Journal of Autoimmunity and Cell Responses
`ISSN 2054-989X | Volume 1 | Article 3
` Review
`
`
`
` Open Access
`Glatiramer acetate and therapeutic peptide vaccines
`for multiple sclerosis
`
`Jill Conner
`
`Correspondence: Jill.Conner@tevapharm.com
`
`CrossMark
`← Click for updates
`
`Global Specialty Medicines, Teva Pharmaceutical Industries, Ltd., Overland Park, KS, USA.
`
`Abstract
`Therapeutic vaccines are antigen-specific agents that inhibit an unwanted immune response to prevent
`progression of, or eliminate, existing disease. Unlike prophylactic vaccines, developing therapeutic
`vaccines poses a significant challenge because they can require multimodal immunomodulatory activity
`to modulate complex pathogenic disease pathways. This is particularly true in autoimmune disorders,
`including multiple sclerosis, which has complex pathogenesis that, although not fully understood, appears
`to involve dysfunction of both innate and adaptive immune processes. The glatiramoid, glatiramer acetate
`(Copaxone®), a nonbiologic, complex, heterogenous mixture of synthetic polypeptides, is currently
`the only approved therapeutic peptide for treatment of multiple sclerosis. Glatiramer acetate has an
`enormous number of potentially active epitopes (estimated to be ~1030) in the polypeptides mixture. The
`epitopes in glatiramer acetate have not been identified, but they appear to act as altered peptide ligands
`of encephalitogenic epitopes within myelin basic protein, a suspected autoantigen implicated in multiple
`sclerosis. Peptide epitopes in glatiramer acetate compete with autoantigens for binding with major
`histocompatability complex molecules on antigen-presenting cells, thereby altering the functional outcome
`of T cell signaling from inflammatory to anti-inflammatory responses. Other therapeutic vaccines designed
`to more selectively compete with myelin antigens for receptor binding have been shown effective in animal
`models of multiple sclerosis but toxic in human patients in clinical trials. The partially random structure
`of glatiramer acetate and potentially huge number of antigenic sequences may be integral to safety and
`efficacy by influencing multifactorial immune processes and surmounting challenges related to inter- and
`intra-individual heterogeneity of T cell responses and the phenomenon of epitope spreading. Follow-on
`generic versions of glatiramer acetate have been made available to multiple sclerosis patients outside the
`United States. Analyses of these glatiramoids indicate they have different biological and immunological
`activity from that of Copaxone®, illustrating the difficulty of replicating complex glatiramoids and of
`developing safe, effective peptide vaccines in general. Reviewed here are some of the major mechanisms of
`glatiramer acetate activity on innate and adaptive immune pathology and considerations for development
`of future therapeutic peptide vaccines for multiple sclerosis.
`keywords: Autoimmune disease, glatiramer acetate, glatiramoid, multiple sclerosis, peptide, copolymer,
`therapeutic, vaccine
`
`Introduction
`Most vaccines are used for prophylaxis of disease; however,
`therapeutic vaccination involves the use of an antigen-specific
`intervention to inhibit an unwanted immune response,
`with the goal of preventing progression of, or eliminating,
`existing disease [1]. Effective therapeutic vaccines are less
`common than preventive vaccines and are more challenging
`to develop because preventive vaccines typically have
`immunologic specificity for an individual infective agent (e.g.,
`
`viral poliomyelitis, smallpox). In contrast, therapeutic vaccines
`may require multimodal immunomodulatory activity to modu-
`late complex pathogenic disease mechanisms already underway.
`Suppression of harmful inflammatory immune responses by
`antigenic peptide vaccination requires induction of tolerance,
`which can be attained by repeated dosing [2].
