Multiple Sclerosis Resource Centre

Welcome to the Multiple Sclerosis Resource Centre. This website is intended for international healthcare professionals with an interest in Multiple Sclerosis. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Comparative efficacy and discontinuation of dimethyl fumarate and fingolimod in clinical practice at 12-month follow-up

Multiple Sclerosis and Related Disorders, Volume 10, November 2016, Pages 44 - 52

Abstract

Background

Dimethyl fumarate (DMF) and fingolimod (FTY) are approved oral disease modifying therapies (DMT) for relapsing multiple sclerosis (MS). Phase 3 trials established these agents as effective and generally well tolerated, though comparative efficacy and discontinuation remain unknown.

Objective

To assess real-world efficacy and discontinuation of DMF and FTY over 12 months in patients with MS.

Methods

We identified 458 DMF-treated and 317 FTY-treated patients in a large academic MS center. Measures of disease activity and discontinuation were compared using propensity score (PS) weighting. Covariates in the PS model included demographics and baseline clinical and MRI characteristics within 12 months of DMT initiation. The primary outcome measure was on-treatment annualized relapse rate (ARR) ratio, which was analyzed using a Poisson regression model. Other measures included time to first relapse, drug discontinuation, time to discontinuation, and new brain MRI lesions at 12 months.

Results

The on-treatment ARR for DMF was 0.16 (95% CI (0.12, 0.18)) and 0.13 (95% CI (0.08, 0.16)) for FTY. PS weighting, which demonstrated excellent covariate balance, showed no differences between groups on ARR (rate ratio=1.56, 95% CI (0.78, 3.14)), overall brain MRI activity defined as new T2 and/or gadolinium enhancing (GdE) lesions (OR=1.38, 95% CI (0.78, 2.42)), new T2 lesions (OR=1.33, 95% CI (0.71, 2.49)), and discontinuation (OR=1.30, 95% CI (0.84, 1.99)). DMF had higher odds of GdE lesions (OR=2.19, 95% CI (1.10, 4.35)), earlier time to discontinuation (HR=1.35, 95% CI (1.05, 1.74)), and earlier relapses (HR=1.64, 95% CI (1.10, 2.46)) compared to FTY.

Conclusion

Assessment in our clinical practice cohort showed comparable clinical efficacy, overall brain MRI activity, and discontinuation between DMF and FTY at 12 months. DMF had increased GdE lesions and intolerability early after treatment initiation.

Highlights

  • DMF and FTY had comparable ARR and overall brain MRI activity at 12 months.
  • DMF- and FTY-treated patients had similar proportions who discontinued therapy.
  • DMF had greater GdE lesions and side effects early after treatment initiation.
  • DMF-treated patients discontinued medication and relapsed earlier compared to FTY.

Keywords: Dimethyl fumarate, Fingolimod, Comparative efficacy, Discontinuation, Multiple sclerosis.

1. Introduction

Fingolimod (FTY) and dimethyl fumarate (DMF) are two oral disease modifying therapies (DMTs) currently approved to reduce disease activity in relapsing forms of MS. FTY (0.5 mg, one tablet daily) (Cohen et al, 2010 and Kappos et al, 2010) was the first oral therapy approved by the United States Food and Drug Administration (FDA) in 2010 and the European Medicines Agency in 2011. DMF (240 mg, one tablet twice daily) was approved to treat MS in 2013 (Fox et al, 2012a and Gold et al, 2012).

In phase 3 trials, FTY showed a 54% reduction of annualized relapse rate (ARR) and slowed disability worsening by 30% in comparison to placebo. MRI measures of disease activity supported these clinical findings (Calabresi et al., 2012). In a one-year head-to-head clinical trial against intramuscular interferon beta-1a, FTY demonstrated a greater reduction in ARR (approximately 33%) and reduction in MRI disease activity but no difference in disability worsening (Cohen et al., 2010). DMF demonstrated clinical efficacy in phase 3 trials with 48%–53% ARR reduction as compared to placebo (Fox et al, 2012a and Gold et al, 2012). Data also showed a reduction in MRI measures of disease activity and reduction in risk of disability worsening (Gold et al., 2012), although the latter finding was only shown in one of two phase 3 studies (Fox et al., 2012a). Overall, both medications have similar efficacy in clinical trials (Phillips and Fox, 2013), with absolute relapse rate reductions similar in magnitude, though treatment decisions in clinical practice are difficult owing to lack of direct comparative data.

During clinical trials, adverse effects reported with FTY were mild to moderate in severity and included upper respiratory tract infections, headache, diarrhea, and back pain. The most concerning adverse effects included cardiac events (bradycardia and atrioventricular block at treatment initiation), elevated liver enzymes, lymphopenia, fatal herpes virus infections, and macular edema (Cohen et al, 2010, Kappos et al, 2010, and Calabresi et al, 2012). By 24-month follow-up, 7.5% of patients in phase 3 trials discontinued FTY due to adverse effects (Kappos et al., 2010). The most frequent adverse effects observed during trials with DMF were gastrointestinal symptoms - including stomach pain, nausea, vomiting, and diarrhea - and transient skin flushing. Gastrointestinal symptoms were generally more prominent during the first several weeks of treatment and usually improved thereafter (Fox et al, 2012a and Gold et al, 2012). However, a sizeable proportion of patients (12%–16%) in phase 3 studies discontinued DMF due to intolerability by 24 months (Fox et al, 2012a and Gold et al, 2012).

Randomized controlled trials (RCTs) are the standard approach for regulatory approval of new therapeutic agents. However, observational studies are valuable when multiple agents are available and head-to-head RCTs may not be feasible. Observational studies also afford the opportunity to examine larger numbers of patients, and their findings may reflect real-world conditions more accurately than RCTs (Yang et al., 2010). However, since observational studies are subject to confounding, comparisons should be made after efforts to reduce baseline imbalances in disease characteristics (Grimes and Schulz, 2002). Propensity score (PS) analysis reduces the impact of confounding as well as selection and indication biases through balancing the distributions of covariates across treatments of interest, approximating a randomized study design (D'Agostino, 1998 and Rosenbaum and Rubin, 1983). Covariates are included in the PS model if they affect treatment decisions, in relation to the agents being studied. Standard methods for observational study analyses, including matching and weighting on the PS, allow investigators to specify and compare treatment groups in such a way as to minimize bias (Rubin, 2001).

