Epilepsy Research 51 (2002) 31 /71
`
`www.elsevier.com/locate/epilepsyres
`
`Progress report on new antiepileptic drugs: a summary of the
`Sixth Eilat Conference (EILAT VI)
`
`M. Bialer a,*, S.I. Johannessen b, H.J. Kupferberg c, R.H. Levy d, P. Loiseau e,
`E. Perucca f
`
`a School of Pharmacy and David R. Bloom Centre for Pharmacy, Faculty of Medicine, Ein Karem, The Hebrew University of Jerusalem,
`Jerusalem 91120, Israel
`b The National Centre for Epilepsy, Sandvika, Norway
`c Kupferberg Consultants, LLC, Bethesda, MD, USA
`d Department of Pharmaceutics and Neurological Surgery, University of Washington, Seattle, WA, USA
`e Department of Neurology, Bordeaux University Hospital Pellegrin, Bordeaux, France
`f Clinical Pharmacology Unit, Department of Internal Medicine and Therapeutics, University of Pavia, Pavia, Italy
`
`Received 20 May 2002; accepted 23 May 2002
`
`Abstract
`
`The Sixth Eilat Conference on New Antiepileptic Drugs (AEDs) took place in Taormina, Sicily, Italy from 7th to
`11th April, 2002. Basic scientists, clinical pharmacologists and neurologists from 27 countries attended the conference,
`whose main themes included dose /response relationships with conventional and recent AEDs, teratogenic effects of
`conventional and recent AEDs, update on clinical implications of AED metabolism, prevention of epileptogesis, and
`seizure aggravation by AEDs. According to tradition, the central part of the conference was devoted to a review of
`AEDs in development, as well to updates on AEDs, which have been marketed in recent years. This article summarizes
`the information presented on drugs in preclinical and clinical development, including carabersat (SB-204269), CGX-
`1007 (Conantokin-G), pregabalin, retigabine (D-23129), safinamide, SPD421 (DP-VPA), SPM 927, talampanel and
`valrocemide (TV 1901). Updates on fosphenytoin, gabapentin, lamotrigine, levetiracetam, oxcarbazepine, tiagabine,
`topiramate, vigabatrin, zonisamide, new formulations of valproic acid, and the antiepileptic vagal stimulator device are
`also presented. # 2002 Elsevier Science B.V. All rights reserved.
`
`Keywords: Antiepileptic drugs; Drug development; Epilepsy; Pharmacology; Clinical trials; Conference
`
`* Corresponding author. Tel.: /972-2-6758610; fax: /972-
`2-675746
`E-mail address: bialer@md.huji.ac.il (M. Bialer).
`
`1. Introduction
`
`The Sixth Eilat Conference on New Antiepilep-
`tic Drugs (AEDs) took place in Taormina, Sicily,
`Italy from 7th to 11th April, 2002. The conference
`saw active participation of basic scientists, clinical
`pharmacologists and neurologists from 27 coun-
`tries, with representatives from academia,
`the
`
`0920-1211/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
`PII: S 0 9 2 0 - 1 2 1 1 ( 0 2 ) 0 0 1 0 6 - 7
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 1
`
`

`
`32
`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`pharmaceutical industry, and regulatory agencies
`from both sides of the Atlantic. Main themes
`included dose /response relationships with conven-
`tional and recent AEDs, teratogenic effects of
`conventional and recent AEDs, update on clinical
`implications of AED metabolism, prevention of
`epileptogenesis, and seizure aggravation by AEDs.
`The central part of the conference was dedicated to
`discussion of drugs in development, as well as to
`updates on AEDs marketed in recent years, new
`formulations of valproic acid (VPA) and vagal
`nerve stimulator device. The following is a sum-
`mary on the contributions related to these topics.
`
`2. Drugs in development
`
`2.1. Carabersat (SB-204269)
`
`J.A. Messenheimer
`GlaxoSmithKline (GSK), Research Triangle Park,
`NC, USA.
