|
Using economic analysis to evaluate the potential of multimodality therapy for elderly patients with locally advanced pancreatic cancer |
Krzyzanowska M K, Earle C C, Kuntz K M, Weeks J 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 health technology examined comprised various strategies for the treatment of elderly patients with locally advanced pancreatic cancer (LAPC). In the first stage, radiation plus concurrent fluorouracil-based chemotherapy (RT-FU) was compared with no treatment. In the second stage, radiation plus gemcitabine (40 mg/m2 twice weekly for 5.5 weeks) (RT-GEM) was compared against RT-FU. After completion of radiation, non-progressing patients were eligible to continue with gemcitabine at a dose of 1,000 mg/m2 weekly, every 3 of 4 weeks.
Study population The model adopted a hypothetical cohort of 65-year-old men with no co-morbidities. The study population data were derived from the Surveillance, Epidemiology and End Results (SEER) Program and comprised a cohort of 1,696 patients with LAPC who were older than 65 years of age. The median age of the cohort was 75 years, 56% were female and 79% were white. Patients with any prior cancers, or with insufficient treatment information due to enrolment in a health maintenance organisation during the study period, were excluded from the study. Patients whose date of diagnosis (or death) differed by more than 2 months in the SEER and Medicare databases were also excluded, as were cases where cancer was first discovered at death or autopsy.
Setting Although the setting was not explicitly stated, it appears to have been outpatient care. The economic study was carried out in the USA.
Dates to which data relate The effectiveness data used to populate the model were obtained from studies published between 1969 and 2003. Resource use was retrospectively estimated using data from the SEER Program and a sample of patients diagnosed with LAPC between January 1991 and December 1996. The price year was 2005.
Source of effectiveness data Numerous clinical parameters were included in the model. However, the authors only reported the probability of 1-year survival without treatment and with RT-FU, and the probability of treatment-related complications when receiving RT-FU or RT-GEM.
Modelling A Markov model was developed to assess the cost-effectiveness of RT-FU versus no treatment and RT-GEM versus RT-FU. The time horizon of the model was not explicitly stated, but it appears to have been the patients' lifetime. The health states, cycle length (1 month), time-dependent transition probabilities and several modelling assumptions used were reported in full. Treatment effectiveness in each arm was estimated using Cox modelling and propensity scores methods. The authors used logistic regression in order to assign each patient a "propensity to receive treatment" based on the patients' baseline characteristics. Sample size power calculations were also reported.
Sources searched to identify primary studies The baseline survival probability was obtained from an observational study (Krzyzanowska et al. 2003, see 'Other Publications of Related Interest' below for bibliographic details). Gemcitabine-related toxicity estimates were obtained from a Phase II trial, while RT-FU related adverse events were derived from three published studies. As clinical effectiveness data on the gemcitabine-based radiation schedule were not available, authors' assumptions were used.
Methods used to judge relevance and validity, and for extracting data The methods used to identify the primary studies and to select the estimates used in the model were neither reported nor discussed. No inclusion criteria were specified for any parameters. The methods used to identify and calculate the estimates of effectiveness in the observational study were not reported in the current paper.
Measure of benefits used in the economic analysis Quality-adjusted life-years (QALYs) were used as the measure of benefit in the economic analysis. The utility values for the "not on active treatment" health state were derived from a published study (Glimelius et al. 1995, see 'Other Publications of Related Interest' below for bibliographic details). Due to a lack of available data, the authors arbitrarily assigned values for the remaining health states. The benefits were discounted at a rate of 3%.
Direct costs The study reported costs to Medicare and to the patient. However, only summary costs were reported, namely, the monthly cost per phase for each intervention (no treatment, RT-FU). For the RT-GEM intervention, only the additional costs of gemcitabine and treatment-related adverse effects were accounted for. Resource use was derived from the SEER database and adjusted Medicare reimbursement costs were used. However, the unit costs and the resources were not reported separately. The patients' monthly out-of-pocket payments were also included in the analysis, but were based on a different sample of patients and, again, only summary costs were reported. The costs were discounted at a rate of 3%. Adjustments for inflation were performed using the medical component of the Consumer Price Index. The costs were reported for the price year 2005.
Statistical analysis of costs The data were treated deterministically. However, power calculations were performed to ascertain an appropriate sample size for the economic analysis.
Indirect Costs No productivity losses were considered.
Sensitivity analysis Parameter uncertainty was investigated using one-way sensitivity analyses around most of the model parameters. Although the ranges used were reported, the methods used to determine these ranges were not discussed. Structural uncertainty of the model was also investigated. Specifically, the base-case analysis was restricted to first-line therapy only while, in the sensitivity analysis, it was assumed that all patients who completed radiation continued treatment with two cycles of single-agent gemcitabine. This assumption was justified with reference to a previous study. Expected value of perfect information analysis was not undertaken.
Estimated benefits used in the economic analysis Compared with no treatment, RT-FU resulted in 9.3 incremental quality-adjusted life-months.
Cost results The reported total costs included the cost of intervention-related adverse effects.
The total costs were $49,242 for the no treatment group and $103,323 for RT-FU.
