Health Technology Assessment
Medicines used for Multiple Sclerosis. A Health Technology Assessment
Health technology assessment
|Updated
This report assessed the effect and cost-effectiveness of disease-modifying medicines used in Norway for patients with relapsing remitting multiple sclerosis.
Download
Key message
This Health Technology Assessment was commissioned by the “National system for the introduction of new health technologies within the specialist health service”. The aim of this report was to assess the effect and cost-effectiveness of the disease modifying medicines used in Norway for patients with relapsing remitting multiple sclerosis (dimethyl fumarate, teriflunomide, interferon beta, peg-interferon, glatiramer acetate, natalizumab, fingolimod, and alemtuzumab).
The key results are:
- We identified 37 randomised clinical trials. The quality of the available evidence ranged from very low to high.
- Alemtuzumab 12 mg had the best effect on annual relapse (for medicines we had evidence of high quality). Dimethyl fumarate 240 mg twice daily and fingolimod oral 0.5 mg were the most effective against disability progression (for medicines we had evidence of high quality).
- Our results indicated that interferon beta-1a 44 mcg and peg-interferon beta-1a were associated with more withdrawal due to adverse events than placebo. The examined treatments had no effect on mortality compared to placebo.
- Our health economic analysis, examining all multiple sclerosis treatment alternatives, indicated that alemtuzumab was more effective (in terms of quality-adjusted life-years (QALY)) and less costly than the other treatment alternatives. We did several scenario analyses and the cost-effectiveness results were robust to variations in the model assumptions.
- The results of a scenario analysis that excluded alemtuzumab (the dominant strategy), showed that three treatments alternatives (interferon beta-1b (Extavia), peg-interferon beta-1a and natalizumab) could be cost-effective depending on the willingness-to-pay (WTP) per QALY. Assuming a WTP below NOK 1,000,000, interferon beta-1b (Extavia) was 40% likely to be the most cost-effective treatment, followed by peg-interferon beta-1a (30% likely).
- The results of our model analysis showed that there is some degree of uncertainty regarding the input parameters. More research on efficacy and epidemiological data would have the greatest impact on reducing decision uncertainty.
- Our budget impact analysis based on the results of our cost-effectiveness analysis, the drugs’ adverse events profile, and current clinical practice showed that there is a substantial potential for cost saving.
Summary
Background
Several disease-modifying therapies are available for the treatment of multiple sclerosis, but the comparative clinical effectiveness of these medicines is unclear. Furthermore, the cost-effectiveness of the different treatments has not been investigated in a Norwegian setting. To ensure the most appropriate multiple sclerosis management, it is important to assess effectiveness and cost-effectiveness of disease modifying medicines used for multiple sclerosis.
Objective
The aim of this project was to compare the effect and cost-effectiveness of the disease modifying medicines used for multiple sclerosis in Norway.
Methods
We conducted a systematic review based on the following conditions: Evidence should come from randomised controlled trials (RCTs) with study populations that included men and women aged 18 years or older were eligible. Modifying medicines used for multiple sclerosis were our intervention of interest (dimethyl fumarate, teriflunomide, interferon beta, peg-interferon, glatiramer acetate, natalizumab, fingolimod, and alemtuzumab). We included studies that compared these medicines to placebo or to each other. We examined the following endpoints: annual relapse, disability progression, mortality, serious adverse events, withdrawal from the study due to adverse events, hospitalisations, and health related quality of life.
We systematically searched the literature for previously published health technology assessment reports or systematic reviews that answered our objectives, and met our inclusion criteria. We conducted a systematic review of randomised controlled trials to supplement the evidence of previously published health technology assessments.
Two persons independently examined the risk of bias of included studies using the Norwegian Knowledge Centre for the Health Services methods. These are based on Cochrane methodology.
We summarised the evidence from the randomised clinical trials quantitavely through network meta-analyses of data on direct and indirect evidence on all relevant comparisons.
Two persons independently assessed the quality of the evidence for each selected endpoint. We used GRADE (Grading of recommendations Assessment, Development, and Evaluation) to assess our confidence in the effect estimates.
In order to assess the cost-effectiveness of disease-modifying therapies in patients diagnosed with relapsing-remitting multiple sclerosis, we developed a decision analytic model. The economic model was developed in the form of a cost-utility analysis and included treatments approved and available in Norway. The model structure and all assumptions were adapted to the Norwegian setting based on Norwegian clinical practice. Efficacy estimates were taken from our network meta-analyses. Transitional probabilities were derived from published sources and clinical experts’ opinions. Quality of life data were extracted from published studies based on a systematic review of the literature. The costs of medications were based on prices obtained through the Drug procurement co-operation (LIS), and other costs were based on official Norwegian unit prices.
We performed probabilistic sensitivity analyses, designed as a Monte Carlo simulation with 10,000 iterations, to explore the uncertainty surrounding our results.
