Advertisement

Clinical outcomes in patients with relapsing-remitting multiple sclerosis who switch from natalizumab to delayed-release dimethyl fumarate: A multicenter retrospective observational study (STRATEGY)

Open AccessPublished:February 26, 2018DOI:https://doi.org/10.1016/j.msard.2018.02.028

      Highlights

      • Assessed patients with RRMS who switched from NAT to DMF in clinical practice.
      • DMF may be a therapeutic option for patients who discontinue NAT.
      • NAT patients, notably non-relapsers, switching to DMF in ≤ 90 days had fewer relapses.

      Abstract

      Background

      Delayed-release dimethyl fumarate (DMF) may be a therapeutic option for patients with relapsing-remitting multiple sclerosis (RRMS) who are treated with natalizumab and require a change in therapy. However, there is limited information regarding predictors of favorable treatment outcomes in patients switching from natalizumab to DMF. Clinical practices and sequencing protocols vary. Herein, we present the clinical results, including annualized relapse rate (ARR) and risk of relapse, of a phase 4 retrospective observational study of patients with RRMS who switched from natalizumab to DMF in a community practice setting (STRATEGY).

      Methods

      STRATEGY was performed through a single time point medical record abstraction; no study visits or procedures were required. Key inclusion criteria included age ≥ 18 years, RRMS diagnosis (McDonald criteria, 2010 revised), ≥ 12 months of continuous treatment with natalizumab monotherapy before DMF initiation, and initiation of DMF ≥ 12 months before enrollment. Patients were eligible to enroll regardless of current DMF use.

      Results

      A total of 530 patients at 45 US sites enrolled, and 506 met the inclusion criteria and were included in the modified evaluable population for analysis. Mean (SD) age at DMF initiation was 47.0 (10.9) years, with a mean (SD) of 12.7 (7.2) years since MS diagnosis. The mean (SD) duration of natalizumab treatment was 3.4 (1.9) years, and the mean (SD) washout from natalizumab discontinuation to DMF initiation (n = 502) was 101.6 (164.0) days. Overall risk of relapse 12 months after DMF initiation was 19.6%. Overall unadjusted ARR was higher during the 12 months following initiation of DMF treatment compared with the 12 months following initiation of natalizumab treatment (rate ratio, 2.32 [95% CI, 1.69–3.18]; p < 0.0001), but was lower compared with that observed in the year before initiation of natalizumab (rate ratio, 0.51 [95% CI, 0.40–0.64]; p < 0.0001). At 1 year following initiation of DMF treatment, the relapse rate was lower for patients who did not experience a relapse during 1 year following initiation of natalizumab treatment than for those who did (rate ratio for relapse rate, 0.47 [95% CI, 0.16–1.38]; p = 0.1664). The relapse rate for patients who did not relapse during natalizumab treatment was significantly lower with a washout period of ≤ 90 days as compared with a washout period of > 90 days (rate ratio for relapse rate, 0.49 [95% CI, 0.26–0.90]; p = 0.0216). A total of 42 (8%) patients reported ≥ 1 adverse event leading to DMF discontinuation during the study; the most commonly reported events were gastrointestinal disorders (n = 21; 4%).

      Conclusions

      Results from this multicenter retrospective observational study suggest that DMF may be an effective treatment option for patients who discontinue natalizumab in routine clinical practice. ARR was lower in patients who initiated DMF within 90 days of natalizumab discontinuation compared with patients who initiated DMF after 90 days of natalizumab discontinuation.

      Trial registration number

      ClinicalTrials.gov identifier NCT02159573.

      Abbreviations:

      AE (adverse event), ARR (annualized relapse rate), DMF (dimethyl fumarate), DMT (disease-modifying therapy), HR (hazard ratio), IPTW (inverse probability of treatment weight), JCV (John Cunningham virus), MS (multiple sclerosis), PML (progressive multifocal leukoencephalopathy), RRMS (relapsing-remitting multiple sclerosis), SAE (serious adverse event), SIPTW (stabilized inverse probability of treatment weight)

      Keywords

      1. Introduction

      Multiple sclerosis (MS) is a chronic inflammatory and degenerative disease of the central nervous system, typically characterized by episodes of neurological dysfunction (
      • Compston A.
      • Coles A.
      Multiple sclerosis.
      ,
      • Noseworthy J.H.
      • Lucchinetti C.
      • Rodriguez M.
      • Weinshenker B.G.
      Multiple sclerosis.
      ). Approximately 90% of patients with MS are initially diagnosed with relapsing-remitting MS (RRMS), involving acute attacks of neurological symptoms followed by partial or complete recovery periods (
      • Tullman M.J.
      Overview of the epidemiology, diagnosis, and disease progression associated with multiple sclerosis.
      ).
      Disease-modifying therapies (DMTs) effectively reduce the frequency and severity of MS relapses. However, as MS is a complex, chronic disorder, multiple approaches are required to control the clinical and pathological progression and minimize long-term DMT risks. When a DMT's benefit-risk profile is no longer ideal, a switch should be considered, accounting for the disease activity, severity, pharmacodynamics, and pharmacokinetic properties of the previous therapy, and the risk of reactivation of disease activity (

      Otero-Romero, S., Amato, M., Chandraratna, D., Clanet, M., Comi, G., Derfuss, T., et al., 2016. ECTRIMS-EAN clinical practice guideline on pharmacological management of multiple sclerosis. Presented at the 32nd Congress of the European Committee for Treatment and Research in Multiple Sclerosis, London, UK, September 14–17.

      ).
      Natalizumab is a highly effective monotherapy approved for relapsing forms of MS in the United States (,
      • Butzkueven H.
      • Kappos L.
      • Pellegrini F.
      • Trojano M.
      • Wiendl H.
      • Patel R.N.
      • et al.
      Efficacy and safety of natalizumab in multiple sclerosis: interim observational programme results.
      ). Treatment with natalizumab is associated with the risk of progressive multifocal leukoencephalopathy (PML), and consideration of factors that increase that risk, such as positive serum anti-John Cunningham virus (JCV) antibody levels, prior exposure to other immunosuppressants, and duration of treatment, may influence treatment recommendations of whether to initiate or continue natalizumab therapy (,

      European Medicines Agency, 2016b. Tysabri 300 mg concentrate for solution for infusion. EU Summary of Product Characteristics. 〈http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000603/WC500044686.pdf〉 (Accessed 31 October 2016).

      ,
      • Plavina T.
      • Subramanyam M.
      • Bloomgren G.
      • Richman S.
      • Pace A.
      • Lee S.
      • et al.
      Anti-JC virus antibody levels in serum or plasma further define risk of natalizumab-associated progressive multifocal leukoencephalopathy.
      ).
      Delayed-release dimethyl fumarate (DMF) is an oral DMT approved for relapsing forms of MS in the United States (

      Biogen Inc. TECFIDERA® (dimethyl fumarate) delayed-release capsules, for oral use: US prescribing Information. 2016a; http://www.tecfidera.com/pdfs/full-prescribing-info.pdf. Accessed May 5, 2016.

      ) and for adults with RRMS in the European Union (

      European Medicines Agency, 2014. Tecfidera 120 mg gastro-resistant hard capsules; tecfidera 240 mg gastro-resistant hard capsules. Summary of Product Characteristics. 〈http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/002601/WC500162069.pdf〉 (Accessed 5 June 2016).

