Cannabidiol with Clobazam for refractory epilepsy in these patients (NG144). We audited patients at a tertiary adult neuroscience centre who were prescribed Cannabidiol for the treatment of LGS and DS. Minimum 3-month follow-up data was obtained for 24 patients with a median age of 22 (range 17-50 years). The majority had LGS (87%). All patients had intellectual disability. The median number of existing anti-epileptic medications was 3 (range: 2 to 5 medications). A third of patients had a Vagal Nerve Stimulator (VNS) in-situ, a further 16% were waiting for the procedure. Post CBD follow-up ranged from 3-15 months. Half of all patients had >50% reduction in convulsive seizures, a fifth experienced a reduction of < 50% in convulsive seizure frequency, and convulsive seizure frequency was unchanged in the remainder. In patients with > 50% reduction in convulsive seizures, 75% also had >50% reduction in at least one other seizure type. For patients with a <50% reduction in convulsive seizure frequency, 46% had a >50% reduction in seizure frequency of at least one other seizure type. Atonic and tonic seizures responded most favourably; seizure reduction with Cannabidiol started at doses of 2.5-5mg/kg/day. Additive improvement was seen with dose increases up to 7.5mg/day. Families noticed increased alertness and improved behaviour in 41.6% of patients. 8.3% of carers observed increased tiredness or sedation which improved in half of these patients over time. 4.2% had diarrhoea which settled and 4.2% was found to have deranged liver function tests. In summary, the combination of Cannabidiol and Clobazam was well tolerated in our patient population with LGS and DS. There appeared to be a sustained seizure reduction at modest doses with secondary benefits in alertness and behaviour.

Daniel White

University Hospitals Birmingham NHS Foundation Trust

We collected the long-term outcome (dead or alive) of 282 patients who had been referred for suspected epileptic seizures 14 to 20 years after the original referral. For deceased patients we collected date and cause of death. We used Kaplan Meier statistics to estimate the risk of death. 166 (59%) patients had epileptic seizures (ES), 59 (21%) had syncope, 26 (9%) had non epileptic attacks and 31 (11%) other conditions (nonES). The mean age in ES at referral was 38.18 years (SD 15.95 years), the mean age in nonES was 36.08 years (SD 16.08 years). Forty one deaths occurred, 33 in ES, 8 in nonES (3 in patients with syncope, 1 in non epileptic attacks, 4 in other). The mean age at death in ES was 55.05 years (SD 16.31), and in nonES it was 63.10 years (SD 15.21) NS. 10% of ES patients had died at 5 years (95% confidence intervals (CI) 5-15), 19% at 15 years (95%CI 14-24), in nonES patients the numbers were 2% (95%CI 0-4) and 7% (95%CI 2-12) respectively (p = 0.0023, Hazard ratio 3.1192, 95% CI 1.6810 to 5.7878). Causes of death in ES were malignancy (n=7 patients, of whom 3 were brain tumours), alcohol related death (n=6), sudden unexpected death in epilepsy patients (SUDEP) (n=5), cardiac disease (n=4), infections (n=3), trauma (n=2), dementia (n=1), unknown (n=5). In non ES the causes of death were dementia (n=2), cardiac disease (n=1), alcohol (n=1), malignancy (n=1), unknown (n=3). Our data showed a significantly higher mortality in patients who had epileptic seizures than in patients who had did not have epileptic seizures. About one in five patients with epileptic seizures died within 15 years of the original referral.

