Multiple sclerosis (MS) is primarily an autoimmune disease of the central nervous system, but also encompasses prominent neurodegenerative aspects. A significant proportion of MS patients will develop neurological disability over time and up until recently treatment options have been limited. However, MS treatment is now at a stage of rapid progress, with several new drugs that have reached the market or will be launched in the near future. This provides new opportunities for individualized treatment, but also creates new challenges regarding monitoring of disease activity, long-term safety issues and efficacy, not least in patients with progressive disease.
Multiple sclerosis (MS) is an autoimmune disease that targets the central nervous system (CNS) leading to demyelination, axonal damage and subsequent loss of neurological functions . Second to trauma, it is the most common cause of neurological disability amongst young adults and is associated with a considerably reduced quality of life. Disease onset is usually in the third or fourth decade of life, and women are affected more than twice as often as men. Most individuals initially present with a relapsing–remitting disease course, where immune cells from the periphery periodically enter the CNS and cause focal inflammation that can be visualized with magnetic resonance imaging (MRI) and is clinically reflected by worsening of neurological functions. A large variety of symptoms can occur depending on anatomical localization, but sensory loss, visual disturbance, muscle weakness, ataxia and impaired balance are common. Resolution of inflammation is followed by a varying degree of recovery and a stable course between attacks. Relapses tend to be more frequent early in the disease course than at later stages. Traditionally MS was diagnosed only after a second relapse involving another area of the CNS, but according to current criteria, clinical signs of disease activity can be replaced by paraclinical MRI evidence of disease activity , in turn leading to earlier diagnosis. A first clinical episode indicative of MS is defined as a ‘clinically isolated syndrome suggestive of MS’ (CIS) . Because in a substantial proportion of patients CIS converts to MS, disease-modifying drugs (DMDs) should be considered already at this stage.
The notion that MS is initiated by aberrant activation of the adaptive immune system is supported by accumulating evidence from large genome scans and epidemiological data. The strongest genetic linkage is to the human leucocyte antigen (HLA) complex and many of the non-HLA risk genes are immune related, in particular genes that affect T cell function, and some also have a role in other autoimmune conditions . Environmental factors also play an instrumental role, and these include common childhood infections, in particular with Epstein–Barr virus and low sunlight exposure/vitamin D deficiency . More recent data also suggest interaction between genetic and environmental factors. For example, a strong interaction between the main MS HLA risk haplotype DRB15*01 and smoking was recently demonstrated . Taken together these data suggest that susceptibility to MS is determined by a combination of numerous genetic and environmental factors that may interact with each other. This is also a plausible explanation for the heterogeneity of MS in terms of clinical presentation, disease activity, disease severity and response to treatment.
With time, a majority of patients with relapsing–remitting MS (RRMS) will according to the natural disease course convert to secondary progressive MS (SPMS). SPMS is defined as unrelenting progression of neurological deficits not explained by relapses; however, patients may also experience occasional relapses and plateau phases . There are large interindividual differences in the time from disease onset to conversion to SPMS, but at the group level, at least half of all RRMS patients have entered the progressive phase 20 years after disease onset . Approximately 10% of patients have primary progressive MS (PPMS) where the condition is progressive from disease onset. PPMS patients share similarities with those with SPMS such as prominent spinal symptoms, onset after the fourth decade of life and a higher relative risk in men [1, 7]. Compared to RRMS, much less is known about relevant disease mechanisms and the genetic and/or environmental determinants of progressive disease. The absence of remission in late-stage RRMS is believed to reflect a failed recovery process due to repeated attacks of inflammatory demyelination and inadequate remyelination, leading to axonal damage, which has now been accepted as the major cause of irreversible disability in MS patients . It has been suggested that the transition from RRMS to SPMS occurs when the CNS can no longer compensate for neuronal loss . However, active disease-related processes such as direct T cell-mediated cytotoxicity, glutamate-induced excitotoxicity and noxious effects of chronic exposure to cytokines, proteolytic enzymes, oxidative products, free radicals and nitric oxide probably also play a role. The neurological decline in the absence of relapses and radiological signs of inflammation and the lack of response to traditional treatments during progressive MS suggest that disease mechanisms other than inflammation driven by an adaptive immune response might be operative in SPMS . A debated issue has been whether inflammation and neurodegeneration are independent, parallel or sequential processes. Accumulating evidence suggests that loss of axons in MS is initiated and maintained by complex inflammatory processes where the inherent susceptibility to neurodegeneration may vary amongst individuals . A schematic illustration of a typical MS disease course is shown in Fig. 1.
In the last two decades, MS has shifted from being a condition with very limited treatment possibilities to one of the most dynamic fields in terms of DMDs in clinical neurology, with a substantial number of novel drugs at various stages of clinical testing. The main focus here is to review the DMDs in late-stage clinical development and consider the potential impact of these agents on the treatment landscape in the coming years, as well as to identify some of the challenges that remain to be addressed.
