Intravenous Platelet Blockade with Cangrelor during PCI

Transcription

1 The new england journal of medicine original article Intravenous Platelet Blockade with during PCI Deepak L. Bhatt, M.D., M.P.H., A. Michael Lincoff, M.D., C. Michael Gibson, M.D., Gregg W. Stone, M.D., Steven McNulty, M.S., Gilles Montalescot, M.D., Ph.D., Neal S. Kleiman, M.D., Shaun G. Goodman, M.D., Harvey D. White, D.Sc., Kenneth W. Mahaffey, M.D., Charles V. Pollack, Jr., M.D., Steven V. Manoukian, M.D., Petr Widimsky, M.D., Dr.Sc., Derek P. Chew, M.B., B.S., M.P.H., Fernando Cura, M.D., Ivan Manukov, M.D., Frantisek Tousek, M.D., M. Zubair Jafar, M.D., Jaspal Arneja, M.D., Simona Skerjanec, Pharm.D., and Robert A. Harrington, M.D., for the CHAMPION PLATFORM Investigators* Abstract Background Intravenous cangrelor, a rapid-acting, reversible adenosine diphosphate (ADP) receptor antagonist, might reduce ischemic events during percutaneous coronary intervention (PCI). Methods In this double-blind, placebo-controlled study, we randomly assigned 5362 patients who had not been treated with clopidogrel to receive either cangrelor or placebo at the time of PCI, followed by 600 mg of clopidogrel. The primary end point was a composite of death, myocardial infarction, or ischemia-driven revascularization at 48 hours. Enrollment was stopped when an interim analysis concluded that the trial would be unlikely to show superiority for the primary end point. Results The primary end point occurred in 185 of 2654 patients receiving cangrelor (7.0%) and in 210 of 2641 patients receiving placebo (8.0%) (odds ratio in the cangrelor group, 0.87; 95% confidence interval [CI], 0.71 to 1.07; P = 0.17) (modified intentionto-treat population adjusted for missing data). In the cangrelor group, as compared with the placebo group, two prespecified secondary end points were significantly reduced at 48 hours: the rate of stent thrombosis, from 0.6% to 0.2% (odds ratio, 0.31; 95% CI, 0.11 to 0.85; P = 0.02), and the rate of death from any cause, from 0.7% to 0.2% (odds ratio, 0.33; 95% CI, 0.13 to 0.83; P = 0.02). There was no significant difference in the rate of blood transfusion (1.0% in the cangrelor group and 0.6% in the placebo group, P = 0.13), though major bleeding on one scale was increased in the cangrelor group, from 3.5% to 5.5% (P<0.001), because of more groin hematomas. The authors affiliations are listed in the Appendix. Address reprint requests to Dr. Bhatt at VA Boston Healthcare System, 1400 VFW Pkwy., Boston, MA 02132, or at dlbhattmd@post.harvard.edu. *Investigators in the versus Standard Therapy to Achieve Optimal Management of Platelet Inhibition (CHAMPION) PLATFORM trial are listed in the Supplementary Appendix, available with the full text of this article at NEJM.org. This article ( /NEJMoa ) was published on November 15, 2009, at NEJM.org. N Engl J Med 2009;361. Copyright 2009 Massachusetts Medical Society. Conclusions The use of periprocedural cangrelor during PCI was not superior to placebo in reducing the primary end point. The prespecified secondary end points of stent thrombosis and death were lower in the cangrelor group, with no significant increase in the rate of transfusion. Further study of intravenous ADP blockade with cangrelor may be warranted. (ClinicalTrials.gov number, NCT ) /nejmoa nejm.org 1

2 The new england journal of medicine Blockade of the platelet adenosine diphosphate (ADP) receptor has been demonstrated to improve cardiovascular outcomes in patients undergoing percutaneous coronar y inter vention (PCI). 1-5 In patients undergoing coronary stenting, ticlopidine given with aspirin was found to decrease the risk of stent thrombosis and other important ischemic complications, as compared with aspirin alone or aspirin plus warfarin. 6 Ticlopidine was ultimately replaced by clopidogrel as the ADP-receptor blocker of choice because of the drug s improved profile for side effects and adverse events. 7 Despite its efficacy, clopidogrel has a delayed onset of action, even when given with a loading dose; it also does not provide complete ADP-receptor inhibition and has substantial variability among patients. 8 Even in the contemporary era, the vexing problem of acute stent thrombosis has not been eliminated. 9,10 Moreover, many physicians refrain from administering clopidogrel before obtaining angiographic definition of the coronary anatomy, since this irreversible platelet inhibitor has been associated with an increased risk of perioperative bleeding if coronary-artery bypass grafting (CABG) is required rather than PCI. In addition, more potent oral ADP-receptor blockers have been tested and found to reduce ischemic outcomes even further but also increase the rate of bleeding Therefore, an even more potent intravenous agent with fast onset and fast offset of action might provide a desirable combination of protection from ischemia without an excessive risk of bleeding. (the Medicines Company) is an adenosine triphosphate analogue that reversibly binds to and inhibits the P2Y 12 ADP receptor. It has a half-life of 3 to 6 minutes and, when given as a bolus plus infusion, quickly and consistently inhibits platelets to a high degree, with normalization of platelet function within 60 minutes after discontinuation. In the versus Standard Therapy