Diagnostic Yield of Comprehensive Reanalysis After Nondiagnostic Short-Read Genome Sequencing in Infants With Unexplained Epilepsy.

Abstract

The highest incidence of epilepsy in childhood occurs in the first year of life. Infantile epilepsies are associated with substantial morbidity and mortality. Although most are presumed to have genetic etiologies, many infants with nonacquired epilepsy remain genetically unsolved after clinical genome sequencing. The yield of reanalysis after nondiagnostic genome sequencing in this population is unknown. We aimed to determine the diagnostic yield of comprehensive reanalysis after nondiagnostic genome sequencing in infants with unexplained epilepsy. This cohort study included infants with unexplained epilepsy or complex febrile seizures who were recruited from 4 pediatric referral centers from September 2021 to March 2024 and had nondiagnostic clinical rapid genome sequencing. We performed comprehensive reanalysis of genome sequencing data from infants and available biological parents using multiple bioinformatics pipelines through July 2025 and clinically confirmed reanalysis findings. The primary outcome was diagnostic yield of genome sequencing reanalysis, defined as the percentage of infants who received genetic diagnoses from reanalysis. The secondary outcome was clinical utility of reanalysis findings. From an initial cohort of 312 infants with unexplained epilepsy who underwent clinical rapid genome sequencing, we performed comprehensive genome reanalysis in 176 infants with initially nondiagnostic results at a median age of 642 days, including 63 female patients (36%) and 30 (17%) with neonatal-onset seizures. The diagnostic yield of reanalysis was 5.1% (9/176, 95% CI 2.4%-9.5%), increasing the overall yield from 43.6% (136/312, 95% CI 38.0%-49.3%) to 46.5% (145/312, 95% CI 40.8%-52.2%). Of the new diagnoses, 6 involved variants not reported by clinical laboratories (2 single nucleotide variants, 2 structural variants, 1 tandem repeat expansion, 1 mosaic variant) and 3 involved previously reported variants of uncertain significance with new evidence. All diagnoses had clinical utility. Comprehensive reanalysis after nondiagnostic rapid genome sequencing has utility for infants with unexplained epilepsy. Our findings support implementation of reanalysis within 1-2 years after nondiagnostic genomic sequencing into routine clinical care of children with unexplained epilepsy and the expansion of clinically accredited genomic sequencing to include complex and noncoding variant detection.