Early childhood is one of the periods of life in which the risk to develop epilepsy is highest. Besides, genetic causes are much more common in the young. Recently, an ever-increasing amount of genes has been found to be involved in numerous early-onset epilepsies. Thanks to next-generation sequencing (NGS), a diagnosis can now be reached in close to 50% of children with epilepsy and developmental delay. This, in turn, has led to the successful application of the concept of individualized treatment in a growing number of children with epilepsy. Genetic investigations have thus been progressively included in the routine work-up of children with early-onset epilepsies throughout the world, mostly in high-income countries up to now. As a result of a scientific collaboration between pediatric neurology divisions at University Hospitals Geneva (HUG), Switzerland, and Children's Hospital 2 in Ho Chi Minh City (HCMC), Viet Nam, genetic testing of children with early-onset epilepsies followed at the pediatric neurology division, Children's Hospital 2 started at the genetics laboratory of the Vietnam National University in 2017.

Aims: Our project aims at establishing the proportion of patients in whom a causal genetic finding can be identified, in a prospective cohort of children with Developmental and Epileptic Encephalopathies (DEE) followed at Children's Hospital 2 (ND2). The investigators also aim at identifying the percentage of these children in whom this approach would change current management.

Methods: A series of children diagnosed with DEE and followed at ND2 Hospital, enrolled consecutively. Exome sequencing was applied to all, with biostatistical analyses of a panel of 671 genes involved in epilepsies and developmental disorders performed in parallel at Ho Chi Minh City Vietnam National University and Geneva Genetic Medicine Division. Sanger sequencing confirmation of potentially causal variants in patients, and in parents for familial segregation. Comparison of Vietnamese and Swiss genetic findings, and multidisciplinary discussions in formal Genome Boards. Additional genetic investigations, if deemed necessary in Genome Board sessions. Clinical management adapted to genetic findings wherever applicable, and follow-up according to standard practice. One-hundred-and-fifty patients are expected to participate during the 3-year study period.

With this prospective cohort study, the investigators wish to know the proportion of children with developmental and epileptic encephalopathies followed at the largest pediatric neurology facility in South Vietnam and enrolled consecutively for whom advanced genetic investigations allow the identification of pathogenic gene variants. The investigators expect this proportion to be 50%.

The investigators will also investigate the proportion of children in whom the genetic results will allow individualized treatment adaptations and global management changes. The investigators expect this proportion to be at least 10 %.

This collaborative project is the first step to demonstrating the clinical utility of high-throughput genetic procedures for all patients with severe early-onset epilepsy and developmental difficulties in Viet Nam. Parallel analyses of patients' exome sequences in Vietnam and Switzerland will allow the optimization of the procedure.

Patient selection:

Based on a birth rate of 16.745 children/1000 people in Vietnam (2018 statistics, data.worldbank.org, consulted Dec 18th, 2020), a mortality rate of 15.9/1'000 live births in Vietnam (2019 statistics, data.worldbank.org, consulted Dec 18th, 2020), a total of 1'394'401 live births in the entire country, and a nationwide population of 97'752'966 of which 8'602'317 live in HCMC (2020 statistics, worldometers.info, worldpopulationreview.org, consulted Dec 18th, 2020), the investigators estimate that 122'708 to 144'046 children are born in the city each year, of which 120'756 to 141'756 will stay alive. On the basis of a recently calculated general incidence of 54/100'000 live births in the State of Victoria, Australia, and a comparable age-adjusted incidence of 44.8/100'000 persons in Ba Vi, a rural district of Vietnam, the investigators estimate that 54-75 children with SEI are born every year in HCMC. Given the fact that a majority of these children are followed at ND2, and that many patients with severe neurological conditions coming from a broader area also consult at ND2, our aim is to enroll 50 children per year.

