4 Consideration of the evidence

The Appraisal Committee reviewed the data available on the clinical and cost effectiveness of obinutuzumab plus chlorambucil, having considered evidence on the nature of chronic lymphocytic leukaemia and the value placed on the benefits of obinutuzumab plus chlorambucil by people with the condition, those who represent them, and clinical experts. It also took into account the effective use of NHS resources.

4.1 The Committee discussed with the patient expert the nature of the condition. It heard that some people with chronic lymphocytic leukaemia have a variety of symptoms, some of which can be disabling, for example, fatigue and concurrent infections. The patient expert stated that because patients are on life‑long treatment, patients who are in remission are concerned about imminent relapse and the need for additional treatment. The Committee agreed that chronic lymphocytic leukaemia can seriously impair health‑related quality of life.

4.2 The Committee discussed the current clinical management of chronic lymphocytic leukaemia and the most likely place in the pathway of care for obinutuzumab plus chlorambucil treatment. It heard from the clinical expert that one‑third of people with chronic lymphocytic leukaemia are asymptomatic and may not need immediate treatment. The Committee heard from the clinical expert that, for people with untreated chronic lymphocytic leukaemia, fludarabine combination therapy is the standard of care when they need immediate treatment. It understood that fludarabine combination therapy may not be suitable for about half the people needing immediate treatment, for example, people who are older or have comorbidities such as impaired renal function, hypertension or diabetes. The Committee heard from the clinical expert that people who cannot have fludarabine combination therapy may have bendamustine either as monotherapy or combined with rituximab. The Committee also heard from the clinical expert that bendamustine may not be appropriate for some people and these people currently have chlorambucil monotherapy or rituximab plus chlorambucil (see section 4.3). The Committee acknowledged that, for people with untreated chronic lymphocytic leukaemia, NICE technology appraisal guidance on rituximab for the first-line treatment of chronic lymphocytic leukaemia recommends rituximab only in combination with fludarabine and cyclophosphamide. It also noted that NICE technology appraisal guidance on bendamustine for the first-line treatment of chronic lymphocytic leukaemia recommends bendamustine for people who cannot have fludarabine. The Committee heard from the clinical expert that obinutuzumab is a significant clinical advance over bendamustine and chlorambucil. Furthermore, some people may prefer to have obinutuzumab instead of bendamustine, because obinutuzumab is associated with fewer adverse events. The clinical expert and patient expert acknowledged that some people may prefer oral treatment with chlorambucil instead of having to attend a day unit for intravenous treatment with obinutuzumab or bendamustine. The Committee recognised that patients value having more treatment options.

4.3 The Committee discussed the company's response to the appraisal consultation document. In its response the company requested that 2 subgroups of people who cannot have fludarabine‑based therapy should be considered: people who can have bendamustine and people who cannot have bendamustine. It heard that there are people who may not be fit enough to tolerate bendamustine, but are fit enough for active treatment with either rituximab plus chlorambucil or chlorambucil alone. The Committee noted that, of the people who cannot have fludarabine‑based treatment, the proportion who would have a bendamustine‑containing treatment was estimated to be around 25% in the company's original submission, whereas the CLL11 trial suggested that this could be as high as 50%. The Committee concluded that, in people who cannot have fludarabine‑based treatment, it was reasonable to consider 2 distinct subgroups: those who can have bendamustine‑based treatment and those who cannot.

