The Prevention And Treatment Of Hdn Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Critically evaluate novel approaches to prevention and treatment of HDN due to anti-D, in terms of the basic underlying principles, current limitations and potential to change the management of Rh D-negative women in pregnancy

In 1609 a French midwife first reported haemolytic disease of the newborn (HDN) in a set of twins. In 1932, Diamond et al described the relationship between foetal hydrops, jaundice, anaemia and erythroblasts in the circulation, a condition given the name erythroblastosis fetalis. Levine and Stetson subsequently determined the cause after Landsteiner and Weiner discovered the Rh blood group system in 1940. In 1953, Chown later confirmed the pathogenesis of Rh alloimmunisation to be the result of passage of Rh-positive foetal red blood cells (RBC) after transplacental haemorrhage into maternal circulation that lacked the corresponding antigen. Currently, anti-D is still one of the most common antibodies found in pregnant women, followed by anti-K, anti-c and anti-E.

In the 1960s, groups from the United Kingdom and the United States independently demonstrated that anti-D immunoglobulin G (IgG) prophylaxis administered soon after delivery prevented sensitisation in Rh-negative women (Finn 1961 & Freda 1964). By 1971, the World Health Organisation (WHO) recommended that a dose of 125IU of anti-D IgG should be given for every 1 mL of fetomaternal haemorrhage of Rh-positive RBCs. In 1998, it was recommended by the American Association of Blood Banks and the American College of Obstetrics and Gynaecologists that anti-D prophylaxis was also given at 28 weeks' gestation. The routine use of prophylaxis resulted in a significant decrease in the incidence of RhD alloimmunisation (Figure x).

Prevention of HDN

Administrating anti-D immunoglobulin to D-negative women during and after pregnancy is highly effective at preventing D immunisation by D-positive foetal red cells (Bowman 2003). Anti-D immunoglobulin is produced by fractionation of IgG from pooled plasma of donors with high titre anti-D (Kumpel 2007). In the 1960s and 1970s most anti-D came from women immunised by pregnancy, however most of these donors have now retired and most of today's donors are deliberately immunised men. Some countries have even stopped manufacturing anti-D because of the ever increasing production costs. In the UK, plasma for anti-D is sourced from North America to minimise any risk of transmission of variant Creutzfeldt-Jakob disease. Therefore, an alternative safe, cheap and unlimited supply of anti-D would be desirable.

Many monoclonal anti-D antibodies (mAb) and some recombinant antibodies (rAb) have been produced by a number of techniques (Kumpel 2007). The first mAbs were produced from Epstein-Barr virus (EBV)-transformed B-lymphoblastoid cell lines (B-LCL) (Kumpel 1989). However, B-LCLs secreting monoclonal anti-D show a huge variability in growth rate and antibody production and most are not suitable for largescale clinical development. Researches then decided to fuse anti-D secreting B cells or B-LCLs with murine myeloma cells to create heterohybridoma (HH) cell lines. These are known to be less stable than B-LCL and require recloning but they grow quicker and secrete higher concentrations of anti-D into the culture media (Thompson 1990).

In clinical trials performed with human B-LCL-derived mAbs results were variable with regards to clearance of D-positive red cells although most were thought to be inferior to polyclonal anti-D. Some did not mediate any clearance whilst in others the removal of red cells was slow. Thomson et al (1990) showed that BRAD-3 appeared to clear precoated autologous red cells quite well but less rapidly when anti-D was injected separately into subjects while MONO-D did not clear it at all (Beliard 2006). It is thought that BRAD-3 and BRAD-5 would be necessary for comparable efficacy to polyclonal anti-D (Kumpel 1995). Anti-D mAbs derived from mouse HH cell lines were either less effective than polyclonal anti-D (Urbaniak 1998) or required very high, saturating levels of anti-D to clear the injected cells (Belkina 1996).

By the mid-1990s many recombinant therapeutic antibodies had been engineered and expressed in rodent cell lines, mostly Chinese hamster ovary (CHO) (Kumpel 2007). Clearance mediated by CHO rAbs was very variable, generally slower than polyclonal anti-D and did not correlate with the amount of anti-D on red cells. The rate of removal of red cells by the rat YB2/0-derived FOG-1 (Armour 2006) and LFB R297 (Cortey 2006) was extremely fast, more rapid than would be expected with polyclonal anti-D and complete by 200 minutes. However, removal of red cells using FOG-1 was associated with release of some radioactivity into the plasma, indicating haemolysis. Clearance to the liver as well as the spleen was recorded in two of three subjects and two subjects had chills 1 hour after injection and aches the following day (Armour 2006).