Current first-line treatments
Recombinant interferon-beta (IFNβ) was first approved for RRMS in 1993 (IFNβ-1b, Betaseron/Betaferon), representing a first, long-awaited breakthrough in the management of MS . Subsequently, two IFNβ-1a preparations (Avonex and Rebif) were also approved. The pivotal studies for the different types of IFNβ consisted of placebo-controlled Phase III trials each with 300–560 patients showing efficacy in reducing the annualized relapse rate (ARR) by 30–40%, for references see . Secondary outcomes included different MRI measures (gadolinium-enhancing lesions, new or enlarging T2 lesions). In addition to proving the feasibility of therapeutic intervention in RRMS, these trials set an example for future MS drug trials using MRI-based outcome measures. The second drug to be approved for RRMS was glatiramer acetate (GA; also known as Co-polymer 1, Cop-1 or Copaxone), a mixture of random oligomers of amino acids enriched in myelin basic protein . Both IFNβ and GA are delivered by self-administered injections, varying in frequency from once daily to once weekly. Subsequent studies have shown that IFNβ and GA are also effective for reducing the risk of conversion to definite MS in CIS patients (for example see ). According to current guidelines, both IFNβ and GA are indicated as first-line drugs for the treatment of RRMS and for patients with CIS with a high risk of conversion to MS. Recent head-to-head studies of IFNβ and GA have demonstrated a similar efficacy for reducing relapse frequency in RRMS [14, 15]. However, a lowering of the risk of permanent disability in RRMS has not been unequivocally demonstrated, and these drugs are not effective for slowing the worsening of progressive forms of MS. It is also uncertain whether IFNβ and GA delay conversion to a progressive disease course. Open long-term follow-up studies of participants in randomized controlled trials generally indicate that both IFNβ and GA have a favourable long-term safety and tolerability profile. However, many of these studies are affected by a substantial dropout rate that makes them inadequate for predicting overall long-term clinical effects. Registry-based open-label follow-up studies of patients treated in clinical practice also show conflicting results. The risk of conversion to SPMS was reduced by half compared to historical controls in a Swedish study with long-term follow-up of IFNβ- and GA-treated patients . By contrast, the findings of a Canadian study failed to demonstrate long-term efficacy differences between treated and nontreated patients . It is also clear from clinical trials, performed more recently than the original RCTs in the 90s, that the current first-line drugs are only modestly effective. Thus, delayed initiation of IFNβ by 2 years in CIS patients was not associated with worse outcome at the 5-year follow-up; half of all patients had developed MS and a quarter had progressed despite initially being allocated to the active treatment arm . In fact, if both clinical and neuroradiological evidence of continued disease activity are taken into account, more than half of the patients who start treatment with first-line drugs have an inadequate response (also termed ‘breakthrough disease’) within the first 2 years . The most important factor is likely to be an insufficient effect on underlying disease mechanisms, with additional factors including lack of treatment adherence and development of neutralizing antibodies (NAbs). This is supported by the finding that patients with more pronounced pretreatment neuroradiological disease activity are more likely to display early breakthrough disease . There is no convincing evidence of a beneficial effect of switching to another first-line DMD, with the exception of patients who develop NAbs for whom GA is an alternative. Lack of treatment adherence (patient compliance) is of particular interest as it may not be adequately captured in clinical trials. In a Canadian retrospective cohort study of patients initiating a first-line DMD, less than half remained on treatment after 2 years . An important reason for poor treatment adherence is drug-related side effects; one of the most common for both GA and IFNβ is irritation at the injection site, which may evolve into local destruction of fat tissue (i.e. lipoatrophy). In addition, IFNβ produces influenza-like symptoms, and some patients treated with GA may experience a postinjection reaction characterized by chest tightness, heart palpitations and breathlessness. Another reason for poor treatment adherence is that it is difficult to perceive a beneficial treatment effect on an individual basis .
Current second-line treatments
Until recently there have been two main options for patients displaying breakthrough disease on first-line drugs: mitoxantrone and natalizumab. Mitoxantrone is a cytostatic agent used in cancer therapy that also displays some degree of immunosuppressive activity. The clinical evidence supporting the use of mitoxantrone in MS is limited. The largest randomized trial in MS included 194 patients with worsening of RRMS or SPMS and demonstrated a moderate effect on relapses and disease progression . Mitoxantrone is approved for worsening RRMS, SPMS and progressive-relapsing MS in the USA and is given by intravenous (IV) infusion once every 3 months for up to 2–3 years. The maximum accumulated lifetime dose is limited by cardiotoxicity and risk of congestive heart failure that may develop years after cessation of treatment. Another severe and potentially fatal side effect is leukaemia. Therefore, the role of mitoxantrone in current MS treatment paradigms is uncertain. The management of breakthrough disease was revolutionized with the approval in 2006 of natalizumab for highly active RRMS. Natalizumab is a humanized monoclonal antibody that binds to the α4-integrin of very late antigen-4 (VLA-4), a surface marker present on immune cells, and is given by monthly IV infusion. Antibody binding severely impedes the transmigration of leucocytes across the blood–brain barrier and thus targets a disease mechanism thought to be at the heart of MS pathogenesis (Fig. 2). VLA-4-dependent lymphocyte transmigration across brain endothelial cells in experimental autoimmune encephalomyelitis (EAE), the