High Concordance of Copy Number Variants Detected by Chromosomal Microarray and Exome Sequencing in Clinical Diagnostics.

Abstract

Exome sequencing (ES), originally developed to detect single nucleotide variants (SNVs), has been increasingly leveraged to detect copy number variants (CNVs) through read-depth analysis, enhancing diagnostic yield with minimal computational overhead. However, chromosomal microarray (CMA) testing remains widely used. To evaluate the utility of ES as a first-tier clinical diagnostic test, we compared the sensitivity of CNV detection by ES to that of CMA in individuals who underwent both tests, and developed triploidy screening based on ES data. ES identified most clinically relevant CNVs, with a 98.91% concordance for regions adequately captured. A retrospective analysis of CNVs detected from ES over a ~3-year period, comprising 1563 prenatal and 4884 postnatal cases, revealed CNVs in 3.8% of prenatal and 4.4% of postnatal samples. The relatively low percentage stems from the fact that most cases underwent CMA before ES. Pathogenic or likely pathogenic variants constituted 78.7% and 78.0% of these subgroups, with the remainder classified as variants of unknown significance. We highlight clinically relevant examples of CNVs across a range of sizes, including cases involving both CNVs and SNVs. The high consanguinity rate of the cohort allowed for systematic analysis of homozygous CNVs. Additionally, we demonstrate that ES can capture other diagnostic utilities traditionally associated with CMA, specifically uniparental disomy (UPD) and triploidy. Overall, our findings support ES as a robust, cost-effective alternative to CMA and advocate for its broader use as a first-tier diagnostic test for neurodevelopmental delay and congenital malformations, particularly until whole genome sequencing becomes more accessible and affordable.