      ). DMF has demonstrated significant efficacy and a favorable benefit-risk profile in the 2-year phase 3 DEFINE and CONFIRM trials of RRMS, along with sustained efficacy and safety in the ENDORSE extension study (
      • Fox R.J.
      • Miller D.H.
      • Phillips J.T.
      • Hutchinson M.
      • Havrdova E.
      • Kita M.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis.
      ,
      • Gold R.
      • Giovannoni G.
      • Phillips J.T.
      • Fox R.J.
      • Zhang A.
      • Marantz J.L.
      Sustained effect of delayed-release dimethyl fumarate in newly diagnosed patients with relapsing-remitting multiple sclerosis: 6-year interim results from an extension of the DEFINE and CONFIRM studies.
      ,
      • Gold R.
      • Kappos L.
      • Arnold D.L.
      • Bar-Or A.
      • Giovannoni G.
      • Selmaj K.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis.
      ). Despite evidence for activation of the nuclear factor (erythroid derived 2)-related factor 2 (Nrf2) and hydroxycarboxylic acid receptor 2 (HCAR2) (
      • Chen H.
      • Assmann J.C.
      • Krenz A.
      • Rahman M.
      • Grimm M.
      • Karsten C.M.
      • et al.
      Hydroxycarboxylic acid receptor 2 mediates dimethyl fumarate's protective effect in EAE.
      ,
      • Fox R.J.
      • Kita M.
      • Cohan S.L.
      • Henson L.J.
      • Zambrano J.
      • Scannevin R.H.
      • et al.
      BG-12 (dimethyl fumarate): a review of mechanism of action, efficacy, and safety.
      ), the basis for its therapeutic effects has not been established. As of September 30, 2017, over 280,000 patients have been treated with DMF worldwide, representing over 464,000 patient-years of exposure. As of September 2017, 5 cases of PML have been reported in patients treated with DMF, each in the setting of prolonged, moderate-to-severe lymphopenia (Biogen data on file). Aside from these occurrences of PML, no overall increased risk for serious infections, including other opportunistic infections, has been observed (
      • Fox R.J.
      • Chan A.
      • Gold R.
      • Phillips J.T.
      • Selmaj K.
      • Chang I.
      • et al.
      Characterizing absolute lymphocyte count profiles in dimethyl fumarate-treated patients with MS: patient management considerations.
      ).
      Although patients discontinuing natalizumab were included in DEFINE and CONFIRM, a 6-month washout between agents was required (
      • Fox R.J.
      • Miller D.H.
      • Phillips J.T.
      • Hutchinson M.
      • Havrdova E.
      • Kita M.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis.
      ,
      • Gold R.
      • Kappos L.
      • Arnold D.L.
      • Bar-Or A.
      • Giovannoni G.
      • Selmaj K.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis.
      ). Return of radiographic and clinical disease activity have been observed as early as 4–12 weeks following natalizumab discontinuation (
      • Fox R.J.
      • Cree B.A.
      • De Seze J.
      • Gold R.
      • Hartung H.P.
      • Jeffery D.
      • et al.
      MS disease activity in RESTORE: a randomized 24-week natalizumab treatment interruption study.
      ). Theoretically, an additive pharmacodynamic effect could occur if DMF was initiated very early after discontinuation of natalizumab. Although the safety of concomitant natalizumab and DMF has not been established, post hoc pivotal trial analyses suggest that the efficacy of DMF on annualized relapse rate (ARR) and magnetic resonance imaging activity first becomes statistically significant after 10–12 weeks of treatment (
      • Kappos L.
      • Giovannoni G.
      • Gold R.
      • Phillips J.T.
      • Arnold D.L.
      • Hotermans C.
      • et al.
      Time course of clinical and neuroradiological effects of delayed-release dimethyl fumarate in multiple sclerosis.
      ). Therefore, an excessive delay in DMF initiation could increase the risk of recurrent disease activity in patients discontinuing natalizumab (
      • Kappos L.
      • Radue E.W.
      • Comi G.
      • Montalban X.
      • Butzkueven H.
      • Wiendl H.
      • et al.
      Switching from natalizumab to fingolimod: a randomized, placebo-controlled study in RRMS.
      ).
      The STRATEGY (Multicenter, Retrospective, Observational Study Evaluating Real-world Clinical Outcomes in Relapsing-remitting Multiple Sclerosis Patients Who Transition from Tysabri® [Natalizumab] to Tecfidera® [Dimethyl Fumarate]) study evaluated the effectiveness of DMF in patients transitioning from natalizumab in the clinical practice setting, as well as the association between washout duration and relapse activity.

      2. Methods

      2.1 Study design and patients

      STRATEGY (NCT02159573) was a phase 4 retrospective observational study of patients with RRMS who switched from natalizumab to DMF. Retrospective data, drawn from patient medical records and other documents at a single time point, were collected by trained site staff from 45 US clinical practice sites and recorded in an electronic case report form. Although the focus was on disease activity 1 year after DMF initiation, data were collected before natalizumab initiation to obtain an estimate of ARR before and during the 1 year after natalizumab initiation, and in the year after DMF initiation (Fig. 1).
      Fig. 1
      Fig. 1Study design. Retrospective observational study performed through a single time point retrospective medical record abstraction. aDMF, delayed-release dimethyl fumarate.
      Eligible patients were ≥ 18 years of age at screening/enrollment, had a diagnosis of RRMS per 2010 revised McDonald criteria (
      • Polman C.H.
      • Reingold S.C.
      • Banwell B.
      • Clanet M.
      • Cohen J.A.
      • Filippi M.
      • et al.
      Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria.
      ), had received ≥ 12 months of natalizumab and no other DMT between natalizumab discontinuation and DMF initiation, initiated DMF ≥ 1 year before study enrollment, and had available medical records to meet study objectives from 12 months before natalizumab treatment through 12 months after DMF initiation. Patients were not required to be receiving DMF during medical record abstraction. It was assumed that natalizumab and DMF were administered according to the corresponding package insert instructions. However, medication usage was solely at the discretion of the physician and was evaluated retrospectively. In the United States, the recommended dose for natalizumab is 300 mg infused intravenously over 1 h every 4 weeks (); for DMF, an initial dose of 120 mg twice daily for the first week is followed by a maintenance dose of 240 mg twice daily (

      Biogen Inc. TECFIDERA® (dimethyl fumarate) delayed-release capsules, for oral use: US prescribing Information. 2016a; http://www.tecfidera.com/pdfs/full-prescribing-info.pdf. Accessed May 5, 2016.

      ). Patients were excluded if they had a progressive form of MS, received another DMT after natalizumab discontinuation and before DMF initiation, received any other formulation of DMF or compounded fumarate before DMF initiation, or had a history of PML while on natalizumab or within 6 months of discontinuation.