animal model of MS, was found to be nonredundant and laid the basis for the successful development of this drug . It has been suggested that treatment with natalizumab sequesters encephalitogenic T cells in the peripheral compartment . Expansion of the effector memory T cell pool, but not the activation state of T cells in blood, has been demonstrated , whilst T regulatory cell numbers and functions are left largely unperturbed . The main mode of action of natalizumab is therefore to impede migration of encephalitogenic cells to the CNS, which may have clinical implications, since treatment interruptions must be carefully managed as discussed below. The marketing approval for natalizumab was based on two pivotal Phase III trials: AFFIRM in which natalizumab monotherapy was compared to placebo and SENTINEL in which the antibody was combined with once-weekly IFNβ [28, 29]. Both studies were conducted in patients with RRMS, a large proportion of whom were treatment naïve. Trial data demonstrated a 67% decrease in ARR and a reduction in gadolinium-enhancing lesions by >90%. The proportion of patients displaying progression of disability was reduced by 40%; this finding represented an important step forwards in terms of clinically relevant outcome measures compared to previous DMDs. The beneficial treatment effect of natalizumab has also led to a better understanding of disease mechanisms in MS, since it has been unclear to what degree immunomodulatory intervention can protect and retain axonal integrity. Thus, in these Phase III trials, the rate of brain atrophy was reduced and in a small open-label study, the levels of neurofilament light, a marker of ongoing neurodegeneration, were normalized upon initiation of natalizumab . Long-term follow-up in the postmarketing setting demonstrated that natalizumab is generally well tolerated with beneficial effects on other treatment-related outcome measures . Unfortunately, follow-up studies also revealed that treatment with natalizumab increases the risk of progressive multifocal leukoencephalopathy (PML), a serious and potentially lethal opportunistic brain infection caused by the JC virus. For this reason, natalizumab is approved only as monotherapy, mainly in MS patients displaying breakthrough disease on first-line drugs. Further, other immunosuppressants have been associated with PML, but with variable risk profiles . Efalizumab (Raptiva) which targets CD11a was withdrawn due to several cases of PML in patients with psoriasis. It is now clear that the risk of natalizumab-associated PML can be assessed by serological testing for JC virus, with approximately half the population giving a positive test result . The risk of PML increases with >2 years of treatment and is also affected by prior use of immunosuppressants, such as mitoxantrone. Therefore, natalizumab provides a good example of how the risk–benefit profile must be determined at an individual level, with the drug representing an effective option with a good safety profile in JC virus serology-negative patients with high disease activity. As mentioned above, interruption of natalizumab treatment has to be managed carefully as disease activity will return upon cessation of drug administration. In a review of almost 2000 patients who had participated in clinical trials, both relapses and neuroradiological disease activity increased shortly after natalizumab interruption and peaked after between 4 and 7 months . Smaller studies have also found evidence of rebound phenomena, with reoccurring disease activity above that of pretreatment levels . This may be explained by the fact that patients on natalizumab in clinical practice generally have a more active disease than those patients recruited to trials. Currently there is no clear consensus on how to reduce the risk of a sometimes sudden return of disease activity, but when switching to another therapy, the washout phase should be kept as short as possible whilst also considering the risk of a ‘carry over’ subclinical PML.
Other immunosuppressants and stem cell transplantation
In addition to mitoxantrone and natalizumab, there is a history of off-label use of a number of oral or IV immunosuppressants in MS, such as azathioprine, mycophenolate mofetil and cyclophosphamide. However, available data on clinical efficacy and side effect profile of these agents in MS are too limited to include them in treatment guidelines. In addition to pharmacological treatments, haematopoietic stem cell transplantation (HSCT) has emerged as a treatment option. The availability of clinical trial data is limited, but promising results have been demonstrated in small case series . Due to the small, but non-negligible mortality risk of HSCT, this treatment is indicated only as a possible third-line option in carefully selected cases of severe and aggressive RRMS.
Fingolimod: the first oral treatment
Fingolimod (Gilenya) was the first oral treatment approved for RRMS and is also the first in class for drugs targeting sphingosine-1-phosphate receptors (S1PRs). This derivate of the fungus Isaria sinclairii and structural analogue of sphingosine acts as an antagonist on S1PRs. There are five known S1PRs, which are expressed on a number of different cell types. However, the main mechanism of action of fingolimod in MS is thought to be the sequestration of lymphocytes in lymph nodes due to blocking of S1PR1, which is present at variable levels on different types of immune cells (Fig. 2) . Pharmacologically, fingolimod causes a pronounced lymphopenia in peripheral blood. The most prominent reduction in lymphocyte numbers is seen for naïve and central memory T cells (TCM), with a smaller reduction in peripheral effector memory T and B cells, and natural killer (NK) cells and monocytes are largely unaffected [38, 39]. Fingolimod reduces Th17-expressing TCMs, which are found mainly in the blood (90% of TCMs) and are believed to be important for MS pathogenesis, without affecting T cell activation . Thus, similar to natalizumab, fingolimod mainly targets lymphocyte trafficking without affecting lymphocyte activation patterns. The half-life of fingolimod is approximately 1 week, and it takes at least 2 months for lymphocyte counts to recover upon stopping drug administration.
