3 The manufacturer's submission
3.1 The manufacturer's submission compared the combination of rituximab plus fludarabine and cyclophosphamide with the combination of fludarabine plus cyclophosphamide. This comparison was based on the REACH trial, a phase III, multicentre, open-label, randomised controlled trial in people with previously treated chronic lymphocytic leukaemia. People were enrolled if they had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1, a life expectancy greater than 6 months and if they had previously received treatment with chlorambucil monotherapy with or without prednisolone, fludarabine monotherapy (or other nucleoside analogue), or an alkylator-containing combination therapy (such as cyclophosphamide plus doxorubicin, vincristine and prednisolone, or cyclophosphamide plus vincristine and prednisolone). People were excluded from the trial if they had previously received treatment with interferon, rituximab or another monoclonal antibody, or fludarabine and cyclophosphamide, either concurrently or sequentially. People were also excluded if they had chronic lymphocytic leukaemia that was refractory to fludarabine (defined as not achieving at least a partial response for a minimum duration of 6 months). A total of 552 people were randomised to receive either rituximab plus fludarabine and cyclophosphamide or fludarabine and cyclophosphamide alone. The median age of people in the trial was 63 years and 67% were men. Most people (90%) had Binet stage B or C disease.
3.2 People in the trial were randomised to six cycles of treatment, with an interim assessment of response after three cycles. At this point, people whose disease showed a partial or complete response continued treatment to six cycles, people with progressive disease discontinued treatment and people with stable disease continued treatment at the investigator's discretion. Each treatment cycle of 28 days consisted of fludarabine and cyclophosphamide chemotherapy (fludarabine 25 mg/m2 and cyclophosphamide 250 mg/m2 on days 1, 2 and 3) with or without rituximab (375 mg/m2 on day 0 of cycle 1, 500 mg/m2 on day 1 of cycles 2–6). All treatments were administered intravenously.
3.3 The primary outcome of the trial was progression-free survival, defined as the time between randomisation and the date of the first documented disease progression, relapse or death by any cause. Secondary outcomes were event-free survival, overall survival, disease-free survival, duration of response, time to new chronic lymphocytic leukaemia treatment and response rates. Quality-of-life data were collected in the first year of the trial using the Functional Assessment of Cancer Therapy – General (FACT-G).
3.4 Demographic characteristics and disease characteristics, including Binet stage B symptoms and prognostic markers such as cytogenetic abnormalities, were well balanced between the trial groups. Of all people in the trial, 59% had Binet stage B disease, 31% had Binet stage C disease, and 10% had Binet stage A disease. The trial enrolled 42 people (8%) with del(17p), a chromosome mutation associated with a poorer prognosis. The manufacturer's submission stated that most people had previously been treated with single-agent chemotherapy (82%), most commonly an alkylating agent (66%) such as chlorambucil or cyclophosphamide. Of the people in the trial 56% were alkylator sensitive, 26% were alkylator refractory, and 16% had previously received fludarabine.
3.5 The trial results reported in the manufacturer's submission are based on a median follow-up of 25.3 months. At this point, the median progression-free survival was 30.6 months in the rituximab plus fludarabine and cyclophosphamide group and 20.6 months in the fludarabine and cyclophosphamide group, with a hazard ratio of 0.65 (95% CI 0.51 to 0.82, p = 0.0002). The best overall response rate was 69.9% in the rituximab plus fludarabine and cyclophosphamide group and 58% in the fludarabine and cyclophosphamide group (p = 0.0034). The median overall survival was 51.9 months in the fludarabine and cyclophosphamide group and was not reached in the rituximab plus fludarabine and cyclophosphamide group (hazard ratio 0.83, 95% CI 0.59 to 1.17, p = 0.2871).
3.6 The manufacturer presented a number of subgroup analyses. For people with the del(17p) mutation, the hazard ratio for progression-free survival was 0.75 (95% CI 0.38 to 1.49). The hazard ratio for progression-free survival for people with Binet stage A disease was 0.75 (95% CI 0.33 to 1.72), Binet stage B disease was 0.65 (95% CI 0.47 to 0.88) and Binet stage C disease was 0.61 (95% CI 0.41 to 0.90). The REACH trial was not powered to detect differences between the treatment groups for any of these subgroups.
