Uncovering etiologic genes through whole exome sequencing in pediatric epilepsy: A case series from Thailand

Introduction

Patients with DEE represent a group of disorders that are characterized by important clinical features. The most common age range for symptoms is birth to late childhood. This group of diseases may cause severe seizures with or without fever (febrile/afebrile status epilepticus), prompting the treating physician to consider a diagnosis. Uncontrolled seizures, including status epilepticus, can occur with full anticonvulsant treatment and care for underlying conditions. Seizures could affect the brain, further beyond the identified cause, resulting in disrupted learning, slowed or regressed mental development, altered mental state, and abnormal behavior. These effects can lead to impaired intelligence and increased abnormal behavior at any age.¹ An increasing number of studies have shown that developmental and epileptic encephalopathies are interrelated and have long-term pathological effects on each other. An event may occur before, during, or after a seizure. Although seizures can be controlled, brain function may still decline. There is a possibility of experiencing more seizures that may recur later in life, leading to further regression.¹ This group displays specific seizure patterns such as epileptic spasms, myoclonic seizures, generalized tonic seizures, and focal clonic seizures.^2–5 A physical examination of the nervous system may reveal abnormal movement, dysmorphic features, and developmental regression. Some individuals have a family history of epilepsy, or siblings with the same disease progression. Electroencephalogram (EEG) displays characteristics such as a disorganized or diffuse slowing background, focal or multifocal epileptiform discharges, and specific epileptiform discharge patterns such as burst suppression or hypsarrhythmia.⁶ Genetic-related epilepsy should be considered, and diagnosis is necessary to determine the cause based on clinical data.³ Diagnostic tests were available to identify single-gene disorders. On the basis of the patient’s clinical characteristics, such as SCN1A, KCNQ2, and PRRT2 abnormalities were identified. Abnormalities in more than one gene locus, including SCN1A, SCN2A, and SCN1B, are found in some epileptic syndromes, such as Dravet syndrome or Ohtahara (EIEE) syndrome, and are associated with genetic abnormalities in KCNQ2, SCN2A, STXBP1, GABRA1, and CDKL5. However, the same abnormal gene can cause clinical signs of varying severity, such as KCNQ2 in the self-limited familial neonatal-infantile seizure group and SCN1A in generalized epilepsy febrile seizure.^7,8 Previous studies in various countries have revealed genetic abnormalities in one out of every four patients through deoxyribonucleic acid (DNA) sequencing testing.³ Although the remaining patients exhibited symptoms, they did not have the expected genetic abnormalities. Numerous studies have attempted to answer this question over the last ten years. There have been methods to simultaneously test many genes at the same time. This sequence is called the next sequence. Genetic testing has become easier and more efficient with the ability to test for multiple genes at once and the use of whole exome sequencing for diagnosing unknown diseases. The cause of these abnormalities was found to be as high as 30-40%.^9,10 Exon testing for epilepsy panel genes can identify additional causes of epilepsy.

Methods

This research project number MTU-EC-PE-1-141/64 was conducted by the Declaration of Helsinki and approved by a full board review of The Human Research Ethics Committee of Thammasat University (Medicine) on 13 September 2021, and written informed consent was obtained from all the patients’ guardian. We used descriptive cross-sectional statistics to measure clinically detected rates. All ten participants who presented with at least two subsequent symptoms of epilepsy were admitted to Thammasat Hospital for treatment at the Pediatric Neurology Clinic. The age of the volunteers ranged from newborns to 15 years.

The exclusion criteria for volunteers consisted of patients who had been referred for difficult-to-control epilepsy consultation due to brain injury-induced brain tumors and diseases of the nervous system caused by infectious pathogens. All recruited patients or guardians provided written informed consent. The patient’s history, age at seizure onset, clinical characteristics including seizure type, a three-generation pedigree, ASMs, comprehensive physical examination, and dysmorphic features evaluated by the geneticist (K.R.), data electroencephalogram (EEG), and neuroimaging results were recorded if feasible. Blood was collected peripherally from patients and biological parents, if accessible. Genomic DNA was isolated from leukocytes using a Puregene^® blood kit. Exome sequencing was performed using the Illumina HiSeq platform (Macrogen, South Korea). All variant annotations were analyzed using a standard protocol (Burrows-Wheeler Alignment tool [BWA]).¹¹ Then, the variants were filtered by a minor allele frequency (MAF) of >0.01 in the database for single nucleotide polymorphisms in the whole 1000 genome data (phase 3). The gene target analysis approach was used for phenotypes based on the Human Phenotype Ontology (HPO) terms¹² of epilepsy (HP:0001250). The gene list is presented in the data availability. The classification used for genetic variant interpretation was based on recommendations from the American College of Medical Genetics and Genomics, and the Association for Molecular Pathology criteria for variant classification.¹³ In silico predictive programs, determining the novelty of variants and determining the supposed evidence of pathogenicity of identified variants were performed using Franklin, Clinvar,gnomAD, and our Thai database Sanger sequencing was performed to confirm the presence of pathogenic and likely pathogenic variants in the patients and their parents. The polymerase chain reaction was used 10 mMol of dNTP-Mix (GeneON^®, lot number 101.183) volume 0.5 microliter (uL), 10xTaq buffer (Thermo Scientific^®, lot number 00892578) volume 2.5 uL, 5 uM/uL of forward and reverse primer (Macrogen^®) volume 1.5 uL per each (sequence of forward and reverse in Supplementary tablexxx), 25 mM MgCl2 (Thermo Scientific, lot number 00892580) volume 1.5 uL, Taq DNA Polymerase recombination (Thermo Scientific, lot number 2667134) volume 0.2 uL, DNA (50 ng/uL) volume 3 uL, and water upto 25 uL per reaction.

