We read with interest the study by Porreco et al on noninvasive prenatal screening (NIPS) of common aneuploidies by massively parallel sequencing of cell-free DNA that is derived from placental trophoblasts. The NIPS results were compared with cytogenetic findings that were obtained from karyotyping of chorionic villi or amniotic fluid samples.

For trisomy 21 (T21), there were 3 false-positive (FP) and no false-negative (FN) cases; for trisomy 18 (T18) and 13 (T13), there were 3 and 2 FN cases, respectively, and no FP cases. The authors comment that discordant or discrepant results may be explained by human error in the chain of custody process.

We would like to add to the discussion that there are various biologic explanations for FP and FN results, with fetoplacental mosaicism being a primary potential mechanism. Mosaicism in the cytotrophoblast, but not the fetus, may lead to a FP NIPS result, whereas a FN may result when the fetus, but not the cytotrophoblast, contains an aneuploid cell line. Clinical reports that have described this have been published recently.

We cannot be sure that all FNs reported by Porreco et al were caused by true fetal mosaicism (TFM) type 5 (normal cytotrophoblast in presence of an abnormal fetus) because the authors did not describe whether cytogenetic methods for placental karyotyping included the combined method of direct preparation of cytotrophoblast with the long-term culture of mesenchyme. They had 3/39 T18, 2/16 T13, and 0/137 T21. This corresponds to a FN rate (FNR; 1 – specificity) of 7.7% (95% confidence interval [CI], 2.7–20.3%) for T18, 12.5% (95% CI, 3.5–36%) for T13, and 0% (95% CI, 0–2.7%) for T21. The expected FNR attributed to TFM5, as per our published report, are 1.6% (95% CI, 0.7–3.4%) for T18, 0.7% (95% CI, 0.1–4%) for T13, and 0.7% (95% CI, 0.4–1.5%) for T21. All 95% CI ranges show an overlap between the 2 studies. Therefore, the result by Porreco et al is within the FNR that we predicted.

In conclusion, TFM5 could have accounted for the FN results by Porreco et al because it corresponds to the biologic findings from our article. We would like to add this explanation, that of fetoplacental mosaicism, to the FNR of Porreco et al because it is an important contribution to the analysis of their results. This biologic phenomenon is one of the reasons that NIPS is not “currently” considered diagnostic and will never allow 100% sensitivity, despite further future technologic/bioinformatic breakthroughs.