FormalPara Key Points

Several high-efficacy immune therapies can reduce the risk of disability progression in relapsing–remitting multiple sclerosis.

A standard definition of disability progression would facilitate comparative evaluation of therapies.

In relapsing multiple sclerosis, and potentially in certain progressive phenotypes, the best outcomes may be afforded by early treatment with the most effective immune therapies.

1 Introduction

Disability progression is a key clinical outcome in patients with multiple sclerosis (MS) that was originally assessed using the Disability Status Scale (DSS) [1]. The DSS was superseded by an expanded version of the scale [Expanded DSS (EDSS)], with which worsening disability can be scored in 0.5-point increments from 0 (normal neurological status) to 10 (death due to MS) [2]. Without treatment, patients with MS accrue moderate levels of disability (DSS score of 3) on average within 8 years of diagnosis, and need assistance walking (DSS score of 6) within 15 years of diagnosis [3]. Compounded by the unpredictable and debilitating nature of relapses experienced by many patients early in the disease course, even moderate levels of disability can be highly disruptive to normal living. A retrospective analysis of the Danish MS patient registry found that the median time from onset of MS to retirement (receipt of an early pension) was 10 years, compared with 24 years among matched control individuals [4]. A study of patients in nine European countries found an unemployment rate of 50 % among patients of working age with an EDSS score of 3.0, and also found a steady decline in utility score [calculated from the 5-dimension European Quality of Life (EuroQol) questionnaire] with increasing EDSS score [5].

The EDSS focuses mainly on motor function and ambulation, but captures cognitive decline poorly and has several other shortcomings [6]: ambiguity in the original rules for scoring affects reproducibility among raters, especially in the range 0–4.0 [7–10]; the scale is non-linear (i.e. the clinical importance of a 1-point increase varies depending on initial score) and a patient’s rate of progression through the scale also depends on baseline score [11]. This situation prompted development of the MS Functional Composite (MSFC), which assesses disability progression based on dexterity (Nine-Hole Peg Test), ambulation (Timed 25-Foot Walk Test) and cognitive function [Paced Auditory Serial Addition Test (PASAT)] [12, 13]. The MSFC has advantages over the EDSS, and is included as an endpoint in many studies, but it also has limitations: z scores (the summary score from the three components) are difficult to interpret; learning effects can skew PASAT on repeated use; and assessment of visual impairment is excluded [14].

Here, we review evidence for the potential of disease-modifying treatments (DMTs) to delay disability progression in patients with MS. We summarize disability data from all completed phase III trials of both approved and experimental therapies in patients with all forms of MS, grouping treatments by route of administration. Disability data from phase II trials are also reported, and a number of failed or inconclusive trials are listed, although these have been reviewed extensively elsewhere [15, 16]. The EDSS is the assessment most commonly used in MS trials (relatively few report changes in MSFC score), therefore we focus on measures related to EDSS score, such as confirmed disability progression (CDP), which is usually based on changes in the score sustained over 3 or 6 months; of these, 6-month CDP is the more robust indicator of permanent disability progression [17, 18]. Generally, CDP is defined as a 1.0-point increase if the EDSS score is less than 5.5 at baseline (sometimes a 1.5-point increase if EDSS score is 0), and as a 0.5-point increase if the baseline EDSS score is at least 5.5, but variations in these criteria are noted.

Reflecting the current treatment landscape, the majority of trials are in patients with relapsing–remitting MS (RRMS), rather than clinically isolated syndrome (CIS), primary progressive MS (PPMS) or secondary progressive MS (SPMS), and although used in the trials discussed here, some of this nomenclature has been recently superseded [19]. Trials are therefore grouped as pertaining to relapsing–remitting or to progressive disease.

2 Therapies in Patients with Relapsing–Remitting MS

Summary trial information and baseline characteristics are shown in Table 1 and disability outcomes from each trial are provided in Table 2.

