INTRODUCTION AND INDICATIONS FOR NON-INVASIVE RH TESTING

Hemolytic disease of the fetus and newborn (HDFN) is caused by maternal IgG alloantibodies against fetal red blood cells (RBCs) expressing paternally inherited antigens for which the mother is negative. Depending on the titer and the effector function of these antibodies, the fetal RBCs can be destroyed. The effector function can be determined by antibody-dependent cellular cytotoxicity (ADCC) testing, or more directly by determining the Fc-glycosylation pattern of the antibodies. Afucosylated antibodies mediate up to 100-fold more effective ADDC. There is huge variation in the level of afucosylated anti-Rh antibodies between women, and there is growing evidence that these differences greatly predict the risk of severe HDFN. Antibodies directed against the D antigen are the major cause of severe HDFN, but immunization against non-D RBC antigens became relatively more important when the incidence of anti-D immunization decreased. The specificities of non-D Rh alloantibodies detected in European pregnant women are, in order of frequency, anti-E, anti-c, anti-Cw, and anti-e. However, severe HDFN, requiring fetal blood transfusion or intensive neonatal treatment, is almost exclusively seen in pregnancies with anti-c antibodies, and rarely in pregnancies with the other (non-D/c) Rh-antibodies.

When the presence of maternal anti-Rh antibodies is detected by screening, the pregnant woman has to be monitored by repeated laboratory tests (antibody titers, antibody quantitation, ADCC) and – if these serum tests are above a certain cut-off level – by clinical investigations in specialized centers such as Doppler sonography to timely detect severe HDFN requiring fetal therapy or delivery with neonatal treatment. For the monitoring of alloimmunized women, it is important to know the fetal Rh status, to prevent unnecessary anxiety, unnecessary referrals, and diagnostic tests in women carrying an antigen-negative child. Since the fetus can become severely anemic from 16 weeks’ gestation onwards, cfDNA testing of the fetal Rh status should be done before 14–15 weeks. In the near future, fetal anemia in alloimmunized mothers may be prevented by treatment with FcRn blockers, an innovation currently investigated in clinical trials (www.clinicaltrial.gov – NCT03842189 and NCT03755128). This would mean that fetal Rh status should be determined even earlier, around 10–12 weeks, in order to start this new treatment at 13–14 weeks.

Before non-invasive Rh typing was available, the biological father was typed to determine the chance on an antigen-positive child. A likely homozygous father was considered sufficient evidence to assume that the fetus was antigen-positive. In the case of a heterozygous father, amniocentesis was performed to assess the fetal antigen status. This invasive test has the risk of fetal loss and ruptured membranes, and moreover it can also boost the immune response of the mother by FMH.

If untreated, HDFN can be fatal or may cause long-term sequelae.¹ Before the introduction of Rh immunoglobulin (RhIg) prophylaxis, HDFN was a major cause of perinatal death.² Routine postnatal administration of RhIg – introduced in the 1960s – greatly decreased maternal alloimmunization detected in a subsequent pregnancy from 15% to 1.6%.³ In the mid-1990s, antenatal RhIg became standard care for D-negative women in many high-income countries and this strategy further halved the risk of anti-D immunization.^4,5,6

The administration of postnatal RhIg is mostly guided by cord blood serology, and is only given to D-negative women who delivered a D-positive child. But for (targeted or routine) antenatal prophylaxis the determination of the fetal D status is more difficult. In many countries, it is still given to all D-negative pregnant women, including the approximately 40% of women carrying a D-negative fetus. However, RhIg is a human blood product, with an inherent risk of infectious disease transmission. In addition, there is a worldwide shortage of this product, produced by only a limited number of companies. Moreover, the success of immunoprophylaxis has resulted in a decreasing number of potential donors, who have been naturally immunized (REF). To date, anti-D plasma is donated in only a few countries in the world. In the absence of insight into the working mechanism, no equally effective recombinant anti-D antibody is available to replace the human-derived polyclonal Ig preparations. The observation of Lo et al. in 1997, that low amounts of cell-free fetal (cff) DNA, mainly derived from apoptotic placental syncytiotrophoblasts, are present in maternal plasma,⁷ made it possible to introduce non-invasive fetal RHD typing for optimized use of RhIg.^8,9 Large-scale studies from 2006 and 2008 showed that non-invasive fetal RHD typing on maternal plasma is feasible and highly reliable.^10,11,12 As the result of these successful clinical trials, several northern European countries introduced nationwide fetal RHD genotyping for the screening of non-immunized D-negative women to guide antenatal RhIg prophylaxis. Denmark was the first in 2010, followed by the Netherlands in 2011, and Finland in 2014. In 2021 there were also government-based fetal RHD genotyping programs in Sweden, Germany, Slovenia, Spain, France, and Norway.^13,14,15,16

