Idelalisib for treating refractory follicular lymphoma

This appraisal focuses on idelalisib as a third-line treatment

3.1 The final NICE scope specified a population with follicular lymphoma that is refractory to 2 previous lines of therapy, that is, third-line treatment. This is in line with the marketing authorisation and the company's submission. The population considered in the appraisal is narrower than the population specified by the marketing authorisation. It relates to a 'double-refractory' population (that is, the follicular lymphoma did not respond, or only showed a limited response to at least 2 previous treatments) and the previous treatment had to include an anti‑CD20 monoclonal antibody (rituximab or obinutuzumab) and chemotherapy containing an alkylating agent (for example, cyclophosphamide). Clinical experts considered this 'double-refractory' population to have an unmet need for treatment options. The committee agreed that the population presented in the company's submission was appropriate.

There is no single standard of care chemotherapy for the third-line treatment of follicular lymphoma

3.2 In the population who would be offered idelalisib (see section 3.1), most people are currently offered chemotherapy including (as listed in the NICE scope) cyclophosphamide- containing regimens, fludarabine-containing regimens, bendamustine and chlorambucil. The clinical experts explained that these vary in effectiveness and toxicity. The choice of chemotherapy regimen depends on individual circumstances, and takes into account previous chemotherapy use and clinician preference. Because of this, there is no single standard of care chemotherapy regimen in this population. The committee concluded that all the chemotherapeutic regimens listed in the scope were appropriate comparators.

Best supportive care is also a comparator for idelalisib

3.3 The clinical experts stated that, at third line, most people would currently be offered chemotherapy, but those who could not have chemotherapy would be offered best supportive care instead. The clinical experts stated that some of those people could take idelalisib because it has a different toxicity profile to chemotherapy. So, while people may not tolerate the specific adverse effects of chemotherapy, they may tolerate those of idelalisib. The committee concluded that best supportive care is an option for people who cannot have chemotherapy.

Stem cell transplantation is used at second and subsequent lines

3.4 The clinical experts stated that response to second-line treatment is consolidated with autologous stem cell transplantation in people who are considered fit enough, in line with NICE's guideline on non-Hodgkin's lymphoma: diagnosis and management. They also noted that autologous stem cell transplantation may be offered at later lines of therapy in people whose disease responded well to treatment. The clinical experts considered stem cell transplantation to be an important prognostic factor and that it could be used after idelalisib. The committee concluded that it was possible for idelalisib to be used as a bridge to stem cell transplantation, if there was a sufficient response to idelalisib.

The key clinical evidence comes from the DELTA study

3.5 The key clinical evidence for idelalisib came from the single-arm phase II DELTA study, which included 125 patients with indolent (slowly growing) non-Hodgkin's lymphoma, 72 of whom had follicular lymphoma refractory to 2 or more lines of treatment. The primary outcome measure was overall response rate. Overall survival and progression-free survival were among the secondary outcomes. Median progression-free survival was 11.0 months; median overall survival was not reached, but the company estimated it at 38.1 months. The company supplemented the DELTA study with 2 other sources of evidence for both idelalisib and the comparators:

It is unclear whether the DELTA population or the CUP cohort more closely reflects clinical practice

3.6 The committee discussed the generalisability of the 2 study populations to the decision problem:

'Self-control' comparisons with previous lines of therapy do not give reliable comparator information

3.7 In the absence of a head-to-head randomised comparison of idelalisib with chemotherapy, the company compared progression and death on idelalisib with the last line of chemotherapy before idelalisib, for each study. Therefore, the company used the previous line of chemotherapy as a proxy for chemotherapy at the next line of treatment (that is, the point at which idelalisib would be used). It got these data from DELTA and the CUP. The committee discussed several issues around this:

The committee recognised that this 'self-control' comparison did not compare the same patients with each other because the company had not done a paired matched analysis. Instead, it compared one population (having chemotherapy) with itself after patients had progressed and survived to have another treatment (having idelalisib). It concluded that there were multiple sources of bias in this type of analysis, and that the overall direction of bias and effect on relative effectiveness could not be determined.

The results from DELTA and the CUP are inconsistent

3.8 The committee noted that comparing idelalisib with previous chemotherapy gave inconsistent results in DELTA and the CUP. The results from DELTA suggested an increase in progression-free survival compared with last previous therapy, whereas the results from the CUP showed similar progression-free survival compared with last previous therapy. The clinical experts commented that this might be because of how progression was determined and how relapse was identified, given the indolent nature of the disease. They also noted that the CUP, being a 'real-world' study, was likely to have defined and assessed disease progression more loosely than DELTA. The committee agreed that, although this approach addressed the issue of defining standard of care, (see section 3.2), the results were associated with a high degree of bias and did not give reliable comparator information.

