NeuroRx威: The Journal of the American Society for Experimental NeuroTherapeutics
`
`Developing Therapeutics for the Treatment of Multiple Sclerosis
`
`Neurology and GI Centre of Excellence for Drug Discovery, GlaxoSmithKline Pharmaceuticals, Harlow, Essex CM19 5AW,
`United Kingdom
`
`David J. Virley
`
`Summary: Multiple sclerosis (MS) is both a complex and
`chronic neurological disease of the CNS. This poses unique
`challenges for drug discovery in terms of delineating specific
`targets related to disease mechanisms and developing safe and
`effective molecules for clinical application. Preclinical animal
`models of MS provide the necessary test bed for evaluating the
`effects of novel therapeutic strategies. Because the clinical
`manifestations and pathological consequences of disease vary
`dramatically from individual to individual, as well as treatment
`response to existing therapies, this creates a significant research
`endeavor in terms of translating preclinical methodologies to
`the clinical domain. Potentially exciting treatments have
`emerged in the form of natalizumab (Tysabri), an ␣4 integrin
`antagonist, and more recently FTY720, a sphinogosine-1 phos-
`phate receptor modulator, providing a compelling proof-of-
`principle from bench to bedside. However, further research is
`
`required to discharge safety concerns associated with these
`therapeutic avenues. Future prospects in the guise of disease-
`modifying therapies that target the inflammatory and neurode-
`generative components of disease have come to the forefront of
`preclinical research with the sole aim of reducing the underly-
`ing irreversible progressive disability of MS. Significant
`progress with novel therapies will be made by implementing
`biomarker strategies that extrapolate robustly from animal
`models to the divergent patient populations of MS. The future
`therapeutic options for MS will depend on improvements in
`understanding the precise factors involved in disease onset and
`progression and subsequently the development of oral thera-
`peutics that translate sustained benefit from the preclinical con-
`text into clinical reality. Key Words: Multiple sclerosis, inflam-
`mation, demyelination, regeneration, experimental autoimmune
`encephalomyelitis, therapeutics.
`
`INTRODUCTION
`
`Multiple Sclerosis (MS) is the most common demy-
`elinating disease of the CNS, affecting young adults in
`their formative years, where current treatments have lim-
`ited effectiveness. MS is typified pathologically by mul-
`tiple inflammatory foci, plaques of demyelination, glio-
`sis, and axonal pathology within the brain and spinal
`cord, all of which contribute to the clinical manifesta-
`tions of neurological disability. Although the causal
`events in precipitating the disease are not fully under-
`stood, most evidence implicates an autoimmune etiology
`together with environmental factors, as well as specific
`genetic predispositions. Functional impairment, disabil-
`ity, and handicap are expressed as paralysis, sensory and
`cognitive disturbances, spasticity, tremor, lack of coor-
`dination, and visual impairment. All these symptoms
`significantly impact on the quality of life of the individ-
`
`Address correspondence and reprint requests to David J. Virley,
`Ph.D., Neurology and GI Centre of Excellence for Drug Discovery,
`GlaxoSmithKline Pharmaceuticals, Third Avenue, Harlow, Essex
`CM19 5AW, United Kingdom. E-mail: David_J_Virley@gsk.com.
`
`ual. The clinical course of MS can vary from individual
`to individual, but invariably the disease can be catego-
`rized into three forms: relapsing-remitting, secondary
`progressive, and primary progressive. In approximately
`85% of patients with MS, the disease starts with alter-
`nating episodes of neurological impairment character-
`ized by relapses with subsequent complete or partial
`remission.1 In the majority of patients over a variable
`period, this course is followed by a secondary progres-
`sive phase where recovery is absent. A minority of pa-
`tients (⬃15%) display primary progressive characteris-
`tics where irreversible worsening of clinical signs
`manifest from disease onset.1 The disease as a whole
`places a huge burden on economic and societal resources
`and highlights the importance of developing novel, safe,
`and effective therapies for MS in treating the underlying
`and progressive course of the disease.
