Are current disease-modifying therapeutics in multiple sclerosis justified on the basis of studies in experimental autoimmune encephalomyelitis?


Address correspondence and reprint requests to Cris S. Constantinescu, Division of Clinical Neurology, University of Nottingham, Queen’s Medical Centre, C Floor South Block, Nottingham NG7 2UH, UK.


J. Neurochem. (2010) 115, 829–844.


The precise aetio-pathology of multiple sclerosis remains elusive. However, important recent advances have been made and several therapies have been licensed for clinical use. Many of these were developed, validated or tested in the animal model, experimental autoimmune encephalomyelitis (EAE). This systematic review aims to assess whether the current disease modifying treatments and those that are the closest to the clinic are justified on the basis of the results of EAE studies. We discuss some aspects of the utility and caveats of EAE as a model for multiple sclerosis drug development.

Abbreviations used:

experimental autoimmune encephalomyelitis


glatiramer acetate


multiple sclerosis

Multiple sclerosis (MS) is an enigmatic autoimmune disease of the nervous system characterised by inflammation with T cells, B cells, activated antigen presenting cells and complement; demyelination because of direct attack of myelin by the immune system or death of myelin producing oligodendrocytes; axonal and neuronal damage and loss, correlating with long term irreversible disability; and gliosis because of the astrocytic reaction to the immune mediated injury (Compston and Coles 2008). There is incomplete remyelination. Our understanding of its pathology and the development of therapies have been heavily linked to developments in the animal model experimental autoimmune encephalomyelitis (EAE) (Wallström and Olsson 2008).

This model of autoimmune disease has its origins in the encephalomyelitis occasionally observed after vaccination with Louis Pasteur’s rabies vaccine in the 19th century. See (Baxter 2007) for an excellent historical overview. The resulting condition had many clinical and histological similarities to MS. Subsequent work by Thomas Rivers and others refined the techniques to deliberately elicit encephalomyelitis by immunisation with CNS emulsions and using Freund’s adjuvant of killed Mycobacterium tuberculosis. Initial work was performed in monkeys and subsequently in rats, mice, guinea pigs, rabbits and other animals.

Since the early development of EAE a wide array of methods have been utilised for induction including active immunisation, adoptive transfer and the use of transgenic animals. Other key variables include the use of different types of adjuvant substances, myelin proteins or their immunodominant peptides, the use of pertussis toxin, and the timing of immunisation. Experimental models of MS have also been diversified through the introduction of viral-induced or toxin-induced demyelination paradigms. Different methods and models give rise to distinct types of EAE varying in clinical presentation, severity and natural history. To a certain extent, these types of disease reflect the variability observed in relapsing-remitting, secondary progressive, and primary progressive MS. Several recent reviews and a monograph are available on types and utility of EAE; see for example Krishnamoorthy and Wekerle 2009; Steinman and Zamvil 2006; Gold et al. 2006; Lavi and Constantinescu 2005.

Many of the currently used MS therapeutic agents and many of the imminently emerging ones (Lim and Constantinescu 2010; Niino and Sasaki 2010) are effective in EAE and in most cases were developed or tested in EAE. Below, we systematically review MS therapeutics in, or approaching clinical use and retrospectively assess how justified they are by EAE studies. We do not appraise EAE studies in toto nor discuss details of drug mechanisms or translational failures.

When EAE studies post-date the clinical licensing of these drugs for MS therapy, they may not be said to provide justification for translation. Nonetheless, these studies underscore the utility of EAE in advancing our understanding of drug mechanisms and for studying novel methods of drug delivery.

Materials and methods

Study identification, data extraction and analysis

Experimental studies of disease modifying drugs in EAE were identified from PubMed and Web of Science by searching all articles from 1970 onwards. Additional publications were identified from the references of identified publications, non-systematic reviews and book chapters. The search strategy employed the following keywords: experimental autoimmune/allergic encephalomyelitis/encephalitis, animal models and multiple sclerosis, demyelination, interferon, interferon-alpha, interferon-beta, type 1 interferon, glatiramer, copaxone, copolymer, copolymer-1, integrin, α4β1, VLA-4, natalizumab, tysabri, mitoxantrone, anthracene dione, novantrone, FTY720, fingolimod, S1P, laquinimod, ABR-215062, roquinimex, linomide, cladribine, leflunomide, teriflunomide, panaclar, fumarate, fumaric acid esters, disease modifying treatments, disease modifying drugs. Conference abstracts were not included.

