Journal Logo

Original Article

Platelet Aggregometry and Receptor Binding to Predict the Magnitude of Antithrombotic and Bleeding Time Effects of Clopidogrel in Rabbits

Wong, Pancras C PhD; Crain, Earl J BS; Watson, Carol A BS; Jiang, Xiaosui PhD; Hua, Ji MS; Bostwick, Jeffrey S MS; Ogletree, Martin L PhD; Schumacher, William A PhD; Rehfuss, Robert PhD

Author Information
Journal of Cardiovascular Pharmacology: May 2007 - Volume 49 - Issue 5 - p 316-324
doi: 10.1097/FJC.0b013e31803e8772
  • Free

Abstract

INTRODUCTION

Clopidogrel (Plavix) is a thienopyridine-based oral antiplatelet prodrug. After oral administration, clopidogrel is converted in the liver to an active metabolite, which binds selectively and irreversibly to the P2Y12 adenosine diphosphate (ADP) receptor on the platelet surface. By virtue of its ability to inhibit ADP binding, receptor activation and subsequent platelet aggregation, clopidogrel has been shown to be effective in reducing the incidence of stroke, myocardial infarction, or cardiovascular death in high-risk patient populations.1

Although newer platelet function tests have been introduced since the discovery of light transmission aggregometry by Born2 in 1962, ex vivo inhibition of platelet aggregation (IPA) induced by ADP is still the current standard assay by which the antiplatelet effects of clopidogrel and other P2Y12 receptor antagonists are measured.1,3-7 It is commonly assumed that ex vivo IPA directly reflects the in vivo effectiveness and thus, the adequacy of the antithrombotic activity of clopidogrel. However, there is still much debate on the target IPA level for clopidogrel to produce a clinically relevant antithrombotic effect or reveal a bleeding liability. This study was designed to address this critical issue in rabbits. We first characterized the antithrombotic and bleeding time (BT) effects of clopidogrel in rabbits and compared these effects with aspirin. We then systematically examined the association between ex vivo IPA effects of clopidogrel and its in vivo effects on arterial thrombosis and BT. [33P]-2MeS-ADP binding to platelet P2Y12 receptors was also assessed in rabbits and compared with IPA for its usefulness to monitor the in vivo effects of clopidogrel.

MATERIALS AND METHODS

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

Compounds

Clopidogrel was obtained from Sanofi-aventis Company. MRS-25008 was synthesized at Bristol-Myers Squibb Company. Aspirin, apyrase, 2MeS-ADP, and MRS-2179 were obtained from Sigma Chemical Co. (St. Louis, MO). [33P]-2MeS-ADP was obtained from Perkin-Elmer (Wellesley, MA). ADP and arachidonic acid were obtained from Chrono-Log Corp. (Havertown, PA).

Antithrombotic Studies

The electrical current-induced arterial thrombosis (ECAT) model, first described by Romson et al9 and modified by Wong,10 was used in this study. Briefly, male New Zealand White rabbits were anesthetized with ketamine (50 mg/kg + 50 mg/kg/h IM) and xylazine (10 mg/kg + 10 mg/kg/h IM). 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. Integrated blood flow over 90 min was calculated by the trapezoidal rule, which measures the area under the carotid blood flow-time curve, and expressed as a percentage of control. In addition, thrombus formed in the injured vessel was weighed. It should be noted that basal blood flow of the sham-injured carotid artery (ie, placing an electrode on the artery without application of current) was not significantly altered during the experimental period in pilot studies.

Nonfasted rabbits were dosed orally once daily with vehicle (0.6% methocel, 2 mL/kg) and clopidogrel (0.3 to 30 mg/kg) for 3 days. Antithrombotic evaluation was performed 2 to 3 h after the last dose. Compounds were suspended evenly in solution by repeated stirring (30 min) and sonicating (5 min). Treatment groups consisted of vehicle (n = 10) and clopidogrel at 0.3 (n = 6), 1 (n = 6), 3 (n = 6), 10 (n = 6), and 30 mg/kg (n = 6).

In a separate study, aspirin or saline-vehicle was given as continuous IV infusion at 6 mL/kg/h, starting 1 h before the antithrombotic evaluation and continuing for 90 min after the electrical stimulation of the carotid artery. Treatment groups consisted of saline-vehicle and aspirin at 0.3, 1, and 3 mg/kg/h (n = 5 to 6 per group).

Bleeding Time Studies

The rabbit cuticle BT model, described previously by Wong et al,10 was used in this study. 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 lactated Ringer solution. BT was defined as the time after transection when bleeding ceased and was measured by averaging the BT values of 3 nail cuticles. The maximum BT recorded was 20 min. Off-scale bleeding was defined as bleeding that was still observable at 20 min after cuticle transection. For the purpose of statistical analysis, a BT value of 20 min was used for the off-scale bleeding. Compound or vehicle was given as described above. Cuticle bleeding was measured 2 to 3 h after the last dose. Treatment groups consisted of vehicle (n = 10) and clopidogrel at 0.3 (n = 6), 1 (n = 6), 3 (n = 6), 10 (n = 6), and 30 mg/kg (n = 5).

