Lenalidomide for the treatment of multiple myeloma in people who have received at least 2 prior therapies

4 Consideration of the evidence

4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of lenalidomide, having considered evidence on the nature of multiple myeloma and the value placed on the benefits of lenalidomide by people with the condition, those who represent them, and clinical specialists. It was also mindful of the need to take account of the effective use of NHS resources.

4.2 The Committee understood that multiple myeloma is an incurable disease. It was aware that the disease and its course are heterogeneous and that for relapsed multiple myeloma the choice of therapy for a particular person is influenced by the initial treatment and their response to it, the inherent characteristics of the disease and the person's performance status and preferences. The Committee heard from clinical specialists and patient experts that lenalidomide is an important advance in the treatment of multiple myeloma and could be considered as an alternative to bortezomib (currently recommended as a treatment option in NICE technology appraisal guidance 129) at first relapse. The Committee noted the importance that patients, their carers and physicians placed on having effective options to treat multiple myeloma at presentation and at subsequent relapses. However, it understood that the optimal sequence of agents to use is as yet unclear and depends on several factors, including a person's treatment history, comorbidities and disease characteristics.

4.3 The Committee understood that, in accordance with current NICE guidance (NICE technology appraisal guidance 129), bortezomib is routinely used in clinical practice for the treatment of progressive multiple myeloma in people who are at first relapse having received only one prior therapy. Therefore it considered bortezomib to be the most appropriate comparator for lenalidomide in people who have had only one prior therapy. The Committee noted that bortezomib is usually used in combination with dexamethasone, but that there is variation in clinical practice and limited formal evidence for the superior efficacy of this combination compared with bortezomib monotherapy.

4.4 The Committee considered the options available for the treatment of multiple myeloma at second and subsequent relapse. Current NICE guidance restricts the use of bortezomib to first relapse because the use of bortezomib at subsequent relapses was found not to be cost effective (NICE technology appraisal guidance 129), with ICERs of £77,000 or more per life year gained. Thalidomide is not licensed for this indication, and an application for a licence for the treatment of relapsed or refractory multiple myeloma was withdrawn. However, thalidomide is used as a treatment option for multiple myeloma within the NHS, although the extent of this use is not known. The Committee also noted a statement from the manufacturer that there is a lack of evidence for the efficacy of thalidomide for this indication, particularly after failure of two or more therapies. The Committee understood that a variety of other regimens could be used at second and subsequent relapse, but that no studies had been identified that have demonstrated the superiority of these compared with dexamethasone alone. Therefore it accepted that in people who have received two or more prior therapies, high-dose dexamethasone was a reasonable comparator for lenalidomide.

4.5 The Committee discussed the RCTs comparing len/dex with dexamethasone alone for the management of relapsed multiple myeloma. It noted that TTP (the primary outcome) was statistically significantly increased in the len/dex arm for the whole trial population as well as in subgroups of people who had received prior therapy with bortezomib or thalidomide. It considered that the RCTs provided evidence that overall survival and response rates were also higher with len/dex compared with dexamethasone alone. The Committee concluded that the len/dex combination improved outcomes in people with relapsed multiple myeloma when compared with dexamethasone. This included people who had received either one or two or more prior therapies, and when prior therapies included the use of thalidomide.

4.6 The Committee next discussed the relative effectiveness of len/dex compared with bortezomib. It noted that the evidence for the effectiveness of len/dex compared with bortezomib monotherapy was derived from an indirect comparison via the common comparator of high-dose dexamethasone. It considered that there was uncertainty in the results of the indirect comparison because of heterogeneity between the studies, such as differences in the regimen of dexamethasone and the definition of response. The Committee noted that there was additional uncertainty in interpreting the context of current practice, as it understood that bortezomib is usually used in combination with dexamethasone in clinical practice.

4.7 The Committee discussed the adverse effects associated with lenalidomide. It noted that from the patients' viewpoint lenalidomide is associated with a more favourable adverse effect profile than most other regimens and agents used in the management of relapsed multiple myeloma. It heard from clinical specialists and patient experts that lenalidomide might be particularly useful for people with pre-existing peripheral neuropathy, in whom the use of bortezomib at first relapse is restricted. However, the Committee noted that lenalidomide is associated with a statistically significant increased risk of venous thrombosis and embolism. It heard from clinical specialists that this risk is usually managed with prophylaxis in the form of low-dose aspirin in people with multiple myeloma. However, in people with a history of venous thromboembolism or other relevant risk factors, the use of warfarin or low-molecular-weight heparin would be considered. The Committee heard that with such prophylaxis the risk would return to baseline levels. The additional cost incurred for the management of people with multiple myeloma would be minimal if low-dose aspirin was used, but could have an impact if either low-molecular-weight heparin or warfarin was needed.

4.8 The Committee considered the manufacturer's economic evaluation of the use of lenalidomide and the critique from the ERG. It accepted that the general structure of the manufacturer's model was reasonable. It considered that the subgroups in the model were those relevant to decision-making in routine clinical practice. It discussed the sensitivity and scenario analyses presented by the manufacturer, as well as those explored by the ERG using the manufacturer's model. In particular, the Committee discussed the methods used for adjustment for the crossover effect in the RCTs, the extrapolation of survival data, the costs of medical management and administration of bortezomib therapy, and the utility values reflecting health-related quality of life for the pre-progression and post-progression states and from adverse effects.

