36-12 in Multiple Sclerosis
`
`j. Theodore Phillips, MD. Pth Robert J. Fox. MD2
`
`1 Multiple Sclerosis Program. Baylor Institute for immunology
`Research. Dallas, Texas
`2Department: of Neurology and Neurological Institute. Mellen Center
`for Multiple Sclerosis Treatment and Research. Cleveland Clinic.
`Cleveland, Ohio
`
`Semin Neurol 2013;33:56 65.
`
`Address for correspondence J. Theodore Phillips. MD. PhD. Multiple
`Sclerosis Program. Baylor Institute for Immunology Research. Dallas.
`TX 75204 (e mail: ted.philiipsGPbaylorhealth.edu).
`
`.Abétrast'
`
`.
`
`
`Dimethyl furnarate (DMF)is anorally:admInIsteredagentthat has been used for over
`__ 40 years for the treatmentofpso asis-R ent.wot de '
`nstratesboth D_MF_-'Immuno— '
`
`
`
`modulator-y andneuroProtectIve _
`”Ions in.vitro andI
`_ animalmoansof autoreactive
`
`I
`MEacts thrhugh
`IlIzatIonandnuclear
`
`__
`5 ._
`
`'
`
`Iofa cascadeofseveralcytoprotect
`'siori oftranscrlptlon factor NF—_I<B—'
`
`
`
`For almost 20 years. disease—modifying therapies have been
`available for treatment ofrelapsing forms of multiple sclero«
`sis (M5). The current standard. first-line therapies for MS are
`interferon beta
`1 (IFNB
`imvonex. Biogen Idec. Weston.
`Massachusetts: BetaseronfBetaferoo. Bayer HealthCare. Lev-
`erkeusen. Germany: Extavia. Novartis. Basel. Switzerland:
`Rebif. EMD Serooo. Inc. Rockland. Massachusetts) and gla-
`tiramer acetate (GA; Copaxone. Teva Pharmaceuticals. Petah
`Tikva. Israel). Phase III clinical trials found each to reduce the
`rate of clinical relapses by approximately 30% relative to
`placebo} 4 Side effects include skin reactions (predominant-
`ly subcutaneous therapies) and flu-like side effects and
`hepatic irritation {IFNB
`l therapies}. A distinct advantage
`of these therapies is their excellent long-term safety and
`reasonable tolerability. although their efficacy is modest and
`
`administered by injection (subcutaneous or
`are
`all
`intramuscular).
`
`therapies have emerged as
`three additional
`Recently.
`important long-term disease—modifying therapies for relaps-
`ing MS. Natalizumab (Tysabri. Biogen ldec. Weston. Massa-
`chusetts) is a monoclonal antibody that inhibits leukocyte
`trafficking into the central nervous system (CNS). Natalizu-
`mab has excellent efficacy in reducing clinical relapses. new
`brain lesions on magnetic resonance imaging (MRI). and
`delaying sustained progression of clisability.5'5 Two draw-
`backs ofnatalizumab include its monthly intravenous admin-
`istration and its association with progressive multifocal
`leukoencephalopathy (PML). a potentially fatal CNS infec-
`tion? Fingolimod (Ciienya, Novartis. Basel. Switzerland). a
`second contemporary therapy.
`is a sphingosine receptor
`
`Issue Theme Current and Emerging
`Therapies for Multiple Sclerosis: Guest
`Editor. B. Mark Keegan. MD, FRCPlC)
`
`Copyright © 2013 by Thiome Medicai
`Publishers. Inc., 333 Seventh Avenue.
`New York. NY 10001, USA.
`Tel: +1912} 584 4662.
`
`DOI http:f{dx.doi.orgf
`10.105513 00331343796.
`ISSN 0271 8235.
`
`
`
`MYL-DMF-00064909
`
`MYLAN PHARMS.I INC. EXHIBIT 1066 PAGE 1
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 1
`
`

`

`30—12 in Multiple Sclerosis
`
`Phillips, Fox
`
`57
`
`modulator that both reduces leukocyte egress from periph-
`eral lymphoid tissue and may have direct immunomodulat-
`ing effects within the CNS.8 Fingolimod shows beneficial
`relapse. disability progression and MRI effects. A head-to-
`head study showed fingolimod is more effective than IFNB
`1.9 However. fingolimod is associated with cardiac side
`effects (mostly initial bradycardia events}. macular edema.
`and increased risk of some infections (most notably respira-
`tory infections and herpes virus infections). A third therapy.