`There is great interest in developing therapeutic peptide vac-
`cines for treatment of autoimmune diseases such as multiple
`sclerosis (MS) [3-8]. Synthetic peptides, in principle, could be
`
`© 2014 Jill Conner; licensee Herbert Publications Ltd. This is an Open Access article distributed under the terms of Creative Commons Attribution License
`(http://creativecommons.org/licenses/by/3.0). This permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
`
`Page 1 of 11
`
`YEDA EXHIBIT NO. 2007
`MYLAN PHARM. v YEDA
`IPR2014-00644
`
`

`
`designed to present selected epitopes to immune cells to
`elicit a desired response. MS is a chronic, progressive,
`neurodegenerative autoimmune disease of the central nervous
`system (CNS) associated with demyelination and axonal
`damage. MS is a disease with no identified specific pathogenic
`target for prophylaxis. The etiology of MS is unknown, and
`pathogenesis is complex and not fully understood, but appears
`to include pathologic mechanisms of innate and adaptive
`immunity working together to exacerbate disease (Figure 1)
`[9-11]. Therefore, an effective vaccine will optimally modulate
`both innate and adaptive immune responses. Currently, the
`glatiramoid, glatiramer acetate (GA, Copaxone®), a non-biologic
`complex drug (NBCD), is the first and only approved therapeutic
`peptide for treatment of MS [1,2,12]. Copaxone® and the
`interferon beta (IFNβ) drugs (protein products produced by
`recombinant DNA techniques, including Avonex®, Rebif®, and
`Betaseron®) were the first disease-modifying therapies (DMTs)
`approved for treatment of MS. The IFNβ drugs have profound
`activities on several components of the process required for
`the migration of inflammatory cells into the CNS. These agents
`have been shown to reduce relapse rate, delay progression
`of neurologic disability, and decrease brain lesions on MRI in
`patients with relapsing-remitting MS (RRMS) [13-16]. Unlike
`GA activity (described below), a primary mechanism of the
`IFNβ drugs is reduction of inflammatory cell trafficking into
`the CNS [10,15,16]. More recently, several oral MS treatments
`have become available, including Aubagio® (teriflunomide),
`Tecfidera® (dimethyl fumarate), and Gilenya® (fingolimod).
`That GA is the only therapeutic peptide approved for MS
`reflects the complexity of the disease and of GA, and the
`challenges of preparing safe and effective therapeutic peptide
`vaccines. Reviewed here are some of the key mechanisms of
`GA activity on innate and adaptive immunity that contribute
`to its efficacy and safety, and considerations for development
`of follow-on versions of GA and new peptide vaccines for MS.
`
`Review
`Glatiramer acetate
`Discovery
`The immunomodulatory benefits of GA were discovered
`serendipitously. In the 1970s, researchers at the Weizmann
`Institute in Israel synthesized a series of amino acid copolymers
`in hopes of developing synthetic antigens that mimic
`myelin basic protein (MBP), the main protein component
`of the lipid-rich myelin sheath and a major autoantigen in
`MS. The copolymers were designed to reproducibly induce
`experimental autoimmune encephalomyelitis (EAE), an
`animal model of MS. However, rather than induce EAE, GA
`immunization proved to be protective against EAE induction
`and reduced symptoms of established EAE in several animal
`species [17-20].
`
`Chemical nature
`GA is a complex, heterogenous, partially random mixture of
`
`Innate
`
`Immature
`dendritic
`cell
`
`TGFβ
`Adaptive
`
`Antigen processing and
`uptake onto MHC class II
`IL-4
`IL-10
`
`IL-4
`IL-10
`
`Mature
`dendritic
`cell
`TGFβ
`IL-6
`
`IL-12
`
`IL-1β
`
`Naive T cell
`
`Naive T cell
`
`TReg cell
`
`Macrophage
`
`TH1 cell IFN-γ
`TNF
`
`B cell
`
`CTL/CD8+ T cell
`
`Innate
`
`Astrocyte
`
`Microglia macrophage
`Reactivation
`
`TH17 cell
`
`IL-17
`
`Blood–brain
`barrier
`
`Central
`nervous
`system
`
`Activation
`TH17 cell
`
`→
`Glutamate
`
`NO
`Neuron
`
`IL-17
`IFN-γ
`TNF
`
`IFN-γ
`TNF
`IFN-γ
`
`TH1 cell
`MMPs
`NO/
`O2•
`
`B cell
`
`CTL/CD8+ T cell
`Plasma
`cell
`
`Autoantibodies
`
`Figure 1. Multiple sclerosis pathogenesis involves innate
`and adaptive immune dysfunction. Reprinted by permission
`from Macmillan Publishers Ltd: Nat Rev Drug Discov.