In a separate study, aspirin or saline-vehicle was given as continuous IV infusion at 6 mL/kg/h, starting 1 h before the cuticle transection and continuing for 90 min after the cuticle transection. Treatment groups included saline-vehicle and aspirin at 0.3, 1, and 3 mg/kg/h (n = 5 to 6 per group).

Platelet Aggregation and Serum Thromboxane B2

In the ex vivo studies, rabbit platelet-rich plasma (PRP) samples were obtained from the antithrombotic and BT studies. To prepare PRP, blood was collected into a 1/10 volume of 3.8% sodium citrate and centrifuged at 250 × g for 6 min. Treatment groups consisted of vehicle (n = 15) and clopidogrel at 0.3 (n = 12), 1 (n = 12), 3 (n = 12), 10 (n = 12), and 30 mg/kg (n = 11). Platelet aggregation was measured with a platelet aggregometer (Model PAP-4D, BioData, Horsham, PA). Two hundred fifty microliters of citrated PRP was incubated for 3 min at 37°C. Aggmax and late aggregation (Agg5min) within 5 min of stimulation with ADP at 10 or 20 μM were measured. We chose to study the ex vivo platelet aggregation responses to ADP at 10 and 20 μM because these concentrations have been frequently used in the clinical studies of clopidogrel responsiveness and new P2Y12 antagonists,6,7,11,12 and they produce the maximal and highly reproducible aggregation responses. In contrast, low concentrations of ADP below 10 μM elicit rapidly and fully reversible aggregation, which is less robust and more variable.13

In a separate group of rabbits treated with vehicle or clopidogrel, ex vivo platelet aggregation-induced by 20 μM ADP was measured in the presence of 1 μM MRS-2500,8 a P2Y1 antagonist. In preliminary studies, MRS-2500 at 1 μM was shown to partially inhibit the in vitro 20 μM ADP-induced Aggmax in rabbits. Rabbit PRP was preincubated first with 1 μM MRS-2500 for 3 min before the addition of 20 μM ADP. Treatment groups in this separate study consisted of vehicle and clopidogrel at 0.3, 1, 3, 10, and 30 mg/kg (n = 6 per group).

In a separate group of rabbits treated with saline-vehicle or aspirin, ex vivo peak platelet aggregation-induced by 250 μM arachidonic acid was measured. In some rabbits, platelet production of thromboxane (TX) B2 in response to endogenous agonists such as thrombin was also determined by allowing a 1 mL tube of whole blood to clot at 37°C for 1 h.14 Serum was stored at −80°C until assay and serum TXB2 was determined by using a commercial assay ELISA kit (Neogen Corp., Lexington, KY).

[33P]-2MeS-ADP Binding

Approximately 10 mL of rabbit PRP obtained from citrated whole blood containing low level apyrase (0.25 U/mL) was incubated with gentle agitation at 37°C with an additional 2.5 U/mL apyrase for 10 minutes. The sample was then diluted to 15 mL with Buffer A [145 mM NaCl, 5.5 mM dextrose, 0.1 mM MgCl2, 5 mM KCl, 15 mM HEPES (pH 7.4), 5 mM EDTA] containing a final concentration of 0.5 μM PGE1. The diluted PRP sample was then centrifuged at room temperature (1000 × g for 10 minutes), supernatant was discarded, and the pelleted platelets were washed once in 15 mL of assay buffer [145 mM NaCl, 0.1 mM MgCl2, 5 mM KCl, 15 mM HEPES (pH 7.4), 5mM EDTA] containing 0.25 U/mL apyrase and 0.5 μM PGE1. The sample was then re-centrifuged (1000 × g, room temperature, 10 min), the supernatant was again discarded, and the pellet was re-suspended in 6 mL of assay buffer. Platelet counts were then determined on a System 9000 cell counter (Serono-Baker Diagnostics, Allentown, PA) and were generally 0.1 to 0.2 × 106/μL.

Binding reactions were conducted in 96-well PTFE filter plates (MABVN0B50; Millipore, Billerica, MA) pre-wet with 200 μL of assay buffer. Reactions consisted of 50 μL of the platelet preparation (~0.5 × 107 platelets/reaction), 12.5 μM of the P2Y1 antagonist MRS-2179 (an amount that had been previously shown to provide a complete P2Y1 receptor block under these conditions), and 1 nM [33P]-2MeS-ADP (~2000 Ci/mol) in 200 μL of assay buffer. Binding was allowed to progress for 1 h at room temperature, and the platelets were then separated from the reaction by filtration and washed 3 times with 200 μL of ice-cold PBS. Plates were air dried, 100 μL scintillation fluid was added (Ultima Gold; Perkin Elmer, Wellesley, MA) and the residual 2MeS-ADP binding was determined by scintillation counting. Total binding (0% receptor occupancy) was determined by the average response of vehicle-treated rabbits (n = 6) and generally corresponded to ~10,000 cpm/107 platelets. Nonspecific binding (100% occupancy) was determined from the same vehicle-treated samples containing 12.5 μM cold 2MeS-ADP. The average non-specific binding was ~1000 cpm/107 platelets. P2Y12 receptor occupancy was defined as % inhibition of specific binding of [33P]-2MeS-ADP.

