Rare autosomal trisomies: comparison of detection through cell-free DNA analysis and direct chromosome preparation of chorionic villus samples

Objective Direct chromosome preparations of chorionic villus samples (CVS) and cell-free DNA (cfDNA) testing both involve analysis of the trophoblastic cell lineage. The aim of this study was to compare the spectrum of rare autosomal trisomies (RATs) detected by these two approaches and assess the available information on their clinical significance. Methods Data from 10 reports on genome-wide cfDNA testing were pooled to determine which chromosomes were most frequently involved in RAT-positive cases, and pregnancy outcome information was reviewed. CVS information was obtained from an updated database of 76 102 consecutive CVS analyses performed over a period of 18 years at TOMA laboratory, in which trophoblastic and mesenchymal layers were analyzed and amniotic fluid cell analysis was recommended for RAT-positive cases. Chromosomes involved and presence of confined placental mosaicism, true fetal mosaicism and uniparental disomy (UPD) for imprinted chromosomes were assessed. Also evaluated were the frequency and types of RATs in products of conception. Results RATs were present in 634 of 196 662 (0.32%) cfDNA samples and 237 of 57 539 (0.41%) CVS trophoblast samples (P < 0.01). The frequency of RATs varied over 8-fold between the cfDNA reports. Confirmation of abnormality through amniocentesis was more likely when RATs were ascertained through cfDNA (14 of 151; 9.3%) than through CVS trophoblasts (seven of 237; 3.0%) (P < 0.01). In cfDNA-ascertained cases, trisomies 15, 16 and 22, which are associated with fetal loss, were identified proportionately more often. Of 151 cases with RAT identified by cfDNA and outcome information available, 41.1% resulted in normal live birth; 27.2% in fetal loss; 7.3% had phenotypic abnormality detected through ultrasound or other follow-up evaluation; 2.0% had a clinically significant UPD; and 14.6% had fetal growth restriction or low birth weight. All autosomes were involved in trisomies in products of conception; the most common RATs detected were trisomies 16, 22 and 15 with a frequency of > 9% each. Conclusions Although there are strong parallels between RATs ascertained through cfDNA analysis and direct chromosome preparation of CVS, caution is needed in applying conclusions from CVS analysis to cfDNA testing, and vice versa. RATs identified through genome-wide cfDNA tests have uncertain risks for fetal loss, growth restriction or fetal abnormality. Copyright (C) 2019 ISUOG. Published by John Wiley & Sons Ltd.