The profusion of EAE methodologies makes analysis difficult. Studies of the same treatment may use different animals, different methods of EAE induction, different routes and dosages of administration, different outcome measures or scoring systems and different time points. Moreover, raw data are frequently unpublished.

We have included in our analysis studies using clinical (EAE) score as an outcome measure. Where studies provided clinical data at a number of time points we chose the time point most relevant to the intervention and outcome. In studies attempting to prevent onset of EAE, we typically used the peak disease score attained. Distinct experiments within a paper were considered separately.

Heterogeneous outcome measures were standardised by converting all results to a percentage of maximum score. In the majority of studies, the maximum score (usually 4 or 5) is defined as terminal paralysis or death. Some older studies define a lower maximum such as complete hindlimb paralysis; these were treated in the same way for the purposes of standardisation.

Some studies use a binary outcome measure ‘incidence’ defined as presence or absence of paralysis; several such incidence studies has been performed with Glatiramer acetate and were therefore analysed separately. For other agents, these studies were excluded. Studies were also excluded if insufficient data were provided for the purposes of our analysis. These studies have nonetheless been included in tables for reference and been marked with a star. We employed the Cochrane Review Manager software version 5 ( to collate and analyse studies.

Data from trials of different sizes were incorporated in our meta-analysis. In order to offset the skewing of data resulting from these disparities, each trial was weighted. Different methods are available for this; our meta-analysis employed a random effects weighting which incorporates heterogeneity in true treatment effect (DerSimonian and Laird 1986).

The resulting data are summarised in forest plots where appropriate. The adjacent tables display the mean standardised clinical score and standard deviation, number of animals, study weighting and a standardised measure of effect, calculated as above. For incidence studies, data are expressed as odds ratio, that is, ratio of the odds of paralysis occurring in the treatment group to the odds of paralysis in the control group.

Funnel plots to assess for publication bias were drawn for the agents most intensively studied in EAE – glatiramer acetate (GA) and type-1 interferons.

Glatiramer acetate

Teitelbaum and colleagues developed a random co-polymer of tyrosine, glutamate, alanine and lysine in ratios resembling myelin basic protein in 1971 initially as a potential encephalitogen. Serendipitously, this substance – Copolymer 1 or GA was found to block the induction of EAE. After clinical trials, GA was approved for the treatment of relapsing-remitting MS some 25 years later (Farina et al. 2005). Studies in EAE using Copolymer 1 are summarised in Table 1 and Figs 1 and 2. As these figures demonstrate, EAE studies consistently demonstrated highly significant effects with the use of GA. These data justified and encouraged translational studies in MS, which resulted in the establishment of GA as a first line disease-modifying treatment.

Table 1.   Glatiramer acetate (GA) studies in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. *Study excluded from analysis because of insufficient data or inappropriate outcome measure.

  2. M, male; F, female; n/s, not specified; MBP, myelin basic protein; PLP, proteolipid protein; MOG, myelin oligodendrocyte glycoprotein; I, incidence; CS, clinical score; CE, cytokine expression; H, histology.

