EPCR knockout: inflaming the discussion

In this issue of Blood, Magisetty et al¹  provide some surprising results from their studies of the role of endothelial protein C receptor (EPCR) in hemophilic arthropathy. There is a rationale for studying EPCR in hemophilia. It is well recognized that EPCR promotes the activation of protein C (PC) to activated PC (aPC) on endothelial surfaces.²  The aPC has antithrombotic effects by inactivating factor Va (FVa) on nearby endothelial surfaces. Like aPC, activated coagulation FVIIa can also bind to EPCR. Pavani et al³  showed that binding to EPCR enhances the hemostatic effect of FVIIa in a mouse hemophilia model, when assayed as ferric chloride–induced thrombosis. The Rao group⁴  showed that hemostatic levels of FVIIa displace PC from EPCR, and blockade of EPCR reduces the level of FVIIa needed for hemostasis in a saphenous vein bleeding assay. However, they reported that EPCR blockade by itself did not reduce the severity of bleeding in hemophilic mice. They concluded that downregulation of aPC generation is the major mechanism by which FVIIa interaction with EPCR contributes to the hemostatic effect of FVIIa in hemophilia.

Hemophilic arthropathy is associated with production of multiple inflammatory mediators, including tumor necrosis factor-α (TNF-α) and interleukin -6 (IL-6), with rapid local upregulation of IL-1β, TNF-α, and IL-6 as well as others. Blocking these inflammatory mediators has been considered a therapy for hemophilic arthropathy. A blocking antibody against IL-6 receptor in combination with factor replacement protects against hemophilic arthropathy in a mouse model.⁵  Inhibition of IL-1β, but not TNF-α, protects cartilage in a cell culture model of blood-induced damage.⁶  However, anti-TNF-α treatment provides significant protection against hemophilic synovitis in a mouse hemarthrosis model.⁷  In addition to its antithrombotic/anticoagulant effects, aPC promotes anti-inflammatory signaling by binding to EPCR. The Rao group⁸  showed that FVIIa also induces EPCR-mediated anti-inflammatory signaling, including promoting endothelial barrier function.

Since EPCR can contribute to both hemostatic and anti-inflammatory effects of FVIIa, it was reasonable to hypothesize that FVIIa interaction with EPCR would reduce the severity of hemophilic arthropathy by inducing an anti-inflammatory effect. Surprisingly, mice with both FVIII and EPCR knocked out had reduced bleeding severity compared with FVIII knockout (KO) mice with normal or even increased EPCR expression. The double KO mice needed only 1 dose of FVIIa to prevent development of arthropathy after joint bleeding, whereas the FVIII KO mice with normal EPCR needed 3 doses of FVIIa. Administration of an EPCR blocking antibody also reduced the severity of arthropathy in FVIII KO mice.

These results lead us to reexamine some assumptions about the development of arthropathy in hemophilia. For example, what are the relative contributions of bleeding and inflammation to hemophilic arthropathy? Certainly joint bleeding in non–hemophilic individuals does not generally lead to significant joint damage, whereas even minor bleeding in hemophilic mouse models results in progressive joint damage. Why is this so? Is the progressive joint damage with even “micro” bleeds in hemophilia due to intense/prolonged inflammation, poor clearance of blood from the joint, or repeated bleeding episodes? If inflammation is a key player, then enhanced anti-inflammatory effects through EPCR should improve outcome. However, if rebleeding is a driver for poor outcomes, it suggests rigorous maintenance of hemostasis following a bleed might be sufficient to prevent subsequent joint damage. There is some evidence that even a minor bleed leads to a cycle of rebleeding, inflammation, and angiogenesis in the setting of hemophilia. A pattern of long-term rebleeding has been observed following soft tissue wounding in a hemophilic mouse model.⁹  Recurrent bleeding has also been observed following joint injury in a mouse hemophilia model.¹⁰ 

In summary, the published data suggest that the short-term hemostatic effect of FVIIa is enhanced by binding to EPCR. However, Magisetty et al show that EPCR KO reduces rebleeding, inflammation, and angiogenesis following joint injury and thereby reduces the severity of the resulting arthropathy. At this point, it is not clear that EPCR-mediated anti-inflammatory effects have any significant ameliorating effect on hemophilic arthropathy. The results reported in the current paper are most consistent with the hypothesis that the procoagulant effects of EPCR KO are responsible for both the reduced bleeding initially and the reduced severity of arthropathy in the longer term. These results give support to the idea that an extended period of hemostasis is required to prevent long-term damage after a joint bleed.