Department of Neurology
In the early 1990s, the first disease-modifying treatment (DMT) for multiple sclerosis (MS), interferon β-1b (Betaseron/Betaferon), was approved by the FDA for the treatment of relapsing MS. Within a few years this was followed by the launch of two interferon β-1a agents (Avonex, Rebif) and of glatiramer acetate (Copaxone). These agents target the early inflammatory component of the disease and do not influence the degenerative component. One of the aims of the new diagnostic criteria published in 2001, the so-called McDonald criteria, was to meet the need for early diagnosis while avoiding false-positive diagnosis.(1,2)
The availability of DMTs for MS changed the field within a few years, both for those affected with the disease and their families and for the neurologists trying to modify the course of the disease. Many countries have developed guidelines for the use of these agents. However, it should be noted that, despite efforts made, there are no firm definitions of responders versus nonresponders or of treatment failure. It is also not known whether the reported benefits relate to a partial response for many or a significant response for a minority. Therefore, the decision to start, change or stop treatment remains a challenge for the patient and the neurologist.
There is strong, albeit indirect, evidence that MS is an autoimmune disease.(3) Histological features are characterised by inflammation, demyelination and axonal degeneration. Loss of the insulating myelin sheath decreases the efficiency of action potential conduction, leading to reduced conduction velocity or conduction block. Restoration of function can occur in two ways: axonal adaptation through redistribution of sodium channels, and remyelination by surviving oligodendrocytes and maturing oligodendrocyte precursors. In severe cases, demyelination can lead to transection of the axon with resulting neuronal death and permanent loss of function.
Disease progression results from the balance between local production of proinflammatory and neurotoxic factors on one hand (leading to demyelination and neurodegeneration) and of anti-inflammatory and neuroprotective factors on the other (leading to myelin repair and neuroprotection).
MS affects young adults and women twice as often as men. The mean age at onset is approximately 30 years, but the disease occurs in adolescents and even, although rarely, in children. In most patients the disease has a relapsing course at onset, with episodes of new symptoms (relapses) followed by complete or partial recovery (remissions); this is known as relapsing–remitting MS (RRMS). The frequency of relapses varies greatly over time in an individual patient and also among patients. One relapse per year is often mentioned as average, but some patients may experience more frequent relapses, whereas in others there might be intervals of several years between clinical episodes. The relapse rate will usually decrease over time and, in a large proportion of patients, the disease will change to a progressive course without acute relapses (secondary progressive MS [SPMS]). Transition from relapsing to progressive course usually indicates worsening prognosis. A small proportion of all patients with MS will have a progressive course from onset (primary progressive MS [PPMS]).
The prognosis of MS is extremely variable. At one extreme, there are individuals who will never experience clinical symptoms in their lifetime but will be found to have had typical MS brain lesions at autopsy. At the other extreme are those with an aggressive course, with many relapses, poor recovery and early transition to progressive course and total dependency within a few years. In some natural history cohorts, approximately half of the individuals had progressive MS within 10–15 years.(4) But several other cohorts have reported that between one-third and one-fourth of the patients were still able to walk without support after 25–30 years.(5–7) Two recent publications have shown that acute onset of optic neuritis (ON) in previously healthy individuals was followed by new relapses and clinically definite MS (CDMS) in only approximately 40% within 10–15 years.(8,9) Even in individuals with one or more MRI lesions or with inflammatory cerebrospinal fluid (CSF) changes (elevated mononuclear cells, increased IgG index or oligoclonal IgG) the risk was only around 50%.
Currently in Europe four agents (Betaferon, Avonex, Rebif, Copaxone) have been licensed by the EMEA for the treatment of RRMS. Neither of these nor any other agent is indicated for the treatment of SPMS or PPMS. Trials are presently underway to test whether new agents or combinations of agents will be more effective without compromising patient safety.
Interferons belong to a group of endogenous glycoproteins with antiviral, antiproliferative, anti-inflammatory and immunomodulating effects. In humans, these cytokines are produced by the immune system as part of the normal immune response. Interferon β (IFN β) is used in the treatment of MS. There are two types of IFN β, 1a and b, both produced by recombinant DNA technology. IFN β-1a is produced in mammalian cells and, therefore, the amino acid sequence (Avonex, Rebif) is identical to the natural human product. IFN β-1b (Betaferon) is produced in bacterial cells (E coli), resulting in a preparation that is lacking methionine at position 1, has a serine residue substituted for cysteine at position 17, and has no glycosylation of the asparagine residue at position 80.
IFN β binds to specific surface receptors on human cells, which initiates a complex cascade of intracellular events, resulting in several IFN-induced gene products and markers, including MHC class I, Mx protein, 2’5′-oligoadenyl synthetase, beta-2-microglobulin and neopterin.
