Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7

Contents lists available at ScienceDirect

Seminars in Fetal & Neonatal Medicine

journal homepage: www.elsevier.com/locate/siny

Review

Hemolytic disease of the fetus and newborn in the molecular era

Ross M. Fasano a, b, *
a
Transfusion, Tissue, and Apheresis, Children’s Healthcare of Atlanta, Atlanta, GA, USA
b
Grady Health System Transfusion Services, Atlanta, GA, USA

s u m m a r y
Keywords: Maternalefetal red cell antigen incompatibility can lead to alloimmunization, maternal immunoglobulin
Hemolytic disease of newborn transplacental transfer, and hemolytic disease of the fetus and newborn (HDFN). The use of routine
Rhesus blood group
antenatal anti-D prophylaxis (RAADP) has sharply decreased the incidence of and mortality from HDFN
Molecular diagnostic testing
due to RhD allosensitization. The ability to identify pregnancies/fetuses at risk of HDFN has significantly
improved due to paternal molecular RHD zygosity testing, and non-invasive fetal molecular diagnostics
for detecting putative antigen(s) (notably RhD) in fetuses utilizing cff-DNA in maternal plasma. Fetal RHD
genotyping using cff-DNA has become increasingly accurate for fetal RHD detection, prompting some
countries to implement targeted RAADP through mass screening programs of RhD-negative pregnant
women. Along with middle cerebral artery Doppler ultrasonography for predicting fetal anemia, non-
invasive fetal molecular diagnostics have greatly decreased the need for invasive diagnostic pro-
cedures in pregnancies at risk for severe HDFN. This review highlights these molecular advancements in
HDFN-related prenatal diagnostics.
© 2015 Published by Elsevier Ltd.

1. Introduction negative phenotype, the immunogenicity of the RhD antigen, and

the hemolysis potential caused by anti-D antibodies [1]. Excluding
Alloimmune hemolytic disease of the fetus and newborn RhD, non-D Rh and Kell antibodies e most notably c, E, and K an-
(HDFN) is a disorder caused by the binding of transplacentally tibodies e are most frequently implicated in severe cases of HDFN,
transmitted maternal immunoglobulin (Ig) G antibodies on pater- followed by antibodies belonging to the Duffy, Kidd, and MNS
nally inherited antigens present on fetal but absent on maternal red systems. Severe fetal anemia and hydrops fetalis are extremely
blood cells (RBCs). Antigen-negative women may have naturally unusual in ABO hemolytic disease, which is generally a less severe
occurring antibodies to certain RBC antigens (anti-A or anti-B) or disease limited to the newborn period (HDN), and therefore will not
may develop antibodies as a result of exposure to foreign RBC an- be discussed further in this review.
tigens through previous blood transfusion or by silent fetomaternal The use of postpartum anti-D Ig (also termed RhIg) in RhD-
hemorrhage during pregnancy or at delivery. Maternal IgG anti- negative women in order to prevent RhD sensitization and HDFN
bodies bind to fetal RBCs, resulting in hemolysis or suppression of in subsequent pregnancies was first successfully trialed in 1967
erythropoiesis. As a consequence, fetal and/or neonatal anemia, [2,3]. Routine use of postpartum anti-D Ig for RhD-negative women
extramedullary hematopoiesis, and neonatal hyperbilirubinemia was implemented in many countries over the next few years which
may result, with severe cases resulting in fetal loss or neonatal resulted in a sharp fall in the rate of alloimmunization to RhD from
death due to hydrops fetalis. There are three main classes of 16% to ~2% [4]. A further reduction in the sensitization rate was
alloimmune HDFN, based on the antigen(s) involved: Rh (rhesus), achieved by introducing routine antenatal anti-D prophylaxis
minor red cell antigens (i.e. Kell, Duffy, Kidd antigens) and ABO. (RAADP) during the third trimester of pregnancy (around 28
Historically, the most widespread form of severe HDFN has been weeks). Consequently, the incidence of HDFN due to anti-RhD at
due to anti-RhD antibodies owing to the frequency of the RhD- the present day is ~0.5% in RhD-negative women in industrialized
countries due to the widespread adoption of guidelines for ante-
natal and postpartum use of anti-D Ig prophylaxis [5]. The reduc-
* Departments of Clinical Pathology and Pediatric Hematology, Emory University tion in HDFN incidence has produced a similar decline in HDFN-
School of Medicine, 7105B Woodruff Memorial Building, 101 Woodruff Circle,
related mortality from 46 per 100,000 births to 1.6 per 100,000
Atlanta, GA 30322, USA. Tel.: þ1 404 727 4026.
E-mail address: [email protected].
births [4].

http://dx.doi.org/10.1016/j.siny.2015.10.006
1744-165X/© 2015 Published by Elsevier Ltd.