      2.2 Clinical effectiveness endpoints

      The primary effectiveness endpoint was the proportion of patients who relapsed during the 12 months after DMF initiation. Secondary endpoints included ARR at 1 year after DMF initiation, the proportion of patients with a relapse requiring treatment with intravenous steroids, and the proportion of patients with MS-related hospitalization at 1 year after DMF initiation. Three exploratory endpoints included the associations between: (1) ARR before the first 12 months of receiving natalizumab treatment, during the first 12 months of receiving natalizumab, and during the first 12 months of receiving DMF; (2) washout duration from natalizumab discontinuation to DMF initiation and relapse occurrence at 1 year after DMF initiation; and (3) relapse resulting in short courses of steroid use during washout period, and ARR while on DMF. Relapses were identified and recorded by site investigators, without independent source confirmation. An assessment was also conducted to examine if an association existed between clinical disease activity during DMF treatment and age at DMF initiation.

      2.3 Safety assessment

      Only adverse events (AEs) or serious AEs (SAEs) leading to DMF discontinuation were proactively collected. All other AEs and SAEs were reported through post-marketing reporting procedures in accordance with their standard spontaneous post-marketing reporting procedures for other non-serious AEs. All identified causes of death from the washout period through 1 year after DMF initiation were reported as SAEs. Results from anti-JCV antibody tests, categorized as positive or negative, were collected for all patients.

      2.4 Matching and statistical analysis

      The proportion of patients who relapsed by 1 year after DMF initiation was assessed using Kaplan-Meier estimates based on time to first relapse survival distributions. Unadjusted ARR was calculated as the total number of relapses divided by the total number of patient-years in a period for all patients; 95% CI for the ARR ratio 1 year before and 1 year after DMF initiation was based on an empirical standard error derived from a generalized estimating equation based on an unadjusted Poisson regression model. Washout duration was defined as the date of first DMF dose minus the date of last natalizumab infusion. The overall comparisons and comparisons by number of relapses used adjusted ARR estimates 1 year after DMF initiation based on a negative binomial regression model adjusted for gender, age, natalizumab treatment duration (≤ vs > median), and relapse during natalizumab treatment and washout (yes vs no).
      To assess the impact of washout on ARR 1 year after DMF initiation, patients were stratified into early (≤ 90 days) and delayed (> 90 days) switcher groups, based on the lower limit suggested for a 12- to 16-week natalizumab washout. Baseline characteristics (age, number of relapses during natalizumab treatment and washout, duration of natalizumab treatment, and steroid use during washout) of these 2 groups were balanced using propensity score. A stabilized inverse probability of treatment weight (SIPTW) estimator was used to assess ARR in these 2 groups. The inverse probability of treatment weight (IPTW) estimates a standardized effect using the entire study population as the standard population (
      • Sturmer T.
      • Rothman K.J.
      • Avorn J.
      • Glynn R.J.
      Treatment effects in the presence of unmeasured confounding: dealing with observations in the tails of the propensity score distribution—a simulation study.
      ). A standardized effect using the population with washout duration ≤ 90 days as the standard population was also estimated (the result was similar to that based on IPTW above [not shown]). The balance in baseline covariate distribution between the 2 groups was assessed using the t-test and Kolmogorov-Smirnov test for continuous variables and the chi-square test for categorical variables. A good balance was achieved in the baseline characteristics for the washout duration categories. The estimated hazard ratio (HR) for impact of washout duration (≤ 90 vs > 90 days) among patients without relapse during natalizumab treatment was based on a Cox proportional hazard model using the SIPTW estimator. Variability of the HR was based on a robust (sandwich) standard error. The estimated HR for occurrence of relapse (0 vs ≥ 1) during natalizumab treatment and washout period was based on a Cox proportional hazard model adjusted for age, gender, and duration of natalizumab treatment (≤ vs > median).
      In addition to overall population analyses, disease activity among patients with or without relapse during natalizumab treatment and washout was evaluated using a Cox proportional hazard model to estimate the HR for comparing relapse risk within subgroups (0 vs ≥ 1 relapse). However, as the study was not primarily designed as a comparative study, all p values should be considered exploratory. As with non-randomized or non-controlled studies, 95% CIs represent a more informative alternative to a significance test based on a p value, and are more relevant to evaluating group differences in observational studies (
      • Wasserstein R.L.
      • Lazar N.
      The ASA's statement on p-values: context, process, and purpose.
      ). A post hoc analysis estimated relapse activity after stratifying by quartiles of washout duration.

      2.5 Ethical considerations

      Investigators obtained ethics committee/institutional review board approval for the study protocol and all amendments. The study was performed in accordance with relevant US Code of Federal Regulations, the International Conference on Harmonisation guidelines on Good Clinical Practice, and the Declaration of Helsinki. Waivers of informed consent were pursued as permitted by local legislation. When informed consent waivers were not granted, institutional review board–approved written informed consent was obtained from each patient before eligibility evaluations were performed. Patients received a copy of the signed and dated informed consent form.