Clinical effectiveness

4.4 The Committee considered the evidence presented by the company on the clinical effectiveness of obinutuzumab plus chlorambucil compared with chlorambucil alone or in combination with rituximab. It noted that the main source of evidence was the CLL11 open‑label randomised controlled trial. The Committee noted that, in CLL11, obinutuzumab plus chlorambucil was associated with statistically significantly greater progression‑free survival than chlorambucil alone (March 2014 data: hazard ratio [HR] 0.19; 95% confidence interval [CI] 0.14 to 0.25) or rituximab plus chlorambucil (March 2014 data: HR 0.41; 95% CI 0.33 to 0.50). The Committee also noted that obinutuzumab plus chlorambucil was associated with statistically significantly greater overall survival compared with chlorambucil monotherapy and that the difference in overall survival between obinutuzumab plus chlorambucil and rituximab plus chlorambucil was not statistically significant. However, the Committee acknowledged that the overall survival data were immature (see section 3.6). The Committee heard from the Evidence Review Group (ERG) that the open‑label design may have biased the primary outcome of investigator‑assessed progression‑free survival. It also noted that, as a result of the different routes of administration of the treatments in each arm, the number of placebo treatments that would be needed to make the study double blind would be unethical. The Committee heard from the ERG and clinical expert that a lower dose of chlorambucil had been used in CLL11 than the dose routinely used in clinical practice in England (see section 3.12). The ERG and clinical expert considered that if a lower dose of chlorambucil did have a lower efficacy it was likely to be similarly lower for all the treatment groups in CLL11. The Committee considered that the lower dose of chlorambucil in CLL11 was unlikely to significantly affect the comparative efficacy of the treatment groups (chlorambucil monotherapy, obinutuzumab plus chlorambucil, and rituximab plus chlorambucil). The Committee concluded that, for progression‑free survival, obinutuzumab plus chlorambucil is a clinically effective treatment for chronic lymphocytic leukaemia compared with chlorambucil alone or with chlorambucil plus rituximab.

4.5 The Committee considered the network meta‑analyses used by the company to estimate the progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine monotherapy (HR 0.399; 95% CI 0.218 to 0.672) in the absence of any head‑to‑head trials. The Committee was aware that the small network meta‑analysis did not compare obinutuzumab plus chlorambucil with bendamustine. Therefore, it was not considered further and the Committee focused on the large network meta‑analysis (see section 3.8). It heard from the ERG that the large network meta‑analysis included studies of patients for whom fludarabine therapy was suitable. These patients were not covered by the marketing authorisation for obinutuzumab or the scope of this appraisal. The Committee acknowledged that patients for whom fludarabine therapy is suitable are likely to be younger and have fewer comorbidities than patients in CLL11, and their disease may respond differently to treatment. The Committee agreed that the hazard ratio calculated in the large network meta‑analysis for progression‑free survival was not reliable when comparing obinutuzumab plus chlorambucil with bendamustine monotherapy.

4.6 The Committee considered adverse events associated with obinutuzumab treatment. It was aware that the Medicines and Healthcare Products Regulatory Agency had issued a warning about serious and fatal infusion‑related reactions associated with the use of ofatumumab and other anti‑CD20 monoclonal antibodies. The Committee heard from the clinical expert that initial infusion‑related reactions in people having obinutuzumab were managed in CLL11 by splitting the first dose into 2 administrations, in line with the summary of product characteristics. The clinical expert also noted that, other than infusion‑related reactions, obinutuzumab seemed to be well tolerated. The Committee took into consideration the summary of product characteristics and concluded that obinutuzumab had an acceptable adverse event profile.

Cost effectiveness

4.7 The Committee considered the company's economic model (without the patient access scheme) and the critique and exploratory analyses from the ERG. The Committee noted that the company presented a comparison of obinutuzumab plus chlorambucil with chlorambucil monotherapy, chlorambucil plus rituximab, bendamustine monotherapy and bendamustine plus rituximab.

4.8 The Committee considered the utility values calculated by the company. The Committee was aware that quality‑of‑life data were collected during CLL11 using the European Organisation for Research and Treatment of Cancer Quality of Life – Core 30 (EORTC‑QLQ‑C30) questionnaire (see sections 3.22 and 3.28). However, these data were not mapped onto the EuroQol (EQ‑5D) questionnaire or presented in the company's submission. The Committee heard from the company that it had considered mapping the quality‑of‑life data from CLL11, but decided to carry out a utility elicitation study instead. This was because the mapping tools were designed for non‑chronic lymphocytic leukaemia disease areas and used different versions of the EORTC‑QLQ, and the global domain scores from the EQ‑5D may not be applicable to people with chronic lymphocytic leukaemia. The Committee noted that the utility elicitation study was done with a sample of the general population and not people who had chronic lymphocytic leukaemia. It was also not stratified by age. The Committee concluded that the utility elicitation study was not the most appropriate source of utility values. The resulting utility values used in the cost‑effectiveness modelling were not reliable because they were not from people with chronic lymphocytic leukaemia and were not mapped onto the EQ‑5D.

4.9 The Committee acknowledged that the ERG had concerns about some of the assumptions made by the company in the base‑case analysis:

  • The utility value while on obinutuzumab treatment after the first cycle of treatment.

