Evaluating the impact of early vs delayed ofatumumab initiation and estimating the long-term outcomes of ofatumumab vs teriflunomide in relapsing multiple sclerosis patients in Spain

Abstract

Objectives

To evaluate the impact of early (at first-line) vs delayed (3-year delay) ofatumumab initiation and long-term clinical, societal, and economic outcomes of ofatumumab vs teriflunomide in relapsing multiple sclerosis (RMS) patients from a Spanish societal perspective.

Methods

A cost-consequence analysis was conducted using an Expanded Disability Status Scale (EDSS)-based Markov model. Inputs were sourced from ASCLEPIOS I and II trials and published literature.

Results

At the end of 10 years, compared with first-line teriflunomide treatment, early first-line ofatumumab initiation was projected to result in 35.6% fewer patients progressing to EDSS ≥ 7 and 27.8% fewer relapses. The ofatumumab cohort required 7.3% reduced informal care time and had 19% fewer disability-adjusted life years (DALYs) than the teriflunomide cohort. A 3-year delay in ofatumumab treatment (3-year teriflunomide + 7-year ofatumumab) was projected to result in 32.2% more patients progressing to EDSS ≥ 7, 20.2% more relapses, 5.4% increased informal care time, and 16.6% more DALYs compared with early ofatumumab initiation. Early ofatumumab initiation was associated with total annual cost savings (excluding disease-modifying-therapies’ acquisition costs) of €35,328 ($34,549; conversion factor 1€= $1.02255) and €24,373 ($23,836) per patient vs teriflunomide and 3-year delayed ofatumumab initiation, respectively.

Conclusions

This study highlights the benefits of early initiation of high-efficacy therapy such as ofatumumab vs its delayed initiation for improving the outcomes in RMS patients (having characteristics similar to those of patients included in the ASCLEPIOS trials). Ofatumumab treatment was projected to provide improved long-term clinical, societal, and economic outcomes vs teriflunomide treatment in RMS patients from a Spanish societal perspective.

PLAIN LANGUAGE SUMMARY

The therapeutic benefits of ofatumumab in relapsing multiple sclerosis (RMS) patients have been reported in the ASCLEPIOS I and II trials. Using an Expanded Disability Status Scale-based Markov model, this study aimed to assess the impact of early (i.e. at first-line, as the first treatment after diagnosis) vs delayed (i.e. 3-year delay) ofatumumab initiation in RMS patients from a Spanish societal perspective. In addition, the long-term clinical, societal, and economic outcomes of ofatumumab vs teriflunomide treatment were assessed. Evidence from this study suggests that early initiation of a high-efficacy therapy such as ofatumumab, compared with its delayed initiation, is an effective and cost-saving strategy for improving outcomes in RMS patients. Furthermore, patients receiving ofatumumab for 10 years are projected to experience comparatively better outcomes (clinical, societal, and economic) than those receiving teriflunomide.

Keywords:

• Cost-consequence model
• disease-modifying therapies
• ofatumumab
• relapsing multiple sclerosis
• societal perspective
• Spain

JEL classification codes:

• I11
• I1
• I
• I00

Introduction

Multiple sclerosis (MS) is a chronic neurodegenerative disorder characterized by inflammatory demyelination, gliosis, and neuronal loss. Estimates suggest that the prevalence of MS ranges between 0.10% and 0.13% of the total population in Spain^1,2. MS is two times more prevalent in women than in men and is a leading cause of neurological disability in young and middle-aged adults, resulting in remarkable health and socioeconomic burden^2,3.

The economic burden imposed by MS in Spain is substantial. The overall annual costs related to MS in Spain are estimated to range from €1,200 to €1,337 million (amounting to €24,272 and €33,457 per patient per year)³. Several moderate-efficacy (beta interferons, glatiramer acetate, dimethyl fumarate, and teriflunomide) and high-efficacy (natalizumab, alemtuzumab, cladribine, ocrelizumab, ofatumumab, fingolimod, and ozanimod) disease-modifying therapies (DMTs) are currently used in clinical practice for treating relapsing MS (RMS) patients^4,5. While escalation strategy is the most applied treatment algorithm for RMS patients, emerging data have shown a positive long-term impact of high-efficacy DMTs when initiated earlier in the disease course without compromising the safety of the patients^5–9. There exists ambiguity across the MS treatment guidelines in terms of classifying DMTs based on their efficacy^10,11. While the 2015 version of the Association of British Neurologists guidelines classifies S1P-receptor (S1P-R) modulators such as fingolimod as a moderate-efficacy DMT¹⁰, the Brazilian guidelines categorizes fingolimod as a high-efficacy DMT11. Given the differences between the MS treatment guidelines, we used evidence from recent studies, most of which report S1P-R modulators (i.e. fingolimod and ozanimod) as high-efficacy DMTs^{4,5,9,12–14}.

