Detection of maternal X chromosome abnormalities using single nucleotide polymorphism–based noninvasive prenatal testing

Abstract

Maternal X chromosome abnormalities may cause discordant results between noninvasive prenatal screening tests and diagnostic evaluation of the fetus/newborn, leading to unnecessary invasive testing. Women with X chromosome abnormalities are at increased risk for reproductive, pregnancy, or other health complications, which may be reduced or ameliorated by early diagnosis, monitoring, and intervention. This study aimed to validate a single nucleotide polymorphism-based noninvasive prenatal test to identify X chromosome abnormalities of maternal origin. All tests unable to evaluate fetal risk for aneuploidy because of uninformative algorithm results were eligible for inclusion. Two groups of cases were prospectively identified: Group A (n=106) where a maternal X chromosome abnormality was suspected and Group B (control group, n=107) where a fetal chromosome abnormality involving chromosome 13, 18, 21, or X was suspected but did not meet criteria for reporting. Maternal DNA was isolated from the plasma-depleted cellular pellet and sent to a reference laboratory for blinded analysis using chromosomal microarray. A chromosome abnormality involving chromosomes 13, 18, 21, or X was reported by the reference laboratory if ≥5 Mb in size and present in ≥20% of the DNA. A maternal X chromosome abnormality was suspected in 1/1305 tests (149/194,385; 0.08%). In Group A, a maternal X chromosome abnormality was confirmed in 100/106 cases (94.3% positive predictive value, 1-sided 97.5% confidence interval, 88.1%-100.0%). Turner syndrome was the most commonly suspected maternal abnormality (58/106, 54.7%), with confirmation of mosaic or nonmosaic 45,X by microarray in 38/58 (65.5%) cases. Noninvasive prenatal screening tests suspected the presence of maternal 47,XXX with or without mosaicism in 40/106 (37.7%) cases, confirmed by microarray in 38/40 (95.0%). In Group B (n=107), no maternal microarray abnormalities were reported, providing a negative predictive value of 100% (1-sided 97.5% confidence interval, 96.6%-100.0%). When noninvasive prenatal testing suspected a maternal X chromosome abnormality, maternal microarray confirmed an X chromosome abnormality with 94.3% positive predictive value. Of the maternal X chromosome abnormalities detected by array, >50% were 45,X. When fetal chromosome abnormalities involving chromosomes 13, 18, 21, or X were suspected, no maternal chromosome abnormalities were reported, yielding a negative predictive value of 100%. Women with maternal X abnormalities suspected with noninvasive prenatal testing may be at increased risk for reproductive and health complications; early evaluation and treatment may prevent long-term consequences or disability.