  • The utility value for progression‑free survival off treatment.

  • The mean doses of rituximab and bendamustine in the bendamustine plus rituximab arm.

  • The progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine monotherapy.

  • The progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine plus rituximab.

    The Committee considered each of these assumptions in turn, as detailed below (see sections 4.10–4.14).

4.10 The Committee discussed the utility values used after the first cycle of treatment while on obinutuzumab. The Committee noted that the company used the same utility value as for progression‑free survival off treatment (0.82). The Committee heard from the ERG that the utility value while on obinutuzumab treatment should have been the same as the utility value for progression‑free survival on intravenous treatment (0.67). The Committee heard from the company that using 0.82 as a utility value was an error. The Committee concluded that the utility value of 0.67 was the more appropriate utility value for after the first cycle of treatment while on obinutuzumab treatment.

4.11 The Committee then considered the utility value for progression‑free survival off treatment. The Committee was aware that the company used a utility value of 0.82 from its utility elicitation study. The Committee heard from the ERG that this utility value was higher than the utility value for members of the general public at a similar age to people with chronic lymphocytic leukaemia. The Committee agreed that it was not plausible that the utility value for progression‑free survival off treatment was higher than the utility value for members of the general public without the disease. The ERG suggested an upper value for the utility value for progression‑free survival off treatment of 0.76. This was based on the mean utility value for the UK general population at the same average age of people with chronic lymphocytic leukaemia whose disease has not progressed and who are off treatment. The Committee heard from the ERG that the utility value would probably be lower than 0.76 for people with chronic lymphocytic leukaemia who have comorbidities and it suggested a utility value of 0.71. This was based on progression‑free survival on intravenous treatment. The Committee accepted that the utility values of 0.76 and 0.71 for progression‑free survival off treatment were more plausible than those estimated by the company, but were still subject to some uncertainty.

4.12 The Committee discussed the drug acquisition costs used in the company's model (without the obinutuzumab patient access scheme) and noted that the company had used an assumed dose intensity of 100% for bendamustine and rituximab. The Committee heard that using dose intensities from trial data would have been more appropriate. The ERG suggested that the dose intensity for bendamustine should be equal to that of bendamustine monotherapy (90%) in Knauf et al. (2009) and for rituximab should be equal to that of rituximab in the rituximab plus chlorambucil arm in CLL11 (98.8%). The Committee concluded that these dose intensities were likely to be more accurate than those suggested by the company because they were based on data rather than assumptions.

4.13 The Committee considered the progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine monotherapy. It recalled that the results of the company's network meta‑analysis were uncertain (see sections 3.8, 3.13 and 4.5). The Committee heard from the ERG that it was possible to do an indirect comparison of obinutuzumab plus chlorambucil with bendamustine using the CLL11 and Knauf et al. (2009) results without having to do a network meta‑analysis. The Committee accepted that the hazard ratio identified by the ERG (0.55) was likely to be more accurate than the hazard ratio calculated in the company's large network meta‑analysis (0.40) because the results of the network meta‑analysis were uncertain. The Committee concluded that the progression‑free survival hazard ratio of 0.55 for obinutuzumab plus chlorambucil compared with bendamustine monotherapy was more plausible than the hazard ratio of 0.40 calculated by the company.

4.14 The Committee considered the progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine plus rituximab. The Committee was aware that the company estimated a hazard ratio for bendamustine plus rituximab compared with rituximab plus chlorambucil using power calculation assumptions from an ongoing randomised controlled trial (the MaBLe trial). This was because the network meta‑analyses did not provide a hazard ratio for this comparison. The Committee noted that the company had calibrated the correlation between the hazard ratio and the proportion of people who had a complete response for bendamustine plus rituximab and rituximab plus chlorambucil comparison using the estimated sample size of MaBLe (see section 3.8). The Committee heard from the ERG that it would have been more appropriate to calibrate the correlation using the interim proportion of patients who had a complete response in the MaBLe trial rather than using the estimated sample size. The Committee accepted the ERG's suggested calibration value and that the resulting progression‑free survival hazard ratio of 0.76 was more reliable than the company's hazard ratio of 0.68, but it was still subject to some uncertainty because it was based on interim data from the MaBLe trial. The Committee concluded that the progression‑free survival hazard ratio of 0.76 for obinutuzumab plus chlorambucil was more plausible than the hazard ratio of 0.68.

