|
Clinical and economic impact of epoetin in adjuvant-chemotherapy for breast cancer |
Fagnoni P, Limat S, Chaigneau L, Guardiola E, Briaud S, Schmitt B, Merrouche Y, Pivot X, Woronoff-Lemsi M C |
|
|
Record Status 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. Health technology The use of epoetin (EPO) for the treatment of anaemia in breast cancer (BC) patients receiving adjuvant chemotherapy, when required. Treatment with EPO was administered when a patient's haemoglobin level was lower than 11.5 g/dL while receiving adjuvant chemotherapy. EPO alpha was given at a dose of 40,000 IU/week. If no response (an increase of at least 1 g/dL in haemoglobin level) was observed after the first 6 weeks of treatment, the EPO dose was increased to 80,000 IU/week. When no response was observed after 6 additional weeks of treatment, EPO was stopped. In all cases, a haemoglobin level below 8 g/dL was treated with red blood cell transfusion.
Study population The study population comprised women with BC who had received 6 cycles of standard adjuvant chemotherapy regimen every 3 weeks, by intravenous infusion. Adjuvant chemotherapy included anthracyclin (adriamycin 50 mg/m2 or epirubicin 100, 75 or 50 mg/m2) and cyclophosphamide (500 mg/m2 with or without fluorouracil 500 mg/m2). Women who did not complete the 6 cycles of chemotherapy during the study period (2 weeks) were excluded, as were patients who received sequential or concomitant taxane-based adjuvant chemotherapy.
Setting The setting was a hospital. The economic study was carried out in France.
Dates to which data relate The effectiveness and resource use data were gathered between 1999 and 2001 for the control group and between 2002 and 2004 for the EPO group. The price year was 2004.
Source of effectiveness data The effectiveness evidence was derived from a single study.
Link between effectiveness and cost data The costing was carried out retrospectively on the same sample of patients as that used in the effectiveness analysis.
Study sample Power calculations were not reported. A sample of 192 consecutive patients was retrospectively identified at the authors' institution over the period 1999 to 2004. There were 91 women in the control group and 101 in the EPO group. The women had a mean age of 52.3 (+/- 10.1) years in the control group and 55.2 (+/- 9.9) years in the EPO group, and their mean weights were 66.0 (+/- 13.8) kg and 64.5 (+/- 12.2) kg, respectively. In terms of tumour size, 36.3% of the control group were T1, 52.7% were T2 and 11% were T3. The corresponding values for the EPO group were 43.6% (T1), 46.5% (T2) and 9.9% (T3). Five patients (1 in the control group and 4 in the EPO group) who did not complete the 6 cycles of chemotherapy were excluded from the analysis.
Study design This was a retrospective cohort study with historical control that was carried out at a single centre (the University Hospital of Besancon, France). The length of follow-up was 22 weeks. No patient was lost to the follow-up assessment.
Analysis of effectiveness All patients who completed the chemotherapy cycles were considered in the analysis of effectiveness. The clinical end points used were:
the actual percentage of patients receiving EPO in the EPO group,
the rate of EPO failure,
red blood cell transfusion requirement,
anaemia-related hospitalisations,
haemoglobin levels, and
the duration of sick leave.
The authors stated that the two groups of patients were similar at baseline in terms of demographic, clinical and therapeutic aspects.
Effectiveness results In the EPO group, 45.5% of patients received EPO with a mean duration of 9.2 weeks. Twelve patients (26.1%) that received EPO experienced treatment failure.
No red blood cell transfusions were carried out in any of the groups.
The rate of anaemia-related hospitalisation was comparable between the groups (2.2% in the control group and 2.0% in the EPO group).
The mean number of weeks spent with haemoglobin levels above 12 g/dL was significantly higher in the EPO group than in the control group (17.6 versus 13.0).
The average number of weeks spent with a haemoglobin level between 10.00 and 11.99 g/dL was significantly lower in the EPO group (4.1 versus 8.1). Similarly, the average number of weeks spent with a haemoglobin level lower than 9.99 g/dL was significantly lower in the EPO group (0.2 versus 0.9 weeks).
The rate of sick leave was similar between groups (54.8% in the control group and 51.4% in the EPO group).
The length of sick leave was 9.4 weeks in the control group and 10.1 weeks in the EPO group (the difference was not statistically significant).
Clinical conclusions The effectiveness analysis showed that the potential use of EPO for the treatment of cancer-related anaemia was more effective in terms of the haemoglobin levels than a strategy without EPO.
