Cost-Effectiveness of Cladribine Tablets and Fingolimod in the Treatment of Relapsing Multiple Sclerosis with High Disease Activity in Spain

Expert Review of Pharmacoeconomics & Outcomes Research

ISSN: 1473-7167 (Print) 1744-8379 (Online) Journal homepage: https://www.tandfonline.com/loi/ierp20

Cost-effectiveness of Cladribine Tablets and fingolimod in the treatment of relapsing multiple sclerosis with high disease activity in Spain

J. L. Poveda, J. L. Trillo, C. Rubio-Terrés, D. Rubio-Rodríguez, A. Polanco & C. Torres

To cite this article: J. L. Poveda, J. L. Trillo, C. Rubio-Terrés, D. Rubio-Rodríguez, A. Polanco & C. Torres (2019): Cost-effectiveness of Cladribine Tablets and fingolimod in the treatment of relapsing multiple sclerosis with high disease activity in Spain, Expert Review of Pharmacoeconomics & Outcomes Research, DOI: 10.1080/14737167.2019.1635014 To link to this article: https://doi.org/10.1080/14737167.2019.1635014

Accepted author version posted online: 21 Jun 2019. Published online: 25 Jul 2019.

Submit your article to this journal

Article views: 56

View Crossmark data

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=ierp20 EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH https://doi.org/10.1080/14737167.2019.1635014

ORIGINAL RESEARCH Cost-effectiveness of Cladribine Tablets and fingolimod in the treatment of relapsing multiple sclerosis with high disease activity in Spain J. L. Povedaa, J. L. Trillob, C. Rubio-Terrésc, D. Rubio-Rodríguezc, A. Polancod and C. Torresd aPharmacy Department, Hospital Universitario y Politécnico La Fe, Valencia, Spain; bPharmacy Department, Hospital Clínico Universitario de Valencia, Valencia, Spain; cPharmacoeconomics Department, Health Value, Madrid, Spain; dCorporate Affairs Department, Merck, Madrid, Spain

ABSTRACT ARTICLE HISTORY Objective: To estimate the cost-effectiveness of Cladribine Tablets in the treatment of relapsing multi- Received 2 May 2019 ple sclerosis (RMS) with high disease activity compared with fingolimod, from the perspective of the Accepted 19 June 2019 National Health System (NHS) in Spain. KEYWORDS Methods: A Markov model was developed. The annual transition probabilities, were adjusted to patients with Cladribine Tablets; RMS with high disease activity. The effect of the treatments compared on the Expanded Disability Status Scale fingolimod; multiple (EDSS) was modeled by hazard ratios for the confirmed progression of disability. The annual relapse rate and sclerosis; relapsing multiple the probability of suffering adverse reactions were obtained from a meta-analysis and the literature. The sclerosis; cost-effectiveness derived costs were calculated from Spanish unit costs. The utilities were obtained from the CLARITY clinical trial and the literature. Deterministic and probabilistic sensitivity analyzes were performed. Results: Cladribine tablets was the dominant treatment: lower costs (−86,536 €) and more effective (+1.11 quality-adjusted life years – QALYs) compared to fingolimod. The probability that Cladribine Tablets was cost- effective compared to fingolimod ranged between 94.6% and 96.1% for willingness to pay from € 20,000 to € 30,000 per QALY gained. Conclusions: Cladribine Tablets is a cost-effective treatment, compared to fingolimod, for the treatment of RMS with high disease activity. Expert Opinion: According to the present study, compared to fingolimod, treatment with Cladribine Tablets of relapsing multiple sclerosis with high disease activity is an option that could generate savings for the Spanish National Health System, with a considerable gain in QALYs. Cladribine Tablets is considered cost-effective and dominant (less costs and more effectiveness) than fingolimod treatment option in this population.

1. Introduction (EMA) has approved fingolimod for the treatment of RMS with high disease activity for the following groups of patients: (i) Multiple sclerosis (MS) is a disease of the central nervous system Patients with highly active disease despite a full and adequate characterized by inflammation, demyelination and degenerative course of treatment with at least one disease modifying therapy; changes [1]. It usually begins at between 20 and 40 years of age and (ii) Patients with rapidly evolving severe relapsing remitting and affects women tree times more as men, being the most multiple sclerosis defined by 2 or more disabling relapses in frequent cause of non-traumatic disability in the young adult one year, and with 1 or more Gadolinium enhancing lesions on population [1,2]. According to three recent studies, the prevalence brain MRI or a significant increase in T2 lesion load as compared to of MS in Spain is estimated at between 65 and 90 cases per 100,000 apreviousrecentMRI[10]. The EMA has also recently approved inhabitants [3–5]. Cladribine Tablets in the treatment of adult patients with RMS with According to a cross-sectional Spanish study [6], 92% and 64% high disease activity defined by clinical or imaging features [11]. of patients with MS analyzed experienced fatigue and cognitive The objective of this study was to assess the cost- problems, respectively, with a mean utility (quality of life as per- effectiveness of both disease-modifying treatments (DMTs), ceived by the patient) of 0.772 and costs per patient per year (from Cladribine Tablets and fingolimod, in the treatment of RMS the societal perspective) ranging between 20,600 € with a score in with high disease activity, from the perspective of the National the Expanded Disability Status Scale (EDSS) of 0 to 3 and 68,700 € Health System (NHS) in Spain. with an EDSS score of 7 to 9. In Spain, the total annual cost of MS has been estimated at 1,4 billion euros (40% direct healthcare cost, 30% direct non-healthcare cost and 30% indirect cost) [7]. 2. Methods The majority of patients with MS, between 85% and 90%, are 2.1. Markov model diagnosed with relapsing MS (RMS) characterized by periods of acute exacerbation (relapses) followed by periods of remission of The characteristics of this model have been published pre- neurological symptoms [6,8,9]. The European Medicines Agency viously [12]. The analysis has been based on a Markov model

