3 Committee discussion
The appraisal committee (section 5) considered evidence submitted by Pierre Fabre and a review of this submission by the evidence review group (ERG). See the committee papers for full details of the evidence.
People with unresectable or metastatic BRAF V600 mutation-positive melanoma would welcome an additional treatment option
3.1 Melanoma often affects people at a younger age than some other cancers. It has a substantial effect on patients, their carers and wider society. Advanced (unresectable or metastatic) melanoma can cause severe and debilitating symptoms and is life threatening. The clinical and patient experts stated that they would welcome the availability of another targeted combination treatment for BRAF V600 mutation-positive melanoma, in addition to the currently available combination of dabrafenib with trametinib. Encorafenib with binimetinib has a different toxicity profile. The patient expert emphasised the value of having a choice of treatments available and that a favourable safety profile would improve the quality of life of patients. The committee concluded that people with unresectable or metastatic BRAF V600 mutation-positive melanoma would welcome an additional treatment option.
The position in the care pathway is uncertain for targeted therapies of BRAF V600 mutation-positive melanoma
3.2 The committee acknowledged that the management of unresectable or metastatic BRAF V600 mutation-positive melanoma is changing rapidly with the availability of new immunotherapy and other treatments. The clinical experts explained that standard treatment is either immunotherapies such as nivolumab, ipilimumab or pembrolizumab (alone or in combination) or targeted therapy. Targeted therapy is usually dabrafenib in combination with trametinib but sometimes dabrafenib, trametinib or vemurafenib are taken alone. The clinical experts explained that there is no consensus on whether first-line treatment should be targeted therapies or immunotherapies. They noted that there may be a preference to use immunotherapies first-line for people who have a good performance status, however this is not standard practice across the NHS. Most patients with BRAF V600 mutation-positive melanoma are likely be offered a targeted therapy at some point in the treatment pathway. The committee concluded that the management of advanced melanoma is evolving, making it difficult to determine the position of targeted therapies in the care pathway for mutation-positive melanoma.
The clinical evidence is relevant to clinical practice in England but does not provide a direct comparison with the comparator specified in the scope
3.3 The clinical evidence comes from COLUMBUS. This is an open-label, randomised trial of encorafenib plus binimetinib compared with vemurafenib in people with unresectable or metastatic BRAF V600 mutation-positive melanoma. It included people who had not had treatment for advanced disease, and people with disease that had progressed on or after first-line immunotherapy. The ERG explained that very few people in COLUMBUS had brain metastases, and that people with a poor performance status (2 or above) were excluded from the trial. However, the clinical experts noted that people with unresectable or metastatic BRAF V600 mutation-positive melanoma generally have good performance status even with high-volume disease. The ERG highlighted that only 6% of people in the trial had had treatment with immunotherapy for metastatic disease, which may not reflect the population in the NHS who would have encorafenib and binimetinib. The committee recalled that there is no consensus on whether immunotherapies or targeted therapy should be used first in advanced mutation-positive melanoma (see section 3.2). It noted that the clinical experts do not consider that immunotherapy will affect the response to subsequent targeted therapies, because of their different mechanism of action. The ERG noted that although COLUMBUS has some limitations, overall it was well conducted and generally representative of patients with unresectable or metastatic BRAF V600 mutation-positive melanoma who would be treated in the NHS. The committee concluded that COLUMBUS was well conducted and representative of clinical practice in England but noted that it does not directly compare encorafenib with binimetinib against dabrafenib with trametinib, which is the comparator specified in the NICE scope.
Encorafenib with binimetinib is more clinically effective than vemurafenib in the clinical trial population
3.4 Progression-free survival, assessed by blinded independent review, is the primary endpoint in COLUMBUS. At a median follow up of 11.8 months, median progression-free survival for encorafenib with binimetinib is 14.9 months (95% confidence interval [CI] 11.0 to 18.5) compared with 7.3 months (95% CI 5.6 to 8.2) for vemurafenib. The committee noted that the hazard ratio (HR) for progression-free survival for encorafenib with binimetinib compared with vemurafenib is 0.54 (95% CI 0.41 to 0.71), representing an estimated 46% reduction in the risk of disease progression. It also noted that investigator-assessed progression-free survival is very similar to the results from blinded independent review. In addition, median overall survival is longer for encorafenib with binimetinib (33.6 months; 95% CI 24.4 to 39.2) compared with vemurafenib (16.9 months; 95% CI 14.0 to 24.5), representing a 39% reduction in the risk of death (HR 0.61, 95% CI 0.47 to 0.79; p<0.0001). The committee concluded that encorafenib with binimetinib is more clinically effective than vemurafenib in the clinical trial population.