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006
2 R.M. Fasano / Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7

2. Molecular testing for HDFN part of exon 3), and 8 to 10 from RHD, but exons 4 to 7 (and often
part of exon 3) from RHCE. Neither RHDj nor the hybrid RHD-CE-D
The current utility of molecular testing for HDFN is broadly gene produces epitopes of the D antigen. Thus, it is critical that RHD
categorized into two distinct clinical scenarios: (a) the identifica- genotyping assays for predicting fetal RhD-positivity are configured
tion of RhD-negative mothers who are carrying RhD-negative fe- to detect wild-type RHD genes but do not detect RHDj or hybrid
tuses and thus not requiring RAADP (“targeted antenatal anti-D RHD-CE-D genes so as to minimize false-positive results, especially
prophylaxis”), and (b) paternal and fetal RhD testing in alloimmu- in populations in which African genes are likely to be present [7,8].
nized mothers to determine potential fetal risk and the need for Molecular testing for the RhC antigen is additionally chal-
further clinical intervention [6]. In addition to RhD testing, fetal lenging. Even though the expression of RhC from Rhc antigen re-
blood group antigen testing is utilized to determine non-D Rh (C/c, sults from a single nucleotide polymorphism (SNP) (307C/T) in
E/e) and minor red cell antigens in alloimmunized mothers in order exon 2 of RHCE which encodes a single amino acid substitution
to predict potential fetal risk of HDFN. (Ser103Pro) in the second extracellular loop of the RhCE protein,
molecular testing for the C antigen is complicated by the fact that
2.1. Complexity of the Rh system and implications on genotyping exon 2 of RHD and of the C allele of RHCE have identical sequences.
The main way to genotype for RhC is to detect a 109 base pair
The Rh blood group system is molecularly quite complex. It insertion in intron 2 of the RHCE gene found only in individuals
consists of numerous antigens other than RhD with more than 50 with C expression. Testing for Rhc antigen, which is the more
Rh antigens identified; the most frequent clinically significant an- widely occurring antigen implicated in severe HDFN, is more
tigens include C, c, E, e, which are encoded on the paired RHCE gene. straightforward as the C307 SNP is unique to the c allele of RHCE
All antigens of the Rh blood group system are encoded by RHD and gene. The determination of RhE/e antigens, as for many other minor
RHCE, two genes each with 10 exons in opposite orientation (50 - red cell antigens, results from a single SNP, and therefore resolving
RHD-30 e30 -RHCE-50 ). RHD and RHCE share 92% nucleotide sequence these phenotypes tends to be more straightforward [7].
homology and 96% amino acid sequence similarity due to the fact
that the two genes were evolutionarily created by a duplication 2.2. Potential of targeted RAADP using non-invasive fetal RHD
event from one gene (Fig. 1). RhD-negative Caucasian individuals genotyping
are most often homozygous for deletion of the RHD gene, which
encompasses the whole of RHD and part of each of the flanking The standard practice throughout Europe, North America, and
rhesus boxes. The resultant RhD-negative phenotype is character- Australia is for all pregnant women to be tested for ABO and RhD
ized by the absence of the whole RhD protein from the red cell type and screened for the presence of RBC antibodies at the time of
membrane; however, the RHCE-encoded antigens are present. In the first prenatal visit, with administration of RAADP at ~28 weeks
contrast, only 18% of RhD-negative black Africans are homozygous of gestation, postpartum, and at the time of any sensitizing event
for the RHD deletion. The majority (66%) of RhD-negative black during pregnancy for women found to be RhD-negative [4,9]. A
Africans have an inactive RHD pseudogene (RHDj) which has a 37 second antibody screen is taken prior to the first routine prophy-
base pair duplication in intron3/exon 4 junction, missense muta- lactic anti-D Ig injection administered, and antibody assessments
tions in exon 5, and a nonsense mutation in exon 6 leading to a should be repeated in each subsequent pregnancy. Although
premature translation stop codon. An additional 15% have a hybrid introduction of RAADP has sharply reduced the incidence of and
RHD-CE-D gene, which contains exons 1, 2, 3 (either the entire or mortality from RhD-HDFN, population genetics (in Caucasians)

Fig. 1. (A) Transmembrane model of the RhD and RhCE proteins in the erythrocyte membrane. Amino acid positions that differ between RhD and RhCE are shown as dark circles on
RhD. The amino acid changes responsible for C/c and E/e polymorphisms are shown on RhCE. (B) Inverted orientation of the RHD and RHCE genes. (C) RH locus structures of the most
frequently occurring RhD phenotypes in normal RhD-positive individuals and RhD-negative individuals based on ethnic origin. Most RhD-negative Caucasians and ~20% of in-
dividuals of African descent possess two deleted RHD genes. The RHD pseudogene (RHDj) is found in ~66% of individuals of African descent; an additional 15% have a hybrid RHD-
CE-D gene. Rhesus boxes are shown as white and gray triangles, with a resulting hybrid Rhesus box in those with the RHD deletion. Mutations in RHDj (37 base pair duplication
intron3/exon 4 junction, missenses in exon 5, nonsense in exon 6) are depicted as arrows. Figure adapted in part from references [7] and [8].