Anti-CD20 B-cell-depleting therapies are high-efficacy DMTs that are increasingly being used to treat MS¹⁵. Ofatumumab is the first fully human anti-CD20 monoclonal antibody approved for the treatment of RMS in Europe¹⁶. Data from pivotal clinical trials (ASCLEPIOS I and II) have demonstrated the therapeutic benefits and comparable safety profile of ofatumumab and teriflunomide in overall and treatment-naive RMS patients^17,18. Moreover, several studies have emphasized the therapeutic benefits of early initiation of high-efficacy therapy (HET) vs an escalation approach in MS patients^{5–9,19–22}. Although the clinical outcomes of HETs such as ofatumumab, in newly diagnosed treatment-naive MS patients has been reported¹⁷, the impact of early ofatumumab vs teriflunomide initiation on the long-term clinical, societal, and economic outcomes in RMS patients remains unexplored. To fill this knowledge gap, in the present study, a cost-consequence analysis was conducted to assess the impact of early (i.e. at first-line, as the first treatment after diagnosis) vs delayed (i.e. 3-year delay) ofatumumab initiation and the long-term clinical, societal, and economic outcomes of ofatumumab vs teriflunomide from a Spanish societal perspective. As part of the clinical outcomes, the present study assessed the distribution of MS patients in the different Expanded Disability Status Scale (EDSS) states, the proportion of wheelchair and bedridden patients (i.e. EDSS ≥ 7), the number of relapses, and the number of disability-adjusted life years (DALYs). The economic outcomes assessed included direct (excluding DMT costs), relapse, and indirect costs. In addition, the societal outcomes assessed were productivity measure (i.e. % employed and % retired early) and informal care days per annum. A cost–consequence analysis is a simplified variant of economic evaluation in which the outcomes are reported separately from all costs (direct and indirect costs), allowing decision-makers to compare the incremental costs with the incremental consequences of the different interventions²³.

Methods

Study population and interventions

The patient population considered in this model was aligned to patients included in the ASCLEPIOS I and II clinical trials¹⁸. Thus, the characteristics of the cohort at the start of the model were assumed to match the baseline characteristics of the ASCLEPIOS I and II trial participants (Supplemental Table S1)¹⁸. The interventions considered for this study were ofatumumab 20 mg administered subcutaneously once every month and teriflunomide 14 mg administered orally once daily^16,24.

Model structure and inputs

A discrete-time Markov model based on the EDSS health states (0 = normal neurological exam and no disability; 10 = death) was developed in Microsoft Excel 2013. A hypothetical cohort of RMS patients with a cycle length of 1 year was tracked as they progressed through the disability states or naturally regressed to lower disability states or to the death state, as well as experienced a relapse (Supplementary Figure S1). The model was run from a societal perspective (i.e. direct [excluding DMT acquisition costs] and indirect MS management costs) and adopted a 10-year time horizon, as forecasting the productivity outcomes beyond 10 years using the current inputs might not be meaningful. The extrapolation of disability progression reported in the current study is in line with several published cost-effectiveness analysis and health technology assessment submissions^25–28.

Figure 1. Patient distribution in MS health states at the end of 10 years.

All model inputs and their respective sources are presented in Supplementary Table S2. For the natural history model, the most used sources of long-term natural history on disease progression in MS are the British Columbia²⁹ and the London Ontario datasets³⁰. This British Columbia dataset measures changes in EDSS from 1980 to 1995 and was used instead of the London Ontario cohort, which may not reflect present-day MS patients as data were collected from the 1970s to the 1980s. The annualized relapse rate (ARR) by EDSS during the untreated course of the disease was based on published natural history data^31,32.