The clinical effectiveness of encorafenib with binimetinib is likely to be similar to dabrafenib with trametinib, but this is associated with uncertainty
3.5 The committee noted that in the absence of trial evidence directly comparing encorafenib with binimetinib against dabrafenib with trametinib, the company did network meta-analyses (NMAs) to indirectly estimate progression-free survival, overall survival, health-related quality of life and incidence of grade 3 or 4 adverse events. The NMAs included 7 randomised controlled trials of BRAF-inhibitor therapies, all reporting clinical efficacy and safety data, of which 5 reported health-related quality of life data. The committee noted that the company's NMAs show no statistically significant difference between the treatment combinations for investigator-assessed progression-free survival, overall survival, adverse events or health-related quality of life, and that the credible intervals are wide for all base-case and sensitivity analyses. It noted that these results should be viewed with caution because of methodological limitations highlighted by both the company and the ERG. However it acknowledged clinical expert opinion suggesting that the clinical effectiveness of encorafenib with binimetinib is likely to be similar to dabrafenib with trametinib. The committee concluded that the clinical effectiveness of encorafenib and binimetinib is likely to be similar to dabrafenib with trametinib, but this is associated with uncertainty.
3.6 The committee noted that the frequency of adverse events in COLUMBUS is similar for encorafenib with binimetinib, compared with vemurafenib. The trial does not include the relevant comparator (dabrafenib with trametinib), however the committee noted that adverse events for encorafenib with binimetinib were infrequent and serious adverse events were low. The clinical experts stated that encorafenib with binimetinib may offer a more favourable side-effect profile than dabrafenib with trametinib. They explained that some people taking dabrafenib with trametinib develop pyrexia during the first month of treatment, which may lead to hospitalisations and dose interruptions. This did not seem to happen as frequently for encorafenib with binimetinib in COLUMBUS. The committee concluded that encorafenib with binimetinib may have a more favourable adverse-effects profile than dabrafenib with trametinib.
3.7 The company presented a 3 state transition model comparing encorafenib with binimetinib against dabrafenib with trametinib. The progression-free and post-progression states included tunnel states to account for people being 'on' or 'off' primary treatment, reflecting clinical practice. Extrapolation of outcomes beyond the trial period was done using parametric curves that were selected based on best fit to the trial data and visual inspection. The committee concluded that the model structure is appropriate for decision making.
3.8 The committee noted that the efficacy and clinical parameters in the model were derived from COLUMBUS data, the company's NMA results and cancer registry data. The committee noted that the clinical effectiveness of encorafenib with binimetinib was based on an indirect comparison against dabrafenib with trametinib, not direct trial evidence (see section 3.5). It acknowledged comments from the ERG that it would be inappropriate to model any difference in efficacy or utility because the company's NMAs show no statistically significant differences between encorafenib with binimetinib and dabrafenib with trametinib for investigator-assessed progression-free survival, overall survival, adverse events and health-related quality of life. The ERG suggested a cost-minimisation approach as their preferred method of analysis, based on an assumption of clinical equivalence. The clinical experts stated that it is biologically plausible that encorafenib with binimetinib and dabrafenib with trametinib are similar in efficacy, in the absence of robust evidence suggesting any differences. However, the committee noted that equivalent efficacy has not been proven. It concluded that the differences in efficacy between treatments using the company's approach are small, and the assumptions in the company's model can be used for decision making alongside the ERG's cost-minimisation approach.
3.9 In the company model, treatment-related costs are calculated from time on treatment, administration costs, relative dose-intensity multipliers and drug costs. The company assumed that the relative dose-intensity multiplier for encorafenib and binimetinib is lower than for dabrafenib with trametinib. The committee noted that the ERG does not consider this analysis to be robust. The ERG stated that both treatment combinations should have the same multipliers, time on treatment and administration costs (given that they have the same mode of delivery). The ERG therefore corrected the company's base case for relative dose intensity. The committee concluded that this approach is appropriate.
Taking into account all the commercial arrangements, encorafenib with binimetinib is a cost-effective use of NHS resources
3.10 The committee saw in confidence the company's corrected base case and the ERG's preferred base case using the confidential patient access schemes for encorafenib and binimetinib and for dabrafenib with trametinib. It concluded that encorafenib with binimetinib is a cost-effective use of NHS resources and can be recommended for routine commissioning for treating unresectable or metastatic BRAF V600 mutation-positive melanoma.