The matching adjusted indirect comparison with the HMRN gives an estimation of the effect of chemotherapy in the DELTA study population

3.9 In addition to the 'self-control' comparison, the company presented an unanchored matching adjusted indirect comparison (MAIC) with the HMRN cohort (see section 3.5). The company matched individual patient-level data from the HMRN cohort to the population-level descriptive characteristics of patients in the DELTA study. This estimated the effects of chemotherapy in the DELTA population. The ERG preferred estimating the effect of idelalisib in the HMRN (UK) population because this gave a larger source of individual patient data and a more robust dataset to the adjustments done in the MAIC. The committee agreed that there was merit in matching the DELTA population to the HMRN cohort to explore uncertainty and give a better representation of clinical practice. It also considered that, because patient-level data were available from both DELTA and the HMRN cohort, the company could develop a propensity score. The committee also recognised that without sufficient information on potential confounders, this method could be biased.

The MAIC does not adjust appropriately for potential confounders

3.10 The committee discussed what factors most affect prognosis and therefore should have been included in the MAIC analysis. The committee was aware it had not been presented with a systematic review of risk factors for progression and death to inform the MAIC. In addition to history of autologous stem cell transplantation (see section 3.4), the committee understood that there are other factors associated with progression and death. These include, but are not limited to, the components of the Follicular Lymphoma International Prognostic Index I and II (FLIPI and FLIPI2), notably age, serum beta 2 microglobulin levels, bone marrow involvement, size of the largest involved lymph node, and haemoglobin levels. Other factors include time in previous remission, time since completing the last therapy, comorbid conditions, and which chemotherapeutic agents the patient has had previously. The clinical experts suggested that the FLIPI index is the best validated prognostic tool, but that it is has limited value at third line. They proposed that a key prognostic indicator would be response to previous therapy, but this input was not captured in the variables chosen in the MAIC. The company matched 5 of 7 variables that it assumed accounted for all prognostic factors and treatment-effect modifiers. The committee was aware that a technical support document published by the Decision Support Unit recommends that, when only single-arm trial data are available, all the characteristics that could influence the outcomes of interest should be adjusted. However, increasing the number of matched characteristics reduced the effective sample size and hence the precision of the estimates. For example, when the company removed the variable 'median time since diagnosis' from the adjustment, estimated 2‑year overall survival reduced by more than 20%. The committee also appreciated that, in any case, there would be unobserved differences between study populations that would not be accounted for. It noted that this would bias the estimates of relative effectiveness if associated with progression or death. The ERG commented that the MAIC should be treated with caution because the equivalent sample size was very small (n=6.9), the adjustment was very sensitive to variable selection and there was no sensitivity analysis done despite these uncertainties. The committee concluded that, based on the company's MAIC, it could not ascertain whether the estimated effect of idelalisib compared with chemotherapy reflected what would be seen in clinical practice.

The 'blended' comparator assumes that different chemotherapeutic treatments are similarly effective but this assumption is not justified

3.11 The committee was aware that the NICE scope listed the chemotherapeutic agents separately, and had heard from clinical experts that the therapies are likely to differ in effectiveness and tolerability. It recognised that the company had combined treatments together, which reflected a 'blended' comparator, and which NICE's guide to the methods of technology appraisal discourages. The committee considered that evidence for the effectiveness of separate chemotherapeutic agents might exist from the trials that provided the evidence for the use of chemotherapy, or from registries other than the HMRN, but also considered that the evidence may be difficult to source. The committee concluded that it had not been presented with evidence that justified the assumption that chemotherapeutic agents could be considered similarly effective to one another and in line with current UK practice.

There are no data for best supportive care

3.12 The committee recognised that the company did not provide a source of data to describe the natural history of disease in patients having best supportive care. The committee asked the company whether the HMRN cohort could provide this, or whether clinical trial data were available from control arms of clinical trials of chemotherapy. The committee concluded that the company had not explored this, but that this would be needed to make a comparison with best supportive care.

