REVIEWS
`
`New avenues for anti-epileptic
`drug discovery and development
`
`Wolfgang Löscher1,2, Henrik Klitgaard3, Roy E. Twyman4 and Dieter Schmidt5
`
`Abstract | Despite the introduction of over 15 third-generation anti-epileptic drugs, current
`medications fail to control seizures in 20–30% of patients. However, our understanding of
`the mechanisms mediating the development of epilepsy and the causes of drug resistance
`has grown substantially over the past decade, providing opportunities for the discovery and
`development of more efficacious anti-epileptic and anti-epileptogenic drugs. In this Review
`we discuss how previous preclinical models and clinical trial designs may have hampered
`the discovery of better treatments. We propose that future anti-epileptic drug development
`may be improved through a new joint endeavour between academia and the industry,
`through the identification and application of tools for new target-driven approaches,
`and through comparative preclinical proof-of-concept studies and innovative clinical
`trials designs.
`
`Epilepsy
`A chronic brain disorder that
`is characterized by partial
`or generalized spontaneous
`(unprovoked) recurrent
`epileptic seizures and,
`often, comorbidities such
`as anxiety and depression.
`
`1Department of
`Pharmacology, Toxicology
`and Pharmacy, University
`of Veterinary Medicine,
`Hannover 30559, Germany.
`2Center for Systems
`Neuroscience, Hannover
`30559, Germany.
`3UCB Pharma, Neurosciences
`Therapeutic Area,
`Braine‑l’Alleud 1420,
`Belgium.
`4Janssen Research &
`Development, Raritan,
`New Jersey 08869, USA.
`5Epilepsy Research Group,
`Berlin 14163, Germany.
`Correspondence to W.L. 
`e‑mail: wolfgang.loescher@
`tiho‑hannover.de
`doi:10.1038/nrd4126
`Published online
`20 September 2013
`
`Epilepsy is a life-shortening brain disorder affecting
`approximately 1% of the worldwide population1. Although
`repeated epileptic seizures are the clinical hallmark
`of epilepsy, the disease process (epileptogenesis) begins
`before the first seizure and may also lead to the progres-
`sion of epilepsy after the onset of seizures. Epilepsy is
`diverse, with over 15 different seizure types and over
`30 epilepsy syndromes2, and is associated with sub-
`stantial comorbidity, including depression, anxiety and
`increased mortality3.
`During the past three decades, the introduction of
`over 15 third-generation anti-epileptic drugs (AEDs) has
`provided physicians and patients with more options
`for the treatment of many types of seizures4. However,
`although approximately 70–80% of patients with new-
`onset epilepsy eventually enter remission with current
`AEDs, these medications fail to control seizures in
`20–30% of patients5,6. Furthermore, no AED has been
`shown to prevent the development of epilepsy in
`patients prior to the first seizure; these drugs seem to
`purely act to symptomatically suppress seizures once
`they occur7,8. For some AEDs, an anti-epileptogenic
`effect has actually been suggested in certain preclinical
`epilepsy models9,10, but this has not been proven in
`humans. Indeed, with the exception of traumatic brain
`injury7, none of the therapies found to be effective in
`preclinical studies has been adequately tested using an
`appropriately designed clinical trial in humans.
`Unfortunately, there are few aetiologically relevant
`animal models used in epilepsy research today that have
`
`been validated at the clinical level — a fact that obviously
`hampers clinical trial design using the appropriate
`patient population.In addition, there is no compelling
`evidence that third-generation AEDs are generally much
`better tolerated11–13. However, individual modern AEDs
`such as gabapentin (Neurontin; Pfizer) or levetiracetam
`(Keppra; UCB Pharma) cause fewer or no dermatological
`hypersensitivity reactions. Also, non-enzyme-inducing
`modern AEDs such as gabapentin or levetiracetam do
`not induce the drug interactions seen with older AEDs
`that have been reported to substantially lower the effi-
`cacy of other medications, including other AEDs given
`in combination14.
`AEDs are also unable to prevent or reverse the devel-
`opment of drug-resistant epilepsy, to treat comorbidities
`or to reduce the burden of disease in a holistic sense4.
`A particularly disquieting aspect of current epilepsy
`treatments is that we have not made substantial pro-
`gress in seizure control over the past 40–50 years since
`the introduction of carbamazepine and valproate to the
`market4,15.