Statistical Analyses

Statistical analyses used were ANOVA and the Student-Newman-Keuls test by the SAS system (SAS for Windows release 8.02A, Cary, NC). A value of P < 0.05 was considered statistically significant. All data are means ± SE. ED50 doses were determined using the 4-parameter logistic equation, y = A + ((B − A)/(1 + ((C/x)D))) where A = minimum y value, B = maximum, C = Log ED50, and D = slope factor, and the logistic fit was analyzed by XLfit© (Microsoft, Redmond, WA). Significant differences in ED50 values were determined by lack of overlap in 95% confidence intervals.

RESULTS

Clopidogrel and Aspirin Effects on Arterial Thrombosis and Hemostasis

Figure 1 (top panel) shows the effects of vehicle and clopidogrel on carotid blood flow after electrical stimulation. In the vehicle-treated group, basal carotid blood flow was 16 ± 1 mL/min, and final thrombus weight was 10.3 ± 0.6 mg (n = 6). After electrical stimulation, blood flow was gradually decreased, and the artery was totally occluded in about 50 min in vehicle-treated animals. Clopidogrel produced dose-dependent increases in the duration of patency. Values of integrated blood flow (%) for the vehicle, clopidogrel at 0.3, 1, 3, 10, and 30 mg/kg were 25 ± 4, 41 ± 3, 50 ± 7, 77 ± 6, 90 ± 6, and 83 ± 8, respectively (n = 6 per group). Compared with the vehicle, clopidogrel at 1 to 30 mg/kg increased integrated blood flow and inhibited thrombus formation significantly (P < 0.05; Figure 1, bottom panel). Similar antithrombotic potencies were obtained using either the integrated blood flow or thrombus reduction as indices of efficacy [ED50s (mg/kg): integrated blood flow, 1.6 ± 0.5; thrombus reduction, 1.14 ± 0.25; n = 5-6 per dose].

Figure 1
Figure 1:
Clopidogrel inhibits arterial thrombosis. (Top) Effects of vehicle and clopidogrel given orally once daily for 3 days on carotid blood flow (expressed as % of control carotid blood flow) after thrombus induction in ECAT rabbits. *P < 0.05 compared with vehicle on the basis of integrated blood flow (area-under-curve). (Bottom) Effects of vehicle and clopidogrel on thrombus formation in ECAT rabbits. *P < 0.05 compared with vehicle. Means ± SEM and n = 6 per group.

Figure 2 (top panel) shows the BT effects of vehicle and clopidogrel in rabbits. Control BT was 184 ± 5 sec in the vehicle-treated group (n = 10). Clopidogrel produced off-scale bleeding in 1 of 6 rabbits at 10 mg/kg and 1 of 5 rabbits at 30 mg/kg. There was a good separation between antithrombotic effect and increase in BT in rabbits treated with low doses of clopidogrel (Figure 2, bottom panel). For instance, the 1 mg/kg dose of clopidogrel produced a significant thrombus weight reduction of 55 ± 2% and 2 ± 0.1-fold increase in BT (P < 0.05).

Figure 2
Figure 2:
Bleeding time (BT) and antithrombotic effects of clopidogrel. (Top) Dose-dependent effects of clopidogrel on bleeding times. Bleeding time in the vehicle-treated group averaged 184 ± 5 sec. Bleeding times were determined about 2 h after the last dose and monitored up to 20 min. Clopidogrel produced off-scale bleeding in 1 of 6 rabbits at 10 mg/kg and 1 of 5 rabbits at 30 mg/kg. *P < 0.05 compared with vehicle. Means ± SEM and n = 5 to 6 per group except n = 10 for vehicle. (Bottom) Combined antithrombotic and BT data. Antithrombotic effect was indicated by % reduction of thrombus weight. BT was expressed as fold-increase over control.

As a comparator, Figure 3 (left panel) shows in vivo effects of aspirin on thrombosis and BT in rabbits. Aspirin at 0.3 and 1 mg/kg/h IV did not show significant effects on thrombus formation and BT. At 3 mg/kg/h IV aspirin significantly reduced thrombus weight by 39% ± 5% (n = 5) and prolonged BT by 1.67 ± 0.03-fold (P < 0.05, n = 6).

Figure 3
Figure 3:
Effects of aspirin. (Left) Antithrombotic and bleeding time (BT) effects of aspirin in rabbits (n = 5 to 6 per group). (Right) Ex vivo effects of aspirin on arachidonic acid-induced platelet aggregation (n = 5 per group) and serum TXB2 production (n = 3 per group). Means ± SEM.

