Pete Lollar

Pathogenic antibodies to coagulation factors. Part one: Factor VIII and Factor IX

Pathogenic antibodies targeting coagulation factors usually are clinically significant because they inhibit function. Circulating anticoagulants were recognized as early as 1906 ([1], cited in [2]). They interfere with the coagulation of normal blood as well as that of the patient, which is the basis of the classical mixing study. Circulating anticoagulants inhibiting factor (F)VIII and factor (F)IX were identified in the early 1940s and 1950s, nearly as soon as these factors were discovered and differentiated from one another [2], and subsequently were shown to be antibodies [3]. Another circulating anticoagulant was identified in this era that did not appear specific for a single coagulation factor and frequently was associated with systemic lupus erythematosus. Although some of these patients had abnormal bleeding, a paradoxical association of the lupus anticoagulant with thrombosis was subsequently discovered [4]. Both the hemostatic and adaptive immune systems developed rapidly during early vertebrate evolution and then produced no major changes in their principle features over the next 450 million years. Fibrinogen, factor XIII, the vitamin K-dependent proenzymes and their procofactors, factor (F)V and FVIII, are present in fish, which were the first vertebrates [5,6]. The adaptive immune system, characterized by antigenspecific lymphocyte receptors, major histocompatibility class (MHC) molecules, and antibody production, is present in jawed, but not jawless, fishes, indicating that it evolved during the 50 million years between the divergence of these two vertebrate classes [7]. The hemostatic challenge to the early vertebrates was to plug hemorrhagic leaks while maintaining overall patency of the vascular system. This includes allowing effector cells of the immune system to cross the endothelial barrier. The immunological challenge was to kill invading microorganisms and neutralize their toxic products using a defense system that did not produce self-inflicted injury. Vertebrate immune systems possess elaborate mechanisms to tolerate self-molecules and cells. It is interesting to consider hemostasis from the point of view of the immune system. A major hemostatic event is associated with deposition of products derived from fibrinogen and other coagulation factors at potentially immunizing levels. Specific and non-specific proteolytic fragmentation exposes possible neoepitopes, often in the setting of an associated inflammatory process with its attendant adjuvant effects. Thus, it seems surprising that autoantibody formation is not a common sequela to hemorrhage. In fact, autoantibodies to coagulation factors are rare: the incidence of clinically significant autoantibodies to FVIII, which is the most commonly targeted coagulation factor in autoimmunity, is only 0.2–1 per million people per year [8] and the incidence of autoantibodies to the other coagulation factors is considerably lower. Although tolerance to the coagulation proteins has not been studied, experimental models using complement component C5 as a model plasma protein have been developed and may be relevant to the development of inhibitors [9, 10]. Conversely, the recent emergence of methods to study the immunogenicity of FVIII and FIX that are described in this review may contribute to the general understanding of immunological tolerance. Inhibitory antibodies to coagulation factors occur more commonly as an immune response to non-self protein, e.g. during factor replacement therapy for hemophilia A or B or during the use of bovine thrombin products in surgery. Here one may wonder why the incidence is not higher. For example, most patients with hemophilia A and apparent null mutations do not develop a detectable immune response after repeated exposure to FVIII. However, the usual peaceful coexistence of the coagulation and immune systems provides little consolation to the practicing hematologist in a tertiary care center, who sees inhibitor patients often enough. Patients with autoantibodies to FVIII or other coagulation factors often present with severe bleeding that is difficult and expensive to control. The development of inhibitory antibodies in hemophilia A and B frequently transforms these disorders from treatable to refractory and now is considered the most significant complication of treatment. Correspondence: P. Lollar, 1365-B Clifton Road, Suite B5100, Atlanta, GA 30322., USA. Tel.: +1 404 727 5569; fax: +1 404 727 4859; e-mail: jlollar@ emory.edu Journal of Thrombosis and Haemostasis, 2: 1082–1095

Factor VIII and factor VIIIa.