The current investigation compares efficacy and discontinuation in patients treated with DMF or FTY in clinical practice over 12 months. While experience in separate clinical trials showed comparable efficacy for DMF and FTY and possibly higher discontinuation in DMF, no head-to-head comparisons of these agents have yet been reported. The current study addresses a common clinical treatment decision question faced by clinicians and seeks to fill this knowledge gap by providing comparative data in a moderately large patient cohort in routine clinical practice.

2. Materials and methods

2.1. Patient population

We conducted a retrospective observational cohort study of patients with MS who were followed at the Cleveland Clinic Mellen Center and treated either with DMF or FTY with 12-month follow-up available. We included patients who started FTY between October 2010 and August 2011 to investigate a cohort similar in size to that of individual treatment arms in phase 3 trials. All patients who started DMF between March and July 2013 with 12-month follow-up were included to ensure a comparable sample size to FTY. Patients were selected from these time points to investigate the earliest experience of FTY and DMF use in clinical practice.

2.2. Data collection

Following institutional review board approval, all patients prescribed DMF or FTY at the Cleveland Clinic Mellen Center during the designated time periods were identified. Review of the electronic medical record (EMR) was conducted to collect baseline and outcomes data 12 months after drug initiation. We incorporated baseline covariates derived from information collected from the EMR in the 12 months prior to treatment exposure, also using a collection of data points readily available in routine clinic notes that are reviewed and approved by an MS specialty provider prior to inclusion (Katzan et al., 2011). The treating neurologist determined a clinical relapse as new or worsening symptoms lasting greater than 24 h without a concomitant illness. All relapses included in the study were confirmed clinically. Number of new T2-hyperintense brain MRI lesions and semi-quantitative assessment of overall lesion burden were manually counted and categorized, respectively, by the author (CH) and Cleveland Clinic neuroradiologists. Neuroradiological data were compared to MRI 6–12 months prior to baseline MRI. Patient reported outcome (PRO) measures including Patient Health Questionnaire-9 (PHQ-9, a measure of depression) (Kroenke et al., 2001), Multiple Sclerosis Performance Scale (MSPS, an assessment tool of vision, hand function, sensation, spasticity, mobility, fatigue, cognition, and bladder and bowel control) (Schwartz et al., 1999), and European Quality of Life-5 Dimensions (EQ5D, a standardized assessment of quality of life) (The EuroQol Group, 1990) were included in the PS model. We used REDCap software to create an encrypted database on secure Cleveland Clinic servers (Harris et al., 2009).

2.3. Statistical analysis

Data were imported for analysis into R version 3.1.1© (Team, 2014). All covariates included in the propensity score model were missing in fewer than 10% of subjects. To account for missingness patterns in the PS, we incorporated a dummy variable to indicate missingness for all variables with missing data, and then used single imputation to impute a random value from among those observed in other subjects for each missing value. As a stability check, we then repeated the process using a different random choice of imputed value to see if any meaningful changes in the balance of the covariates after application of the propensity model might be observed, which would have implicated the missingness pattern as potentially important in terms of confounding.

Covariates with proportion of missing information >10% (e.g. John Cunningham Virus antibody) were excluded from the model.

Analyses were conducted “on-treatment.” The primary outcome was the ARR ratio (DMF vs. FTY). Secondary outcome measures included: time to first relapse; time to discontinuation; proportion of patients with new T2 lesions and/or gadolinium enhancing (GdE) lesions; proportion who discontinued therapy; proportion of patients with 20% worsening on timed 25 foot walk (T25FW, a quantitative measure of walking speed) (Rudick et al., 2002) and 9-hole peg test (9-HPT, a quantitative measure of dexterity) (Kragt et al., 2006); and proportion of patients with depression defined as PHQ-9 score ≥10. Post-baseline follow-up (e.g. visit/MRI frequency and protocols) did not differ between the two cohorts.

The PS model was built as a logistic regression to calculate the likelihood of DMF initiation, as compared to FTY, using pre-specified demographics and baseline disease characteristics (Table 1). A PS was derived for each patient and subsequently used in Average Treatment effect on the Treated (ATT) weighting (Hirano et al., 2003) to identify samples of patients in the DMF and FTY groups who were similar at baseline except for their treatment assignment. In this case, ATT weighting using the PS assigned each DMF patient a weight of 1, while FTY patients were weighted by PS/(1-PS) so as to mirror the distribution of covariates in the DMF group to the FTY group. This approach reduced the impact of selection bias on our comparisons of DMF to FTY while still retaining information from all patients. In addition, we developed a 1:1 matched sample, formed by matching DMF and FTY patients without replacement on their PS, with the same goal of comparing DMF to FTY patients, while reducing treatment selection bias. Prior to observing outcomes across the two groups, we selected the weighted approach on the basis of its more complete balance between the DMF and FTY groups on the variables included in the PS (Austin, 2011).

Table 1

Baseline characteristics of dimethyl fumarate and fingolimod patient cohorts.a

 

DMF Fingolimod p-value*
n=458 n=317
n % or SD n % or SD
Age (years, SD) 47.1 11.2 43.9 9.2 0.001*
Female 320 69.9% 223 70.3% 0.950
Race 0.014*
White 381 83.2% 287 90.5%
Black 57 12.4% 23 7.3%
Other 20 4.4% 7 2.2%
Disease Duration (years, SD) 14.3 8.9 16.4 8.1 0.001*
Relapsing-Remitting MS 337 73.6% 259 81.7% 0.011*
Number of Relapses 0.020*
0 276 62.3% 188 59.7%
1 119 26.9% 111 35.2%
2 33 7.4% 13 4.1%
≥3 15 3.3% 3 1.0%
DMF/FTY used as first-line agent 35 7.6% 16 5.0% 0.170
Direct Switch from prior DMT 330 72.1% 243 76.7% 0.043*
Reason for switch from prior DMT
Clinical relapse 59 12.9% 60 18.9% 0.04*
MRI activity 59 12.9% 41 12.9% 1.000
Disability progression 85 18.6% 81 25.6% 0.038*
Intolerance 208 45.4% 156 49.2% 0.333
Cost/insurance coverage 13 2.9% 8 2.5%
Last therapy prior to DMF or FTY 0.001*
Interferon 132 28.8% 120 37.9%
Glatiramer 142 31.0% 90 28.4%
Natalizumab 54 11.8% 47 14.8%
Immunosupression 29 6.3% 32 10.1%
# Prior DMTs (number, SD) 2.0 1.4 2.1 1.4 0.069
Interferon 297 64.8% 245 77.3% 0.001*
Glatiramer 228 49.8% 162 51.1% 1.000
Natalizumab 91 19.9% 68 21.5% 0.770
Immunosupression 74 16.2% 71 22.4% 0.051
John Cunningham Virus Antibodyb 0.001*
Positive 144 31.4% 49 15.5%
Negative 76 16.6% 33 10.4%
Not assessed 238 52.0% 235 74.1%
Laboratory Values
Mean WBC (×109/L) 7.26 3.2 7.24 2.9 0.923
Mean ALC (×109/L) 2.04 1.7 3.45 2.2 0.211
MRI available for review 442 96.5% 313 98.7%
Disease burden on MRI 0.001*
Mild 237 53.6% 110 34.7%
Moderate 169 38.2% 157 50.2%
Severe 36 8.1% 46 14.7%
Gadolinium enhancement on MRI 104 23.5% 75 24.0% 0.468
New T2 Lesions on MRI 120 27.1% 62 19.8% 0.035*
Objective Measures
T25FW (sec, SD) 7.3 7.9 8.1 9.4 0.249
Ambulation assistance 0.146
None 343 74.9% 241 76.0%
Unilateral 31 6.8% 31 9.8%
Bilateral 53 11.6% 25 7.9%
9 HPT (sec, SD) 26.64 16.2 26.36 14.2 0.851
Patient Reported Outcomes
MSPS score (mean, SD) 11.8 7.6 11.4 7.3 0.528
EQ5D score (mean, SD) 714.1 209.9 735.6 187.6 0.192
PHQ-9 score (mean, SD) 6.9 5.9 9.3 5.8 0.330