Udo Wieshmann

The Walton Centre for Neurology and Neurosurgery

Background: Health-related quality of life (HRQOL) in children with epilepsy (CWE) is multifactorial. In addition to child-specific factors, parental factors can impact a child’s HRQOL. Parents of CWE are at increased risk of mental health difficulties including anxiety and depression, as well as sleep difficulties. From both the child’s and parent’s perspective, HRQOL has been shown to be strongly related to parental mental health, however, there is no literature on parental sleep as a predictor of child HRQOL. In addition, co-occurring neurodevelopmental traits play a significant role in child HRQOL. We aimed to understand whether parental factors play a stronger role in determining child HRQOL compared to neurodevelopmental traits, which can aid in guiding epilepsy management in clinical practice. Methods: Parents of 33 CWE aged 4-16 years old completed a battery of questionnaires assessing child sleep disturbances, neurodevelopmental traits, seizure severity, and parental mental health and sleep. Principal component analysis reduced five child variables to two main components forming composites of sleep/neurodevelopmental traits (PC1) and seizure severity (PC2). Two regression analyses were used to predict child HRQOL from PC1 in combination with parental anxiety and depression, respectively. Results: 36.3% and 12.1% of parents scored above the clinical cut-off rates for anxiety and depression respectively, which is consistent with prior research. Similarly, 71.4% of parents scored above the threshold for significant sleep quality disturbance. In the first regression, parental sleep quality and anxiety were not predictive of child HRQOL, but PC1 was. In the second regression, parental depression and PC1 were significantly predictive of child HRQOL. Conclusion: Our findings suggest that a systemic approach to epilepsy management is required in order to improve child HRQOL. Within this, both co-occurring conditions in children and parental depression contribute independently to a child’s HRQOL.

Alice Winsor

Centre for Human Brain Health, University of Birmingham

Background: Health-related quality of life (HRQOL) in children with epilepsy (CWE) is affected not only by the chronicity of the disorder, its seizures, and the side effects of antiepileptic medication (AEDs), but by the impact of co-occurring conditions. These can include sleep disturbances and neurodevelopmental conditions such as autism or ADHD, all of which are prevalent in CWE. Consideration of the impact of these conditions on child HRQOL is scarce within research and clinical practice. Moreover, as sleep problems are common in both epilepsy and neurodevelopmental conditions, it is important to understand their independent contribution to HRQOL in CWE. Methods: Parents of 33 CWE aged 4-16 years old completed a battery of questionnaires assessing child sleep disturbances, autistic traits, ADHD traits, seizure severity, and epilepsy-specific HRQOL. Clinical records of patients were reviewed for information on antiepileptic medication. Results: Hierarchical multiple regressions investigated the contribution of these measures to child HRQOL. A first analysis indicated that sleep disturbances but not epilepsy-specific factors (seizure severity, number of AEDs) contribute to poor HRQOL. However, in a second analysis, it was found that the inclusion of ADHD and autism traits resulted in these neurodevelopmental conditions predicting HRQOL, but sleep disturbances were no longer a significant predictor of HRQOL. Conclusion: Our findings suggest that co-occurring neurodevelopmental traits play a stronger role in determining HRQOL in CWE than epilepsy characteristics or sleep disturbances. Currently, the role of neurodevelopmental traits on HRQOL in epilepsy management is not largely considered. Recognition of the role of co-occurrences in CWE can help to increase access to pathways in services and support for parents to improve overall HRQOL.

Alice Winsor

Centre for Human Brain Health, University of Birmingham

Introduction: Dravet Syndrome is an inherited childhood epilepsy caused by a mutation in the SCN1A gene, which encodes the voltage-gated sodium channel NaV1.1. Patients suffer from cognitive impairment and generalized treatment-refractory seizures which frequently progress to status epilepticus and are associated with a high mortality. Knockout mice from an existing 129S-Scn1a strain develop seizures but do not breed well, thereby presenting a challenge. Here we characterise the neurological phenotype of a Scn1a-/- mouse model bred onto a CD1 background which possesses enhanced fertility when compared to the existing model. Methods: Scn1a-/- mouse were generated by crossing heterozygous 129S-Scn1atm1Kea/Mmjax with CD1 mice. The resulting heterozygous F1 mice were used to obtain F2 CD1x129Sv Scn1a-/- pups whose survival and behaviour were characterised. Scn1a and NaV1.1 expression were assessed using quantitative PCR and western blot. We additionally assessed CD68 and GFAP expression in brain tissue. Results: F2 CD1x129Sv Scn1a-/- pups reach humane end-point by post-natal day 14 (P14), with weight loss starting around P12, associated with a continuous decrease in their motor activity, frequent changes in directions when walking resembling ataxia, and take significantly longer to return to prone position in the righting reflex test. Seizures were observed at P13 and P14. There was a significant decrease in the expression of Scn1a and NaV1.1 in the cortex, hippocampus and diencephalic region. We did not find an increase in inflammatory markers in the brain. Conclusions: Scn1a-/- CD1 mice show a similar disease progression to the previously described strain but importantly possess an additional advantage of having large litters. Interestingly, here we have shown that Scn1a is expressed throughout the brain in a mouse model of Dravet Syndrome and that there is a significant decrease in its expression in Scn1a-/- mice, not only in the hippocampus, but also in the cortex and diencephalon. These findings highlight the need to design new treatment approaches that not only focus on the hippocampal expression of NaV1.1.

