The Scottish Medicines Consortium issued on 7th December, 2008 an advice on bivalirudin (Angiox):
Bivalirudin (Angiox) is accepted for restricted use within NHS Scotland for the treatment of adult patients with acute coronary syndromes (unstable angina/non-ST segment elevation myocardial infarction) planned for urgent or early intervention. It is restricted to use in patients who would otherwise have been considered for heparin in combination with a glycoprotein IIb/IIIa antagonist. In these patients bivalirudin monotherapy may be a suitable alternative. It should not be used as an alternative to heparin alone.
Bivalirudin should be administered with aspirin and clopidogrel.
Bivalirudin showed a reduced risk of bleeding compared to a heparin-based anticoagulant strategy in patients with moderate and high risk acute coronary syndromes undergoing early invasive management.
This is an excerpt of the detailed advice on bivalirudin:
Treatment of adult patients with acute coronary syndromes (ACS: unstable angina/non-ST segment evaluation myocardial infarction (UA/NSTEMI)) planned for urgent or early intervention. Bivalirudin should be administered with aspirin and clopidogrel.
Start with an intravenous bolus of 0.1mg/kg followed by an infusion of 0.25mg/kg/hour.
Subsequent dosing varies according to the intervention (medical, percutaneous or surgical) undertaken for ACS.
Product availability date
Summary of evidence on comparative efficacy
Bivalirudin is direct inhibitor of thrombin. It was originally licensed as an anticoagulant for use in patients undergoing percutaneous coronary intervention (PCI) and the licence has been extended to include the treatment of ACS planned for urgent or early intervention.
The pivotal evidence for this indication comes from a multicentre, prospective randomised, open-label, parallel-group trial and the most relevant comparison is between 4,603 patients randomised to heparin (unfractionated heparin (UFH) or enoxaparin) plus a glycoprotein IIb/IIIa inhibitor (GPI) and 4,612 randomised to bivalirudin alone (without GPI). Eligible patients were adults with moderate and high risk ACS. Angiography was performed in all patients within 72 hours of randomisation, followed by triage at the physician’s discretion to further treatment (percutaneous, surgical, or medical). A third arm received bivalirudin and GPI but the results are not presented because the addition of GPI conferred no clinical advantage. All patients were recommended to receive clopidogrel or ticlopidine, though this, and the regimen used, was at the discretion of the investigator.
At 30-days’ follow-up (± 5 days) three primary endpoints were tested sequentially in the following order for the comparison between bivalirudin alone and heparin/GPI: 1) superiority of the bleeding endpoint (major bleeding according to a trial-specific scale); 2) non-inferiority then superiority of a composite net clinical outcome endpoint; 3) non-inferiority then superiority of a composite ischaemic outcome endpoint. The non-inferiority margin was
25%. The ischaemic endpoint was a composite of death from any cause, MI or unplanned revascularisation for ischaemia. The net clinical outcome endpoint was a composite of the ischaemic and major bleeding outcomes. All analyses were on the intention to treat population and adjustment was made for multiple comparisons. One-year data are also available for the ischaemic and major bleeding outcomes.
The European Medicines Agency (EMEA) considered bleeding to be a safety outcome and questioned the validity of the net clinical outcome therefore results are presented for the composite ischaemic endpoint only in this section. At 30-days, bivalirudin was non-inferior to heparin plus GPI for the incidence of this endpoint but there was a non-significant numerical advantage for heparin/GPI for the composite and all of its individual components (Table 1).
At one year the rates of composite ischaemia were 15.4% and 16.2% for heparin/GPI and bivalirudin alone respectively, representing a relative risk of 1.06 (95% CI 0.95 to 1.17).
Heparin/GPI was also numerically superior to bivalirudin for MI and revascularisation but not for death from any cause. None of these differences were significant.
Summary of evidence on comparative safety
Although defined as an efficacy variable, major bleeding was considered by the EMEA to be a safety outcome. It was reported both as a trial-specific scale and according to the Thrombolysis in Myocardial Infarction (TIMI) definition.
