Use of porcine factor VIII in the treatment of patients with acquired hemophilia

Data have been collected from 47 centers in Europe and North America on the treatment with porcine factor VIII concentrate of 74 acute bleeding episodes in 65 patients with acquired hemophilia. The median initial anti-human factor VIII auto-antibody inhibitor level was 38 Bethesda unit (BU)/mL (range 1.2 to 1,024) whereas that against porcine was 1 BU/mL (range 0 to 15). The mean initial dose of porcine factor VIII infused was 84 IU/kg, which increased the plasma factor VIII:C activity by 0.85 IU/mL. Therapy was continued for a mean of 8.5 days during which time the average number of infusions was 11. Objective clinical responses were rated as good or excellent in 78% of recipients. Side effects were uncommon; only one patient experienced a severe anaphylactic reaction necessitating the discontinuation of porcine FVIII therapy. After therapy, no increase in the median level of anti- human FVIII or anti-porcine antibody was noted in the group as a whole, although 13 patients showed individual increases in either anti-human or anti-porcine antibody levels or both of more than 10 BU/mL. Of the 7 patients who subsequently rebled, 5 were successfully re-treated and 2 did not respond to further porcine factor VIII treatment. Porcine factor VIII is safe and clinically effective treatment for bleeding episodes associated with acquired hemophilia and should be considered as first-line therapy for patients whose acquired anti-factor VIII:C antibody cross-reacts with porcine factor VIII:C at low levels.     

Haemophilia due to factor VIII inhibitors in a patient suffering from an autoimmune disease: Treatment with intravenous immunoglobulin

Summary This paper describes a case of haemophilia due to factor VIII inhibitors occurring in a 13-year-old boy suffering from an autoimmune disease. The patient had autoantibodies to factor VIII. The haemophilia was controlled by vincristine and steroids, but this regimen had to be discontinued because of side effects, whereupon the haemophilia recurred. Treatment with intravenous immunoglobulin (IgG i.v.) produced a slow rise in factor VIII, and the factor VIII inhibitors disappeared. Although factor VIII activity was raised for only a few months and factor VIII inhibitors reappeared, immunoglobulin treatment was continued and the patient remained remained free of clinical symptoms. The mechanism of action of treatment with IgG is discussed.     

Disappearance of factor VIII:C antibodies in patients with haemophilia A upon frequent administration of factor VIII in intermediate or low dose

In 18 haemophilia A patients with antibodies against factor VIII:C (F VIII:c) the effect of regular treatment with factor VIII (F VIII) in intermediate or low dose was studied. All patients with previous maximal F VIII:c antibody levels between 5 and 60 Bethesda Units per millilitre (BU/ml) showed a decrease of antibody level and normal F VIII recovery within 1-2 months. From nine patients with previous maximal antibody levels above 60 BU/ml four showed a decrease of antibody level within 2-26 months. In four young patients F VIII prophylactic therapy was started or continued as soon as there was evidence of F VIII:c antibody activity. In three of these patients F VIII recovery normalized within a few months.     

Characterization of an antibody to factor VIII in a patient with acquired hemophilia with circulating immune complexes

A 73-year-old previously healthy woman was admitted because of severe bleeding from esophagitic lesions and intraabdominal bleeding following hysterectomy. Acquired hemophilia, probably due to an IgG antibody to factor VIII (64 inhibitor units/ml) was noticed, the VIII:C in the patient's plasma being 18% or normal. Immune complexes isolated by polyethylene glycol precipitation had only a weak factor VIII inhibiting activity whereas IgG purified from the complexes and monomeric IgG present in her plasma exerted a strong inhibition. Removal of the complexes from plasma had no effect on the inhibitor titer thus indicating that only a minor part of the antibody was circulating as immune complexes. Plasma or purified IgG from the patient decreased the VIII:C of normal plasma to 18 og 14%, respectively, total inhibition being impossible to achieve even in antibody excess, probably reflecting residual activity of factor VIII bound to the patient's antibodies. The ristocetin cofactor activity of normal plasma was unaffected by the antibodies. Transfusion of factor VIII concentrate to the patient resulted in therapeutic levels of circulating factor VIII and transfused factor VIII circulated longer than usual. Partial remission of the disease with adequate levels of VIII:C occurred after 3 months.     