      3. Results

      3.1 Patients and study design

      Between July 30, 2014 and April 16, 2016, 530 US patients were enrolled and 518 (98%) met the inclusion criteria and received ≥ 1 dose of DMF. Patients who had distinguishable relapse and DMF start dates were included in the modified evaluable population (96%; 506/530; Fig. 2). Table 1 summarizes the baseline and clinical characteristics, including relapse history, of the modified evaluable population.
      Fig. 2
      Fig. 2Patient disposition. Disposition of patients enrolled in STRATEGY. aPatients could have had multiple reasons for ineligibility and study exclusion. bDMF, delayed-release dimethyl fumarate; MS, multiple sclerosis.
      Table 1Patient demographic characteristics and MS history.
      CharacteristicModified evaluable population
      (n = 506)
      Mean (SD) age at time of DMF initiation, years47.0 (10.9)
      Age range, years, n (%)
       18–191 (< 1)
       20–2930 (6)
       30–39101 (20)
       40–49163 (32)
       50–59138 (27)
       ≥6073 (14)
      Female, n (%)347 (69)
      Race, n (%)
       Asian2 (< 1)
       Black or African American46 (9)
       White413 (82)
       Other45 (9)
      Mean (SD) time since first MS symptoms,
      n = 490.
      years
      15.0 (8.3)
      Mean (SD) time since MS diagnosis,
      n = 498.
      years
      12.7 (7.2)
      Mean (SD) no. of relapses ≤ 1 year before natalizumab initiation0.5 (0.7)
       Min, max0, 4
      Relapses ≤ 1 year before natalizumab initiation, n (%)
       0317 (63)
       1144 (29)
       235 (7)
       38 (2)
       ≥ 42 (< 1)
      Mean (SD) duration of natalizumab treatment, years3.4 (1.9)
      ARR (95% CI) during first year of natalizumab treatment0.11 (0.08–0.14)
      ARR (95% CI) during overall natalizumab treatment0.11 (0.09–0.14)
      Mean (SD) no. of relapses during natalizumab treatment0.4 (0.8)
       Min, max0, 4
      Relapses during natalizumab treatment, n (%)
       0387 (77)
       185 (17)
       ≥ 234 (6)
      Proportion completing JCV testing, n (%)500 (99)
       Positive
      n = 500.
      458 (92)
      ARR, annualized relapse rate; DMF, delayed-release dimethyl fumarate; JCV, John Cunningham virus; MS, multiple sclerosis.
      a n = 490.
      b n = 498.
      c n = 500.
      Patients discontinued natalizumab during this study due to investigator decision (43%), patient's preference (34%), and other (16%). Nearly all patients (99%; 500/506) had an anti-JCV antibody test result; 92% (458/500) of patients had ≥ 1 positive anti-JCV antibody test result.
      Most (92%) patients had received an MS treatment, most commonly interferon beta-1a (72%) and/or glatiramer acetate (39%), before natalizumab (Table 2). The mean (SD) duration of natalizumab treatment was 3.4 (1.9) years, and the mean (SD) washout from natalizumab discontinuation to DMF initiation was 101.6 (164.0) days. The washout duration was ≤ 90 days for 359 patients compared with > 90 days for 147 patients. A total of 33 (7%) patients experienced relapse during washout and 2 (0.4%) patients experienced ≥ 2 relapses. Most (83%) evaluable patients received DMF for ≥ 1 year.
      Table 2Prior treatment, duration of natalizumab treatment, duration of washout, and reasons for DMF discontinuation.
      CharacteristicModified evaluable population
      (n = 506)
      MS treatment before natalizumab,
      Patients could have taken > 1 MS therapy.
      n (%)
       Any466 (92)
       Interferon beta-1a364 (72)
       Interferon beta-1b94 (19)
       Glatiramer acetate195 (39)
       Chronic MS-related steroid23 (5)
       Fingolimod13 (3)
       Other
      Teriflunomide, mitoxantrone, rituximab, estriol, laquinimod, methotrexate, naltrexone, or daclizumab.
      15 (3)
      Mean (SD) duration of natalizumab treatment, years3.4 (1.9)
       Median (25th, 75th percentile)2.9 (1.8, 5.0)
      Reason for natalizumab discontinuation,
      Patients could have had > 1 reason for discontinuation.
      n (%)
       Adverse event13 (3)
       Efficacy reasons; patient perception9 (2)
       Efficacy reasons; investigator perception32 (6)
       Investigator decision218 (43)
       Patient preference174 (34)
       Other80 (16)
      Washout duration, days, n (%)
       0–30129 (25)
       31–60133 (26)
       61–9097 (19)
       91–12049 (10)
       121–15027 (5)
       > 15071 (14)
      Mean (SD) washout duration,
      n = 502.
      days
      101.6 (164.0)
      Patients with relapses during washout, n (%)33 (7)
      Relapses during washout, n (%)
       0473 (93)
       ≥ 133 (7)
       > 12 (0.4)
      Duration of DMF treatment, months, n (%)
       ≥ 1499 (99)
       ≥ 3484 (96)
       ≥ 6456 (90)
       ≥ 12418 (83)
      Reason for DMF discontinuation,
      Patients could have had > 1 reason for discontinuation.
      n (%)
       Adverse event32 (6)
       Efficacy reasons; patient perception15 (3)
       Efficacy reasons; investigator perception9 (2)
       Investigator decision3 (< 1)
       Patient preference16 (3)
       Other9 (2)
      DMF, delayed-release dimethyl fumarate; MS, multiple sclerosis.
      a Patients could have taken > 1 MS therapy.
      b Teriflunomide, mitoxantrone, rituximab, estriol, laquinimod, methotrexate, naltrexone, or daclizumab.
      c Patients could have had > 1 reason for discontinuation.
      d n = 502.

      3.2 Relapse activity

      3.2.1 Risk of relapse and ARR on DMF treatment

      The Kaplan-Meier estimate of overall risk of relapse 1 year after DMF initiation was 19.6%. Over this timespan, 82% (413/506) of patients experienced 0 relapses, 15% (75/506) experienced 1 relapse, 3% (15/506) experienced 2 relapses, and 0.6% (3/506) experienced 3 relapses. Overall, unadjusted ARR was higher for the first year following DMF initiation compared with the first year following natalizumab initiation. The unadjusted ARR for the 1 year following DMF initiation was lower than that observed 1 year before natalizumab initiation (Fig. 3).
      Fig. 3
      Fig. 3Overall unadjusted ARR 1 year before natalizumab initiation, 1 year following initiation of natalizumab treatment, and after 1 year following initiation of DMF treatment. n = 506 for all periods. a Unadjusted ARR for each period calculated as the total number of relapses that occurred during the period for all patients, divided by the total number of patient-years followed in that period. 95% CI based on an empirical (robust) standard error from a generalized estimating equation using an unadjusted Poisson regression model. ARR, annualized relapse rate; DMF, delayed-release dimethyl fumarate.