Platelet Aggregation and Serum Thromboxane B2

Figure 3 (right panel) shows ex vivo inhibitory effects of aspirin on arachidonic acid-induced platelet aggregation and serum TXB2 production in rabbits. The Aggmax response to arachidonic acid averaged 57% ± 2% (n = 15), and serum TXB2 production averaged 414 ± 103 ng/mL (n = 3). Aspirin at 1 and 3 mg/kg/h IV produced almost complete inhibition of both the platelet aggregation response to arachidonic acid and serum TXB2 production (P < 0.05).

Ex vivo effects of clopidogrel on platelet aggregation induced by 10 and 20 μM ADP are shown in Figure 4. Clopidogrel at top doses tested only partially inhibited 10 and 20 μM ADP-induced Aggmax but had a significantly greater effect on Agg5min. We noted that the inhibitory effects of clopidogrel against 10 and 20 μM ADP-induced platelet aggregation were almost identical (n = 11-12 per dose). As a result, only aggregation data with 20 μM ADP are shown in later figures.

Figure 4
Figure 4:
Inhibition of ex vivo peak platelet aggregation (Aggmax and Agg5min) responses to 10 and 20 μM ADP in clopidogrel-treated rabbits. Vehicle or clopidogrel was given orally once daily for 3 days. The platelet aggregation response to ADP was measured about 2 to 3 h after the last dose. Means ± SE and n = 11-12 per dose.

Figure 5 shows the dose-response relationship of ex vivo IPA with thrombus reduction and BT in clopidogrel-treated rabbits on the same axis. In terms of proportionate response, IPA based on Aggmax underestimated the antithrombotic efficacy of clopidogrel at all doses, whereas IPA based on Agg5min better matched the antithrombotic efficacy only at high clopidogrel doses of 10 and 30 mg/kg (Figure 5, left panel). We also noted that IPA, especially Agg5min, tracked better with BT (Figure 5, right panel). As shown in Figure 6 (left panel), clopidogrel at its maximal dose achieved a high level of antithrombotic efficacy of 85% ± 1% and a BT prolongation of 6.0 ± 0.4 times control but an IPA of 57% ± 5% assessed by Aggmax. At a low IPA of 18% ± 5% assessed by Aggmax, clopidogrel significantly decreased thrombus weight by 55% ± 2% and increased BT moderately to 2.0 ± 0.1 times control (P < 0.05). A similar pattern of results was also noted with IPA based on Agg5min, where a low IPA level of 21% ± 4% was also associated with a significant antithrombotic effect (P < 0.05; Figure 6, right panel). The thrombus weight reduction with this dose of clopidogrel was significantly greater than the maximal thrombus weight reduction observed with aspirin (P < 0.05).

Figure 5
Figure 5:
Matching in vivo antithrombotic (left panel) and bleeding time (BT) effects (right panel) to ex vivo platelet aggregation responses (Aggmax and Agg5min) to 20 μM ADP in rabbits. Vehicle or clopidogrel was given orally once daily for 3 days. The platelet aggregation response to ADP was measured 2 to 3 h after the last dose. Means ± SE. n = 5-6 per group in thrombosis and bleeding time studies and n = 11 to 12 in platelet aggregation studies.
Figure 6
Figure 6:
Correlation of ex vivo inhibition of 20 μM ADP-induced platelet aggregation, Aggmax (left panel) and Agg5min (right panel), with antithrombotic and bleeding time effects in clopidogrel-treated rabbits. Vehicle or clopidogrel was given orally once daily for 3 days. The platelet aggregation response to ADP was measured 2 to 3 h after the last dose. Means ± SE. n = 5 to 6 per group in thrombosis and bleeding time studies and n = 11 to 12 in platelet aggregation studies.

As both P2Y1 and P2Y12 contribute to the ADP-induced platelet aggregation, we, therefore, modified this platelet aggregation response by the addition of 1 μM MRS-2500, a P2Y1 antagonist. Unlike the results shown in Figure 5, we noted that this modified IPA tracked better with antithrombotic than BT effects in the clopidogrel-treated rabbits (Figure 7). Control Aggmax and Agg5min measured in the presence of 1 μM MRS-2500 were still very robust and averaged 43% ± 6% and 41% ± 6%, respectively (n = 6). Interestingly, in the presence of MRS-2500, maximal inhibition of Aggmax and Agg5min by clopidogrel was increased to 90% ± 4% and 100% ± 0%, respectively (Figure 7). Additionally, Aggmax measured in the presence of MRS-2500 appears to be preferred over standard IPA as a measure of clopidogrel antithrombotic efficacy as the ED50s for the modified IPA assay and in vivo efficacy were almost identical [antithrombotic ED50 = 1.14 ± 0.25 mg/kg and Aggmax ED50 (+MRS-2500) = 1.12 ± 0.30 mg/kg]. In contrast, IPA measured by Agg5min in the presence of MRS-2500 produced an ED50 of less than 0.3 mg/kg, thereby making it too sensitive to be an optimal monitor of clopidogrel efficacy (Figure 7).

