Author + information
- Received June 25, 2013
- Revision received August 21, 2013
- Accepted August 30, 2013
- Published online April 1, 2014.
- David J. Schneider, MD∗ (, )
- Zubin Agarwal, MD,
- Naveen Seecheran, MD,
- Friederike K. Keating, MD and
- Prospero Gogo, MD
- Cardiology Unit, and Cardiovascular Research Institute, Department of Medicine, University of Vermont, Burlington, Vermont
- ↵∗Reprint requests and correspondence:
Dr. David J. Schneider, University of Vermont, 208 South Park Drive, Colchester, Vermont 05446.
Objectives This study sought to determine pharmacodynamic effects during transition from intravenous cangrelor to oral ticagrelor and from oral ticagrelor to intravenous cangrelor.
Background Cangrelor is an intravenous antagonist of P2Y12 and its use will require transition to and from oral agents.
Methods Patients (n = 12) with stable coronary artery disease who were taking aspirin 81 mg daily were recruited. On study day 1, they received a bolus plus 2-h infusion of cangrelor plus a 180-mg dose of ticagrelor at either 0.5 h (n = 6) or 1.25 h (n = 6). Pharmacodynamic effects (light transmission platelet aggregation in response to 20 and 5 μmol/l adenosine diphosphate, VerifyNow, P2Y12 assay (Accumetrics, San Diego, California), vasodilator-stimulated phosphoprotein index, and flow cytometry) were assessed during and after the cangrelor infusion. Patients took 90 mg of ticagrelor twice daily for either 6 (n = 6) or 7 (n = 6) doses. On study day 5, pharmacodynamic effects were assessed before and during a bolus plus 2-h infusion of cangrelor.
Results During cangrelor infusion, extensive inhibition of platelet function reflected by limited residual platelet reactivity was apparent. After cangrelor was stopped, the antiplatelet effects of ticagrelor were preserved despite a modest increase in platelet reactivity.
Conclusions Ticagrelor given before or during infusion of cangrelor did not attenuate the pharmacodynamic effects of cangrelor. The pharmacodynamic effects of ticagrelor were preserved when ticagrelor was given during infusion of cangrelor. Consistent with the reversible binding of ticagrelor, this oral P2Y12 antagonist can be administered before, during, or after treatment with cangrelor.
In patients with myocardial infarction, increased platelet reactivity has been associated with a greater burden of thrombus (1). Persistently, high platelet reactivity has been associated with a greater risk of early and later cardiovascular events (2). Consistent with these observations, antiplatelet therapy reduces the risk of subsequent cardiovascular events. The bioavailability of orally administered antiplatelet agents may be delayed in patients with sympathetic activation or ST-segment elevation myocardial infarction (3,4). Accordingly, treatment with cangrelor, a parenteral P2Y12 antagonist, is likely to be advantageous during selected intervals in which a rapid onset and offset of effect and consistent suppression of platelet reactivity is desired. The CHAMPION PHOENIX (Cangrelor Versus Standard Therapy to Achieve Optimal Management of Platelet Inhibition) trial (5) demonstrated that cangrelor reduced the incidence of ischemic events during percutaneous coronary intervention, without increasing the incidence of severe bleeding. The BRIDGE (Maintenance of Platelet Inhibition With Cangrelor After Discontinuation of Thienopyridines in Patients Undergoing Surgery) study (6) demonstrated that cangrelor can be used to maintain antiplatelet effects in patients scheduled for surgery.
The objective of the present study was to determine pharmacodynamic effects during the transition from intravenous cangrelor to oral ticagrelor and from oral ticagrelor to intravenous cangrelor. Previous studies have demonstrated that cangrelor blocks access to the P2Y12 receptor of the metabolites of thienopyridines (7,8). Administration of clopidogrel immediately after discontinuation of cangrelor was not associated with an increased incidence of early thrombotic complications among patients treated in the CHAMPION studies (5,9–11). By contrast, ticagrelor does not require metabolism to become active, and it binds reversibly to the P2Y12 receptor (12). Accordingly, the characteristics of ticagrelor suggest that it can be given before, during, and after an infusion of cangrelor without altering pharmacodynamic effects. We recruited 12 patients with stable coronary artery disease (CAD) to determine whether pharmacodynamic effects would be maintained if ticagrelor was given during infusion of cangrelor and whether previous treatment with ticagrelor altered the pharmacodynamic effects of cangrelor.
