As the genetic etiologies of an expanding number of epilepsy syndromes are revealed, the complexity of the phenotype genotype correlation increases. As our review will show, multiple gene mutations cause different epilepsy syndromes, making identification of the specific mutation increasingly more important for prognostication and often more directed treatment. Examples of that include the need to avoid specific drugs in Dravet syndrome and the ongoing investigations of the potential use of new directed therapies such as retigabine in KCNQ2-related epilepsies, quinidine in KCNT1-related epilepsies, and memantine in GRIN2A-related epilepsies.
Key points
- •
Epilepsy can be secondary to single gene, or much more commonly, to multifactorial and remote symptomatic etiologies, such as stroke or hypoxia.
- •
Causative single gene mutations have been identified that result in multiple genetic, often overlapping, epilepsy syndromes each with distinctive clinical phenotypes.
- •
Causative mutations causing epilepsy have either been in ion channel genes, neurotransmitter receptor genes, structural genes, or signal transduction pathway genes.
- •
Discovery of the causative mutations is allowing for investigation of more targeted and precision individualized therapies of patients with epilepsy owing to such mutations.
Introduction and general principles
The role of genetics in epilepsy has been contemplated since the time of Hippocrates. Subsequently, twin studies, family studies, linkage studies, and more recently copy number variation and whole exome sequencing studies on larger populations have provided modern added evidence for the role of genetics in epilepsy. Whereas the role of genetic factors in idiopathic epilepsies has long been suspected, the role of these factors in cryptogenic and symptomatic epilepsies has been demonstrated in a number of other studies. It is now thought that genetic factors account for about 40% of the etiologic causes of epilepsy. Mendelian epilepsies, which can be demonstrated to have such inheritance, however, account only for 1% of epilepsies. Epidemiologic studies have estimated the risk of epilepsy for offsprings and siblings of patients with epilepsy of any cause to be about 2% to 5%. Twin studies continue to show higher concordance in monozygotic compared with than dizygotic pairs (case-wise concordance of 0.62 vs 0.18, respectively). With the recent advance in modalities of genetic testing, multiple gene mutations have been discovered to be the cause of a wide spectrum of genetic epilepsies. The term ‘genetic epilepsies’ has expanded to include conditions caused by a genetic defect where seizures are the core manifestation. It has been hard to parcel out the role of environmental factors in such epilepsies, but it is thought that the genetic defect is the major determinant of the phenotype in such entities. While sorting out the interplay of environmental and genetic factors, the following 2 types of epilepsies have been recognized. The Mendelian epilepsies, which are often “monogenic, simple, and rare” epilepsies account for about 1% of all epilepsy cases. These epilepsies include monogenic syndromes where single gene mutations produce the specific phenotype, including, for example, benign familial neonatal epilepsy syndrome. Monogenic epilepsies, however, may show variable penetrance and variable degrees of severity of the epilepsy and some could be owing to de novo mutations like the majority of cases of Dravet syndrome. On the other hand, the majority of epilepsy patients have what is referred to as “complex epilepsy,” which are common and multigenic. It is believed that genetic and environmental factors interact to various degrees to play a role in the susceptibility to epilepsy. The genetic alterations in these complex epilepsies are being identified using collaborative, multicenter, and multinational projects like the Epilepsy Genome Phenome Program, EPICURE Consortium, EuroEPINOMICS consortium, EMINet Consortium, and the Epi4K consortium. These and similar projects have permitted the identification of specific and recurrent copy number variants (which are deletions or duplications of the genetic information in excess of a kilobase) that occur at genomic hot spots and that increase likelihood of certain epilepsies. These copy number variants tend to occur in genes that are intolerant to mutations, which are also phylogenetically preserved. Continued research shows also that a monogenic etiology could not be established for most of the generalized idiopathic epilepsies, including childhood absence epilepsies and most cases of juvenile myoclonic epilepsies, so susceptibility genes are being investigated. Ongoing studies are also starting to concentrate on whole genome sequencing, because this may identify possible contribution of intronic mutations to the occurrence of epilepsy. All this constitutes a rapidly growing body of knowledge that is compiled in the Online Mendelian Inheritance in Man database (OMIM). To start to address some of these questions, experimental research on animal models of epilepsy has been able to identify the additive effects of genetic mutations to environmental insults and the complex interaction between different genes resulting in different epilepsy phenotypes. The combination of 2 mild alleles of monogenic epilepsy genes in mutant mice, namely in sodium and potassium channel genes, can lead to a much more severe phenotype, whereas in other situations combining 2 genetic mutations that predispose to epilepsy does not necessarily enhance epileptogenicity, but actually ameliorates it. Additionally, it has been demonstrated that mouse strain can influence seizure threshold. Finally, we have recently shown that the presence of an epilepsy predisposing gene mutation (in the potassium channel KCN1a) and of mild hypoxia can result in spontaneous seizures and in predisposition to other consequences when either one of these insults alone was not sufficient to result in either.
The clinical utility of genetic testing in pediatric epilepsy is an important issue. The yield from genetic testing of the usual cases is so low that it is not justifiable to do such testing unless there is family history to suggest it, or if there is drug resistance or developmental delay and an abnormal neurologic examination not explained by other findings. In such situations, genetic testing is justifiable because the yield is relatively high (as high as 34.5%) and there is more of a chance of uncovering potentially treatable genetic conditions, such as vitamin-responsive genetic epilepsies. Genetic testing in patients with intractable epilepsy and developmental delay usually involves, in addition to metabolic testing for inborn errors of metabolism and vitamin dependent conditions, 1 or more of the following: karyotype, chromosomal microarray, specific gene or gene panel sequencing, or whole exome sequencing. This testing often leads to closure of the search for an underlying etiology and helps physicians to better prognosticate and families to better accept and deal with the problem. However, such genetic testing also often raises many other questions, such as dealing with the uncertainty of variants of unknown significance, which is seen in at least one-third of patients tested with gene panels or whole exome sequencing, not to mention the often very high cost of the testing and the need to decide how to deal with possible incidental unrelated findings. In some situations, finding the underlying etiology may help in the choice of therapy, such as in the case of Dravet syndrome, the avoidance of use of valproate in patients with polymerase G mutations to avoid risk of hepatotoxicity, and the diagnosis of glucose transporter deficiency in rare patients with normal cerebrospinal fluid glucose levels, leading to initiation of the ketogenic diet. Such testing can also help to detect and confirm the diagnosis of vitamin-responsive (B 6 , folinic acid, biotin) epilepsies suspected clinically or on metabolic testing. Such testing may also, at times, lead to the investigation of novel therapies ( Fig. 1 ; see Fig. 3 ).
