Prophylactic factor VIII (FVIII) replacement therapy in hemophilia A requires frequent administration because of the short half-life of FVIII. Polyethylene glycol (PEG) conjugation is thought to extend FVIII half-life by decreasing hepatic clearance. BAY 94-9027 is a rationally designed B-domain–deleted (BDD) FVIII molecule with a 60-kDa PEG molecule attached to a specific amino acid (1804) to increase circulating half-life and reduce the exposure to epitopes reported to cause immunogenicity in the A3 domain while preserving full biological function. BAY 94-9027 is currently in clinical trials and has prolonged half-life and improved efficacy in animal models and humans.
To determine whether half-life extension with BAY 94-9027 is related to PEG steric hindrance, we first investigated whether PEG impacts BAY 94-9027 binding interactions. Direct binding of hybrid of kidney and B cells (HKB)11-derived FVIII, BAY 94-9027 or BDD-FVIII, was assessed by measuring the ability of a panel of immobilized monoclonal antibodies directed toward different FVIII domains to capture FVIII. Interactions with more physiologic partners were indirectly assessed by thrombin generation assay (TGA) and by an in vitro hepatocyte clearance assay.
Our results indicate that the presence of A3-directed PEG reduced BAY 94-9027 capture by immobilized antibodies directed toward FVIII regions at or near the site of conjugation. Capture by antibodies directed toward the A3 and C2 domains were most impacted, while those directed toward A1 and A2 still bound BAY 94-9027. The A3-specific C7F7 antibody showed ~50% lower capture of BAY 94-9027 vs BDD-FVIII at 20 ng/mL of FVIII. C7F7 capture of PEG-BDD-FVIII was further reduced when a di-PEG conjugate of BDD-FVIII was subjected to the same assay, again confirming that PEG sterically modulates PEG-BDD-FVIII reactivity to the antibody. To determine whether steric effects observed with PEG may impact FVIII function globally, TGA was performed with BAY 94-9027 spiked into FVIII-deficient plasma and subjected to 1 pM tissue factor initiation. By TGA, both BDD-FVIII and BAY 94- 9027 generated comparable peak thrombin levels, with EC50 values of 3.9 and 3.2 nM for BDD-FVIII and BAY 94-9027, respectively. These results indicate that the PEG did not disrupt activated PEG-BDD-FVIII interactions with its partners in the factor Xase enzyme complex, consistent with published PEG-BDD-FVIII efficacy. By hepatocyte clearance assay, PEG- BDD-FVIII clearance was reduced ~30-40% compared with BDD-FVIII, regardless of whether von Willebrand factor was present. This reduction in hepatocyte clearance is likely to contribute to the prolonged plasma half-life reported for BAY 94-9027.