Nivolumab in combination with ipilimumab for treating advanced melanoma

Clinical effectiveness

4.3 The committee discussed the clinical effectiveness of nivolumab plus ipilimumab. The clinical‑effectiveness evidence is in the company's submission (pages 38–124) and in the ERG report (pages 29–114). The committee noted that overall survival data from the CheckMate‑067 and 069 trials, which compared nivolumab plus ipilimumab with ipilimumab alone (and also nivolumab alone in Checkmate‑067), were immature because the number of events (deaths) pre‑specified in the statistical analysis plan had not been reached at the time of the company submission. The committee heard from the company that an early exploratory analysis of CheckMate‑069 requested by the European Medicines Agency showed an 18‑month overall survival rate of 69% in patients taking nivolumab plus ipilimumab, irrespective of BRAF mutation status in the intention‑to‑treat population. This was nearly double the 18‑month overall survival rate of 35% for ipilimumab alone in a pooled analysis of historical trials.The committee noted that the final overall survival data from CheckMate‑067 (which compared nivolumab plus ipilimumab with both nivolumab and ipilimumab alone) were not yet available. It was therefore difficult to draw any firm conclusion on relative overall survival benefit, although the committee noted that the interim overall survival results for the combination regimen compared with ipilimumab alone looked promising.

4.4 The committee considered the Kaplan–Meier curves for progression‑free survival from CheckMate‑067 in the intention‑to‑treat population (all people who were randomised), which included people with BRAF mutation‑negative melanoma (n=213 in the nivolumab plus ipilimumab arm and n=218 in the ipilimumab arm) and mutation‑positive melanoma (n=101 in the nivolumab plus ipilimumab arm and n=97 in the ipilimumab arm). The committee was aware that treatment with nivolumab plus ipilimumab resulted in a significant extension in progression‑free survival compared with ipilimumab alone (hazard ratio 0.42, 99.5% confidence interval [CI] 0.31 to 0.57, p<0.001) for the intention‑to‑treat population. The median progression‑free survival was 11.5 months (95% CI 8.9 to 16.7) for nivolumab plus ipilimumab and 2.9 months (95% CI 2.8 to 3.4) for ipilimumab alone.

4.5 The committee noted that in CheckMate‑067 treatment with nivolumab plus ipilimumab resulted in an unweighted objective response rate difference of 38% compared with ipilimumab alone (57.6% in the nivolumab plus ipilimumab arm and 19% in the ipilimumab alone arm, odds ratio 6.11, 95% CI 3.59 to 10.38, p<0.001) for the intention‑to‑treat population. The committee noted that in CheckMate‑069, the investigator‑assessed objective response rate was 59% in the nivolumab plus ipilimumab group compared with 11% in the ipilimumab group (odds ratio for response 12.19, 95% CI 4.41 to 33.68, p<0.0001) for the overall population, regardless of BRAF mutation status. The committee concluded that nivolumab plus ipilimumab is more effective in the short term than ipilimumab alone, but the long‑term benefit of nivolumab plus ipilimumab remains uncertain until further follow‑up data are available. The committee had not been presented with any data to establish the relative benefit of nivolumab plus ipilimumab compared with nivolumab alone.

4.6 The committee discussed whether the clinical effectiveness of nivolumab plus ipilimumab would vary depending on the expression of programmed death receptor ligand 1 (PD‑L1) as referred to in the summary of product characteristics (see section 2.2). The committee noted that the Committee for Medicinal Products for Human Use (CHMP) stated that 'relative to nivolumab monotherapy, an increase in progression‑free survival for the combination of nivolumab with ipilimumab is established only in patients with low tumour PD‑L1 expression'. The committee heard from a clinical expert that this was an exploratory analysis requested by the European Medicines Agency and that PD‑L1 expression is not routinely assessed in clinical practice. Furthermore, there is no universally agreed threshold for PD‑L1 expression. The committee concluded that PD‑L1 expression may be one of the factors that influence clinical decision making, but it would not be appropriate for NICE to base recommendations on PD‑L1 expression at present.

4.7 The committee discussed the adverse events associated with nivolumab plus ipilimumab. It noted that, in the trials, nivolumab plus ipilimumab was associated with a higher rate of high‑grade or serious adverse events than ipilimumab alone. The committee heard from clinicians that the presence of elevated transaminases (commonly alanine transaminase and aspartate transaminase) indicating liver damage is one of the most common serious adverse effects related to treatment with nivolumab plus ipilimumab, and that clinicians routinely monitor transaminases and can deal with this complication if it occurs. Other serious adverse events included diarrhoea and colitis requiring hospitalisation, which are recognised complications of ipilimumab, and there were a few treatment‑related deaths in the nivolumab plus ipilimumab groups in the trials (none in CheckMate‑067 and 3 in CheckMate‑069). The committee concluded that although the adverse events related to nivolumab plus ipilimumab were significant and could be severe, the additional effectiveness of this treatment is likely to outweigh the potential risk of serious adverse events in patients who are fit enough and willing to tolerate this combination immunotherapy regimen.