Many of the clinical trials of red cell clearance were followed by tests to detect serological anti-D responses for at least 6 months in the recipients, to determine whether the mAb or rAb had protected the D-negative subjects from immunisation by D-positive red cells. Again variable results were achieved. Two human anti-D mAbs (BRAD-3 and BRAD-5) prevented D immunisation (Chapman 2007), however in contrast, several anti-D mAbs produced by HH cell lines increased the immune response (Kumpel 2007).

The results of these clinical trials have left researchers with more questions than answers. Ultimately the main one, "why is no anti-D mAb or rAb quite as good as polyclonal anti-D?" still needs a lot more investigating. Kumpel suggests that unnatural glycosylation of monoclonal anti-D may cause these unsatisfactory results. For some antibodies, unusual oligosaccharides on anti-D may affect binding to Fc receptors leading to reduced red cell clearance. For others, non-human glycoforms of anti-D may bind to innate immune recognition molecules increasing inflammatory reactions. Either way polyclonal anti-D is still the gold standard and, if logistically possible, will be used until a better solution is found.

Identifying foetus' at risk of HDN - foetal blood group genotyping of DNA from maternal plasma

When a D-negative woman with anti-D is pregnant, it is beneficial to know the D phenotype of her foetus. Currently, women at risk of alloimmunisation undergo an indirect antiglobulin test and antibody quantitation or titre at their first antenatal visit. If positive, a paternal blood group and genotype should be obtained. If the father is negative for the corresponding antigen, the foetus is not at risk from HDN and no further invasive procedures are required. Serial maternal quantitation/titres are suggested if the father is positive for the corresponding antigen (BCSH guidelines).

Foetal genotyping from maternal plasma was introduced at the International Blood Group Reference Laboratory (IBGRL) in May 2001. In 1998, Lo et al demonstrated cell-free foetal DNA in plasma and serum from pregnant women. This has led to the development of non-invasive real-time polymerase chain reaction (PCR) assays to detect foetal RHD during pregnancy when the father is heterozygous for the D antigen. Determining the foetal Rh genotype is also possible by using PCR on DNA from amniocytes obtained by amniocentesis, however this procedure is highly invasive and involves a 0.5-1% risk of spontaneous abortion. Furthermore, amniocentesis carries a 17% risk of causing transplacental haemorrhage, which could ultimately boost the maternal antibody, increasing the risk of severe HDN (Daniels, 2004).

Of individuals who are D+, 45% are homozygous and 55% are heterozygous for the RHD gene. In Caucasians, deletion of the entire RHD gene is the most common cause of the D- phenotype (Avent 2006). In contrast, a D- phenotype in Africans and Asians is most often the result of a silent non-functional RHD gene, termed the RHD-pseudogene (RHDΨ). 67% of D- South African blacks and 25% of African Americans have the RHDΨ, which contains a 37 base pair insert in exon 4 and a nonsense mutation in exon 6, which introduce stop codons and results in a lack of transcription (Singleton et al, 2000). Furthermore, 15% of African Americans carry a hybrid RHD-CE-D gene in which the RHCE segments encompass exons 4 through 7.

The current technology for determination of foetal RHD genotype utilising the presence of cell-free foetal DNA adapts real-time PCR, which is quantitative, making it easy to distinguish foetal and maternal contributions; moreover, the amplification and analysis take place in closed tubes, which reduces the risk of contamination. The method routinely used at the IBGRL incorporates Taqman primers and probes to detect, by real-time PCR, exons 4, 5 and 10 of RHD. The exon 10 reaction also amplifies the RHDΨ psuedogene and the hybrid RHD-CE-D gene. To evaluate the efficiency of the DNA extraction procedure and to confirm that male foetal DNA is present in the maternal plasma, CCR5 (the universal chemokine receptor gene) and the Y-chromosome linked SRY gene sequence is also amplified. A foetus is characterised as D+ if at least one of two replicates are positive for all three RHD exons. If there are discrepancies after a second amplification, the results are considered inconclusive and a new sample is obtained for re-analysis. To prevent false-positives due to RHDΨ or RHD-CE-D hybrid genes, foetuses are considered negative when no amplifications are detected in both RHD exon 4 and RHD exon 5 whatever the PCRs of exon 10. Between 2001 and 2004 the IBGRL tested 359 foetuses: 233 were predicted to be D+ and 114 D-. Of 247 of these the result was confirmed, either by testing DNA from amniotic fluid (during the trial period) or by subsequent serological testing on the baby's red cells. In one case an incorrect result was obtained: a D- foetus was predicted, from tests at 19 weeks of gestation, to be D+. The reason for this error may be due to the fact that the presence of the CCR5 gene confirms the presence of DNA in the sample but it does not distinguish between maternal and foetal DNA, therefore there is the theoretical possibility that no foetal DNA was isolated from the maternal sample. In three cases the foetus was predicted to be D+, but at birth the baby was serologically D- as a result of multiple intrauterine transfusions. In 12 cases the results were reported as inconclusive. 99 remained untested. In 2008, Minon et al reported a concordance rate of foetal RHD genotypes in maternal plasma and newborn D phenotypes at delivery as 99.8%. For this reason, clinicians may also provide an amniotic fluid sample when an RhD negative result is obtained. RhD typing from amniotic fluid or chorionic villus uses a multiplex PCR to amplify intron 4 and exon 7 of the RHD gene because these regions show differences from the related RHCE gene. Thus, upon electrophoresis, the presence of PCR products representing intron 4 and exon 7 indicates the presence of the RHD gene and hence, a RhD-positive fetus. 