Teitelbaum et al. (1971)Guinea pig (n/s)Bovine spinal cordIntravenous GAI, HProphylaxis
Teitelbaum et al. (1973)Guinea pig and rabbit (n/s)Bovine spinal cord/purified human antigenIntravenous/intramuscular/intraperitoneal GAI, HProphylaxis
Teitelbaum et al. (1974)Monkey (n/s)Bovine spinal cordIntramuscular GAITherapy
Keith et al. (1979)Guinea pig (n/s)Guinea pig spinal cordSubcutaneous GAI, CS, HProphylaxis and therapy
Lisak et al. (1983)Guinea pig (M)MBPIntramuscular GAI, CS, HProphylaxis and therapy
Aharoni et al. (1993)Mouse (F)Mouse spinal cordGA induced T suppressor cells from hybridomaI, CSProphylaxis. Analysis used incidence
Teitelbaum et al. (1996)Mouse (n/s)PLPGA with encephalitogenic inoculumI, CSProphylaxis
Aharoni et al. (1997)Mouse (n/s)Mouse spinal cordAdoptive transfer of GA induced T cells pre-immunisationCS, CEProphylaxis
Aharoni et al. (1998)Mouse (F)PLPAdoptive transfer of GA induced T cells pre-immunisationI, CS, CEProphylaxis
Teitelbaum et al. (1999)Rat (F)Mouse spinal cordOral GA or adoptive transfer of oral GA induced T cellsI, CSProphylaxis
Gilgun-Sherki et al. (2003)Mouse (n/s)MOGSubcutaneous GAI, CSProphylaxis
Teitelbaum et al. (2004)*Mouse, rat, rhesus monkey (n/s)MBP/PLP/MOGOral GA or adoptive transfer of GA induced T cells pre-immunisationCS, CEProphylaxis
Illes et al. (2004)Humanised mouse (F)MBPSubcutaneous GACS, CEProphylaxis and therapy
Stern et al. (2004)Humanised mouse (F)MBPSubcutaneous GA or adoptive transfer of GA induced T cellsCSProphylaxis and therapy
Giuliani et al. (2005)Mouse (n/s)MOGSubcutaneous GA pre-immunisationCS, H, Th2 shiftProphylaxis with glatiramer. Additive effect with minocycline (not shown)
Aharoni et al. (2005)*Mouse (F)MOGSubcutaneous GACS, HProphylaxis/therapy
Jee et al. (2007)Mouse (F)MOGAdoptive transfer of GA induced CD4+CD25+ T cells pre-immunisationCSProphylaxis
Begum-Haque et al. (2008)Mouse (F)MOGDaily subcutaneous GACS, CEProphylaxis
Stern et al. (2008)*Mouse (n/s)MOGDaily subcutaneous GACS, H (electron microscopy)Prophylaxis
Kala et al. (2010)Mouse (n/s)MOGAdoptive transfer of GA induced B cellsCSProphylaxis
Figure 1.

 Forest plot of studies of glatiramer acetate in EAE using clinical score as an outcome measure. See Table 1 for details of excluded studies. Data illustrated as mean standardised difference of clinical score and confidence intervals. All studies showed a beneficial treatment of effect.

Figure 2.

 Forest plot of studies of glatiramer acetate in EAE using disease incidence as an outcome measure. Data illustrated as odds ratio and confidence intervals. All studies showed a beneficial treatment of effect.

A funnel plot of a measure of precision (standard error) against the mean difference was drawn to assess for the presence of publication bias in EAE studies of glatiramer acetate. This is shown in Fig. 3; visual inspection of the plot did not demonstrate major asymmetry, arguing against the presence of publication bias.

Figure 3.

 Funnel plot of a measure of precision (standard error) against treatment effect (standardised mean difference) for EAE studies of glatiramer acetate using clinical score. No gross asymmetry observed suggesting no evidence for publication bias.

Type-1 interferons

Longstanding interest in the role of cytokines in the pro-inflammatory milieu of autoimmune nervous system disease led to the first trials of interferon in EAE (Abreu 1982; Abreu et al. 1983). This trial demonstrated a significant beneficial effect; a large number of trials followed with various interferon subtypes and administration techniques. The promising results led on to clinical trials in MS and the eventual establishment of interferon-beta as a widely used disease modifying treatment for MS. Several formulations of interferon are now licensed in MS – interferon beta 1a (Avonex/Rebif) and interferon beta 1b (Betaferon/Betaseron/Extavia).

Currently interferon beta is well established as a first line treatment of relapsing remitting MS, along with glatiramer acetate (Bermel and Rudick 2007). The role of interferons in EAE – both type 1 such as interferons alpha and beta, and type 2 such as interferon gamma – has been reviewed (Heremans and Biliau 2005; Gran 2007). Knockout mouse studies have provided further evidence for the importance of type-1 interferons in EAE. Deletion of the interferon-beta gene augments EAE scores (Teige et al. 2003). Mice with defects or deletions of the type-1 interferon receptor (IFNAR) or associated receptor domains also demonstrate more severe EAE than controls (Guo et al. 2008; Prinz et al. 2008).

We include in our analysis studies using both interferon alpha and interferon beta. Both are type 1 interferons and use the same receptor. Moreover, several clinical trials have been carried out into the use of interferon alpha in MS and its use is well established in Cuba (see e.g. Knobler et al. 1984; Durelli et al. 1996; Cabrera-Gomez and Lopez-Saura 1999).

Figure 4 illustrates the EAE data for the use of type-1 interferons, and further details are provided in Table 2. Although some studies produced equivocal results or even significant worsening of disease with interferons, our meta-analysis indicates that overall the data support the hypothesis that type-1 interferon administration improves clinical score, justifying subsequent translational studies.

Figure 4.

