SEARCH

SEARCH BY CITATION

Keywords:

  • Factor VIII inhibitor;
  • Factor VIII genotype;
  • HLA system;
  • Inhibitor prediction;
  • Major histocompatibility complex (MHC) molecules;
  • T-cell receptor

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References

Summary.  Advances in molecular immunology over the past two decades permit a better understanding of why antibodies develop to peptide antigens like factor VIII and the events that lead to the development of these antibodies. Two important variables that are critical in antibody formation are (i) the molecular defect in FVIII and the consequences of that defect on translation and protein production, and (ii) the major histocompatibility complex (MHC) molecules which bind specific peptide sequences and present those peptides to CD4 T lymphocytes to initiate the cellular cascade leading to B-cell stimulation and differentiation, and ultimately to antibody formation. Inhibitors develop in hemophilia because transfused FVIII can be seen as a foreign protein and elicits an immune response in much the same way that any other foreign protein might elicit an immune response. However, not all hemophiliacs generate an immune response, either because they do not recognize FVIII as foreign or because their MHC phenotype is such that a cellular immune response is not initiated. In this model, it is the combination of molecular defect and MHC phenotype that determines inhibitor formation. The interplay of these two variables in the context of why some but not all hemophiliacs develop antibodies after treatment with replacement factor is reviewed.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References

Inhibitors in hemophilia A are a serious complication that renders normal therapy with replacement factor VIII ineffective [1–3]. Inhibitors are antibodies, frequently IgG4 antibodies, which are directed against epitopes in FVIII that are involved in the interaction of FVIII with other coagulation factors in the formation of the ‘tenase’ complex. Inhibitors against other coagulation factors in other inherited coagulation factor deficiencies have been reported, but for reasons that remain uncertain, inhibitors in hemophilia A are at least 10-times more common than in any other inherited coagulation factor deficiency [4,5]. Interestingly, spontaneous coagulation inhibitors in normal individuals are also much more commonly directed against FVIII than against other factors. Fortunately, inhibitors do not occur in all patients with hemophilia A, although it is not possible presently to accurately predict those who are and are not at risk for inhibitor development.

Understanding how and in whom inhibitors occur in hemophilia is important, not only to be able to inform patients at risk for inhibitor development but also for the therapeutic approaches that might come from such an understanding. From a broader perspective, inhibitors are also an important model for studying the immune system. They occur with a high frequency, assays for detecting inhibitors are available and reproducible, FVIII structure and function are known, and mouse and large animal models of hemophilia exist and can be used to study inhibitor development. In addition, hemophilia is the only clinical disorder for which it is possible to induce tolerance [6]. Understanding the molecular immunology of tolerance induction might lead to methods for inducing tolerance in other immunological diseases.

Critical questions to understanding how and in whom inhibitors occur are (i) why do any hemophiliacs develop inhibitors, (ii) why do not all hemophiliacs develop antibodies, and (iii) is it possible to predict inhibitor development in a given individual?

Why do any hemophiliacs develop inhibitors?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References

At the simplest level, hemophiliacs develop inhibitors because replacement FVIII is seen as a foreign or ‘non-self’ protein that triggers an adaptive immune response. Development of the normal self-tolerance repertoire, including tolerance to FVIII, starts during embryogenesis around the end of the first trimester when primitive lymphoid precursors start to transit and mature in the thymus, is generally completed by birth, and is maintained during adult life by maturation of T cells in the thymus. The mechanism of self-tolerance is believed to be through deletion of autoreactive T cells. During their maturation in the thymus, immature T lymphocytes are highly sensitive to negative selection through apoptosis. The presentation of self-antigens to these sensitive immature T cells results in deletion from the repertoire of the T cells that are reactive to self. In hemophilia, the absence of FVIII prevents the generation of self-tolerance to FVIII and, as a result, FVIII that is infused for treatment is seen as a foreign protein and an immune response ensues.

