Diagnostic accuracy of random massively parallel sequencing for non-invasive prenatal detection of common autosomal aneuploidies: a collaborative study in Europe

Abstract

The objective of this study is to validate the diagnostic accuracy of a non-invasive prenatal test for detecting trisomies 13, 18, and 21 for a population in Germany and Switzerland. Random massively parallel sequencing was applied using Illumina sequencing platform HiSeq2000. Fetal aneuploidies were identified using a median absolute deviation based z-score equation. A bioinformatics algorithm based on guanine-cytosine normalization was applied after the data were unblinded. Results of massively parallel sequencing and invasive procedures were compared. Overall, 40/42 samples were correctly classified as trisomy 21-positive, including a translocation trisomy 21 [46,XY,der(13;21),+21] and a structural aberration of chromosome 21 [46,XX,rec(21)dup(21q)inv(21)(p12q21.1)] but not including a low percentage mosaic trisomy 21 [47,XY,+21/46,XY], [sensitivity: 95.2%; one-sided lower confidence limit: 85.8%]; 430/430 samples were correctly classified as trisomy 21-negative (specificity: 100%; one-sided lower CL: 99.3%). Using a new bioinformatics algorithm with guanine-cytosine normalization, detection of trisomy 21 was facilitated, and five of five trisomy 13 cases and eight of eight trisomy 18 cases were correctly identified. Our newly established non-invasive prenatal test allows detection of fetal trisomies 13, 18, and 21 with high accuracy in a population in Germany and Switzerland.