4 Consideration of the evidence

4.1 The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of topotecan, having considered evidence on the nature of recurrent and stage IVB cervical cancer and the value placed on the benefits of topotecan by women with the condition, those who represent them, and clinical specialists. It also took into account the effective use of NHS resources.

Clinical effectiveness

4.2 The Committee considered current clinical practice for treating recurrent and stage IVB cervical cancer. The Committee heard from clinical specialists that there is currently no nationally agreed standard treatment for women with this condition. Treatment may consist of topotecan plus cisplatin, cisplatin alone, or paclitaxel plus either carboplatin or cisplatin. The clinical specialists considered that combination therapies were generally more effective than single-agent therapies. They also stated that the main reason for there being no single established treatment regimen is because clinical trials in the past had not shown clinically significant advantages in terms of response rates or overall survival for any single regimen in this patient group. The Committee heard from patient experts that they considered it was important to have a number of treatment options because one may be more suitable than others for the individual patient. For example, the choice of treatment may be influenced by comorbidities such as renal dysfunction. Patient experts also highlighted that for some women topotecan plus cisplatin may be considered a final treatment option.

4.3 The Committee specifically considered the use of cisplatin to treat recurrent and stage IVB cervical cancer. It heard from clinical specialists that approximately 90–95% of women within the licensed population will have previously received cisplatin because it is standard UK clinical practice to use cisplatin either with radiotherapy or as chemotherapy alone as first-line treatment for cervical cancer. It heard how cervical screening in the UK enables early identification of disease and so initial presentation with stage IVB disease is unusual. The Committee also heard that the dose of cisplatin used in chemotherapy and in chemoradiotherapy is the same.

4.4 The Committee heard from clinical specialists that previous cisplatin use has a significant effect on response rates to subsequent cisplatin-containing chemotherapy regimens. In cisplatin-naive women, response rates to cisplatin were approximately 45%, which could be higher if combination therapy was used. However, for women who had previously received cisplatin, response rates could be as low as 10%. In addition, the response rates were found to increase as the duration of remission after initial cisplatin treatment increased.

4.5 The Committee discussed the clinical effectiveness of topotecan plus cisplatin compared with cisplatin alone presented in the main trial. The Committee considered that combination therapy was shown to be more effective than cisplatin alone in the GOG-0179 trial population. The Committee noted the results from the subgroup analyses suggesting that topotecan plus cisplatin was more clinically effective in women who were cisplatin naive than in women who had previously received cisplatin. The Committee considered that the reduced response to topotecan plus cisplatin was evident even when the sustained cisplatin-free interval was longer than 180 days.

4.6 The Committee examined the trial comparing four combination treatments (GOG-0204), including topotecan plus cisplatin. The Committee was aware that the trial had closed early because none of the other treatment combinations were likely to show a significant benefit over paclitaxel plus cisplatin. The Committee noted that hazard ratios from this trial suggested that paclitaxel plus cisplatin was more effective than the other cisplatin combination therapies, but this difference did not reach statistical significance. The Committee heard from clinical specialists that they did not consider there to be any differences in effectiveness between the different combinations that had been used in this trial. The Committee noted that in this trial approximately 70% of women had received cisplatin as prior chemoradiotherapy. The Committee understood that no specific data for cisplatin-naive women from this trial had been provided. It noted that the manufacturer had also included an indirect comparison of topotecan plus cisplatin with paclitaxel plus cisplatin. The Committee recognised that this comparison suggested that topotecan plus cisplatin was more effective than paclitaxel plus cisplatin, but again the difference did not reach statistical significance. The Committee concluded that there was uncertainty about the differences in effectiveness among combination chemotherapy regimens.

4.7 When considering the comparative evidence the Committee was aware that there were differences in the trial populations. The Committee considered that the trial of combination therapies (GOG-0204) appropriately reflected the majority of the clinical population in England and Wales, where women often received chemoradiotherapy that included cisplatin. However, the Committee noted that for the subgroup of women who were cisplatin naive, the trial of combination therapies was not representative of this population. The Committee concluded that there was additional uncertainty about the efficacy of topotecan in comparison with paclitaxel and other combination regimens for this subgroup.

