Hormonal therapies for the adjuvant treatment of early oestrogen-receptor-positive breast cancer

4.1 Clinical effectiveness

4.1.1 The Assessment Group identified seven prospective randomised controlled trials (RCTs) that it considered to meet the inclusion criteria. Anastrozole was investigated in four studies, letrozole in two studies, and exemestane in one study.

4.1.2 Each study examined one of four different treatment strategies:

4.1.3 Meta-analysis of the RCTs was not possible because of the heterogeneity of trial designs.

4.1.4 In addition to the RCTs outlined in sections 4.1.1 and 4.1.2, two secondary studies were identified that compared anastrozole with tamoxifen in women who had already undergone 2 to 3 years of adjuvant tamoxifen therapy.

4.2 Cost effectiveness

4.2.1 The Assessment Group reviewed the literature and the submitted economic evidence, and generated its own economic model.

4.2.2 One cost-effectiveness study of anastrozole from the US was identified, which did not include endometrial cancer as an outcome, and it included a 25% reduction in benefit for anastrozole because of the effect on hip fracture. This was associated with an incremental cost per quality-adjusted life year (QALY) of $76,000 over a 20-year horizon or $202,000 over an 8-year horizon.

4.2.3 Economic models were submitted by the manufacturers of each drug for use in primary adjuvant, unplanned switch and extended adjuvant treatments, as appropriate. All submitted models were Markov models with 1000 simulations, a mean patient age of 61–64 years and a 25-year or lifetime horizon, and they all contained broadly similar health states and model structures. Differences between the models included the assumptions on the duration of benefit, the source of the utility data, and the inclusion of adverse events. The individual models were based on the data from the relevant clinical studies.

4.2.4 The Assessment Group developed a probabilistic Markov model that used trial evidence for all three treatment strategies and drugs separately, and five sub-models were developed: anastrozole (primary adjuvant and unplanned switch, the latter being an unlicensed indication), letrozole (primary adjuvant and extended adjuvant) and exemestane (unplanned switch). Health-related utilities were taken from published literature. The model used an annual cycle length and was run until patients reached 100 years of age. The comparator for the primary adjuvant and the unplanned switch strategies was 5-year tamoxifen therapy immediately following surgery, and for extended adjuvant therapy was placebo. The health states modelled were disease-free survival, loco-regional or contralateral recurrence, remission from loco-regional or contralateral recurrence, distant recurrence, death from breast cancer (via metastatic disease) and death from other causes.

4.2.5 An important assumption in the model was related to the duration of benefit after the treatment period: in the base-case analysis the benefits of aromatase inhibitors, in terms of relative risk, gradually declined over the next 10-year period to the extent that, by year 15 after surgery, the number of patients in disease-free survival was the same in both arms. The Assessment Group considered this to be a conservative assumption. An alternative assumption was tested in a sensitivity analysis (referred to as 'benefits maintained'). In the 'benefits maintained' scenario the rate of recurrence was the same for both arms after the treatment period, based on a long-term follow-up study on tamoxifen, thereby preserving the benefits of aromatase inhibitors.

4.2.6 The following adverse events of tamoxifen were included in the model: vaginal bleeding, endometrial cancer, venous thromboembolic events and ischaemic cerebrovascular disorders. Hypercholesterolaemia was excluded as an adverse event in the model because of the current uncertainty in the evidence base. Cardiovascular events were not modelled. Although the effect of aromatase inhibitors on hip fracture was not statistically significant, it was included in the model because of the potential long-term risks of hip and other fractures in postmenopausal women. The impact of fracture on the cost effectiveness was modelled separately using the Assessment Group's model developed for the appraisal on bisphosphonates, raloxifene, strontium ranelate and teriparatide for the prevention of osteoporotic fractures. The base case included a differential effect on fracture between treatment arms extending 5 years beyond the treatment period, with the relative risk for fracture being constant during the time of endocrine therapy, and declining gradually to control level over the next 5 years. The impact of doubling the fracture risk, a constant fracture risk over 10 years and the addition of treatment for osteoporosis with bisphosphonates was also estimated.

4.2.7 Generally, treatment with aromatase inhibitors was associated with increased drug costs and slightly decreased follow-up costs (for example, the costs of treating disease recurrence) compared with tamoxifen. Adverse events made a very minor contribution to the costs.