Publisher Summary This chapter focuses on the role of factor VIII and factor VIIIa present in blood. The factor VIIIa apparently alters the active site structure of factor IXa to make it recognize the rate-limiting transition state for factor X activation, because its dominant kinetic effect is to increase the kcat. This chapter discusses the structure of factor VIII. Electrophoretic analysis of factor VIII preparations is commonly used to evaluate impurities and to determine whether factor VIII has been unintentionally activated. Assays for factor VIII coagulant function continue to play an important role in the assessment of factor VIII preparations. These assays are based on the ability of factor VIII to shorten the clotting time of plasma derived from a patient with hemophilia A. Two types of assays are commonly employed: the one-stage assay and the two-stage assay. The ability to modify a protein covalently at unique locations with fluorescence reporter groups can provide information on intra- and inter-molecular interactions. Factor VIII possesses no active site so that labeling at a single, unique position is difficult.

Involvement of Thrombin Anion-binding Exosites 1 and 2 in the Activation of Factor V and Factor VIII

The role of anion-binding exosites of thrombin in the activation of factor V and factor VIII was studied using thrombin Arg93 → Ala, Arg97 → Ala, and Arg101 → Ala (thrombin RA), a recombinant exosite 2 defective mutant, and a synthetic N-acetylated dodecapeptide, Ac-Asn-Gly-Asp-Phe-Glu-Glu-Ile-Pro-Glu-Glu-Tyr-O-SO4Leu (hirugen), which competitively inhibits binding of macromolecules to exosite 1. The catalytic efficiency of the activation of factor VIII or of the first step of factor V activation by thrombin RA was approximately 10% that of wild-type thrombin. The overall rate of conversion to factor Va was not influenced by the mutation. In contrast to factor V, the slow activation of factor VIII by thrombin RA was associated with a decreased rate of cleavage at all three proteolytic sites (Arg372, Arg740, and Arg1689). Hirugen inhibited factor V and factor VIII activation. These results indicate that both anion-binding exosites of thrombin are involved in the recognition of factor V and factor VIII.

Analysis of the Human Factor VIII A2 Inhibitor Epitope by Alanine Scanning Mutagenesis

Antibodies directed to the A2 domain of factor VIII (fVIII) are usually an important component of the polyclonal response in patients who have clinically significant inhibitory antibodies to fVIII. A major determinant of the A2 epitope has been located by homolog scanning mutagenesis using recombinant hybrid human/porcine fVIII molecules to a sequence bounded by Arg484–Ile508 (Healey, J. F., Lubin, I. M., Nakai, H., Saenko, E. L., Hoyer, L. W., Scandella, D., and Lollar, P. (1995) J. Biol. Chem.270, 14505–14509). Within this region, human residues Arg484, Pro485, Tyr487, Ser488, Arg489, Pro492, Val495, Phe501, and Ile508 differ from porcine fVIII. We stably expressed in mammalian cells nine active B-domainless human fVIII molecules containing single alanine substitutions at these sites. Their inhibition by a murine anti-A2 monoclonal antibody, monoclonal antibody (mAb) 413, and by three A2-specific alloimmune and two A2-specific autoimmune human inhibitor plasmas was measured by the Bethesda assay. The inhibition of Arg484 → Ala, Tyr487 → Ala, Arg489 → Ala, and Arg492 → Ala by mAb413 was reduced by greater than 90% compared with wild-type, B-domainless human fVIII. mAb413 inhibited the most severely affected mutant, Arg489 → Ala, 0.01% as well as wild-type fVIII. For all five patient plasmas, the Tyr487 → Ala mutant displayed the greatest reduction in inhibition. The inhibition of the Tyr487 → Ala mutant by these antibodies ranged from 10% to 20% that of wild-type fVIII. The inhibition of the Ser488 → Ala, Arg489 → Ala, Pro492 → Ala, Val495 → Ala, Phe501 → Ala, and Ile508 → Ala mutants by most of the plasmas also was significantly reduced. In contrast, the Arg484 → Ala and Pro485 → Ala mutants were relatively unaffected. Thus, although mAb413 binds to the same region as human A2 inhibitors, it recognizes a different set of amino acid side chains. The side chains recognized by human A2 inhibitors appear to be similar, despite the differing immune settings that give rise to fVIII alloantibodies and autoantibodies.