to Achieve Optimal Management of Platelet Inhibition (CHAMPION) PLATFORM trial, which we report here, we examined the efficacy of cangrelor versus placebo administered to patients during PCI, with patients in the placebo group subsequently receiving a loading dose of 600 mg of clopidogrel at the end of the revascularization procedure and patients in the cangrelor group receiving 600 mg of clopidogrel at the end of the infusion. The results of the CHAMPION PCI trial, 16 which compared the use of cangrelor with the use of 600 mg of clopidogrel given at the start of the PCI procedure, are also reported in this issue of the Journal. Methods Patients A total of 5362 patients who provided consent were enrolled at 218 sites in 18 countries from October 2006 through May Patients underwent randomization according to a double-blind, placebo-controlled, double-dummy design to receive either cangrelor (in a bolus of 30 μg per kilogram of body weight and an infusion of 4 μg per kilogram per minute) or a placebo bolus and infusion for the duration of the PCI procedure, with a minimum infusion duration of 2 hours and a maximum of 4 hours. Patients in the cangrelor group received 600 mg of clopidogrel after the end of the cangrelor infusion, and those in the placebo group received 600 mg of clopidogrel at the end of the procedure (Fig. 1). Inclusion criteria for the trial were an age of at least 18 years, diagnostic coronary angiography revealing at least one atherosclerotic lesion amenable to PCI with or without stent implantation, and evidence of either myocardial infarction without ST-segment elevation or unstable angina. Patients with stable angina were initially eligible at the beginning of the trial before a protocol amendment. The diagnosis of non ST-segment elevation myocardial infarction required a level of troponin I or troponin T that was higher than the upper limit of the normal range within 24 hours before randomization (or if troponin results were unavailable at that time, a level of creatine kinase myocardial band [CK-MB] isoenzyme that was higher than the upper limit of the normal range). The diagnosis of unstable angina required ischemic chest discomfort occurring at rest and lasting for at least 10 minutes within 24 hours before randomization and dynamic electrocardiographic changes; an age of 65 years or older or the presence of diabetes mellitus (or both) was also required. All patients provided written informed consent. Exclusion criteria included the use of any thienopyridine in the previous 7 days, a planned staged PCI procedure in which the second stage would occur within 30 days after the first procedure, admission planned within 12 hours after /nejmoa nejm.org

3 Intravenous Platelet Blockade with during PCI PCI, the occurrence of myocardial infarction with ST-segment elevation within 48 hours before randomization, known or suspected pregnancy, lactation, increased bleeding risk (the occurrence of ischemic stroke within the previous year or any previous hemorrhagic stroke), intracranial tumor, cerebral arteriovenous malformation, intracranial aneurysm, trauma or major surgery (including CABG) within the previous month, current warfarin use, active bleeding, a known international normalized ratio of more than 1.5, a past or present bleeding disorder, a platelet count of less than 100,000 per cubic millimeter, severe hypertension (systolic blood pressure, >180 mm Hg; or diastolic blood pressure, >110 mm Hg), or the use of fibrinolytic therapy or a glycoprotein IIb/ IIIa inhibitor in the 12 hours preceding randomization. Primary and Secondary End Points The primary efficacy end point was a composite of death, myocardial infarction, or ischemia-driven revascularization 48 hours after PCI. The primary analysis was performed on a modified intention-to-treat population. Confirmatory analyses were performed on an intention-to-treat population. Secondary end points included the individual rates of death, myocardial infarction, new Q-wave myocardial infarction, ischemia-driven revascularization, abrupt vessel closure, or stroke at 48 hours. Rates of death at 30 days and 1 year were also recorded. The clinical events committee adjudicated the rates of myocardial infarction, Q-wave myocardial infarction, ischemia-driven revascularization, stent thromboses, and stroke (ischemic or hemorrhagic). The definition of stent thrombosis was based on the Academic Research Consortium s definition of definite stent thrombosis. After review of the prespecified analyses, we examined two exploratory end points that were less reliant on the measurement of periprocedural biomarkers. The exploratory end points, which were composed of prespecified and adjudicated end points, were a composite of death, Q-wave myocardial infarction, or ischemia-driven revascularization and a composite of death, Q-wave myocardial infarction, or stent thrombosis. We compared the rates of bleeding and adverse events at 48 hours and examined several bleeding end points. None of these events were prespecified as a primary bleeding end point, since we wanted 2656 Received cangrelor, 30 µg/kg bolus and 2654 received 4 µg/kg/min infusion for 2 hr or the duration of the procedure, whichever was longer 2627 Received 600-mg placebo capsules immediately after procedure 2620 Received 600-mg clopidogrel capsules immediately after infusion 48 hr after randomization 2654 Included in modified intention-to-treat analysis 2662 Included in safety analysis 2691 Included in intention-totreat analysis 5301 Patients requiring PCI (with or without stent, excluding STEMI) underwent randomization to discern a bleeding effect if present and provide full disclosure of such bleeding. The definitions of all these end points were consistent with those used in the CHAMPION PCI trial. 