All three IFN β preparations have to be injected, but doses and intervals vary. Betaferon is given subcutaneously every other day, Avonex intramuscularly once weekly, and Rebif subcutaneously three times weekly. Higher dose and more frequent administration result in improved efficacy, at least in the short term, as shown both in vitro and in direct comparative studies.(10–12)
All three agents can claim efficacy on activity measures (clinical relapses and active lesions on MRI), whereas benefits on progression measures (disability and MRI lesion area) have been less consistent.(13–18) Compared with placebo, relapse rate was reduced by 34% (Betaferon, p=0.0001), 32% (Rebif, p<0.0002) and 18% (Avonex, p=0.04). Treatment with Betaferon and Rebif reduced the number of active lesions by 90% (p<0.001). Avonex and Rebif also slowed progression of disability.
Comparisons across studies cannot be made readily because of differences in study patients, study design and outcome measures. In addition, the pivotal study on Avonex(15,16) raises some concerns, because only 57% of enrolled patients completed the preplanned two-year study and those not completing the study had considerably worse outcomes than those who did. Therefore, data from the two-year subset must be viewed cautiously.(10)
Evidence of efficacy beyond two years comes from extensions of the pivotal studies. The original IFN β-1b (Betaferon)(13) study was extended as a double-blind, placebo-controlled study.(19) Each year, the relapse rate continued to be reduced in patients receiving Betaferon compared with placebo (relative reduction 24–33%), but due to a decreasing number of patients remaining in the study, these differences were not statistically significant. However, the reduction of accumulation of MRI lesions in the active treatment arm was significant.
The original study with subcutaneous IFN β-1a (Rebif)(17) was extended for two years as a double-blind controlled study.(20) Patients randomised to placebo during years 1–2 were re-randomised to active treatment for years 3–4, while those originally randomised to active treatment continued without change. All patients remained blind as to dose. Patients on active treatment during all four years had 0.72 relapses per year compared with 1.02 for those who switched from placebo to IFN β-1a (relative reduction 32%, p<0.001) with no apparent loss of treatment effect over time. The proportion of patients who did not progress during the four years was also greater among those on active treatment for all four years compared with those who switched from placebo to active treatment.
Long-term data beyond five years come from a follow-up approximately 7.5 years after enrolment into the study with subcutaneous IFN β-1a (Rebif).(21) Only 68% of patients returned, data were partly retrospective, and the examining neurologist was not blind as to treatment. Therefore, results have to be viewed with caution. Despite these limitations, it appears that the proportion of patients who had developed SPMS (around 20%) was less than expected from natural history cohorts (35–50%).(4,22)
Recombinant IFN β products may induce formation of antibodies, and it has become clear that these antibodies neutralise some or all of the treatment benefit. Neutralising antibodies (NAbs) appear in a proportion of patients, usually within 6–24 months.(23) Consensus on how to measure NAbs is only slowly developing and results vary across studies and between different laboratories. Nevertheless, there are considerable differences between the preparations. Treatment with Avonex induces NAb formation in only a small proportion of patients, whereas treatment with Betaferon and Rebif induces it in much higher proportions. Over time, despite continued treatment, a small proportion of patients (usually those with low titres) revert to antibody-negative status. It is not possible to predict who will develop NAbs. Therefore, the treating neurologist is faced with the dilemma that the two preparations (Betaferon and Rebif) that, in direct comparative studies with Avonex, have shown superior efficacy up to two years induce NAbs in a much greater proportion of patients, which will reduce or abrogate the treatment benefit.
The main adverse effects are local injection site reactions, particularly with subcutaneous preparations, and flu-like symptoms. Injection site reactions (flare, subcutaneous infiltration and, rarely, abscess formation and necrosis) are more common with subcutaneous preparations. Blood count, liver and thyroid function should be monitored regularly. IFN β should not be used during pregnancy or nursing and in patients with uncontrolled epilepsy or depression.
Glatiramer acetate (GA) is a standardised mixture of polypeptides containing four amino acids (L-glutamic acid, L-tyrosine, L-alanine and L-lysine) designed to mimic human myelin basic protein. The beneficial effect in MS is thought to depend on a redirection of the autoimmune process towards an anti-inflammatory pathway by generation in the periphery of GA-reactive T-cell clones with a Th2 phenotype that migrate into the central nervous system (CNS). In the CNS, T-cells are reactivated by myelin-related antigens, anti-inflammatory cytokines are released, and inflammation and demyelination are suppressed. There is some evidence that GA-reactive T-cells may release brain-derived neurotrophic factor (BDNF), a factor promoting neuronal repair.