`This article, will review key challenges for drug dis-
`covery in MS, based initially on the existing clinical
`outcome measurements, available preclinical models to
`simulate the disease process, and treatment response to
`current
`therapeutics. Specific emphasis will
`then be
`placed on novel therapeutic challenges for MS, drawing
`
`638
`
`Vol. 2, 638 – 649, October 2005 © The American Society for Experimental NeuroTherapeutics, Inc.
`
`Page 1 of 12
`
`YEDA EXHIBIT NO. 2003
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`FUTURE THERAPEUTIC CHALLENGES IN MS
`
`639
`
`on the weight of evidence from natalizumab (Tysabri)
`and examples of novel anti-inflammatory, neuroprotec-
`tive, and regenerative approaches. Improving the trans-
`lational quality of candidate compounds from bench to
`bedside, involving the utility of biomarkers will also be
`highlighted to help guide the future development of ro-
`bust treatment options for MS.
`
`EVALUATION OF CLINICAL OUTCOME
`IN MS
`
`Due to the fluctuating nature and breadth of symp-
`toms, robust measurement of the clinical manifestations
`of MS is problematic. For appropriate assessment of
`efficacy of drug treatment within clinical trials, the mea-
`surement tool(s) should be sensitive and reproducible
`enough to detect a significant treatment effect. The Ex-
`panded Disability Status Scale (EDSS) is considered the
`most widely used instrument
`to evaluate therapeutic
`strategies in MS, despite drawbacks of reproducibility
`and inadequate representation of upper limb function and
`cognitive decline.2 More recent developments by the
`National MS Society’s clinical outcomes task force in
`MS have provided a more quantitative and sensitive tool
`in the MS Functional Composite (MSFC).3 The MSFC
`provides more objective measures of leg function, arm
`and hand function, and cognitive function. This instru-
`ment comprises of three specific tests that probe walking
`speed (timed 25 foot walk), fine upper limb dexterity (9
`hole peg test), and cognitive processing (paced auditory
`serial additional test) on a continuous scale. The MSFC
`has been shown to be more sensitive to change than
`EDSS, and during and after treatment with the cortico-
`steroid, methylprednisolone, clinical improvements from
`acute relapses were more consistently measured.4 Fur-
`thermore, MSFC scores have been found to correlate
`with EDSS, magnetic resonance imaging (MRI) lesion
`load, and self-reported quality of life. In terms of appli-
`cation to clinical trials, the MSFC has been reported to be
`strongly predictive of clinical and MRI status in relaps-
`ing-remitting patients and may offer improved sensitivity
`to assessing progression in the course of disease and
`ultimately, the effects of novel disease-modifying thera-
`pies for MS. Recent concurrent validation of the MSFC
`with MRI has been established to determine biological
`sensitivity to disease severity.5 In this study, the EDSS
`was directly compared with the MSFC in relation to MRI
`measurements of lesion load. The EDSS was not shown
`to correlate with MRI measures of disease, whereas the
`MSFC was shown to correlate with both T1 and T2
`lesion load, especially in both relapsing-remitting and
`secondary progressive MS patients. Although MRI pro-
`vides valuable primary end-points in phase II clinical
`trials and supportive outcome measures to phase III clin-
`ical trials, as a putative surrogate marker of disease ac-
`
`FIG. 1. Schematic view of the putative pathogenic steps in MS.
`1: Activation of autoreactive T cells by antigen presenting cells in
`the periphery. 2: Migration of T cells and monocytes through the
`blood brain barrier. 3: Amplification of local inflammation and
`activation of resident microglia. 4: Release of toxic mediators
`damages myelin and oligodendrocytes with the culmination of
`axonal loss.
`
`tivity, further developments on assessing additional MRI
`parameters are required to improve the association with
`clinical disability. The evolving development of more
`sensitive, predictive, and practical measures of impair-
`ment and disability aligned with more comprehensive
`and quantitative assessments of MRI tissue signatures in
`the brain and spinal cord will help evaluate novel ther-
`apeutic strategies for MS.
