|
The cost-effectiveness of etanercept and infliximab for the treatment of patients with psoriatic arthritis |
Bravo Vergel Y, Hawkins N S, Claxton K, Asseburg C, Palmer S, Woolacott N, Bruce I N, Sculpher M J |
|
|
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. CRD summary This study compared two tumour necrosis factor antagonists, etanercept and infliximab, with palliative care for the treatment of patients with psoriatic arthritis. The authors concluded that only etanercept was within the acceptable cost-effectiveness thresholds defined by the National Institute of Clinical Excellence. The study was comprehensive and clearly reported, despite the analysis being hindered by data limitations. The conclusions should be considered in the context of these limitations, which were fully outlined by the authors. Type of economic evaluation Study objective This study compared three management options for the treatment of patients, with psoriatic arthritis (PsA), who had not responded to treatment with at least two different conventional disease-modifying anti-rheumatic drugs. Interventions The treatment options included two tumour necrosis factor (TNF) antagonists, namely etanercept (Enbrel®) at a dosage of 25mg, and infliximab (Remicade®) four vials at 5mg per kg dosage. These were compared with palliative care. Methods Analytical approach:A probabilistic decision analytic model was constructed to compare the cost-effectiveness of the treatment options. A 10-year time horizon was used in the base case. The authors reported that the study perspective was that of the UK National Health Service.
Effectiveness data:The authors stated that a systematic review was undertaken to identify all the relevant clinical data. Bayesian evidence synthesis methods were used to obtain the estimates of the main clinical effect parameters. These were the respective response rates of the two comparators, at three months, measured using PsA response criteria, and the mean reduction in functional status, measured using the Health Assessment Questionnaire (HAQ). The synthesis consisted of two linked meta-analyses. The model also considered a natural history of the progression rate, mortality and withdrawal rates. Withdrawal was possible at any time during the model, with all treatment failures moving on to palliative care.
Monetary benefit and utility valuations:The utility values were derived from a published study which combined individual patient level data on the HAQ and the European Quality of Life (EQ-5D) questionnaires using a linear-regression model.
Measure of benefit:The measure of benefit was quality-adjusted life-years (QALYs). These were discounted at an annual rate of 1.5%.
Cost data:The cost categories included drug acquisition costs, administration costs and monitoring costs. The drug costs were taken from the British National Formulary and other unit costs were derived from national databases. The other direct costs associated with PsA were estimated as a function of the patients’ HAQ scores and were derived from a published study on rheumatoid arthritis. All costs were discounted at an annual rate of 6% and were reported for the price year 2004 to 2005 in UK pounds sterling (£).
Analysis of uncertainty:The uncertainty in the model parameters was investigated using Monte Carlo simulations in a probabilistic sensitivity analysis. Probabilities assigned to the input parameters were reported. One-way sensitivity analyses around the discount rate and using three vials of infliximab (instead of four) were also conducted. Two alternative rebound scenarios were also tested. In the first, the rebound was assumed to be equal to the gain, which meant that the patient’s HAQ score reduced by the amount that it had initially gained and they then continued from that level. In the second, the rebound after treatment failure was assumed to be equal to the natural history of the progression of the disease, which meant that the patient's HAQ score reduced to where it would have been had treatment not been initiated and they then continued from that level. Lastly, the expected value of perfect information was calculated for the whole model and for specific parameters (short-term effectiveness, treatment failure rate, mortality, direct health care costs and utilities). Results An incremental cost-effectiveness analysis was performed.
Over a 10-year horizon, the incremental cost-effectiveness ratio (ICER) for etanercept compared with palliative care was £26,361 for the rebound equal to gain scenario (best case) and £30,628 per QALY for the rebound to natural history scenario (worst case).
The ICER for infliximab compared with etanercept was £165,363 for the best case and £205,345 per QALY gained for the worst case.
At a threshold of £30,000 to £40,000 per QALY in the best-case scenario and in the worst-case at £40,000, etanercept had the greatest probability of being cost-effective, which ranged from 0.69 to 0.93. At a £20,000 threshold and in the worst-case at £30,000, palliative care had the greatest probability of being cost-effective.
The longer lifetime horizon resulted in greater costs as well greater effectiveness lowering the ICERs for both TNF antagonists. Using the same discount rate of 3.5% for both costs and QALYs resulted in higher costs and lower QALYs increasing the ICER for the two TNF antagonists. Using three vials of infliximab did not alter the results of the incremental analysis. Authors' conclusions The authors concluded that only etanercept could be regarded as cost-effective based on acceptable thresholds for treatments defined by the National Institute of Clinical Excellence. They stated that further research was needed into the uncertainty surrounding the estimates of short-term effectiveness, the utility parameters, and the assumptions on the effect of rebound. CRD commentary Interventions:Although the interventions that comprised palliative care were not clearly reported, the TNF antagonists were described in detail, including their dosage. The authors did not discuss the existence of alternative therapies, which may have been relevant comparators.
Effectiveness/benefits:The effectiveness data were mainly obtained from three Phase-III randomised controlled trials, which were identified by a systematic review. No details of this review were presented. To fully assess its quality, the reader would need to refer to the full report or the review paper publication. The benefits were presented in the form of QALYs, which were taken from the literature. The details of how they were derived and the patient characteristics were reported.
Costs:The costs appeared to reflect the perspective. The direct costs associated with PsA were taken from a published study and not presented in any detail, so an assessment on their completeness is not possible from the paper. The costs associated with the TNF antagonists were reported in more detail, along with their sources. Both discounting and the price year were reported, enhancing the possibility of revaluing the results in future years.
Analysis and results:The model structure was presented graphically and relevant details on the modelling assumptions were reported. In addition, the authors referred the reader to the full Health Technology Assessment for further information on the model and the methods used for the synthesis of the data. The evidence synthesis techniques were reported clearly and in enough detail to allow the reader to understand what was undertaken. The synthesis would appear to have been comprehensive and conducted in an appropriate manner. A probabilistic sensitivity analysis was conducted to identify the uncertainty in the model input parameters. Details of the distributions applied were reported. All the cost parameters were incorporated as fixed parameters. The results of the sensitivity analysis were presented clearly and fully. Further, the authors reported their perceived limitations to their study, and appropriately acknowledged the need for further research.
Concluding remarks:The study was comprehensive and clearly reported although the analysis was hindered by data limitations. The conclusions reached should be considered in the context of these limitations. Funding Funded by the Health Technology Assessment Programme. Bibliographic details Bravo Vergel Y, Hawkins N S, Claxton K, Asseburg C, Palmer S, Woolacott N, Bruce I N, Sculpher M J. The cost-effectiveness of etanercept and infliximab for the treatment of patients with psoriatic arthritis. Rheumatology 2007; 46: 1729-1735 Indexing Status Subject indexing assigned by NLM MeSH Adult; Antibodies, Monoclonal /economics /therapeutic use; Antirheumatic Agents /economics /therapeutic use; Arthritis, Psoriatic /drug therapy /economics; Bayes Theorem; Clinical Trials, Phase III as Topic; Cost-Benefit Analysis; Drug Costs /statistics & Etanercept; Female; Humans; Immunoglobulin G /economics /therapeutic use; Infliximab; Male; Middle Aged; Palliative Care /economics; Receptors, Tumor Necrosis Factor /therapeutic use; State Medicine /economics; Treatment Outcome; Tumor Necrosis Factor-alpha /antagonists & inhibitors; numerical data AccessionNumber 22007002779 Date bibliographic record published 03/02/2009 Date abstract record published 13/05/2009 |
|
|
|