4.9 The Committee noted that the trial results included a crossover effect and considered whether it was appropriate to use data from historical MRC trials to predict survival for people treated with dexamethasone in this population in the absence of an unbiased estimate from the trials of lenalidomide. The Committee was aware that the MRC data were derived from trials of agents in first-line therapy for multiple myeloma. Despite this, it accepted that these data represented the best available survival data for people with multiple myeloma to be used in extrapolation of overall survival in the current analysis. The Committee also noted that use of these data assumed that dexamethasone monotherapy was a suitable proxy (in the absence of more specific evidence) for all anti-myeloma therapies used in relapse. The Committee also considered that there was no evidence to indicate that the effectiveness of dexamethasone in relation to survival had changed over time since the MRC trials. It accepted the statements from the clinical specialists indicating that where improvements were noticed these were likely to be attributable to the use of the newer agents and stem-cell transplantation.

4.10 The Committee considered the ERG's exploratory reanalysis with an improved fit of the len/dex overall survival curve to the trial data and calibration of the dexamethasone overall survival curve to predict mean (and not median) overall survival based on a risk equation for survival derived from the MRC trials. The Committee considered that the ERG's approach to modelling overall survival in both the len/dex and dexamethasone arms was valid and resulted in more plausible estimates of cost effectiveness than those presented by the manufacturer. The Committee noted that these adjustments to the modelling of survival may have different effects in different subgroups and that the ERG's adjustments had been made separately to subgroups defined according to number of prior therapies.

4.11 The Committee considered the base-case ICERs resulting from the manufacturer's economic analysis, as well as the results of the ERG's exploratory analysis using the alternative approach to the modelling of overall survival. It noted that, in the manufacturer's base case, none of the ICERs for lenalidomide for the subgroups with only one prior therapy were within the range that would normally be considered a cost-effective use of NHS resources. The Committee noted that the comparison of len/dex with bortezomib in the subgroup of people who had received only one prior therapy resulted in a high ICER. The Committee also considered that the ICER for the comparison with bortezomib would increase further if the model took into account: the bortezomib response-based rebate scheme (as described in NICE technology appraisal guidance 129); the lower costs of bortezomib administration suggested by the ERG; the higher maximum number of cycles of bortezomib; and the likely dosage reduction for bortezomib. When the ERG's approach to modelling overall survival was used, the ICER was more than £69,000 per QALY gained for the comparison of lenalidomide with dexamethasone in people who had received one prior therapy only. For the comparison with dexamethasone in people who had received one prior therapy and that therapy was thalidomide, the ICER was more than £56,000 per QALY gained.

4.12 The Committee considered other issues with the base-case analysis. The model did not fully include costs and utility decrements owing to adverse effects, and the Committee considered that if appropriate costs and disutilities for adverse effects and anti-thrombosis prophylaxis were used in the model, the ICERs for lenalidomide would increase. It also noted that the utility in the model for the pre-progression state was that of the normal population at age 54 years, and that this is considerably younger than the average age of people who usually develop multiple myeloma. The Committee noted the results of the exploratory analysis by the ERG, which showed that using lower administration costs for bortezomib, using higher costs for routine medical management and modelling the bortezomib response-based rebate scheme all had the effect of increasing the ICERs for lenalidomide for the subgroup of patients who had received one prior therapy. The Committee concluded that, in the light of these additional issues, the most plausible ICERs in all subgroups would be higher than those stated in 4.11.

4.13 The Committee discussed the updated analysis presented by the manufacturer. It noted that the manufacturer had chosen not to present any new analysis for people who had received only one prior therapy. The Committee discussed whether there were any further factors that would have a bearing on its considerations about the cost effectiveness of lenalidomide in this patient group. These included the degree of certainty in the ICERs, the severity of the illness experienced by people with multiple myeloma who have received one prior therapy and the innovative nature of lenalidomide. It did not identify any factors that would alter its conclusions based on the evidence currently available. Overall, the Committee concluded that the use of lenalidomide for the treatment of multiple myeloma in people who had received only one prior therapy would not be a cost-effective use of NHS resources.

4.14 The Committee considered the subgroups of people who had received two or more prior therapies, including the subgroup who had received thalidomide as one of these therapies. When the ERG's approach to modelling overall survival was used, the ICERs for lenalidomide for these subgroups increased to at least £47,100 per QALY gained for those who had received two or more prior therapies, and to at least £43,600 per QALY gained for those who had received two or more prior therapies of which one was thalidomide.