Table 1 Baseline characteristics of patients with relapsing forms of MS in phase II and III trials that report disability outcomes
Full size table
Table 2 Phase II and III trials reporting disability outcomes in patients with relapsing forms of multiple sclerosis (MS)
Full size table

2.1 Approved Oral Therapies

2.1.1 Fingolimod

Fingolimod (Gilenya®, Novartis) was the first oral therapy used in the treatment of patients with RRMS, and has shown evidence of reducing disability progression in phase III trials [20–22]. It is approved in the USA for treating patients with relapsing forms of MS [23], and in the EU either as first-line therapy in patients with rapidly evolving severe RRMS or as second-line treatment in patients with RRMS and high disease activity despite treatment with at least one DMT [24]. Compared with placebo, fingolimod 0.5 mg (approved daily dose) reduced the respective risk of 3-month and 6-month CDP by 30 and 37 %, respectively, in FREEDOMS (p < 0.05, both) [20], but had a non-significant effect on CDP in FREEDOMS II [21]. The finding in FREEDOMS II was attributed to high variability in disability progression among patients with baseline EDSS scores of 0, and an exploratory analysis that excluded these patients found a significant reduction in the risk of 3-month CDP with fingolimod 0.5 mg versus placebo [hazard ratio (HR), 0.70; 95 % confidence interval (CI): 0.50, 0.98; p = 0.040]. Proportionately more patients had 3-month CDP in FREEDOMS II than in FREEDOMS, and the proportions with 6-month CDP were similar in both studies. A relatively small proportion of patients had 3-month CDP in the 1-year TRANSFORMS trial, and neither the 29 % reduction in risk of 3-month CDP with fingolimod 0.5 mg versus intramuscular interferon (IFN) beta-1a, nor the between-group difference in EDSS score reached significance [22, 24]. However, changes in MSFC scores were beneficial on fingolimod 0.5 mg compared with controls in all three trials [20–22].

2.1.2 Teriflunomide

Teriflunomide (Aubagio®, Genzyme) was the second oral drug approved in both the EU and the USA for use in adults with RRMS [25, 26]. Two pivotal placebo-controlled, phase III trials of teriflunomide have been conducted in patients with RRMS (TEMSO [27] and TOWER [28]) and another trial in CIS (TOPIC [29]). In the 2-year TEMSO trial, patients receiving teriflunomide 14 mg had a 30 % reduction in risk of 3-month CDP compared with those receiving placebo (p < 0.05) [27], and at the 14-mg dose in TOWER there was a 32 % risk reduction in time to sustained disability progression (a key secondary endpoint based on 3-month CDP; p < 0.05). Reductions in the risk of 6-month CDP failed to reach significance in either study [25] and no significant changes in MSFC were reported in TEMSO [30]. At 12 months in TOWER there was a small decrease from baseline EDSS score in the 14-mg group (p < 0.05) compared with the placebo group [28], and in TOPIC there was a significantly greater reduction in EDSS scores at 24 months with teriflunomide 14 mg compared with placebo (p < 0.05) [29]. Reductions in 3-month CDP, however, were not significant in the TOPIC study [29], which used the same CDP criteria as TEMSO. A fourth phase III trial, TENERE, which compared teriflunomide with subcutaneous IFN beta-1a, assessed a composite endpoint of time to disease relapse or treatment discontinuation, rather than evaluating disability as a separate endpoint [31]. Accrual of at least a 0.5-point increase in EDSS score, however, was one possible criterion for disease relapse. No differences were reported for the composite endpoint comparing teriflunomide with subcutaneous IFN beta-1a.