CONTACT D. Rubio-Rodríguez [email protected] Health Value, Health Economics & Research of Outcomes Consulting, C/Virgen de Aránzazu, 21. 5º B., Madrid 28034, Spain This article has been republished with minor changes. These changes do not impact the academic content of the article. © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. 2 J. L. POVEDA ET AL.

status); (iii) maintain their status (remain in the EDSS status in Article Highlights which they are); or (iv) die. Table 1 shows the annual probabilities of transition between ● The economic model simulated the evolution of a cohort of patients with relapsing multiple sclerosis (RMS) with high disease activity the EDSS statuses, adjusted to the RMS with high disease activity treated with Cladribine Tablets or fingolimod. [12,16]. Table 2 offers a summary of the other premises and values ● Treatment with Cladribine Tablets was cost-effective, generating adopted in the model [10,12,15,17–33]. At the start of the analysis, lower costs (−86,536 €) and demonstrating greater effectiveness (+1.11 QALYs) than treatment with fingolimod, demonstrating that the cohort of patients was distributed among the different health Cladribine Tablets is a dominant treatment in each patient with RMS statuses (EDSS scores) according to the distribution observed in with high disease activity. the CLARITY phase III clinical study [20]. The population parameters ● The saving was mainly due to the lower drug related cost: acquisition (in line with the Summary of Product Characteristics, Cladribine of the cohort were adjusted to the patients of the CLARITY study, Tablets is administered a maximum of 20 days in two courses of with an average age of 37.1 years[20]. The average body weight treatment), administration (Cladribine Tablets does not require the was adjusted to the specific Spanish population, taking into use of specific health resources for its administration) and monitoring (the use of resources needed to monitor Cladribine Tablets is account the average weight (68 kg) of three Spanish cohorts of reduced). patients with MS (total N = 701) [3,5,34](Table 2). ● The gain in QALYs was due to the delay in Expanded Disability Status It was assumed that patients with EDSS statuses between 0 Scale (EDSS) progression with Cladribine Tablets vs. fingolimod. ● The probability that Cladribine Tablets was cost-effective compared and 6 were receiving treatment with a DMT, unless discontin- to fingolimod ranged between 94.6% and 96.1% for willingness to ued due to tolerability issues or the appearance of relapses € € pay from 20,000 to 30,000 per QALY gained. (development of secondary progressive MS [SPMS]). Patients ● According to the present study, compared to fingolimod, treatment with Cladribine Tablets of RMS with high disease activity is an option with EDSS statuses between 7 and 9 were assumed to not be that could generate savings for the Spanish National Health System receiving treatment with a DMT, in line with the usual clinical (NHS), with a considerable gain in QALYs in RMS patients. Cladribine practice and current recommendations [35]. Tablets is considered a cost-effective and dominant (generating less costs and more effectiveness than fingolimod) treatment option in Although it may be the case that in routine clinical practice this patient population. patients may be switched to a different DMT following discontinuation (depending on the patient’s clinical history prior to the interruption), the model did not consider such sequen- cing in the absence of evidence to simulate the efficacy and simulating the evolution of a cohort of patients with RMS safety of these therapies in patients who stop treatment. with high disease activity through different clinical situations The time horizon (duration) of the model was the life of the (statuses of health), defined by the EDSS scale (10 statuses), MS patient (50 years in an initial population with 37 years of whichestablishesthedegreeofdisabilityandtheprogres- age). An annual discount rate of 3% was applied both to the sion of the disease. Each health status has associated costs costs and to future effects (quality-adjusted life years [QALYs]) and specific clinical consequences. The health statuses of the following the current recommendations in Spain [36]. model are mutually exclusive; each patient in a hypothetical cohort can only be in one health status at any given time. 2.2. Clinical data Figure 1 shows the simplified scheme of the model, in which the transition between statuses and the relevant clinical and The probabilities of transition between statuses were based on economic consequences were evaluated in cycles of 1-year the natural history of the disease [16] and were adjusted with [12–15]. For each annual cycle, the patients in each status the data from the CLARITY study [20] to represent the sub- can: (i) worsening in their illness (transition to a higher EDSS group of patients with RMS with high disease activity (Table 1) status); (ii) improve their status (transition to a lower EDSS [12]. The effect of both DMTs (Cladribine Tablets and

Under treatment On DMD

EDSS EDSS EDSS EDSS EDSS EDSS EDSS EDSS EDSS EDSS 0 1 2 3 4 5 6 7 8 9

EDSS 10 Other causes of discontinuation Discontinuation due to DEATH SPMS/EDSS progression Without treatment

Off DMD

EDSS EDSS EDSS EDSS EDSS EDSS EDSS EDSS EDSS EDSS 0 1 2 3 4 5 6 7 8 9

Figure 1. Markov Model of 11 statuses (Hettle, 2018). Patients with RMS with high disease activity. EDSS: Expanded Disability Status Scale; RMS: relapsing multiple sclerosis; SPMS: Secondary progressive multiple sclerosis. EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 3