`
`ANIMAL MODELS OF MS
`
`A major thrust of preclinical research is to identify and
`validate novel targets within appropriate disease-relevant
`models that mimic the clinical situation as closely as
`possible. Animal models form an essential part of the
`drug development process to assess the validity of the
`target for therapeutic intervention and provide proof-of-
`concept for clinical progression. Although there is no
`gold standard model of MS, experimental autoimmune/
`allergic encephalomyelitis (EAE) models simulate the
`clinical and pathological hallmarks of MS in various
`guises and can provide the necessary predictive index for
`clinical therapeutic application.6 EAE is primarily in-
`duced by generating T-cell-mediated immunity to CNS
`antigens and is commonly modeled in rodents (mice,
`rats, and guinea pigs). The range of autoantigen prepa-
`rations used to induce EAE range from whole CNS ho-
`mogenate (spinal cord) to purified protein and peptides.
`Myelin basic protein (MBP), proteolipid protein, myelin
`oligodendrocyte glycoprotein (MOG), S100␤, and glial
`fibrillary acidic protein as well as specific peptides from
`respective parent proteins are encephalitogenic in the
`appropriate host, as the major histocompatability com-
`plex (MHC) is one of the major determinants of immune
`
`NeuroRx威, Vol. 2, No. 4, 2005
`
`Page 2 of 12
`
`YEDA EXHIBIT NO. 2003
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`640
`
`DAVID J. VIRLEY
`
`responsiveness and disease susceptibility to these self-
`antigens. The pathogenic autoimmune steps that are
`thought to initiate and amplify tissue damage in EAE and
`MS are described in Figure 1. The key steps are: 1)
`activation of autoreactive CD4⫹ T-cells in the periphery
`to an antigen; 2) transmigration of proinflammatory T-
`cells and monocytes through the blood brain barrier
`(BBB); 3) amplification of local inflammation and acti-
`vation of resident antigen-presenting cells (APCs), such
`as microglia; and 4) destruction of oligodendrocytes,
`myelin sheath, and axons culminating in demyelination
`and axonal pathology. Neurological deficits in rodent
`EAE models are typically manifested in an ascending
`manner, beginning with loss of tail tone and progressing
`to hind limb paralysis, hind and forelimb paralysis, and
`death. However, the clinical course of EAE is greatly
`dependent on the type of CNS antigen used, immuniza-
`tion protocols, species and strain of animal used to in-
`duce disease. For a valid model of EAE to adequately
`mimic the clinical condition of acute or chronic progres-
`sive MS, enduring pathological signatures such as in-
`flammation, gliosis, oligodendrocyte degeneration, de-
`myelination, and axonal loss should be readily observed
`within the brain and spinal cord. A number of EAE
`models possess some but not all these characteristic fea-
`tures, each of which can provide valuable insight into
`target identification and validation for drug discovery in
`MS. Depending on the hypothesis being tested for a
`specific target of interest, the choice of model in the
`appropriate species allows assessment of the target in the
`pathological process and the putative mode of action of
`a therapeutic acting at that specific target. Therefore, a
`number of rodent EAE models can recapitulate different
`phases of the disease process such as a rapidly progress-
`ing acute monophasic disease, a relapsing-remitting clin-
`ical course or a chronic progressive outcome with vary-
`ing degrees of inflammation, gliosis, oligodendrocyte
`degeneration, demyelination, gliosis, and axonal pathol-
`ogy in the CNS.