Glatiramer acetate (Copaxone) has to be injected subcutaneously once daily. It can claim efficacy on activity measures (clinical relapses and enhancing lesions on MRI),(24,25) but benefits on progression measures (disability and MRI lesion burden) have been less convincing. The number of relapses in treated patients compared with those receiving placebo was reduced but not statistically significant (p=0.055). However, using multivariate statistics, the relative reduction in relapse rate was 29% (p<0.007). Treatment reduced the total number of enhancing lesions by 29% (p=0.003), but a large number of lesions were still seen in treated patients. Treatment effect became apparent only after approximately six months, which is assumed to be due to the mode of action of this agent.
The majority of patients experience pain at the injection site, but local irritation is uncommon. Up to 15% of patients may experience an unpredictable, systemic, self-limiting reaction with facial flushing, chest tightness, dyspnoea and palpitations. Laboratory abnormalities do not occur. Treated patients may develop antibodies to glatiramer acetate, but experimental and clinical evidence does not suggest that they reduce treatment benefits.(26) Glatiramer acetate should not be used during pregnancy.
This drug was licensed by the FDA for reducing disability or frequency of relapses in patients with progressive or worsening relapsing MS despite limited evidence of efficacy.(27,28) It is administered via intravenous infusion. Although not licensed in Europe, it is used in some centres in patients with clinical or MRI evidence of active disease despite treatment with interferons or glatiramer acetate. Serious side-effects have been reported, including dose-dependent cardiomyopathy, which may appear several months after the last dose, as well as acute leukaemia. Two recent reports estimate the risk of leukaemia at 0.1–0.2%. (29,30)
These drugs are not FDA or EMEA approved for the treatment of MS. Nevertheless, treatment with intravenous methylprednisolone 1g daily for three days hastens recovery(31) and has become standard for acute relapses with neurological dysfunction. Treatment should be started as early as possible to protect axons from irreversible injury by inflammatory mediators.(32) Until recently it was thought that corticosteroids had no effect on the natural course of MS. However, it has been reported that methyl-prednisolone given on a regular schedule for five years to patients with relapsing MS resulted in clinical and MRI improvements.(33) The evidence for a treatment benefit in progressive MS is less certain.(34) Our own experience includes a number of patients with PPMS who have been stable over a 10-year period while receiving monthly or bimonthly methyl-prednisolone (unpublished results).
Clinically isolated syndromes (CIS)
For some years, as available DMTs are effective in the inflammatory phase of MS, focus has been on “clinically isolated syndromes” (ie, symptoms that, in an individual, could mark the onset of MS). Results of two studies have been published,(35,36) and results of a third study have been reported in abstract form.(37)
The CHAMPS (Controlled High Risk Avonex Multiple Sclerosis Study) trial enrolled patients within four weeks of onset of a clinically isolated symptom if they had two or more asymptomatic MRI lesions, and randomised patients to intramuscular IFN β-1a (Avonex) once weekly or placebo.(35) The ETOMS (Early Treatment Of MS) study recruited patients within three months of an initial clinical episode if they had four or more MRI lesions, and randomised patients to subcutaneous IFN β-1a (Rebif) once weekly or placebo.(36) In both studies a second clinical episode establishing the diagnosis of CDMS was delayed in patients on active treatment and taken as evidence of clinical efficacy (p=0.002 and p=0.034, respectively).
Although there is no reason to believe that these agents would be less effective at the time of the first attack than after several more attacks, the studies raise a number of concerns. The primary outcome (ie, a new clinical episode) was the patient’s report of new symptoms suggesting a relapse. It did not require MRI examination at the time of clinical conversion to MS. The time to conversion to CDMS was very short (18% within one month and 26% within four months in the placebo group of CHAMPS), especially compared with recent publications on optic neuritis demonstrating that only around 50% of patients with at least one MRI lesion or inflammatory CSF changes convert to CDMS within 10–15 years.(8,9)
New agents, ongoing trials and developments
Natalizumab is a humanised monoclonal antibody blocking the α4-integrin adhesion molecule, thereby reducing cell migration across the blood–brain barrier.
Highly significant treatment effects on relapse rate and MRI activity were demonstrated in two studies – natalizumab versus placebo and IFN β-1a (Avonex) with natalizumab or placebo as add-on treatments – leading the FDA to license natalizumab (Tysabri) after patients had received treatment for a median duration of only 13 months. Soon after, however, the pharmaceutical companies (Biogen Idec and Elan Pharmaceuticals) withdrew the agent because three cases of progressive multifocal leukoencephalopathy were reported. At this time it remains uncertain whether this agent will find a place in the treatment of MS.