Table 1. Probabilities (%) per year of transition between EDSS statuses (age of applying the deduction according to the Royal Decree-Law 8/ ≥ MS onset 28 years), adjusted to the RMS with high disease activity [12,16]. 2010 [39]. The annual use of health resources (drug adminis- EDSS: Expanded Disability Status Scale; RMS: relapsing multiple sclerosis. tration and monitoring) was estimated according to the EDSS from/ recommendations of the Summary of Product Characteristics to 0 1–1.5 2–2.5 3–3.5 4–4.5 5–5.5 6–6.5 7–7.5 8–8.5 9–9.5 (SmPC) of Cladribine Tablets and fingolimod [10,11] and vali- 0 58.3 27.8 9.9 3.0 0.6 0.2 0.2 0.0 0.0 0.0 dated by an Advisory Board (Economic value of Cladribine – 1 1.5 5.8 59.8 22.0 8.5 2.3 0.6 0.9 0.1 0.0 0.0 Tablets: 4 July 2018 Bilbao; 17 July 2018 Valencia). Unit costs 2–2.5 1.6 12.1 50.9 23.3 6.2 2.6 3.0 0.2 0.1 0.0 3–3.5 0.6 5.0 12.0 43.8 12.6 8.1 16.1 1.4 0.5 0.0 of associated healthcare resources were obtained from 4–4.5 0.2 2.2 6.7 11.5 37.7 14.2 23.0 3.5 0.9 0.1 Spanish sources [22,23,31,40](Table 2). 5–5.5 0.1 0.5 2.9 5.9 8.7 36.8 37.1 5.3 2.6 0.1 6–6.5 0.0 0.1 0.4 2.5 3.1 4.1 67.4 15.5 6.2 0.6 7–7.5 0.0 0.0 0.1 0.2 0.7 0.4 11.7 69.3 16.1 1.6 8–8.5 0.0 0.0 0.0 0.0 0.1 0.1 1.9 5.6 90.3 2.1 2.5. Sensitivity analysis 9–9.5 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.6 17.4 81.8 Deterministic and probabilistic analyses were carried out. Each parameter in the model varied between the limits of their 95% fingolimod) on the natural history of the MS was derived from confidence intervals or of their credible intervals, or 50% of the results of a network meta-analysis [15] in which we esti- the means when the variances were not available [12]. The mated the hazard ratio of 3-month confirmed disability pro- probabilistic analysis was performed using a Monte Carlo gression for Cladribine Tablets or fingolimod versus placebo, simulation of the second order, with a hypothetical cohort of as well as the annualized relapse rate and the probability of 1,000 patients. treatment withdrawal [15,20](Table 2). This meta-analysis There were also deterministic analyses for different scenar- included four clinical trials of fingolimod [19,21,24,26] and ios: (i) time horizon of 10, 20, 30 and 40 years; (ii) annual a clinical trial of Cladribine Tablets [20], all of which were discounts of costs and benefits of 0% and 6%; and (iii) exclud- controlled with placebo. [15] The transition probabilities ing the loss of utilities of caregivers of patients with RMS with were assumed to be constant over time. The occurrence of high disease activity. adverse events related to the treatments (progressive multifocal leukoencephalopathy, serious infections, macular oedema, cancer) were also considered in the analysis, obtain- 3. Results ing their odds from several previously published studies 3.1. Deterministic base case [12,15,18,19,21,24,26,27]. The probability of death from any cause was assumed to be Treatment with Cladribine Tablets was cost-effective, generat- dependent on age, starting from the general mortality of the ing lower costs (−86,536 €) and demonstrating greater effec- Spanish population [37] and was adjusted according to the tiveness (+1.11 QALYs) than treatment with fingolimod, specific mortality of the MS. [38] demonstrating that Cladribine Tablets is a dominant treatment in each patient with RMS with high disease activity (Table 3). Table 4 presents the disaggregated results of the analysis. The 2.3. Utilities saving was mainly due to the lower drug related cost: acquisi- Utility is a measure of the preference for, or desirability of, tion (in line with the SmPC, Cladribine Tablets is administered a specific level of health status or specific health outcome. The a maximum of 20 days in two courses of treatment [10]), health state utility is a cardinal number, between 0 (death) and administration (Cladribine Tablets does not require the use 1 (perfect health) associated with a particular health state. of specific health resources for its administration [10]) and Utilities for the EDSS statuses 0 to 5 were obtained from the monitoring (the use of resources needed to monitor CLARITY clinical trial [20]. Utilities for the EDSS statuses 6–8 Cladribine Tablets is reduced [10]). The gain in QALYs was and 9 were taken from the studies of Hawton et al. [29] and due to the delay in EDSS progression with Cladribine Tablets Orme et al. [17], respectively. The loss of utilities associated vs. fingolimod. with the adverse effects was also obtained from the medical literature [28,30,33]. Finally, the loss of utilities that occurs in caregivers of patients with MS was taken from the study of 3.2. Deterministic sensitivity analysis Acaster et al. [25](Table 2). As can be seen in the tornado diagram shown in Figure 2, for all the variables analyzed, there was a net positive health 2.4. Costs benefit, compared to fingolimod, with Cladribine Tablets being the dominant treatment. Costs included in the model were the following: drug related Table 5 presents the results of the deterministic sensitivity costs (acquisition, administration and monitoring), costs based analysis for different scenarios for the time horizon, the annual on disability and adverse events management costs. discount costs and QALYs, excluding the loss of utilities in The annual and four-year cost of the DMTs was obtained caregivers of patients with RMS with high disease activity from the authorized public sale prices of each pharmaceutical from the analysis. In all analyses Cladribine Tablets was the company [32], the recommended dosing regimens [10,11] and dominant treatment compared to fingolimod. 4 .L OEAE AL. ET POVEDA L. J.