The current analyses do not represent a reliable source of comparator information

3.13 The committee considered that there was high degree of clinical uncertainty in the evidence base because the company had not provided direct comparative evidence and the 2 analyses comparing idelalisib with standard of care were not suitable for decision-making. The committee queried the company at the meeting as to whether other observational data for idelalisib exists globally, and understood that such data might exist from CUPs in other countries. The committee discussed how the evidence could be improved by:

The model structures are appropriate for modelling

3.14 The company presented 4 economic analyses using different model structures and sources of clinical data, which are summarised in table 1.

Table 1 Summary of comparisons used in the economic modelling

Comparisons A, C and D reflected state transition models in which the movements between states (clinical events) are governed by 'transition probabilities', and expected costs and quality-adjusted life years (QALYs) are estimated by assigning values to the time spent by patients in each state. Comparison B used the more conventional 'partitioned survival analysis' approach in which the proportion of patients in each state at each time point is determined from a set of survival curves, rather than transition probabilities. For this approach, an overall survival curve is needed to estimate the proportion of people alive over time directly. This was not available in comparisons A and C because the comparator data came from the 'self-control' comparisons that, by definition, excluded patients who had died. The committee concluded that it was important to bear in mind the different modelling approaches used when comparing the resulting cost-effectiveness estimates.

One key area of uncertainty in the economic models is how previous line chemotherapy effectiveness was adjusted using a 'hazard ratio'

3.15 In comparisons A and C, which used 'self-control' data, the company applied a 'hazard ratio' to adjust for the expected decline in the effectiveness of chemotherapy later in the treatment pathway (see section 3.7). Specifically, it estimated that effect declines by 25%. The ERG commented that it could not verify the data source underpinning this assumption. The committee recalled that a range of chemotherapy regimens was being used and these varied in effectiveness (see section 3.1). This meant that applying a single estimate reflecting the decrease in treatment effect across the whole range of regimens was unlikely to be reliable. Furthermore, the hazard ratio adjustment affected the cost-effectiveness estimate for comparison C to a much greater extent than comparison A, and the committee questioned why this was the case. The ERG commented that, when comparing progression-free survival for those having idelalisib against an unadjusted previous line of therapy, DELTA had a much greater difference than the CUP. Therefore, the addition of a hazard ratio had a greater effect on the CUP cost-effectiveness estimates. The committee concluded that using the hazard ratio introduced additional uncertainty to the model that needs to be explored in sensitivity analyses.

The different modelling structures influence which clinical inputs drive the modelled survival differences and therefore the cost effectiveness

3.16 The committee agreed that it would focus on comparisons A (company's base case) and B. It was not presented with a clear rationale for using both DELTA and CUP sources of data without adjustment for using different populations to inform the clinical inputs of comparison C. It noted that the company did not provide best supportive care data for comparison D (see section 3.12). However, it agreed that it could use comparisons C and D to evaluate its preferred analyses:

The model estimates of survival are uncertain

3.17 The committee considered the survival estimates produced by the different comparisons:

The committee agreed that the survival modelling reflected a high degree of uncertainty in the evidence. Given the lack of direct comparator data, the absence of evidence for individual chemotherapeutic regimens or for best supportive care, and the paucity and immaturity of observed data, the committee considered that it was important that the survival estimates aligned as closely as possible with whatever data were observed. In the current analyses, the Markov model (comparisons A, C and D) consistently overestimated the effect of idelalisib and the approach (B) using partitioned survival analysis appeared to have markedly underestimated the effect of chemotherapy. Therefore, the committee concluded that the model predictions were not reliable for decision-making.

The 'blended comparator' masks ineffective treatments

3.18 The committee considered that using a blend of chemotherapeutic regimens as a comparator meant averaging the cost effectiveness of the treatments included, and potentially masked cost-ineffective individual treatments. The committee concluded that the cost effectiveness of idelalisib therefore needed to be considered against each chemotherapy treatment individually.

Utility values based on the DELTA study should be explored

3.19 The company presented utility values to reflect health-related quality of life from a published study and the ERG presented scenario analyses with alternative utility values from other published studies. The clinical experts agreed that all values seemed reasonable because people with follicular lymphoma can expect a high utility value once the bulk of the disease has decreased. The ERG questioned why the company did not use health-related quality-of-life data (FACT‑G instrument) from DELTA. The company responded that there were no mapping algorithms available for the target population and that the FACT‑G instrument does not capture key elements of quality of life in patients with follicular lymphoma. The committee concluded that, nevertheless, this analysis would be useful as validation of the chosen utility inputs.

