Precision in pediatric epilepsy

Introduction

Epilepsy is one of the most common medical problems that affect infants and children. An estimated 5% of children will have a seizure¹. A recent meta-analysis study found that the active period prevalence of epilepsy (the number of annual new and existing cases of epilepsy) was 4.8/1000 worldwide and in the USA was 0.94/1000 for children less than 18 years of age². In the same study, the lifetime prevalence of epilepsy (the number of active and existing cases of epilepsy between birth and the time of assessment) for children less than 18 years of age was 7.2/1000². Epilepsy and the conditions that cause epilepsy impact children and their families in many different ways, impacting cognition, behavior, and socioeconomic status. Uncontrolled epilepsy and the conditions that cause seizures leave indelible changes that can affect a child for life and even increase the risk of sudden death. We have had a limited understanding of the pathophysiology of epilepsy and limited tools to diagnose epilepsy and its cause. In recent years, however, the diagnosis of the cause of epilepsy in infants and children has dramatically improved. The classification of epilepsy has changed; we no longer use the same terminology to describe seizures and epilepsy. New methods to identify the cause of seizures and epilepsy are now available; as a result, we have found new ways to treat seizures and epilepsy. We have new tools to distinguish types of epilepsy and to localize the seizure focus in focal epilepsy. With a better understanding of epilepsy pathophysiology and new ways to diagnose and treat epilepsy, the International League Against Epilepsy (ILAE) has developed a new classification system³. Epilepsy in infants and children has entered the age of precision medicine. We are closing in on the goal to identify and treat the cause of epilepsy rather than the symptoms. How has our ability to identify an epilepsy etiology changed? What new tests are available for making diagnoses in each of the etiological categories in the ILAE classification? How do you use these tools? This review will answer some of these questions.

Changes in epilepsy classification

The ILAE recently reclassified seizures and epilepsies. The classification still depends primarily on clinical and electroencephalography (EEG) features but with modest changes in some of the terminology³. The term focal onset replaces the term partial onset, and focal onset seizures are placed into two categories: aware and impaired awareness. Focal seizures that become generalized seizures are now referred to as focal to bilateral tonic-clonic instead of partial with secondary generalization. Many of the descriptions of the motor and non-motor activity that occurs during the seizure remain the same. The most significant change has been the move toward classifying epilepsies according to epilepsy etiology and a better definition of epilepsy syndromes. Figure 1 lists the new classification system and the six etiological categories in the ILAE classification. This change recognizes some of the important advances in defining etiologies for epilepsy, particularly in children and infants with seizures.

Figure 1. Classification of seizures and epilepsy.

Epilepsy: At least two unprovoked (or reflex) seizures occurring more than 24 hours apart or one unprovoked or reflex seizure and an at least 60% chance of recurrent seizures over the next 10 years or the diagnosis of an epilepsy syndrome.

The etiological classification has six categories (Figure 1). As Falco-Walter et al. noted in describing the changes in the ILAE classification of seizures and epilepsy, “from first contact with the patient, the clinician is encouraged to consider the etiology of their seizures”³. This is an important move, as epileptologists have long known that establishing a specific etiology was important for optimizing treatment and that defining epilepsy syndromes was an important first step in the process of defining specific etiologies and treatments. The advances in basic and clinical sciences have provided many new diagnostic avenues and tools that allow even greater diagnostic and therapeutic precision. This can be compared with the importance of etiology in infectious disease.

Epilepsy diagnosis

A recent population-based study from Australia provides important new data regarding the etiology of severe epilepsies of infancy⁴. In this study, the authors analyzed records of infants whose seizures began before age 18 months and were frequent, associated with an epileptiform EEG, and failed to respond to two or more antiepileptic drugs. The authors were able to identify 114 cases that were born during 2011–2013 and found that 14% had acquired brain insults, 54% had genetic/presumed genetic etiologies, and 33% had unknown etiologies. Of the 54% (62) that had genetic/presumed genetic etiologies, 31 had structural abnormalities (usually dysplasias or cortical migrational defects), six had metabolic etiologies, nine had chromosomal etiologies, and 16 had single-gene etiologies. This study shows the importance of the subsequent advances in diagnosis that we will describe. The study also highlights how whole exome sequencing (WES) has changed the financial value of genetic testing, increasing diagnostic yield at a lower cost.

