The effect of elinogrel on high platelet reactivity during dual antiplatelet therapy and the relation to cyp 2c19*2 genotype: first experience in patients


Paul A Gurbel, Sinai Hospital of Baltimore, Cardiac Catheterization Laboratory, 2401 W. Belvedere Ave, Baltimore, MD 21215, USA.
Tel.: +1 410 601 9600; fax: +1 410 601 9601.


Summary.  To study the effect of a new direct acting reversible P2Y12 inhibitor, elinogrel (PRT060128), and the relation to cytochrome P450 (CYP) polymorphisms in patients with high platelet reactivity (HPR) on standard dual antiplatelet therapy. Methods and Results: We studied the pharmacodynamic and pharmacokinetic effects of a single 60-mg oral dose of elinogrel in 20 of 45 previously stented stable patients with HPR. We also genotyped for CYP2C19*2,3,5,17 and CYP3A5*3. Platelet reactivity fell within 4 h of dosing, the earliest time point evaluated as measured by the following assays: maximum 5 and 10 μm ADP LTA (P < 0.001 for both vs. predosing); maximum 20 μm ADP LTA (P < 0.05); VerifyNow (P < 0.001); thrombelastography (P < 0.05); VASP phosphorylation (P < 0.01); and perfusion chamber assay (P < 0.05); this was reversible within 24 h in these same assays (P = ns vs. predosing for all assays). CYP2C19*2 was present in 44% of all patients but was more frequent in HPR patients (77% vs. 16%, P = 0.0004). Conclusions: HPR is reversibly overcome by a single 60-mg oral dose of elinogrel, a drug now being investigated in a phase 2 trial. CYP2C19*2 was associated with HPR during conventional dual antiplatelet therapy.


P2Y12 receptor inhibition by clopidogrel, together with aspirin therapy, is a major treatment strategy in patients with high risk coronary artery disease. Currently, the only available P2Y12 inhibitors are the thienopyridines, clopidogrel, ticlopidine and prasugrel has been recently approved. Clopidogrel is routinely administered as a daily 75-mg dose along with 81–325 mg aspirin in patients treated with coronary artery stents and after acute coronary syndromes, and in stable patients with prior myocardial infarction, or established peripheral arterial disease [1].

Pharmacodynamic studies have demonstrated wide variability in response to clopidogrel therapy and a substantial percentage of patients exhibit non-responsiveness and high on-treatment platelet function in stented patients treated with clopidogrel and aspirin, whereas translational research studies have produced concordant results demonstrating that high on-treatment platelet reactivity (HPR) to adenosine diphosphate (ADP) is a risk factor for recurrent ischemic events [2–7]. High on-treatment platelet reactivity has been associated with specific polymorphisms of cytochrome P450 (CYP) genes, especially the common CYP 2C19*2 variant that has also been associated with an increased risk for post-stenting ischemic events [8–12]. Therefore, a major goal of ongoing antiplatelet therapy research is to develop an agent associated with a reliable effect that can overcome high platelet reactivity in the high-risk patient and not be influenced by CYP gene polymorphisms. Recently, prasugrel, a new thienopyridine that exhibited greater platelet inhibition than clopidogrel in phase II studies, was also associated with lower ischemic event occurrence in a major phase III acute coronary syndrome trial, but with a concurrent increase in bleeding risk [13,14]. Moreover, a new oral direct acting P2Y12 inhibitor, ticagrelor, is also associated with greater platelet inhibition compared with clopidogrel and is undergoing comparison with clopidogrel in a phase III investigation [15]. Elinogrel (Portola Pharmaceuticals Inc., South San Francisco, CA, USA) is an investigational, direct acting (non-prodrug) reversible P2Y12 receptor inhibitor that has both oral and parenteral formulations. Phase I single ascending dose studies in healthy subjects have demonstrated high levels of inhibition of ADP-induced platelet aggregation following either single oral doses or after intravenous bolus administration [16,17].

The aim of the current study was to investigate the antiplatelet effect of a single 60 mg oral dose of elinogrel administered to stented patients screened for high platelet reactivity (HPR) to ADP during standard dose clopidogrel and aspirin therapy and examine the relation of platelet reactivity to specific cytochrome P450 (CYP) polymorphisms.


