3 Committee discussion
The appraisal committee (section 6) considered evidence submitted by Clovis Oncology, a review of this submission by the evidence review group (ERG), and the technical report developed through engagement with stakeholders. See the committee papers for full details of the evidence.
The appraisal committee was aware that 1 issue was resolved during the technical engagement stage, and agreed that:
For the long-term extrapolation of progression-free survival for the subgroup of people without a BRCA mutation it is most plausible to use the lognormal distribution, because this is more aligned with time to treatment discontinuation.
For the subgroup of people with a BRCA mutation who have had 2 lines of platinum-based chemotherapy it is most plausible to use the lognormal distribution, because this is more aligned with what was seen in Study 19 at 6‑year follow up.
It recognised that there were remaining areas of uncertainty associated with the analyses presented (see technical report, table 2, page 34), and took these into account in its decision making. It discussed the following issues (issues 1, 2, 3, 4, 5, 7 and 8), which were outstanding after the technical engagement stage.
3.1 The patient experts explained in their written submissions that ovarian cancer negatively affects many aspects of life including physical and mental wellbeing, self-esteem and body image. The disease is often diagnosed after the cancer has spread beyond the ovary, making curative treatment difficult, and the diagnosis is often associated with devastation, shock, disbelief and fear. People with advanced disease are likely to face a future of recurrent disease, needing multiple rounds of treatment to manage it. Living under the shadow of the disease and not knowing when it will recur can significantly affect their quality of life. The committee understood these factors and concluded that there is a high disease burden for people with recurrent, platinum-sensitive disease.
3.2 The patient experts explained that the risk of developing resistance to platinum is high, and treatments for platinum-resistant disease are extremely limited. Recurrence and the development of resistance to platinum need to be delayed for as long as possible. Maintenance treatment with a poly‑ADP-ribose polymerase (PARP) inhibitor such as rucaparib may extend time between recurrences, allowing people more freedom to lead a normal life. For some people, PARP inhibitors can have a long-lasting effect. There are currently no PARP inhibitors routinely commissioned for maintenance treatment after response to second-line platinum-based chemotherapy, although niraparib is currently available through the Cancer Drugs Fund. Olaparib capsules are recommended for routine commissioning after third-line chemotherapy for people with a BRCA mutation. There is also an ongoing NICE appraisal of olaparib tablets for maintenance treatment in people with relapsed disease that has responded to platinum. The committee concluded that the availability of a PARP inhibitor after response to second-line platinum would be greatly valued by patients and their families.
3.3 ARIEL3 is a double-blind randomised controlled trial of rucaparib compared with placebo in people with platinum-sensitive, high-grade serous or endometroid ovarian, fallopian tube or primary peritoneal carcinoma who have had 2 or more platinum-based chemotherapy regimens with a complete or partial response to the last regimen. The company submitted results for the overall intention-to-treat (ITT) population and also a subgroup of people who have a BRCA mutation and have had 3 or more courses of platinum-based chemotherapy (the BRCA 3L+ population) for the comparison of rucaparib with olaparib. At clarification stage, the ERG requested additional subgroup analyses of people without a BRCA mutation, and people with a BRCA mutation who have had 2 courses of platinum-based chemotherapy (the BRCA 2L population). However, the company commented that these analyses are not robust because they are post hoc and based on small sample sizes. The clinical experts explained that PARP inhibitors have been shown to have greater benefit in the subgroup of people with a BRCA mutation, but there are some people without a BRCA mutation whose disease responds in a similar way. Some people without a BRCA mutation may gain long-term benefit from PARP inhibitors, as seen in a trial of olaparib (Study 19). The clinical experts therefore considered that the ITT population is the most relevant population for the decision problem. The committee concluded that the results for the subgroups were not robust, and that the ITT population is the most relevant for decision making.
3.4 The clinical experts explained that ARIEL3 is representative of patients treated in the UK. Importantly, inclusion in ARIEL3 was not restricted based on the extent of residual disease, unlike some previous studies of PARP inhibitors. The clinical experts also explained that the proportions of people whose disease had a partial or complete response to platinum were similar to what would be seen in clinical practice. The committee noted that the proportion of people with a BRCA mutation is higher in ARIEL3 than in clinical practice (35% compared with 20%). The clinical experts explained that although this is higher than would be expected in the UK population, it is closer to the UK percentage than some other studies of PARP inhibitors. For example, 50% of people in Study 19 had a BRCA mutation. The committee concluded that ARIEL3 is broadly generalisable to clinical practice in England.