4.15 The Committee considered the company's response to the appraisal consultation document together with their new evidence and patient access scheme (see section 3.36). It agreed that the population who cannot have fludarabine could be divided into 2 subgroups of people: those who can have treatment with bendamustine and those who cannot (see section 4.3). The Committee noted that the company had accepted all the ERG's suggested amendments and the Committee's preferred amendments to some of the assumptions made by the company in the base‑case analysis (outlined in sections 4.10–4.14). The Committee noted that the company had decreased the neutropenia costs to more accurately reflect the proportion of people needing hospital stays for neutropenia, but it acknowledged the ERG's comments that the alternative costs made little difference to the ICERs. The Committee discussed the company's sensitivity analysis that explored an alternative utility value for progression‑free survival off treatment (0.76) rather than the ERG's preferred estimate of 0.71. It was aware that the alternative utility value was from the COMPLEMENT‑1 study (see section 3.36), which reflected the patient population being considered as part of this appraisal. The Committee noted that the ERG's preference for a 0.71 utility value was because COMPLEMENT‑1 only partly represented the time period of interest and because of evidence from a large study that estimated a utility value of 0.76 for the general population in England (Ara and Brazier, 2010). The Committee agreed with the ERG that if the best evidence available suggests that the utility value for the general population for this age group is 0.76, then it is logical that for people with chronic lymphocytic leukaemia and multiple comorbidities the utility value for progression-free survival off treatment would be less than 0.76. The Committee recalled that people with chronic lymphocytic leukaemia who are in remission are concerned about imminent relapse that could affect their quality of life (see section 4.1). The Committee concluded that the company's revisions to its economic model, including the patient access scheme, the ERG's suggested amendments and updated neutropenia costs, were appropriate and that the company's revised base‑case cost‑effectiveness estimates were the most appropriate for its decision‑making.

4.16 The Committee considered the cost‑effectiveness estimates for people who can have bendamustine. It noted that both the company's and the ERG's ICERs (including the patient access scheme) for obinutuzumab plus chlorambucil compared with both bendamustine alone and with bendamustine plus rituximab were above the top end of the range that would normally be considered a cost‑effective use of NHS resources (£20,000–30,000 per QALY gained). The Committee also recognised that all the comparisons with bendamustine‑containing treatments were based on very weak evidence (see section 4.5). The Committee concluded that it could not recommend obinutuzumab plus chlorambucil for people who can have bendamustine, especially when this group has the option of treatment with a bendamustine‑based therapy.

4.17 The Committee considered the cost‑effectiveness estimates for people who cannot have bendamustine. It noted that both the company's and the ERG's ICERs (including the patient access scheme) for obinutuzumab plus chlorambucil compared with both chlorambucil alone and with rituximab plus chlorambucil were all in the range considered to be a cost‑effective use of NHS resources (£20,000–30,000 per QALY gained). The Committee considered that obinutuzumab plus chlorambucil is a clinically effective treatment for people who have untreated chronic lymphocytic leukaemia that is unsuitable for fludarabine‑based therapy (see section 4.4). It acknowledged comments received during consultation about the lack of alternative effective treatments in people who cannot have bendamustine‑based treatment. The Committee recommended obinutuzumab plus chlorambucil for untreated chronic lymphocytic leukaemia in people who cannot have fludarabine‑based therapy, only if bendamustine is not suitable and the company provides obinutuzumab with the discount agreed in the patient access scheme.

4.18 The Committee considered whether obinutuzumab plus chlorambucil was innovative. It noted the company's comments that obinutuzumab plus chlorambucil results in improved progression‑free survival. The Committee concluded that there were no additional benefits with obinutuzumab plus chlorambucil that were not already captured in the QALY estimate in the modelling.

4.19 The Committee considered the potential equality issue raised by a consultee that failure to consider the population who cannot have fludarabine as 2 separate groups (those who can have bendamustine and those who cannot) may be interpreted as discriminatory. This is because people who cannot have bendamustine would not have access to alternative effective treatments if obinutuzumab was not recommended. The Committee decided that this was not an equality issue under the equality legislation. Therefore its recommendations did not lead to discrimination and it did not need to add to, or change, its recommendations.

4.20 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:

  • 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 criteria into account, the Committee must be persuaded that the estimates of the extension to life are robust and that the assumptions used in the reference case of the economic modelling are plausible, objective and robust.