2.1.3 Dimethyl Fumarate

Dimethyl fumarate (DMF; BG-12; BG00012; Tecfidera®, Biogen Idec) is approved in the EU and USA for the treatment of adult patients with RRMS [32, 33]. Two 2-year, placebo-controlled phase III trials (CONFIRM [34] and DEFINE [35]) have been reported; CONFIRM also included a reference group randomized to subcutaneous glatiramer acetate (GA). In CONFIRM, reduction in risk of 3-month CDP was not significant compared with placebo, regardless of whether DMF was administered twice or three times a day [34]. In the DEFINE trial, a 38 % reduction in risk of 3-month CDP was noted for the approved regimen of 240 mg twice a day (p < 0.01) [35]; neither trial reported a reduction in the risk of 6-month CDP [32, 36]. Both trials used slightly more stringent criteria for CDP than were generally used elsewhere (a 1.0-point increase, or a 1.5-point increase if EDSS score is 0 at baseline) [34, 35]. Improvements in MSFC score were observed in both trials but were not statistically significant [36].

2.2 Oral Therapies Not Currently Approved

2.2.1 Laquinimod

Laquinimod (Nerventra®, Teva), an oral immunomodulator as yet unapproved for use in patients with MS, has been compared with placebo for the treatment of RRMS in the 2-year, phase III placebo-controlled ALLEGRO [37] and BRAVO [38] trials; BRAVO also included a comparator group receiving intramuscular IFN beta-1a. Significant reductions in the risk of 6-month CDP among patients receiving laquinimod 0.6 mg/day compared with placebo were reported in both ALLEGRO (49 %; p < 0.01) and BRAVO (39 %; p < 0.05), and the risk of 3-month CDP was also reduced in ALLEGRO (36 %, p < 0.05). In BRAVO there was a non-significant 31 % reduction in 3-month CDP; changes in MSFC scores from baseline were non-significant versus placebo in both studies [37, 38].

2.2.2 Cladribine

Cladribine (Movectro®, Merck Serono), an oral therapy approved for treatment of patients with hairy cell leukemia, was evaluated in patients with MS in the phase III CLARITY [39] and ORACLE MS [40] trials; however, clinical development was suspended in 2011 pursuant to safety concerns [41] and ORACLE MS terminated before disability outcomes were assessed [40]. The 2-year CLARITY trial used the same 3-month CDP criteria as did the phase III trials of DMF (a 1.0-point increase in EDSS score from baseline, or a 1.5-point increase if EDSS score is 0). At both doses of cladribine tested, the risk of 3-month CDP was approximately 50 % lower than with placebo (p < 0.05, both) [39].

2.2.3 Azathioprine

Azathioprine has been used to treat patients with MS for over 30 years, but its use was largely superseded by the advent of IFN beta-based therapies; however, a recent trial demonstrated its non-inferiority to IFN beta in patients with RRMS, including the finding that there was no significant between-group difference in the proportion of patients with 6-month CDP [42].

2.3 Approved Intravenous Therapies

2.3.1 Alemtuzumab

Alemtuzumab [Lemtrada®, Genzyme (Sanofi)] is an anti-CD52 monoclonal antibody (mAb) originally indicated for second-line therapy in patients with B-cell chronic lymphocytic leukemia, that was re-licensed in the EU (2013) and in the USA (2014) for intravenous treatment of relapsing forms of MS, with active disease defined by clinical or imaging features [43, 44]. At the approved dose of 12 mg/day for 5 days, alemtuzumab was compared with subcutaneous IFN beta-1a in the 2-year, phase III CARE-MS I [45] and CARE-MS II [46] trials (some patients were also randomized to alemtuzumab 24 mg/day in CARE-MS II, but this was discontinued following a protocol amendment designed to accelerate recruitment into the other treatment groups). Disability was a co-primary endpoint in both trials, defined as 6-month CDP with a 1.0-point increase in EDSS score from baseline, or a 1.5-point increase if EDSS score was 0. Compared with subcutaneous IFN beta-1a, a 42 % reduction in risk of CDP was seen among patients receiving alemtuzumab in CARE-MS II (p < 0.01) [46], but the 30 % reduction seen in CARE-MS I was not significant [45]. In CARE-MS II, the improvement in mean EDSS score (p < 0.001) was significant compared with that observed with subcutaneous IFN beta-1a [46], but there was no between-group difference in EDSS score in CARE-MS I [45]. Patients eligible for CARE-MS I were treatment-naïve, unlike those eligible for CARE-MS II, who had disease breakthrough on first-line IFN beta or GA. Also in CARE-MS I, a lower proportion of patients in the subcutaneous IFN beta-1a control group had 6-month CDP (11 %) [45] than in CARE-MS II (21 %) [46]. Changes in MSFC scores in both studies were significant when considered in isolation, but not when considered in the context of a pre-specified hierarchical analysis of secondary endpoints.

