A method for improving the accuracy of non-invasive prenatal screening by cell-free foetal DNA size selection

Background Non-invasive prenatal screening (NIPS) using cell-free foetal DNA (cfDNA) has been widely used for identifying common foetal aneuploidies (e.g. trisomy 21 (T21), trisomy (T18) and trisomy 13 (T13)) in clinical practice. The sensitivity and specificity of NIPS exceeds 99%, but the positive prediction value (PPV) is approximately 70% (combined T21, T18 and T13). Thus, some 30% of pregnant women who have positive NIPS results are eventually identified as normal by amniocentesis. These women therefore must undertake needless invasive tests and risk miscarrying healthy babies because of false positive NIPS results. Methods In order to achieve higher accuracy, we amended the standard NIPS (s-NIPS) protocol with an additional cfDNA size selecting step in agarose-electrophoresis. The advantage of the new method (named e-NIPS) was validated by comparing the results of e-NIPS and s-NIPS using 114 retrospective cases selected from 15,930 cases. Results Our results showed that the foetal cfDNA fraction can be enriched significantly by a size selection step. With this modification, all 98 negative cases and 9 of 11 false positive cases of s-NIPS were correctly identified by e-NIPS, resulting in an increased PPV from 71% to 77%. Additionally, a simulation test showed that e-NIPS is more reliable than s-NIPS, especially when the foetal cfDNA concentration and sequencing coverage are low. Conclusion cfDNA size selection is an important step in improving the accuracy of non-invasive prenatal screening for chromosomal abnormalities.