Olaparib for maintenance treatment of relapsed, platinum-sensitive ovarian, fallopian tube or peritoneal cancer after 2 or more courses of platinum-based chemotherapy

3.1 The patient and clinical experts explained that relapsed, platinum-sensitive ovarian, fallopian tube and peritoneal cancer (from here ovarian cancer refers to ovarian, fallopian and peritoneal cancer) are devastating conditions. After relapse, the goal of treatment is to manage rather than cure the condition. Maintenance treatment helps to extend the time before progression. This extends the time between courses of chemotherapy, which 1 patient expert described as gruelling. The patient experts said that the diagnosis of the first relapse can be more devastating than the initial diagnosis of ovarian cancer. They said that it is a huge emotional burden knowing that the cancer is likely to keep coming back and that the treatment outcomes are worse for each successive relapse. The clinical experts explained that survival rates and outcomes for ovarian cancer are worse in the UK than in other developed countries. The committee understood these factors and concluded that there is a high disease burden for people with relapsed, platinum-sensitive ovarian cancer.

3.2 The clinical experts explained that using poly-ADP-ribose polymerase (PARP) inhibitors, such as olaparib, as maintenance treatment after platinum-based chemotherapy has become standard care in the NHS. But not all PARP inhibitors, in all lines of treatment, are available for routine commissioning. Some are only available through the Cancer Drugs Fund. The specific PARP inhibitor available depends on how many courses of chemotherapy the person has had before. Also, they are only available for people who have not had treatment with a PARP inhibitor before. The clinical and patient experts explained that olaparib has manageable side effects, can be taken orally at home and is effective at delaying disease progression. They noted that people may have a psychological benefit from taking a maintenance treatment. One of the patient experts said that olaparib "massively improves quality of life" and that they can "live an amazing life" as a result of the treatment. The clinical experts said that the life expectancy for people with relapsed ovarian cancer has dramatically improved since PARP inhibitors became widely available. The committee concluded that the continued availability of olaparib, to extend periods of remission and improve quality of life, would be greatly valued by people with the condition and their families.

3.3 The committee noted that the clinical pathway for relapsed ovarian cancer has changed significantly since olaparib entered the Cancer Drugs Fund. After their first course of chemotherapy, people will usually have a PARP inhibitor through the Cancer Drugs Fund (see NICE's technology appraisal guidance on niraparib, olaparib and olaparib plus bevacizumab). In the NHS, people are only offered a PARP inhibitor if they have not had one previously. But there is a need for PARP inhibitors to be available after later courses of chemotherapy. For example, the clinical expert explained that people with stage 1 or stage 2 ovarian cancer cannot have a PARP inhibitor after the first course of chemotherapy. If their cancer relapses, they can have a PARP inhibitor after the second course of chemotherapy. Also, for some people, PARP inhibitors may not have been available on the NHS after their previous course of chemotherapy. But over time, this population is expected to reduce as more people have PARP inhibitors after the first course of chemotherapy. One of the patient experts explained that they started treatment with olaparib after fourth-line chemotherapy and that it had significantly extended their life. So, they felt that PARP inhibitor maintenance treatment should be available whenever it is needed, regardless of the number of previous courses of chemotherapy. The committee concluded that despite the small number of people who are currently eligible for olaparib after their second course of chemotherapy, it remains a much-valued treatment option for those who need it.

3.4 The committee recalled that when olaparib entered the Cancer Drugs Fund, its preferred source of overall survival data for olaparib (from the SOLO2 clinical trial, a randomised, double-blind study comparing olaparib with placebo after platinum-based chemotherapy in people with a BRCA mutation) was not yet mature. So, olaparib's entry into the Cancer Drugs Fund was based on data from Study 19, but the committee felt that this data should be considered with caution. This was because most people in Study 19 with a BRCA mutation had had 3 or more courses of platinum-based chemotherapy before olaparib. This is more than would be expected in clinical practice. Also, some people in the study had a BRCA mutation and some did not, and their mutation status was determined retrospectively. The committee noted that the overall survival data from SOLO2 is now mature, and it is more relevant to the population being considered in this appraisal. The committee reviewed the baseline characteristics of the SOLO2 trial population. It noted that the baseline performance status was potentially slightly better than might be expected in the NHS. The committee also noted that more people's cancer had a complete response to their previous treatment than might be expected (that is, there were no signs of cancer on their scans or tests). But it concluded that the population was broadly generalisable to the NHS population in England.