There are both ethical and economical arguments for implementing non-invasive fetal RHD typing in these countries. From an ethical perspective, one can argue that pregnant women should not unnecessarily be given a blood product.¹⁷ Also, from the perspective of the correct use of blood products, it is a waste of the gift of donors if RhIg is given to women carrying a D-negative child. The economic arguments are different between countries.^18,19 An add-on test in countries with a centralized pregnancy screening Rh-immunization program is more rapidly cost-effective than countries with a more decentralized approach, as the equipment needed for high-throughput non-invasive screening contributes largely to the costs of the test. In a country like the Netherlands, with a single national screening lab for 30.000 D-negative pregnant women/year, the extra costs of fetal RHD typing are saved by decreased costs for RhIg and of cord blood. In contrast, in the US the consensus is still that the current approach of antenatal RhIg prophylaxis for all D-negative pregnant women should not be changed, because administration of antenatal prophylaxis is considered safe and there is no shortage of RhIg in the US. Moreover, it is feared that implementation of PCR-guided antenatal prophylaxis will cause more alloimmunizations unless the sensitivity of the cffDNA testing is 100%.²⁰ It should be emphasized however that in the US, the costs of fetal RHD testing substantially outweigh these costs in Europe.

The goal of this chapter is to give a review on the current status of non-invasive fetal RH typing, thereby providing the reader with the knowledge needed for design and/or interpretation of non-invasive fetal RH typing assays.

PRE-ANALYTICAL CONDITIONS

No loss of cffDNA was observed up to 5 days of transport, showing that cffDNA is highly stable in vitro after blood drawing.²¹ Moreover, as per definition the paternally inherited Rh-allele encoding the antigen to which the anti-Rh antibodies are present is lacking from the genome of an alloimmunized woman, fetal RH typing assays are not influenced much by the fraction of cffDNA.²² Therefore, there is, for most of the highly specific Rh assays, no need to apply tubes containing stabilizing agents, or to reject material because of hemolysis, or to monitor transport temperature. Anticoagulated blood is preferred above serum. In conclusion, the high stability of cffDNA makes non-invasive fetal RH typing feasible without extra costs for tubes or transport.

The major challenge for non-invasive RH typing is the extremely low and highly variable cffDNA concentration (up to 100-fold differences between women).²³ The failure to isolate sufficient cffDNA can result in false-negative (FN) results. The concentration of cffDNA gradually rises during pregnancy.^24,25,26 The diagnostic accuracy of non-invasive RH typing depends therefore on the plasma equivalent used as input for the assay, as well as on the timing of blood drawing. In general, it is recommended to isolate cff DNA from 500–1000 μl of maternal plasma. The first studies on non-invasive RHD typing were all performed relatively late in pregnancy, around 24–30 weeks, because that enabled the combination with the second anti-D antibody screening test and this was still in time for guiding routine antenatal prophylaxis. In more recent years, data have become available, which show that fetal RHD typing on blood samples drawn in the first trimester after 11 weeks is reliable,^27,28,29,30 and nowadays many countries (Belgium, Sweden, Spain, Germany, Poland, Ireland, and France) perform the test as early as possible, but not earlier than 10–11 weeks, to also guide targeted antenatal prophylaxis in situations at risk of fetomaternal hemorrhage at earlier gestational age.¹³ Italy and Switzerland test during the second trimester, and only the countries that were the first to introduce non-invasive RHD typing (Denmark, the Netherlands, Finland) still test just before the third trimester.

