Rituximab for the first-line maintenance treatment of follicular non-Hodgkin's lymphoma

The Committee considered the data presented by the manufacturer on the clinical effectiveness of rituximab maintenance treatment after first-line induction with rituximab plus chemotherapy. The Committee noted that the manufacturer derived data on efficacy primarily from the PRIMA trial that compared rituximab maintenance with observation in people whose disease had responded to first-line induction therapy. The Committee noted that the most recent data from this trial were available from the post‑study observational follow‑up period, which had a median follow-up of 38 months, and indicated that progression-free survival was statistically significantly improved in people who had been randomised to rituximab maintenance treatment compared with people who had been randomised to observation. The Committee was aware that the manufacturer could not estimate time to progression for patients randomised to rituximab, because too few people had progressed during the trial, and the manufacturer therefore had to extrapolate this value using a statistical distribution. The Committee noted the concerns of the ERG that because progression-free survival had been estimated from the period after the end of the trial, patients may have received other therapies, which in turn could have affected the chance of disease progression. The Committee also noted that despite following patients beyond the end of the trial, the manufacturer could not estimate the overall survival associated with rituximab maintenance treatment because of the small number of deaths during this period. The Committee was aware that the trial stopped earlier than originally planned on advice from a Data and Safety Monitoring Committee (section 3.3), and heard from the ERG that there is evidence suggesting that studies that have stopped earlier than planned often overestimate the clinical benefit. However, the Committee was satisfied, after advice from the clinical specialists, that progression-free survival for people treated with rituximab maintenance therapy in the PRIMA trial reflected the clinicians' observations from clinical practice. The Committee concluded that the available evidence shows that first‑line maintenance treatment with rituximab significantly improves progression-free survival compared with observation (36 months' median PFS: 74.9% versus 57.6% respectively; HR 0.55; p<0.0001), but that the size of the overall survival benefit could not be determined.

The Committee was aware that the PRIMA trial was the only trial that directly addressed the decision problem, and included the relevant population, intervention, comparison and outcomes. The Committee heard from the clinical specialists that the results from the PRIMA trial inform clinical practice in the UK. The Committee learned from the manufacturer that another trial (ECOG 1496) demonstrated that rituximab maintenance led to longer progression-free survival compared with observation, but that this trial had been conducted in people who had not previously had first-line induction treatment with rituximab-containing chemotherapy. The Committee understood that another trial (SAKK 35/98) had observed a longer event-free survival for people randomised to rituximab maintenance compared with those who had been randomised to no treatment (observation), but that the participants of this trial had either not been treated before rituximab maintenance or had only received an induction regimen with rituximab monotherapy, not rituximab combined with chemotherapy.

The Committee considered the adverse-event profile of rituximab. It noted that the incidence of grade 3 or 4 adverse events was significantly higher in the rituximab maintenance arm than in the observation arm of the PRIMA trial (section 3.6). However, the Committee heard from the clinical specialists and patient experts that rituximab maintenance treatment is generally well tolerated and that adverse events are easily managed. The patient experts also considered the adverse effects associated with rituximab maintenance therapy to be less severe than those experienced with chemotherapy on relapse of disease. The Committee concluded that, overall, most adverse events associated with rituximab treatment are not severe, and that using rituximab to extend remission may delay the need for chemotherapy and, in turn, delay the associated adverse events.

The Committee reviewed the original and revised economic analyses provided by the manufacturer and the exploratory sensitivity analyses performed by the ERG. It heard from the ERG that inconsistencies and errors were identified in the manufacturer's revised model, but that correcting them had only a small effect on the manufacturer's base-case results. The Committee noted that in the PRIMA trial the median age at randomisation was 57 years. However, it heard from the clinical specialists that the mean age at the start of first-line treatment in the UK is usually between 60 and 65 years. The Committee acknowledged that people in clinical trials tend to be younger and fitter than those in clinical practice, but it noted from sensitivity analyses conducted by the ERG that the ICER varied depending on the age assumed at the start of treatment. Therefore, the Committee considered that the average age of people with advanced non-Hodgkin's lymphoma seen in UK clinical practice should be used in the analysis to provide a more accurate estimate. The Committee considered the revised base-case ICER from the manufacturer of £15,400 per QALY gained, which assumed that the mean age at induction was 62.5 years. The Committee heard from the ERG that the method that the manufacturer had used to adjust for age in its economic model did not reflect the prognostic importance of incident age. The Committee considered exploratory sensitivity analyses from the ERG in which the hazard ratios for progression-free survival were adjusted for specific patient ages to reflect a reduction in clinical effect with the increase of age. The Committee noted that age was not the only variable that had an impact on prognosis and, therefore considered that the ERG's adjustment of the hazard ratios for different age groups was not needed. The Committee was satisfied that the manufacturer's base-case analysis had appropriately adjusted for age and reflected the average patient population seen in UK clinical practice.