* Statistically significant with p<0.05.

a Covariates used in the PS model.

b JCV Ab status was left out of the covariate model since there was far greater than 10% missingness in the FTY group. Since JCV antibody stratification was not yet commercially available in routine practice during the 12-month period prior to FTY initiation (JCV antibody status was not assessed in 74.1% of the FTY cohort), inclusion as a covariate would have been misleading as it would not have represented a true predictor of FTY versus DMF initiation. Among those with JCV antibody status available, seropositivity rates were relatively similar in both groups (65.5% DMF, 59.8% FTY) with rates as expected in the general MS population (about 55%).

The pre-baseline period over which the number of relapses was recorded was 12 months.

9 HPT: 9 hole peg test; ALC: absolute lymphocyte count; DMF: dimethyl fumarate; DMT: disease modifying therapy; EQ5D: European Quality of Life- 5 Dimensions; FTY: fingolimod; MRI: magnetic resonance imaging; MS: multiple sclerosis; MSPS: Multiple Sclerosis Performance Scale; PHQ-9: Patient Health Questionaire-9; SD: standard deviation; sec: seconds; T25FW: timed 25 foot walk; WBC, white blood cell.

The strength of weighting and matching techniques was evaluated by how effectively they balanced the two treatment groups, as determined by comparing standardized differences in covariates before and after propensity adjustment. Excellent covariate balance was defined as absolute standardized difference <10% on the means of the covariate across the two treatments. Following ATT weighting and 1:1 greedy matching without replacement on the linear PS, treatment groups were then compared using conditional logistic regression to obtain odds ratio estimates for binary outcomes, linear regression to obtain difference estimates for continuous outcomes, and Cox-proportional hazards model and Kaplan-Meier survival curves to obtain estimates for survival outcomes. On-treatment relapse rates were analyzed using a Poisson regression model. Unadjusted and PS-adjusted outcome measures were compared. Odds and hazards ratios were calculated as patients treated with DMF compared to patients treated with FTY. The main outcome (ARR ratio) was based on one two-tailed test of statistical significance with α=0.05. Assuming 80% power with a total sample size=775 patients, the minimum detectable effect size (MDES) was determined to be 0.24.

3. Results

3.1. Baseline characteristics

In total, 458 patients started DMF, and 317 patients started FTY and were available for 12-month follow-up. Fifty-five DMF patients and 23 FTY patients were excluded due to lack of follow-up. Baseline demographic and disease characteristics are presented in Table 1. Both groups comprised a typical MS clinic population in terms of age and sex distributions, though patients in the current study were 10 years older compared to patients in DMF and FTY clinical trials. Additionally, patients starting DMF were older compared to those starting FTY. Both groups had similar proportion of patients starting DMF or FTY as first-line treatment, similar number of prior DMTs, and comparable T25FW, 9-HPT, and PRO measures. Patients in the two groups differed by race, disease duration, proportion of patients with RRMS, number of prior relapses, DMTs used prior to DMF or FTY, T2 lesion burden, and new T2 lesions on baseline brain MRI. Patients with progressive forms of the disease were included in this study to reflect our real-world experience of DMT use in clinical practice. Subgroup analysis of RRMS patients also was conducted to ascertain treatment effects using adjusted PS analysis.

3.2. Propensity model

The propensity model was built using covariates listed in Table 1. The missing data among covariates in the PS model did not meaningfully change overall covariate balance following PS analysis. The model appropriately assigned higher PS to DMF versus FTY, and there was adequate overlap of PS between the two treatment groups (Fig. A1). Prior to propensity adjustment, the treatment groups of interest were not well balanced (absolute value of the standardized difference of the linear PS, comparing DMF to FTY=86.8%). Thus, we favored the need for propensity adjustment over unadjusted comparisons to account for observed selection bias. 1:1 greedy matching on the linear PS resulted in a sample size of 317 patients in each group. PS matching only produced a partial balance of covariates between treatments with 15 covariates having absolute standardized differences greater than 10% (Fig. A2). Further, 1:1 greedy matching did not yield treatment groups with similar linear PS distributions, as the standardized difference in linear PS between groups was 86.4%, well over the 50% standard proposed by Rubin (Rubin, 2001).

ATT weighting more effectively balanced treatment groups with only 5 covariates having absolute standardized differences greater than 10% (Fig. A3). Weighting approach also yielded similar linear PS distributions with a standardized difference of 14.1%. Therefore, we selected PS weighting as preferable for incorporating information on selection bias to obtain outcome estimates for the entire cohort and RRMS subgroup analysis.

3.3. Outcome estimates

A summary of unadjusted outcomes is presented in Table 2 and unadjusted, post-matching, and post-weighting outcome estimates are presented in Table 3. Since PS weighting showed more favorable balance between treatment groups compared to 1:1 greedy matching, we highlight post-weighting outcome measures.

Table 2

Summary of unadjusted outcomes at 12-month follow-up.