Precision in structural causes of epilepsy and surgical management

Advances in precision medicine: advances in metabolic and genetic testing

Neurology and genetics have made amazing progress in identifying the etiology of epilepsy in infants and children. We will describe some of the changes in metabolic and genetic testing and provide some guidelines for their use.

Metabolic tests

Metabolic testing for diseases that cause epilepsy has greatly expanded in the last few years, and new methods for testing many metabolites are now available. Table 2 provides a list of the metabolic tests for diseases that cause epilepsy in pediatric patients. Many of the tests listed in the table are panels organized by the type of metabolic disease. These panels include tests for different metabolic diseases in the same metabolic pathway. Different medical institutions and companies provide these tests. An internet search using the terms in Table 2 will provide a list of the companies and institutions that offer the tests, the sample requirements, the cost of the test, and how to send the samples. The metabolic tests in Table 2 are only screening tests, and confirmation of a disease would require additional testing. Interpretation of the tests may require help from an expert in metabolic diseases. The list is very extensive but the choice of tests usually can be reduced by considering the patient’s clinical and laboratory features.

The tests in Table 2 are biochemical tests for specific diseases in specific pathways. Metabolomics/metabonomics use “quantitative measurement of dynamic multiparametrics metabolic response of living systems to pathophysiological stimuli or genetic modification”¹⁹, an approach pioneered by Nicholson et al.²⁰. Metabolomics analyzes the metabolome, organic substances occurring from metabolism in the body, and xenobiotics, foreign substances that are not naturally found in an organism. Metabolomics is closer to the phenotype of an organism than genomics, transcriptomics, or proteomics. Metabolomics complements genomics. Metabolomics is a relatively new method that analyzes plasma, urine, saliva, cerebrospinal fluid, and other samples by using mass spectrometry (metabonomics) or nuclear magnetic resonance spectroscopy (metabonomics) or both to quantify, for clinical diagnostic purposes, many metabolites. This testing is only for small molecules, so metabolomic testing does not test for large-molecule diseases like lysosomal storage diseases, mucopolysaccharidosis, mucolipidosis, or congenital disorders of glycosylation. Metabolomic testing is available through several laboratories. Interpretation of the results requires a skilled biochemist. Currently, we do not know with certainty whether metabolomics will play a larger role in the diagnoses of diseases causing epilepsy in pediatric patients, but the field is rapidly growing and has the same attraction as WES or whole genome sequencing (WGS) does. Currently, metabolomics is used when other diagnostic testing has failed to identify an etiology and to confirm an abnormal finding in metabolic testing or to help determine whether a variant of unknown significance in genetic testing is significant.

Genetic testing

Genetic abnormalities reportedly cause or influence more than 70% of epileptic conditions (that is, conditions in which epilepsy presents as a core or associated symptom)²¹. Determining whether a gene mutation causes epilepsy is complicated because single-gene mutations can cause variable phenotypes and different gene mutations cause the same phenotype. As genetic epilepsy syndromes tend to manifest in infancy or childhood, genetic testing has become an integral part of pediatric epilepsy work-ups. The evolution of diagnostic techniques due to massively parallel sequencing has allowed a rapid evolution from narrowly applicable tools (for example, fluorescence in situ hybridization [FISH] and single-gene testing) to multigene panels, clinical exome sequencing, clinical genome sequencing, and chromosomal microarrays for testing infants and children with epilepsy. Table 3 lists the types of genetic testing for genetic causes of epilepsy and their advantages and disadvantages.

