Idelalisib for treating chronic lymphocytic leukaemia

Economic model

4.8 The Committee considered the model presented by the company, the associated assumptions and the critique presented by the ERG. It noted that the company had submitted an economic model which addressed the population in Study 116 – that is, people with previously treated chronic lymphocytic leukaemia which is high‑risk relapsed or refractory. The Committee recalled the clinical specialists' advice that treatments suitable for high‑risk relapsed disease were: fludarabine, cyclophosphamide and rituximab; chlorambucil plus rituximab; or bendamustine plus rituximab. It was aware that there was no direct evidence for these comparators and an indirect comparison was not possible (see section 4.7). It further noted that the company had assumed that best supportive care had equal efficacy to rituximab alone, and heard from the clinical experts that this was a fair assumption. The Committee discussed the assumptions and concluded that the model was appropriate to inform its decision‑making for people with high‑risk relapsed and refractory disease. The Committee further noted that the company had not submitted any cost‑effectiveness evidence for people with lower risk relapsed disease (that is, disease which has relapsed more than 24 months since treatment). The Committee therefore concluded that it could not make any recommendations for people who were outside the scope of Study 116.

4.9 The Committee considered the company's rationale for not submitting an economic model for people with untreated chronic lymphocytic leukaemia and a 17p deletion or TP53 mutation. It heard from the company that due to the limited evidence base it was not possible to conduct an economic evaluation using the results of Study 101‑08 (see section 3.22). The Committee discussed the limitations of the available evidence base for the untreated population and concluded that an economic model could have been presented by the company. However, it noted that a proportion of patients in Study 116 (43.2%) had a TP53 mutation or 17p deletion (see section 4.5). It further noted the clinical experts' advice that in people with a 17p deletion or TP53 mutation, treatment with chemotherapy is ineffective and potentially harmful (see section 4.3). The Committee therefore concluded that it would consider its recommendations for the untreated group in the light of its decision of the cost‑effectiveness results from the company's model for the high‑risk relapsed groups.

4.10 The Committee considered the clinical effectiveness parameters used in the company's model. It heard from the ERG that the company's model contained an assumption that time on treatment was restricted, but that the benefits of treatment with idelalisib plus rituximab continued after both the time horizon of the trial and treatment discontinuation up to a maximum of 5 years. The Committee heard from the ERG that these assumptions may have overestimated the treatment effect of idelalisib plus rituximab. The Committee noted that the company had measured the effect of this assumption in its sensitivity analyses (see section 3.56). The Committee heard from the clinical experts that patients would continue to have treatment for chronic lymphocytic leukaemia until disease progression or until the treatment stopped working but that after stopping treatment, rebound deterioration could occur. The Committee concluded that the treatment benefit of idelalisib plus rituximab is unlikely to continue beyond treatment discontinuation, and therefore considered the most plausible assumption was to limit the treatment benefit of idelalisib plus rituximab, after treatment discontinuation, to 3 years.

4.11 The Committee considered the resource use assumptions in the company's model. It heard from the ERG that the company had assumed in the proposed base case that 45% of patients whose disease did not respond to treatment would have intravenous immunoglobulin therapy (see section 3.58). It heard from the clinical experts that this was an overestimate and that in clinical practice, up to 20% of patients whose disease did not respond to therapy would have intravenous immunoglobulin therapy. The clinical experts also noted that intravenous immunoglobulin therapy is administered in up to 10% of patients whose disease did respond to treatment. The Committee concluded that in order to inform its decision‑making, the most plausible estimate for the proportion of patients who would have intravenous immunoglobulin therapy is 20% for those whose disease does not respond treatment and 10% for those whose disease does respond to treatment.

4.12 The Committee considered the utility parameters used by the company in its economic model. It heard from the ERG that the company had used EQ‑5D data collected alongside Study 116 for its pre‑progression states and utilities from Dretzke et al. (2010) in its post‑progression states (see section 3.32). The ERG highlighted that in the company's base case, patients who discontinued treatment maintained a permanently higher utility than in the rituximab alone group (see section 3.47). The ERG stated that it had explored using the Dretzke et al. (2010) values for both the pre‑progression and post‑progression states in its amendments to the company's base case, but that these changes did not have a notable effect on the incremental cost‑effectiveness ratios (ICERs) for idelalisib plus rituximab compared with its comparators. Following the appraisal consultation the company submitted a proposed base case which used the EQ‑5D trial data but corrected the errors identified by the ERG (see sections 3.51–3.56). The Committee heard from the ERG that these changes were appropriate (see section 3.58). The Committee therefore concluded that it was acceptable to use Study 116 EQ‑5D data to inform the pre‑progression utility values.

