Payne, Christopher D MS*; Li, Ying Grace MS†; Small, David S PhD†; Ernest, C Steven II MS†; Farid, Nagy A PhD†; Jakubowski, Joseph A PhD†; Brandt, John T MD†; Salazar, Daniel E PhD‡; Winters, Kenneth J MD†

INTRODUCTION

Platelets play a critical role in atherothrombosis; accordingly, antiplatelet therapy is widely prescribed for primary and secondary prevention of atherothrombotic events.^1-3 The antiplatelet agents aspirin and clopidogrel are the most commonly used, often in combination.^4-10 Clopidogrel is metabolized by 2 consecutive cytochrome P-450 (CYPs)-dependent steps to a thiol-containing active metabolite (clopidogrel-AM) that irreversibly binds to the P2Y₁₂ receptor of platelets, thereby inhibiting adenosine diphosphate (ADP)-mediated platelet activation and aggregation.^11-13 Concerns have been raised regarding the interindividual variability of platelet inhibition with clopidogrel, and the potential risk for recurrent atherothrombotic events.^14-18 To address these concerns, higher clopidogrel loading doses (LD) of 600-mg and 900-mg have been investigated.^14,19-21 A clopidogrel 600-mg LD achieves greater levels of inhibition of platelet aggregation (IPA) than the approved clopidogrel 300-mg LD, albeit still with substantial response variability.^14,19 In addition, the level of IPA observed after a clopidogrel 900-mg LD was not significantly higher than after a clopidogrel 600-mg LD.²²

Prasugrel is a novel and potent thienopyridine that targets the same P2Y₁₂ ADP receptor as clopidogrel.^23-25 Unlike clopidogrel, conversion of prasugrel to its active metabolite (prasugrel-AM) involves rapid hydrolysis by esterases followed by a single CYP-dependent step.^26,27 The CYPs contributing to prasugrel-AM formation are located predominantly in the gut lumen (CYP3A4 and CYP3A5) and the liver (CYP3A4, CYP3A5, CYP2B6, CYP2C9, and CYP2C19).^27,28 Previous studies have shown that prasugrel is absorbed quickly after dosing, with concentrations of its active metabolite peaking approximately 30 minutes after dosing.²⁹ On a molar basis, the active metabolites of clopidogrel and prasugrel are equipotent platelet inhibitors.^30,31

The present study compared the level of platelet inhibition and degree of response variability achieved by a prasugrel 60-mg LD and a 10-mg MD to those obtained with a clopidogrel 300-mg/75-mg and 600-mg/75-mg regimen. In addition, the pharmacokinetics of the active metabolites for both drugs and the relationship between active metabolite exposure and extent of platelet inhibition were characterized. The prasugrel 60-mg/10-mg LD/MD regimen used in this study is currently under investigation in a large Phase 3 clinical trial in patients with acute coronary syndrome undergoing percutaneous coronary intervention.³²

METHODS

Subject Population

Forty-one healthy subjects, 25 men and 16 nonpregnant women aged 20 to 63 (mean 38.3) years, inclusive, with body mass index of 19 to 32 kg/m², and a screening baseline maximal platelet aggregation (MPA) response ≥70% to 20 μM ADP and 1.5 mM arachidonic acid, participated in this study. Of the 41 subjects, 32 were of Caucasian descent and 9 were of African descent.

