3 Committee discussion

The appraisal committee considered evidence submitted by AstraZeneca, a review of this submission by the evidence review group (ERG), and responses from stakeholders. See the committee papers for full details of the evidence.

The condition

Chronic lymphocytic leukaemia has substantial effects on quality of life

3.1 Chronic lymphocytic leukaemia (CLL) is a malignant disorder of white blood cells and is the most common type of leukaemia in England. The patient experts explained that the physical and psychological effects of CLL have a debilitating effect on their daily lives. The committee noted the increase in prevalence of CLL with age and the additional effect of the condition on family and carers. It concluded that CLL substantially affects both physical and psychological aspects of quality of life.

Treatment pathway and comparators

There is an unmet need for more effective treatments for CLL and a new treatment option would be welcome

3.2 The clinical and patient experts noted that people with untreated CLL are a heterogeneous population, with different mutation status and comorbidities. They agreed that there is an unmet need for more effective, targeted treatments with fewer side effects than existing NHS treatments. They considered that this unmet need is particularly high for people with a 17p deletion or TP53 mutation. This is because ibrutinib and idelalisib plus rituximab are the only available treatments, and idelalisib is poorly tolerated and not widely used. However, for people without a 17p deletion or TP53 mutation there is also a need for greater treatment choice. Around one-third of this population are offered fludarabine, cyclophosphamide and rituximab (FCR) or bendamustine plus rituximab (BR), which have many long-term side effects. The patient experts also noted that access to treatments other than these chemo-immunotherapies is limited. Chlorambucil plus obinutuzumab is the only other option, so targeted treatments such as acalabrutinib are needed. The committee acknowledged that for previously treated CLL that has progressed, the treatment options are currently limited to either ibrutinib or venetoclax plus rituximab. The patient experts explained that acalabrutinib is generally well tolerated and causes fewer side effects than current treatments. Also, it is an option when ibrutinib is not suitable for some people with cardiovascular comorbidities. The committee concluded that acalabrutinib would be welcomed as a new treatment option for people with CLL.

Treatment varies depending on mutation status and comorbidities

3.3 The clinical experts confirmed that mutation status and comorbidities affect the treatment options for people with untreated CLL. People without a 17p deletion or TP53 mutation who have comorbidities that make FCR or BR unsuitable for them, would be offered chlorambucil plus obinutuzumab. People with a 17p deletion or TP53 mutation would usually be offered ibrutinib. Idelalisib plus rituximab is rarely used in clinical practice because it has an intensive dosing regimen and is associated with increased infection risk. The clinical experts also stated that ibrutinib is the most used treatment for previously treated CLL that has progressed, but venetoclax plus rituximab is also used. The committee agreed that it was appropriate to model different treatments depending on mutation status and comorbidities.

For previously treated CLL, venetoclax plus rituximab is a relevant comparator

3.4 The company did not present evidence comparing acalabrutinib with venetoclax plus rituximab for the previously treated relapsed or refractory CLL population. It did not consider venetoclax plus rituximab to be a commonly used treatment in this population in the NHS. Instead, it regarded ibrutinib to be the established treatment for relapsed or refractory CLL in the NHS and was the only comparator in its cost-minimisation analysis for this population. The committee noted that a proportion of this population would likely have venetoclax plus rituximab, but the economic analysis did not include costs for this combination. The committee concluded that venetoclax plus rituximab is a relevant comparator for previously treated relapsed or refractory CLL.

The company does not present any evidence for a population with untreated CLL for which FCR or BR is suitable

3.5 The company's submission did not include people with untreated CLL for which FCR or BR is suitable, although this population was in the NICE scope and is included in the marketing authorisation for acalabrutinib. The patient experts suggested that acalabrutinib should have been presented in the company's submission as an alternative to chemo-immunotherapy for this population. However, the company's clinical experts suggested excluding this population from its clinical trial, ELEVATE-TN, in line with expected clinical practice. The company explained that ELEVATE-TN did include people with untreated CLL for which FCR or BR is suitable, but it presented no clinical or cost evidence for this group. The committee acknowledged that the company was not seeking reimbursement for acalabrutinib for this population and that no comparative evidence was presented. It concluded that although people with untreated CLL for which FCR or BR is suitable is an important subgroup, it could not make a recommendation for this group because no evidence had been presented.