 

DMF Fingolimod
n=458 n=317
n % or SD n % or SD
Discontinued drug at 12 months 134 29.3% 79 24.9%
 Disease Activity 38 8.3% 27 8.5%
  Clinical relapse 10 2.2% 10 3.2%
  MRI activity 19 4.1% 6 1.9%
  Disability progression 20 4.4% 15 4.7%
 Intolerance 105 22.9% 58 18.3%
Mean time to discontinuation (months, SD) 3.92 3.52 6.51 4.17
 Median time to discontinuation (months) 3.0 6.0
Relapse data available for review 421 91.9% 317 100%
Clinical relapse at 12 months (number of patients) 59 14.0% 36 11.4%
 Relapses per patient (mean, SD) 0.16 0.44 0.12 0.35
Mean time to relapse (months, SD) 3.85 2.97 7.65 4.38
MRI available for review 370 80.8% 290 91.5%
Disease activity on MRI at 12 months 77 20.8% 49 16.9%
 Gadolinium enhancement 36 9.7% 19 6.6%
 New T2 lesions 65 17.6% 37 12.8%
MRI available for review while on DMT 281 61.4% 248 78.2%
Disease activity on MRI at 12 months while on DMT 54 19.2% 39 15.7%
 Gadolinium enhancement 25 8.9% 15 6.0%
 New T2 Lesions 46 16.4% 28 11.3%
Adverse effects (number of patients) 329 71.4% 79 24.9%
 Mean WBC (×109/L) 6.23 2.25 5.06 2.99
 Mean ALC (×109/L) 1.48 0.95 0.63 0.54
Measures of neurologic impairment
 T25FW (mean sec, SD) 7.61 (n=383) 5.25 8.27 (n=274) 10.70
 9 HPT- dominant (mean, SD) 26.51 (n=272) 17.44 25.45 (n=173) 14.48
 9 HPT- non-dominant (mean, SD) 28.15 (n=272) 17.43 27.30 (n=173) 14.75
Patient Reported Outcomes
 PHQ-9 score (mean, SD) 6.42 (n=403) 5.68 6.30 (n=249) 5.42

9 HPT: 9-hole peg test; ALC: absolute lymphocyte count; DMF: dimethyl fumarate; DMT: disease modifying therapy; FTY: fingolimod; MRI: magnetic resonance imaging; PHQ-9: Patient Health Questionaire-9; SD: standard deviation; sec: seconds; T25FW: timed 25 foot walk; WBC; white blood cell.

Table 3

Unadjusted and adjusted discontinuation and efficacy outcomes.

 

Unadjusted and adjusted discontinuation and efficacy outcomes Unadjusted 1:1 Matching ATT Weighting
Discontinuation Outcomes at 12 Months
Discontinuation 1.39 1.78** 1.30
OR (95% CI) (0.99, 1.92) (1.10, 2.89) (0.84, 1.99)
Intolerability 1.53* 1.98** 1.27
 OR (95% CI) (1.06, 2.20) (1.19, 3.23) (0.78, 2.07)
Breakthrough Disease 1.05 1.35 1.28
 OR (95% CI) (0.63, 1.76) (0.64, 2.82) (0.72, 2.28)
Time to Discontinuation 1.41** 1.43** 1.35**
Relative Hazard Ratio (95% CI) (1.07, 1.86) (1.04, 1.96) (1.05, 1.74)
 
Efficacy Outcomes at 12 Months: Clinical Measures of Disease Activity
Annualized Relapse Rate (ARR) 1.28 1.29 1.56
ARR ratio (95% CI) (0.85, 1.91) (0.65, 2.56) (0.78, 3.14)
Time to First Relapse 1.31 1.06 1.64**
Relative Hazard Ratio (95% CI) (0.87, 1.99) (0.66, 1.71) (1.10, 2.46)
Timed-25 Foot Walk −0.66 −0.67 0.01
Difference (95% CI) (−1.90, 0.59) (−1.34, 1.94) (−1.22, 1.23)
T25FW 20% Worsening 1.22 1.24 0.92
 OR (95% CI) (0.86, 1.72) (0.74, 2.08) (0.57, 1.49)
9 Hole Peg Test 1.06 0.40 0.01
Difference (95% CI) (−2.07, 4.19) (−3.01, 3.89) (−0.06, 0.08)
9 Hole Peg Test 20% Worsening 1.17 0.18 1.06
 OR (95% CI) (0.55, 2.50) (0.02, 1.98) (0.47, 2.42)
 
Efficacy Outcomes at 12 Months: MRI Measures of Disease Activity
Brain MRI Activitya 1.29 1.56 1.18
OR (95% CI) (0.87, 1.92) (0.76, 3.17) (0.69, 1.99)
Brain MRI Gad-Enhancing Lesions 1.54 2.08 2.19**
 OR (95% CI) (0.86, 2.74) (0.73, 5.98) (1.10, 4.35)
Brain MRI New T2 Lesions 1.47 1.59 1.24
 OR (95% CI) (0.95, 2.27) (0.78, 3.20) (0.72, 2.13)
Brain MRI Activity on DMT 1.27 1.04 1.38
OR (95% CI) (0.81, 2.00) (0.40, 2.72) (0.78, 2.42)
Brain MRI Gad-Enhancing Lesions 1.52 2.05 2.90*
 OR (95% CI) (0.78, 2.95) (0.42, 9.99) (1.24, 6.57)
Brain MRI New T2 Lesions 1.54 1.99 1.33
 OR (95% CI) (0.93, 2.55) (0.77, 5.12) (0.71, 2.49)
 
Efficacy Outcomes at 12 Months: Patient Reported Outcome
PHQ-9 0.12 −0.14 −0.14
Difference (95% CI) (−0.76, 1.01) (−1.16, 0.89) (−1.27, 1.00)
Depressed (PHQ9≥10) 1.02 0.68 1.00
 OR (95% CI) (0.70, 1.50) (0.35, 1.35) (0.63, 1.60)

** p<0.01.

* p<0.05.

a Includes MRI data from the entire cohort, including patients who discontinued DMF or FTY.

DMT: disease modifying therapy; Gad: gadolinium; MRI: magnetic resonance imaging; PHQ-9: Patient Health Questionaire-9; T25FW: timed 25 foot walk.

