Fibroblast transcriptomics uncovers pathogenic genomic variants in individuals with exome-negative childhood onset epilepsy

Objective: This study aims to improve genetic diagnosis in childhood onset epilepsy with neurodevelopmental problems by utilizing RNA sequencing of fibroblasts to identify pathogenic variants that may be missed by exome sequencing and copy number variation analysis. Methods: We enrolled 41 individuals with childhood onset epilepsy and neurodevelopmental problems who previously had inconclusive genetic testing. Fibroblast samples were cultured and analyzed using RNA sequencing to detect aberrant expression, aberrant splicing, and monoallelic expression using the Detection of RNA Outlier Pipeline (DROP) pipeline. Detected events were correlated with phenotypes, and long-read genome sequencing was performed on individuals with strong candidate events to identify the causal genomic variant. A systematic literature review on RNA sequencing in rare disorders was conducted to contextualize our findings. Results: RNA sequencing identified five strong candidate events in four individuals, affecting the genes QRICH1, TSC1, SMARCA1, GNAI1, and PTEN. (Likely) pathogenic genomic variants affecting expression of QRICH1, TSC1, and SMARCA1 were detected, resulting in a diagnostic yield of 7% (3/41). Two variants were not covered in the initial exome sequencing data but were revealed through long-read sequencing. The identification of a pathogenic TSC1 variant led to a previously unrecognized diagnosis of tuberous sclerosis complex. This prompted guideline-based screening, which revealed tuberous sclerosis lesions in the brain and lung, directly impacting clinical care. Notably, two of the three pathogenic events would not have been detected using whole blood due to the lack of expression of the involved genes. The lower yield of this study compared to studies in other rare disorders reflects the genetic heterogeneity of epilepsy and neurodevelopmental disorders, and the inaccessibility of affected tissue. Significance: This research underscores RNA sequencing of cultured fibroblasts as a valuable tool in genetic diagnostics for childhood onset epilepsy, particularly when conventional methods fail. Expanding the control dataset with age-matched samples and incorporating RNA sequencing with nonsense-mediated decay inhibition could further enhance diagnostic yield.