Inhibitor development: patient-determined risk factors


Jan Astermark, MD, PhD, Associate Professor, Department for Coagulation Disorders, Malmö University Hospital, SE-205 02 Malmö, Sweden.
Tel.: +46 40 332 392; fax: +46 40 336 255;


Summary.  The reasons that inhibitory factor VIII antibodies develop in only a fraction of patients with haemophilia A remain unclear, but studies of genetically related subjects have indicated that the immunological outcome of replacement therapy is to a large extent determined by patient-related risk factors. Non-genetic factors will also influence the inhibitor risk, since events challenging the immune system will elicit and stimulate immune regulatory processes with the potential of modifying the immune response. Further insight into the immunological pathways and risk factors involved will be important in order to better predict and prevent this complication. This review will briefly summarize the data obtained to date in unrelated and related subjects in the Malmö International Brother Study (MIBS) regarding genetic factors and discuss how these factors might interact with non-genetically determined factors and events.


The formation of inhibitory factor VIII antibodies in patients with haemophilia A is a T-helper (TH) cell dependent event that involves antigen presenting cells (APC) and B-lymphocytes [1]. Several risk factors for development of these antibodies have been discussed, but studies of genetically related subjects with haemophilia A have shown that the immunological outcomes are mainly determined by patient-related risk factors [2,3]. The risk of developing these antibodies is associated with the severity of the disease, and the highest incidence (20–30%) occurs in those with the severe form (fVIII activity <1%) [4]. Consequently, the type of fVIII mutation is associated with the risk of inhibitor development [5]. However, family studies have also shown that the mutation itself will not provide information sufficient to understand why inhibitory antibodies develop. An overall concordance of 78% between siblings with severe haemophilia A and a two-fold higher frequency of inhibitors in African–Americans compared with Caucasians was observed in the Malmö International Brother Study (MIBS) [3,6]. The major histocompatibility complex (MHC) class II molecules play a central role in which they determine the peptides to be bound and presented to the T-cells. Several class II alleles have been suggested to influence the risk of inhibitor development, but associations identified to date are weak, and the overall impact of the MHC has yet to be fully established [7,8]. Additional genetic risk factors that have been suggested include polymorphisms in the genes coding for interleukin 10 (IL-10), tumour necrosis factors (TNF)-alfa (α) and cytotoxic T-lymphocyte antigen-4 (CTLA-4) [9–11]. This review will briefly summarize the available data on these genetic risk factors and discuss how they might interact with immune system challenges in the complex process of inhibitor development.

Factor VIII mutation

The most extensively studied genetic risk factor for inhibitor is the type of causative fVIII mutation and from the German data and the data reported to the HAMSTeRS database, we know that patients with large gene deletions, nonsense mutations and intrachromosomal aberrations suffer a relatively high risk, whereas those with missense mutations, small deletions/insertions and splice site mutations experience this side-effect less frequently [5]. However, when evaluating the impact of the causative mutation, studies of related subjects offer advantages, since the type of mutation will be the same in all patients. Among the families enrolled in MIBS, approximately 70% were found to be concordant in which either all or none of the siblings had a history of inhibitors [6]. The con- and discordancies in each subgroup of mutation are shown in Fig. 1. The concordance in the families with inhibitors was approximately 40%. The corresponding figures for the families with intron 22 inversions were 63% and 40%, respectively. In two families with large gene deletions, none of the siblings had an inhibitor history, and although only a small proportion of the families with missense mutations, small deletions/insertions and splice site mutations experienced inhibitors, all siblings in some of these families had high-responding inhibitors. The family data clearly indicate that additional inherited genetic determinants, other than the type of causative fVIII mutation, will be of major importance in predicting the immunological outcome of replacement therapy.

Figure 1.

 The frequency of concordant and discordant siblings with and without inhibitors in 209 MIBS families and various types of fVIII mutations. The number of families in each subgroup is given in the brackets.

MHC class II

The HLA class I alleles A3, B7 and C7, as well as the class II alleles DQA0102, DQB0602, DR15 have all been associated with higher risk for inhibitor development in unrelated patients [relative risk (RR) of 1.9–4.0], whereas the HLA C2, DQA0103, DQB0603 and DR13 alleles seem to be protective [7,8]. The reported associations were, however, weak and not statistically consistent. In the MIBS study, these alleles were equally distributed between the two patient groups [10]. Instead, significant associations were identified for two of the other class I alleles, i.e. HLA A26 and B44, but after correction for multiple comparisons no significant differences remained.