`
`2.1.1. Introduction
`Carabersat (CRB; SB-204269) is a novel fluor-
`obenzoylamino benzopyran compound which ex-
`hibits
`excellent
`anticonvulsant potency
`and
`efficacy relative to phenytoin (PHT), carbamaze-
`pine (CBZ), DZP, lamotrigine (LTG) and GBP in
`a wide range of in vivo (pentylenetetrazol (PTZ)
`
`infusion, maximal electroshock (MES) seizure
`threshold,
`supramaximal electroshock) and in
`vitro (elevated potassium and zero magnesium
`brain slice assays) rodent seizure models (Upton
`et al., 1997). Overall, the profile of activity in these
`models strongly suggests that the compound acts
`by preventing seizure spread. This particular mode
`of action is also shared by AEDs such as CBZ and
`LTG and may translate into clinical utility for
`symptomatic control of partial (with and without
`generalization) and generalized tonic-clonic sei-
`zures. In addition, CRB appears to slow the
`development of amygdala-kindled seizures
`in
`rats,
`indicating that it may be able to inhibit
`epileptogenesis. Importantly, the anticonvulsant
`properties of CRB are of long duration and there
`is no evidence of a decline in efficacy following
`repeated administration.
`
`2.1.2. Mechanism of action
`Although the precise molecular mechanism(s) of
`CRB remain(s) to be determined, a specific bind-
`ing site for the compound (first discovered using
`[3H]-CRB) has been identified in the brain tissue of
`several species including mouse, rat, cat, dog,
`marmoset and, most importantly, man. The high-
`est levels of binding are found in the superficial
`layers of the cerebral cortex and granule cell layer
`of the cerebellum, with moderate levels in CA
`fields and dentate gyrus of
`the hippocampus
`(Herdon et al., 1997). Several lines of evidence
`now strongly support the relevance of this binding
`site to the anticonvulsant properties of CRB. For
`example,
`there is a good correlation between
`binding affinity and anticonvulsant activity in the
`mouse MES seizure threshold test for a wide range
`of analogues of CRB. To date, no activity of CRB
`(pKi5/5) has been identified in over 50 radioli-
`gand-binding assays. These include binding to
`amino acid receptors and ion channels, sodium
`and potassium channels, purinergic and aminergic
`receptors, and opioid and other peptidergic recep-
`tors. In addition, CRB did not demonstrate
`significant activity in a range of in vitro functional
`assays. These findings indicate that CRB interacts
`selectively with its own binding site and, unlike
`other AEDs, has no known effect on sodium
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 2
`
`

`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`33
`
`channels or GABAergic or glutamatergic path-
`ways.
`
`2.1.3. Pharmacokinetics
`A total of 136 healthy volunteers have com-
`pleted dosing with CRB in eight separate studies.
`CRB was well tolerated. Pharmacokinetic data
`from these studies have shown that the current
`formulation exhibits dose proportional changes in
`maximum concentration (Cmax) and area-under-
`the-curve (AUC), a variable time to maximum
`concentration (tmax) of several hours, and an
`apparent terminal half-life of 24 h. Oral bioavail-
`ability is enhanced by food. Radiolabeled studies
`with [14C]-carabersat have shown that the com-
`pound is predominantly cleared by hepatic meta-
`bolism in man.
`
`2.1.4. Efficacy
`A parallel-design, multicenter, double-blind,
`placebo-controlled evaluation of the safety and
`efficacy of CRB given as add-on therapy in 305
`patients (228 randomized to CRB, 77 to placebo)
`with refractory simple or complex partial seizures
`(with or without secondary generalization) has
`been completed. The parallel groups received 400,
`800 or 1200 mg/day CRB or placebo. This study,
`which used the initial formulation of CRB, showed
`a mean reduction in seizure frequency of 20 /30%
`in the 1200 mg/day dose group compared with
`placebo (P B/0.05). In general, CRB was well
`tolerated at all dose groups. These findings suggest
`that CRB has biological activity in epilepsy and
`displays a favorable safety profile. A novel for-
`mulation of CRB has been developed to provide a
`more advantageous pharmacokinetic profile (de-
`creased Cmax, delayed tmax, plasma levels sustained
`above a Cmin level). Future clinical trials will be
`conducted using this new formulation.
`
`2.2. CGX-1007 (Conantokin-G)
`
`T. McCabea, S.C. Schachterb
`aCognetix, Salt Lake City, UT, USA, and bBeth
`Israel Deaconess Medical Center and Harvard
`Medical School, Boston, MA, USA.
`
`2.2.1. Introduction
`CGX-1007 (Conantokin-G) is a synthetic ver-
`sion of a conopeptide, derived from Conus geo-
`graphus cone snail venom,
`that has exhibited
`anticonvulsant activity in preclinical
`studies.
