Clinical experience across the fetal-fraction spectrum for a non-invasive prenatal screen with low test-failure rate.

Objective To describe our clinical experience across the entire range of fetal-fraction (FF) measurements of a non-invasive prenatal screen (NIPS) that uses wholegenome sequencing (WGS). Methods We analyzed retrospectively results from 58 105 singleton pregnancies that underwent NIPS on a customized WGS platform during an 8-month period and assessed clinical test performance for trisomy 21, trisomy 18 and trisomy 13. Pregnancy outcomes were sought for all screen-positive patients and for 18% of screen-negative patients. As differences in outcome-collection response rates could artificially impact test-performance calculations, we computed inferred sensitivity, specificity, positive predictive values (PPV) and negative predictive values adjusted for ascertainment bias. Results The screening test yielded a result for 99.9% (n=58 048) of patients, meaning that approximately 1 in 1000 patients received a test failure (i.e. test failure rate=0.1%). Of pregnancies with a test result, 572 (1%) screened positive for one of the common aneuploidies (362 for trisomy 21, 142 for trisomy 18 and 68 for trisomy 13). Informative outcomes were received for 237 (41.4%) patients with a screen-positive result and 3258 (5.7%) of those with a screen-negative result. In the full cohort, inferred sensitivities for trisomy 21, trisomy 18 and trisomy 13 were 99.7%, 96.8% and 94.3%, respectively, and PPVs were 93.1%, 85.2% and 48.4%, respectively. If a FF threshold of 4% had been employed to guard against false negatives, calculated sensitivities for the three aneuploidies would not have changed significantly, yet, importantly, the overall test-failure rate would have increased to 6.6% (n=3829), impacting 1 in 15 women. Conclusions Our clinical experience demonstrates that a customized WGS-based NIPS without a FF threshold achieves high accuracy while maintaining a low test-failure rate of 0.1%. As such, alternative strategies to ensure high accuracy of detection of common aneuploidies in samples with low FF (such as redraw after test failure, redrawing at a later gestational age, risk scoring based on FF) are not necessary for this screening approach. (C) 2019 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of the International Society of Ultrasound in Obstetrics and Gynecology.