Juan Antinao-Diaz

EGA Institute for Women's Health, University College London, WC1E 6HX, UK

Spike and wave discharges (SWD), generated by the cortico-thalamo-cortical (CTC) network, are pathological, large amplitude oscillations and the hallmark of absence seizures. SWD begin in cortical initiation networks in both humans and animal models, including the Genetics Absence Epilepsy Rats from Strasbourg (GAERS), where they initiate in the primary somatosensory cortex (S1). The mechanisms of how SWD rapidly spread across the brain is unknown and cannot be explained within the principal CTC network. We have now investigated the role of higher-order thalamic (HO) nuclei in the generalisation of SWD in freely moving GAERS. The diffuse connectivity of HO nuclei, their known physiological interactions with S1 and their altered synaptic anatomy in GAERS make these nuclei serious potential candidates for the rapid cortical spreading of SWD. Local field potential recordings in HO nuclei and different cortical regions revealed a novel feature of SWD. The synchrony in cortical regions located far from S1 (such as primary visual cortex, V1) transiently disappeared, i.e. there are short breaks in SWD (named SWD-breaks). Additionally, relatively brief SWD would occur only in S1 or only in S1 and the neighbouring primary motor cortex (M1), but not elsewhere. These spontaneous events provided a unique mechanistic insight since they represent unsuccessful maintenance and generalisation of SWD. Local inhibition of different HO thalamic nuclei with muscimol increased the delay of SWD propagation and the occurrence of SWD-breaks. Moreover, HO nuclei single unit recordings revealed three groups of putative excitatory neurons with different burst firing dynamics during SWD. Notably, all three neuronal groups exhibited a switch from tonic to burst firing before onset of SWD, but they responded differently during V1 SWD-breaks and non-generalised SWD. These results support the view that trans-thalamic cortical communication is utilised in the initial propagation of SWD and has an active role in maintaining cortical synchrony throughout the paroxysmal activity.

Zoe Atherton

Cardiff University

The use of medical/recreational marijuana and its legal market size are on the rise worldwide, including in the UK. Medical Marijuana has potential benefits for different diseases, but the lack of biomedical research and its popularity based on anecdotal evidence poses a risk to society. For instance, medical use of cannabis and derived cannabinoids show promising results in severe paediatric epilepsies. However, cannabinoids acting on the endogenous cannabinoid system can aggravate different types of seizures. Here we have clarified how the cannabinoids affect absence seizure, the most common childhood epilepsy. The failure of first and second monotherapy with anti-absence drugs in 30% of children with absence seizures and the high rate of comorbidity, including attention-deficit-hyperactivity-disorders and learning/memory impairment (which persist in seizure-free children), demand novel therapeutic approaches and the cannabinoid system may be a promising target. Using cutting-edge in vivo and in vitro techniques, we have investigated the effect of drugs that increase the concentration of endogenous cannabinoids in absence seizures GAERS model by blocking key catabolic enzymes. Moreover, we have characterized the effect of exogenous cannabinoids in this absence seizure model. Our preliminary data indicate a complex role of exogenous and endocannabinoids in controlling absence seizures. Since endocannabinoid modulators with an excellent safety record in humans are already available, our data may identify new targets that in the medium-term lead to a novel treatment for absence and their comorbidity.