`CGX-1007 is 17 amino acids in length. Among
`the approaches for epilepsy management is N -
`methyl-D-aspartate (NMDA) receptor modula-
`tion. CGX-1007 is a selective NMDA receptor
`antagonist (Donevan and McCabe, 2000). It is
`proposed that CGX-1007 be administered in-
`trathecally (i.t.) to patients with epilepsy using
`the SynchroMed† Infusion System. The Medtro-
`nic SynchroMed† Infusion System consists of a
`small pump implanted in the abdominal region
`and a catheter that delivers medication to a
`specific site within the body. The system bypasses
`the digestive system and the blood /brain barrier,
`two factors important for effective delivery of a
`central nervous system (CNS) active peptide. The
`SynchroMed† Infusion System is approved by the
`FDA for the chronic intraspinal infusion of sterile
`preservative-free morphine sulfate for intractable
`chronic pain,
`including cancer pain;
`for
`the
`intrathecal infusion of baclofen for severe spasti-
`city of spinal and cerebral origin; as well as
`intravascular infusion of floxuridine for the treat-
`ment of primary or secondary metastatic cancer.
`In animal seizure models, CGX-1007 has been
`shown to exhibit broad-spectrum anticonvulsant
`activity with very low behavioral toxicity. More-
`over, CGX-1007 has been demonstrated to be
`effective in animal models of complex partial
`seizures. The objective is to develop the compound
`for the treatment of patients with partial seizures
`with or without secondary generalization refrac-
`tory to available AEDs.
`In humans, CGX-1007 appears to be safe and
`well tolerated via the intravenous (i.v.) route.
`While CGX-1007 possesses a broad-spectrum
`anticonvulsant profile, it is also recognized that
`direct administration to the CNS will be required
`for treatment of patients with epilepsy. The i.t.
`route represents the most feasible delivery ap-
`proach. Given the novel molecular mechanism of
`action, broad spectrum of activity and low beha-
`vioral toxicity, CGX-1007 may represent a unique
`and novel anticonvulsant agent and may provide a
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 3
`
`

`
`34
`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`2.2.3. Toxicology
`In support of a completed phase I clinical i.v.
`study, 14 day i.v. toxicology studies were con-
`ducted in both the rat and dog. An additional 13
`week subcutaneous study was conducted in the rat
`to evaluate the effect of
`longer-term systemic
`exposure as well as any local effects at the site of
`repeated injections. A battery of genetic toxicity
`tests
`(Ames, mouse
`lymphoma
`and mouse
`micronucleus) and immunogenicity studies also
`were conducted. To support delivery of CGX-
`1007 to the brain by direct infusion into the
`cerebrospinal fluid (CSF), i.c.v. and i.t. toxicology
`studies of 28 days and 13 weeks duration
`in both the rat and dog were conducted. It was
`also shown that,
`in general, pharmacokinetic
`parameters were similar after i.c.v. or i.t. admin-
`istration. Based on these studies, CGX-1007
`should be safe for i.t. administration at doses
`many times greater than those required to reduce
`seizure activity.
`
`2.2.4. Drug interactions
`In animal models, based on the protection
`against MES-induced seizures, there was no po-
`tentiation, but a strictly additive interaction when
`CGX-1007 was combined with PHT, CBZ, or
`VPA.
`
`2.2.5. Planned studies
`Based on the data described above, it is pro-
`posed that CGX-1007 be administered to patients
`with
`epilepsy
`by
`i.t.
`delivery
`using
`the
`SynchroMed† Infusion System.
`
`2.3. Pregabalin
`
`A.R. Kuglera, C.M. de Meynardb, L.E. Knappa
`aPfizer Global Research and Development, Ann
`Arbor, MI, USA and bPfizer Global Research and
`Development, Fresnes, France.
`
`significant advance in the treatment of patients
`with uncontrolled seizures.
`
`2.2.2. Pharmacology
`CGX-1007 is a specific NMDA receptor an-
`tagonist that does not interact with any other
`receptors or binding sites
`examined in the
`NovaScreen † receptor profiling study (concentra-
`tion tested was 10 mM). CGX-1007 exhibits wide
`and uniform brain distribution when given i.t. in
`rats and dogs.