Figure 7
Figure 7:
Matching ex vivo inhibition of platelet aggregation in the presence of MRS-2500 with in vivo antithrombotic (left panel) and bleeding time (BT) effects (right panel) in clopidogrel-treated rabbits. Antithrombotic and BT data were obtained fromFigures 2 and 3. Ex vivo platelet aggregation was measured in a separate group of rabbits treated with clopidogrel. Peak aggregation (Aggmax) and late aggregation (Agg5min) responses induced by 20 μM ADP were measured in the presence of 1 μM MRS-2500. Means ± SE. n = 5 to 6 per group in thrombosis and bleeding studies and n = 6 per group in the platelet aggregation studies.

[33P]-2MeS-ADP Binding

Figure 8 shows ex vivo effects of clopidogrel on the specific binding of [33P]-2MeS-ADP. Specific binding of [33P]-2MeS-ADP to the washed platelets from vehicle-treated control rabbits averaged 611 ± 70 cpm/106 platelets (n = 11). As shown in Figure 8, clopidogrel inhibited [33P]-2MeS-ADP binding in a concentration-dependent manner with ED50 of 1.8 ± 0.2 mg/kg for clopidogrel (n = 5-6 per group except n = 11 for clopidogrel at 30 mg/kg), which was similar to its antithrombotic ED50 of 1.14 ± 0.25 mg/kg. We noted that receptor occupancy tracked better with antithrombotic than BT effects in the clopidogrel-treated rabbits (Figure 8).

Figure 8
Figure 8:
Effects on ex vivo receptor binding, thrombosis, and bleeding time (BT). Matching in vivo antithrombotic and BT responses to platelet P2Y12 receptor occupancy. Effects of clopidogrel on specific [33P]-2MeS-ADP binding (expressed as % P2Y12 receptor occupancy) to rabbit washed platelets. Vehicle or clopidogrel was given orally once daily for 3 days. Washed platelets from these rabbits were prepared 2 to 3 h after the last dose. Means ± SE and n = 5 to 6 per group except n = 11 for clopidogrel at 30 mg/kg.

DISCUSSION

Inhibition of the ex vivo platelet aggregation response to ADP is considered the gold standard in assessing the antiplatelet effects of P2Y12 receptor antagonists.1,3-7 Unfortunately, conclusive clinical event data are still lacking for calibrating different IPA levels with antithrombotic efficacy and bleeding liability of P2Y12 receptor antagonists. Our study was the first preclinical evaluation of ex vivo antiplatelet activity markers as predictors of in vivo antithrombotic and bleeding effects of the selective P2Y12 antagonist clopidogrel in rabbits. Potent antithrombotic activities of clopidogrel and other P2Y12 receptor antagonists have also previously been reported in rabbits.15-17

The antithrombotic and BT effects of clopidogrel were evaluated in the rabbit ECAT and cuticle BT models, respectively.10 To appreciate the complete dose-response relationship between ex vivo platelet activity biomarkers and in vivo effects of clopidogrel, we treated rabbits with clopidogrel with a full range of doses from 0.3 to 30 mg/kg. As shown in this study, clopidogrel produced dose-dependent effects in both models, and there is a good separation between doses that produce the antithrombotic and the BT effects. The 10-mg/kg dose of clopidogrel appears to be the maximal antithrombotic dose; it almost completely preserved integrated blood flow and reduced thrombus formation significantly. In contrast, increasing the clopidogrel dose to 30 mg/kg did not improve antithrombotic efficacy but caused a further increase in BT. Our study, therefore, cautions that the use of near-to-maximal or maximal antithrombotic doses of clopidogrel and possibly other P2Y12 antagonists could increase bleeding complications. We noted that integrated blood flow tracked very well with thrombus reduction in rabbits treated with clopidogrel, and thus may serve as an efficacy marker for clopidogrel treatment in this model.

As aspirin is a widely used antiplatelet agent in patients for the prevention of atherothrombosis,18 it was included as a comparator in this study. We observed that aspirin produced a consistent inhibition of ex vivo arachidonic acid-induced platelet aggregation and serum TXB2 production in rabbits. Complete inhibition of platelet aggregation and serum TXB2 production were obtained with aspirin infusions at 1 and 3 mg/kg/h. As complete suppression of arachidonic acid-induced platelet aggregation and serum TXB2 formation is needed for optimal clinical antithrombotic effects of aspirin,18 aspirin infusions at 1 and 3 mg/kg/h in the rabbit may be considered as clinically relevant doses. Though 1 and 3 mg/kg/h aspirin produced maximal ex vivo antiplatelet activities, it produced moderate in vivo effects on thrombosis and hemostasis. In this regard, clopidogrel is a more effective antithrombotic agent than aspirin. At clopidogrel and aspirin doses that produced equivalent, approximately 2-fold increase in BT, clopidogrel appears to be a more effective antithrombotic agent than aspirin. Similar results were also observed in the CAPRIE trial showing that clopidogrel was superior to aspirin in reducing the risk of major cardiovascular events among patients at high risk of secondary ischemic events.1,19

