ABSTRACT
OBJECTIVE: To validate and evaluate the performance metrics of the high-throughput semiconductor sequencing platform, Ion Proton®, in non-invasive prenatal genetic screening (NIPS) for common fetal aneuploidies in a clinical setting. METHODS: This prospective cohort study included 2505 pregnant women from eight academic genetics laboratories (695 high risk for trisomy 21 (risk ≥ 1/250) pregnancies in a validation study, and 1810 such pregnancies, without ultrasound anomalies, in a real-life NIPS clinical setting). Outcome was available for all cases in the validation cohort and for 521 in the clinical cohort. Cell-free DNA from plasma samples was sequenced using the Ion Proton sequencer, and sequencing data were analyzed using the open-access software, WISECONDOR. Performance metrics for detection of trisomies 21, 18 and 13 were calculated based on either fetal karyotype result or clinical data collected at birth. We also evaluated the failure rate and compared three methods of fetal fraction quantification (RASSF1A assay, and DEFRAG and SANEFALCON software). RESULTS: Results from both cohorts were consistent and their gestational age was not significantly different so their data were combined to increase the sample size for analysis. Sensitivities and specificities, respectively, were as follows: for trisomy 21, 98.3% (95% CI, 93.5-99.7%) and 99.9% (95% CI, 99.4-100%); for trisomy 18, 96.7% (95% CI, 80.9-99.8%) and 100% (95% CI, 99.6-100%); and for trisomy 13, 94.1% (95% CI, 69.2-99.7%) and 100% (95% CI, 99.6-100%). Our failure rate was 1.2% initially and as low as 0.6% after retesting some of the failed samples. Fetal fraction estimation by the RASSF1A assay was consistent with DEFRAG results, and both were adequate for routine diagnosis. CONCLUSIONS: We describe one of the largest studies evaluating Ion Proton-based NIPS and the first clinical study reporting pregnancy outcome in a large series of patients. This platform is highly efficient in detecting the three most common trisomies. Our protocol is robust and can be implemented easily in any medical genetics laboratory. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd.