16 The statistical group at the Duke Clinical Research Institute (DCRI) received regular transfers of data from the sponsor. The DCRI was responsible for coordinating activities and analyses 2645 Received placebo, 30 µg/kg bolus and 4 µg/kg/min infusion for 2 hr or the duration of the procedure, whichever was longer 2620 Received 600-mg clopidogrel capsules immediately after procedure 2607 Received 600-mg placebo capsules immediately after infusion 48 hr after randomization 2641 Included in modified intention-to-treat analysis 2650 Included in safety analysis 2664 Included in intention-totreat analysis Follow-up analysis Primary and secondary end points at 30 days Death at 6 mo and 1 yr Figure 1. Enrollment and Outcomes in the Modified Intention-to-Treat Population. All patients who underwent randomization, received at least one dose of a study drug, and underwent the index percutaneous coronary intervention (PCI) were included in the modified intention-to-treat population. All efficacy analyses of the primary and secondary end points were performed in this population. The safety population included all patients who received at least one dose of a study drug. Intention-to-treat analyses were also reported. STEMI denotes ST-segment elevation myocardial infarction /nejmoa nejm.org 3

4 The new england journal of medicine Table 1. Baseline Characteristics of the Patients, According to Study Population.* Characteristic Modified Intention-to-Treat Population Intention-to-Treat Population Safety Population (N = 2656) (N = 2645) (N = 2693) (N = 2669) (N = 2662) (N = 2650) Median age (interquartile range) yr 63.0 ( ) 63.0 ( ) 63.0 ( ) 63.0 ( ) 63.0 ( ) 63.0 ( ) Sex no. (%) Male 1909 (71.9) 1863 (70.4) 1938 (72.0) 1877 (70.3) 1915 (71.9) 1866 (70.4) Female 747 (28.1) 782 (29.6) 755 (28.0) 792 (29.7) 747 (28.1) 784 (29.6) Race or ethnic group no. (%) White 2015 (75.9) 2006 (75.8) 2039 (75.7) 2024 (75.8) 2017 (75.8) 2009 (75.8) Asian 475 (17.9) 473 (17.9) 482 (17.9) 476 (17.8) 477 (17.9) 474 (17.9) Black 75 (2.8) 72 (2.7) 80 (3.0) 73 (2.7) 76 (2.9) 73 (2.8) Hispanic 74 (2.8) 84 (3.2) 75 (2.8) 85 (3.2) 75 (2.8) 84 (3.2) Other 8 (0.3) 5 (0.2) 8 (0.3) 5 (0.2) 8 (0.3) 5 (0.2) Missing data 9 (0.3) 5 (0.2) 9 (0.3) 6 (0.2) 9 (0.3) 5 (0.2) Median weight (interquartile range) kg 80.0 ( ) 80.0 ( ) 80.0 ( ) 80.0 ( ) 80.0 ( ) 80.0 ( ) Median height (interquartile range) cm ( ) ( ) ( ) ( ) ( ) ( ) Stable angina no. (%) 139 (5.2) 140 (5.3) 145 (5.4) 142 (5.3) 138 (5.2) 141 (5.3) Unstable angina no. (%) 939 (35.4) 909 (34.4) 949 (35.2) 918 (34.4) 940 (35.3) 911 (34.4) Myocardial infarction without ST-segment elevation no. (%) Medical history no./total no. (%) 1578 (59.4) 1596 (60.3) 1599 (59.4) 1609 (60.3) 1584 (59.5) 1598 (60.3) Diabetes mellitus 812/2655 (30.6) 862/2642 (32.6) 828/2692 (30.8) 868/2666 (32.6) 815/2661 (30.6) 862/2647 (32.6) Current smoker 842/2639 (31.9) 799/2628 (30.4) 850/2675 (31.8) 806/2651 (30.4) 845/2645 (31.9) 799/2633 (30.3) Hypertension 1972/2647 (74.5) 1962/2633 (74.5) 1994/2684 (74.3) 1979/2656 (74.5) 1974/2653 (74.4) 1966/2638 (74.5) Hyperlipidemia 1324/2472 (53.6) 1332/2472 (53.9) 1342/2508 (53.5) 1347/2494 (54.0) 1325/2477 (53.5) 1335/2477 (53.9) Stroke or transient ischemic attack 159/2644 (6.0) 158/2639 (6.0) 162/2681 (6.0) 160/2662 (6.0) 160/2650 (6.0) 158/2644 (6.0) Family history of coronary artery disease 902/2489 (36.2) 890/2479 (35.9) 918/2525 (36.4) 901/2502 (36.0) 907/2495 (36.4) 891/2484 (35.9) Myocardial infarction 640/2642 (24.2) 679/2636 (25.8) 645/2678 (24.1) 683/2659 (25.7) 641/2648 (24.2) 680/2641 (25.7) PTCA or PCI 374/2645 (14.1) 409/2637 (15.5) 381/2682 (14.2) 411/2659 (15.5) 377/2651 (14.2) 409/2642 (15.5) CABG 199/2655 (7.5) 221/2644 (8.4) 203/2692 (7.5) 223/2667 (8.4) 200/2661 (7.5) 221/2649 (8.3) Congestive heart failure 206/2644 (7.8) 191/2641 (7.2) 210/2681 (7.8) 192/2664 (7.2) 208/2650 (7.8) 191/2646 (7.2) Peripheral artery disease 122/2604 (4.7) 142/2598 (5.5) 126/2641 (4.8) 143/2619 (5.5) 124/2610 (4.8) 142/2603 (5.5) /nejmoa nejm.org