Adverse effects with idelalisib are different than those with chemotherapy

3.20 The committee noted the potential serious adverse effects in people having idelalisib outlined in the European Medicine Agency's Risk Management plan, which include severe drug-related colitis, pneumonitis and organising pneumonia, and serious infections. The company did not provide relative safety data for chemotherapy. The clinical experts commented that adverse events with chemotherapy are qualitatively different to those with idelalisib, so it is difficult to comment on the comparative safety profiles. The committee considered that the adverse effects of chemotherapy are unlikely to differ between haematological malignancies. The company included adverse events as disutilities within all the economic models, these disutility values were based on various sources and the incidence was estimated using data from DELTA. The company models assumed the same incidence of disutilities for both idelalisib and chemotherapy. The clinical experts commented that idelalisib and chemotherapies will all have different toxicities, but idelalisib has generally been well tolerated within its risk management plan. The committee noted that adverse event disutility is not a key driver in the model. It concluded that idelalisib is unlikely to have the same adverse event disutility as the average disutility for possible chemotherapy treatments, but is unlikely to markedly change the outcomes from the model.

There are no plausible cost-effectiveness estimates

3.21 The company presented a base-case incremental cost-effectiveness ratio (ICER) for comparison A of £26,076 per QALY gained. However, this increased to £32,882 per QALY gained with the ERG's corrections to the model. The company also presented an ICER of £47,011 per QALY gained for comparison C, which increased to £58,754 per QALY gained with the ERG's corrections. Because of the substantial uncertainty around the clinical and cost inputs to the model, the committee considered that none of these estimates were valid. Furthermore, all ICERs presented were deterministic, which does not reflect the uncertainty in the evidence base. The committee also recognised, in line with NICE's guide to the methods of technology appraisal, that given the high level of uncertainty associated with the analyses it would be inappropriate to consider idelalisib to be a cost-effective use of NHS resources unless the most plausible ICER values were towards the lower end of the maximum acceptable ICER range (that is, £20,000 per QALY gained). The committee concluded that none of the ICERs presented, which ranged between £16,855 and £95,120 per QALY gained, were reliable or accurate estimates, so idelalisib could not be considered cost effective.

The current analyses do not represent a reliable source of cost effectiveness but could be improved

3.22 The committee agreed that the key uncertainties in the model came from the modelling of survival outcomes, namely:

To improve the quality of the evidence and further characterise the uncertainty, the committee agreed that additional analyses around the key model parameters should be presented. These could include:

The committee concluded there was not enough information about end-of-life considerations

3.23 The committee considered the advice about life-extending treatments for people with a short life expectancy in NICE's guide to the methods of technology appraisal:

Idelalisib is not a candidate for the Cancer Drugs Fund

3.24 Having concluded that idelalisib could not be recommended for routine use, the committee then considered whether it could be recommended for treating follicular lymphoma within the Cancer Drugs Fund. The committee discussed the arrangements for the Cancer Drugs Fund agreed by NICE and NHS England in 2016, noting NICE's Cancer Drugs Fund methods guide (addendum). It also noted that the company had not expressed an interest in providing idelalisib through the Cancer Drugs Fund. The company confirmed that there are no ongoing comparative trials that will be able to provide more robust, controlled evidence. The committee appreciated that the Cancer Drugs Fund is designed to resolve uncertainties, and that the key uncertainty in this appraisal were the assumptions surrounding the effectiveness and adverse effects associated with chemotherapy. It was aware that the Cancer Drugs Fund could collect data on the intervention (idelalisib), but not the comparator (chemotherapy), so would not be able to resolve this uncertainty. Also, the committee could not determine whether idelalisib had 'plausible potential' to be cost effective (see section 3.21). It concluded that idelalisib could not be recommended for use within the Cancer Drugs Fund as an option for follicular lymphoma that had not responded to 2 previous lines of therapy in adults.

It is not clear if idelalisib is innovative

3.25 The committee noted that idelalisib is innovative in that it has a different mechanism of action to other available treatments for 'double-refractory' follicular lymphoma, and addresses an area of unmet clinical need (see section 3.1). It concluded that the model captured all quality-of-life improvements, including its oral route of administration. However, the committee could not determine whether it reflected a step change in treatment, given the uncertain estimates of comparative effectiveness.

Idelalisib is not recommended

3.26 The committee was not presented with enough reliable information to be able to address the decision problem. Therefore, it could not recommend idelalisib for the routine treatment of follicular lymphoma that has not responded to 2 previous lines of treatment in adults.