`Nov;9(11):883-97, copyright 2010 [115].
`
`synthetic polypeptides containing four naturally occurring
`amino acids: L-glutamic acid, L-alanine, L-lysine, and L-tyrosine
`[21]. The polypeptides chains in GA vary in length from 20
`to 200 amino acid residues, with an average of approximately
`60 residues, and molecules in the colloidal GA sus-pension
`range in size from 1.5 nm to 550 nm, making GA one of the
`first true nanomedicines [22,23]. The average molecular
`weight of the polypeptides in GA is 7000-9000 daltons. GA
`has an enormous number of potentially active epitopes
`(estimated to be ~1030 [24]) in the polypeptides mixture, which
`may be integral to drug efficacy and safety [25]. However, the
`complexity of GA makes isolation and identification of the
`active epitopes in the GA polypeptides mixture impossible.
`Nevertheless, it is possible to differentiate among glatiramoids
`(i.e., synthetic copolymer mixtures comprising the four amino
`acids in GA in a defined molar ratio [22]) in analytical and bio-
`logical tests (as described below).
`
`Clinical usage
`Copaxone® is indicated for the treatment of patients with
`relapsing forms of MS [21]. Copaxone® is currently available
`as a daily 20 mg SC injection (approved in the US in 1996)
`and recently approved in the US in January 2014 as a 40 mg
`SC injection administered three times per week.
`
`GA mechanisms of action
`Numerous effects of GA on cellular and humoral immune
`cells have been identified over the last 4 decades, though
`the complete activity of GA remains unknown and new
`mechanisms continue to be discovered (Figure 2) [22,25,26].
`Perhaps the best-known therapeutic mechanisms of GA
`activity involve its effect on adaptive immune cells (T and
`
`2
`
`Jill Conner, Journal of Autoimmunity and Cell Responses 2014,
`http://www.hoajonline.com/journals/pdf/2054-989X-1-3.pdf
`
`doi: 10.7243/2054-989X-1-3
`
`Page 2 of 11
`
`YEDA EXHIBIT NO. 2007
`MYLAN PHARM. v YEDA
`IPR2014-00644
`
`

`
`B cell
`
`Induction of GA-specific antibodies,
`especially GA-specific lgG4 antibodies
`
`T cell
`
`Induction of GA-specific Th2 CD4+ cells
`(Th1→Th2shift of CD4+ T cells)
`Decrease of GA-specific CD4+ T cell proliferation
`Induction of GA-specific CD8+ suppressor T cell
`Induction of GA-specific CD4+CD25+ regulatory
`T cells
`
`APC
`
`More Th2-like differentiation of APC
`Modulation of monocytic cell function
`(e.g., decrease of TNF-α-producing cells)
`Modulation of dendritic cell function
`e.g., restoration of pDC phenotype and function)
`Figure 2. Glatiramer acetate modulates innate and adaptive
`immune cells. Reprinted from Int Rev Neurobiol Vol. 79,
`Ziemssen and Schrempf, Glatiramer acetate: mechanisms
`of action in multiple sclerosis, pp. 537-570 (2007), with
`permission from Elsevier [20].
`
`B lymphocytes). More recently, GA effects on cells of the
`innate immune system (e.g., antigen-presenting cells
`[APCs], natural killer [NK] cells) have been discovered.
`
`GA effects on innate immune cells
`Antigen-presenting cells: The main functions of APCs are
`phagocytosis, antigen presentation, and cytokine production.
`In the peripheral immune compartment, APCs provide three
`sequential signals to activate antigen-specific T cells:
`1) Toll-like receptors (TLRs) on activated APCs recognize
` bacteria, viruses, and other antigens [27,28]. Upon
` phagocytosis, antigenic peptide fragments bound to
` major histocompatibility complex (MHC) molecules on
` the surface of APCs facilitate T cell recognition of the
` cognate antigen by T cell receptors (TCR) on CD4+ cells
` (antigen presentation on MHC class II molecules) or
` CD8+ cytotoxic T cells (antigen presentation by MHC
` class I molecules);
`2) Costimulatory molecules on APCs activate the T cells to
` initiate proliferation and differentiation; and
`3) Cytokine secretion by APCs polarizes differentiation of
` activated CD4+ T cells into different effector (T helper
` Th; e.g., Th1, Th2, Th3, Th17) and regulatory (e.g.,
` CD4+CD25+FoxP3+) T cell subtypes; influences
` cytotoxic CD8+ T cell phenotypes [29]; and instructs
` plasma cell differentiation.