`CGX-1007 has been characterized as an antic-
`onvulsant compound in an in vivo reflex model of
`epilepsy. CGX-1007 dose-dependently blocked
`sound-induced tonic extension following i.t. and
`intracerebroventricular (i.c.v.) injection in Frings
`audiogenic seizure-susceptible mice. In addition,
`CGX-1007 did not
`induce behavioral
`toxicity
`(TD50, median behaviorally toxic dose or minimal
`motor impairment as measured by rotarod) in
`animals until
`significantly greater doses were
`administered. The separation between the ED50
`and TD50 doses was greater than that of many
`established AEDs, therefore yielding a greater
`protective index (PI/TD50/ED50). In addition,
`CGX-1007 displayed a better PI than either the
`non-competitive NMDA antagonist MK-801 or
`the polyamine antagonist ifenprodil in the audio-
`genic, chemically- and electrically-induced seizure
`models. In the Frings model, CGX-1007 displays a
`rapid onset (within 1 /3 min) and a prolonged
`duration (2 /4 h) of action following i.c.v. admin-
`istration of a single dose.
`CGX-1007 was also found to display a broad-
`spectrum anticonvulsant profile that was similar to
`that of VPA. Thus,
`it was effective at non-
`behaviorally toxic doses against tonic extension
`seizures induced by both threshold and MES as
`well as clonic seizures induced by PTZ, bicuculline
`(BIC) and picrotoxin (PIC). CGX-1007 is effective
`in rat kindling models of partial seizures when
`given by either i.t. or i.c.v. administration. Collec-
`tively, these results indicate that CGX-1007 has
`the potential to be effective in a variety of human
`seizure disorders including generalized tonic-clonic
`seizures (anti-MES activity), generalized absence/
`myoclonic seizures (anti-PTZ activity), and partial
`seizures (kindling models).
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 4
`
`

`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`35
`
`2.3.1. Introduction
`Pregabalin (PGB), a structural analogue of the
`mammalian neurotransmitter gamma-aminobuty-
`ric acid (GABA), is the pharmacologically active
`S -enantiomer of 3-aminomethyl-5-methylhexa-
`noic acid or 3-(S )-isobutyl GABA.
`
`2.3.2. Pharmacology
`
`2.3.2.1. Anticonvulsant profile. PGB is active in a
`number of animal models of epileptic seizures
`including MES seizures, chemical convulsant sei-
`zures (PTZ, BIC, PIC), kindled seizures in rats,
`and audiogenic seizures in genetically susceptible
`animals.
`PGB prevents electroshock tonic extensor sei-
`zures when given per os (p.o.) and i.v. in mice
`low-
`(maximal stimulus ED50, 20 mg/kg, p.o.;
`intensity stimulus ED50, 1.4 mg/kg, i.v.) and rats
`(ED50, 1.8 mg/kg, p.o.). The ED50 values obtained
`following i.v. and p.o. administration were quite
`similar. PGB did not cause ataxia except at high
`doses (ED50, 60 /260 mg/kg, p.o.). PGB prevented
`threshold clonic seizures induced by PTZ in mice,
`with an ED50 value of 100 mg/kg, p.o. or
`intraperitoneally (i.p.). In mice, threshold seizures
`from BIC or PIC were not completely blocked
`(maximal 70% protection, 250 mg/kg, i.p.), and
`seizures from strychnine were not blocked. Spon-
`taneous absence seizures (6 Hz spike/wave dis-
`charges in neocortical EEG) were not altered by
`10, 40, or 100 mg/kg PGB, but were slightly
`increased by 200 or 400 mg/kg i.p. PGB decreased
`seizures in DBA/2 audiogenic mice after dosages
`of 3 and 10 mg/kg p.o. PGB exhibits a high
`potency in animal seizure models compared to
`other anticonvulsant agents, based on mg/kg
`doses.
`
`2.3.2.2. Mechanisms of action. The mechanism of
`action of PGB is unknown, but it appears to be
`different from that of conventional AEDs. PGB
`does not appear to have any direct action at ion
`channels (Na, Ca2) or transmitter responses
`(glutamate, NMDA, GABA), does not change
`neurotransmitter uptake (glutamate, GABA), and
`does not displace radioligand binding at a variety
`of
`receptors
`(glutamate, GABA, monoamine,
`
`adenosine, cholinergic, opiate) or Na and Ca2
`channels. PGB increases GABA content in neuro-
`nal tissues and enhances glutamic acid decarbox-
`ylase activity. In vitro studies show that PGB
`interacts with an auxiliary subunit (a2-d protein)
`of voltage-gated calcium channels in brain, po-
`tently displacing [3H]-gabapentin or [3H]-L-leu-
`cine. Studies with the R-enantiomer of PGB and
`a number of structural derivatives of pregabalin
`indicate that binding at the a2-d site is required for
`analgesic and anticonvulsant activity in animal
`models.
