Cost-effectiveness of Delayed-Release Dimethyl Fumarate Compared With Glatiramer Acetate and Fingolimod for the Treatment of Relapsing-Remitting Multiple Sclerosis Josephine Mauskopf,1 Monica Fay,2 Ravi Iyer,2 Sujata Sarda,2 Terrie Livingston2 1 RTI Health Solutions, Research Triangle Park, NC, United States; 2 Biogen Idec Inc., Weston, MA, United States

OBJECTIVE Table 4. Multiple Sclerosis Health State (EDSS) Annual Costs and QALYs

• To estimate the cost-effectiveness of delayed-release dimethyl fumarate compared with glatiramer EDSS State acetate and fi ngolimod in treatment of relapsing-remitting multiple sclerosis (RRMS) in the United 0 1 2 3 4 5 6 7 8 9 States (US). Costsa $6,297 $7,072 $7,847 $8,623 $9,397 $10,172 $14,806 $19,441 $24,075 $28,708

QALYs (RRMS, no 0.8752 0.8342 0.7802 0.6946 0.6253 0.5442 0.4555 0.3437 0.0023 –0.1701 BACKGROUND relapse)b

a Source: Kobelt et al. (2006)5 infl ated to 2013 values using linear interpolation from values for EDSS < 4, EDSS 4-6, and • In a prespecifi ed integrated analysis of two phase 3 clinical trials, delayed-release dimethyl fumarate EDSS > 6. was shown to be associated with statistically signifi cant improvements in annualized relapse rates b Source: Delayed-release dimethyl fumarate clinical trial data (DEFINE and CONFIRM) and Orme et al. (2007).6 and risk of Expanded Disability Status Scale (EDSS) progression. Note: QALYs in those with SPMS were assumed to be 0.0092 lower than for those with RRMS, and occurrence of a relapse during the year was assumed to reduce QALYs by 0.0437 for both those with RRMS and SPMS.

METHODS • One-way sensitivity analyses were performed changing input parameter values and model assumptions. • A cohort-based Markov model tracking patients through EDSS health states with a cycle time of 1 year was developed in Microsoft Excel. Each year patients can transition to better or worse EDSS health states, have a relapse, transition from RRMS to secondary progressive multiple sclerosis (SPMS), and/or die. Disease-modifying treatment (DMT) was assumed to reduce the rate of EDSS RESULTS progression and the annual relapse rate but not the rate of transition from RRMS to SPMS (Figure 1). • Annual discontinuation rates were assumed to be the same for all DMTs (10% in years 1 and 2 and 3% • Results were presented for a 10-year time horizon. in all subsequent years). All patients were assumed to stop DMT when their EDSS level reached 7. • Compared with glatiramer acetate and fi ngolimod, delayed-release dimethyl fumarate increased • The model was designed to estimate the discounted (at 3%) cost and quality-adjusted life-years QALYs by 0.205 and 0.156 QALYs, respectively, and was less costly by $9,879 and $16,462, (QALYs) associated with treatment with delayed-release dimethyl fumarate compared with respectively (Table 5). glatiramer acetate or fi ngolimod in RRMS. • Thus, delayed-release dimethyl fumarate was dominant (had lower costs and higher QALYs gained) compared with glatiramer acetate and fi ngolimod (Table 5).

Figure 1. Schematic of the Model Table 5. Discounted (3.0%) Cost-effectiveness Model Outcomes per Patient: 10-Year Costs and Outcomes

Delayed-Release Dimethyl Fumarate EDSS 0 0.5 to 1 1.5 to 2 2.5 to 3 3.5 to 4 4.5 to 5 5.5 to 6 6.5 to 7 7.5 to 8 8.5 to 9.5 Total Cost Total Total vs. Other DMT Drug (US $) LYs a QALYs Incremental QALYs ICER Costs Gained (Cost/QALY) Death ssion/Reg Delayed- gre ressi Pro on release Relapse $486,220 8.411 5.044 NA NA NA dimethyl RRMS RRMS fumarate Delayed-release EDSS Conversion EDSS Glatiramer $496,099 8.408 4.839 –$9,879 0.205 dimethyl fumarate State N to SPMS State N + acetate dominantb SPMS SPMS Delayed-release Relapse Fingolimod $502,682 8.408 4.888 –$16,462 0.156 dimethyl fumarate dominantb Progression ICER = incremental cost-effectiveness ratio; LY = life years; NA = not applicable. a 10 life-years discounted at 3% is equal to 8.79 discounted life-years. b Delayed-release dimethyl fumarate is less costly and more effective (lower right quadrant in the cost-effectiveness plane). Source: Adapted from Gani et al., 2008.1