Patients and study design

This is a single center study conducted at Sinai Hospital of Baltimore between 1 February 2008 and 15 September 2008. The study was approved by the Investigational Review Board. Inclusion criteria were clinically stable patients older than 18 years of age who had undergone previous coronary artery stenting and were treated with chronic daily 75 mg clopidogrel and 81 mg aspirin therapy. Fifty patients were screened for HPR (five patients with HPR had been previously identified at our center). The definition of HPR was based on our previous investigation in patients on maintenance clopidogrel and aspirin therapy that identified the upper tertile of 5 μm ADP-induced platelet aggregation (≥ 43%). In that study HPR was a marker of post-stenting ischemic risk [18]. Patients were instructed to take clopidogrel at least 12 h but no more than 16 h prior to the initial screening visit (Fig. 1). Exclusion criteria were: any history of a bleeding diathesis or gastrointestinal bleeding; stroke or transient ischemic attack of any etiology within 30 days of screening; illicit drug or alcohol abuse; consumption of grapefruit or grapefruit juice 48 h prior to dosing; coagulopathy; major surgery within 6 weeks prior to screening; planned surgical procedure within 30 days of anticipated dosing; enrollment in an investigational drug study within 30 days of screening; a medical or surgical condition that may impair drug absorption or metabolism; platelet count < 100 000 mm−3; hematocrit < 30%; creatinine > 2mg dL−1; or current use of non-steroidal anti-inflammatory drugs, anticoagulants, dietary supplements, herbal products or antiplatelet drugs other than aspirin or clopidogrel within 2 weeks of screening.

Figure 1.

 Study design. HPR, high platelet reactivity; ASA, aspirin; ADP, adenosine diphosphate; qd, once daily.

Seven to 14 days after the screening visit patients underwent platelet function testing and were administered a single 60-mg oral dose of elinogrel between 12 and 16 h after the previous day’s dose of clopidogrel. All subjects were instructed to continue their maintenance aspirin and clopidogrel therapy for the duration of the investigation. Throughout the study, patients continued their daily therapy of clopidogrel and aspirin and compliance was assessed. Follow-up visits were at 24 h and 7 days post-dosing with elinogrel.

Elinogrel (PRT060128) drug product

Elinogrel (PRT060128 potassium) was supplied by Portola Pharmaceuticals, Inc. as a powder and was reconstituted with sterile water by an in-hospital pharmacist as a solution containing 2.28 mg mL−1. The solution was stored at room temperature to be used within 24 h of preparation; 30 mL (60 mg) were administered orally to patients together with 100 mL of water.

Blood and urine collection

Blood was collected from the antecubital vein with an 18 gauge needle at screening and at predosing 12–16 h after the previous day’s clopidogrel dose and then at 4 h, 6 h, 24 h and 7–10 days after dosing with elinogrel. Three Vacutainer® tubes (Becton-Dickinson, Franklin Lakes, NJ, USA) containing 3.2% trisodium citrate were used for LTA (5 and 20 μm ADP and 2 mm arachidonic acid), thrombelastography (TEG) and vasodilator-stimulated phosphoprotein (VASP) phosphorylation measurements; two tubes containing a proprietary anticoagulant CT921-78 (factor Xa inhibitor; Portola Pharmaceuticals) were used for LTA (10 μm ADP and 4 μg mL−1 collagen) and for the perfusion chamber assay (PCA) to ensure physiologic calcium concentrations; one tube containing lithium heparin (Becton-Dickinson) was used for TEG PlateletMapping; and one tube containing 3.2% sodium citrate (Greiner Bio-One Vacuette® North America, Inc., Monroe, NC, USA) was used for the VerifyNow™ assay. One tube containing ethylene diamine tetraacetate (1.8 mg mL−1) was used for elinogrel plasma concentration measurements (Becton-Dickinson). Urine samples were collected at screening and on the day of dosing prior to administration of elinogrel to test for pregnancy in women of childbearing potential and to assess for amphetamine, methamphetamine, cocaine, opiate and tetrahydrocannibinol (THC) use.

Light transmittance aggregometry

Platelet aggregation was assessed as previously described using a Chronolog Optical Aggregometer (Model 490-4D) with the Aggrolink software package (Chronolog, Havertown, PA, USA) [2]. Maximum aggregation was expressed as the maximum percent change in light transmittance from baseline in platelet-rich plasma, using platelet-poor plasma as a reference. Final aggregation was measured at 6 min after the addition of an agonist.

VerifyNow™ P2Y12 assay

The VerifyNow™ is a turbidimetric-based optical detection assay designed to measure platelet aggregation that is based upon the ability of activated platelets to bind to fibrinogen. Light transmittance increases as activated platelets bind and aggregate fibrinogen-coated beads. The change in optical signal is reported as P2Y12 Reaction Units (PRU).

Thrombelastograph® (TEG) Hemostasis System with PlateletMapping™

The TEG Hemostasis Analyzer with PlateletMapping assay (Haemoscope Corporation, Niles, IL, USA) relies on the measurement of platelet-fibrin clot strength to enable a quantitative analysis of platelet function [4]. In heparinized blood, reptilase and factor XIIIa are used to generate a cross-linked fibrin clot. The contribution of the P2Y12 receptor to platelet-fibrin clot strength is measured by the addition of ADP and is expressed as MAADP (mm).