4.2.8 For primary adjuvant treatment the Assessment Group's modelling of 5year treatment with anastrozole compared with 5-year tamoxifen was associated with incremental costs per QALY gained of £31,965 and £12,310, using the base-case assumptions on the duration of benefit and the 'benefits maintained' scenario respectively. The corresponding result for anastrozole from the submission model was £12,463 for an analysis similar to the 'benefits maintained' scenario. The Assessment Group's modelling of 5-year treatment with letrozole compared with 5-year treatment with tamoxifen was associated with incremental costs per QALY gained of £21,580 and £9325, using the base-case assumptions on the duration of benefit and the 'benefits maintained' scenario respectively. The corresponding result for letrozole from the submission model was £10,286 for an analysis similar to the 'benefits maintained' scenario. Because of the shorter follow-up in the letrozole study underpinning this modelling, the results for letrozole were associated with a higher uncertainty than the results for anastrozole.

4.2.9 For unplanned switching the Assessment Group's modelling of 2- to 3-year tamoxifen therapy followed by a 2- to 3-year treatment with exemestane (compared with 5-year tamoxifen therapy) was associated with incremental costs per QALY gained of £19,170 using the basecase assumption on the duration of benefit, and £1638 using the 'benefits maintained' scenario – the latter value being lower than the corresponding result from the manufacturer's submission (£6817). The results from the model that compared anastrozole and tamoxifen in unplanned switching are not included because they relate to the use of anastrozole outside its licensed indication at the time of the appraisal.

4.2.10 For extended adjuvant treatment the Assessment Group's base-case analysis of 5 years of treatment with letrozole after 5 years of tamoxifen therapy was associated with an incremental cost per QALY gained of £9760 when compared with 5-year tamoxifen therapy followed by placebo. When the 'benefits maintained' scenario was used the incremental cost per QALY gained was £3306. The corresponding result from the manufacturer's submission was £7725 for an analysis similar to the 'benefits maintained' scenario.

4.2.11 When alternative assumptions on the impact of increased fracture risk and treatment for osteoporosis were included in the Assessment Group's model the incremental cost per QALY gained increased to more than £35,000 for anastrozole used as primary adjuvant treatment. However, using these assumptions in the 'benefits maintained' scenario the incremental cost per QALY gained remained less than £20,000 for all drugs and treatment strategies.

4.2.12 The manufacturer's model for letrozole in extended adjuvant treatment showed that the incremental cost-effectiveness ratios (ICERs) for node-negative and node-positive patients were £11,784 and £5373 respectively.

4.2.13 The Assessment Group carried out probabilistic sensitivity analyses for the base-case scenario, with the following results. For primary adjuvant treatment at a willingness to pay of £20,000 per additional QALY anastrozole and letrozole are 15–25% likely to be cost effective, while at a willingness to pay of £30,000 per additional QALY they are 50–60% likely to be cost effective. For unplanned switch treatment, exemestane is approximately 45% and 80% likely to be cost effective at a willingness to pay of £20,000 and £30,000 per additional QALY respectively. For extended adjuvant treatment letrozole is more than 95% likely to be cost effective at a willingness to pay of £20,000 per additional QALY. A probabilistic sensitivity analyses for the 'benefits maintained' scenario was not carried out. The probability of being cost effective can be expected to be higher with the 'benefits maintained' scenario.

4.3 Consideration of the evidence

4.3.1 The Committee reviewed the data available on the clinical and cost effectiveness of hormonal therapies for the adjuvant treatment of early oestrogen-receptor-positive breast cancer, having considered evidence on the nature of the condition and the value placed on the benefits of hormonal therapies by people with early oestrogen-receptor-positive breast cancer, 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 the evidence on the clinical effectiveness of aromatase inhibitors. It noted that there were important differences among the clinical trials, particularly in the timing of treatment with an aromatase inhibitor in relation to any previous treatment with tamoxifen. Furthermore, the Committee was aware of the differences in the licensed indications among the aromatase inhibitors.

4.3.3 The Committee agreed that there is convincing evidence that all three aromatase inhibitors, within their respective licensed indications, provide clinical benefit over tamoxifen in primary adjuvant or unplanned switch treatment, and over placebo in extended adjuvant treatment. The Committee considered the differences between the clinical trials and agreed that there is insufficient evidence to conclude that any one aromatase inhibitor (used within the licensed indications) or treatment strategy is more clinically effective than another.