Patients were enrolled in a protocol approved by the University of Vermont/Fletcher Allen institutional review board, and provided written informed consent. Eligible patients were 18 to 75 years of age, had CAD documented by a previous myocardial infarction or coronary revascularization, and were taking 81 mg of aspirin daily. Exclusion criteria included the following: an acute coronary syndrome within the past 12 months; treatment with an anticoagulant or antiplatelet agent other than aspirin; a history of a bleeding diathesis; anemia (hematocrit <35%); severe renal insufficiency (creatinine clearance <30 ml/min); and moderate or severe hepatic insufficiency. Prohibited concomitant medications included strong and potent CYP3A inhibitors, simvastatin and lovastatin at doses more than 40 mg/day, omeprazole or esomeprazole, and digoxin. Use of nonsteroidal anti-inflammatory agents was discouraged during study participation but was not prohibited.
The study, which is outlined in Figure 1, was designed to determine whether cangrelor and ticagrelor each alter the antiplatelet effects of the other agent. On study day 1, initial treatment was with cangrelor. We determined whether addition of ticagrelor alters antiplatelet effects of cangrelor and whether previous treatment with cangrelor alters antiplatelet effects of ticagrelor. On study day 5, patients have been treated previously with ticagrelor. By stopping therapy 12 or 24 h before day 5, we determined whether previous treatment with ticagrelor alters the antiplatelet effects of cangrelor.
On study day 1, an intravenous catheter was placed in each arm. One arm was used for cangrelor infusion and the contralateral arm for blood sampling (this catheter was kept patent by infusion of normal saline at 50 ml/h). Baseline pharmacodynamic assessment was performed. A 30 μg/kg bolus of cangrelor was administered followed immediately by a 2-h infusion at a rate of 4.0 μg/kg/min. A loading dose of ticagrelor (180 mg) was given after 0.5 or 1.25 h (n = 6 for each). Blood for pharmacodynamic platelet function studies was taken after 0.5 or 1.25 h (corresponding to the time of ticagrelor load, n = 6 for each), and then at 1.75, 2, 2.25, 2.5, 2.75, 3, 4, and 5.25 h. Pharmacodynamic assessment included light transmission aggregometry (LTA), VerifyNow P2Y12 assay (Accumetrics, San Diego, California), vasodilator-stimulated phosphoprotein (VASP) index, and platelet activation measured with the use of flow cytometry. Patients were assigned randomly to receive either 6 (n = 6) or 7 (n = 6) doses of ticagrelor to be taken every 12 h. On study day 5, as performed on day 1, intravenous catheters were inserted into each arm and blood was taken for baseline pharmacodynamic assessment. A 30-μg/kg bolus of cangrelor was administered followed immediately by a 2-h infusion at a rate of 4.0 μg/kg/min. Blood for pharmacodynamic assessment was taken after 1 and 2 h. Adverse events were queried throughout study participation that ended with a telephone interview performed on study days 10 to 12.
Assessment of LTA, VerifyNow, and platelet activation with the use of flow cytometry was performed within 30 min of blood being taken. In the case of flow cytometry, samples were processed to fixation and then batched for analysis. For VASP index, samples were batched and processed as recommended by the vendor within 2 h of blood being taken. This approach limited the time from fixation of VASP index to flow analysis to <1 h.
LTA induced by 5 μmol/l and 20 μmol/l of adenosine diphosphate (ADP) was quantified ex vivo (i.e., in nonadjusted platelet-rich plasma). Platelet-poor plasma was set as 100% aggregation, and both maximal (peak) and terminal (at 300 s) aggregation were measured with a PAP4 aggregometer (BioData, Horsham, Pennsylvania). The VerifyNow P2Y12 assay, which measures the effects of drugs on the P2Y12 receptor by activating platelets with prostaglandin E1 (PGE1) in addition to ADP, was used in accordance with the instructions provided by the manufacturer. Platelet reactivity was expressed in platelet reactivity units.
Activation of platelets was identified with the use of flow cytometry as previously described (13). In brief, whole blood was added to N-2-hydroxyethylpiperazine-N-2-ethanesulfonic–Tyrode buffer containing fluorochrome-labeled antibodies. Activation was induced by 1 μmol/l ADP. After 15 min, platelets were fixed and red blood cells lysed with the use of Optilyse-C (Immunotech, Westbrook, Massachusetts). Flow cytometric analysis was performed with the use of a Beckman Coulter Epics Elite instrument (Miami, Florida). Platelets were identified on the basis of size (forward and side scatter) as well as the binding of an activation-independent ligand (anti-CD42b). Activation was identified by the surface expression of P-selectin with a phycoerythrin-conjugated anti-CD62, and activation of glycoprotein IIb-IIIa was identified with the use of fluorescein isothiocyanate–conjugated PAC-1 (fluorochrome antibodies from Becton Dickinson, San Jose, California).
