Deep genome sequencing identifies additional candidate variants in epileptogenic brain lesions

Malformations of cortical development and low-grade epilepsy-associated tumors can cause drug-resistant lesional focal epilepsy. In a study published in Epilepsia, researchers performed deep whole genome sequencing at greater than 300× coverage on resected brain lesions from 33 individuals. Lesions were selected for a high prior probability of a genetic finding, despite previous negative deep exome sequencing at greater than 350× coverage and array-based genotyping.

Candidate somatic variants in known lesional focal epilepsy-associated genes were identified in seven participants (21%), including two with two candidate variants each. Three participants (9%) had variants detectable only by genome sequencing: two TSC2 deletions, two NPRL3 deletions, and an exonic FGFR1 inversion. Three individuals had germline polymorphisms potentially representing modulators or risk factors. Exploratory analysis also identified clocklike mutational signatures and correlations with age at seizure onset and histopathology.

The findings support the feasibility of deep genome sequencing for detecting low-frequency mosaic variants and complex structural changes missed by exome sequencing. However, the small, selected sample limits generalizability, and candidate findings do not establish clinical utility or treatment benefit. The authors describe genome sequencing as a potential complement to other tissue-testing technologies and call for further work on conceptual and technical challenges in somatic variant detection.