Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Blood coagulation factor Xa (fXa) plays a critical role in the blood coagulation cascade, serving as the juncture between the intrinsic and extrinsic system (Davie et al., 1991). fXa binds to phospholipid in membranes, primarily phosphatidyl serine and along with fVa and calcium ions forms the prothrombinase complex, which is responsible for the generation of thrombin from prothrombin. Thrombin plays key roles in both coagulation and platelet activation. Thrombin cleaves fibrinopeptides from fibrinogen allowing precipitation of insoluble fibrin, participates in positive feedback reactions by activating fV and fVIII, and activates fXIII, which stabilizes clots. Furthermore, thrombin binds to and proteolytically activates a surface receptor on platelets and is the most potent activator of platelets known. Thus, inhibitors of fXa are expected to produce anticoagulant and antithrombotic effects by decreasing the conversion of prothrombin to proteolytically active thrombin, thereby diminishing thrombin-mediated activation of both coagulation and platelets. Experimental evidence suggests that agents that act by this mechanism have antithrombotic efficacy without an increase in bleeding risk in animals when compared with other antithrombotic agents, such as heparin and direct thrombin inhibitors (for review see Hauptmann and Stürzebecher, 1999; Leadley, 2001).

Recently, we reported a novel series of potent and orally bioavailable pyrazole fXa inhibitors, exemplified by DPC423 (Pinto et al., 2001). DPC423, the hydrochloride salt of 1-[3-(aminomethyl)phenyl]-N-[3-fluoro-2'-(methylsulfonyl)[1,1'-biphenyl]-4-yl]-3-(trifluoromethyl)-1H-pyrazole-5-carboxamide (Fig. 1), is a potent, direct inhibitor of human fXa with high selectivity [K_i (nanomolar): fXa, 0.15; trypsin, 60; thrombin, 6,000; plasma kallikrein, 61; activated protein C, 1,800; fIXa, 2,200; fVIIa, >15,000; chymotrypsin, >17,000; urokinase, >19,000; plasmin, >35,000; tissue plasminogen activator, >45,000; complement factor I, 44,000 (IC₅₀)] (Pinto et al., 2001). DPC423 given to dogs produced a pharmacokinetic profile with an oral bioavailability of 57%, a plasma clearance of 0.24 l/kg/h, and an apparent terminal elimination half-life of 7.5 h (Pinto et al., 2001). Given intravenously, DPC423 is a potent antithrombotic agent in the rabbit model of arteriovenous shunt thrombosis (Pinto et al., 2001; Wong et al., 2002). Preliminary human data showed that DPC423 was well tolerated and orally bioavailable, with a plasma half-life ranged from 27 to 35 h, supporting once daily dosing in humans (Barrett et al., 2001). To our knowledge, DPC423 is the first reported orally bioavailable, small-molecule fXa inhibitor in humans.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Structural formula DPC423.

Since the efficacy of DPC423 for the prevention of arterial thrombosis and its bleeding risk have not been evaluated, the objective of this study was to compare effects of DPC423 with those of the currently used anticoagulants enoxaparin (a low-molecular-weight heparin; Noble and Spencer, 1998) and argatroban (a direct thrombin inhibitor; McKeage and Plosker, 2001) on arterial thrombosis and hemostasis in rabbit models of the electrically induced carotid artery thrombosis (ECAT; Wong et al., 2000a) and cuticle bleeding (Himber et al., 1997), respectively. We also examined the oral antithrombotic activity of DPC423 in the rabbit ECAT model. Since aspirin is widely used in patients for the prevention and treatment of acute and chronic artery diseases (for review, see Awtry and Loscalzo, 2000), it is likely that DPC423 would be combined with aspirin in the clinical setting. Therefore, we also evaluated the antithrombotic effect of the combination of aspirin and DPC423.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

All experiments were conducted in accordance with the regulations of the Animal Care and Use Committee of the Bristol-Myers Squibb Company.

