Coronary revascularisation: Cangrelor

Drug action

Cangrelor is an intravenous, direct P2Y12 platelet receptor inhibitor that blocks adenosine diphosphate (ADP)‑induced platelet activation and aggregation. Cangrelor provides rapid‑onset P2Y12 inhibition (within a few minutes) with reversible binding and has a half‑life of 3 to 6 minutes.

Licensed therapeutic indication

Cangrelor (Kengrexal) is licensed for reducing the risk of thrombotic cardiovascular events in adults with coronary artery disease undergoing PCI who have not received an oral P2Y12 inhibitor prior to the PCI procedure and in whom oral therapy with P2Y12 inhibitors is not feasible or desirable. It is licensed only for co‑administration with aspirin. Oral P2Y12 inhibitors include clopidogrel, prasugrel and ticagrelor.

Course and cost

The Kengrexal summary of product characteristics (SPC) states that the recommended dose of cangrelor is a 30 micrograms/kg intravenous bolus followed immediately by 4 micrograms/kg per minute intravenous infusion. The bolus and infusion should be initiated prior to the procedure and continued for at least 2 hours or for the duration of the procedure, whichever is longer. At the discretion of the physician, the infusion may be continued for a total duration of 4 hours. For continuation of long‑term treatment, people should be changed onto oral P2Y12 therapy post‑procedure. For transition to this, a loading dose of oral P2Y12 therapy (clopidogrel, prasugrel or ticagrelor) should be administered immediately following discontinuation of cangrelor infusion. Alternatively, a loading dose of ticagrelor or prasugrel, but not clopidogrel, may be administered up to 30 minutes before the end of the infusion. No dose adjustment of cangrelor is required in people aged 75 years or older or in people with renal or hepatic impairment, but caution is advised when administering cangrelor to people with severe renal impairment (creatinine clearance 15–30 mL/minute).

Cangrelor is available as a 50 mg powder for concentrate for solution for infusion and costs £250 per vial.

Table 1 Summary of CHAMPION PHOENIX (Bhatt et al. 2013)

Table 2 Safety data from a combined pooled analysis of patient level data of the 3 CHAMPION studies (Steg et al. 2013)

Clinical effectiveness

Cangrelor bolus plus infusion followed by transition to a loading dose of clopidogrel statistically significantly reduced the risk of periprocedural ischaemic events compared with a single loading dose of clopidogrel in a large RCT of people receiving periprocedural aspirin who underwent PCI for mixed indications without P2Y12 inhibitor pre‑treatment. It did not statistically significantly reduce mortality, and the positive results were mainly driven by a reduction in the rate of acute periprocedural MI and stent thrombosis. The number needed to treat for the primary outcome of death from any cause, MI, ischaemia‑driven revascularisation or stent thrombosis at 48 hours was 84 (Bhatt et al. 2013). The incidence of MI and of stent thrombosis at 48 hours was also statistically significantly lower in the cangrelor group compared with the clopidogrel group (absolute risk reduction [ARR] 0.9%, p=0.02 and 0.6%, p=0.01, respectively, see table 1, Bhatt et al. 2013). Benefits for people in the cangrelor group compared with clopidogrel, in terms of reduction in the odds of ischaemic events following PCI, were maintained at 30 days (see table 1, Bhatt et al. 2013).

The first 2 studies in the phase III clinical evaluation of the efficacy of cangrelor, CHAMPION PLATFORM (Bhatt et al. 2009) and CHAMPION PCI (Harrington et al. 2009) were terminated early as interim analysis showed that cangrelor did not statistically significantly reduce the primary outcome of periprocedural events in PCI (a composite of death from any cause, MI or ischaemia‑driven revascularisation at 48 hours). The third, larger RCT, CHAMPION PHOENIX (Bhatt et al. 2013) used a narrower definition of MI. The authors state that this increased the sensitivity of the analysis to discriminate between events occurring before and after randomisation. They suggest that this is why the clinical benefits of cangrelor in the 2 previous studies were limited to secondary outcomes (primarily stent thrombosis at 48 hours), whereas in CHAMPION PHOENIX cangrelor was statistically significantly superior to clopidogrel for the primary composite outcome (see table 1 above).

