Health technology assessment
A health technology assessment of the new drugs for inoperable or metastatic malignant melanoma patients
Health technology assessment
|Updated
In this health technology assessment we have compared the relative effectiveness and cost-effectiveness of seven new drugs used for the treatment of advanced malignant melanoma patients in the Norwegian setting.
Key message
In this health technology assessment we have compared the relative effectiveness and cost-effectiveness of seven new drugs used for the treatment of advanced malignant melanoma patients in the Norwegian setting. The drugs are: cobimetinib, dabrafenib, ipilimumab, nivolumab, pembrolizumab, trametinib and vemurafenib. The clinical endpoints are overall survival, progression free survival, health related quality of life and serious adverse events.
Our results are based upon 17 randomized controlled trials. Our conclusions for the relative effectiveness of the included drugs or combinations of drugs are based upon network meta-analyses using both direct and indirect evidence with dacarbazine as a common comparator. We ranked the different treatments in terms of their likelihood of leading to the best results for each endpoint. The rankings were interpreted cautiously taking into account the quality of the evidence. The cost-utility analysis was based on a probabilistic discrete-time Markov cohort model. Our findings:
- For overall survival: Nivolumab and pembrolizumab in monotherapy, as well as nivolumab combined with ipilimumab, vemurafenib combined with cobimetinib, and dabrafenib combined with trametinib seemed to have a higher probability of good performance than the other available treatment strategies.
- For progression free survival: Dabrafenibin combination withtrametinib and vemurafenib combined with cobimetinib seemed to have a higher probability of good performance than the other available treatment strategies.
- For health related quality of life:Evidence from pairwise comparisons for four interventions reported better health related quality of life in the intervention groups.
- For serious adverse events: We could not establish any differences between the treatment strategies. However, pembrolizumab and nivolumab seemed to have a higher probability of fewer serious adverse events than the other treatment strategies.
- We assessed the quality of the evidence for overall survival and progression free survival from the network meta-analyses to be moderate or high for the majority of our comparisons. For serious adverse events, we assessed the quality to be low or very low in most of our assessments.
- The analysis of cost-effectiveness was conducted using the maximum pharmacy retail prices, due to the fact that negotiated discounts are hidden from the general public as per contract between the Drug Procurement Cooperation system and the manufacturers.
None of the interventions are cost-effective at the maximum pharmacy retail prices, and the budget impact if the interventions are accepted in clinical practice are substantial. Drug price reductions in the region of 63 to 84 percent would be necessary to improve the cost-effectiveness and reduce the budget impact.
Background
The Commissioners Forum, in the “National system for the introduction of new health technologies within the specialist health service” has requested a health technology assessment to compare effectiveness and cost-effectiveness of the new drugs used for inoperable or metastatic malignant melanoma patients in the Norwegian setting.
The drugs are: cobimetinib, dabrafenib, ipilimumab, nivolumab, pembrolizumab, trametinib and vemurafenib. These can be used as monotherapy or in combination with each other.
The incidence of malignant melanoma in Norway is among the highest in the world with approximately 1,500 persons diagnosed annually.
Objective
To assess the effectiveness and cost-effectiveness of seven new drugs used for inoperable or metastatic malignant melanoma patients relative to each other in the Norwegian setting.
Method
In this health technology assessment, clinical effectiveness was measured in terms of overall survival, progression free survival, health-related quality of life and serious adverse events. In the economic evaluation the primary endpoint was the incremental cost-effectiveness ratio with effectiveness measured in quality-adjusted life-years. Results were also presented in life years gained, in net health benefits, scatterplots, probability of being cost-effective and value of information analysis.
We performed a systematic search for randomized controlled trials in February 2015 in relevant bibliographic databases, Google Scholar and websites of selected health technology assessment agencies. We updated the search in September 2015. We contacted relevant pharmaceutical companies to obtain additional information.
Two reviewers worked independently to identify relevant publications. One review author extracted data from the included references and another review author verified the data.
We performed network meta-analyses where appropriate according to population, intervention, control and outcome. We ranked the different treatments in terms of their likelihood of leading to the best results for each endpoint. This we did by help of the surface under the cumulative ranking curve (SUCRA).
The quality of the direct evidence, indirect evidence, and the combined evidence from the network meta-analyses were evaluated by two review authors using the GRADE working group approach for network meta-analysis.
Our cost-utility analysis were based on a probabilistic discrete-time Markov cohort model with three health states, progression free survival, progressed disease and death. We adjusted the baseline transition probabilities with the hazard ratios from the network meta-analysis. Clinicians in the field provided information relevant for the estimation of costs as well as modelling assumptions.
