Abstracts

Rationale:
Despite growing recognition of the important role of genetics in pediatric epilepsy, many children with unexplained epilepsy do not obtain genetic diagnoses through routine neurological care. We seek to demonstrate the high yield and utility of trio whole exome sequencing (WES) in children with unexplained epilepsy through a prospectively enrolling, single-center research study that combines phenotypic and genomic expertise to deliver clinically returned genetic results.
Method:
We enrolled children with unexplained epilepsy at Boston Children’s Hospital, and their family members, in the Children’s Rare Disease Cohorts Initiative (CRDCI). We recruited patients seen by neurologists at our institution, who were made aware that clinically relevant results would be confirmed and returned if parents opted in for results return. WES data were generated for probands and parents and uniformly processed through a standard alignment and variant calling pipeline. Detailed medical history, including seizure semiology, age of onset, family history, EEG and MRI findings, were reviewed by at least 2 epileptologists; seizure type, epilepsy type, and epilepsy syndrome were classified according to the 2017 ILAE Seizure and Epilepsy classification. Mendelian inheritance analysis and variant evaluation for pathogenicity prediction were conducted using in silico tools. We identified de novo, homozygous, or compound heterozygous variants that were rare (allele frequency < .0001 and < .000001 for mono- and bi-allelic variants, respectively). ACMG criteria were annotated to each variant. Additionally, we incorporated each patient’s phenotypic information.
Results:
To date, we have recruited 487 children with epilepsy and thus far have DNA samples from 284 trios, 169 duos, and 34 probands only. Analyzing 303 cases, we have identified clinically significant variants for 52 children with unexplained epilepsy with a range of phenotypes, including generalized epilepsies and focal epilepsies, with and without intellectual disability (Fig. 1). We identified 35 missense or splice-affecting de novo variants in following genes associated with autosomal dominant conditions, 21 biallelic variants and 2 X-linked variants. Candidate findings in an additional 128 cases require further confirmation and additional case identification before clinical return of results will be pursued. Our overall yield of diagnostic variants that could be clinically confirmed is 17%, and we expect this to increase with on-going iterative re-analysis that incorporates recognition of new epilepsy genes. The average age at the time of genetic diagnosis was 10 compared to 3 years old, the average age of onset, and the reasons noted for the long delay between epilepsy diagnoses and genetic diagnoses were observed to be lack of clinician suspicion for genetic etiology and lack of insurance coverage or personnel resources to undertake clinical testing.
Conclusion:
We demonstrate that an institutional platform to phenotype and sequence children with unexplained epilepsy allows for the discovery and clinical return of results to many families who would otherwise continue without clinical genetic diagnoses. In this era of emerging rational and precision treatments based on genetic diagnoses, our results emphasize the need to pursue genetic evaluation for children with unexplained epilepsy. In our case, an institutional effort allowed us to narrow the gap between research and clinical sequencing. Future efforts should include the development of systematic testing guidelines for subpopulations of children with epilepsy and professional efforts to advocate for genetic testing in patients with unexplained epilepsy.
Funding:
:BCH Children's Rare Disease Cohorts Initiative