4 Consideration of the evidence

The appraisal committee reviewed the data available on the clinical and cost effectiveness of dabrafenib, having considered evidence on the nature of unresectable or metastatic BRAF V600 mutation‑positive melanoma and the value placed on the benefits of dabrafenib by people with the condition, those who represent them, and clinical specialists. It also took into account the effective use of NHS resources.

4.1

The committee considered the clinical need for treatment in people with unresectable or metastatic BRAF V600 mutation‑positive melanoma. The committee heard from the patient expert that metastatic melanoma can be associated with severe and debilitating symptoms. Dabrafenib may sometimes have very rapid therapeutic effects such as improving poor performance status and slowing down the progression of high volume disease, even in those who are severely ill or bed‑ridden. In some cases, this enables people to resume everyday tasks and activities. The patient expert also stated that people may experience very few side effects with dabrafenib. The clinical specialist explained that the side‑effect profile of dabrafenib differed from that of vemurafenib in that there were much lower rates of photosensitivity with dabrafenib, but higher rates of non‑specific fever. The committee concluded that an alternative BRAF‑targeted treatment with a manageable adverse reaction profile, such as dabrafenib, was a potentially valuable additional treatment option for people with unresectable or metastatic melanoma.

4.2

The committee considered the clinical management of unresectable or metastatic BRAF V600 melanoma and discussed the relevant comparators for this appraisal. It was aware that in its evidence submission the company compared dabrafenib with dacarbazine and vemurafenib, and not with any other comparators specified in the scope (including temozolomide in people with brain metastases and ipilimumab in previously treated people). The committee noted the consultees' statement, confirmed by the clinical specialist, that the management of metastatic melanoma is rapidly evolving, with several ongoing clinical trials, and there is uncertainty about how these treatments would be sequenced in future. The committee heard from the clinical specialist that vemurafenib is the current standard of care for people with BRAF V600 mutation‑positive melanoma because it is a targeted therapy. The committee heard from the clinical specialist and noted the statements from consultees that ipilimumab, which is now available for first‑ and second‑line use, is more commonly used in a subgroup of patients with good performance status and in whom the disease is not progressing rapidly. The committee also heard that dacarbazine is no longer routinely used as a first‑line treatment in clinical practice and would now be used only further down the treatment pathway, probably as a third‑line option. The clinical specialist also stated that temozolomide is not a standard drug used in people with brain metastases and that there was no evidence that it results in a better outcome than the targeted therapies. The committee accepted the views of the clinical specialists that in those patients being considered for dabrafenib treatment, the relevant comparator is the alternative BRAF‑targeted therapy vemurafenib.

Clinical effectiveness

4.3

The committee considered the clinical effectiveness of dabrafenib and noted that the key clinical evidence in the company's submission came from the BREAK‑3 trial, which compared dabrafenib with dacarbazine. It discussed the relevance of the BREAK‑3 trial, given that dacarbazine is no longer considered established clinical practice in the UK. It heard from the clinical specialist that dabrafenib and vemurafenib were developed at around the same time and that dacarbazine was the standard of care when the BREAK‑3 trial was conducted. It also noted the comment in the company's submission that although dacarbazine is no longer widely used in this setting, results from BREAK‑3 allow an indirect comparison of dabrafenib with vemurafenib. The committee concluded that the BREAK‑3 trial was relevant for this appraisal.

4.4

The committee examined the results of BREAK‑3 presented by the company. It noted that dabrafenib reduces the risk of progression compared with dacarbazine, with a statistically significant difference of 4.2 months in median time to progression. The committee was aware that for overall survival, the point estimate for the hazard ratio suggested that dabrafenib results in an overall survival benefit, but that the difference did not reach statistical significance. The committee noted the high level of crossover (57%) in the trial and was aware that even after adjusting for crossover using the RPSFT method (see section 3.22) the company's estimate of overall survival gain did not reach statistical significance. The committee was aware of the ERG's concerns about the appropriateness of the RPSFT method (see section 3.22); however, it did not think it was necessary to discuss the issue further because it considered vemurafenib to be the relevant comparator for this appraisal. The committee noted that the relatively low numbers in the dacarbazine arm of the trial (3:1 randomisation) and the high rates of crossover made it very difficult to draw a firm conclusion about the precise effect on overall survival. The committee concluded that compared with dacarbazine, dabrafenib significantly improved progression‑free survival and probably improved overall survival, but it was unable to draw firm conclusions about the magnitude of overall survival benefit.