Vasodilator stimulated phosphoprotein phosphorylation

VASP phosphorylation was determined in whole blood using a flow cytometric assay [Platelet VASP; Diagnostica Stago (Biocytex), Asnieres, France] as previously described [5]. The platelet reactivity index (PRI) was calculated from the mean fluorescence intensity (MFI) according to the formula: PRI = [(MFI(PGE1)−MFI(PGE1+ADP))/MFI(PGE1)] × 100%.

Capillary perfusion chamber preparation and real-time thrombosis profiler (RTTP)-2

Rectangular capillaries with 0.2 × 2 mm sections (Vitrocom, Mountain Lakes, NY, USA) were coated with human type III fibrillar collagen (Chronolog Corp) as previously described [19]. Evaluation of thrombotic deposits was performed at 8 mm from the proximal end of the capillary. Experiments were completed within 1 h of blood sampling.

The RTTP-2 consists of an epifluorescence microscope to monitor thrombus formation and a syringe pump (Harvard Apparatus, Holliston, MA, USA) to establish the desired flow and wall shear rate in the capillary perfusion chamber. Platelets were labeled by incubating rhodamine 6G (final concentration 1.25 μg mL−1; EMD, Gibbstown, NJ, USA) in whole blood at 37 °C for 15 min. A high-power light emitting diode with a spectral maximum at 530 nm and a spectral half width of 35 nm (Luxeon V; Lumileds Lighting, San Jose, CA, USA) excited the dye. Excitation and emission light were filtered with a set of fluorescence filters (31002; Chroma Technologies, Rockingham, VT, USA). A microscope objective images an area of 360 × 270 μm on the internal wall of the capillary onto a Sony XCD X-710 digital camera (resulting magnification c. 13 ×). Images were recorded at a frequency of 1 Hz. A personal computer with custom software was used to control the camera and the syringe pump, to display experimental conditions, and record images.

Thrombus size was represented as the measurement of the fluorescence intensity divided by total area. Segmentation, partitioning of an image into non-overlapping regions, was accomplished based on a method proposed by Otsu [20]. A watershed algorithm was applied to identify individual thrombi in the image [21]. Once the image was segmented, total object volume, area and perimeter were computed. Total volume was computed as sum of intensity values of pixels inside the foreground objects. Total area was computed as number of pixels inside the foreground objects. Data were expressed as fluorescence intensity (pixels)/total area (μm2).

Analysis of plasma concentration

Blood samples were centrifuged and plasma was harvested and stored at −20 °C until analysis. Plasma was extracted with acetonitrile (protein precipitation) and analyzed for elinogrel concentration using liquid chromatography followed by tandem mass spectrometry. The analytical range was 0.500–500 ng mL−1.

Genotype analysis

Genotyping of the known common loss of function CYP2C19*2 variant (rs4244285), as well as other functional variants of CYP2C19 [*3 (rs4986893), *5 (rs56337013), *17 (rs12248560)] and CYP3A5*3 (rs776746), was performed using TaqMan® SNP genotyping assays (Applied Biosystems, Foster City, CA, USA).

Adverse events

Adverse events (AE) were defined as any untoward medical occurrence in a subject that may or may not have been due to treatment with elinogrel, including any unfavorable or unintended sign, symptom or disease temporally associated with the use of the study drug, whether or not it was considered study-drug related. This included any newly occurring event or previous condition that had increased in severity or frequency since the administration of the study drug.

Serious adverse events (SAE) were defined as the occurrence of any event regardless of causality that resulted in death, was life-threatening, required inpatient hospitalization, resulted in persistent or significant disability/incapacity, or was an important medical event (defined as an event that might not have resulted in death, or been life-threatening or required hospitalization, but could be considered an SAE if, based upon appropriate medical judgment, it could have jeopardized the subject and required medical or surgical intervention).