`The consequences of the standstill in the development
`of more efficacious drugs for the treatment of epilepsy are
`several-fold. Patients and physicians are increasingly dis-
`appointed and have thus become less interested in using
`recently developed, pricier AEDs. Payers are hesitant to
`pay premium prices for drugs that do not differentiate
`from established low-cost generic medications, and the
`pharmaceutical industry is losing interest in developing
`novel compounds for epilepsy (BOX 1).
`
`NATURE REVIEWS | DRUG DISCOVERY
`
` VOLUME 12 | OCTOBER 2013 | 757
`
`© 2013 Macmillan Publishers Limited. All rights reserved
`
`MYLAN - EXHIBIT 1034
`
`

`
`R E V I E W S
`
`Box 1 | Business challenges and opportunities for anti-epileptic drug development
`
`In the 1990s, epilepsy presented an opportunity to enter a therapeutic space in which there was a good chance for return
`on investment. Drivers for this included a significant unmet need with few treatment options (especially for patients with
`refractory epilepsy), good potential for reimbursement at competitive pricing with few competitors in the field, as well as
`manageable technical and regulatory hurdles.
`The adjunctive or add‑on treatment paradigm in the clinical management of refractory epilepsy was well suited for
`bringing forward new agents to the market. The placebo‑controlled adjunctive model for evaluating the efficacy of a test
`compound in refractory patients established efficacy and tolerability at an early stage and could be performed using
`cost‑efficient short‑term clinical studies. Furthermore, following the introduction of felbamate (Felbatol; MedPointe)
`to the market, a new regulatory path existed for the clinical development and labelling of anti‑epileptic drugs (AEDs).
`Together, these commercial, scientific, technical and regulatory factors drove confidence and reduced the risk
`associated with developing and obtaining a value‑returning marketable product for epilepsy. This template provided
`an incentive for several companies to confidently invest in bringing new AEDs to the market.
`Loss of industry interest in AEDs
`Prior incentives for investment in AED development are now negatively balanced by the drug development challenges
`facing industry overall144–146. Payer reimbursement requires that future AEDs bring additional value or differentiation
`(principally an improvement in efficacy) to an already crowded, highly generic AED field. No AED to date has convincingly
`been demonstrated to be superior in efficacy to any other AED in adjunctive therapy for partial seizures, and
`differentiation by safety profile for new AEDs is not a principal component for optimizing pricing and reimbursement.
`New regulatory hurdles have also evolved over the past 15–20 years. A generally lower risk tolerance for new drugs and
`recent class labelling regarding safety signals (that is, suicide) have affected opportunities in non‑epilepsy indications
`and had an impact on the overall value proposition for AEDs. New AEDs can require commitments for long‑term safety
`data in a variety of age populations, and paediatric investigational plans necessitate the development and testing of
`new formulations in very young patients (babies who are ≥1 month old). Commercialization models indicate that the
`adjunctive indication alone for a marginally differentiated product is not adequate. Product promotion for additional
`uses requires those specific indications to be established in the label. A monotherapy indication can move an AED earlier
`into the epilepsy treatment paradigm. However, the approval of a monotherapy has so far required the prior approval
`of an adjunctive therapy and this causes a considerable time delay.
`Future business opportunities for AED development
`Interesting business cases seem to exist for the very disabling epilepsy syndromes — which are associated with an
`increased risk of premature death — such as infantile spasms and Lennox–Gastaut syndrome. These may present viable
`business opportunities for orphan indications, for which tax incentives are provided, investments are smaller and there
`is a potentially less demanding path for approval.
`Another more immediate business opportunity may involve the repurposing of drugs from other therapeutic areas that
`possess either relevant disease‑modifying properties for epilepsy or a novel mechanism of action that provides substantial
`synergistic efficacy against drug‑resistant epilepsy when combined with an existing AED therapy. This would markedly
`reduce the level of investment necessary for discovery and development, and also potentially lower the technical hurdles
`and regulatory data requirements, thereby improving the premises for a very positive business case.
`A substantial level of investment, beyond that required for traditional AED development, will be necessary for the future
`development of new AEDs that have evidence of superior efficacy against a relevant standard of care for the treatment
`of drug‑resistant epilepsy, or that have the ability to markedly alter the course or the prognosis of epilepsy. However, as these
`types of new epilepsy therapies address a major unmet medical need, they also offer a promising business case to drive
`incentive for future AED development.