5 Intravenous Platelet Blockade with during PCI Periprocedural medications no./total no. (%) Bivalirudin 559/2656 (21.0) 555/2644 (21.0) 565/2692 (21.0) 561/2666 (21.0) 561/2662 (21.1) 556/2649 (21.0) Unfractionated heparin 1699/2656 (64.0) 1695/2644 (64.1) 1714/2692 (63.7) 1709/2666 (64.1) 1701/2662 (63.9) 1699/2649 (64.1) Low-molecular-weight heparin 481/2654 (18.1) 497/2645 (18.8) 487/2690 (18.1) 501/2667 (18.8) 484/2660 (18.2) 497/2650 (18.8) Glycoprotein IIb/IIIa 241/2653 (9.1) 244/2642 (9.2) 245/2689 (9.1) 247/2664 (9.3) 242/2659 (9.1) 244/2647 (9.2) No. of target vessels no./total no. (%) /2654 (83.6) 2201/2643 (83.3) 2231/2667 (83.7) 2211/2653 (83.3) 2217/2653 (83.6) 2202/2644 (83.3) 2 414/2654 (15.6) 412/2643 (15.6) 414/2667 (15.5) 412/2653 (15.5) 414/2653 (15.6) 412/2644 (15.6) 3 19/2654 (0.7) 29/2643 (1.1) 19/2667 (0.7) 29/2653 (1.1) 19/2653 (0.7) 29/2644 (1.1) Missing data 3/2654 (0.1) 1/2643 (<0.1) 3/2667 (0.1) 1/2653 (<0.1) 3/2653 (0.1) 1/2644 (<0.1) Type of stent no./total no. (%) Drug-eluting 1033/2654 (38.9) 1021/2643 (38.6) 1037/2667 (38.9) 1023/2653 (38.6) 1032/2653 (38.9) 1022/2644 (38.7) Non drug-eluting 1509/2654 (56.9) 1510/2643 (57.1) 1514/2667 (56.8) 1515/2653 (57.1) 1509/2653 (56.9) 1510/2644 (57.1) No stent or other 112/2654 (4.2) 112/2643 (4.2) 116/2667 (4.3) 115/2667 (4.3) 112/2653 (4.2) 111/2644 (4.2) Angiographic complication (site reported) no. (%) Threatened abrupt closure 10 (0.4) 9 (0.3) 10 (0.4) 9 (0.3) 10 (0.4) 9 (0.3) Unsuccessful procedure 81 (3.1) 95 (3.6) 84 (3.1) 97 (3.7) 81 (3.1) 95 (3.6) Abrupt vessel closure 13 (0.5) 16 (0.6) 13 (0.5) 16 (0.6) 13 (0.5) 16 (0.6) New thrombus or suspected thrombus 14 (0.5) 15 (0.6) 14 (0.5) 15 (0.6) 14 (0.5) 15 (0.6) Acute stent thrombosis 1 (<0.1) 5 (0.2) 1 (<0.1) 5 (0.2) 1 (<0.1) 5 (0.2) Need for urgent CABG 5 (0.2) 3 (0.1) 5 (0.2) 4 (0.2) 5 (0.2) 3 (0.1) Intravenous study drug administered no. (%) 2656 (100.0) 2645 (100.0) 2663 (98.9) 2649 (99.3) 2662 (100.0) 2650 (100.0) Bolus administered no. (%) 2656 (100.0) 2645 (100.0) 2663 (98.9) 2649 (99.3) 2662 (100.0) 2650 (100.0) Infusion administered no. (%) 2654 (99.9) 2645 (100.0) 2659 (98.7) 2649 (99.3) 2658 (99.8) 2650 (100.0) Median duration of infusion (interquartile range) hr 2.1 ( ) 2.1 ( ) 2.1 ( ) 2.1 ( ) 2.1 ( ) 2.1 ( ) Oral study drug administered no. (%) 2629 (99.0) 2625 (99.2) 2630 (97.7) 2626 (98.4) 2629 (98.8) 2627 (99.1) * Percentages may not total 100 because of rounding. CABG denotes coronary-artery bypass grafting, PCI percutaneous coronary intervention, and PTCA percutaneous transluminal coronary angioplasty. Race or ethnic group was self-reported. Data on unsuccessful procedures were available for 2653 patients in the placebo group in the intention-to-treat population, for 2654 patients in the cangrelor group in the modified intention-to-treat population, and for 2653 patients in the cangrelor group in the safety population /nejmoa nejm.org 5

6 The new england journal of medicine for the independent data and safety monitoring committee and an interim analysis review committee. At the conclusion of the trial, the full database was transferred to the DCRI for the primary and secondary analyses. These analyses were performed independently but in collaboration with the sponsor. One of the trial s principal academic investigators prepared the first draft of the manuscript, which was then reviewed and edited by the executive committee and selected members of the steering committee and site investigators. The sponsor had the right to review but not approve the final manuscript. The trial s principal investigators accept full responsibility for the accuracy and completeness of the reported analyses and interpretations of the data. Statistical Analysis All efficacy analyses were performed in the modified intention-to-treat population, which consisted of all patients who underwent randomization, received at least one dose of a study drug, and underwent the index PCI. All safety-related analyses were performed on the safety population, which included all patients who received at least one dose of a study drug. Patients in the safety analyses were assigned to a study group on the basis of the treatment they actually received, not as randomized. Intention-to-treat analyses are also presented for full disclosure of results. The between-group comparison of the primary end point was performed by calculating an odds ratio with accompanying 95% confidence intervals with the use of logistic regression. Logistic regression was also used to analyze the majority of the remaining secondary end points. All reported P values are two-sided, with a P value of less than 0.05 considered to indicate statistical significance. The trial had a power of 85% to detect a 25% relative reduction in the primary end point, on the assumption that the event would occur in 7.7% of patients in the placebo group, with a projected sample size of 6400 patients. While the trial was under way and the results were still blinded, an adaptive design was prospectively implemented. 17 This design allowed for early termination for either lack of efficacy or superiority of the primary end point, for a sample-size increase in the entire trial population, or for a sample-size increase in specified subgroups. 17 After a second interim analysis, trial enrollment was terminated because the review committee decided that the trial was unlikely to show the superiority of cangrelor. Results Patients A total of 5362 patients were included in the intention-to-treat population; 61 of these patients did not receive a study drug or undergo PCI, which left 5301 patients in the primary modified intention-to-treat population (Fig. 1 in the Supplementary Appendix, available with the full text of this article at NEJM.org). Baseline characteristics were well matched in the two study groups (Table 1). The majority of patients (59.8%) had received the diagnosis of myocardial infarction without ST-segment elevation. During PCI, unfractionated heparin was the most frequently used antithrombin agent (in 63.9% of patients), and