From gene discovery to precision intervention in epilepsy: almost the end of the beginning.

Abstract

This commentary is on the case report by Papandreou et al. on pages 416–420 of this issue. An unprecedented rate of gene discovery is occurring in the most severe of the epilepsies, the early onset epileptic encephalopathies (EOEEs). In these rare conditions, underlying epileptic activity is associated with, and at least partially responsible for, profound and progressive neurodevelopmental impairment. These syndromes are increasingly identified as monogenic. This is in contrast to more common epilepsies in which it is hypothesized that multiple genetic factors, each with relatively small effect (along with environmental factors), contribute to the phenotype. Early reports of monogenic early life epilepsies established that mutations in genes encoding ion channels such as the potassium channel KCNQ2 and sodium channel SCN1A, result in benign familial neonatal seizures and severe myoclonic epilepsy of infancy (Dravet syndrome) respectively. Yet the genotype–phenotype correlations became complicated. KCNQ2 mutations also cause EOEE, and SCN1A mutations also cause genetic epilepsy with febrile seizures-plus (GEFS+). It is complex distinguishing between mutations that lead to benign phenotypes, for which a clinician can confidently offer reassurance, and those resulting in a bleak prognosis. The calculus involves analysis of the mutation type and location, conservation of nucleotide position across species, conservation of amino acid effect, de novo versus inherited transmission, and overall effect on protein function. The presence of modifier and regulatory genes and epigenetic factors play putative roles. The list of EOEE genes expands beyond channelopathies and includes a long list of cellular signaling, synaptic vesicle, transporter, enzyme, and membrane receptor genes. The latter include glutamatergic NMDA receptor (GRIN2A, GRIN2B) and GABA-A receptor (GABRA1, GABRB3) genes. Papandreou et al. report a novel de novo heterozygous missense mutation of GABRB3, which encodes the beta-3-subunit of the pentameric GABA-A receptor, in a child with EOEE.1 This gene has previously been implicated in other neurodevelopmental disorders, including intellectual disability, autism spectrum, Angelman syndrome, and epilepsy, including both absence epilepsy and EOEE.2 The reported case presented with neonatal hypotonia and failure-to-thrive, followed by refractory seizure onset at 3 months; the child had profound intellectual disability and acquired microcephaly by 3 years. The evaluation highlights the utility of gene panels for selected conditions to establish a diagnosis in a relatively non-specific phenotype. Whole exome sequencing will increase diagnostic yield even further and is expected to subsume disease specific gene panels as state-of-the-art. The state of the field is moving from mutation discovery to functional assessment with an eye toward targeted therapy. This is taking place for individual EOEE genes with glimmers of progress. Zebrafish provide a high throughput drug screening platform and identify the antibiotic clemizole as potential therapy for SCN1A-related epilepsies.3 The observation that quinidine can reverse large potassium currents emanating from gain-of-function KCNT1 mutations led to rational intervention in a patient with epilepsy of infancy with migrating focal seizures, among the most severe of the EOEEs.4 Following exome sequencing which identified a GRIN2A mutation in a proband with EOEE, functional modeling and drug screening led to successful clinical intervention using memantine.5 These studies provide proof of principle that we are on the verge of employing rational, targeted, precision therapy for the severe monogenic epilepsies. Ezogabine/retigabine is an FDA-approved anti-seizure medicine and a KCNQ2-channel opener. Yet the identification of a KCNQ2 gene variation first requires analysis whether the patient has a benign polymorphism or an epilepsy syndrome. This involves increasingly large population studies which are in progress; hence, we are in the age of ‘big data’. The next distinction is to establish whether the diagnosis is the syndrome of benign familial neonatal/infantile epilepsy or instead EOEE. Furthermore, the functional impact of the mutation is integral because ezogabine may be expected to be beneficial if there is loss of channel function but deleterious in the case of gain of function. Furthermore, whether a salutary effect on seizure activity will in fact improve the overall neurodevelopmental outcome is the ultimate clinical question. In a nod to Papandreou et al. (a UK group), the timing is Churchillian, i.e. almost ‘the end of the beginning’.