Most plausible ICER considerations

4.13 The Committee considered the most plausible ICER for idelalisib plus rituximab in people whose disease is high‑risk relapsed. The Committee noted the results of the proposed base case and exploratory analysis for the comparisons of idelalisib plus rituximab (using the simple discount agreement price for idelalisib) with the relevant comparators for people with high‑risk relapsed disease (see section 3.60). It noted that the ICERs ranged from £25,100 per quality‑adjusted life year (QALY) gained (compared with fludarabine, cyclophosphamide and rituximab) to £34,900 per QALY gained (compared with chlorambucil plus rituximab). The Committee was aware that these analyses included the assumption that 45% of people whose disease does not respond to treatment go on to have intravenous immunoglobulin therapy. It noted that when its preferred assumptions on intravenous immunoglobulin therapy usage were applied (see section 4.11), the ICER for the comparison with fludarabine, cyclophosphamide and rituximab increased to £26,500 per QALY gained, and for the comparison with chlorambucil plus rituximab the ICER increased to £36,400 per QALY gained (see section 3.61). The Committee was also aware that when its preferred assumption for treatment benefit of idelalisib plus rituximab following treatment discontinuation was applied (see section 4.10), the ICERs would likely increase further, up to a maximum of about £36,000 per QALY gained for the comparison with fludarabine, cyclophosphamide and rituximab and £46,000 per QALY gained for the comparison with chlorambucil plus rituximab. The Committee therefore concluded that the most plausible ICERs for idelalisib plus rituximab compared with fludarabine, cyclophosphamide and rituximab, chlorambucil plus rituximab and bendamustine plus rituximab were within the range of approximately £36,000 per QALY gained and £46,000 per QALY gained.

4.14 The Committee considered the most plausible ICER for idelalisib plus rituximab in people with previously treated refractory chronic lymphocytic leukaemia. The Committee was aware that in this group the relevant comparator was rituximab alone or best supportive care. The Committee recognised that the benefits for this group would probably be better than for the relapsed group given the paucity of other treatments. Nevertheless, the Committee explored the calculated cost‑effectiveness for this group separately. It noted that in the company's proposed base‑case analysis, after using the simple discount agreement price for idelalisib, the ICER was £26,400 per QALY gained for the comparison with rituximab alone and £35,300 per QALY gained for the comparison with best supportive care (see section 3.54). The Committee noted that if its preferred assumptions on intravenous immunoglobulin therapy usage were applied (see section 4.11), the ICER for the comparison with rituximab alone increased to £31,300 per QALY gained and for the comparison with best supportive care the ICER increased to £40,200 per QALY gained. The Committee was also aware that when its preferred assumption on treatment benefit of idelalisib plus rituximab following treatment discontinuation was applied (see section 4.10), the ICERs would likely increase further, up to a maximum of £41,000 per QALY gained for the comparison with rituximab alone and £50,000 per QALY gained for the comparison with best supportive care. The Committee concluded that the most plausible ICER for idelalisib plus rituximab compared with rituximab alone is between approximately £31,000 per QALY gained and £41,000 per QALY gained, and the most plausible ICER for the comparison with best supportive care is between approximately £40,000 per QALY gained and £50,000 per QALY gained.

End‑of‑life considerations

4.15 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.

4.16 The Committee discussed each end‑of‑life criterion in turn for the populations that the company had provided evidence for (that is, people with previously treated high‑risk relapsed and refractory chronic lymphocytic leukaemia). For the short life expectancy criterion, the Committee noted the results of Study 116 which showed that that mean overall survival in the rituximab plus placebo group was less than 24 months. The Committee then considered the mean overall survival for the comparators relevant to the high‑risk relapsed group (disease relapse less than 24 months since previous treatment). It noted that the modelled mean overall survival for the comparators was less than 24 months (see section 3.31). The Committee further noted the clinical experts' view that life expectancy is approximately 12–24 months for people with high‑risk relapsed disease and less than 12 months for people with refractory disease. The Committee therefore concluded that the short life expectancy criterion was met for people with high‑risk relapsed and refractory chronic lymphocytic leukaemia.

