Author + information
- Somjot S. Brar, MD, MPH∗ ()
- ↵∗Reprint requests and correspondence:
Dr. Somjot S. Brar, 4867 Sunset Boulevard, 3rd Floor, Room 3755, Los Angeles, California 90027.
- anticoagulant agents
- coronary artery disease
- percutaneous coronary intervention
- platelet aggregation inhibitors
Pharmacotherapy around the time of percutaneous coronary intervention (PCI) remains an area of intense research. This is evident by the fact that there are approximately 30 potential combinations of anticoagulant and antiplatelet agents, both oral and intravenous, which can be encountered during PCI. With Food and Drug Administration approval of cangrelor in 2015, the number of potential strategies doubled. What value does an additional medication confer in an already kaleidoscopic mix of therapeutic options?
Cangrelor, an adenosine triphosphate analog, is an intravenous direct-acting P2Y12 platelet receptor inhibitor that blocks adenosine diphosphate–induced platelet activation and aggregation. After initiation, platelet inhibition occurs within 2 min, and on discontinuation the antiplatelet effect normalizes within 1 h. These effects, the rapid and predictable platelet inhibition, and an average elimination half-life of approximately 3 to 6 min, are pharmacological characteristics that make cangrelor particularly appealing for use during PCI. For emerging medications like cangrelor, it is important to contextualize the efficacy and safety in the setting of established therapies, such as glycoprotein IIb/IIIa inhibitors (GPIs).
The report by Vaduganathan et al. (1) in this issue of the Journal provides important insight into the complex interplay among GPIs, cangrelor, and clopidogrel in patients undergoing PCI. The investigators performed a patient level meta-analysis, aggregating results from 3 large-scale contemporary trials (CHAMPION PCI [A Clinical Trial to Demonstrate the Efficacy of Cangrelor (PCI)], PLATFORM [Cangrelor Versus Standard Therapy to Achieve Optimal Management of Platelet Inhibition (Platform)], and PHOENIX [A Clinical Trial Comparing Cangrelor to Clopidogrel Standard Therapy in Subjects Who Require Percutaneous Coronary Intervention (PCI) (CHAMPION PHOENIX)]), yielding a pooled cohort of 25,384 patients. The aggregation of patient-level data often yields the highest quality meta-analysis by allowing for analyses of time to event data on the basis of allocated treatment using standardized outcomes. This approach also increases statistical power and is perhaps the only situation where it is reasonable to evaluate whether there are differences between treatments (with some caution) in clinically relevant or common patient subgroups.
There are noteworthy limitations to this approach as well. A point of concern is that GPI use in a clinical trial, as in clinical practice, can be driven by various factors, which can confound the results. The negative effects of confounding by indication, for GPI use in this case, can be lessened through stratification; that is, by separately analyzing the efficacy and safety of cangrelor versus clopidogrel for upfront and bailout GPI use. Stratified analyses are likely to be more informative and better align with established clinical practice patterns than those in which all GPI users, irrespective of indication, are grouped together. Appropriately, the investigators provide these analyses in their report and supplement. Also, among the trials in the pooled analysis, only the CHAMPION PCI trial allowed for upfront GPI use; thus, the pooled analysis provides little added insight into upfront GPI use with cangrelor. In contrast, approximately 88% of the pooled cohort was GPI-naive and eligible for bailout GPI use; thus, it is this group that most benefits from pooling of the trials.
Both cangrelor and GPI achieve >90% steady-state inhibition of platelet aggregation, albeit through different mechanisms. Therefore, the rationale for using 2 potent intravenous antiplatelet medications simultaneously, particularly in a planned fashion, is not clear. These data suggest that there does not seem to be any significant advantage to upfront GPI use with cangrelor. There were no significant differences in the primary efficacy or safety endpoints between cangrelor and clopidogrel, although there were few patients with upfront GPI use. Unlike with upfront GPI use, there was a significant reduction in the primary efficacy outcome with cangrelor in the GPI-naive group. The absolute risk difference for the primary efficacy endpoint was 0.8% in favor of cangrelor (4.0% vs. 4.8%; odds ratio: 0.82; 95% confidence interval: 0.72 to 0.93). It is reassuring that the event rates were numerically lower for all 4 of the components comprising the composite endpoint, and statistically significant for myocardial infarction (MI) and stent thrombosis. The absolute reduction in Q-wave MI, although statistically significant, was only 0.1%, suggesting that much of the benefit may have been caused by a reduction in post-procedural MI, characterized more so by biomarker elevation. Also, the stent thrombosis definition was protocol-specific; the number of Academic Research Consortium definite stent thrombosis events is not clearly reported. In aggregate, these data suggest cangrelor is most beneficial for reducing ischemic events in GPI-naive patients.