4.8 The committee considered the likely duration of treatment with nivolumab plus ipilimumab in clinical practice. It noted that after the initial course of combined therapy, the summary of product characteristics recommends continuing treatment with nivolumab 'as long as clinical benefit is observed or until treatment is no longer tolerated by the patient'. The company stated in its submission that nivolumab would not be expected to be given beyond 2 years. The committee heard from the clinical experts that there is no evidence that treatment with nivolumab should stop at 2 years. It heard from a patient expert that patients who are currently having nivolumab expressed concern that their treatment might be stopped after 2 years regardless of whether they were still benefiting from it. The committee concluded that the 2‑year treatment duration cap proposed by the company was arbitrary and not based on clinical evidence, and that there was considerable uncertainty about the optimum duration of treatment. However, based on the discontinuation rate in clinical trials, the committee considered that only a small number of patients would still be having treatment after 2 years. The committee appreciated that there is considerable uncertainty about the optimum duration of treatment with nivolumab. The committee also expressed the view that a review of this guidance after 2 years (to coincide with the review of pembrolizumab guidance) should be recommended, at which time overall survival data will be more mature and the optimum duration of treatment may have been clarified.

Cost effectiveness

4.9 The committee discussed the cost‑effectiveness evidence presented by the company and its critique by the evidence review group (ERG). The cost‑effectiveness evidence is in the company's submission (pages 125–216), in the appendices to the company's submission and in the ERG report (pages 115–232). The company's model compared nivolumab plus ipilimumab with ipilimumab alone in BRAF mutation‑negative disease, and with ipilimumab, vemurafenib and dabrafenib in BRAF mutation‑positive disease, for people with previously untreated advanced (unresectable or metastatic) melanoma. The committee noted that the ERG considered the company's economic evaluation to be complex and reliant on too many assumptions. The committee noted that the ERG's main concern related to the company's assumptions about post‑progression survival in the model (see the ERG report, pages 185–186 and 195–197). However, the committee was aware that the ERG was able to change the parameters in the model to compensate for that assumption (see the ERG report, pages 236–237). Therefore, the committee accepted that the structure of the company's model for the BRAF mutation‑negative population comparing nivolumab plus ipilimumab with ipilimumab monotherapy could be used for the purposes of decision‑making in this appraisal.

4.10 The committee went on to look specifically at the assumptions made by the company in its modelling of the BRAF mutation‑positive population comparing nivolumab plus ipilimumab, with ipilimumab alone, dabrafenib and vemurafenib. The committee considered that the modelling used for the BRAF mutation‑positive population was very complex, and also made it very difficult to compare the effectiveness versus ipilimumab alone and the BRAF inhibitors (see the ERG report, page 198). The committee recalled their earlier view that BRAF inhibitors were not relevant comparators for the patient population for whom the combination of nivolumab with ipilimumab might be used and that immunotherapy was used regardless of BRAF mutation status in most patients. The committee therefore concluded that the most clinically relevant comparison was the cost effectiveness of nivolumab plus ipilimumab in the mixed population (BRAF mutation‑positive and mutation‑negative) for whom immunotherapy was considered appropriate. The committee further concluded that the ERG's exploratory modelling comparing nivolumab plus ipilimumab against ipilimumab in the mixed population (BRAF mutation‑positive and mutation‑negative) could be used for the purposes of decision‑making in this appraisal.

4.11 The committee considered the ERG's preferred base‑case scenario comparing nivolumab plus ipilimumab with ipilimumab alone. This included both BRAF mutation‑negative and mutation‑positive disease in the same ratio as that in CheckMate‑067, with a number of changes. The committee noted that, using the list prices, this increased the company's base case incremental cost‑effectiveness ratio (ICER) for nivolumab plus ipilimumab compared with ipilimumab alone from £10,400 per quality‑adjusted life year (QALY) gained (in the BRAF mutation‑negative population) to £19,300 per QALY gained (for the mixed population). The committee further considered the cost effectiveness when the patient access scheme discounts were applied and noted that the ICER remained below £30,000 per QALY gained.