`terifiunomide (Aubagio. Genzyme. Cambridge. Massachu-
`setts) has been recentiy approved for relapsing MS in the
`United States. Teriflunomide inhibits new pyrimidine syn-
`thesis and rapid proliferation of activated lymphocytes
`through its action on dihydroorotate dehydrogenase.” Teri-
`flunomide shows beneficial relapse. disability progression.
`and MRI effects. Gastrointestinal disturbances. hair thinning.
`and mild—moderate elevations in alanine aminotransferase
`
`leveis have been noted more commonly with teriflunomide
`than placebo. Based on animal studies. the US. Food and Drug
`Administration (FDA) has also issued an alert concerning
`possible teratogenic effects in humans.11
`Despite the availability of many disease—modifying thera-
`pies for relapsing MS. none have the ideal treatment triad:
`strong efficacy. good tolerability. and excellent safety. The BC-
`12 preparation of dimerhyl fumarate (DMF) may come closer
`to offering that ideal treatment triad.
`
`Initial Clinical Experience with Fumaric Acids
`
`Furnaric acid (FA) is an intermediate in the citric acid cycle
`used by cells to produce energy. In 1946. food chemists
`recognized its acid reguiatory properties anti fruit—like taste.
`FA is still routinely used as a food aciduient in beverages.
`baking powders. and candy (--Table 1).
`
`is credited to
`The medicinal use of fumaric acids (FA)
`German chemist Waiter Schweckendiek. who suffered from
`psoriasis. hi
`the late i9505. Schwecltendiek turned to his
`professional expertise in chemistry to search for treatments
`for his skin condition. He proposed that psoriasis resulted
`from a defect in carbohydrate metaboiism involving the citric
`acid cycle and its oxidative steps. and predicted improvement
`through ingestion of FA.32 However. his first
`treatment
`attempts with oral FA were disappointing owing to significant
`gastrointestinai irritation. To improve gastrointestinal ab-
`sorption and tolerability. Schwecltenciiek13 synthesized a
`mixture of FA ester (RAE) salts that did prove effective in
`treating his psoriasis. and also demonstrated less irritant side
`effects. Gunther Schafer. a general practitioner. later promot—
`ed a standardized treatment regimen using an FAE mixture
`and restricted diet. and subsequently reported encouraging
`open-label resuits with over 70% of 900 psoriasis patients
`showing improvement.14 Despite the lack ofdata supporting
`the metabolic defect hypothesis. efficacy and safety ofFAE in
`psoriasis Were subsequently demonstrated in two random-
`ized. double-banded. piacebo-controlled studies.”‘-15 These
`studies were performed using a proprietary preparation of
`FAE (120 mg dimethyl fumarate [DMFL 87 mg monoethyl
`fumarate calcium salt. 5 mg monoethyl fumarate magnesium
`salt. and 3 mg monoethyl fumarate zinc salt per ente‘ricu
`coated tabiet).
`in 1994. the Swiss company Fumapharm
`(acquired by Biogen ldec. Weston. Massachusetts) received
`German regulatory approval for this FAB preparation {called
`Fumaderm) for the treatment of psoriasis. Furnaderm is
`currently one of the most widely prescribed systemic thera-
`pies for psoriasis in Germany. with visible improvement
`typically noted within 6 weeks. Furnaderrn is generally weil
`tolerated: cutaneous flushing and gastrointestinal upset are
`the most commonly reported side effects. A small study has
`
`Table 1 Timeline of development of 36-1 ZIDMF in multiple sclerosis
`
`1946: Fumaric acids are recognized by food chemists for their acid regulatory properties and fruit—like taste.
`19605—19805: German chemist Walter Schweckendiek and general practitioner Gunther Schafer develop fumaric acid ester-
`based regimen to treat psoriasis.”-I4
`1994: The Swiss company Fumapharm receives German regulatory approval for Fumad arm. which is a proprietary formulation
`of dimethyl fumarate and several fumarate ester salts
`~ 2000: Bochum neurologlst Horst Przuntek observes that MS patients With psoriasrs appear to have stabilized MS after
`starting Fumaderm
`2002: A patent is fiied for the use of fumarzc acids in the treatment of MS
`
`2003: Fumapharrn establishes collaboration with Slogen ldec to develop fumaric acids in MS.