The generation of antibodies to peptide antigens like FVIII involves a cellular cascade from antigen-presenting cell to T lymphocyte to B lymphocyte (see Fig. 1). The initial processing of peptides derived from proteins is mediated by major histocompatibility complex (MHC) molecules on antigen-presenting cells (APCs). For extracellular proteins like exogenously administered FVIII in patients with hemophilia, MHC class II molecules mediate the processing of peptide antigens. Intracellular proteins, such as influenza peptides, are processed through MHC class I molecules.

image

Figure 1. MHC Class II cellular immune cascade. Exogenous peptide antigens such as FVIII are processed through MHC class II mechanisms. Proteins are bound to surface immunoglobulin on APCs and taken into endocytic vesicles where they are bound to MHC class II molecules. Bound peptides are then presented on the surface of the cell to specific T-cell receptors (TCR) on CD4+ T lymphocytes. Proper T-cell presentation requires a costimulatory interaction, typically provided by the interaction of CD80 (B7.1) or CD86 (B7.2) with CD28 or CTLA4 on the T lymphocyte. In response to antigen presentation, T lymphocytes elaborate cytokines and upregulate a number of surface molecules, including CD2, CD30, CD40 ligand (CD40L), and CD28. These surface molecules interact with corresponding proteins (LFA-3, CD30 ligand (CD30L), CD40, and CD80/86) on B lymphocytes, leading to maturation of B cells and antibody formation. The remarkable specificity of the cellular immune response derives from two reactions: the specific recognition of peptide sequences by MHC class II molecules and the specific recognition of MHC-bound peptide by the TCR. APC, antigen-presenting cell. CD4, CD4 positive T lymphocyte. B cell, B lymphocyte. MHC, major histocompatibility complex. TCR, T-cell receptor. CD, cluster of differentiation. IL, interleukin. (copyright G. C. White II).

Download figure to PowerPoint

When exogenous FVIII is administered to a hemophiliac, some of the FVIII binds to surface immunoglobulin on APCs. The bound FVIII is taken into endocytic vesicles where the FVIII is proteolytically cleaved and the resulting peptides are bound to class II molecules in the endosome [7]. Each MHC molecule has a unique peptide recognition sequence, generally about nine amino acids in length, allowing it to bind a unique repertoire of peptides. Class II molecules are more promiscuous than class I molecules and bind a broader range of peptide sequences. After FVIII peptides are bound to MHC molecules, the MHC-peptide complex transits to the surface of the APCs where the peptide is presented to T-cell receptors (TCRs) on the surface of CD4 lymphocytes. Proper presentation of antigen to the TCR requires a second interaction between the APCs and the T cell, typically through CD80/86 on the APCs and CD28 on the T cell [7,8]. In the presence of both signals, the T cell is activated and there is secretion of cytokines and upregulation of a number of proteins on the surface of the CD4 lymphocyte, including CD2, CD30, CD40 ligand (CD40L), and CD28 [9]. These events promote the interaction of the T cell with B lymphocytes, leading to B-cell proliferation, immunoglobulin isotype switching, and differentiation into antibody-secreting plasma cells. Blocking costimulation prevents the T-cell response to antigen and leads to a form of tolerance (see [10]). This has been achieved experimentally in hemophilic mice using monoclonal antibodies that block CD86 (anti-B7-2) [11].

An individual's MHC class II phenotype determines which, if any, FVIII-derived peptides will be bound and presented to T lymphocytes and thereby determines if the cellular cascade leading to B-cell activation and antibody formation is initiated. There are multiple MHC class II isoforms grouped in three families: DR, DQ, and DP. Each isoform binds a specific and essentially unique peptide sequence [12]. Table 1 illustrates consensus recognition sequences identified for a limited number of DR, DQ, and DP molecules [see 13]. For example, an individual who is HLA DR B1*0402 will recognize peptides that have the amino acid sequence L/I L R R/L H N/QR/K F H/A, where the underlined sequences are ‘anchoring’ sequences that are required for the interaction with the MHC molecule. The non-underlined, intervening sequences are generally on the opposite side of the peptide and are the sequences that interact with the TCR. A given MHC molecule may recognize more than one sequence in a protein or might recognize no sequences in a protein.