Statistical analysis and sample size determination

Categorical variables were expressed as absolute numbers and percentages and compared using the chi-square test. Continuous variables were expressed as mean ± SD. Overall comparisons of platelet function measurements were performed by one-way repeated measures anova. Comparisons between predosing and postdosing platelet function measurements were performed by one-way repeated measures anova followed by post-hoc Bonferroni multiple comparisons t-test or Kruskal–Wallis anova on ranks followed by post-hoc Dunn’s multiple comparisons method as appropriate based on normality test and equal variance test. Platelet function measurements at screening in screen-failure and elinogrel-treated patients were compared by t-test; P < 0.05 was considered significant (SigmaStat software, Point Richmond, CA, USA). Correlation analysis of 5 μm ADP-induced platelet aggregation (maximum) vs. P2Y12 reaction units (VerifyNow), PRI (VASP-P), MA-ADP (TEG) and fluorescence intensity (pixels)/total area (μm2) (RTTP-2) was performed by Pearson’s product moment correlation coefficient. The relationship between plasma concentration and the ADP-induced maximum aggregation at the ADP concentrations of 5-, 10- and 20-μm was evaluated using WinNonlin program (version 5.2; Pharsight Corp., Cary, NC, USA). Platelet reactivity measurements were compared in *2 and *17 genotype groups by one-way anova.

The primary objective was to evaluate the change in 5 μm ADP-induced aggregation from pre- to 4 h post-elinogrel treatment in patients with HPR currently on clopidogrel therapy. In order to detect an absolute change (mean of upper tertile minus lower tertile cut-point) of 20% (52–32%) in 5 μm ADP-induced platelet aggregation with a standard deviation of 10, a sample size of approximately 10 subjects is required to give a 95% power with an alpha of 0.05 [18].

The authors at the Sinai Center for Thrombosis Research had full access to and take full responsibility for the integrity of the data. The sponsor did not have any role in the analysis of the assays, performance of the statistical analysis, or interpretation of the results. Portola provided the drug, as well as expertise and equipment for the proprietary RTTP-2 assay. All authors have read and agreed to the manuscript as written.


Patients and demographics

Fifty patients were screened; five patients were excluded at screening. Screening platelet function analyses were performed in 45 patients; 20 patients had HPR (Fig. 1). Blood sampling was complete for all time points except for two patients who did not return for the follow-up visit (7–10 days). Compliance with antiplatelet therapy was 100% based on patient confirmation. Demographics are shown in Table 1. Most patients were Caucasian and cardiovascular risk factors were common. Patients with HPR were more often diabetic, and had a greater incidence of prior myocardial infarction, coronary intervention, restenosis and cerebrovascular accident than screen-failure patients. Elinogrel was well tolerated in all patients and there were no adverse events.

Table 1.   Patients demographics
 Elinogrel-treated patients (n = 20)Screen-failure patients (n = 25)P-value
Age (years)61 ± 1265 ± 120.22
Gender n, (male %)13 (65)11 (44)0.08
Caucasians n, (%) 14 (70)18 (72)0.44
African-Americans n, (%)6 (30)7 (28)0.44
Body mass index (kg m−2)32 ± 629 ± 60.11
Systolic blood pressure (mm Hg)128 ± 16133 ± 740.77
Diastolic blood pressure (mm Hg)71 ± 1274 ± 160.49
Medical history
 History of smoking n, (%)7 (35)8 (32)0.42
 Current smokers n, (%)0 (0)3 (12)0.06
 Hypertension n, (%)17 (85)19 (76)0.23
 Hyperlipidemia n, (%)16 (80)17 (68)0.19
 Diabetes n, (%)12 (60)5 (20)< 0.005
 Myocardial infarction n, (%)12 (60)5 (20)< 0.005
 Coronary artery bypass graft n, (%)7 (35)6 (24)0.21
 Percutaneous transluminal coronary angioplasty n, (%)17 (85)14 (56)0.02
 History of restenosis n, (%)14 (83)2 (8)< 0.0001
 Chronic heart failure n, (%)3 (15)3 (12)0.39
 Transient ischemic attack n, (%)0 (0)3 (12)0.06
 Cerebrovascular accident n, (%)3 (15)0 (0)0.03
 Peripheral arterial disease n, (%)2 (10)1 (4)0.21
 Family history of coronary artery disease n, (%)3 (70)20 (80)0.22
Baseline medications
 Beta-blocker n, (%)17 (85)19 (76)0.23
 Nitrate n, (%)6 (30)1 (4)0.01
 Angiotensin converting enzyme inhibitor n, (%)10 (50)14 (56)0.35
 Calcium blockers n, (%)5 (25)7 (28)0.41
 Lipid-lowering agent n, (%)17 (85)20 (80)0.33
 Angiotensin receptor blocker n, (%)60 (30)0 (0)< 0.005
 Diuretic n, (%)6 (30)9 (36)0.34
 Proton pump inhibitor n, (%)4 (20)8 (32)0.19
 Insulin n, (%) 7 (33)0 (0)< 0.005
Laboratory data
 White blood cells × 1000 mm−36.6 ± 1.97.2 ± 1.90.29
 Platelets × 1000 mm−3230 ± 82212 ± 440.35
 Hemoglobin g dL−113.2 ± 1.612.7 ± 1.70.32
 Creatinine mg dL−11.1 ± 0.40.9 ± 0.30.12