Reagents. The following drugs and chemicals were used in this study: chromogenic substrates S-2222 and S-2238 (Chromogenix AB products distributed by DiaPharma Group, Inc., West Chester, OH), human -thrombin and fXa (Enzyme Research Laboratories, Inc., South Bend, IN), human -thrombin (ICN Biomedicals, Inc., Costa Mesa, CA), activated partial thromboplastin time (APTT) reagent, prothrombin time (PT) reagent (thromboplastin with calcium), and ADP (Sigma-Aldrich, St. Louis, MO), and argatroban (SmithKline Beecham Pharmaceuticals, Philadelphia, PA). DPC423 was synthesized at Bristol-Myers Squibb Company.

Antithrombotic Studies. The rabbit ECAT model, described by Wong et al. (2000a), was used in this study. Briefly, male, New Zealand, white rabbits were anesthetized with ketamine (50 mg/kg + 50 mg/kg/h i.m.) and xylazine (10 mg/kg + 10 mg/kg/h i.m.). These anesthetics were supplemented as needed. An electromagnetic flow probe was placed on a segment of an isolated carotid artery to monitor blood flow. Thrombus formation was induced by electrical stimulation of the carotid artery for 3 min at 4 mA using an external stainless steel bipolar electrode. Carotid blood flow was measured continuously over a 90-min period to monitor thrombus-induced occlusion. Enoxaparin, argatroban, DPC423, or saline vehicle (6 ml/kg/h) was infused intravenously 1 h before the electrical stimulation of the carotid artery and continuously during the 90-min period. Doses of DPC423 were expressed as its free base equivalent.

Bleeding Times Studies. The rabbit cuticle bleeding time model, described previously by Himber et al. (1997), was used in this study with some modifications. Briefly, rabbits were anesthetized as described above, and their hind paws were shaved. A standard cut was made at the apex of the cuticle with a razor blade. Blood was allowed to flow freely by keeping the bleeding site in contact with 37°C warm Lactated Ringer's solution. Bleeding time was defined as the time after transection when bleeding was ceased. It was measured by averaging the bleeding time of three nail cuticles in the control period and at 60 min of the treatment period. Compound or vehicle was infused i.v. 1 h before the cuticle bleeding and continuously during the bleeding time measurement period.

Coagulation Assays. Arterial blood samples for the determination of ex vivo APTT, PT, thrombin time (TT), anti-fXa, and antithrombin activity were taken from the femoral arterial catheter and collected in tubes containing one-tenth the volume of 0.109 M sodium citrate before and at the end of the test.

Ex Vivo Platelet Aggregation. In some experiments, arterial blood samples were collected before and after DPC423 at 2.5 mg/kg/h i.v. for the determination of ex vivo platelet aggregation. Platelet aggregation was measured with a platelet aggregometer (Model PAP-4D; BioData, Horsham, PA). Two hundred microliters of platelet rich plasma was incubated for 3 min at 37°C. Percentages of platelet aggregation (percentage of light transmission) were determined 4 min after the addition of 20 µl of the agonist (ADP at 10 µM, -thrombin at 35 nM, final concentration).

Statistical Analysis. Statistical analyses used were correlation, linear regression, analysis of variance, and Duncan's new multiple-range test (Cody and Smith, 1991). A value of P < 0.05 was considered statistically significant. All data are means ± S.E.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Antithrombotic Effects of DPC423, Enoxaparin, and Argatroban in Rabbits. Figure 2 shows effects of vehicle and DPC423 on carotid blood flow after electrical stimulation. Control carotid blood flow in these animals averaged 21 ml/min. After electrical stimulation, blood flow was gradually decreased, and the artery was totally occluded in about 30 min in vehicle-treated animals. DPC423 at 0.08 to 2.5 mg/kg/h i.v. caused a dose-dependent increase in duration of the patency of the artery. At 0.82 and 2.5 mg/kg/h i.v., there was no occlusion in all the animals up to 90 min. Figure 2 shows that enoxaparin at 0.03 to 3 mg/kg/h i.v. and argatroban at 0.03 to 1 mg/kg/h i.v. caused similar dose-dependent increases in duration of the patency of the injured artery.