4.5

The committee went on to discuss the clinical effectiveness evidence for dabrafenib compared with vemurafenib, which is based on an indirect comparison using data from the BREAK‑3 and BRIM‑3 trials. The committee understood the ERG's concern that the proportional hazards assumption that this analysis relied upon was not supported by the trial data. It noted the ERG's comment that lactate dehydrogenase levels differed between BREAK‑3 and BRIM‑3, which suggests that patients in BRIM‑3 had worse prognoses than patients in BREAK‑3 (see section 3.23), and that the median overall survival in the dacarbazine arm in BREAK‑3 was better than in BRIM‑3 (15.6 months compared with 9.6 months).The committee heard from the clinical specialist that although lactate dehydrogenase level is one of many measures of disease severity used in clinical practice, it was not a marker of disease prognosis that could be relied on in isolation. The committee noted that patients were otherwise well matched in the BREAK‑3 and BRIM‑3 trials, including for the stage of disease. The clinical specialist highlighted that BRIM‑3 was conducted in the UK, where there may be a more conservative approach to treatment initiation than outside the UK where BREAK‑3 was conducted, and that the crossover criteria differed in the 2 trials. The committee noted the statement from consultees that 'the BREAK‑3 trial convincingly shows that dabrafenib is as active as vemurafenib, in terms of response rate and progression‑free and overall survival hazard ratios compared with the standard arm, dacarbazine'. The committee also heard from the clinical specialist that the clinical effectiveness of dabrafenib and vemurafenib were not considered to differ in clinical practice and that the choice between the 2 treatments would be largely based on their adverse reaction profiles. The committee considered that BRIM‑3 was a large trial and could show an overall survival benefit with vemurafenib, whereas BREAK‑3 was a smaller trial and it would be more difficult to demonstrate a statistically significant overall survival benefit with dabrafenib. The committee also noted that the progression‑free survival gains in the 2 trials were very similar, and concluded that there was no clear evidence that dabrafenib and vemurafenib differed in clinical effectiveness and that it would not be unreasonable to assume that they have similar effect.

4.6

The committee considered the adverse reactions associated with dabrafenib. It noted that pyrexia, cutaneous squamous cell carcinoma and back pain were the most commonly reported grade‑3 adverse reactions with dabrafenib. It heard from the clinical specialist that cutaneous squamous cell carcinoma was also an issue with vemurafenib. The clinical specialist stated that these targeted therapies are not given to people with serious medical conditions and are administered only in specialist centres with the ability to manage the skin malignancies that may occur. The committee heard from the clinical specialist that some clinicians may prefer dabrafenib to vemurafenib because of lower rates of photosensitivity, which may be a major problem for some patients. The clinical specialist indicated that photosensitivity could also occur with dabrafenib, but in their experience the incidence was much lower than with vemurafenib. The committee concluded that the current evidence suggests that adverse reactions from dabrafenib treatment were not a major concern when compared with those from alternative treatments and that the relative adverse reaction profile would be taken into account in discussions between patients and oncologists on the choice of the most appropriate therapy.

Cost effectiveness

4.7

The committee discussed the company's cost‑effectiveness analysis. The committee had previously heard from the clinical specialist that dacarbazine was not an appropriate comparator and that people eligible for treatment with a BRAF inhibitor would not be given dacarbazine in clinical practice. Therefore it did not consider the cost effectiveness of dabrafenib compared with dacarbazine any further; it restricted its cost‑effectiveness discussion to the comparison with vemurafenib.