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006
 R.M. Fasano / Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7 3

indicate that ~38e40% of RhD-negative women will carry an RhD- antigen. This previously was done by blood group genotyping using
negative fetus, and therefore receive unnecessary administration of DNA isolated from uncultured amniocytes collected by amniocen-
antenatal anti-D Ig [10]. The advent of non-invasive methods of tesis; however, the development of cff-DNA molecular techniques
prenatal diagnosis using fetal DNA extracted from maternal plasma allows for a non-invasive alternative diagnostic option.
as early as the first trimester of pregnancy has led several countries Paternal zygosity can be determined serologically for all com-
in Europe to study the potential use of targeted antenatal anti-D mon blood group antigens implicated in HDFN, with the exception
prophylaxis through mass screening of RhD-negative pregnant for RhD. Historically, RHD zygosity has been determined through
women for RhD-positive circulating fetal DNA using circulating serologic testing for the gene products of the closely linked RHCE
cell-free fetal DNA (cff-DNA) molecular techniques. gene. Once the entire Rh phenotype of the father was determined,
population statistics specific to ethnicity were used through pub-
2.3. The role of molecular testing in determining fetal risk in lished frequency tables to indicate the likelihood of a heterozygous
allosensitized pregnancies state. This methodology could only estimate RHD zygosity in RhD-
positive individuals, but was necessary prior to the cloning of the
In the event that a clinically significant alloantibody is identified, RH genes in the early 1990s with the subsequent elucidation of the
or if there is a history of a previous fetus or neonate affected by molecular basis to RhD antigen expression [11e13] and the haplo-
HDFN, the initial management is to determine fetal risk of HDFN by types responsible for RhD-negative phenotypes. These discoveries
ascertaining the presence or absence of the corresponding antigen have led to the development of more direct and robust methods of
on fetal RBCs. Fig. 2 illustrates a general clinical management algo- determination of RHD zygosity molecularly which are used today.
rithm of RhD-negative non-alloimmunized, and RhD and non-RhD RHD zygosity testing by quantitative fluorescence polymerase
alloimmunized pregnancies. Determining the paternal zygosity is chain reaction (QF-PCR) is commercially available and utilizes RHD-
optimally the first step in the management of an alloimmunized specific (exons 5 and 7) to RHCE-specific (exon 7) amplification
pregnancy, because if the father is homozygous for the putative ratios along with the 37 base pair insertion associated with RHDj to
antigen, the fetus is 100% at risk for HDFN and therefore needs determine RHD copy number. Although suitable for use in both
further antenatal monitoring with serial antibody titers and Doppler Caucasian and ethnic African individuals, this molecular technique
determinations of the middle cerebral artery peak systolic velocity has a false-positive rate of 1% due to rare RHD alleles that are not
for predicting fetal anemia. Conversely, a fetus of an antigen- expressed, and a false-negative rate of 1% due to rare partial D al-
negative mother and a heterozygous antigen-positive father has a leles (i.e. DBT type 1, type II) which lack RHD exons 5 and 7 but still
50% chance of being antigen positive and thus being affected by express RhD epitopes [14e16]. Alternatively, RBC genotyping can be
maternal alloimmunization. In these cases, further evaluation is used when typing for antigen systems where serological reagents
required to determine whether the fetus carries the implicated are rare or non-existent.

Fig. 2. Algorithm for the clinical management of non-alloimmunized RhD-negative, and alloimmunized RhD and non-RhD pregnancies. *Targeted RAADP is a proposed alternative
option to universal RAADP for all RhD-negative pregnant women. yIn addition to RBC antibody identification, antibody titers are determined and repeated every four weeks until 24
weeks of gestation, and every two weeks thereafter when a fetus is determined at risk. zPaternal antigen detection and zygosity testing can be done either serologically or
molecularly for most non-D Rh and minor RBC antigens. xNon-invasive fetal molecular diagnostic approaches are preferred to invasive methods which include: amniocentesis,
cordocentesis, or chorionic villus sampling. Cff-DNA, cell-free fetal DNA; RAADP, routine antenatal anti-D prophylaxis; RBC, red blood cell; HDFN, hemolytic disease of the fetus and
newborn; MCA, middle cerebral artery; MoM, multiples of the median; Ag, antigen; Hct, hematocrit [9,11].

