|Cost utility analysis of early adjuvant letrozole or anastrozole versus tamoxifen in postmenopausal women with early invasive breast cancer: the UK perspective
|Karnon J, Delea T, Barghout V
This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.
This study compared two aromatase inhibitors (anastrazole and letrozole) with tamoxifen as adjuvant therapy for postmenopausal women with early breast cancer. The authors concluded that five years of letrozole or anastrazole treatment was cost-effective, with letrozole being probably more cost-effective at a £20,000 willingness to pay threshold. The study was adequately conducted, and, on the whole, was clearly and transparently reported. The conclusions reached by the authors reflected the scope of their analysis.
Type of economic evaluation
The aim was to evaluate the cost-effectiveness of two aromatase inhibitors (AIs, letrozole and anastrazole) compared with tamoxifen for early adjuvant therapy in postmenopausal women with early, invasive, hormone receptor positive, breast cancer.
The interventions compared were five years of treatment with letrozole, anastrazole or tamoxifen following primary surgery. The adverse events associated with the interventions were also analysed. No further details of the interventions were given. The data were mainly extracted from two trials (Breast International Group, BIG, 1-98 Collaborative Group. 2005, and ATAC Trialists’ Group. 2005, see 'Other Publications of Related Interest' below for bibliographic details).
A Markov model was adapted, from a previous study, to evaluate the lifetime costs and benefits of the interventions, on a yearly basis. The authors stated that the perspective was that of the UK National Health Service. The effect of treatment was not considered to carry on beyond the fifth year and, after the five years of treatment, all women returned to the general population age-adjusted incidence of adverse events.
The main effectiveness data came from the Breast International Group (BIG) 1-98 Collaborative Group and ATAC Trialists’ Group studies. These were both multi-centre randomised trials with more than 4,000 patients in total, and a median follow-up of over two years. The main effectiveness estimates were the reduction in breast cancer with letrozole (hazard ratio, HR: 0.72, 95% confidence interval, CI: 0.61, 0.86, p=0.001) and with anastrazole (HR: 0.74, 95% CI: 0.64, 0.87, p=0.0002) compared with tamoxifen, and similar reductions in distant metastases. The other estimates were extracted from other sources using the authors’ criteria and no systematic literature review was reported. The other parameters, besides treatment effectiveness, were tumour recurrences or contralateral tumours, locoregional recurrences, distant metastases, breast cancer, and general mortality. The incidence of adverse events (endometrial cancer, hip or other fractures, myocardial infarction, heart failure, unstable angina, arthralgias, and venous thromboembolism, VTE) was also considered. The authors used the mean figures from the original trials for VTE and arthralgias and all other baseline event incidence rates were from UK data.
Monetary benefit and utility valuations:
The utility values for all breast cancer health states were taken from a cohort of 26 breast cancer patients, using the standard gamble method. No further details were provided. The decrease in utility due to treatment related adverse events during the first year was derived from the literature, and, because no data were found on the rate of decreases after the first year, this was assumed to be a half.
Measure of benefit:
The two main measures of benefit were life-years (LYs) and quality-adjusted life-years (QALYs) gained. A discount rate of 3.5% was applied to both measures. The authors also reported relapse rates, metastatic disease rates, and disease free survival rates.
The cost categories included the costs of adjuvant therapy, surveillance, treatment of locoregional recurrence, distance recurrence, or metastases, terminal costs, and adverse event costs. The recurrence costs, during each of the first five years, were taken from a patient-level database of 199 women. The mean cost and bootstrapped 95% confidence intervals were calculated. Other sources were used for treatment related adverse event costs. The unit costs came from nationally available sources. The price year was 2005, and the currency was UK pounds sterling (£). Although the authors reported that the costs were discounted at an annual rate of 3.5%, they also stated that they used no discounting of costs in the individual patient data analysis (which had a five-year time horizon).
Analysis of uncertainty:
A series of one-way and two-way sensitivity analyses were performed, together with a probabilistic sensitivity analysis using a second-order Monte Carlo simulation. In addition, a scenario analysis investigated the effects of the relative risk carrying over from the first five years to an additional five year period and the effects of including the costs of osteoporosis screening and treatment with letrozole or anastrazole. Probability distributions were adequately reported.
The discounted LYs gained with tamoxifen were 13.20. With letrozole they were 13.55 (0.35 LYs benefit over tamoxifen), and with anastrazole they were 13.45 (0.25 LYs benefit over tamoxifen).