Measure of benefits used in the economic analysis The summary benefit measure used was the number of quality-adjusted life-years (QALYs). The authors described the approach used to calculate the QALYs. Specifically, QoL was estimated according to haemoglobin levels found in the clinical study. For each week, these haemoglobin levels were transformed into QoL data according to the Linear Analog Scale Assessment (LASA) relationship between changes in haemoglobin level and QoL. The continuous haemoglobin level was divided into sections. There was one sub-set with a haemoglobin level lower than 7.49 g/dL, one sub-set with a haemoglobin level higher than 13.5 g/dL, and six sub-sets defined by 1 g/dL unit value between 7.5 and 13.49 g/dL. For each haemoglobin level section, a LASA QoL score was attributed. QoL scores for each haemoglobin level section were then multiplied by the number of weeks spent by section, added for every patient and transformed into QALYs.
Direct costs The analysis of the costs was carried out from the perspective of the French Public Health Insurance. Thus, only direct costs such as EPO, red blood cell transfusions and anaemia-related hospitalisations were included. Other resources were assumed to have been similar between the groups. Hospitalisation costs referred to medical and non-medical personnel costs, hospital structure costs, logistics costs, small equipment and medical resources (e.g. medications, medical devices, biology and radiology). The unit costs were reported but data on the quantities of resources used were not presented separately. Resource use was based on data derived from hospital records for the patients included in the effectiveness analysis. The costs of EPO were derived from wholesale price lists from the authors' institution, while other costs came from the accounting system of the hospital. The cost of EPO was calculated according to the exact dose that was administered. Since the costs were incurred during a short timeframe, discounting was not relevant. The price year was 2004.
Statistical analysis of costs The non-parametric Mann-Whitney test was used to assess the statistical significance of differences in total costs between the groups.
Indirect Costs The indirect costs were included only in the sensitivity analysis. Such costs were calculated using data from the French National Institute of Statistics and Economic studies. The duration of sick leave came from the sample of patients included in the effectiveness analysis. Since the French Public Insurance only refunds 50% of sick leave payments, only this percentage of unit costs was applied to the number of days lost from work.
Sensitivity analysis A univariate sensitivity analysis was carried out to assess the robustness of cost-utility ratios to variations in the approach used to measure the relationship between haemoglobin levels and QoL (other scales were used), the cost of individual items (+/- 30%) and the inclusion of the indirect costs. Two sub-group analyses, one according to the use of 100 mg/m2 of epirubicin and other chemotherapy regimens and one according to age (above or below the age of 55 years), were also performed.
Estimated benefits used in the economic analysis The additional QALYs per patient associated with EPO in comparison with no EPO were 0.0052. This difference was statistically significant.
Cost results The mean health care cost per patient was EUR 34 in the control group and EUR 1,649 in the EPO group. The difference (EUR 1,615) was statistically significant.
Synthesis of costs and benefits Incremental cost-utility ratios were calculated in order to combine the costs and benefits of the alternative strategies. The incremental cost per QALY gained with EPO over a strategy without EPO was EUR 310,577.
The base-case results were confirmed in the sensitivity analysis, with the incremental cost per QALY gained ranging from EUR 163,131 (changing sources for utility weights) to EUR 403,846 (increasing EPO costs). In particular, the inclusion of the indirect costs did not substantially alter the base-case results (EUR 292,692 per QALY), while the sub-group analysis revealed very similar cost-effectiveness results in women aged younger or older than 55 years.
Authors' conclusions A strategy in which epoetin (EPO) could be used, when required, for the treatment of cancer-related anaemia in women with breast cancer (BC) undergoing adjuvant chemotherapy was not cost-effective in comparison with a strategy without EPO from the perspective of the health insurance system in France.
CRD COMMENTARY - Selection of comparators The choice of the comparator, a strategy without EPO, was appropriate as it allowed the added value of EPO to be assessed. Dosages and administration schedules of EPO were reported and were consistent with international recommendations. You should decide whether they are valid comparators in your own setting.
