4 Consideration of the evidence

4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of apixaban for the prevention of venous thromboembolism after elective total hip or knee replacement surgery in adults having considered evidence on the nature of the condition and the value placed on the benefits of apixaban 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 discussed the decision problem and agreed that this appraisal would focus on choice among drugs for preventing venous thromboembolism. The Committee discussed 'Venous thromboembolism: reducing the risk' (NICE clinical guideline 92). It noted the recommendation that in addition to mechanical prophylaxis, people having elective hip replacement or elective knee replacement should be offered LMWH, dabigatran etexilate, rivaroxaban or fondaparinux. It also noted the written comments received from one of the clinical specialists that the newer oral anticoagulant agents for the prevention of venous thromboembolism are not used by all surgical units because some surgeons are concerned that they may increase the incidence of bleeding at the operation site, increasing the risk of infection, delayed wound healing and delayed mobilisation. The Committee was aware of these concerns, but concluded that any recommendations made would be limited to situations in which drugs for the prevention of venous thromboembolism were already recommended in 'Venous thromboembolism: reducing the risk' (NICE clinical guideline 92).

4.3 The Committee discussed the benefits of oral treatments compared with subcutaneous injection, in particular the greater acceptability and ease of management of oral administration. The Committee noted the written evidence from one of the clinical specialists and the patient expert that for treatments given by subcutaneous injection, people would have to be taught to self-inject before discharge from hospital or a district nurse would have to continue the treatment after discharge. The Committee heard from patient experts and the clinical specialist present at the Committee meeting that people prefer oral dosing to subcutaneous injection and therefore adherence to treatment after discharge might improve, although this remained uncertain. The Committee noted the twice-daily regimen for apixaban compared with the once-daily regimens for rivaroxaban and dabigatran etexilate. The Committee heard from the patient experts that if people received enough information about the effectiveness of apixaban and the importance of preventing venous thromboembolism they would be likely to adhere to apixaban treatment even though they needed to take it twice a day.

4.4 The Committee discussed the clinical effectiveness of apixaban compared with LMWH, rivaroxaban, dabigatran etexilate and fondaparinux in people having elective hip or knee surgery. It noted the direct evidence from randomised controlled trials of apixaban and enoxaparin, and the indirect and mixed treatment comparison of apixaban versus rivaroxaban, dabigatran etexilate and fondaparinux. It also noted that the direct evidence was limited to a comparison of apixaban and enoxaparin and that the manufacturer assumed that all LMWHs were equivalent in terms of their clinical effectiveness. The Committee heard that enoxaparin is the most widely used LMWH in the UK, and agreed that the comparison using enoxaparin as the only LMWH was appropriate. It discussed the applicability of the clinical trials to UK clinical practice, understanding that there is variation in strategies for preventing venous thromboembolism. The Committee agreed that data from the ADVANCE 2 and ADVANCE 3 randomised controlled trials, in which the patients in the control arm received 40 mg enoxaparin once daily, were applicable to UK clinical practice. It agreed that the ADVANCE 1 and APROPOS trials, which used an alternative dosing regimen of 30 mg enoxaparin twice daily, did not use the UK licensed dosage, and that they were less relevant to the evaluation of clinical effectiveness of apixaban than ADVANCE 2 and 3.

4.5 The Committee discussed the outcome data from these trials. It noted the written concerns of one of the clinical specialists about the use of surrogate markers as valid predictors of clinically relevant outcomes. The clinical specialist highlighted that there were limited data available to show a relationship between one of the major components of the composite primary outcome of the studies (asymptomatic deep vein thrombosis) and clinically relevant venous thromboembolic events, and that the data that were available did not suggest that asymptomatic deep vein thrombosis was a good predictor of clinical thromboembolic events after joint-replacement surgery. The clinical specialist present at the meeting stated that in his opinion there was sufficient evidence to suggest an association between asymptomatic deep vein thrombosis and acute pulmonary embolism. The Committee acknowledged the difference of opinion expressed by clinician specialists about asymptomatic deep vein thrombosis but agreed that asymptomatic deep vein thrombosis was widely used as an outcome measure in research studies, and was relevant for consideration.