`
`2.3.3. Pharmacokinetics
`Plasma and brain concentrations of PGB in rats
`were compared with the time course of antic-
`onvulsant effects (MES). In the rat, PGB pharma-
`cokinetics was dose proportional and PGB was
`not significantly metabolized. Maximal MES ac-
`tion was observed approximately 2 /4 h after i.v.
`dosing. Plasma concentrations decreased mono-
`exponentially.
`Phase I clinical studies indicate oral bioavail-
`ability to be approximately 90%, plasma half-life
`(t1/2) 6 h, time to maximum concentration (tmax) 1
`h, and Cmax and AUC to be dose proportional.
`PGB is not significantly metabolized in man, with
`approximately 98% of an absorbed dose being
`excreted unchanged in the urine. PGB is not
`bound to plasma proteins. AUC and t1/2 remain
`unchanged following administration with a stan-
`dardized meal, while Cmax is reduced by approxi-
`25 /30% and tmax
`is
`increased to
`mately
`approximately 3 h.
`
`2.3.4. Drug interactions and tolerability
`PGB is well tolerated, with dose-related CNS
`adverse events (headache, dizziness, somnolence)
`most commonly reported by normal volunteers
`with doses up to 900 mg/day. PGB may be used as
`adjunctive therapy without changes in the blood
`concentrations of other AEDs or PGB itself (no
`drug /drug interactions identified or expected).
`Therapeutic drug monitoring is not required for
`dosing PGB. It is recommended that patients with
`impaired renal
`function have their PGB dose
`reduced.
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 5
`
`

`
`36
`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`2.3.5. Efficacy data
`PGB has been investigated in three multicenter,
`12 week, double-blind, placebo-controlled studies
`as add-on therapy in patients with partial seizures
`uncontrolled by currently available AEDs. These
`studies investigated dose /response (50 /600 mg/
`day), dose regimen (two times a day (b.i.d.) or
`three times a day (t.i.d.) dosing), and titration
`rates. The mean age at epilepsy diagnosis was 13.7
`years (range 0 /73.5) and mean duration of epi-
`lepsy was 25.0 years. Mean and median baseline
`seizure rates were 22.5 and 10.0 seizures per 28
`days, respectively. Seventy-five percent of
`the
`patients were taking two or more concurrent
`AEDs at baseline. Patient demographics were
`well distributed among the treatment groups.
`PGB administered b.i.d. or
`t.i.d. was highly
`effective in reducing seizure frequency compared
`with placebo in all three add-on trials and a clear
`dose /response relationship was shown. Responder
`rates (]/50% reduction in seizure frequency from
`baseline) ranged from 14 /31% with 150 mg/day to
`44 /51% with 600 mg/day. In an additional analy-
`sis of pooled data from all effective dose groups
`from the three controlled studies, statistically
`significant efficacy was seen as early as day 2 of
`treatment. PGB was well tolerated overall by these
`patients. The most frequently reported adverse
`events were CNS-related (dizziness, somnolence,
`and ataxia), most of which were mild to moderate
`in intensity. CNS adverse events tended to increase
`with increasing PGB doses, particularly 300 and
`600 mg/day. Overall, PGB was efficacious and well
`tolerated as reflected by the fact that 83% of PGB-
`treated patients enrolled in open-label extensions
`to the double-blind treatment phases. (French et
`al., 1999; Robbins et al., 2001; Ramsay et al.,
`2001). A presurgical, inpatient monotherapy study
`provided initial evidence of PGB anticonvulsant
`efficacy.
`In conclusion, PGB is a novel CNS active
`compound that has shown anticonvulsant activity
`as adjunctive therapy in epilepsy patients with
`partial seizures. PGB has been well tolerated and
`safe in clinical studies to date, following exposure
`in a total of approximately 1600 patients with
`epilepsy.
