4 Consideration of the evidence

4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of bivalirudin, having considered evidence on the nature of ST-segment-elevation myocardial infarction (STEMI) and primary percutaneous coronary intervention (PCI) and the value placed on the benefits of bivalirudin by people with the condition, those who represent them, and clinical specialists. It also took into account the effective use of NHS resources.

4.2 The Committee considered the nature of the condition, and noted evidence submitted and presented by the patient experts and clinical specialists on the clinical signs and symptoms associated with STEMI. The Committee heard that bivalirudin is already in use in the UK and that both patients and clinicians are in favour of its continued use. It was noted that bivalirudin is easier to use than abciximab because the solution is easier to prepare and administer. The patient expert stressed the importance of reducing major bleeding, which was thought to be an important benefit of bivalirudin. Compared with anticoagulants, bivalirudin may require less monitoring and therefore be more convenient for patients.

Clinical effectiveness

4.3 The Committee discussed the evidence for the clinical effectiveness of bivalirudin in combination with aspirin and clopidogrel. It agreed that the results of the single but large HORIZONS-AMI trial (both the primary outcome of major bleeding, and the secondary outcomes of all-cause mortality and cardiac mortality) showed statistically significant advantages for treatment with bivalirudin over treatment with heparin plus glycoprotein IIb/IIIa inhibitor. In particular it noted that all-cause mortality was 1.3% and 1.8% lower in the bivalirudin group than in the comparator group after 1 and 3 years of follow-up, respectively.

4.4 The Committee discussed whether similar outcomes to those in the HORIZONS-AMI trial can be expected for patients seen in UK clinical practice. The Committee heard from the clinical specialists that the trial had some limitations: it only compared bivalirudin with heparin and glycoprotein IIb/IIIa inhibitors, not with heparin alone; in the UK, more people have PCI via the radial access site than in the trial; and a substantial proportion of the trial population received pre-procedural heparin, which is not standard UK practice.

4.5 The Committee heard how the use of the radial access site rather than femoral access reduces the incidence of access site bleeding. Radial access is more common in UK practice than in the trial, and hence this could reduce the benefit of reduced access site bleeding with bivalirudin shown in the trial. However, in the trial bivalirudin was also shown to reduce bleeding not related to the access site
in comparison with glycoprotein IIb/IIIa inhibitor plus heparin. The Committee accepted that similar outcomes for bleeding not
related to access site could be expected in UK clinical practice as in the trial.

4.6 The Committee discussed the omission of a comparison against heparin alone in the manufacturer's submission. The clinical specialists stated that there is a question around the necessity to include a glycoprotein IIb/IIIa inhibitor with heparin, because there is some evidence to support the use of heparin alone from the BRAVE-3 study. However, the clinical specialists stated that clinicians still feel that this question is largely unanswered. Treatment with heparin alone is not standard practice in the UK. The Committee was satisfied that it was not necessary to compare bivalirudin against treatment with heparin alone.

4.7 The Committee discussed the use of bivalirudin in combination with aspirin and prasugrel as opposed to aspirin and clopidogrel. The clinical experts stated a prasugrel/bivalirudin combination was likely to be better than a clopidogrel/bivalirudin combination because prasugrel reduces stent thrombosis. This was considered particularly important because the bivalirudin group in the trial had a higher incidence of stent thrombosis within the first 24 hours of PCI than the comparator group. The manufacturer, however, explained that when the trials for bivalirudin were designed, and at the time of the first licence application for bivalirudin, prasugrel had not received a marketing authorisation. For this reason, bivalirudin has only been studied with aspirin and clopidogrel, and therefore the marketing authorisation is specifically in combination with these two drugs. The Committee was satisfied that treatment with bivalirudin in combination with aspirin and prasugrel is outside the marketing authorisation for bivalirudin and therefore beyond the remit of this appraisal.

4.8 The Committee explored the reasons why in the bivalirudin group, the rates of major adverse cardiovascular events and stent thrombosis were lower in patients treated with pre-procedural heparin. The clinical specialists explained that it is hard to understand clinically why taking pre-procedural heparin would reduce the rate of major adverse cardiovascular events. They suggested that the population who received pre-procedural heparin may be different from those who did not, and it is this difference that may have affected the rate of major adverse cardiovascular events. The Committee heard from the ERG that there was no interaction between pre-procedural heparin and the major adverse cardiovascular events outcome. The clinical specialists also explained that it is not common practice in the UK to administer pre-procedural heparin, and anecdotally this has not reduced the effectiveness of bivalirudin in clinical practice. The Committee was satisfied that differences in the use of pre-procedural heparin between the trial and UK clinical practice would not significantly alter the favourable outcomes for bivalirudin.

4.9 The Committee considered whether there were any subgroups of patients who would be expected to benefit from treatment with bivalirudin more than other groups. It heard from the clinical specialists that theoretically it might be possible to reserve bivalirudin for those who are at the greatest risk of bleeding, but that in reality the same patients would be at a high risk of other symptoms and may not be a discrete subgroup for the purposes of targeting treatment. The Committee was persuaded that no subgroups could be selected on the basis of the greatest potential to benefit from bivalirudin.