Safety and tolerability

The safety population from the pooled analysis (Steg et al. 2013) provides data for a large number of people (n=25,107) approximately half of whom were exposed to cangrelor (see table 2). All participants in the safety population were followed up for 30 days and more than half were followed up for up to 1 year. The EPAR states that the safety database is adequate for the licensed indication and the maximum duration (4 hours) of treatment.

The CHAMPION studies evaluated bleeding events according to 3 definitions: GUSTO, TIMI and ACUITY. GUSTO and TIMI criteria have been used for several decades and were initially used to identify significant bleeding in the setting of fibrinolytic therapy for STEMI. ACUITY criteria have been developed by adapting the components of TIMI major and GUSTO severe/moderate bleeding that are relevant to patients undergoing PCI (Mehran et al, 2011).

The EPAR states that the results from the CHAMPION studies show that there was a statistically significant higher bleeding risk associated with cangrelor (17.5%) compared with clopidogrel (13.5%) at 48 hours (see table 2). There was no statistically significant difference in the rate of GUSTO severe non‑CABG‑related bleeding at 48 hours (see table 2). However, there was a statistically significant increase in ACUITY major bleeding in the cangrelor group (4.2% compared with 2.8%, absolute risk difference 1.4%, OR 1.53, 95% CI 1.34 to1.76, p<0.0001). This was in part due to an increased number of haematomas at the puncture site that were 5 cm or larger, but persisted after exclusion of these haematomas (1.3% compared with 1.0%, OR 1.38, 95% CI 1.09 to 1.74, p=0.007). The number needed to harm for ACUITY major bleeding at 48 hours was 71, and for ACUITY major bleeding at 48 hours excluding puncture site haematomas 5 cm or larger it was 333.

There was no statistically significant difference in GUSTO moderate bleeding, in TIMI major bleeding, or in the rate of transfusions. Cangrelor did increase the rate of less severe bleeding events such as GUSTO mild bleeding (ARR 3.8%, OR 1.35, 95% CI 1.26 to 1.45, p<0.0001), TIMI minor bleeding or ACUITY minor bleeding compared with clopidogrel. In the pooled analysis (Steg et al. 2013) the number needed to harm for GUSTO mild bleeding at 48 hours was 26. Fatal bleeding occurred in 0.1% of participants in each group.

Transient dyspnoea was also reported more frequently with cangrelor compared with clopidogrel (1.1% compared with 0.4%, p<0.0001). The EPAR states that the higher incidence of breathlessness reported with cangrelor than with clopidogrel is a class effect for direct P2Y12 inhibitors and has been previously reported with ticagrelor. Serious adverse events were infrequent and reported at the same frequency in both the cangrelor and comparator groups (2.2%).

The SPC states that in pivotal studies conducted in patients undergoing PCI, there were more intracranial bleeds, including fatal bleeds, at 30 days with cangrelor (0.07%) than with clopidogrel (0.02%), and a higher rate of cardiac tamponades at 30 days with cangrelor (0.12%) than with clopidogrel (0.02%). In patients with severe renal impairment (creatinine clearance 15–30 mL/min) a higher rate of worsening in renal function (3.2%) was reported with cangrelor compared to clopidogrel (1.4%) and a higher rate of GUSTO moderate bleeding was reported with cangrelor group (6.7%) compared to clopidogrel (1.4%).

Cangrelor is contraindicated in people with active bleeding or increased risk of bleeding and in people with any history of stroke or transient ischaemic attack. The SPC advises that cangrelor should be used with caution in people with severe renal impairment (creatinine clearance 15–30 mL/min). The SPC gives further information on contraindications, potential interactions and adverse effects of cangrelor.