4.8

The committee considered the company's economic model and the ERG's critique of the model. It noted the ERG's concerns regarding the structure of the model; in particular, the way the company modelled the clinical effectiveness estimates using a 3‑stage approach, resulting in survival curves that did not appear clinically plausible. The committee also noted the ERG's concerns with the company's indirect comparison (see section 3.23) and noted that the ERG had not carried out exploratory analyses of the clinical or cost effectiveness of dabrafenib compared with vemurafenib.

4.9

The committee considered whether dabrafenib was a cost‑effective use of NHS resources. It noted that the company's base case ICER was £11,000 per QALY gained for dabrafenib compared with vemurafenib, but was much lower than this if a class effect was assumed for dabrafenib and vemurafenib (see section 3.20). The committee noted the ERG's comments on the limitations of the company's model but was aware that the ERG did not conduct exploratory analyses because of its concerns that any estimated ICERs generated from the results of the company's indirect comparison would be unreliable. In the absence of any further numerical analysis by the ERG, the committee could not give an estimate of the most plausible ICER for the comparison of dabrafenib with vemurafenib. However, having considered the lack of evidence of difference in clinical effectiveness between the treatments, and that the overall costs were not different in the economic analysis, the committee concluded that any difference in the cost effectiveness would be small and would fall within the range considered to be a cost-effective use of NHS resources. As a result, dabrafenib should be recommended as an option for treating unresectable or metastatic BRAF V600 mutation‑positive melanoma.

Summary of appraisal committee's key conclusions

Key conclusion

  • Section 1.1: Dabrafenib is recommended, within its marketing authorisation, as an option for treating unresectable or metastatic BRAF V600 mutation‑positive melanoma only if the company provides dabrafenib with the discount agreed in the patient access scheme.

  • Section 4.4: The committee concluded that compared with dacarbazine, dabrafenib improved progression‑free survival and probably improved overall survival, but it was unable to draw firm conclusions about the magnitude of overall survival benefit.

  • Section 4.5: The committee concluded that it was likely that dabrafenib and vemurafenib did not differ in clinical effectiveness and that it would not be unreasonable to assume that they have similar effect.

  • Sections 4.2 and 4.7: The committee concluded that vemurafenib was the most appropriate comparator, and therefore restricted its cost‑effectiveness discussions to the comparison with vemurafenib.

  • Section 4.9: Having considered the lack of evidence of difference in clinical effectiveness between dabrafenib and vemurafenib, and that the overall costs were not different in the economic analysis, the committee concluded that any difference in the cost effectiveness would be small and, would fall within the range considered to be a cost effective use of NHS resources.

Current practice

Clinical need of patients, including the availability of alternative treatments
  • Sections 4.1 and 4.2: The committee heard from the patient expert that metastatic melanoma can be associated with severe and debilitating symptoms. It heard from the clinical specialist that vemurafenib is the current standard of care for people with BRAF V600 mutation‑positive melanoma because it is a targeted therapy.

The technology

How innovative is the technology in its potential to make a significant and substantial impact on health‑related benefits?
  • Section 4.1: The committee heard that dabrafenib may sometimes have very rapid therapeutic effects such as improving poor performance status and slowing down the progression of high-volume disease, even in those who are severely ill or bed‑ridden, in some cases enabling them to resume everyday tasks and activities. The patient expert also stated that people may experience very few side effects with dabrafenib. The clinical specialist explained that the side‑effect profile of dabrafenib differed from that of vemurafenib in that there were much lower rates of photosensitivity with dabrafenib, but higher rates of non‑specific fever. The committee concluded that an alternative BRAF‑targeted treatment with a manageable adverse reaction profile, such as dabrafenib, was a potentially valuable additional treatment option for people with unresectable or metastatic melanoma.

What is the position of the treatment in the pathway of care for the condition?
  • Section 4.2: The committee noted the consultees' statement, confirmed by the clinical specialist, that the management of metastatic melanoma is rapidly evolving, with several ongoing clinical trials, and there is uncertainty about how these treatments would be sequenced in future. The committee accepted the views of the clinical specialists that in those patients being considered for dabrafenib treatment, the relevant comparator is the alternative BRAF targeted therapy vemurafenib.

