Search

Childhood epilepsy is an important, largely unrecognised public health problem with pervasive economic and psychosocial impact in developing countries. Controversy over the side-effects of phenobarbitone, the challenges of case-finding for a stigmatising disorder, attrition from treatment, and lack of meaningful evaluation are major obstacles to achieving control. This study assessed the acceptability and relative efficacy of phenobarbitone in rural India and evaluated the effectiveness of an intervention within a community-based rehabilitation programme, using broad measures of outcome. A population of 117,000 in rural West Bengal was screened for childhood epilepsy, by key informant and survey methods, compared for cost, sensitivity and acceptability. Efficacy and side effects of phenobarbitone and phenytoin were compared in a randomised controlled trial of 94 untreated children. Parental adjustment, behavioural problems and social integration were blindly assessed at 12 months. Reasons for dropout were investigated. The survey was four times more sensitive but less acceptable to the community. There was no evidence of a difference in efficacy between phenobarbitone and phenytoin (hazard ratio 0.97; 95% CI: 0.32-2.78). There was no evidence of excess behavioural problems with phenobarbitone in parental reports (p=0.72), or by objective rating (odds ratio 0.51; 95% Cl: 0.16-1.60). Social integration was greatly impaired at baseline (p<0.01), intervention was associated with improvements, especially for girls. Parental adjustment was independently associated with low severity of impairment (p=0.01). Both child behavioural problems (p=0.03) and parental adjustment (p=0.06) were independently predicted by maternal satisfaction with social support. Ascertainment cost $8-$11 per case, and programme costs were $1 per child per month. Access and symptom resolution accounted for half the dropouts. Phenobarbitone is an acceptable first-line drug for childhood epilepsy in India. Ascertainment by key informants is more appropriate in community-based rehabilitation, itself an appropriate framework for intervention which resulted in important psychosocial changes, and at feasible cost.

Juvenile Myoclonic Epilepsy (JME) is a common Idiopathic Generalised Epilepsy (IGE) with an adolescent onset and characteristic myoclonic jerks. The phenotypic spectrum of JME is broad, with some patients also experiencing absence seizures, executive function deficits and poor seizure prognosis. Alongside the hallmark generalised spike-/polyspike-and-wave epileptiform discharges observed on the EEG of patients with JME, previous studies have also shown alterations in the resting- state/interictal EEG of patients with IGE including alpha oscillatory activity and brain network topology, with evidence of these alterations also existing in unaffected relatives. Due to its complex genetic nature, results from previous genetic studies of JME have been inconsistent and therefore somewhat inconclusive. The use of endophenotypes can help to decrease the heterogeneity of complex disorders such as JME to give more statistical power in genetic association studies. In this thesis, I investigate potential neuropsychological, electrophysiological and clinical endophenotypes, or intermediate phenotypes of JME, for use in genetic association studies of the condition. These include impulsivity, as measured by the Barratt Impulsiveness Scale (BIS), resting-state EEG features and in-depth exploration of the JME clinical phenotype, including seizure type and prognostic group stratification. Participants with JME were recruited via the Biology of Juvenile Myoclonic Epilepsy (BIOJUME) study, whereby BIS-brief questionnaires, genetic, clinical and EEG data were collected. Strict eligibility criteria and a phenotyping panel confirmed a diagnosis of JME. I used multivariable, stratified analysis to explore clinical phenotypes, such as seizure types and a sensitivity to seizure precipitants, and factors influencing prognosis in 765 individuals with JME. Further, I compared JME BIS-brief scores (n=322) to matched controls and other clinical cohorts from the literature and explored clinical associations of impulsivity in JME. A comparison of resting-state EEG features, such as power spectral density, network topology and a dynamic measure of network ictogenicity (Brain Network Ictogenicity), between JME (n=147) and control EEGs (n=95) was carried out to assess their use as biomarkers and endophenotypes. Finally, a genome-wide association study (GWAS) and hypothesis-driven candidate gene study of three endophenotypes (trait impulsivity, measured by the BIS-brief, and two resting- state EEG features) in JME was carried out in collaboration with members of the BIOJUME consortium. Regarding the clinical phenotype of JME, I observed that poor seizure prognosis is associated with experiencing absence seizures in both males and females, while a susceptibility to stress-related precipitants negatively modifies disease course in females only. Alongside the association with poor seizure control, absence seizures are also associated with increased trait impulsivity which is significantly higher in JME than in matched controls, to a level equivalent to