Fingolimod: efficacy and safety issues
Fingolimod was approved in the USA in 2010 (first line) and the EU in 2011 (second line) based on two pivotal Phase III studies: FREEDOMS to assess fingolimod 1.25 mg or 0.5 mg versus placebo over 2 years , and TRANSFORMS with the same doses of fingolimod versus weekly IFNβ over 1 year . These two studies showed a reduction in ARR of 53–60% compared to placebo and of 38–52% compared to IFNβ, respectively. Secondary outcome measures in the FREEDOMS study were also significant, and close to 90% of treated patients were free of gadolinium-enhancing lesions compared to 65% in the placebo group. The deaths of two study subjects in the 1.25-mg arm of the TRANSFORMS study, one due to herpes encephalitis and the other to a primary varicella zoster virus (VZV) infection, caused concerns over the risk of opportunistic infections. It is therefore recommended that VZV immunity status should be determined before initiating therapy. Another safety concern is the fact that S1PRs are present in the heart and that pharmacological blockade perturbs electroconduction. In fact, bradycardia and heart block are known side effects, and fingolimod should therefore be used with care in patients with known heart conditions, especially as several cases of sudden death in MS patients treated with fingolimod have been reported . A third safety concern is the risk of macular oedema, which has been detected in 0.5% (13 of 2564) of patients in the registration studies, although only in two patients receiving the approved 0.5-mg dose . It is therefore recommended that patients undergo ophthalmological examination after initiating therapy. The risk of macular oedema is increased in patients with diabetes, and fingolimod should only be used after carefully considering alternative therapies for these patients. Fingolimod has been shown to mediate neuroprotective effects and to enhance remyelination in preclinical models of MS, which has fuelled hopes that the drug may also be effective in progressive forms of MS. This notion needs confirmation in clinical studies, and currently fingolimod is being tested in the ongoing INFORMS study, which is the largest PPMS trial to date . In addition, several second-generation S1PR antagonists are in clinical trials for the treatment of MS. The two main advantages of these drugs are (i) a higher selectivity for the S1PR1 receptor subtype preferentially expressed on lymphocytes, thereby possibly reducing the risk of side effects and (ii) shorter half-lives, thus enabling faster washout if therapy needs to be interrupted.
Emerging oral treatments
In spite of the much more robust efficacy data for natalizumab and fingolimod in RRMS, injectable therapies still command the major share of the market for MS drugs due to their first-line status. However, this may change as new first-line oral drugs for RRMS are in early stages of marketing or in the regulatory phase of registration: teriflunomide (Aubagio), dimethyl fumarate (DMF; Tecfidera) and laquinimod (Nerventra). Teriflunomide and DMF have already been approved and launched in the USA, and they were recently also approved by the European Medicines Agency. However, it now seems unlikely that laquinimod will be approved based on existing clinical trial data in the EU, since it recently received a negative opinion by the CHMP expert committee. Meanwhile a new Phase III study to compare the currently recommended 0.6-mg dose with a higher 1.2-mg dose (CONCERTO) has been initiated as requested by the US Food and Drug Administration.
Emerging oral treatments; teriflunomide
Teriflunomide is a selective immunosuppressant with anti-inflammatory properties that exerts its therapeutic effect by blocking the mitochondrial enzyme dihydroorotate dehydrogenase (DHO-DH), a necessary enzyme for de novo pyrimidine synthesis in proliferating lymphocytes (Fig. 2). It is the active metabolite of leflunomide (Arava), a drug approved for the treatment of rheumatoid arthritis. Teriflunomide also appears to have additional actions independent of DHO-DH inhibition, including effects on cytokine production, expression of cell-surface molecules and cellular migration (Fig. 2). The most commonly reported adverse events are upper respiratory and urinary tract infections, diarrhoea, nausea, hair thinning and increases in liver enzymes. However, safety data pooled from several studies showed a low incidence of serious infections, which was not different between the active and placebo arms . In the TEMSO trial, once-daily doses of 7 and 14 mg teriflunomide were compared to placebo over 2 years in patients with RRMS . The results showed a 31% reduction in ARR with both doses compared to placebo, whilst an effect on disability progression was only observed with the higher dose (26% relative risk reduction). The primary MRI outcome measure, increase in total lesion volume, was reduced by 39% and 67% in the 7- and 14-mg dose groups, respectively, versus placebo . The proportion of patients free from gadolinium-enhancing lesions was 51% in the 7-mg group and 64% in the 14-mg group, compared with 39% in the placebo group. A similar design was used in the TOWER study, in which teriflunomide at 14 mg daily reduced ARR by 36% and disability progression by 31% compared with placebo . A smaller Phase III study, TENERE, comparing teriflunomide with three times weekly IFNβ1a, failed to meet its primary outcome measure, the proportion of patients experiencing treatment failure . The TOPIC study of teriflunomide in patients with CIS is ongoing, and the TERACLES trial of teriflunomide as add-on therapy in IFNβ-treated patients was recently interrupted prematurely.