4.21 The Committee considered the criterion for short life expectancy. It was aware that the overall survival data from CLL11 were immature and so it considered the overall survival estimates from CLL5, which were used to validate the extrapolation of the overall survival curves in the company's economic model. It noted that the median overall survival of patients with chronic lymphocytic leukaemia in the chlorambucil arm of CLL5 was 64 months. The Committee agreed that treatment with obinutuzumab plus chlorambucil does not fulfil the criterion for short life expectancy and did not consider it necessary to form a view on the remaining end‑of‑life criteria. The Committee concluded that treatment with obinutuzumab plus chlorambucil does not fulfil the criteria for special consideration in the supplementary advice from NICE.

Summary of Appraisal Committee's key conclusions

TA343

Appraisal title: Obinutuzumab in combination with chlorambucil for untreated chronic lymphocytic leukaemia

Section

Key conclusion

Obinutuzumab, in combination with chlorambucil, is recommended as an option for adults with untreated chronic lymphocytic leukaemia who have comorbidities that make full‑dose fludarabine‑based therapy unsuitable for them, only if:

  • bendamustine‑based therapy is not suitable and

  • the company provides obinutuzumab with the discount agreed in the patient access scheme.

1.1

There was uncertainty in several of the model parameters, namely: the utility value while on obinutuzumab treatment after the first cycle of treatment, the utility value for progression‑free survival off treatment, the mean dose of rituximab and bendamustine in the bendamustine plus rituximab arm of the analysis, the progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine monotherapy, and the progression‑free survival hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine plus rituximab.

4.9

In response to consultation, the company submitted a patient access scheme and revised cost‑effectiveness analyses that incorporated all the ERG's suggested amendments to assumptions made by the company in the base‑case analysis (see sections 4.10–4.14). The company also amended the cost of treating neutropenia. The Committee concluded that the company's revised base‑case cost‑effectiveness estimates were the most appropriate for its decision‑making.

4.15

The Committee agreed that the population who cannot have fludarabine could be divided into people who can have bendamustine and those who cannot. For people who cannot have bendamustine, it noted that the most likely ICERs (including the patient access scheme) for treatment with obinutuzumab plus chlorambucil when compared with both chlorambucil monotherapy and with chlorambucil plus rituximab were in the range considered to be a cost‑effective use of NHS resources (£20,000–30,000 per QALY gained).

4.15, 4.17

For people who can have bendamustine, the Committee noted that the most likely ICERs (including the patient access scheme) for treatment with obinutuzumab plus chlorambucil when compared with both bendamustine alone and with rituximab plus bendamustine therapy were above the top end of the range that would normally be considered a cost‑effective use of NHS resources (£20,000–30,000 per QALY gained).

4.16

Current practice

Clinical need of patients, including the availability of alternative treatments

Around one‑third of people with chronic lymphocytic leukaemia are asymptomatic and may not need immediate treatment. Fludarabine combination therapy is the standard of care for people needing immediate treatment, but may be unsuitable for around half the people needing treatment. If fludarabine combination therapy is not appropriate, people may have bendamustine either as monotherapy or combined with rituximab. If bendamustine is not appropriate, people may have chlorambucil monotherapy or rituximab plus chlorambucil.

4.2

The technology

Proposed benefits of the technology

How innovative is the technology in its potential to make a significant and substantial impact on health‑related benefits?

There were no additional benefits with obinutuzumab plus chlorambucil that were not already captured in the QALY estimate in the modelling.

4.18

What is the position of the treatment in the pathway of care for the condition?

For people for whom fludarabine combination therapy is unsuitable (for example people who are older or have comorbidities such as impaired renal function, hypertension or diabetes), some may prefer to have obinutuzumab instead of bendamustine because obinutuzumab is associated with fewer adverse events.

4.2

Adverse reactions

The Committee heard from the clinical expert that initial infusion‑related reactions in people having obinutuzumab were managed in CLL11 by splitting the first dose into 2 administrations in line with the summary of product characteristics. Other than infusion‑related reactions, obinutuzumab seemed to be well tolerated.

4.6

Evidence for clinical effectiveness

Availability, nature and quality of evidence

The company presented evidence from 1 trial, CLL11, which was an open‑label randomised controlled trial of obinutuzumab plus chlorambucil compared with chlorambucil monotherapy and chlorambucil plus rituximab. The primary outcome was progression‑free survival.

