Rituximab for aggressive non-Hodgkin's lymphoma

4.1 Clinical effectiveness

4.1.1 The Assessment Group found a single randomised controlled trial comparing the combination of rituximab plus CHOP (R-CHOP) with CHOP alone in people with stage II, III or IV DLBCL. The study involved 399 previously untreated people aged 60–80 years who were judged to have good to fair performance status.

4.1.2 Participants were randomly assigned to receive either eight cycles of CHOP every 3 weeks (n = 197) or eight cycles of CHOP plus rituximab at a dosage of 375 mg/m² of body surface area on day 1 of each of the eight cycles of CHOP (n = 202). The CHOP regimen was:

4.1.3 The primary endpoint in the trial was event-free survival, where an event was defined as disease progression, relapse, death or initiation of different treatment. Within the median follow-up period of 24 months, 86 (43%) participants receiving R-CHOP experienced at least one event, as did 120 (61%) of those in the CHOP group. A proportional hazards model showed that the relative risk (RR) of an event with the addition of rituximab to the CHOP regimen was 0.58 (95% confidence interval [CI], 0.44 to 0.77). Adjusting for a number of baseline prognostic factors did little to change this result (RR 0.55; 95% CI, 0.41 to 0.75).

4.1.4 Secondary outcomes were overall survival, response rates and adverse events. Health-related quality of life was not measured in the trial.

4.1.5 Fifty-nine (29%) of the group receiving R-CHOP died within a median of 2 years' follow-up, compared with 81 (41%) of those receiving CHOP alone. The relative risk of death from any cause with the addition of rituximab was 0.64 (95% CI, 0.45 to 0.89). After adjusting for a number of baseline prognostic factors between the treatment groups, this fell to 0.53 (95% CI, 0.37 to 0.77).

4.1.6 Significantly more participants achieved complete response or unconfirmed complete response in the group receiving R-CHOP compared with the group receiving CHOP (76% compared with 63%, p = 0.005). The relative increase in the chance of a complete or unconfirmed complete response was 20% (95% CI, 5 to 37). Complete response was defined as the disappearance of all lesions and of radiological or biological abnormalities observed at diagnosis and the absence of new lesions; unconfirmed complete response was defined as a complete response except for the persistence of some radiological abnormalities that had regressed in size by at least 75%.

4.1.7 The incidence of severe adverse events (grade 3 or 4 according to the US National Cancer Institute Common Terminology Criteria for Adverse Events, plus grade 2 infections) was 74% in the CHOP arm, compared with 79% in the R-CHOP arm (RR 1.08; 95% CI, 0.96 to 1.20).

4.1.8 No comparative data in younger people were identified, but in a small uncontrolled phase II study designed to establish the safety and efficacy of R-CHOP in people with newly diagnosed aggressive lymphoma (n = 33), the response rate to R-CHOP was at least as good in people younger than 60 years of age as in those older than 60 years of age.

4.1.9 Data are lacking on the usefulness of rituximab in people with localised disease.

4.2 Cost effectiveness

4.2.1 One economic evaluation of R-CHOP versus CHOP was supplied by the manufacturer. However, the Assessment Group incorporated a number of different assumptions into the framework of the manufacturer's model as part of the review process.

4.2.2 Both versions of the model included only the costs to the NHS, expressed health benefits in terms of quality-adjusted life-years (QALYs), and used a 15-year time horizon. Both versions of the model also estimated utilities from the same unpublished study. The estimates of the proportion of people achieving complete response and the overall duration of survival in people receiving the CHOP regimen were based on observational data. The estimate of the relative treatment effect for R-CHOP was based on the single randomised controlled trial. Both versions of the model also used these data to estimate cost effectiveness separately for people younger than 60 years of age.

4.2.3 The manufacturer's version of the model produced a cost per life-year gained of approximately £4500 and a cost per quality-adjusted life-year gained (QALY) of £6100 for people aged 60 years and older. For people younger than 60 years, these figures were approximately £4700 and £6800 respectively. Sensitivity analysis showed that these incremental cost-effectiveness ratios (ICERs) were relatively robust to changes in the input assumptions. However, the ICERs approximately doubled when the time horizon was reduced to 5 years.

4.2.4 The Assessment Group's version of the model differed from the manufacturer's mainly in the interpretation of the survival curves for people receiving CHOP or R-CHOP and the inclusion of other costs associated with treatment failure (second-line therapies and palliative care costs). The results for people younger than 60 years were slightly less favourable than those from the manufacturer: approximately £8500 per life-year gained and £7500 per QALY gained. In people aged 60 years and older, the ICERs were less favourable: about £9700 per life-year gained and £10,500 per QALY gained. Extensive sensitivity analyses found these results to be robust to changes in the input assumptions. Probabilistic sensitivity analysis estimated that there was only a 5% chance that the cost per additional QALY would exceed £23,400 in people aged 60 years and older, or £19,000 in people younger than 60 years.

4.2.5 Both versions of the model suggest that rituximab in combination with each of eight cycles of CHOP is cost effective relative to CHOP used alone.

4.3 Consideration of the evidence

4.3.1 The Committee reviewed the data available on the clinical and cost effectiveness of rituximab, having considered evidence on the nature of the condition and the value placed on the benefits of rituximab from people with DLBCL, those who represent them, and clinical experts. It was also mindful of the need to take account of the effective use of NHS resources.

4.3.2 The Committee considered that the evidence reviewed in the Assessment Report – primarily the results of a single randomised trial in people between the ages of 60 and 80 years – supported the case that rituximab in combination with CHOP is clinically and cost effective in people with stage II, III or IV DLBCL.

4.3.3 The Committee considered the use of rituximab in people with stage II, III or IV DLBCL who were younger than 60 years. In the absence of a randomised study in this group, the Committee was persuaded by the evidence from experts that it was biologically plausible that rituximab was likely to have the same effect on the disease process in people younger than 60 years as for those aged 60 years and older. In addition, the Committee was aware that the baseline cost-effectiveness estimates from the Assessment Group's version of the model had assumed the same relative risk of disease progression (0.6) in both age groups but the sensitivity analysis explored a reduced effect (0.8) in the under-60 years age group, thus taking account of possible lesser effectiveness. In considering the clinical effectiveness in this younger age group, the Committee also took into account the evidence from a small phase II study, which supported this view. The Committee members considered that it was appropriate to use the results of the randomised trial in the 60–80 years age group to estimate the clinical and cost effectiveness in people younger than 60 years. The results of the economic evaluations suggest that R-CHOP is likely to be cost effective in both age groups compared with CHOP alone.

4.3.4 Evidence from the experts suggested that CHOP is an intensive course of chemotherapy that is associated with significant toxicity. There was no evidence showing that rituximab is effective in people for whom CHOP is contraindicated. Therefore it was concluded that rituximab should be used in these circumstances only as part of ongoing or new clinical studies.

4.3.5 For people with localised (stage I) disease, the standard treatment is three cycles of CHOP followed by involved-field radiotherapy. The Committee heard from experts that this regimen has been shown to be better than CHOP alone in this patient group, and that it is associated with a 5-year survival of more than 80%. Additionally, there is currently no evidence to support the use of rituximab in people with stage I disease. The Committee concluded that the use of rituximab in this group of patients should therefore be restricted to ongoing or new clinical studies.

4.3.6 There is little directly comparative evidence on the effectiveness of R-CHOP in people with AIDS-related lymphoma. After discussion with experts, the Committee concluded that this group should not be treated differently from other people with stage II, III or IV DLBCL.