VASP, an intracellular actin regulatory protein, is a substrate of both cyclic adenosine monophosphate–dependent and cyclic guanosine monophosphate–dependent protein kinases. Dephosphorylation of VASP occurs after activation of the P2Y12 receptor. Conversely, inhibition of the P2Y12 receptor and stimulation of PGE1-activated adenylyl cyclase induce phosphorylation of VASP by cyclic adenosine monophosphate–dependent protein kinases. The ratio of VASP phosphorylation/dephosphorylation reflects P2Y12 inhibition/activation (14). To determine the VASP index, a commercially-available kit (Diagnostica Stago, Inc., Parsippany, New Jersey) was used. Blood samples were incubated in vitro with ADP and/or PGE1 before fixation. Each sample was indirectly immunolabeled by incubation with 16C2 monoclonal antibody followed by staining with a goat antimouse fluorescein isothiocyanate polyclonal reagent. Flow cytometric analysis was performed using a Coulter Epics Elite cytometer (Beckman Coulter, Inc., Fullerton, California). Platelet population was identified based on forward and side scatter distributions, and 3,000 platelet events were gated and analyzed for mean fluorescence intensity (MFI) using EPICS XL software (Beckman Coulter). The MFI corresponding to each experimental condition was determined to establish a ratio directly correlated with the VASP phosphorylation state. The ratio 100 × [(MFIPGE1 – MFIADP+PGE1)/MFIPGE1], was used to define the VASP phosphorylation index (VASP index) corresponding to a ratio of the VASP phosphorylation of activated platelets versus resting platelets and was expressed as a percentage of platelet reactivity (14).
The primary objective of the study was to assess the preservation of pharmacodynamic effects during the transition from cangrelor to ticagrelor (assessed on study day 1) and during the transition from ticagrelor to cangrelor (assessed on study day 5). The primary endpoint was the aggregation (LTA) of platelets after 5 min (final aggregation) in response to 20 μmol/l ADP. A change in platelet reactivity and the extent of inhibition was determined relative to reference points at which the pharmacodynamic effect of cangrelor (7) or ticagrelor (12) alone would be apparent. The cangrelor reference was at 0.5 or 1.25 h (before administration of ticagrelor). The ticagrelor reference was at 5.25 h of day 1. This time point was chosen because the interval from cangrelor cessation to this time point was >5× the elimination half-life (∼25 min) of cangrelor. To demonstrate preservation of treatment effect, changes from reference time points and percentage changes were calculated. The mean changes will be estimated with 95% confidence intervals (CIs).
The study was powered based on the estimated 95% CI of mean difference in the extent of inhibition of platelet aggregation. With an assumed SD that was 20% of the mean and a sample of 12 patients, the 95% CI for the mean difference extended 13% in each direction (half width) from the mean difference when switching from cangrelor to ticagrelor. A half width of 21% could be detected when switching from ticagrelor to cangrelor, with the sample size of 6 in each group. This extent of variation was considered to be within the expected biological variation in effect (9,15,16).
The clinical characteristics of the patients are shown in Table 1. The median time from previous myocardial infarction was 6 years (range 3 to 13 years) and the median time from previous revascularization was 6 years (range 2 to 17 years). No adverse events (ischemic or bleeding) occurred during the trial.
Final LTA (after 5 min, primary endpoint) aggregation measured by LTA was extensively and consistently inhibited during the infusion of cangrelor (Fig. 2). Residual platelet reactivity was <4% and the extent of inhibition was >95% when cangrelor was being infused before and after ticagrelor was administered on study day 1. On study day 5, previous treatment with ticagrelor did not alter the inhibitory effect of cangrelor. Consistent pharmacodynamic effects were apparent with each of the secondary measures of platelet function (Table 2, Fig. 3).
On study day 1, a modest increase in platelet reactivity was apparent during the first hour after discontinuation of cangrelor (Figs. 2 and 3). This reflected the offset of cangrelor and the onset of ticagrelor. Pharmacodynamics effects during that hour demonstrated no significant increase in platelet reactivity compared with the pre-defined reference time point (5.25 h). Consistent with the pharmacokinetics (12), administration of ticagrelor earlier (at 0.5 h rather than 1.25 h) appeared to attenuate the increase in residual platelet reactivity and augment the extent of inhibition that was apparent. To provide another assessment of the potential clinical importance of the increment in platelet reactivity, the residual platelet reactivity seen at 2.5 h when ticagrelor was given at 0.5 h (16 ± 15%) was comparable to the residual platelet reactivity seen 12 h after the last dose of ticagrelor measured on study day 5 (12 ± 9%). Once again, each of the secondary measures of platelet function demonstrated results consistent with the primary endpoint (Table 3, Fig. 3).