2.3.2 Natalizumab

Natalizumab (Tysabri®, Biogen Idec) is a humanized anti-α4 integrin mAb approved in the USA for the treatment of relapsing forms of MS and in the EU for the treatment of rapidly evolving severe RRMS and highly active RRMS among patients with disease breakthrough on IFN beta or GA [47, 48]. Administered intravenously at the approved dose of 300 mg every 4 weeks, natalizumab has been tested in two 2-year, phase III trials: AFFIRM [49], which was placebo-controlled; and SENTINEL [50], in which all patients continued intramuscular IFN beta-1a after randomization to natalizumab or placebo. The rate of 3-month CDP (defined as a 1.0-point increase in EDSS score from baseline, or a 1.5-point increase if EDSS score was 0) was the co-primary endpoint in both studies. Risk reductions of 42 % (AFFIRM; p < 0.001) and of 24 % (SENTINEL; p < 0.05) were demonstrated; and a 54 % reduction in the risk of 6-month CDP (p < 0.001) was reported in AFFIRM [48–50].

2.3.3 Mitoxantrone

Mitoxantrone (Novantrone®, EMD Serono) is an antineoplastic type II topoisomerase inhibitor used to treat patients with worsening or aggressive RRMS. Not currently approved throughout the EU, it is a therapy option in Germany and is approved for use in the USA [51]. Controlled trials of mitoxantrone involving patients with RRMS or worsening RRMS include: a 6-month study in which patients received methylprednisolone with or without mitoxantrone [52]; a 1-year placebo-controlled trial with a 1-year follow-up [53]; the 2-year placebo-controlled MIMS trial [54] ( please also see Sect. 3.1.1); and, more recently, a 3-year study by the French–Italian Mitoxantrone IFN beta-1b Trial Group [55], which randomized patients either to mitoxantrone and intravenous methylprednisolone monthly for 6 months followed by subcutaneous IFN beta-1b for 27 months, or to subcutaneous IFN beta-1b for 3 years combined with monthly methylprednisolone for the first 6 months. Compared with controls, mitoxantrone treatment reduced the proportion of patients with at least a 1.0-point increase in EDSS score from baseline in the 6-month study (p < 0.01), and reduced the risk of 3-month CDP (defined as a 1.0-point increase in EDSS score) by 30 % (p < 0.05) at 2 years [53], and by 65 % at 3 years [55].

2.4 Approved Injectable Therapies

2.4.1 Glatiramer Acetate

Approved in the USA and EU in patients with CIS and RRMS [56, 57], GA (Copaxone®, Teva) has been evaluated in six phase III trials [34, 58–62]. It was compared with placebo in two trials (GALA in patients with RRMS [58] and PreCISe in individuals with CIS [59]), with subcutaneous IFN beta-1a in REGARD [60], was an active comparator in two trials in patients with RRMS (the placebo-controlled CONFIRM trial of DMF [34] and the BEYOND trial of subcutaneous IFN beta-1b [61]), and was evaluated in combination with intramuscular IFN beta-1a in a placebo-controlled trial in patients with RRMS (CombiRx [62]). No large trials have reported an effect of GA on CDP relative to control treatment.