Emerging oral treatments; dimethyl fumarate
DMF (also known as BG-12) is the methyl ester of fumaric acid, an intermediate in the citric acid cycle. Different combinations of fumaric acid esters, including DMF, have been used for topical and/or oral use in Germany for many years, mainly for the treatment of psoriasis. Initially, interest in fumarates for MS therapy was provoked by the findings of a small academic open-label study . The subsequent clinical development programme for RRMS included a Phase II trial with primary MRI outcome measures and DMF at three different doses, 120 mg once daily and 120 or 240 mg three times daily, compared to placebo. A significant reduction in gadolinium-enhancing lesions was only noted for the highest dose . Two different DMF regimens were examined in Phase III trials, 240 mg twice or three times daily. In the Phase III DEFINE study, ARR was reduced to 0.17 and 0.19 in the twice- and thrice-daily DMF groups, compared with 0.36 in the placebo group, corresponding to relative reductions of 53% and 48%, respectively, in the active arms . The Phase III CONFIRM trial also included an active, open-label comparator arm with GA, although the study was not powered to compare the two active treatments directly . The ARR values were 0.22, 0.20, 0.29 and 0.40 with twice-daily DMF, thrice-daily DMF, GA and placebo, respectively, corresponding to relative reductions compared to placebo of 44% for twice-daily DMF, 51% for thrice-daily DMF and 29% for GA. Several MRI measures were also significantly affected by treatment, including a reduction in new or enlarging T2 hyperintense lesions by 85% and 74% with DMF twice and three times daily, respectively . At 2 years, 93% and 86% of patients in the twice- and thrice-daily DMF groups, respectively, were free from gadolinium-enhancing lesions compared with 62% in the placebo group . The relative risk of disability progression compared to placebo was reduced by 34–38% in DEFINE, whereas the corresponding comparison in CONFIRM was not significant, in part due to a lower proportion of patients progressing in the placebo group. DMF treatment was not associated with increased risk of serious adverse events compared to placebo. However, at least initially many patients starting DMF experience flushing and gastrointestinal events, even though symptoms in most cases are mild to moderate. Low-dose salicylic acid may be used to reduce gastrointestinal discomfort.
The mode of action of DMF in MS has not been characterized in detail, but it is not a typical immunomodulatory or immunosuppressive drug, despite the fact that mild lymphopenia is common [53, 54]. Preclinical data suggest that the therapeutic effects of DMF are mainly through activation of the nuclear factor (erythroid-derived 2)-like 2 (Nrf2) pathway, which is a cellular response pathway induced by stress, affecting the expression of a number of different antioxidant genes (Fig. 2) . In the EAE animal model of MS, DMF reduces the infiltration of immune cells in the CNS and affects their activation response to myelin antigen, as well as increasing the preservation of myelin and axons . It is interesting that DMF also seems to mediate neuroprotective effects in noninflammatory preclinical models of neurodegeneration (for example see ). The relevance of this for human diseases is currently not known.
Emerging oral treatments; laquinimod
The third oral drug close to completing clinical development is laquinimod, a small synthetic molecule with once-daily administration. Laquinimod is a quinoline-3-carboxamide derivative of the parent compound linomide (Roquinimex), which was in late-stage clinical trials for RRMS and SPMS in the late 1990s when the clinical development programme was suspended because of several deaths due to cardiopulmonary adverse events . Due to concerns over similar safety issues with laquinimod, the clinical development programme was initiated using low doses of 0.1 and 0.3 mg daily in the first Phase II study . A relatively modest effect on MRI outcome measures was revealed, and therefore, the higher daily doses of 0.3 and 0.6 mg laquinimod were tested in the second Phase II trial . The 0.6-mg dose reduced the frequency of gadolinium-enhancing lesions in RRMS patients by 40% compared to placebo. Based on this result, two Phase III trials were initiated: the 2-year ALLEGRO study to compare 0.6 mg daily oral laquinimod with placebo and the BRAVO study in which the same oral dose of laquinimod was compared to both placebo and an active comparator of weekly IFNβ. In ALLEGRO, laquinimod showed a relatively modest effect on ARR (0.30 vs. 0.39 in the placebo group; relative risk reduction of 23%) and a 37% relative reduction in gadolinium-enhancing lesions . It is interesting, however, that the effect on confirmed disability was more robust, with a relative risk reduction of 29% (11% vs. 16%). Tolerability was good, and no serious safety concerns were raised, though a mild increase in liver enzymes was relatively common. By contrast, BRAVO, the results of which are still not reported in full, failed to demonstrate a significant effect on the primary study end-point ARR in the initial analysis (platform presentation at ECTRIMS 2012 in Lyon, France). However, brain atrophy was significantly reduced compared to placebo-treated patients. A post hoc correction was performed due to imbalances regarding MRI characteristics in the allocation of study subjects to the different arms, leading to slightly different and significant results: 21% relative reduction in ARR, a 33% reduction in confirmed disability and a highly significant reduction in brain atrophy (27%; P < 0.0001). The position of the regulatory authorities regarding the statistical analysis of the BRAVO trial is not yet clear, but in the light of the favourable safety and tolerability profile, a third Phase III study examining a higher 1.2-mg dose has been initiated. The mechanism by which laquinimod ameliorates autoimmune neuroinflammation has not been clarified in detail, but is likely to be more dependent on immunomodulatory than immunosuppressive effects. In a recent preclinical study, the administration of laquinimod to mice with EAE prevented further relapses and reduced CNS infiltration of immune cells . Further studies using both murine and human cells indicated that dendritic cell function is modulated, with reduced potential to induce CD4+ T cell proliferation and secretion of pro-inflammatory cytokines, possibly through inhibition of the nuclear factor kappa light-chain enhancer of activated B cells (NF-κB) pathway (Fig. 2) . Similarly, in another study, it has been shown that laquinimod protects against toxic demyelination by modulating the NF-κB pathway in astrocytes .