The discounted QALYs gained were 12.78 with tamoxifen, 13.14 with letrozole (0.36 QALYs benefit over tamoxifen), and 13.04 with anastrazole (0.26 QALYs benefit over tamoxifen).
The total costs were £10,314 for tamoxifen, £14,034 for letrozole (£3,721 more than tamoxifen), and £13,243 for anastrazole (£2,929 more than tamoxifen).
For letrozole over tamoxifen, the incremental cost per LY was £10,502, and the incremental cost per QALY was £10,379 (95% CI: £6,654, £24,369).
For anastrazole over tamoxifen, the incremental cost per LY was £11,703, and the incremental cost per QALY was £11,428 (95% CI: £6,131, £53,125).
As expected, assuming the benefits lasted longer, improved the cost-effectiveness of both anastrazole and letrozole. The most sensitive parameters were the relative risk of recurrence with anastrazole or letrozole. The two AIs were less cost-effective in the elderly cohorts. The cost-effectiveness acceptability curves showed that if the willingness to pay threshold was £20,000, it was 95% certain that letrozole was cost-effective over tamoxifen and, if the threshold was £30,000, it was 99% certain. At £20,000 the probability was 85% that anastrazole was cost-effective over tamoxifen, and at £30,000 the probability was 94%.
The authors concluded that five years of letrozole or anastrazole treatment was cost-effective in post-menopausal women with early breast cancer. Although the results showed significant overlap, letrozole had a 95% probability of being more cost-effective than tamoxifen at a £20,000 willingness to pay threshold, while anastrazole had an 85% probability. Further research to define the relative benefits of the aromatase inhibitors had already begun.
The authors reported the interventions clearly and justified their selection. However, it was not clear whether other AIs or other classes of drug were available and could have been included.
The authors relied heavily on two large trials that compared the two AIs with tamoxifen. This study analysed the AIs' relative effects to tamoxifen, using a model with a tamoxifen arm in common. Although this was an indirect comparison, it could inform decision making until a direct comparison trial has been completed. The authors clearly reported their sources, the data and their assumptions, although it was not clear whether a systematic review was performed. As the authors acknowledged, there were only a few published utility studies on breast cancer, but the inputs they used seem to have been reasonable.
Most of the costs categories relevant to the perspective stated were included. An important part of the cost data came from an study of a Scottish cohort of women with breast cancer. This study was not described in detail, but the authors judged that it could be extrapolated to the rest of the UK.
Analysis and results:
The authors improved a previous model and they reported the study transparently. The impact of uncertainty was adequately investigated, and the model was validated against external data, showing similar long-term survival rates. The generalisability to settings other than the UK was not addressed. One limitation of the study is that the authors performed two independent comparisons for each AI with tamoxifen, because the AIs were part of the same class of drug. An incremental analysis of both drugs would have been interesting had it also compared each one with tamoxifen in order to select the most cost-effective alternative.
This study was adequately conducted, and, on the whole, was clearly and transparently reported. The conclusions reached by the authors reflected the scope of their analysis.
Funded by Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA.
Karnon J, Delea T, Barghout V. Cost utility analysis of early adjuvant letrozole or anastrozole versus tamoxifen in postmenopausal women with early invasive breast cancer: the UK perspective. European Journal of Health Economics 2008; 9: 171-183
Other publications of related interest
The Breast International Group (BIG) 1-98 Collaborative Group. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. New Eng J Med 2005;353:2747-57.
ATAC Trialists' Group. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years' adjuvant treatment for breast cancer. Lancet 2005;365:60-2.
Subject indexing assigned by NLM
Antineoplastic Agents /administration & Antineoplastic Agents, Hormonal /administration & Breast Neoplasms /drug therapy /economics /epidemiology; Cost-Benefit Analysis; Female; Great Britain /epidemiology; Humans; Markov Chains; Neoplasm Recurrence, Local; Neoplasms, Hormone-Dependent /drug therapy /economics; Nitriles /administration & Postmenopause; Quality of Life; Randomized Controlled Trials as Topic; Tamoxifen /administration & Triazoles /administration & dosage /adverse effects /economics; dosage /adverse effects /economics; dosage /adverse effects /economics; dosage /adverse effects /economics; dosage /adverse effects /economics
Date bibliographic record published
Date abstract record published