Campath-1H is a humanised monoclonal antibody targeting the CD52 antigen present on all lymphocytes and some monocytes. It activates complement and mediates antibody-dependent cell-mediated cytotoxicity, resulting in prolonged lymphopenia. Treatment effects on relapse rate and new MRI lesion formation were reported in preliminary open-label studies. Recently, these results were confirmed in a phase II randomised study, when interim analysis at one year demonstrated a 75% treatment effect on relapse rate and 60% effect on accumulation of disability compared with high-dose IFN β-1a (Rebif). The future fate of this compound for MS treatment appears uncertain after the occurrence of three cases of idiopathic thrombocytopenic purpura, one of which proved fatal.
Oral agents are not yet available for the treatment of MS, although several are undergoing clinical trials (phases II and III).
The availability of DMTs for MS immediately generated hope among those affected with the disease and their families, as well as neurologists. Patient care and prescribing practice have almost certainly been affected. Agents currently used (INF β-1b, IFN β-1a, glatiramer acetate) target the early inflammatory component of the disease and reduce disease activity (both clinical relapses and activity demonstrated by MRI). The slowing-off of the accumulation of disability is less convincing. Effect on the degenerative component has not been demonstrated. All approved preparations are given by self-injection. Efficacy is partial, there are no predictive definitions of responders/nonresponders or of treatment failure, and it is not known whether the reported benefits relate to a partial response for many or a significant response for a few patients. Treatment with IFN β induces production of antibodies in some patients, and these antibodies reduce or abrogate the benefit of treatment. No oral agents are approved, but several are in phase II and phase III trials. Combinations of agents are also being tested. Many countries have developed guidelines for the use of these agents, defining which patient populations should be treated. The new McDonald criteria allow CDMS to be diagnosed early by including MRI and CSF in the diagnostic scheme, thereby increasing the number of patients that could benefit from treatment. Despite progress made, the decision to start, change or stop treatment remains a challenge for the patient and the neurologist.
- McDonald WI, et al. Ann Neurol 2001;50:121-7.
- Polman CH, et al. Ann Neurol 2005;58:840-6.
- Compston A, Coles A. Lancet 2002;359:1221-31.
- Weinshenker BG, et al. Brain 1989;112:133-46.
- Compston ND. Lancet 1953;ii:271-5.
- Confavreux C, et al. N Engl J Med 2000;343:1430-8.
- Skoog B, et al. Mult Scler 2004;10 Suppl:S156.
- Beck RW, et al. Arch Ophthalmol 2003;121:944-9.
- Nilsson P, et al. J Neurol 2005;252:396-402.
- Goodin DS, et al. Neurology 2002;58:169-78.
- Panitch H, et al. Neurology 2002;59:1496-506.
- Durelli L, et al. Lancet 2002;359:1453-60.
- IFN-β Multiple Sclerosis Study Group. Neurology 1993;43:655-61.
- Paty DW, Li DK. Neurology 1993;43:662-7.
- Jacobs LD, et al. Ann Neurol 1996;39:285-94.
- Simon JH, et al. Ann Neurol 1998;43:79-87.
- PRISMS Study Group. Lancet 1998;352:1498-504.
- Li DK, Paty DW. Ann Neurol 1999;46:197-206.
- IFN-β Multiple Sclerosis Study Group and the University of British Columbia MS/MRI Analysis Group. Neurology 1995;45:1277-85.
- PRISMS-4. Neurology 2001;56:1628-36.
- Paty DW. Mult Scler 2003;9 Suppl 1:S138. Abstract P555.
- Runmarker B, Andersen O. Brain 1993;116:117-34.
- Sorensen PS, et al. Lancet 2003;362:1184-91.
- Johnson KP, et al. Neurology 1995;45:1268-76.
- Comi G, et al. Ann Neurol 2001;49:290-7.
- Teitelbaum D, et al. Mult Scler 2003;9:592-9.
- Hartung HP, et al. Lancet 2002;360:2018-25.
- Goodin DS, et al. Neurology 2003;61:1332-8.
- Ghalie RG, et al. Mult Scler 2002;8:441-5.
- Voltz R, et al. Mult Scler 2004;10:472-4.
- Durelli L, et al. Neurology 1986;36:238-43.
- Redford EJ, et al. Brain 1997;120:2149-57.
- Zivadinov R, et al. Neurology 2001;57:1239-47.
- Milligan NM, et al. J Neurol Neurosurg Psychiatry 1987;50:511-6.
- Jacobs LD, et al. N Engl J Med 2000;343:898-904.
- Comi G, et al. Lancet 2001;357:1576-82.
- Kappos L, et al. Mult Scler 2005;11 Suppl 1:S10 (Abstract 50).