4.10 The Committee also considered other limitations of the trial. It noted that there was only a single trial and that it was an open-label design with some risk of bias, but agreed that the results suggested an advantage for bivalirudin. The Committee concluded that there was sufficient evidence to demonstrate that bivalirudin was more clinically effective than glycoprotein IIb/IIIa inhibitor plus heparin, leading to lower rates of major bleeds and mortality.

Cost effectiveness

4.11 The Committee discussed the manufacturer's economic model in which the main factor affecting cost effectiveness is the reduced mortality risk with bivalirudin treatment compared with treatment with glycoprotein IIb/IIIa inhibitor plus heparin. It noted that in the base case, treatment with bivalirudin was both more effective and less expensive than treatment with a glycoprotein IIb/IIIa inhibitor plus heparin. This finding appeared robust in the face of both deterministic and probabilistic sensitivity analyses.

4.12 The Committee discussed the robustness of the manufacturer's model. First the Committee discussed the two-stage structure of the model: an initial stage, in which clinical events could occur, followed by a stage representing the rest of the lifetime. The manufacturer presented one model in which the first stage is 1 year, and another model in which the first stage is 3 years. The Committee was satisfied that the results from the 3-year model were consistent with those from the 1-year model. Second, it questioned the appropriateness of mutually exclusive events within the first stage of the model and suggested that this was an oversimplification. The manufacturer explained that there were no common combinations of events which could have been modelled. The Committee noted that the ERG's rebuilding of the model found no discrepancies. The Committee was satisfied with the approach taken by the manufacturer in its model design.

4.13 The Committee then discussed key input assumptions used in the model, concerning firstly vial use and, secondly, choice of glycoprotein IIb/IIIa inhibitor in the comparator group, and thirdly radial arterial access. With regards to vial use, it heard from the clinical specialists that treatment with bivalirudin would take place at the time of PCI, and that usually not more than one vial would be used. The mean use of 1.23 vials from the trial was used for calculating cost of treatment. The Committee questioned whether vials could be split and heard from the clinical specialists that if more than 1 vial were needed, a second vial would be used and any extra discarded. Sensitivity analysis was conducted by the ERG on the effect of including 2 vials instead of 1.23 vials in the model, but did not lead to any changes in the results. The Committee was satisfied with the way that vial use was incorporated into the economic model.

4.14 The Committee considered the assumptions around the choice of glycoprotein IIb/IIIa inhibitor used in the manufacturer's model. It noted that abciximab, which is the most costly glycoprotein IIb/IIIa inhibitor, was assumed to be used in 73% of people, tirofiban in 19% and eptifibatide, which is the least costly glycoprotein IIb/IIIa inhibitor, in only 8.1%. The Committee heard from the clinical specialists that this was largely because abciximab has been the favoured glycoprotein IIb/IIIa inhibitor for many years, and some clinicians are so familiar with its use that they are reluctant to change to eptifibatide or tirofiban. The Committee noted that 100% use of eptifibatide in the manufacturer's sensitivity analysis had a small impact on the ICER, with the ICER for bivalirudin rising to £1764 per QALY gained. The Committee accepted that 100% eptifibatide use was an extreme position and was satisfied that the high usage of abciximab in the model had a limited impact on the cost effectiveness of bivalirudin.

4.15 The Committee considered the assumptions around radial arterial access site in the model. In the base-case analysis, radial arterial access was assumed in 42.5% of cases, in line with UK practice. A sensitivity analysis from the manufacturer which increased the usage to 100% led to no changes in the results, with bivalirudin remaining the dominant treatment option. The Committee was satisfied that the model results were robust to changes in
access site.

4.16 The Committee discussed the results of the combined sensitivity analysis, in which 100% eptifibatide use had been combined with the assumptions of 100% radial arterial access and equal length of hospital stay in those treated with bivalirudin and those treated with glycoprotein IIb/IIIa inhibitors. In this analysis, which was considered by the manufacturer to be the most unfavourable scenario for bivalirudin, the ICER increased to £4106 per QALY gained. The Committee accepted that this unfavourable position was clinically unrealistic, but noted that the ICER produced with this analysis was still well within the range normally considered to be a cost-effective use of NHS resources.

4.17 The Committee concluded the discussion by noting the robustness of the manufacturer's base case, in which treatment with bivalirudin dominated treatment with a glycoprotein IIb/IIIa inhibitor plus heparin (that is, was less costly and more effective) and that the results of the model are robust to the various sensitivity analyses. The Committee concluded that the model is associated with a very low degree of decision uncertainty and that bivalirudin should be recommended for the treatment of adults with STEMI undergoing PCI.

4.18 No equalities issues were raised at any point in the appraisal.

Summary of Appraisal Committee's key conclusions

TA230

Appraisal title: Bivalirudin for the treatment of ST-segment-elevation myocardial infarction

Section

Key conclusion

Bivalirudin in combination with aspirin and clopidogrel is recommended for the treatment of adults with ST-segment-elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PCI).

