Comparison of noninvasive prenatal screening for defined pathogenic microdeletion/microduplication syndromes and nonsyndromic copy number variations: a large multicenter study

Abstract

Background: This retrospective study assessed the precision of noninvasive prenatal testing (NIPT) in detecting microdeletion/microduplication syndromes (MMSs) and nonsyndromic copy number variations (CNVs). Methods: The study included 19,086 singleton pregnancies screened on NIPT using high-throughput sequencing. Pregnancies with CNVs on NIPT underwent amniocentesis for karyotyping and CNV sequencing (CNV-seq). We analyzed pathogenic MMSs and nonsyndromic CNVs separately, dividing the CNVs into subgroups based on fragment size and fetal ultrasound findings. Results: A total of 170 abnormalities were detected by NIPT, of which 113 (66.5%) underwent invasive testing. The positive predictive value (PPV) of CNV-seq for all types of CNV detected by NIPT was 35.4%, with PPVs of 61.5 and 27.6% for pathogenic MMSs and nonsyndromic CNVs, respectively. PPVs for NIPT showed different values depending on gestational characteristics, with the highest PPV for NIPT in the group with increased nuchal thickness (66.7%) and for the abnormal ultrasound group (57.1%). CNVs ≤5 Mb with normal ultrasound findings were generally associated with a healthy fetus. Conclusion: NIPT can detect chromosomal aberrations in the first trimester, with high performance for MMSs. However, due to the low PPV for nonsyndromic CNVs, and the good pregnancy outcome in most cases, the introduction of expanded NIPT would cause an increase in unnecessary invasive procedures and inappropriate terminations of pregnancy.

The accuracy of cell‐free DNA screening for fetal segmental copy number variants: A systematic review and meta‐analysis

“…In total, 7845 search results were identified, of which 1862 were duplicates. After screening the remaining 5983 results, 63 articles satisfied the inclusion criteria 12,14–16,18,26–83 . The study selection process is shown in Figure 1.…”

“…Of the 63 included studies, PPV results were calculated from 59 (48 case‐series, 12,14–16,18,26–28,30,33–38,40–46,48,49,52–59,61–66,68–71,74,78–82 and 11 cohort studies 29,47,50,60,67,72,73,75–77,83 ). Sensitivity and specificity were reported from nine (four case‐controls, 31,32,39,51 five cohort studies with complete diagnostic confirmation 29,60,75–77 ).…”

“…The most frequent high-risk bias arose in the 'Flow and Timing' domain as a result of incomplete diagnostic confirmation of cfDNA results, with 31 studies (49.2%) deemed highrisk. 15,18,[26][27][28]30,[33][34][35]37,38,42,[45][46][47][48][49]52,54,55,58,[62][63][64][65]70,71,73,74,82,83 This is most likely attributable to loss of follow-up of study participants or maternal decline of diagnostic testing. The most frequent applicability concern was in 'Reference Standard', with 15.9% (n = 10/63) of studies being classified as high-risk.…”

Prenatal Screening for Microdeletions and Rare Autosomal Aneuploidies

“…The overall screen-positive rate for CNVs, regardless of size or pathogenicity, ranges from 0.1% to 0.9%, with reported sensitivities of 20% to 100% 11,[13][14][15][16][17][18][19][20][22][23][24][25]38,40,[53][54][55][56][57][58][59] .…”

“…Some studies report a true positive when the fetus is demonstrated by diagnostic genetic testing to harbor the CNV in question, whereas others consider the detection of the genetic anomaly in placental, maternal, or fetal tissues as a true positive 15,62,63 . Unfortunately, VUS and nonsyndromic CNVs were excluded from the calculation of PPVs in some trials but not in others 55 . In one report, VUS/benign CNVs accounted for ~19% of those confirmed to be fetal in origin 23 .…”