Giuseppe Di Giovanni

University of Malta

Introduction: Single gene disorders of the autophagy pathway are an emerging, novel and diverse group responsible for multisystem diseases in children. Clinically, these disorders prominently affect the central nervous system at various stages of development, leading to brain malformations, developmental delay, intellectual disability, epilepsy, movement disorders, and neurodegeneration. Variants in RUBCN have so far been identified in Saudi patients suffering from autosomal recessive spinocerebellar ataxia (Salih ataxia, SCAR15, OMIM # 615705). However, the molecular mechanism underlying its implication in ataxia is still unknown. Methods: We employed NGS techniques, primary exome sequencing, to investigate a patient cohort presenting with paroxysmal neuromuscular conditions, collected as part of the SYNAPS Study Group at UCL. We have generated plasmids expressing each RUBCN variant and checked the interaction with Rubicon-binding proteins (e.g. RAB7, Beclin1, UVRAG) by transiently transfecting into HeLa cells. Results: We identified 5 novel patients from 5 unrelated families with biallelic variants in the RUBCN gene found by trio exome sequencing. In all 5 patients identified, psychomotor development was severely delayed in all domains. Affected individuals did not communicate well and did not achieve complete speech. Early onset seizures were present in cases 1 and 2 which exhibited the most profound phenotype. Two affected individuals found to carry a RUBCN homozygous truncating mutation (Arg466Ter) presented a severe phenotype in our preliminary cellular analysis. When the interaction and colocalization with RAB7 and the Beclin1-PI3K-UVRAG complex was investigated, we found that Arg466Ter is a short fragment (70 kDa) mutant lacking Rab7 binding region, which escapes NMD. Fluorescent microscopy confirmed that Arg466Ter was not colocalized with RAB7 or the Beclin1-PI3K-UVRAG complex. Rescue experiments using siRNA-treated cells which test the effect on autophagy flux by measuring tfLC3 (tandem fluorescent LC3; mRFP-EGFP-LC3) showed that with variants Glu107Lys and Arg466Ter, autophagic activity was not restored compared to wild-type expression. Discussion: Although mutations in RUBCN have been so far identified in patients suffering from autosomal recessive spinocerebellar ataxia, it is still unclear whether RUBCN is related to symptoms in these patients. To find the missing link between the patients’ phenotypes and the cell culture model established here, we would like to further establish a conditional knockout mouse model. We are planning to generate central nerve system-specific conditional knockout mice by mating with Nestin-cre mouse. The wider academic community would benefit from greater understanding of the function of Rubicon because of the fundamental role it plays both in autophagy and likely other pathways.

Stephanie Efthymiou

UCL Institute of Neurology, Queen Square, WC1N 3BG, UK

Mutations in the SYNGAP1 gene are associated with non-syndromic intellectual disability (ID) and autism spectrum disorder (ASD). Most patients with SYNGAP1 pathogenic variants display some form of epilepsy, with high prevalence of absence seizures. SYNGAP1 haploinsufficiency is an epileptic encephalopathy (DEE) with cognitive and behavioural developmental plateauing occurring upon seizure onset. Additionally, impairments of the sleep-wake cycle have been described in many forms of ASD and are common in a high proportion of patients with SYNGAP1 mutations. We previously found that 75% of Syngap+/∆Gap rats, which are heterozygous for deletion of C2 and GAP SYNGAP1, displayed spontaneous spike and wave discharges (SWDs) with simultaneous head bobbing, consistent with an absence seizure phenotype. We were able to block SWDs by acute treatment with the T-type calcium channel blocker, ethosuximide, a standard anti seizure drug used to treat absence seizures. In this study we assessed the full-day circadian distribution of the SWDs along with sleep-wake states. We found an increased number of SWDs through the entire circadian cycle as well as a significant decrease in the amount of REM sleep in Syngap+/∆Gap rats when compared to littermate wild-type controls. The REM decrease was present during both the light and dark phases of the day and was a result of shorter duration REM episodes. To assess a possible relationship between the seizure phenotype and sleep impairments, we treated animals with a single dose of systemic ethosuximide. The treatment successfully blocked SWDs, although the effect was limited to a 10 to 16 hr period, equivalent to the reported half-life of the drug. Strikingly, REM sleep deficits were rescued both during the drug half-life and during the subsequent 12 hours, corresponding to the dark phase of the circadian cycle. The number of REM sleep episodes were increased, resulting in an equivalent total duration compared to controls. Saline treatment had no effect on SWDs or REM sleep. Our results suggest that sleep is altered by seizure activity and may contribute to cognitive and social deficits in SYNGAP1 haploinsufficiency.

