Mousa, Rimal
(2017)
Cost-effectiveness of primary endocrine therapy comparing against surgery as the initial treatment strategy in older women with primary breast cancer.
PhD thesis, University of Nottingham.
Abstract
Background
Breast cancer is highly prevalent in older women. Although surgery is the main initial treatment for younger population, research has demonstrated that older women with primary breast cancer have different tumour biology, comorbidities, and patient preferences from younger patients, and hence may benefit from primary endocrine therapy (PETx), such as Tamoxifen and aromatase inhibitors (AIs). However, there is no cost-effectiveness study that compares the use of different treatment strategies in older women.
Aim of study
This study aimed to evaluate the clinical and cost-effectiveness of using PETx against surgery as the initial treatment strategy in older women with primary breast cancer.
Methods
This research included four stages. First, a retrospective cohort study was conducted to describe the treatment pathways, and evaluate the effectiveness and costs associated with using different treatment strategies by using a longitudinal database of 1,759 older women, who were diagnosed with primary breast cancer at ages over 70 years old in breast cancer units in Nottingham. Regression analyses, including survival analysis and generalised linear model, were used to identify the covariates, the predict death rates and costs, respectively.
Second, a systematic review of economic evaluation studies was conducted using NHS Economic Evaluation Database, Cochrane Library, Ovid Medline, PubMed, and EMBASE to identify full economic evaluations that compared different treatment strategies in postmenopausal women with primary breast cancer. Quality and modelling methodologies of the included studies were assessed and summarised.
Finally, a Markov model was conducted from the systematic review to estimate the lifetime costs and quality adjusted life year (QALYs) associated with the use of PETx as against surgery in a hypothetical cohort of older women with primary breast cancer and ER-positive status. The analysis took the UK NHS perspective with a lifetime horizon. Transition probabilities were estimated using parametric survival models derived from the longitudinal database. Resource use and costs were assessed in British pounds of the year 2014 using the longitudinal database. Utilities were derived from literature, including Prescott et al. (2007) (1), Fallowfield et al. (1994), and Peasgood et al. (2010)(2). Both costs and outcomes were discounted by 3.5% annually. A subgroup analysis was conducted for patients with higher oestrogen receptor content (H-score >250 out of 300). PSA was conducted and aggregated results were presented in cost-effectiveness plane and cost-effectiveness acceptability curve. Sensitivity analysis was also conducted to investigate the impact of uncertain parameters and model assumptions.
Results
Retrospective analysis of the cohort study found that most patients received surgery (n=350; 52.08%) or PETx (n=322; 47.92%) as the initial treatment strategy. The most common adjuvant therapy was endocrine therapy (n=164; 46.86%). Comparing to those received surgery as an initial treatment, patients who received PETx had a significantly higher adjusted risk of breast cancer-related death (HR: 2.15; 95%CI: 1.03, 4.46), and lower costs (mean difference: -£3558.87; 95%CI: -£4018.11, -£3105.39).
Overall, 29 economic evaluations were included from the systematic reviews and studies that assessed surgery and none assessed PETx as the initial treatment. Most of the included economic studies used a Markov model with lifetime horizon and one-year cycle length. Nine studies which included subgroup analysis for older women (over 65 years old) used similar economic models and transition states to those that were used for younger women (50 to 65 years old) with primary breast cancer. The key disease-related health states were disease-free, recurrence, and death. Recurrence was mostly separated into locoregional and distant recurrence. These findings were used to design a Markov model in the following decision modelling.
In the base-case analysis of decision analytical modelling, the deterministic results showed that patients who received PETx as the initial treatment had lower costs and QALYs compared with patients who received surgery; the mean ICUR was £194 per QALY. However, for patients with H-score >250, PETx was associated with higher costs and lower QALYs; the mean ICUR was -£1849 per QALY.
In the probabilistic sensitivity analysis, PETx was associated with lower costs and QALYs. The mean ICUR was £363 per QALY, 95%CI: -6014.99, 5,756.86). However, for patients with H-score >250, PETx was associated with higher costs and lower QALYs. The mean ICUR was -£390.67 per QALY (95%CI: -7,295, 7776.54). The probability of PETx being cost-effective in a situation where the decision-maker is willing to accept (WTA) £21,000 per QALY was 24%. In the subgroup analysis, PETx was associated with slightly lower QALYs and higher costs. The probability of PETx being cost-effective in a situation where the decision-maker is willing to accept £21,000 per QALY was 43%.
Conclusion
The main findings from this study showed that PETx was not cost-effective compared to surgery. However, this study was mainly based on the effectiveness and cost from a single observational study in a secondary care setting, and other information (evidence) that may bias the effectiveness and cost estimates, including frailty status, adherence, comorbidities, and other unmeasured confounders were not recorded, and thus, may bias the cost-effectiveness results. Future research is recommended to collect further information on patients’ characteristics, including frailty, adherence, and adverse events that may influence the cost-effectiveness results. Moreover, collecting detailed estimation of the resource use in order to provide an accurate estimate for costs.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of 170717_Final_Thesis_RM.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "170717_Final_Thesis_RM.pdf")