3.5 People in the SOLO2 trial could not switch treatment from placebo to olaparib. But, people in both groups could have a PARP inhibitor after disease progression as part of clinical practice. A substantial proportion of people in the placebo arm had a subsequent PARP inhibitor. The exact percentage is considered confidential by the company and cannot be reported here. In the company's original base case, it adjusted the placebo arm data to remove the benefit of subsequent PARP inhibitor use. This was to better reflect current NHS practice, in which very small numbers of people are now eligible for a PARP inhibitor after their third or later course of platinum-based chemotherapy. The NHS England Cancer Drugs Fund clinical lead (from here, Cancer Drugs Fund lead) noted that under the terms of reference for this Cancer Drugs Fund review, the cost-effectiveness analysis should be based on the clinical pathway at the point of Cancer Drugs Fund entry, not current NHS practice. So, it was not appropriate to assume zero use of subsequent PARP inhibitors in the placebo arm. The clinical experts said that if someone had not had a PARP inhibitor previously and their cancer responded to the third course of chemotherapy, they would be offered one. The committee concluded at the first committee meeting that the high levels of subsequent PARP inhibitor use in SOLO2 may be more reflective of NHS practice at the point of Cancer Drugs Fund entry than the adjusted data. So, it asked the company and ERG to use the unadjusted SOLO2 data for the placebo arm in their updated base cases. The ERG noted that there were some limitations with this approach, because it was necessary to make additional assumptions about the data. People in SOLO2 had access to PARP inhibitors, some of which are not routinely available on the NHS at that line of treatment. They may also have had retreatment with a PARP inhibitor after multiple courses of chemotherapy (which does not reflect NHS practice). So, it was necessary to assume that all subsequent PARP inhibitor use in SOLO2 was olaparib and that it was taken after the third course of chemotherapy. The Cancer Drugs Fund lead said that the PARP inhibitor treatments have the same mode of action, so they could be expected to have similar efficacy and tolerability to olaparib. Noting the limitations of the unadjusted data outlined by the company and the ERG, the committee confirmed that it preferred the unadjusted overall survival data from SOLO2 for the placebo arm.

3.6 Some people in the olaparib arm had a subsequent PARP inhibitor. The exact percentages are considered confidential by the company and cannot be reported here. The company did a scenario analysis showing that retreatment in the olaparib arm had a limited effect on overall survival. The ERG confirmed that this analysis was appropriate. The committee agreed that the olaparib arm did not need to be adjusted to remove any benefit of future treatment with a PARP inhibitor.

3.7 The unadjusted data from SOLO2 shows that olaparib significantly delays disease progression after the second course of chemotherapy. It also improves median overall survival. The clinical expert noted that about 20% of people have a very good response to treatment. So, in addition to the median overall survival, it is also important to consider the tail of the curve and the hazard ratios, which better reflect this large benefit in a proportion of people. The exact data for progression-free survival and overall survival are considered confidential by the company and cannot be reported here. The committee concluded that olaparib extends progression-free survival and overall survival compared with placebo.

3.8 The company model used the same structure as in TA620 (see the committee discussion for TA620 in the evidence section for this appraisal on the NICE website), which was a 3‑state (progression-free, progressed disease and death) partitioned survival model. The inputs were updated in line with the terms of engagement of Cancer Drugs Fund entry. So, the Study 19 data on overall survival, progression-free survival, time to stopping treatment, subsequent treatments and baseline characteristics were replaced with data from SOLO2. Also, the time horizon changed from 30 years to 50 years, in line with the committee's preference. The ERG agreed that the changes to the model inputs were appropriate and aligned with the terms of engagement for Cancer Drugs Fund entry.

