Development and Validation of a SNP-Based cfDNA Test for Vanishing Twin Pregnancies: The VANISH Trial

Document Type

Conference Proceeding

Publication Date

1-2026

Publication Title

Pregnancy

Abstract

Objective: Develop and validate a SNP-based non-invasive prenatal cell free DNA (cfDNA) screening algorithm to identify the cfDNA signal for the living twin in a vanishing twin (VT) pregnancy and assess that twin's aneuploidy risk. Study Design: Prospective study including 108 patients with VTs at 21 US sites (8/21–8/23). Inclusion: >18 years old, >9 wks’ gestation and ultrasound-confirmed VT. SNP-based cfDNA screening occurred at enrollment and twice more at 3-week intervals. Neonatal buccal swabs were collected from the living twin for next generation sequencing, to confirm that the SNPs identified were those of the living fetus. Exclusion: lost to follow-up, had a pregnancy loss or missing study samples. Using SNP-based technology, one or two fetoplacental cfDNA signals, in addition to that from the pregnant patient, were identified. Fetal fractions were calculated for each cfDNA signal and an algorithm to identify the living fetus in a VT pregnancy was developed using samples from 47 participants. Concordance between the fetal genotype derived from cfDNA in the maternal plasma and the buccal swab genotype from 61 subjects was used to validate the algorithm in correctly identifying the living fetus. Common aneuploidies and fetal sex were determined for the living fetus in these samples. Results: Median gestational age was 10.8 wks (range: 8.6–16.1) at VT diagnosis. In 57/61 (93.4%; 95% CI: 84.3%, 97.4%) validation cases, we identified the living fetus and made the correct aneuploidy call confirmed by newborn buccal swabs (Figure 1). Aneuploidy results included 56 low-risk and 1 high-risk result for trisomy 21. No false positive or false negative results were reported. Validation results were obtained on 55/57 living twins within the first 2 maternal blood draws; 2 cases required 3 draws. At the time the living fetus was identified, 39 samples had two fetal cfDNA signals and 18 had one signal. Conclusion: The SNP-based algorithm accurately identified the living fetus in VT pregnancies. Aneuploidy risk assessment was validated in all cases where the cfDNA signal for the living twin was identified.

Volume

2

Issue

S1

First Page

7

Comments

Society for Maternal-Fetal Medicine (SMFM) 2026 Pregnancy Meeting, February 8-13, 2026, Las Vegas, NV

Last Page

7

DOI

10.1002/pmf2.70157

Link to Full Text

Share

COinS