4.6 The Committee discussed the results of the ADVANCE trials and concluded that apixaban was significantly more effective than enoxaparin in preventing venous thromboembolism. The Committee considered adverse events such as bleeding, noting that the bleeding rates were lower for apixaban than for enoxaparin, although the difference was not statistically significant. It concluded that apixaban could, using the evidence available, be considered more clinically effective than enoxaparin in preventing venous thromboembolic events, but was broadly comparable to enoxaparin in terms of short-term adverse effects.

4.7 The Committee considered the evidence on the clinical effectiveness of apixaban compared with rivaroxaban, dabigatran etexilate and fondaparinux through the indirect and mixed treatment comparisons. The Committee noted the ERG's critique of the indirect and mixed treatment comparisons and the manufacturer's response to the clarification requested by the ERG. It also noted that the manufacturer had undertaken three indirect and two mixed treatment comparison analyses which took account of the differences in UK and USA dosages of enoxaparin. The Committee agreed with the manufacturer and the ERG that the indirect and mixed treatment comparisons using the UK dosage of enoxaparin were most relevant to UK clinical practice and therefore it was reasonable to consider the results of these comparisons. The Committee was aware of the uncertainty associated with the results from the indirect comparison and the mixed treatment comparison as demonstrated by the wide confidence intervals (surrounding the point estimates from the indirect comparison analysis) and the credibility intervals (surrounding the point estimates from the mixed treatment comparison). The Committee was also aware of the inconsistency of some results between the indirect comparison and the mixed treatment comparison and between the indirect comparison and the direct randomised controlled trials, and the manufacturer's explanation for these inconsistencies. The Committee agreed that based on the evidence presented it was not possible to estimate the relative effectiveness of apixaban compared with rivaroxaban, dabigatran etexilate or fondaparinux.

4.8 The Committee then considered the factors that clinicians take into account when prescribing anticoagulant therapy for people having total hip and knee replacement surgery. The Committee discussed the concerns raised by orthopaedic surgeons about the effect of anticoagulants on local bleeding, infection and wound healing. The Committee noted the written statement from one clinical specialist that bleeding caused by anticoagulation is clinically important because it can contribute to deep surgical site infection. The Committee also noted from the written statement that some orthopaedic surgeons believed that new oral antithrombotic agents such as rivaroxaban and dabigatran etexilate may be associated with more treatment-related bleeds than enoxaparin and result in worse patient outcomes. The Committee heard from the clinical specialist present at the meeting that there was a 'trade-off' to be made between the benefits of reduced thrombotic events and increased risk of bleeding.

4.9 The Committee discussed the risk of bleeding into the joint and the association between the time between surgery and anticoagulant therapy taking effect, that is, the greater the time between surgery and anticoagulant therapy taking effect the lower the risk of bleeding. The Committee was aware of the different starting times in the summaries of product characteristics for each of the anticoagulants and noted that apixaban had the longest time window for administration following surgery (12–24 hours after surgery, compared with rivaroxaban 6–10 hours after surgery, dabigatran etexilate 1–4 hours after surgery and enoxaparin 6–12 hours after surgery). The Committee had not seen any evidence assessing the association between the different drug regimens and the risk of bleeding. The Committee agreed that the longer time window for administration of apixaban could allow clinicians to assess a patient's bleeding risk before starting thromboprophylaxis after total hip or knee replacement surgery.

4.10 The Committee then discussed the choice of anticoagulants for people with renal disease. The Committee heard from the clinical specialist that in patients with severe renal disease all anticoagulant therapy carries additional risks and oral agents would be unsuitable. The Committee noted that for mild-to-moderate renal disease, the summary of product characteristics for apixaban did not specify a dose adjustment whereas the summary of product characteristics for dabigatran etexilate did. The Committee agreed that although there was insufficient evidence to determine the relative clinical effectiveness of the oral anticoagulants, apixaban and rivaroxaban shared a potential benefit for clinical practice in that there is no need to modify the dose in people with mild to moderate kidney function impairment.

