Validation of Copy Number Variants Detection from Pregnant Plasma Using Low-Pass Whole-Genome Sequencing in Noninvasive Prenatal Testing-Like Settings

Abstract

Detection of copy number variants as an integral part of noninvasive prenatal testing is increasingly used in clinical practice worldwide. We performed validation on plasma samples from 34 pregnant women with known aberrations using cell-free DNA sequencing to evaluate the sensitivity for copy number variants (CNV) detection using an in-house CNV fraction-based detection algorithm. The sensitivity for CNVs smaller than 3 megabases (Mb), larger than 3Mb, and overall was 78.57%, 100%, and 90.6%, respectively. Regarding the fetal fraction, detection sensitivity in the group with a fetal fraction of less than 10% was 57.14%, whereas there was 100% sensitivity in the group with fetal fraction exceeding 10%. The assay is also capable of indicating whether the origin of an aberration is exclusively fetal or fetomaternal/maternal. This validation demonstrated that a CNV fraction-based algorithm was applicable and feasible in clinical settings as a supplement to testing for common trisomies 21, 18, and 13.

Keywords: CNV detection; low-pass whole-genome sequencing; noninvasive prenatal testing (NIPT).

Figure 1

Visualization of CNV for chromosome 18. Normalized read counts per bin are depicted as gray dots. The red horizontal line shows the 18p11.32–p11.21 microdeletion approximately of 12.8 Mb. The light gray vertical band depicts an unmappable region around the centromere. Black horizontal bands signify bins that did not pass quality metrics and are thus excluded from the analysis. The light red region highlights the detected pathogenic region. The approximated z-score of deletion is displayed over the red segment. The estimated level of aberration detection based on the fetal fraction of this sample (12.8%) is visualized as a red dashed line, while the magenta dashed line represents the estimated level of maternal aberration detection.

Figure 2

Detection sensitivity based on the fetal fraction and aberration size. Full markers depict the detected samples, while the empty ones represent the non-detected samples. The dashed gray line shows the estimated detection range.