cohorts with neurotic- type disorders. GWAS analysis of trait impulsivity in 373 individuals with JME gave 3 significantly associated loci, annotated to SLCO5A1, DUSP10 and PARD3 genes, exemplifying the strength of using endophenotypes to uncover novel genetic associations. In addition, gene-set enrichment analysis indicated an 8.5-fold enrichment of genes involved in the regulation of presynapse assembly/organisation in the GWAS of trait impulsivity. Analysis of resting-state EEG features showed significant differences between JME and control EEGs, particularly features relating to activity in the low-alpha frequency range (6-9Hz) and network ictogenicity, whilst network topology measures showed significant associations with JME clinical characteristics and outcomes. Genetic association analysis of two EEG endophenotypes (Brain Network Ictogenicity (BNI) and low-alpha power spectral density) with associated genes in the impulsivity GWAS (SLCO5A1, DUSP10 and PARD3) and genes involved in the regulation of presynapse assembly/organisation (n=33 genes), showed an association of SNPs at the PTPRD locus (a gene involved in the regulation of presynapse organisation/assembly) with BNI. Overall, this thesis verifies the heterogeneity of clinical, electrophysiological and neuropsychological phenotypes of JME and supports the use of endophenotypes to increase the power to detect novel and significant genetic associations in complex diseases. In addition, results from genetic association analyses suggest further research into presynapse organisation/assembly in the pathophysiology of JME.

Introduction: Rolandic Epilepsy (RE), an idiopathic focal epilepsy, is one of the most common epilepsies in school-age children, which frequently co-occurs with a heterogeneous mix of cognitive impairments (Pellock et al. 2016). These cognitive deficits predominantly involve reading disability, developmental language disorder and developmental coordination disorder (Overvliet et al. 2011, Smith et al. 2015, Vega et al. 2015). The seizures of this epilepsy spontaneously remit when the child reaches adolescence. However, it is unclear whether the same occurs with the cognitive impairments (Camfield and Camfield 2014, D'Alessandro et al. 1990, Garcia-Ramos 2015, Metz-Lutz and Fillipini 2006, Northcott 2006, Overvliet 2010). Furthermore, it is unknown how brain structure and function changes in respect to remission of seizures and whether this differs from the development of healthy children, which could provide a better understanding of the process to seizure remission. Methods: In this longitudinal study, children with active Rolandic epilepsy and healthy children were invited to have structural neuroimaging and neuropsychological testing at baseline and 4.5 years later at follow-up, in seizure remission. Furthermore, at follow-up, a non-clinical sleep electroencephalogram (EEG) was performed. Details of the investigations are below: Neuroimaging: Longitudinal structural magnetic resonance imaging (MRI) scans were used to assess cortical thickness and subcortical volumes. The a priori hypothesis proposed at baseline increased cortical thickness within the inferior frontal gyrus, supramarginal gyrus and inferior parietal lobe and the transverse temporal and superior temporal gyrus at baseline with reduced thinning at follow-up. Furthermore, the putamen would be larger in volume at baseline and would increase in volume at follow-up. Also included, were exploratory investigations of cortical thickness within the whole brain and parcellated regions. Neuropsychology: Longitudinal neuropsychological assessments were used to assess fluid intelligence, single-word reading, central auditory processing, indication for developmental coordination disorder (DCD) and attention deficit hyperactivity disorder (ADHD). At follow-up, additional tasks were included to investigate phonological processing and the nature of fine-motor deficits. The a priori hypothesis proposed that DCD would be the most prevalent disorders, and there would be additional layers of cognitive problems. The first layer would be epilepsy specific, and the second layer suspected to be familial in origin. In seizure remission, it was proposed that the epilepsy-specific cognitive problems, which include motor and attentional issues, would reduce, whereas familial features such as dyslexia would remain. Electroencephalography: Clinical EEG reports were reviewed, and the appearance of Rolandic spikes (RS) and their location were recorded. At follow-up, a non-clinical, sleep-deprived EEG was recorded to see if there was any evidence of RS in seizure remission. Furthermore, using global power analysis and topographic power maps, sections of awake resting-state EEG were compared with similar EEG samples from clinical EEG recordings during active epilepsy. The a priori hypothesis derived from Bouma et al. (1997), proposed that RS would be apparent in 10% of the cohort in seizure remission. An additional, exploratory component investigated changes in global EEG power and peak-dominant frequency between active epilepsy and seizure remission. Results: Neuroimaging: The longitudinal neuroimaging data demonstrates patchy regions of cortical thinning across the cortex; however, in the