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

Approximately 30% of patients with hemophilia A develop detectable antifactor VIII (fVIII) antibodies in response to infusions of fVIII.^1-4  The immune response to fVIII currently is the most significant complication in the management of patients with hemophilia A. In addition, autoimmune antibodies to fVIII can develop in nonhemophiliacs, producing acquired hemophilia A, which frequently produces life- or limb-threatening bleeding. Most inhibitory antibodies are directed at either the 40-kDa A2 or the 15-kDa C2 domains of the A1-A2-B-ap-A3-C1-C2 fVIII sequence.⁵  fVIII inhibitors can either inhibit fVIII completely or incompletely at saturating concentrations, corresponding to type I and type II behavior, respectively.⁶  Classical anti-C2 antibodies inhibit binding of fVIIIa to negatively charged phospholipid membranes.^7-9  The binding of fVIII to phospholipid membranes and to von Willebrand factor (VWF) is mutually exclusive, and antibodies have been shown to block binding to both phospholipid and/or VWF.^10-14  In addition, murine anti-C2 monoclonal antibodies (mAbs)^15,16  and anti-C2 antibodies in 2 polyclonal patient plasmas^16,17  have been identified that interfere with the activation of fVIII by thrombin or factor Xa

We recently characterized the diversity of a large panel of murine anti-C2 mAbs.¹⁸  Five groups of structural epitopes were defined based on patterns of overlapping epitopes. Group A, AB, and B antibodies correspond to classical inhibitors that inhibit the binding of fVIII to phospholipid and VWF. Group BC antibodies are the most frequent and are type II inhibitors with inhibitory titers usually greater than 10 000 Bethesda units per mg immunoglobulin G. These antibodies inhibit the activation of fVIII by thrombin and factor Xa in the presence and absence of VWF. ESH8, a well-characterized murine anti-C2 mAb, which blocks the release of VWF from fVIII after thrombin activation, is a group C mAb.¹⁶  In this study, we used murine group-specific antihuman C2 mAbs in a competition enzyme-linked immunosorbent assay (ELISA) to determine whether nonclassical group BC and C antibodies are present in human fVIII inhibitor patients.

fVIII inhibitor plasmas from 26 patients with congenital hemophilia A or acquired hemophilia A were obtained either as described previously^19,20  from the Emory Comprehensive Hemophilia Center or from George King Bio-Medical (Overland Park, KS). Recombinant full-length human fVIII was a gift from Baxter Biosciences (Duarte, CA). mAbs ESH-4 (group A) and ESH-8 (group C) were purchased from American Diagnostica (Greenwich, CT). mAbs 3E6 (group A), I109 (group AB), 1B5 (group B), 2-77 (group BC), and 2-117 (group C) were isolated as described previously.¹⁸  mAbs were biotinylated as previously described.¹⁸ 

Anti-fVIII ELISAs were performed as a modification of previously described procedures.¹⁸  Briefly, ELISA plates were coated with fVIII, preincubated with 3 μg/mL of a nonbiotinylated murine antihuman C2 “blocking” mAb, followed by addition of various concentrations of biotinylated antihuman C2 mAb diluted one-ninth in test inhibitor plasma or control (severe hemophilia A noninhibitor) plasma (Figure 1). The blocking mAbs used were 2-77, 3E6/2-117, and I109 for biotinylated mAbs ESH4, 1B5, and ESH8, respectively. Bound biotinylated mAb was quantitated using alkaline-phosphatase conjugated streptavidin and p-nitrophenyl-phosphate.

ELISA titration curves of bound biotinylated mAb binding were fitted to the 4-parameter logistic equation. The mAb concentration required to produce an A₄₀₅ of 0.5 (EC_0.5) was calculated by interpolation on the fitted curve. The corresponding mAb titer is defined as EC_0.5⁻¹. The normal range of EC_0.5 values for the binding of biotinylated mAbs was estimated by performing 8 replicate mAb titrations of the control plasma. EC_0.5 values (mean ± SD) for ESH4, 1B5, and ESH8 in control plasma were 285 (± 39) ng/mL, 32.6 (± 4.0) ng/mL, and 23.1 (± 2.3) ng/mL, respectively. The corresponding normal range of control plasma ELISA titers was defined using EC_0.5 values within 2 SDs from the mean.