Alfredo Gonzalez-Sulser

Simons Initiative for the Developing Brain, Patrick Wild Centre

Background: Mitochondrial diseases are common neurogenetic disorders that arise due to mutations in either mitochondrial DNA or nuclear DNA and are characterised by oxidative phosphorylation (OXPHOS) defects. Approximately 1 in 4 adult patients develop epilepsy providing a direct link between mitochondrial dysfunction and development of epilepsy. Seizures play a major role in the development of stroke-like episodes which are a well-recognised feature of several forms of mitochondrial disease (particularly in mitochondrial encephalomyopathy, stroke-like episodes and lactic acidosis (MELAS)). Seizure-associated stroke-like episodes correspond to cortical T2-hyperintensities on MRI head and focal epileptic discharges on EEG. The process by which seizures are triggered and spread is not well understood. However, there are opposing theories of the aetiology. Hypothesis: Compromised inhibitory restraint due to selective vulnerability of y-aminobuytric acid (GABA)-inhibitory interneurons to OXPHOS dysfunction underlies seizure-associated stroke-like episodes. Methods: Neuropathological reports were reviewed from 26 patients (16 with documented ante-mortem stroke-like episodes). Formalin-fixed paraffin-embedded brain tissues were acquired from the Newcastle Brain Tissue Resource for 10 patients with clinically and genetically defined mitochondrial disease and 10 cognitively normal, aged-matched controls. Cerebellum, occipital and, temporal cortex were subjected to histopathological staining to determine neuronal population density and immunofluorescence staining to interrogate OXPHOS subunit expression within interneurons, astrocytes, capillaries and arterioles. Results: Macroscopic necrotic cortical lesions were evident upon brain dissection and histology showed neuronal cell loss, ranging from selective neuronal dropout to cortical laminar necrosis, with presence of reactive astrocytes and macrophages. This was a common neuropathological finding identified across all genotypes. The brain regions most affected by laminar necrosis included cerebellum, brainstem, temporal and occipital lobes. Immunofluorescently labelled OXPHOS subunits; complex I (NDUFB8) and IV (COXI) were quantified relative to mitochondrial mass marker (porin) in inhibitory interneurons, astrocytes, capillaries and arterioles. Analysis revealed that inhibitory interneurons show the greatest loss of OXPHOS subunits relative to astrocytes and microvessels. Discussion: This study represents the largest cohort reporting clinical and pathological correlates of seizure-associated stroke-like episodes in patients with mitochondrial disease. There is clear evidence that the neuronal deficit is more pronounced and given the preponderance of inhibitory interneuron involvement, provides further evidence that these paroxysmal stroke-like events are seizure-induced, and are not driven by angiopathic changes. Whether these paroxysmal events result from concomitant dysfunctional neuroinflammatory response to various triggers, involving the interaction of neuronal, vascular and innate immune processes, as in other refractory seizure disorders, that is ultimately predicated on mitochondrial dysfunction remains to be determined.

