Broadening the scope of multigene panel analysis for adult epilepsy patients

Abstract

Objective: Epilepsy is a suitable target for gene panel sequencing because a considerable portion of epilepsy is now explained by genetic components, especially in syndromic cases. However, previous gene panel studies on epilepsy have mostly focused on pediatric patients.

Methods: We enrolled adult epilepsy patients meeting any of the following criteria: family history of epilepsy, seizure onset age ≤ 19 years, neuronal migration disorder, and seizure freedom not achieved by dual anti-seizure medications. We sequenced the exonic regions of 211 epilepsy genes in these patients. To confirm the pathogenicity of a novel MTOR truncating variant, we electroporated vectors with different MTOR variants into developing mouse brains.

Results: A total of 92 probands and 4 affected relatives were tested, and the proportion of intellectual disability (ID) and/or developmental disability (DD) was 21.7%. As a result, twelve probands (13.0%) had pathogenic or likely pathogenic variants in the following genes or regions: DEPDC5, 15q12-q13 duplication (n = 2), SLC6A1, SYNGAP1, EEF1A2, LGI1, MTOR, KCNQ2, MEF2C, and TSC1 (n = 1). We confirmed the functional impact of a novel truncating mutation in the MTOR gene (c.7570C > T, p.Gln2524Ter) that disrupted neuronal migration in a mouse model. The diagnostic yield was higher in patients with ID/DD or childhood-onset seizures. We also identified additional candidate variants in 20 patients that could be reassessed by further studies.

Significance: Our findings underscore the clinical utility of gene panel sequencing in adult epilepsy patients suspected of having genetic etiology, especially those with ID/DD or early-onset seizures. Gene panel sequencing could not only lead to genetic diagnosis in a substantial portion of adult epilepsy patients but also inform more precise therapeutic decisions based on their genetic background.

Plain language summary: This study demonstrated the effectiveness of gene panel sequencing in adults with epilepsy, revealing pathogenic or likely pathogenic variants in 13.0% of patients. Higher diagnostic yields were observed in those with neurodevelopmental disorders or childhood-onset seizures. Additionally, we have shown that expanding genetic studies into adult patients would uncover new types of pathogenic variants for epilepsy, contributing to the advancement of precision medicine for individuals with epilepsy. In conclusion, our results highlight the practical value of employing gene panel sequencing in adult epilepsy patients, particularly when genetic etiology is clinically suspected.

Keywords: adult epilepsy; gene panel sequencing; next‐generation sequencing; precision medicine.

FIGURE 1

Pathogenic or likely pathogenic variants identified in our patients. (A) Proportion of pathogenic (P) or likely pathogenic (LP) variants and deleterious variants of unknown significance (dVUS). (B) Increased coverage depth in the 15q12‐q13 region for EP88 and EP91, calculated as log₂(fold change) in comparison with the average coverage depth of study samples. Gray lines in the background indicate the coverage depth ratios of other patients in the 15q12‐q13 region.

FIGURE 2

Diagnostic yields differing by clinical subgroups. (A) Proportions of genetically diagnosed patients for different clinical conditions. (B) Diagnostic yields for clinical subgroups subdivided by onset age. The clinical subgroups were determined by seizure type, family history, and intellectual disability (ID)/developmental disability (DD). ^†This compares patients with and without ID/DD among those with seizure onset at ≤10 years.

FIGURE 3

Pathogenicity confirmation of a novel nonsense mutation of MTOR. (A) In ClinVar, most of the pathogenic or likely pathogenic variants in MTOR are missense variants, except one frameshift variant located in the N‐terminal region. In contrast, EP27 had a nonsense variant (p.Gln2524Ter) in the C‐terminal region of MTOR. (B) Representative images of E18.5 mouse brain slices electroporated at E14.5 with vectors containing each MTOR variant. Neuronal migration was characterized by Ctip2 staining and quantified by the percentages of GFP⁺ cells in each layer (vector, n = 3; MTOR ^WT, n = 3; MTOR ^Q2524*, n = 6; MTOR ^C1483Y, n = 3; MTOR ^KD, n = 4; MTOR ^{Q2524* KD}, n = 6; by Kruskal–Wallis test with Dunn's multiple comparison test). The error bar represents the mean ± SD. p < 0.05 (*), p < 0.01 (**), p < 0.001 (***); scale bars, 50 μm; KD, kinase‐dead; loCP, lower cortical plate; mCP, middle cortical plate; uCP, upper cortical plate.