Erlotinib for the first-line treatment of locally advanced or metastatic EGFR-TK mutation-positive non-small-cell lung cancer

Clinical practice

4.2 The Committee discussed the clinical need of patients with locally advanced or metastatic EGFR-TK mutation-positive NSCLC. It heard from the clinical specialists that the main aim of treatment is to extend progression-free and overall survival with the fewest adverse reactions and with the best quality of life possible for the remaining months of life. The clinical specialists also highlighted that for this patient population an oral treatment with a tyrosine kinase inhibitor, such as gefitinib or erlotinib, is usually associated with an improved quality of life compared with platinum doublet chemotherapy.

4.3 The Committee heard from the clinical specialists that current UK clinical practice for the first-line treatment of locally advanced or metastatic EGFR-TK mutation-positive NSCLC is to use gefitinib as recommended in NICE technology appraisal guidance 192. The Committee also heard that chemotherapy with pemetrexed plus carboplatin or cisplatin may be used as a second-line treatment and is rarely used as first-line treatment for this patient population. The Committee accepted that gefitinib is current standard practice in England and Wales for the first-line treatment of locally advanced or metastatic EGFR-TK mutation-positive NSCLC.

4.4 The Committee discussed the availability of EGFR testing to inform the first-line treatment of locally advanced or metastatic NSCLC. It heard from the clinical specialists that EGFR testing is standard practice for this patient population across almost all the NHS. The Committee accepted that EGFR testing is standard practice in England and Wales when making decisions about the first-line treatment of locally advanced or metastatic NSCLC.

4.5 The Committee discussed the use of tyrosine kinase inhibitors in clinical practice for the first-line treatment of locally advanced or metastatic EGFR-TK mutation-positive NSCLC. It heard from clinical specialists that in their opinion erlotinib and gefitinib are very similar treatments with similar efficacy and levels of adverse reactions. The clinical specialists highlighted that having the choice of two similar treatments enables better management of adverse reactions. The Committee also heard from the clinical specialists that the adverse reactions associated with both these treatments are much less common than those associated with chemotherapy but may vary (for example, rash may be more common with erlotinib and interstitial lung disease may be more common with gefitinib). Erlotinib offers the advantage of being able to vary the dosage by using tablets of different dose strength. The Committee also heard that the patient access scheme for gefitinib is not straightforward and that hospitals may find the patient access scheme for erlotinib easier to administer. The Committee concluded that further first-line treatment options for patients with locally advanced or metastatic EGFR-TK mutation-positive NSCLC would be valuable for clinical practice.

Clinical effectiveness

4.6 The Committee considered the evidence submitted by the manufacturer on the clinical effectiveness of erlotinib. The Committee agreed with the manufacturer that although pemetrexed plus cisplatin or carboplatin was listed as a comparator in the scope, recent changes in the clinical pathway since the publication of NICE technology appraisal guidance 192 in 2011 have resulted in the use of gefitinib for first-line treatment for most patients with EGFR-TK mutation-positive NSCLC, as confirmed by the clinical specialists (see section 4.3). The Committee concluded that gefitinib is the appropriate comparator for this appraisal.

4.7 The Committee noted that the evidence of clinical effectiveness of erlotinib in locally advanced or metastatic EGFR-TK mutation-positive NSCLC was based on the EURTAC trial with supporting evidence from the OPTIMAL trial. The Committee noted that both trials provided evidence of increased progression-free survival compared with doublet chemotherapy. The Committee agreed that the EURTAC trial provided evidence relevant to clinical practice in the NHS in England and Wales. The Committee concluded that the evidence from the EURTAC trial demonstrated that erlotinib increased progression-free survival compared with doublet chemotherapy.

4.8 The Committee considered the indirect comparison presented by the manufacturer. The hazard ratio for progression-free survival used in the model (0.82, 95% CI 0.54 to 1.26) was obtained by comparing the EURTAC trial with the gefitinib meta-analysis. The Committee noted the wide confidence intervals around the estimated hazard ratio for progression-free survival, but recognised the difficulties in constructing a robust indirect comparison given the limited number of studies in this patient population and the heterogeneity between the studies. The Committee discussed the heterogeneity between the trials and the possible prognostic factors that may have influenced heterogeneity, such as ethnic group and class of mutation (exon 19 deletion compared with mutation in exon 21 L858R). It heard from the clinical specialists that the difference in the response rate in the chemotherapy arms between the EURTAC and OPTIMAL trials was within the acceptable range for this group of patients. The Committee noted that the ERG had pointed out the similarities in the curves for progression-free survival from the EURTAC and OPTIMAL trials and the difference between the results from the IPASS trial and the other gefitinib trials after 12 months of treatment. The Committee heard from the ERG that the gefitinib trials had not been uniformly reported so it was not possible to be certain whether the differences were caused by factors such as differing variables in multivariate analyses or small patient numbers. The Committee also discussed the ERG's comments about the difficulties associated with using the proportional hazards assumption for these data and the possibility of using revised efficacy outcomes. The Committee was not convinced that an indirect comparison could be used with the existing data, to support the assumption that erlotinib was more effective than gefitinib, given the heterogeneity of the populations included and the variations in prognostic factors within the populations. In addition, the Committee noted the clinical specialists' view that erlotinib and gefitinib are very similar treatments with similar efficacy for locally advanced or metastatic EGFR-TK mutation-positive NSCLC (see section 4.5). The Committee concluded that there was insufficient evidence to suggest a difference in clinical effectiveness between erlotinib and gefitinib in the model and therefore the most appropriate value for the hazard ratio for progression-free survival between the treatments is 1.