glycoprotein IIb/IIIa inhibitors were used sparingly (in 9.2% of patients). Drug-eluting stents were used less often than bare-metal stents (in 38.7% of patients vs. 56.9% of patients). The time from hospital admission to PCI was short (median, 7.9 hours; interquartile range, 3.3 to 24.1). Primary End Point The primary end point occurred in 185 of 2654 patients receiving cangrelor (7.0%) and in 210 of 2641 patients receiving placebo (8.0%) (odds ratio in the cangrelor group, 0.87; 95% confidence interval [CI], 0.71 to 1.07; P = 0.17) (Table 2 and Fig. 2A). (Data were missing for two patients in the cangrelor group and four in the placebo group.) There was no significant between-group difference in the overall rate of myocardial infarction, Q-wave myocardial infarction, or is che mia-driven revascularization. The rate of stent thrombosis at 48 hours was significantly lower in the cangrelor group (0.2%) than in the placebo group (0.6%) (5 vs. 16 patients) (odds ratio, 0.31; 95% CI, 0.11 to 0.85; P = 0.02), and the difference was still significant at 30 days (Fig. 2B). The rate of death at 48 hours was significantly lower in the cangrelor group (0.2%) than in the placebo group (0.7%) (6 vs. 18 patients) (odds ratio, 0.33; 95% CI, 0.13 to 0.83; P = 0.02), though at 30 days, this difference was no longer significant (Fig. 2C, and Table 1 in the Supplementary Appendix). Somewhat counterintuitively, in the subgroup of 1659 patients who did not have an elevation in the troponin level at baseline, the primary end point was reduced in the cangrelor group (4.6%), as compared with the placebo group (7.2%) (odds ratio, 0.62; 95% CI, 0.41 to 0.95; P = 0.03) /nejmoa nejm.org

7 Intravenous Platelet Blockade with during PCI Table 2. Major End Points at 48 Hours, According to Study Population.* Population and End Point no. (%) Odds Ratio (95% CI) P Value Modified intention-to-treat population No. of patients evaluated Adjudicated end points Death, myocardial infarction, or ischemia-driven revascularization 185 (7.0) 210 (8.0) 0.87 ( ) 0.17 (primary end point) Myocardial infarction 177 (6.7) 191 (7.2) 0.92 ( ) 0.42 Ischemia-driven revascularization 19 (0.7) 24 (0.9) 0.79 ( ) 0.44 Death from any cause 6 (0.2) 18 (0.7) 0.33 ( ) 0.02 Stroke 7 (0.3) 5 (0.2) 1.39 ( ) 0.57 Stent thrombosis 5 (0.2) 16 (0.6) 0.31 ( ) 0.02 Q-wave myocardial infarction 4 (0.2) 8 (0.3) 0.50 ( ) 0.25 Exploratory end points Death, Q-wave myocardial infarction, or ischemia-driven revascularization 23 (0.9) 41 (1.6) 0.55 ( ) 0.02 Death, Q-wave myocardial infarction, or stent thrombosis 13 (0.5) 34 (1.3) 0.38 ( ) Intention-to-treat population No. of patients evaluated Adjudicated end points Death, myocardial infarction, or ischemia-driven revascularization 187 (6.9) 213 (8.0) 0.86 ( ) 0.15 Myocardial infarction 177 (6.6) 192 (7.2) 0.91 ( ) 0.36 Ischemia-driven revascularization 19 (0.7) 26 (1.0) 0.72 ( ) 0.28 Death from any cause 8 (0.3) 19 (0.7) 0.42 ( ) 0.04 Stroke 7 (0.3) 6 (0.2) 1.16 ( ) 0.80 Stent thrombosis 5 (0.2) 16 (0.6) 0.31 ( ) 0.02 Q-wave myocardial infarction 4 (0.1) 9 (0.3) 0.44 ( ) 0.17 Exploratory end points Death, Q-wave myocardial infarction, or ischemia-driven revascularization 25 (0.9) 44 (1.7) 0.56 ( ) 0.02 Death, Q-wave myocardial infarction, or stent thrombosis 15 (0.6) 36 (1.4) 0.41 ( ) Safety population No. of patients evaluated Adjudicated end points Death, myocardial infarction, or ischemia-driven revascularization 185 (7.0) 212 (8.0) 0.86 ( ) 0.14 Myocardial infarction 176 (6.6) 193 (7.3) 0.90 ( ) 0.33 Ischemia-driven revascularization 19 (0.7) 25 (0.9) 0.75 ( ) 0.36 Death from any cause 7 (0.3) 18 (0.7) 0.39 ( ) 0.03 Stroke 7 (0.3) 5 (0.2) 1.39 ( ) 0.57 Stent thrombosis 5 (0.2) 16 (0.6) 0.31 ( ) 0.02 Q-wave myocardial infarction 4 (0.2) 9 (0.3) 0.44 ( ) 0.17 Exploratory end points Death, Q-wave myocardial infarction, or ischemia-driven revascularization 24 (0.9) 42 (1.6) 0.56 ( ) 0.03 Death, Q-wave myocardial infarction, or stent thrombosis 14 (0.5) 35 (1.3) 0.40 ( ) * Data were missing for two patients in the cangrelor group and three in the placebo group in the modified intention-to-treat population, for two patients in the cangrelor group and five in the placebo group in the intention-to-treat population, and for two patients in the cangrelor group and four in the placebo group in the safety population /nejmoa nejm.org 7

8 The new england journal of medicine Therefore, exploratory analyses were performed in the overall study population to evaluate the following two composite clinical end points: death, Q-wave myocardial infarction, or stent thrombosis; and death, Q-wave myocardial infarction, or ischemia-driven revascularization. These end points were significantly reduced in the cangrelor group (Table 2). Bleeding According to the criteria for major or minor bleeding in the Thrombolysis in Myocardial Infarction (TIMI) study or of severe or moderate bleeding in the Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries (GUSTO) study, the rates of bleeding did not differ significantly between the two groups. However, according to criteria of the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial and the criteria for minor bleeding in the GUSTO trial, the rates of major and minor bleeding were significantly higher in the cangrelor group (Table 