`The identification of a target antigen that significantly
`contributes to clinical severity, lesion topography and the
`extent of demyelination in animal models of EAE has
`been attributed to MOG. MOG is a quantitatively minor
`myelin protein (less than 0.05% of total myelin proteins),
`with an Ig-like extracellular domain that is expressed in
`abundance on the outer most layer of myelin sheaths,
`which may render it accessible to antibody attack. Au-
`toantibodies against MOG have been shown to enhance
`demyelination in several EAE models and localized to
`disintegrating myelin around axons in lesions of acute
`MS patients on pathological inspection. Furthermore, an-
`ti-MOG antibodies have been demonstrated within the
`peripheral blood and CSF of MS patients, further asso-
`ciating MOG in the pathogenesis of the disease. On the
`weight of this evidence, MOG (35–55 peptide)-induced
`
`NeuroRx威, Vol. 2, No. 4, 2005
`
`EAE in the C57BL/6 mouse is a robust model of EAE
`with a chronic clinical course of disease with accompa-
`nying pathological hallmarks of inflammation, gliosis,
`and demyelination.7 The consistency of disease inci-
`dence and severity is usually maintained with the addi-
`tion of Bordetella pertussis toxin, which is thought to
`open BBB and facilitate the entry of autoreactive T-cells
`primed by MOG.8,9 The clinical and pathological signs
`of MOG-induced EAE are thought to mimic the chronic
`sustained and progressive phase of MS, particularly rel-
`evant to secondary progressive and primary progressive
`clinical courses of MS.
`Due to the majority of MS patients presenting relaps-
`ing-remitting symptoms before progressing onto a
`chronic phase, a number of animal models of EAE have
`been designed to simulate the more dynamic clinical and
`pathological features of relapsing-remitting MS. One
`such model using the Biozzi AB/H mouse,10 involves the
`inoculation of homologous spinal cord homogenate (or
`more specifically MOG peptide)11,12 in adjuvant without
`the additional use of Bordetella pertussis toxin, and re-
`producibly induces a chronic relapsing-remitting demy-
`elinating disease. The dynamic chronicity of symptoms
`is expressed as an acute induction of disease (loss of tail
`tone and hindlimb paralysis), followed by reduced sever-
`ity (remission) and then a relapse disease episode. The
`development of clinical signs in this model are preceded
`by a loss in weight, whereas remission periods are asso-
`ciated with an increase in body weight,
`implicating
`changes in weight as surrogate markers of disease status.
`Reductions in the degree of inflammation and evidence
`for remyelination are thought to reflect the remission
`period in this EAE model, whereas relapses are thought
`to be indicative of an amplified inflammatory response,
`gliosis, and demyelination within the CNS.
`A key challenge for investigators using rodent EAE
`models in preclinical drug development for MS is the
`assessment of neurological deficits in a more sensitive,
`objective, and quantifiable manner as opposed to the
`more traditional, qualitative clinical-grading scales.
`More specific functional measures assessed in rodent
`EAE models over time, such as hindlimb sensorimotor
`behavior13 and fine motor coordination, may provide a
`more powerful and sensitive means in extrapolating
`more closely to the clinical situation (such as the MSFC
`outcome measure) and provide a more comprehensive
`assessment of novel therapeutics targeted for MS.
`The utility of nonhuman primate EAE models has
`provided an improved insight into CNS autoimmunity
`and ensuing pathology due to their close evolutionary
`relationship with humans.14 Nonhuman primate models
`of EAE have advantages over rodent models in that they
`simulate more closely the relapsing-remitting and pro-
`gressive course of disease and have a more sophisticated
`neuroanatomy, with a greater ratio of white to gray mat-
`
`Page 3 of 12
`
`YEDA EXHIBIT NO. 2003
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`FUTURE THERAPEUTIC CHALLENGES IN MS
`
`641
`
`ter, similar to humans. Furthermore, monkeys are out-
`bred in nature, unlike rodents which are inbred, making
`the individual response to EAE more variable, similar to
`human MS. The use of the common marmoset (Callithrix
`jacchus), a small new-world monkey, allows a practical
`and more sophisticated functional and pathological anal-
`ysis of EAE disease progression, as well as providing
`essential middle-ground for the development of novel
`putative therapeutic agents from rodent models to human
`clinical trials. The incidence of EAE in marmosets im-
`munized with whole myelin, myelin proteins (MOG),
`recombinant human MOG 1-125 or specifically MOG
`14-36 peptide in adjuvant is 100%, with clinical signs
`following a relapsing-remitting or chronic progressive
`course.14 The pathological hallmarks relating to large
`foci of demyelination surrounding perivascular infiltrates
`(inflammation, gliosis, and remyelination) can be readily
`visualized by serial in vivo MRI in this animal species,
`providing valuable pathological correlates to human
`MS.15 Clinical signs are usually preceded by weight loss,
`and include motor weakness, visual defects and paralysis
`usually scored on a qualitative grading scale. However, a
`thorough objective characterization of quantitative func-
`tional deficits, particularly locomotor activity, fine-motor
`movement, visuo-spatial neglect and cognitive function,
`has yet to be interrogated in the marmoset. The EAE
`model in the marmoset may bridge the gap for novel
`therapeutic strategies being progressed for clinical trials,
`such as humanized antibody approaches (e.g., CD40),16
`and provide definitive MRI surrogate markers of disease
`activity and treatment response to help guide phase II
`proof-of-principle clinical trials.
