The appraisal committee reviewed the data available on the clinical and cost effectiveness of rituximab, having considered evidence on the nature of relapsed or refractory chronic lymphocytic leukaemia and the value placed on the benefits of rituximab by people with the condition, those who represent them, and clinical specialists. It also took into account the effective use of NHS resources.
The appraisal committee discussed current standard clinical management of relapsed or refractory chronic lymphocytic leukaemia. The committee heard from clinical specialists that the most frequently used first-line treatments are: fludarabine plus cyclophosphamide with or without rituximab; and chlorambucil for people unable to have fludarabine because they have a poor performance status. However, for relapsed or refractory chronic lymphocytic leukaemia there is no single standard treatment option. The choice of treatment depends on a number of factors, including the presence of genetic abnormalities such as del(17p) mutation, previous treatments the person has received, whether a response was achieved from previous treatments, and if so, the duration of response. Clinical specialists noted that for these reasons, they considered it important to have a range of treatment options available. The committee heard that, for relapsed disease, treatments used previously may be administered again either with or without the addition of another therapeutic agent, or alternatively a different agent may be used. When additional or different treatments were used, these could include fludarabine and cyclophosphamide with the addition of mitoxantrone, alemtuzumab and stem cell transplantation.
The committee noted that the clinical effectiveness evidence for this appraisal was based mainly on a single unpublished randomised controlled trial (the REACH trial). In this trial rituximab plus fludarabine and cyclophosphamide was compared with fludarabine and cyclophosphamide. The committee heard from clinical specialists that people in the REACH trial were younger and had a better performance status than people with chronic lymphocytic leukaemia seen in routine practice in the NHS in England and Wales. However, the clinical specialists commented that the people in the trial were representative of the people who would be eligible for treatment with fludarabine plus cyclophosphamide. The committee discussed the inclusion in the trial of people who had Binet stage A chronic lymphocytic leukaemia. It heard from clinical specialists that the decision to treat chronic lymphocytic leukaemia would depend on symptoms and progression of disease rather than specific staging.
The committee discussed the exclusion from the REACH trial of people who were previously treated with fludarabine combination therapy, people who were previously treated with rituximab and people who had chronic lymphocytic leukaemia that was refractory to fludarabine. However, it heard from clinical specialists that, if suitable, people often had fludarabine combination regimens as a first-line treatment. It also heard that the publication of NICE's technology appraisal guidance on rituximab for the first-line treatment of chronic lymphocytic leukaemia recommending rituximab plus fludarabine and cyclophosphamide meant that in the future an increasing number of people with relapsed or refractory disease will have had rituximab and fludarabine combination therapy as a first-line treatment. The committee considered the exclusion of these groups from the clinical trial was a limitation for decision making because it meant that the trial population did not reflect all the people with relapsed or refractory disease who would be eligible for rituximab plus fludarabine and cyclophosphamide in the NHS.
The committee accepted that the REACH trial demonstrated that the addition of rituximab to fludarabine and cyclophosphamide improved progression-free survival and complete response rates. The committee noted there was potential for bias in outcome assessment because of the open-label design of the trial. The committee discussed the results of an interim analysis of the trial data. This was an independent assessment of response that was provided as academic-in-confidence. The committee noted that there was a difference between the investigator and independent assessments but was aware that the interim analysis was conducted 1 year before the investigator assessment. The committee heard from clinical specialists that assessment of progression-free survival was subjective and could change depending on familiarity with assessment tools. The committee considered that the differences in these assessments led to uncertainty in estimating the additional benefit of rituximab.
The committee noted that in the REACH trial median overall survival had not been reached in the rituximab group, and that survival curves for patients in the 2 treatment groups hardly diverged until 30 months. The committee heard from clinical specialists and patient experts that it is difficult for studies of chronic lymphocytic leukaemia to demonstrate an effect of treatment on overall survival because of the long natural history of the disease and because people with the disease often receive multiple treatments. It also heard that progression-free survival and response rates were often accepted as surrogates for overall survival. Furthermore, clinical specialists commented that longer term trial evidence is emerging that demonstrates an overall survival benefit of first-line treatment with rituximab plus fludarabine and cyclophosphamide. On balance, the committee was persuaded that the improvements observed in progression-free survival and response rates were likely to lead to at least some gain in overall survival, although this gain could not be quantified.