Sensitivity analyses showed that delayed-release dimethyl fumarate was less costly and more • Population characteristics matched those in the delayed-release dimethyl fumarate phase 3 clinical • trials (Table 1). effective than glatiramer acetate for most of the input parameter values and assumptions tested (Figure 2). • The untreated rate of transition between the EDSS health states and annualized relapse rates were estimated using data from the placebo arms of the phase 3 clinical trials. • Sensitivity analyses for fi ngolimod also showed that delayed-release dimethyl fumarate was less costly and more effective for most of the input parameter values and assumptions tested. In • The extent to which the DMTs slowed disease progression and reduced annualized relapse rates particular, incremental costs were most sensitive to delayed-release dimethyl fumarate acquisition was estimated using a mixed-treatment comparison (MTC) analysis (Table 2). costs (+10% +$21,356; –10% –$54,280), including informal care costs and productivity losses (–$25,163), and using annual discontinuation rates from the MTC (–$36,630). QALYs were most sensitive to changes in delayed-release dimethyl fumarate impact on disease progression (Upper 95% CI –0.031; Lower 95% CI +0.304) and use of Kobelt utility values (+0.022).5 Table 1. Population Characteristics

Demographic Variable Value Source Figure 2. Sensitivity Analyses for Costs (A) and QALYs (B) for Delayed-Release Dimethyl Delayed-release dimethyl fumarate Fumarate Compared With Glatiramer Acetate Cohort starting age 37.9 clinical trial data (DEFINE and CONFIRM) A. Alternate Values Campbell et al., 20142 Female/male ratio 3.27 DR DMF acquisition costs (+10% or –10%) (using the US MEPS database) MTC discontinuation rates EDSS state direct + informal care costs + productivity losses EDSS 0: 5.05% EDSS 1: 8.52% Discontinue at transition to SPMS Delayed-release dimethyl fumarate EDSS 2: 34.08% EDSS state direct + informal care costs clinical trial data, intent-to-treat EDSS distribution EDSS 3: 22.94% 7 population Goldberg relapse costs ($5,477) EDSS 4: 20.64% (DEFINE and CONFIRM) DR DMF relapse relative risk (upper and lower 95% CIs) EDSS 5: 8.65% DR DMF disability progression hazard ratio (upper and lower 95% CIs) EDSS 6: 0.12% No EDSS regression MEPS = Medical Expenditure Panel Survey. Patzold and Pocklington natural history relapse rates8 EDSS state costs (+10% or –10%) No costs or QALYs lost from adverse events First or only alternate value London Ontario natural history disease progression rates Second alternate value

–50,000 –40,000 –30,000 –20,000 –10,000 0 10,000 20,000 30,000 Table 2. Impact of DMTs on Annualized Relapse Rate and Disease Progression Cost ($)

