TNF-inhibitors for rheumatoid arthritis Part III Cost-effectiveness analysis
Health technology assessment
|Updated
Download
Key message
Background:
Treatment with tumour necrosis factor alpha (TNF-α, or simply TNF) inhibitors is considered to be an alternative to the use of traditional disease-modifying anti-rheumatic drugs (DMARDs) in patients with different rheumatic diseases, i.e. rheumatoid arthritis (RA). There are three TNF-inhibitor drugs currently available on the market (brand names in brackets): adalimumab (Humira), etanercept (Enbrel) and infliximab (Remicade). The Norwegian Knowledge Centre for the Health Services has previously summarised the evidence on the drugs’ efficacy and safety (in randomised clinical trials and observational studies) while the present report considers cost-effectiveness of the drugs for rheumatoid arthritis. After considerable growth over several years, the aggregate sales of the three drugs amounted to 860 million NOK in 2006.
RA is a serious disease, not least from an economic perspective. No cost-of-illness studies have been found for Norway, but studies from Sweden suggest that the costs of the disease are substantial with a large proportion related to loss of work capacity.
Methods:
We undertook a review of economic evaluations of TNF-inhibitors against RA, and considered an analysis of health-related quality of life data for patients on TNF-inhibitors and DMARD users from a Norwegian observational study.
Results:
A total of twelve studies from six countries was included in the literature review. The studies were based on health economic models, which were diverse in their characteristics, and therefore the estimates of cost-effectiveness varied significantly.
Conclusions:
In our review of economic evaluations of TNF-inhibitors, we found significant variation in the type and features of the models used, which led to a wide range of estimates. The potential for direct comparisons of results between the studies, and thus transferability of results into Norwegian setting, is limited. With this in mind, our main conclusions are as follows:
First line therapy:
TNF inhibitors seem not to be cost-effective as first line therapy, based on the one study in which this was considered.
Second line therapy:
We cannot draw any conclusions, since no relevant studies were found.
Third line therapy:
TNF-inhibitors may be cost-effective, particularly in the case of patients in early disease. The drugs are also likely to be more cost-effective for patients who experience a good rather than a moderate response.
Indirect costs:
Prevention of productivity loss may account for considerable savings, but has only been accounted for in a few of the economic evaluations.
Analysis of six month quality of life data:
Data from Norway indicate that RA patients on TNF + methotrexate (MTX) on average experience a larger improvement in health-related quality of life than patients on MTX monotherapy who in turn had a larger improvement than those on TNF monotherapy. Differences in patient groups concerning severity of disease and MTX tolerance should however, be taken into consideration.
Summary
Introduction
Treatment with tumour necrosis factor alpha (TNF-α, or simply TNF) inhibitors is considered to be an alternative to the use of traditional disease-modifying anti-rheumatic drugs (DMARDs) in patients with different rheumatic diseases, i.e. rheumatoid arthritis (RA). There are three TNF-inhibitor drugs currently available on the market (brand names in brackets): adalimumab (Humira), etanercept (Enbrel) and infliximab (Remicade). The drugs are often referred to as biologics, along with other medicines derived from living organisms that target specific receptors in the immune system.
The Norwegian Knowledge Centre for the Health Services’ review of TNF-inhibitors for rheumatoid arthritis
The Norwegian Knowledge Centre for the Health Services (NOKC) has conducted a systematic review of the available evidence on effectiveness and safety connected to the use of TNF inhibitors for the treatment of rheumatoid arthritis, psoriasis arthritis and Bechterew’s Disease (ankylosing spondylitis). The review has been published in two reports. These were concerned with data on effectiveness and adverse events from randomised clinical trials and observational studies respectively.
Objectives
The third and present report’s mandate states that "the cost-effectiveness of TNF-inhibitors shall also be considered". Cost-effectiveness analysis is an economic tool that can be used to guide priority-setting in the health care sector. In this context it is taken to mean the assessment of whether the drugs represent value for money compared to alternative treatment. The value or benefit refers to the potential impact the drugs have on health-related quality of life and survival. Money refers to the effect on costs, or resource use, both inside and outside the health sector. Given the significantly higher prices of TNF-antagonists compared to those of traditional DMARDs, the primary objective is therefore to examine whether the additional resources spent following the use of the TNF-inhibitors are in reasonable proportion to the expected added benefits.
The secondary objective is to investigate whether the TNF-inhibitors are cost-effective earlier rather than later in the treatment sequence. In other words, are there any significant differences between first, second and third line therapy?
Structure of the report
The main focus of the analysis is a literature review of economic evaluations of TNF-inhibitors for RA (not psoriasis arthritis or Bechterew’s). In addition, the report comprises sections on the economic burden of RA. The purpose is to illustrate the extent of resources lost to society due to the disease, and the magnitude and composition of TNF-inhibitor and DMARD consumption in Norway. The section on economic evaluation is followed by an analysis of the 6-month difference in health related quality of life associated with TNF and DMARD use in Norwegian clinical practice, as recorded in the NOR-DMARD observational study.