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006
4 R.M. Fasano / Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7

When the father is heterozygous or paternal zygosity is un- Although incorrectly genotyping a fetus as RHD positive (false
known, fetal RBC antigen determination is required through fetal positive) is undesirable, a more consequential scenario is falsely
blood group genotyping in order to plan further management. genotyping a fetus as RHD negative (false negative), which can lead
There are several sources of fetal DNA for fetal blood group geno- to deferring RAADP resulting in potentially allosensitizing an RhD
typing. These traditionally included fetal cells obtained by amnio- negative woman; or falsely reassuring and thus precluding appro-
centesis, cordocentesis, and chorionic villus sampling (CVS) and priate antenatal monitoring in an allosensitized pregnancy. False-
cervical tissue obtained by transvaginal lavage. Each has risks to the negative results may occur due to low levels of cff-DNA or from
fetus and issues related to quality of sample. Whereas cervical fetal inheritance of partial D or weak D phenotypes. False-negative
tissue obtained by transvaginal lavage is limited by the low quan- results stemming from low levels of cff-DNA have been addressed
tity of fetal DNA and the possibility of contamination by maternal by incorporating internal controls for detecting the presence of
DNA, amniocentesis, cordocentesis, and CVS carry a risk of feto- fetal DNA. Internal controls such as Y-chromosome gene sequences
maternal hemorrhage with an increased risk of augmenting (SRY, DBY, and TTTY2) in RhD-negative male fetuses, and short
maternal sensitization, and of fetal loss [17,18]. Amniocytes have tandem repeats and hypomethylated promoter sequences of fetal
historically been the preferred source of DNA for fetal blood group genes have been effectively used as internal fetal markers [6,11,28].
genotyping due to an improved safety profile of amniocentesis Furthermore, RTePCR-based fetal RHD detection analysis is done in
relative to the alternative methods; however, even amniocentesis duplicate at minimum, and sometimes triplicate using at least two
carries a 0.3e1% risk of fetal loss and 3e17% transplacental hem- RHD-specific exon primers (most commonly exons 5, 7, and/or 10)
orrhage which can boost the maternal antibody titers in alloim- which maximizes accuracy of the testing [29]. Fetal inheritance of
munized pregnancies [10,19]. The advent of non-invasive methods partial D or weak D phenotypes may be the result of various hybrid
of prenatal diagnosis using fetal DNA extracted from maternal RHD-RHCE genes or by SNPs within the RHD gene. Although most of
plasma as early as the first trimester of pregnancy has obviated these partial D or weak D phenotypes are poorly immunogenic
those concerns and has greatly improved the ability to perform during pregnancy, there are some notable partial D phenotypes
molecular testing on fetal tissue [20]. (DVa, DVI type 3) that are potentially immunogenic and have been
associated with severe HDFN [30,31]. However, by using at least
3. Fetal DNA testing from maternal blood two RHD-specific exon primers, a positive or at least inconclusive
result will prevent a false-negative event in the vast majority of
The concept of non-invasive fetal blood group genotyping for events.
targeted antenatal anti-D prophylaxis and fetal blood group antigen
determination in alloimmunized mothers became a possibility in 4. Accuracy and large-scale feasibility of non-invasive fetal
the late 1990s when Lo and colleagues demonstrated that circu- RBC genotyping
lating fetal DNA, released by villous trophoblasts within the
placenta, was present in maternal plasma. Indeed, it was observed The advancements in non-invasive fetal RHD genotyping using
that cff-DNA was present at 3.4% (range: 0.39e11.9%) and 6.2% cff-DNA at 16e20 weeks of gestation have allowed for fetal RhD