The committee noted that in the REACH trial there were slightly more grade 3 or 4 adverse events and treatment-related deaths in the rituximab plus fludarabine and cyclophosphamide group than in the fludarabine and cyclophosphamide group. It heard from clinical specialists that people with chronic lymphocytic leukaemia are aware of the risks of treatments and are willing to accept these risks because of the severity of the condition. The committee discussed the 6 cases of hepatitis B seen in the trial in the rituximab plus fludarabine and cyclophosphamide group. However, it heard from clinical specialists that this would be unlikely to happen in the UK because all people with chronic lymphocytic leukaemia are screened for hepatitis B before treatment, and so hepatitis B reactivation would be rare.
People who have previously received treatment with rituximab
The committee discussed the use of rituximab plus fludarabine and cyclophosphamide in people who have previously received treatment with rituximab-containing regimens. These people were excluded from the REACH trial, and the committee heard from clinical specialists that there was uncertainty about the degree of benefit of retreatment with rituximab. However, patient experts indicated that there was anecdotal evidence that people retreated with rituximab may have a good response to treatment. The committee also noted comments received at consultation that retreatment with rituximab is common in other lymphoproliferative conditions where there has been a good response, and that the same could be expected for chronic lymphocytic leukaemia. It was also aware that over the next few years there would be an increasing number of people who would be treated with rituximab and who would require further treatment following relapse.
The committee considered the evidence from uncontrolled phase 2 studies reporting the benefits of retreatment with rituximab and noted the methodological limitations of these studies. It discussed the MDACC data provided during consultation, reporting that there was a similar response rate and progression-free survival in people who have previously received rituximab compared with people who have not. However, it noted this study had limitations in its design, for example, it was open label and uncontrolled (and therefore not randomised). The study included 100 people who had previously been treated with a rituximab-containing regimen. However, limited data were provided about these regimens and they included rituximab monotherapy and rituximab plus chemotherapy other than fludarabine and cyclophosphamide. The committee was not persuaded that the results from this study could be considered reflective of the UK population, of whom an increasing number will have previously received rituximab plus fludarabine and cyclophosphamide.
The committee considered the use of rituximab plus fludarabine and cyclophosphamide in people who have previously received treatment with fludarabine. It first discussed people who had previously had a response to treatment with fludarabine (that is, people with fludarabine-sensitive disease). The committee discussed evidence from the REACH trial which included people whose disease was sensitive to fludarabine monotherapy. It noted that the REACH trial did not include people who had previously received fludarabine combination therapy. However, the committee considered that the clinical effectiveness was likely to be similar for people who were sensitive to fludarabine monotherapy and for people who were sensitive to fludarabine combination therapy. Therefore the committee was persuaded that data from the REACH trial could apply to people who were sensitive to fludarabine combination therapy.
The committee then discussed the evidence for the use of rituximab plus fludarabine and cyclophosphamide in people who have chronic lymphocytic leukaemia that is refractory to fludarabine. It noted the methodological limitations of the non-comparative studies provided by the manufacturer. The committee understood that clinical specialists did not consider that people with fludarabine-refractory disease should be retreated with the same fludarabine-containing regimen. The committee considered that the results of the MDACC study indicated a lower response to treatment with rituximab plus fludarabine and cyclophosphamide in chronic lymphocytic leukaemia refractory to fludarabine than in disease that was sensitive to fludarabine. The committee concluded that although people with fludarabine-refractory disease may derive some benefit from retreatment with fludarabine-containing chemotherapy regimens such as rituximab plus fludarabine and cyclophosphamide, the benefit was likely to be less than if the disease was fludarabine sensitive.