By 12-month follow-up, 14.0% of DMF patients experienced a clinical relapse compared to 11.4% of FTY patients. Patients treated with DMF experienced a total of 69 relapses during a total of 423.5 patient-years of treatment. The ARR was 0.16 (95% CI (0.12, 0.18)). Patients treated with FTY experienced a total of 38 relapses during a total of 298 patient-years of treatment, resulting in an ARR of 0.13 (95% CI (0.08, 0.16)). There was no significant difference in ARR (rate ratio=1.28, 95% CI (0.85, 1.91)). After ATT weighting using the linear PS, there was also no significant difference in ARR (rate ratio=1.56, 95% CI (0.78, 3.14)), though DMF had earlier time to first relapse compared to FTY (HR=1.64, 95% CI (1.10, 2.46)). Mean time to first relapse was 3.85 months for DMF vs. 7.65 months for FTY (Fig. 1(a)). Other clinical measures of disease activity were comparable between treatment groups, including proportion with 20% worsening on T25FW or 9-HPT.

Fig. 1

Fig. 1

(a). Kaplan-Meier plot of relapse-free status through 12-month follow-up. (b). Kaplan-Meier plot of DMT discontinuation through 12-month follow-up. DMF: dimethyl fumarate; DMT: disease modifying therapy; FTY: fingolimod.

 

Of all patients who started DMT, odds of MRI activity, a combined measure of new T2 lesions and/or GdE lesions (OR=1.18, 95% CI (0.69, 1.99)), and new T2 lesions (OR=1.24, 95% CI (0.72, 2.13)) were comparable between DMF and FTY. Of those who underwent brain MRI while on DMT (DMF n=281, FTY n=248), 19.2% of patients on DMF demonstrated MRI activity (16.4% new T2 lesions) compared to 15.7% on FTY (11.3% new T2 lesions), but these differences were not significant (p>0.2).

There was a greater likelihood of developing GdE lesions on DMF compared to FTY (OR=2.19, 95% CI (1.10, 4.35)). In one sensitivity analysis, we re-assessed PS analysis only including patients with MRI data available (DMF n=370; FTY n=290) at 12 months. Overall covariate balance did not meaningfully change post-matching and post-weighting in this subset. After excluding patients with missing MRI data, DMF still had higher odds of developing GdE lesions compared to FTY (OR=2.04, 95% CI (1.04, 3.99)). The subgroup of patients with MRI while still on DMT (DMF n=281, 61%; FTY n=248, 78%) confirmed increased likelihood of GdE lesions at 12-month follow-up (OR=2.90, 95% CI (1.24, 6.57)).

At 12 months, 134 patients (29.3%) had discontinued DMF and 79 patients (24.9%) had discontinued FTY. The majority of patients discontinued DMT due to intolerability (DMF n=105, 22.9%; FTY n=58, 18.3%), with fewer discontinuations due to disease activity (DMF n=38, 8.3%; FTY n=27, 8.5%). A sizeable proportion of patients with progressive MS discontinued therapy (DMF=40%, FTY=38%). There were no significant differences in discontinuation rate or reason for discontinuation between groups. DMF patients discontinued medication earlier than FTY patients (HR=1.35, 95% CI (1.05, 1.74)): mean time of discontinuation in DMF=3.9 months vs. FTY=6.5 months. Kaplan-Meier survival estimates confirmed these findings (Fig. 1(b)).

Overall, there was no significant difference in mean PHQ-9 scores between DMF and FTY at 12-months: DMF=6.42 and FTY=6.30. Further, the proportions of patients with reported depression (defined as PHQ-9≥10) were comparable (OR=1.00, 95% CI (0.63, 1.60)).

A summary of unadjusted outcomes for the RRMS subgroup is presented in Table A1. Adjusted subgroup comparative analysis of RRMS patients on DMF versus FTY showed findings consistent with the larger cohort (Table A2); specifically no difference in ARR (DMF=0.17 with a total of 310.5 patient-years; FTY=0.14 with a total of 242 patient-years) or new T2 lesions, and higher proportion with GdE lesions and earlier time to discontinuation in the DMF group. There was a trend towards earlier first relapse with DMF (HR=1.55, 95% CI (0.98, 2.44)) but did not reach statistical significance.

4. Discussion

In the current study, we present a comparison of 12-month efficacy and discontinuation of DMF and FTY. We investigated the earliest clinical experience of each agent by focusing on the period immediately following FDA approval of each therapy. Through this analysis, we were able to investigate a large number of patients from a single academic center using cohorts similar in size to those of the individual treatment arms in phase 3 trials of the respective therapies (Cohen et al, 2010, Kappos et al, 2010, Fox et al, 2012a, Gold et al, 2012, and Calabresi et al, 2012).

No significant differences in ARR were found between treatment groups. However, patients treated with DMF experienced first clinical relapses earlier than FTY patients, possibly reflecting issues with early DMF adherence due to intolerance that may have resulted in early relapses. Earlier time to discontinuation in the DMF group (discussed in more detail below) also supports this finding. Overall, ARR in the treatment groups (DMF 14.0%, FTY 11.4%) were similar to those in phase 3 RCTs (DMF 17.0%–22.0%, FTY 16.0%–18.0%) (Cohen et al, 2010, Kappos et al, 2010, Fox et al, 2012a, and Gold et al, 2012), supporting moderate efficacy of each treatment in clinical practice at 12 months.

We found no difference in a composite measure of MRI activity or new T2 lesions between groups, though patients treated with DMF had two-fold higher odds of developing GdE lesions by 12 months compared to patients on FTY. Given earlier time to first relapse and higher proportion of GdE lesions in DMF, we should consider that our results showing no difference in ARR might have resulted from insufficient power to detect a difference, especially with the overall trend favoring FTY. However, it appeared the study design was robust enough to capture a true lack of effect with MDES=0.24. Alternatively, relapses were measured over one year, and DMF may take longer to become effective.

Proportion of patients with GdE lesions on FTY (6.0%) was similar to phase 3 RCT results (9.9%) (Cohen et al., 2010). Our cohort treated with DMF overall had a lower proportion of patients with GdE lesions (8.9%) compared to phase 3 RCT results (20%) (Fox et al., 2012b), likely reflective of our older population and greater proportion of patients with secondary progressive MS, as well as more frequent MRIs in clinical trials. However, only 40% of DMF patients in a phase 3 RCT had follow-up MRI evaluated (Fox et al., 2012a). Overall, our data supported high MRI efficacy at 12 months in clinical practice in both treatment groups.

No significant differences in clinical measures of neurologic impairment were observed between DMF and FTY. However, longer follow-up is needed to more properly power estimates of worsening disability (e.g. T25FW and 9-HPT).