2.4.2 IFN beta

The three IFN beta-based therapies are the most long-standing approved DMTs for patients with RRMS and CIS, subcutaneous IFN beta-1b originally being licensed in the USA in 1993 [63] then in the EU in 1995 [64]. Intramuscular IFN beta-1a (Avonex®, Biogen Idec [65, 66]), subcutaneous IFN beta-1a (Rebif®, Merck Serono [67, 68]) and subcutaneous IFN beta-1b (Extavia®, Novartis [69, 70]; Betaferon®, Bayer [71, 72]) have been included in at least 18 phase III trials, either as the focus of investigation [61, 62, 73–82] or as reference compounds [22, 31, 38, 45, 46, 50, 55]. In the 2-year MSCRG study, proportionately fewer patients with RRMS had 6-month CDP on intramuscular IFN beta-1a than on placebo, and time to 6-month CDP was greater than with placebo (p = 0.02) [73]; disability endpoints were not reported in CHAMPS in patients with CIS [74]. In the PRISMS trial, there was a smaller increase in EDSS score over 2 years in both subcutaneous IFN beta-1a dose groups than in the placebo group [75], and subcutaneous IFN beta-1a was associated with non-significant reductions in the risk of 3-month and 6-month CDP compared with intramuscular IFN beta-1a in EVIDENCE [76, 77]; disability endpoints were not reported in IMPROVE [78] or among patients with CIS in REFLEX [79]. In the ADVANCE study [83] of subcutaneous pegylated IFN beta-1a (Plegridy®, Biogen Idec [84]), there was a 38 % lower risk of 6-month CDP (p < 0.05) among patients treated every 2 weeks for 2 years compared with those who received placebo in year 1 followed by treatment every 2 weeks in year 2 [85]. Finally, with subcutaneous IFN beta-1b there was a non-significant reduction in 3-month CDP compared with placebo in the IFNB study in patients with RRMS [80]. Generally, there were no between-group differences in disability measures at the 8-year follow-up [86] of the BENEFIT study [81], which examined the impact of delaying treatment initiation by 1 year in patients with CIS. The exception was that throughout the study period, the early-treatment group scored higher than the delayed-treatment group on PASAT (p < 0.05). There were no between-group differences in the proportions of patients with 3-month CDP in the BEYOND study of subcutaneous IFN beta-1b and GA [61]. These findings are somewhat contradicted by the 2-year INCOMIN study [82], in which the risk of 6-month CDP was 56 % lower with subcutaneous IFN beta-1b than with intramuscular IFN beta-1a (p < 0.01).

2.5 Injectable Therapies Not Currently Approved

2.5.1 Daclizumab High-Yield Process

The randomized DECIDE trial compared subcutaneous daclizumab high-yield process (DAC HYP; Biogen Idec and Abbvie) with intramuscular IFN beta-1a administered over 96–144 weeks, in patients with RRMS [87]. Provisional results ahead of publication indicated no significant between-group difference in 3-month CDP [88].

3 Therapies in Patients with Progressive Forms of MS

Summary trial information and baseline characteristics are shown in Table 3 and disability outcomes from each trial are provided in Table 4.

Table 3 Baseline characteristics of patients with progressive forms of multiple sclerosis (MS) in phase II and III trials
Full size table
Table 4 Disability outcomes of patients with progressive forms of multiple sclerosis (MS) in phase II and III trials
Full size table

3.1 Approved Intravenous Therapies

3.1.1 Mitoxantrone

Approved in Germany and the USA for the treatment of patients with SPMS and progressive relapsing MS [51], mitoxantrone 12 mg/m2 was associated with a small reduction in EDSS score from baseline in the 2-year placebo-controlled MIMS trial. Although interpretation is somewhat confounded by the fact that the trial was conducted in a mixed population of patients with progressive relapsing MS or SPMS, this reduction in EDSS score represented a significant treatment benefit compared with placebo (p = 0.0194). Time to 3-month and to 6-month CDP (increase in EDSS score of ≥1 point) was also greater in these patients than in those receiving placebo (p = 0.03, both) [54].