The transition probabilities for the on-treatment model (i.e. those receiving DMTs) were derived by applying the treatment efficacy (time to 6-month confirmed disability progression) observed during clinical trial to natural history disability progression. Disability progression efficacy was applied using hazard ratio (HR) for time to 6-month confirmed disability progression (CDP), and relapse efficacy was applied using the rate ratio (RR) for ARR. Based on evidence from the ASCLEPIOS trials, it is expected that patients on ofatumumab would take more time to progress to high disability states and achieve a lower relapse rate than those on other less efficacious DMTs. The inputs for the treatment-adjusted model, i.e. HR for time to 6-CDP and RR for ARR, were sourced from a network meta-analysis (NMA)³³. Efficacy estimates were considered from the NMA for ofatumumab vs teriflunomide because of its robustness over the effect estimate from the ASCLEPIOS trials. The annual discontinuation probabilities for ofatumumab and teriflunomide were sourced from the ASCLEPIOS trials and the above-mentioned NMA^18,33. The all-cause mortality rates for the general population were derived from the age- and gender-specific mortality rates for Spain³⁴ and adjusted for the MS population using the mortality multipliers reported in the literature³⁵. MS-specific disability weights to calculate the DALYs were considered from Cho et al.’s study³⁶. MS-related productivity loss data (i.e. proportion of patients employed or retired early and the number of informal care days per annum by EDSS) were retrieved from published literature^37,38. Retirement age and working days per year were assumed as 65 years and 251 days, respectively. Cost inputs, namely, annual disease-related costs stratified according to the EDSS state and adverse event (AE) costs (general clinical visit, neurology visit, and medication), were sourced from the Spanish Health Costs and Cost-Effectiveness Ratios database³⁹. Additionally, annual relapse costs were based on the severity of the relapse and adapted from Hawton et al.’s study⁴⁰.

Model assumptions

Treatment effects were applied in the model in the form of delaying disability progression and reducing the number of relapses over the 10-year period to assess the clinical, societal, and economic outcomes of the treatment. Patients were assumed as fully adherent before treatment discontinuation. Patients who discontinued ofatumumab/teriflunomide treatment were moved to receive best supportive care either when they reached EDSS ≥ 7 or during all-cause discontinuation in line with the ASCLEPIOS trials¹⁸.

Scenarios

Three scenarios with a 10-year time horizon were simulated. The two base scenarios evaluated ofatumumab (i.e. 10 years on ofatumumab at first-line) vs teriflunomide (i.e. 10 years on teriflunomide at first-line). The third scenario simulated a 3-year delay in ofatumumab treatment (i.e. 3-year teriflunomide treatment, followed by 7-year ofatumumab treatment).

Model outcomes

Clinical outcomes, including the distribution of MS patients in the different EDSS states, the proportion of wheelchair and bedridden patients (EDSS ≥ 7), the number of relapses, and the number of DALYs, were estimated. DALY was calculated as the sum of the years of life lost (YLL) due to premature mortality and years lived with disability (YLD). Societal outcomes that were reported included productivity measure (% employed and % retired early) and informal care days per annum. Economic outcomes included direct, relapse, and indirect costs. Direct costs comprised healthcare costs (disease management, drug administration and monitoring, AE management, and non-medical) and excluded DMT acquisition costs. The drug acquisition costs were not considered in the model because the net drug acquisition cost of any drug (in this case ofatumumab and teriflunomide) is confidential in Spain. Relapse costs were those associated with the management of relapse events. Indirect costs referred to costs due to productivity losses incurred by patients and caregivers. All cost estimates were reported in 2020 euros using the Spanish consumer price index data⁴¹.

Results

Clinical outcomes

At the end of 10 years, the proportion of patients in a mild disability state (EDSS 0–3) were projected to be higher in the ofatumumab cohort than in the teriflunomide cohort (57.0% vs 44.0%). Compared with early ofatumumab initiation, a 3-year delay in ofatumumab initiation (i.e. as the first treatment after diagnosis) was found to result in an 11.4% reduction in the proportion of patients in a mild disability state (Figure 1).

Over 10 years, the proportion of patients requiring a wheelchair or bedridden (EDSS ≥ 7) was projected to be lower in the ofatumumab cohort than in the teriflunomide cohort (11.3% vs 17.6%) (Figure 2). Furthermore, patients in the ofatumumab cohort were projected to experience 27.8% fewer relapses (3.8 vs 5.3) and 19% fewer DALYs (2.0 vs 2.5) vs those in the teriflunomide cohort. The difference in the number of DALYs was mainly due to differences in YLD, which was considerably lower in the ofatumumab cohort than in the teriflunomide cohort (1.46 vs 1.90) (Figure 3). Compared with early ofatumumab initiation, a 3-year delay in initiation was estimated to result in 32.2% more patients progressing to EDSS ≥ 7 (Figure 2), 20.2% more relapses (4.6 vs 3.8), and 16.6% more DALYs over 10 years (Figure 3).