Platelet function analyses

As defined by the inclusion criteria, patients selected for elinogrel treatment at the screening visit had higher maximum platelet aggregation induced by 5, 10 and 20 μm ADP than screen-failure patients (53 ± 10% vs. 33 ± 14%, 51 ± 13% vs. 37 ± 13%, 63 ± 11% vs. 44 ± 15%, respectively; P < 0.001 for all measurements). Arachidonic acid-induced aggregation was 8 ± 19% in elinogrel-treated patients and is indicative of aspirin responsiveness. Measurement of ADP-induced platelet aggregation at the predosing time demonstrated that the HPR phenotype was stable for 7–14 days (P = ns vs. screening for all concentrations of ADP). There was a significant change in platelet function over time after elinogrel administration as measured by all methods (P ≤ 0.006).

ADP-induced platelet aggregation (LTA)

The individual patients’ response to elinogrel measured by 5 μm ADP-induced platelet aggregation is shown in Fig. 2. The mean absolute decrease in maximum platelet aggregation in response to 5 μm ADP in elinogrel-treated patients was ∼22% at 4 h postdosing and was also significantly reduced at 6 h. The mean extent of maximum aggregation at 4 h induced by 5 μm ADP met the predetermined endpoint of dropping from the highest to lowest tertile (< 32% maximum aggregation). The majority of patients had a significant decrease in platelet aggregation at 4 or 6 h, relative to their baseline value, and most of the patients had a stable HPR phenotype that persisted from the screening visit to the predose visit, and was reproduced at the follow-up visit, despite ongoing dual antiplatelet therapy. The antiplatelet effect of elinogrel was reversible within 24 h (Figs 3–5).

Figure 2.

 Individual patients’ response to elinogrel as measured by 5 μm ADP-induced aggregation. Thick line indicates the mean and dotted line represents the 43% cut-off value used for HPR.

Figure 3–10.

 Pharmacodynamic response to a single oral 60-mg elinogrel dose in patients with high platelet reactivity as measured by light transmittance aggregometry (5,10 and 20 μm ADP-induced aggregation (Figs 3–5)), 4 μg mL−1 collagen-induced aggregation (Fig. 6); ADP-induced platelet-fibrin clot strength (MA-ADP, Fig. 7); VerifyNow P2Y12 assay (Fig. 8); vasodilator stimulated phosphoprotein phosphorylation assay (Fig. 9); and real-time thrombosis profiler assay (Fig. 10). P-values are relative to mean value at predose measurement.

Collagen-induced aggregation (LTA)

There was a non-significant decrease in collagen aggregation at 4 and 6 h postdosing (Fig. 6).

Thrombelastography (TEG) platelet mapping assay

The rapid and reversible antiplatelet effect of elinogrel was also demonstrated by thrombelastography; MAADP decreased at 4 and 6 h postdosing and at 24 h postdosing did not differ from predosing values (Fig. 7).

VerifyNow P2Y12 receptor assay

PRU also decreased at 4 and 6 h postdosing and the effect was reversible at 24 h postdosing (Fig. 8).

Vasodilator stimulated phosphoprotein (VASP)

VASP PRI significantly decreased from baseline at 4 and 6 h post-treatment and again the effect was reversible at 24 h postdosing (Fig. 9)

Real-time thrombosis profiler (RTTP)-2

Concordant with the other platelet function analyses, there was a significant decrease in thrombus size [fluorescence intensity/area (μm2)] at 4 and 6 h postdosing and the inhibitory effect on thrombus size was reversible at 24 h (Fig. 10).

Pharmacokinetics/pharmacodynamics relationship

The mean (±SD) observed maximum plasma concentration of elinogrel achieved at the median Tmax (4 h) was 2459 (± 1460) ng mL−1 following the administration of 60 mg elinogrel (Fig. 11). The relationship between plasma concentration and the ADP-induced maximum aggregation could be described by an inhibitory effect Emax model. The IC50 (inhibitory constant) increased with increasing ADP concentrations, as expected for a competitive, reversible P2Y12 antagonist such as elinogrel. The IC50 for the 5-, 10- and 20 μm ADP-induced aggregation was 2230, 2412 and 5852 ng mL−1, respectively. As seen from the mean PK curve following a single dose of elinogrel (Fig. 11), the time course of peak plasma concentration was 4–6 h, followed by a decrease in the plasma concentration to negligible levels at 24 h. The time course of inhibition observed in the various PD assays reflected the time course of the PK of elinogrel, in that maximum inhibition was observed at 4–6 h followed by a return to baseline levels at 24 h.

Figure 11.