Clinical effectiveness

The clinical-effectiveness evidence is largely relevant to NHS clinical practice in England

3.6 The company presented results for the population with untreated CLL from ELEVATE-TN, an open‑label randomised controlled trial comparing acalabrutinib monotherapy (n=179) with chlorambucil plus obinutuzumab (n=177). ELEVATE-TN included people over 18 years with untreated CLL whose multimorbidities made FCR or BR unsuitable. People in ELEVATE-TN had to have a Cumulative Illness Rating Scale score greater than 6, or a creatinine clearance of less than 70 ml/minute (low creatinine clearance levels indicate serious kidney damage). The company considered that these criteria meant that FCR or BR would be unsuitable for similar patients in NHS clinical practice. ELEVATE-TN also included an acalabrutinib plus obinutuzumab arm, but this was not part of the company's submission and was not considered further. Of the 356 people in ELEVATE-TN, 35 had a 17p deletion or TP53 mutation. For the population with previously treated CLL, the company presented clinical effectiveness evidence from ASCEND. This was an open-label, randomised controlled trial comparing acalabrutinib (n=155) with either idelalisib plus rituximab or BR (n=155). The ERG considered that the population in ELEVATE-TN broadly represented the population with untreated CLL for which FCR or BR was unsuitable as seen in the NHS in England. Also, the population in ASCEND broadly represented the population with previously treated relapsed or refractory CLL who would be eligible for treatment with acalabrutinib. The committee was satisfied that the clinical-effectiveness evidence was largely relevant to NHS clinical practice.

It is acceptable to use the full trial data from ELEVATE-TN in the untreated CLL model

3.7 The company used data from ELEVATE-TN to inform the economic analysis for the populations with untreated CLL. ELEVATE-TN included 35 people with a 17p deletion or TP53 mutation (high-risk CLL). The ERG explained that the company's economic model for untreated CLL used the full population from ELEVATE-TN. Because there is a separate model for people with high‑risk CLL, this resulted in the same population with high‑risk CLL being included in 2 different models with different comparators. The ERG noted that the effect of including the population with high‑risk CLL in the untreated CLL model was uncertain. The clinical experts explained that it was reasonable to assume a similar treatment effect of acalabrutinib for the populations with untreated CLL whether or not they had high-risk CLL. They considered that it was therefore acceptable to include both populations in the same model. The committee agreed and concluded that it was acceptable to use the full trial data from ELEVATE-TN in the untreated CLL model.

For untreated CLL when FCR or BR is unsuitable, acalabrutinib improves progression-free survival but the overall survival benefit is uncertain

3.8 After a median follow up of 28 months, there was a statistically significant increase in progression‑free survival for acalabrutinib compared with chlorambucil plus obinutuzumab (hazard ratio [HR] 0.20, 95% confidence interval [CI] 0.13 to 0.30, p<0.0001). Median progression‑free survival was not reached in the acalabrutinib arm and was 22.6 months in the chlorambucil plus obinutuzumab arm. Median time to next treatment was not reached in either treatment arm but the trend was towards this being longer with acalabrutinib. Median overall survival was not reached in either treatment arm and there was no difference in overall survival between the 2 arms (HR 0.60, 95% CI 0.28 to 1.27, p=0.1556). The committee concluded that the trial data showed that acalabrutinib increased progression-free survival and time to next treatment compared with chlorambucil plus obinutuzumab. The committee concluded that the benefit of acalabrutinib on overall survival was uncertain, noting the immaturity of the data (that is, the endpoints had not been reached).