ANALYTICAL CONDITIONS

The introduction of non-invasive prenatal testing (NIPT) for fetal aneuploidy in 2011³¹ has rapidly changed the landscape of prenatal testing, and therefore nowadays various DNA extraction systems have been validated for cffDNA. Because cffDNA is small in fragment size (less than 150 bp),³² originally DNA isolation methods developed for virus DNA isolation resulted in optimal recovery.^33,34 Nowadays, commercial kits dedicated for the isolation of cell-free DNA are widely available. For screening, automated DNA extraction is preferable since it involves less error-prone handling than manual extraction. It might be considered to add a fixed amount of exogenous DNA to the plasma samples prior to extraction to control for total DNA isolation, which serves as a better control than a genomic target as the amount of maternal cfDNA largely varies between women.^35,36

For detection of fetal RH sequences, several technical approaches can be used. All initial large-scale studies on the feasibility of fetal RHD typing were based on multiplexed real-time PCR. Because D-negativity is in most individuals caused by deletion of the RHD gene, and 6.2% of the RHD genome is different from the RHCE genome, it is relatively easy to design highly RHD-specific PCR-assays and to reach the maximal limit of detection of 1 copy. In addition, the number of negative replicates around the detection limit due to Poisson’s statistics can even be further decreased by applying the same fluorochrome for multiple RHD probes.³⁷ Other methods applied to fetal RHD typing involve single-base extension, followed by demonstration of the specific products by GeneScan³⁸ or mass spectrometer.³⁹ Although these latter approaches are highly valuable for fetal antigen typing in a diagnostic setting of alloimmunized women, as they allow for including paternally inherited markers to control for the presence of fetal DNA, these platforms are not optimal with regard to costs in a screening setting. Massively parallel sequencing of cell-free DNA in pregnant women has shown to be reliable, but is not fast and still costly.^40,41 Eryilmaz et al. have successfully described a chip-based digital PCR for fetal blood group typing.⁴² Recently, the group of Hyland described a droplet digital PCR (ddPCR) approach for fetal blood group genotyping.⁴³ This is now the method of choice in alloimmunized women. In a ddPCR each droplet contains only a single copy of the maternal or fetal gene, and the presence of a fetal copy will therefore always result in a strongly positive signal that can easily be discriminated from droplets containing maternal DNA copies. A major advantage of the ddPCR above real-time PCR is that it always provides an accurate quantitative result for both fetal RH- as for fetal control-targets, even at low cffDNA concentrations. In a considerable number of pregnant women carrying a D-positive fetus, the RHD real-time PCR results are outside the quantitative range of the assay, with Ct values above 35, whereas also in women carrying a D-negative fetus, Ct values within that range might be observed. Therefore, discrimination between D-positive and D-negative results can be arbitrary, and some laboratories report in cases with high Cts an inconclusive result,⁴⁴ or stay in a screening situation at the safe side and report those as D-positive in order to prevent false-negative results.⁴⁵ This approach is highly reliable for guiding immunoprophylaxis, as false-positive results have no major clinical consequences.⁴⁶ However, false and inconclusive results should be circumvented in alloimmunized women. These false-positive or inconclusive results are not seen in the ddPCR assay, because a low level of amplification from an excess of maternal RHCE DNA is not hampering this assay, as a droplet contains maximally only one copy of the maternal RHCEgene. Theoretically, no positive droplets should be generated in plasma without RHD sequences.

There is much debate on whether a positive control for the presence of fetal DNA should be included in the screening assay.⁴⁷ Unfortunately, to date there is no universal fetal identifier that can easily be implemented in a screening setting. In several countries, negative results need to be confirmed. Genomic targets from the Y-chromosome are applicable in only 50% of the pregnancies, and some countries have chosen to try to confirm negative results with a Y-chromosome marker.⁴⁸ Hypermethylated RASSF1a is an epigenetic marker, which can discriminate between fetal and maternal DNA in all pregnancies, but its low sensitivity and specificity compared to RHD-PCRs hampers its application in a screening assay, although it is highly useful in diagnostic ddPCR assays in alloimmunized women.^49,50 An alternative possibility is the amplification of paternally inherited markers. Several in/del polymorphisms have been successfully applied in real-time PCR settings,⁵¹ but again very time consuming to implement in a screening setting based on real-time PCR.