View larger version (40K): [in this window] [in a new window]   Fig. 2.   Effects of vehicle, DPC423 at 0.08 to 2.5 mg/kg/h i.v., enoxaparin at 0.03 to 3 mg/kg/h i.v., and argatroban at 0.03 to 1 mg/kg/h i.v. on carotid blood flow (expressed as a percentage of control carotid blood flow) after thrombus induction in ECAT rabbits. Top panel, n = 12/group; middle panel, n = 5 to 6/group; bottom panel, n = 6/group. Means ± S.E.

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Dose-response curves of enoxaparin (0.03 to 3 mg/kg/h i.v.; n = 5-6/dose), DPC423 (0.08 to 2.5 mg/kg/h i.v.; n = 12/dose), and argatroban (0.01 to 1 mg/kg/h i.v.; n = 6/dose) in ECAT rabbits. V denotes vehicle-treated group. Means ± S.E. *, P < 0.05 compared with vehicle.

Effects of DPC423, Enoxaparin, and Argatroban on Coagulation Parameters in Rabbits. Figure 4 shows ex vivo effects of DPC423, enoxaparin, and argatroban on APTT, TT, and PT. DPC423 slightly elevated APTT and PT at higher doses and did not change TT. Enoxaparin caused a greater increase in TT than APTT but did not change PT. Argatroban also produced a greater increase in TT than APTT and had a moderate effect on PT.

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Ex vivo APTT, PT, and TT effects of DPC423, enoxaparin, and argatroban in ECAT rabbits from Fig. 3. The ratio is the end of the experiment value over the control value. *, P < 0.05 compared with the vehicle. Means ± S.E.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Top panel, ex vivo anti-fXa and antithrombin effects of DPC423 in ECAT rabbits from Fig. 3. *, P < 0.05 compared with the vehicle. Means ± S.E. Bottom panel, relationship between the effect of DPC423 on carotid blood flow (as a percentage of control) and its ex vivo anti-fXa activity in ECAT rabbits (n = 6/group). Means ± S.E.

Cuticle Bleeding Time Effects of DPC423, Enoxaparin, and Argatroban in Rabbits. The cuticle bleeding time effects of DPC423, enoxaparin, argatroban, and heparin are shown in Fig. 6. The control cuticle bleeding time averaged 160 s. Values of cuticle bleeding time (percent change over control) determined at the maximum antithrombotic dose were 88 ± 12 for argatroban, 4 ± 3 for enoxaparin, and 5 ± 4 for DPC423 compared with 3 ± 2 for the vehicle (n = 5-6/group). Although heparin was a poor antithrombotic agent in ECAT rabbits and at 100 U/kg/h i.v. produced only a moderate antithrombotic effect (Wong et al., 2000a), it still significantly increased the cuticle bleeding time by 69 ± 13% (Fig. 6).

View larger version (20K): [in this window] [in a new window]   Fig. 6.   Cuticle bleeding time effects (expressed as the percent change over control) of vehicle, DPC423, enoxaparin, argatroban, and heparin in rabbits. The percent change is based on the difference in bleeding time in the treatment period compared with that in the control period. *, P < 0.05 compared with vehicle.

Ex Vivo Effects of DPC423 on Platelet Aggregation in Rabbits. At 2.5 mg/kg/h i.v., DPC423 did not change the ex vivo platelet aggregation induced by either ADP or -thrombin (ADP, 31 ± 1% for the control and 30 ± 5% for DPC423; or -thrombin, 41 ± 5% for the control and 41 ± 8% for DPC423; n = 4/group).

Oral Antithrombotic Effects of DPC423 in Rabbits. Figure 7 shows effects of vehicle and DPC423 given orally on carotid blood flow after electrical stimulation. Control carotid blood flow before electrical stimulation was 19 ± 1, 20 ± 2, 20 ± 2, and 19 ± 2 ml/min for the vehicle and DPC423 at 1, 3, and 10 mg/kg p.o., respectively, in these animals (n = 6/group). Values of carotid blood flow (expressed as a percentage of control) at 45 min after electrical stimulation were 1 ± 1, 10 ± 4, 24 ± 6, and 74 ± 7 for the vehicle and DPC423 at 1, 3, and 10 mg/kg p.o., respectively. Compared with the vehicle, DPC423 at 3 and 10 mg/kg p.o. significantly increased carotid blood flow at 45 min after electrical stimulation (P < 0.05).