 Forest plot of type-1 interferon studies in EAE. Data illustrated as mean standardised difference of clinical score and confidence intervals. See Table 2 for details of excluded studies. Five subgroups across four studies demonstrated an exacerbation of disease with treatment; meta-analysis shows an overall beneficial effect of treatment.

Table 2.   Interferon studies in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. See Table 1 footnote for symbols and abbreviations.

Abreu (1982)Rat (M)Guinea pig spinal cordIntravenous rat interferonCS, IProphylaxis
Hertz and Deghenghi (1985)Rat (F)Guinea pig spinal cordIntravenous or intracerebral rat interferon or human interferon betaCSProphylaxis with rat interferon
No effect with human interferon
Abreu (1985)Rat (M)Adoptive transfer MBP-induced lymph cellsRat interferon added to cultures prior to adoptive transferCSProphylaxis
Abreu et al. (1986)Rat (M)Guinea pig spinal cord/adoptive transfer MBP induced lymph cellsIntraventricular rat interferonCS, HNo significant effect as prophylaxis or therapy
Brod and Burns (1994)Mouse (F)Mouse spinal cordOral murine type 1 interferonCS, HSuppresses relapse
Brod et al. (1995b)Rat (F)Guinea pig MBPOral rat/human type 1 interferonCS, HProphylaxis
Brod et al. (1995a)Mouse (F)Mouse spinal cordOral human or murine interferon alphaCS, I (of adoptive transfer)Therapy, prophylaxis of adoptive transfer
Yu et al. (1996)Mouse (F)PLPSubcutaneous murine interferon betaCS, H, exacerbation rateTherapy
Brod and Khan (1996)Mouse (F)Mouse spinal cordOral/subcutaneous murine interferon alphaCSTherapy
Ruuls et al. (1996)Rat (F)MBPSubcutaneous rat interferon betaCSTherapy. Withdrawal exacerbated disease.
Croxford et al. (1998)Mouse (M)Mouse spinal cordIntracerebral mouse-interferon-beta expressing plasmidCS, CEProphylaxis
van der Meide et al. (1998)Rat (F)Guinea pig spinal cord and MBPSubcutaneous interferon betaCS, lymphocyte proliferationIncreased severity of EAE
Yasuda et al. (1999)Mouse (F)MBPIntraperitoneal interferon betaCS, CEProphylaxis
Luca et al. (1999)Mouse (F)PLPSubcutaneous/intraperitoneal interferon betaCS, CENo significant effect as prophylaxis or therapy
Brod et al. (2000)Mouse (F)Mouse spinal cordAdoptive transfer of oral interferon-alpha induced CD8+ T cellsCSTherapy
Wender et al. (2001)*Rat (F)Guinea pig spinal cordRecombinant human interferon beta 1aI, H, CEProphylaxis
Schaefer et al. (2006)Mouse (F)PLPIntramuscular mouse-interferon-betaexpressing plasmidCSProphylaxis
Jaini et al. (2006)Mouse (F)PLPIntramuscular mouse interferon-beta expressing plasmidCSTherapy
Martin-Saavedra et al. (2007)Mouse (F)Bovine MBPIntraperitoneal interferon-betaCS, CEProphylaxis
Makar et al. (2008)Mouse (F)PLPIntravenous bone-marrow stem cells transduced to express interferon betaCS, CEProphylaxis

A funnel plot of standard error against mean difference was drawn (Fig. 5); visual examination does not demonstrate significant asymmetry, arguing against the presence of publication bias.

Figure 5.

 Funnel plot of a measure of precision (standard error) against treatment effect (standardised mean difference) for EAE studies of type-1 interferons using clinical score. No gross asymmetry observed suggesting no evidence for publication bias.

Natalizumab and other anti-adhesion molecule compounds

The experiments of Steinman and Yednock in EAE led to the development and production of the monoclonal antibody natalizumab. Demonstrating that the alpha-4 beta-1 integrin (very late activation antigen VLA-4) was the adhesion molecule most critical to lymphocyte adhesion to the blood brain barrier endothelial cells and migration into the inflamed brain, these investigators produced a monoclonal antibody against it, eliciting an improvement in various models of EAE (Yednock et al. 1992). EAE studies using clinical score as an outcome measure are presented in Fig. 6 and Table 3.

Figure 6.

 Forest plot of EAE studies of Natalizumab and other anti-adhesion-molecule agents in EAE. See Table 3 for details of exclusions. Data illustrated as mean standardised difference of clinical score and confidence intervals. Three subgroups from two studies showed an exacerbation of disease with treatment; overall meta-analysis demonstrates beneficial effect of treatment.