Table 2. Premises and values adopted in the economic model. Item Cladribine Tablets Fingolimod Source Cohort of patients Population parameters of the cohort Mean age 37.1 ± 0.83 years Giovannoni, 2010[20] Women/men ratio 1.709 “ Number of relapses in the previous year 2.02 ± 0.4 “ Mean weight of the patient 68,0 Bártulos, 2015[15];Carreón, 2016[5]; Llaneza, 2016[34]; Initial distribution of patients Giovannoni, 2010[20] EDSS 0 2.77% “ EDSS 1 2.77% “ EDSS 2 32.53% “ EDSS 3 21.45% “ EDSS 4 23.53% “ EDSS 5 11.07% “ EDSS 6 5.88% “ EDSS 7 0.00% “ EDSS 8 0.00% “ EDSS 9 0.00% Probabilities Hazard ratio for confirmed progression of disability at 0.281 0.620 Siddiqui, 2017[15] 3 months (vs. placebo) (95% CrI 0.148–0.532) (95% CrI 0.368–1.041) Annualized Relapse Rate (vs. placebo) 0.350 0.375 Siddiqui, 2017[15] (0.243–0.505) (0.269–0.507) Probability of treatment withdrawal Giovannoni, 2010[20]; 0–1 years 4.9% 10.3% Siddiqui, 2017[15] 1–2 years 0.0% 10.3% Following years 0.0% 10.3% Likelihood of suffering an AR Tan, 2009[18]; Cohen, 2010[19]; Kappos, 2010[21]; Saida, 2012[24]; Calabresi, 2014[26]; Pakpoor, 2015[27]; Progressive multifocal leukoencephalopathy 0.00% 0.01% Siddiqui, 2017[15]; Hettle, 2018[12] Serious infections 2.82% 2.23% Macular oedema 0.00% 0.39% Cancer 0.60% 9.60% Utility Utility of the patients according to the status EDSS 0 0.906 Giovannoni, 2010[20] EDSS 1 0.845 “ EDSS 2 0.804 “ EDSS 3 0.701 “ EDSS 4 0.655 “ EDSS 5 0.565 “ EDSS 6 0.496 Hawton, 2016[29] EDSS 7 0.392 “ EDSS 8 0.025 “ EDSS 9 –0.195 Orme, 2007[17] (Continued) Table 2. (Continued). Item Cladribine Tablets Fingolimod Source Loss of utility of the caregiver EDSS 0 −0.002 Acaster, 2013[25] EDSS 1 –0.002 “ EDSS 2 –0.002 “ EDSS 3 –0.045 “ EDSS 4 –0.142 “ EDSS 5 –0.160 “ EDSS 6 –0.173 “ EDSS 7 –0.030 “ EDSS 8 –0.095 “ EDSS 9 –0.095 “ Loss of utility by AR Progressive multifocal leukoencephalopathy −0.0510 NICE TA312[33]; Serious infections –0.0073 Shingler, 2015[28]; Macular oedema –0.0092 NICE TA312[33]; Cancer –0.1160 Trogdon, 2016[30] Healthcare resources Annual use of healthcare resources per patient Administration All years Year 1/Years 2+ Mavenclad, 2018[11]; Day hospital visit 0 1/0 Gilenya, 2018[10] Monitoring Year 1/Year 2 Year 1/Years 2+ Basic analysis (biochemistry) 0/0 6/2 Mavenclad, 2018[11]; Complete blood count (CBC) 3/3 4/2 Gilenya, 2018[10] XETRVE FPAMCEOOIS&OTOE RESEARCH OUTCOMES & PHARMACOECONOMICS OF REVIEW EXPERT MRI 1/0 1/0 Consulting the neurologist 2/1 3/1 Consulting the ophthalmologist 0/0 1/0 QuantiFERON 1/0 0/0 Costs Drug acquisition cost (1) Years 1 and 2 All years BotPlus, 2018[32]; Mavenclad, 2018[11]; Gilenya, 2018[10] Annual 25,197.00 € 19,292.86 € Four-year 50,394.00 € 77,171.00 € Drug administration costs (2017) Day hospital visit € 201 eSalud, 2017[31] Drug monitoring costs (2017) Basic analysis (biochemistry) € 90.70 eSalud, 2017[31] Complete blood count (CBC) € 4.16 “ MRI € 291.77 “ Consulting the neurologist € 84.32 “ Consulting the ophthalmologist € 84.32 “ QuantiFERON* € 8.55 “ Cost based on disability EDSS 0 to EDSS 3 € 2,833.65 Karampampa, 2012[23] EDSS 4 to EDSS 7 € 5,355.25 e-Salud, 2017[31] EDSS 8 and EDSS 9 € 6,487.40 Gubieras, 2011[22] Adverse events costs Progressive multifocal leukoencephalopathy (2) € 7,070 e-Salud, 2017[31] Serious infections (3) € 4,484 “ Macular oedema (4) € 2,457 “ Cancer (5) € 6,113 “ (1) Ex-factory price (EFP) with 7.5% deduction; (2) CIE09: 046.3; (3) GRD 79/80; (4) CIE09: 362.83; (5) GRD 694. 95% CrI: 95% credible interval; AR: Adverse reactions; EDSS: Expanded Disability Status Scale; MRI: magnetic resonance imaging; vs.: compared with. *The cost of the tuberculin test was assumed [40]. 5 6 J. L. POVEDA ET AL.