      3.2.2 ARR and risk of relapse stratified by relapse and washout duration

      For patients who relapsed in the year before natalizumab treatment initiation, the adjusted ARR 1 year after DMF initiation was 0.34 (95% CI, 0.25–0.46) compared with 0.24 (95% CI, 0.18–0.32) for patients without relapses in the year before natalizumab treatment initiation (Table 3). The relapse rate ratio was 1.40 (95% CI, 0.94–2.08; p = 0.0953). For patients who relapsed during natalizumab treatment and washout, the adjusted ARR 1 year after DMF initiation was 0.46 (95% CI, 0.34–0.63) compared with 0.17 (95% CI, 0.13–0.23) for patients without relapses during these periods (relapse rate ratio, 0.38 [95% CI, 0.25–0.57]; p < 0.0001). The Kaplan-Meier estimate of risk of relapse 1 year after DMF initiation was 14.4% for patients without relapse during natalizumab treatment compared with 37.6% for patients who experienced ≥ 1 relapse during natalizumab treatment (HR, 0.35 [95% CI, 0.23–0.53]; p < 0.0001; Fig. 4a).
      Table 3Summary of ARR 1 year after DMF initiation stratified by relapse 1 year before natalizumab initiation.
      Includes modified evaluable population for sensitivity analysis.
      CharacteristicRelapse 1 year before natalizumab initiation
      RelapseNo relapse
      (n = 189)(n = 317)
      Adjusted DMF ARR (95% CI)
      Adjusted ARR estimate based on negative binomial regression model adjusted for gender, age, natalizumab duration group (≤ vs > median), and relapse during natalizumab treatment and washout (yes vs no).
      0.34 (0.25–0.46)0.24 (0.18–0.32)
      ARR, annualized relapse rate; DMF, delayed-release dimethyl fumarate.
      a Includes modified evaluable population for sensitivity analysis.
      b Adjusted ARR estimate based on negative binomial regression model adjusted for gender, age, natalizumab duration group (≤ vs > median), and relapse during natalizumab treatment and washout (yes vs no).
      Fig. 4
      Fig. 4Time to first relapse 1 year after DMF initiation stratified by (a) relapse during the first year of natalizumab treatment and (b) washout duration category (≤ 90 vs > 90 days) in patients without relapse during natalizumab treatment. In (a), estimated HR (0 vs ≥ 1 relapse) was based on a Cox proportional hazard model adjusted for age and number of relapses during washout. In (b), estimated HR for impact of washout duration (≤ 90 vs > 90 days) among patients without relapse during natalizumab treatment was based on a Cox proportional hazard model using the stabilized inverse probability of treatment weight estimator; the variability of HR was based on a robust (sandwich) standard error. DMF, delayed-release dimethyl fumarate; HR, hazard ratio.
      Similarly, patients without relapse during natalizumab treatment and washout (n = 369) had a lower risk of relapse after DMF initiation than those with relapse (n = 137; HR, 0.36 [95% CI, 0.24–0.54]; p < 0.0001).
      Baseline characteristics of each of the populations stratified by washout duration category (≤ 90 vs > 90 days) were balanced (Table 4). The adjusted ARR 1 year after DMF initiation was 0.25 (95% CI, 0.19–0.32) in patients with washout duration ≤ 90 days (n = 359) compared with 0.30 (95% CI, 0.21–0.45) in patients with washout duration > 90 days (n = 147; relapse rate ratio, 0.81 [95% CI, 0.51–1.28]; p = 0.36). In patients without relapse during the first year of natalizumab treatment (n = 387), the Kaplan-Meier estimate of risk of relapse 1 year after DMF initiation was 12.4% in patients with shorter washout (≤ 90 days) compared with 19.5% in patients with longer washout (> 90 days; HR, 0.59 [95% CI, 0.33–1.04]; p = 0.07; Fig. 4b).
      Table 4Summary of ARR 1 year after DMF initiation stratified by washout duration category using SIPTW estimator.
      CharacteristicWashout duration
      ≤ 90 days> 90 days
      (n = 359)(n = 147)
      Mean (SD) age at therapy change, years47.0 (10.7)47.0 (11.1)
      Female, n (%)249 (69)98 (67)
      Mean (SD) duration of natalizumab treatment, days1229.7 (678.2)1226.6 (687.1)
      Mean (SD) no. of relapses during natalizumab treatment and washout0.5 (1.1)0.5 (1.0)
      Adjusted ARR (95% CI)
      Adjusted ARR estimate based on negative binomial regression model using SIPTW estimator; variability of adjusted ARR based on robust (sandwich) standard error.
      0.25 (0.19–0.32)0.30 (0.21–0.45)
      ARR, annualized relapse rate; DMF, delayed-release dimethyl fumarate; SIPTW, stabilized inverse probability of treatment weight.
      a Adjusted ARR estimate based on negative binomial regression model using SIPTW estimator; variability of adjusted ARR based on robust (sandwich) standard error.
      Among patients without relapse during the first year of natalizumab treatment (n = 387), the adjusted ARR 1 year after DMF initiation was significantly lower in those with a washout duration ≤ 90 days versus a washout duration > 90 days (relapse rate ratio, 0.49 [95% CI, 0.26–0.90]; p = 0.0216; Fig. 5a). For patients with relapse during the first year of natalizumab treatment, the ARR 1 year after DMF initiation did not differ significantly between patients with a washout duration ≤ 90 versus > 90 days (ARR, 0.55 [95% CI, 0.40–0.77) vs ARR, 0.44 [95% CI, 0.26–0.73], respectively; p = 0.4415). Further stratification of washout into shorter time intervals did not reveal any further between-group differences.
      Fig. 5
      Fig. 5ARR 1 year after DMF initiation stratified by washout duration category. Washout duration categories (≤ 90 vs > 90 days) were balanced on age at therapy change, gender, duration of natalizumab treatment, and number of relapses during the first year of natalizumab treatment and washout using propensity score. Adjusted ARR estimate was based on a negative binomial regression model using stabilized inverse probability of treatment weight estimator; the variability of adjusted ARR was based on a robust (sandwich) standard error. ARR, annualized relapse rate; DMF, delayed-release dimethyl fumarate.
      Analyses were conducted to assess whether there was an association between clinical disease activity, as measured by ARR, during DMF treatment and (i) age at treatment initiation (< 55 [n = 372] vs ≥ 55 [n = 134] years) and (ii) steroid use during washout. At 1 year of DMF treatment, adjusted ARR appeared to be higher among patients < 55 years of age (ARR, 0.327 [95% CI, 0.252–0.424]) versus patients ≥ 55 years of age (ARR, 0.172 [95% CI, 0.092–0.322]); the rate ratio for relapse was 1.90 (95% CI, 0.91–3.99), but it did not reach statistical significance (p = 0.0894). Adjusted ARR at 1 year of DMF treatment was not associated with steroid use during washout (yes [n = 28] vs no [n = 478]; relapse rate ratio, 0.59 [95% CI, 0.25–1.38]; p = 0.2225).

      3.3 Safety data

      Of the 518 eligible patients who received ≥ 1 dose of DMF (safety population), 42 (8%) reported ≥ 1 AE leading to DMF discontinuation. The most commonly reported AEs were gastrointestinal disorders (n = 21 [4%]; Table 5), consistent with observations in clinical trials (
      • Fox R.J.
      • Miller D.H.
      • Phillips J.T.
      • Hutchinson M.
      • Havrdova E.
      • Kita M.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis.
      ,
      • Gold R.
      • Kappos L.
      • Arnold D.L.
      • Bar-Or A.
      • Giovannoni G.
      • Selmaj K.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis.
      ). The proportion of patients who experienced ≥ 1 AE leading to discontinuation was not different when stratified by washout duration (≤ 90 days, 8% [29/364] vs > 90 days, 8% [13/154]), whereas incidence of gastrointestinal disorders was higher in patients with washout duration > 90 days (6% [9/154] vs 3% [12/364] in those with washout duration ≤ 90 days). No SAEs leading to discontinuation were noted. One percent (7/506) of patients were hospitalized due to relapse in the year after DMF initiation. Less than 1% of patients discontinued due to decreases in lymphocyte counts (2/518) or infections (1/518; upper respiratory tract infection); no opportunistic infections were reported.
      Table 5AEs in ≥ 2 patients leading to DMF discontinuation.
      AE, n (%)Safety population
      (n = 518)
      SAE0
      AE42 (8)
       Gastrointestinal disorders21 (4)
        Abdominal discomfort/pain/pain upper/tenderness10 (2)
        Nausea6 (1)
        Diarrhea5 (1)
        Vomiting2 (< 1)
       Vascular disorders9 (2)
        Flushing9 (2)
       Nervous system disorders5 (1)
        Dysstasia2 (< 1)
       Psychiatric disorders5 (1)
        Depression3 (< 1)
       General disorders (fatigue)4 (< 1)
       Investigations4 (< 1)
        Lymphocyte count decreased2 (< 1)
        Weight decreased2 (< 1)
       Blood and lymphatic system disorders
      One case each of leukopenia and lymphadenopathy.
      2 (< 1)
       Renal and urinary disorders
      One case each of bladder pain, pollakiuria, and urinary incontinence.
      2 (< 1)
       Skin and subcutaneous tissue disorders
      One case each of pruritus and rash.
      2 (< 1)
      AE, adverse event; DMF, delayed-release dimethyl fumarate; SAE, serious adverse event.
      a One case each of leukopenia and lymphadenopathy.
      b One case each of bladder pain, pollakiuria, and urinary incontinence.
      c One case each of pruritus and rash.