Despite the recent introduction of a variety of platelet function assays, ADP-induced platelet aggregation is still the most widely used ex vivo marker to measure the effects of clopidogrel. The platelet aggregation response to ADP can be characterized as maximal (ie, Aggmax) or late (ie, Agg5min) aggregation.20,21 Aggmax is the most widely used index of aggregation in clinical studies of clopidogrel responsiveness and new P2Y12 antagonists.6,7,11,12 In this study, clopidogrel at maximal doses only partially inhibited ADP-induced Aggmax by about 55% in rabbits. A similar degree of partial blockade of the 20 μM ADP-induced Aggmax was also reported in humans treated with the maximal doses of P2Y12 antagonists.6,22 A complete inhibition of Aggmax was achieved only when MRS-2500,8 a selective P2Y1 receptor antagonist, was added, indicating that the residual Aggmax is mainly driven by ADP-induced activation of the P2Y1 receptor. The opposite is true for the late aggregation response to ADP. Clopidogrel inhibited Agg5min, by more than 80%, suggesting the Agg5min response is primarily due to the activation of the P2Y12 receptor. Previous reports also showed that the ADP-induced peak aggregation in human PRP is influenced by activities of both the P2Y1 and P2Y12 receptors, whereas the late aggregation is determined mainly by the activity of the P2Y12 receptor.17,20,21

It has been argued that clopidogrel at the maintenance dose of 75 mg used in the clinic is a submaximal dose, which yields incomplete P2Y12 receptor blockade.5,23 Consequently, clopidogrel achieves only moderate inhibition of Aggmax to 20 μM ADP by about 30% in humans.6,7 Our study demonstrated that even at this moderate level of IPA, clopidogrel still produced a significant and greater antithrombotic effect than aspirin in rabbits. Thus a low IPA is not necessarily associated with a low antithrombotic effect in rabbits-treated with clopidogrel. Therefore, our study raises doubts about the sensitivity and translational value of IPA based on Aggmax in response to 20 μM ADP as a predictive marker for quantifying clopidogrel efficacy. Our study also indicates that greater degrees of IPA achieved by higher doses of clopidogrel were associated with marginal gains in antithrombotic efficacy but with greater increases in BT. Our finding is consistent with the clinical report by Husted et al6 showing that AZD6140, a selective P2Y12 antagonist, at doses that produced maximal IPA increased BT to a greater extent than clopidogrel at 75 mg that produced low IPA. Thus both preclinical and clinical studies suggest that excessive inhibition of platelet aggregation with complete P2Y12 antagonism may increase the risk of bleeding.

In matching ex vivo IPA with in vivo effects of clopidogrel, we observed that IPA based on Aggmax underestimated the antithrombotic efficacy of clopidogrel at the doses studied in rabbits. Since Aggmax depends on the activities of both the P2Y1 and P2Y12 receptors, it is conceivable that standard IPA is not a sensitive and specific test for measuring clopidogrel efficacy. However, Agg5min, which depends mainly on the activity of the P2Y12, is expected to be more specific for monitoring clopidogrel efficacy. Interestingly, IPA assessed by Agg5min also underestimated the antithrombotic efficacy of clopidogrel except at the maximal doses of clopidogrel that also substantially increased BT. It is possible that multiple mechanisms in addition to antiaggregating activity could account for the antithrombotic efficacy of clopidogrel and thus ADP-induced platelet aggregation per se is not sufficient to measure clopidogrel efficacy. Indeed, the P2Y12 receptor has been shown to be involved in the procoagulant activity of platelets, the amplification of platelet aggregation induced by agonists such as thrombin, TXA2, and collagen, and shear-induced platelet aggregation, and these P2Y12 activities may all contribute to thrombosis.24 We noted that platelet aggregometry, especially Agg5min, tracked better with BT. Since the P2Y12 receptor plays a key role in the stabilization of platelet aggregates,24 our study suggests that P2Y12-mediated stabilization of platelets may also play a major role in the BT effect of clopidogrel in rabbits.

