Rare Autosomal Trisomies Detected by Noninvasive Prenatal Testing: Performance, Outcomes, and Exploratory Analysis of the Theoretical Mosaicism Ratio.

Rare autosomal trisomies detected in maternal cell-free DNA by genome-wide NIPT, seldom confirmed in the fetus, with a proportion of cases corresponding to uniparental disomy or runs of homozygosity revealed by invasive testing.

This single-centre retrospective cohort study covered 64,889 singleton pregnancies screened by genome-wide NIPT between March 2021 and July 2024, in order to characterise the rare autosomal trisomies detected and assess their clinical significance. Screening was positive in 112 pregnancies (0.17%); among the 79 that proceeded to invasive testing, the positive predictive value for fetal confirmation was 5.1% (4/79) and the diagnostic yield 8.9% (7/79) once uniparental disomy and runs of homozygosity were counted. Follow-up of 104 pregnancies recorded 63 uncomplicated pregnancies, 34 adverse outcomes and 7 terminations of pregnancy. In the primary analysis of 93 pregnancies, the platform-specific theoretical mosaicism ratio (TMR) discriminated adverse outcomes poorly (AUC 0.643, 95% CI 0.531-0.756), although a TMR of 0.63 or above remained associated with adverse outcomes after adjustment (adjusted OR 3.66, 95% CI 1.33-10.05; p = 0.012). The authors conclude that a positive result carries meaningful pregnancy risk despite the low fetal confirmation rate, and that TMR should only be used as a supplementary marker, read alongside the chromosome involved, invasive diagnosis, ultrasound and obstetric findings.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

Two messages for the prenatal clinic: a positive NIPT result for a rare autosomal trisomy is rarely confirmed in the fetus, yet it is far from harmless, since roughly a third of the followed pregnancies had an adverse outcome. With an AUC of 0.643 and wide overlap between groups, TMR does not perform well enough to guide a decision on its own and should not be presented to couples as a prognostic marker. Multicentre external validation, independent of the sequencing platform, would be needed before any use in practice.

Analysis by Dr Thibaut Benquey