Table 1.  Consensus MHC Class II recognition sequences
HLA typeConsensus recognition sequence
123456789
  1. Anchoring sequences, e.g. sequences required for recognition are generally in positions P1, P4, P6, and P9 and are underlined. Table developed from data presented by Rammensee [12].

DR B1*0101Y/VRFL/ARA/GYHV/L
DR B1*0301L/IRMDSK/RTKY/L
DR B1*0401F/YVRF/ADS/ND/AAA/Q
DR B1*0402L/ILRR/LHN/QR/KFH/A
DR B1*0404M/LRYF/IHT/NAMS/K
DR B1*1101F/YTCI/VTR/KARF
DR B1*1201L/IRL/MYDV/NNSY/F
DR B1*1501L/VRAY/FLDF/VTG
DR B5*0101Y/MRAV/QTPQGK/R
DQ A1*0501L/MMQAV/LPM/LGA
DP A1*0201Y/AFAATQF/YEC

While the MHC phenotype on the APC determines if a signal is initiated, the T-cell functions as the gatekeeper in the cascade. For the signal initiated by the binding of peptide to MHC molecules to be transmitted to B lymphocytes capable of making antibody, it must pass through T-helper cells. In this gatekeeper function, the binding of the MHC-peptide complex to the TCR is the critical interaction and is very specific. In normal individuals, autoreactive T cells – such as T cells reactive to peptide sequences in FVIII – have been deleted, so there would be no T cells able to interact with the FVIII peptide. As a consequence, no immune response to FVIII would occur. However, in hemophilia, T cells able to recognize FVIII peptides are not deleted and are able to recognize the FVIII peptide on the APC and interact through the TCR, leading to T-cell activation and, subsequently, B-cell activation with the formation of antibodies to FVIII.

Why do not all hemophiliacs develop inhibitors?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References

If the absence of FVIII during development leads to a failure to generate self-tolerance to FVIII and thereby leads to an immune response to infused FVIII, why do not all hemophiliacs develop inhibitors when they are treated? Demographic studies show that anywhere from 8% to up to 52% of patients with severe hemophilia A develop inhibitors, but not all patients will do so [4,5,14–19]. What accounts for this failure of some individuals who lack detectable FVIII to respond immunologically to the infusion of FVIII?

One important variable that affects the probability of inhibitor formation is the nature of the molecular defect in the individual with hemophilia and the consequences of that defect on protein production. The extent of the FVIII molecule that comes to be recognized as ‘self’ depends on how much of the molecule is synthesized and able to be presented as peptide sequences to the immature T cells in the thymus. Any part of FVIII that is synthesized will induce tolerance to that part of the molecule. An example might be a mutation in FVIII that results in truncation of FVIII such that the heavy chain is synthesized but not the light chain. Even though no functional protein might be detected, the heavy chain peptides could be presented to the immature T cell in the thymus and be included in the self-recognition repertoire. Upon later infusion of full-length FVIII, only peptide sequences derived from the light chain of FVIII would be recognized as foreign while sequences from the heavy chain would not. Another example might be an individual with a missense mutation. The individual might have no detectable plasma FVIII activity, but is likely to synthesize a full-length dysfunctional FVIII protein that is recognized as self. As a result, when normal FVIII is infused, the only sequence that would be seen as foreign would be the sequence containing the missense mutation; the remainder of the molecule would be tolerated.