`
`2.4. Retigabine
`
`R. Hermanna, E. Schneidera, M. Mentha, N.
`Knebelb, G.M Ferronc, J. Borlakd, K. Erbe, A.
`Partiotc, J-M. Germainc, M. Locherb, R. Porterc
`aClinical Development VIATRIS GmbH & Co.
`KG, Frankfurt am Main, Germany, bEarly Phase
`Development VIATRIS GmbH & Co. KG,
`Frankfurt am Main, Germany, cWyeth Research,
`Radnor, PA, USA and Paris, France, dFraunhofer
`Institute of Toxicology, Drug Research and Clin-
`ical Inhalation, Center for Drug Research and
`Medical Biotechnology, Hannover, Germany and
`eCliphase, Frankfurt, Germany.
`
`2.4.1. Introduction
`Retigabine (RGB; D-23129), N -(2-amino-4-(4-
`fluorobenzyl-amino)-phenyl) carbamic acid ethyl
`ester,
`is a compound structurally unrelated to
`other marketed AEDs. Its remarkable antiepileptic
`potential was identified in the course of a screening
`at the National Institute of Health (NIH).
`
`2.4.2. Pharmacology
`
`2.4.2.1. Anticonvulsant profile. RGB exerts potent
`anticonvulsant activity in a broad variety of
`animal models including models with electrical
`and chemical induction of convulsions and genetic
`models of epilepsy (Rostock et al., 1996). RGB has
`shown a potent effect in the kindling model of
`epileptogenesis (Tober et al., 1996) and is effective
`in a rat model of status epilepticus.
`The drug is active after oral administration, and
`anticonvulsant effects can be clearly separated
`from adverse neurological effects. Unlike standard
`anticonvulsants, RGB is more active in the amyg-
`dala kindling model, a model of complex partial
`seizures in humans, than in models of generalized
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 6
`
`

`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`37
`
`seizures like the MES. PI calculations based on
`effective doses in the amygdala kindling model are
`remarkable. The N -acetyl metabolite of retigabine
`was also found to have some anticonvulsant
`activity.
`
`2.4.2.2. Other pharmacological properties. RGB
`improved learning performance in a model of
`cerebral ischemia and during electroshock amne-
`sia, indicating possible neuroprotective activity in
`models of cerebral deficit. RGB showed dose-
`dependent suppression of chronic neuropathic
`pain in two animal models (formalin test, spinal
`nerve injury test).
`
`2.4.2.3. Mechanism of action. RGB is a first-in-
`class new antiepileptic compound. The primary
`mechanism of action was shown to be a selective
`and potent M-current potassium channel opening
`effect at the KCNQ2/3 and KCNQ3/5 potassium
`channels, which are involved in the control of
`neuronal
`excitability (Rundfeldt and Netzer,
`2000). Mutations in these channels have recently
`been linked to benign neonatal familial convul-
`sions (Hirsch et al., 1999). An ancillary mode of
`action of RGB is the potentiation of GABA-
`evoked currents, which, however, occurs at much
`higher concentrations.
`
`2.4.3. Toxicology
`Repeated-dose toxicity studies of up to 12
`months duration in different animal species in-
`dicate that RGB is well tolerated. Predominant
`central side effects are sedation accompanied by
`hyperexcitability and a decrease in body tempera-
`ture. No dependence liability was observed during
`chronic treatment.
`
`2.4.4. Pharmacokinetics and metabolism
`Single- and multiple-dose pharmacokinetics of
`RGB were assessed in male and female young and
`elderly white subjects and in young male black
`subjects.
`The absolute bioavailability (F ) of oral immedi-
`ate release RGB dosage forms is 50% (data on
`file). Pharmacokinetics are linear for daily doses
`up to 700 mg and are not modified on multiple
`
`administrations. In patients, dose linearity of RGB
`could be demonstrated up to 1200 mg/day. RGB is
`rapidly absorbed, with mean maximum concentra-
`tions of 387 ng/ml (normalized to a 100 mg dose)
`occurring within 1.5 h after intake of a single dose.
`In young white subjects the mean terminal half-life
`of RGB is 8 /9 h and the apparent oral clearance
`0.70 l/h/kg. In black subjects, RGB clearance and
`volume of distribution (Vss) are 25 and 30% lower,
`respectively,
`leading to higher exposure in this
`population (Ferron et al., 2002).