3.2 Intravenous Therapies Not Currently Approved

3.2.1 Cladribine

Intravenous cladribine has been evaluated in two phase III placebo-controlled trials in patients with progressive MS, but neither study reported values for 3-month or 6-month CDP [89, 90]. Evidence for an effect on disability progression was presented in the earlier, smaller trial (n = 48) [89], but no effect (based on changes in EDSS score) was seen in the subsequent larger trial of patients with PPMS (n = 48) or SPMS (n = 111) [90].

3.2.2 Rituximab

The OLYMPUS trial of intravenous rituximab in patients with PPMS found no effect of treatment on time to 3-month CDP. A significant delay in time to 3-month CDP was identified in planned subgroup analyses of patients with gadolinium-enhanced lesions on magnetic resonance imaging (MRI) at baseline (HR 0.41; p = 0.007) and in patients aged younger than 51 years (HR 0.52; p = 0.010), and the effect was augmented in the subgroup with both characteristics (HR 0.33; p = 0.009) [91].

3.2.3 Dirucotide (MBP8298)

The 2-year MAESTRO-01 trial examined the efficacy of dirucotide (MBP8298; Eli Lilly) in patients with SPMS and with human leukocyte antigen (HLA) haplotypes DR2+ or DR4+ [92]. No effect of treatment on disability progression was seen in MAESTRO-01, leading to the termination of phase III trials MAESTRO-02 and MAESTRO-03.

3.2.4 Intravenous Immunoglobulin

A randomized, 2-year, placebo-controlled trial of intravenous immunoglobulin (IVIG) in a mixed population of patients with PPMS (n = 34) or SPMS (n = 197) found that time to sustained disability progression was 12 weeks longer among patients receiving treatment than in those taking placebo (p = 0.0406) [93].

3.3 Approved Injectable Therapies

3.3.1 IFN Beta

Among the IFN beta-based therapies, only subcutaneous IFN beta-1b has received approval in the EU for treatment of progressive disease, specifically in relapsing forms of SPMS [70, 72]. Placebo-controlled trials of subcutaneous IFN beta-1b in Europe [94] and in North America [95] among patients with SPMS yielded different findings. There was a significant increase in time to 3-month CDP with treatment in the European trial compared with placebo (p = 0.0008) [94], but no effect on 6-month CDP was reported in the North American trial [95]. However, a post hoc meta-analysis of the two trial populations found an overall reduction in the risk of 6-month CDP (HR, 0.79; 95 % CI: 0.66, 0.93; p = 0.0076) [96]. Unfortunately, no such benefit was seen among patients with PPMS in a subsequent 2-year placebo controlled trial of subcutaneous IFN beta-1b [97]. Subcutaneous IFN beta-1a is a treatment option in Germany for relapsing forms of SPMS, although no effect on disability measures was reported in either of two 3-year placebo-controlled trials in patients with SPMS (SPECTRIMS [98]; the Nordic SPMS trial [99]); no phase III trials of subcutaneous IFN beta-1a in patients with PPMS have been reported.

3.4 Injectable Therapies Not Currently Approved

3.4.1 Intramuscular IFN Beta-1a

No improvement in time to 3-month CDP with intramuscular IFN beta-1a was seen among patients with PPMS [100], or among patients with SPMS in the 2-year IMPACT trial [101]. However, a significant improvement in MSFC scores (the primary endpoint) was seen at 2 years with intramuscular IFN beta-1a compared with placebo (p = 0.033).

3.4.2 Glatiramer Acetate

The placebo-controlled PROMiSe trial of GA in patients with PPMS was terminated when an interim analysis revealed no treatment effect on time to 3-month CDP, the primary outcome [102].

4 Discussion

The goal of DMT in patients with MS is to prevent the accrual of physical and cognitive deficits associated with disease worsening. Here, we have collected evidence of the effects of MS DMTs, with the aim of identifying those which offer the greatest benefit in mitigating disability progression. In addition to summarizing reported trial data (Tables 1, 2, 3, 4), we have listed for completeness any ongoing trials of MS DMTs and any trials that have been withdrawn or stopped (Table 5).