It is important to emphasize here that it is the peptide sequences that are synthesized and able to be presented to the immature T cells in the thymus that determine whether tolerance occurs, not the level of FVIII activity in the plasma. Even if there is no detectable FVIII in plasma, all or portions of the protein may be synthesized and induce tolerance. Thus, it is important to know what portions of FVIII are transcribed and synthesized as protein if one is to understand how much of the molecule can be expected to be foreign. Unfortunately, the location and type of mutation in FVIII do not necessarily provide this information, as some mutations – missense, nonsense, or deletion – might result in an unstable message in which none of the protein is produced even though the location of the mutation might predict that substantial portions of the protein could be produced. Conversely, sequences that are not expected to be synthesized could be. Recent findings indicate that a phenomenon termed exon skipping may occur with nonsense mutations, permitting sequences distal to the site of truncation to be expressed [20]. The occurrence of exon skipping in hemophiliacs with nonsense mutations might result in tolerance to sequences distal to the termination sequence. In addition, with the common intron 22 inversion mutations that occur in hemophilia, it has been speculated that synthesis of the carboxyterminal light chain fragment derived from exons 23–26 may occur, although no evidence currently exists for this. In the end, the only way to determine sequences of FVIII that are synthesized in individuals with hemophilia may be through in vitro expression studies.

Another important variable is the MHC type of the individual. As described above, MHC molecules bind short, proteolytic fragments of FVIII and present these peptide fragments to the TCR on T cells, initiating the cellular immune response cascade. This binding of FVIII peptides to MHC molecules is very specific, and each MHC molecule recognizes a unique nine amino acid peptide sequence. The likelihood that a protein like FVIII has a consensus recognition sequence for a given MHC molecule is variable, with most having at least one and many having multiple recognition sequences. Figure 2 shows consensus recognition sequences in FVIII for a number of MHC molecules. In the example given, there are up to 13 potential recognition sequences for HLA-DRB1*1501 in FVIII, with at least one in each domain, whereas there are only two for HLA-DRB1*1101, both in the heavy chain. The overall likelihood that an individual would be able to recognize and process FVIII peptides is determined by the number of class II molecules an individual has, where an individual may have as few as three (one DR, one DQ, and one DP if they are homozygous at each allele) or as many as 12 (four DR, four DQ, and four DP), and the specificity of each of their class II molecules.

image

Figure 2. Location of MHC motifs in FVIII. The number and location of potential consensus recognition motifs in FVIII for various HLA-DR molecules are shown. HLA DRB1*1501 contains 13 potential recognition motifs: one in the A1 domain, four in the A2 domain, three in the B domain, two in the A3 domain, one in the C1 domain, and two in the C2 domain. In contrast, HLA DRB1*1101 contains only two motifs: one in the A1 domain and one in the A2 domain. The two examples on the bottom show the additive effects of two MHC molecules. These are computer-identified recognition sequences based on the known consensus recognition sequences for each molecule and are not experimentally confirmed; real recognition motifs might differ from those shown. A1, A2, B, A3, C1, C2, domains of FVIII. (copyright G. C. White II).

Download figure to PowerPoint

Ultimately, the development of an inhibitor depends on (i) what sequences in infused FVIII are seen as foreign based on the molecular defect of the individual and (ii) whether or not the individual has an MHC phenotype capable of interacting with sequences in the portion of the molecule seen as foreign. In a sense, the molecular defect and the MHC phenotype are codependent variables in the generation of an immune response. The way they work together is illustrated in Fig. 3. The molecular defects in two hemophiliacs are shown. One has a nonsense mutation in the A3 domain. The second has a nonsense mutation in the A2 domain. Assuming that the proteins are translated up to the truncation defect, the first hemophiliac would see the A1, A2, B, and part of the A3 domains as self while the second hemophiliac would see the A1 and part of the A2 domains as self. A nine amino acid sequence, YRAVTPQGR, which is a consensus recognition motif for HLA DRB5*0101, is present in the A3 domain of FVIII, within the sequence recognized as self in the first hemophiliac but in the sequence not recognized as self in the second hemophiliac. When the first hemophiliac is infused with FVIII, the FVIII binds to surface immunoglobulin on APCs and is internalized and proteolytically degraded. In an individual who is HLA DRB5*0101 positive, peptides containing YRAVTPQGR bind to DRB5*0101 in endosomes and are subsequently presented on the surface of the APC to the TCR on CD4 lymphocytes. However, because of the molecular defect in this individual, T-cell clones recognizing the YRAVTPQGR sequence in FVIII were deleted during development of the self-recognition repertoire, so no T cells able to recognize the YRAVTPQGR peptide exist and no immune response occurs. In contrast, in the second hemophiliac, T-cell clones recognizing the YRAVTPQGR sequence have not been deleted from the self-recognition repertoire and exist, able to interact with the peptide-bearing APC and generate a cellular immune response. Suppose, instead, that the second hemophiliac was homozygous for HLA DQA1*0302, which has no consensus recognition sequences in FVIII. In this situation, even though T-cell clones recognizing the YRAVTPQGR peptide exist, there would be no MHC molecule able to recognize the YRAVTPQGR or any other peptide in FVIII, so no stimulation of the TCR and no immune response would occur.