4.4

The company also presented 2 network meta‑analyses to estimate the hazard ratio for obinutuzumab plus chlorambucil compared with bendamustine monotherapy. The Committee was aware that the small network meta‑analysis did not compare obinutuzumab plus chlorambucil with bendamustine. Therefore it did not consider it further and focused on the large network meta‑analysis. The Committee agreed that the hazard ratio calculated for treatment with obinutuzumab plus chlorambucil compared with bendamustine monotherapy from the large network meta‑analysis was unreliable because the network included studies of patients who could have fludarabine therapy. These patients were not covered by the marketing authorisation for obinutuzumab or the scope of this appraisal. The Committee acknowledged that patients for whom fludarabine therapy is suitable are likely to be younger and have fewer comorbidities than patients in CLL11, and their disease may respond differently to treatment.

4.5

Relevance to general clinical practice in the NHS

The ERG and clinical expert believed that a lower dose of chlorambucil was used in CLL11 than the dose routinely used in clinical practice in England. However, they acknowledged that if a lower dose of chlorambucil did have a lower efficacy, it would be similarly lower for all treatment groups in CLL11.

4.4

Uncertainties generated by the evidence

The ERG felt that the open‑label design of CLL11 may have biased the primary outcome. The Committee accepted that as a result of the different routes of administration of the treatments in each arm, the number of placebo treatments needed to make the study double blind would be unethical.

4.4

The Committee believed that the hazard ratio calculated from the large network meta‑analysis was unreliable because the network included studies of patients who could have fludarabine therapy. These patients were not covered by the marketing authorisation for obinutuzumab or the scope of this appraisal. The Committee acknowledged that patients for whom fludarabine therapy is suitable are likely to be younger and have fewer comorbidities than patients in CLL11, and their disease may respond differently to treatment.

4.5

Are there any clinically relevant subgroups for which there is evidence of differential effectiveness?

The Committee heard that there are people who may not be fit enough to tolerate bendamustine, but are fit enough for active treatment with either rituximab plus chlorambucil or chlorambucil alone. It concluded that in people who cannot have fludarabine‑based treatment, it was reasonable to consider 2 distinct subgroups: those who can have bendamustine‑based treatment and those who cannot. No clinical evidence was provided for these 2 subgroups.

4.3

Estimate of the size of the clinical effectiveness including strength of supporting evidence

In CLL11, treatment with obinutuzumab plus chlorambucil was associated with statistically significantly greater progression‑free survival than chlorambucil monotherapy or treatment with rituximab plus chlorambucil. The Committee also noted that obinutuzumab plus chlorambucil therapy was associated with statistically significantly greater overall survival compared with chlorambucil monotherapy and that the difference in overall survival between treatment with obinutuzumab plus chlorambucil and rituximab plus chlorambucil was not statistically significant. The overall survival data from CLL11 were immature.

4.4

The Committee believed that the hazard ratio calculated from the large network meta‑analysis was unreliable because the network included studies of patients who could have fludarabine therapy. These patients were not covered by the marketing authorisation for obinutuzumab or the scope of this appraisal. The Committee acknowledged that patients for whom fludarabine therapy is suitable are likely to be younger and have fewer comorbidities than patients in CLL11, and their disease may respond differently to treatment.

4.5

Evidence for cost effectiveness

Availability and nature of evidence

The company presented a comparison of obinutuzumab plus chlorambucil therapy with chlorambucil monotherapy, chlorambucil plus rituximab therapy, bendamustine monotherapy, and bendamustine plus rituximab therapy.

4.7

European Organisation for Research and Treatment of Cancer Quality of Life (EORTC‑QLQ) data were collected in CLL11, but were not mapped to the EuroQol (EQ‑5D) or used in the model. Instead the company carried out a utility elicitation study to determine the utility values.

4.8

Uncertainties around and plausibility of assumptions and inputs in the economic model

Utility values were determined from a sample of the general population and not from people who had chronic lymphocytic leukaemia, and were not stratified by age. The Committee believed that the utility values used in the company's model were not reliable.

Several assumptions in the company's model were queried:

  • The company used the incorrect utility value while on obinutuzumab treatment after the first cycle of obinutuzumab treatment.