Nichola Lax

Newcastle University

Approximately 20-30% of people diagnosed with epilepsy have dissociative seizures (also known as psychogenic non-epileptic seizures). This is particularly problematic in the acute setting where misdiagnosis can result in in unnecessary treatment and morbidity. However, the gold standard diagnostic tool of video-EEG at the time of the attacks is not readily available in the emergency setting and, delayed, interictal EEG is poor at distinguishing those with epilepsy from those without this condition. Serum prolactin has been used as a marker of epilepsy, but because of poor sensitivity and specificity, it is no longer recommended in national guidelines. We propose to use non-invasive detection of volatile organic compounds (VOCs) within breath and sweat samples after seizure and non-seizure events to distinguish the two. A VOC is a carbon-containing compound that is sufficiently volatile to be detectable in the gas phase at room temperature. Therefore, gas chromatography–mass spectrometry (GC-MS) is the analytical method of choice based on a targeted approach. Exhaled VOCs have already been used by applicants to detect a number of human diseases including lung, breast, bladder, prostate and gastrointestinal cancers. Our study takes place on a video-EEG telemetry unit where we test people’s breath and sweat before and at specific time points after a generalised, focal or psychogenic seizure. We have shown in a preliminary study of 8 patients that this is feasible, and that VOCs change after seizure events. In this study, we aim to determine the unique breath and sweat VOC signature of patients with different seizure types, and to use it as a biomarker for epilepsy to develop a non-invasive diagnostic model for these disorders. This tool could be used at the point of care in the community, ambulance and emergency room.

Eleonora Lugara

UCL, DCEE & National Hospital for Neurology & Neurosurgery, Queen Square, London

Multiple clinical and experimental studies have reported the concurrence of cortical seizures and the neurophysiological correlate of migraine aura, cortical spreading depression (CSD). Although observed in multiple environments, the relationship between these two phenomena remains enigmatic. CSD waves propagate across the cortex at a velocity of ~3mm/min and lead to the destabilisation of ion homeostasis, resulting in the prolonged suppression of subsequent neuronal activity. During epileptic paroxysms, this suppression of activity has been observed to result in seizure termination. However, many other studies have also observed an increase in excitability and epileptiform discharges during the period immediately following CSD. This study aims to elaborate on the nature of this elusive relationship through the in-depth characterisation of interactions between propagating interictal discharges, seizures and CSD waves. Paramount to deciphering these relationships is understanding the spatial propagation and interaction of these events on a cortex-wide scale. Many studies to date have investigated these phenomena by sampling single, or few, point measurements from the cortex using DC-coupled electrodes. Using bilateral GCaMP7a-based widefield Ca2+ imaging, our study aims to capture information regarding the propagation of both seizures and CSDs across different cortical regions. Moreover, to provide high temporal resolution and achieve mapping in greater detail, we simultaneously utilise 16-channel graphene solution-gated field effect transistor (gSGFET) arrays to record full-bandwidth electrophysiology with exceptional stability. Following induction of epileptiform discharges, we first examine the temporal relationship between seizures and CSD bilaterally in the awake murine cortex. Using our custom-built automated analysis pipeline, for each event, we are able to classify them based on cortical origin and propagation characteristics. For each CSD event, we extract multiple properties (such as amplitude, waveform, co-occurring hyperpolarisation, duration and laterality) and perform unbiased statistical testing with seizure-relevant parameters. Using gSGFET-acquired data, we resolve the full-bandwidth electrophysiology signals into physiologically-relevant frequency components and compare the power of each during different events, thus allowing us to correlate cortical oscillations with activity throughout both hemispheres. Altogether, the study presented here provides a previously unseen view into the relationship between CSD and cortical seizures. We envision that these outcomes will largely progress our understanding of seizure initiation and termination, facilitating the development of novel therapeutic avenues.