      4. Discussion

      Natalizumab has proven to be effective in the control of relapsing forms of MS. However, the risk of developing PML has made strategies for transitioning patients to other DMTs an important topic, particularly in light of a potential resurgence in disease activity following its discontinuation. STRATEGY examined the utility of DMF as an alternative DMT for maintaining disease control after natalizumab discontinuation, as well as the impact of duration of post-natalizumab washout.
      In this study, patients transitioned from natalizumab to DMF had acceptable clinical disease control and AE profiles, and disease activity was less likely to return with a shorter (≤ 90 days) natalizumab washout. Patients without relapse activity during the first year of natalizumab treatment had a significantly lower ARR 1 year after DMF initiation than patients with ≥ 1 relapse during natalizumab treatment. The lower ARR (0.133) of patients who were relapse free while taking natalizumab and with washout duration ≤ 90 days, compared with patients with washout duration > 90 days (ARR, 0.272), suggests these may potentially be used to guide transition from natalizumab to DMF, and suggest that DMF may be a reasonable option for clinically stable patients who electively discontinue natalizumab, and that a shorter post-natalizumab washout duration may be associated with reduced relapse risk. Indeed, washout durations of 0–1 month between natalizumab and other DMTs, in previously reported studies with other DMTs, have been associated with reduced risk of post-natalizumab breakthrough disease activity (
      • Alvarez E.
      • Vollmer B.
      • Jace B.
      • Corboy J.
      • Vollmer T.
      • Sillou S.
      • et al.
      Effectiveness of switching to rituximab over fingolimod or dimethyl fumarate after natalizumab in preventing disease activity in multiple sclerosis.
      ,
      • Cohan S.
      • Edwards K.
      • Chen C.
      • Gervasi T.
      • O’Connor J.
      • Smoot K.
      • et al.
      Rebound disease activity reduction in relapsing multiple sclerosis patients transitioned from natalizumab to teriflunomide.
      ).
      Although the ARR for the entire cohort was higher while receiving DMF, compared with during natalizumab therapy, it was significantly lower than in the year before initiating natalizumab treatment. Our study did not address the relationship between lymphocyte counts and DMF relapse risk; however, no such association was noted in previous studies (
      • Fox R.
      • Chan A.
      • Gold R.
      • Phillips J.
      • Yang L.
      • Liu S.
      • et al.
      Absolute lymphocyte count in patients with relapsing-remitting multiple sclerosis treated with delayed-release dimethyl fumarate: considerations for patient management.
      ,
      • Longbrake E.E.
      • Naismith R.T.
      • Parks B.J.
      • Wu G.F.
      • Cross A.H.
      Dimethyl fumarate-associated lymphopenia: risk factors and clinical significance.
      ).
      The optimal washout duration when switching from natalizumab to DMF has not been established. However, in patients discontinuing natalizumab without beginning a new DMT, a resurgence in disease activity was observed ≤ 12–16 weeks following the last natalizumab infusion, some of which had aggressive rebound disease (
      • West T.W.
      • Cree B.A.
      Natalizumab dosage suspension: are we helping or hurting?.
      ). Although some patients have reportedly experienced mild relapses following discontinuance of natalizumab (
      • Jokubaitis V.G.
      • Li V.
      • Kalincik T.
      • Izquierdo G.
      • Hodgkinson S.
      • Alroughani R.
      • et al.
      Fingolimod after natalizumab and the risk of short-term relapse.
      ,
      • Kappos L.
      • Radue E.W.
      • Comi G.
      • Montalban X.
      • Butzkueven H.
      • Wiendl H.
      • et al.
      Switching from natalizumab to fingolimod: a randomized, placebo-controlled study in RRMS.
      ), moderate-to-severe relapses (
      • Fagius J.
      • Feresiadou A.
      • Larsson E.M.
      • Burman J.
      Discontinuation of disease modifying treatments in middle aged multiple sclerosis patients. First line drugs vs natalizumab.
      ,
      • Vidal-Jordana A.
      • Tintore M.
      • Tur C.
      • Perez-Miralles F.
      • Auger C.
      • Rio J.
      • et al.
      Significant clinical worsening after natalizumab withdrawal: predictive factors.
      ,
      • West T.W.
      • Cree B.A.
      Natalizumab dosage suspension: are we helping or hurting?.
      ) and a case of fatal fulminant relapse (
      • Larochelle C.
      • Metz I.
      • Lecuyer M.A.
      • Terouz S.
      • Roger M.
      • Arbour N.
      • et al.
      Immunological and pathological characterization of fatal rebound MS activity following natalizumab withdrawal.
      ) have been observed. The pharmacokinetic and pharmacodynamic profiles of both natalizumab and DMF () are consistent with our observations that patients starting DMF ≤ 12 weeks after discontinuing natalizumab had fewer relapses. Although shorter washout periods may also be of benefit for patients discontinuing drugs other than natalizumab, the result of this study, which focuses on natalizumab withdrawal relapse risk, emphasizes the need to balance concomitant exposure of natalizumab and DMF with the risks of increased MS disease activity associated with extended periods of no DMT use.
      The overall safety profile of DMF in STRATEGY was similar to that observed in the phase 3 trials and extension study (
      • Fox R.J.
      • Miller D.H.
      • Phillips J.T.
      • Hutchinson M.
      • Havrdova E.
      • Kita M.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis.
      ,
      • Gold R.
      • Giovannoni G.
      • Phillips J.T.
      • Fox R.J.
      • Zhang A.
      • Marantz J.L.
      Sustained effect of delayed-release dimethyl fumarate in newly diagnosed patients with relapsing-remitting multiple sclerosis: 6-year interim results from an extension of the DEFINE and CONFIRM studies.
      ,
      • Gold R.
      • Kappos L.
      • Arnold D.L.
      • Bar-Or A.
      • Giovannoni G.
      • Selmaj K.
      • et al.
      Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis.
      ), with no difference in AE and SAE incidence in subgroups stratified by washout duration. The safety and efficacy analyses are potentially limited by the absence of independent certified study monitors to assure consistency in reporting.
      One reason patients transition from natalizumab to DMF is an elevated risk of PML, which increases with prior JCV exposure, natalizumab treatment duration (especially ≥ 2 years), and prior treatment with immunosuppressants (,

      European Medicines Agency, 2016b. Tysabri 300 mg concentrate for solution for infusion. EU Summary of Product Characteristics. 〈http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000603/WC500044686.pdf〉 (Accessed 31 October 2016).

      ). In patients who have received natalizumab therapy for 2–3 years, the risk of PML increases from 0.2/1000 patients in patients with anti-JCV antibody index values ≤ 0.9, to 2.6/1000 patients in patients with anti-JCV antibody index values > 1.5 (

      European Medicines Agency, 2016a. EMA Confirms Recommendations to Minimise Risk of Brain Infection PML with Tysabri [news release]. 〈http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/2016/02/WC500202389.pdf〉 (Accessed 17 April 2017).