Table 3. The overall results of the deterministic cost-effectiveness analysis. Net Health Benefit. According to the willingness to QALYs Cost of gaining a QALY with pay Treatment Total cost per patient per patient Cost difference Difference of QALYs Cladribine Tablets(1) (€ per QALY gained)(2) 20,000 € 25,000 € 30,000 € Cladribine Tablets 157,856 € 10.39 −86,536 € 1.11 Dominant 4.327 3.461 2.885 Fingolimod 244,392 € 9.28 Dominated (1) The dominant medication (Cladribine Tablets) is more effective, obtains more QALYs, with lower costs than the medicine dominated (fingolimod). (2) A positive net health benefit indicates that Cladribine Tablets is a cost-effective treatment compared to fingolimod, for willingness to pay 20,000 €, 25,000 € and 30,000 per QALY gained with the most effective treatment (Cladribine Tablets), which will get more QALYs per patient than fingolimod (Δ Effectiveness – Δ Costs/willingness to pay).

Table 4. Detailed results of the deterministic cost-effectiveness analysis. a patient with RMS with high disease activity would be cost- Cladribine Cladribine Tablets effective between 94.6% and 96.1% of the patients for will- Items Tablets Fingolimod vs. Fingolimod ingness to pay 20,000 € to 30,000 € per QALY gained with the Costs (€) most effective treatment (Cladribine Tablets). In other words, Drug purchase costs 48,264 € 127,247 € −78,983 € the probability that fingolimod is cost-effective in comparison Administration of 0 € 201 € −201 € Medication with Cladribine Tablets would range between 3.9% and 5.4% Medication monitoring 570 € 2,676 € −2,106 € (Table 6). Totals for the drug 48,834 € 130,124 € −81,290 € Totals by adverse effects 955 € 957 € −2 € Totals by relapses 3,693 € 4,336 € −643 € Totals based on disability 104,374 € 108,975 € −4.60.1 € 4. Discussion (EDSS) Total cost 157,856 € 244,392 € −86,536 € According to this analysis, compared to fingolimod, Cladribine QALYs Tablets would generate both economic savings as well as Totals by relapses −0.11 −0.12 0.01 a gain of QALYs in patients with RMS and high disease activity. Totals based on disability 12.24 11.21 1.03 (EDSS) The volume of savings per patient in a lifelong treatment Total losses of caregivers −1.74 −1.81 0.07 would be around 86,000 €. This saving is mainly due to drug Total QALYs 10.39 9.28 1.11 acquisition cost (maximum of 20 days in two courses of treatment), drug administration (no use of specific health resources 3.3. Probabilistic sensitivity analysis for its administration) and drug monitoring cost (reduced associated monitoring needed) [41]. According to the probabilistic analysis (Figure 3, Table 6), com- The gain of QALYs per patient in a lifelong treatment would pared to fingolimod, treatment with Cladribine Tablets of be 1.11 QALYs according to the probabilistic analysis. It should

Figure 2. Deterministic sensitivity analysis (tornado diagram). Net Health Benefit. ARR: annualized relapse rate. A positive net health benefit indicates that Cladribine Tablets is a cost-effective treatment compared to fingolimod. EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 7

Table 5. Results of the deterministic analysis of scenarios. Item Cost difference Difference of QALYs Cost per AVAC gained (1) Base case −86,536 € 1.12 Cladribine Tablets is dominant Time horizon 40 years (10 years less) −83,108 € 1.03 Cladribine Tablets is dominant 30 years (20 years less) −84,525 € 0.78 Cladribine Tablets is dominant 20 years (30 years less) −79,172 € 0.38 Cladribine Tablets is dominant 10 years (40 years less) −56,823 € 0.01 Cladribine Tablets is dominant Discount costs and QALYs 0% −117,023 € 2.43 Cladribine Tablets is dominant 6% −67,064 € 0.56 Cladribine Tablets is dominant Exclusion from loss of utility on caregivers −86,536 € 1.05 Cladribine Tablets is dominant (1) The dominant medication (Cladribine Tablets) is more effective, obtains more QALYs, with lower costs than the medicine dominated (fingolimod).

Figure 3. Probabilistic analysis. QALY: quality-adjusted life year. The likelihood that Cladribine Tablets is cost-effective compared to fingolimod ranged between 94.6% and 96.1% for willingness to pay between € 20,000 and € 30,000 per QALY gained.