 Pharmacokinetic response to single oral 60-mg elinogrel dose in patients with high platelet reactivity. This plot represents the mean plasma concentration achieved at 4, 6 and 24 h.

Correlation between pharmacodynamic measurements

Platelet aggregation (5-μm ADP, maximum) was significantly correlated with other pharmacodynamic parameters. Besides platelet aggregation measurements, the strongest correlation was observed between 5 μm ADP-induced platelet aggregation (maximum) and P2Y12 reaction units (VerifyNow) (Table 2).

Table 2.   Correlation between 5 μm ADP-induced platelet aggregation (maximum) and other pharmacodynamic measurements
 Correlation coefficientP-value
Light transmittance aggregometry (%)
 5 μm ADP (final)0.863< 0.0001
 10 μm ADP (maximum)0.736< 0.0001
 20 μm ADP (maximum)0.892< 0.0001
 2 μg mL−1 Collagen (maximum)0.373< .0.001
 VerifyNow P2Y12 assay-platelet reactivity units0.720< 0.0001
 VASP assay-platelet reactivity index0.601< 0.0001
 TEG-ADP-induced platelet-fibrin clot strength (mm)0.470< 0.0001
 RTTP-2 (fluorescence intensity (pixels)/total area (μm2).0.466< 0.0001

Genotype analysis and relation to platelet function

Seventeen patients with HPR and 19 without HPR consented to genetic screening. The results are shown in Table 2. At least one CYP2C19*2 allele was present in 44% of all patients genotyped: the *2 allele frequency was higher in HPR patients than in patients without HPR (13/17 (77%) vs. 3/19 (16%), P = 0.0004). All patients with HPR who were CYP2C19*2 negative were diabetic and 3/4 were CYP2C19*17 positive. The presence of CYP2C19*17 carriers was numerically less in HPR patients (29% vs. 53%, P = 0.08). The CYP 2C19*3 and *5 and CYP 3A5*3 alleles were not present in any patient. Platelet reactivity was higher in CYP2C19*2 carriers relative to non-carriers during clopidogrel and aspirin therapy alone (Table 3). Platelet reactivity fell in CYP2C19*2 carriers and non-carriers after elinogrel administration (Fig. 12).

Table 3.   Pre-elinogrel platelet function in relation to genotype
 Patients with CYP2C19*2 Allele (n = 17)Patients without CYP2C19*2 Allele (n = 19)P-value
5 μm ADP-induced platelet aggregation (%)52 ± 1336 ± 140.001
20 μm ADP-induced platelet aggregation (%)61 ± 1250 ± 130.03
VASP assay-platelet reactivity index72 ± 1258 ± 200.05
VerifyNow P2Y12 assay-platelet reactivity units259 ± 68214 ± 820.14
VerifyNow P2Y12 assay-per cent inhibition19 ± 1935 ± 240.07
TEG-ADP-induced platelet-fibrin clot strength (mm)57 ± 849 ± 120.18
TEG-per cent inhibition23 ± 936 ± 200.07
 Patients with CYP2C19*17 allele (n = 15)Patients without CYP2C19*17 allele (n = 21)P-value
  1. ADP, adenosine diphosphate; VASP, vasodilator stimulated phosphoprotein; TEG, thrombelastography.

5 μm ADP-induced platelet aggregation (%)41 ± 1945 ± 140.30
20 μm ADP-induced platelet aggregation (%)51 ± 1458 ± 130.18
VASP assay-platelet reactivity index60 ± 2367 ± 150.35
VerifyNow P2Y12 assay-platelet Reactivity units206 ± 91251 ± 650.13
VerifyNow P2Y12 assay-per cent inhibition35 ± 2824 ± 180.24
TEG-ADP-induced platelet-fibrin clot strength (mm)51 ± 1253 v 110.83
TEG-per cent inhibition35 ± 2824 ± 180.17
Figure 12.

 Five micromolar ADP-induced aggregation in individual patients measured pre- and 4 h postelinogrel treatment and CYP2C19*2 allele status. HPR, high platelet reactivity.