(range: 2.3e11.4%) of the concentration of maternal DNA in determination with exceptionally reliable diagnostic accuracy. A
maternal plasma in the first and third trimester respectively [21], meta-analysis published in 2006 reviewed 37 publications report-
and that cff-DNA could be detected as early as 4 weeks of gestation ing on non-invasive RHD genotyping using either cff-DNA or fetal
[22]. Confirming these findings, investigators in the UK and The DNA from fetal-derived cells within maternal blood, and included
Netherlands subsequently showed that this amount of cff-DNA was results from more than 3000 maternal blood samples from Rh-
present in sufficient quantities in maternal plasma at 20 weeks of negative alloimmunized (n ¼ 783) and non-alloimmunized
gestation to determine the presence of the RHD gene employing (n ¼ 2478) pregnant patients. The meta-analysis demonstrated an
real-time quantitative polymerase chain reaction (RTePCR) tech- overall accuracy of 91.4% (2980/3261 correctly typed). However, the
nology and RHD-specific primers [1,23e25]. Importantly, the accuracy was 94.8% [32] when excluding: studies reporting on <10
demonstration that the mean half-life for circulating fetal DNA is on patients (28/3261; 0.86%), samples where there was an absence of
average less than 30 min by Lo and colleagues [26] suggests that fetal DNA (56/3261; 1.7%), and two studies' samples from women
plasma fetal DNA analysis is quite insusceptible to false-positive exhibiting nonfunctional/rearranged RHD genes because they did
results from previous pregnancies, especially when samples are not include RHD-specific primers to discriminate between wild-
processed fresh by centrifugation followed by filtration or micro- type RHD and RHDj or hybrid RHD-CE-D pseudogenes (44/3261;
centrifugation when producing cell-free plasma, so as to remove all 1.3%). The diagnostic accuracy using cff-DNA from maternal plasma
cellular material which may have very small amounts of engrafted was noted to be significantly better (96.5%) than when DNA (or
fetal cells from previous pregnancies in maternal lymphoid organs RNA) from fetal cells in maternal blood were used as the source. In
or bone marrow [27]. addition, the highest accuracy of testing was demonstrated in the
RTePCR-based detection of fetal RHD has subsequently become first trimester e the time this testing would be most clinically
remarkably reliable despite the complexity of the RH gene. Fetal useful [32].
DNA detection has been enhanced by improved and automated Presently, the level of accuracy is much higher in reference
DNA extraction techniques, and an increased understanding of the laboratories which provide this testing as a routine service, such as
genetic background of RhD has facilitated strategically designed those used in Denmark, the Netherlands, and the UK. In these
assays to account for the many variants of the RHD gene in many settings, the accuracy of detection is particularly high, and the risk
different ethnic backgrounds. False-positive results can occur due of a false-negative RHD result with current cff-DNA techniques has
to presence of the inactive African pseudogenes (RHDj or the become exceedingly small with accuracy and false-negative rates
hybrid RHD-CE-D) in either the maternal or fetal genetic back- reported between 95.7e99.4% and 0.1e0.2% respectively in recent
ground. In order to circumvent this potential for false-positives in large-scale feasibility studies from the Netherlands [33], UK [34],
individuals carrying either RHDj or the hybrid RHD-CE-D, RHD- and Denmark [29]. In fact the most recent studies show >99.8%
specific primers in either exons 4 or 5 are used that discriminate sensitivity when testing is performed at 16 weeks of gestation or
between wild-type RHD nucleotide sequences and SNPs in those later [1,33e35]. As such, RHD cff-DNA testing is commercially
exons in both RHDj or the hybrid RHD-CE-D. available, and is being increasingly utilized in the USA [36,37], the