Due to the fact that negotiated discounts are hidden from the general public, as per contract between the Drug Procurement Cooperation system and the manufacturers, the analysis of cost-effectiveness was conducted using the official maximum pharmacy retail prices.
Results
Our results are based upon 17 randomized controlled trials, presented in 40 publications. Our conclusions for the relative comparisons of effectiveness for the included drugs or combinations of drugs are based upon network meta-analyses using both direct and indirect evidence with dacarbazine as a common comparator. We ranked the different treatments in terms of their likelihood of leading to the best results for each endpoint. The rankings were interpreted cautiously taking into account the quality of the evidence.
Our findings:
- For overall survival: Nivolumab and pembrolizumab in monotherapy, as well as nivolumab combined with ipilimumab, vemurafenib combined with cobimetinib, and dabrafenib combined with trametinib seemed to have a higher probability of good performance than the other available treatment strategies. We assessed the quality of the evidence to be moderate for nivolumab and vemurafenib combined with cobimetinib; low for nivolumab combined with ipilimumab, and dabrafenib combined with trametinib, and very low for pembrolizumab.
- For progression free survival: Dabrafenib combined with trametinib and vemurafenib combined with cobimetinib seem to have a higher probability of good performance than the other available treatment strategies. We assessed the quality of the evidence to be moderate in both cases.
- For health related quality of life: Due to insufficient data we did not perform a network meta-analysis for health related quality of life. Evidence from pairwise for four comparisons reported better health related quality of life in the intervention groups.
- For serious adverse events: We could not establish any differences between the available treatment strategies. However, the ranking suggests that pembrolizumab and nivolumab have a higher probability of fewer serious adverse events than the other available treatment strategies. We assessed the quality of the evidence to be low in both cases.
- We assessed the quality of the evidence for overall survival and progression free survival from the network meta-analyses to be moderate or high for the majority of our comparisons. For serious adverse events, we assessed the quality to be low or very low in most of our assessments.
- The economic model predicted a median survival of about 12.5 months for ipilimumab and about 19 months for nivolumab, pembrolizumab and the combination nivolumab and ipilimumab. The median survival for the BRAF/MEK inhibitors dabrafenib, vemurafenib and trametinib in monotherapy was about 11 months, and for the combinations dabrafenib and trametinib as well as vemurefanib and cobimetinib, 17.5 months. In comparison, the median survival of dacarbazine was 9 months.
- None of the interventions were cost-effective at the maximum pharmacy retail prices. The ranking of the interventions and the budget impacts may however change as a result of price changes.
- The first analysis included all the interventions from the network meta-analysis. Nivolumab had an incremental effect of 0.82 quality adjusted life years and an incremental cost-effectiveness ratio against dacarbazine of about NOK 1.1 million per quality adjusted life year gained. The combination vemurafenib and cobimetinib had an incremental effectiveness of 0.07 quality adjusted life years and an incremental cost-effectiveness ratio of about NOK 19.8 million per quality adjusted life year gained against nivolumab.
- When we restricted the analysis to the BRAF and MEK inhibitors, dabrafenib had an incremental effect of 0.36 quality adjusted life years and an incremental cost-effectiveness ratio compared to dacarbazine of approximately NOK 2.2 million per quality adjusted life year gained. The combination vemurafenib and cobimetinib had an incremental effect of 0.53 quality adjusted life years and incremental cost-effectiveness ratio compared to dabrafenib of about NOK 2.9 million per quality adjusted life year gained. The BRAF and MEK inhibitor monotherapies (dabrafenib, vemurafenib, trametinib) all had very similar costs and effectiveness. The same applied to the BRAF/MEK combinations (dabrafenib and trametinib or vemurafenib and cobimetinib), but at a higher level of costs and effectiveness.
- When the analysis was limited to the immunotherapies, nivolumab in monotherapy was not clearly separable from pembrolizumab in monotherapy, which had similar effectiveness and costs. The combination nivolumab and ipilimumab had about the same level of effectiveness as nivolumab and pembrolizumab in monotherapy, but at a higher level of costs.
- The expected value of partial perfect information analysis identified the efficacy data used in the model as the dominating source of uncertainty, followed by the health related quality of life data, costs and serious adverse events.