`
`2011—2012: Both phase lil trials report positive results on Citrilcai and MRI outcomes
`
`2005: Schilling and coileagues show DMF or MMF effectiveiy prevents chronic EAE.24
`2005: Schlmngk and colleagues publish a case series of 10 MS patients treated with furnatlerrn.51
`2007: A piacebo-controlled phase II trial shows a robust reduction in MRI lesions with BIS-12.70 and two phase Ell trials in
`reiapsing remitting MS get. underway.52'53
`2011:Linker and colleagues demonstrate fumaric acid esters activate the Nrf—2 antioxidant pathway and exert neuroprotective
`effects in EAE-associated neuroinflarnmation.25
`
`°
`
`~
`
`52.53
`
`Abbreviations: DMF. dimethyl fumarate; EAE, experimental autoimmune encephalomyelitis; MMF, monomethyi fumarate; MRI. magnetic resonance
`imaging: MS. multiple sclerosis.
`
`Seminars in Neurology Vol. 33 ND. ”2013
`
`MYL-DMF-00064910
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 2
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 2
`
`

`

`58
`
`56-12 in Multiple Scierosis
`
`Phillips. Fox
`
`found acceptable long-term safety in psoriasis patients
`treated for 10 to 14 years.17
`
`DMF Mechanisms of Action
`
`Pharmacokinetics
`
`Comparative studies have shown that DMF (C6H304; MW
`141.1) is the key FAE component accounting for therapeutic
`benefit in psoriasis.18 DMF is rapidly hydrolyzed to mono-
`methyl furnarate (MMF; also known as MHF. metiiyihydrogen
`fumarate) by esterases in the alkaline environment of the
`small intestine.19 in healthy volunteers and psoriasis patients.
`time to detectable MMF in blood is somewhat delayed with
`food intake. but without effect on maximum blood concen-
`tration 5 to 6 hours later.20 Plasma FA levels remain un-
`changed by EAE ingestion. MM? is detectable in blood after a
`variable period of 60 minutes or more, whereas DMF is
`subject to a strong first—pass metabolism and is not systemi-
`caliy detectable at any time after oral DMF ingestionfm'x
`MMF is further metabolized (50% protein-bound; serum haif—
`life 36 3122) in the citric acid cycle to yield water and carbon
`dioxide. DMF is highly lipophiiic and may be absorbed by
`intestinal tissue. where it is at ieast partially metabolized to
`MMF. 9MP can also conjugate directly to intracellular gluta—
`thione and be released systemically as a (possibly bioactive)
`glutathione conjugate. which is further metabolized to DMF-
`mercapturic acid and detectable in urine.23 Additionally. DMF
`may also be released unaltered into the portal circulation and
`taken into venous blood cells. or hydrolyzed to MMF in the
`plasma. These scenarios suggest that DMF could still have
`biologically refeVant activities in vivo (in addition to MM?)
`despite undetectable freenplasma DMF after ingestion. A
`small amount of unaltered MMF is excreted in urine and
`feces. FAEs do not appear to be nephrotoxic. Cytochrome
`MSG-dependent pathways are not involved in FAE metabo-
`lism. and drug interactions have not been reported.
`
`In Vivo Effects of DMF and Experimental Modeis of
`Demyelination and Neurodegeneration
`Efficacy of FAEs in the prevention of myelin—oligodendrocyte
`glycoprotein {MOG)-induced experimental autoimmune
`encephaiomyelitis (EAE)
`in mice was first
`identified in
`2006.24 MOG-EAE is an animai model of inflammatory de-
`myelination that also displays features resembling several
`neurodegenerative aspects of MS. Either DMF or MMF given
`preventatively in acute MOG-EAE led to decreased CNS
`inflammatory infiltrates and increased serum interleukin-
`10 {IL-10: an anti-inflammatory cytokine). Linker and col-
`leagues later confirmed the preventative effects ofDMF given
`during the acute phase of MOG-EAE. and extended these
`findings to demonstrate a therapeutic effect in established.
`chronic lVlOGuEAE.25 However. in contrast to findings in acute
`phase EAE. DMF did not significantiy decrease inflammatory
`infiltrates when administered during the chronic phase of
`MOG—EAE. Interestingly. ciinical benefit in this modei ap~
`peared to result primarily from reduction in demyelination
`with relative preservation of myelin arid axons. This interest-
`ing, but unexpected. outcome was suggested to reflect the
`
`SeminalsinNeurology Vol.33 No. 1:21:13
`
`rescue of neurons and gliai (astrocytes) cells from oxidative
`stress-induced cell death through neuroprotection that is
`mediated by DMF—induced activation of the nuclear factor-
`(erythroid—derived 2—) reiated factor-2 {Nrf2) pathway. Simi-
`larly. other studies have now also demonstrated DMF~
`induced. NrQ-dependent cytoprotection of neurons and
`astrocytes.26 23
`Evidence of oxidative stress is readily apparent within the
`CNS in EAE and MS.29 3‘ Glutathione. a major intracellular
`reactive oxygen species (ROS) scavenger. is decreased within
`CNS inflammatory ford};2 and a variety of antioxidant pro-
`teins.
