We evaluated the interaction between protease-activated receptor-1 antagonist vorapaxar and concomitant glycoprotein (GP) IIb/IIIa receptor inhibitors in patients with non–ST-segment elevation acute coronary syndromes who underwent PCI. In Thrombin Receptor Antagonist for Clinical Event Reduction in Acute Coronary Syndrome trial, 12,944 patients with non–ST-segment elevation acute coronary syndromes were randomized to vorapaxar or placebo. Administration of GP IIb/IIIa receptor inhibitors was allowed at the treating physician’s discretion. We investigated whether use of GP IIb/IIIa receptor inhibitors modified vorapaxar’s effect on non–coronary artery bypass grafting (CABG)-related bleeding at 7 days and ischemic events at 30 days. In total, 7,455 patients underwent PCI during index hospitalization. Of these, 2,023 patients (27.1%) received inhibitors and 5,432 (72.9%) did not. Vorapaxar was associated with a numerically higher rate of non–CABG-related moderate/severe Global Use of Strategies to Open Occluded Arteries (GUSTO) bleeding at 7 days compared with placebo in those who did (1.3% vs 1.0%) and did not (0.6% vs 0.4%) receive GP IIb/IIIa receptor inhibitors. Ischemic end point rates at 30 days were not significantly lower with vorapaxar versus placebo. Increased rates of non-CABG GUSTO moderate/severe bleeding were observed in patients who received GP IIb/IIIa receptor inhibitors versus those who did not (adjusted hazard ratio [HR] 1.77, 95% confidence interval [CI] 0.43 to 7.35 in placebo arm; adjusted HR 2.02, 95% CI 0.62 to 6.61 in vorapaxar arm) and in those who received vorapaxar versus placebo (adjusted HR 1.54, 95% CI 0.36 to 6.56 in the GP IIb/IIIa group; adjusted HR 1.34, 95% CI 0.44 to 4.07 in the no–GP IIb/IIIa group). No interaction was found between vorapaxar and inhibitor use up to 7 days (P interaction = 0.89) nor at the end of the treatment (P interaction = 0.74); however, the event rate was low. Also, no interaction was observed for efficacy end points after PCI at 30 days or at the end of the treatment. In conclusion, GP IIb/IIIa receptor inhibitor use plus dual antiplatelet therapy in a population with non–ST-segment elevation myocardial infarction planned for PCI was frequent but did not interact with vorapaxar’s efficacy or safety. Nonetheless, GP IIb/IIIa receptor inhibitors and vorapaxar were associated with increased bleeding risk, and their combined use may result in additive effects on bleeding rates.
The Thrombin Receptor Antagonist for Clinical Event Reduction in Acute Coronary Syndrome (TRACER) trial was a large, international, phase III, clinical trial that evaluated the efficacy and safety of vorapaxar—an oral, selective, protease-activated receptor (PAR)-1 antagonist—compared with placebo in patients presenting with non–ST-segment elevation acute coronary syndromes (NSTE ACS) who were treated with standard of care. Vorapaxar administration results in a potent inhibition of PAR-1–mediated platelet aggregation. In TRACER, vorapaxar significantly increased the risk of major bleeding, and the primary end point (a composite of cardiovascular death, myocardial infarction [MI], stroke, coronary ischemia with hospitalization, or urgent coronary revascularization) was not significantly reduced; however, vorapaxar reduced the rate of MI, particularly spontaneous (type 1).
In TRACER, the decision to use concomitant antiplatelet therapies, including aspirin, P2Y 12 inhibitors, and glycoprotein (GP) IIb/IIIa receptor inhibitors, was left to the treating physician, based on applicable guidelines. Current guidelines suggest that GP IIb/IIIa receptor inhibitor use in selected patients who underwent percutaneous coronary intervention (PCI) is reasonable, depending on high ischemic and bleeding risk (class Ib recommendation according to the European Society of Cardiology, class IIa/IIb recommendation according to the American Heart Association). However, the challenge is to balance the effectiveness of suppression of ischemic events with any additional bleeding risk, especially because of increasing understanding of the importance of bleeding and its effects on long-term outcomes. We designed the current analysis to evaluate the interaction between GP IIb/IIIa receptor inhibitors and PAR-1 antagonist vorapaxar and effects on early bleeding and ischemic events in patients with NSTE ACS who underwent PCI.