image

Figure 3. Molecular defect and HLA type are important in inhibitor development. Two hypothetical individuals, one with a nonsense mutation at the end of the A3 domain (top, arrow marks the site of mutation), the other with a nonsense mutation in the A2 domain (bottom, arrow marks the site of mutation), are illustrated. In both individuals, the truncated molecule is expressed up to the point of the nonsense mutation. Both are homozygous for MHC class II HLA-DRB5, where HLA-DRB5 recognizes the sequence YRAVTPQGR that is present in the A3 domain of human FVIII. In the individual shown on top, one might expect that this sequence in the FVIII molecule would be recognized as self. As a consequence, when exogenous FVIII is infused to that individual, the YRAVTPQGR sequence in the exogenous FVIII would be recognized as self and would not generate an immune response. Conversely, in the second individual, the YRAVTPQGR sequence is not expressed and would be recognized as non-self. Following infusion of exogenous FVIII, the YRAVTPQGR peptide would interact with HLA-DRB5 molecules in the endocytic vesicles of the APC, be presented to reactive T cells, and lead to T-cell activation and B-cell stimulation, resulting in antibody formation. Conversely, if the individuals were HLA DQA1, there is no consensus recognition sequence for HLA DQA1 anywhere in the FVIII molecule, so that no matter where the defect is located in FVIII, no immune response would develop. A1, A2, B, A3, C1, C2, domains of FVIII. MHC, major histocompatibility complex. APC, antigen-presenting cell. CD6, CD4 positive T lymphocyte. B cell, B lymphocyte. (copyright G. C. White II).

Download figure to PowerPoint

If MHC molecules are important for the generation of an immune response, why have previous studies failed to show a clear correlation between HLA type and risk of inhibitor development? An increased prevalence of HLA DRB1*1501, DQA1*0102, and DQB1*0602 in inhibitor patients vs. non-inhibitor patients was observed by both Oldenburg et al. and by Hay et al. but the correlation was weak [21,22]. Other studies have shown other correlations but none have been confirmed. Nevertheless, these studies do not mean that HLA type is unimportant; they mean that the role that HLA type plays is complex. None of these studies have taken the molecular defect of the hemophiliacs into account. More importantly, the remarkable MHC diversity in man makes it very difficult to see meaningful correlations in a sample as small as the hemophilia population.