Hazard Ratio Disability Relative Risk Annualized B. Treatment Progression Relapse Rate Alternate Values Mean (95% CI) Mean (95% CI) DR DMF disability progression hazard ratio (upper and lower 95% CIs) MTC discontinuation rates Delayed-release dimethyl fumarate 0.60 (0.43-0.84) 0.53 (0.45-0.62) London Ontario natural history disease progression rates 5 Glatiramer acetate 0.83 (0.64-1.09) 0.65 (0.59-0.72) Kobelt patient utilities No costs or QALYs lost from adverse events Fingolimod 0.80 (0.63-1.01) 0.45 (0.36-0.50) Discontinue at transition to SPMS CI = confi dence interval. DR DMF relapse relative risk (upper and lower 95% CIs) Source: Estimated using an MTC.3,4 No EDSS regression First or only alternate value Patzold and Pocklington natural history relapse rates8 Second alternate value 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 QALY • Costs included drug acquisition costs (wholesale acquisition costs [WACs]), administration and DR DMF = delayed-release dimethyl fumarate. monitoring costs (based on resource use estimates and US unit costs), and the cost and disutility associated with adverse events (based on adverse event rates, resource use for treatment of adverse events, disutility for each serious and nonserious adverse event, and US unit costs) (Table 3). CONCLUSIONS • Cost and patient utility in each EDSS health state were based on published US data (Table 4). • Delayed-release dimethyl fumarate, a new oral drug indicated for RRMS, is a cost- Table 3. DMT-Related Annual Costs and QALYs Lost effective treatment when compared with glatiramer acetate and fi ngolimod. • Sensitivity analyses support the robustness of the model results. Adverse Adverse Treatment Administration Monitoring DMT Event Event Cost (WAC)a Costb Costb Costc QALY Lossc

Delayed-release Year 1 $0 Year 1 $416 $60,122 $165 0.0037 REFERENCES dimethyl fumarate Year 2+ $0 Year 2 $416 1. Gani R, Giovannoni G, Bates D, et al. Cost-effectiveness analyses of 5. Kobelt G, Berg J, Atherly D, Hadjimichael O. Costs and quality of life in multiple sclerosis: a cross-sectional study in the United States. Year 1 $105 Year 1 $327 natalizumab (Tysabri) compared with other disease-modifying Glatiramer acetate $61,736 $143 0.0061 therapies for people with highly active relapsing-remitting multiple Neurology. 2006;66(11):1696-702. Year 2+ $35 Year 2 $327 sclerosis in the UK. Pharmacoeconomics. 2008;26(7):617-27. 6. Orme M, Lerrigan J, Tyas D, et al. The effect of disease, functional 2. Campbell JD, Ghushchyan V, McQueen B, et al. Burden of multiple status, and relapses on the utility of people with multiple sclerosis in Year 1 $0 Year 1 $628 sclerosis on direct, indirect costs and quality of life: National US the UK. Value Health. 2007;10:54-60. Fingolimod $63,270 $230 0.0048 Year 2+ $0 Year 2 $218 estimates. Mult Scler and Relat Disord. 2014;3:227-36. 7. Goldberg LD, Edwards NC, Fincher C, et al. Comparing the cost- 3. Hutchinson M, Fox RJ, Havrdova E, et al. Effi cacy and safety of effectiveness of disease-modifying drugs for the fi rst-line treatment a Drug acquisition costs were estimated as the WAC cost per pack multiplied by the number of packs per year. BG-12 (dimethyl fumarate) and other disease-modifying therapies of relapsing-remitting multiple sclerosis. J Manag Care Pharm. 2009 Sep;15(7):543-55. b Administration and monitoring costs for each DMT were estimated using resource use assumed to be needed to train and for the treatment of relapsing-remitting multiple sclerosis: a systematic review and mixed treatment comparison. Curr Med Res monitor self-administration for the injectable drugs, as well as to monitor for effi cacy and safety through physician visits and 8. Patzold U, Pocklington PR. Course of multiple sclerosis. First results Opin. 2013 Nov 26. Epub ahead of print. laboratory tests as recommended in the product prescribing information; standard US unit costs were applied to the resource of a prospective study carried out of 102 MS patients from 1976- use estimates. 4. Biogen Idec. Cost-effectiveness analysis of Tecfi dera® in patients 1980. Acta Neurol Scand. 1982;65(4):248-66. with relapsing remitting multiple sclerosis. Version 18.0 Technical c Adverse event costs and QALY loss were estimated based on estimated incidence rates for each DMT for the most common Report. Data on fi le. October 2013. adverse events observed in the delayed-release dimethyl fumarate clinical trials, percentage of those adverse events that were serious for each DMT, and resource use and disutility for each serious and nonserious adverse event based on expert opinion; standard US unit costs were applied to the resource use estimates.

ISPOR 19th Annual International Meeting; May 31-June 4, 2014; Montreal, QC, Canada