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006
 R.M. Fasano / Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7 5

UK and Europe [38], with Denmark and the Netherlands already analysis by mass spectrometry, and it is suitable for high-
having introduced the routine use of fetal RHD typing for all RhD- throughput analysis of several thousand samples per day [43].
negative pregnant women [4]. This approach has been shown to accurately determine fetal
Although there are only a few published reports on non-invasive RhD status using cff-DNA. In a report validating the implementa-
fetal genotyping using cff-DNA from maternal plasma to determine tion of MALDI-TOF MS-based, non-invasive fetal RHD genotyping
for Rh C, c, E, e, and Kell (K) status, the emerging data suggests that within a CLIA-certified, CAP-accredited, specialty reference labo-
these antigens can be predicted with equally high accuracy rates as ratory, Bombard and colleagues reported an overall test accuracy of
RhD [39]. In a meta-analysis published in 2008, which reviewed 99.5%, with no false-negative and one false-positive result in 199
RhC/c and E/e genotyping protocols using DNA obtained from plasma samples from pregnant women at 6e30 weeks of gestation
maternal blood for a total of 369 samples, the combined accuracy of [38]. This MALDI-TOF MS-based assay utilizes PCR and extension
fetal genotype was 96.3% for C/c and 98.2% for E/e. However, RhCE reaction primers that detect RHD exons 4, 5, and 7, the RHDj-
determination using cff-DNA obtained from maternal plasma was pseudogene, as well as internal fetal DNA controls, and is now
100% accurate, albeit with a small number of samples analyzed commercially available in the USA [36,37]. Fetal KEL1 antigen
(nTotal ¼ 300; nC/c ¼ 147, nE/e ¼ 153) [40]. Results of fetal blood detection by MALDI-TOF MS-based single allele-based extension
group testing for Kell using cff-DNA in maternal plasma at the In- reaction from cff-DNA in maternal plasma has been described in a
ternational Blood Group Reference Laboratory (IBGRL) in the UK small group of pregnant women (n ¼ 32) in their second and third
from 2002 to 2009 demonstrated one false positive and one false trimesters (median: 21.5 weeks) with a reported accuracy of 94%
negative in 554 samples tested (accuracy 99.6%) using locked [44], but requires further evaluation on a larger scale prior to being
nucleic acid allele-specific K/k primers. The authors mentioned that clinically applicable.
the false-negative sample was obtained at 17 weeks of gestation Other diagnostic approaches that may provide additional op-
and had a particularly low yield of total DNA, which prompted a portunities for accurate assessment of fetal risk for HDFN include
change in their policy that K testing not be performed before 20 digital droplet PCR, multiplex ligation-dependent probe amplifi-
weeks of gestation and that a K-negative result obtained before 28 cation, and possibly next-generation sequencing which currently is
weeks be repeated at 28 weeks or later [10]. comparatively expensive but continues to fall in cost [6].
Non-invasive detection of the pregnancies at risk of HDFN using
cff-DNA has become a clinical reality because of continued ad- 6. Conclusions
vances in molecular techniques and fine tuning of genotyping
protocols since 1997. These refinements have improved accuracy Although HDFN remains a clinically significant problem that
and have made large-scale non-invasive fetal molecular blood may potentially affect any pregnancy, there have been many ad-
group testing more cost effective. In fact, it has been suggested that vances in the methods for identifying and monitoring fetuses at risk
the cost of cff-DNA testing from maternal blood using automated of severe hemolytic disease which have tremendous impact on
technology may be less than the cost of antenatal anti-D IgG perinatal morbidity and mortality. In particular, serial middle ce-
[32,41]. Thus, numerous clinical services employing non-invasive rebral artery (MCA) Doppler ultrasonography evaluations have
prenatal diagnostics (NIPD) have arisen across Europe and many replaced serial amniocenteses for the prediction of fetal anemia
large-scale NIPD genotyping studies are being conducted using because there are virtually no risks to the fetus or mother from this
maternal plasma. Because of the rapidly expanding interest in non-invasive method and because peak systolic velocity MCA
NIPD, the Special Non-Invasive Advances in Fetal and Neonatal measurements have been proven to be a more accurate assessment
Evaluation (SAFE) Network of Excellence was developed to imple- of anemia than indirect measures of amniotic fluid bilirubin. In
ment routine, cost-effective NIPD and neonatal screening through addition, non-invasive fetal molecular diagnostic approaches have
the creation of long-term international partnerships to standardize advanced the care of fetuses at risk for HDFN. DNA determination of
methodology and ensure universal quality of results. The SAFE paternal zygosity for RhD has replaced the serologic-based
network of excellence has and continues to play a major role in the methods which only estimate zygosity based on population sta-
standardization of non-invasive RHD genotyping methodology and tistics specific to ethnicity, and which are imprecise and particu-
establishing international external quality assurance policies larly problematic in cross-ethnic couples. The use of non-invasive
through International Society of Blood Transfusion (ISBT) work- fetal molecular diagnostic approaches utilizing cff-DNA for
shops [1,42]. detecting putative antigen(s) in fetuses (notably RhD) of alloim-
munized pregnancies reduces the need for invasive procedures for
5. New non-invasive fetal molecular diagnostic approaches fetal blood group antigen determination, which carry a risk of
for clinical management of HDFN fetomaternal hemorrhage that can augment maternal sensitization,
and of fetal loss. Furthermore, non-invasive fetal RHD genotyping
Quantitative RTePCR using allele-specific primers has been the has become increasingly accurate for RhD detection with
diagnostic methodology of choice for the determination of fetal adequately low false-negative rates, which is necessary for imple-
blood group genotype because of its high degree of accuracy, low menting targeted antenatal anti-D prophylaxis through mass
cost, ease of use, widespread abundance of real-time PCR instru- screening programs of RhD-negative pregnant women for RhD-
mentation, and ability to automate and expand for high- positive fetal DNA using cff-DNA. In fact, some European coun-
throughput application. However, alternative methods have also tries have already introduced the routine use of fetal RHD typing for
emerged for clinical use. A method combining PCR, base extension all RhD-negative pregnant women based on justifications of the
reaction, and matrix-assisted laser desorption ionization/time of cost-effectiveness of targeted RAADP (which may be debated) but
flight mass spectrometry (MALDI-TOF MS) has been shown to be an also based on ethical grounds of preventing the unnecessary
effective alternative non-invasive fetal molecular diagnostic administration of a human-derived product to ~40% of women who
approach for prenatal blood group determination. The MALDI-TOF do not require it. Lastly, although there are a number of strategies in
MS approach offers several advantages over quantitative RTePCR place to prevent RhD HDFN, few options exist to prevent the
methodology for NIPD. It permits the reliable detection of many development of non-RhD HDFN. Researchers continue to investi-
fetal DNA point mutations from the maternal counterpart by gate strategies to prevent primary maternal RBC alloimmunization
examining a large number of multiplex PCR products in a single to RhD and to other blood group antigens, as well as to mitigate the

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006
6 R.M. Fasano / Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7