`including N112. are increased in MS lesions.33 The
`importance ofthe Ner pathway in regulating CNS inflamma-
`tion is emphasized by the observation of significantly wors-
`ened EAE in Ner (
`i
`) knockout mice.25‘34 Also. DMF
`neuroprotective effects are not inducible in Nrf‘2(
`I
`) mice.
`The effects of DMF or MMF have also been examined in
`other modeis ofdemyelination and neurodegeneration. in the
`cuprizone model of toxic. noninflammatory demyelination.
`DMF or MMF demonstrated minimal or no significant impact
`on demyelination or remyelination.35 in contrast to EAE. no
`obvious protective effects of DMF or MMF on CNS celiular
`elements were found in this model. However. in an experi—
`mental model of neurodegeneration with similarities to
`Huntington's disease.
`the transgenic mouse strains R612
`and YACi 28 showed preservation of striatal and motorcortex
`neurons and improved short- and long-term motor outcomes
`with DMF treatment.“5 Apparent DMF-associated neuropro-
`tection has also been observed in a rat model of excitotoxic
`neurodegeneration induced by intrastriatal malonate injec-
`tion.” In this model. DMF treatment resulted in reduction of
`striatal lesionvoiume and improved abnormal apomorphine—
`induced rotational behavior compared with vehicle controls.
`
`DMF Effects on the Immune System
`Afterdismissa‘i of the initial citric acid cycle defect hypothesis.
`early explanations of FAE therapeutic effects focused on the
`proinflammatory Tucell (Tht) theory of certain autoimmune
`disorders.
`including psoriasis33 and M599 In addition to
`antiproliferative effects on keratinocytes.40 FAEs were found
`to have direct effects on multipie components of the immune
`system. MMF induces anti-inflammatory cytokines IL—4.
`lL—S.
`”.4 0. and iL—l RA production by human peripheral blood
`mononuclear cells. without stimulation of proinflammatory
`Cytokines 1L-2. interferon gamma arm). or it—iz.“ 43 Other
`studies found that DMF or MMF may decrease production of
`IFNy and 11.42.44 as weli as decrease expression oicytokine-
`induced endothelial cell adhesion moiecuies lCAM—1.VCAM-
`i. and E—selectin."’5 In vitro differentiation of antigen-pre-
`senting. monocyte-derived dendritic ceils (mDCs) is inhibited
`by DMF or MMF. resulting in decreased mDC production of IL—
`6. lei 0.
`lL—i2. mm. and decreased T—cell production of
`proinflammatory IFNY and iL—17.45 49
`DMF. but not MMF. may induce apoptosis in cultured mDCs
`and activated T cells.50 Possibly related. a transient 4% increase
`in 034+ T cell apoptosis was reported in Fumaderm-treated
`RRMS patients during the first 12 weeks of therapy.“ Apoptosis
`may therefore expiain in part the mild decrease in circulating
`
`MYL-DMF-000649’l1
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 3
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 3
`
`

`

`86-12 in Multiple Sclerosis: [Phillipsfroir - 59
`
`lymphocyte count observed in psoriasis16 and M35253 patients
`treated with Fumadenn or 36-12. respectively.
`
`Subceliular Mechanisms of Action
`
`Twenty years ago. a systematic study of different compounds
`capable of inducing phase [I detoxifying enzymes {e.g.. NAD
`(Phi quinone oxidoreductase-i; NQOT)
`identified DMF.
`among others. to be highly effectives‘j A common structural
`feature among these inducing compounds is that of a so-
`called Michael acceptor. which ref] acts the compounds rela-
`tively eiectrophilic nature. As such. DM? and several other
`Michael acceptor compounds were shown to activate an
`electrophile-responsive element (EpRE)
`located in the 5‘
`flanking region of the mouse glutathione S-transferase Ya
`subunit gene.55 EpRE is also known. perhaps more commonly.