Study Design

This single site, open-label, randomized, 3-treatment, 3-sequence crossover study (unique study identifier H7T-EW-TAAZ) was conducted from July 2005 to January 2006. Subjects received a LD and 7 days of daily MD; dosing regimens (LD/MD) were clopidogrel 300-mg/75-mg, clopidogrel 600-mg/75-mg, and prasugrel 60-mg/10-mg. A minimum 14-day washout period separated each treatment. On the day before each LD, subjects were admitted to the research unit. On Day 1 of each treatment period, subjects received LD of clopidogrel 300-mg (4 × 75-mg tablets), clopidogrel 600-mg (8 × 75-mg tablets), or prasugrel 60-mg (6 × 10-mg tablets). Commercially available clopidogrel bisulfate (Plavix® 75-mg tablets, Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership, New York, NY) and tablets containing 10-mg of prasugrel as the HCl salt (Eli Lilly and Company, Indianapolis, IN) were administered in the morning after an overnight fast with approximately 200 mL of water. Subjects remained in the research unit for 24 hours of observation and blood sampling, and were discharged on Day 2. Subjects returned as outpatients to the research unit each day for 7 days to receive either clopidogrel 75-mg or prasugrel 10-mg MD. On Days 1, 2, and 8 of each treatment period, subjects fasted overnight for at least 10 hours before dosing. During the days of outpatient dosing, a light lowfat breakfast was allowed before dosing, except as noted above. Consumption of grapefruit-containing products was prohibited from 7 days before enrollment through the final follow-up visit 14 days after the final dose. Subjects did not receive concomitant aspirin, nonsteroidal antiinflammatory drugs known to affect platelet function, or other medications that are inhibitors or inducers of CYP-450 enzymes from 30 days before admission and through 14 days after the final dose. While resident in the unit, subjects were not permitted to use nicotine-containing products, and were asked to refrain from smoking for at least 10 hours before each outpatient visit until all procedures had been completed for outpatient visits on Days 3 to 9. The institutional review board approved the protocol, and the study was conducted in accordance with regulatory standards and Good Clinical Practice guidelines rooted in the Declaration of Helsinki. All subjects provided signed, informed consent.

Assessment of Platelet Aggregation

Blood samples were collected into 3.8% sodium citrate tubes predose and at 0.25, 0.5, 1, 2, 4, and 6 hours after the prasugrel and clopidogrel doses on Day 1, predose on Days 2 to 7, and predose and 24 hours after dosing on Day 8 for assessment of platelet aggregation. Platelet-rich and platelet-poor plasma were prepared by differential centrifugation, at room temperature, as previously described.³³ The platelet count in platelet-rich plasma was adjusted to approximately 250 × 10⁹/L by addition of autologous platelet-poor plasma. Maximum platelet aggregation (MPA) in platelet-rich plasma was recorded during the 6-minute monitoring period after addition of agonists (20 μM or 5 μM ADP) using a Chrono-Log 4-channel optical aggregometer (Chrono-Log Corporation, Havertown, PA) with temperature maintained at 37°C.

Pharmacokinetic Assessments

Blood samples for the determination of plasma concentrations of the active metabolites of clopidogrel and prasugrel were collected at predose (Day 8) and at 0.25, 0.5, 1, 1.5, 2, 4, 8, 12, and 24 hours after the prasugrel or clopidogrel doses on Days 1 and 8. Since the active metabolites of prasugrel and clopidogrel are not stable in blood, 3′-methoxy-phenacylbromide was added within 30 seconds of blood sample collection in ethylenediaminetetraacetic acid (EDTA) to derivatize and stabilize the metabolites.²⁶ After extraction, the active metabolites of prasugrel and clopidogrel were quantified by validated liquid chromatography with tandem mass spectrometry detection (LC-MS/MS) methods.^26,34 Pharmacokinetic parameter estimates for prasugrel-AM and clopidogrel-AM were calculated using noncompartmental methods of analysis using the log-linear trapezoidal method of Model 200 in WinNonlin (Pharsight Corporation, Cary, NC). The primary pharmacokinetic parameters were maximum observed concentration (C_max) and area under the plasma concentration-time curve (AUC) from the time of dosing through the sampling time of the last quantifiable concentration (AUC_0-tlast) for the active metabolites. The molecular weights of clopidogrel-AM (356 g/mol) and prasugrel-AM (349 g/mol) differ by only 2%, thus direct comparison is acceptable.

Due to improper blood collection or shipping procedures, 163 plasma samples from 19 subjects were lost to pharmacokinetic analysis. These samples represented 8% of all PK samples. Samples lost to the analysis were distributed evenly across the different treatments. The losses reduced the number of samples available for pharmacokinetic analysis but did not affect any single treatment group disproportionally or impact the derived conclusions.