Table 6. Detailed results of the deterministic cost-effectiveness analysis. Probability of cost-effectiveness according to the Mean QALYs Mean cost Difference of mean Cost of gaining 1 willingness to pay Treatment Costs mean (CI 95%) (CI 95%) difference QALYs QALY(1) (€ per QALY gained)(2) 20,000 25,000 30,000 Cladribine 158,149 € 10.57 −86,328 € 1.13 Dominant 94.6% 96.1% 94.8% Tablets (146,420–170,985 €) (7.39–13.18) Fingolimod 244,477 € 9.44 Dominated 5.4% 3.9% 5.2% (232,998–257,097 €) (7.15–11.77) (1) The dominant medication (Cladribine Tablets) is more effective, you get more QALYs, with lower costs than the medicine dominated (fingolimod). (2) Probability that Cladribine Tablets is a cost-effective treatment compared to fingolimod, for the willingness to pay 20,000 €, 25,000 € and 30,000 € for each QALY gained with the most effective treatment (Cladribine Tablets), which achieves more QALYs per patient than fingolimod (Δ Effectiveness – Δ Costs/ willingness to pay). be noted that a difference of 1.11 QALYs in favour of first, it must be remembered that this is a theoretical model Cladribine Tablets versus fingolimod is very notable, given that is, by definition, a simplified simulation of reality. that generally it is thought that the minimally clinically impor- A limitation of the model is the fact that the hazard ratios of tant difference of utilities (i.e., that the patient is able to confirmed progression of disability for Cladribine Tablets and detect) between two interventions observed with the instru- fingolimod versus placebo were obtained through a network ments EQ-5D, HUI2, HUI3 and SF-6D would be 0.074, 0.030, meta-analysis [15], which analysed the subgroups of patients 0.030 and 0.033 QALYs, respectively [41–44]. with RMS with high disease activity, because there are no The economic model simulated the evolution of a cohort of clinical trials comparing both drugs in these patients. patients with RMS with high disease activity treated with Another weakness of the study, common to many economic Cladribine Tablets or fingolimod. For the adequate assessment models, is that it was necessary to simulate the evolution of of the results of the study we need to bear in mind both the a hypothetical cohort of patients over the long term (between possible limitations and their consistencies. With regard to the 10 and 50 years) on the basis of the data of clinical trials that 8 J. L. POVEDA ET AL. had a maximum duration of 2 years [15]. As in previously study. C Torres works in the PR & HEOR Area of Corporate Affairs published analyses [14,45–49] and in line with the SmPC for Department of Merck. A Polanco leads the Department of Corporate Cladribine Tablets, the study was modelled assuming that the Affairs of Merck. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or results of effectiveness in 2 years would be extrapolated over financial conflict with the subject matter or materials discussed in the manu- the longer term. The evidence of the comparative effective- script apart from those disclosed. ness in the longer term of the DMTs versus placebo is limited for obvious ethical reasons [12]. The greatest strength of the study lies in the stability of the Disclosure statement economic model (the maintenance of the sense of the deter- The abstract of this paper was presented at the ISPOR 21st Annual European ministic base case outcome in the sensitivity analyses), which Congress, 2018, Barcelona, as a poster presentation with interim findings. The was confirmed in all scenarios analysed. According to the poster’s abstract was published in Value in Health. 2018; 21: S339 (https:// probabilistic analysis, compared to fingolimod, treatment www.valueinhealthjournal.com/article/S1098-3015(18)35331-2/fulltext). with Cladribine Tablets of a patient with RMS with high disease activity would be cost-effective between 94.6% and Reviewer Disclosures 96.1% of patients for willingness to pay 20,000 € to 30,000 € per QALY gained with Cladribine Tablets. Peer reviewers on this manuscript have no relevant financial or other An analysis of the cost-effectiveness of the treatment of relationships to disclose. RMS with high disease activity in the United Kingdom (UK) has been previously published [12]. According to this study, Author contribution statement Cladribine Tablets was the dominant treatment, with a probability of cost-effectiveness of 93% for a willingness to JL Poveda, JL Trillo, C Rubio-Terrés, D Rubio-Rodríguez, A Polanco and C Torres made the adaptation of the economic model. C Rubio-Terrés and pay 30,000 £ per QALY gained. The findings reported in this D Rubio-Rodríguez wrote the first draft. All authors interpreted the data work are consistent with those reported in the UK cost- and commented on the first draft. All authors revised the first draft. All effectiveness study. No other analysis comparing the cost- authors agreed with the final version. effectiveness of Cladribine Tablets versus fingolimod in the RMS with high disease activity has been identified. References