Methods
The details of the TRACER trial have been published and are briefly summarized here. The TRACER trial randomized 12,944 patients presenting with NSTE ACS. Key inclusion criteria were acute coronary ischemia–related symptoms within 24 hours of hospital presentation and at least 1 of the following: (1) cardiac troponin (I or T) or creatine kinase-MB greater than the local upper limit of normal, (2) new ST-segment depression >0.1 mV, or (3) transient (<30 minutes) ST-segment elevation >0.1 mV in at least 2 contiguous leads. In addition, ≥1 of the following 4 criteria were required: age ≥55 years; previous MI, PCI, or coronary artery bypass grafting (CABG); diabetes mellitus; or peripheral artery disease (PAD). Patients were randomized to receive either vorapaxar (40-mg loading dose and a 2.5-mg maintenance dose) or matching placebo in addition to standard of care with stratification according to the intention to use a GP IIb/IIIa receptor inhibitor (vs none) and the intention to use a parenteral direct thrombin inhibitor (vs other antithrombin agents). Study treatment was to be continued for the entire duration of the trial and for a minimum of 1 year. Decisions regarding concomitant treatment and need for invasive procedures and revascularization were left to the discretion of the treating physician, who was encouraged to follow applicable guidelines. Importantly, there was no restriction on the use of concomitant antiplatelet medications, including aspirin, thienopyridines (mostly clopidogrel), or GP IIb/IIIa receptor inhibitors. The primary end point of TRACER was a composite of cardiovascular death, MI, stroke, rehospitalization for coronary ischemia, or urgent coronary revascularization. The key secondary end point was a composite of cardiovascular death, MI, or stroke. The primary safety end points were Global Use of Strategies to Open Occluded Arteries (GUSTO) moderate or severe bleeding or Thrombolysis In Myocardial Infarction (TIMI) bleeding classified as major, minor, or requiring medical attention (TIMI clinically significant bleeding). A blinded central clinical events committee assessed all suspected efficacy and bleeding events.
To investigate whether use of GP IIb/IIIa receptor inhibitors modified vorapaxar’s effect, patients who received at least 1 dose of the study medication and underwent a PCI for treatment of index NSTE ACS were studied. Administration of GP IIb/IIIa receptor inhibitors was allowed at the treating physician’s discretion. Because GP IIb/IIIa receptor inhibitors have a short duration of effect and are known to have an effect on short-term outcomes (up to 30 days), we investigated whether use of GP IIb/IIIa receptor inhibitors modified vorapaxar’s effect on the occurrence of non–CABG-related bleeding at 7 days and ischemic events at 30 days. The 7-day end point for bleeding was chosen to answer the question of whether GP IIb/IIIa receptor inhibitors have influenced the early bleeding signal seen with vorapaxar. Although the study design included prerandomization stratification based on planned use of GP IIb/IIIa receptor inhibitors, we observed a large amount of crossover between planned and actual use of GP IIb/IIIa receptor inhibitors in the study, and therefore, patients were classified and analyzed according to 2 groups based on the actual use of GP IIb/IIIa receptor inhibitors between randomization and PCI. This analysis was prespecified based on GP IIb/IIIa receptor inhibitor stratification and subsequently modified given the crossover rate. GP IIb/IIIa receptor inhibitor use was defined as infusion of eptifibatide, tirofiban, or abciximab between randomization and PCI, including upstream use and provisional use during PCI. Because the time in which use of GP IIb/IIIa receptor inhibitors would occur included up until PCI, time for event accrual started at the time of PCI. As a consequence, end points were not included before PCI. At 7 days after PCI, the following primary safety end points were assessed: composite of non–CABG-related moderate and severe GUSTO bleeding and non-CABG TIMI clinically significant bleeding. GUSTO moderate bleeding is bleeding requiring transfusion of whole blood or packed red blood cells without hemodynamic compromise. GUSTO severe bleeding includes bleeding that was fatal and intracranial or that caused hemodynamic compromise requiring intervention (e.g., systolic blood pressure <90 mm Hg that required blood or fluid replacement, vasopressor/inotropic support, or surgical intervention). TIMI clinically significant bleeding includes major, minor, and bleeding requiring medical attention. For the purpose of this analysis, the following efficacy end points were assessed at 30 days after PCI: (1) composite of cardiovascular death, MI, stroke, rehospitalization for coronary ischemia, or urgent coronary revascularization; (2) cardiovascular death, MI, or stroke; and (3) composite of types 1 and 4a MI. As secondary safety and efficacy analyses, all end points were also evaluated from the index PCI procedure up until the end of the treatment period.
