We studied the yield of commercial next generation sequencing (NGS) panels in epilepsy and explored whether any clinical features were associated with diagnostic yield.

Genetics are playing an expanding role in clinical practice. While previous studies have examined the utility of NGS panels in diagnosing epilepsy in research settings, we studied the yield of such panels in actual clinical practice.

We performed a retrospective chart review for patients for whom a NGS epilepsy panel was ordered at our center. ACMG criteria was used to classify the pathogenicity of variants. Chi-square analysis with Bonferroni correction was used to assess whether any clinical features were associated with diagnostic yield.

Our cohort consisted of 84 (11 adult and 73 pediatric) patients. Two (2%) patients had a family history of consanguinity and 18 (21%) had a family history of seizures in a first-degree relative. 16 (19%) patients had an epileptic encephalopathy, 25 (30%) had generalized epilepsy, 33 (39%) had focal epilepsy, four (5%) had febrile seizures alone and six (7%) had unclassifiable epilepsy. Gene panel identified the pathogenic variant in 16 (19%). Eight were autosomal dominant (AD) de novo mutations, four were likely AD, but parent sequencing was unavailable, two were AD inherited from a parent, one was X-linked and one was autosomal recessive. 12 of the 16 patients with positive sequencing had an onset of seizures prior to the age of two years old. The mean age of seizure onset was significantly lower in patients with positive sequencing. No patient characteristics were significantly associated with a higher diagnostic yield. However, patient characteristics that, though not statistically significant, appeared to predict a higher yield included developmental delay, being unable to ambulate without assistance and history of status epilepticus.

NGS epilepsy gene panels have a diagnostic yield of 19% when used in a clinical setting.