Table 5 Trials of multiple sclerosis (MS) therapies that are active but no longer recruiting, or that have been withdrawn, suspended or terminated. Trials are only listed if disability endpoints were specified
Full size table

There are difficulties when trying to compare the efficacy of different DMTs in relation to disability outcomes. Definitions of CDP vary across trials, both in the magnitude of the change in EDSS score that constitutes progression, and in the time over which this change must be sustained. A 1.5-point change from a baseline EDSS score of zero is a more robust measure of permanent disease worsening than a 1.0-point change, because recovery from a relapse is likely in the early stages of MS. Given that recovery is also likely if assessments are made over 3 months [18], the European Medicines Agency has recommended that assessments of disability should be made at least 6 months apart [17]; however, this recommendation was made subsequent to the time when the trials reviewed here were planned.

These challenges notwithstanding, trials of fingolimod, alemtuzumab, natalizumab laquinimod and subcutaneous IFN beta-1b in patients with RRMS have demonstrated significant reductions in the risk of 6-month CDP, effect sizes being in the range 40–60 % [20, 37, 38, 46, 49, 82]. A similar effect size is also implied by the proportions of patients free from 6-month CDP reported for intramuscular IFN beta-1a [73]. Patient demography at baseline was broadly similar in the phase III trials that demonstrated these treatment effects (Table 1), although the period since symptom onset was shortest in the alemtuzumab trials. Laquinimod is unusual among this group, in that its impact on disability was significant but its effect on relapse rates was modest [37, 38].

Evidence for mitigation of disease worsening in progressive forms of MS remains sparse, particularly in patients with PPMS. Treatment effects have been reported for mitoxantrone in a mixed population of patients with progressive disease [54], and for subcutaneous IFN beta-1b in individuals with SPMS [94, 96] but the general lack of reported values for risk reduction in trials in patients with progressive MS tends to confound comparisons. Details of the phase III INFORMS trial of fingolimod in PPMS are awaited, but it is known that the primary endpoint was not met [103].

Considerable advances have been made in the last 5 years in the treatment of relapsing MS, but the immunomodulatory nature of the majority of DMTs means that they may have little influence on the pathophysiological mechanisms associated with progressive disease [104]. Although inflammation is known to occur [105], it remains unclear whether it precedes or follows neurodegenerative tissue injury. However, there is some evidence that targeting inflammatory activity may be beneficial in certain patients with progressive disease. In subgroup analyses of the OLYMPUS trial of rituximab in PPMS [91], there was a significant effect of treatment on 3-month CDP among patients with gadolinium-enhanced MRI lesions at baseline. This treatment effect was driven by the fact that disability progression was faster among patients who received placebo in this subgroup (patients with disease now classified as ‘active and with progression’ [13]), than among those on placebo who were free from inflammatory activity at baseline (‘not active but with progression’ [13]). Age, but not disease duration, also affected outcomes in OLYMPUS: the effect of treatment on 3-month CDP was seen in patients younger than 51 years of age, but not in older patients [91]. These findings tend to support the use of the most effective immunomodulatory therapies early in the disease course among young patients whose MS is both active and progressive. The lack of effect among older patients may be symptomatic of age-related functional changes in immunity, or may be because recovery mechanisms are overwhelmed once a certain level of CNS damage has been accumulated by the combined effects of disease progression and aging. Either scenario adds another layer of complexity to the problem of identifying effective treatments in progressive MS.

Figure 1 illustrates schematically how different factors might combine in different MS disease types; analogous schemes have been proposed previously [106, 107]. The relapsing–remitting disease course is initially driven by inflammation, characterized by relapses, focal demyelinated lesions and axonal loss in the central nervous system (CNS), the effects of which are to some degree mitigated by neuronal plasticity and repair mechanisms [108, 109]. Coinciding with this, diffuse neurodegenerative damage in normal-appearing white and grey matter (‘underlying progression’) becomes increasingly apparent as inflammatory focal disease recedes [105, 109]. In addition to permanent focal CNS damage, this diffuse neurodegeneration contributes to the long-term brain atrophy characteristic of disease progression.