`as ARE (antioxidant-responsive element}.55'57
`Further work showed that the effect of DMF and MMF on
`
`ARE/EpRE is actually indirect. Under normal conditions. ARE,l
`EpRE is not active; however. under cellular stress conditions.
`N112 activates AREprRE.5g No? is normally not available for
`downstream regulatory interactions as it is targeted for rapid
`ubiquitin £3 ligase—(Cu13-) directed proteasomaldegradation
`through its usual bound interaction with the cytosolic re-
`pressor Keapi (Kelch-like ECH-associated protein 1). DMF or
`MMF S-alkylates Cy5151 of cysteine-rich Keapi.
`thereby
`blocking the interaction of Keapi and MD. which prevents
`N112 ubiquitination and leads to N rl2 stabilization. phosphor-
`ylation. nuclear translocation. and subsequent er‘Z nuclear
`activation ofAREiEpRE.59 Therefore. Keapi can be considered
`a critical intracellular redox sensor; therefore. it is sensitive to
`inorganic and organic hydroperoxides. peroxynitrite. and
`other electrophilic molecular species generated by oxidative
`stress and tissue damage.50 AREprRE activation results in an
`increase in synthesis of many antioxidation-related proteins
`such as NQOI. heme oxygenase-1{HO-1).glutamate—cysteine
`ligase catalytic subunit
`(GCLC). glutamate—cysteine ligase
`modifier subunit (GCLM). peroxiredoxin—l (Prxi). glutathi—
`one S-transferase Mu-1 (GSTMT). thioredoxin (Trx). .thiore-
`doxin reductase (Ter). and various heat shock proteins?”
`ARElEpRE activation also enhances other prosurvival pl'O-
`cesses such as phase II detoxification. nucleotide excision
`repair. and autophagy inhibition. Nrfl signaling is ultimately
`terminated through a combination of Nrfz nuclear export
`processes dependent on Fyn—mediated Nri2 phosphotylation,
`as well as removal of NrQ—activating signals through the
`action of the induced antioxidant enzymes?0
`DMF and MMFalso enhance glutathione synthesis through
`the er2 signaling pathway. despite initial reduction ofintra-
`cellular glutathione stores through direct reaction with glu-
`tathione.26 2351 5"'Importantly.initialglutathione reduction
`in mDCs results in preferential induction of type ll mDCs
`promoting Tl12 differentiation. and also blocks STATl phos-
`phorylation. thereby reducing proinflammatory mDC lL—iZ
`production.49
`NF—ltB. a key redox—sensitive. proinilammatory nuclear
`transcription factor.
`is inhibited by DMF or MM? through
`Ner-rnediated induction of phase II antioxidants. with sub-
`sequent reduction of intracellular ROS accumulation in im-
`
`mune~activated cells. and inhibition of NF—kB upstream signal
`mediators Akt. not. and 11<1<F°r55 Through these indirect
`inhibitory actions of DMF and MMF on NF-kB.proinf1arnrna—
`tory gene expression is suppressed. DMF also inhibits NF-kB
`activity through suppression of ERKUE and downstream
`MSKI kinase activities. and subsequent reduction of p65
`(RelA) phosphorylation. nuclear translocation. and target
`gene ti'ansactlvation.43*55 In addition. recent work suggests
`that DMF may be capable of suppressing inflammation
`through one or more Ner-independent pathwaysm
`MMF. but not DME also competitively binds and activates
`the high-affi nity nicotinic acid (NA) receptor GPRi USA. with
`subsequent production of prostanoids.GB This interaction
`appears to account for commonly encountered MMF- (and
`NA—) induced cutaneous flushing. COX—1 and COX-2 inhibitors
`were found to decrease early— and late-phase flushing. re-
`spectively. which prompted a recent successful study with
`aspirin to lessen flushing after DMF ingestion.fig
`
`Summary of DMF Mechanisms of Action
`DMF is a small electrophilic molecule that is rapidly con—
`verted in the gut to MME its major bioactive metabolite. MMF
`alkylates the intracellular redox sensor Keapl. resulting in
`stabilization and nuclear translocation of Ner. which in turn
`activates the gene regulatory element AREIEpRE with subse-
`quent downstream induction of a variety of antioxidant and
`cytoprotective proteins. Through reduction of intracellular
`ROS by several of these induced phase II proteins. proin-
`flammatory activities of NF—kB are reduced. DlVlF direct
`effects on lymphocytes and dendritic cells induce NriZ—relat-
`ed pathways. inhibit proinflammatory cytokine production.