RH TYPING ASSAYS

The Rh blood group system consists of two highly homologous genes RHD and RHCE on chromosome 1 (1p36.11), positioned in opposite directions.⁵² Assays specific for RHD, RHC/c and RHE/e are based on allele-specific nucleotides. Thirty-seven nucleotides are specific for the coding sequence of RHD and not present in any of the RHCE alleles. Five nucleotides (in exon 2) are specific for RHc, of which the 307C encoding 103Pro is best correlated with c expression.^53,54 Exon 2 of RHC is identical to exon 2 of RHD, therefore only 48C (in exon 1) (16Cys) is specific for RHC. However, 16Cys is not correlated to C-expression and 74% of African blacks have 48C in a ce-allele with normal expression of Rhc.⁵⁵ Therefore, RHC genotyping assays have to be based on a 109 bp insert in intron 2, that is only present in RHC.⁵⁶ The E/e polymorphism is caused by a single nucleotide variation 676C>G (Pro226Ala) in exon 5 of RHCE. The 676G is also present in the RHD allele, but is in RHD surrounded by seven D-specific nucleotides.⁵⁷

However, fetal RH testing is challenged by genetic variation of RH alleles. The RH locus is highly polymorphic, and many hybrid RHD/RHCE gene variants have been described. In women carrying variant RHD genes, false-positive (FP) results can be observed due to maternal RHD sequences, whereas the presence of variant RHD genes in fetuses can result in both FP or FN results. The main mechanism responsible for the generation of these so-called hybrid RH genes is thought to be gene conversion, which is explained by the opposite orientation of the homologous RHD and RHCE genes. Multiple exons can be converted, but also micro-conversion events can lead to only single amino acid changes. Furthermore, gene conversion events can also be associated with untemplated mutations, in which the mutated nucleotides are not derived from the other RH gene.⁵⁸ RHD variant genes were originally subdivided into so-called null, DEL, weak D, and partial D genes. However, since it can be difficult to recognize partial D variants serologically and also some weak D individuals, for example with weak D type 4.2 (DAR), type 11, 15, 21, or 57 (reviewed by Sandler and colleagues).⁵⁸ DEL individuals carrying the RHD*DEL8 allele or the RHD*DEL5⁵⁹ can also become immunized against D.⁶⁰ The DEL/weak D/partial D nomenclature is misleading and nowadays the term “D variant” is applied for all RHD alleles that result in qualitatively and/or quantitatively aberrant expression.⁶¹

Especially in African blacks, RH genes are highly variable. In individuals of European descent, D-negativity is almost invariably caused by the absence of the RHD gene, but in the African population deletion of the RHD gene is not the only mechanism responsible for D-negativity. About 66% of African blacks carry the RHDΨ gene and 15% of them carry a RHD-CE-Ds (r’s) hybrid allele.⁶² The RHDΨ gene (RHD*08N.01) contains a 37 base-pair duplication causing a frameshift, consisting of the last nucleotides of intron 3 and the first nucleotides of exon 4, and a nonsense mutation in exon 6 (807T>G, Tyr269stop).⁵⁷ The African RHD-CE-Ds allele consists of exon 1, 2, and 3 of RHD*DIIIa, exons 4–7 of RHCE and exons 8–10 of RHD (type 1 = RHD*03N.01),⁶³ but occasionally it is a hybrid RHD (exon1–2)-RHCE(exon3–7)-RHD(exon8–10) allele (type 2 = RHD*01N.06).⁶⁴ These two latter genes produce no D, but do produce an abnormal C antigen. Because the RHD-CE-Ds alleles miss intron 2 on which RHC genotyping assays are based, reliable RhC prediction is also complicated in African blacks. The observation that in Africa different mechanisms resulting in D-negativity have emerged might indicate that there has been an ancient (unknown) selective pressure in Africa.⁶⁵

In Asian populations (China, Korea, Japan), 15–30% of serologically typed D-negative individuals carry the so-called Asia type DEL gene: RHD*01EL.01 (RHD*1227A) gene, which has a single mutation (rs549616139) causing a splice site defect, which results in the DEL phenotype.^66,67 In DEL individuals, D expression on RBCs is extremely weak and can only be recognized by adsorption-elution. These individuals are not at risk for RhD immunization,⁶⁷ whereas DEL RBCs are immunogenic.⁶⁸ Between 3–8% of truly D-negative Asian individuals carry the D-negative hybrid RHD*D-CE(2–9)-D allele (RHD*01N.03). In addition, in Chinese D-negative individuals, the silent RHD*711delC (RHD*01N.16) allele is relatively frequent (>1%).^66,69,70,71

In women of European descent, the RHD*06 (RHD*DVI) gene is the most frequent variant RHD gene,⁷² and these women can be immunized by a D-positive fetus.