It can be seen from the foregoing that a reason that not all hemophiliacs develop inhibitors is that some hemophiliacs do not have the right combination of molecular defect and HLA type to generate an immune response. This model for inhibitor formation in hemophilia is consistent with what is presently known about immunological processing of peptide antigens such as FVIII, but it does not explain some of the clinical features of inhibitor formation. For example, why are inhibitors 10-times more common in hemophilia A than in other disorders? Factor VIII is a large protein and presents more peptides that increase the likelihood of recognition by MHC molecules, but does this account for the difference? Why do some patients develop low titer inhibitors while other develop high titer inhibitors? Is this primarily a property of the B-cell clone that is stimulated or is there something in the T-cell response that affects this? Why do some hemophiliacs develop inhibitors after 10 exposure days while others develop inhibitors after 60 exposure days? Why do not all hemophiliacs with the right combination of molecular defect and MHC type develop inhibitors after their first exposure to FVIII? Are there genetic modifiers that influence these aspects of inhibitor formation? How do epigenetic factors, such as infection, inflammation, immunization, cytokine production, etc., play a role in inhibitor formation and the characteristics of the antibodies that form? Do these factors feed in through the cellular cascade or do they exert their influence elsewhere? These are all important questions that are not completely addressed by the model. Ongoing studies to use single nucleotide polymorphism (SNP) linkage analysis and haplotype maps to perform genome-wide scans to identify potential modifier genes (C. Winkler and the HIGS investigators, NCI-Frederick, Frederick, MD, unpublished studies) may help clarify some of these questions.

On the other hand, some of what we know of inhibitor formation fits well with the model. The increased correlation for inhibitor development between siblings [23,24] is consistent with the model as brothers have the same molecular defect, are more likely to share MHC haplotypes, and would be predicted to be concordant for inhibitor development if they have the same FVIII defect and MHC type. As expected, the concordance in monozygotic twins is very high, although there is a report of one set of twins who are discordant for inhibitor formation, an occurrence that cannot be fully explained by the model. An important and testable corollary of this model is that brother-pairs who are discordant for inhibitor development should have different MHC haplotypes. Further support for the model comes from the previous demonstration that the molecular defect has predictive value. The molecular defect determines the structure of the FVIII molecule – what is expressed, what is not, whether the message is stable, whether it is unstable – and ultimately what is recognized as ‘self’ and what is not. Based on this, one might expect that nonsense mutations, partial inversion mutations, and large deletions would have larger portions of the molecule that are not expressed and not recognized as ‘self’, and therefore have a higher rate of inhibitor formation than missense mutations. This is indeed what is observed. About one-third of individuals with nonsense mutations, partial inversion mutations, or large deletions develop inhibitors whereas only about 3% of those with missense mutations experience inhibitors [25–27]. The risk of inhibitor development can be further correlated with the extent of the molecular defect. For example, among individuals with large deletions, those with multidomain deletions have a higher rate of inhibitor formation, and those with single domain deletions have a lower rate. Nevertheless, although it is an important determinant of antibody formation, the molecular defect is not the only determinant, as evidenced by the frequent siblings who are discordant for inhibitor development. Even individuals with complete deletion of the FVIII gene, who would be expected to recognize the entire molecule as ‘non-self’ and develop an inhibitor, may not do so. The MHC type of such individuals would be important to know and might be expected to be a type that fails to recognize FVIII peptides. The predictive value of disease severity is also supportive as mild and moderate hemophiliacs generally have missense mutations and would be predicted by this model to have a lower risk of inhibitor development. While inhibitors are less frequent in mild hemophiliacs, these individuals provide an excellent opportunity to show that the model proposed here is correct. In these cases, the individual's MHC molecules should recognize and bind the normal FVIII sequence (to which the individual should not be tolerant) but not the mutant peptide and T-cell responsiveness to the wild type peptide but not the mutant peptide should be demonstrable. Interestingly, Jacquemin et al. showed this in a mild hemophiliac with an Arg2150His substitution in the C1 domain that developed an inhibitor [28]. Wild type but not mutant C1 domain peptides encompassing the Arg2150 mutation presented to T-cell clones derived from the patient resulted in T-cell stimulation as measured by interferon-γ production.

Is it possible to predict inhibitor development in a given individual?