Efficacy of porcine factor VIII in the management of haemophiliacs with inhibitors

We report our experience with polyelectrolyte-fractionated highly purified porcine factor VIII concentrate (Hyate:C). Porcine factor VIII concentrate was used to treat 14 haemorrhagic episodes including seven severe haemarthroses, three severe haematomata, two episodes of oral bleeding, one traumatic and the other after multiple dental extractions and two life-threatening intestinal haemorrhages. Altogether 60 infusions of Hyate:C were given to five haemophiliacs with inhibitors. At the time of the first infusion with porcine factor VIII the anti-human inhibitor level ranged from 60 to 2-5 modified Bethesda units/ml (MBU 4 h incubation). The porcine cross-reactivity of these inhibitors was 25-6% and 5-4% of the human inhibitor level in two patients and scarcely measurable in the other three patients. We injected an amount of porcine factor VIII concentrate corresponding to the calculated porcine neutralizing units plus the units required for haemostasis. The clinical efficacy was excellent and no adverse effects were encountered, apart from two episodes of mild pyrogenic reactions well controlled by hydrocortisone. The anamnestic antibody response against human factor VIII was of slight degree in all cases, while the porcine cross-reactivity showed no significant variation. Our results emphasize the role of porcine factor VIII in the treatment of haemophiliacs with inhibitors.     

Immunoblot cross-reactivity of factor VIII inhibitors with porcine factor VIII

Porcine factor VIII has been used successfully to treat factor VIII inhibitor patients whose plasmas have minimal cross-reactivity to porcine factor VIII. However, some inhibitor plasmas do inhibit porcine factor VIII, and the extent of procoagulant inhibition often increases after treatment with porcine factor VIII. Because there is no information about the porcine factor VIII epitopes with which these antibodies react, we have compared the immunoblot and enzyme-linked immunosorbent assay (ELISA) reactivities with porcine and human factor VIII for 20 inhibitor plasmas (11 from hemophilia A patients and 9 autoantibodies). Immunoblots identified binding to porcine factor VIII for only 2 of the 12 plasmas from patients who had not received porcine factor VIII, but this reactivity could not be predicted from the inhibitor titer to porcine factor VIII. Immunoblot reactivity with porcine factor VIII was detected for 7 of 8 inhibitor plasmas from patients who had been previously treated with porcine factor VIII, and the strength of this reactivity was generally related to the inhibitor titer. Of the 5 plasmas that were immunoblot positive with the porcine factor VIII A2 domain, 4 had inhibitor titers greater than 45 Bethesda units when tested with porcine factor VIII, whereas only 1 of 15 of the other plasmas had this level of inhibitor activity with porcine factor VIII. In contrast, immunoblot reactivity to the porcine factor VIII A1 domain did not correlate with the antiporcine VIII inhibitor titer. We also determined the effect of preincubation with human or porcine factor VIII on immunoblot reactivity. In one case, immunoblot reactivity with porcine factor VIII was absorbed with porcine, but not human, factor VIII, which is consistent with antibody formation after treatment with porcine factor VIII. In no cases did human factor VIII reduce the reactivity of inhibitor plasmas with the porcine A1 domain, suggesting that these antibodies are directed at unique porcine factor VIII determinants. The reactivity to porcine A2 in 2 plasmas probably represented cross-reactivity of similar A2 determinants, because it was absorbed by both human and porcine factor VIII. Although the ELISA assays with porcine factor VIII detected antibodies in some plasmas that could not be identified by inhibitor assay or immunoblot, the level of ELISA reactivity was generally consistent with the titers of the other assays.     

The Relevance of Factor VIII Inactivation Characteristics in the Treatment of Patients with Antibodies Directed against Factor VIII

The management of a patient with acquired haemophilia who sustained a compound fracture of the tibia is described. Haemostasis was achieved with infusions of large doses of porcine antihaemophilic globulin (AHG) and human cryoprecipitate. Investigations of the patient's antibody showed that when it was in excess it initially inactivated both porcine and human factor VIII rapidly, and then at a rate which decreased with progressive incubation. This decreasing rate of factor VIII inactivation differs from the action of a classically described factor VIII antibody which inactivates factor VIII in a linear manner when the antibody is in excess. It is suggested that the patient was more amenable to treatment with concentrated preparations of factor VIII than a patient with the classical type of factor VIII antibody, infused factor VIII persisting for longer in the patient's circulation than might be expected.     

