The Impact of Anti-Factor VIII A1 and A3 Domain Antibodies in the Immune Reponse to Factor VIII in Patients with Hemophilia a

Introduction: Up to 30% of treated patients with severe hemophilia A will develop inhibitory antibodies to factor VIII (fVIII inhibitors). These patients present primarily with a polyclonal antibody response targeted against the A2 and C2 domains of fVIII, although antibodies to the entire fVIII molecule have been reported, including the A1 and A3 domains. Although thought not to be a major contributor to the overall inhibitor titer as in a model system, they make up a combined 25% of antibodies with the majority being non-inhibitory. In this study, we compared the inhibitory titers using human fVIII or ET3i (Expression Therapeutics), a hybrid B domain deleted protein in which the A1 and A3 domains have been replaced by porcine A1 and A3. We hypothesize that removal of these binding epitopes within the A1 and A3 domain will reduce anti-fVIII Bethesda titers in a subset of high titer patients and provide a deeper understanding of the role of these domains in the overall immune response. Previously published data has shown ET3i to have enhanced biosynthetic properties as well as provide full therapeutic correction in a murine model of hemophilia A. Using patient samples in the Emory IRB-approved congenital hemophilia inhibitor bank we evaluated the potential of ET3i to reduce the inhibitory titer compared to full length human fVIII (hfVIII).

Methods: Samples in the inhibitor bank with an inhibitory titer greater than or equal to 10 BU/ml were tested with a single sample chosen for any given patient with a total of 21 samples meeting these criteria. A modified Bethesda assay previously described by our lab was used to determine anti-fVIII inhibitory antibody titers with 1 unit/mL hfVIII or ET3i in fVIII deficient plasma as the antigen.

Results: Of the 21 patients tested Bethesda titers ranged from 12.1 to 3337 BU/ml for hfVIII and 4.92 to 2431 BU/ml for ET3i. Inhibitor plasma exposed to ET3i had a median titer of 28.6 BU/ml and a mean of 154 BU/ml, while the hfVIII had a median of 42 BU/ml and a mean of 209 BU/ml. ET3i was found to significantly decrease the inhibitory titer (p < 0.02, Wilcoxon test) as compared to hfVIII. Of patient plasma samples tested 62% showed a decrease of at least 20% in their inhibitory titer indicating that majority of patients have significant inhibitory anti-A1 and/or anti-A3 antibodies.

Conclusion: These results suggests that antibodies against the A1 and A3 domains of fVIII may contribute more significantly to a patient's inhibitory titer than what was previously recognized, and thus needs to be considered when designing a less antigenic and/or less immunogenic fVIII molecule.