Adverse reactions
  • Section 4.6: The committee concluded that the current evidence suggests that adverse reactions from dabrafenib treatment were not a major concern when compared with those from alternative treatments and that the relative adverse reaction profile would be taken into account in discussions between patients and oncologists on the choice of the most appropriate therapy.

Evidence for clinical effectiveness

Availability, nature and quality of evidence
  • Section 4.3: The committee noted that the key clinical evidence in the company's submission came from the BREAK‑3 trial, which compared dabrafenib with dacarbazine.

  • Section 4.5: The committee discussed the clinical effectiveness evidence for dabrafenib compared with vemurafenib; which is based on an indirect comparison using data from the BREAK‑3 and BRIM‑3 trials.

Relevance to general clinical practice in the NHS
  • Section 4.3: The committee noted that dacarbazine was the standard of care when the BREAK‑3 trial was conducted and that the results of the BREAK‑3 trial allow an indirect comparison of dabrafenib with vemurafenib. The committee concluded that the BREAK‑3 trial was relevant for this appraisal.

Uncertainties generated by the evidence
  • Sections 4.4 and 4.5: The committee noted the high level of crossover in the BREAK‑3 trial and was aware that even after adjusting for crossover using the rank preserving structural failure time (RPSFT) method, the company's estimate of overall survival gain did not reach statistical significance. The committee understood the ERG's concern that the proportional hazards assumption that the indirect comparison relied upon was not supported by the trial data.

Estimate of the size of the clinical effectiveness including strength of supporting evidence
  • Sections 4.4 and 4.5: The committee concluded that, compared with dacarbazine, dabrafenib significantly improved progression‑free survival by 4.2 months and probably improved overall survival, but it was unable to draw firm conclusions about the magnitude of overall survival benefit. In light of the available evidence, the committee concluded that it was likely that dabrafenib and vemurafenib did not differ in clinical effectiveness and that it would not be unreasonable to assume that they have similar effect.

Evidence for cost effectiveness

Availability and nature of evidence
  • Section 4.7: The committee did not consider the cost effectiveness of dabrafenib compared with dacarbazine, having noted that dacarbazine was not an appropriate comparator as it was no longer used in clinical practice. It restricted its cost‑effectiveness discussion to the comparison with vemurafenib.

Uncertainties around and plausibility of assumptions and inputs in the economic model
  • Section 4.8: The committee noted the ERG's concerns regarding the structure of the model; in particular, the way the company modelled the clinical effectiveness estimates using a 3‑stage approach, resulting in survival curves that did not appear clinically plausible. The committee also noted the ERG's concerns with the company's indirect comparison and noted that the ERG had not carried out exploratory analyses of the clinical or cost effectiveness of dabrafenib compared with vemurafenib.

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?
  • Section 3.16: Treatment‑specific EQ‑5D utility data for pre‑progression and post‑progression, derived directly from BREAK‑3, were used in the model for dabrafenib and dacarbazine. In the absence of comparable EQ‑5D utility data for vemurafenib, the company assumed that the vemurafenib utility values would be the same as those for dabrafenib.

What are the key drivers of cost effectiveness?
  • Section 3.20: The company conducted a series of deterministic sensitivity analyses. For the comparison of dabrafenib with dacarbazine, overall survival was the key driver of the cost‑effectiveness results. The key driver of the cost‑effectiveness result for the comparison of dabrafenib with vemurafenib was the progression‑free survival assumption.

Most likely cost‑effectiveness estimate (given as an ICER)
  • Section 4.9: The committee noted that the company's base case ICER was £11,000 per QALY gained for dabrafenib compared with vemurafenib, but was much lower than this if a class effect was assumed for dabrafenib and vemurafenib. In the absence of any further numerical analysis by the ERG, the committee could not give an estimate of the most plausible ICER for the comparison of dabrafenib with vemurafenib.

Additional factors taken into account

Patient access schemes (PPRS)
  • Section 2.3: Novartis has agreed a patient access scheme with the Department of Health that makes dabrafenib available with a discount applied at the point of purchase or invoice. The size of the discount is commercial in confidence.

Equalities considerations and social value judgements
  • No equality issues relevant to the committee's recommendations were raised.