The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of pemetrexed having considered evidence on the nature of non-small-cell lung cancer and the value placed on the benefits of pemetrexed by clinical specialists. It also took into account the effective use of NHS resources.
The Committee considered current UK practice for the treatment of people with non-squamous non-small-cell lung cancer. The Committee heard from clinical specialists that patients undergo induction with a platinum doublet of carboplatin or cisplatin in combination with gemcitabine, paclitaxel, vinorelbine or docetaxel. The Committee was also aware of the recommendation in NICE's technology appraisal guidance on pemetrexed for the first-line treatment of non-small-cell lung cancer. After induction, patients are monitored but receive no chemotherapy until progression. Patients whose disease progresses only receive second-line chemotherapy if they have a good performance status. In the UK, this is normally docetaxel or erlotinib. Patients who do not receive second-line chemotherapy receive best supportive care, which can include palliative radiotherapy.
The Committee heard that maintenance treatment after first-line treatment is a new concept in lung cancer and is not currently practised in the UK. The Committee also heard from clinical specialists that pemetrexed has fewer adverse events associated with its use compared with many other chemotherapies offered for the treatment of non-small-cell lung cancer. The aim of maintenance treatment with pemetrexed is to prolong the period of remission after first-line chemotherapy and possibly increase eligibility for second-line chemotherapy.
The Committee noted that the clinical effectiveness evidence for pemetrexed for the maintenance therapy of non-small-cell lung cancer was based on the JMEN trial, and noted that the overall survival achieved with pemetrexed was higher than for people receiving best supportive care. The Committee considered the trial to be generally well designed but had a number of concerns over the interpretation of the trial results (see section 4.5 to 10).
The initial primary endpoint was changed from overall survival to progression-free survival during the course of the trial. The Committee heard from the manufacturer that this was done after consultation with regulatory authorities in the USA and the change was implemented before any trial data had been analysed.
The evidence in the manufacturer's submission was from the non-squamous histological subgroup of the trial but histology was not a factor in the randomisation process. However, the Committee heard from the clinical specialists that the histological groups were reasonably balanced between the two arms of the trial and that lack of histological testing as part of the randomisation would not have significantly affected the outcome of the analysis. The clinical specialists also told the Committee that although definitive histological testing is variable in practice, the trial strategy was a reasonable reflection of what would be done in non-trial conditions.
In the trial treatment cycles were only limited by disease progression. The Committee heard from clinical specialists that patients would continue to receive pemetrexed while they were responding to treatment and so the trial did reflect the likely UK clinical practice. Therefore, the Committee was concerned that the capping in the manufacturer's original economic model was not consistent with clinical practice (see section 4.14).
None of the trial centres were located in the UK and one third of the trial population was Asian (from China, Korea, Taiwan and India). The Committee heard from the clinical specialists that although this ethnic group has a relatively favourable prognosis for non-small-cell lung cancer, it would have the same relative benefit from treatment with pemetrexed as the UK population.
There was an imbalance in the use of second-line treatments in the trial, and the Committee was concerned about how this was used in the manufacturer's economic model (see section 4.16).
The Committee was concerned that insufficient health-related quality-of-life data had been collected from the trial to enable their inclusion in the economic modelling. The Committee heard from clinical specialists and the manufacturer that trial patients who are in progressive disease after first- and second-line treatment are less likely to complete quality-of-life surveys, making it hard to get health-related quality-of-life data.
The Committee considered the population eligible for maintenance treatment. The Committee heard from clinical specialists that patients who receive first-line treatment usually have a good performance status, and that approximately one third of patients will progress while on first-line chemotherapy. The Committee also considered how patients are monitored in UK clinical practice. The Committee heard that although computer tomography (CT) scanning is not routinely used to monitor patients in UK clinical practice, it is likely that patients receiving pemetrexed maintenance treatment would undergo more CT scans to confirm that they have not progressed.
Cost effectiveness
The Committee considered the manufacturer's submitted cost-effectiveness analysis and the ERG's critique. The manufacturer's base case stated that the incremental cost of pemetrexed compared with best supportive care was £9,137 and the incremental QALY was 0.27, giving an ICER of pemetrexed compared to best supportive care of £33,732 per QALY gained. However, the Committee was aware of several concerns that the ERG had described in the calculation of this base case. These included: the modelling of overall survival, the capping of the number of treatment cycles but not the associated benefits, the different utilities assigned to patients in the same initial health state, the handling of second-line treatment effects and the absence of a probabilistic sensitivity analysis.