Our 3-month experience showed 3-fold increased odds of discontinuation in DMF compared to FTY (OR=3.39, 95% CI (1.63, 7.03)) with greater odds of discontinuation due to intolerability (OR=4.84, 95% CI (2.06, 11.36)) (Cohn et al., 2014). Conversely, at 12 months, discontinuation rates and reason for discontinuation (intolerability and breakthrough disease activity) in DMF and FTY were comparable. The current study confirmed earlier time to discontinuation at 12 months in DMF, as we also observed in our 3-month investigation. These effects were related to earlier intolerability to gastrointestinal and flushing events, as was similarly seen in phase 3 trials (Fox et al, 2012a and Gold et al, 2012), while FTY patients discontinued medication at later time periods that we later captured in our 12-month assessment.

Our 12-month experience showed higher rates of drug discontinuation due to adverse effects compared to RCTs: 23% of DMF patients discontinued drug vs. 12%–16% in phase 3 trials (Fox et al, 2012a and Gold et al, 2012) and 18% of FTY patients discontinued drug vs. 5.6%–7.5% in phase 3 trials (Cohen et al, 2010 and Kappos et al, 2010). These findings are not unexpected given the differences between a highly selected and motivated clinical trial population and those treated in routine practice in a less regimented format with different motivations to continue treatment. Another consideration for increased discontinuation rates in our experience compared to clinical trials may be related to differences in disease course. Specifically, only patients with RRMS were eligible to participate in pivotal phase 3 RCTs, whereas our patient cohort included patients with progressive forms of the disease (DMF n=121, 26.4%; FTY n=58, 18.3%). A higher percentage of patients with progressive forms of MS, in whom FTY was not shown to have significant benefit (Novartis, 2014), may have perceived an increased risk: benefit ratio compared to patients with relapsing forms of MS and were therefore more inclined to discontinue medication earlier. This is supported by our findings that a sizeable proportion of our cohort with progressive MS discontinued therapy. Discontinuation may have also been increased in DMF owing to first clinical experience with unclear understanding of how to mitigate adverse effects. These differences highlight the importance of studies investigating patients in a real-world setting. Similar mean PHQ-9 scores between these groups demonstrated comparable perceptions of mood during treatment and did not explain discontinuations in general or any differences between the groups.

We used PS methods to improve the balance of baseline covariates between DMF and FTY. While PS analysis may lessen effects of certain biases (e.g. selection and indication biases), it does not reduce the effect of other types of biases (e.g. ascertainment and attrition biases) or account for unmeasured covariates. Overall, we believe the variables used in our PS model accurately reflect the characteristics healthcare providers use when deciding between MS treatments. Further, analysis of survival outcome measures, in addition to ARR ratio as the primary endpoint, allowed estimation of treatment effects that are more robust to attrition bias.

This study had several limitations. The patients were followed in a single large academic MS referral center, so the results may not be representative of the general population. The cohort included progressive patients, and their relatively low inflammatory activity may have obscured differences in treatment effects on the outcomes studied, though similar results were seen in the RRMS subgroup. The duration of follow-up was relatively short which also may have obscured differences in treatment effects. Limited availability of EDSS measures at baseline and routine follow-up resulted in exclusion of certain disability domains in the PS model not captured by ambulation metrics (e.g. T25FW and walking assistance). Finally, the number of patients with missing MRI or PRO data was substantial, which illustrates a common limitation of real world retrospective observational studies. However, analysis of the subgroup with complete data showed similar findings, which was reassuring.

5. Conclusions

Our study showed no differences in ARR and overall brain MRI activity in patients treated with DMF vs. FTY over 12 months. Although there was a consistent pattern across all measures favoring FTY, and several exploratory outcomes did reach statistical significance, including proportion with GdE lesions, time to first relapse, and time to discontinuation; the magnitude of any potential differences appears likely to be small. The authors are currently collaborating with another MS tertiary referral center to study a larger cohort to better address this question. Additional follow-up at 24 months will also be important to evaluate the longer term implications of these findings.

Declaration of conflicting interests

Dr. Carrie Hersh is supported by National Multiple Sclerosis Society Sylvia Lawry Physician Fellowship Award FP 1788-A-1.

Dr. Thomas Love- there is no conflict of interest.

Dr. Samuel Cohn – there is no conflict of interest.

Claire Hara-Cleaver has received consulting or speaking fees from Biogen Idec, TEVA, EMD Serono, Acorda, Novartis, and Genzyme.

Dr. Robert Bermel has received consulting or speaking fees from Biogen Idec, Novartis, TEVA, Genzyme, and Questcor.

Dr. Robert Fox has received consulting fees from Biogen Idec, MedDay, Novartis, Questcor, TEVA, and Xenoport.

Dr. Jeffrey Cohen has received consulting fees from Biogen Idec, EMD Serono, Genzyme, Novartis, Receptos, Synthon, TEVA, and Vaccinex.

Dr. Daniel Ontaneda is supported by KL2 TR000440/TR/NCATS NIH Grant.

Acknowledgments

Dr. Carrie Hersh conducted this study while supported by National Multiple Sclerosis Society Sylvia Lawry Physician Fellowship Award FP 1788-A-1. We thank the clinicians at the Cleveland Clinic Mellen Center who helped formulate and implement the fingolimod and dimethyl fumarate treatment protocols, as well as our patients. We also thank Maria Stadtler, CCRP for data management.

Appendix

See Fig A1, Fig A2, and Fig A3 and Table A1 and Table A2.

Fig. A1

Fig. A1

Density plot of propensity scores (propensity for DMF) showing adequate overlap of propensity scores between DMF and FTY. DMF: dimethyl fumarate; FTY: fingolimod.

 

Fig. A2

Fig. A2

Standardized difference plot comparing baseline covariates between DMF and FTY before and after greedy matching. Positive values represent higher standardized effect sizes for DMF. DMT: disease modifying therapy; EQ5D: European Quality of Life- 5 Dimensions; GA: glatiramer acetate; GdE: gadolinium-enhancing; IFN: interferon; intol: intolerability; IS: immunosuppressive therapy; IVMP: intravenous methylprednisolone; linps: linear propensity score; MRI: magnetic resonance imaging; MS: multiple sclerosis; MSPS: Multiple Sclerosis Performance Scale; NTZ: natalizumab; PHQ9: Patient Health Questionnaire-9; PS, propensity score; T25FW: timed 25 foot walk; Terifl: teriflunomide.

