Factor VIII Structure, Not Activity, Is the Primary Determinant of Immunogenicity in Hemophilia A and Combined Hemophilia A/Von Willebrand Disease Mice

Abstract 1187

A major complication in the treatment of severe hemophilia A is the development of anti-fVIII antibodies in approximately 30% of patients. In these patients and in animal models injection of fVIII by an intravenous route, which traditionally is considered a tolerogenic route of protein delivery, not only fails to induce tolerance but induces a brisk T- and B- cell immune response. The role of fVIII structure and function in the immunogenicity of fVIII remains unclear. In this study we tested four interrelated hypotheses: (i) FVIII is immunogenic due to its role in promoting production of thrombin, leading to downstream events that provide an immunogenic milieu. (ii) FVIII is immunogenic because it is exposed to the immune system in the context of active inflammation (i.e. at the site of a clot). (iii) Structural determinants intrinsic to the fVIII molecule are immunogenic. (iv) FVIII is protected from the immune system until it is released from its large carrier protein von Willebrand factor (VWF).

To address these hypotheses we constructed wild-type B domain deleted fVIII (wt fVIII) and 2 structurally intact inactive fVIII molecules, R372A/R1689A fVIII and an A2 domain point mutant, V634M fVIII. R372A/R1689A fVIII is inactive due to substitutions at thrombin and factor X proteolytic activation sites. It is not released from VWF, and thus may not be present at the site of a clot. V634M fVIII undergoes normal thrombin cleavage but has specific procoagulant activity that is less than 1% of wt fVIII.

The immunogenicity of the fVIII molecules was compared in 3 protocols. In a low dose protocol, fVIII deficient mice were injected with 6 weekly tail vein injections of 0.2 μg followed by 2 injections of 0.5 μg wt fVIII or R372A/R1689A fVIII. In a varying dose protocol, the immunogenicity of wt fVIII, R372A/R1689A fVIII, and V634M fVIII was determined in fVIII deficient mice following 4 weekly tail vein injections of 0.5 μg, 1.0 μg, 1.5 μg, or 2.0 μg fVIII per dose followed by 1 boost at twice the dose. Finally, the immunogenicity of wt fVIII, R372A/R1689A fVIII, and V634M fVIII was compared in fVIII/VWF deficient mice following 6 weekly injections at 0.6 μg followed by 2 boost injections at 1.5 μg.

In the low dose protocol 68% of fVIII deficient mice injected with wt fVIII had positive ELISA titers with a median titer of 400 compared with 40% of those injected with R372A/R1689A fVIII with a median titer of 0. Mice injected with wt fVIII had a median inhibitor titer of 10 BU/ml compared with a median titer of 0. Although R372A/R1689A fVIII was statistically less immunogenic with lower ELISA and inhibitor titers (p=0.027 and 0.018, respectively, Mann-Whitney test) this may not be clinically relevant as 40% of the mice mounted an immune response.

In the varying dose protocol, there was no difference in median ELISA fVIII inhibitor titers at any dosing level. At the 2.0 μg dose all mice except for 1 in the V634M fVIII cohort mounted an immune response. The median ELISA titers at 2.0 μg were1760 for wt fVIII, 447 for R372A/R1689A fVIII, and 1480 for V634M fVIII. The median inhibitor titers at the 2.0 μg dose were 310 BU/ml for wt fVIII, 103 BU/ml for R372A/R1689A fVIII, and 288 BU/ml for V634M fVIII. There was no significant difference between wt fVIII, R372A/R1689A fVIII and V634M fVIII in either ELISA or inhibitor titers (p=0.2 and p=0.35, respectively, Kruskal-Wallis test).

In the fVIII/VWF deficient mouse protocol, 85% of mice had positive ELISA titers in the wt fVIII cohort compared with 79% for R372A/R1689A fVIII and 85% for V634M fVIII. The median ELISA titers were similar for each group at 354 for wt fVIII, 179 for R372A/R1689A fVIII, and 363 for V634M fVIII. Inhibitor titers were similar for each group with a median inhibitor titer of 107 BU/ml for wt fVIII, 46 BU/ml for R372A/R1689A fVIII, and 198 BU/ml for V634M fVIII. There was no significant difference between wt fVIII, R372A/R1689A fVIII and V634M fVIII in either ELISA or inhibitor titers (p=0.46 and p=0.32, respectively, Kruskal-Wallis test).

In conclusion, there was no significant difference in the immunogenicity of wt fVIII and V634M fVIII in fVIII deficient mice. R372A/R1689A fVIII was slightly less immunogenic in fVIII deficient mice in 1 of 2 protocols tested. In the absence of VWF, wt fVIII, R372A/R1689A fVIII, and V634M fVIII were equally immunogenic. This suggests that the immunogenicity of fVIII is intrinsic to fVIII structure and not its cofactor activity, while VWF may have a small protective effect.