3). The difference in major bleeding according to the ACUITY criteria was due to an excess of groin hematomas but not in more serious forms of bleeding in the cangrelor group. The rates of redcell transfusion were not significantly different (0.9% in the cangrelor group vs. 0.6% in the placebo group, P = 0.12). Notably, among patients who were at increased risk for bleeding, such as the elderly and those who had a history of stroke or transient ischemic attack, the rate of transfusion was not significantly higher in the cangrelor group than in the placebo group (Fig. 2 in the Supplementary Appendix). There was no significant difference in the rate of arrhythmia in the cangrelor group, as compared with the placebo group (2.3% vs. 2.4%, P = 0.77). There was a higher incidence of dyspnea in the cangrelor group (1.4%) than in the placebo group (0.5%), with 37 patients and 14 patients, respectively (P = 0.002) (for details on adverse events, see Table 2 in the Supplementary Appendix). Discussion was not superior to placebo in reducing the composite of death, myocardial infarction, or ischemia-driven revascularization at 48 hours after PCI, the primary end point of this study. However, important prespecified secondary end points, which are considered exploratory, Figure 2 (facing page). Kaplan Meier Curves for the Rate of the Primary End Point, Stent Thrombosis, and Death. Panel A shows the rate of the primary efficacy end point a composite of death, myocardial infarction, or ischemia-driven revascularization at 48 hours after percutaneous coronary intervention (PCI) in the cangrelor group and the placebo group. Panel B shows the rate of stent thrombosis at 48 hours (at left) and at 35 days (at right) after PCI. Panel C shows the rate of death at 48 hours (left) and 35 days (right). In Panels B and C, which both show landmark analyses of the Kaplan Meier curves, the value at day 2 has been reset to zero in the right-hand panels. In all three panels, data are shown for the end points at 30±5 days, as specified in the protocol. including rates of stent thrombosis and death from any cause, were significantly reduced in the cangrelor group at 48 hours. Given the mechanism of action of cangrelor, these reductions in important clinical outcomes are plausible. In light of a recent trial of another reversible ADP-receptor antagonist (the oral agent ticagrelor), the potential mechanistic benefits of cangrelor seem even more likely to affect important ischemic outcomes. 14 The Study of Platelet Inhibition and Patient Outcomes (PLATO) (ClinicalTrials.gov number, NCT ) showed a significant reduction in mortality among patients receiving ticagrelor, as compared with those receiving clopidogrel, with 6 to 12 months of therapy. 14 Although a periprocedural infusion of cangrelor would not be expected to have as profound a reduction in mortality against an active control (600 mg of clopidogrel, as was administered in the CHAMPION PCI trial), it is conceivable that such benefits might be apparent as compared with administering placebo, as was done in our study. Our study raises questions about the appropriate definition for and prognostic use of periprocedural myocardial infarction in patients entering a study with raised biomarkers at baseline. In such patients short times from admission to PCI prevent a clear delineation between myocardial infarction that occurred before randomization and myocardial infarction that occurred after randomization. In addition, although the inclusion of periprocedural myocardial infarction as an end point is useful to increase the number of events in clinical trials, its clinical relevance continues to be widely debated Our results do not show an effect of cangrelor on periprocedural myocardial infarction (as defined in this /nejmoa nejm.org

9 Intravenous Platelet Blockade with during PCI A Primary End Point Event Rate (%) No. at Risk P= Days % 8.9% B Stent Thrombosis % Event Rate (%) No. at Risk P=0.02 Hours % P= Days % 0.38% C Death P= P= % Event Rate (%) % % % 0.2 No. at Risk Hours Days trial) except in the subgroup of patients with a negative troponin test at baseline but do show an effect of cangrelor on more important outcomes (e.g., stent thrombosis and death). These findings should lead to further research on the most appropriate end points for clinical trials involv /nejmoa nejm.org 9

10 The new england journal of medicine Table 3. Bleeding Events at 48 Hours (Safety Population). Event Access-site bleeding requiring radiologic or surgical intervention Hematoma at puncture site (N = 2662) no. (%) (N = 2650) Odds Ratio (95% CI) P Value 8 (0.3) 10 (0.4) 0.80 ( ) cm 115 (4.3) 71 (2.7) 1.64 ( ) <5 cm 150 (5.6) 119 (4.5) 1.27 ( ) 0.06 Intracranial hemorrhage 2 (0.1) 1 (<0.1) 1.99 ( ) 0.57 Intraocular hemorrhage 0 0 Bleeding requiring surgery 1 (<0.1) 1 (<0.1) 1.00 ( ) 1.00 Retroperitoneal hemorrhage 2 (0.1) 1 (<0.1) 1.99 ( ) 0.57 Ecchymosis 95 (3.6) 57 (2.2) 1.68 ( ) Epistaxis 6 (0.2) 12 (0.5) 0.50 ( ) 0.16 Oozing at puncture site 125 (4.7) 91 (3.4) 1.39 ( ) 0.02 Thrombocytopenia 2 (0.1) 3 (0.1) 0.66 ( ) 0.65 Hemodynamic compromise 7 (0.3) 5 (0.2) 1.40 ( ) 0.57 Transfusion Any blood 26 (1.0) 16 (0.6) 1.62 ( ) 0.13 Platelet 4 (0.2) 2 (0.1) 1.99 ( ) 0.43 Red-cell 25 (0.9) 15 (0.6) 1.67 ( ) 0.12 Drop in hemoglobin or