3.7 In the REACH trial, 80% of people in the rituximab plus fludarabine and cyclophosphamide group experienced a grade 3 or 4 adverse event compared with 74% in the fludarabine and cyclophosphamide group. The most common grade 3 or 4 adverse events, with at least a 2% higher incidence in the rituximab plus fludarabine and cyclophosphamide group, were neutropenia, febrile neutropenia, granulocytopenia and hepatitis B infections. In the rituximab plus fludarabine and cyclophosphamide group, there were 19 treatment-related deaths (7%) and 51% of people had their treatment modified or interrupted for safety reasons. In the fludarabine and cyclophosphamide group, there were 14 treatment-related deaths (5%) and 39% of people had their treatment modified or interrupted for safety reasons.
3.8 The manufacturer provided supporting data from 20 non-comparative studies. These studies examined the efficacy and tolerability of rituximab plus different chemotherapy regimens, and of rituximab-containing regimens in people with fludarabine-refractory chronic lymphocytic leukaemia, and in people previously treated with rituximab (both groups had been excluded from the REACH trial). Of these 20 studies, 19 were uncontrolled, phase II studies and one was a randomised phase II trial of fludarabine, cyclophosphamide and mitoxantrone with or without rituximab in people with previously treated chronic lymphocytic leukaemia (n = 52). However, the small number of people included in each group in the randomised trial did not allow a statistical comparison to be made. Seven of the 20 trials investigated the use of rituximab outside the terms of the marketing authorisation (either rituximab monotherapy or rituximab plus non-chemotherapy regimens).
3.9 The largest non-comparative study was a single-arm, open-label, phase II study of 177 people with relapsed or refractory chronic lymphocytic leukaemia (median follow-up 28 months) treated at the MD Anderson Cancer Centre (MDACC). Of the people in the study, 82% had previously received treatment with fludarabine monotherapy or combination therapy (of whom 108 people were fludarabine sensitive and 37 were fludarabine refractory) and 18% had received prior alkylating agents only. Twenty-two of the 177 people in the study had received rituximab monotherapy or combination therapy. The overall response rate for all people in the study was 73% and the complete response rate was 25%.The overall and complete response rates were 58% and 6% respectively for the group with fludarabine-refractory chronic lymphocytic leukaemia compared with 77% and 33% for the group with fludarabine-sensitive disease. For the group who had previously received rituximab monotherapy or combination therapy, the overall response rate was 64% and the complete response rate was 18%.
3.10 During consultation on the appraisal consultation document, the manufacturer provided new data from the MDACC study described above. This included longer term results for a total of 284 people who received rituximab plus fludarabine and cyclophosphamide after previous treatment for chronic lymphocytic leukaemia. One hundred of these people had previously received a rituximab-containing regimen, which may have been rituximab monotherapy or one of various rituximab combination regimens. The number of people receiving each type of treatment was not reported. The overall response rate for all people in the study was 75% and the complete response rate was 31%. For people with fludarabine-refractory chronic lymphocytic leukaemia, the overall and complete response rates were 57% and 8% respectively compared with 80% and 36% for people with fludarabine-sensitive disease. For people who had previously received rituximab, the overall and complete response rates were 73% and 32% respectively compared with 76% and 30% for people who had not previously received rituximab. There was no difference in progression-free survival between people who had previously received rituximab and those who had not (hazard ratio 1.13, p = 0.431).
3.11 The manufacturer submitted an economic analysis comparing rituximab plus fludarabine and cyclophosphamide with fludarabine and cyclophosphamide. The manufacturer used a three-state Markov model with a cycle length of 1 month and a 25-year time horizon (to represent a lifetime horizon). The health states in the model were 'progression-free survival', 'progressed', and 'death'. People entered the model in the progression-free survival health state. The probability of transition from the progression-free survival to the progressed health state was taken from the groups in the REACH trial. For the transition from the progression-free survival to the death health state, trial data were used and supplemented with Office of National Statistics data to inform the background mortality rate. Transition from the progressed to the progression-free survival health state was not possible. For the transition from the progressed to the death health state, people from both groups of the trial were assumed to have equal risk of death. This assumption was based on a non-significant (p = 0.5596) difference in Kaplan–Meier curves for post-progression survival.