`Compared with young men, higher peak RGB
`concentrations and exposure, but similar clearance
`normalized by body weight, were observed in
`young women,
`suggesting that differences
`in
`body weight between sexes may explain these
`findings. In the elderly (66 /81 years) of both
`sexes, peak plasma concentrations were not altered
`as compared with young subjects, but RGB was
`eliminated more slowly (30% lower weight-nor-
`malized clearance), resulting in higher exposure
`(42%) and a longer terminal half-life (30%) (Her-
`mann et al., 2000).
`In humans, RGB undergoes extensive and
`virtually exclusively phase II biotransformation
`and renal elimination. There is no indication that
`oxidative metabolic pathways, e.g. cytochrome
`P450 isozymes, may play any role in RGB
`biotransformation (Hempel et al., 1999). RGB is
`predominantly metabolized by N -glucuronidation,
`resulting in the formation of two distinct N -
`glucuronides, and to a considerably lesser extent
`by N -acetylation to form a pharmacologically
`active N -acetyl derivative.
`Recently, the specific UGT isozymes 1A1, 1A9,
`and to a lesser extent 1A4 were identified as the
`main isozymes involved in RGB major metabolic
`pathway (Hiller et al., 1999). The kidney contri-
`butes to the formation of both RGB N -glucur-
`onides, indicating that the metabolism of the drug
`is not exclusively confined to the liver. UGT1A1
`and 1A4 are also expressed in all parts of the
`gastrointestinal tract and UGT1A9 is expressed in
`the colon, which suggests that the gut may also
`contribute to the metabolism of RGB. The pre-
`sence of RGB’s major metabolic enzymes in
`various organs further suggest that the pharmaco-
`kinetics of the drug may be not be greatly altered
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 7
`
`

`
`38
`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`in patients with impaired liver function. Specific
`studies in this regard, however, are yet to be
`performed.
`Systemic cleavage of N -glucuronides with sub-
`sequent recycling of the parent drug has been
`suggested. After drug intake, a considerable frac-
`tion of the RGB dose is initially converted to the
`inactive N -glucuronides, with a subsequent gra-
`dual release of free parent drug from the N -
`glucuronide pool. This suggests the modeling of
`a favorable slow-release mechanism from endo-
`genous N -glucuronide sources.
`
`2.4.5. Interactions
`RGB shows no interaction with food (data on
`file) and does not alter the phramcokinetics of the
`oral contraceptive steroids ethinylestradiol/norges-
`trel (Hiller et al., 1999). Lack of pharmacokinetic
`interaction with phenobarbital
`(PB) has been
`shown, suggesting that no dosage adjustment is
`necessary when RGB and PB are coadministered.
`While the pharmacokinetics of RGB is not
`altered by LTG, LTG clearance appears to be
`slightly increased and LTG terminal half-life
`slightly decreased after multiple dosing with
`RGB.
`RGB pharmacokinetics is not altered by VPA,
`topiramate (TPM), PHT or CBZ, and RGB does
`not alter the pharmacokinetics of these drugs in
`epileptic patients. PHT and CBZ, however, mod-
`erately increase the clearance of RGB. Thus,
`increased RGB doses may be required when co-
`administered with PHT and/or CBZ, and caution
`should be taken when reducing or removing co-
`administered CBZ and PHT since RGB concen-
`trations may increase.
`
`2.4.6. Efficacy and tolerability data
`More than 200 patients aged 16 /64 years were
`treated in five phase IIa trials, and in long-
`term extension studies with daily doses of 600 /
`2400 mg. This phase IIa studies were of explora-
`tory character and were designed to assess the
`safety, tolerability, and preliminary efficacy of
`RGB.
`
`2.4.6.1. Efficacy. A confirmatory randomized,
`double-blind, placebo controlled, 4-parallel-group,
`
`multinational multicenter, phase IIb dose ranging
`study involved a total of 399 patients according to
`an add-on study design. This trial confirmed
`statistically significant superiority of 900 and
`1200 mg daily doses in the primary and secondary
`efficacy variables as compared to placebo. The
`median% changes
`in monthly
`total partial
`seizure rates from baseline were 13, 23, 29 and
`35% for placebo and for daily RGB doses of 600,
`900, and 1200 mg, respectively. Corresponding
`responder rates (% of patients showing at least
`50% seizure reduction), assessed as a secondary
`efficacy measure, were 16, 23, 32, and 33%,
`respectively.