Figure 2. Proportion of RMS patients progressing to EDSS ≥7 (wheelchair or bedridden).

Figure 3. Disability-adjusted life years.

Societal outcomes

At the end of 10 years, the proportion of patients who were employed was projected to be higher (40.0% vs 35.2%) and that of patients who retired early was relatively lower (13.0% vs 15.4%) in the ofatumumab cohort than in the teriflunomide cohort. Additionally, patients in the ofatumumab cohort were projected to require 7.3% reduced informal care time than those in the teriflunomide cohort (1,542 vs 1,664 days). A 3-year delay in ofatumumab initiation was projected to result in lower productivity (i.e. 6% less employed; 9.1% retired early) and 5.4% increased informal care time (1,625 vs 1,542 days) in patients receiving delayed vs early ofatumumab initiation.

Economic outcomes

Over 10 years, patients in the ofatumumab cohort were estimated to incur 17.4% lower costs than those in the teriflunomide cohort (€167,327 vs €202,655 per patient). Additionally, a 3-year delay in ofatumumab initiation was projected to result in an incremental cost of €24,373 per patient compared with early ofatumumab initiation (191,700 vs 167,327 per patient over 10 years) (Figure 4). In line with the clinical and societal outcomes, a 3-year delay in ofatumumab initiation was projected to result in 14.6% more costs compared with early ofatumumab initiation (€191,700 vs €167,327 per patient). Indirect costs accounted for two-thirds of the total costs, whereas direct costs (excluding DMT acquisition costs) accounted for one-third. Across the three scenarios, disease management and non-medical costs were the key contributors to the direct costs, while costs associated with productivity loss (short- and long-term absence from work, invalidity, and early retirement) accounted for most of the total indirect costs (Supplementary Table S3).

Figure 4. Total annual cost (per patient) at the end of 10 years.

Discussion

Based on the results from this cost-consequence economic modeling analysis, early initiation of HET such as ofatumumab, compared with its delayed initiation, was found to be an effective and cost-saving strategy for improving the long-term (i.e. 10 years) outcomes in RMS patients. Furthermore, patients receiving ofatumumab for long-term (i.e. 10 years) were projected to experience better outcomes (clinical, societal, and economic) than those receiving teriflunomide among RMS patients from a Spanish societal perspective.

The therapeutic landscape of MS has witnessed a discernible shift in the last decade with the availability of multiple HETs⁴². Evidence supporting the early initiation of HETs, contrary to a treatment escalation approach, continues to accumulate with benefits to both the patients and society^{5–9,19–22}. A population-based Danish Multiple Sclerosis Registry study showed a lower risk of 6-month EDSS worsening (16.7% vs 30.2%) and first relapse (HR: 0.50, 95% confidence interval: 0.37–0.67) in patients initiated on HET as the first therapy vs a matched cohort initiated on a moderate-efficacy DMT¹⁹. Data from two cohort studies from the UK and Italy reported that long-term trajectories are more favorable with an early intensive therapy strategy than with moderate-efficacy DMTs^5,6. Similarly, data from a retrospective international observational study showed that commencement of HETs within 2 years of disease onset was associated with lesser disability after 6–10 years compared with commencement later in the disease course in relapsing-remitting multiple sclerosis patients⁷. Despite the growing body of evidence in favor of early initiation of HETs, widespread use of high-efficacy DMTs as the first-line treatment in newly-diagnosed MS patients has not yet been established as the standard treatment paradigm. Moreover, most treatment guidelines recommend initiating moderate-efficacy DMTs in patients with mild-to-moderate disease activity, while HETs are often reserved for later use in the disease course or for highly active patients with a rapidly evolving aggressive disease^43–46.

Data from this cost-consequence analysis showed that, over 10 years, patients in the ofatumumab cohort experienced better health outcomes, i.e. a lower proportion of patients required a wheelchair or were bedridden (EDSS ≥ 7), reduced relapse rates, lower disability states (i.e. fewer number of DALYs), lesser informal care time, and higher productivity, compared with those in the teriflunomide cohort. This was also reflected by a higher proportion of patients in the ofatumumab cohort maintaining a mild disability state over the 10 years than those in the teriflunomide cohort. The results of this study are in line with those of previous studies that reported favorable long-term health benefits (delayed disability progression, reduced relapse rates, increased likelihood of achieving no evidence of disease activity, increased work attendance and work productivity, and lower work productivity activity impairment) with the initiation of HETs compared with other moderate-efficacy DMTs^{5,6,9,21,47–50}. In addition to the health benefits, our model estimated that 10-year ofatumumab treatment led to a considerable cost-saving compared with teriflunomide treatment (Δ €35,328 per patient).

