Transcriptomic Sequencing of Primary Tissue Improves Genetic Diagnosis in Bicuspid Aortopathy by Identification and Validation of Splicing Variants

The biccuspid aortic valve complex (BAV) comprises a set of disorders characterized by aortic malformations. These are heritable conditions and may occur in isolation or as a syndromic disease. Clinical genetic testing is available for patients in the form of gene panel testing, however, with relatively low molecular diagnosis rates of 10 to 27%. Combining variant calling algorithms from RNAseq data with expression level measurements, transcriptomic approaches have shown strong potential to contribute to detection and interpretation of genetic variants missed by gene panels.

A study cohort of 101 patients with bicuspid aortopathy confirmed at surgery was recruited prospectively into the Stanford Aortic Biobank, where 34 had stenosed, 62 patients had regurgitant, and 5 patients had normal functioning BAV. Tissue from 49 matched nonsyndromic patients was selected for sequencing. Tissue from five cardiac transplant donors was included as control. Another 56 patients from Beijing, China were included as confirmatory cohort. We isolated genomic DNA and whole RNA from all samples. Whole exome sequencing (WES) was performed on Illumina HiSeq 2500, PE100bp. Whole transcriptome sequencing (RNAseq) was performed on Illumina HiSeq 4000, PE150bp.

Within the Stanford discovery cohort of 49 individuals undergoing WES, pathogenic or likely pathogenic mutations were identified in 6 individuals (12%) within genes related to BAV and aortopathy. Despite excluding patients with syndromic disease, mutations causing connective tissue disorders (Ehlers–Danlos’ syndrome, Marfan’s syndrome) were discovered. Of the subset undergoing RNAseq, four additional variants missed by WES were detected in three individuals without a causal variant identified by exome sequencing. No variants were detected in the RNAseq data from the five healthy controls. Among the Chinese cohort undergoing RNAseq, we observed 18 pathogenic variants. Of the 56 individuals, 22 individuals (39%) received a molecular diagnosis from RNA sequencing.

Here, we describe the addition of RNAseq of primary tissue to molecular diagnoses using WES in patients with BAV aortopathy. Combining all patients in both cohorts who received RNA sequencing, the diagnostic rate was 32% (25 out of 77), which exceeds reported rates of molecular diagnosis of 10 to 27% for gene-based testing. Given the modest overlap in diagnosis between exome and RNAseq, these data suggest that genome sequencing may offer the most comprehensive clinical diagnostic approach in patients with BAV and aortopathy.