`and promote a shift from Th1 {and Th1?) to'i‘hZ—like immune
`activities. Other non-N112 mediated pathways may be in-
`volved in DMF anti-inflammatory effects as well. DMF direct
`effects on neurons and glia {primarily astrocytes) induce
`Nrf2-related pathways that promote neuroprotection and
`cell survival. Therefore. DMF—associated net beneficial effects
`in MS are expected to result from combined anti—inflamma-
`tory and neuroprotective actions.
`
`Early Studies of 86-12 in Multiple Sclerosis
`
`The idea of using FAEs therapeutically in MS was reportedly
`born in the late 19905 at St. Josef Hospital. Ruhr University.
`Bochum. Germany. when neurology chair Horst Przuntek
`noticed that the disease course of two of his MS patients
`appeared to stabilize after they started treatment of their
`concomitant psoriasis with Fumaderm [R. Gold. personal
`communication. 2012i. His colleague Sebastian Schimrigk
`then conducted an open-label. phase l trial of Fumaderm in
`19 MS patients. which found a 95% reduction in number of
`gadolinium-enhancing (Gd-H lesions over 48 weeks.51
`Gastrointestinal side effects from Fumaderm led to its
`further refinement into 36—12. which is an enteric—coated.
`microtablet preparation of DMF. A US. patent was filed in
`2002 for the use of FAE in the treatment of MS. and a formal
`collaboration with between Fumapharm and Biogen Idec was
`established in October 2003.
`
`Seminars in Neurology
`
`Vol. 33 No. ”2013
`
`MYL-DMF-00064912
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 4
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 4
`
`

`

`60 BG-12in Multiple Sclerosis
`
`ehllllps'.jrox.- '-
`
`Subsequently. a 48-week. 257 patient phase llb clinical
`trial in relapsing-remitting MS was conducted (ClinicalTrials.
`gov identifier: NCT00168701). Participants were randomized
`to three different doses of 36-12 (120 mg daily. or 120 mg
`three times daily [360 lug/d). or 240 mg three times daily
`[720 mgfdl) or placebo for 24 weeks. followed by a 24-week
`dose—blinded safety extension (placebo switched to BG-12
`240 mg three times daily)?“ The primary outcome was its
`effect on new Cd—l— lesions. where the highest dose of 86-12
`showed a 69% reduction (p < 0.0001)
`in lesion number
`compared with placebo The middle dose group (E20 mg
`three times daily) did not show a significant reduction in
`lesions. but was probably handicapped by a 56% higher
`baseline MRI disease activity relative to the placebo group.
`When the primary outcome is instead expressed as the
`within-group reduction in MRI disease activity compared
`with each group's prestudy baseline. a consistent dose re—
`sponse is obselved: 29%. 41%. 69%. and 73% lesion reduction
`in the placebo. 120 mg/d. 350 mgid. and 720 mgi'd groups.
`respectively. BC—IZ was tolerated relatively well. with 91% of
`patients completing the fuIiy blinded first 24. weeks. and 97%
`completing the dose-blinded. 24—week safety extension. As
`seen with Fumaderm. the most common 36—12 associated
`adverse effects were skin flushing and gastrointestinal events.
`
`Phase III Trials of 86—12 in Multiple Sclerosis
`
`Based upon these encouraging phase II trial results. two phase
`III trials were conducted: Determination of the Efficacy and
`Safety of Oral Fumarate in Relapsing-Remitting Multiple Scle—
`rosis (DEFINE: CIinicalTrialsgov identifier: l\lCT004'20212)53
`and Comparator and an Oral Fumarate .in Relapsing-Remitting
`Multiple Sclerosis (CONFIRM: ClinicaiTrials.gov identifier:
`ncroo451451).52 Both DEFINE and common were 2—year.