In 2 subjects, plasma active metabolite concentrations were atypically low (<1 ng/mL) at all sampling times after the last prasugrel MD, although they were consistent with those in other subjects after the prasugrel LD and after the clopidogrel LD and MD. The atypically low concentrations were considered unreliable and were excluded from the pharmacokinetic analysis. All pharmacodynamic data in these subjects were consistent with those in other subjects and were included in the pharmacodynamic analysis.

Safety Monitoring

Adverse events, vital signs, clinical laboratory evaluations, fundoscopy, and physical examinations for signs of bleeding or petechiae were recorded as safety measures.

Statistical Methods

The crossover study design allowed the extent of IPA of a subject after each clopidogrel dosing regimen to be contrasted with the same subject's response to prasugrel. Thirty-three subjects provided 90% power to detect a significant difference in IPA among treatments at a given time point. This sample size was calculated based on a 1-sided α = 0.05 test and an assumption that the standard deviation of IPA response for each study drug LD treatment is 20% and that there is a 15-percentage point magnitude of difference in the IPA level among treatments.

All statistical analyses were performed with SAS software (version 8.2; SAS Institute Inc., Cary, NC). A linear mixed effect model was used to compare the IPA to 20 μM ADP among treatments at each time point without any multiplicity adjustment. In this model, Day 1 predose MPA was a covariate; treatment, time, and treatment-by-time were fitted as fixed effects; subject, subject-by-treatment, and subject-by-time were random effects. In addition, the model allows for different residual errors for different treatments.

The IPA for each post dose sample was calculated with the following formula:

Equation (Uncited)

Statistical analysis of the 3 sets of baseline MPA values indicated no period effect; accordingly, to minimize baseline variability, MPA₀ is the mean of the three Day 1 predose MPA values from each of the 3 periods; MPA_t is the MPA measured at time t, and IPA_t is the IPA calculated at time t.

A likelihood ratio test was used to compare the IPA variability among the treatments at each time point with the aid of 2 nested linear mixed effect models with treatment as a fixed effect.³⁵ One model allowed different variance for different treatments; the other forced common variance for all treatments. This test was based on a difference in the log-likelihoods with a χ² distribution having 2 degrees of freedom. Similar analyses were conducted to compare the mean and variability of IPA using 5 μM ADP.

On the basis of published Bayesian threshold criteria, pharmacodynamic poor responders were identified as those with IPA less than 20% (20 μM ADP) or IPA less than 25% (5 μM ADP) at either 4 or 24 hours after the LD (or at 24 hours after the last MD).^36,37 By this definition, a poor pharmacodynamic response could not be statistically differentiated from the response observed in the absence of drug administration.

RESULTS

Subject Demographics and Disposition

Of the 41 subjects, 33 completed the study. Seven subjects withdrew for personal reasons (4, 2, and 1 during clopidogrel 300-mg/75-mg, clopidogrel 600-mg/75-mg, and prasugrel 60-mg/10-mg, respectively). Six subjects withdrew during or after their first treatment sequence, 2 subjects withdrew during or after their second treatment sequence, and no subjects withdrew during the third treatment sequence. One of the subjects who completed only the prasugrel treatment period withdrew from the study after experiencing a syncopal episode 6 days after the final dose of study drug. The syncopal episode, assessed as unrelated to the study drug or procedures, resulted in serious adverse events related to head trauma sustained during a fall; this subject did not receive clopidogrel. Protocol noncompliance for 1 subject was strongly suspected when all but one plasma concentration of active metabolites during all clopidogrel and prasugrel treatments were below quantitation limits. In addition, MPA responses did not deviate from the predose baseline level after any prasugrel or clopidogrel LD or MD. This subject was excluded from all pharmacodynamic and pharmacokinetic analyses.

Pharmacodynamics: Inhibition of Platelet Aggregation to 20 μM ADP

Consistency of Baseline MPA Over Time

No period effect was observed for baseline MPA across the 3 treatment periods. For one subject; however, a single baseline MPA measurement (15%) fell outside the acceptable lower limit of the assay range, and was excluded from analysis. MPA data from the other 2 treatment sequences for this subject were averaged to establish the baseline. The IPA profiles determined with 20 μM and 5 μM ADP were qualitatively similar for each thienopyridine.