The ability of human fVIII inhibitors to compete with the binding of biotinylated murine antihuman fVIII mAbs from groups A (mAb ESH4), B (mAb 1B5), or C (mAb 2-117) to fVIII was evaluated by ELISA (Figure 2). A second, nonbiotinylated mAb also was added to target the specificity of the test plasmas. The principle of the assay is illustrated in Figure 1 for the identification of antibodies with group BC/C specificity. Competition with ESH4 was carried out by preincubating the fVIII-coated ELISA plate with 2-77, a group BC mAb. Group BC antibodies inhibit the binding of group B, BC, and C antibodies. Thus, inhibition of biotinylated ESH4 binding to fVIII under these conditions is limited to group A or group AB antibodies. Similarly, competition with 1B5 was carried out by preincubating the fVIII-coated ELISA plate with 3E6, a group A mAb, and 2-117, a group C mAb. The group A and C mAbs collectively inhibit the binding of antibodies from groups A, AB, BC, and C. Thus, inhibition of biotinylated 1B5 binding to fVIII is limited to group B antibodies. Finally, competition with ESH8 was carried out by preincubating the fVIII-coated ELISA plate with I109, a group AB mAb, which inhibits the binding of antibodies from groups A and AB. Thus, inhibition of biotinylated ESH8 binding to fVIII is limited to nonclassical group BC or group C anti-C2 antibodies.

Figure 2C indicates that most inhibitory plasmas with C2 specificity contain nonclassical group BC or C antibodies. Human structural epitopes typically are larger in size than murine structural epitopes.^21,22  Antibodies with these larger footprints could theoretically block both binding to VWF and/or phospholipid and the activation of fVIII. Given the preincubation step in this protocol, it is possible that antibodies with large footprints, which overlapped both the preincubation mAb and the biotinylated mAb, did not bind to fVIII and therefore could not inhibit the binding of the biotinylated mAb. Thus, this method may underestimate the frequency of antibodies with nonclassical behavior.

Murine group BC antibodies inhibit the activation of fVIII and have high specific inhibitory activity in the Bethesda assay,¹⁸  in which the inhibitor titer is defined as the dilution of plasma that produces 50% inhibition of fVIII. However, they are type II inhibitors that do not completely inhibit fVIII. Patients with type II inhibitors often have spontaneous, severe bleeding despite having residual fVIII activities that typically are 10% to 30% at saturating inhibitor concentrations. In contrast, fVIII levels as low as 1% provide significant hemostatic benefit in noninhibitor patients. This indicates that the Bethesda assay underestimates the anticoagulant activity of type II inhibitors. The Bethesda assay is based on the one-stage clotting assay, in which the concentration of thrombin produced may be greater than that during in vivo hemostasis. High concentrations of thrombin may overcome the inhibition of fVIII activation by saturating levels of type II inhibitors. Further investigation is needed into the mechanism of action and the in vivo pathogenicity of nonclassical C2 antibodies.

This work was supported by grants from the National Institutes of Health (HL082609 and HL40921) and Hemophilia of Georgia, Inc (P.L.), National Hemophilia Foundation Clinical Fellowship (S.L.M.), and Hemophilia and Thrombosis Research Society Research Fellowship (S.L.M.)

National Institutes of Health

Contribution: S.L.M. and J.F.H. designed and performed research, analyzed data, and cowrote the paper; E.T.P. and R.T.B. designed and performed research; and P.L. designed research, analyzed data, and cowrote the paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Pete Lollar, Emory Children's Center, Room 426D, 2015 Uppergate Drive, Atlanta, GA 30322; e-mail: jlollar@emory.edu.