`randomized. placebo-controlled trials. evaluating clinical re—
`lapses. progressive disability. and (in a subset of patients) MRI
`disease activity. Two doses of 36-12 were evaluated—240 mg
`three times per day (720 mg/d). and 240 mg twice a day (480
`
`mgid with placebo given for the midday dosel—and were
`compared with placebo given three times per day. The CON-
`FIRM trial also included giatiramer acetate (CA) as an open-
`Iabel (but still randomized and evaluator-blinded) treatment
`arm. as required by the European Medicines Agency. to provide
`a risk—benefit comparison to a standard therapy. Because of
`potential partial treatment unblinding due to skin flushing and
`gastrointestinal side effects. several steps were taken to ensure
`study validity: blinded examining neurologists were unaware
`ofeither treatment assignment( oral therapy VS. CA) or patient;
`patients were instructed to not take study medication within 4
`hours of each study visit to reduce potential unblinding of
`study investigators; an independent. blinded adjudication
`committee confirmed all relapses; and image analysis was
`conducted blinded to all study information. Enrollments of
`DEFINE and CONFIRM studies started in 2007. and the last
`patient completed follow-up in 201i. The DEFINE study en—
`rolled anci dosed i234 patients; the CONFIRM study enrolled
`and dosed 1.417 (including the additional GA arm). Both
`studies enrolled patients from North and Central America.
`Europe. and Asia. Of note. these trials were one of the first to
`enroll MS patients in India. Seventy—seven percent of the study
`participants in the DEFINE study completed the study (68% on
`originally assigned study medication}. and 80% of the study
`patients in the CONFIRM study completed the study (71% on
`originally assigned study medication).
`
`Efficacy
`In the DEFINE and CONFIRM studies. both dosing regimens of
`136-12 reduced the annualized relapse rate (ARR; the primary
`outcome in the CONFIRM study) by 44 to 53% (p < 0.00] for
`all; r-Table 2)?”3 Various sensitivityanalyses in both studies
`showed consistent results. Similariy. the risk of relapse (the
`primary outcome in the DEFINE study) was also reduced by 34
`to 50% in both closing regimens (p < 0.001; hTabie 2). 36—12
`reduced the risk ofsustai ned progression ofdisability by 34 to
`38% in the DEFINE study (p < 0.05). and 21 to 24% in the
`CONFERM study. which was not significant. In the CONFIRM
`
`Table 2 Key outcomes in 86-12 phase III studies
`
`
`
`Reduction in;
` Annualized relapse rate
`
` 63"
`
`New Tl-hypointense lesions
`
`5'73
`65‘1
`41"
`New gadolinium-enhancing lesions m 73’
`
`
`Abbreviations: BID. twice daily dosing; GA. glatiramer acetate: TID. three times daily dosing.
`Note: Main clinical and MRI outcomes from the DEFINE (Determination of the Efficacy and Safety of Oral Fun-larate in Relapsing Remitting Multiple
`ScEerosis) and CONFIRM (Comparator and an Oral Fumarate in Reiapsing Remitting Multiple Sclerosis) phase III clinical trials.
`2p < 0.001.
`l’p 5 0.01.
`‘p < 0.05.
`dNot significant.
`
`72“
`
`Seminars in Neurology Vol. 33 No. IIJOIB
`
`MYL-DMF-«00064913
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 5
`
`MYLAN PHARMS. INC. EXHIBIT 1066 PAGE 5
`
`

`

`36—12 in Multiple Sclerosis
`
`Phiilips. Fox‘
`
`61
`
`study. only 13 to 17% of patients in each group had sustained
`progression ofdisability. which was lower than the 16 to 27%
`of patients with progressive disability in the DEFINE study.
`This difference may contribute to the difference in progres-
`sion outcomes between the two studies. B012 also showed a
`robust effect on MRI measures of inflammatory disease
`activity in both studies: the number ofnew or newly enlarg-
`ing'i“) lesions on MRI was reduced by 71 to 85%; the number
`ochl+ lesions were reduced by 65 to 90%. and new Tl black
`holes were reduced by 57 to 72% [ail p < 0.001: -Table 2).
`Compared with placebo. GA performed similariy to that
`seen in previous clinical trials.4'7‘ with a 29% reduction in
`ARR and a 29% reduction in the risk of relapse.The proportion
`with sustained progression of disability was not different
`from placebo (7% reduction. p
`0.70). GA reduced new or
`newlyenlarging'fz lesions by 54%(p < 0.001). Cd'l lesions by
`61% (p < 0.001). and T1 black holes by 41% (p
`0.002).
`Compared with placebo. the treatment effect of BG~12 was
`numerically greater than GA for all clinical and imaging
`outcomes. In a post hoc direct comparison. BG-lZ showad a
`greater treatment effect than GA on ARR (three times daily
`only. p
`0.02): new or enlarging T2 lesions (twice daily.
`p
`0.007: three times daily. p
`0.002); and new T1 lesions
`(three times daily only. p
`0.003).