Loading Dose Pharmacodynamics

At 30 minutes after LD, prasugrel 60-mg achieved greater mean IPA to 20 μM ADP (54%) than either clopidogrel 300-mg (3%) or 600-mg (6%) (Figure 1, P < 0.001). The prasugrel 60-mg LD also achieved greater IPA by 1 hour (82%) and 2 hours (91%) than the 6-hour peak IPA for clopidogrel 300-mg (51%) or 600-mg (69%) (P < 0.01). Compared to a clopidogrel 300-mg LD, a clopidogrel 600-mg LD achieved significantly higher IPA at 1-, 2-, 4-, 6-, and 24-hours after dosing (Figure 1).

Figure 1

Maintenance Dose Pharmacodynamics

The mean IPA to 20 μM ADP after the prasugrel 10-mg MD exceeded 75% throughout the MD phase and was significantly higher (P < 0.001) than that achieved in either clopidogrel 75-mg MD sequence (Figure 1). Steady-state IPA during clopidogrel MD (IPA ∼ 56%) was observed on Day 4 for both clopidogrel dosing regimens and on Day 4 for prasugrel MD (82%). As anticipated, no significant differences were observed in IPA at steady-state during MD between the 2 clopidogrel LD/MD regimens.

Intrasubject Response Variability in Platelet Inhibition

The IPA for each subject at specific timepoints during the crossover between prasugrel and the 2 clopidogrel LD/MD regimens is shown in Figure 2. Overall, the IPA achieved with prasugrel LD was higher and less variable than observed after either clopidogrel LD. From 2 hours through 24 hours after LD administration, the variability in IPA with prasugrel is significantly lower (P < 0.01) than that in either clopidogrel LD. Of note, compared to the clopidogrel 300-mg LD, the clopidogrel 600-mg LD did not significantly reduce intersubject variability in IPA even though mean IPA was somewhat higher with the 600-mg LD. During MD, the variability in IPA was significantly less with prasugrel 10-mg than with clopidogrel 75-mg MD (P < 0.01).

Figure 2

Pharmacodynamic Poor Responder Rates

Using the Bayesian criteria for 20 μM ADP as defined above, 39% (14 of 33) and 6% (2 of 33) of subjects were classified as pharmacodynamic poor responders to clopidogrel 300-mg and 600-mg LDs, respectively, whereas all subjects responded to the prasugrel 60-mg LD. Using 5 μM ADP, the percentage of poor responders was 19% (7 of 36) and 3% (1 of 33) with clopidogrel 300-mg and 600-mg LDs, respectively; while all subjects responded to the prasugrel LD. During MD, all subjects on prasugrel 10-mg were classified as responders using 20 μM and 5 μM ADP. However, during the clopidogrel MD phase, 9% (3 of 33), and 12% (4 of 33) of subjects in the 300-mg/75-mg and 600-mg/75-mg MD sequences were poor responders with 20 μM ADP. With 5 μM ADP, 9% (3 of 33) and 6% (2 of 33) of subjects showed a poor IPA response in the clopidogrel 300-mg/75-mg and 600-mg/75-mg sequences, respectively, during the MD phase.

Active Metabolite (AM) Pharmacokinetics

Tables 1 and 2 represent the respective LD and MD exposure estimates to prasugrel-AM and clopidogrel-AM, measured as C_max and AUC_0-tlast after the 3 treatment regimens. After the LDs, the AUC_0-tlast and C_max of prasugrel-AM were significantly greater than those for clopidogrel-AM after a clopidogrel 600-mg LD. The mean C_max and AUC_0-tlast of clopidogrel-AM show a less-than-proportional increase with dose over the clopidogrel dose range of 75- to 600-mg. As clopidogrel dose changes 4-fold from 75- to 300-mg and then doubles from 300- to 600-mg, clopidogrel-AM C_max changes only 2.2-fold and 1.2-fold, respectively, while AUC_0-tlast changes only 2.8-fold and 1.4-fold, respectively. After the 7th MD, the C_max and AUC_0-tlast of prasugrel-AM were 43% and 29% higher (P < 0.01) than the C_max and AUC_0-tlast of clopidogrel-AM during clopidogrel MD. After the seventh MD, the C_max and AUC_0-tlast of clopidogrel-AM were comparable between the clopidogrel 300-mg/75-mg and 600-mg/75-mg regimens.