Nonclassical anti-C2 domain antibodies are present in patients with factor VIII inhibitors

The antihuman factor VIII (fVIII) C2 domain immune response in hemophilia A mice consists of antibodies that can be divided into 5 groups of structural epitopes and 2 groups of functional epitopes. Groups A, AB, and B consist of classical C2 antibodies that inhibit the binding of fVIII to phospholipid and von Willebrand factor. Groups BC and C contain nonclassical C2 antibodies that block the activation of fVIII by thrombin or factor Xa. Group BC antibodies are the most common and display high specific inhibitory activity and type II kinetics. The C2 epitope groups recognized by 26 polyclonal human anti-fVIII inhibitor plasmas were identified by a novel competition enzyme-linked immunosorbent assay using group-specific murine monoclonal antibodies. Most of the anti-C2 inhibitor plasmas inhibited the binding of both classical and nonclassical antibodies. These results suggest that nonclassical anti-C2 antibodies contribute significantly to the pathogenicity of fVIII inhibitors.

The humoral response to human factor VIII in hemophilia A mice

Background: Inhibitory antibodies (Abs) to factor VIII (FVIII inhibitors) constitute the most significant complication in the management of hemophilia A. The analysis of FVIII inhibitors is confounded by polyclonality and the size of FVIII. Objectives: The goal of this study was to dissect the polyclonal response to human FVIII in hemophilia A mice undergoing a dosage schedule that mimics human use. Methods: Splenic B‐cell hybridomas were obtained following serial i.v. injections of submicrogram doses of FVIII. Results of a novel, anti‐FVIII domain‐specific enzyme‐linked immunosorbent assay were compared to Ab isotype and anti‐FVIII inhibitory activity. Results: The robust immune response resulted in the production of ∼300 hybridomas per spleen. We characterized Abs from 506 hybridomas, representing the most comprehensive analysis of a protein antigen to date. Similar to the human response to FVIII, anti‐A2 and anti‐C2 Abs constituted the majority of inhibitors. A novel epitope was identified in the A2 domain by competition ELISA. Anti‐A2 and anti‐C2 Abs were significantly associated with IgG1 and IgG2a isotypes, respectively. Because the IgG2a isotype is associated with enhanced Fc receptor‐mediated effector mechanisms, this result suggests that anti‐C2 Abs and inflammation may be linked. Additionally, we identified a novel class of Abs with dual specificity for the A1 and A3 domains. Forty per cent of the Abs had no detectable inhibitory activity, indicating that they are prominent and potentially pathologically significant. Conclusion: The expanded delineation of the humoral response to FVIII may lead to improved management of hemophilia A through mutagenesis of FVIII B‐cell epitopes.

Comparative immunogenicity of recombinant B domain‐deleted porcine factor VIII and Hyate:C in hemophilia A mice presensitized to human factor VIII

Summary.  Hyate is a commercial plasma‐derived porcine factor (F)VIII concentrate that is used in the treatment of patients with inhibitory antibodies to FVIII. OBI‐1 is a recombinant B domain‐deleted form of porcine FVIII that is in clinical development for the same indication. Hemophilia A mice were presensitized with human FVIII to simulate clinical inhibitory antibody formation and then were randomized to receive OBI‐1 or Hyate:C in a comparative immunogenicity trial. OBI‐1 or Hyate:C were given in a series of four intravenous injections at weekly intervals at doses of 1, 10, or 100 U kg−1. Inhibitory antibodies to porcine FVIII were not detected by Bethesda assay in most of the mice given OBI‐1 or Hyate:C at doses of 1 or 10 U kg−1, but were identified in 81% and 94% of mice given 100 U kg−1 of OBI‐1 or Hyate:C, respectively. There was no significant difference between OBI‐1 and Hyate:C in inhibitory antibody formation at any dose, although there was a trend toward a lower Bethesda titer in OBI‐1‐treated mice at 10 U kg−1 (P = 0.09). Total anti‐FVIII antibodies to Hyate:C and OBI‐1 were also measured by ELISA using immobilized purified plasma‐derived porcine FVIII and OBI‐1, respectively, as antigens. At the 10 and 100 U kg−1 doses, the mean anti‐FVIII response was higher in Hyate:C‐treated‐mice than in OBI‐1‐treated mice (P = 0.02 and P = 0.004, respectively). The results using this model suggest that OBI‐1 may be less immunogenic and safer than Hyate:C in FVIII inhibitor patients.

Non-classical anti-factor VIII C2 domain antibodies are pathogenic in a murine in vivo bleeding model.