There were significant advantages for bivalirudin alone versus heparin/GPI for both scales, but the incidence rates and absolute difference were smaller in the TIMI analysis.
Other than for bleeding, the analysis of adverse events combined results from the bivalirudin alone and bivalirudin plus GPI arms. Up to the Day 30 visit, 416 (8.9%) heparin and 772 (8.6%) bivalirudin patients experienced one or more serious adverse events (SAEs, including SAEs with an outcome of death). The system order classes (SOCs) most commonly associated with SAEs up to the Day 30 visit were cardiac disorders, general disorders and administration site conditions, and infections and infestations (in ≥1.0% of all heparin and bivalirudin patients). The overall incidence of SOCs associated with SAEs up to the Day 30 visit was comparable for the heparin and bivalirudin treatment groups. More heparin than bivalirudin patients experienced cardiac disorder (3.4% versus 3.2%), respiratory, thoracic and mediastinal disorder (1.1% versus 0.9%), and vascular disorder (1.0% versus 0.8%) SAEs (≥0.2% difference between treatment groups), but the incidence of these events was low.
Summary of clinical effectiveness issues
The key efficacy outcome for registration was considered to be the composite ischaemic events for which bivalirudin alone was shown to be non-inferior but not superior to heparin/GPI at 30-days’ follow-up. Efficacy was also similar between groups after one year.
The EMEA considered the non-inferiority margin of 25% to be rather wide but noted that the one-year results were ‘reassuringly tight’ – effectively within a 15% margin.
The primary outcome of major bleeding was considered by the EMEA to be a safety outcome and significantly favoured bivalirudin alone over heparin/GPI. This advantage disappeared when bivalirudin was combined with GPI such that addition of GPI to bivalirudin was not considered to offer any clinical advantage.
Feedback and data obtained during expert consultation indicate that the trial design differs substantially from Scottish practice. For example, GPI are used more conservatively in Scottish practice (in about 44% of patients overall and less than 7% as an early intervention). Radial access is the default route for PCI in Scotland and is associated with a lower risk of bleeding than the femoral route used in 87% of patients in the trial. The trial-specific definition of major bleeding was argued to be more specific to ACS/PCI than the Thrombolysis in MI scale, but it was strongly influenced by access site bleeding, particularly haematoma, and the EMEA had reservations about its validity to reflect major bleeding.
Major bleeding rates continued to favour bivalirudin using the TIMI scale and after exclusion of access site haematoma, but the incidence rates, and hence the absolute difference, were smaller than with the specific trial scale.
The pivotal trial incorporated a further randomisation in the heparin/GPI and bivalirudin/GPI arms. This was between routine ‘up-stream’ introduction of GPI and selective deferred use – only in patients undergoing PCI. The primary hypothesis of this analysis did not involve bivalirudin, however in the light of expert comment it is notable that deferred GPI was associated with a significantly reduced rate of major bleeding at 30 days. Over-representation of GPI use, particularly upstream, may have inflated the incidence of major bleeding in the heparin/GPI arm compared with Scottish practice.
About 64% of patients received heparin pre-randomisation, and the incidence was similar in all treatment arms, including those in which patients received bivalirudin.
Additional analyses showed that the results of the composite ischaemic endpoint were more favourable to bivalirudin, especially bivalirudin alone, in patients who also receive aspirin and clopidogrel and the bleeding advantage was retained in those patients. Thus the indication for bivalirudin specifies co-administration with those agents.
Bivalirudin was associated with a significant advantage over heparin and GPI for the net clinical benefit outcome. The EMEA did not consider this for registration purposes as it was of the strong opinion that efficacy and safety variables should not be mixed in a single outcome. Nevertheless, with similar efficacy between bivalirudin alone and heparin/GPI and significantly reduced rates of bleeding, it reflects a net clinical advantage for bivalirudin when
combined with aspirin and clopidogrel as licensed (but not when combined with GPI).
The trial was open-label but hard endpoints, including death and MI, were adjudicated by blinded assessors to minimise bias.