3.12 In the model, the drug costs were calculated assuming a body surface area of 1.86 m2, which reflected the average body surface area of the people in the REACH trial. The REACH trial used fludarabine and cyclophosphamide administered intravenously, but it is more common to use oral chemotherapy in the UK. In the model it was assumed that the efficacy of fludarabine and cyclophosphamide was the same regardless of the route of administration if the dosage was adjusted to ensure equivalent bioavailability. The costs of fludarabine and cyclophosphamide treatment in the model were adjusted to allow for the difference in the route of administration. In the base case, the drug doses and costs were reduced according to the proportion of people expected to progress or die each month. The average undiscounted drug cost for rituximab was £9078 for all six cycles of treatment. The average undiscounted drug costs of fludarabine were £2569 for people in the rituximab plus fludarabine and cyclophosphamide group and £2510 for people in the fludarabine and cyclophosphamide group. The average undiscounted drug costs of cyclophosphamide were calculated as £21 and £20 for each group respectively.
3.13 The model included costs for supportive care. Supportive care consisted of quarterly outpatient consultations, blood transfusions and bone marrow transplants in the progression-free survival health state and monthly outpatient consultations and second-line therapies for the progressed health state. The cost for intravenous administration of rituximab was £307 per cycle of treatment and the cost for an appointment to prescribe oral fludarabine and cyclophosphamide chemotherapy was £201. It was assumed that oral chemotherapy could be prescribed in the same appointment as rituximab so no additional cost of prescribing oral chemotherapy was included for the rituximab treatment group. Costs were also added for the pharmacist's time to prepare the infusion and one consultation with a clinical oncologist.
3.14 The utility values used in the manufacturer's submission were taken from a health technology assessment report that assessed the cost effectiveness of fludarabine as a first-line treatment for chronic lymphocytic leukaemia. A utility value of 0.8 was attached to the progression-free survival health state and 0.6 to the progressed health state. The estimates of utility were not preference based, and were estimated by the authors of the report from condition-specific health-related quality-of-life data. No disutility for adverse events was included in the model. The manufacturer provided an interim analysis of 34 people from an observational study of utility in people with chronic lymphocytic leukaemia. The value for the progression-free survival health state was consistent with that used in the manufacturer's submission. No conclusions could be drawn about the utility value appropriate for the progressed health state because only data for two people were available.
3.15 The manufacturer provided a base-case estimate of incremental cost effectiveness of rituximab plus fludarabine and cyclophosphamide compared with fludarabine and cyclophosphamide. The incremental quality-adjusted life year (QALY) gain was 0.585 at an incremental cost of £9128, giving an incremental cost-effectiveness ratio (ICER) of £15,593 per QALY gained. The probabilistic sensitivity analysis presented suggested that rituximab plus fludarabine and cyclophosphamide had a 75% probability of being cost effective at £20,000 and a 94% probability of being cost effective at £30,000 when compared with fludarabine and cyclophosphamide.
3.16 During consultation on the appraisal consultation document, the manufacturer provided an estimate of the cost effectiveness of rituximab plus fludarabine and cyclophosphamide compared with fludarabine and cyclophosphamide, for the subgroup of people who had previously received rituximab. This was calculated using the same model described in section 3.11 with an adjustment to the progression-free survival in the rituximab plus fludarabine and cyclophosphamide group using the hazard ratio (1.13) estimated from the MDACC study. This adjustment to the model resulted in a QALY gain of 0.406 at an incremental cost of £9134, giving an ICER of £22,519 per QALY gained.