4.8 The Committee discussed the manufacturer's network meta-analysis that was provided after consultation on the appraisal consultation document. The Committee noted that the meta-analysis combined direct and indirect evidence that, when considered individually, did not show consistent effects. In addition, there were differences in the trial populations in terms of prior cisplatin exposure, performance status and disease stage. The Committee heard from the ERG that they did not consider that the data from the trials were exchangeable, and therefore it was inappropriate to carry out a meta-analysis of the data. The Committee also heard from the ERG that the manufacturer's analyses that pooled the data for paclitaxel suggested an estimate of effect similar for both the licensed population and the cisplatin-naive population that was not consistent with the clinical trial or clinical specialists' evidence or biological plausibility. The Committee concluded that in principle a network meta-analysis was an appropriate method of calculating efficacy, but the nature of the evidence available in this situation meant that it could not be considered appropriate as a basis on which to make a decision about the cost effectiveness of topotecan.

4.9 The Committee considered the adverse event profile of topotecan and recognised that women receiving topotecan plus cisplatin may have more adverse events compared with those receiving cisplatin alone. The Committee heard from clinical specialists specifically about neutropenia and febrile neutropenia. It heard how febrile neutropenia may lead to hospital admission, and may be a more frequent occurrence than for other regimens such as paclitaxel plus cisplatin. However, patient experts mentioned that they considered the safety profile of topotecan plus cisplatin to be manageable, although they were concerned about reported deaths following chemotherapy. They also indicated that quality of life may not be worse for women receiving combination therapy than for women receiving monotherapy, although they were specifically concerned about fatigue.

Cost effectiveness

4.10 The Committee considered the evidence on the cost effectiveness of topotecan plus cisplatin presented in the manufacturer's submission. The Committee recognised that the manufacturer considered their main analysis to be the within-trial comparison and not the model-based comparison. The Committee noted that the ERG could not completely validate the within-trial comparison and that they considered the manufacturer's model-based comparison to have greater external validity. The ERG had therefore used the model-based comparison as the basis for their exploratory analyses.

4.11 The Committee noted that in the within-trial comparison the base-case ICER provided by the manufacturer for topotecan plus cisplatin compared with cisplatin alone was £18,000 per QALY gained in the licensed population, £11,000 per QALY gained in the cisplatin-naive population and £32,500 per QALY gained in the sustained cisplatin-free interval population. The Committee noted that the results of the model-based comparison using the hazard ratio derived from the indirect comparison suggested that topotecan plus cisplatin had greater efficacy and lower costs than paclitaxel plus cisplatin. However, when the hazard ratio from the trial comparing different combination therapies directly was used, paclitaxel plus cisplatin was more effective and less costly than topotecan plus cisplatin.

4.12 The Committee considered the utility estimates provided by the manufacturer. The Committee heard from the ERG that the manufacturer had incorrectly mapped disease-specific quality of life to utility and that correcting this led to a lower starting utility of 0.72 instead of 0.79. The ERG also expressed concerns about the mapping equation used in the base-case analysis, including the transparency of the analysis and imputation methods. The ERG suggested that combining their amended starting utility from the main clinical trial with data identified by the manufacturer from a study of metastatic breast cancer could be more appropriate. The Committee recognised that neither approach to estimating utilities reflected the reference case and considered that both sets of utility estimates were associated with uncertainty. However, on balance the Committee considered that utility values suggested by the ERG, which led to more favourable ICERs, may be more appropriate than those provided by the manufacturer.

4.13 The Committee considered the manufacturer's assumptions about the number of topotecan vials required in clinical practice and the administration costs. The Committee heard that the manufacturer may have underestimated the administration costs for topotecan on days 2 and 3 because they had used an assessment report from a previous appraisal that had built up the costs without the appropriate health resource group codes being available. The ERG stated that these codes were now available and that the cost for administering the second and third infusion of topotecan would be £195 rather than £51. The Committee considered that the revised administration costs proposed by the ERG were appropriate. The Committee also heard from the ERG that the manufacturer had not assumed minimum or maximum wastage of excess topotecan, but used a midpoint. The ERG considered that an assumption of maximum wastage may be more consistent with the SPC. The Committee heard from clinical specialists that although women were grouped so that drug wastage could be reduced, there was less opportunity to group women receiving topotecan because of the small number of women who receive the drug. The Committee considered that although there was uncertainty about the manufacturer's estimate of topotecan wastage, assumptions of either minimum or maximum wastage may also not be accurate.

4.14 The Committee noted that there appeared to be inconsistencies between the mean cost estimates in the within-trial comparison and the model-based comparison. The Committee heard from the ERG that this may be because dose reduction related to adverse events was included in the within-trial comparison but not in the model-based comparison. However, without a breakdown of costs, the ERG was unable to confirm this. In addition, there were no data on how the use of paclitaxel may be affected by dose reduction. The Committee considered that dose reduction could be important and the ERG's exploratory analyses showed that dose reduction could lower the ICER. However, the Committee was not persuaded that the cost estimates including dose reduction were sufficiently robust for these to form the basis of their examination of the cost effectiveness of topotecan.