4.15 The Committee discussed the updated analysis from the manufacturer and the exploratory reanalysis by the ERG for people who had received two or more prior therapies. The Committee concluded that the changes to the len/dex curve, utilities and costs had been implemented appropriately. The Committee noted that the variable that had the greatest impact on cost effectiveness was the method of calibrating the dexamethasone overall survival curve in the economic model (that is, to the predicted median or mean survival for the trial population) from the risk equation for survival derived from the MRC trials. The Committee noted that the data from the MRC trials were complete, with most participants having reached the outcome of interest (that is, there was very little censoring of the data), and that in such a situation the mean was a better estimate of average survival. It considered that, since the mean overall survival was used in the calculations of cost effectiveness, calibrating the overall survival predicted by the MRC data calibration model to the mean was more representative of the costs and benefits for this population. In addition, the Committee noted that when median survival was used to calibrate the survival curve, the improvement in overall survival predicted by the model was out of proportion to the observed improvement in progression-free survival from the lenalidomide trials. This relationship between the progression-free survival and overall survival from the lenalidomide trials was, however, maintained when calibrating the overall survival curve in the dexamethasone arm to the mean survival predicted from the MRC trials. The Committee understood that the choice between the use of mean or median survival was a scientific judgement, but concluded that for the purposes of decision-making in this situation, the ICER estimates from using the mean were most appropriate.

4.16 The Committee accepted that the patient access scheme for lenalidomide was correctly implemented by the manufacturer in the economic evaluation. It noted that, in the economic model, the patient access scheme was included by capping the maximum cost of lenalidomide for an individual patient at 26 cycles of 28 days each, equivalent to 2 years. The Committee noted that the manufacturer stated that treatment interruptions within cycles were generally short, and that no patients missed entire cycles in the clinical trial. The cost of lenalidomide per cycle in the model was adjusted, as in the base case, to take into account the treatment reductions and interruptions noted in the first 23 cycles in the trials. The Committee concluded that the relevant and appropriate ICERs upon which to make a decision were £43,800 per QALY gained for the subgroup of patients who had received two or more prior therapies and £41,300 per QALY gained for the subgroup who had received two or more prior therapies including thalidomide. These ICERs represented the cost effectiveness of lenalidomide with the patient access scheme triggered after 26 cycles, regardless of any dose reductions or treatment interruptions that may occur during a cycle, and were the basis for the Committee's decisions.

4.17 The Committee considered supplementary advice from NICE that should be taken into account when appraising treatments that may extend the life of patients with a short life expectancy and that are licensed for indications that affect small numbers of people with incurable illnesses. For this advice to be applied, all the following criteria must be met.

4.18 The Committee next discussed whether the subgroup of people with multiple myeloma who had received two or more prior therapies, and the benefit provided by lenalidomide, fulfilled the criteria for consideration as an appraisal of a life-extending, end-of-life treatment. The Committee noted from the clinical trials and the MRC data that normal life expectancy without lenalidomide was unlikely to be greater than 24 months and was potentially as low as 9 months. The Committee considered that evidence from the lenalidomide trials suggested that lenalidomide increased survival by more than 3 months compared with dexamethasone, and that crossover in the dexamethasone arm means that this benefit is likely to have been underestimated. The Committee considered that the potential alternatives, thalidomide and bortezomib, were unlikely to be routinely available on the NHS, as discussed in section 4.4. The Committee noted from the manufacturer's submission that the estimated eligible population was approximately 2100. In summary, the Committee was satisfied that the population and the technology of interest meet the criteria for accepting that this is an appraisal of a life-extending, end-of-life treatment and that the evidence presented for this consideration was supported by robust data.

4.19 The Committee subsequently considered the ERG's cost-effectiveness estimates using the manufacturer's model in the context of a life-extending, end-of-life treatment. It noted that the QALY increment associated with the most plausible cost-effectiveness estimates was approximately 1.24 QALYs. The Committee considered that the magnitude of the additional weight that would need to be assigned to the original QALY benefit for the cost effectiveness of lenalidomide to fall within the currently applied ICER threshold range was acceptable. Additionally, the Committee noted from the model that the extension to life with lenalidomide was approximately 1.81 years. The Committee considered that the impact of giving greater weight to QALYs achieved in the later stages of terminal diseases, using the assumption that the extended survival period is experienced at the full quality of life anticipated for a healthy person of the same age, was acceptable.

4.20 In summary, the Committee accepted that, for people with multiple myeloma who had received two or more prior therapies, the most plausible ICERs were those suggested by the ERG on the basis of exploration of the manufacturer's model and with the implementation of the patient access scheme (where the manufacturer would bear the costs of lenalidomide beyond 26 cycles [normally 2 years] for people whose disease had not progressed at this time). For the purpose of the recommendations, the patient access scheme would be triggered by the completion of 26 cycles (which normally takes 2 years), regardless of treatment interruptions and dose reductions within those cycles. The Committee accepted that the benefits provided by lenalidomide fitted the criteria for consideration for appraising a life-extending, end-of-life treatment. The Committee concluded that the additional weights that need to be attached to the QALYs to achieve ICERs within the normal threshold range are acceptable in these circumstances. Consequently the Committee recommended lenalidomide, within its licensed indication, as an option for the treatment of multiple myeloma in people who have received two or more prior therapies. The Committee noted that some people who have not received two or more prior therapies may be currently receiving lenalidomide for the treatment of multiple myeloma, and recommended that these people should have the option to continue treatment until they and their clinicians consider it appropriate to stop.