The first 2-year prospective audit of prenatal cell-free deoxyribonucleic screening using single nucleotide polymorphisms approach in a single academic laboratory

Abstract

Objectives: We reported a performance during an implementation of prenatal cell-free (cf) DNA screening using single nucleotide polymorphism (SNP) approach in our accredited laboratory. Methods: Prospective audit with prompt intervention was set for the processing of 2,502 samples from singleton pregnancy from August 2017 to July 2019. Risks of trisomy 21 (T21), T18, T13, monosomy X (XO), and other sex chromosome aneuploidies (SCAs) were clarified by a proprietary bioinformatics algorithm. Results: Laboratory failure occurred in 192 samples (7.7 %) as a result of inadequate sequencing (n=144), fundamental limitation of the testing (n=19), and obvious human error (n=29). Faulty setting of the calibration curve was the most common human error (n=22/29). After a redraw (n=110), 79 (71.8 %) were settled. Overall, 2,389/2,502 samples (95.5 %) were reportable. Thirty-five samples were high-risk for T21 (n=19), T18 (n=5), T13 (n=1), XO (n=3), and other SCAs (n=7), respectively. Positive predictive values calculated from either prenatal confirmatory tests or postnatal findings were 93.8 % (n=16), 100 % (n=4), 50 % (n=2), and 83.3 % (n=6) for T21, T18, XO, and other SCAs, respectively, with high sensitivity and specificity (>99.9 %). Vanishing twin was detected from 1 out of 4 samples with detected additional haplotypes. Conclusions: An overall performance of SNP-based prenatal cf-DNA screening during our initial implementation can be optimized with proactive approach. The technical know-how was a significant limiting factor for adopting the technology. The lessons learnt may be of interest to the academic laboratory considering adopting their own test instead of sending blood to a testing service of a supplier.