IL-10 is an important anti-inflammatory cytokine exerting a broad spectrum of activities. IL-10 also enhances the in vitro production of all types of immunoglobulins by peripheral blood mononuclear cells in patients with autoimmune diseases and the serum concentration of IL-10 has been correlated to the disease activity in these patients [12,13]. The most interesting polymorphism with a functional implication described in the IL-10 gene is a 134 bp long variant of a CA microsatellite in the promoter region (IL-10.G) [14–16]. In the MIBS study, the allele 134 bp was identified in 44 of all 164 patients with haemophilia A (26.8%) [9]. Thirty-two of these 44 patients (72.7%) developed inhibitors compared with 45 of the 120 patients (37.5%) without the allele. Among all 77 patients with a history of inhibitors, allele 134 was found in 32 patients (41.6%) compared with 12 of the 87 inhibitor negative patients (13.8%; P < 0.001). This corresponds to an odds ratio (OR) of 4.4 with a 95% confidence interval (CI) of 2.1–9.5. A significant association between the allele and the development of inhibitors was also found in a subgroup analysis of patients with severe haemophilia A, i.e. in 26 of 63 inhibitor patients (41.3%) and in 7 of 61 (11.5%) patients without inhibitors, corresponding to an OR of 5.4 (95% CI; 2.1–13.7, P < 0.001).

TNF-α -308 polymorphism

TNF-α is an important mediator of inflammatory responses and has crucial immunomodulatory activities. The TNF-α locus is located in the HLA class III region of the MHC complex and several polymorphic sites have been described [17–19]. The most extensively studied polymorphism with pathophysiological effects is a bi-allelic polymorphism at position -308 in the promoter region of the TNF-α gene consisting of a substitution of an A (allele 2, A2) instead of a G (allele 1, A1). The polymorphism induces increased levels of TNF-α [20,21]. In the MIBS study, 142 patients (86.6%) were allele A1 carriers and 86 (52.4%) were allele A2 carriers compared with frequencies of 97% and 76% of alleles A1 and A2, respectively [10]. The most common genotype A1/A1 was identified in 78 (47.6%) of the patients, and homozygocity for the A2 allele (A2/A2) in 22 individuals (13.4%). Thirty-one of the 78 subjects (39.7%) with the A1/A1 genotype had inhibitors compared with 30 of 64 patients (46.9%; ns) with the A1/A2 genotype. Sixteen of 22 patients (72.7%) homozygous for allele 2 had inhibitors yielding an OR of 4.0 (95% CI 1.4–11.5, P = 0.008) with A1/A1 as the reference group. The association between the genotype A2/A2 and the development of inhibitors was consistent in subgroup analysis of the 124 patients with severe haemophilia A (OR 19.2, 95% CI 2.4–156.5, P < 0.001), as well as in the smaller group of 75 patients with inversions (OR 11.8, 95% CI 1.3–105.1, P = 0.013). Logistic regression analysis revealed that the TNF-α genotype A2/A2 was associated with inhibitors after adjustment for the presence of allele 134 in the IL-10.G microsatellite in the entire cohort (OR 4.0, 95% CI 1.3–11.8, P = 0.013) as well as in the subgroup of patients with severe haemophilia (OR 19.3, 95% CI 2.3–162.1, P = 0.007).

Cytotoxic T-lymphocyte associated protein-4

Cytotoxic T-lymphocyte associated protein-4 (CTLA-4) is a receptor mainly displayed on activated T-cells. It mediates a down-regulation of the T-cell activity by competing with CD28 for the binding of the B7 molecules [20,21]. Consequently, blockade of this interaction by CTLA-4-antibodies enhances T-cell proliferation and B-cell activity. Several polymorphisms in the CTLA-4 gene have been found to modulate the immune response in antibody-mediated autoimmune diseases, including two single nucleotide polymorphisms (SNPs) [22–26]. The first of these SNPs is located in the promoter region of the gene at position -318 (C or T). The T-allele has been associated with an up-regulation of the CTLA-4 activity on the activated T-cells, thereby counteracting the co-stimulatory signal provided by the B7-CD28 interaction required to elicit an immune response. The second SNP is located at position +49 in the coding sequence (CDS) 1 (A or G) encoding a threonine to alanine substitution in the leader peptide.