`In MS, peptide epitopes from putative MS autoantigens, in-
`cluding MBP, myelin oligodendrocyte glycoprotein (MOG),
`and proteolipid protein (PLP) molecules, bind to MHC class II
`molecules on APCs. Upon T cell recognition, pro-inflammatory
`(Type I) cytokines are expressed and stimulate differentiation
`and proliferation of pathogenic autoreactive CD4+ Th1 cells
`[10,30]. Reciprocally, APCs exposed to pro-inflammatory T
`cells in vitro tend to express a range of adhesion molecules,
`costimulatory molecules, and cytokines favoring further T
`
`cell differentiation of pro-inflammatory Th1 cells. Activated
`myelin-reactive Th1 cells in the periphery move from the
`systemic circulation across the blood-brain barrier (BBB) and
`enter the CNS [12]. Once Th1 cells establish an inflammatory
`milieu there, pro-inflammatory Th17 cells can enter the CNS
`[31]. Th17 cells play an important role in the pathogenesis of
`inflammatory and autoimmune diseases [32]. Once in the CNS,
`antigenic peptide fragments bound to dendritic cells (DCs),
`microglia, and astrocytes reactivate autoreactive T cells, leading
`to pro-inflammatory cytokine release and myelin damage.
`Circulating APCs of the myelomonocytic lineage, i.e., mono-
`cytes/macrophages, and DCs, may be the primary targets
`of GA immunomodulation. GA activity on monocytes, DCs,
`and microglia is antigen-nonspecific, which may help explain
`why GA has shown efficacy in animal models of several in-
`flammatory and neurodegenerative conditions [33-39].
`GA polypeptides are promiscuous binders to MHC class II
`molecules on APCs, with or without antigen processing [40].
`GA peptides compete with myelin peptides for binding to
`MHC class II molecules on APCs and can preferentially displace
`peptides from MBP [41], MOG [42], and PLP [43] from the MHC
`binding site, but cannot be displaced by them [44]. As a result,
`differentiation of autoaggressive Th1 cells is reduced and GA
`mediates a shift in T cell phenotypes from Th1 to Th2/3 cells.
`Monocytes cultured in anti-inflammatory supernatants push
`T cells toward Th2 differentiation in vitro, which is associated
`with secretion of anti-inflammatory (Type II) cytokines [45].
`GA induces Type II monocyte and microglia differentiation
`in EAE models and in MS patients [45-48]. Adoptive transfer
`of monocytes from GA-treated mice to GA-naïve mice
`with EAE directs T cell differentiation toward Th2 cells and
`CD4+CD25+FoxP3+ Tregs, an important subclass of regulatory
`cells that attenuate autoreactive T cell responses [47].
`Continued GA treatment reduces the ability of APCs to
`present antigen or to respond to various stimuli [45-49]. In
`mice with EAE, GA reduced expression of TLRs on DCs [50],
`and down-regulated osteopontin, a protein implicated in
`chronic inflammatory diseases that induces DC maturation
`toward the Type I inflammatory phenotype. In another study,
`GA inhibited monocyte activation and production of pro-
`inflammatory tumor necrosis factor alpha (TNFα) by human
`DCs in vitro [48], and compared with untreated patients,
`DCs and monocytes from MS patients who received GA for
`1 year showed less activation, accompanied by reduced risk
`of relapse [51].
`The ability of APCs to penetrate the BBB is essential to the
`establishment of autoimmune inflammatory CNS diseases.