The committee recognised that the marketing authorisation for rituximab allowed its use with any chemotherapy regimen. It discussed the evidence on rituximab plus chemotherapy regimens other than fludarabine and cyclophosphamide. The committee discussed comments received on the appraisal consultation document that suggested that people who cannot take fludarabine and people with chronic lymphocytic leukaemia that is refractory to fludarabine may benefit from treatment with rituximab plus other chemotherapy. The committee was aware of the lack of treatment options available to these people. However, the committee noted the methodological limitations of the non-comparative evidence provided. It heard from the manufacturer that a study of rituximab plus chlorambucil for first-line treatment was under way and that preliminary data from a cross-trial analysis indicated that response rates were better for people treated with rituximab plus chlorambucil than with chlorambucil alone. Overall, the committee considered that there was significant uncertainty about the relative benefit of adding rituximab to chemotherapy regimens other than fludarabine and cyclophosphamide and therefore more research was needed.
The committee reviewed the economic model submitted by the manufacturer and the ERG's analysis of the model. It was aware that the model did not allow transition from the progressed health state to the progression-free survival health state. The committee considered that this did not appropriately reflect the disease process because people may receive later treatments with further periods of progression-free survival. The committee was aware that a similar model had been used in NICE's technology appraisal guidance on rituximab for the first-line treatment of chronic lymphocytic leukaemia. On balance, the committee agreed that the model could be used as a basis for considering the cost effectiveness of rituximab.
The committee considered how the costs of rituximab had been incorporated into the economic model. It noted that the ERG considered the assumption that costs were spread throughout the cycle in the base-case analysis inappropriate because rituximab was provided on the first day of each cycle. Therefore, the ERG explored remodelling rituximab costs so that costs were incurred at the start of each treatment cycle. The ERG re-analysis was corrected after consultation on the appraisal consultation document, and concluded that the ICER increased from £15,600 per QALY gained in the base case to £16,600 per QALY gained, which the committee accepted.
The committee discussed the utilities used in the economic model and noted that the evidence base for these estimates did not reflect the NICE reference case; in particular, preference-based methods were not used. It was aware that a utility study was under way in people with chronic lymphocytic leukaemia in the UK but detailed results from this study for people who had progressed following treatment were not yet available. The committee heard from patient experts that they considered an assumption of only a small difference in utility between the progressed and progression-free survival health states was not realistic. People greatly value being progression free and asymptomatic – it is associated with a marked improvement in quality of life. The committee considered the lack of appropriate utility data contributed to uncertainty in the economic model.
The committee discussed whether the modelled gains in overall survival from the economic model appropriately reflected the data from the clinical trial. It noted that the outputs from the manufacturer's economic analysis modelled a difference in overall survival between treatment groups from the start of treatment that did not reflect the trial data. The overall survival curves from the clinical trial provided by the manufacturer showed no difference in overall survival between the treatment groups before around 30 months, although, beyond this time, the extrapolated curves began to diverge. The committee considered that there was little evidence from the REACH trial to support the validity of the analysis provided by the manufacturer and that the manufacturer's base-case analysis was likely to have overestimated the benefits associated with rituximab.
The committee considered the estimates of cost effectiveness provided by the manufacturer and the additional exploratory analyses performed by the ERG that examined the impact on the ICER of reducing the survival advantage of treatment with rituximab. It noted that using an assumption of no overall survival advantage had the effect of increasing the cost-effectiveness estimates from £15,600 per QALY gained in the base case to £41,000 per QALY gained. Furthermore it recognised that when there was no modelled gain in overall survival the results became very sensitive to the difference between the utility values used for the progression-free survival health state and those used for the progressed health state which were uncertain. However, based on comments from the clinical specialists, the committee was persuaded that it was appropriate to assume at least some gain in overall survival in the economic model. Overall, the committee considered that the most plausible ICER was likely to be at the upper end of the range of £20,000 to £30,000 per QALY gained, which was higher than the ERG's corrected base case of £16,600 per QALY gained.
On balance, the committee was persuaded that even taking into account the uncertainty about utility values and the uncertainty about a gain in overall survival from treatment with rituximab, the use of rituximab plus fludarabine and cyclophosphamide was a cost-effective use of NHS resources for the population represented in the REACH trial; that is, people who have not previously received rituximab or fludarabine combination therapy and those whose disease is not refractory to fludarabine monotherapy. Additionally, the committee was persuaded that the cost effectiveness of rituximab plus fludarabine and cyclophosphamide could be generalised from people whose chronic lymphocytic leukaemia was sensitive to fludarabine monotherapy to those whose disease was sensitive to fludarabine combination therapy (section 4.10).