4.9 The Committee discussed the overall survival data from the trials. It noted that the data for overall survival were incomplete (either not available for all patients or not known) for the EURTAC and OPTIMAL trials and therefore no comparison of overall survival benefit for erlotinib and gefitinib was available. It also noted the ERG's concerns about whether there was an overall survival benefit for treatment with a tyrosine kinase inhibitor compared with doublet chemotherapy in light of the recently published final results from the IPASS trial (gefitinib compared with doublet chemotherapy). Because of the similarities in the treatments and the lack of data on overall survival, the Committee was not convinced of a survival benefit for erlotinib compared with gefitinib for patients with locally advanced or metastatic EGFR-TK mutation-positive NSCLC.

4.10 The Committee considered the adverse reactions experienced by patients receiving treatment for locally advanced or metastatic NSCLC. It noted that data from the EURTAC trial demonstrated that fewer patients in the erlotinib arm experienced grade 3 or 4 events compared with the chemotherapy arm. Low grade skin reactions (rash grade 3, 5%) and diarrhoea (grade 3, 4%) were the most commonly reported adverse reactions associated with erlotinib. The clinical specialists confirmed that the adverse reactions associated with erlotinib and gefitinib were generally modest but slightly different. The Committee concluded that the adverse reactions associated with erlotinib were relatively mild in most patients and that from a clinical perspective there may be some advantage to having a choice of tyrosine kinase inhibitors for this patient group.

4.11 The Committee noted the lack of quality of life data from the EURTAC trial and heard from the clinical specialists that a common problem with studies in this patient population is the failure to complete questionnaires. The Committee was disappointed that there were insufficient quality of life data from the EURTAC trial for analysis. Because erlotinib and gefitinib are both oral tyrosine kinase inhibitors with similar efficacy and comparable adverse reactions, the Committee concluded that the health-related quality of life of patients would be similar for the two treatments.

Cost effectiveness

4.12 The Committee considered the manufacturer's original cost-effectiveness analysis and the ERG's critique. It noted that the manufacturer used a semi-Markov model to evaluate the cost effectiveness of erlotinib compared with gefitinib. The clinical data used in the model were derived mainly from the EURTAC trial and the indirect comparison of data from the EURTAC trial with the gefitinib meta-analysis described by Ku et al. (2011). The Committee was aware of the ERG's concerns that the structure of the model allowed the benefit in progression-free survival to be translated into an overall survival benefit in the economic model. Given the uncertainties associated with the hazard ratio for progression-free survival obtained from the indirect comparison (described in sections 3.22 to 3.25 and 4.8), as well as the lack of evidence demonstrating an overall survival benefit for erlotinib compared with gefitinib (see section 4.9), the Committee concluded that the cost effectiveness of erlotinib compared with gefitinib could be best assessed from an analysis which assumes equal clinical benefit between the treatments and focuses on their differential costs.

4.13 The Committee discussed the utility values used within the original model and noted that the utility value for the progression-free survival health state was 0.661 for erlotinib and 0.656 for gefitinib. It noted that the difference was mainly a result of difference in the response rates (58% for erlotinib compared with 28% for gefitinib) used in the calculation. The Committee heard from the ERG that the response rate from the gefitinib meta-analysis was 71.5% (Ku et al. 2011). The Committee heard from the manufacturer that the difference in utility values (0.005, <1%) used for the two treatments made little difference to the results from the model. However, the Committee saw little clinical justification for the difference in the utilities in the manufacturer's original model, and concluded that an analysis incorporating identical utility values for patients receiving erlotinib and patients receiving gefitinib in the progression-free survival health state should be used as a basis for its decision making.

4.14 The Committee acknowledged that, following its request for further clarification in the appraisal consultation document, the manufacturer had provided an updated economic model which incorporated equal progression-free survival and utilities for erlotinib and gefitinib. The results from the model depended on the costs of the drugs, the cost of administering the gefitinib patient access scheme and the proportion of patients on gefitinib who incurred the fixed charge on day 60. The Committee discussed the uncertainties in the clinical evidence which led to this request for an economic model based on there being no difference in the clinical benefit between the treatments. The Committee acknowledged the limitations of this type of economic model which incorporates no uncertainties about survival. The Committee concluded that, although the assumption of equal clinical benefit could be a conservative estimate of the clinical effectiveness of erlotinib, the updated economic model was in line with clinical opinion (see section 4.5), reflected the absence of any clinical data from direct comparisons, and allowed a direct comparison of the costs of the two treatments.