Aggmax is a widely used aggregation index in the clinic; it is, therefore, important to improve the sensitivity of this clinical marker of clopidogrel responsiveness. Because both P2Y1 and P2Y12 contribute to the ADP-induced Aggmax response, one approach to make Aggmax more sensitive and specific for P2Y12 is to measure it in the presence of P2Y1 blockade. However, this approach is complicated by the fact that maximal P2Y1 blockade with a P2Y1 antagonist, such as MRS-2500, completely inhibits the Aggmax response in rabbits (Pancras C. Wong, unpublished data). Nevertheless, a solution was provided by the study of Hardy et al,25 who reported that P2Y12 might positively regulate the P2Y1-regulated calcium level, whereas P2Y1 might negatively regulate this action of P2Y12 in human platelets. This downregulation of the P2Y12-mediated signal by P2Y1 could be blocked by a submaximal concentration of a P2Y1 antagonist.25 We reasoned that P2Y1 might also downregulate the P2Y12-induced platelet aggregation and that a submaximal concentration of a P2Y1 antagonist might provide a useful means to unmask most of the P2Y12 components of the ADP-induced platelet aggregation. This is indeed the case. In the presence of 1 μM of MRS-2500, which inhibits the 20 μM ADP-induced Aggmax and Agg5min by less than 30%, clopidogrel at the maximal antithrombotic dose almost completely inhibited the remaining ex vivo platelet aggregation response to 20 μM ADP. Moreover, there is a very good match between this modified IPA, especially based on Aggmax, and the level of antithrombotic efficacy in clopidogrel-treated rabbits. Interestingly, this modified IPA overestimated the BT effects of clopidogrel, especially at the submaximal doses of clopidogrel. Reasons accounting for this finding are not known. However, we noted that the duration of thrombosis and bleeding responses in the rabbit ECAT and BT models is different (ie, 40 min versus 3 min, respectively), suggesting the involvement of different mechanisms. It is possible that thrombosis and BT effects may not utilize all P2Y12 activities to the same extent and thus may account for the differences in their relationship to IPA.

To substantiate our findings with IPA, we assessed [33P]-2MeS-ADP binding to platelet P2Y12 receptors and compared this with IPA to assess its value as a measure reflecting the in vivo effects of clopidogrel. Previous studies showed that clopidogrel produced a partial ex vivo inhibition of the binding of the radiolabeled 2MeS-ADP on platelets, which is attributed to the presence of unblocked P2Y1 receptors, as 2MeS-ADP is a potent ligand for both the P2Y1 and P2Y12 receptors.1,26,27 Thus the binding assay in the present study was carried out in the presence of complete P2Y1 blockade. As a result, the percent inhibition of specific binding of [33P]-2MeS-ADP represented the percentage of P2Y12 receptor occupancy. Clopidogrel, via the irreversible binding of its hepatic-generated active metabolite to platelet P2Y12 receptors, produced a dose-dependent reduction of the specific binding of [33P]-2MeS-ADP and achieved greater than 80% receptor occupancy at its maximal antithrombotic dose. Similar to the modified IPA, receptor occupancy for clopidogrel tracked well with antithrombotic efficacy, but less so with BT. As the active metabolite of clopidogrel formed in vivo is short-lived and difficult to detect in blood, our study suggests that ex vivo receptor occupancy may be used as a pharmacodynamic marker for productive active metabolite activity of clopidogrel.

CONCLUSION

Our study demonstrates that clopidogrel at a dose that caused little inhibition of standard ADP-induced platelet aggregation and low receptor occupancy could produce significant antithrombotic efficacy, which was greater than that of aspirin, with a moderate BT increase in rabbits. In addition, the magnitude of standard IPA consistently underestimated antithrombotic efficacy but tracked well with BT effects. However, modified IPA, which was measured in the presence of partial P2Y1 blockade, quantified well the magnitude of clopidogrel efficacy. The same was true for P2Y12 receptor occupancy. If these findings translate into humans, our study suggests that low standard IPA values may be associated with meaningful antithrombotic effects of clopidogrel. Moreover, modified IPA and P2Y12 receptor occupancy appear to predict the magnitude of clopidogrel's efficacy, and standard IPA may be a better predictor of bleeding liability.

ACKNOWLEDGMENTS

We thank Jergen Baumann and Qimin Wu for technical assistance.