dangers of existing maternal RBC alloantibodies. Through the [11] Moise Jr KJ, Argoti PS. Management and prevention of red cell alloimmuni-
zation in pregnancy: a systematic review. Obstet Gynecol 2012;120:1132e9.
continued collaborative efforts of maternalefetal medicine spe-
[12] Garratty G, Glynn SA, McEntire R. ABO and Rh(D) phenotype frequencies of
cialists, hematologists, transfusion medicine physicians, radiolo- different racial/ethnic groups in the United States. Transfusion 2004;44:703e6.
gists, neonatologists, and researchers in molecular diagnostics and [13] Eder AF. Update on HDFN: new information on long-standing controversies.
red cell alloimmunization, the care of fetuses and infants at risk for Immunohematol/Am Red Cross 2006;22:188e95.
[14] Pirelli KJ, Pietz BC, Johnson ST, Pinder HL, Bellissimo DB. Molecular determi-
HDFN, and their mothers, will likely continue to improve in future nation of RHD zygosity: predicting risk of hemolytic disease of the fetus and
years. newborn related to anti-D. Prenat Diagn 2010;30:1207e12.
[15] Blood Center of Wisconsin Diagnostics. RhD zygosity testing. Available at:
http://www.bcw.edu/cs/groups/public/documents/documents/mdaw/mda0/
~edisp/rhd_zygosity_desc.pdf [accessed 01.08.15].
[16] ARUP Laboratories. RhD antigen (RHD) genotyping. Available from: http://ltd.
Practice points aruplab.com/Tests/Pdf/242 [accessed 27.07.15].
[17] Jansen MW, Brandenburg H, Wildschut HI, Martens AC, Hagenaars AM,
Wladimiroff JW, et al. The effect of chorionic villus sampling on the number of
 Advances in non-invasive methods for identifying and fetal cells isolated from maternal blood and on maternal serum alpha-
monitoring fetuses at risk of severe HDFN have made a fetoprotein levels. Prenat Diagn 1997;17:953e9.
[18] Bowman JM, Pollock JM. Transplacental fetal hemorrhage after amniocentesis.
tremendous impact on perinatal morbidity and mortality.
Obstet Gynecol 1985;66:749e54.
 Non-invasive fetal molecular diagnostic approaches uti- [19] Denomme GA, Fernandes BJ. Fetal blood group genotyping. Transfusion
lizing cff-DNA in maternal plasma have advanced the care 2007;47:64Se8S.
of fetuses at risk for HDFN by reducing the need for [20] Lo YM, Bowell PJ, Selinger M, Mackenzie IZ, Chamberlain P, Gillmer MD, et al.
Prenatal determination of fetal RhD status by analysis of peripheral blood of
invasive procedures for detecting putative antigen(s) rhesus negative mothers. Lancet 1993;341:1147e8.
(notably RhD) in fetuses of alloimmunized pregnancies. [21] Lo YM, Tein MS, Lau TK, Haines CJ, Leung TN, Poon PM, et al. Quantitative
 Non-invasive fetal RHD genotyping using cff-DNA in analysis of fetal DNA in maternal plasma and serum: implications for
noninvasive prenatal diagnosis. Am J Hum Genet 1998;62:768e75.
maternal plasma has become increasingly accurate for [22] Illanes S, Denbow M, Kailasam C, Finning K, Soothill PW. Early detection of
RhD detection with adequately low-false negative rates, cell-free fetal DNA in maternal plasma. Early Hum Dev 2007;83:563e6.
which has allowed for the implementation of targeted [23] Finning KM, Martin PG, Soothill PW, Avent ND. Prediction of fetal D status
from maternal plasma: introduction of a new noninvasive fetal RHD geno-
antenatal anti-D prophylaxis for RhD-negative pregnant typing service. Transfusion 2002;42:1079e85.
women in some countries. [24] Grootkerk-Tax MG, Soussan AA, de Haas M, Maaskant-van Wijk PA, van der
 Few options exist to prevent the development of non-RhD Schoot CE. Evaluation of prenatal RHD typing strategies on cell-free fetal DNA
from maternal plasma. Transfusion 2006;46:2142e8.
HDFN. [25] Faas BH, Beuling EA, Christiaens GC, von dem Borne AE, van der Schoot CE.
 Further research is needed to investigate strategies to Detection of fetal RHD-specific sequences in maternal plasma. Lancet
prevent primary maternal RBC alloimmunization to RhD 1998;352:1196.
[26] Lo YM, Zhang J, Leung TN, Lau TK, Chang AM, Hjelm NM. Rapid clearance of
and to other blood group antigens, as well as to develop
fetal DNA from maternal plasma. Am J Hum Genet 1999;64:218e24.
approaches to mitigate the dangers of existing maternal [27] Chiu RW, Poon LL, Lau TK, Leung TN, Wong EM, Lo YM, et al. Effects of blood-
RBC alloantibodies. processing protocols on fetal and total DNA quantification in maternal plasma.
Clin Chem 2001;47:1607e13.
[28] Scheffer PG, de Haas M, van der Schoot CE. The controversy about controls for
fetal blood group genotyping by cell-free fetal DNA in maternal plasma. Curr
Opin Hematol 2011;18:467e73.
Conflict of interest statement [29] Clausen FB, Christiansen M, Steffensen R, Jørgensen S, Nielsen C, Jakobsen MA,
