|A new decision model for cost-utility comparisons of chemotherapy in recurrent metastatic breast cancer
|Hutton J, Brown R, Borowitz M, Abrams K, Rothman M, Shakespeare A
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.
Docetaxel versus paclitaxel as second line therapy to patients with anthracycline-resistant metastatic breast cancer.
The study considered a hypothetical, but representative, female patient with the following characteristics: Breast mass with axillary lymph node involvement was discovered six years previously; the patient had radial mastectomy with axillary clearance and received cyclophosphamide, methotrexate and fluorouracil (CMF). Five years later she developed bone pain and metastatic breast cancer was diagnosed. Further chemotherapy was given containing anthracycline (FAC) and had a partial response lasting 6 months followed by further progression resulting in lung metastases. The age of the chosen patient was not stated.
Hospital. The economic study was conducted in the UK.
Dates to which data relate
Effectiveness studies conducted between 1995 and 1996. Resource and cost data were obtained from a national database and literature published in 1996. Prices were converted to 1994 levels.
Source of effectiveness data
The effectiveness data were derived from a review/synthesis of previous studies in combination with estimates based on expert opinion.
A Markov model was utilised to estimate benefits and costs.
Outcomes assessed in the review
The review assessed the following positive outcomes and side-effects associated with each treatment regimen: Overall response rate, stable disease, early progressive disease, infections, hospitalised for infection, febrile neutropenia (FN), hospitalised for FN (3 days), severe neurotoxicity, severe fluid retention, severe arthralgia/myalgia or skin reactions, and death associated with infection and FN.
Study designs and other criteria for inclusion in the review
The authors did not state any specific inclusion criteria and the studies reviewed were only classified asclinical trials'. The authors included previous studies with the same disease profile as the hypothetical patient used in the model and rejected others.
Sources searched to identify primary studies
Criteria used to ensure the validity of primary studies
Methods used to judge relevance and validity, and for extracting data
Number of primary studies included
The authors reviewed approximately 13 studies of which data from 5 were ultimately used to provide base-line inputs to the model.
Methods of combining primary studies
The method of pooling studies is unclear.
Investigation of differences between primary studies
The authors demonstrated that they had investigated the results of previous studies and eliminated those which did not comply with the model criteria. Two examples of this were the inclusion of trials that only examined patients with anthracycline-resistant metastatic breast cancer, and the selection of dosage rates appropriate to this category of patient.
Results of the review
The following outcome results, for docetaxel and paclitaxel respectively,were selected for inclusion in the model:
Overall response rate = 47% and 21%, stable disease = 33% and 59%, early progressive disease = 20% for both therapies, infections = 25% and 23%,hospitalised for infection = 25% of infections, febrile neutropenia (FN) = 22% and 12.5%, hospitalised for FN (3 days) = 25% of FN, severe neurotoxicity = 3.5% and 7%, severe fluid retention = 7% and 0%, severe arthralgia/myalgia or skin reactions = 12% and 16%, and death associated with infection and FN = 0.8% and 1%.
Measure of benefits used in the economic analysis
The benefit measure was the Quality-Adjusted-Life-Year (QALY). A decision tree was used to estimate QALYs. A valuation matrix was utilised using opinions of nurses caring for the patients valued at the time of the study using standard gamble techniques.
Some costs and quantities were reported separately. Direct costs included drugs, hospital day medical and nursing staff, diagnostic tests, therapeutic procedures and outpatient appointments. These were calculated by using one oncologist to identify the resources needed for the four stages of the disease considered (early progressive stage, late progressive stage, stable disease and terminal disease), which were reviewed by four more oncologists before producing the final decision. Costs were analysed separately and taken from national databases and published literature, including the Monthly Index for Medical Specialties (MIMS). Discounting was not applied. The price year was 1994. Final costs were calculated using a decision tree.
Sensitivity analyses were carried out on the response rates for docetaxel and paclitaxel, the utility data, and the acquisition cost of docetaxel. In all cases, variability of data was investigated by means of one-way analyses.
Estimated benefits used in the economic analysis
The higher response rate associated with docetaxel produced an improvement in utility to the patient at an incremental-QALY rate equal to 0.0905, also expressed in incremental-Quality-Adjusted-Days (QAD) as 33.03 QAD. The benefits were considered from the commencement of treatment to the death of the patient. Side effects were considered in the economic analysis.
Total per-patient costs for the base-case solution were 8233 for docetaxel and 8013 for paclitaxel.
Synthesis of costs and benefits
The costs and benefits were combined in the summary measure of cost/QALY using incremental analysis. The results show that when comparing docetaxel with the comparator paclitaxel there was an incremental cost-utility ratio of 2431 per QALY (7 per healthy day). This was sensitive to the efficacy of docetaxel: it falls to 1,186 if docetaxel response rate increases from 47% to 56%.
The authors concluded that response rate was the key parameter determining the utility and cost-utility of treatments for metastatic breast cancer. Docetaxel produced a higher response rate when compared withpaclitaxel, resulting in an improvement in the quality of life of patients receiving docetaxel, which clearly outweighed the side-effects caused by the toxicity of the treatment. Docetaxel further produced a substantially larger utility benefit than paclitaxel at a smaller additional cost of 2431 per QALY gained, an incremental health care cost that was acceptable according to available defined limits and which did not alter in terms of docetaxel-dominance for all scenarios tested by the model.
The limitations of this study derive from those assumptions which were made during the construction of the decision model. The model, for example, assumed that neither docetaxel nor paclitaxel resulted in survival gains for patients (but only quality of life gains), that the median survival is the same (43%) for both treatment options, etc. However, none of these assumptions was based on clear evidence. The pooling of the individual studies was unclear, as is whether the search mechanism was comprehensive or systematic.
Implications of the study
The study provides interim data that will assist clinical decision-making in this area but there is a need for good data from prospective clinical trials.
Hutton J, Brown R, Borowitz M, Abrams K, Rothman M, Shakespeare A. A new decision model for cost-utility comparisons of chemotherapy in recurrent metastatic breast cancer. PharmacoEconomics 1996; 9(2): 8-22
Subject indexing assigned by NLM
Breast Neoplasms /therapy; Decision Support Techniques; Female; Humans; Markov Chains; Neoplasm Metastasis
Date bibliographic record published
Date abstract record published