We read with interest the recent article by Butenas et al,1 who conclude that the prohemostatic effect of high-dose factor VIIa in hemophilic blood or a synthetic plasma system is strictly dependent on tissue factor (TF). The authors note that their proposed mechanism differs from our previously published conclusion that high-dose factor VIIa can, in the absence of TF, generate factor Xa on the surface of activated platelets at concentrations sufficient to ameliorate hemophilia.2
Butenas et al propose that differences in the results from our 2 labs are due to differences in experimental design or protein preparations. We feel it is very likely that most of the differences in results between our 2 labs are due to our use of tissue factor–expressing cells as the initiating stimulus in our model system rather than tissue factor protein reconstituted into phospholipid vesicles. For example, Mann's group has previously reported that the presence of zymogen factor VII in a synthetic plasma model competes with factor VIIa for binding to TF on phospholipid vesicles.3 By contrast, we see no effect of zymogen factor VII on thrombin generation in a model system initiated by cell-associated TF.4
Butenas et al suggest that high-dose factor VIIa therapy works in hemophilia to directly enhance activation of factor X through the factor VIIa/TF complex, resulting in an enhanced activation of platelets. Our data indicates that, although the factor VIIa/TF complex is essential to initiate coagulation, once platelets have been activated, high-dose factor VIIa binds to the surface of activated platelets and replaces the absent factor IXa/VIIIa complex to generate factor Xa, which then boosts platelet-surface thrombin generation.2 Our view is consistent with the observation that the tissue factor pathway is intact in hemophilic patients5 and, indeed, is responsible for platelet activation, accounting for the tendency of hemophilics to initially stop bleeding as a normal platelet plug forms6 7 but then suffer severe delayed rebleeding.
These disparate views of the mechanism of action of factor VIIa have important implications for dosing. The TF-dependent effect described by Butenas et al is saturated at levels below those now used therapeutically. By contrast, the binding of factor VIIa to the activated platelet surface is nowhere close to being saturated at therapeutically relevant concentrations. Thus, our mechanism predicts that escalating the dose of factor VIIa should enhance its hemostatic effect, while the mechanism of Butenas et al suggests that escalating the dose would waste money without benefiting patients. Taken together, we believe the data supports a platelet-dependent, TF-independent mechanism for high-dose factor VIIa in hemophilia.
Mechanism of factor VIIa–dependent coagulation in hemophilia blood
There are 3 areas in which the experimental results and their interpretations are in discord between our 2 groups: (1) whether factor VII and factor VIIa compete for tissue factor, (2) whether tissue factor is required for the function of factor VIIa in the condition of hemophilia A or B, and (3) whether vigorous propagation of thrombin generation by factor VIIa can occur on activated platelets in the absence of tissue factor, factor VIII, and factor IX. We are unsure of the basis for the discrepancies between the results. But we can offer the following facts in order to rationalize some potential bases for the differences:
First, the competition of factor VII and factor VIIa for tissue factor (TF) can only be observed at low tissue factor concentrations relative to the factor VII and factor VIIa concentrations. As the concentration of tissue factor is increased, the feedback activation of factor VII by factor VIIa–TF, factor Xa–membrane, and thrombin obliterates this phenomenon.1(Fig2)
Second, the tissue factor requirement essential to trigger the reaction appears to be indisputable. The relative rate of factor VIIa to factor VIIa–TF in factor X activation is approximately 0.0 001-fold,1-2 and the relative efficiency of factor VIIIa–factor IXa compared to factor VIIa–TF in factor X activation is approximately 50-100 to1.1-3 Our experimental data in synthetic system and in whole blood1-4 are consistent with these experimental ratios determined by other laboratories. During the propagation phase of thrombin generation, the rate of factor X activation by factor VIIa without tissue factor in the hemophilia A or B situation would be depressed by approximately 1 000 000-fold relative to that observed in normal blood in the presence of tissue factor.
Third, the data of our recent publication1-4 indicate that factor VIIa at supraphysiologic concentrations has little effect on thrombin generation either in congenital hemophilia A or “acquired” hemophilia B blood. This observation is valid for experiments conducted in the absence of tissue factor and for those conducted at fixed TF concentrations. A prolonged clotting time in “acquired” hemophilia B blood (>40 min)4(Fig4A) and a lack of response to the anti-TF antibody in normal blood (see comments to Figure 24) indicate that the blood used in our experiments had no functional tissue factor. In contrast to the observation by Monroe et al,1-5 in our experiments thrombin generation in hemophilia blood is almost not affected by the initial rate of platelet activation. Data show4(Fig4C) that additions of 10 to 50 nM factor VIIa to “acquired” hemophilia B blood in the absence of tissue factor restores platelet activation to the levels observed in normal blood. But thrombin generation,4(Fig4B) however, is almost not affected by these factor VIIa additions. Similarly, while supraphysiologic concentrations of factor VIIa in congenital hemophilia A blood triggered with tissue factor lead to an increased rate of platelet activation,4(Fig5D) robust thrombin generation is not observed,4(Fig5A) even when platelets are completely activated. These observations would seem to contradict the notion that factor VIIa activity on platelets can provide for the rapid activation of factor X in the absence of the factor VIIIa–factor IXa complex and tissue factor.
The experimental systems used by the Roberts group and our group differ in the source of tissue factor that is used to trigger the reaction. It is conceivable that the tissue factor source or the environment in which tissue factor is presented could be the cause of some of the differences in our results.