4.11 The Committee discussed the evidence submitted by the manufacturer on the cost effectiveness of apixaban for the prevention of venous thromboembolism in people having hip or knee replacement, the ERG's critique of the manufacturer's submission, and the manufacturer's response to the clarification requested by the ERG. The Committee noted the two-phase structure of the economic model and accepted that the modelling approach used by the manufacturer was appropriate for modelling the prevention of venous thromboembolism.

4.12 The Committee discussed the base-case analysis in the manufacturer's submission. It noted that the manufacturer had not included fondaparinux in the economic evaluation because there was insufficient clinical evidence available for the indirect comparison of apixaban and fondaparinux in patients having total knee replacement surgery. It also noted that the manufacturer had highlighted that fondaparinux was used in less than 1% of people in current UK clinical practice. The Committee was aware that the cost of dabigatran etexilate used in the base-case analysis did not reflect the current NHS list price of dabigatran etexilate. It was also aware that the 40% reduction in the cost of dabigatran etexilate occurred after the manufacturer of apixaban had provided its submission to NICE, but that the submission included a one-way sensitivity analysis that anticipated the price reduction (see section 3.24). The Committee noted that in the base-case analysis apixaban dominated enoxaparin and dabigatran etexilate in both total hip and knee replacement, that is, apixaban was associated with larger QALY gains and lower costs. It further noted that for total knee replacement the ICER for rivaroxaban compared with apixaban was £21,700 per QALY gained, but for total hip replacement rivaroxaban dominated apixaban. The Committee then considered the one-way sensitivity analysis provided by the manufacturer, and in particular that relating to a 50% reduction in the price of dabigatran etexilate. The Committee noted that for patients having total hip replacement or total knee replacement, apixaban still remained dominant compared with enoxaparin and dabigatran etexilate for all changes in the sensitivity analysis. The Committee then considered the plausibility of the cost-effectiveness results, in particular in relation to the strength of the evidence for clinical effectiveness. The Committee recognised that there were uncertainties associated with the ICERs for rivaroxaban and dabigatran etexilate because the clinical data for these drugs included in the model were originally derived from the indirect comparison, unlike the ICER for enoxaparin which was derived from direct head-to-head trials. The Committee therefore agreed that the cost-effectiveness results for apixaban compared with enoxaparin were the most robust, and that there were significant uncertainties in the comparisons of cost effectiveness with rivaroxaban and dabigatran etexilate.

4.13 The Committee concluded that apixaban was more clinically effective and cheaper than enoxaparin. It also concluded that there was insufficient clinical evidence to determine whether or not apixaban was more or less clinically effective than rivaroxaban or dabigatran etexilate and therefore the ICERs presented for these anticoagulants had to be interpreted with caution. However, the Committee did accept that any differences in clinical effectiveness between the agents were likely to be small, and apixaban may provide some clinical advantages compared with the other agents. The Committee therefore concluded that apixaban should be recommended as an option for preventing venous thromboembolism in adults after elective total hip and total knee replacement surgery.

4.14 The Committee considered whether its recommendation was associated with any issues related to equality legislation and the requirement for fairness. The Committee noted that no issues had been highlighted during the scoping exercise or during the course of the appraisal. The Committee was aware that consultees and commentators had raised the issue of providing non-injectable medication for people with needle phobia but concluded that the recommendations do not affect access to the technology for any specific groups.

Summary of Appraisal Committee's key conclusions

TA245

Appraisal title: 'Apixaban for the prevention of venous thromboembolism after total hip or knee replacement in adults'

Section

Key conclusion

Apixaban is recommended as an option for the prevention of venous thromboembolism in adults after elective hip or knee replacement surgery.

1.1

The Committee concluded that apixaban was more clinically effective and cheaper than enoxaparin. It also concluded that there was insufficient clinical evidence to determine whether or not apixaban was more or less clinically effective than rivaroxaban and dabigatran etexilate and therefore the ICERs presented for these anticoagulants had to be interpreted with caution.