Clinical experience with polyelectrolyte-fractionated porcine factor VIII concentrate in the treatment of hemophiliacs with antibodies to factor VIII

Circulating antibodies to factor VIII (anti-VIII, "inhibitors") occurring in patients with hemophilia neutralize porcine factor VIII less readily than human factor VIII in vitro. Over an 18-mo period, 8 patients with anti-VIII were treated with 45 courses (297 infusions) of polyelectrolyte-fractionated porcine factor VIII concentrate (PE porcine VIII). Where no anti-PE porcine VIII was detectable, mean post- infusion rise in plasma factor VIII was 1.29 U/dl/units infused/kg. Above 13 Old Oxford units of anti-PE porcine VIII and 48 Bethesda units of anti-human VIII, there were no postinfusion rises in plasma factor VIII. Where postinfusion rises were detected, clinical responses were good and conventional methods could be used to guide dosage. Ten percent of infusions were followed by febrile reactions, but these were usually mild and decreased in frequency and severity with increasing exposure. Multiple and prolonged courses of therapy were given to some patients without evidence of loss of clinical or laboratory efficacy. PE porcine VIII could provoke anamnestic rises of anti-VIII in susceptible patients, but appeared to have a lower immunogenic potential than human VIII. PE porcine VIII is a rational and effective therapeutic alternative for patients with anti-VIII, particularly those with intermediate level inhibitors who cannot be managed effectively using human factor VIII.     

Reduction of the inhibitory antibody response to human factor VIII in hemophilia A mice by mutagenesis of the A2 domain B-cell epitope

Approximately 25% of patients with hemophilia A develop inhibitory antibodies after treatment with factor VIII. Most of the inhibitory activity is directed against epitopes in the A2 and C2 domains. Anti-A2 inhibitory antibodies recognize a 25-residue segment bounded by R484-I508. Several antigenic residues in this segment have been identified, including R484, R489, and P492. The immunogenicity of purified recombinant B domain-deleted (BDD) human factor VIII molecules containing mutations at R484A/R489A or R484A/R489A/P492A was studied in hemophilia A mice. Inhibitory antibody titers in mice receiving the R484A/R489A/P492A mutant, but not the R484A/R489A mutant, were significantly lower than in mice receiving control human BDD factor VIII. The specific coagulant activity and the in vivo clearance and hemostatic efficacy in hemophilia A mice of the R484A/R489A/P492A mutant were indistinguishable from human BDD factor VIII. Thus, the inhibitory antibody response to human factor VIII can be reduced by mutagenesis of a B-cell epitope without apparent loss of function, suggesting that this approach may be useful for developing a safer form of factor VIII in patients with hemophilia A.     

Molecular characterization of commercial porcine factor VIII concentrate

Commercial porcine factor VIII concentrate (Hyate:C) is effective in treatment of patients with hemophilia A who have circulating antibodies to factor VIII. The molecular forms of factor VIII in the concentrate were identified and evaluated in light of the known properties of porcine and human factor VIII. The factor VIII in the concentrate was isolated by tandem chromatography on gelatin-Sepharose and monoclonal anti-factor VIII-Sepharose. The factor VIII was 1% of the protein mass of the concentrate when calculated by either quantity of protein recovered or by radioimmunoassay. Both functional assay and Western blotting of the crude concentrate indicated that maximum coagulant function was achieved by proteolytic activation of procofactor forms of factor VIII. The factor VIII can be fractionated by cation-exchange high-performance liquid chromatography (HPLC) into two or three species of heterodimers depending on the lot. The specific activity of the purified porcine factor VIII was 550 U/mg using pooled porcine plasma at 1 U/mL as a standard. From this value, a factor VIII concentration in normal pig plasma of 2 micrograms/mL was calculated. This agreed well with a value of 3 micrograms/mL obtained by radioimmunoassay (RIA) of factor VIII in porcine plasma. In contrast, reported values for human factor VIII average 5800 U/mg, resulting in a calculated concentration in plasma of 0.2 microgram/mL. The finding that porcine plasma contains a significantly higher circulating mass of factor VIII than human plasma appears to explain previous difficulties in comparing porcine and human factor VIII in standard assays.     