`
`CURRENT THERAPIES FOR MS
`
`The treatment of MS is still in its infancy with limited
`therapeutic options, where the main-stay therapies in-
`volve the utility of corticosteroid and immunosuppres-
`sive interventions. There are currently only five Food and
`Drug Administration (FDA)-approved treatments for re-
`lapsing-remitting MS: two interferon (IFN)-␤1a agents
`(Avonex and Rebif), one IFN-␤1b (Betaseron), glati-
`ramer acetate (GA) (Copaxone) and Mitoxanthrone (No-
`vantrone). For patients with secondary progressive MS,
`cyclophosphamide (Cytoxan) and mitoxanthrone17 are
`prescribed, although provide only modest benefit with
`significant
`toxicity. There are currently no available
`treatment options for primary progressive MS. The main
`therapeutic options for patients with MS will now be
`discussed (i.e., corticosteroids, IFN␤, and GA).
`
`Corticosteroids
`Corticosteroid treatment is extensively used in MS for
`promoting a hastened recovery following a period of an
`acute attack.18 High-dose methylprednisolone, via the
`
`intravenous route, is now more popular than oral pred-
`nisone, as it provides a stable therapy for MS patients at
`the onset of an acute relapse. Although short-term ther-
`apy has shown benefit to varying degrees, long-term
`administration is more useful in the treatment and man-
`agement of relapsing-remitting MS patients. Dramatic
`improvement in the clinical course of secondary progres-
`sive MS has not been shown with corticosteroid treat-
`ment. Although the mechanism of action of corticoste-
`roids in MS is not completely understood, evidence from
`preclinical research has highlighted a number of putative
`mechanisms: reduction in BBB disruption, an inhibition
`of the Th1 immune response, a dampening of T-cell
`migration and the response to antigens, suppression in
`the expression of adhesion molecules, and protection of
`oligodendrocytes from cytokine-induced cell death. EAE
`models have confirmed suppressive actions of cortico-
`steroid treatment on the clinical course of disease and the
`use of the anti-glucocorticoid, RU 38486 (mifepristone),
`has been shown to intensify and reverse steroid-induced
`inhibition of disease.19,20 However, the side effects of
`corticosteroid treatment should not be underestimated.18
`Short-term treatment can induce transient changes in
`mood, headache, gastrointestinal pain, and myalgias.
`Chronic treatment may decrease bone density, leading to
`osteoporosis with risk of fractures, and infections making
`the suspension of treatment more appropriate for man-
`agement of the patient.
`
`IFN-␤
`The IFN-␤-based therapies have been established after
`25 years of clinical development. The original rationale
`for exploring the effects of IFNs in MS was based on the
`premise that MS was thought to be a virally mediated
`disease. However, this antiviral hypothesis was untena-
`ble based on a clinical trial assessing IFN␥where clinical
`symptoms worsened, suggesting that IFN␥ played a role
`in the pathological process of MS. IFN␤, like other IFNs,
`is a species-specific glycoprotein that has numerous bi-
`ological properties. Although its mechanism of action is
`still poorly understood, immunomodulatory as opposed
`to antiviral and antiproliferative effects seem to predom-
`inate. IFN␤-1a is identical to the natural IFN-␤, whereas
`IFN␤-1b differs by two amino acids and is not glycosy-
`lated. Irrespective of these subtle structural differences
`IFN␤-1b shows similar biological activity to IFN␤-1a.