For untreated high-risk CLL, an indirect treatment comparison in a different population is acceptable for decision making

3.9 The company's economic model for the population with untreated high-risk CLL used data from an indirect treatment comparison in the population with relapsed or refractory disease (see section 3.10). The company considered that the results from the indirect comparison could apply to the population with high-risk CLL and that acalabrutinib is clinically equivalent to ibrutinib. The ERG explained that because the data did not specifically relate to this population, there was uncertainty in assuming clinical equivalence based on the separate relapsed or refractory CLL population analysis. The clinical experts explained that there was no reason to consider acalabrutinib to be clinically inferior to ibrutinib and that assuming equivalent effectiveness was reasonable. The committee noted that there was no direct evidence presented for the population with high-risk CLL. Although there was uncertainty, it concluded that it was plausible that clinical equivalence between acalabrutinib and ibrutinib could be assumed in both populations and this was acceptable for decision making.

For previously treated CLL, the company's indirect treatment comparison with ibrutinib is acceptable for decision making

3.10 No direct evidence was presented comparing acalabrutinib with ibrutinib for the population with previously treated relapsed or refractory CLL. The company did an unanchored matching adjusted indirect comparison (MAIC) using data from the acalabrutinib arm of ASCEND and the ibrutinib arm of RESONATE. Individual patient data were weighted to match baseline characteristics between arms and all observed effect modifiers and prognostic variables accounted for in the analysis. Kaplan–Meier estimates for progression-free survival and overall survival were found to be similar for both interventions (the exact hazard ratios and statistical values are confidential and cannot be reported here). The results of the MAIC were used to inform the cost-minimisation approach for the population with previously treated relapsed or refractory CLL. The company considered that the results justified the assumption of equivalent efficacy between acalabrutinib and ibrutinib in the populations with previously treated relapsed or refractory CLL and untreated high-risk CLL. The ERG considered that the methods for the indirect comparison were appropriate, concluding that it was reasonable to assume clinical equivalence of acalabrutinib and ibrutinib in the population with previously treated relapsed or refractory CLL. The committee concluded that the indirect treatment comparison was acceptable for decision making.

Adverse effects

Acalabrutinib is generally well tolerated compared with current treatments

3.11 The results of ELEVATE-TN showed that acalabrutinib had an acceptable tolerability profile compared with chlorambucil plus obinutuzumab. The patient experts highlighted that acalabrutinib was associated with fewer adverse effects and was generally well tolerated. They explained that some people noted reduced adverse effects after changing to acalabrutinib from other treatments. The committee agreed that acalabrutinib was likely to be generally well tolerated compared with current treatments.

Cost-effectiveness model structure

The model structure is appropriate for decision making

3.12 For the population with untreated CLL, the company submitted a semi-Markov model with 3 states (progression-free, progressed disease and death). The company used data from ELEVATE‑TN to estimate progression-free survival, overall survival and time to next treatment using parametric curves fitted to Kaplan–Meier data. Post-progression survival was estimated from the overall survival data from the MURANO and RESONATE trials in previously treated CLL. Data from the venetoclax plus rituximab arm of MURANO were applied to people whose disease progressed on first-line acalabrutinib. Data from the ibrutinib arm of RESONATE were applied to people whose disease progressed on chlorambucil plus obinutuzumab. Acalabrutinib treatment was assumed to continue until disease progression or death. Chlorambucil plus obinutuzumab was given for 6 cycles or until disease progression or death. Following disease progression after initial treatment, the model included a delay between disease progression and starting subsequent treatment. The ERG highlighted that the model structure did not allow for a second progression event and subsequent treatment costs were applied from the start of the second treatment until death or the maximum treatment duration. The committee noted the uncertainty about the duration of subsequent treatment (see section 3.13 and section 3.14) but concluded that the model structure was appropriate for decision making.