Transmission of Human Parvovirus B19 by Coagulation Factor Concentrates

The prevalence of antibody to human parvovirus B19 was determined in 86 children with congenital bleeding disorders. Forty-seven of 53 boys (89%) receiving non-heat-treated factor VIII or prothrombin complex concentrates were anti-B19 IgG positive compared with 38% of their age-matched controls and 48% of children treated with cryoprecipitate. Acute B19 virus infection occurred in 2 boys 3-4 weeks after they had received the same batch of commercial factor VIII concentrate. Of 11 susceptible children who had only received heat-treated National Health Service factor VIII concentrate (8Y), 1 acquired anti-B19 IgG. This suggests that 8Y heat-treated concentrate has a much reduced risk of transmitting B19 virus and, by implication, other less heat-stable viruses such as human immunodeficiency virus.     

ENHANCEMENT OF THE ACTIVITY OF THYROID-STIMULATING ANTIBODIES BY ANTI-HUMAN IgG ANTIBODIES IN VITRO

In an attempt to determine whether or not anti-human IgG antibodies could influence the activity of thyroid-stimulating antibodies (TSAb), we investigated the effects of anti-human Fc antibody (anti-Fc antibody) and IgG from rheumatoid arthritis patients (RA-IgG) on TSAb activity using FRTL-5 cells. It was found that these anti-human IgG antibodies enhanced the TSAb activity in vitro. FRTL-5 cells were first incubated with Graves' disease IgG for 30 min at 37 degrees C, then washed and incubated in Hanks' balanced salt solution with anti-human IgG antibodies for 60 min at 4 degrees C, and then for a further 120 min at 37 degrees C. The level of cAMP accumulated in the medium was determined by RIA. Anti-Fc antibody significantly augmented the cAMP formation stimulated by 16 out of 24 Graves' IgGs, whereas anti-F(ab')2 antibody did not potentiate cAMP accumulation. Three of five RA-IgGs, which are usually defined as specific antibodies for the Fc fragment of human IgG, mimicked these stimulatory effects. Protein A also potentiated the cAMP formation stimulated by Graves' IgGs. Furthermore, there was a significant correlation between the TSAb titres of these Graves' IgGs and the potentiating effects of anti-Fc antibody (r = 0.495, P less than 0.05, n = 21). These results suggest that the interaction of TSH receptor antibody with anti-human IgG antibodies might modulate thyroid function in Graves' disease.     

Acquired von Willebrand syndrome due to an inhibitor specific for von Willebrand factor antigens

A patient with acquired von Willebrand syndrome associated with polycythemia rubra vera is described. Her plasma factor VIII procoagulant activity (67 U/dl) and factor VIII-related antigen (117 U/dl) were normal but no von Willebrand factor activity could be detected. Factor VIII crossed immunoelectrophoresis revealed decreased levels of less anodic polymeric forms of factor VIII. Mixture of her plasma or immunoglobulin G (IgG) fraction with normal plasma resulted in complete recovery of factor VIII activity and related antigen but no measurable von Willebrand factor activity, confirming the presence of an unique inhibitor. The limited specificity of this inhibitor to antigenic sites solely on the von Willebrand portion of the factor VIII bimolecular complex is distinct from all previous reports of this syndrome. This unique inhibitor offers a molecular probe to examine the von Willebrand factor: platelet interaction.     