areas where thinning occurred, there was a greater reduction in cortical thickness compared to controls. The greatest thinning occurred in the bilateral frontal lobes, insular and anterior temporal regions. The cross-sectional analysis shows regions of predominantly thinner cortex at baseline, whereas, at follow-up, altered regions were likely to have thicker cortex compared to controls. In particular, a large region of thicker cortex was identified around the right post-central gyrus. There was no difference in the size of putamen at baseline or follow-up, whereas the longitudinal analysis found an increase in the volume of the putamen over time. In the control group, putamen size decreased with time. Neuropsychology: At baseline, an indication for DCD was the predominant cognitive problem; this frequently co-occurred with evidence for dyslexia and ADHD. At follow-up, the number of participants with an indication for DCD was reduced compared to baseline, but the numbers at follow-up were greater than the controls. The longitudinal analysis found an improvement in group cognition, in particular in the processing of filtered words; there was one exception, a deterioration in matrix reasoning scores. Despite group improvements, specific cognitive problems persisted in some individuals, and in some, new ones appeared in seizure remission. Electroencephalography: The review of clinical EEG reports found a predominance for spikes over the right hemisphere. Furthermore, there were individuals with spikes before seizures and no spikes recorded during active epilepsy. At follow-up, in seizure remission, 50% had evidence of EEG abnormalities, of which, 28.6% had evidence of RS. In the remaining individuals with abnormalities, there was evidence of poorly formed, short duration, generalised spike and slow-wave discharges. Quantitative analysis of EEG resting state between active epilepsy and seizure remission found a relationship between relative delta power and time to final seizure furthermore; there was a significant difference in the topography of absolute delta power between active epilepsy and seizure remission. Conclusions: This longitudinal controlled study has demonstrated altered cortical thickness and abnormal cognition in both active epilepsy and seizure remission. In those individuals with cognitive problems, the cognitive profile of RE is predominantly based on motor and coordination problems which co-occur in with dyslexia and ADHD. Specific cognitive problems can persist or appear in seizure remission, and thus a rethink of the educational assessment of individuals with RE is proposed. Furthermore, large amounts of thinning occur in the frontal lobes in seizure remission; this would suggest that the maturity of this structure is required for seizure remission, and this presents a new therapeutic target. The appearance of new generalised spike and waves discharges in seizure remission needs to be replicated and investigated to see how prevalent they are and whether they influence cognition. Finally, the EEG evidence lessens the role of the RS and implicates the power of resting delta wave activity in the generation of seizures; which could improve our understanding and treatment of this common childhood epilepsy.

Myoclonic astatic epilepsy (MAE) is a rare generalised childhood epilepsy with variable but poorly described neurodevelopmental outcome. Family studies suggest a major genetic influence as up to two thirds of relatives have seizures, or electroencephalographic (EEG) abnormalities. MAE is associated with 10 different genes, yet these genes account for less than 20% of the genetic aetiology of MAE leaving the majority unexplained. The aims of this thesis were (1) describe the epilepsy and neurodevelopmental phenotype of MAE cases, (2) perform EEG studies on first degree family members for familial EEG abnormalities and compare occurrence of epileptiform features to population prevalence and (3) to collect DNA and identify MAE causative genetic variants through exome sequencing. I assembled the largest MAE cohort (n=123) to date. The epilepsy phenotype is remarkably similar to previously published cohorts. I identified a severe neurodevelopmental phenotype: intellectual disability was reported in 64.9%, autism spectrum disorder in 21.3% and attention deficit hyperactivity symptoms in 41.0%. Additionally, extremely low adaptive behavioural scores were identified in 69.4% of cases. I performed EEG studies on 38 first-degree relatives of 13 MAE families, and found an excess of epileptiform EEG features in adults (>16 years), compared to controls (P=0.05, RR 6.82). I identified likely pathogenic or candidate variants in 11 of 109 cases. This comprised known genes associated with MAE: CHD2 n=1, SYNGAP1 n=2, SLC6A1 n=1, KIAA2022 n=1; epilepsy associated genes novel for MAE: KCNB1 n=1, MECP2 n=1, KCNH5 n=1, and three new candidate genes; SMARCA2 n=1, ASH1L n=1 and CHD4 n=1. Lastly, I highlight phenotypic features which help correlate with known and novel specific gene associations, discuss that MAE is a phenotypic and genetic nosological bridge between genetic generalised epilepsy and epileptic encephalopathy, and discussion applications and future directions leading on from this project.