`
`2.4.6.2. Adverse events. In patients with epilepsy,
`RGB has been administered as add-on therapy
`and as monotherapy using a gradual dose-titra-
`tion. The maximum tolerated dose (MTD) in
`patients was usually in the range from 1200 to
`1600 mg/day although some patients have toler-
`ated higher doses. The minimum daily dose
`considered to be effective ranged from 400 to 600
`mg (administered b.i.d.). It also appears that a
`t.i.d. dose regimen is equally or better tolerated
`compared with b.i.d. The most frequent adverse
`events that were reported in patients include
`asthenia, dizziness, headache, somnolence, tremor,
`speech disorder, amnesia, vertigo, abnormal think-
`ing (a COSTART/WHO preferred term used to
`indicate symptoms related to mental slowing, not
`psychosis), lack of coordination, nervousness, and
`paresthesia. With regard to cardiac safety, RGB is
`devoid of adverse effects on cardiac repolarization,
`such as prolongation of the QTc interval of the
`ECG.
`
`2.5. Safinamide
`
`R.G. Farielloa, C. Cattaneoa, S. Wischerb, R.
`Majb, F. Tergaub, W. Paulusb
`aNewron Pharmaceuticals Gerenzano, Varese,
`Italy and bDepartment of Clinical Neurophysiol-
`ogy University of Goettingen Medical School,
`Goettingen, Germany.
`
`Actavis v. Research Corp. Techs.
`IPR2014-01126
`RCT EX. 2015 page 8
`
`

`
`M. Bialer et al. / Epilepsy Research 51 (2002) 31 /71
`
`39
`
`via an electrode stereotactically implanted in the
`basolateral amygdala, both the electrical after-
`discharges and the behavioral component of the
`seizures were reduced more prominently by SAF
`than by PHT (Fariello et al., 2000).
`
`2.5.2.2. Behavioral pharmacology. SAF does not
`affect locomotor activity in rats until toxic doses
`are reached (700 mg/kg p.o.
`in rats, which
`correspond to 55 times the MES ED50). At a
`dose 40-fold higher than the ED50, SAF did not
`affect passive avoidance in mice.
`
`2.5.2.3. Mechanisms of action. SAF shows an IC50
`of 8.2 mM at the batrachotoxin sensitive site of the
`Na channel. In electrophysiological tests, safina-
`mide shows use- and frequency-dependent inhibi-
`effects on Na currents, with more
`tory
`pronounced effects compared with PHT, CBZ
`and LTG (Salvati et al., 1999). SAF inhibits
`sustained repetitive firing at concentrations of 3
`mM, without affecting the first action potentials of
`the burst. Safinamide also modulates Ca2 cur-
`rents of the L- and N-type. Perhaps as a con-
`sequence of these actions, safinamide inhibits in
`hippocampal slices glutamate release induced by
`addition of veratridine as well as by high mM K,
`a situation that more closely mimics pathological
`events during seizure activity.
`SAF has no affinity for noradrenaline, dopa-
`mine, serotonin, glutamate or GABA receptors
`(Ki/100 mM).
`SAF is a selective and reversible MAO B
`inhibitor and through this mechanism it increases
`neostriatal dopamine levels in primates, while
`decreasing the levels of dopamine metabolites.
`Other MAO inhibitors have demonstrated antic-
`onvulsant activity in rodent models of epilepsy and
`seizures (Loscher and Honack, 1995, Loscher and
`Lehmann, 1996). In addition, MAO B inhibitors
`reduce production of free radicals and may con-
`tribute to inhibit epileptogenesis, particularly fol-
`lowing post- traumatic or post hemorrhagic injury.
`
`2.5.3. Toxicology
`tolerated. Rats and
`SAF is generally well
`monkeys were chosen for acute and chronic
`studies. Six- and nine-month studies are approach-
`
`2.5.1. Introduction
`Safinamide (SAF; formerly NW-1015) is the S -
`isomer of the fluorobenzyloxy-benzylamino pro-
`panamide methansulfonate salt, which has been
`selected out of a medicinal chemistry program on
`new sodium channel blockers (Pevarello et al.,
`1998). Safinamide uniquely combines use and
`frequency-dependent blockade of Na channels,
`antagonism of N- and L-type Ca2 channels,
`inhibition of glutamate