Subsequent treatment

For untreated CLL, the distribution of subsequent treatments is uncertain

3.13 In the company's untreated CLL model, it assumed that the type of second-line treatment would depend on their first-line treatment. After disease progression, the company assumed that people in the acalabrutinib group would have second-line treatment with venetoclax plus rituximab and people in the chlorambucil plus obinutuzumab group would have ibrutinib. This sequence was assumed because, in clinical practice, it is unlikely that people would have a Bruton's Tyrosine Kinase (BTK) inhibitor such as acalabrutinib after another BTK inhibitor such as ibrutinib. The company considered that ibrutinib was mainly used after chlorambucil plus obinutuzumab in the NHS. The company assumed that in the chlorambucil plus obinutuzumab group, 13% go on to have venetoclax plus rituximab. The ERG highlighted that there was some uncertainty in the proportion of people who would have venetoclax plus rituximab in the chlorambucil plus obinutuzumab group, but it also considered 13% reasonable. In response to consultation, the company provided more evidence to support its initial assumption. The clinical experts explained that venetoclax plus rituximab was a relatively recent treatment option. At the second committee meeting, they agreed that it was likely to currently account for between 13% and 20% of second-line treatment after chlorambucil plus obinutuzumab, but noted this proportion may increase over time. The committee agreed that the distribution of subsequent treatments after disease progression in the untreated CLL model was uncertain and considered scenarios with a range of proportions. It concluded that it was plausible that venetoclax plus rituximab currently accounts for up to 20% of second-line treatment after chlorambucil plus obinutuzumab, but that this may increase over time.

For untreated CLL, the ERG's model for costing subsequent treatments is appropriate

3.14 In the company's original economic model, subsequent treatment and its associated costs were modelled to continue from the start of subsequent treatment until death. However, the ERG highlighted that the model incorrectly applied second-line treatment costs because people would only have second-line treatment until progression. At this point treatment, and costs, would stop. In response to comments submitted by the company as part of their check of the factual accuracy of the ERG report, the ERG developed a second-line treatment costing model. In the ERG's model, mean progression-free survival for second-line ibrutinib treatment was extrapolated from the RESONATE progression-free survival data for 1 to 2 previous lines of treatment. The company agreed with this approach, but disagreed with the ERG on 2 points:

  • The company preferred a log-normal parametric curve for second-line treatment duration, which estimated a duration of 5.56 years for ibrutinib treatment. The ERG preferred the Weibull curve, which estimated a duration of 4.78 years.

  • The company assumed that the delay from progressing on first-line treatment through to starting second-line treatment would be 1 cycle. But the ERG used the company's original assumption of a 14-cycle delay.

    The ERG highlighted that the subsequent treatment costs accounted for a substantial proportion of the overall costs of the chlorambucil plus obinutuzumab comparator group. Therefore, increasing the duration of second-line ibrutinib treatment in the chlorambucil plus obinutuzumab group substantially increased the overall costs compared with the acalabrutinib group. The company considered that the log-normal distribution provided the most clinically realistic second-line ibrutinib treatment duration and provided the best statistical fit to the data. The clinical experts suggested that a second-line treatment duration of about 4.5 years for ibrutinib was reasonable, which corresponded to the ERG's approach. The ERG also highlighted that reducing the delay from progression to second-line treatment led to a much greater increase in the subsequent treatment cost of ibrutinib in the chlorambucil plus obinutuzumab group than in the subsequent treatment cost of venetoclax plus rituximab in the acalabrutinib group. The company explained that the 14-cycle delay was included to account for the assumption that people would have subsequent treatments at different ages depending on when they progressed. The company reduced the delay to 1 cycle because it considered that the ERG's model used a more personalised approach. The company also explained that the 14-cycle delay was based on immature data from ELEVATE‑TN. It discussed this with its clinical advisers who suggested that the 14-cycle delay was not clinically plausible because people would not have to wait for 1 year before starting second-line treatment. The ERG noted that the company's submission did not clearly identify the rationale for assuming a 14-cycle delay and that data from ELEVATE‑TN could be used to determine the mean delay, but those data were not presented. The clinical experts explained that in practice it is sometimes reasonable to wait 1 year before starting second-line treatment after progression with chlorambucil plus obinutuzumab. The committee considered the log-normal parametric model to be plausible but preferred the Weibull model because it was less constrained by overall survival gains. It agreed that the treatment duration with second-line ibrutinib was uncertain, with the most plausible estimate likely to be between that estimated using the log-normal and the Weibull distributions. It noted some uncertainty about the appropriate delay between progression and starting second-line treatment and separate scenarios were considered for the 14-cycle and the 1-cycle delays. The committee also acknowledged the effect of sequencing on costs of subsequent treatments (see section 3.13). It concluded that the ERG's model for costing subsequent treatments was appropriate, but that it would consider scenarios using the log-normal and Weibull distributions.