IN-VITRO CONVERSION OF BLOCKING TYPE ANTI-TSH RECEPTOR ANTIBODY TO THE STIMULATING TYPE BY ANTI-HUMAN IgG ANTIBODIES

We examined the relationship between blocking type anti-TSH receptor antibody (TRAb) and stimulating type TRAb by trying to convert blocking type TRAb to the stimulatory type in vitro. Immunoglobulins (IgGs) purified from sera of six patients with primary hypothyroidism blocked bovine TSH (100 μU/ml)-induced cAMP production (58-3-82-1% inhibition) in cultured porcine thyroid cells. None of these IgGs showed stimulating activity even after their dilution. In the conversion experiment, thyroid cells were first incubated with these IgGs at 34°C for 30 min and then washed with incubation buffer. They were then incubated with various kinds of anti-human IgG antibodies first for 1 -5 h at 4°C and then for 18 h at 34°C, and the cAMP concentrations in the supernatants were measured. All six IgGs showed strong, dose-dependent thyroid-stimulating activity after addition of antibodies against human whole IgG, or Fab or Fc fragments of IgG. The Fab or F(ab')2 fragments of goat anti-human IgG antibody also had these converting activities, although less than whole IgG. Addition of normal IgG in the first incubation or anti-human IgG antibody alone had no thyroid-stimulating activity. Anti-human IgA or IgM antibodies did not have these converting activities. These results show that blocking type anti-TSH receptor antibodies can be converted to the stimulating type by anti-human IgG antibodies in vitro. The results suggest that the blocking and stimulating types bind to the same epitope(s) of TSH-receptor related antigens. The same anti-TSH receptor antibody may act as a stimulator or blocker by the influence of other factors, such as anti-idiotype antibody.     

Noncoagulation inhibitory factor VIII antibodies after induction of tolerance to factor VIII in hemophilia A patients

We recently described tolerance induction with factor VIII/IX, cyclophosphamide, and high-dose intravenous IgG in hemophilia A or B patients with coagulation inhibitory antibodies. Circulating noninhibitory antibodies complexed with factor IX have been demonstrated in tolerant hemophilia B patients. Similar findings are now described in six tolerant hemophilia A patients. Complexes between factor VIII and the 'tolerant' antibody were demonstrated by subjecting plasma to gel filtration chromatography, void fractions containing factor VIII/vWF complexes being collected and adsorbed to protein A. Using 125I-labeled F(ab')2 fragments against IgG subclass and factor VIII antigen, complexes between an IgG4 antibody and factor VIII were found to adsorb to protein A. After infusion of factor VIII to tolerant patients, all factor VIII circulated in complex with IgG4 antibody. In three of the patients, the 'tolerant' antibodies inhibited an ELISA specific for factor VIII light chain but, unlike the pretolerant antibodies, did not bind radiolabeled factor VIII heavy chain. Although after induction of tolerance the patients still have circulating IgG4 antibodies against factor VIII, the antibodies differ in specificity, lack coagulation inhibitory activity, and do not enhance the rate of elimination of factor VIII.     

Pathogenesis of antibody-induced acquired von willebrand syndrome

A patient with clinical and laboratory evidence of von Willebrand syndrome is described in association with an IgG-kappa immunoglobulin and Bence-Jones proteinuria due to a probable lymphoproliferative disorder. He had a persistently prolonged bleeding time of greater than 20 minutes, factor VIII related antigen (VIII: R. Ag), factor VIII procoagulant activity (VIII:C) and factor VIII ristocetin co-factor (VIIIR: Rcof) below 10%. Following cryoprecipitate or high purity factor VIII concentrate infusion, he had the expected immediate rise in VIII:C, VIII: R. Ag, and VIIIR: Rcof, but there was a rapid decline in all three components within two hours. The larger forms of VIII: R. Ag were preferentially removed from the plasma, and this paralleled the fall in plasma VIIIR: Rcof level. However, no inhibitory activity could be demonstrated in vitro using the patient's plasma or IgG. Using protein A it was possible to demonstrate that his plasma or IgG bound factor VIII and that this complex retained its biological activity in vitro. It is postulated that the monoclonal IgG forms complexes with factor VIII in vivo and these are rapidly removed by the reticuloendothelial system (RES).     

Comparison of Anti-Human and Anti-Porcine Factor VIII Inhibitor Levels in 63 Patients with Severe Haemophilia A

The levels of anti-human and anti-porcine factor VIII inhibitors, measured in 63 severe haemophilia A patients, lay in the ranges of <0.2-2,600 and <4.2-1,300 Bethesda units per ml (BU/ml), respectively, with a median cross-reactivity of 33%. In 4 patients, human and porcine inhibitor levels were determined using both plasma, either human or porcine, and factor VIII concentrate, either very high purity human or porcine (Hyate:C®). A good correlation between titres was found, whatever the source of factor VIII (plasma or concentrate). The cross-reactivity varies from 0 to over 100%, indicating that the evaluation of both human and porcine inhibitors should be mandatory before any treatment with Hyate:C. Results show that of the 46 patients with human inhibitor of more than 5 BU/ml, 21 (46%) with a low porcine inhibitor (<5 BU/ml) could benefit from Hyate:C.     

