Vinflunine for the treatment of advanced or metastatic transitional cell carcinoma of the urothelial tract

Clinical effectiveness

4.3 The Committee considered the clinical evidence on the use of vinflunine for the second-line chemotherapy of patients with advanced or metastatic transitional cell carcinoma of the urothelial tract. It noted there was only 1 randomised clinical trial (study 302), and that this compared vinflunine with best supportive care alone. The Committee was aware that best supportive care was the only comparator listed in the scope for the appraisal. It was also aware that there are no proven standard agents for second-line chemotherapy (see section 4.2). For these reasons, the Committee concluded that best supportive care was the appropriate comparator for vinflunine.

4.4 The Committee discussed whether the population in study 302 was representative of patients with advanced or metastatic transitional cell carcinoma of the urothelial tract who would receive vinflunine in UK clinical practice. It heard from the clinical specialists that the study population was younger, fitter and had better renal function than the general population of UK patients with advanced or metastatic transitional cell carcinoma of the urothelial tract. The Committee was also aware that neoadjuvant chemotherapy, adjuvant chemotherapy and concurrent chemotherapy and radiotherapy are all used as part of radical treatment for localised muscle-invasive transitional cell carcinoma of the urothelial tract. The Committee noted that patients treated in this way had been excluded from study 302. The Committee heard from the clinical specialists that many patients in the UK who are eligible to receive second-line palliative chemotherapy will already have received 2 lines of treatment (that is, neoadjuvant chemotherapy, adjuvant chemotherapy or concurrent chemotherapy and radiotherapy, plus first-line palliative chemotherapy). Finally, the Committee noted that the manufacturer considered the trial population to be only people with a poor prognosis. The Committee understood that this was because 74% of people in the trial had visceral involvement, and because it was unlikely that people with a better prognosis would be willing to be randomised to a trial in which one of the treatment options was best supportive care. The Committee noted that the marketing authorisation for vinflunine is for all patients with advanced or metastatic transitional cell carcinoma of the urothelial tract after failure of a prior platinum-containing regimen. The manufacturer was invited to submit further evidence for vinflunine in the whole licensed patient population compared with best supportive care, but no data were submitted for the Committee to consider. The Committee considered whether the evidence from study 302 might be generalisable to the full licensed population. It was mindful of clinical specialists' comments regarding the differences in the characteristics and treatment pathway of patients in the trial compared with patients in UK clinical practice. The Committee was not persuaded that the evidence for the effectiveness of vinflunine would be generalisable to the whole population who might receive vinflunine in UK clinical practice, compared with best supportive care. However the Committee was aware that study 302 was the only available evidence on which a decision could be based. The Committee concluded that the nature and availability of the evidence base would result in significant uncertainty regarding the effectiveness of vinflunine for the whole licensed population compared with best supportive care.

4.5 The Committee discussed the results of study 302. It noted that vinflunine was associated with improved progression-free survival and a higher disease control rate (defined as the percentage of patients with a complete response, a partial response or stable disease) compared with best supportive care alone. The Committee also noted that the difference in overall survival between the study arms was not statistically significant for the ITT population, but was significant for the eligible ITT population. The Committee was aware that the difference between the 2 analyses resulted from the exclusion of 13 patients from the ITT analysis because they had not been shown to have progressive disease after prior therapy. A greater proportion of ineligible patients came from the best supportive care arm than from the vinflunine arm (8% versus 2%) and this lowered the overall survival in the best supportive care arm in the eligible ITT analysis. The Committee considered that the results from the ITT population were the most appropriate basis for its deliberations because randomisation had not been broken. It also noted that there were no significant differences in health-related quality of life between patients receiving vinflunine and those receiving best supportive care alone. The Committee concluded that the extent of the clinical effectiveness of vinflunine compared with best supportive care had not been conclusively demonstrated because of the uncertainty in the overall survival results.

4.6 The Committee discussed the most common adverse events associated with vinflunine, namely constipation, anaemia, stomatitis and infusion-site reactions. It noted that grade 3 or 4 constipation occurred in 16% of patients receiving vinflunine. It was aware that grade 4 constipation can lead to intestinal obstruction or acute abdominal distension requiring hospitalisation. The Committee also noted the 6% incidence of febrile neutropenia in the vinflunine arm of the study. The clinical specialists stated that the safety profile of second-line chemotherapy in this setting needed to be predictable, acceptable to patients and manageable, and that they had concerns about vinflunine in this regard. The Committee concluded that there were concerns about the tolerability of vinflunine.