Evidence strengths and limitations

The large phase III RCT CHAMPION PHOENIX, (Bhatt et al. 2013) provides patient orientated outcome data suggesting clinical benefit for some people undergoing PCI receiving cangrelor bolus plus infusion followed by transition to a loading dose of clopidogrel compared with a single loading dose of clopidogrel. This benefit was described as modest by the Committee for medicinal products for human use (CHMP, CHMP meeting minutes, January 2015). Treatment with cangrelor did not increase the risk of severe bleeding events according to 1 definition (GUSTO), but rates of severe bleeding were increased according to a different definition that is particularly intended for use in people undergoing PCI (ACUITY). Mild and moderate bleeding and transient breathlessness were more common in people receiving cangrelor compared with clopidogrel or placebo.

The EPAR states that the study design of CHAMPION PHOENIX was not optimal and may have favoured cangrelor, although it was accepted as reliable and representative of the diversity of current EU clinical practice. The EPAR notes that inclusion in the study population of 3 subsets of people with differential diagnosis is questionable due to the different risk factors and standard of care for each condition. It states that the results for the composite outcome were only statistically significantly superior with cangrelor for the people with stable angina. However, CHAMPION PHOENIX was not powered to show superiority for all 3 main subsets of participants: the EPAR states that the size of the effect of cangrelor appeared consistent irrespective of the initial presentation and that the results can be generalised to all 3 patient groups. Nevertheless, the EPAR also states that the exclusion of people with a history of stroke limits the validity of the study to clinical practice.

Furthermore, the loading dose of clopidogrel in the control arm was either 300 mg or 600 mg at the discretion of the site investigator, but a fixed 600 mg was given in the cangrelor group. Clopidogrel 300 mg is the licensed loading dose however the EPAR states that a higher, 600 mg dose has been shown to be more effective and this is advocated in ESC guidelines. Specialists involved in the production of this evidence summary advise that in UK clinical settings a loading dose of clopidogrel 600 mg is common practice when this drug is used. Moreover, in CHAMPION PHOENIX the clopidogrel loading dose was given at the end of the PCI procedure in 30% of people in the control arm. According to Steg et al. 2013, deferred administration of clopidogrel at the end of the PCI procedure could have magnified the difference in platelet inhibition between the study arms, although the EPAR states that the effect of cangrelor seemed consistent irrespective of the dose or timing of clopidogrel administration.

In addition to these issues with the study design, the treatment pathway in CHAMPION PHOENIX (Bhatt et al. 2013) differs from usual UK practice regarding choice of antiplatelet drug. Specialists involved the production of this evidence summary advise that in UK practice prasugrel and ticagrelor have superseded clopidogrel as the standard of care for people with unstable angina, NSTEMI and STEMI. This limits the applicability to UK practice of evidence from CHAMPION PHOENIX and the wider CHAMPION programme. Prasugrel and ticagrelor have positive NICE technology appraisals and their use is advocated in ESC guidelines as the standard of care in people at higher risk of ischaemic events, requiring immediate PCI. An editorial (Mehta 2013) accompanying the pooled analysis (Steg et al. 2013) points out that there are no data for how cangrelor compares with either prasugrel or ticagrelor and the EPAR states that these may have been more valid comparators.

Alternative treatments

Cangrelor, co‑administered with aspirin, is indicated for the reduction of thrombotic cardiovascular events in adults with coronary artery disease undergoing PCI who have not received an oral P2Y12 inhibitor prior to the PCI procedure and in whom oral therapy with P2Y12 inhibitors is not feasible or desirable.

The oral P2Y12 inhibitors clopidogrel, prasugrel and ticagrelor are licensed for use, in combination with aspirin, for the prevention of atherothrombotic events in people with NSTE‑ACS and STEMI including those undergoing PCI (see summaries of product characteristics for specific licensed indications). Price comparisons with these agents are included for context although cangrelor is an option only when use of these agents is not feasible or desirable.