4.12 The committee considered the cost effectiveness of nivolumab plus ipilimumab compared with pembrolizumab alone, which they considered to be the most clinically relevant comparator in the scope based on the testimony of clinical experts. The committee was disappointed that the company did not include this comparison in its original submission and only provided it during the clarification response stage of the appraisal process. In its clarification response, the company produced a meta‑analysis of the hazard ratios for overall survival and progression‑free survival based on data from CheckMate‑067 and 2 pembrolizumab trials (Keynote‑006 and Keynote‑002) to compare nivolumab plus ipilimumab with ipilimumab alone and with pembrolizumab alone. In the BRAF mutation‑negative population, ipilimumab alone was dominated by pembrolizumab. In the BRAF mutation‑positive population, dabrafenib, vemurafenib and ipilimumab were dominated by pembrolizumab and nivolumab plus ipilimumab and therefore excluded from the analysis. The committee noted that, compared with pembrolizumab, nivolumab plus ipilimumab resulted in an increase of 1.63 and 1.64 QALYs and an ICER of £29,900 and £27,900 per QALY gained for the BRAF mutation‑negative and mutation‑positive populations respectively when list prices were used. The committee was aware that the ERG considered this comparison to be unreliable due to the assumptions made by the company (see section 5.6.2.3 of the ERG report, pages 225–226). The committee noted that the company had access to data from CheckMate‑067 which included a comparison of nivolumab plus ipilimumab with nivolumab alone. If the company had presented that comparison (even though nivolumab monotherapy was not included in the scope) it would have been very helpful to reduce the uncertainty about the cost effectiveness of nivolumab plus ipilimumab compared with the PD‑1 inhibitors, which are regarded as having similar effectiveness in clinical practice. The committee considered that the most appropriate cost‑effectiveness analysis within the scope would have been the comparison of nivolumab plus ipilimumab with pembrolizumab alone in the mixed population (that is, BRAF mutation‑positive and mutation‑negative) using the ERG's preferred model assumptions. The committee concluded, that within the scope, the comparison with pembrolizumab was the most relevant to current UK clinical practice, and there remained some uncertainty about the robustness of this comparison.

4.13 The committee considered the ICERs from the ERG's preferred base case and the company's comparison of nivolumab plus ipilimumab against pembrolizumab, recalculated to include the discounted prices in the patient access schemes for 2 comparators (pembrolizumab and ipilimumab), which are commercial in confidence. The committee took into account uncertainties in the clinical and cost‑effectiveness evidence. The committee recalled that the using the ERG's preferred base case, the ICER for nivolumab plus ipilimumab compared with ipilimumab was less than £30,000 per QALY gained in the mixed population of BRAF mutation‑positive and mutation‑negative advanced melanoma when the patient access scheme prices were used. Although it noted that the ERG's preferred base case had not been applied to the comparison of nivolumab plus ipilimumab against pembrolizumab, it concluded that, on balance, the ICER for nivolumab plus ipilimumab compared to pembrolizumab is likely to be less than £30,000 per QALY gained in the mixed population of BRAF mutation‑positive and mutation‑negative advanced melanoma when the patient access scheme prices were used in the model. It therefore considered that nivolumab plus ipilimumab could be considered a cost‑effective use of NHS resources.

4.14 The committee noted that the company had stated that nivolumab plus ipilimumab was innovative and a step change in the management of advanced melanoma because it treats a life‑threatening and seriously debilitating condition, meets a high unmet need and provides a significant advantage over other treatments used in England. Although the committee accepted that the combination of nivolumab and ipilimumab was associated with a higher response rate than ipilimumab alone, it noted that the increased toxicity made it a promising new advance only for people who are fit enough to tolerate the considerable related adverse effects. The committee did not identify any specific health‑related benefits which had not been captured in the QALY calculation.

4.15 The committee did not formally consider whether the end‑of‑life criteria applied because the technology was considered to be a cost‑effective use of NHS resources without this. However, the committee was aware that as pembrolizumab was considered to be the most clinically relevant comparator and it had not been presented with extension to life evidence for nivolumab and ipilimumab compared with pembrolizumab, it would have been unable to decide whether this criterion had been met.

4.16 The committee was aware of NICE's position statement on the Pharmaceutical Price Regulation Scheme (PPRS) 2014, and in particular the PPRS payment mechanism. It accepted the conclusion 'that the 2014 PPRS payment mechanism should not, as a matter of course, be regarded as a relevant consideration in its assessment of the cost effectiveness of branded medicines'. The committee heard nothing to suggest that there is any basis for taking a different view about the relevance of the PPRS to this appraisal. It therefore concluded that the PPRS payment mechanism was not relevant in considering the cost effectiveness of the technology in this appraisal.

Summary of appraisal committee's key conclusions