`lmportantly. except for the additional GA treatment arm in
`the CONFIRM study. both the DEFINE and CONFIRM studies
`were identically designed. with the same inclusion and
`exclusion criteria, treatment assignments. clinicai manage-
`ment. safety monitoring. and outcome criteria. This identical
`study design allowed a subsequent. preplanned combined
`analysis of both studies. comparing each BG-lz dose to
`placebo. This combined analysis found a reduction in the
`proportion of patients with sustained progression of disabili-
`tyof32%(BG-12twicedaily.p
`0.003)and30%(BG—12three
`times daily. 13
`0.005).?2 in addition. the combined analysis
`showed a 49% reduction in ARR with both BG-12 dosing
`regimens (p < 0.0001 ), and a reduction in the risk of relapse
`of 43% (twice daily. p < 0.0001) and 47% (three times daily.
`p < 0.0001). MRI lesion reduction in the combined analysis
`was similarly robust. with 73 to 78% reduction in new or
`enlarging T2 lesions (p < 0.0001 for both doses). 54 to 65%
`reduction in new T1 lesions in < 0.0001 ). and 70 to 83%
`reduction in the odds of having a greater number of Gd+
`lesions vs placebo (13 < 0.0001 ).73
`
`Safety and Tolerability
`BG«12 was generally well tolerated in both phase III trials.
`with skin flushing and gastrointestinal events reported as the
`most common 30-12 associated adverse events. Skin flushing
`was reported in 28 to 38% of 36—12 treated patients in both
`studies. compared with 5 to 6% of placebo patients. Skin
`flushing occurs approximately 30 to 45 minutes after dosing.
`typically invoives the chest. neck. and face. and usually
`resolves spontaneously after 15 to 30 minutes. In a small
`study with healthy volunteers. aspirin was shown to almost
`completely biock BG~12-related flushing.69 Gastrointestinal
`events were reported in 36 to 4% ofBG-12 treated patients in
`both studies. compared with 25% of placebo patients. These
`
`gastrointestinal events include nausea. vomitingabdominal
`pain. and diarrhea. Flushing led to treatment discontinuation
`in 1 to 4% of 136—12 treated patients. and nausea or diarrhea
`each led to treatment discontinuation in 0 to 2% of 30-12
`
`treated patients. The incidence of both flushing and gastro-
`intestinai events peak in the first month of treatment. and
`significantly decline thereafter. The incidence of serious
`adverse events was similar between 36-12 groups and pla-
`cebo. There was an equal. low rate of cancer across placebo
`and treatment groups in both studies.
`Infections were observed in 56 to 68% of BC-lZ treated
`
`patients. compared with 50 to 65% ofplacebo~treated patients
`in both studies.Theincidence of serious infections was 1 to 2%
`
`across all groups. No opportunistic infections were reported
`in BG-i2—t1‘eated patients. There was a reduction in mean
`white cell count (11—12%) and mean lymphocyte count (28—
`32%) in both BG—12 treatment groups during the first year.
`which then plateaued and remained within the normal range.
`White blood cell counts < 3.0 x 1091L were seen in 7 to 10%
`of BG-12—treated patients {and 1% of placebo). while lym-
`phocyte counts < 0.5 x 109/Lwere seen in 4 to 5% ofBG-12u
`treated patients (and < 1% ofplacebo); none ofthese patients
`were discontinued from 86-12 due to these laboratory values.
`An increased incidence of mildly elevated liver aminotrans—
`ferase levels was seen during the first few months of treat-
`ment in the DEFINE (6% in 130—12 groups. and 3% in placebo)
`study. but not in the CONFIRM study.
`
`Ongoing Studies
`At the conclusion of both the DEFINE and CONFIRM studies.
`patients who successfully completed 2 years of follow-up
`were offered enrollment in an open~label extension study.
`Patients originally randomized to 86—12 continued their
`previous regimen (twice daily or three times daily). and
`patients originally randomized to placebo or CA were re-
`randomized to either 86-12 twice daily or 80-12 three times
`daily. Foliow—up includes clinical. MRI. and safety assess‘
`ments. Aithough regulatory submission included only twice
`daily closing. the inclusion of three times daily in the long-
`term follow-up will identify whether the three times daily
`regimen may have long-term benefits over a twice daily
`regimen that were not apparent over the first 2 years of
`treatment.
`
`In addition. a 6-month. open—labei. add-on study has been
`conducted in patients receiving either IFNB
`1 or GA.
`in—
`tended primarily to evaluate the safety of adding BG-12 to
`standard first-line disease modifying therapies. This combi-