The copy number variants (deletions and duplications) test was additionally performed in patients with severe and uncontrolled epilepsy with negative sequencing results, using the next-generation technique (PacBio sequencing).

Results

Participating in this research are ten volunteers who meet the inclusion criteria. These patients included seven males and three females. Upon enrollment, their ages ranged from 10 months to 10 years and 9 months with a median age of 70 months. Every individual is referred to a pediatric neurology clinic on account of their primary ailments, which consist of seizures that are uncontrollable and have developmental delays. The age of seizure onset ranged from 3 months to 10 years and 9 months, with a median of 11 months. We report the clinical symptoms and main findings of each patient and summarize their clinical characteristics in Table 1.

Genomic study

We conducted Whole Exome Sequencing (WES) on all patients, followed by Sanger sequencing, to confirm the presence of pathogenic and likely pathogenic variants in our patients and their parents, if available. We identified four pathogenic variants: SCN2A (Patient no. 1) (Figure 1a), CHD2 (Patient no. 6) (Figure 1b), SCN1A (Patient no. 7) (Figure 1c), and PCHD19 (Patient no. 9) (Figure 1d). Sadly, undetected pathologic variants were observed in patients no. 2, no. 3, no. 4, no. 5, no. 8, and no. 10. Sanger sequencing results are shown in Figure 1. However, due to epileptic spasms, global developmental delay (GDD), and abnormal EEG in Patient no. 5, we also performed copy number variant (deletions and duplications) testing using the next-generation technique (PacBio sequencing). Unfortunately, the patient’s results were inconclusive, and this case was classified as a variant of uncertain significance.

Figure 1. Sanger positive sequencing.

a: SCN2A (Patient no. 1). b: CHD2 (Patient no. 6). c: SCN1A (Patient no. 7). d: PCHD19 (Patient no. 9).

Discussion

Developmental and epileptic encephalopathy is a clinical condition that involves epilepsy and developmental issues. This is the most common reason for consultations in pediatric neurology clinics. The symptoms include uncontrolled seizures, specific seizure types such as epileptic spasm or myoclonic seizure with concomitant developmental delay or regression, intellectual disabilities (ID), autistic features, aggressive mood, behavioral problems, abnormal movement or gait, and dysmorphic features. However, these symptoms are indicative of many diseases and syndromes. In developing countries, resources are limited, and it may not be possible to investigate all possible diseases simultaneously. Our study aimed to find a way to accurately diagnose and provide specific treatments for patients with DEE. Each participant underwent EEG, brain MRI, or CT, as well as basic metabolic, electrolyte, or thyroid function tests based on their symptoms and signs. However, these investigations yielded no significant findings. A small number of volunteers may not be representative samples for the statistical evaluation of the clinical characteristics of this group of diseases. However, we were able to identify pathological variants in 4 of 10 cases. These patients share the same clinical symptoms as those of delayed development, which may not be normal in the age range before a seizure occurs. If uncontrolled seizures occur, learning abilities are affected. Internally, seizures may be somewhat controlled, but their development may not be the same. SCN1A and PCHD19 variants have a similar clinical spectrum that is linked to both febrile seizures and epilepsy. The onset of febrile seizures occurred during infancy in both groups; however, SCN1A tended to occur earlier. Seizure duration may be prolonged in SCN1A, while PCHD19 usually has shorter durations and occurs in clusters of brief durations. However, the incidence of febrile status epilepticus did not differ between the two conditions.

SCN1A can occur in both males and females, whereas PCHD19 related to epilepsy is mostly found in female patients and rarely in male patients with somatic mosaicism. Approximately one-third of PCHD19 patients may have a normal intelligence quotient (IQ), and those affected may range from mild to severe intellectual disability. In contrast, cognition is more severely affected in SCN1A, and there is a rapid worsening of either cognition or motor skills.^14–16 Both our patients had clinical spectrums identical to those in previous studies.