3.5 The primary endpoint of ARIEL3 was progression-free survival (PFS). A statistically significant improvement in PFS was seen at data cut-off for the overall ITT population. Median PFS was 10.8 months in the rucaparib arm and 5.4 months in the placebo arm (hazard ratio 0.36, 95% confidence interval 0.30 to 0.45). The committee concluded that rucaparib improves PFS compared with placebo.
3.6 Overall survival (OS) was a secondary endpoint in ARIEL3. At data cut-off, 88% of patients were still alive. Median OS was not reached and no statistically significant difference between the treatment arms had been seen. The clinical experts explained that there is a high possibility of a class effect for PARP inhibitors, and they expect rucaparib to show a broadly similar improvement in OS to that shown by olaparib in Study 19. The committee concluded that rucaparib is expected to provide a similar survival benefit to other PARP inhibitors.
3.7 Because of its immaturity, the company did not use ARIEL3 OS data in the cost-effectiveness analysis. It used OS data from Study 19 to model the long-term outcomes of rucaparib compared with routine surveillance. The committee noted that there were several differences between ARIEL3 and Study 19 in terms of trial design and patient characteristics, which may have influenced the results. For example, BRCA mutation status, a known prognostic factor, was a stratification factor at randomisation in ARIEL3 but it was confirmed retrospectively in Study 19. Also, a lower proportion of people had a BRCA mutation in ARIEL3 (35% compared with 50% in Study 19). The clinical experts highlighted that a higher proportion of people in ARIEL3 had rucaparib after 2 lines of platinum-based chemotherapy rather than after 3 or more lines (63% compared with 43% in Study 19), and that earlier use of PARP inhibitors is associated with better outcomes. Conversely, the use of a PARP inhibitor in subsequent treatment lines in the placebo arm was substantially higher in ARIEL3 than in Study 19, which probably reduced the magnitude of the difference in OS between rucaparib and placebo in ARIEL3. The committee appreciated that the populations in the trials were not strictly comparable. However, it noted that Study 19 provides the most mature data available for a PARP inhibitor, with over 6 years of follow up. The committee concluded that Study 19 provides the best OS data currently available for a PARP inhibitor, and that it is reasonable to use this data for modelling in the absence of mature OS data from ARIEL3.
3.8 The company submitted a partitioned survival model with 3 states (progression-free, progressed disease and death) to estimate the cost effectiveness of rucaparib compared with routine surveillance. The committee considered that the model is suitable for decision making.
Post-progression survival is uncertain, but the ERG's modelling of post-progression survival is the most plausible
3.9 The way post-progression survival (PPS) is modelled is one of the key drivers of the model results. Time spent in the progression-free health state is informed by ARIEL3 data. However, to model PPS for rucaparib, the company used the difference between Study 19 PFS and OS outcomes, assuming that PPS outcomes for rucaparib are equivalent to those for olaparib in Study 19. The ERG considered that this method is unconventional because the calculation of PPS is disconnected from the PFS used elsewhere in the model, and results in OS benefits that are implausible. The ERG's preferred approach is to calculate PPS as the difference between OS in Study 19 and PFS in ARIEL3. However, the company considered the ERG's approach to be inappropriate because it leads to a higher rate of death after progression for rucaparib than for olaparib. This results in shorter PPS outcomes for rucaparib, because PFS in ARIEL3 is longer than in Study 19. The committee considered that the company's approach is optimistic because it combines a greater PFS benefit for rucaparib from ARIEL3 and all the PPS benefit of olaparib from Study 19, resulting in a higher OS with rucaparib than for olaparib. The committee questioned the plausibility of this and recalled its earlier conclusion that rucaparib is expected to provide a similar survival benefit to other PARP inhibitors (see section 3.6). Therefore, the committee concluded that the ERG's approach is preferable, although it acknowledged that there is uncertainty associated with the modelling that will not be fully resolved until more mature OS data from ARIEL3 are available.