 

Fig. A3

Fig. A3

Standardized difference plot comparing baseline covariates between DMF and FTY before and after Average Treatment Effect on the Treated (ATT) weighting using the linear propensity score. Positive values represent higher standardized effect sizes for DMF. DMT: disease modifying therapy; EQ5D: European Quality of Life- 5 Dimensions; GA: glatiramer acetate; GdE: gadolinium-enhancing; IFN: interferon; intol: intolerability; IS: immunosuppressive therapy; IVMP: intravenous methylprednisolone; linps: linear propensity score; MRI: magnetic resonance imaging; MS: multiple sclerosis; MSPS: Multiple Sclerosis Performance Scale; NTZ: natalizumab; PHQ9: Patient Health Questionnaire-9; PS: propensity score; T25FW: timed 25 foot walk.

 

Table A1

Summary of unadjusted outcomes for RRMS patients at 12-month follow-up.

 

DMF Fingolimod
n=337 n=259
n % or SD n % or SD
Discontinued drug at 12 months 100 29.7% 58 22.4%
Disease Activity 28 8.3% 19 7.3%
Clinical relapse 9 2.7% 9 3.5%
MRI activity 17 5.0% 6 2.3%
Disability progression 10 3.0% 8 3.1%
Intolerance 73 21.7% 40 15.4%
Mean time to discontinuation (months, SD) 3.73 3.46 6.50 4.31
Median time to discontinuation (months) 3.0 6.0
Relapse data available for review 311 92.3% 259 100%
Clinical relapse at 12 months (number of patients) 45 13.4% 32 12.4%
Relapses per patient (mean, SD) 0.16 0.44 0.12 0.35
Mean time to relapse (months, SD) 3.99 3.25 7.84 4.27
MRI available for review 270 80.1% 240 92.7%
Disease activity on MRI at 12 months 68 20.2% 44 17.0%
Gadolinium enhancement 31 9.2% 18 6.9%
New T2 lesions 57 16.9% 33 12.7%
MRI available for review while on DMT 209 62.0% 210 81.1%
Disease activity on MRI at 12 months while on DMT 56 16.6% 36 13.9%
Gadolinium enhancement 27 8.0% 14 5.4%
New T2 Lesions 39 11.6% 26 10.0%
Adverse effects (number of patients) 235 69.8% 62 23.9%
Mean WBC (×109/L) 6.29 2.36 5.11 3.16
Mean ALC (×109/L) 1.52 1.01 0.61 0.52
Measures of neurologic impairment
T25FW (mean sec, SD) 6.45 (n=292) 3.97 7.51 (n=241) 10.52
9 HPT- dominant (mean, SD) 26.53 (n=156) 17.32 25.52 (n=101) 14.64
9 HPT- non-dominant (mean, SD) 25.15 (n=156) 11.59 24.15 (n=101) 7.60
Patient Reported Outcomes
PHQ-9 score (mean, SD) 6.38 (n=297) 5.83 6.29 (n=206) 5.55

9 HPT: 9-hole peg test; ALC: absolute lymphocyte count; DMF: dimethyl fumarate; DMT: disease modifying therapy; FTY: fingolimod; MRI: magnetic resonance imaging; PHQ-9: Patient Health Questionaire-9; SD: standard deviation; sec: seconds; T25FW: timed 25 foot walk; WBC; white blood cell.

Table A2

Unadjusted and adjusted tolerability and efficacy outcomes for RRMS patients.a

 

Causal effect of treatment Unadjusted ATT Weighting
Discontinuation Outcomes at 12 Months
Discontinuation 1.63* 1.42
OR (95% CI) (1.12, 2.37) (0.87, 2.32)
Intolerability 1.66* 1.34
 OR (95% CI) (1.09, 2.55) (0.77, 2.33)
Breakthrough Disease 1.24 1.40
 OR (95% CI) (0.68, 2.28) (0.62, 3.14)
Time to Discontinuation 1.61** 1.44**
Relative Hazard Rate (95% CI) (1.17, 2.23) (1.07, 1.94)
 
Efficacy Outcomes at 12 Months: Clinical Measures of Disease Activity
Annualized Relapse Rate (ARR) 1.22 1.59
ARR ratio (95% CI) (0.78, 1.90) (0.47, 5.41)
Time to First Relapse 1.29 1.55
Relative Hazard Rate (95% CI) (0.82, 2.02) (0.98, 2.44)
Timed-25 Foot Walk −1.06 −0.29
Difference (95% CI) (−2.37, 0.25) (−1.40, 0.81)
T25FW 20% Worsening 1.28 1.21
 OR (95% CI) (0.86, 1.90) (0.73, 2.02)
9 Hole Peg Test 0.10 0.02
Difference (95% CI) (−0.92, 1.12) (−0.07, 0.10)
9 Hole Peg Test 20% Worsening 1.11 1.20
 OR (95% CI) (0.73, 1.70) (0.38, 3.75)
 
Efficacy Outcomes at 12 Months: MRI Measures of Disease Activity
Brain MRI Activityb 1.44 1.66
OR (95% CI) (0.94, 2.20) (0.97, 2.83)
Brain MRI Gad-Enhancing Lesions 1.54 1.89
 OR (95% CI) (0.84, 2.83) (0.86, 4.18)
Brain MRI New T2 Lesions 1.62* 1.61
 OR (95% CI) (1.01, 2.59) (0.90, 2.90)
Brain MRI Activity on DMT 1.36 1.75
OR (95% CI) (0.84, 2.22) (0.95, 3.20)
Brain MRI Gad-Enhancing Lesions 1.48 2.63*
 OR (95% CI) (0.73, 3.02) (1.07, 6.47)
Brain MRI New T2 Lesions 1.62 1.74
 OR (95% CI) (0.95, 2.78) (0.89, 3.40)
 
Efficacy Outcomes at 12 Months: Patient Reported Outcome
PHQ-9 0.10 −0.49
Difference (95% CI) (−0.92, 1.12) (−1.91, 0.92)
Depressed (PHQ910) 1.11 0.91
 OR (95% CI) (0.73, 1.70) (0.53, 1.57)

* p<0.05.

** p<0.01.

a Total n= 596 (DMF= 337, FTY= 259)

b Includes MRI data from the entire cohort, including patients who discontinued DMF or FTY.

DMT: disease modifying therapy; Gad: gadolinium; MRI: magnetic resonance imaging; PHQ-9: Patient Health Questionaire-9; T25FW: timed 25 foot walk.