Survival extrapolations

For untreated CLL, the overall survival data are immature, leading to highly uncertain survival estimates

3.15 The data from ELEVATE-TN showed a trend towards improved overall survival for acalabrutinib compared with chlorambucil plus obinutuzumab. However, the data were immature, with median follow up at 28 months, and the difference between the groups was not statistically significant. The company estimated overall survival as a function of time to progression, pre-progression mortality and post-progression survival. Parametric survival models were fitted to data from ELEVATE-TN to model time to progression and pre-progression mortality. Post-progression survival in the acalabrutinib arm used overall survival data from the venetoclax plus rituximab arm of the MURANO trial. In the chlorambucil plus obinutuzumab group, data from the ibrutinib arm of RESONATE were used in a similar way but resulted in a lower overall survival rate. The ERG considered this approach favoured the sequence using venetoclax plus rituximab but that it was possible the difference in overall survival was because of confounding due to the unadjusted arm‑based comparison. The company considered that the MURANO and RESONATE trials reflected clinical practice for subsequent treatments as indicated by its clinical advisers. It therefore considered the post-progression survival estimates reasonable. The ERG highlighted that when modelling post-progression survival using MURANO, the overall survival hazard converged with that for the general population. This led to most people having acalabrutinib followed by venetoclax plus rituximab having similar survival to the general population. The ERG preferred to assume equal rates of post-progression survival for the acalabrutinib and chlorambucil plus obinutuzumab groups based on post-progression survival data from RESONATE because this leads to less optimistic projections of overall survival. The clinical experts suggested that overall survival was likely to be longer when starting treatment with acalabrutinib followed by venetoclax plus rituximab. This is because it is more effective and less toxic than chlorambucil plus obinutuzumab followed by ibrutinib. However, long-term data confirming overall survival benefit are lacking at present. They considered it reasonable to use MURANO because it accurately reflects the most likely treatment sequence of acalabrutinib followed by venetoclax plus rituximab. One clinical expert also explained that it was reasonable to expect that many people will reach the life expectancy of the general population after treatment with acalabrutinib and will be functionally cured. The other clinical expert did not consider this plausible. The committee concluded that there was considerable uncertainty in the overall survival estimates for acalabrutinib because of the extrapolation using data from trials for other treatments and the immature data from ELEVATE‑TN.

Cost-effectiveness results

For untreated CLL when FCR or BR is unsuitable, the incremental cost-effectiveness ratios are likely to be in the range normally considered acceptable

3.16 The incremental cost-effectiveness ratios (ICERs) used by the committee for decision making took account of all available confidential discounts, including those for comparators and follow-up treatments. Because of these confidential discounts, exact results cannot be reported here. The company's revised base-case ICER for acalabrutinib compared with chlorambucil plus obinutuzumab for untreated CLL when FCR or BR is unsuitable was within the range normally considered cost effective. Incorporating the ERG's preferred assumptions on applying a 14-cycle delay and using the Weibull model for subsequent treatment (see section 3.14) and using RESONATE post-progression survival for both treatment arms (see section 3.15) increased the ICER but it remained below £30,000 per quality-adjusted life year (QALY) gained. The committee used the ERG's base case for decision making. However, it also considered that further assumptions should be considered:

  • The proportion of people having second-line venetoclax plus rituximab was up to 20% (see section 3.13).