hematocrit* 33 (1.2) 35 (1.3) 0.94 ( ) 0.79 Category of bleeding ACUITY criteria Minor bleeding 320 (12.0) 246 (9.3) 1.34 ( ) Major bleeding 147 (5.5) 93 (3.5) 1.61 ( ) <0.001 GUSTO criteria Mild bleeding 427 (16.0) 310 (11.7) 1.44 ( ) <0.001 Moderate bleeding 20 (0.8) 13 (0.5) 1.54 ( ) 0.23 Severe or life-threatening bleeding 9 (0.3) 6 (0.2) 1.50 ( ) 0.45 TIMI criteria Minor bleeding 22 (0.8) 16 (0.6) 1.37 ( ) 0.34 Major bleeding 4 (0.2) 9 (0.3) 0.44 ( ) 0.17 * A drop in hemoglobin or hematocrit was defined as a decrease of at least 3 g per deciliter in hemoglobin or a 9% decrease in hematocrit after treatment, as compared with baseline, as reported by investigators. The bleeding options under each criterion are not mutually exclusive. For example, a patient may have had both major and minor bleeding on the basis of criteria from the Acute Catheterization and Urgent Intervention Triage Strategy (ACUITY) trial if more than one bleeding episode occurred. Each patient was counted only once for each criterion level, regardless of the number of bleeding events that were identified under each criterion. Listed bleeding events include bleeding associated with coronary-artery bypass grafting. GUSTO denotes Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries, and TIMI Thrombolysis in Myocardial Infarction. ing patients with acute coronary syndromes and positive biomarkers at baseline who undergo PCI. Taken together, the two CHAMPION trials may provide insight into the optimal timing of periprocedural antiplatelet blockade with clo pido grel. In the examination of secondary composite end points, there was a more robust effect in the CHAMPION PLATFORM trial (600 mg of /nejmoa nejm.org

11 Intravenous Platelet Blockade with during PCI clopidogrel at the end of the procedure) than in the CHAMPION PCI trial (600 mg of clopidogrel at the beginning of the procedure). Therefore, if one contrasts the results of the two CHAMPION trials, it appears that the 600 mg loading dose of clopidogrel may provide incremental benefit when given at the start of the procedure versus only at the end, though this conclusion remains speculative. However, even when clopidogrel is given at the start of the procedure, the additional antiplatelet blockade conferred by cangrelor may provide clinical benefit, as suggested by the reduction in secondary end points in the CHAMPION PCI trial and by the results of the platelet substudy in the two CHAMPION trials. 16 did not cause more major or minor bleeding on the basis of GUSTO and TIMI criteria. 22,23 However, it did cause more bleeding on the basis of the more sensitive ACUITY criteria for major bleeding and GUSTO criteria for mild bleeding, as one might expect for a potent antiplatelet agent, as compared with placebo. 24 Reassuringly, there was no significant increase in the rate of blood transfusion in the cangrelor group, as compared with the placebo group, and the excess in major bleeding on the basis of ACUITY criteria was because of an increased number of groin hematomas in the cangrelor group. The occurrence of transient dyspnea is intriguing, given a similar observation in the PLATO trial. 14 Perhaps this effect reflects the similar mechanisms of action of cangrelor and ticagrelor, which are both reversible platelet ADP-receptor antagonists that may act through other adenosine-dependent pathways as well. The early termination of this study does decrease its statistical power. The primary end point of this trial was negative; therefore, any additional end points, even prespecified ones, should be interpreted with caution but may serve to generate hypotheses for future investigations. The definition of periprocedural myocardial infarction that was used in this trial may not have been a good end point to detect an effect of cangrelor for patients who had an elevated troponin level at baseline and had rapid times to PCI. In conclusion, although the use of cangrelor in our study did not result in a significant reduction in the primary end point of death, myocardial infarction, or ischemia-driven revascularization in patients undergoing PCI, in exploratory analyses the rates of important prespecified secondary end points of stent thrombosis and death were reduced. Given the rapid effect on platelet inhibition seen in the CHAMPION platelet substudy, the reductions in these secondary end points are biologically plausible. Supported by the Medicines Company. Financial and other disclosures provided by the authors are available with the full text of this article at NEJM.org. We thank Penny Hodgson and Elizabeth Cook of the Duke Clinical Research Institute for their editorial support; and Medicines Company employees Jayne Prats and Meredith Todd for their assistance in constructing figures and appendices and Bo Gao for his collaboration on statistical analyses. Appendix The affiliations of the authors are as follows: the VA (Veterans Affairs) Boston Healthcare System and Brigham and Women s Hospital (D.L.B.) and Beth Israel Deaconess Medical Center (C.M.G.) all in Boston; Cleveland Clinic, Cleveland (A.M.L.); Columbia University Medical Center and the Cardiovascular Research Foundation, New York (G.W.S.); Interventional Cardiology and Cardiac Research, Hudson Valley Heart Center, Poughkeepsie, NY (M.Z.J.); Duke Clinical