4.15 The Committee considered how adverse events had been included in the model-based comparison. The Committee heard from clinical specialists that the manufacturer's assumption that an adverse event lasted only a week was appropriate. The Committee noted that only the most severe adverse event was taken into account in the manufacturer's analysis even if two or more adverse events were experienced concurrently. The clinical specialists agreed that there may be a further negative impact on quality of life for women who have two concurrent adverse events. Overall the Committee considered that the manufacturer may have underestimated the reduction in quality of life associated with multiple adverse events.

4.16 The Committee first considered the fully incremental exploratory analysis for the licensed population undertaken by the ERG that incorporated amended administration costs and utility values. The Committee noted that, for the licensed population, using the hazard ratio from the indirect comparison, the ICER for topotecan plus cisplatin compared with cisplatin alone was £59,000 per QALY gained when minimum wastage was assumed, and the ICER for topotecan plus cisplatin compared with paclitaxel plus cisplatin was £117,000 per QALY gained when maximum wastage was assumed. The Committee was also aware that when the hazard ratio derived from the trial of different combination therapies was used, topotecan plus cisplatin was dominated by paclitaxel plus cisplatin. The Committee considered that the trial of different combination therapies (GOG-0204) was more representative of the patient population in England and Wales than the other available evidence. The Committee therefore concluded that for the licensed population, the cost-effectiveness data suggested that topotecan in combination with cisplatin was not a cost-effective use of NHS resources.

4.17 The Committee considered 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.

4.18 The Committee considered that the life expectancy for women with recurrent and stage IVB cervical cancer would normally be less than 24 months. The Committee noted that topotecan was licensed for multiple indications, but it could still be considered to be indicated for a small patient population. The Committee discussed the additional benefits provided by topotecan in comparison with other therapies available on the NHS. It noted the clinical trial results from GOG-0179 and agreed that for the licensed population topotecan plus cisplatin had demonstrated a gain in life expectancy of more than 3 months in comparison with cisplatin alone. However, the Committee was aware that the majority of women receive combination therapies in the NHS. The Committee noted the results of the trial of combination therapies (GOG-0204), which had closed early because none of the treatments including topotecan were likely to show a significant benefit over paclitaxel plus cisplatin. The Committee also noted the clinical specialists' comments that they considered there to be equal efficacy among the different combination treatments in GOG-0204. On balance, the Committee considered that, for the licensed population, topotecan plus cisplatin compared with other combination therapies currently available in the NHS had not shown an additional benefit of 3 months. The Committee therefore concluded that topotecan plus cisplatin did not fulfil the criteria for consideration of NICE's supplementary advice on end of life and agreed that topotecan in combination with cisplatin could not be recommended as a cost-effective use of NHS resources.

4.19 The Committee then considered the subgroup of women who had not previously received cisplatin. The Committee noted that the manufacturer's estimates suggested topotecan plus cisplatin may be cost effective in this group. The Committee considered the ERG's exploratory analyses for this subgroup. When the hazard ratios from the indirect comparison were used the ICER for topotecan plus cisplatin compared with cisplatin alone was £26,800 per QALY gained assuming minimum wastage, and £34,000 per QALY gained assuming maximum wastage. The Committee recognised that this did not include dose reduction, which the ERG had suggested could further reduce these ICERs. The Committee was aware that when the hazard ratios from the trial of different combination therapies were used, topotecan plus cisplatin was dominated by paclitaxel plus cisplatin, but that this evidence was predominantly from a population who had received cisplatin as a radiosensitiser before. Because the indirect comparison was the only data available in which the majority of women were cisplatin naive, the Committee was persuaded that topotecan plus cisplatin could be considered an appropriate use of NHS resources for the treatment of women who have not previously received cisplatin.

4.20 In light of the duty to have due regard to the need to eliminate unlawful discrimination and promote equality, the Committee discussed the higher prevalence of cervical cancer among women living in the most socioeconomically deprived areas, as outlined by the patient expert statements. It also discussed comments received during consultation on the appraisal consultation document. The Committee noted that a negative recommendation for topotecan in combination with cisplatin for the group of women with prior exposure to cisplatin does not impact particularly on any group protected by the equalities legislation. In addition, given the uncertainty about whether topotecan in combination with cisplatin is more clinically effective than other combination therapies for the treatment of cervical cancer in women with prior exposure to cisplatin, and the availability of alternative treatment options, the Committee was satisfied that its recommendation was consistent with NICE's obligations under the equalities legislation and the requirement for fairness.

  • National Institute for Health and Care Excellence (NICE)