Baseline characteristics were summarized for the 2 groups (determined by GP IIb/IIIa receptor inhibitor use between randomization and PCI). Continuous variables are presented as medians (interquartile ranges), and categorical variables are presented as counts (proportions). Event counts and unadjusted Kaplan-Meier estimates are presented by GP IIb/IIIa receptor inhibitor use and randomized treatment for each primary and secondary efficacy and safety end point. Time zero for all end points is the date of PCI. Before Cox modeling, an assumption of a linear association between continuous covariates and the outcomes was tested, and splines were fit as appropriate. To assess the adjusted relation between GP IIb/IIIa receptor inhibitor use and clinical outcomes and to determine whether GP IIb/IIIa receptor inhibitor use modifies treatment effect, Cox proportional hazard models were fit. The models included a term for the interaction of GP IIb/IIIa receptor inhibitors and vorapaxar and additional covariates for efficacy (weight, diabetes, stroke, PAD, creatinine clearance, positive biomarkers, previous CABG, systolic blood pressure, previous MI, baseline Killip class, and heart rate) and safety (age, gender, weight, hypertension, diabetes, stroke, PAD, current smoking, statin, clopidogrel, aspirin, creatinine clearance, ST-segment depression, Killip class, and hemoglobin). These variables were derived from a literature search and from clinical judgment. All variables for each end point were included in the models collectively, plus the interaction term for vorapaxar and GP IIb/IIIa receptor inhibitors, without a formal selection process.
All statistical tests were performed at a significance level of 0.05. All analyses were performed by an independent statistician (RMC) at the Duke Clinical Research Institute (Durham, North Carolina) using SAS software, version 9.3 (SAS Institute Inc., Cary, North Carolina).
Results
In total, 7,466 patients underwent PCI during index hospitalization. Patients treated with thrombolytic therapy were excluded (n = 11). Among the remaining 7,455 patients, 2,023 patients (27.1%) received GP IIb/IIIa receptor inhibitors before or on the same day of PCI start, and 5,432 (72.9%) did not ( Table 1 ). Table 2 lists treatment during hospitalization by GP IIb/IIIa receptor inhibitor use. Among patients who did receive GP IIb/IIIa receptor inhibitors, infusion was started upstream at least 1 day before PCI in 689 patients (34%) (mean 1.8 ± 1.8 days before PCI), whereas inhibitors were started on the same day as the PCI procedure in 1,334 patients (66%). The loading dose of the randomized treatment in TRACER was given a median of 3.6 hours (interquartile range 1.9 to 21.6 hours) before PCI. Predominantly high-risk patients with ACS were treated with PCI (>99% had a TIMI risk score ≥3). GP IIb/IIIa receptor inhibitor use was higher in North America, and patients who received GP IIb/IIIa receptor inhibitors were younger, had less chronic kidney disease, and were more often men.
| Variable | Glycoprotein IIb/IIIa Use | |
|---|---|---|
| No (N = 5432) | Yes (N = 2023) | |
| Age (years) | 64 (58, 72) | 63 (57, 70) |
| ≥75 | 993 (18.3%) | 265 (13.1%) |
| Women | 1389 (25.6%) | 427 (21.1%) |
| White | 4604 (84.8%) | 1851 (91.5%) |
| Body weight (kg) | 80 (70, 91) | 84 (74, 95) |
| Hemoglobin (g/dL) | 14 (13, 15) | 14 (13, 15) |
| Region | ||
| North America | 971/1770 (54.9%) | 799/1770 (45.1%) |
| South America | 286/358 (79.9%) | 72/358 (20.1%) |
| Western Europe | 2562/3544 (72.3%) | 982/3544 (27.7%) |
| Eastern Europe | 894/989 (90.4%) | 95/989 (9.6%) |
| Asia/Australia/New Zealand | 719/794 (90.6%) | 75/794 (9.4%) |
| Cardiovascular risk factors/history | ||
| Hypertension | 3791 (69.8%) | 1370 (67.7%) |
| Hyperlipidemia | 3303/5429 (60.8%) | 1310 (64.8%) |
