Objective: To assess the performance of chromosome-selective sequencing of maternal plasma cell-free DNA (cfDNA) in non-invasive prenatal testing for trisomy 13.
Methods: Two-phase case-control study on a single plasma sample per case. The first phase was used to optimize the trisomy 13 algorithm, which was then applied to a second dataset to determine the risk score for trisomy 13 by laboratory personnel who were blinded to the fetal karyotype.
Results: In the first phase, trisomy 13 risk scores were given for 11 cases of trisomy 13 and 145 euploid cases at 11-13 weeks' gestation. The test identified seven (63.6%) cases of trisomy 13 with no false positives. The trisomy 13 algorithm was subsequently modified and the trisomy 13 risk score was > 99% in all 11 cases of trisomy 13 and < 0.01% in all 145 euploid cases. In the second phase, the new algorithm was used to generate trisomy 13 risk scores for 10 cases of trisomy 13 and 1939 euploid cases. The trisomy 13 risk scores were > 99% in eight (80.0% (95% confidence interval (CI), 49.0-94.3%)) cases of trisomy 13. In the 1939 euploid cases the risk score for trisomy 13 was < 0.01% in 1937 (99.9%), 0.79% in one, and > 99% in one. Therefore, at the predefined risk cut-off of 1% for classifying a sample as high or low risk, the false-positive rate (FPR) was 0.05% (95% CI, 0.0-0.3%).
Conclusions: Chromosome-selective sequencing of cfDNA can detect the majority of cases of trisomy 13 at an FPR of less than 0.1%.
Copyright © 2012 ISUOG. Published by John Wiley & Sons, Ltd.