Inclusion criteria: Consecutive patients, treated at ND2 Hospital during the study period, will be asked to take part in the study and will be requested to sign a consent form if they fulfill all of the following inclusion criteria:

1. Drug-resistant epilepsy, either from the onset or at follow-up (according to ILAE 2010 criteria)
2. Severe developmental delay or regression, according to pediatric neurologist evaluation (or developmental quotient <50, if formally assessed)
3. Age of onset of principal symptoms (Seizures, Developmental delay): 0-36 months
4. Agreement to take part in the study and signed consent form All of the pre-selected patients will be discussed in monthly teleconferences between Nhi Dong 2 hospital and Geneva University hospitals before blood samples are taken. Blood samples will be taken from the patients and their both parents, when available.

Exclusion criteria: Patients with identified structural, infectious, or inflammatory causes (such as perinatal hypoxic-ischemic encephalopathy, Cerebrovascular disorders, Sequelae of trauma or encephalitis, neurocutaneous disorders, etc…) will not be asked to take part in the study.

Genetic investigations:

Based on what precedes, the investigators intend to start our genetic investigations by Whole Exome Sequencing (WES), and keep karyotyping and CGH-array for cases in which no causal variant is found. WES is a multistep procedure that includes DNA extraction, exome sequencing, processing and mapping, variant calling, annotation, and prioritization, as described below. In case of negative WES results, the utility of performing further tests will be evaluated in the planned Genome Board (GB) sessions.

Genomic DNA Extraction and Exome Sequencing In brief, for each participant and both his/her parents, 2-5 ml of whole blood will be sampled and stored in tubes coated with K2/K3-Ethylenediaminetetraacetic acid (EDTA) anticoagulants. Genomic DNA will be extracted using a DNA Mini Blood Isolation kit (QIAGEN DmbH, Netherlands). DNA quantity will be measured by Quant-iTTM PicoGreenTM dsDNA Assay Kit (Invitrogen), an ultra-sensitive fluorescent nucleic acid stain method for quantitating double-stranded DNA, using Victor 3 fluorometry, and the condition of DNA will be assessed by 1% gel electrophoresis to ensure the quality and quantity of DNA samples before the sequencing step.

Whole Exome Sequencing targets all protein-coding regions and accounts for 85% of highly penetrant mutations in the human genome. Whole Exome Sequencing will be performed in a CLIA-certified laboratory at Macrogen Ltd., Seoul, Korea. The constructed libraries will be sequenced using the Agilent SureSelect V5 exome capture kit (Agilent Technologies, Santa Clara, California, USA), which utilizes 120-mer biotinylated cRNA baits to capture and enrich exome regions (357,999 exons in total), on a NovaSeq 6000 sequencing system (Illumina, San Diego, CA, USA).

DNA sequences obtained by WES (FASTQ files) will be analyzed "in silico" in parallel, but blinded to each other, in Ho Chi Minh City and in Geneva, as will decisions for additional, focused analyses based on the phenotypic description (e.g., Sanger sequencing of some poorly covered exons in specific genes). To that end, coded aliquots of patients and parental DNAs will be transferred to Geneva.

Raw data processing and mapping Raw data will be analyzed in Ho Chi Minh City using an in-house bioinformatics pipeline, and independently in Geneva. The quality of FASTQ raw data will be ensured by Phred-score assigning, GC content, read length, read quality, and sequence duplication level. Preprocessing step using Trimmomatic will be applied to trim out any low-quality reads and adapter contaminant sequences. The algorithms by BWA-MEM, PICARD, and GATK will be used for mapping and additional post-alignment processing, including removing duplicates and base quality recalibration to mark or remove any artifacts.

Variant calling and annotation The refined mapping data will be then further processed for variant calling and annotation by GATK Haplotype Caller and ANNOVAR. Functional prediction and annotation of all potential candidates will be then carried out by dbNSFP including 29 algorithms, 9 conservation scores, and related information observed in many population databases such as 1000 Genomes, ExAC, gnomAD, and the NHLBI Exome sequencing project ESP6500 Variant prioritization The suspected etiological variants filtered by a targeted epilepsy panel of 671 genes will be thoroughly analyzed according to the 2015 recommendations of the American College of Medical Genetics and Genomics (ACMG). Sanger sequencing will also be used to validate the family segregation of all detected candidate variants, as recommended. The variants are classified into five categories: 1) benign, 2) likely benign, 3) uncertain significance, 4) likely pathogenic, and 5) pathogenic. The variants with minor allele frequencies >0.05 in any of these databases: dbSNP, 1000 Genomes Project, ExAC, and NHLBI Exome Sequencing Project (ESP6500) will be filtered out. The analyses performed will include computational analysis of DANN, Genomic Evolutionary Rate Profiling++, Likelihood Ratio Test, Mutation Taster, Functional Analysis through Hidden Markov Models-MKL coding, Combined annotation-dependent depletion, and EIGEN score.