The Committee noted the 29-month overall survival data from the trial were extrapolated to 6 years in the model. It noted that the exponential curve applied in the base case did not fit the data well. The Committee noted that the manufacturer's analysis using a Weibull model was also plausible and that the ICERs were higher when Weibull models were used, suggesting that the figure in the base case might be at the lower end of the likely range. The Committee also expressed concern that the primary outcome in the trial (progression-free survival) was not captured in the model and number of cycles of treatment was used as a proxy.
The Committee considered the capping of costs in the manufacturer's original model at a maximum of 17 cycles. The Committee was informed by the manufacturer that clinical advice suggested that most benefit is derived in the first 8 to 10 cycles of treatment, which informed their decision to cap the cycles at 17 (1 standard deviation above the mean of 8 cycles). However, the Committee heard that in other cancers where patients receive maintenance treatment, cycles are not capped. The Committee considered that when capping was assumed, it had the effect of constraining the costs of maintenance therapy without a corresponding effect on the benefits accrued from use of pemetrexed, therefore building an essential bias in the economic evaluation in favour of pemetrexed.
The Committee considered the utility estimates assigned to patients in different arms of the model. It noted that in the manufacturer's original analysis, patients who entered the trial in the same health state were assigned higher utilities in the pemetrexed arm than in the placebo arm of the model – biasing the model in favour of pemetrexed. The Committee also noted that the disutilities of adverse events associated with pemetrexed were not modelled in the base case. Although the clinical specialists said that a minority of patients may feel better on pemetrexed maintenance treatment because their tumour shrank, the Committee was not persuaded that this justified the manufacturer's difference in utility between the two arms. The Committee considered the ERG's re-analysis of the model, which used a slightly lower utility for progression-free disease in the pemetrexed arm compared with the placebo arm, to be more appropriate. The Committee noted that this approach was adopted by the manufacturer in the revised analysis.
The Committee considered the six scenarios of the revised analysis presented by the manufacturer. The Committee also considered the ERG analysis of scenario 5, which corrected the utility estimates, removed cycle capping, performed an approximate probabilistic sensitivity analysis and also corrected discounting errors. The Committee considered scenario 5 to represent the most plausible assumptions for modelling the cost effectiveness of pemetrexed maintenance treatment compared with best supportive care. The Committee considered that the manufacturer's revised analysis had adequately addressed the main concerns identified in the original model. The Committee considered the updated ICERs presented for scenario 5 by the manufacturer (£47,000 per QALY gained) and the ERG (which ranged from £47,000 per QALY gained with the exponential model to £51,000 per QALY gained with the Weibull model) to be reliable.
The Committee considered the supplementary advice from NICE that should be taken into account when appraising treatments which may extend the life of patients with a short life expectancy and which 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:
The treatment is indicated for patients with a short life expectancy, normally less than 24 months.
There is sufficient evidence to indicate that the treatment offers an extension to life, normally of at least an additional 3 months, compared with current NHS treatment.
The treatment is licensed or otherwise indicated for small patient populations.
In addition, when taking these into account the Committee must be persuaded that the estimates of the extension to life are robust and the assumptions used in the reference case economic modelling are plausible, objective and robust.
The Committee discussed whether the benefit provided by pemetrexed for the maintenance treatment of non-small-cell lung cancer fulfilled the criteria for consideration as a life-extending, end-of-life treatment. The Committee understood that patients with stage IIIB or IV non-small-cell lung cancer who receive no treatment usually survive for about 7 to 10 months. The Committee considered the evidence from the pemetrexed randomised controlled trial (the JMEN trial) that showed a median survival benefit of 5.2 months for pemetrexed versus placebo. The Committee agreed that the data from the trial were sufficiently robust and that maintenance treatment with pemetrexed would increase overall survival by more than 3 months. The Committee considered that the estimated population for whom pemetrexed is licensed is currently small enough to allow the end-of-life advice to apply. The Committee concluded that the evidence submitted by the manufacturer was robust enough to show that maintenance treatment with pemetrexed fulfilled the criteria for the supplementary advice from NICE (see section 4.19).
The Committee considered the evidence presented by the manufacturer in the revised analysis to be robust. The Committee also considered the ERG's exploratory analysis, which demonstrated that the ICER for pemetrexed compared with best supportive care was about £47,000 per QALY gained. The Committee was persuaded that the most plausible ICER for pemetrexed compared with best supportive care was approximately £47,000 per QALY gained and, with reasonable certainty, was below £50,000 per QALY gained. The Committee considered this ICER, taking into account the end-of-life criteria. The Committee considered that the additional weight that would need to be assigned to the QALY benefits for the ICER to fall within the plausible range was acceptable. Therefore, the Committee recommended pemetrexed as an option for the maintenance treatment of people with locally advanced or metastatic non-small-cell lung cancer other than predominantly squamous cell histology, if disease has not progressed immediately following platinum-based chemotherapy in combination with gemcitabine, paclitaxel or docetaxel.