View larger version (37K): [in this window] [in a new window]   Fig. 7.   Oral effects of vehicle and DPC423 at 1 to 10 mg/kg on carotid blood flow (expressed as a percentage of control carotid blood flow) after thrombus induction in ECAT rabbits (n = 6/group). Means ± S.E. , vehicle; , DPC423 (1 mg/kg p.o.); , DPC423 (3 mg/kg p.o.); , DPC423 (10 mg/kg p.o.).

Antithrombotic Effect of a Combination of Aspirin and DPC423 in Rabbits. The antithrombotic effects of aspirin (1 mg/kg/h i.v.), DPC423 (0.08 and 0.25 mg/kg/h i.v.), and their combination are shown in Fig. 8. When ineffective doses of aspirin (1 mg/kg/h) and DPC423 (0.08 mg/kg/h) were combined, carotid blood flow was significantly enhanced (Fig. 8, upper panel). A similar enhancement of carotid blood flow was also seen after coadministration of an ineffective dose of aspirin (1 mg/kg/h) and a moderately effective dose of DPC423 (0.25 mg/kg/h) (Fig. 8, upper panel). Carotid blood flow (as a percentage of control) after vehicle and aspirin at 1 mg/kg/h was 21 ± 5 and 18 ± 2%, respectively (Fig. 8, lower panel). Carotid blood flow (as a percentage of control) after DPC423 at 0.08 and 0.25 mg/kg/h i.v. was significantly increased by aspirin at 1 mg/kg/h i.v. from 16 ± 2 and 37 ± 6% to 61 ± 9 and 88 ± 9%, respectively (n = 6/group and P < 0.05; Fig. 8, lower panel).

View larger version (34K): [in this window] [in a new window]   Fig. 8.   Top panel, effects of vehicle (n = 6), aspirin at 1 mg/kg/h i.v. (n = 6), DPC423 at 0.08 mg/kg/h i.v. (n = 5), DPC423 at 0.25 mg/kg/h i.v. (n = 6), DPC423 at 0.08 mg/kg/h i.v. + aspirin at 1 mg/kg/h i.v. (n = 6), and DPC423 at 0.25 mg/kg/h i.v. + aspirin at 1 mg/kg/h i.v. (n = 6) on carotid blood flow (expressed as a percentage of control carotid blood flow) after thrombus induction in ECAT rabbits. Means ± S.E. Bottom panel, dose-dependent effects of DPC423 on carotid blood flow (as a percentage of control) in the vehicle- and aspirin-treated ECAT rabbits. Means ± S.E. , P < 0.05 compared with the corresponding dose of DPC423 in the vehicle-treated group.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we evaluated antithrombotic and bleeding time effects of DPC423, enoxaparin, and argatroban in rabbits. We also examined whether the addition of aspirin might influence antithrombotic and bleeding time effects of DPC423 in rabbits. We selected the rabbit as our animal model because DPC423 has similar potency in inhibiting human and rabbit fXa (K_i: 0.15 nM in humans, 0.3 nM in rabbit; Wong et al., 2002). In contrast, rat and dog fXa are much less sensitive to DPC423 (K_i: 2.35 nM in rat, 1.2 nM in dog; Wong et al., 2002). Other investigators have also noted that rat and dog fXa are much less sensitive than human and rabbit fXa to small-molecule fXa inhibitors (Hara et al., 1995; Taniuchi et al., 1998; Abendschein et al., 2000; McClanahan et al., 2001). We showed that DPC423 given either i.v. or orally was an effective antithrombotic agent in ECAT rabbits. Furthermore, in contrast to argatroban but similar to enoxaparin, DPC423 at maximum antithrombotic dose did not increase bleeding time in rabbits. In addition, the combination of aspirin and DPC423 at ineffective antithrombotic doses produced a significant antithrombotic effect.