Research Institute, Duke University Medical Center, Durham, NC (S.M., K.W.M., R.A.H.); Institut de Cardiologie, Pitié Salpêtrière Hospital, Paris (G.M.); Methodist DeBakey Heart Center, Methodist Hospital, Houston (N.S.K.); Terrence Donnelly Heart Centre, Division of Cardiology, St. Michael s Hospital and the Canadian Heart Research Centre, Toronto (S.G.G.); Green Lane Cardiovascular Service, Auckland City Hospital, Auckland, New Zealand (H.D.W.); Pennsylvania Hospital, University of Pennsylvania, Philadelphia (C.V.P.); Sarah Cannon Research Institute and Hospital Corporation of America, Nashville (S.V.M.); Third Faculty of Medicine, Charles University, Prague, Czech Republic (P.W.); Flinders University/Flinders Medical Centre, Adelaide, SA, Australia (D.P.C.); the Department of Interventional Cardiology, Instituto Cardiovascular de Buenos Aires, Buenos Aires (F.C.); Clinic of Invasive Cardiology, St. George s University Hospital, Plovdiv, Bulgaria (I.M.); Regional Hospital, České Budějovice, Czech Republic (F.T.); Arneja Heart Institute, Nagpur, Maharashtra, India (J.A.); and the Medicines Company, Parsippany, NJ (S.S.). References 1. Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med 2001;345: Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with clopidogrel and aspirin followed by long-term therapy in patients undergoing percutaneous coronary intervention: the PCI-CURE study. Lancet 2001;358: Sabatine MS, Cannon CP, Gibson CM, et al. Addition of clopidogrel to aspirin and fibrinolytic therapy for myocardial infarction with ST-segment elevation. N Engl J Med 2005;352: Sabatine MS, Cannon CP, Gibson CM, et al. Effect of clopidogrel pretreatment before percutaneous coronary intervention in patients with ST-elevation myocardial /nejmoa nejm.org 11

12 Intravenous Platelet Blockade with during PCI infarction treated with fibrinolytics: the PCI-CLARITY study. JAMA 2005;294: Steinhubl SR, Berger PB, Mann JT III, et al. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention: a randomized controlled trial. JAMA 2002;288: [Erratum, JAMA 2003;289:987.] 6. Leon MB, Baim DS, Popma JJ, et al. A clinical trial comparing three antithrombotic-drug regimens after coronaryartery stenting. N Engl J Med 1998;339: Bhatt DL, Bertrand ME, Berger PB, et al. Meta-analysis of randomized and registry comparisons of ticlopidine with clopidogrel after stenting. J Am Coll Cardiol 2002;39: Meadows TA, Bhatt DL. Clinical aspects of platelet inhibitors and thrombus formation. Circ Res 2007;100: Stone GW, Lansky AJ, Pocock SJ, et al. Paclitaxel-eluting stent versus bare-metal stents in acute myocardial infarction. N Engl J Med 2009;360: Bavry AA, Bhatt DL. Appropriate use of drug-eluting stents: balancing the reduction in restenosis with the concern of late thrombosis. Lancet 2008;371: Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2007;357: Bhatt DL. Intensifying platelet inhibi- tion navigating between Scylla and Charybdis. N Engl J Med 2007;357: Idem. Prasugrel in clinical practice. N Engl J Med 2009;361: Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med 2009;361: Schömig A. Ticagrelor is there need for a new player in the antiplatelettherapy field? N Engl J Med 2009;361: Harrington RA, Stone GW, McNulty S, et al. Platelet inhibition with cangrelor in patients undergoing PCI. N Engl J Med 2009;361. DOI: /NEJMoa Mehta C, Gao P, Bhatt DL, Harrington RA, Skerjanec S, Ware JH. Optimizing trial design: sequential, adaptive, and enrichment strategies. Circulation 2009; 119: Harrington RA, Lincoff AM, Califf RM, et al. Characteristics and consequences of myocardial infarction after percutaneous coronary intervention: insights from the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT). J Am Coll Cardiol 1995;25: Bhatt DL, Topol EJ. Does creatinine kinase-mb elevation after percutaneous coronary intervention predict outcomes in 2005? Periprocedural cardiac enzyme elevation predicts adverse outcomes. Circulation 2005;112: Prasad A, Gersh BJ, Bertrand ME, et al. Prognostic significance of periprocedural versus spontaneously occurring myocardial infarction after percutaneous coronary intervention in patients with acute coronary syndromes: an analysis from the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) trial. J Am Coll Cardiol 2009;54: Prasad A, Rihal CS, Lennon RJ, Singh M, Jaffe AS, Holmes DR. Significance of periprocedural myonecrosis on outcomes after percutaneous coronary intervention: an analysis of preintervention and postintervention troponin T levels in 5487 patients. Circ Cardiovasc Intervent 2008;1: Medina HM, Bhatt DL. Evolution of anticoagulant and antiplatelet therapy: benefits and risks of contemporary pharmacologic agents and their implications for myonecrosis and bleeding in percutaneous coronary intervention. Clin Cardiol 2007;30:Suppl 2:II-4 II Rao AK, Pratt C, Berke A, et al. Thrombolysis in Myocardial Infarction (TIMI) Trial phase I: hemorrhagic manifestations and changes in plasma fibrinogen and the fibrinolytic system in patients treated with recombinant tissue plasminogen activator and streptokinase. J Am Coll Cardiol 1988;11: Stone GW, McLaurin BT, Cox DA, et al. Bivalirudin for patients with acute coronary syndromes. N Engl J Med 2006; 355: Copyright 2009 Massachusetts Medical Society /nejmoa nejm.org