Daman Rathore

UCL Queen Square Institute of Neurology, University College London, London, UK

Purpose: To pilot the use of a completely non-invasive auditory sleep aid, shown to safely modulate neural activity in healthy participants, in a cohort of patients with temporal lobe epilepsy. Closed Loop Auditory Stimulation (CLAS) has additionally been shown to enhance sleep and memory in healthy controls possibly by strengthening memory consolidation circuits, but as yet has not been used in these patient populations, although sleep and memory problems are often co-morbid. 15 outpatients (age 20-41) with temporal lobe epilepsy (TLE) and 22 (age 19-30) neurotypical controls preformed memory tests before and after one night of sleep and wore a portable EEG band during sleep to record sleep parameters and deliver stimuli. Patients underwent two rounds of testing, with either 50(±5) decibels of CLAS delivered during slow wave sleep, or no auditory stimulus. The study was carried out completely remotely due to the COVID pandemic of 2020. Methods: During slow wave sleep, neuronal activity is dominated by 0.5-2 cycles per second (Hz) oscillating signals that can be easily recorded by scalp EEGs. CLAS synchronizes 100ms bursts of pink noise with the peaks of slow oscillations. In this study, we used portable 5-sensor EEG bands to record sleep under stim and sham nights, spaced at least one week apart. Each participant underwent pre- and post- sleep memory testing of a Paired Associates Learning task of paired words and a serial reaction task. Participants had a clinical diagnosis of TLE, with fewer than 3 generalized tonic clonic seizures per month for the last three months and were taking fewer than 3 anti-seizure medications. Results: No seizures or serious adverse effects were reported on any night of the study, and the study was well-tolerated (87% participants completing). EEG analysis of patient data showed that CLAS modulated neural activity in a specific sub-sample of patients only. Effect size (of the subsequent slow wave amplitude) was correlated with having a higher proportion of slow wave sleep R2=0.76, but not higher sleep duration R2 =0.11. Neither correlation was seen in neurotypical controls (R2 <0.01). Additionally, patients with the largest effect of CLAS were correlated with higher baseline memory scores (R2=0.67). Conclusion: These data suggest that CLAS in patients with epilepsy may be strongest in those with already relatively good sleep and memory and has limited effects in those with more pronounced sleep and memory issues, as circuits may be more fundamentally disrupted, or simply not be as malleable. This supports a ‘rich get richer’ model of closed loop stimulation.

Jennifer Roebber

CUBRIC, School of Psychology, Cardiff University, Cardiff, UK

More than 30% of all people with epilepsy are eventually considered refractory: their seizures are not well-controlled with anti-epileptic drug treatments. For this significant cohort of people with epilepsy, surgery could offer a potentially transformative treatment in terms of reducing the frequency of seizures. However, only a minority of people with refractory epilepsy are usually considered suitable for surgery, with long-term seizure freedom is estimated to be achieved in half the cases. In particular, people with epilepsy may display a reduction in seizure frequency immediately after the surgery, but their seizures may often return over time and may even be different in nature to those before the surgery (De Tisi et al. 2011; Mohan et al. 2018). However, despite these challenges, epilepsy surgery is established a cost-effective solution, and recently there has been an increase in calls for more widespread acceptance of surgery as a treatment for people with refractory epilepsy (Sheikh et al. 2020; Engel 2016). Recently, several computational approaches have been proposed to provide support pre-surgical planning (Goodfellow et al. 2016; Sinha et al. 2016; Khambhati et al. 2016; Jirsa et al. 2017). Typically, these approaches use a dynamic network model to explore the potential impact of surgical resection in computer simulations. The network component of the model is typically informed by clinical imaging data and is considered fixed or static after the intervention. This assumption likely critically overlooks the plasticity of the brain and, therefore, how continued evolution of the brain network post-surgery may impact upon the success of a resection in the longer term. In this work, we use a simplified dynamic network model, which describes transitions to seizures, to systematically explore how the network structure influences the likelihood of seizures occurring in the model simulations, both before and after virtual resections. We illustrate key results in small networks (four or five nodes), before extending our findings to larger networks. We demonstrate how the evolution of brain networks post resection can result in a return to increased seizure propensity. Our results effectively provide a way to determine the robustness of a given candidate resection to possible network reconfigurations and so provide a potential strategy for optimizing long-term seizure freedom. The next step for this research is to apply this theoretical framework to network structures and model parameters inferred from clinical data.