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References

Does the current understanding of the mechanisms of cellular immune responses make it possible to predict inhibitor development in a given individual with accuracy? In other words, if one knows the molecular defect, the sequence of the translation product of the mutant FVIII gene, and the MHC phenotype of an individual and the peptide sequences recognized by each MHC molecule, would it be possible to predict inhibitor development with precision? The thesis of this presentation is that it would be possible to predict inhibitor formation with significant precision if these parameters were known.

If one were able to predict inhibitor development in an individual, what would one do with that information? Would one withhold FVIII to prevent inhibitor formation? Almost certainly not. Would one start an immune tolerance type regimen with the initial treatment? Perhaps. Would treatment with a bypassing agent be a reasonable approach? This has been suggested as an approach by some centers in Europe. While only time will tell if this approach is reasonable, it seems likely that subsequent treatment with FVIII would still have the same risk of inhibitor formation, so the gain of using the bypassing agent is uncertain.

In conclusion, the development of inhibitors in hemophilia can be best viewed through an understanding of current immunological mechanisms involved in cellular processing of peptide antigens from APCs to T lymphocytes to antibody-producing B lymphocytes. Hemophilia provides an important way to test and validate this model through a careful correlation of genetic defect and MHC type with inhibitor risk. Equally important, the model suggests important ways in which the processing of antigens can be interrupted and modified to reduce the risk of inhibitor formation.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References