Validity of estimate of measure of effectiveness The clinical evidence came from a retrospective study. This type of study is usually associated with some limitations given the lack of random allocation of patients to the study groups and the retrospective design. Thus, despite the inclusion of consecutive patients, some selection bias might have affected the results of the analysis. However, the authors pointed out that an unselected cohort of patients could be considered more representative of common clinical practice. Since the two groups of patients were identified in two different time periods, some time-related bias could have had an impact on the clinical outcomes. The authors did not consider the potential impact of factors other that the study interventions. However, a strength of the analysis was the baseline comparability of the study groups. Power calculations were not reported and no justification for the sample size was provided. Thus, it was unclear whether the study was appropriately powered to detect statistically significant differences between the groups. The evidence came from a single centre, which may not be representative of other institutions. The potential uncertainty in the clinical results was not addressed in the sensitivity analysis in which only other items (costs and utilities) were varied. These issues tend to limit the internal validity of the analysis.
Validity of estimate of measure of benefit The benefit measure used in the analysis was appropriate as QALYs capture the impact of the interventions on the most relevant dimension of health for women with BC, which is QoL. In addition, QALYs can be compared with the benefits of other health care interventions. QALYs were estimated from QoL data on the basis of the relationship between QoL and haemoglobin levels. Alternative approaches to calculate the QALYs were considered in the sensitivity analysis.
Validity of estimate of costs The authors explicitly stated the viewpoint chosen in the economic analysis. A broader perspective was also considered in the sensitivity analysis in which indirect costs associated with productivity losses were also included. The unit costs were reported for all items, but there was limited information on resource consumption. The sources from which the costs were derived were consistent with the perspective adopted in the analysis. The cost estimates were specific to the study setting but the impact of using alternative economic estimates was investigated in the sensitivity analysis. Statistical analyses of the costs were also performed. The price year was given, thus facilitating reflation exercises in different time periods.
Other issues The authors not only stated that the final cost-effectiveness results of their analysis were comparable with those from other studies, but also that some clinical results of the analysis such as failure rate or days of sick leave were similar to those reported in other studies. The issue of the generalisability of the study results to other settings was not explicitly addressed, although the use of sensitivity analysis for some economic estimates might have gone some way to enhancing the external validity of the study. However, most of the cost estimates should be considered country-specific and will be difficult to transfer to other settings. The analysis referred to women with BC and this was reflected in the authors' conclusions. The results of the base-case analysis and the sensitivity analysis were extensively reported and clearly presented.
Implications of the study The study results do not support the use of EPO for the treatment of anaemia in women with BC being administered adjuvant chemotherapy. However, the authors noted that further discussion would be useful in defining an acceptable threshold for the implementation of health technologies.
Bibliographic details Fagnoni P, Limat S, Chaigneau L, Guardiola E, Briaud S, Schmitt B, Merrouche Y, Pivot X, Woronoff-Lemsi M C. Clinical and economic impact of epoetin in adjuvant-chemotherapy for breast cancer. Supportive Care in Cancer 2006; 14(10): 1030-1037 Other publications of related interest Because readers are likely to encounter and assess individual publications, NHS EED abstracts reflect the original publication as it is written, as a stand-alone paper. Where NHS EED abstractors are able to identify positively that a publication is significantly linked to or informed by other publications, these will be referenced in the text of the abstract and their bibliographic details recorded here for information.
Marchetti M, Barosi G. Clinical and economic impact of epoetins in cancer care. Pharmacoeconomics 2004;22:1029-45.
Cremieux PY, Finkelstein SN, Berndt ER et al. Cost-effectiveness, quality adjusted life-years and supportive care. Pharmacoeconomics 1999;16:459-72.
Bokemeyer C, Aapro MS, Courdi A, et al. EORTC guidelines for the use of erythropoietic proteins in anaemic patients with cancer. Eur J Cancer 2004;40:2201-16.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Anemia /chemically induced /drug therapy /economics; Antineoplastic Combined Chemotherapy Protocols /adverse effects; Biomarkers /blood; Breast Neoplasms /drug therapy /economics; Chemotherapy, Adjuvant; Cost-Benefit Analysis; Cyclophosphamide /adverse effects; Doxorubicin /adverse effects; Epirubicin /adverse effects; Epoetin Alfa; Erythropoietin /economics /therapeutic use; Female; Fluorouracil /adverse effects; France; Hematinics /economics /therapeutic use; Hemoglobins /drug effects; Humans; Infusions, Intravenous; Middle Aged; Quality of Life; Quality-Adjusted Life Years; Recombinant Proteins; Retrospective Studies; Sensitivity and Specificity; Treatment Outcome AccessionNumber 22006002079 Date bibliographic record published 30/04/2007 Date abstract record published 30/04/2007 |
|
|
|