The massive sequence data, limited known disease mechanisms, and genetic heterogeneity contribute to the complexity of many conditions, such as DEEs. Thus, conclusive variant interpretation will be carried out in multidisciplinary discussions among geneticists, clinicians, laboratory technicians, and biostatisticians during regular formal genome board (GB) sessions, throughout the study period.

GBs are multidisciplinary sessions hosted by the HUG Center of Genomic Medicine (CGEM), where files of patients are reviewed, who so far eluded diagnosis. Pediatric neurology GBs put together neuro pediatricians, pediatric radiologists, clinical geneticists, biologists and bioinformaticians from the Genetic Medicine division, as well as other specialists as requested by the cases, e.g. adult neurologists, laboratory medicine specialists, immunologists, or other specialists. Together, they review the patient's genetic variants in genes that might explain the disease, and either reject them or propose to order additional, extremely focused investigations for a posteriori validation of the phenotype. If no diagnosis can be reached, the patient is invited to come back to the clinic for a new GB 18-24 months later. Our in-house series shows that, since being implemented in February 2019, GBs have allowed the HUG to increase its diagnostics rate of DEE from approximately 45% to 55%. GBs take place in a videoconference room equipped for the safe and confidential exchange of patient data under medical secrecy over long distances. Genetic data of the patients are discussed but not copied nor transferred from one center to another. Data consist of a list of genetic variants in different genes of the targeted WES, with annotations from public and in-house databases (e.g., GnomAD) along with computations from bioinformatics tools (e.g., CADD score). Typically, the neuro pediatrician will present the clinical data of the patient, the pediatric radiologist will show and comment on the MRI, the geneticist will present the clinically relevant genes that carry candidate variants, and the molecular diagnostics laboratory team of biologists and bioinformaticians will qualify the variants as pathogenic, likely pathogenic, or of unknown significance. In the latter case, the team will review the literature, during the GB, to decide if an additional test might definitely confirm a putative diagnosis, i.e., confirm the phenotype caused by the gene carrying a variant so far qualified as of unknown significance.

The investigators have expertise in next-generation sequencing approaches to support clinical diagnostic of genetic diseases since 2012 and supplies bioinformatics platforms and other translational SNP screening based on genome-wide association studies. The team is strongly interested in contributing Vietnamese genomic rare disease database.

Inclusion Criteria:

• Drug-resistant epilepsy, either from onset, or at follow-up, according to ILAE 2010 criteria
• Severe developmental delay or regression, according to paediatric neurologist evaluation (or developmental quotient <50, if formally assessed)
• Age of onset of principal symptoms (Seizures, Developmental delay): 0-36 months
• Agreement to take part in the study and signed consent form

Exclusion Criteria:

• Patients with identified structural, infectious or inflammatory causes (such as perinatal hypoxic-ischemic encephalopathy, Cerebrovascular disorders, Sequelae of trauma or encephalitis, neurocutaneous disorders, etc…) will not be asked to take part to the study.

Based on a birth rate of 16.745 children/1000 people in Vietnam (VN), a mortality rate of 15.9/1'000 live births in VN, a total of 1'394'401 live births in VN, and a nationwide population of 97'752'966 which 8'602'317 live in Ho Chi Minh City (HCMC), we estimate that 122'708 to 144'046 children are born in the city each year, of which 120'756 to 141'756 will stay alive. On the basis of a recently calculated general incidence of 54/100'000 live births in Australia, and a comparable age-adjusted incidence of 44.8/100'000 persons in Ba Vi, a rural district of VN, we estimate that 54-75 children with DEE are born every year in HCMC. Given the fact that a majority of these children are followed at Nhi Dong 2 hospital, our aim is to enroll 50 children per year.