However, at present finding satisfactory internal controls is problematic, due to the large amount of maternal DNA. Amplification of the Y-linked gene, SRY, provides an internal control, but is only applicable when the foetus is male. Other potentials for internal controls that detect DNA sequences in foetal DNA inherited from the father and not present in the mother are insertion and deletion polymorphisms and short tandem repeat sequences, or involve the detection of differentially methylated regions of DNA (.

After anti-D, the next most common causes of HDN are anti-c and anti-K, although in both the prevalence of severe disease is much lower than that caused by anti-D. The c/C polymorphism is attributed to a 307C>T single nucleotide polymorphism (SNP) in exon 2 of RHCE, encoding a P103S substitution in the second extracellular loop of the RhCcEe protein (Avent xxxx). Molecular testing for C is difficult because exons 2 of RHD and of the C allele of RHCE have identical sequences, but testing for c is more straightforward as C307 is unique to the c allele of RHCE. The k/K polymorphism is due to a 698C>T SNP in exon 6 of KEL encoding T193M. Consequently, it has proved relatively straightforward to devise methods, involving restriction enzymes or allele-specific primers, for the prediction of c and K phenotypes.

Many other blood group polymorphisms, including S/s, E/e, Kpa/Kpb, Jsa/Jsb, Fya/Fyb, Jka/Jkb, Dia/Dib and Coa/Cob, result from SNPs and are rarely associated with HDN. It would be feasible to develop non-invasive methods for predicting foetal phenotypes from maternal blood for all these polymorphisms, but the demand would be extremely low owing to the rarity of severe disease. Likewise, it should be relatively straightforward to develop non-invasive foetal genotyping methods for the HPA-1 and HPA-5 platelet polymorphisms, associated with neonatal alloimmune thrombocytopenia (NAIT).

Management of the foetus

The management of severely affected cases of HDN involves controlled early delivery (with risks of prematurity) and intrauterine transfusions (with risks of foetal exsanguination and boosting of maternal antibody level) (Urbaniak 2008). Due to technical limitations, foetal transfusions are not safe before 20 weeks gestation and are associated with a 1 to 3 percent foetal loss therefore less conventional approaches have been examined to prolong the gestation prior to the need for intrauterine transfusions.

Various types of therapy to target the maternal RBC alloimmunisation in pregnancy have been tried. These include serial plasmapheresis procedures (Urbaniak 2000), oral administration of D-positive red cell stroma in an effort to induce D-specific immunologic tolerance in the mother (xxxx), or the use of promethazine to decrease phagocytosis by the fetal reticuloendothelial system (xxxx). None of these have proven consistently beneficial in subsequent clinical trials.

Some success has been seen following the infusion into the mother of high-dose intravenous IgG immunoglobulin (IVIG) (Margulies 1991) which has led to one group combining plasma exchange and IVIG to achieve temporary immunomodulation in cases of severe alloimmunisation (Ruma 2007). The mechanism of action of IVIG has not yet been determined, although inhibition of maternal antibody synthesis, increased catabolic rate of IgG, partial blockade of antibody transport across the placenta and Fc blockade at the level of the foetal reticuloendothelial system have been proposed (Urbaniak 2008). Evidence exists that suggests that the level of maternal IgG needs to be maintained above 20g/L by IVIG administration to prevent placental transfer of anti-D (Urbaniak 1997). However, this is an expensive and labour intensive process and cheaper alternative approaches would be welcomed.

Administration of a D-specific non-destructive antibody could be an attractive alternative. It has been proposed that non-destructive antibody could be administered to the mother, cross the placenta and block the binding of haemolytic maternal anti-D (Armour 1999). Maternal IgG (both IgG1 and IgG3 subclasses) crosses selectively into the foetal circulation early in the pregnancy and by 20 weeks of gestation maternal IgG1 is detectable in foetal blood at levels equal to those in the mother (Saji 1999). In contrast to IgG1 and IgG3, IgG2 and IgG4 are known to be ineffective at inducing RBC destruction (Urbaniak 2008). Futhermore, it is believed that IgG crosses the placenta by binding to FcRn receptors present in syncytiotrophoblast granules and is transported across the syncytiotrophoblast where the antibody is then released into the foetal circulation (Simister 2003). It is also thought that His435 determines efficient IgG1 transfer (Firan 2001) therefore any mutations of recombinant IgG anti-D designed to stop foetal red cell haemolysis must avoid interfering with the FcRn-binding properties otherwise the mutated anti-D would not survive in the circulation for the normal plasma half-life of 21 days and would not transfer efficiently across the placenta. Furthermore, in vitro immunologic studies have shown that the Fc portion of IgG interacts with mononuclear cells via a specific receptor FcγR (Van de Winkel 1993). FcγRIIIa mediates red cell lysis and is thought likely to be the primary receptor for in vivo lysis of IgG anti-D-coated red cells in the foetal spleen.