et al. Report of the first nationally implemented clinical routine screening for
fetal RHD in D-pregnant women to ascertain the requirement for antenatal
None declared. RhD prophylaxis. Transfusion 2012;52:752e8.
[30] Mayne K, Bowell P, Woodward T, Sibley C, Lomas C, Tippett P. Rh immuni-
zation by the partial D antigen of category DVa. Br J Haematol 1990;76:537e9.
Funding sources
[31] Leschek E, Pearlman SA, Boudreaux I, Meek R. Severe hemolytic disease of the
newborn caused by anti-Gonzales antibody. Am J Perinatol 1993;10:362e4.
None. [32] Geifman-Holtzman O, Grotegut CA, Gaughan JP. Diagnostic accuracy of
noninvasive fetal Rh genotyping from maternal blood e a meta-analysis. Am J
Obstet Gynecol 2006;195:1163e73.
References [33] Van der Schoot CE, Soussan AA, Koelewijn J, Bonsel G, Paget-Christiaens LG,
de Haas M. Non-invasive antenatal RHD typing. Transfus Clin Biol 2006;13:
[1] de Haas M, Finning K, Massey E, Roberts DJ. Anti-D prophylaxis: past, present 53e7.
and future. Transfus Med (Oxf, Engl) 2014;24:1e7. [34] Finning K, Martin P, Summers J, Massey E, Poole G, Daniels G. Effect of high
[2] Zipursky A, Israels LG. The pathogenesis and prevention of Rh immunization. throughput RHD typing of fetal DNA in maternal plasma on use of anti-RhD
Can Med Assoc J 1967;97:1245e57. immunoglobulin in RhD negative pregnant women: prospective feasibility
[3] Hamilton EG. Prevention of Rh isoimmunization by injection of anti-D anti- study. BMJ 2008;336:816e8.
body. Obstet Gynecol 1967;30:812e5. [35] Scheffer PG, van der Schoot CE, Page-Christiaens GC, de Haas M. Noninvasive
[4] Qureshi H, Massey E, Kirwan D, Davies T, Robson S, White J, et al. BCSH fetal blood group genotyping of rhesus D, c, E and of K in alloimmunised
guideline for the use of anti-D immunoglobulin for the prevention of pregnant women: evaluation of a 7-year clinical experience. Br J Obstet
haemolytic disease of the fetus and newborn. Transf Med (Oxf, Engl) 2014;24: Gynaecol 2011;118:1340e8.
8e20. [36] SensiGene® Fetal RHD Genotyping. Sequenom Laboratories. Available at:
[5] Urbaniak SJ, Greiss MA. RhD haemolytic disease of the fetus and the newborn. https://laboratories.sequenom.com/providers/fetal-rhd-genotyping/[accessed
Blood Rev 2000;14:44e61. 27.07.15].
[6] Avent ND. Prenatal testing for hemolytic disease of the newborn and fetal [37] ARUP Laboratories. Fetal non-invasive prenatal testing for RhD genotyping.
neonatal alloimmune thrombocytopenia e current status. Expert Rev Hematol Available at: http://ltd.aruplab.com/Tests/Pub/2009077[accessed 27.07.15].
2014;7:741e5. [38] Bombard AT, Akolekar R, Farkas DH, VanAgtmael AL, Aquino F, Oeth P, et al.
[7] Westhoff CM. The structure and function of the Rh antigen complex. Semin Fetal RHD genotype detection from circulating cell-free fetal DNA in maternal
Hematol 2007;44:42e50. plasma in non-sensitized RhD negative women. Prenat Diagn 2011;31:802e8.
[8] Avent ND. RHD genotyping from maternal plasma: guidelines and technical [39] Finning K, Martin P, Summers J, Daniels G. Fetal genotyping for the K (Kell)
challenges. Meth Mol Biol 2008;444:185e201. and Rh C, c, and E blood groups on cell-free fetal DNA in maternal plasma.
[9] American College of Obstetricians and Gynecologists. ACOG Practice Bulletin Transfusion 2007;47:2126e33.
No. 75: management of alloimmunization during pregnancy. Obstet Gynecol [40] Geifman-Holtzman O, Grotegut CA, Gaughan JP, Holtzman EJ, Floro C,
2006;108:457e64. Hernandez E. Noninvasive fetal RhCE genotyping from maternal blood. Br J
[10] Daniels G, Finning K, Martin P, Massey E. Noninvasive prenatal diagnosis of Obstet Gynaecol 2009;116:144e51.
fetal blood group phenotypes: current practice and future prospects. Prenat [41] Daniels G, Finning K, Martin P. Noninvasive fetal blood grouping: present and
Diagn 2009;29:101e7. future. Clin Lab Med 2010;30:431e42.

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006
 R.M. Fasano / Seminars in Fetal & Neonatal Medicine xxx (2015) 1e7 7

[42] Daniels G, van der Schoot CE, Olsson ML. Report of the fourth International [44] Li Y, Finning K, Daniels G, Hahn S, Zhong X, Holzgreve W. Noninvasive
Workshop on molecular blood group genotyping. Vox Sang 2011;101:327e32. genotyping fetal Kell blood group (KEL1) using cell-free fetal DNA in
[43] Ding C. Maldi-TOF mass spectrometry for analyzing cell-free fetal DNA in maternal plasma by MALDI-TOF mass spectrometry. Prenat Diagn 2008;28:
maternal plasma. Meth Mol Biol 2008;444:253e67. 203e8.

Please cite this article in press as: Fasano RM, Hemolytic disease of the fetus and newborn in the molecular era, Seminars in Fetal & Neonatal
Medicine (2015), http://dx.doi.org/10.1016/j.siny.2015.10.006