3.10 At the appraisal committee meeting, the company's base-case incremental cost-effectiveness ratio (ICER) for rucaparib compared with routine surveillance in the ITT population was above the range that is normally considered a cost-effective use of NHS resources (that is, £20,000 to £30,000 per quality-adjusted life year [QALY] gained). None of the company's scenario analyses substantially changed the results. The committee concluded that it could not recommend rucaparib as a cost-effective use of NHS resources for people with relapsed platinum-sensitive epithelial ovarian, fallopian tube or primary peritoneal cancer. After consultation, the company agreed to provide rucaparib to the NHS with a higher discount. The company and the ERG presented revised base-case ICERs including the agreed discount for rucaparib. The estimated ICERs for the ITT population incorporating the updated patient access scheme range from £29,138 (company) to £32,455 (ERG) per QALY. Most of the difference is because of the alternative approaches used for modelling PPS (see section 3.9). The committee considered that the results are uncertain because of the immaturity of the OS data and concluded that this uncertainty could only be resolved with the availability of more mature data from ARIEL3. Therefore, the committee was not confident that rucaparib represents a cost-effective use of NHS resources and could not recommend it for routine use in the NHS. However, if the company projections for PFS and OS prove to be correct when further evidence becomes available, then rucaparib would have the potential to be cost effective.
Rucaparib has not been shown to be cost effective compared with olaparib in people with a BRCA mutation who have had 3 lines of platinum-based chemotherapy
3.11 Olaparib capsules are a treatment option for people with a BRCA mutation who have had 3 lines of platinum-based chemotherapy. The company presented a cost-effectiveness analysis for this subgroup in which it was assumed that rucaparib and olaparib have equivalent efficacy. The results of the analysis showed that rucaparib was dominated by olaparib (that is, rucaparib costs more and worked equally as well). The committee noted that there is currently no robust efficacy data to inform this comparison. Therefore, the cost effectiveness of rucaparib compared with olaparib in this subgroup of patients is uncertain. The committee concluded that rucaparib cannot be recommended for routine use in people with a BRCA mutation who have had 3 lines of platinum-based chemotherapy.
Rucaparib meets the criteria for inclusion in the Cancer Drugs Fund for the maintenance treatment of relapsed platinum-sensitive ovarian, fallopian tube or peritoneal cancer
3.12 Having concluded that rucaparib cannot be recommended for routine use, the committee considered if it could be recommended for use within the Cancer Drugs Fund. The committee discussed the arrangements for the Cancer Drugs Fund agreed by NICE and NHS England in 2016, noting NICE's Cancer Drugs Fund methods guide (addendum). The committee recognised that PARP inhibitors are innovative treatments for recurrent disease. The key uncertainty associated with rucaparib is the immature OS data, and this could be addressed through the collection of additional data from ARIEL3. The committee took the view that if mature OS data from ARIEL3 supports the survival estimate in the company's model, the ICERs for rucaparib could be within the range normally considered to be a cost-effective use of NHS resources (£20,000 to £30,000 per QALY gained). The committee therefore concluded that the criteria for inclusion in the Cancer Drugs Fund are met because:
rucaparib cannot be recommended for routine commissioning based on current clinical data but there is plausible potential for cost effectiveness
there is outstanding clinical uncertainty about the overall survival with rucaparib
the uncertainty is likely to be resolved by further data from ARIEL3.
Rucaparib is recommended for use within the Cancer Drugs Fund as an option for the maintenance treatment of relapsed platinum-sensitive ovarian, fallopian tube or peritoneal cancer
3.13 Rucaparib increases PFS compared with routine surveillance but the benefit in OS is uncertain because mature data are not available yet from the trial. Because of the uncertainty about the OS benefit, the estimates of cost effectiveness are very uncertain and rucaparib cannot be recommended for routine use in the NHS. If mature OS data from ARIEL3 support the survival estimates in the company's model, rucaparib has the potential to be cost effective. Therefore, rucaparib is recommended for use within the Cancer Drugs Fund as an option for the maintenance treatment of relapsed platinum-sensitive epithelial ovarian, fallopian tube or primary peritoneal cancer after response to platinum-based chemotherapy.