The present study demonstrated that HPR accompanying standard maintenance clopidogrel and aspirin therapy in stable patients who had undergone prior coronary stenting is reversibly overcome by a single 60-mg oral dose of elinogrel. The additive antiplatelet effect of elinogrel was demonstrated by concordant results from multiple assays that measure P2Y12 receptor reactivity. The pharmacodynamic effect followed a concordant pharmacokinetic response. Platelet inhibition occurred within 4 h of drug administration, the earliest time point evaluated, and corresponded to maximal plasma concentrations of the drug, and platelet function returned to predose levels within 24 h when drug levels were minimal as indicated by plasma concentrations of 200–300 ng mL−1. Our observations correspond to the time of maximum drug concentration of elinogrel observed in previous studies in healthy subjects [22]. The CYP2C19*2 allele was associated with HPR observed during clopidogrel and aspirin therapy. However, a single 60-mg oral dose of elinogrel overcame HPR in patients who were either wild type or who had at least one CYP2C19*2 allele. The current study is the first pharmacokinetic and pharmacodynamic evaluation of oral elinogrel in patients with coronary artery disease, the first investigation of platelet reactivity in patients treated simultaneously with a thienopyridine and a reversible oral P2Y12 inhibitor, and the first study specifically designed to evaluate antiplatelet therapy responsiveness in patients with HPR during standard maintenance dual antiplatelet therapy, an established cardiovascular risk factor.

Elinogrel is a direct-acting, non-prodrug, competitive inhibitor of the P2Y12 receptor available in both oral and parenteral formulations. It has a terminal half-life of approximately 12 h, is cleared by both renal and hepatic routes and undergoes limited metabolism. When given as an intravenous bolus, immediate and full platelet inhibition of ADP-induced platelet aggregation was observed [17]. Moreover, intravenous bolus doses up to 60 mg administered concurrently with standard therapy were well tolerated in patients undergoing primary angioplasty for ST-elevated myocardial infarction (data on file).

The effect of a clopidogrel loading dose in patients with HPR during clopidogrel maintenance therapy is unknown. Additional platelet inhibition has been demonstrated in patients treated with chronic maintenance clopidogrel and aspirin therapy by administration of a loading dose of clopidogrel [23]. Our study was designed to evaluate the antiplatelet effect of a single elinogrel dose on platelet function in patients with HPR on standard maintenance dose clopidogrel. Bonello et al. [24] demonstrated additional platelet inhibition by measuring VASP phosphorylation levels in patients with high platelet reactivity who were repeatedly treated with 600 mg clopidogrel up to a total dose of 2400 mg over 3 days.

The third generation thienopyridine, prasugrel, has addressed the limitations of delayed and variable inhibition by clopidogrel. However, irreversible platelet inhibition often precludes prepercutaneous intervention administration of thienopyridines until the coronary anatomy is known. A greater frequency of bleeding was observed with prasugrel vs. clopidogrel, with a 4-fold increase seen in patients undergoing coronary bypass graft surgery as well as increased bleeding during chronic prasugrel administration [14]. However, emerging data with direct-acting, non-thienopyridine, reversible agents such as elinogrel demonstrate that these agents can achieve immediate high-level blockade following parenteral bolus administration, and may have a broader therapeutic index than the irreversible thienopyridine inhibitors, in that they have less effect on hemostasis at equivalent levels of antithrombotic activity in preclinical studies [25]. Current observations of a readily reversible pharmacodynamic effect suggest that CABG bleeding may be reduced if performed 24 h after a single 60-mg dose of elinogrel as compared with surgery performed at the same time after cessation of thienopyridine therapy. The reversibility after an intravenous dose may be even more rapid due to a < 1 h distribution half-life after an intravenous bolus. Moreover, in patients undergoing surgery, the hiatus in antiplatelet therapy may be shortened considerably with a reversible agent such as elinogrel compared with the present guidelines recommending withholding clopidogrel therapy for 5–7 days [26]. Thus, a rapid-acting, reversible agent would reduce bleeding risk in both the chronic and acute settings, and allow for easier management of surgical procedures for patients with drug eluting stents who require dual antiplatelet therapy.

The pharmacokinetic and pharmacodynamic variability observed with elinogrel was likely influenced by two factors. First, the simple liquid formulation used (drug powder dissolved in water) in this study was not optimized and may have affected the level of gut absorption. In contrast, an immediate release tablet formulation is being studied in the ongoing phase II Randomized, Double-Blind, Active-Controlled Trial to Evaluate Intravenous and Oral PRT060128, a Selective and Reversible P2Y12-Receptor Inhibitor, vs. Clopidogrel, as a Novel Antiplatelet Therapy in Patients Undergoing Non-Urgent Percutaneous Coronary Intervention (INNOVATE-PCI) trial. The 60-mg dose in the present study achieved plasma concentrations that are at the low end of the range being studied in the INNOVATE-PCI trial (50, 100 and 150 mg twice daily). Also, the protocol mandated that the drug be administered to the patient while fasting, as food may also decrease drug exposure. However, 60% of the patients were diabetic and we could not withhold food for a prolonged period. Thus, some patients were allowed to eat a snack after drug administration that could also have an impact on drug levels in those patients and increase the variability of plasma concentrations. Second, because only a single dose of drug was administered in this ‘proof of concept’ study, drug levels did not reach steady state (as the half-life of elinogrel is ∼12 h, it would require > 2 days of b.i.d. dosing to reach steady state). It is anticipated that the plasma concentrations achieved with a traditional formulation method such as the immediate release form presently being studied in Phase II, as well as the fact that achieving steady state will reduce overall PK variability, will result in improved consistency of the PK (and PD) relative to the formulation used in the current study.