      ,
      • Koendgen H.
      • Chang I.
      • Sperling B.
      • Bloomgren G.
      • Haddock B.
      • Richman S.
      • et al.
      New algorithm to estimate risk of natalizumab-associated progressive multifocal leukoencephalopathy (PML) in anti-JCV antibody positive patients: analyses of clinical trial data to provide further temporal precision and inform clinical practice.
      ). In this study, 92% of patients were JCV positive and the mean duration of natalizumab treatment was 3.4 years; anti-JCV antibody index levels were not available for further risk stratification. No cases of PML were observed in patients transitioned to DMF. Although the duration of follow-up on average was only 12 months, and one might argue, that this is too short a period of observation, we are unaware of any patient in this study having developed PML, whether or not they had also been previously exposed to chemotherapeutic agents, ≥ 12 months after natalizumab discontinuation.
      Dividing the post-natalizumab washout into intervals < 90 days was not associated with lower ARR during DMF treatment compared with the full 90-day interval, possibly due to the small number of patients in these intervals (data not shown). Alternatively, washouts of ≤ 90 days might provide the maximum achievable benefit in reducing the risk of washout duration–related disease breakthrough.
      The benefits of this observational study design include the use of a real-world patient population, which reduces the impact of recruitment restriction inherent in most prospective clinical trials. STRATEGY captured natalizumab switch patients, some of whom might have been excluded from randomized controlled clinical trials, and are more representative of patients with MS encountered in routine clinical practice. This design also may increase opportunities for hypothesis generation focused on improving patient selection and future switch study protocols.
      Potential limitations of this study are those attributable to retrospective observational studies, most notably the lack of randomization and a controlled design to assess the effect of different washout schedules. Although observed variables were balanced, the impact of unmeasured confounders and hidden bias, including selection bias, cannot be ruled out. Unmeasured confounders include, but are not limited to, reasons for the different durations of washout, which may impact ARR. This becomes important when the unmeasured confounder is associated with the outcome. In this case, relapse may be associated with patient or drug selection. Treatment decisions, such as the initial switch to DMF and the ideal washout duration, may be influenced by patient demographics or disease characteristics and thus serve as a source of bias. Furthermore, patients were categorized as early versus delayed switchers after patients had transitioned from natalizumab to DMF. The time interval for transitioning may have been influenced by difficulties in obtaining insurance payer approval, the intensity of MS disease activity experienced by a patient before initiating natalizumab, patient apprehension about transitioning off natalizumab, and physician preconceptions regarding the relationship between duration of natalizumab washout and patient safety. Relapse data were the focus of this study, primarily because they were available in patient medical records; however, ARR in this study was based on clinical practice diagnoses, rather than independently confirmed, protocol-defined criteria such as those employed in prospective controlled clinical trials, and may even have included some patients who were misdiagnosed and had secondary progressive MS. However, this potential shortcoming is applicable to, and might encumber, previously published community-based studies. A further limitation of this study is its short duration of observation while on DMF, thus failing to detect potential later-onset AEs; however, as noted above, we are unaware of any patient in this study developing PML and, as natalizumab-associated PML has only been detected for up to 6 months after natalizumab discontinuation, the observed follow-up should be sufficient to capture events caused by any concomitant exposure of natalizumab and DMF. Disability progression and standardized magnetic resonance imaging data were not available in this study, given the community practice source of the data, but those measurements would have been potentially valuable information and may have provided additional valuable insights into disease activity; they should be included in future observational studies.

      5. Conclusions

      These results, reflecting community-based clinical practice, suggest that DMF may be an acceptable treatment option for patients transitioning from natalizumab, and that DMF initiation ≤ 90 days after natalizumab discontinuation is associated with a lower risk of relapse, particularly among patients whose relapse activity was well controlled during natalizumab treatment. Although prospective investigations that include more rigorous definitions of relapses and disease phenotypes are desirable, attempts to further clarify the duration of natalizumab washout to achieve optimal control of MS disease activity, particularly studies employing placebo controls, may be limited by ethical considerations for patient safety.

      Funding and acknowledgments

      We thank Allie McGinty, the clinical operations lead, and her counterpart, Michelle Leavy, from the contract research organization (Quintiles), who managed the STRATEGY study. This study was funded by Biogen (Cambridge, MA, USA). Biogen provided funding for medical writing support in the development of this paper; Karen Spach from Excel Scientific Solutions wrote the first draft of the manuscript based on input from authors, and Elizabeth Cassell from Excel Scientific Solutions copyedited and styled the manuscript per journal requirements. Biogen reviewed and provided feedback on the paper to the authors. The authors had full editorial control of the paper, and provided their final approval of all content.

      Conflict of interest

      SLC: research support from Biogen, Genzyme, Mallinckrodt, Novartis, Opexa, and Roche; advisory boards and steering committees for Biogen, Genzyme, and Novartis; speaker honoraria from Acorda, Biogen, Genzyme, Novartis, and Roche.
      HM: consulting/speaker fees and research support from Avanir, Bayer, Biogen, EMD Serono, Genzyme, Novartis, and Teva.
      JC: advisory/consulting/speaker activities for Acorda, Bayer, Biogen, EMD Serono, Genzyme, Mylan, Novartis, Questcor, and Teva; grant/research activities for Biogen, Genzyme, Novartis, Receptos, Roche, and Xenoport.
      CT: consulting fees and research support from Biogen, Genzyme, and Novartis; speaker bureau for Biogen.
      CL: advisory/consulting/speaker activities for Acorda, Bayer, Biogen, EMD Serono, Genzyme, Novartis, Pfizer, Questcor, and Teva Neuroscience; research support from Bayer, Biogen, Genzyme, GlaxoSmithKline, Novartis, Pfizer, Teva Neuroscience, and Vaccinex.
      KES: advisory/speaker activities for Acorda, Biogen, EMD Serono, Genzyme, Novartis, and Teva.
      CH, JPM, and MKM: employees of and hold stock/stock options in Biogen.
      MO, VM, and LM: former employees of and hold stock/stock options in Biogen.

      References

        • Alvarez E.
        • Vollmer B.
        • Jace B.
        • Corboy J.
        • Vollmer T.
        • Sillou S.
        • et al.
        Effectiveness of switching to rituximab over fingolimod or dimethyl fumarate after natalizumab in preventing disease activity in multiple sclerosis.
        Neurology. 2015; 84 (P3.288)
      1. Biogen Inc. TECFIDERA® (dimethyl fumarate) delayed-release capsules, for oral use: US prescribing Information. 2016a; http://www.tecfidera.com/pdfs/full-prescribing-info.pdf. Accessed May 5, 2016.

      2. Biogen Inc. TYSABRI (natalizumab) prescribing information. 2016b; https://www.tysabri.com/content/dam/commercial/multiple-sclerosis/tysabri/pat/en_us/pdfs/tysabri_prescribing_information.pdf. Accessed October 16, 2016.

        • Butzkueven H.
        • Kappos L.
        • Pellegrini F.
        • Trojano M.
        • Wiendl H.
        • Patel R.N.
        • et al.
        Efficacy and safety of natalizumab in multiple sclerosis: interim observational programme results.
        J. Neurol. Neurosurg. Psychiatry. 2014; 85: 1190-1197
        • Chen H.
        • Assmann J.C.
        • Krenz A.
        • Rahman M.
        • Grimm M.
        • Karsten C.M.
        • et al.
        Hydroxycarboxylic acid receptor 2 mediates dimethyl fumarate's protective effect in EAE.
        J. Clin. Investig. 2014; 124: 2188-2192
        • Cohan S.
        • Edwards K.
        • Chen C.
        • Gervasi T.
        • O’Connor J.
        • Smoot K.
        • et al.
        Rebound disease activity reduction in relapsing multiple sclerosis patients transitioned from natalizumab to teriflunomide.
        Mult. Scler. 2016; 22: 313
        • Compston A.
        • Coles A.
        Multiple sclerosis.
        Lancet. 2008; 372: 1502-1517
      3. European Medicines Agency, 2014. Tecfidera 120 mg gastro-resistant hard capsules; tecfidera 240 mg gastro-resistant hard capsules. Summary of Product Characteristics. 〈http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/002601/WC500162069.pdf〉 (Accessed 5 June 2016).