Cost effectiveness

4.7 The Committee reviewed the economic model submitted by the manufacturer and the ERG's critique of the model. The Committee was aware that the costs for the intravenous administration of vinflunine included in the manufacturer's model were based on out-of-date NHS HRG figures which were lower than current estimates. The Committee considered the manufacturer's lack of inclusion of vial wastage in the model to be inappropriate because of the small number of patients who would be treated with vinflunine at any one centre and time. The Committee concluded that the costs of treatment with vinflunine had been underestimated in the manufacturer's model.

4.8 The Committee discussed the utility values used in the manufacturer's model. It noted that different methods of estimating utilities were used for the post-progression and pre-progression health states. It noted that the utility for the post-progression health state was taken from a study of patients with lung cancer. The pre-progression utility was based on answers to 1 of the 30 questions in the EORTC questionnaire, which asked patients to rate their overall quality of life during the past week. The questionnaire was administered at the end of each treatment cycle. The Committee noted clinical specialist opinion that quality of life varies considerably between 2 consecutive clinic visits. It therefore considered that this question may have to be interpreted with caution because a patient's quality of life in the last week of a treatment cycle may not reflect their quality of life for the whole period before disease progression. It also noted that established algorithms for mapping EORTC responses to EQ-5D exist but were not used by the manufacturer. The Committee noted that neither utility used in the economic model conformed to the preferred NICE reference case and concluded that the lack of appropriate utility data contributed to uncertainty in the model.

4.9 The Committee discussed the data on clinical effectiveness used in the model. It was aware that various hazard ratios of overall survival had been reported depending on the population analysed and the type of analysis used (multivariate analysis or extended multivariate analysis). The Committee noted that the modelled hazard ratios were based on the multivariate analysis of the results for the eligible ITT population and that these results were more favourable for vinflunine than those obtained from the ITT population. The Committee had previously concluded (see section 4.5) that the results from the ITT population were the most appropriate for this appraisal. It therefore concluded that the survival benefit of vinflunine compared with best supportive care alone was likely to be overestimated in the manufacturer's model.

4.10 The Committee discussed the inclusion of adverse events in the model and noted that although the costs of adverse events were included, the disutility associated with them was not. It discussed the costs for grade 3 and 4 constipation, and considered that these were likely to be significantly higher than the cost for constipation used in the model (£39).

4.11 The Committee discussed the manufacturer's base-case ICER of £100,100 per QALY gained (incremental cost of £13,100 and incremental QALYs of 0.131). It noted that in the manufacturer's sensitivity analyses the inclusion of vial wastage and the use of a lower pre-progression utility value increased the ICER significantly from the base case (to £121,100 and £133,100 per QALY gained respectively). It also noted that in the ERG's exploratory analysis, based on Kaplan–Meier estimates of survival from the ITT population rather than the eligible ITT population, the ICER was £126,400 per QALY gained. The Committee considered the most plausible ICER to be above £120,000 per QALY gained. It further considered that additional uncertainties around the costs of adverse events and the modelling of survival data would increase the ICER.

4.12 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:

4.13 The Committee considered that the life expectancy of patients with advanced or metastatic transitional cell carcinoma of the urothelial tract whose disease has progressed after first-line chemotherapy is usually less than 6 months. It discussed the number of UK patients for whom vinflunine is licensed, estimated by the manufacturer to be about 800–1500, and concluded that this could be considered a small patient population. The Committee discussed the extension to life offered by vinflunine in the study populations. In the manufacturer's model, the overall survival benefit of vinflunine was 3.2 months. However, the overall survival benefit based on the trial results was 2.3 months in the ITT population (not statistically significant) and 2.6 months in the eligible ITT population. The Committee was not persuaded that an extension to life of at least 3 months had been proven, and therefore concluded that the end-of-life advice did not apply to this appraisal. The Committee further noted that even if the end-of-life considerations were taken into account, the most plausible ICER for vinflunine compared with best supportive care (above £120,000 per QALY gained) was substantially higher than would normally be considered cost effective. The Committee was mindful of the limitations of the evidence base for vinflunine in the whole licensed population of patients who may receive treatment with vinflunine in UK clinical practice compared with best supportive care. However, the Committee was conscious of uncertainty in the overall survival results for vinflunine from the available evidence (see section 4.5), and the exceptionally high ICER which was based on this evidence. On balance, the Committee did not consider it plausible that additional evidence, even if available, would demonstrate the magnitude of survival gain which would be required to bring the cost effectiveness of vinflunine to a level considered an acceptable use of NHS resources. Therefore the Committee concluded that vinflunine could not be considered a cost-effective use of NHS resources for the treatment of advanced or metastatic transitional cell carcinoma of the urothelial tract that has progressed after treatment with platinum-based chemotherapy.

Summary of Appraisal Committee's key conclusions