`The putative effects of IFN␤ on MS progression primar-
`ily relate to antiinflammatory effects: dampening the
`IFN␥,
`stimulatory effects of
`tumor necrosis factor
`(TNF)␣, interleukin (IL)-12, and lymphotoxin secretion;
`inhibiting monocyte activation; preventing the disruption
`of the BBB and thereby reducing the entry of lympho-
`cytes into the CNS; reducing antigen presentation to
`T-cells; and up-regulation of anti-inflammatory cyto-
`kines such as TGF␤ and IL-10. EAE models have dem-
`
`NeuroRx威, Vol. 2, No. 4, 2005
`
`Page 4 of 12
`
`YEDA EXHIBIT NO. 2003
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`642
`
`DAVID J. VIRLEY
`
`onstrated that IFN␤ reduces the progression of disease,
`delays the exacerbation onset and rate,21 and may mod-
`ulate the IL10/IL-12 circuit reducing the effect of epitope
`spreading and disease severity.22
`In the clinical trials that have been conducted with
`both IFN␤ formulations, the key efficacy findings relate
`to: one third reduction in relapse rate at higher doses;
`rapid onset of effect, within 1 year for relapse rate and
`within a few weeks for MRI disease activity; dispropor-
`tionately large effects on inflammation as measured by
`MRI activity; slowing of the accumulation of MRI bur-
`den of disease; and a tendency for a reduction in the
`number of patients with observed progression of disabil-
`ity.23–25 However, adverse effects are associated with
`IFN␤ therapy, such as flu-like symptoms and injection
`site reactions. Discontinuation of IFN␤ therapy is war-
`ranted where patients show no improvement over a
`6-month period, and where disability progresses or more
`relapses occur with three or more courses of corticoste-
`roids over a 1-year period. Additional concerns over
`severe depression or suicidal ideation, drug toxicity, and
`noncompliance highlight the need for alternative classes
`of drug with a better therapeutic index. Issues concerning
`the effects of neutralizing antibodies on IFN␤ efficacy
`also need to be elucidated in relation to the potential
`long-term complications for MS patients on IFN␤
`treatment.
`
`GA
`GA is non-IFN, nonsteroidal therapy that constitutes a
`mixture of synthetic random base copolymers of four
`amino acids (alanine, glutamic acid,
`lysine, and ty-
`rosine), in a highly specific molar ratio. Original research
`investigated the potential encephalitogenic role of GA in
`animal models of EAE, but unexpectedly GA suppressed
`the acute and chronic clinical and pathological hallmarks
`of EAE in a number of animal species.26 These effects
`translated into clinical benefit, in that an initial phase II
`trial demonstrated GA to reduce relapse rates by 76% in
`relapsing-remitting MS patients.27 Further clinical devel-
`opment confirmed reductions in relapse rates by a third
`and a higher preponderance of patients relapse-free.28
`These effects were confirmed on follow-ups for more
`than 5 years on treatment and demonstrated sustained
`efficacy for GA in slowing the progression of disability.
`Lesion burden assessed by MRI has shown a beneficial
`profile for GA in relapsing-remitting patients, in that
`treatment reduced the frequency of new enhancing le-
`sions and lesion load compared to baseline pretreatment
`measures.29 However, no significant improvement in the
`course of the disease with GA has been demonstrated for
`secondary progressive MS patients.26 A number of
`mechanisms have been proposed related to its biological
`activity in relapsing-remitting MS: induction of antigen-
`specific suppressor T cells and competitive inhibition of
`
`NeuroRx威, Vol. 2, No. 4, 2005
`
`MBP and related-peptides from antigen-presenting cells.