Table 1

Table 2

The ∼8-hour terminal elimination half-lives of clopidogrel-AM and prasugrel-AM are much shorter than the 24-hour dosing interval; therefore the active metabolites do not accumulate.³⁸ Thus, the LD and MD treatments were analyzed together for the assessment of dose proportionality for each metabolite. The mean C_max and AUC_0-tlast of clopidogrel-AM showed a less-than-proportional increase with dose over the clopidogrel dose range of 75-mg to 600-mg. As the clopidogrel dose changed 8-fold from 75-mg to 600-mg, the clopidogrel-AM AUC_0-tlast only changed 4.4-fold, and C_max changed 2.8-fold. The exposure response to increasing dose was nonlinear, with less of an increase between 300-mg and 600-mg than between 75-mg and 300-mg. Conversely, as the prasugrel dose changed 6-fold from 10- to 60-mg, prasugrel-AM C_max changed 5.9-fold while AUC_0-tlast changed 7.2-fold consistent with dose proportionality.

Relationship Between IPA and Active Metabolite Generation

Panels A and B of Figure 3 show the relationship between the AUC_0-tlast of the active metabolite and the IPA 24 hours after LDs and 24 hours after the 7th MDs, respectively, of prasugrel and clopidogrel. After a LD (Figure 3A), IPA was related to exposure to the active metabolite of either prasugrel or clopidogrel. Across all LDs, IPA rises sharply with increasing AUC_0-tlast, approaching an IPA asymptote that appears to lie between 90% and 100%. Above an AUC_0-tlast of 300 ng·h/mL, greater than 50% IPA was observed, whereas the response was quite variable below an AUC_0-tlast of 300 ng·h/mL. A similar relationship was observed for the MD phase of each treatment sequence (Figure 3B), albeit at lower AUC_0-tlast values. Across the LD and MD phases of each dosing sequence, subjects with the lowest exposure as measured by AUC_0-tlast, also generally had the lowest IPA.

Figure 3

Tolerability

All regimens were well tolerated. Mild spontaneous bruising was more common with the clopidogrel 600-mg/75-mg (7 of 35 subjects; P = 0.035) and prasugrel 60-mg/10-mg (8 of 36 subjects; P = 0.038) than with clopidogrel 300-mg/75-mg (1 of 36 subjects). There was no difference (P = 0.98) between prasugrel 60-mg/10-mg and clopidogrel 600-mg/75-mg. Minor bleeding at venipuncture sites did not differ significantly across the 3 treatment regimens: 8 of 36 subjects (22%) for prasugrel 60-mg/10-mg; 7 of 35 subjects (20%) for clopidogrel 600-mg/75-mg, and 6 of 36 subjects (17%) for clopidogrel 300-mg/75-mg.

DISCUSSION

In the present study, prasugrel 60-mg/10-mg (LD/MD) achieved faster onset, higher levels, and less variability of IPA than did either the clopidogrel 300-mg/75-mg or 600-mg/75-mg dosing regimens. On the basis of previously defined Bayesian criteria, 3% to 36% of the subjects were pharmacodynamic poor responders to the clopidogrel LDs and MD, whereas all subjects were classified as responders to the prasugrel LD and MD.³⁶ This greater antiplatelet activity was related to higher and more consistent exposure to the active metabolite of prasugrel.