Sodium channel blockers may cause seizures in some patients with SCN1A. This may be more effective for certain SCN2A variants. Based on a previous study conducted on SCN2A variant subtypes,¹⁷ our patient with de novo SCN2A presented with epileptic spasms that began at the age of 10 months and did not respond to vigabatrin and/or prednisolone treatment. According to the categorization of previous studies, our patient falls under the infantile epileptic encephalopathy onset category, with an onset period of > 3 months. This group is associated with de novo loss-of-function (LOF) missense, protein-truncating, and splice site variants or missense variants with mixed gain-of-function (GOF) and LOF characteristics.^17,18 This might be due to the poor response to sodium channel blockers. However, another study found that in 50% of SCN2A patients in this age group, seizure onset was improved by oxcarbazepine.¹⁹ Our patient seemed to improve slightly after receiving sodium channel blockers (lamotrigine and topiramate). In terms of prognosis, most de novo SCN2A cases exhibit developmental delays.¹⁸ Patients with seizure onset after 11 months of age may be the furthest non-missense variation group and lead to LOF, which exhibits delayed development or autism, and then has epilepsy in late infancy or early childhood.²⁰ Patients with de novo SCN2A variants usually have poor clinical outcomes and lack specific anti-seizure medications.

Some patients may exhibit delayed development and behavioral issues for a long time before they experience seizures. They may not receive a specific diagnosis until they experience seizures in late childhood. Patient no. 6 was born prematurely at 35 weeks, and her family was made aware of the potential developmental delays associated with this condition. Her seizures began at 10 years of age and were accompanied by aggressive behavior. She experienced generalized tonic seizures and brief absence seizures that were difficult to control. After three months of treatment with valproic acid and topiramate, her behavior returned to baseline, and she stopped experiencing seizures. Interestingly, her interictal EEG showed fewer generalized epileptiform activities, but there were still some photo-paroxysmal activities. In a previous study, CHD2 variants were found to be associated with neurodevelopmental disorders. This disorder should be considered in patients with intellectual disability and/or autism spectrum disorders who have drop attacks, myoclonus, atonic-myoclonic-absence seizure, rapid onset of multiple seizure types associated with generalized spike-wave on EEG, and clinical photosensitivity.^21,22 Further studies to better understand genotype and phenotype are necessary for prognosis, specific medication, and genetic counseling. All patients were informed of the gene results by a medical geneticist (Table 2). They followed up and re-evaluated their seizures and other medical problems at the pediatric neurology clinic. Our team provides developmental pediatric consultation, all aspects of physical medicine, and rehabilitation care by physicians and physical therapists.

Conclusion

According to this study, the detection rate of pathologic variants was similar to that in the previous WES study, standing at 4 out of 10 (40%). All patients in this study had difficult-to-treat epilepsy and required two or more anti-seizure medications. The most common type of seizure observed was generalized seizures. In this type of epilepsy, specific patterns, such as epileptic spasms, atypical absences, and mixed-type seizures, should be considered. Furthermore, EEG patterns such as hypsarrhythmia, burst suppression, generalized epileptiform, and photosensitivity epilepsy should be evaluated. However, neuroimaging findings initially appeared unremarkable in this study. Family history remains an important gathering, and a normal family history does not exclude the possibility of developmental and epileptic encephalopathy. Those who exhibited the same clinical symptoms, but had not been diagnosed with any abnormalities. Consequently, the low detection rate of WES for pathogenic variations is due to its inability to detect non-coding areas, minor deletions or duplications, and repeat expansions.²³ Thus, further examinations using other methods should be considered.

Author contributions

KK designed the conceptualization, project administration, and funding acquisition; wrote the original draft; and reviewed and edited the manuscript. SP. prepares and performs genetic testing under supervision of KR. KR. interpreted the genetic tests, and reviewed and edited the manuscript. All authors contributed to and approved the manuscript.

Data availability

Accession numbers

Additional data is available in the Clinvar database, which can be accessed through the following link: https://www.ncbi.nlm.nih.gov/clinvar/.

The relevant accession numbers are detailed as follows:

ClinVar: patient no.1 SCV004848932 NM_001040142.2(SCN2A):c.719C>T (p.Ala240Val); Accession number; VCV000449164.4 http://identifiers.org/clinvar:449164

ClinVar: patient no. 6 SCV004848930 NM_001271.4(CHD2):c.1719+1G>A; Accession number; VCV003076079.1 http://identifiers.org/clinvar:3076079

ClinVar: patient no. 7 SCV004848931 NM_001165963.4(SCN1A):c.955C>T (p.Gln319Ter); Accession number; VCV003076079.1 http://identifiers.org/clinvar:3076080

ClinVar: patient no. 9 SCV004848933 NM_001184880.2(PCDH19):c.1019A>G (p.Asn340Ser); Accession number; VCV000206364.34 http://identifiers.org/clinvar:206364