Abstracts

Rationale: Genetic testing panels for epilepsy are ordered more frequently in recent years, particularly for patients with early-life epilepsies. At this time, there are no guidelines for how and when to integrate genetic testing into the epilepsy work-up; however, a recent publication suggested that all patients with early-life epilepsy (<3 years old) should have genetic testing incorporated into their initial evaluation (Berg et al., JAMA Pediatrics 2017; 171(9); 863-871). The goal of this study was to investigate the age at molecular diagnosis for patients with pathogenic variants in genes causing infantile-onset epilepsy who underwent clinical diagnostic testing over a 6-year period. Methods: This retrospective study reviewed results of molecular diagnostic testing from 2833 individuals with epilepsy who underwent testing in a clinical diagnostic laboratory for a panel of genes causing infantile-onset epilepsy.  Testing was completed between 2012 and 2017 and included Next Generation sequencing (NGS) and exon-level array comparative genomic hybridization (aCGH).  The age at molecular diagnosis was calculated for all individuals who had testing, including individuals with pathogenic or likely pathogenic (P/LP) variants in specific genes (CDKL5, KCNQ2, PRRT2, and STXBP1) associated with infantile-onset epilepsy typically with onset in the first year of life. Results: The overall positive diagnostic yield for the infantile-onset epilepsy panel was 23% (664/2833 cases). P/LP variants in the CDKL5, KCNQ2, PRRT2, and STXBP1 genes accounted for approximately 39% (256/664) of the positive cases. The mean age at molecular diagnosis was under 1 year of age for the PRRT2 gene but not for the KCNQ2, CDKL5, or STXBP1 genes.  The ages in years at molecular diagnosis by gene were: PRRT2-mean 0.90, median 0.56, range 0.14-5.4; KCNQ2-mean 1.63, median 0.31, range 0.05-39.5; CDKL5-mean 3.48, median 1.35, range 0.25-22.6; STXBP1-mean 4.07, median 2.09, range 0.06-27.6. The number of individuals who had clinical diagnostic testing at our laboratory using an epilepsy panel increased 3-fold from 2012-2017.  The mean age at molecular diagnosis did not change significantly during this timeframe for patients with pathogenic findings in the PRRT2, KCNQ2, or CDKL5 genes, while the mean age at diagnosis for patients with STXBP1 pathogenic variants decreased from 7.5 years in 2012 to 0.5 years in 2017. Conclusions: An early molecular diagnosis may benefit patients with epilepsy by ending the lengthy and expensive diagnostic odyssey, enabling accurate genetic counseling about recurrence risks, and allowing for precision medicine or participation in clinical trials in some cases. Our data demonstrate that although genetic testing panels for infantile-onset epilepsy are ordered more frequently in recent years, testing is often pursued months to years after the initial onset of seizures and is often not incorporated into the initial evaluation of new-onset infantile seizures. Funding: None