`Suppressed expression of chemokines with CNS chemotactic
`properties may also contribute to GA therapeutic activity
`[50]. Macrophage inflammatory protein (MIP)-1 increases
`DC transmigration across endothelial cells and is elevated
`in mice with EAE compared with control mice. Daily GA trea-
`tment suppressed MIP-1α and MIP-1β expression in mice with
`EAE, and also decreased expression of RANTES [50,52], a poly-
`
`3
`
`Jill Conner, Journal of Autoimmunity and Cell Responses 2014,
`http://www.hoajonline.com/journals/pdf/2054-989X-1-3.pdf
`
`doi: 10.7243/2054-989X-1-3
`
`Page 3 of 11
`
`YEDA EXHIBIT NO. 2007
`MYLAN PHARM. v YEDA
`IPR2014-00644
`
`

`
`doi: 10.7243/2054-989X-1-3
`
`peptide with strong chemoattractant activity for T cells, mon-
`ocytes and macrophages associated with pathogenesis of MS
`lesions in the CNS [52].
`Attesting to the primacy of GA effects on APCs, researchers
`showed that GA did not influence T cell polarization when
`added to naïve Th cells activated in an APC-free system. More-
`over, induction of Type II monocytes by GA does not require
`reciprocal signalling by T cells, [53]. In STAT6-deficient mice,
`which cannot generate interleukin (IL)-4-producing Th2 cells,
`monocytes from GA-treated mice exhibited increased Type II
`anti-inflammatory cytokines, IL-10, and transforming growth
`factor beta (TGFβ) [47]. Similarly, monocytes from GA-treated
`RAG-1 mice, which lack mature T and B cells, also showed an
`anti-inflammatory cytokine pattern [47].
`Natural killer cells: NK cells are lymphocytes of the innate
`immune system responsible for immunosurveillance and
`regulating immune responses [54]. NK cells limit damage in
`MS by inhibiting autoreactive T cell responses, and increased
`frequency or functional competence of circulating NK cells
`is associated with decreased disease activity in MS patients
`[55]. NK cells exposed to GA kill both immature and mature
`DCs, impairing the ability of mature DCs to present antigens
`to autoreactive T cells [54,56,57]. This mechanism may be
`involved in GA effects on inhibiting graft-vs-host (GvH) disease
`in laboratory animals [36,58].
`
`GA effect on adaptive immune cells
`CD4+ T cells: Therapeutic effects of GA have been attributed
`in part to the ability of antigenic sequences in hydrolyzed
`GA peptides to act as altered peptide ligands (APLs) of
`encephalitogenic epitopes in MBP [1,12,59,60]. An APL serves
`as a receptor ligand that contains a substitute of one or more
`amino acids of the native ligand to change the functional
`outcome of TCR signaling [61]. An APL must be close enough in
`chemical composition to the native peptide antigen to trigger
`an immune response that is cross-reactive with the infectious
`agent or autoantigen, but must itself be biologically harmless.
`Accordingly, GA is cross-reactive with myelin antigens at both
`the humoral and cellular levels, without being encephalitogenic
`[62-67]. Repeated GA immunization modifies the GA-reactive
`T cell repertoire by skewing them from the Th1 phenotype
`toward the Th2 phenotype [59,66,68,69]. GA-reactive Th2
`cells release anti-inflammatory cytokines and neurotrophic
`factors [69-71] and suppress neighboring auto-aggressive Th1
`cells by means of “bystander suppression” in the CNS [68,69].
`GA treatment also increases the number and suppressive
`capacity of CD4+CD25+FoxP3+ Tregs in MS patients [72,73].
`Tregs dampen autoimmune responses, and are functionally
`impaired in MS patients [74].
`CD8+ T cells: Untreated patients with MS initially show
`low GA-specific CD8+ cell responses compared with healthy
`controls [75], but GA therapy produces proliferative responses
`in CD8+ T cells [75,76]. Adoptive transfer of GA-induced CD8+
`T cells results in amelioration of EAE [77]. GA-reactive CD8+
`
`T cells appear to regulate proliferation of myelin-reactive
`CD4+ cells [75,78]. Over time, continued treatment with GA
`results in decreasing numbers of GA-reactive CD4+ T cells
`while the number of GA-reactive CD8+ T cells increases [79].
`It has been proposed that GA effects on CD8+ T cells may be
`an indispensable component of its therapeutic activity [77].