The ECAT model has been used to evaluate antiplatelet agents (Schumacher et al., 1993; Herbert et al., 1998) and anticoagulants (Herbert et al., 1996b; Kawasaki et al., 1998; Wong et al., 2000a) in rats and rabbits. Thrombus formed in this model consists mainly of platelets and fibrin (Schumacher et al., 1993; Kawasaki et al., 1998; Wong et al., 2000a) and thus mimics clinical arterial thrombosis. Although the electrolytic injury of an artery to induce thrombosis is unrelated to clinical thrombosis, thrombus morphology and antithrombotic efficacy of antithrombotic agents suggest that thrombus growth in the ECAT model may be clinically relevant even though the mechanism of thrombus initiation is not (Schumacher et al., 1993). For instance, we noted in this study that argatroban prevented arterial thrombosis in ECAT rabbits at doses similar to its clinical antithrombotic doses (McKeage and Plosker, 2001), which supports some clinical relevance of the rabbit ECAT model.

We demonstrated that DPC423 (fXa K_i: 0.15 nM in humans, 0.3 nM in rabbit) was as effective as enoxaparin and argatroban in preventing arterial thrombosis in the ECAT rabbits with an ED₅₀ value of 0.6 mg/kg/h (0.77 µmol/kg/h). Previously, we also showed that SK549 (fXa K_i: 0.52 nM in humans, 0.26 nM in rabbit) has an ED₅₀ value of 0.02 mg/kg/h (0.035 µmol/kg/h) in the ECAT rabbit model (Quan and Wexler, 2001; Wong et al., 2000a). Interestingly, even though DPC423 was as potent as SK549 in terms of rabbit fXa K_i, it was 22-fold less potent than SK549 as an antithrombotic agent in ECAT rabbits. Since it is the plasma concentration of the fXa inhibitor rather than the dose that determines the anticoagulation, we, therefore, determined the effective plasma concentration of DPC423 in ECAT rabbits. Our study showed that the EC₅₀ value for DPC423 was 137 nM. As the EC₅₀ for SK549 was 97 nM in the ECAT rabbits (Wong et al., 2000a), these data indicate that the relative antithrombotic potency of DPC423 and SK549 in terms of EC₅₀ was quite similar to their relative in vitro inhibitory constants for fXa.

Since fXa is at the convergent point of the intrinsic and extrinsic pathways of coagulation (Davie et al., 1991), inhibition of fXa by DPC423, as expected, prolonged APTT, the intrinsic coagulation pathway-dependent clotting time, and PT, the extrinsic pathway-dependent clotting time. We also showed that antithrombotic efficacy of DPC423 could be achieved with less than a 2-fold increase in APTT and PT. Other studies have also showed moderate increases in APTT and PT with direct fXa inhibitors at antithrombotic doses in animals (for review, see Hauptmann and Stürzebecher, 1999; Leadley et al., 2001). These results suggest that antithrombotic efficacy of direct fXa inhibitors could be achieved at doses that produced moderate increases in systemic anticoagulation. As pointed out by Leadley et al. (2000), the sensitivity of these clotting tests for a given fXa inhibitor seems to vary among compounds from different chemical series, which makes it difficult to compare compounds and to evaluate effective plasma concentrations of fXa inhibitors with these assays. It appears that anti-fXa activity is a good method to monitor antithrombotic effect of DPC423 since the antithrombotic activity of DPC423 in ECAT rabbits was significantly correlated with its ex vivo anti-fXa activity. Dyke at al. (2002) also reported that anti-fXa activity is a better method than PT and APTT to monitor the plasma concentration of the direct fXa inhibitor DX-9065a in patients.

DPC423 inhibited ex vivo fXa activity and did not change TT and ex vivo thrombin activity, supporting that the antithrombotic effect of DPC423 is consistent with fXa inhibition but not related to thrombin inhibition. It should be noted that the antithrombotic effect of DPC423 is not likely due to the direct inhibition of platelet aggregation since DPC423 at maximal antithrombotic dose did not affect ex vivo platelet aggregation responses to ADP and thrombin. This does not, however, rule out a role of decreased platelet activation in vivo in the antithrombotic effect of DPC423 and other fXa inhibitors since inhibition of thrombin generation may lead to reduced platelet activation.