1. Montalban X, Gold R, Thompson AJ, et al. ECTRIMS/EAN Guideline 5. Conclusions on the pharmacological treatment of people with multiple sclerosis. Mult Scler. 2018;24(2):96–120. According to the present study, compared to fingolimod, 2. MSSSI: Ministerio de Sanidad, Servicios Sociales e Igualdad. treatment with Cladribine Tablets of RMS with high disease Estrategia en Enfermedades Neurodegenerativas del Sistema activity is an option that could generate savings for the Nacional de Salud; 2016 [cited 2018 Jun]. Available from: http:// Spanish NHS, with a considerable gain in QALYs in patients. www.msps.es/organizacion/sns/planCalidadSNS/pdf/Est_ Cladribine Tablets is considered a cost-effective and dominant Neurodegenerativas_APROBADA_C_INTERTERRITORIAL.pdf 3. Bártulos M, Marzo ME, Estrella LA, et al. Estudio epidemiológico de (generating less costs and more effectiveness than fingolimod) la esclerosis múltiple en La Rioja. Neurología. 2015;30(9):552–560. treatment option in this patient population. 4. Izquierdo G, Venegas A, Sanabria C, et al. Long-term epidemiology This evaluation, together with the convenient treatment of multiple sclerosis in the Northern Seville District. Acta Neurol regimen of Cladribine Tablets (maximum of 20 days in two Scand. 2015;132(2):111–117. courses of treatment, without specific health resources for its 5. Carreón-Guarnizo E, Andreu-Reinón E, Cerdán-Sánchez M, et al. Prevalencia de la esclerosis múltiple en la Región de Murcia. Rev administration and its associated low monitoring consump- Neurol. 2016;62(9):396–402. tion) may be useful to decision-makers in their assessment of 6. Oreja-Guevara C, Kobelt G, Berg J, et al. The European Multiple the advantages of Cladribine Tablets as a new and valued Sclerosis Platform. New insights into the burden and costs of multiple option for RMS with high disease activity patients. sclerosis in Europe: results for Spain. Mult Scler J. 2017;23(2S):166–178. 7. Fernández O, Calleja-Hernández MA, Meca-Lallana J, et al. Estimate of the cost of multiple sclerosis in Spain by literature review. Expert Funding Rev Pharmacoecon Outcomes Res. 2017;17(4):321–333. 8. Weinshenker BG, Bass B, Rice GP, et al. The natural history of This study was supported by a grant from Merck, S.L.U., an affiliate of multiple sclerosis: A geographically based study. I. Clinical course Merck KGaA Darmstadt, Germany. This supported development and per- and disability. Brain. 1989;112(Pt1):133–146. formance of the analysis and writing and editorial support for this manu- 9. Confavreux C, Vukusic S. Natural history of multiple sclerosis: script. Medical writing support and development and performance of the A unifying concept. Brain. 2006;129(Pt3):606–616. analysis was provided by Darío Rubio Rodríguez and Carlos Rubio Terrés 10. Gilenya 0,5 mg cápsulas duras. Ficha técnica o Resumen de of of Health Value, Madrid, Spain and was funded by Merck, S.L.U. In características del producto; [cited 2018 Apr 13]. Available from: addition, we also thank Schiffon L. Wong for her assistance in the original https://cima.aemps.es/cima/dochtml/ft/11677005/FT_11677005.html Markov model. 11. Mavenclad 10 mg comprimidos. Ficha técnica o Resumen de características del producto; [cited 2018 Apr 13]. Available from: https://cima.aemps.es/cima/pdfs/es/ft/1171212001/FT_ Declaration of interest 1171212001.pdf 12. Hettle R, Harty G, Wong SL. Cost-effectiveness of cladribine D Rubio Rodríguez is a Senior Consultant of Health Value, a company that tablets, alemtuzumab and natalizumab in the treatment of has received fees in relation to the present study. C Rubio Terrés is Director relapsing-remitting multiple sclerosis with high disease activity of Health Value, a company that has received fees in relation to the present in England. J Med Econ. 2018;21(7):676–686. EXPERT REVIEW OF PHARMACOECONOMICS & OUTCOMES RESEARCH 9

13. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an 34. Llaneza MA Epidemiología de la esclerosis múltiple en el Área expanded disability status scale (EDSS). Neurology. 1983;33 sanitaria de Ferrol. Tesis doctoral Universidad de la Coruña, 2016 (11):1444–1452. [cited 2017 Sep 14] . Available from: http://hdl.handle.net/2183/ 14. Palace J, Duddy M, Bregenzer T, et al. Effectiveness and 17565 cost-effectiveness of interferon beta and glatiramer acetate in the 35. Scolding N, Barnes D, Cader S, et al. Association of British UK multiple sclerosis risk sharing scheme at 6 years: a clinical Neurologists: revised (2015) guidelines for prescribing cohort study with natural history comparator. Lancet Neurol. disease-modifying treatments in multiple sclerosis. Pract Neurol. 2015;14(5):497–505. 2015;15(4):273–279. 15. Siddiqui MK, Khurana IS, Budhia S, et al. Systematic literature 36. López-Bastida J, Oliva J, Antoñanzas F, et al. Propuesta de guía para review and network meta-analysis of cladribine tablets versus alter- la evaluación económica aplicada a las tecnologías sanitarias. Gac native disease-modifying treatments for relapsing-remitting multi- Sanit. 2010;24(2):154–170. ple sclerosis. Curr Med Res Opin. 2017;34(8):1361–1371. 37. The human mortality database; [cited 11 de abril de 2018]. 16. Palace J, Bregenzer T, Tremlett H, et al. UK multiple sclerosis Available from: http://www.mortality.org/ risk-sharing scheme: a new natural history dataset and an 38. Rodríguez-Antigüedad Zarranz A, Mendibe Bilbao M, Llarena improved Markov model. BMJ Open. 2014;4(1):e004073. González C, et al. Mortality and cause of death in multiple sclerosis: 17. Orme M, Kerrigan J, Tyas D, et al. The effect of disease, functional findings from a prospective population-based cohort in Bizkaia, status, and relapses on the utility of people with multiple sclerosis Basque Country, Spain. Neuroepidemiology. 2014;42(4):219–225. in the UK. Value Health. 2007;10(1):54–60. 39. Real Decreto-ley 8/2010, de 20 de mayo, por el que se adoptan 18. Tan K, Roda R, Ostrow L, et al. PML-IRIS in patients with HIV medidas extraordinarias para la reducción del déficit público. BOE infection: clinical manifestations and treatment with steroids. Nº 126, Lunes 24 de mayo de 2010: 45070. Neurology. 2009;72(17):1458–1464. 40. Resolución de adjudicación. Suministro de productos y equipos 19. Cohen JA, Barkhof F, Comii G, et al. Oral fingolimod or intramus- necesarios para la realización de pruebas mediante cular interferon for relapsing multiple sclerosis. N Engl J Med. Enzimoinmunoanálisis en el Laboratorio Central de la OSI 2010;362(5):402–415. Araba; [cited 2018 Sep 5]. Available from: http://www.contrata 20. Giovannoni G, Comi G, Cook S, et al. A placebo-controlled trial of cion.euskadi.eus/w32-1084/en/v79aWar/comunJSP/ oral cladribine for relapsing multiple sclerosis. N Engl J Med. v79aObtenerFichero.do;jsessionid= 2010;362(5):416–426. fSLO1l59MGryeVEuGZ4GUYTIAj3qlo8pipVaKmOlpQOCubFtKA-E!- 21. Kappos L, Radue EW, Connor P, et al. A placebo-controlled trial of 2033055975!-1619630890?identificador=140721&idTabla= oral fingolimod in relapsing multiple sclerosis. New Engl J Med. 007&R01HNoPortal=true 2010;362(5):387–401. 41. Brazier J, Roberts J, Tsuchiya A, et al. A comparison of the EQ-5D 22. Gubieras L, Casado V, Romero-Pinel L, et al. Cost of the relapse of and SF-6D across seven patient groups. Health Econ. 2004;13 multiple sclerosis in Spain. Value Health. 2011;14(5):A320. (9):873–884. 23. Karampampa K, Gustavsson A, Miltenburger C, et al. Treatment 42. Horsman J, Furlong W, Feeny D, et al. The Health Utilities Index experience, burden and unmet needs (TRIBUNE) in MS study: (HUI): concepts, measurement properties and applications. Health results from Spain. Mult Scler. 2012;18(2 Suppl):35–39. Qual Life Outcomes. 2003;1:54. 24. Saida T, Kikuchi S, Itoyama Y, et al. A randomized, controlled trial of 43. Walters SJ, Brazier JE. What is the relationship between the mini- fingolimod (FTY720) in Japanese patients with multiple sclerosis. mally important difference and health state utility values? The case Mult Scler. 2012;18(9):1269–1277. of the SF-6D. Health Qual Life Outcomes. 2003;1:4. 25. Acaster S, Perard R, Chauhan D, et al. A forgotten aspect of the 44. Wee H-L, Machin D, Loke W-C, et al. Assessing differences in utility NICE reference case: an observational study of the health related scores: a comparison of four widely used preference-based quality of life impact on caregivers of people with multiple instruments. Value Health. 2007;10(4):256–265. sclerosis. BMC Health Serv Res. 2013;13:346. 45. Tappenden P, Chilcott J, O’Hagan T, et al., [Internet]. Cost- 26. Calabresi PA, Radue EW, Goodin D, et al. Safety and efficacy of in effectiveness of beta interferons and glatiramer acetate in the patients with relapsing-remittingmultiple sclerosis (FREEDOMS II): a management of multiple sclerosis: final report to the National double-blind, randomised, placebo-controlled, phase 3 trial. Lancet Institute for Clinical Excellence; 2001 [cited 2018 Jan 31]. Neurol. 2014;13(6):545–556. Available from: http://www.academia.edu/8908699/Cost_effective 27. Pakpoor J, Disanto G, Altmann DR, et al. No evidence for higher risk ness_of_beta_interferons_and_glatiramer_acetate_in_the_manage of cancer in patients with multiple sclerosis taking cladribine. ment_of_multiple_sclerosis Neurol Neuroimmunol Neuroinflamm. 2015;2(6):e158. 46. Tappenden P, McCabe C, Chilcott J, et al. Cost-effectiveness of 28. Shingler S, Fordham B, Evans M, et al. Utilities for treatment-related disease-modifying therapies in the management of multiple sclero- adverse events in type 2 diabetes. J Med Econ. 2015;18(1):45–55. sis for the Medicare population. Value Health. 2009;12(5):657–665. 29. Hawton A, Green C. Health utilities for multiple sclerosis. Value 47. Peninsula Technology Assessment Group (PenTAG) [Internet]. The Health. 2016;19(4):460–468. effectiveness and cost-effectiveness of natalizumab for multiple 30. Trogdon JG, Ekwueme DU, Chamiec-Case L, et al. Breast cancer in sclerosis: an evidence review of the submission from Biogen; young women: health State utility impacts by race/ethnicity. Am 2007 [cited 2018 Jan 31]. Available from: https://www.nice.org.uk/ J Prev Med. 2016;50(2):262–269. guidance/ta127/resources/multiple-sclerosis-natalizumabevalua 31. e-salud. Base de datos de costes sanitarios en España; 2017 [cited tion-report-evidence-review-group-report2 2017 Nov]. Available from: http://www.oblikue.com/bddcostes 48. Gani R, Giovannoni G, Bates D, et al. Cost-effectiveness analyses of 32. BotPlus 2.0. Consejo General de Colegios Oficiales de natalizumab (Tysabri) compared with other disease-modifying Farmacéuticos (CGCOF); 2018 [cited 2018 Jan 11]. Available from: therapies for people with highly active relapsing-remitting multiple www.portalfarma.com sclerosis in the UK. PharmacoEconomics. 2008;26(7):617–627. 33. NICE TA312. National Institute for Health and Clinical Excellence 49. Chilcott J, McCabe C, Tappenden P, et al. Modelling the (NICE). Alemtuzumab for treating relapsing-remitting multiple cost-effectiveness of interferon beta and glatiramer acetate in the sclerosis. NICE technology appraisal guidance 312; 2014 [cited management of multiple sclerosis. Commentary: evaluating disease 2016 Nov]. Available from: https://www.nice.org.uk/guidance/ modifying treatments in multiple sclerosis. BMJ. 2003;326 ta312?unlid=106213045220161014202829 (7388):522.