Table 3.   Anti adhesion-molecule agents in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. See Table 1 footnote for symbols and abbreviations.

Yednock et al. (1992)*Rat (n/s)MBP-specific T cell clonesIntraperitoneal anti-VLA4 antibodyIProphylaxis
Kent et al. (1995)Guinea pig (F)Guinea pig CNSIntracardiac anti-VLA4 antibodyCS, HProphylaxis – delayed onset. Therapy temporarily efficacious
Leger et al. (1997)*Guinea pig (F)Guinea pig CNSSubcutaneous humanised mouse antibody against alpha4 integrinCSTherapy (complete reversal)
Soilu-Hanninen et al. (1997)*Mouse (F)Mouse spinal cord + Semliki Forest VirusAnti-VLA4 antibodyIProphylaxis
Brocke et al. (1999)Mouse (F)Adoptive transfer of MBP induced T cellsAnti-alpha4 integrin antibodyCSProphylaxis
Theien et al. (2001)Mouse (F)PLP or adoptive transfer of PLP induced T cellsAnti-VLA4 antibodyCSTreatment inhibits induction but exacerbates ongoing EAE
Piraino et al. (2002)Guinea pig (F)Guinea pig CNS tissueVLA-4 inhibitor CT301CS, H, CESustained benefit in chronic EAE
van der Laan et al. (2002)Rat (M)MBP or adoptive transfer of MBP induced T cellsModified peptide inhibitor of VLA-4CSNot significant in actively induced EAE
Theien et al. (2003)Mouse (F)PLPVLA-4 antagonist BIO 5192CSTreatment at height of acute disease temporarily beneficial, otherwise exacerbates disease
Cannella et al. (2003)Mouse (F)Adoptive transfer of MBP induced T cellsVLA-4 antagonist TBC-3486CS, H, CEBenefit in acute EAE only
Leone et al. (2003)Rat (F)MBPAnti-VLA4 antibody or antagonist BIO5192CSAntibody and BIO5192 similarefficacy in delaying onset andreducing peak severity
Myers et al. (2005)Mouse (F)PLPAntisense blockade of alpha4 integrinCSEffect as therapy > prophylaxis

Although at least two studies (Theien et al. 2001, 2003) demonstrated a significant exacerbation of disease with anti-VLA4 agents and advised caution in translation to MS, our meta-analysis suggests a significant overall beneficial treatment effect.

Clinical license for MS was obtained in 2004 for the human monoclonal antibody Natalizumab (Tysabri) after good results were obtained in Phase-III clinical trials. Unfortunately, the drug was withdrawn 3 months later following reports of progressive multifocal leukoencephalopathy, particularly when Natalizumab was used in association with interferon beta (Mix et al. 2008). Progressive multifocal leukoencephalopathy results from reactivation of latent JC virus infection. After safety review, the drug was re-licensed in 2006 under a special prescription program as the clinical benefits are thought to outweigh the risks.

The story of natalizumab clearly illustrates both some of the benefits and the hazards of translating a therapy from animal models to human clinical use. JC virus is a neurotropic virus that infects only humans, therefore this side-effect was entirely unpredictable on the basis of EAE experiments (Steinman and Zamvil 2005). On the other hand, natalizumab was also a success story of translation of results from EAE into clinical practice, leading to the establishment of a new generation of disease modifying treatments for MS. It is currently considered the most potent licensed disease-modifying treatment.

Further studies in EAE have investigated the use of inhibitors of VLA-4 other than monoclonal antibodies. Some of these inhibitors have shown very impressive results; none as yet have been translated to clinical use.


Fujita and colleagues first isolated a potent immunosuppressive compound, ISP-1 from cultures of the fungus Isaria sinclairii (Fujita et al. 1994). Chemical modification to enhance immunosuppression and reduce toxicity produced the novel synthetic agent FTY720 (Fingolimod). This agent has demonstrated potential in the survival of transplanted allografts in rats. Brinkmann and colleagues demonstrated that fingolimod could prevent the development of EAE in rats in 2002 (Brinkmann et al. 2002) (Fig. 7).

Figure 7.

 Forest plot of EAE studies of fingolimod. Data illustrated as mean standardised difference of clinical score and confidence intervals. All studies showed a significant beneficial effect of treatment.

These promising results encouraged translation to clinical use; two phase-III clinical trials have demonstrated impressive clinical effect (Cohen et al. 2010; Kappos et al. 2010). EAE studies are summarised in Fig. 4 below and Table 4.