This report also highlights the critical importance of the bleeding definition used in evaluating drug safety. Using the ACUITY (Acute Catheterization and Urgent Intervention Triage Strategy) bleeding definition, the absolute increase in major bleeds with cangrelor was 1.6%, corresponding to a 71% increase in the odds compared with clopidogrel (odds ratio: 1.71; 95% confidence interval: 1.46 to 1.99; p = 0.003) in GPI-naive patients. Using this definition, the bleeding rate exceeds the 0.8% absolute reduction in ischemic events with cangrelor. In addition, cangrelor was also associated with significant increases in GUSTO (Global Use of Strategies to Open Occluded Arteries) severe or moderate bleeding, TIMI (Thrombolysis in Myocardial Infarction) major or minor bleeding, and ACUITY major or minor bleeding. Blood transfusions were also greater with cangrelor in GPI-naive subjects. These observations are consistent with the more potent antiplatelet effect of cangrelor compared with clopidogrel. Thus, although cangrelor did not increase the rate of GUSTO severe or life-threatening bleeding, the primary safety endpoint, the totality of these data suggest there is some increase in bleeding in GPI-naive patients, who represent most of the patients in this analysis and are the patients most likely to be treated with cangrelor. It should be noted that these trials excluded patients at heightened risk for bleeding.
The pooled analysis also shows that GPIs significantly increase bleeding with both cangrelor and clopidogrel. The bleeding rates across multiple definitions and the need for blood transfusion were significantly greater with upfront GPI use compared with bailout use. This may be explained, in part, by the prolonged infusion period with GPI. The low rates of GPI use during PCI in these studies are consistent with a more general global trend. Factors that may be driving this include a greater appreciation for the significance of bleeding, combined with the advent of newer, more potent antiplatelet medications, and uncertainty regarding the prognostic significance of periprocedural MI. The present report shows significant regional variations in practice patterns, with about two-thirds of GPI use in the pooled cohort occurring in the United States—a disparity that may be caused, in part, by tepid recommendations for upfront GPI use in international guidelines (2).
Further investigation is needed to better contextualize the role of cangrelor in the setting of other medications used during PCI. Whether there is added benefit to using cangrelor with newer antiplatelet agents, such as ticagrelor and prasugrel, remains unknown. Also, decision making at the point of care may be challenging without a better accounting of the prognostic significance of the endpoints of interest. For the primary endpoints, the pooled analysis shows that the ischemic benefit of cangrelor over clopidogrel exceeded the bleeding rate in the GPI-naive study patients. However, the prognostic significance of these endpoints, periprocedural MI in particular, is the subject of ongoing controversy. The ischemic benefit with cangrelor seems to be driven, to a significant extent, by a reduction in periprocedural MI. Although limiting MI during PCI is likely of benefit, the threshold for a worsening prognosis associated with increasing cardiac biomarkers without procedural complications is ill-defined. Recent analyses from multiple acute coronary syndrome trials show no association of periprocedural MI with long-term mortality, and a position paper suggests a higher threshold of biomarker elevation be considered than was used in the present analysis (3–5). Moreover, these data also suggest that early bleeding post-PCI may be a much stronger predictor of long-term mortality than periprocedural MI, and second only to spontaneous MI (4). Thus, clarity around these issues will help better contextualize the role of cangrelor during PCI. With unique and advantageous pharmacokinetics, cangrelor is a welcome addition to the multiplex of pharmacological options available during PCI, but further work is needed to better understand what role it will play in PCI pharmacology.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Dr. Brar has reported he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Vaduganathan M.,
- Harrington R.A.,
- Stone G.W.,
- et al.,
- on behalf of the CHAMPION Investigators
- Roffi M.,
- Patrono C.,
- Collet J.P.,
- et al.
- Damman P.,
- Wallentin L.,
- Fox K.A.,
- et al.
- Prasad A.,
- Gersh B.J.,
- Bertrand M.E.,
- et al.
- Moussa I.D.,
- Klein L.W.,
- Shah B.,
- et al.