Recently data from about 60.000 study participants were pooled in a large meta-analysis for the purpose to determine the diagnostic accuracy of NIPT RhD . However, in this systematic review, only studies were included, which evaluated lab developed (in-house) tests. In contrast, the literature dealing with the validation of commercial NIPT RhD test kits is less comprehensive. The SensiGene® RHD assay (Sequenom, San Diego, CA, USA) detects exons 4, 5 and 7 of the RHD gene, RHD psi and three sequences on the Y chromosome (SRY, TTTY, DBY) for the control of fetal DNA in male fetuses. Initially Bombard et al. described a sensitivity of 97.2%, and a specificity of 96.8% (n = 207) of the SensiGene® RHD assay taking the newborn’s serological RhD result as a reference. However, in comparison with genotype reference they observed an increased sensitivity (100.0%) and specificity 98.3% (n = 199) . In a subsequent observational study Moise et al. found one false negative result in 324 RhD positive fetuses due to mislabeling of a collection tube and 2 false positive results in 136 RhD negative fetuses when testing between 11 and 29 wg .
Protected by an exclusive patent license the Free DNA Fetal Kit® RhD (Institut de Biotechnologies Jacques Boy, Reims, France) was until recently the only real-time NIPT RhD PCR assay commercially available in Europe. In a validation study performed to receive CE-approval of this test kit Roullica-Le Sciellour et al. described a sensitivity of 100% and a specificity of > 99% (two false-positive) in 300 plasma specimen tested between 10 to 34 weeks of gestation . In addition, Londero et al. found more recently a complete concordance between the Free DNA Fetal Kit® RhD result obtained from week 11 + 6 of gestation until term and RhD phenotype at birth in all 133 cases investigated . In comparison with these studies, the FetoGnost RhD assay was evaluated with more pregnant women during this continuous performance evaluation in Vienna. Diagnostic accuracy of this test kit was as high as the diagnostic accuracy determined in larger nation-wide screening studies with in-house NIPT for RhD [14, 29, 30]. One false positive result could be explained by a maternal non-coding RHD variant, in a second case two fetuses vanished during a pregnancy with triplets and only an RhD negative fetus survived and for the third false positive result no conclusive reason was found. In this case and in one false negative case with a positive repeat test either the first test was wrong due to a technical failure or a wrong sample was tested as a consequence of human error.
Beyond RAADP anti-D administration is also required earlier in RhD negative pregnancy whenever there are clinical signs or risks for fetomaternal hemorrhage . Therefore, we were especially interested in the sensitivity and specificity of the FetoGnost assay in the first trimester. Wikman and co-workers described in 4,118 pregnancies an increase of sensitivity of a single-exon fetal RHD assay during the course of pregnancy. After exclusion of samples analyzed before 10 wg, the sensitivity was 99.3% and it increased up to 100% when results from 22 wg or later were included into the calculation . In a large study performed in seven maternity units in England Chitty et al. described a sensitivity of 99.83%, 99.67%, 99.82% and 100% at 11–13, 14–17, 18–23 and > 23 completed wg, respectively. Although false negative results were rare, they were mainly observed earlier in gestation . In contrast, another study performed in 10–14 wg did not observe false negative results in 416 serum samples (2.2% inconclusive). However, the specificity of an assay based on replicate testing of RHD exon 10 with 95.2% was lower than in the previous studies . In our performance evaluation study the sensitivity of 99.93% and specificity of 99.61% were as high as sensitivities and specificities determined in the second trimester with large-scale screening studies [14, 21, 29, 33]. Notably, our single false negative result was observed at the beginning of the second trimester.
In multiple pregnancies no false negative NIPT RhD results were described in 3 studies with overall 92 cases [30, 34, 35]. With this publication we add another 203 cases, where NIPT for RhD correctly predicted the risk of anti-D alloimmunization if that at least one newborn was RhD positive or weak D whereas no false positive genotyping results were reported in 61 cases. Therefore, targeted RAADP is reasonable also in multiple pregnancies.
In a recent review Yang et al. concluded, that false negative results are rather rare after 13 wg. . Due to the false negative case at 13 + 4 wg in our diagnostic accuracy study we support the conclusion of the German Association for Transfusion Medicine and Immunohematology (DGTI) that a single NIPT for RhD test beyond 19 wg is a safe procedure to apply targeted RAADP in the second trimester . If testing for fetal markers was not been performed in Vienna, one more false negative case would have occurred at 11 + 3 wg. Therefore we suggest, that before 20 wg a control for fetal DNA should confirm the presence of cff DNA if NIPT for RhD predicts an RhD negative fetus and if no second sample for confirmation is available due to public health economic considerations. The suitability of testing hypermethylated RASSF1A with real-time PCR as fetal marker has also been demonstrated by other groups [37–39].
In conclusion, NIPT RhD performed with the FetoGnost RhD assay delivers reliable results in the first and second trimester both in singleton and multiple pregnancies, respectively. However, even if an analytical process is 100% reliable through a maximum of automation, human errors during blood collection or labelling of blood tubes never can be completely excluded. Based on the analysis of underlying risks for the individual patient, a single NIPT for RhD can be considered sufficient for targeted anti-D prophylaxis. In contrast, if the management of an alloimmunized woman with antibodies against paternally inherited blood group antigens has to be stratified, a negative NIPT for RhD has to be confirmed from a second, independent blood drawing because errors due to mislabeling of blood tubes can never be excluded.