The primary objective was to determine whether platelet reactivity could be suppressed consistently during a transition from an intravenous P2Y12 antagonist, cangrelor, to an oral P2Y12 antagonist, ticagrelor, and from oral ticagrelor to intravenous cangrelor. Previous pharmacodynamic studies with the thienopyridine clopidogrel demonstrated that this agent must be given after discontinuation of cangrelor (7). Because ticagrelor binds reversibly (12), we tested whether ticagrelor could be administered during infusion of cangrelor.
Pharmacodynamic effects were assessed in patients with stable CAD who were taking 81 mg of aspirin daily. Because no single measure of platelet function is able to fully characterize platelet reactivity, multiple tests were performed at each time point. The frequency of blood drawing, the complexity of the pharmacodynamic assay, and the total amount of blood taken were not well suited to combine with coronary intervention. Accordingly, this was a standalone study in patients with known CAD.
The pharmacodynamic assessment demonstrated that residual platelet reactivity during infusion of cangrelor was limited, regardless of whether ticagrelor was given during the infusion (study day 1) or ticagrelor had been given before infusion (study day 5). With respect to the primary endpoint, terminal aggregation of platelets in response to 20 μmol/l ADP, the residual platelet reactivity was <5% and the extent of inhibition was >95% during cangrelor treatment. Parallel results were seen with each of the secondary endpoints.
During the transition from cangrelor to ticagrelor, a nonsignificant increase in platelet reactivity was observed during the first hour after cangrelor was stopped. Earlier administration of ticagrelor appeared to attenuate the increase in platelet reactivity. This observation is consistent with the pharmacokinetics that predict the onset of steady-state concentrations of ticagrelor approximately 1.5 h after administration (12).
Although this trial is not of sufficient size to determine clinical implications, the variation in pharmacodynamic effects appeared to be within the range of expected variation observed during regular administration. We found that extensive inhibition (>80%) of platelet aggregation was apparent both 12 and 24 h after the last dose of ticagrelor. Residual platelet reactivity seen on study day 1 after cangrelor was discontinued was not greater than that seen 12 h after the last dose of ticagrelor (Fig. 2). Further, the 95th percentile of the PRU measured by the VerifyNow assay was <190 when ticagrelor was administered either 0.5 or 1.25 h earlier. Results from previous pharmacodynamic studies with ticagrelor (15,16) support the observation that the variation in the extent of inhibition observed during the first hour after stopping cangrelor is consistent with variation seen during regular dosing.
No significant interaction between the binding of cangrelor and ticagrelor was evident; however, this study was not sufficiently sized to exclude a limited interaction. Comparison of results obtained after an equivalent interval raises the possibility of a modest interaction. For patients given ticagrelor at 1.25 h (which is 0.75 h before cangrelor was stopped), the residual platelet reactivity and extent of inhibition seen 1.75 h after the loading dose (at 3.0 h) were 4.5 ± 3.4% and 93 ± 7%. For patients given ticagrelor at 0.5 h, results 1.75 h later (at 2.25 h) were 13 ± 9.3% and 82 ± 13% and trended up rather than down at 2.5 h. A similar observation was apparent with each of the secondary endpoints. Our results were consistent with a previous study in dogs in which a substantial interaction between cangrelor and ticagrelor was not apparent (17).
Ticagrelor given before or during infusion of cangrelor did not attenuate the pharmacodynamic effects of cangrelor. In addition, the pharmacodynamic effects of ticagrelor were preserved when ticagrelor was given during infusion of cangrelor. Consistent with the reversible binding of ticagrelor, this oral P2Y12 antagonist can be administered before, during, or after treatment with cangrelor. Consistent with its pharmacokinetics, the pharmacodynamic effects will be greater when ticagrelor is given earlier.
The authors thank Michaelanne Rowen, RN, Linda Chadwick, RN, Joseph Miller, and Heidi Taatjes for the assistance in completing this study.
This study was sponsored by and funded by a grant from The Medicines Company. Dr. Schneider has received honorarium (<$10,000) from The Medicines Company and AstraZeneca. Dr. Gogo has received consulting fees from The Medicines Company. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- adenosine diphosphate
- coronary artery disease
- confidence interval
- light transmission aggregometry
- mean fluorescence intensity
- prostaglandin E1
- vasodilator-stimulated phosphoprotein
- Received June 25, 2013.
- Revision received August 21, 2013.
- Accepted August 30, 2013.
- American College of Cardiology Foundation
- Parodi G.,
- Valenti R.,
- Bellandi B.,
- et al.
- Angiolillo D.J.,
- Schneider D.J.,
- Bhatt D.L.,
- et al.
- Bonello L.,
- Camoin-Jau L.,
- Arques S.,
- et al.
- Gurbel P.A.,
- Bliden K.P.,
- Butler K.,
- et al.
- Storey R.F.,
- Angiolillo D.J.,
- Patil S.B.,
- et al.