Emerging monolconal antibodies
Emerging monoclonal antibodies; CD20 antagonists
Several monoclonal antibodies are currently in clinical trials for the treatment of RRMS. One of the most interesting targets of immunomodulation is the CD20 antigen expressed on B cells (Fig. 2). Antibodies targeting this antigen include the mouse–human chimera rituximab, the humanized monoclonal antibody ocrelizumab and the fully human monoclonal antibody ofatumumab. Rituximab is approved for the treatment of B-cell lymphoma and rheumatoid arthritis, but has also been used off-label for neuroimmunological conditions such as myasthenia gravis, neuromyelitis optica and antibody-mediated encephalitides. In the Phase II HERMES study in patients with RRMS, two doses of 1000 mg rituximab led to a very robust drop in newly appearing gadolinium-enhancing lesions and a trend towards a reduction in relapse frequency over 48 weeks compared to placebo, despite a relatively small study population . In the subsequent OLYMPUS study, the effect of two twice-yearly infusions of 1000 mg rituximab or placebo on disease progression over 2 years in a much larger cohort of PPMS patients was investigated . Although the primary end-point of time to confirmed progression was not met, a subgroup analysis showed a significant effect in patients aged <51 years with ≥1 gadolinium-enhancing lesion at randomization. Further clinical development of rituximab for MS has been stopped by the sponsor; however, several smaller academic studies have been conducted. For example, in a small open-label academic study, the effect of a single 100-mg dose of rituximab twice yearly in RRMS patients led to a considerable reduction in neuroradiological and clinical disease activity compared to the run-in period .
Ocrelizumab has completed a Phase II study for use in the treatment of patients with RRMS, and a Phase III study programme is in progress. In the Phase II study, two doses of ocrelizumab, 600 and 2000 mg were compared to placebo or weekly IFNβ . The low and high doses of ocrelizumab led to reductions of 86% and 96%, respectively, in gadolinium-enhancing lesions compared to placebo and decreases in ARR of 80% and 73%, respectively. The drug was generally well tolerated, but one death occurred in the high-dose ocrelizumab treatment arm; a woman developed disseminated intravascular coagulopathy and multiple organ dysfunction syndrome after a bee stings to the face . In the ongoing Phase III programme, two studies (OPERA 1 and 2) are examining the effect of 600 mg ocrelizumab twice yearly compared to thrice-weekly IFNβ for RRMS and a third study ORATORIO is investigating the effect of 600 mg ocrelizumab twice yearly compared to placebo in PPMS.
Ofatumumab is the third CD20 antagonist to undergo testing for the treatment of MS. A dose-finding Phase II study with 100, 300 or 700 mg has been completed, but has not yet been published in full. In another Phase II study, subcutaneous delivery in RRMS patients will be investigated. Ofatumumab is also in Phase III studies for the treatment of rheumatoid arthritis.
The role of B cells in MS remains unclear, however, it has been suggested that the interaction between B cells and the Epstein–Barr virus drives inflammation in the intrathecal compartment, especially in advanced disease . Treatment with rituximab has been shown to deplete B cells and also to reduce inflammatory mediators and T cells in cerebrospinal fluid (CSF) . In addition, the efficacy of B cell-depleting therapies may depend on reduced antigen-presenting and T cell regulatory activities .
Emerging monoclonal antibodies; alemtuzumab
Alemtuzumab is a humanized monoclonal antibody directed against CD52, a surface antigen expressed on various subtypes of lymphocyte, and originally approved for the treatment of haematological malignancies (Fig. 2). The clinical development of alemtuzumab in MS has been tortuous, with the first study already published in 1999 . In 27 patients with relatively advanced disease, a single treatment course over 5 days resulted in cessation of inflammatory activity, but no clear effect on worsening of progressive disability and continued brain atrophy. In a subsequent Phase II study (CAMMS223), RRMS patients with high disease activity and <3 years disease duration were assigned randomly to a yearly treatment course with 12 or 24 mg alemtuzumab or thrice-weekly IFNβ . The results demonstrated a remarkable superiority of alemtuzumab compared to the standard treatment in reducing the risk of progressing disability over 3 years. However, a small number of patients developed immune thrombocytopenia, one of whom died. Also, a high rate of thyroid disorders and a case of Goodpasture's syndrome were noted in the alemtuzumab arms. Although further alemtuzumab treatment was suspended, the antibody showed superiority over IFNβ even at the 5-year follow-up . In two parallel Phase III studies, CARE-MS I and II, the effect of alemtuzumab compared to IFNβ was examined over 2 years, with relapse rate and time to 6-month sustained accumulation of disability as co-primary end-points [74, 75]. Treatment-naïve patients with a disease duration of <5 years were eligible for enrolment in CARE-MS I, whereas previously treated patients with <10 years' disease duration were also included in CARE-MS II. In CARE-MS I, alemtuzumab demonstrated a 55% reduction in relapses and a nonsignificant reduction in risk of sustained progression compared to IFNβ . A similar reduction in relapses of 49% and a 42% reduced risk of sustained disability were observed in CARE-MS II . Both studies confirmed a relatively high rate of thyroid disorders, including two cases of thyroid papillary carcinoma, a low rate of immune thrombocytopenia and an increased risk of herpes infections. Taken together, alemtuzumab demonstrates very good efficacy, but a high rate of adverse events requiring close monthly monitoring for at least 4 years after the last infusion. Alemtuzumab was recently approved by regulatory authorities in the EU. However, the US Food and Drug Administration declined approval based on concerns that benefits do not outweigh the risk of serious adverse events.
