Non-invasive prenatal testing in the management of twin pregnancies

Abstract

Twin pregnancies are common and associated with pregnancy complications and adverse outcomes. Prenatal clinical management is intensive and has been hampered by inferior screening and less acceptable invasive testing. For aneuploidy screening, meta-analyses show that non-invasive prenatal testing (NIPT) through analysis of cell-free DNA (cf-DNA) is superior to serum and ultrasound-based tests. The positive predictive value for NIPT is driven strongly by the discriminatory power of the assay and only secondarily by the prior risk. Uncertainties in a priori risks for aneuploidies in twin pregnancies are therefore of lesser importance with NIPT. Additional information on zygosity can be obtained using NIPT. Establishing zygosity can be helpful when chorionicity was not reliably established early in pregnancy or where the there is a concern for one versus two affected fetuses. In dizygotic twin pregnancies, individual fetal fractions can be measured to ensure that both values are satisfactory. Vanishing twins can be identified by NIPT. Although clinical utility of routinely detecting vanishing twins has not yet been demonstrated, there are individual cases where cf-DNA analysis could be helpful in explaining unusual clinical or laboratory observations. We conclude that cf-DNA analysis and ultrasound have synergistic roles in the management of multiple gestational pregnancies.

FIGURE 1

Heterozygosity plots for single nucleotide polymorphisms (SNPs) in euploid twin pregnancies. (A) monozygotic twins. (B) dizygotic twins. Heterozygosity plots are visual graphical representations of the alleles present in the maternal plasma. In these figures, each SNP type is denoted as either an “A” or “B” allele. The x‐axis, left to right, shows the alleles along chromosomes 13, 18, 21, X and a set of homologous non‐recombinant regions (HNR) from the X and Y chromosomes. The HNR SNPs are designated “A” when they map to the X‐chromosome and “B” when they map to the Y‐chromosome. The Y‐axis shows, for each locus present, the overall percentage of A type, that is, (A/[A + B])%. The Y axis also specifies the corresponding allele combinations for the mother and the fetuses. The data is shown in green if the maternal genotype is AA, blue BB, or orange AB. (A) Euploid male monozygotic twins. The patten is the same as that which would be present for a singleton pregnancy. The extent to which the lower green band departs from 100%, the two orange bands depart from 50%, and upper blue band departs from 0% is determined by the fetal fraction. Thus, all informative SNPs provide a measure of the fetal fraction. The SNP pattern for the X‐chromosome indicates that there are no paternally derived X‐chromosomes present in the cell‐free DNA (cf‐DNA), consistent with a male fetal sex. Moreover, the HNR SNPs indicate presence of Y‐chromosome alleles at a level consistent with both fetuses being male. (B) Euploid female dizygotic twins. There are additional allelic combinations, compared to the monozygotic pattern, that are attributable to the presence of a second paternally derived haplotype. Some of the paternally derived SNPs will be identical in the two fetuses and others will differ. Although band patterns appear to be more diffuse and overlapping in the graphical representation, the allele contributions from each fetus can be computationally resolved (allowing for linkage and recombination) and the individual fetal fractions can be separately determined, similar to that for singleton pregnancies. The X‐chromosome pattern is similar to the autosomes when the sex of the fetuses is female. The HNR SNPs indicate no Y‐chromosome contribution. For SNP‐based NIPT in twins, initially evaluating zygosity is necessary in the assessment of the presence or absence of aneuploidy. Presence of aneuploidy or a specific microdeletion would be indicated by a larger change in the A/(A + B) ratios, compared to that seen for the disomic chromosome regions (not shown). Illustration constructed from material provided by Natera, Inc

FIGURE 2

Scatterplot of the two fetal fractions in dizygotic twin pregnancies. Each fetal fraction (FF) in dizygotic pregnancies was randomly assigned as “Fetus A” or “Fetus B” and plotted. The plot illustrates that some pregnancies show large differences between the paired results (overall R ² = 0.66). Points below the horizontal line or to the left of the vertical line have a least one fetal fraction below 2.8%, the cut‐off used for a reportable test. Data is based on 3161 dizygotic pregnancies previously reported by Hedriana et al.