4.15 The Committee considered the impact of the cost of administering the gefitinib patient access scheme on the results from the updated economic model. When the mean administration cost changed from £438 per patient in the manufacturer's updated model to £111–118 in the ERG's exploratory analysis, erlotinib was cost effective when the proportion of patients incurring the fixed charge for gefitinib increased from 91% to 95%. The Committee acknowledged that the time taken to complete the online forms for the gefitinib patient access scheme was much shorter than that estimated by the manufacturer and that the typical administration costs for patient access schemes of this type were likely to be nearer to the ERG's estimate rather than the manufacturer's. This remained despite the possibility of additional reporting costs associated with the gefitinib patient access scheme which were not included in the ERG's analyses. The Committee understood that there may not be complete uptake of the gefitinib patient access scheme across the NHS, that some trusts may pay the list price for gefitinib and that this has not been considered in the updated model, which assumed all patients on gefitinib for more than 60 days would incur the fixed charge. The Committee concluded that the administration costs of the gefitinib patient access scheme were likely to be nearer the ERG's estimates rather than the manufacturer's, and that there may be some additional savings not included in the updated model because of the incomplete uptake of the gefitinib patient access scheme across the NHS.

4.16 The Committee considered the impact of the proportion of patients who receive gefitinib for more than 60 days (and would therefore incur the fixed charge) on the results from the updated economic model. Both the manufacturer and the ERG presented sensitivity analyses incorporating different costs for administering the gefitinib patient access scheme (see section 4.15). The Committee discussed the evidence on the proportion of patients who receive gefitinib for more than 60 days (and incur the fixed charge). The Committee was disappointed that there was not more reliable evidence available from the NHS. The Committee heard from clinical specialists that nearly all patients on gefitinib survive until day 60 when the third pack is issued. The Committee noted that the base-case analysis presented by the manufacturer used a proportion of 80%, which was the proportion of patients in the EURTAC trial who had completed 60 days of erlotinib treatment. There were about 88% of erlotinib patients in the EURTAC trial who were in the progression-free survival health state on day 60 but this included some patients who had not completed 60 days of treatment. The sensitivity analysis presented by the manufacturer identified that erlotinib became cost effective compared with gefitinib when 91% of patients incurred the fixed charge for gefitinib. The ERG's exploratory analysis estimated this proportion to be 95%. The Committee discussed the results from the updated analyses and on balance agreed that the sums of money either saved or spent are small given the uncertainties associated with the analysis. The Committee concluded that at the price agreed under the patient access scheme (as revised in 2012) erlotinib should be recommended as an option for the first-line treatment of locally advanced or metastatic EGFR-TK mutation-positive NSCLC.

Other considerations

4.17 The Committee discussed whether it needed to consider the supplementary advice from NICE that should be taken into account when appraising treatments that may extend the life of patients with a short life expectancy and that are licensed for indications that affect small numbers of people with incurable illnesses. It noted that the manufacturer did not make a case for erlotinib to be considered as an end-of life treatment in the submission. The Committee also heard from the manufacturer that in its view erlotinib does not meet the criteria for an end-of-life treatment. The Committee noted that in Erlotinib monotherapy for the maintenance treatment of non-small-cell lung cancer (NICE technology appraisal guidance 227) erlotinib did not meet the end-of-life criteria because the cumulative population for erlotinib was not considered small. The Committee therefore concluded that erlotinib did not need to be considered as a life-extending, end-of-life treatment.

4.18 The Committee discussed whether erlotinib should be considered an innovative technology, or if there were any significant and substantial health benefits which were not included in the economic model. It noted that the manufacturer did not make a case for erlotinib to be considered innovative, and did not identify any additional health benefits not included in the economic model. The Committee heard from the manufacturer that erlotinib is not considered a major change in treatment for locally advanced or metastatic EGFR-TK mutation-positive NSCLC, but is an incremental advance. The manufacturer stated that the oral administration and the straightforward patient access scheme gave value to erlotinib. The Committee concluded that erlotinib could not be considered to show significant innovation and that no additional health benefits had been identified which had not been adequately captured by the economic model.

4.19 The Committee considered whether NICE's duties under the equalities legislation required it to alter or to add to its recommendations. The Committee noted that no equality issues were included in the manufacturer's submission. It also noted that the reduced adverse reactions associated with tyrosine kinase inhibitors compared with those associated with chemotherapy raised during the scope consultation was not an equalities issue for this appraisal. No equalities issues were identified by the Committee. Given that the recommendations did not differentiate between any groups of people, the Committee concluded that its recommendations did not limit access to the technology for any specific group compared with other groups and that there was no need to alter or add to its recommendations.

Summary of Appraisal Committee's key conclusions