`Although GA is well tolerated in MS patients, adminis-
`tration by the subcutaneous route induces localized in-
`jection site reactions in the majority. Generally, it is
`viewed that GA has the most favorable adverse effect
`profile in that there is a reduced propensity to develop
`depression, menstrual disorders, neutralizing antibodies
`compared with the other therapeutic options available for
`MS. However, there is clear need to develop more im-
`proved treatment options for MS patients, which offer
`sustained relief with greater efficacy without associated
`risks. This poses a huge challenge for the pharmaceutical
`and biotechnology industry. A number of alternative dis-
`ease-modifying strategies will now be presented each of
`which exert different modes of action and target different
`phases of the disease process.
`
`EXAMPLES OF NOVEL THERAPEUTIC
`CHALLENGES FOR MS
`
`Blockade of lymphocyte migration
`Very late antigen-4: natalizumab (Tysabri) and
`small molecule antagonists. A key step in the early
`phase of EAE and MS is the binding of leukocytes to the
`vascular endothelium of the BBB, before their penetra-
`tion through it by diapedesis to enter the brain paren-
`chyma (FIG. 1). A substantial body of evidence has now
`been accumulated that implicates very late antigen-4
`(VLA-4, ␣4-␤1 integrin) in this process, via its interac-
`tion with receptors such as vascular cell adhesion mole-
`cule 1 (VCAM-1) and the CS1 domain of fibronec-
`tin.30,31 For example, surface expression of VLA-4 has
`been shown to be essential for the entry of T-cell clones
`into the brain,32 and in a number of different EAE mod-
`els treatment with anti-VLA-4 monoclonal antibodies
`has been effective in suppressing the clinical signs of
`disease and T cell infiltration into the CNS.32–35 Peptide
`blockers of VLA-4 have likewise been shown to be
`effective in EAE,36 preventing the development of clin-
`ical signs and cellular infiltration.37 Direct in vivo evi-
`dence has suggested that VLA-4 may be important not
`only in the capture and adhesion of T cells to microvas-
`cular endothelium through interaction with VCAM-1,38
`but also in facilitation of T-cell entry into the brain
`parenchyma (by the induction of metalloproteinase-2)
`and in maintenance of the residency of T cells within the
`CNS.39 However, caution should be exercised based on
`the preclinical EAE relapsing-remitting model data gen-
`erated with the PS/2 VLA-4 antibody, demonstrating
`that, although prophylactic administration suppressed
`onset and severity of EAE, therapeutic administration at
`the peak of acute disease or during remission exacer-
`bated disease relapses and increased the accumulation of
`CD4⫹ T cells and VCAM-1 expression in the CNS.35
`The concerns highlighted by the authors35 were that the
`
`Page 5 of 12
`
`YEDA EXHIBIT NO. 2003
`MYLAN PHARM. v YEDA
`IPR2017-00195
`
`

`

`FUTURE THERAPEUTIC CHALLENGES IN MS
`
`643
`
`PS/2 VLA-4 antibody could have enhanced costimula-
`tory signals or amplified signals by increased adhesion in
`the CNS, ultimately augmenting immune responses and
`clinical severity.
`Based on these observations, alternative strategies
`have been adopted to develop a therapy for MS that is not
`antibody based. Piraino et al.40 recently reported reversal
`of chronic EAE with a small molecule inhibitor of ␣-4
`integrin (CT301), and Cannella et al.41 showed efficacy
`of a synthetic nonpeptide VLA-4 antagonist (TBC 3486)
`in the acute phase of EAE, but both these agents still had
`the disadvantage of requiring administration via an in-
`jectable route. Furthermore, recent research on highly
`potent and selective ␣4␤1 integrin small molecular
`weight inhibitors, BIO519242 and 2a-PEG (polyethylene
`glycol modification of BIO5192),43 has demonstrated
`them to be effective in suppressing EAE with improved
`pharmacokinetic and pharmacodynamic characteristics.