The active metabolites of prasugrel and clopidogrel are comparably potent; therefore, higher exposure to either active metabolite will result in greater platelet inhibition up to a maximum effect (E_max).^30,31 However, as demonstrated in the crossover design of this study, consistent and significantly higher exposure to prasugrel-AM, compared to clopidogrel-AM in the same subject, indicates that prasugrel-AM is produced more efficiently than clopidogrel-AM, resulting in the higher IPA shown in Figure 1.^39,40 Compared to clopidogrel, the faster onset of prasugrel's effect on platelet aggregation can be attributed to differences in the metabolic pathways leading to formation of the respective active metabolites. The esterase-mediated step for prasugrel occurs mainly in the intestine, as does the CYP-mediated oxidative step leading to the active metabolite formation. However, hydrolysis of clopidogrel by esterases in the intestine and/or liver leads to formation of an inactive metabolite and conversion of the remaining clopidogrel to its active metabolite requires 2 CYP-mediated steps that occur mainly in the liver.⁴¹ The slower metabolic conversion of clopidogrel to clopidogrel-AM, along with the shunting of a large percentage of a clopidogrel dose to an inactive metabolite, lowers both the rate and extent of clopidogrel-AM formation compared to those of prasgurel-AM.³⁰

The impact of higher clopidogrel loading doses has been a topic of discussion within the cardiovascular literature, and the results of the present study are consistent with and help explain the results of 2 previous studies that also investigated the effect of higher clopidogrel loading doses.^20,22 One is the ALBION study, in which 101 non-ST-elevation acute coronary syndrome patients received clopidogrel LDs of 300-mg, 600-mg, or 900-mg on an aspirin background.²⁰ When compared with a clopidogrel 300-mg LD, the clopidogrel 600-mg and 900-mg LDs achieved significantly greater platelet inhibition, but the IPA after the 900-mg LD was only marginally higher than the IPA after the 600-mg LD. The maximal IPA for all LDs was not reached until 4 to 6 hours after dosing.

The second study, ISAR-CHOICE, used different analytical methods than those used in the present study to measure clopidogrel-AM concentrations, but the results paralleled those in our study.²² Patients in ISAR-CHOICE were allocated to 300-mg, 600-mg, or 900-mg clopidogrel LDs, and plasma concentrations of the active metabolite were measured at baseline and serially. Platelet inhibition with 5 and 20 μM ADP was determined using optical aggregometry at baseline and at 4 hours. A clopidogrel 600-mg LD produced higher plasma concentrations of active metabolite than did the 300-mg LD, with lower MPA at 4 hours after dose, but the 900-mg LD showed no additional effect over the 600-mg dose. The authors of ISAR-CHOICE concluded that because of dose-limited clopidogrel absorption, doses of clopidogrel higher than 600-mg were not associated with significant additional suppression of platelet function.

Pharmacokinetic data from the current study are consistent with those from ALBION and ISAR-CHOICE. The AUC_0-tlast of clopidogrel-AM changed only 3-fold as the clopidogrel dose changed 4-fold from 75- to 300-mg, and increased only 44% as the dose doubled from 300- to 600-mg. The mean C_max of clopidogrel-AM increased only 16% as the clopidogrel dose doubled from 300- to 600-mg. These results indicate that changes in clopidogrel-AM exposure were increasingly less than proportional over the dose range studied, and they suggest saturable absorption of clopidogrel and/or saturable metabolism of clopidogrel to its active metabolite. The results also suggest that exposure to clopidogrel's active metabolite will reach an asymptote beyond which increasing the clopidogrel dose will not produce increased exposure. Therefore, increasing the clopidogrel dose to 900-mg would result in relatively small increase in exposure to clopidogrel-AM, and thus only a marginal increase in the inhibition of platelet aggregation.

The current study also expands the literature regarding the intersubject variability of platelet inhibition with higher clopidogrel LDs.^14-21 Similar to the variability observed in this study, Hochholzer et al reported substantial MPA variability in response to a clopidogrel 600-mg LD in 1001 patients.¹⁴ In the current study, prasugrel in healthy subjects produced a more consistent IPA response than clopidogrel. Several hypotheses have been proposed to account for the interindividual variability in response to clopidogrel, including variation in absorption of the drug, differences in metabolism of the prodrug to the active metabolite, variation in the platelet P2Y₁₂ receptor, and variations in laboratory methodologies.^42,43 The results of this study suggest that absorption and/or metabolic conversion of clopidogrel to its active metabolite is a more important determinant of the level of inhibition of platelet aggregation achieved than any differences in P2Y₁₂ receptor function. As demonstrated by the crossover design of this study, even in subjects with poor response to clopidogrel, platelets were successfully inhibited following administration of prasugrel, thereby minimizing variations in P2Y₁₂ function as a potential cause of clopidogrel response variability.

