Abstracts

Rationale:
To understand the clinical characteristics in the diagnosis of epilepsy that determine the diagnostic yield of an epilepsy gene focused exome analysis (EGFE) as compared to Chromosomal Microarray (CMA) testing.   
Method:
We conducted a retrospective study of 375 patients diagnosed with epilepsy who had an EGFE and CMA completed between September 30, 2016 and February 11, 2020 at Children’s Hospital Los Angeles (CHLA).  All 375 extracted DNA samples were analyzed at Center for Personalized Medicine at CHLA. The EGFE was designed to reliably detect single nucleotide variants and small deletions in genes associated with epilepsy.  The original assay included 166 genes, while the updated assay upon addition of new genes associated with epilepsy contained 244 genes.  CMA analysis was performed using the Affymetrix CytoScan HD Microarray. Nine clinical characteristics were collected by medical record review for correlation with pathogenic and likely pathogenic mutations in CMA and EGFE.  The data was analyzed by first performing comparison between the correlation coefficients of the CMA and EGFE with other clinical characteristics and, second, by applying a logistic regression analysis using statistical software Stata 15.
Results:
Analysis of the 375 EGFEs revealed 66 (17.6%) with pathogenic or likely pathogenic mutations. CMA was performed in 134 out of the 375 patients; only 7 (5.22%) revealed pathogenic mutations.   All patients with pathogenic EGFE results had normal CMA. After further review, twenty-one patients with variants of unknown significance (VUS) on the EGFE, were felt to likely have disease causing genotypic variants based on the presenting epilepsy phenotype.  We hypothesize that the diagnostic yield of an EGFE could therefore be as high as 23.2%. The 66 patients with a pathogenic EGFE exhibited more comorbid clinical characteristics compared to the patients with a positive CMA. Using a correlation analysis, statistically significant results for EGFE were obtained for developmental delay, intellectual disabilities, age, and microcephaly.  Using a logistic regression model, the clinical characteristics showing strong statistical significance predicting pathogenic EGFE were developmental delays (2.5) and microcephaly (3.0).  For CMA only hypotonia was significant in the regression model (10).
Conclusion:
In our study, EGFE was shown to have a higher yield compared to CMA in the diagnosis of epilepsy.  We delineated different clinical characteristics associated with the patients who were more likely to have a pathogenic EGFE vs a pathogenic CMA.  The diagnostic yield of the EGFE is likely higher if any reported VUS are evaluated by a comprehensive team for reclassification after a proper clinical evaluation.   We found that CMA also has value for some patients with a negative EGFE.  This information may help guide providers when deciding which test to order.
Funding:
:no funding