`B cells: B cells can serve as efficient APCs for T cells and can
`activate autoreactive T cells; likewise, activated T cells can
`trigger B cell activation and formation of autoreactive anti-
`bodies [12,80]. Demyelinating antibodies can travel from the
`systemic circulation across the BBB and into the CNS. Antibody-
`mediated patterns of demyelination are detected in more
`than 50% of MS patients [81].
`All patients treated with GA develop GA-reactive antibodies
`[67,82]. As an antigen-based therapy, by definition anti-GA
`antibodies are not neutralizing and do not appear to interfere
`with MHC class II binding or induction and proliferation of
`suppressor T cell lines and clones [82]. GA treatment produces B
`cell secretion of the Type II cytokine, IL-10, thereby augmenting
`suppression of autoreactive T cells, and promoting induction
`of GA-reactive Th2 cells [83]. GA effects on B cells may also
`be essential to therapeutic activity. Adoptive transfer of B
`cells from GA-treated mice with active EAE inhibited the
`proliferation of autoreactive T cells, whereas GA treatment
`had no effect on EAE in B cell-deficient mice [84]. In a clinical
`study, relapse-free patients tended to develop higher anti-GA
`antibody titers than patients who relapsed [67], suggesting
`beneficial antibody activity. Additionally, remyelination of
`spinal cord axons was promoted by antibodies to GA in mice
`with EAE [85].
`Anti-GA antibody levels peak between 3 and 6 months of
`treatment initiation, and then gradually decline, but remain
`higher than baseline levels [67,82]. Anti-GA antibodies are
`mainly immunoglobulin G1 (IgG1), IgG2, IgG4, and IgA isotypes
`[67,82,86,87]. IgG1 antibody levels remain relatively consistent
`over time. IgG2 antibodies decline with chronic treatment
`[86] and a gradual shift to IgG1 and IgG4 is observed [87]. GA
`was not cross-reactive with MBP when exposed to polyclonal
`antibodies, but cross-reactivity with MBP was evident with
`monoclonal antibodies, in vitro [63]. RRMS patients in clinical
`trials developed GA-reactive antibodies with very low reactivity
`to MBP [67,82].
`
`Selective peptide vaccine attempts
`Given the APL activity of the random GA mixture, selective
`APLs were developed with the intent of creating more potent
`inhibition of autoreactivity. Two APLs of the immunodominant
`peptide region (83–99) of MBP (alanine was substituted for
`lysine at position 91 in both cases) were developed for
`use as therapeutic vaccination for MS. While effective for
`preventing and treating EAE in laboratory animals, they were
`less successful for MS patients, primarily due to toxicity
`problems [88,89]. The selective APL, NBI 5788, was evaluated in
`MS patients randomized to receive once-weekly SC injections
`
`4
`
`Jill Conner, Journal of Autoimmunity and Cell Responses 2014,
`http://www.hoajonline.com/journals/pdf/2054-989X-1-3.pdf
`
`Page 4 of 11
`
`YEDA EXHIBIT NO. 2007
`MYLAN PHARM. v YEDA
`IPR2014-00644
`
`

`
`doi: 10.7243/2054-989X-1-3
`
`of doses ranging from 5 to 50 mg for up to 4 months in a phase
`II clinical trial [88]. The study was stopped prematurely when
`9% of patients developed immediate-type hypersensitivity
`reactions. There was a robust Th2-like immune response in
`NBI 5788-treated patients. Th2 cell responses are associated
`with the promotion of humoral responses, and anti-NBI
`5788 antibody titers were relatively high in patients with
`hypersensitivity reactions, particularly in patients who received
`the highest tested dose. Thus, the strength of the Th2 response
`is an essential consideration in peptide vaccine design; i.e.,
`optimization of the suppression of Th1-related autoimmunity
`must be balanced against the risk of hypersensitivity reactions
`driven by Th2 responses [88].
`Weekly administration of the APL, CGP77116, was assessedin
`another small phase II study [89]. Of 8 patients treated with
`CGP77116, 2 showed improved disease activity, 3 had stable
`disease, and 3 showed worsened disease activity during
`and immediately after treatment. CGP77116 was poorly to-
`lerated and the trial was discontinued before completion. An
`investigation showed activated CGP77116-reactive T cells
`skewed toward the inflammatory Th1 phenotype rather than
`Th2, and on-study relapses were ascribed to CGP77116-induced
`expansion of encephalitogenic T cells specific for MBP(83–99) [89].
