Search for a command to run...
Abstract The Protein C (PC) system maintains proper control of fibrin formation in hemostasis by downregulating the coagulation propagation phase, making its inhibition a candidate for “rebalancing” approaches in hemophilia treatment. It is also known that decreased PC-function in non-hemophilia individuals causes purpura fulminans, a severe thrombotic condition. In people with hemophilia, coagulation potential sometimes reaches non-hemophilic levels during on-demand therapeutic intervention for breakthrough bleeds. With the expectation of achieving both prophylaxis of bleeding and mitigation of the purpura fulminans risks, we developed a partial neutralizing antibody to human PC (termed PPC85N), and evaluated it in nonhuman primates. In in vitro enzymatic assays, PPC85N demonstrated incomplete inhibition of up to approximately 75% against human activated PC (APC) activity in the presence of Protein S, and up to approximately 60% against human PC activation by thrombin on endothelial cell sheets. Consistent with these results, PPC85N partially restored thrombin generation capacity in PC-activated human and primate plasma in vitro. For in vivo evaluation in a hemophilia A state, we used our established sub-acute bleeding model of primates with acquired hemophilia A, in which specific bleeding symptoms progressed over 3 days following artificial injury procedures. Previously, it was prospectively demonstrated that in vivo results of emicizumab against such bleeding symptoms roughly predicted its clinical efficacy. Results of a small-scale primate study showed that preventive dosing of PPC85N (3 mg/kg IV on Days 0 and 2, n = 2) tended to suppress the sub-acute bleeding symptoms, hemorrhagic anemia (indicated by % hemoglobin (Hb) levels relative to pre-values) and bruised area on body surface (cm²): 82.8% [mean of 2 animals] vs 42.5 ± 7.3% [mean ± SE] in the control group (n = 8), and 62.0 cm² vs 78.4 ± 20.6 cm², respectively. The degree of suppression was comparable to twice-daily dosing of recombinant porcine factor (F)VIII (rpoFVIII) at 1 or 3 U/kg IV (n = 3 for each) that would maintain its plasma levels equivalent to moderate or mild-to-moderate disease: 80.1 ± 6.3% or 92.9 ± 2.8%, and 65.6 ± 8.1 cm² or 28.5 ± 22.7 cm², respectively. Additionally, we induced joint bleeds in their right knees by intra-articular needling and evaluated joint swelling. While peak increase of knee joint diameter (% increase from pre-values) was 10.2 ± 3.5% in the control group, it was 2.6%, 3.0 ± 1.4% or 1.4 ± 0.7% in the PPC85N group, the rpoFVIII 1 U/kg group, or the 3 U/kg group, respectively, suggesting that PPC85N tended to suppress joint bleeds. To more rigorously examine these suppressive effects against sub-acute progressive bleeding symptoms observed in primates, we reproduced such bleeding symptoms in FVIII-deficient mice. The mice were crossbred with a nude strain to visualize bruising without damage from hair removal procedure. In this model, preventive dosing of anti-mouse PC antibody (3 or 10 mg/kg IV, n = 9 for each) significantly improved both Hb level and bruised area: 12.1 ± 0.7 g/dL* or 11.6 ± 1.2 g/dL vs 8.0 ± 1.6 g/dL in the vehicle group (n = 8), and 1.11 ± 0.48 cm²* or 0.73 ± 0.46 cm²* vs 5.44 ± 1.21 cm², respectively [*P < 0.05 by Dunnett's test]. These mouse results supported the effects of PPC85N observed in primates, although this anti-mouse PC antibody had a potential to fully neutralize APC activity, unlike PPC85N. Finally, we examined whether PPC85N would induce purpura fulminans-like symptoms in non-hemophilic primates. To minimize immune reactions caused by PPC85N itself, we prepared and used a chimeric antibody with human IgG1 in which reactivity to Fcγ receptors and C1q was silenced (termed chPPC85N). As a result, chPPC85N (1, 3, or 10 mg/kg IV, n = 3 for each) induced purpura fulminans-like symptoms (skin redness, irritation, wounds etc.) in more than half of the animals at each dosage, and also thrombotic pathological changes in several animals. In conclusion, we demonstrated the bleeding preventive effects of a partial neutralizing antibody to PC, PPC85N, in non-human primates with acquired hemophilia A. However, we also found in the non-hemophilic primate study that the purpura fulminans risk was not avoided even with this partial inhibition approach. Our in vivo research may highlight difficulties in balancing efficacy and safety for clinical application of anti-PC intervention therapy in hemophilia.