| Tobacco use | 1759/5430 (32.4%) | 654 (32.3%) |
| Diabetes mellitus | 1636/5431 (30.1%) | 634 (31.3%) |
| Stroke | 246 (4.5%) | 66 (3.3%) |
| Myocardial infarction | 1484 (27.3%) | 575 (28.4%) |
| Percutaneous coronary intervention | 1331/5430 (24.5%) | 498/2022 (24.6%) |
| Coronary bypass | 576 (10.6%) | 288 (14.2%) |
| Peripheral artery disease | 409 (7.5%) | 137/2023 (6.8%) |
| Creatinine clearance (mL/min) | 95 (73, 121) | 102 (80, 127) |
| Positive troponin or creatinine kinase-MB | 5065/5392 (93.9%) | 1934/2010 (96.2%) |
| Thrombolysis In Myocardial Infarction risk score | ||
| 0–2 | 23 (0.4%) | 13 (0.6%) |
| 3–4 | 2835 (52.2%) | 1001 (49.5%) |
| 5–7 | 2574 (47.4%) | 1009 (49.9%) |
| Variable | Glycoprotein IIb/IIIa Use | |
|---|---|---|
| No (N = 5432) | Yes (N = 2023) | |
| Randomized treatment | 2732 (50.3%) | 1019 (50.4%) |
| Intent to use GPI | 527 (9.7%) | 1080 (53.4%) |
| Clopidogrel | 5315 (97.8%) | 1987 (98.2%) |
| Aspirin | 5395 (99.3%) | 2006 (99.2%) |
| PCI ≤24 hours of randomization | 4182 (77.0%) | 1711 (84.6%) |
| Stenting | 5117 (94.2%) | 1936 (95.7%) |
| Drug-eluting stent | 2810 (51.7%) | 1191 (58.9%) |
| Bare-metal stent | 2521 (46.4%) | 839 (41.5%) |
| Coronary bypass | 29 (0.5%) | 12 (0.6%) |
At 7 days, the rates of non–CABG-related moderate or severe GUSTO bleeding and non-CABG TIMI clinically significant bleeding were numerically higher in patients who received GP IIb/IIIa receptor inhibitors compared with those who did not (GUSTO 1.1% [23 of 2,023] vs 0.5% [26 of 5,432], p = 0.002; TIMI 3.2% [64 of 2,023] vs 2.6% [139 of 5,432], p = 0.15).
In patients who received a GP IIb/IIIa receptor inhibitor, non–CABG-related moderate or severe GUSTO bleeding event rates at 7 days were 1.3% with vorapaxar compared with 1.0% with placebo. In patients who did not receive a GP IIb/IIIa receptor inhibitor, rates were 0.6% versus 0.4% with vorapaxar and placebo, respectively ( Table 3 , Figure 1 ). Unadjusted Kaplan-Meier curves for the occurrence of non–CABG-related moderate or severe GUSTO bleeding at 7 days and the composite of cardiovascular death, MI, or stroke at 30 days are presented in Figure 1 . Rates of the ischemic end points at 30 days between vorapaxar and placebo showed a small nonsignificant treatment effect in the group that did not receive GP IIb/IIIa receptor inhibitors, which was not present in the GP IIb/IIIa receptor inhibitor group. The primary composite end point rate at 30 days was 2.8% with vorapaxar and 3.6% with placebo in those who did not receive a GP IIb/IIIa inhibitor. In patients who were treated with a GP IIb/IIIa receptor inhibitor, the primary composite end point rate was 4.1% with vorapaxar and 3.4% with placebo ( Table 3 ). After adjusting for known risk factors, the interaction between randomized treatment and GP IIb/IIIa inhibitor use was not significant (p = 0.14).
| Variable | Glycoprotein IIb/IIIa Use | |
|---|---|---|
| No | Yes | |
| N, Kaplan-Meier rate (95% CI) | ||
| Non-CABG GUSTO moderate or severe bleeding | ||
| Vorapaxar | 16, 0.6 (0.4–1.0) | 13, 1.3 (0.7–2.2) |
| Placebo | 10, 0.4 (0.2–0.7) | 10, 1.0 (0.5–1.9) |
| Non-CABG TIMI clinically significant bleeding | ||
| Vorapaxar | 78, 2.9 (2.3–3.6) | 37, 3.6 (2.7–5.0) |
| Placebo | 61, 2.3 (1.8–2.9) | 27, 2.7 (1.9–3.9) |
| CV death, MI, stroke, rehospitalization for coronary ischemia, or urgent coronary revascularization | ||
| Vorapaxar | 76, 2.8 (2.2–3.5) | 41, 4.1 (3.0–5.5) |
| Placebo | 95, 3.6 (2.9–4.3) | 34, 3.4 (2.5–4.8) |
| CV death, MI, or stroke | ||
| Vorapaxar | 61, 2.3 (1.8–2.9) | 29, 2.9 (2.0–4.1) |
| Placebo | 71, 2.7 (2.1–3.3) | 27, 2.7 (1.9–3.9) |
| Type 1 and 4a MI | ||
| Vorapaxar | 36, 1.3 (1.0–1.8) | 11, 1.1 (0.6–2.0) |
| Placebo | 26, 1.0 (0.7–1.4) | 14, 1.4 (0.8–2.4) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
