Genome-Wide Fetal Aneuploidy Detection by Maternal Plasma DNA Sequencing

OBJECTIVE: To prospectively determine the diagnostic accuracy of massively parallel sequencing to detect whole chromosome fetal aneuploidy from maternal plasma. METHODS: Blood samples were collected in a prospective, blinded study from 2,882 women undergoing prenatal diagnostic procedures at 60 U.S. sites. An independent biostatistician selected all singleton pregnancies with any abnormal karyotype and a balanced number of randomly selected pregnancies with euploid karyotypes. Chromosome classifications were made for each sample by massively parallel sequencing and compared with fetal karyotype. RESULTS: Within an analysis cohort of 532 samples, the following were classified correctly: 89 of 89 trisomy 21 cases (sensitivity 100%, 95% [confidence interval] CI 95.9-100), 35 of 36 trisomy 18 cases (sensitivity 97.2%, 95% CI 85.5-99.9), 11 of 14 trisomy 13 cases (sensitivity 78.6%, 95% CI 49.2-99.9), 232 of 233 females (sensitivity 99.6%, 95% CI 97.6 to more than 99.9), 184 of 184 males (sensitivity 100%, 95% CI 98.0-100), and 15 of 16 monosomy X cases (sensitivity 93.8%, 95% CI 69.8-99.8). There were no false-positive results for autosomal aneuploidies (100% specificity, 95% CI more than 98.5 to 100). In addition, fetuses with mosaicism for trisomy 21 (3/3), trisomy 18 (1/1), and monosomy X (2/7), three cases of translocation trisomy, two cases of other autosomal trisomies (20 and 16), and other sex chromosome aneuploidies (XXX, XXY, and XYY) were classified correctly. CONCLUSION: This prospective study demonstrates the efficacy of massively parallel sequencing of maternal plasma DNA to detect fetal aneuploidy for multiple chromosomes across the genome. The high sensitivity and specificity for the detection of trisomies 21, 18, 13, and monosomy X suggest that massively parallel sequencing can be incorporated into existing aneuploidy screening algorithms to reduce unnecessary invasive procedures.