Table 4.   Fingolimod studies in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. See Table 1 footnote for symbols and abbreviations.

Brinkmann et al. (2002)Rat (n/s)Bovine spinal cordOral FTY720CSProphylaxis
Fujino et al. (2003)Rat (M)MBPOral FTY720/adoptive transfer of FTY720 induced splenocytesCS, HEffective prevention of acute EAE by both methods
Rausch et al. (2004)*Rat (F)Guinea pig spinal cordOral FTY720CS, H, MRIEffective prevention, diminished relapse
Webb et al. (2004)Mouse (F)PLP or adoptive transfer of PLP induced T cellsIntraperitoneal FTY720 or FTYPCS, CEEffective in established EAE
Kataoka et al. (2005)Rat and Mouse (M)MBP in rats
PLP in mice
Oral FTY720CS, HEffective prevention and therapy
Balatoni et al. (2007)Rat (F)MOGFTY720CS, electrophysiologyEffective prevention and therapy
Foster et al. (2007)Rat (F)Rat brain/spinal cordOral FTY720CS, lymphocyte countTherapy
Foster et al. (2009)Rat (F)Rat brain/spinal cordOral FTY720CS, H, CEProphylaxis, therapy and rescue
Papadopoulos et al. (2010)Rat (F)MOGOral FTY720CS, CE, antibody productionProphylaxis and therapy


The novel chemical agent Laquinimod was developed after phase 3 trials of the immunomodulator Roquinimex (Linomide) were halted because of side effects (Noseworthy et al. 2000). A systematic program was commenced to define immunomodulators effective in MS but without the usually associated side effects. This was achieved by specific chemical modification of the Roquinimex structure (Brunmark et al. 2002).

Experimental autoimmune encephalomyelitis studies have been performed using laquinimod (see Table 5). The first, in 2002 demonstrated significant and dose-dependent suppression of acute EAE in SJL/N mice (90% suppression of mean maximum score in the treated group). In addition, orally-administered laquinimod inhibited relapses in a chronic-relapsing EAE model in B10.RIII mice. A study in Lewis rats demonstrated significant suppression of active EAE, greater than that achieved by Roquinimex (Yang et al. 2004). These encouraging data justified translational studies.

Table 5.   Laquinimod studies in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. See Table 1 footnote for abbreviations.

Brunmark et al. (2002)Mouse (F)Mouse spinal cord/MBPOral laquinimodCS, flow cytometryProphylaxis, therapy
Yang et al. (2004)Rat (F)MBPOral laquinimodCS, CEProphylaxis

Laquinimod has demonstrated efficacy in MS (Comi et al. 2008; Tselis 2010) and was granted Fast Track designation by the FDA; it is currently undergoing a second phase 3 trial and may enter clinical use in 2011. If so, this would be clinical translation almost three times faster than Glatiramer acetate.


Mitoxantrone is a synthetic cytotoxic chemotherapeutic agent and an immunosuppressant. These effects were thought to be promising for MS therapeutics. EAE studies were initially carried out and are summarised in Table 6. All studies showed significant beneficial effects of treatment, reiterated by our meta-analysis (not shown because of small number of studies). After clinical trials, Mitoxantrone was licensed for use in several forms of advancing MS (Neuhaus et al. 2006). Thus, EAE studies would clearly justify the use of mitoxantrone in MS.

Table 6.   Mitoxantrone studies in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. See Table 1 footnote for abbreviations.

Ridge et al. (1985)Rat (M)Active EAE/adoptive transfer with Guinea pig MBPIntraperitoneal mitoxantroneCS, HProphylaxis, therapy
Levine and Saltzman (1986)Rat (both)Guinea pig spinal cordIntraperitoneal/subcutaneous mitoxantroneCS, HProphylaxis, therapy
Lublin et al. (1987)Rat (n/s)Mouse spinal cordIntraperitoneal mitoxantroneI, CSProphylaxis/delayed onset of relapse

Other agents

Cyclophosphamide is a potent cytotoxic agent and immunosuppressant which impairs both humoral and cell-mediated immunity and has well-established use in the treatment of cancer and autoimmune disease. Open-label studies have given conflicting results but it has been suggested to be beneficial in non-responders to first-line therapies or those with particularly aggressive MS (Weiner and Cohen 2002). A recent study has tested its use in four different EAE rodent models (Mangano et al. 2009). No significant effects were demonstrated in SJL or C57Bl/6 mice however prophylactic and therapeutic effects were obtained in Dark Agouti rats. Cyclophosphamide also delayed onset when given as prophylaxis in Lewis rats. These disparities suggest that different immunological mechanisms may be at work in different models of EAE and underscore the importance of testing interventions in a range of models.