Fig. 1
figure 1

Schematic representation of the relationship between disability progression and the underlying pathological and rescue processes during a the typical relapsing–remitting disease course, which can ultimately transition into progressive disease, and b the purely progressive disease course. Disease worsening in patients with relapsing MS is a consequence of incomplete recovery from what is mostly focal inflammatory disease; disease progression is attributable to chronic diffuse neurodegenerative damage, a portion of which is caused by permanent focal damage. The colored boxes indicate typical eligibility criteria, as a range of EDSS scores, for recruitment into phase III trials. CIS clinically isolated syndrome, EDSS Expanded Disability Status Scale, MRI magnetic resonance imaging, MS multiple sclerosis, PPMS primary progressive MS, R–SPMS relapsing secondary progressive MS, RR relapsing–remitting, RRMS relapsing–remitting MS, SPMS secondary progressive MS

Full size image

Brain atrophy begins early in patients with MS [110], is one of the best predictors of long-term disability [111, 112] and correlates with worsening disability [113]. Analyses conducted post hoc in a population of patients with relapsing MS pooled from the three phase III trials of fingolimod [20–22] revealed a strengthening correlation between loss of brain volume and increase in EDSS score over 4 years [113]. During this period, the correlation between these parameters was also generally stronger in the group of patients with 6-month CDP than in the overall analysis population; consistent with these findings, the average increase in EDSS score over 4 years was greatest in the quartile of patients with the most brain volume loss. In the same set of analyses, significant correlations were also seen between brain volume at baseline and both EDSS score and T2 lesion volume, the former correlation supporting an association between loss of brain volume and accrual of disability before enrolment, and the latter correlation supporting an association between accumulated inflammatory focal CNS damage and brain atrophy [113]. These relationships between CNS damage and disability progression emphasise the need to initiate treatment to arrest focal inflammatory and diffuse neurodegenerative processes as early as possible. Although there are both procedural and methodological challenges in conducting routine monitoring of brain volume in patients with MS, these challenges should be tackled because treatment to reduce brain volume loss has been shown to correlate with beneficial effects on disability [114]. The effect of different MS DMTs on brain atrophy is reviewed elsewhere [115].

The evolution of the pathophysiology of MS during the disease course serves to emphasise the difficulty in comparing treatment effects, and confounds long-term extrapolation of effects seen during a 2- or 3-year clinical trial. Furthermore, changes in EDSS score tend to be modest over such a period, and long-term follow-up data, which might corroborate short-term effects on disability, are currently unavailable for many DMTs; the outcomes of several ongoing phase III trial extensions will be interesting. Beyond clinical trials, statistical modelling of real-world data offers a further means to assess the benefits of different treatments. Although still relatively short-term, a recent MS registry analysis of 3326 propensity-matched patients with a median follow-up time of 3.7 years found no difference among IFN beta therapies and GA in terms of 12-month CDP [116].

Despite great progress in the treatment of patients with relapsing MS, there remains an urgent need for drugs that modify the progressive disease course. From the patient’s perspective, physical and cognitive deterioration is probably the greatest concern. It is therefore important that clinical and paraclinical measures that accurately predict and track disease worsening and progression continue to be optimized and standardized to inform prescribing decisions.

5 Conclusions

In relapsing MS, increased numbers of approved DMTs give clinicians much greater scope than was possible only 5 years ago to select therapy options that their patients tolerate and which are likely to slow disability progression. Treatment options for patients with progressive MS remain scant, but immunomodulatory therapies may be effective in certain patient phenotypes. Challenges remain in standardizing how the efficacy of new and existing treatments is assessed, and assessment of subclinical disease as well as of relapses and disability progression should become part of routine disease monitoring. For the best outcomes in relapsing MS, evidence is accumulating that the most effective treatments should be used early in the disease course to reduce inflammatory disease, slow the accumulation of CNS damage and atrophy, and thereby delay the accrual of disability.