Thirty-two of the patients (25.8%) in the MIBS cohort were allele T carriers at position -318 and 10 (31.2%) of these subjects had a history of inhibitor compared with 53 (57.6%) of the 92 patients without allele T [11]. This corresponded to an OR of 0.3 (95% CI 0.1–0.8, P = 0.012), indicating that allele T in the promoter might be protective against inhibitor development. The genotype TT was found in only three patients (2.4%) with intron 22 inversions. None of these patients developed inhibitors. The association between the T-allele and inhibitor formation was also observed in a subgroup analysis of 75 patients with an inversion as the causative mutation (OR 0.3, 95% CI 0.1–0.9, P = 0.032). Interestingly, in 11 families in our cohort discordant with respect to T-allele carriage and inhibitor history, the sibling carrying the T-allele was the one unaffected by inhibitors. No clear association was found for the +49 SNP A/G at +49 in the leader sequence.


There remains a long way to go in the identification of determinants for inhibitor development. Gaining insight into this issue is of great importance, as patients with haemophilia complicated by inhibitors are continually at risk for severe bleeds with potentially detrimental effects on quality of life, and life itself. From a societal perspective, there is a great deal to gain as the treatment and management of these patients, and the often serious outcome in cases of trauma, is extremely costly. In this era of gene therapy, there is still no indication that the inhibitor problem will be solved. Therefore, additional research in the area of inhibitor development, such as the multicentre international Haemophilia Inhibitor Genetics Study (HIGS) is warranted [27]. Thus so far, studies of related and unrelated subjects clearly indicate that the development of inhibitory antibodies is a complex process involving both genetic and non-genetic factors. Family history of inhibitors is a strong determinant for the outcome, hence genetic factors seem to be of major importance. As there are monozygotic twins discordant for inhibitor status and patients who develop inhibitors after many years of exposure to the deficient factor, it is clear that non-genetic factors also have an impact. The MHC class II molecules and the causative fVIII mutation, together with the APCs, T- and B-cell repertoires, will form the platform for the inhibitory antibodies to develop, either as a ‘safe’ or ‘unsafe’ platform (Fig. 2a,b). In patients with a ‘safe’ platform, i.e. patients with a causative fVIII mutation with the potential to delete T-cell clones recognizing dominant immunogenic fVIII epitopes and MHC class II alleles that will bind only non-immunogenic peptides, risk of inhibitor development will be low, even in the case of challenges providing ‘danger signals’ for the immune system (Fig. 2a). On the other hand, in patients with an ‘unsafe’ platform, immune system challenges might add activity sufficient to reach the ‘threshold’ for inhibitors to develop (Fig. 2b). The threshold might be reached by an interaction between genetic and non-genetic factors. The genetic factors consist of polymorphisms in the genes coding for various immune regulatory molecules and cytokines, polymorphisms known to induce levels that will promote and stimulate the immune system to form an immune response. If the most crucial genetic markers induce levels high enough to promote the immune response itself, additional pro-immunogenic ‘danger signals’ elicited by non-genetic factors or events, such as surgical procedures, traumatic bleeds and severe infections, might not be needed, a scenario typical for patients developing inhibitors at young age only after a few exposures to fVIII.

Figure 2.

 Schematic model of patients with a ‘safe’ (a) or ‘unsafe’ (b) platform for inhibitors to be developed, and the potential impact of immune system challenges (danger signals) in association with fVIII replacement therapy to reach the detrimental ‘threshold’.

The suggested schematic model is, of course, a bit simplified, but may nonetheless provide a better understanding of the complexity of the immune response to fVIII. Modifying pathways, including those not yet fully described, and T-regulatory cells will further add to the complexity of the system. A better knowledge about risk factors for inhibitors might allow clinicians to calculate for each patient an inhibitor risk score that, after the identification of additional markers, could permit adjust of their clinical management with the goal of minimizing the risk of an inhibitor response.


The MIBS study was supported by grants from Wyeth and the Research Fund at Malmö University Hospital, from the European Commission Fifth Framework Programme (QLG1-CT-2001-01918), the Swedish Research Council (05646), the foundations of the Karolinska Institutet, and the Palle Ferb foundation.


The author stated that he had no interests which might be perceived as posing a conflict or bias.