3.9 The updated company base case used the unadjusted overall survival data for the routine surveillance arm (see section 3.5), with a lognormal curve extrapolation curve fitted. The lognormal curve was selected based on statistical goodness-of-fit, visual inspection and external clinical validation. The company considered other parametric models but ruled these out because the survival estimates were not consistent with clinical opinion. For example, the percentage of people still alive at 20 years was considered too pessimistic. The company noted that estimates based on the lognormal curve are conservative because they overestimate survival in the routine surveillance arm between years 2 and 3 compared with the observed data from SOLO2. The ERG confirmed that the lognormal curve was appropriate and aligned with clinical expert opinion. The company also presented a scenario that used survival data from Study 19 for the routine surveillance arm. In Study 19 there were lower levels of subsequent PARP inhibitor treatment than in SOLO2. The company said this made the cost-effectiveness estimate more generalisable to current NHS practice, in which very few people would have olaparib after their third course of chemotherapy. It also noted that unadjusted data from Study 19 was used in NICE's technology appraisal guidance on niraparib. But the ERG said that using the Study 19 data would introduce more uncertainty because the population is less relevant than the SOLO2 population. The committee recalled its conclusion from TA620 that because the Study 19 subgroup analysis of people with a BRCA mutation was retrospective, the results should be considered with caution (see section 3.4). It also noted the terms of reference for Cancer Drugs Fund entry, which specified that the company should update the overall survival estimate using SOLO2 trial data. The committee concluded that its preferred cost-effectiveness analysis was based on the extrapolated unadjusted data from SOLO2.

3.10 In the updated company base case, the committee requested that all subsequent PARP inhibitor use in SOLO2 was assumed to be olaparib (see section 3.5). The ERG identified that a small number of people who had taken PARP inhibitors had not been included in the company's updated analysis. To be consistent with the committee preferences, the ERG made a correction to the company base case to include these people. This made a modest improvement to the incremental cost-effectiveness ratio (ICER) in favour of olaparib. The committee agreed with the ERG that all subsequent PARP inhibitor use from SOLO2 should be included in the cost-effectiveness analysis and concluded that the ERG's correction to the company's base case was appropriate.

3.11 The updated company model used the unadjusted overall survival data from SOLO2 for the routine surveillance arm. This data included people who had a PARP inhibitor as subsequent maintenance treatment. So, the company model also included costs for subsequent maintenance treatment with a PARP inhibitor. The ERG confirmed that the company's approach was appropriate. The committee concluded that the costs of subsequent maintenance treatment with PARP inhibitors should be considered alongside the benefits.

3.12 NICE's guide to the methods of technology appraisal 2013 (section 6) notes that above a most plausible ICER of £20,000 per quality-adjusted life year (QALY) gained, judgements about the acceptability of a technology as an effective use of NHS resources will take into account the degree of certainty around the ICER. The committee will be more cautious about recommending a technology if it is less certain about the ICERs presented. The committee noted that the updated company and ERG base cases were broadly consistent with the committee's preferences. It preferred the ERG-corrected company base case (also referred to as the ERG base case) because it included all subsequent PARP inhibitor use (see section 3.10). The committee noted that the data used in the cost-effectiveness model was high quality and generalisable to the NHS at the time of Cancer Drugs Fund entry. It also considered the company's view that estimates based on the lognormal distribution likely represent the upper bound of the cost-effectiveness estimate (see section 3.9). For these reasons, the committee considered that the maximum acceptable ICER would be towards the higher end of the range normally considered a cost-effective use of NHS resources (£20,000 to £30,000 per QALY gained). Because of confidential commercial arrangements for subsequent treatments in the pathway, the ICERs are confidential and cannot be reported here. But, when all confidential discounts were taken into account, the ERG's corrected company base-case ICER was within the range considered cost-effective. So, the committee recommended olaparib after 2 courses of platinum-based chemotherapy for routine use in the NHS.

3.13 In TA620 the committee considered the advice about life-extending treatments for people with a short life expectancy in NICE's guide to the methods of technology appraisal 2013 (section 6). It agreed that the end of life criteria were not met for people who have olaparib after 2 courses of platinum-based chemotherapy.

3.14 The committee concluded that the ICER is within what NICE considers a cost-effective use of NHS resources for olaparib as a maintenance treatment for relapsed, platinum-sensitive, high-grade epithelial ovarian, fallopian tube, or primary peritoneal cancer in adults whose cancer has responded to 2 courses of platinum-based chemotherapy and who have a BRCA1 or BRCA2 mutation. So, olaparib is recommended for routine use in the NHS.