The Committee noted that the manufacturer had assumed in the base case that the clinical benefit of rituximab maintenance would last for 6 years (2 years of treatment and 4 years of sustained benefit once treatment was stopped). The Committee heard from the ERG that the manufacturer's extrapolation of the clinical benefit of rituximab beyond the period observed in the PRIMA trial assumed a proportional increase in survival with time, which may not reflect the true effect. The Committee also noted the ERG's concerns that patient-level data from the PRIMA trial indicated that the duration of effect from rituximab maintenance treatment appears to be 28 months, after which time patients treated with rituximab maintenance therapy experience a rate of progression no better or worse than that of patients not treated with rituximab maintenance therapy. The Committee heard from the clinical specialists that data from the PRIMA trial demonstrated that rituximab maintenance treatment is clinically effective to 36 months at least and without evidence that the effect diminishes over time; therefore, assuming a duration of benefit of only 28 months, as suggested by the ERG, may underestimate the actual effect of treatment. The Committee also heard from the clinical specialists that the period over which rituximab is likely to have an additional benefit over observation is probably 3 to 4 years (that is, 1 to 2 years beyond treatment). However, it further heard from the clinical specialists that it was not possible to predict a definite time period, and a duration of effect of up to 6 years, as seen in the EORTC 20981 study for second-line rituximab maintenance treatment, could be plausible. The Committee considered sensitivity analyses conducted by the manufacturer that assumed a duration of treatment effect of 28 months, 36 months and 48 months and noted that the ICERs ranged from £17,300 to £27,400 per QALY gained. The Committee noted that these estimates were lower than those calculated by the ERG for the same scenarios (range: £24,600 to £43,900 per QALY gained) but acknowledged that the manufacturer and the ERG had used different modelling approaches to calculate their results (section 3.29). The Committee considered that the duration of clinical benefit of rituximab maintenance was a key driver of cost effectiveness, but was satisfied that the manufacturer's sensitivity analyses presented the most plausible range of estimates for the treatment effect in line with clinical opinion and the available data.

The Committee noted that the manufacturer's revised base-case analysis assumed that most (89.2%) of the progression-free survival benefit translated to an overall survival gain in its model. The Committee heard from the clinical specialists that it was not possible to verify the specific conversion rate from progression-free survival to overall survival from the literature or clinical experience, but that they would expect a conversion rate of at least 70%. The Committee also heard from the clinical specialists that patients with non-Hodgkin's follicular lymphoma live longer than in the past, and it is reasonable to assume that this is at least partly due to the introduction of treatment with rituximab. The Committee considered that the manufacturer should have sought data from patient registries or observational data to validate the conversion rate assumed for the base-case estimate, and to confirm the degree to which rituximab maintenance treatment might prolong life. However, it was satisfied that the manufacturer's sensitivity analyses, which assumed conversion rates of 70%, 80% and 90%, provided a plausible range of conversion rate estimates. The Committee heard from the manufacturer that the conversion rate was not an actual input in the model and could only be adjusted by artificially modifying other parameters. As such, the manufacturer was concerned that its revised analyses, which were requested by the Committee, were driven by implausible assumptions. The Committee noted the manufacturer's concerns but was satisfied that the sensitivity analyses addressed the uncertainty that the Committee initially had about the translation from progression-free survival to overall survival gain in the original analysis.

The Committee considered the concerns of the ERG that the early closure of the PRIMA trial may have overestimated the benefit from rituximab, and therefore the hazard ratio for progression-free survival (0.55) derived from the PRIMA trial should be increased to adjust for this bias. The Committee noted the revised sensitivity analyses from the ERG, which took account of the adjusted hazard ratio, but considered that adjusting for early reporting bias is not routinely included in technology appraisals and is not a current requirement in the NICE methods guide. The Committee therefore concluded that the manufacturer had used an appropriate hazard ratio and that the ERG's revised analyses using the higher hazard ratio would not be considered.

The Committee discussed the utility values used in the manufacturer's model. The Committee appreciated that no differences in health-related quality of life were observed between the arms of the PRIMA trial. The Committee considered sensitivity analyses from the ERG that showed that changes to the gains in utility in different health states in the model had a marginal effect on the base-case ICER, and the Committee was therefore persuaded that the ICERs presented by the manufacturer were largely driven by gains in overall survival.

The Committee noted that the ICERs for rituximab maintenance compared with observation in the manufacturer's submission and sensitivity analyses were less than £30,000 per QALY gained for most scenarios. The Committee also noted that the ERG's exploratory sensitivity analyses, which assumed a duration of clinical benefit from rituximab maintenance treatment of 36 to 48 months (in line with clinical opinion), resulted in ICERs ranging from £24,600 to £35,000 per QALY gained, depending on the conversion rate of progression-free survival to overall survival gain assumed. The Committee was aware that the model did not include the utility associated with delaying chemotherapy, and that if it were included, it would decrease the ICER (that is, improve the cost effectiveness) to an estimate which would be considered as a cost-effective use of NHS resources. Therefore, the Committee considered that rituximab maintenance therapy should be recommended as an option for the treatment of people with follicular non-Hodgkin's lymphoma that has responded to first-line induction treatment with rituximab in combination with chemotherapy.