1.1

Current practice

Clinical need of patients, including the availability of alternative treatments

Bivalirudin is already in use in the UK and both patients and clinicians are in favour of its continued use. Bivalirudin is easier to use than abciximab (a glycoprotein IIb/IIIa inhibitor) because the solution is easier to prepare and administer. The patient expert stressed the importance of reducing major bleeding, which was thought to be an important benefit of bivalirudin. Compared with anticoagulants, bivalirudin may require less monitoring and therefore be more convenient for patients.

4.2

The technology

Proposed benefits of the technology

How innovative is the technology in its potential to make a significant and substantial impact on health-related benefits?

The primary outcome of major bleeding and the secondary outcomes of all-cause mortality and cardiac mortality showed statistically significant advantages for treatment with bivalirudin over treatment with heparin plus glycoprotein IIb/IIIa inhibitor. In particular, all-cause mortality was 1.3% and 1.8% lower in the bivalirudin group than in the comparator group after 1 and 3 years of follow-up, respectively.

4.3

What is the position of the treatment in the pathway of care for the condition?

Bivalirudin is used as an anticoagulant in adult patients with STEMI undergoing primary PCI.

2.1

Adverse effects

The bivalirudin group had a higher incidence of stent thrombosis within the first 24 hours of PCI than the comparator group

4.7

Evidence for clinical effectiveness

Availability, nature and quality of evidence

A single trial (HORIZONS-AMI) formed the evidence base. It was an open-label trial with some risk of bias. However, the Committee concluded that there was sufficient evidence to demonstrate that bivalirudin was more clinically effective than glycoprotein IIb/IIIa inhibitor plus heparin.

4.10

Relevance to general clinical practice in the NHS

In the UK, more people have PCI via the radial access site than in the trial. The use of the radial access site, compared with femoral access, reduces the incidence of access site bleeding. Hence the reduced access site bleeding observed with bivalirudin in the trial may be attenuated in UK practice. The Committee accepted that similar rates of non-access site bleeding could be expected in UK clinical practice as in the trial.

In the trial a substantial number of patients were treated with pre-procedural heparin but this is not common practice in the UK. The Committee was satisfied that differences in the use of pre-procedural heparin between the trial and UK clinical practice would not significantly alter the favourable outcomes for bivalirudin.

4.4

4.5

4.8

Uncertainties generated by the evidence

The Committee discussed the use of bivalirudin in combination with aspirin and prasugrel as opposed to aspirin and clopidogrel. It was satisfied that this treatment combination is outside the marketing authorisation for bivalirudin and therefore beyond the remit of this appraisal.

4.7

Are there any clinically relevant subgroups for which there is evidence of differential effectiveness?

The Committee was persuaded that no subgroups could be selected on the basis of the greatest potential to benefit from bivalirudin.

4.9

Estimate of the size of the clinical effectiveness including strength of supporting evidence

The primary outcome of major bleeding and the secondary outcomes of all-cause mortality and cardiac mortality showed statistically significant advantages for treatment with bivalirudin over treatment with heparin plus glycoprotein IIb/IIIa inhibitor. In particular, all-cause mortality was 1.3% and 1.8% lower in the bivalirudin group than in the comparator group after 1 and 3 years of follow-up, respectively.

4.3

Evidence for cost effectiveness

Availability and nature of evidence

The manufacturer submitted an economic analysis that employed an economic model based primarily on the analysis of patient-level data from the HORIZONS-AMI trial.

3.13

Uncertainties around and plausibility of assumptions and inputs in the economic model

The Committee considered that the results from the model were associated with a very low degree of decision uncertainty.

4.17

Incorporation of health-related quality-of-life benefits and utility values

Have any potential significant and substantial health-related benefits been identified that were not included in the economic model, and how have they been considered?

There were no issues raised about health-related quality-of-life values that were thought to be relevant. No health-related benefits were identified that were not included in the economic model.

Are there specific groups of people for whom the technology is particularly cost effective?

The Committee was persuaded that no subgroups could be selected on the basis of the greatest potential to benefit from bivalirudin.

4.9

What are the key drivers of cost effectiveness?

The main factor affecting cost effectiveness in the model is the reduced mortality risk with bivalirudin treatment compared with treatment with glycoprotein IIb/IIIa inhibitor plus heparin.

4.11

Most likely cost-effectiveness estimate (given as an ICER)

In the base case, treatment with bivalirudin dominated the comparator in that it was both more effective and less expensive than treatment with a glycoprotein IIb/IIIa inhibitor plus heparin.

4.17

Additional factors taken into account

Patient access schemes (PPRS)

No patient access schemes were submitted.

End-of-life considerations

End-of-life considerations were not discussed.

Equalities considerations and social value judgements

No equalities issues were raised at any point in the appraisal.

4.18

  • National Institute for Health and Care Excellence (NICE)