Wessel Woldman

University of Birmingham

Novel way of multi-disciplinary work is important for effective management of children with Neurodisability. We present a case diagnosed with ‘Metabolic Syndrome’ following rapid weight gain. 15-year-old boy with autism, learning difficulties, and epilepsy was started on Valproate (AED). He had been on this medication for 4 years uneventfully. Since March 2020, he started gaining weight and his weight was above 99th centile. He was hospitalised following deranged liver function and hypertension. He was diagnosed with type 2 diabetes (HbA1c 86 mmol/mol) and started on insulin followed by Metformin. His AED was changed to Levetiracetam with good seizure control. His sudden weight gain was multifactorial with sedentary lifestyle, unhealthy eating habits and limited exercise during lock down as his parents kept him off school in fear of COVID-19. His significant sensory issues, puberty and Valproate could be contributing factors. Metabolic Syndrome is a constellation of abnormalities with Hypertension, Dyslipidaemia, Insulin Resistance, and Impaired Glucose Tolerance and it is common in children with Obesity. Though exact cause is unknown, insulin resistance is thought to be central to its development. Multi-disciplinary input involving community paediatrician, diabetic, renal and liver team, dietician, physiotherapist, sensory team and school teachers was provided. We convened 3 virtual meetings using social media involving all the professionals and his parents for effective management. His liver function, blood glucose levels and weight were stabilised following strict diet, exercise regimen and parental education. His discharge HbA1c levels and blood pressure were within normal limits with medications. Weight monitoring is important especially in children with Autism and Epilepsy on Valproate during COVID-19. Early recognition of Metabolic syndrome symptoms is vital for appropriate intervention, management and prevention of complications. This pandemic has forced professionals to adapt novel ways of team working for effective management of children with Neurodisability.

Subramanian Ganesan

Leicestershire Partnership Trust NHS

Introduction: Chromodomain Helicase DNA Binding Protein 2 (CHD2) allows chromatin remodelling by binding nucleosome-free promotor regions and separating divergent transcription. Mutations in CHD2 are one of the more common causes of genetic epilepsy and developmental epileptic encephalopathy. We currently know that de novo point mutations in CHD2 are associated with intellectual disability and multiple seizure types, dependent on age, with some showing exquisite photosensitivity. We compiled an international cohort of patients with copy number variation or point mutation of CHD2 to further characterise the phenotypes of this cohort. Materials and Methods: Cases were collected from international collaborators, from the DDD consortium and from DECIPHER. Adults and children were included. We included cases with both point mutations and genomic deletions. Where available we collected: demographic data; neurological, psychiatric and seizure features; information about development and dysmorphology. Results: Results are given as a percentage total, where data were provided. We describe 49 patients, of whom 47% are male; median age was 15.5 years (range 3 – 57 years). 22% were copy number losses, 71% were point mutations. Of these, 68% were loss of function and 32% missense. Where the inheritance status was known, all mutations were de novo (57%). 92% of patients had either developmental delay or an intellectual disability (ID), or both. Regarding ID, 50% of those were mild, 3% mild-moderate, 7% moderate-severe, 20% moderate and 20% severe. Regarding behaviour, salient features were challenging behaviour (49%), aggression (35%), anxiety (15%) and hyperactivity (12%). Psychiatric complications of note included autism spectrum disorder or autistic traits (50%) and attention deficit hyperactivity disorder (33%). Epilepsy was diagnosed in 93%, of which 66% had generalised tonic-clonic seizures, 54% absence, 50% myoclonic, 32% febrile, 28% atonic and 22% focal. Other seizure types were also described. Convulsive status occurred in 23% and 45% were photosensitive. A range of dysmorphic features were described including short philtrum (29%), microcephaly (24%), prominent lower lip (24%), down-slanting palpebral fissures (19%) and short stature (17%). 41% had a normal appearance. Regarding intracranial imaging, 67% had no abnormality and, of the abnormal scans, 50% displayed atrophy – other features were less often described. 91% had an abnormal EEG. 27% of patients with epilepsy, who had medication data available, were maintained on sodium valproate monotherapy, but polytherapy and drug resistance were common. Conclusion: We present the largest cohort of patients with CHD2 mutations. Our findings demonstrate insights into neurodevelopment, behaviour, psychiatric complications, seizure type, patient phenotype and treatment. We suggest that CHD2 patients often display epilepsy, behavioural problems, ADHD and/or ASD and neurodevelopmental delay.

Alice Willison

Translational and Clinical Research Institute, University of Newcastle, UK