Previously used in Rheumatoid Arthritis, leflunomide is a non-competitive reversible inhibitor of the enzyme dihydroorotate dehydrogenase, involved in pyrimidine synthesis. Blockade of the enzyme inhibits T and B cell clonal expansion and antibody production (Korn et al. 2004).

One EAE study has been performed using leflunomide and one using its active metabolite, teriflunomide (Korn et al. 2004; Merrill et al. 2009). The studies showed successful prophylaxis and therapy of EAE and are summarised in Table 7. Teriflunomide is undergoing clinical trials in MS.

Table 7.   Studies of other agents in EAE
Study/referenceExperimental animal (gender)EncephalitogenInterventionOutcome measure(s)Effect/notes
  1. See Table 1 footnote for abbreviations.

Mangano et al. (2009)DA Rat (F)
Lewis rat (M)
Mouse (F)
Spinal cord homogenate
Guinea pig MBP
Intraperitoneal cyclophosphamideCS, CE, HNo effect in mice. Therapy and prophylaxis in DA rat. Delay of induction in Lewis rat.
Korn et al. (2004)Rat (n/s)Guinea pig MBPOral leflunomideCS, CE, T cell chemotaxisProphylaxis
Merrill et al. (2009)Rat (M)Rat spinal cordOral teriflunomideCS, H, electrophysiologyProphylaxis, therapy
Schilling et al. (2006)Mouse (F)MOGOral fumaric acid estersCS, H, CEProphylaxis

Fumaric acid esters are effective in the treatment of psoriasis where they have been reported to induce a Th1–Th2 shift, though their mechanism of action is incompletely understood. In neuroinflammation, they are thought to have a dual beneficial action, immunomodulatory and neuroprotective. The fact that this class of drugs is already in clinical use facilitated clinical trials; an open-label study has demonstrated improvement in radiological (magnetic resonance imaging) signs of MS (Schimrigk et al. 2006). Recently, a phase IIb study demonstrated a significant effect or oral fumarate in relapsing-remitting MS (Kappos et al. 2008). A single animal study has been performed demonstrating effective prophylaxis of EAE induction and an increase in interleukin-10 production (see Table 7). However, interestingly, fumaric acid treatment was ineffective in a cuprizone-induced demyelination model, suggesting that in EAE (and perhaps in MS) it acts primarily through its effects on the immune system rather than through enhancing remyelination by oligodendrocyte stimulation (Moharregh-Khiabani et al. 2010).

Discussion and summary

As this review demonstrates, the majority of MS therapies in current clinical use were extensively investigated in EAE. Glatiramer acetate, interferons, natalizumab, fingolimod, laquinimod and mitoxantrone are all justified on the basis of EAE results. Further development of the newer drugs is in progress and most of these are justified by EAE studies, although some, such as alemtuzumab, have not been tested in EAE models for reasons that are discussed below. Systematic reviews are useful to collate the different studies that have been carried out and perform meta-analyses of available data. This task is complicated by the heterogeneity of EAE, the timing of intervention and the outcome measures utilised. Some of this heterogeneity is inevitable and indeed beneficial in illuminating different aspects of patho-immunology and encouraging novel therapeutic efforts. Standardising outcome measures such as clinical scoring of EAE would not detract from this and would greatly facilitate meta-analysis and comparison of novel against established treatments (Fleming et al. 2005). Clinical scores that take into account a larger range and finer gradation of clinical states will be more useful than cruder scores.

This review has been selective in that it has reviewed agents used in MS and retrospectively evaluated the role of EAE in their journey to clinical use, or the congruence between their success in MS and the EAE results. A number of agents have been successful in EAE and nonetheless fallen by the wayside before reaching clinical practice. An example of a failed translation is anti-tumour necrosis factor (TNF) alpha (Infliximab), although attention to studies with knockout mice would have suggested that TNF-alpha and lymphotoxin are not required for EAE induction (Frei et al. 1997). Other translations, which failed for a variety of reasons, include anti-CD3 and anti-CD4 antibodies, oral tolerogens, Sulfasalazine and Linomide (Mix et al. 2008).