- The maximum pharmacy retail prices would have to be reduced by approximately 79 percent for dabrafenib, 83 percent for the combination dabrafenib and trametinib, 81 percent for vemurafenib, 84 percent for the combination vemurafenib and cobimetinib, 83 percent for trametinib, 75 percent for ipilimumab, 63 percent for nivolumab, 76 percent for the combination nivolumab and ipilimumab, and 64 percent for pembrolizumab in order to achieve incremental cost-effectiveness ratios of NOK 500.000 per quality adjusted life year gained against dacarbazine.
- If the prices for the new interventions were reduced by 63 to 84 percent (depending on intervention) from the maximum pharmacy retail prices, the annual budgetary savings could be about NOK 277 million and the accumulated budgetary savings over a 5 year period NOK 1,387 million.
Discussion
We found only two head to head comparison for the included drugs as monotherapies, and five direct comparisons of combination treatment versus monotherapy. None of the included trials compared a BRAF inhibitor (dabrafenib or vemurafenib) head to head with a drug acting on the immune system. The best available comparisons are the indirect evidences via dacarbazine as a common comparator. All the interventions could be included in the network meta-analyses for overall survival, progression free survival and serious adverse events. Of the endpoints studied, we consider overall survival to be of higher importance than progression free survival, since progression free survival is a surrogate endpoint. Health related quality of life and serious adverse events are of importance for the patients. However, from the available literature we were not able to find data usable for our network meta-analysis for health related quality of life, and the quality of the evidence for serious adverse events were low or very low in most of our assessments.
We only included randomized controlled trials. Our endpoints were all well-defined and harmonized in their definitions across the trials.
Based on a qualitative assessment, the results of the pairwise estimates and network meta-analyses are consistent.
The number of available interventions for patients with advanced malignant melanoma is evolving rapidly at the moment. Many of the interventions in this health technology assessment have just reached marketing authorization in Norway, and the available evidence from randomized controlled trials is quite limited. Hence, the clinical efficacy data in our report have the uncertainty that the majority of the evidence for the included comparisons were based upon a single study. It cannot be ruled out that new evidence from randomized controlled trials have the potential to change the ranking of the interventions both with regards to effectiveness and cost-effectiveness.
We believe that the economic model distinguished the interventions fairly well with regards to costs and overall survival, but not so well with regards to health related quality of life, which is a crucial input for life prolonging interventions. This emphasises the need to make separate judgments and not relying on the cost-effectiveness evidence alone.
We are extrapolating effectiveness data beyond the clinical trial follow-up period for nivolumab, pembrolizumab, the combinations nivolumab and ipilimumab and vemurafenib and cobimetinib. There is uncertainty with regards to the correct treatment duration, both for the new immunotherapies and the BRAF and MEK inhibitors. Also, the results are dependent on that the treatment effects are the same across the three incremental cost-effectiveness analyses.
To our knowledge a relative comparison for the different new drugs used for inoperable or metastatic malignant melanoma patients has not been done by any others, neither for effectiveness nor for cost-effectiveness.
Conclusion
All conclusions are given with respect to the current state of the evidence and with the reservation that new evidence from randomized controlled trials can change the ranking of the interventions both with regards to effectiveness and cost-effectiveness (one of the interventions still do not have marketing authorization).
None of the interventions are cost-effective at the maximum pharmacy retail prices. The budgetary impact of accepting some or all of the new interventions in clinical practice can be substantial, potentially diverting resources away from other interventions or treatment areas with better cost-effectiveness. The budgetary impact and incremental cost-effectiveness ratios can however be reduced through price reductions. We believe that drug price reductions in the region of 63 to 84 percent, depending on drug, would be necessary for the interventions to represent cost-effective use of resources in the Norwegian setting.
We find it difficult to separate the new PD-1 immunotherapies nivolumab and pembrolizumab with respect to cost-effectiveness. If the new immunotheraphies in monotherapy are accepted in clinical practice, we expect increased effectiveness compared to ipilimumab in monotherapy, but at an increased cost. The potential budgetary savings with price reductions from the maximum pharmacy retail price may be as high as NOK 131 million per year across the immunotherapies, if a cost-effectiveness level of NOK 500,000 per gained quality adjusted life year is assumed.
Based on the cost-effectiveness results, we cannot argue that any of the BRAF or MEK inhibitor monotherapies (dabrafenib, vemurafenib, trametinib), should be preferred over another, or that any BRAF/MEK combination (dabrafenib and trametinib or vemurafenib and cobimetinib), should be preferred over another. However, the combination therapies are more likely to give the highest quality adjusted life year gains in the long run, at an increased cost. For the BRAF/MEK inhibitors, the potential budgetary savings with price reductions may be as high as NOK 147 million per year.