The combination of intravenous and oral elinogrel will allow for a seamless transition from the acute to the chronic setting, and avoids the issues of transitioning from an intravenous reversible inhibitor such as cangrelor, to a thienopyridine prodrug, where the presence of the competitive reversible inhibitor has been shown to block the ability of the active metabolite of clopidogrel or prasugrel to inhibit platelets by irreversibly binding the P2Y12 receptor [27,28].

Numerous studies have focused on the influence of genetic polymorphisms on clopidogrel response variability, especially genes encoding CYP P450 isoenzymes [8–12,29]. The influence of the CYP3A5 non-expressor (*3 allele) genotype on the antiplatelet effect of clopidogrel and its relation to post-PCI ischemic event occurrence was previously demonstrated [29]. This allele was not present in our study population. Recent clinical trials have demonstrated that patients carrying at least one CYP2C19*2 allele, present in 25–50% of various ethnic populations and present in 44% of our patients, are less inhibited by clopidogrel and accordingly experience increased clinical events [8–12]. In the present study, we demonstrated the association of the CYP2C19*2 allele with the HPR phenotype in patients treated with clopidogrel and aspirin. The other loss of function alleles, CYP2C19*3 and *5, were not present in our patients. In line with its direct inhibition of the P2Y12 receptor and lack of requirement for metabolism by CYP P450 isoenzymes for activity, the antiplatelet effect of elinogrel was observed irrespective of CYP2C19*2 status. The increased function allele, CYP2C19*17, was frequent in our study and patients with HPR had a numerically lower incidence of this allele.


The current study focused only on stented patients with HPR on chronic clopidogrel and aspirin therapy. The effect of a 60-mg dose in other patient groups cannot be extrapolated from these data. Moreover, we did not evaluate the effect of a 60-mg dose in the absence of clopidogrel therapy. Finally, only one dose was studied. The low baseline collagen-induced aggregation in patients treated with chronic aspirin is consistent with previous investigations and may have limited our ability to observe a significant effect of elinogrel at the current dose [30]. Higher elinogrel dosing may have further lowered platelet reactivity to all agonists. A larger study will be necessary to confirm our observations regarding the relation of CYP 2C19*2 incidence and HPR, and the response to elinogrel in CYP 2C19*2 carriers.


Elinogrel overcomes HPR in the majority of patients on clopidogrel and aspirin therapy within 4 h of dosing and its effect is fully reversible within 24 h. Our data support the association of the CYP 2C19*2 allele with HPR observed during conventional dual antiplatelet therapy. The pharmacodynamic properties of elinogrel observed in the current investigation are being further studied in an ongoing phase 2 trial.


P.A. Gurbel, K.P. Bliden, U.S. Tantry: contributed to concept and design, enrollment of subjects, data acquisition and analysis, handled funding and supervision, drafted manuscript and made critical revisions of manuscript. M.J. Antonino, R.E Pakyz: data acquisition and analysis. A.R. Shuldiner: contributed to concept and design, drafted manuscript and made critical revisions of manuscript. P.B. Conley: contributed to the concept and design, interpretation of data and revising the intellectual content. G. Stephens: contributed technical expertise and training on the RTTP assay. M.M. Jurek, D.D. Gretler: contributed to study management and critical revision.


We are grateful to P. Andre for valuable advice on the RTTP assay, to A. Hutchaleelaha for advice on interpretation of pharmacokinetic data and A. Singla, T. Saurez and B. Kanda for clinical evaluation of patients.

Disclosure of conflict of interests

The study was supported by Portola Pharmaceuticals Inc., South San Francisco, CA, USA. P.A. Gurbel has received research funding from Astra Zeneca, Daiichi Sankyo, Lilly, Schering-Plough, Pozen, Portola Pharmaceuticals, Bayer Healthcare, Sanofi-Aventis and Haemoscope. P.A. Gurbel has received honoraria and consulting fees from Astra Zeneca, Daiichi Sankyo, Lilly, Schering-Plough, Pozen, Portola Pharmaceuticals, Bayer Healthcare and Sanofi-Aventis. P.B. Conley, G. Stephens, D.D. Gretler, M.M. Jurek and A. Hutchaleelaha are employees and shareholders of Portola Pharmaceuticals. All other authors state that they have no conflict of interest.