Next-generation sequencing (NGS) is expanding the list of epilepsy-related genes as well as uncovering variant patterns that potentially could predict pathogenicity, disease pathophysiology and evolution, and therapeutic targeting. Table 4 shows the results from studies involving pediatric patients with epilepsy. Most of the studies involve epilepsy gene panels or WES sometimes in combination with an epilepsy gene panel. Because of the expense, only one of the studies used WGS. The number of genes in the epilepsy panels is rapidly increasing, and an epilepsy panel that includes mitochondrial gene testing and that covers a total of 553 genes is now available from MNG Laboratories (Atlanta, GA, USA)²². The cost of whole exome and whole genome testing is decreasing rapidly enough that these genetic tests may be the next test after imaging. As Table 4 shows, genetic testing has the highest yield in pediatric patients, especially infants with epilepsy syndromes or infant epileptic encephalopathies.

In addition to NGS, epigenetic biomarkers are starting to play a role in enriching diagnostic precision in genetic and non-genetic pediatric epilepsies. Epigenetic factors are vast, and DNA methylation, histone modification, and non-coding RNA are some of the most impactful in pediatric epilepsy. Such factors provide a more detailed diagnostic value, as evidence of their presence can clarify disease manifestations and evolution. A prime example is the possibility of using DNA methylation as a diagnostic epigenetic biomarker in TLE, a non-genetic condition, after recent studies have shown DNA methylation to be widespread in TLE. DNA methylation as an epigenetic biomarker can also enhance disease progression and prognostic predictions for many genetic epilepsies (for example, Dravet, benign familial neonatal seizures, and epileptic neurodevelopmental disorders, or NDDs) as they regulate the associated disease and disease-modifying gene expression and functioning (for example, KCNQ3, SCN3A, and GABRB2, respectively).

In choosing metabolic or genetic tests, the clinician has many challenges that include deciding the sequence of testing, choosing the type of screening test, considering the cost of the tests, and comparing the quality of the laboratory tests. We have discussed some of these issues above. Confirming that a patient has epilepsy by the clinical history and an EEG must be the first step followed in most cases by imaging to uncover structural causes of epilepsy. The subsequent sequence of biochemical or genetic testing to define an etiology is changing. Finding a treatable cause of epilepsy should always accompany trying to control the seizures and in some cases control of the seizures will only be possible when you find the cause of the epilepsy. A pediatric patient with epilepsy should receive a therapeutic trial with pyridoxine, folinic acid, and biotin even if you have done biochemical and genetic testing. Table 5 lists treatment and gene mutations for some of the treatable causes of epilepsy. Early diagnosis and treatment of these disorders are essential to improve long-term outcome.

The choice of testing is highly dependent on how well a pediatric patient fits an epilepsy syndrome or known genetic disorder. As noted by Myers et al., recognizable syndromes like tuberous sclerosis may not require genetic testing unless the testing is for genetic or prognosis⁴³. In some cases, like Rett syndrome, the symptoms and signs would help with the diagnosis but would not be enough to confirm the diagnosis. Genetic testing could be limited to testing for an MECP2 gene mutation and, if this were negative, could be expanded to a Rett-like gene panel⁴³. In the case of Dravet syndrome, a gene panel that includes the SCN1A gene should be the first genetic testing given that single-gene tests may be as expensive as gene panels or more so. The cost of genetic testing sometimes may limit the choices. Interpretation of the test results when using comprehensive genomic testing can be challenging and often requires the help of a geneticist or genetic councilor or both²³. However, a recent cost analysis reported by Howell et al.⁴ found that, in a population-based study of severe epilepsies of infancy, the early use of WES omitting some metabolic testing provided the highest diagnostic yield for the lowest cost. This study strongly suggests that the improvement in diagnostic precision in a patient with epilepsy increasingly calls for the use of gene panels, exome sequencing, or genome sequencing. The exome and genome testing improves when done as a trio, testing the parents of the patient along with the patient, but this sometimes adds cost. Several recent studies have shown that exome and genome sequencing tests significantly improve diagnosis in patients with epilepsy^4,43,44.