Summary.  Objective: The pathogenicity of anti‐human factor (F) VIII monoclonal antibodies (MAbs) was tested in a murine bleeding model. Methods: MAbs were injected into the tail veins of hemophilia A mice to a peak plasma concentration of 60 nm, followed by injection of human B domain‐deleted FVIII at 180 U kg−1, producing peak plasma concentrations of ∼2 nm. At 2 h, blood loss following a 4‐mm tail snip was measured. The following MAbs were tested: (i) 4A4, a type I anti‐A2 FVIII inhibitor, (ii) I54 and 1B5, classical type I anti‐C2 inhibitors, (iii) 2–77 and B45, non‐classical type II anti‐C2 inhibitors, and (iv) 2–117, a non‐classical anti‐C2 MAb with inhibitory activity less than 0.4 Bethesda Units per mg IgG. Results: All MAbs except 2–117 produced similar amounts of blood loss that were significantly greater than control mice injected with FVIII alone. Increasing the dose of FVIII to 360 U kg−1 overcame the bleeding diathesis produced by the type II MAbs 2–77 and B45, but not the type I antibodies, 4A4, I54, and 1B5. These results were consistent with the in vitro Bethesda assay in which 4A4 completely inhibited both 1 U mL−1 and 3 U mL−1 FVIII, while there was 40% residual activity at saturating concentrations of 2–77 at either concentration of FVIII. Conclusions: For patients with an inhibitor response dominated by non‐classical anti‐C2 antibodies both the in vivo and in vitro results suggest that treatment with high‐dose FVIII rather than bypassing agents may be warranted.

Pathogenic antibodies to coagulation factors. Part II. Fibrinogen, prothrombin, thrombin, factor V, factor XI, factor XII, factor XIII, the protein C system and von Willebrand factor

In Part I of this review, studies of antibodies to factor (F)VIII and factor IX were reviewed [1]. In Part II, antibodies to the other components of the coagulationmechanism are described. Because the coagulationmechanism contains antihemostatic as well as prohemostatic factors, antibodies to these factors may produce thrombosis or bleeding, depending on which way they tip hemostatic balance. Hemostasis is regulated primarily by control of the substrate specificity of thrombin and by serpindependent inactivation of thrombin. The prohemostatic functions of thrombin, which are mediated by its conversion of fibrinogen to fibrin and activation of platelet receptors and factor (F)V, FVIII, factor (F)XI and factor (F)XIII, are balanced by antihemostatic functions that are mediated by its activation of protein C and release of prostacyclin and tissue plasminogen activator from endothelial cells. In the protein C system, thrombomodulin is a cofactor for the activation of protein C by thrombin on the surface of endothelial cells [2]. Thrombomodulin inhibits the recognition by thrombin of its prohemostatic substrates and has anti-inflammatory activity by decreasing endothelial cytokine formation and by decreasing leukocyte–endothelial cell adhesion.The endothelial cell protein receptor (EPCR) binds protein C and promotes the binding of protein C to the thrombin–thrombomodulin complex. Activated protein C (APC) binds protein S on the surface of platelets and other cells and inactivates FVa and FVIIIa. Because of the dynamic interplay within the network of the coagulation mechanism, clinically significant antibodies to its components often present difficult diagnostic and therapeutic challenges. Anti-von Willebrand factor (VWF) antibodies are also discussed, because, although VWF is not a coagulation factor, it is intimately related to the coagulation mechanism through its interaction with FVIII. Acquired thrombotic thrombocytopenic purpura due to autoantibodies that inhibit the function of ADAMTS-13 is not discussed in this review. Antibodies to platelet factor 4, which are responsible for heparin-induced thrombocytopenia, also logically fall within the domain of pathogenic antibodies to coagulation factors. They have recently been reviewed in this journal [3].

Factor VIII inhibitors.

The differential diagnosis in the bleeding patient includes inhibitory antibodies to blood coagulation proteins. Factor VIII (fVIII) is the most commonly targeted coagulation protein by the immune system. FVIII inhibitors arise as alloantibodies in transfused hemophiliacs and as autoantibodies in nonhemophiliac populations (1, 2, 3, 4). They develop in response to fVIII infusions in approximately 25% of patients with hemophilia A (5). In nonhemophiliacs, fVIII autoantibodies develop in a variety of clinical settings, including the postpartum period, systemic erythematosus, and chronic lymphocytic leukemia. Interestingly, autoantibody patients tend to present with more severe bleeding than hemophilia A patients with a de novo inhibitor. In both settings, fVIII inhibitors are polyclonal IgG populations directed against multiple epitopes.