`Based on these studies, it was suggested that when relying on
`TCR- or MHC-mediated therapeutic effects (i.e., MHC blockade,
`TCR antagonism or partial agonism, clonal deletion or anergy),
`a single selected APL will be ineffective across individuals with
`different genetic backgrounds, and even intra-individually
`because TCR diversity may preclude a consistent therapeutic
`response [61]. Humans may express up to 8 different class II
`alleles; consequently, there is considerable diversity in the
`number of different antigenic peptides that can be presented
`to T cells among different individuals [61]. Varying autoantigens
`may be important in different individuals and phenotypic T
`cell response to an antigen may be heterogeneous. Further,
`by the time MS becomes clinically evident in humans, it is
`likely that the immune response has spread to more than
`one antigen due to “epitope spreading” [61]. Thus, adverse
`effects seen with NBI 5788 and CGP77116 were likely related
`to the many clonotypes that exist within and among different
`patients against the targeted epitope, MBP(83-99) [90].
`Immune dysregulation at multiple levels appears to cause
`MS, and this may explain an observed resistance to therapeutic
`modalities with high immune selectivity. GA is a pool of
`antigenic peptides of different lengths and composition with
`complex partially random structure and potentially millions
`of antigenic peptides [24,25]. An antigenic epitope in GA for
`a distinct TCR would be present at much lower concentration
`than if the GA structure were defined and not random [61].
`GA administration is thought to lead to anergy of T cells for
`which TCR-binding motifs in GA occur with higher frequency
`and to persistence or expansion of T cells with TCR-binding
`motifs in GA that occur with low frequency [61]. With GA,
`this implies the induction of anergy for Th1/Th0 GA-specific
`
`T cells and persistence/expansion of Th2 GA-specific T cells.
`
`Follow-on glatiramoids-special considerations
`The challenge of creating peptide vaccines is exemplified by
`attempts to create follow-on versions of Copaxone®. GA and
`other NBCDs are heterogenous mixtures of closely related,
`macromolecular, nanoparticulate components that cannot
`be entirely characterized physicochemically using available
`analytical technology [91]. The consistent activity and the
`quality of NBCDs rely on strictly controlled, proprietary
`manufacturing procedures [91-93]. Even small variations to
`the manufacturing method or of ingredient quality used to
`synthesize GA can increase the risk of safety problems or of
`reduced therapeutic efficacy [94-96]. While Copaxone® is not
`encephalitogenic and does not induce autoreactive antibodies,
`the same cannot be assumed for a follow-on GA-like product.
`Several purported “generic” GA products have been mar-
`keted in countries outside of the United States (US), and
`generic GA products await US FDA approval at this writing
`[97,98]. While it is not possible to show that a generic product
`is the same as Copaxone®, demonstration of “similarity” may
`be adequate for generic approval. Thus far, comparisons of
`physicochemical and biological activity have shown differences
`between purported generic GA products and Copaxone®
`[99,100]. A purported generic GA product is currently marketed
`in India (Glatimer®, Natco Pharma, Ltd., Hyderabad, India) [99].
`No information about the safety, efficacy, or immunogenicity
`of this product in RRMS patients has been published at this
`time. However, in analytical tests, Glatimer® has demonstrated
`physicochemical differences from GA [95,96], and gene
`expression studies, which provide a “snapshot” of biological
`processes stimulated by glatiramoid treatment, show Glatimer®
`also has different biologic activities from those of GA [99,100].
`In 2 separate studies, activated splenocytes from GA-treated
`mice showed distinctly different gene transcription profiles
`when reactivated ex vivo by Glatimer® or GA [99,100]. As
`previously described, among GA effects on APCs is down-
`regulation of macrophage and monocyte activation, and
`among GA effects on T cells is skewing T cell differentiation
`toward a Treg phenotype that limits autoimmune activity
`[73,101]. Figure 3 shows the relative expression of Treg-specific,
`macrophage-specific, and monocyte-specific genes in
`splenocyte samples reactivated by GA compared with samples
`activated by Glatimer® [100]. A cell-type enrichment algorithm
`showed the gene expression profile produced by splenocytes
`activated with Glatimer® was significantly enriched in genes
`associated with macrophages and monocytes compared with
`the gene expression profile of splenocytes activated by GA.