Armour et al designed an IgG1 hybrid Fc region, which was inactive in binding to receptors FcγRI-III and complement C1q by incorporating the inactive CH2 sequences 233 to 236 from IgG2, substituting IgG1 residues at 237, 330 and 331 with those from IgG4 and three further modifications to remove allotypic residues, which may result in an immune response in recipients. The hybrid IgG1 was attached to the Fab antigen-binding variable region of a monoclonal anti-D (FOG-1). The mutated anti-D was tested in vivo in D-positive volunteers with autologous red cells coated with the anti-D, and compared to polyclonal anti-D, clearance of the red cells was slower and there was no evidence of haemolysis of the infused anti-D-coated red cells.

Nielson et al used a mutant IgG3 anti-D with the long hinge region of the heavy chain deleted and a further mutation (cys131ser) to link, via a disulphide bond, the light chains to maintain stability. By removing the hinge region it stops complement-mediated RBC lysis, Fcγ receptor-mediated haemolysis and phagocytosis of RBCs; in this study the mutant IgG3 anti-D was effective in blocking the haemolytic activity of recombinant and polyclonal anti-D in vitro.

Although the above reports are of interest, more questions need to be answered; because the plasma half-life is not known, can sufficiently high, sustained levels of mutant anti-D be achieved in vivo to be of any therapeutic value? Because repeated doses would be required, is there a possibility of inducing an anti-IgG immune response that could interfere with efficacy? Can the in vitro demonstration of competitive inhibition by haemolytic anti-D in artificial bioassays be translated into effective blockade of haemolysis in vivo in the foetus? It will also be a challenge to persuade regulators to allow clinical studies with mutant recombinant anti-D in pregnant women (Urbaniak 2008).

It is thought that because IgG can only produce a temporary effect it will need to be repeatedly given to sustain adequate levels. Therefore an alternative approach, currently under clinical trial in several immunologic disorders, is to induce an active immune response to suppress antibody formation. Many IgG antibody responses rely on T-helper cells which recognise short antigen-derived peptides displayed by HLA-class II molecules expressed on the surface of antigen presenting cells (APCs). The Th cells become activated and induce antigen specific B cells to produce antibody directed against that antigen. The circumstances in which such T cell-APC recognition takes place controls whether a specific immune response is activated or tolerated (Bai 2000); for example, an epitope that induces an immune response when given by the mucosal route can induce tolerance. HLA-DR15 is the major restricting locus for Th cells specific for RhD protein and expression of this HLA-DR transgene was found to confer on mice the ability to respond to immunisation with purified RhD protein (Stott 2000). Studies of the nature of the immune response to RhD and its regulation have been limited by lack of a suitable animal model because wild-type rodents do not produce RhD-specific antibody (Hall 2005). A humanised HLA-DR15-transgenic mouse model has recently been developed which show a number of similarities with the human IgG response to the RhD protein. Treatment with 4 dominant epitopes from the RhD protein that induce proliferation of Th cells, RhD52-66, RhD97-111, RhD117-131 and RhD177-191, appeared to inhibit T-cell priming and prevented antibody responses to the RhD protein. The success of this selection of peptides at inducing tolerance now forms the basis for further studies and clinical trials are ongoing?.


Severe HDN is still a source of clinical concern despite the introduction of anti-D prophylaxis and even in developed countries the incidence of maternal alloimunistaion remains at 1-1.5 percent of at-risk D-negative women, due mainly to failures of adequate RhD prophylaxis administration or antepartum transplacental haemorrhage. Conventional management of the foetus, although good, still results in 10 percent of these immunised women having a baby severely affected in utero. Whilst there is no solution that fits all, improvements are always being sought. Identifying at-risk pregnancies utilising high-tech PCR is now routinely used in the UK. Novel approaches such as artificially blocking the active site on foetal cells with inactive, mutant anti-D or "switching off" an antibody response to red cell and inducing tolerance in an individual are attractive alternatives to premature delivery and foetal intrauterine transfusions were the risks to the infant are high. However, ultimately a huge amount of work is still required before a suitable safe alternative can be introduced.