This study compared the bleeding potential of DPC423, enoxaparin, argatroban, and heparin in rabbits. The bleeding model chosen for this study was the cuticle bleeding time model, which was first established by Giles et al., (1982) to monitor hemostatic function in hemophilic dogs. Other investigators have adapted this model in rabbits to monitor the bleeding potential of inhibitors of fXa, fIXa, tissue factor, and thrombin (Hollenbach et al., 1994; Himber et al., 1997; Sinha et al., 2000; Refino et al., 2002). We showed that argatroban and heparin increased the cuticle bleeding time significantly in rabbits. In contrast, DPC423 and enoxaparin could achieve antithrombotic efficacy without significant effect on the cuticle bleeding time in rabbits. Sinha et al. (2000) also observed that enoxaparin at a dose similar to ours, increasing APTT by 2-fold, did not increase cuticle bleeding in rabbits. Other investigators have also noted that direct fXa inhibitors, but not direct thrombin inhibitors and heparin, at antithrombotic doses have little or no effect on the bleeding time in rats, rabbits, and dogs (Herbert et al., 1996a; Himber et al., 1997; Morishima et al., 1997; Sato et al., 1997; Abendschein et al., 2000; Sinha et al., 2000; McClanahan et al., 2001). It is not clear why fXa inhibitors prevent thrombosis without increasing the bleeding time in animals. A possible explanation suggested by some investigators is that reversible and competitive fXa inhibitors might not completely suppress the production of thrombin and might result in the generation of small amounts of thrombin (Morishima et al., 1997; Sato et al., 1997; Leadley, 2001). Because thrombin has a 10,000-fold higher affinity for platelet than for fibrinogen, minimum amounts of thrombin might be sufficient to activate platelets to induce normal hemostasis. It remains to be seen, however, whether the antithrombotic effects of direct fXa inhibitors can be achieved without bleeding complications in humans.

Since many patients with coronary artery diseases are taking aspirin for secondary prevention of stroke and myocardial infarction (Awtry and Loscalzo, 2000), it is likely that DPC423 would be combined with aspirin in the clinical setting. Therefore, we examined the antithrombotic effects of aspirin, DPC423, and their combination in ECAT rabbits. We observed that the addition of the ineffective-dose DPC423 to the ineffective-doses aspirin produced a very significant antithrombotic effect but did not increase the bleeding time. The reason for this enhanced antithrombotic effect is not known. It is well known that aspirin inhibits platelets by inhibiting cyclooxygenase (for review, see Awtry and Loscalzo, 2000). In addition, it has been demonstrated that aspirin depresses thrombin formation in healthy subjects (Szczeklik et al., 1992). Thus, it is possible that the combination of the antiplatelet and antithrombin activities of aspirin and the fXa inhibitory activity of DPC423 may result in an additive or better antithrombotic effect. Other investigators also reported that addition of an anticoagulant agent, such as a fIX antibody or a vitamin K antagonist, to aspirin produced an enhancement of their antithrombotic efficacy in models of arterial thrombosis (Bossavy et al., 1999; Feuerstein et al., 1999).

In summary, we showed that DPC423 was as effective as currently used anticoagulants, enoxaparin and argatroban, in preventing arterial thrombosis in rabbits. Enoxaparin and argatroban, however, are not orally active. In this regard, DPC423 is better than enoxaparin and argatroban. Unlike argatroban and heparin, DPC423 could achieve antithrombotic efficacy without an increase in bleeding in rabbits. The antithrombotic action of DPC423 may be related to fXa but not to thrombin inhibition. Coadministration of low doses of aspirin and DPC423 enhanced their antithrombotic efficacy but not their effect on bleeding time. Therefore, these results suggest that DPC423 may be a clinically useful oral anticoagulant for the prevention of arterial thrombosis.