Conversely, the novel agent Cladribine has been through successful clinical trials (Giovannoni et al. 2010) without being published as validated in EAE. This agent is a good example of a ‘borrowing’ from another clinical field. Cladribine has well-characterised use in haematological malignancy, an established safety profile and known lymphocytotoxic effects (Brousil et al. 2006; Leist and Vermersch 2007). In this context its use in MS appears reasonable. The example suggests that EAE studies, though helpful, may not always be strictly necessary to attempt novel therapies in MS.

The humanized monoclonal anti-CD52 antibody Alemtuzumab (Campath) is undergoing phase-III clinical trials in MS. Again, this agent has not been tested in EAE though in this case the reasons go beyond the fact that it is established in the treatment of haematological malignancy: the CD52 peptide varies markedly between species and Alemtuzumab does not cross-react with murine CD52 (Hale 2001). A recent study has produced a transgenic mouse expressing human CD52 (Hu et al. 2009). Alemtuzumab treatment of this mouse produced lymphocyte depletion and cytokine induction comparable to that observed in humans. This suggests a means by which the problem of absent cross-reactivity may be approached. As yet, no EAE studies have been performed in these transgenic mice.

Criticisms of EAE as a model for MS are numerous and some may be well-founded (see e.g. (Sriram and Steiner 2005): extrapolating results from EAE to human MS can be problematic, particularly with regard to pharmacological studies. EAE is not ‘animal MS’; its immuno-aetiology is by necessity artificial and will not correspond precisely to that of MS – which itself is incompletely described.

Experimental autoimmune encephalomyelitis in mice is in general characterised by major global brain injury with axonal and neuronal damage and relatively limited primary demyelination. Most EAE models demonstrate a predominance of CD4+ T cells contrasting with the CD8+ shift observed in MS (Gold et al. 2006). Rat EAE, particularly when induced by myelin oligodendrocyte glycoprotein in the Dark Agouti (DA) strain comes closer to replicating the pathological and clinical phenotypes of MS. In the same strain, EAE has been induced by immunising with spinal cord homogenate in the absence of adjuvant substances (Stosic-Grujicic et al. 2004; see also the review Wallström and Olsson 2008).

As a recent review (Kap et al. 2010) demonstrates, the benchmark model is marmoset EAE with its anatomical, physiological, pathological and phylogenetic proximity to humans. Recent work suggests the possibility of EAE induction without the use of the traditional complete Freund’s adjuvant which elicits Th1 skewing and neutralising antibodies absent in MS (Jagessar et al. 2010). The marmoset model suffers its own difficulties however, particularly in cost, time, stringent ethical requirements, difficulty of genetic manipulation, and limited availability of reagents and probes (Gold et al. 2006; Kap et al. 2010). The outbred population better reflects the genetic heterogeneity of MS populations but makes interpretation of data more difficult.

Other difficulties with EAE and translation include establishing relative pharmacokinetics for drug dosing, the poor prediction of drug toxicity, the use of inbred animal strains, the small numbers of test subjects, the frequent use of a single gender of animal and the highly compressed natural history relative to MS (Steinman and Zamvil 2006). The problem of differing molecular targets in human and animal studies has been discussed with regard to Alemtuzumab. Everything that works in EAE may not work in MS.

Understanding the limitations of EAE and keeping these criticisms in mind will allow us to learn from it and apply it in a productive manner. The use of EAE to model some phenomenon in MS immunology, neurobiology or clinical therapy is heavily dependent on the question that the experiment is expected to answer. The greatest utility of EAE is in understanding general immunological mechanisms and phenomena, as well as in vivo interactions between the immune and the nervous system. Though translation of basic research will always involve an element of uncertainty, confidence in EAE results can be improved by replication across a range of species, induction regimes and treatment paradigms.

Though EAE is certainly an imperfect mirror of MS, many immuno-pathological and histological findings are impressively replicated by animal models. This is particularly clear in marmoset models which further demonstrate clinical courses comparable to MS including spontaneous relapses. We believe these parallels make EAE invaluable in elucidating the basic immunopathological mechanisms of MS and providing a testing ground for novel therapies.


CC has received research support and travel support to scientific meetings from Bayer-Schering, Biogen Idec, Merck-Serono, and TEVA, consultancy fees from Novartis, and has been on the advisory board of Bayer-Schering, Biogen Idec, and TEVA. BG has received research support from Bayer-Schering, Biogen Idec, and Merck-Serono, and travel support to scientific meetings from Bayer-Schering, Biogen Idec, Merck-Serono, and TEVA. NF has nothing to disclose.