4.13

Current practice

Clinical need of patients, including the availability of alternative treatments

The Committee noted the recommendation in the NICE clinical guideline 'Venous thromboembolism' (NICE clinical guideline 92) that people having elective hip replacement or elective knee replacement should be offered LMWH, dabigatran etexilate, rivaroxaban or fondaparinux.

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?

Apixaban appeared to provide a potential benefit for clinical practice. The dose does not need to be adjusted for people with compromised renal function.

4.10

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

Apixaban has a marketing authorisation for the 'prevention of venous thromboembolic events (VTE) in adult patients who have undergone elective hip or knee replacement surgery'.

2.1

Adverse effects

Common adverse events include anaemia, haemorrhage, contusion and nausea. See the summary of product characteristics for full details.

2.3

Evidence for clinical effectiveness

Availability, nature and quality of evidence

Four good quality randomised controlled trials were identified comparing apixaban with enoxaparin. Fifteen randomised controlled trials were identified for inclusion in the indirect comparison of apixaban with dabigatran etexilate, rivaroxaban and fondaparinux. The manufacturer included 43 studies for the mixed treatment comparison.

3.2

3.9

3.12

Relevance to general clinical practice in the NHS

Of the fifteen studies identified for inclusion in the indirect comparison, nine compared the treatment of interest with enoxaparin at the UK dosage of 40 mg once daily.

3.9

The Committee agreed that the ADVANCE 2 and ADVANCE 3 randomised controlled trials, in which patients in the control arm received enoxaparin 40 mg once daily, were applicable to UK clinical practice.

4.4

Uncertainties generated by the evidence

The manufacturer highlighted inconsistencies among the results of the mixed treatment comparisons and the head-to-head trials.

3.13

The Committee was aware of the inconsistency of some results between the indirect comparison and the mixed treatment comparison and between the indirect comparison and the head-to-head trials, and the manufacturer's explanation for these inconsistencies.

4.7

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

No subgroups were identified.

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

The Committee agreed that based on the evidence presented it was not possible to determine an estimate of relative effectiveness of apixaban compared with rivaroxaban, dabigatran etexilate or fondaparinux.

4.7

Evidence for cost effectiveness

Availability and nature of evidence

The manufacturer submitted an economic model assessing the cost effectiveness of apixaban compared with enoxaparin, dabigatran etexilate and rivaroxaban. The manufacturer adopted a two-stage modelling approach. A decision tree to model treatment in the prophylactic phase (from the time of surgery to 90 days after surgery) and a Markov model to model the long-term events (90 days after surgery).

3.15

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

The Committee recognised uncertainties associated with ICERs for rivaroxaban and dabigatran etexilate because the clinical data were derived from indirect comparison analysis, unlike the ICER for enoxaparin, which was derived from direct head-to-head trials.

4.12

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?

The QALY differences between the treatments were small. For people having total hip replacement surgery, total QALYs ranged from 9.520 for enoxaparin to 9.536 for rivaroxaban. For patients having total knee replacement surgery, total QALYs ranged from 9.023 for enoxaparin to 9.090 for rivaroxaban.

3.22

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

No subgroups were identified.

What are the key drivers of cost effectiveness?

Rivaroxaban was cost effective compared with apixaban, except when the time horizon was 10 years or less, when the age at surgery was 80 years or when a smaller relative difference in the risk of the primary end point (total venous thromboembolism and death from any cause) was assumed.

3.24

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

In base-case analyses apixaban dominated enoxaparin. The Committee agreed that because the ICER for enoxaparin was derived from data from direct head-to-head trials, the cost-effectiveness results for apixaban compared with enoxaparin were the most robust.

4.12

Additional factors taken into account

Patient access schemes (PPRS)

The manufacturer did not submit a patient access scheme.

End-of-life considerations

The supplementary advice was not relevant to this appraisal.

Equalities considerations and social value judgements

No equalities issues were raised in this appraisal.