`Although these compounds were administered via the
`subcutaneous and intravenous routes, respectively, the
`need for less frequent dosing with 2a-PEG at a 30 fold-
`lower dose may provide a reasonable delivery paradigm
`for therapeutic intervention.43 However,
`the ultimate
`goal for the future development of small molecule ␣4␤1
`integrin antagonists is oral delivery with improved phar-
`macological properties that translate into safe and effec-
`tive interventions for MS.
`The clinical significance of VLA-4 in multiple sclero-
`sis has been demonstrated by a 3- to 4-fold increase in its
`expression on CSF and blood lymphocytes of patients
`compared with healthy controls.44 Encouragingly, clini-
`cal treatment with the humanized anti-VLA-4 monoclo-
`nal natalizumab (Tysabri), raised against human ␣ 4 ␤1
`integrin, significantly reduced the number of new active
`enhancing lesions seen by MRI over 6 months by about
`90%.45 A reduction in lesions was seen in both relapsing
`and progressive patients, and the therapy also reduced
`the number of relapses by 50%.45 Although a previous
`smaller study suggested an increased incidence of re-
`lapse following the clearance of the antibody from the
`circulation,46 the subsequent data suggested that there
`was no rebound effect following drug withdrawal.45
`These phase II clinical trials complemented the preclin-
`ical findings with the murine form of the antibody
`(AN100266m) before humanization, in that the murine
`form was a potent inhibitor of in vitro interactions be-
`tween ␣4␤1 integrin and VCAM-1, and it suppressed
`and reversed rodent EAE.47 Further clinical trials with
`natalizumab, involving two randomized, placebo-con-
`trolled, double-blind phase III studies were conducted in
`MS patients who had experienced one relapse in the year
`before trial enrolment.48 The first study demonstrated
`that after 1 year of treatment with natalizumab mono-
`therapy, the percentage of patients relapse-free was 76%
`compared to 53% in the placebo group. When assessing
`
`the MRI end-points 96% patients on natalizumab treat-
`ment had no enhancing lesions, whereas 68% of patients
`on placebo had no visible enhancing lesions. The second
`study showed that patients that had experienced one re-
`lapse on IFN ␤-1a (Avonex) before trial enrolment were
`less likely to relapse on natalizumab with Avonex, com-
`pared to Avonex with placebo treatment after a study
`duration of 1 year, i.e., a 54% reduction in relapse rate.
`The MRI data also revealed a positive add-on benefit of
`natalizumab with Avonex, relative to Avonex and pla-
`cebo, in that 96% of patients on the natalizumab/Avonex
`arm showed no enhancing lesions compared to 76% pa-
`tients on the Avonex/placebo arm. These results from the
`1-year interim analysis were later confirmed after the full
`2-year study duration by Elan (Dublin, Ireland) and Bio-
`gen IDEC (Cambridge, MA), in that natalizumab re-
`duced the risk of disability progression by 42% com-
`pared with patients on placebo.48
`Based on the encouraging results from the 1-year in-
`terim analysis, the FDA approved natalizumab for the
`treatment of relapsing-remitting MS. However, on 28
`February 2005, 3 months after FDA approval, Biogen
`IDEC and Elan voluntarily suspended all ongoing clini-
`cal trials and marketing of natalizumab.48 The reasons
`for the suspension were that two patients developed pro-
`gressive multifocal leukoencephalopathy (PML), one fa-
`tal, both receiving natalizumab and Avonex for 2 years.49
`A month later, a third and fatal case of PML was dis-
`closed in patient enrolled in a clinical trial for natali-
`zumab in Crohn’s disease.48 This information suggested
`that natalizumab was associated with PML, irrespective
`of its combination with Avonex and the MS patient
`population. Although speculation on the likely causes of
`PML in MS patients in combination with Avonex was
`fuelled by possible activation of the polyoma JC virus
`interacting with the antibody for ␣4␤1 integrin, this hy-
`pothesis was not unequivocal based on the case reported
`in the Crohn’s disease trial.48 Nevertheless, natalizumab
`has provided a clear proof-of principle for the ␣4␤1
`integrin target in MS, as preclinical efficacy signals dem-
`onstrated in EAE models ha