  • The utility value for progression‑free survival off treatment in the company's model was based on the utility elicitation study.

  • The company assumed a dose intensity of 100% for bendamustine and rituximab.

  • The company used its large network meta‑analysis to estimate a hazard ratio of 0.40 for the comparison with bendamustine monotherapy.

  • The company used the estimated sample size from an ongoing trial (MaBLe) to calibrate the correlation between the number of people who had a complete response and the progression‑free survival hazard ratio for treatment with bendamustine plus rituximab and treatment with rituximab plus chlorambucil.

The company submitted a patient access scheme and revised cost‑effectiveness analyses that incorporated all of the ERG's suggested amendments and the Committee's preferred amendments to some of the assumptions made by the company in the base‑case analysis. The Committee concluded that the company's revisions to its economic model, including the patient access scheme, the ERG's suggested amendments and updated neutropenia costs, were appropriate and that the company's revised base‑case cost‑effectiveness estimates were the most appropriate for its decision‑making.

4.8–4.15

Incorporation of health‑related quality‑of‑life benefits and utility values

Have any potential significant and substantial health‑related benefits been identified that were not included in the economic model, and how have they been considered?

EORTC‑QLQ data were collected in CLL11 but were not mapped to the EQ‑5D or used in the model. Utility values were determined from a sample of the general population and not from people who had chronic lymphocytic leukaemia and were not stratified by age. The Committee believed that the utility values used in the company's model were not reliable.

The Committee queried:

  • The utility value used after the first cycle of obinutuzumab treatment in the company's model, which was based on progression‑free survival off treatment. The Committee heard from the company that it had used the incorrect value. The ERG used a value based on progression‑free survival on intravenous treatment.

The utility value for progression‑free survival off treatment in the company's model, which was based on the utility elicitation study. The ERG highlighted that this utility value was higher than the utility value for the UK general population at a similar age to people with chronic lymphocytic leukaemia whose disease has not progressed and who are off treatment.

4.8, 4.10, 4.11

No additional benefits with obinutuzumab plus chlorambucil that were not already captured in the QALY estimate in the modelling were identified.

4.18

Are there specific groups of people for whom the technology is particularly cost‑effective?

The Committee agreed that the population who cannot have fludarabine could be divided into people who can have bendamustine and those who cannot. It noted that for people who cannot have bendamustine, the most likely ICERs (including the patient access scheme) for obinutuzumab plus chlorambucil compared with both chlorambucil alone and with rituximab plus chlorambucil were all in the range considered cost effective (£20,000–30,000 per QALY gained).

4.15, 4.17

What are the key drivers of cost‑effectiveness?

The company identified the key drivers of the model as the long‑term projection of progression‑free survival, the post‑progression death rate, the results of the large network meta‑analysis and the utility values used.

3.27

Most likely cost‑effectiveness estimate (given as an ICER)

For people who cannot have bendamustine, the Committee noted that the most likely ICERs (including the patient access scheme) for obinutuzumab plus chlorambucil compared with chlorambucil alone and with rituximab and chlorambucil were within the range considered cost effective (£20,000–30,000 per QALY gained).

4.17

For people who can have bendamustine, the Committee noted that the most likely ICERs (including the patient access scheme) for obinutuzumab plus chlorambucil compared with both bendamustine alone and with rituximab plus bendamustine were above the top end of the range that would normally be considered cost effective (£20,000–30,000 per QALY gained).

4.16

Additional factors taken into account

Patient access schemes (PPRS)

The company has agreed a patient access scheme with the Department of Health that makes obinutuzumab available with a discount. The size of the discount is confidential.

2.3, 5.4

End‑of‑life considerations

The Committee concluded that treatment with obinutuzumab plus chlorambucil does not fulfil the criterion for short life expectancy. The Committee did not consider it necessary to form a view on the remaining criteria.

4.21

Equalities considerations and social value judgements

One consultee commented that failure to consider the population who cannot have fludarabine as 2 separate groups (those who can have bendamustine and those who cannot) may be interpreted as discriminatory. This is because people who are ineligible for bendamustine would not have access to alternative effective treatments if obinutuzumab was not recommended. The Committee decided that this was not an equality issue under the equality legislation. Therefore its recommendations did not lead to discrimination and it did not need to add to, or change, its recommendations.

4.19

  • National Institute for Health and Care Excellence (NICE)