References

  • Austin, 2011 P. Austin. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivar. Behav. Res.. 2011;46:399-424 Crossref
  • Calabresi et al., 2012 Calabresi P.A., Goodin D.S., Jeffery D., et al., 2012. Efficacy and safety of fingolimod versus placebo: primary outcomes from the phase 3 FREEDOMS II study in patients with relapsing–remitting multiple sclerosis.
  • Cohen et al., 2010 J.A. Cohen, F. Barkhof, G. Comi, et al. Oral fingolimod or intramuscular interferon for relapsing multiple sclerosis. New Engl. J. Med.. 2010;362:402-415 Crossref
  • Cohn et al., 2014 Cohn S., Bermel R.A., Hara-Cleaver C., et al., 2014. Comparative tolerability and efficacy of dimethyl fumarate and fingolimod in multiple sclerosis. In: European Committee for Treatment and Research in Multiple Sclerosis, Boston, MA, 10–13 June.
  • D'Agostino, 1998 R. D'Agostino. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat. Med.. 1998;17:2265-2281 Crossref
  • Fox et al., 2012a R. Fox, D. Miller, J.T. Phillips, et al. Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis. New Engl. J. Med.. 2012;367:1087-1097 Crossref
  • Fox et al., 2012b R. Fox, D. Miller, J.T. Phillips, et al. Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis: supplementary appendix. New Engl. J. Med.. 2012;367:1087-1097 Crossref
  • Gold et al., 2012 R. Gold, L. Kappos, D.L. Arnold, et al. Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis. New Engl. J. Med.. 2012;367:1098-1107 Crossref
  • Grimes and Schulz, 2002 D.A. Grimes, K.F. Schulz. Bias and causal associations in observational research. Lancet. 2002;359:248-252 Crossref
  • Hirano et al., 2003 K. Hirano, G.W. Imbens, G. Ridder. Efficient estimation of average treatment effects using the estimated propensity score. Econometrica. 2003;71:1161-1189 Crossref
  • Kappos et al., 2010 L. Kappos, E.-W. Radue, P. O'Connor, et al. A placebo-controlled trial of oral fingolimod in relapsing multiple sclerosis. New Engl. J. Med.. 2010;362:387-401 Crossref
  • Katzan et al., 2011 I. Katzan, M. Speck, C. Dopler, et al. The knowledge program: an innovative, comprehensive, electronic data capture system and warehouse. AMIA Annu. Symp. Proc.. 2011;:683-692
  • Kragt et al., 2006 J.J. Kragt, F.A. van der Linden, J.M. Nielsen, B.M. Uitdehaag, C.H. Polman. Clinical impact of 20% worsening on timed 25-foot walk and 9-hole peg test in multiple sclerosis. Mult. Scler.. 2006;12:594-598 Crossref
  • Kroenke et al., 2001 K. Kroenke, R.L. Spitzer, J.B. Williams. The PHQ-9: validity of a brief depression severity measure. J. Gen. Intern. Med.. 2001;16:606-613 Crossref
  • Novartis, 2014 Novartis. Novartis provides update oln fingolimod phase III trial in primary progressive MS (PPMS). Novartis Glob.. 2014;
  • Harris et al., 2009 P.A. Harris, R. Taylor, R. Thielke, J. Payne, N. Gonzalez, J.G. Conde. Research electronic datacapture (REDCap) - A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377-381 Crossref
  • Phillips and Fox, 2013 J. Phillips, R. Fox. BG-12 in multiple sclerosis. Semin. Neurol.. 2013;33:56-65
  • Rosenbaum and Rubin, 1983 P.R. Rosenbaum, D.B. Rubin. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41-55 Crossref
  • Rubin, 2001 D.B. Rubin. Using propensity scores to help design observational studies: application to the tobacco litigation. Health Serv. Outcomes Res. Methodol.. 2001;2:169-188 Crossref
  • Rudick et al., 2002 R.A. Rudick, G. Cutter, S. Reingold. The multiple sclerosis functional composite: a new clinical outcome measure for multiple sclerosis trials. Mult. Scler.. 2002;8:359-365 Crossref
  • Schwartz et al., 1999 C.E. Schwartz, T. Vollmer, H. Lee. Reliability and validity of two self-report measures of impairment and disability for MS. North American research consortium on multiple sclerosis outcomes study group. Neurology. 1999;52:63-70
  • Team, 2014 R.C. Team. R: A Language and Environment for Statistical Computing. (R Foundation for Statistical Computing, Vienna, Austria, 2014)
  • EuroQol, 1990 The EuroQol Group, 1990. EuroQol--a new facility for the measurement of health-related quality of life, 16, pp. 199–208.
  • Yang et al., 2010 W. Yang, A. Zilov, P. Soewondo, O.M. Bech, F. Sekkal, P.D. Home. Observational studies: going beyond the boundaries of randomized controlled trials. Diabetes Res. Clin. Pract.. 2010;88:S3-S9 Crossref

Footnotes

a Lou Ruvo Center for Brain Health, Cleveland Clinic, 888 W. Bonneville Ave, Las Vegas, NV, 89106 USA

b Department of Epidemiology and Biostatistics, Case Western Reserve University, 10900 Euclid Ave, Cleveland, 44106 USA

c Department of Neurology, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH, 44195 USA

d Mellen Center for Multiple Sclerosis Treatment and Research, Cleveland Clinic, 9500 Euclid Ave, Cleveland, OH, 44195 USA

Corresponding author.


Search this site

Stay up-to-date with our monthly e-alert

If you want to regularly receive information on what is happening in MS research sign up to our e-alert.

Subscribe »

About the Editors

  • Prof Timothy Vartanian

    Timothy Vartanian, Professor at the Brain and Mind Research Institute and the Department of Neurology, Weill Cornell Medical College, Cornell...
  • Dr Claire S. Riley

    Claire S. Riley, MD is an assistant attending neurologist and assistant professor of neurology in the Neurological Institute, Columbia University,...
  • Dr Rebecca Farber

    Rebecca Farber, MD is an attending neurologist and assistant professor of neurology at the Neurological Institute, Columbia University, in New...

This online Resource Centre has been made possible by a donation from EMD Serono, Inc., a business of Merck KGaA, Darmstadt, Germany.

Note that EMD Serono, Inc., has no editorial control or influence over the content of this Resource Centre. The Resource Centre and all content therein are subject to an independent editorial review.

The Grant for Multiple Sclerosis Innovation
supports promising translational research projects by academic researchers to improve understanding of multiple sclerosis (MS) for the ultimate benefit of patients.  For full information and application details, please click here

Journal Editor's choice

Recommended by Prof. Brenda Banwell

Causes of death among persons with multiple sclerosis

Gary R. Cutter, Jeffrey Zimmerman, Amber R. Salter, et al.

Multiple Sclerosis and Related Disorders, September 2015, Vol 4 Issue 5