The committee was not persuaded of the clinical effectiveness of rituximab plus fludarabine and cyclophosphamide for people who have already been treated with rituximab. Nevertheless, the committee discussed the cost-effectiveness estimate provided by the manufacturer during consultation of £22,500 per QALY gained for people who had previously received rituximab. It noted that this did not include the correction for the timing of rituximab costs. It recognised that there was considerable uncertainty in the manufacturer's original base-case ICER because of the uncertainties in the gains in overall survival and the limitations in the health-related quality of life data available. The committee noted that even for the REACH trial population the most plausible ICER was likely to be at the upper end of the range of £20,000 to £30,000 per QALY gained. It considered that in the rituximab-pretreated population, for which there was little research, the manufacturer's estimated ICER could not provide a basis for decision making. The committee concluded that rituximab plus fludarabine and cyclophosphamide could not be recommended as an appropriate use of NHS resources for everyone who had previously been treated with rituximab. However, because of the uncertainty about the benefits of retreatment with rituximab, the committee concluded that rituximab plus fludarabine and cyclophosphamide should be recommended in the context of research for people with relapsed or refractory chronic lymphocytic leukaemia that has previously been treated with rituximab. The committee was aware of comments from consultees that some people in clinical trials had received rituximab in combination with treatments other than fludarabine and cyclophosphamide and at doses of rituximab lower than the licensed dose. The committee considered that this technology appraisal guidance should not adversely affect recruitment to future or ongoing clinical trials. Therefore the committee concluded that rituximab plus fludarabine and cyclophosphamide could be recommended for people with relapsed or refractory chronic lymphocytic leukaemia when rituximab had previously been given in the context of a clinical trial, either at a dose lower than currently licensed for chronic lymphocytic leukaemia, or in combination with chemotherapy other than fludarabine and cyclophosphamide.
After concluding that rituximab plus fludarabine and cyclophosphamide was cost effective for people with chronic lymphocytic leukaemia that was sensitive to fludarabine (section 4.18), the committee then considered its use in people with fludarabine-refractory disease. It noted the lower clinical response in people who were refractory to fludarabine than in people who were sensitive to it (section 4.10), and that clinical practice was not to retreat these people with the same fludarabine regimen. It also noted that the manufacturer had not provided an estimate of the cost effectiveness of rituximab plus fludarabine and cyclophosphamide in this population. The committee heard from the manufacturer that there were difficulties identifying baseline event rate data and that the relative efficacy of rituximab therapy in this group was uncertain. On this basis the committee considered that the use of rituximab plus fludarabine and cyclophosphamide for the treatment of people who had already had fludarabine could only be considered a cost-effective use of NHS resources when the chronic lymphocytic leukaemia remained fludarabine sensitive and not when it was fludarabine refractory.
The committee understood the potential need for other rituximab combinations for people whose disease is refractory to fludarabine or is not suitable for treatment with fludarabine. However, it concluded that there was significant uncertainty about the relative benefit of adding rituximab to chemotherapy regimens other than fludarabine and cyclophosphamide and therefore more research was needed. Furthermore, the committee noted there was no current basis for estimating the cost effectiveness of such combinations, or for considering them to be cost effective. The committee was aware that people with chronic lymphocytic leukaemia that is not suitable for treatment with fludarabine plus cyclophosphamide might be treated with rituximab plus other chemotherapy. It was also aware that this group might be older and include people with poor performance status or comorbidities. The committee considered whether equalities legislation and the requirement for fairness meant that it should make a positive recommendation for rituximab plus other chemotherapy for this group. However, given the lack of evidence for both the clinical and cost effectiveness of this treatment, the committee concluded that rituximab plus chemotherapy other than fludarabine and cyclophosphamide should only be used for the treatment of relapsed or refractory chronic lymphocytic leukaemia in the context of research.