Results
All examined treatments were more effective than placebo against annual relapse. The effect was best for alemtuzumab 12 mg (based on high quality evidence). Fingolimod oral 0.5 mg and dimethyl fumarate 240 mg twice daily were also associated with a reduction in annualised relapse rate. For disability progression, dimethyl fumarate 240 mg twice daily and fingolimod 0.5 mg were more effective than placebo (high quality evidence).
For withdrawal due to adverse events, the conclusion is unclear due to the low quality of the available evidence. However, our results indicate that interferon beta-1a 44 mcg, and peg-interferon beta-1a are associated with more withdrawal due to adverse events than placebo.
For the outcomes change in expanded disability status scale, serious adverse events, and mortality; we did not assess the quality of the available evidence. Our results indicate that interferon beta-1a 30 mcg is associated with a reduction in expanded disability status scale. Interferon beta-1a 30 mcg is associated with fewer serious adverse events. Finally, our results showed that none of the examined treatments increased or decreased mortality compared to placebo.
Our health economic analysis indicated that alemtuzumab dominated all other disease-modifying therapies, as it was more effective in terms of quality-adjusted life-years (QALY) and less costly than the other treatment alternatives.
A scenario analysis that excluded alemtuzumab (the dominant strategy) showed that three treatment alternatives (interferon beta-1b (Extavia), peg-interferon beta-1a and natalizumab) could be cost-effective depending on the willingness-to-pay (WTP) threshold. Interferon beta-1b was likely to be the cost-effective choice for a WTP per QALY below NOK 1,658,000. Peg-interferon was the cost-effective option for a WTP from NOK 1,658,450 to NOK 10,619,960, and natalizumab was the cost-effective alternative for a WTP above NOK 10,619,960. Assuming a WTP below NOK 1,000,000 per QALY, interferon beta-1b (Extavia) was approximately 40% likely to be the most cost-effective treatment, followed by peg-interferon beta-1a (approximately 30% likely).
The results of probabilistic analysis showed that there is some degree of uncertainty regarding the input parameters. More research on efficacy and epidemiologic input parameters would have the greatest impact on reducing decision uncertainty.
We performed several scenario analyses to test the uncertainty around the model assumptions. The results showed that, while there were numerical changes to the incremental cost-effectiveness ratio, the cost-effectiveness results were robust to variations in the model assumptions and the conclusions of the analysis would not change.
Our budget impact analysis based on the results of our cost-effectiveness analysis, the drugs’ adverse events profile, and current clinical practice showed that there is a substantial potential for cost saving.
Discussion
We used a systematic methodology to search for evidence, extract data, and assess the risk of bias of studies and quality of evidence for important outcomes. The systematic review included evidence on both established and emerging treatments. We examined the effect of these treatments on clinical endpoints relevant for patients with multiple sclerosis. We have analysed direct and indirect evidence through network meta-analyses. The consistency of results using different methods indicates that our results are robust.
Our systematic review has some limitations, due more to the weakness of the available evidence than to the methods used in this report. These limitations are related to the paucity and quality of the available evidence, and to the methodologies used in the included randomised controlled trials.
We used a probabilistic Markov-model, considered the appropriate approach for simulating the natural history of multiple sclerosis. The model structure and all assumptions have been adapted to the Norwegian setting based on Norwegian clinical practice with close assistance of experts in this field.
For transitional probabilities, we did not find Norwegian data that were compatible with the developed model, so these were based on estimates reported in the published literature.
Study designs of published trials did not permit separate analyses of first and second line treatments, or conclusions regarding the sequential use of first and second line treatments. Therefore, we did not perform separate cost-effectiveness analyses for first or second line treatments. In addition, based on expert opinion, we did not include combination therapy in our model, as it is not relevant to current Norwegian clinical practice.
Conclusion
Alemtuzumab 12 mg had the best effect against annual relapse. Dimethyl fumarate 240 mg twice daily and fingolimod oral 0.5 mg were the most effective against disability progression. Results indicate that some treatments are associated with more withdrawals due to adverse events than placebo. Our results showed that the examined treatments had no effect on mortality.
Our health economic analysis indicated that alemtuzumab was more effective and less costly than the other treatment alternatives. A scenario analysis that excluded alemtuzumab indicated that three treatment alternatives (interferon beta-1b (Extavia), peg-interferon beta-1a and natalizumab) could be cost-effective depending on the WTP. For a WTP below NOK 1,000,000 per QALY, interferon beta-1b (Extavia) was approximately 40% likely to be the most cost-effective treatment, followed by peg-interferon beta-1a (approximately 30% likely).
The results of probabilistic analysis showed that there is some degree of uncertainty regarding the input parameters. More research on efficacy and epidemiologic input parameters would have the greatest impact on reducing decision uncertainty.
Our budget impact analysis showed that there is a substantial potential for cost saving.