Emerging monoclonal antibodies; daclizumab
The final monoclonal antibody in late-stage clinical development for MS is daclizumab, which targets the CD25 antigen (α-chain of the interleukin-2 receptor) and has been approved for prevention of transplant rejections. A first small Phase II open-label study in 10 IFNβ-treated MS patients with signs of breakthrough disease demonstrated a strong reduction in new contrast-enhancing lesions . In the larger Phase II CHOICE study, patients who were being treated with IFNβ were assigned randomly to receive add-on subcutaneous daclizumab 2 mg kg−1 every 2 weeks or 1 mg kg−1 every 4 weeks, or placebo; the results showed a decrease in gadolinium-enhancing lesions in the high- and low-dose daclizumab treatment groups of 72% and 25%, respectively, compared to placebo . Interestingly, it was found that daclizumab led to a conspicuous expansion of CD56(bright) NK cells, which has been suggested to relate to the mechanism of action of the drug (Fig. 2) . Common adverse events were distributed equally across groups. The findings of a second Phase II study, SELECT, in which the effect of 150 or 300 mg daclizumab monthly as monotherapy was compared to placebo, were recently published . A 50–54% reduction in ARR was observed compared to placebo. The treatment was generally well tolerated, but cutaneous events were more common in the daclizumab group, including in one patient who first had a serious rash and later died from a psoas abscess detected at autopsy. The 2-year Phase III study DECIDE to compare 150 mg daclizumab monthly as monotherapy and weekly IFNβ is ongoing.
The drugs reviewed here are listed in Table 1. In addition, a summary of key clinical study results is shown in Table 2.
Table 1. List of MS drugs
Dose and administration
250 μg s.c. eod
250 μg s.c. eod
30 μg i.m. weekly
22 or 44 μg s.c. 3×/week
20 mg s.c. daily
0.5 mg p.o. daily
300 mg i.v. monthly
7 or 14 mg p.o. daily
240 mg p.o. twice daily
0.6 (1.2) mg p.o. once daily
Lemtrada (formerly Campath)
12 mg i.v. ×5/×3 given twice
N/A (formerly Zenapax)
150 mg s.c. monthly
600 mg i.v. twice yearly
Table 2. Summary of clinical study results for novel MS therapies
Study population/ treatment arms
Key clinical outcomes
Key MRI outcomes (risk ratio versus control unless otherwise stated)
ARR, annualized relapse rate; BID, twice daily; GA, glatiramer acetate; Gd, gadolinium-enhanced MRI; HR, Hazard ratio; IFN-b, interferon-beta; MRI, magnetic resonance imaging; n.a, not applicable; n.d, not determined; n.s, not significant; PPMS, primary progressive MS; RRMS, relapsing-remitting MS; SPMS, secondary progressive MS; TID, three times a day.
After post-hoc correction.
HR 0.33 for patients aged <51 with ≥1 Gd lesion at baseline.
24 mg group prematurely stopped, included only in safety analyses.
Clinical Trials.gov number
RRMS n = 581 2:1
12 mg vs. IFNβ 2 years
Disability progression HR n.s.
ARR 0.18 vs. 0.39
Patients with active lesions at w96 7% vs. 19%
Patients with new or enlarging T2 lesions 48% vs. 58%
Whilst new drugs clearly provide new opportunities for individualized treatment, they also create new challenges, not least which drug should be given to a certain patient and at what stage of the disease. In the current treatment paradigm, IFNβ and GA are first-line drugs and fingolimod and natalizumab are mainly second-line choices. Of the drugs now in late-stage clinical development, DMF and teriflunomide are likely to become first-line alternatives, whereas based on safety data, the monoclonal antibodies will mainly become second-line options. A common problem in communicating with patients in the MS clinic is explaining that treatments target inflammation rather than neurological symptoms. Many older patients with advanced disease are willing to accept higher risks than younger patients with more limited symptoms, even though inflammatory activity as reflected by relapse rates, MRI activity or CSF biomarker analyses generally decrease with age [80, 81]. This fact is underscored in the trials with alemtuzumab, arguably amongst the most effective drugs for dampening MS inflammation, where a striking effect in younger patients with active disease and short disease duration is contrasted with much more modest efficacy in older patients in the later stages of disease. An age-dependent loss of relative efficacy has also been noted for other drugs used to treat MS [82, 83]. From a mechanistic perspective, drugs targeting T and/or B cells or those that affect lymphocyte migration have proven efficacy in relapsing–remitting stages of MS. The effect in progressive disease is still uncertain, although there is some support for the efficacy of B cell depletion, at least in younger patients with signs of inflammation on MRI or in CSF biomarker analyses [65, 84]. Further data are awaited from the ongoing trials of fingolimod and natalizumab in progressive MS. However, there are inherent difficulties associated with studies in patients with progressive disease, as the condition often evolves slowly over time and the natural disease course may include plateau phases with more stable function. Thus, an important task is to refine and develop new disability outcome measures in MS, in particular for progressive disease . One approach that may prove successful is the use of biomarkers that better reflect disease processes in progressive disease. However, a more detailed understanding of the nature of these disease processes still needs to be achieved. It will also be necessary to validate potential biomarkers against clinically relevant outcome measures. In most instances, the trials reviewed here have been conducted in patients with RRMS, usually aged 18–55 years with mild-to-moderate disability, and sometimes also with a restriction on disease duration. Therefore, postmarketing follow-up studies that can provide additional information on efficacy and safety in the real-world setting are a high priority. Of the drugs reviewed here, both laquinimod and DMF may be of interest for progressive disease, at least on experimental grounds, because of their modes of action.