`Similarly, compared with GA, the list of genes down-regulated
`by splenocyte activation with Glatimer® was significantly enr-
`iched in genes associated with Tregs. Thus, Glatimer® could have
`very different immunomodulatory effects than GA. Another
`important finding of gene expression studies was poor batch-to-
`batch reproducibility among Glatimer® batches (Figure 4) [99,100].
`
`5
`
`Jill Conner, Journal of Autoimmunity and Cell Responses 2014,
`http://www.hoajonline.com/journals/pdf/2054-989X-1-3.pdf
`
`Page 5 of 11
`
`YEDA EXHIBIT NO. 2007
`MYLAN PHARM. v YEDA
`IPR2014-00644
`
`

`
`GA
`
`Generic
`
`doi: 10.7243/2054-989X-1-3
`
`GA
`N=24
`
`Generic
`N=105
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Highly Variable Probes By F Test
`
`Figure 4. Biological impact of glatiramer acetate is
`significantly more consistent than that of the purported
`generic product.
`Among probes with variability induced by activation, 4-fold
`more probes had significant variation by F-test across 11
`generic-activated samples from 5 batches, when compared
`with the number of probes with significant variation by F-test
`across 34 GA-activated samples from 30 batches [100].
`
`Under current regulatory statutes in the US, the first clinical
`exposure to generic GA products in MS patients could occur
`after regulatory approval [108]. Adequate testing of the clinical
`safety, efficacy, and immunogenicity of uncharacterized
`follow-on glatiramoid products in MS patients should
`precede regulatory approval and marketing of these products,
`given their complexity, uncharacterized epitopes, unknown
`mechanism of action, and strong immunogenicity. Moreover,
`the inter- and intra-patient variability of antibody responses to
`these immunogenic drugs should be assessed [109]. To ensure
`that the immunologic and immunogenic safety of a generic
`productare comparable to those of Copaxone® will likely
`require clinical trials in MS patients. Antibody responses to a
`generic product should establish that anti-drug antibodies do
`not neutralize drug efficacy or bind to endogenous proteins.
`Risks of free substitution of a follow-on product for Copaxone®
`should be established in a clinical cross-over study that
`provides direct evidence that repeated switching between
`Copaxone® and a generic drug has no negative impact.
`
`Conclusion and future therapeutic peptide vaccines
`Currently, novel synthetic copolymers designed on the basis
`of the immunological properties of GA are under study for
`use in MS [110]. IP-2301, a peptide copolymer currently in
`pharmaceutical development, has 3 of the same amino acid
`constituents as GA (tyrosine, alanine, and lysine) but replaces
`glutamic acid with phenylalanine. This vaccine has been shown
`
`6
`
`FoxP3+ T Cells
`
`Macrophages
`
`Monocytes
`
`1.00
`0.67
`0.33
`0.00
`-0.33
`-0.67
`-1.00
`
`Figure 3. Cell-type specific differences in the biological
`impact of glatiramer acetate and a purported generic
`glatiramer acetate-like product. Differential gene expression
`in mouse splenocytes activated ex vivo with GA or a generic
`product show GA induces higher expression of Treg-associated
`genes than the generic product, and the generic product
`induces higher expression of macrophage and monocyte-
`associated genes [100].
`
`Differences in gene expression profiles have also been detected
`among other purported generic GA products, including Pro-
`bioglat® (Probiomed S.A. de C.V.; Mexico City, Mexico [102]),
`Escadra® (MR PHARMA S.A., Buenos Aires, Argentina), and
`a generic product marketed by several companies in China
`(“Hangzhou”) [96]. Progesterone receptor membrane com-
`ponent 1 (PGRMC1) is significantly down-regulated by Pro-
`bioglat® relative to medium (i.e., the gene expression profile
`from non-reactivated murine splenocytes), but not by Escadra®
`or Hangzhou (Figure 5). Progesterone is thought to play
`a role in myelin repair [103,104] and to affect the balance
`between Tregs and Th cells [105]. Another relevant gene