Low levels of platelet inhibition in patients treated with clopidogrel have been linked to adverse clinical events, despite the use of higher clopidogrel loading doses.^20,44 The ALBION study, while underpowered to show a significant clinical difference, suggested a potential relationship among the clopidogrel LD, the IPA achieved, and patient outcomes.²⁰ In the EXCELSIOR study, patients treated with a clopidogrel 600-mg LD who had platelet aggregation above the median at the time of elective coronary stent placement carried a 6.7-fold risk (95% CI 1.52 to 29.41; P = 0.003) of 30-day major adverse cardiovascular events.⁴⁴

Strengths and Limitations

The current study incorporated a randomized 3-way crossover design in healthy subjects at a single study site; each subject served as her or his own control. Also, this study used robust and validated analytical methods for measuring prasugrel-AM and clopidogrel-AM concentrations in plasma. Approximately 8% of all PK samples were lost to analysis; no treatment group was disproportionally affected with no impact on the derived conclusions regarding active metabolite exposure for prasugrel or clopidogrel. While the study was conducted in healthy subjects not receiving concomitant aspirin, and thus may limit direct application to clinical practice, the IPA results and the pharmacokinetic findings can be clearly attributed to the different responses in the same subject to the clopidogrel and prasugrel LD/MD treatments. Differences between platelet function measurement and methods used to measure clopidogrel-AM plasma concentrations limit direct comparison across studies, but they do not limit the ability to compare results between subjects within each of these studies.

In addition, the current study did not evaluate collagen- and TRAP-induced platelet aggregation. Previous observations in aspirin-treated subjects demonstrated that, as with ADP, partial inhibition of collagen-induced and TRAP-induced aggregation was significantly greater after a prasugrel 60-mg LD than after a clopidogrel 300-mg LD.⁴⁵ In the current study, we therefore chose to focus on a range of concentrations of the specific ligand (ADP) of the target receptor (P2Y₁₂). Finally, at present, a unified standard for defining pharmacodynamic poor response to platelet inhibition in patients treated with thienopyridines and its relationship to clinical outcome has not yet been developed; our model-based methodology for defining poor responders must be carefully considered among the variety of definitions of clopidogrel response variability reported in the literature.

CONCLUSIONS

In a direct crossover comparison in healthy subjects, prasugrel provided faster, higher, and more consistent IPA compared to either the approved 300-mg or a 600-mg high-dose clopidogrel LDs. Greater IPA with less response variability was also observed during prasugrel MD administration. The results further indicated that the exposure to clopidogrel active metabolite is limited at higher doses due to saturable absorption and/or metabolism to clopidogrel's active metabolite. The antiplatelet profile for prasugrel compared with clopidogrel appears to be directly related to higher exposure to prasugrel's active metabolite. Whether prasugrel will demonstrate an improved benefit-risk profile for patients is the subject of the ongoing Phase 3 clinical trial comparing prasugrel (60-mg/10-mg) to clopidogrel (300-mg/75-mg) in patients with acute coronary syndromes undergoing PCI: Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel (TRITON) - TIMI 38.³²

ACKNOWLEDGMENTS

We thank Dr. Jennie Francis, MD, Principal Investigator at the Lilly Laboratories, for Clinical Research, IUMC, for screening, enrollment, and monitoring of subjects. We thank Kendra E. Jones, MS, Xiaomei Peng, MS, and Yu Li, MS, for statistical analysis. We thank Kenneth E. Robertson, PharmD, RPh, for preparing the first and subsequent drafts of the manuscript. Finally, we thank Julie A. Sherman, AAS, for editorial assistance with the manuscript.

REFERENCES