The general consensus is that pregnant women with weak D type 1, 2, and 3 and Asian type DEL should be considered as RhD positive and not at risk for alloimmunization.⁷³ Pregnant women carrying any of the other RHD variants should be categorized as D-negative and included in prophylaxis programs.⁷⁴

For fetal RHD typing various RHD targets have been used, mainly RHD exons 4, 5, 7, and 10, either alone or in combination. There is no clear evidence for a superior RHD exon detection strategy. To increase the chance of a positive detection of fetal RHD variants, detecting at least two RHD exons is recommended. The majority of data are obtained with a duplex assay targeting exons 5 and 7, which cover the RHD sequence that encodes the Rh-region harboring all known exofacial D epitopes, ensuring that fetuses carrying variant genes are not missed.⁷⁵ However, it should be mentioned that in women carrying the most common RHD variant genes, RHD*Ψ in Africans and RHD*DVI, this assay amplifies maternal exon 7 sequences and this amplification inhibits in real-time PCR assays the detection of fetal exon 5 (unpublished results). Furthermore, in these mothers, fetuses carrying a variant gene lacking exon 5 (e.g. RHD*05 (RHD*DVa) or oRHD*06 (RHD*DVI), which is relatively frequent in African blacks or Europeans, respectively, will be missed. Therefore, with this exon5/7 PCR, prediction of a D-negative fetus is not reliable in D-negative women carrying the most common RHD variant genes (1% in Dutch pregnant population⁷² and at least 66% of D-negative women of African descent).^76,77 Therefore, in most screening studies, antenatal prophylaxis is advised to these women, and cord blood serology to guide postnatal prophylaxis to women with negative exon 5 results.⁷⁸

To date, large-scale screening studies have been performed in populations of mainly European background. Before implementation of fetal RHD screening in a more mixed ethnic population, the RHD-PCR assays should be adapted to the variant RHD genes seen in especially the African population,^79,80 in order to reach the same specificity; but please note that these assays still have to be validated in large-scale studies in mixed populations.

ACCURACY

In different European countries, large-scale clinical studies have been performed to validate fetal RHD typing before it could be implemented. Invariably, these trials demonstrate high sensitivities, above 99% for testing at 10–11 weeks and 99.9% for testing at 25–28 weeks.^{81,82,83,84,85,86,87,88,89} It is furthermore important to note that similarly high sensitivities were reported in the three countries Denmark, the Netherlands, and Finland where fetal RHD typing and targeted prophylaxis were implemented in a nationwide program.^86,87,88 In these three countries, cord blood serology is now stopped and postnatal prophylaxis is only guided by the prenatal test result. In the three studies evaluating the nationwide implemented screening programs, in almost 50,000 screening tests only 21 false-negative cases (0.04%) were observed. False-negative (FN) results are typically caused by low cffDNA levels, failed DNA extraction, or human error.^87,88,90 None of the FN results were caused by RHD variants in the fetus. It seems feasible to perform antenatal RHD screening early in pregnancy, although the sensitivity decreases in very early pregnancy.^91,92 Implementation of early testing enables the targeted use of prophylaxis for potential sensitizing events, such as amniocentesis and chorionic villus sampling.

In the Dutch study, nine screening results were false-negative and in 225 cases fetal RHD results were positive and cord blood serology was negative. Because in a screening program especially FN have to be prevented, the scoring algorithm is less stringent towards false-positive (FP) results, and the specificity of the test is therefore lower than the sensitivity. Low levels of non-specific amplification, or from DNA derived from a vanishing twin,⁹³ will be reported as RHD positive. Another cause of FP or inconclusive results are variant genes in the mother or the child. And eventually, discrepancies between fetal RHD typing and cord blood serology can be caused by FN cord blood serology results.⁸⁸ In the Dutch study, in ten of the 225 presumed FP cases it was shown, by analysis of back-up plasma, that the “cord blood serology” had mistakenly been performed on maternal blood samples or cord blood mixed with maternal blood. In an additional 22 FP cases with negative cord blood serology, the serology result appeared to be false as the newborn carried an RHD variant gene with weak D expression.⁹⁰ Thus, despite nine FN tests, the net effect of implementation of non-invasive RHD typing in the Netherlands might even be a lower immunization risk.

CONCLUSIONS

Non-invasive fetal RHD typing is highly accurate and large clinical studies support its implementation for routine use. Fetal RHD typing in populations with mixed ethnicity still needs large-scale validation. The major clinical benefit of implementing fetal RHD typing is that approximately 95% of the D-negative women carrying a D-negative fetus can avoid the unnecessary administration of RhIg.

PRACTICE RECOMMENDATIONS