3.17 A sensitivity analysis was presented in the manufacturer's submission using different parametric models for the progression-free survival extrapolation. Additional sensitivity analyses were completed as follows:
increasing and decreasing adverse event costs by 50%
increasing and decreasing supportive care costs for the health states by 50%
assuming utility values for the health states such that the difference in the values between the health states was 0.4 and 0.1
assuming upper and lower quartiles for drug administration costs (from reference costs 2007/08)
assuming differential probabilities of death after progression between treatment arms.
One-way sensitivity analyses suggested that the results were not sensitive to a variety of parameter assumptions including adverse events costs, monthly supportive care costs, and drug administration costs. The results were also not sensitive to the function used to extrapolate progression-free survival. The results were sensitive to assumptions about utilities and assumptions about the probability of death after progression. The highest ICER reported (using both differential mortality rates between treatment arms and adjusting utilities) was £23,790 per QALY gained.
3.18 The manufacturer's submission also included a scenario analysis to explore the impact on the ICER of combining rituximab with chemotherapy other than fludarabine and cyclophosphamide. The results of this analysis suggested that the QALY gain from combining rituximab with chemotherapy would need to decrease to about 45% of that in the base case, all else remaining the same, for the ICER for rituximab to increase to over £30,000 per QALY gained.
3.19 The ERG considered that all the relevant trials had been identified. The ERG noted that the manufacturer's submission was based on only one clinical trial, and this trial was unpublished. The ERG considered this trial had adequate randomisation and allocation concealment. However, it noted that the trial was open label and therefore assessments might be biased. The ERG noted that an independent assessment of response was made during a pre-planned interim analysis of the trial data (conducted when about two thirds of the total 284 events had occurred). It stated that there were differences in progression-free survival between the trial groups when assessed by the blinded independent panel and the unblinded trial investigators (independent panel data were provided as academic-in-confidence). The ERG considered that the trial population was relatively young compared with the UK population who would be eligible for rituximab and 10% of people had mild stage disease (Binet stage A), a stage at which people were not commonly treated in the UK. The ERG also noted that people with fludarabine-refractory chronic lymphocytic leukaemia were excluded from the trial although they could be eligible for rituximab. It considered that the comparator used in the cost-effectiveness analysis (that is, fludarabine and cyclophosphamide) was appropriate.
3.20 The ERG noted that in the manufacturer's economic model people in the progressed health state could not move back into the progression-free survival health state. They considered that this did not appropriately reflect the disease process because people with chronic lymphocytic leukaemia receive a series of treatments and therefore they may have periods of progression-free survival after relapse and further treatment. The ERG commented that not all adverse events were assigned costs in the model. In particular, hepatitis B, for which there were six cases in the rituximab plus fludarabine and cyclophosphamide group and no cases in the fludarabine and cyclophosphamide group.
3.21 The ERG completed a series of exploratory analyses. It remodelled rituximab costs so that full costs were incurred at the start of each cycle rather than spread throughout the cycle. This amendment increased the base-case analysis from £15,593 to £18,129 per QALY gained. The ICER of £18,129 was corrected to £16,607 per QALY gained during consultation on the ACD. The ERG conducted an analysis using progression-free survival curves based on the independent assessment of progression (from the interim trial analysis) rather than non-blinded, investigator-assessed progression. This increased the base-case ICER to £17,507 per QALY gained. The ERG also explored the effect on the ICER of assuming no overall survival benefit of treatment with rituximab plus fludarabine and cyclophosphamide. It used two methods for this; it used the mortality rate from the fludarabine and cyclophosphamide group and applied it to the rituximab plus fludarabine and cyclophosphamide group and vice versa. The resulting ICERs were £40,568 and £42,444 per QALY gained for each method respectively compared with £15,593 per QALY gained in the manufacturer's base case.
3.22 The ERG identified that if it is assumed there is no difference in overall survival between the rituximab plus fludarabine and cyclophosphamide and fludarabine and cyclophosphamide groups, the model outputs become sensitive to the assumed utility differences between the progression-free and the progressed health states. If the difference in utility between the health states is decreased by 0.1 (that is from a difference of 0.2 to 0.1), the ICER increases to between £81,135 and £84,889 per QALY gained.
3.23 Full details of all the evidence are in the manufacturer's submission and the ERG report.