While cell-free DNA (cfDNA) testing for fetal aneuploidy has been shown to have a high detection rate for trisomy, most large studies have included only high-risk populations. This blinded, prospective study tested the performance of screening with cfDNA for common trisomies in the general population.Maternal blood samples were collected from women at 35 centers in the United States, Canada, and Europe with a singleton pregnancy between 10.0 and 14.3 weeks' gestation. Samples were sent to a single clinical laboratory for cfDNA testing. The patients were also screened for aneuploidy using measurement of nuchal translucency and biochemical analytes. Participants and providers were blinded to the cfDNA testing results but were informed of the standard screening results. Medical records of physical examination at birth and genetic testing were used to inform newborn outcomes. The area under the receiver operating characteristic curve (AUC) for cfDNA testing for trisomy 21 compared with standard testing was the primary outcome. The AUC of the 2 types of testing for trisomies 13 and 18 were the secondary outcomes.A total of 18,955 women were enrolled between March 2012 and April 2013, and after all exclusions, 15,841 patients were included in the cohort. Mean gestational age was 12.5 weeks, and mean maternal age was 31 years. In the cohort, 1 in 236 pregnancies (n = 68) had chromosomal abnormalities. For trisomy 21, the AUC for cfDNA testing was 0.999, and for standard screening, the AUC was 0.958 (P = 0.001). Cell-free DNA identified all 38 cases of trisomy 21 with 100% sensitivity (95% confidence interval [CI], 90.7%-100%), and standard screening had a sensitivity of 78.0%, identifying 30 of the 38 cases (95% CI, 62.7%-90.4%; P = 0.008). However, if one considered the 3 trisomy 21 cases among those patients with "no results," the sensitivity was really 38 of 41, or 92%, which is not that different from traditional combined screening. In women without trisomy 21, cfDNA had a false-positive rate of 0.06% (95% CI, 0.03%-0.11%), and standard screening had a falsepositive rate of 5.4% (95% CI, 5.1%-5.8%; P < 0.001). In the 11,994 low-risk pregnancies (maternal age <35 years), cfDNA had 6 false-positive results for trisomy 21 but identified all 19 cases. For trisomy 18, cfDNA had a positive predictive value of 90.0% (95% CI, 55.5%-99.7%) and a false-positive rate of 0.01% (95% CI, 0%-0.04%) compared with a 14.0% positive predictive value for standard screening (95% CI, 6.3%-25.8%) and a false-positive rate of 0.31% (95% CI, 0.23%-0.41%). For trisomy 13, cfDNA identified both confirmed cases, and standard screening identified 1 of the 2.In the detection of trisomy 21, cfDNA was found to have a higher sensitivity and specificity with a far lower false-positive rate than standard screening. Further studies on the cost-effectiveness of cfDNA are recommended.