REFERENCES

1. Savi P, Herbert J-M. Clopidogrel and ticlopidine: P2Y12 adenosine diphosphate-receptor antagonists for the prevention of atherothrombosis. Semin Thromb Hemost. 2005;31:195-204.
2. Born GVR. Aggregation of blood platelets by adenosine diphosphate and its reversal. Nature. 1992;194:927-929.
3. Niitsu Y, Jakubowski JA, Sugidachi A, et al. Pharmacology of CS-747 (prasugrel, LY640315), a novel, potent antiplatelet agent with in vivo P2Y12 receptor antagonist activity. Semin Thromb Hemost. 2005;31:184-194.
4. Packham MA, Mustard JF. Platelet aggregation and adenosine diphosphate/adenosine triphosphate receptors: a historical perspective. Semin Thromb Hemost. 2005;31:129-138.
5. van Giezen JJJ, Humphries RG. Preclinical and clinical studies with selective reversible direct P2Y12 antagonists. Semin Thromb Hemost. 2005;31:195-204.
6. Husted S, Emanuelsson H, Heptinstall S, et al. Pharmacodynamics, pharmacokinetics, and safety of the oral reversible P2Y12 antagonist AZD6140 with aspirin in patients with atherosclerosis: a double-blind comparison to clopidogrel with aspirin. Eur Heart J. 2006;27:1038-1047.
7. Jernberg T, Payne CD, Winters KJ, et al. Prasugrel achieves greater inhibition of platelet aggregation and a lower rate of non-responders compared with clopidogrel in aspirin-treated patients with stable coronary artery disease. Eur Heart J. 2006;27:1166-1173.
8. Cattaneo M, Lecchi A, Ohno M, et al. Antiaggregatory activity in human platelets of potent antagonists of the P2Y1 receptor. Biochem Pharmacol. 2004;68:1995-2002.
9. Romson JL, Haack DW, Lucchesi BR. Electrical induction of coronary artery thrombosis in the ambulatory canine: a model for in vivo evaluation of anti-thrombotic agents. Thromb Res. 1980;17:841-853.
10. Wong PC, Crain EJ, Watson CA, et al. Nonpeptide factor Xa inhibitors III: Effects of DPC423, an orally-active pyrazole antithrombotic agent, on arterial thrombosis in rabbits. J Pharmacol Exp Ther. 2002;303:993-1000.
11. Dörr G, Schmidt G, Grafe M, et al. Effects of combined therapy with clopidogrel and acetylsalicylic acid on platelet glycoprotein expression and aggregation. J Cardiovasc Pharmacol. 2002;39:523-532.
12. Gurbel PA, Bliden KP, Samara W, et al. Clopidogrel effect on platelet reactivity in patients with stent thrombosis: results of the CREST Study. J Am Coll Cardiol. 2005;46:1827-1832.
13. Rand ML, Leung R, Packham MA. Platelet function assays. Transfus Apher Sci. 2003;28:307-317.
14. Patrono C, Ciabattoni G, Pinca E, et al. Low dose aspirin and inhibition of thromboxane B2 production in healthy subjects. Thromb Res. 1980;17:317-327.
15. Herbert J-M, Dol F, Bernat A, et al. The antiaggregating and antithrombotic activity of clopidogrel is potentiated by aspirin in several experimental models in the rabbit. Thromb Haemost. 1998;80:512-518.
16. Schlitt A, Hauroeder B, Buerke M, et al. Effects of combined therapy of clopidogrel and aspirin in preventing thrombus formation on mechanical heart valves in an ex vivo rabbit model. Thromb Res. 2002;107:39-43.
17. van Gestel MA, Heemskerk JW, Slaaf DW, et al. In vivo blockade of platelet ADP receptor P2Y12 reduces embolus and thrombus formation but not thrombus stability. Arterioscler Thromb Vasc Biol. 2002;23:518-523.
18. Patrono C, Garcia Rodriguez LA, Landolfi R, et al. Low-dose aspirin for the prevention of atherothrombosis. N Engl J Med. 2005;353:2373-2383
19. CAPRIE Steering Committee. A randomized, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet. 1996;348:1329-1339.
20. Jarvis GE, Humphries RG, Robertson MJ, et al. ADP can induce aggregation of human platelets via both P2Y1 and P2T receptors. Br J Pharmacol. 2000;129:275-282.
21. Labarthe B, Théroux P, Angioï M, et al. Matching the evaluation of the clinical efficacy of clopidogrel to platelet function tests relevant to the biological properties of the drug. J Am Coll Cardiol. 2005;46:638-645.
22. Asai F, Jakubowski JA, Naganuma H, et al. Platelet inhibitory activity and pharmacokinetics of prasugrel (CS-747) a novel thienopyridine P2Y12 inhibitor: a single ascending dose study in healthy humans. Platelets. 2006;17:209-217.
23. Aleil B, Ravanat C, Cazenave JP, et al. Flow cytometric analysis of intraplatelet VASP phosphorylation for the detection of clopidogrel resistance in patients with ischemic cardiovascular diseases. J Thromb Haemost. 2005;3:85-92.
24. Hechler B, Cattaneo M, Gachet C. The P2 receptors in platelet function. Semin Thromb Hemost. 2005;31:150-161.
25. Hardy AR, Jones ML, Mundell SJ, et al. Reciprocal cross-talk between P2Y1 and P2Y12 receptors at the level of calcium signaling in human platelets. Blood. 2004;104:1745-1752.
26. Savi P, Laplace MC, Maffrand JP, et al. Binding of [3H]-2-methylthio ADP to rat platelets-effect of clopidogrel and ticlopidine. J Pharmacol Exp Ther. 1994;269:772-777.
27. Sugidachi A, Asai F, Ogawa T, et al. The in vivo pharmacological profile of CS-747, a novel antiplatelet agent with platelet ADP receptor antagonist properties. Br J Pharmacol. 2000;129:1439-1446.
Keywords:

ADP receptor; clopidogrel; hemostasis; platelet aggregation; receptor occupancy; thrombosis

© 2007 Lippincott Williams & Wilkins, Inc.