We further explored the implications of early vs delayed ofatumumab initiation. As expected, a 3-year delay in ofatumumab initiation was found to impose an incremental health (32.2% more patients progressing to EDSS ≥ 7, 20.2% more relapses, 5.4% increased informal care time, and 16.6% more DALYs) and cost (14.6% more healthcare costs) burden compared with its early initiation. These results highlight the potential health and socioeconomic merits of early vs delayed ofatumumab initiation. A direct comparison of our findings with those of other studies is not possible as this is the first study to explore the impact of early vs delayed ofatumumab initiation. However, we can benchmark the results with a few existing studies that have reported the positive benefits of early vs delayed initiation of HETs^{5–8,19,21,22}.

This study has a few limitations. First, the baseline characteristics, natural history data, and efficacy were based on data from the ASCLEPIOS I and II trials, which included patients from multiple countries. Moreover, Spanish MS patients accounted for approximately 5% of the population in ASCLEPIOS trials; thus, the patient population included in this analysis and their characteristics may not be representative of the broader Spanish population. The model can be updated once Spanish-specific data are available. Productivity-related inputs were sourced from a cross-sectional retrospective study³⁷, which had limitations including small subsets of patients in moderate (29%) and severe (11%) disability states and data for moderate and severe states being underestimated compared to those reported in the UK study³⁸. Moreover, Oreja-Guevara et al.’s study³⁷ reported estimates as per mild (EDSS 0–3), moderate (EDSS 4–6.5), and severe (EDSS 7–9) and does not provide inputs as per EDSS (as required for cost-consequence analysis); therefore, regression was fitted to derive inputs. Although this method provides EDSS-wise inputs, this is further hampered by the small sample size of the study. The disability weights used to derive DALYs were from a Korean study and may not be generalizable to Spanish settings³⁶. The Markov model is based on EDSS health states, which focus on functional mobility and are insensitive to cognitive impairments; therefore, there are limitations inherent to the Markov model, such as being memoryless and the transition probabilities used in the model might not reflect actual patient movement in clinical practice. These issues are relevant to most current models of MS disease. The data on age- and gender-specific mortality rates considered in the model were derived from Office for National Statistics reference 2021³⁴ and are assumed to be the same over the entire time horizon of the model. This is a limitation that commonly exists in cost-effectiveness and cost-consequence analysis whose time horizon is long. However, its impact on results may be negligible as same age- and gender-specific mortality rates were applied to both cohorts (i.e. patients on ofatumumab and teriflunomide). Finally, the cost savings presented in this study may vary as we did not include drug acquisition costs, which means that, while all other medical costs can be assessed, the overall treatment costs cannot be compared.

Conclusions

In conclusion, this study highlights the benefits of early initiation of HETs such as ofatumumab vs its delayed initiation for improving outcomes in RMS patients (having characteristics similar to those of patients included in the ASCLEPIOS trials). Based on the results of this economic modeling analysis, ofatumumab treatment was projected to improve the long-term clinical, societal, and economic outcomes, compared with teriflunomide treatment, in RMS patients from a Spanish societal perspective.

Transparency

Declaration of financial/other interests

UV, MP, KG, EK, and NA are employees of Novartis.

Author contributions

UV, KG, EK, and NA were responsible for the conception and design of the project. UV, MP, and KG were responsible for the analysis and development of the model. All authors participated in critically reviewing and interpreting the data. All authors critically reviewed the manuscript for intellectual content and all authors approved the final version for publication submission. All authors agree to be accountable for all aspects of the work.

Reviewer disclosures

A reviewer on this manuscript has disclosed that they consult for MS pharma including Novartis and Genzyme Sanofi; they have also advised on ASCLEPIOS. All the peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

Previous presentations

Study results were previously presented at the 37th Congress of the European Committee for Treatment and Research in Multiple Sclerosis (virtual) held between 13 and 15 October 2021.

Acknowledgements

The authors would like to acknowledge Santosh Tiwari (Novartis Healthcare Pvt. Ltd., Hyderabad) for providing medical writing support for this manuscript.

Additional information

Funding

This work was sponsored by Novartis Pharma AG, Basel, Switzerland and the sponsors were involved in all stages of the work and in the manuscript preparation.