The authors acknowledge the helpful comments of Drs Jack Gorski and Deb Newman (Milwaukee) and the long-term collaboration of Dr Jeff Frelinger (Chapel Hill).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Why do any hemophiliacs develop inhibitors?
  5. Why do not all hemophiliacs develop inhibitors?
  6. Is it possible to predict inhibitor development in a given individual?
  7. Acknowledgements
  8. References
  • 1
    Scandella DH. Properties of anti-factor VIII inhibitor antibodies in hemophilia A. Semin Thromb Hemost 2000; 26: 13742.
  • 2
    Dimichele D. Inhibitors: resolving diagnostic and therapeutic dilemmas. Haemophilia 2002; 8: 2807.
  • 3
    Key NS. Inhibitors in congenital coagulation disorders. Br J Haematol 2004; 127: 37991.
  • 4
    Ljung R, Petrini P, Lindgren AC, Tengborn L, Nilsson IM. Factor VIII and factor IX inhibitors in haemophiliacs. Lancet 1992; 339: 1550.
  • 5
    Ehrenforth S, Kreuz W, Scharrer I, Linde R, Funk M, Gungor T, Krackhardt B, Kornhuber B. Incidence of development of factor VIII and factor IX inhibitors in haemophiliacs. Lancet 1992; 339: 5948.
  • 6
    Brackmann H-H, Gormsen J. Massive factor-VIII infusion in haemophiliac with factor VIII inhibitor, high responder. Lancet 1977; 2: 933.
  • 7
    Lechler R, Aichinger G, Lightstone L. The endogenous pathway of MHC class II antigen presentation. Immunol Rev 1996; 151: 5179.
  • 8
    Lenschow DJ, Walunas TL, Bluestone JA. CD28/B7 system of T cell costimulation. Annu Rev Immunol 1996; 14: 23358.
  • 9
    Van Kooten C, Banchereau J. Function of CD40 on B cells, dendritic cells and other cells. Curr Opin Immunol 1997; 9: 3307.
  • 10
    Judge TA, Tang A, Turka LA. Immunosuppression through blockade of CD28:B7-mediated costimulatory signals. Immunol Res 1996; 15: 3849.
  • 11
    Qian J, Collins M, Sharpe AH, Hoyer LW. Prevention and treatment of factor VIII inhibitors in murine hemophilia A. Blood 2000; 95: 13249.
  • 12
    Rammensee H-G, Friede T, Stevanovic S. MHC ligands and peptide motifs: first listing. Immunogenetics 1995; 41: 178228.
  • 13
    Rammensee HG, Bachmann J, Emmerich NP, Bachor OA, Stevanovic S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 1999; 50: 2139.
  • 14
    Addiego J, Kasper C, Abildgaard C, Hilgartner M, Lusher J, Glader B, Aledort L. Frequency of inhibitor development in haemophiliacs treated with low-purity factor VIII. Lancet 1993; 342: 46264.
  • 15
    Bray GL, Gomperts ED, Courter S, Gruppo R, Gordon EM, Manco-Johnson M, Shapiro A, Scheibel E, White GC II, Lee M. A multicenter study of recombinant factor VIII (Recombinate): safety, efficacy, and inhibitor risk in previously untreated patients with hemophilia A. The Recombinate Study Group. Blood 1994; 83: 242835.
  • 16
    de Biasi R, Rocino A, Papa ML, Salerno E, Mastrullo L, De Blasi D. Incidence of factor VIII inhibitor development in hemophilia A patients treated with less pure plasma derived concentrates. Thrombos Haemostas 1994; 71: 54447.
  • 17
    Lusher JM, Lee CA, Kessler CA, Bedrosian CL, Refacto phase 3 Study Group. The safety and efficacy of B-domain deleted recombinant factor VIII concentrate in patients with severe haemophilia A. Haemophilia 2003; 9: 3849.
  • 18
    Wight J, Paisley S. The epidemiology of inhibitors in haemophilia A: a systematic review. Haemophilia 2003; 9: 41835.
  • 19
    Lusher J, Abildgaard C, Arkin S, Mannucci PM, Zimmerman R, Schwartz L, Hurst D. Human recombinant DNA-derived antihemophilic factor in the treatment of previously untreated patients with hemophilia A: final report on a hallmark clinical investigation. J Thrombos Haemostas 2004; 2: 57483.
  • 20
    Valentine CR. The association of nonsense codons with exon skipping. Mutat Res 1998; 411: 87117.
  • 21
    Oldenburg J, Picard JK, Schwaab R, Brackmann HH, Tuddenham EG, Simpson E. HLA genotype of patients with severe haemophilia A due to intron 22 inversion with and without inhibitors to factor VIII. Thromb Haemost 1997; 77: 23842.
  • 22
    Hay CR, Ollier W, Pepper L, Cumming A, Keeney S, Goodeve AC, Colvin BT, Hill FG, Preston FE, Peake IR. HLA class II profile: a weak determinant of factor VIII inhibitor development in severe hemophilia A. A UKHCDO Inhibitor Working Party. Thromb Haemost 1997; 77: 2347.
  • 23
    Astermark J, Berntorp E, White II GC, Kroner BL, the MIBS Group. The Malmö International Brother Study (MIBS). Further support for genetic predisposition to inhibitor development. Haemophilia 2001; 7: 26772.
  • 24
    Astermark J, Oldenburg J, Escobar M, White II GC, Berntorp E, and the MIBS Group. The Malmö International Brother Study (MIBS). Genetic defects and inhibitor development in families with severe hemophilia A. Haematologica, in press.
  • 25
    Schwaab R, Brackmann HH, Meyer C, Seehafer J, Kirchgesser M, Haack A, Olek K, Tuddenham EG, Oldenburg J. Haemophilia A mutation type determines risk of inhibitor formation. Thromb Haemost 1995; 74: 14026.
  • 26
    Oldenburg J, Brackmann H-H, Schwaab R. Risk factors for inhibitor development in hemophilia A. Haematologica 2000; 85: 710.
  • 27
    Goodeve A. The incidence of inhibitor development according to specific mutations – and treatment? Blood Coagul Fibrinolysis 2003; 14: S1721.
  • 28
    Jacquemin M, Vantomme V, Buhot C, Lavend'homme R, Burny W, Demotte N, Chaux P, Peerlinck K, Vermylen J, Maillere B, van der Bruggen P, Saint-Remy JM. CD4+ T-cell clones specific for wild-type factor VIII: a molecular mechanism responsible for a higher incidence of inhibitor formation in mild/moderate hemophilia A. Blood 2003; 101: 13518.