Synthesis of costs and benefits Compared with no treatment, RT-FU resulted in an incremental cost of $68,724 per QALY gained. Accounting for the two-cycles of gemcitabine chemotherapy resulted in an incremental cost of $76,548 per QALY gained. The sensitivity analysis demonstrated that the results were sensitive to the extreme values of the mortality rate and the costs of RT-FU, and when low utility values were assigned to the health state "not on active treatment".
The results for the cost-effectiveness of the RT-GEM strategy compared with RT-FU were presented as a function of the hazard ratio (HR). Assuming an HR for death of 0.85 or less for RT-GEM compared with RT-FU resulted in an incremental cost of less than $100,000 per QALY gained. This implied a 5% absolute difference in survival at 1 year between the two interventions. The sensitivity analyses demonstrated that the results were most sensitive to variation of the utility value of the "not on active treatment" state, drug acquisition cost and the cost of treating treatment-related adverse events.
Authors' conclusions Radiation with fluorouracil-based chemotherapy (RT-FU) would appear to be a cost-effective option in comparison with no treatment for elderly patients with locally advanced pancreatic cancer (LAPC). On the other hand, radiation with gemcitabine (RT-GEM) incurs higher costs and is associated with greater toxicity, while its cost-effectiveness is highly affected by its efficacy, quality of life values, drug acquisition costs and costs of adverse events.
CRD COMMENTARY - Selection of comparators A justification was provided for the technologies compared. RT-GEM is a newly developed technology and, although its efficacy has not been yet documented, the authors wished to model its threshold efficacy and its potential to become a cost-effective alternative treatment regimen. You should decide if these represent valid comparators in your own setting.
Validity of estimate of measure of effectiveness The parameters in the model were derived from published research. The authors did not indicate whether a systematic search for data was performed. It is not possible to judge the validity of the data given the information reported in this paper (see Krzyzanowska et al. 2003 for further details). The authors did not report any search methods or inclusion criteria, and did not provide any justification for the estimates they selected. In addition, the efficacy of the gemcitabine-based radiation schedule was based on authors' assumptions, owing to a lack of relevant documented information, and the selection of the estimates was not justified. Validity of estimate of measure of benefit The estimation of health benefits (QALYs) was modelled using a Markov model. The benefits were appropriately discounted. However, given the lack of available data, the authors assigned arbitrary utility weights to some health states used in the model.
Validity of estimate of costs Although it was reported that the study had been conducted from a societal perspective, the productivity costs were not accounted for in the analysis. The authors only reported summary costs; it is therefore impossible to know whether all the relevant costs have been included in the analysis. Appropriately adjusted Medicare reimbursement costs were used for health service costs, but treatment-related complications were based on authors' assumptions. In addition, resource use and out-of-pocket patient costs were obtained from a different sample of patients and no relevant details were provided. The authors evaluated uncertainty in the cost data using a one-way sensitivity analysis, but no justification for the ranges selected was provided. All costs were appropriately discounted and converted to reflect 2005 prices.
Other issues The authors compared their findings with those from published studies that had confirmed that treatment is more cost-effective than no treatment. However, there were no studies in the literature that had investigated the relative cost-effectiveness of different treatment options. The authors acknowledged that the SEER study sample was limited to older patients (age above 65 years), thus restricting the generalisability of the results to younger patient populations. The impact of variation in the patient population on the economic results was not investigated through a sensitivity analysis. The authors do not appear to have presented their results selectively.
The authors reported a number of limitations to their study. First, owing to the lack of randomised trials, effectiveness data on chemoradiation were derived from observational studies, thus introducing uncertainty about the quality of the data. Second, there were problems with data availability, especially for quality of life estimates. Finally, certain model assumptions might not have reflected clinical practice. However, these assumptions were investigated in the sensitivity analysis.
Implications of the study The authors indicated that more research is needed to assess the health-related quality of life of patients with pancreatic cancer. Although the authors stated that the two intervention strategies are cost-effective, their incremental cost-effectiveness ratios were both higher than $50,000 per QALY, which other decision-makers may consider to be above their willingness-to-pay threshold.
Source of funding Supported by a fellowship from Cancer Care Ontario.
Bibliographic details Krzyzanowska M K, Earle C C, Kuntz K M, Weeks J C. Using economic analysis to evaluate the potential of multimodality therapy for elderly patients with locally advanced pancreatic cancer. International Journal of Radiation Oncology, Biology, Physics 2007; 67(1): 211-218 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.
Krzyzanowska MK, Weeks JC, Earle CC. Treatment of locally advanced pancreatic cancer in the real world: Population-based practices and effectiveness. J Clin Oncol 2003;21:3409-14.
Glimelius B, Hoffman K, Graf W, et al. Cost-effectiveness of palliative chemotherapy in advanced gastrointestinal cancer. Ann Oncol 1995;6:267-74.
Indexing Status Subject indexing assigned by NLM MeSH Aged; Antimetabolites, Antineoplastic /economics /therapeutic use; Combined Modality Therapy /economics /methods; Cost-Benefit Analysis; Deoxycytidine /analogs & Fluorouracil /economics /therapeutic use; Humans; Markov Chains; Medicare /economics; Models, Economic; Pancreatic Neoplasms /drug therapy /mortality /radiotherapy; Quality-Adjusted Life Years; Survival Analysis; derivatives /economics /therapeutic use AccessionNumber 22007000222 Date bibliographic record published 31/07/2007 Date abstract record published 31/07/2007 |
|
|
|