4.17 The Committee discussed the life extension criterion. For people with high‑risk relapsed disease, the life expectancy was greater than 3 months with idelalisib plus rituximab (data supplied academic in confidence and so cannot be reported here) compared with the modelled mean overall survival for the relevant comparators (see section 3.31). For people with refractory chronic lymphocytic leukaemia, the mean overall survival with idelalisib plus rituximab in Study 116 demonstrated a life expectancy greater than 3 months compared with rituximab plus placebo (data supplied academic in confidence and so cannot be reported here). The Committee concluded that the life extension criterion was met for people with high‑risk relapsed chronic lymphocytic leukaemia and for people with refractory chronic lymphocytic leukaemia.

4.18 The Committee discussed the evidence for the small population size. It noted evidence provided by the company which showed that fewer than 7000 people are likely to have idelalisib plus rituximab. The Committee also heard from the clinical experts that the number of people with previously treated chronic lymphocytic leukaemia is likely to be around 1400. The Committee concluded that both the relapsed and refractory chronic lymphocytic leukaemia groups fulfil the small population end‑of‑life criterion.

Innovation

4.19 The Committee discussed how innovative idelalisib plus rituximab is in its potential to make a significant and substantial impact on health‑related benefits. It understood that idelalisib is a novel agent and that there was a high level of unmet need in this disease area, and it agreed that idelalisib offered a step change in the treatment of chronic lymphocytic leukaemia. Despite this, no additional evidence was submitted to suggest that these elements of innovation offered demonstrable and distinctive benefits of a substantial nature, which had not already been adequately captured in the QALY gain for idelalisib. The Committee concluded that no special considerations were needed to account for any health‑related quality‑of‑life benefits that had not already been captured in the economic model.

Committee conclusions

4.20 The Committee considered its recommendations for the 4 relevant populations in turn (see section 4.3). In people with previously treated relapsed chronic lymphocytic leukaemia (that is, disease relapse 24 months or more after previous treatment), it noted that no evidence had been submitted to support making recommendations for this population. In the absence of evidence, the Committee concluded that idelalisib plus rituximab is not recommended as a cost‑effective use of NHS resources for people with previously treated relapsed chronic lymphocytic leukaemia.

4.21 The Committee considered the use of idelalisib plus rituximab in people with previously treated high‑risk relapsed chronic lymphocytic leukaemia (that is, disease relapse less than 24 months after previous treatment). The Committee noted that the most plausible ICERs for the comparisons with fludarabine, cyclophosphamide and rituximab, chlorambucil with rituximab and bendamustine with rituximab were within the range of £36,000 per QALY gained and £46,000 per QALY gained. It further noted that this population could be considered under the supplementary advice to the Committee on end‑of‑life treatments. The Committee therefore concluded that idelalisib plus rituximab represents a cost‑effective use of NHS resources and should be recommended for people with previously treated high‑risk relapsed chronic lymphocytic leukaemia.

4.22 The Committee considered the use of idelalisib plus rituximab in people with previously treated refractory chronic lymphocytic leukaemia (that is, treatment failure or disease relapse less than 6 months after previous treatment). The Committee noted that the most plausible ICER for the comparison with rituximab alone is between £31,000 per QALY gained and £41,000 per QALY gained and the most plausible ICER for the comparison with best supportive care is between £40,000 per QALY gained and £50,000 per QALY gained. It further noted that this population could be considered under the supplementary advice to the Committee on end‑of‑life treatments. The Committee therefore concluded that idelalisib plus rituximab represents a cost‑effective use of NHS resources and should be recommended for people with previously treated refractory chronic lymphocytic leukaemia.

4.23 The Committee considered the use of idelalisib plus rituximab in people with untreated chronic lymphocytic leukaemia with a 17p deletion or TP53 mutation. It was aware that no cost‑effectiveness evidence had been submitted for this group. The Committee recalled the comments it had received during the appraisal consultation, which highlighted that the potential benefits of idelalisib plus rituximab for this group were sizeable because of the poor prognosis associated with these types of genetic mutations. The Committee noted that this group would, in effect, have access to idelalisib if the disease became relapsed because of its conclusion in section 4.22 above; however, given the discussions with the clinical experts who advised that denying access to idelalisib plus rituximab in this group could be harmful (because chemotherapy is ineffective and increases the risk of further deletions and mutations), the Committee felt it was unreasonable to postpone access until their disease had relapsed. The Committee therefore concluded that idelalisib plus rituximab represents a cost‑effective use of NHS resources and should be recommended for people with untreated chronic lymphocytic leukaemia with a 17p deletion or TP53 mutation.