The choice of a certain drug depends on a risk–benefit analysis, on an individual basis, taking into account MS disease characteristics and concomitant diseases. Biomarkers also have a potential role in this context; one example already in use is JCV serology testing for assessing PML risk. In addition, it has been suggested that analyses of cytokine levels may reflect the risk of developing secondary autoimmunity after alemtuzumab treatment . Biomarkers such as the B cell chemokine CXCL13 may also be used to assess the degree of MS-related inflammation, which may in turn be of relevance for predicting the potential benefit of immune-directed therapy .
A common problem in the MS clinic is that many patients with active disease are women of child-bearing potential. Most of the drugs discussed here are given as chronic treatment, with potential adverse effects on the foetus, and information on safety issues is limited . Careful postmarketing surveys are needed in order to collect important information on side effects and pregnancy outcomes that can guide clinical decisions on treatment choice.
With several new drugs entering clinical practice, two different strategies have become available, induction and chronic treatment (Fig. 3). Alemtuzumab and HSCT, and to some extent CD20 antagonists, are examples of the first strategy, whereas most of the remaining represent the latter approach. In current treatment paradigms, patients are expected to start first-line treatment that can be escalated or changed to a second-line therapy if signs of breakthrough disease become evident. In particular for younger patients with highly active disease, induction therapy may be preferable to prevent further inflammation-induced nerve damage and the need for chronic treatment. Also, as suggested by preclinical evidence, drugs such as DMF or laquinimod may have the potential for follow-up treatment of residual innate inflammation. Clearly, long-term data on safety and outcome in the real-world setting need to be recorded.
Finally, with increasing efficacy, but also perhaps increasing risk of long-term adverse effects, there is a great need for studies of how and when MS therapy can be de-escalated (Fig. 3). From an ethical perspective, unnecessary long-term treatment of patients with costly drugs that may cause adverse effects should be avoided. However, this is an issue that is not a priority for the pharmaceutical industry and therefore requires academic initiatives and broader societal support. This also underscores the importance of postmarketing follow-up studies as well as biomarker or MRI measures that can provide guidance for immunotherapy de-escalation.
The novel treatments discussed above provide hope of better options for therapeutic management of MS; however, many challenges remain in order to use these agents to maximize benefit and minimize risks for patients. One of the primary challenges is to achieve a better understanding of the mechanisms of neurodegeneration in both early- and late-stage MS as well as to identify relevant imaging techniques and biomarkers that reflect these processes, thus laying the basis for studies of the therapeutic effects of existing and future drugs. In addition, high-quality postmarketing studies will be required to provide information on long-term safety and efficacy measures in the real-world setting. Paradoxically, the progress of the last decade creates the need for increasing efforts in refining MS treatment.
Relapsing–remitting MS (RRMS) – the most common presentation of MS, with a disease course characterized by recurrent episodes of new clinical neurological symptoms followed by a varying degree of recovery and a stable course between attacks.
Secondary progressive MS (SPMS) – with time the majority of RRMS patients enter a disease phase characterized by unrelenting progression of neurological deficits with or without occasional relapses and plateau phases.
Primary progressive MS (PPMS) – a minority of MS patients (approx 10%) in whom the disease course is already progressive from the start.
Disease-modifying drugs (DMDs) – drugs used to reduce neuroradiological disease activity, clinical relapses and the risk of developing permanent neurological disability.
Breakthrough disease – continued occurrence of clinical relapses or neuroradiological disease activity despite DMD treatment.
Annualized relapse rate (ARR) – the most common clinical outcome measure in RRMS trials.
Confirmed disability or disability progression – an increase in neurological symptoms that is sustained over 12 or 24 weeks and is thought to better reflect long-term prognosis.
The author acknowledges Drs Mohsen Khademi and Anders Svenningsson for help with the schematic illustration and critical comments on the text, respectively.
Conflict of interest statement
The author has participated in sponsored clinical trials with teriflunomide, dimethyl fumarate, natalizumab, ocrelizumab, ofatumumab, fingolimod, daclizumab, laquinimod and others. Karolinska University Hospital on behalf of the author has received unrestricted academic research grants from Biogen Idec and compensation for lectures and/or participation in advisory boards from Biogen Idec, Merck Serono, Novartis, Sanofi-Aaventis and Teva, which have been exclusively used exclusively for the support of research activities.