Economic burden of RA
Estimates of the total cost to society due to RA vary from country to country but no published studies are as yet available for Norway. Moreover, few specific statistics are readily available with regard to the costs associated with the disease in this country. We thus have to look to
our neighbouring country Sweden for more comprehensive cost-of-illness reviews. A study carried out at the University of Linköping for the Swedish RA patient association, found that the total cost associated with RA (broadly defined as inflammatory joint disease)1 amounted to SEK 8,5 billion (NOK 7,4 billion) in 2001, or NOK 7.8 billion in 2005 money terms. The bulk of the costs was indirect in kind and related to sick leave and early retirement
TNF inhibitors in Norway
The use of TNF-inhibitors in RA and related diseases has grown rapidly over the last decade. They were used by approximately 23% of RA patients in Norway in 2005. A report estimated the number of potential recipients of TNF inhibitors across all diagnoses in Norway to be 11 500. A crude estimate based on TNF-inhibitor sales statistics, assuming continuous consumption of doses at a given level (defined daily dose), estimated some 7 200 patients (all diagnoses) to be actively treated in 2006. In retail prices, the sales of the three biologics added up to approx. NOK 860 million in 2006, which represents approximately 5 % of total prescription drug sales that year.
Literature review
Cost-effectiveness
Cost-effectiveness is synonymous with the notion of "value for money". Whether an intervention is cost-effective can only be judged by comparing costs and outcomes of that intervention with those associated with an alternative course of action, or simply the status quo, "doing nothing". To determine whether something is cost-effective or not, one may employ economic evaluation. This is the process of identifying, measuring, valuing and comparing costs and outcomes of alternative interventions or strategies, The costs do not only encompass the price of the drugs, but also the wider consequences for the health services (e.g. adverse events, consultations and hospitalisation) or for society at large (ability to go to school or work, need for assistance from friends and family). The outcomes may be expressed in several ways, but the most relevant in this context is a combined measure of expected survival time and health-related quality of life (HRQoL) indicators known as a quality-adjusted life year (QALY). The HRQoL, or utility, component of a QALY involves assigning values to a health states in the range between 0 (corresponding to death) to 1 (corresponding to perfect health).
Results
The results of an economic evaluation are often expressed as the incremental cost-effectiveness ratio (ICER), the change in costs per additional unit of benefit brought about by moving from one strategy (the comparison strategy) to another (the intervention strategy). In a cost-utility analysis, the ICERs are the extra costs incurred for an additional QALY gained compared to those resulting from the standard strategy.
We have reviewed twelve studies, some of which included results for more than one TNF-inhibitor-. The studies were from the UK, Sweden, Canada, the Netherlands, Japan and the US. The primary objective was to assess whether TNF-inhibitors in general are likely to be cost-effective compared to DMARDs in patients with RA. The median ICERs from all the studies included was NOK 443 000, which means that half of the results were within the range of what is by many considered cost-effective. There is a great deal of variation in the
results, as the ICERs ranged from NOK145 000 to above 8.1 million. The biologics may hence not be cost-effective at all stages of treatment and for all patient subpopulations.
The secondary objective was to investigate if there were any significant differences in cost-effectiveness between first, second and third line therapy with TNF-inhibitors. Our results suggest the following:
First line therapy:
Based on the one study that covered this question, TNF-inhibitors do not appear to be cost-effective as first line therapy, compared to MTX.
Second line therapy:
We have not found evidence in the literature with regard to the health economic consequences of second line treatment with TNF-inhibitors.
Third line therapy:
The results vary significantly, but it seems that TNF-inhibitors may be cost-effective compared to traditional DMARDs particularly if used by patients in the early stages of disease (3 years or less), or by patients with a good response.
Analysis from the NOR-DMARD observational study
To supplement the results with summary measures of benefit relating to Norwegian clinical practice, we commissioned an analysis of data from the NOR-DMARD observational study. We sought to find out whether there were any significant differences in the HRQoL and work capacity of patients on TNF-monotherapy, TNF + MTX and MTX alone or in combination with other DMARDs. Analysis of six-month follow up data based on the propensity score method revealed that, on average, patients on TNF + MTX and on MTX monotherapy experiences a clinically relevant improvement in quality of life, whereas patients on TNF monotherapy and MTX + DMARDs did not. TNF + MTX was associated with a slightly larger improvement than MTX alone, which in turn involved a larger improvement than TNF monotherapy. The latter is however, given to selected patient groups, who may be MTX intolerant. Employment status after 6 months was investigated, and presented as the share of patients above and below a 50 % employment threshold, using the last observation carried forward method. No significant changes could be observed.
Conclusion
In our review of economic evaluations of TNF-inhibitors, we found significant variation in the type and features of the models used, which led to a wide range of estimates. The potential for direct comparisons of results between the studies, and thus transferability of results into Norwegian setting, is limited. With this in mind, our main conclusions are as follows:
First line therapy:
Based on the results of a single study, TNF inhibitors do not appear to be cost-effective as first line therapy
Second line therapy: We
cannot conclude, since no relevant studies were found.
Third line therapy:
TNF-inhibitors may be cost-effective, particularly in the case of patients in early disease. The drugs are also likely to be more cost-effective for patients who experience a good rather than a moderate response.
Indirect costs:
Prevention of productivity loss may account for considerable savings, but has only been accounted for in a few of the economic evaluations.
Analysis of six-monthly change in quality of life:
Data from Norway indicate that RA patients on TNF + MTX on average experience a larger HRQoL improvement than patients on MTX monotherapy, who in turn had a larger improvement than those on TNF monotherapy. Differences in patient groups concerning severity of disease and MTX tolerance should however, be taken into consideration.