Advances in immune disease testing

In the last decade, the list of autoimmune neurological diseases has rapidly expanded. Many of them cause seizures. Table 6 provides a list autoimmune encephalitides and the antigen targets for the autoantibodies that cause seizures. Autoimmunity as a cause for seizures and epilepsy should be considered in pediatric patients who were previously normal and then present with acute or subacute onset of seizures and signs of infection, additional behavioral symptoms, or movement disorders. Brain imaging abnormalities and an abnormal EEG can also support the diagnosis of presumed autoimmune encephalitis^45,46. Antibody testing for autoimmune encephalitis is available, but prompt immunomodulatory treatment (corticosteroid and intravenous immunoglobulin, plasma exchange, and rituximab) and testing for paraneoplastic tumors should come before the results are available^45,46. Prompt treatment can be very effective.

End

We have described many of the advances in testing to find the cause of epilepsy in a pediatric patient. As the use of NGS continues, so does our ability to capture epileptic significance in many genes that had either no prior seizure association or limited seizure association. A prime example is the substantiation of GABRB2, SNAP25, CACNA1E, and KCNQ3 as epileptic NDD-related genes (KCNQ3 has prior association with benign familial neonatal seizures⁴⁴). This in turn will not only help tailor and increase diagnostic rates for genetic panels but also will help our understanding of the complex pathogenesis underlying genetically mediated epilepsy.

The use of the new diagnostic tools begins with a good history and physical examination and that may provide most of the information needed to find the cause of epilepsy in a child. The history and examination are also essential for choosing tests and interpreting test results. An infant with a history of a perinatal stroke confirmed by MRI imaging and clinical focal seizures, confirmed by an EEG has a clear etiology for focal onset epilepsy, a stroke, and does not require further laboratory testing to find the cause of the seizures. Many infants and children, however, do not have a history or physical examination features to allow a specific diagnosis and these are often the patients with intractable epilepsy, who (as noted above) most benefit from the advances in laboratory diagnosis. The laboratory testing that is now available has dramatically improved our ability to make specific diagnoses. The improvements have been in more and better methods for identifying metabolic abnormalities and genetic abnormalities. The explosion of new tests presents a challenge for clinicians caring for infants and children with epilepsy. Clinicians have to make difficult choices in choosing the right tests, interpreting the results, and using the information to provide earlier and better treatment. These decisions include considering the cost of tests, and at least one study suggests that in early-onset epileptic encephalopathies or drug-resistant epilepsy, genetic testing should be used early in the course of diagnosis.

Improved methods to identify structural, metabolic, genetic, and autoimmune diseases affecting the nervous system have also given us new tools to treat epilepsy. The disorders listed in Table 5 are only the start of precision treatment for epilepsy. The precise definition of the cause of epilepsies opens the door for new drugs, protein replacement, gene therapy, and many other promising treatments for the most devastating epilepsies that affect infants and children. Improvements in localization of developmental/structural changes that cause focal epilepsy are also on the horizon. We are at one of the most exciting times in epilepsy, truly a time of precision medicine.

Abbreviations

DKI, diffusion kurtosis imaging; DTI, diffusion tensor imaging; EEG, electroencephalography; FCD, focal cortical dysplasia; fMRI, functional magnetic resonance imaging; HD-EEG, high-density electroencephalography; ILAE, International League Against Epilepsy; MECP2, methyl-CpG binding protein 2; MEG, magnetoencephalography; MRI, magnetic resonance imaging; NDD, neurodevelopmental disorder; NGS, next-generation sequencing; NODDI, neurite orientation dispersion and density imaging; SCN1A, sodium voltage-gated channel alpha subunit 1; SEEG, stereoelectroencephalography; TLE, temporal lobe epilepsy; VBM, voxel-based morphometry; WES, whole exome sequencing; WGS, whole genome sequencing