4.3.4 The Committee noted that aromatase inhibitors are associated with increased fracture risk. They may also be associated with increased cardiovascular risk compared with tamoxifen, but there is still uncertainty about the long-term adverse effects because of the short follow-up in most of the studies. The Committee also noted that tamoxifen, but not the aromatase inhibitors, is linked to an increased risk of endometrial cancer and other gynaecological conditions. The Committee heard from the patient experts that other side effects related to menopausal symptoms and general well-being, experienced with aromatase inhibitors or tamoxifen, can be unpredictable and vary considerably from one woman to another, having an important effect on quality of life. Therefore, the Committee accepted that tamoxifen and aromatase inhibitors are not necessarily interchangeable therapies for all women.

4.3.5 The Committee considered all the evidence on cost effectiveness of aromatase inhibitors and, in particular, the assumptions made on the duration of benefit after the treatment period. It heard from the clinical specialists that the 'benefits maintained' scenario, used by the Assessment Group in a sensitivity analysis, provides the most relevant analysis. This is because this assumption is biologically more plausible than the base-case assumption and can be inferred from the long-term follow-up data for tamoxifen compared with placebo. Therefore, the Committee based its discussion on the cost-effectiveness analysis using the 'benefits maintained' assumption, and it noted that the incremental cost per QALY gained for aromatase inhibitors, compared with tamoxifen, was less than £20,000 for all treatment strategies. It further noted that the incremental cost per QALY gained did not increase to more than £20,000 when the predicted fracture risk was increased. The Committee concluded that the licensed strategies for the use of the aromatase inhibitors are cost effective when individually compared with tamoxifen. It also agreed that in extended adjuvant treatment letrozole is cost effective compared with placebo.

4.3.6 The Committee was mindful of the fact that both anastrozole and letrozole are currently licensed for primary adjuvant treatment. However, because of the differences between the clinical trials, particularly in the length of follow-up, the Committee agreed that any comparison of cost effectiveness between anastrozole and letrozole was associated with considerable uncertainty. The Committee agreed that it would not be appropriate to differentiate between the two drugs for primary adjuvant treatment. Therefore, the Committee concluded that all three aromatase inhibitors, within their licensed indications, should be recommended as options for the adjuvant treatment of early oestrogen-receptor-positive invasive breast cancer in postmenopausal women.

4.3.7 The Committee considered the evidence relating to the clinical and cost effectiveness of aromatase inhibitors in different subgroups of women with early breast cancer. It noted that the Assessment Report did not include any analysis of cost effectiveness in such subgroups but instead indicated that the use of aromatase inhibitors as primary adjuvant treatment was cost effective for the whole population of women with early breast cancer. The Committee noted that the recurrence rate of primary tumour is greatest for all women within the first 2 years after initial therapy and that, for women with node-positive cancer, this risk of early recurrence is especially high. The Committee heard from the clinical experts that, on this basis, the use of aromatase inhibitors as primary adjuvant treatment may have maximum benefit among women at highest risk of early recurrence, whereas for women who have a low risk of recurrence the potential for clinical benefit from all hormonal therapy was considered to be much smaller.

4.3.8 Therefore, the Committee considered the notion that for primary adjuvant treatment an aromatase inhibitor might be preferable to tamoxifen on the basis of cost effectiveness in women in whom the risk of early recurrence is particularly high. However, because of the lack of definitive evidence on the relative clinical and cost effectiveness of the use of the aromatase inhibitors in different risk groups, the Committee did not feel able to issue guidance on the relative cost effectiveness of the aromatase inhibitors for the different subgroups. The Committee noted that clarification on the definition of different risk groups, and potentially a consequent variation in treatment strategies, is likely to be reflected in the clinical guideline for early breast cancer currently under development at NICE.

4.3.9 The Committee was aware that the hormone receptors relevant to prognostic and therapeutic decision-making in early breast cancer included both oestrogen receptors and progesterone receptors. The clinical specialists informed the Committee that women who have oestrogen-receptor-negative but progesterone-receptor-positive cancer might also respond well to the use of the aromatase inhibitors. However, the Committee concluded that on the basis of the evidence before it, and the remit from the Department of Health, the current guidance applies to the group of women with early oestrogen-receptor-positive invasive breast cancer.

4.3.10 Finally, the Committee agreed that the choice of treatment should be made after discussion between the responsible clinician and the woman about the risks and benefits of the options available. It agreed that consideration of the treatment to be adopted should include whether the woman has received tamoxifen as part of her treatment so far, the licensed indications of the individual drugs, the side-effect profiles of the individual drugs and, in particular, the assessed risk of recurrence.