      4. European Medicines Agency, 2016a. EMA Confirms Recommendations to Minimise Risk of Brain Infection PML with Tysabri [news release]. 〈http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/2016/02/WC500202389.pdf〉 (Accessed 17 April 2017).

      5. European Medicines Agency, 2016b. Tysabri 300 mg concentrate for solution for infusion. EU Summary of Product Characteristics. 〈http://www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000603/WC500044686.pdf〉 (Accessed 31 October 2016).

        • Fagius J.
        • Feresiadou A.
        • Larsson E.M.
        • Burman J.
        Discontinuation of disease modifying treatments in middle aged multiple sclerosis patients. First line drugs vs natalizumab.
        Mult. Scler. Relat. Disord. 2017; 12: 82-87
        • Fox R.
        • Chan A.
        • Gold R.
        • Phillips J.
        • Yang L.
        • Liu S.
        • et al.
        Absolute lymphocyte count in patients with relapsing-remitting multiple sclerosis treated with delayed-release dimethyl fumarate: considerations for patient management.
        Neurology. 2017; 88 (P5.391)
        • Fox R.J.
        • Chan A.
        • Gold R.
        • Phillips J.T.
        • Selmaj K.
        • Chang I.
        • et al.
        Characterizing absolute lymphocyte count profiles in dimethyl fumarate-treated patients with MS: patient management considerations.
        Neurol. Clin. Pract. 2016; 6: 220-229
        • Fox R.J.
        • Cree B.A.
        • De Seze J.
        • Gold R.
        • Hartung H.P.
        • Jeffery D.
        • et al.
        MS disease activity in RESTORE: a randomized 24-week natalizumab treatment interruption study.
        Neurology. 2014; 82: 1491-1498
        • Fox R.J.
        • Kita M.
        • Cohan S.L.
        • Henson L.J.
        • Zambrano J.
        • Scannevin R.H.
        • et al.
        BG-12 (dimethyl fumarate): a review of mechanism of action, efficacy, and safety.
        Curr. Med. Res. Opin. 2014; 30: 251-262
        • Fox R.J.
        • Miller D.H.
        • Phillips J.T.
        • Hutchinson M.
        • Havrdova E.
        • Kita M.
        • et al.
        Placebo-controlled phase 3 study of oral BG-12 or glatiramer in multiple sclerosis.
        N. Engl. J. Med. 2012; 367: 1087-1097
        • Gold R.
        • Giovannoni G.
        • Phillips J.T.
        • Fox R.J.
        • Zhang A.
        • Marantz J.L.
        Sustained effect of delayed-release dimethyl fumarate in newly diagnosed patients with relapsing-remitting multiple sclerosis: 6-year interim results from an extension of the DEFINE and CONFIRM studies.
        Neurol. Ther. 2016; 5: 45-57
        • Gold R.
        • Kappos L.
        • Arnold D.L.
        • Bar-Or A.
        • Giovannoni G.
        • Selmaj K.
        • et al.
        Placebo-controlled phase 3 study of oral BG-12 for relapsing multiple sclerosis.
        N. Engl. J. Med. 2012; 367: 1098-1107
        • Jokubaitis V.G.
        • Li V.
        • Kalincik T.
        • Izquierdo G.
        • Hodgkinson S.
        • Alroughani R.
        • et al.
        Fingolimod after natalizumab and the risk of short-term relapse.
        Neurology. 2014; 82: 1204-1211
        • Kappos L.
        • Giovannoni G.
        • Gold R.
        • Phillips J.T.
        • Arnold D.L.
        • Hotermans C.
        • et al.
        Time course of clinical and neuroradiological effects of delayed-release dimethyl fumarate in multiple sclerosis.
        Eur. J. Neurol. 2015; 22: 664-671
        • Kappos L.
        • Radue E.W.
        • Comi G.
        • Montalban X.
        • Butzkueven H.
        • Wiendl H.
        • et al.
        Switching from natalizumab to fingolimod: a randomized, placebo-controlled study in RRMS.
        Neurology. 2015; 85: 29-39
        • Koendgen H.
        • Chang I.
        • Sperling B.
        • Bloomgren G.
        • Haddock B.
        • Richman S.
        • et al.
        New algorithm to estimate risk of natalizumab-associated progressive multifocal leukoencephalopathy (PML) in anti-JCV antibody positive patients: analyses of clinical trial data to provide further temporal precision and inform clinical practice.
        Mult. Scler. 2016; 22: 659
        • Larochelle C.
        • Metz I.
        • Lecuyer M.A.
        • Terouz S.
        • Roger M.
        • Arbour N.
        • et al.
        Immunological and pathological characterization of fatal rebound MS activity following natalizumab withdrawal.
        Mult. Scler. 2017; 23: 72-81
        • Longbrake E.E.
        • Naismith R.T.
        • Parks B.J.
        • Wu G.F.
        • Cross A.H.
        Dimethyl fumarate-associated lymphopenia: risk factors and clinical significance.
        Mult. Scler. J. Exp. Transl. Clin. 2015; https://doi.org/10.1177/2055217315596994
        • Noseworthy J.H.
        • Lucchinetti C.
        • Rodriguez M.
        • Weinshenker B.G.
        Multiple sclerosis.
        N. Engl. J. Med. 2000; 343: 938-952
      6. Otero-Romero, S., Amato, M., Chandraratna, D., Clanet, M., Comi, G., Derfuss, T., et al., 2016. ECTRIMS-EAN clinical practice guideline on pharmacological management of multiple sclerosis. Presented at the 32nd Congress of the European Committee for Treatment and Research in Multiple Sclerosis, London, UK, September 14–17.

        • Plavina T.
        • Subramanyam M.
        • Bloomgren G.
        • Richman S.
        • Pace A.
        • Lee S.
        • et al.
        Anti-JC virus antibody levels in serum or plasma further define risk of natalizumab-associated progressive multifocal leukoencephalopathy.
        Ann. Neurol. 2014; 76: 802-812
        • Polman C.H.
        • Reingold S.C.
        • Banwell B.
        • Clanet M.
        • Cohen J.A.
        • Filippi M.
        • et al.
        Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria.
        Ann. Neurol. 2011; 69: 292-302
        • Sturmer T.
        • Rothman K.J.
        • Avorn J.
        • Glynn R.J.
        Treatment effects in the presence of unmeasured confounding: dealing with observations in the tails of the propensity score distribution—a simulation study.
        Am. J. Epidemiol. 2010; 172: 843-854
        • Tullman M.J.
        Overview of the epidemiology, diagnosis, and disease progression associated with multiple sclerosis.
        Am. J. Manag. Care. 2013; 19: S15-S20
        • Vidal-Jordana A.
        • Tintore M.
        • Tur C.
        • Perez-Miralles F.
        • Auger C.
        • Rio J.
        • et al.
        Significant clinical worsening after natalizumab withdrawal: predictive factors.
        Mult. Scler. 2015; 21: 780-785
        • Wasserstein R.L.
        • Lazar N.
        The ASA's statement on p-values: context, process, and purpose.
        Am. Stat. 2016; 70: 129-133
        • West T.W.
        • Cree B.A.
        Natalizumab dosage suspension: are we helping or hurting?.
        Ann. Neurol. 2010; 68: 395-399