Costs of alternative treatments

Likely place in therapy

During assessment of the marketing authorisation application for cangrelor, CHMP stated that cangrelor is a suitable alternative for the existing antiplatelet agents, especially in situations where an ad hoc PCI would be considered in people who have not yet received dual antiplatelet therapy. CHMP stated that the intravenous administration of cangrelor and its fast offset of action can be useful in selected people. The benefit of cangrelor was considered modest by the CHMP, mainly a reduction of periprocedural MI and stent thrombosis with an acceptable bleeding risk. CHMP agreed that, for people already pre‑treated with an oral P2Y12 inhibitor, cangrelor was unsuitable (CHMP meeting minutes. January 2015). Any potential clinical benefits must be considered in the context of potential harms and results from the 3 CHAMPION RCTs show an increased risk of bleeding and breathlessness with cangrelor compared with clopidogrel. Specialists involved in the production of this evidence summary advise that cangrelor, co‑administered with aspirin, maybe an alternative option for some people undergoing PCI and for whom oral P2Y12 inhibitors are not feasible or desirable (as required within the licensed indication). For example:

Two further caveats apply to the use of cangrelor. Firstly, in CHAMPION PHOENIX (Bhatt et al. 2013) almost all participants received periprocedural aspirin and this is reflected in the licensed indication. Only oral forms of aspirin are available as licensed products in the UK, thus the person would usually be conscious and able to take aspirin orally (or have taken it that day) and have no contraindications to its use. Administering cangrelor without periprocedural aspirin would be an off‑label use of the drug. (Note that prasugrel and ticagrelor are also both licensed only for use when co‑administered with aspirin).

Secondly, the SPC states that when clopidogrel is administered during infusion of cangrelor, the expected inhibitory effect of clopidogrel on platelets is not achieved. Consequently it is important that the loading dose of clopidogrel is delayed until the infusion finishes. The SPC states that no clinically relevant interruption of P2Y12 inhibition was observed in phase III studies when 600 mg clopidogrel was administered immediately after discontinuation of the cangrelor infusion, but this is an unlicensed dose of clopidogrel.

As well as efficacy, safety and individual user factors, local decision makers will need to take cost into account when considering the likely place in therapy of cangrelor.

No estimate of the current UK use of oral antiplatelets specifically for use during PCI procedures was available at the time this evidence summary was prepared. In research conducted with 45 UK interventional cardiologists by The Medicines Company in 2014, 14–17% of ACS patients were documented as "oral antiplatelet therapy not feasible or desirable" and therefore not likely to receive an oral platelet inhibitor. According to the British Cardiovascular Intervention Society audit, there were 92,589 PCIs in 2013, 65.6% of which were acute (n=60,738); 14 to 17% of these would result in a potential population of 8,503 to 10,325 people who may undergo treatment with cangrelor over a 12‑month period (personal communication, The Medicines Company June 2015).

Expert advisers

Mr S Antoniou, Consultant Pharmacist in Cardiovascular Medicine, Bart's Health NHS Trust and UCL Partners

Professor A H Gerschlick, Professor of Interventional Cardiology, University of Leicester NHS Trust

Dr R Khattar, Consultant Cardiologist and Honorary Clinical Senior Lecturer, Royal Brompton Hospital and Imperial College, London

Declarations of interest

Sotiris Antoniou has acted as an advisor and received honoraria from the following companies: AstraZeneca, Eli Lilly / Daiichi Sankyo, Correvio, Bayer and Boehringer Ingelheim

Anthony Gerschlick has acted as an advisor and received honoraria, travel reimbursement and research support from the following companies: Medicines Company, AstraZeneca, Eli Lilly/ Diachii Sankyo, Abbott Laboratories Ltd, Medtronic, Boston Scientific

Rajdeep Khattar has no interests to declare

Copyright

© National Institute for Health and Care Excellence, 2015. All rights reserved. NICE copyright material can be downloaded for private research and study, and may be reproduced for educational and not‑for‑profit purposes. No reproduction by or for commercial organisations, or for commercial purposes, is allowed without the written permission of NICE.

ISBN: 978-1-4731-1511-8