Structure-Function of the Factor VIII Complex Studied with an Immobilized Heteroantisera to VIII:C

An antibody was raised in rabbits to the small active fragment of human factor VIII, obtained by Ca²⁺ dissociation of a human factor VIII preparation made from a multidonor plasma pool. After absorption, the antibody neutralized the factor VIII coagulant activity of normal human plasma, but did not precipitate with any plasma or plasma fractions or neutralize von Willebrand factor (vWF) activity as measured by ristocetin aggregation of fixed washed platelets. Immune beads were prepared by CNBr binding of the partially purified rabbit antibody to 1% agarose beads. Non-immune beads were prepared with IgG fractions obtained from the rabbits before immunization and used throughout as a control. The amount of factor VIII coagulant activity (VIII:C) removed from plasma by immune beads was time-dependent and proportional to the amount of beads used, but all of the VIII:C could not be readily removed. Removal of VIII:C by immune beads parallelled removal of factor VIII:antigen, but less vWF activity was removed. Immune beads could be blocked or saturated by treatment with large amounts of normal plasma, but not by von Willebrand disease plasma and only by some haemophilic plasmas.     

Prevention of occlusive coronary artery thrombosis by a murine monoclonal antibody to porcine von Willebrand factor.

A murine monoclonal antibody (mAb) against porcine von Willebrand factor (vWF) induced an antithrombotic state in normal pigs. Thrombosis was induced by a standard procedure of stenosis and mechanical injury of the artery. The mAb was an IgG1 kappa that inhibited vWF-induced platelet aggregation at a titer of 1:6250 and bound to immobilized vWF at a maximal dilution of 1:512,000. The antibody did not affect two other vWF functions, platelet adhesion and binding of coagulant factor VIII (factor VIII:C). The antithrombotic state was characterized by a prolonged bleeding time and lack of plasma vWF activity, but with near-normal levels of factor VIII:C and von Willebrand antigen. The circulating Ag.mAb complex demonstrated a multimeric distribution comparable to that of native plasma vWF. Three groups of pigs were studied: group A consisted of nine untreated animals, eight of which developed occlusive coronary thrombosis; group B, four treated animals with a long bleeding time, none of which developed occlusive thrombosis; and group C, two animals with preexisting thrombosis treated with mAb, in which stable blood flow was reestablished. Morphologically, the group B animals showed adherent platelets covering the injured intima but no thrombosis. This mAb is an antithrombotic agent that prevents platelet thrombosis without affecting intrinsic platelet function.     

Immunosuppressive Treatment in Haemophiliacs with Inhibitors to Factor VIII and Factor IX

9 patients with severe haemophilia A and inhibitors (inhibitor levels between 0.1 to 5.8 U/ml) and 3 patients with severe haemophilia B and inhibitors (inhibitor levels between 0.1 to 11 U/ml) were treated on a total of 16 and 13 occasions, respectively, with a large dose of antigen (factor VIII or factor IX) and cyclophosphamide (10–15 mg/kg b.w. i.v. initially and then 2–3 mg/kg b.w. orally for 7–10 days) in connection with severe bleeding and surgery. All the patients had proved not to respond to treatment with factor VIII or factor IX concentrate alone, and all except one had shown strong secondary antibody increases. In 6 of the patients with haemophilia A the treatment (11 occasions) had a satisfactory haemostatic effect and even permitted neurosurgery without bleeding complications. The inhibitor level remained at zero for 5–10 days, after which it gradually began to return towards its original level. In these cases it was possible to give factor VIII in amounts which neutralised the inhibitor and afterwards raised the factor VIII initially to at least 50%. In the 3 patients with haemophilia B treatment (13 occasions) was successful except on one occasion, and surgery was performed without abnormal bleeding. The factor IX level was initially raised to at least 50% except in the one failure. The inhibitor level remained at zero for 12 days to 3 months, after which it gradually rose towards its original level. One patient was treated on 8 occasions.