  • Adjusting the overall survival gain for acalabrutinib compared with chlorambucil plus obinutuzumab so that it was 50% lower, reflecting uncertainty about the immature survival data in ELEVATE-TN (see section 3.15).

    In all the combinations of scenarios the committee considered, the ICER remained below £30,000 per QALY gained, except the combination using the most pessimistic assumptions. The committee recalled that the most plausible duration of subsequent treatment was likely between that estimated by the log-normal and Weibull distributions (see section 3.14) which reduced the ICER considerably. So, it was satisfied that acalabrutinib is likely to be a cost-effective use of NHS resources for untreated CLL that is not high risk and when FCR or BR is unsuitable.

For people with a 17p deletion or TP53 mutation, acalabrutinib is likely to be cost saving compared with ibrutinib

3.17 In the company's base case from its cost-minimisation analysis, costs for acalabrutinib were lower than costs for ibrutinib for people with untreated CLL with a 17p deletion or TP53 mutation (high‑risk CLL). The cost-minimisation analysis considered if the acquisition and management of adverse events costs for acalabrutinib were lower than for ibrutinib. The ERG's analysis made no substantial changes to the company's base case and resulted in mostly unchanged cost savings for acalabrutinib treatment. Final costs considered by the committee took account of all available confidential discounts, including those for comparators. The committee concluded that for people with untreated CLL with a 17p deletion or TP53 mutation, acalabrutinib is likely to be cost saving compared with ibrutinib.

For previously treated CLL, acalabrutinib is likely to be cost saving compared with ibrutinib

3.18 In the company's base case from its cost-minimisation analysis, costs for acalabrutinib were lower than costs for ibrutinib for people with previously treated CLL. The cost-minimisation analysis considered if the acquisition and management of adverse events costs for acalabrutinib were lower than for ibrutinib. The ERG's analysis made no substantial changes to the company's base case and resulted in mostly unchanged cost savings for acalabrutinib treatment. The final costs considered by the committee took account of all available confidential discounts, including those for comparators. The committee concluded that for people with previously treated CLL, acalabrutinib was likely to be cost saving compared with ibrutinib. However, the committee considered that venetoclax plus rituximab was a relevant comparator for this population and because the company did not present a comparison with acalabrutinib, the cost effectiveness is unknown (see section 3.4).

End of life

Acalabrutinib does not meet the criteria to be considered a life-extending treatment at the end of life

3.19 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. It considered that the short life expectancy criterion of less than 24 months was not met because people with CLL have a life expectancy of more than 2 years. The committee concluded that acalabrutinib does not meet the criteria to be considered a life-extending treatment at the end of life.

Conclusions

Equality considerations

There are no equality issues relevant to the recommendations

3.23 The company's submission did not include people with untreated CLL for which FCR or BR is suitable. Patient submissions highlighted that this would potentially deny younger and fitter people access to a new treatment option that is well tolerated. However, the committee could not make a recommendation about the clinical and cost effectiveness of acalabrutinib for this population because the company did not present any evidence. Therefore, the committee did not consider this an equality issue it could resolve.

Innovation

There are no additional benefits that are not captured in the QALY calculations

3.24 The company considered acalabrutinib to be an innovative treatment because it is a highly selective BTK inhibitor that addresses a significant unmet need in first-line CLL treatment. Also, it offers an alternative option to the first-generation BTK inhibitor in previously treated CLL. Its targeted mechanism of action means it offers improved safety and tolerability compared with current treatments. The committee concluded that acalabrutinib would be a beneficial additional treatment option. However, it noted that it had not been presented with evidence of any additional benefits that were not captured in the measurement of QALYs.

  • National Institute for Health and Care Excellence (NICE)