`The figure illustrates a hypothetical investment example for the development of an AED: a new molecular entity
`(NME) transitions from discovery into clinical development to be ultimately approved for marketing authorization.
`From discovery, the lead molecule passes through late‑stage preclinical toxicology testing and chemistry scale‑up into
`clinical testing at a cost of US$10 million and a success rate of 70%. The NME passes through each stage with an overall
`success rate of about 5% at a total cost of $350 million. A key inflection point is at the Phase II stage prior to the most
`significant spending investment in Phase III. A reduction of risk at this stage can greatly influence the overall success
`rate and total expenditure for the development of an AED. Note that a cost‑effective proof‑of‑differentiation step
`early in Phase II can further reduce the investment risk, cost and time. Sales and marketing costs add to the investment
`and can be of a similar
`magnitude to the development
`costs. Following marketing
`approval, there are costs
`for sales and marketing,
`launch, sales force, Phase IV
`medical affairs studies and
`post‑marketing regulatory
`commitments. Investments
`in the initial monotherapy
`indication and an alternative
`non‑epilepsy indication could
`add up to approximately
`$50–250 million.
`
`Epileptogenesis
`The gradual process (also
`termed latent period) by
`which epilepsy develops in the
`normal brain following brain
`insults or gene mutations.
`
`Anti-epileptic drugs
`(AEDs). Also termed
`anti convulsant or anti-seizure
`drugs. Compounds that, when
`administered systemically in
`animal models or to patients,
`inhibit or control seizures that
`are associated with epilepsy
`or other conditions.
`
`Stage
`Preclinical
`Phase I
`NME
`success
`Chance of progressing
`
`70%
`success
`
`×
`
`50%
`success
`
`Cost (millions of US$)
`
`$10
`
`+
`
`$15
`
`Time (years)
`
`2
`
`+
`
`1
`
`Phase II
`success
`
`Phase III
`success
`
`Regulatory
`success
`
`=
`
`Marketing
`approval
`
`35%
`success
`
`$93
`
`2
`
`×
`
`+
`
`+
`
`50%
`success
`
`$225
`
`4
`
`×
`
`+
`
`+
`
`×
`
`+
`
`+
`
`80%
`success =
`
`Overall
`success
`rate: 5%
`
`$10
`
`1
`
`=
`
`=
`
`Total cost:
`$350 million
`
`Total time:
`10 years
`
`758 | OCTOBER 2013 | VOLUME 12
`
`Nature Reviews | Drug Discovery
` www.nature.com/reviews/drugdisc
`
`© 2013 Macmillan Publishers Limited. All rights reserved
`
`

`
`R E V I E W S
`
`MES seizure test
`(Maximal electroshock seizure
`test). A model in which a short
`(0.2-second) transcorneal or
`transauricular application of a
`50 or 60 Hz electrical stimulus
`in rodents induces generalized
`tonic–clonic seizures that are
`mediated by brainstem
`structures.
`
`Pentylenetetrazole
`(PTZ). A chemical convulsant
`that, when administered
`systemically to rodents, induces
`characteristic myoclonic and
`clonic convulsions that
`are mediated by forebrain
`structures.
`
`Amygdala kindling
`Repeated administration of an
`initially subconvulsive electrical
`stimuli via a depth electrode in
`the amygdala, which induces
`seizures that progressively
`increase in severity and
`duration; once established, the
`increased susceptibility to the
`induction of kindled seizures is
`a permanent phenomenon.
`
`GAERS rat
`(Genetic absence epileptic rat
`from Strasbourg). A genetic
`rat model that displays
`characteristic 6–7 Hz
`spike-wave electrographic
`seizures and a pharmacological
`profile that is consistent with
`generalized absence epilepsy.
`
`6-Hz psychomotor
`seizure model
`A seizure model in which
`a prolonged (4-second)
`transcorneal application of a
`6-Hz electrical stimulus in mice
`induces limbic seizures that
`are characterized by a stun,
`vibrissae chomping, forelimb
`clonus and a Straub tail;
`these seizures are resistant
`to phenytoin and some
`other anti-epileptic drugs.
`
`Non-inferiority trial design
`A clinical trial design that
`determines whether a test
`compound is inferior to
`another compound; the lower
`limit (95% confidence
`interval) of a test compound’s
`treatment efficacy or
`effectiveness is to be
`compared to a preset lower
`boundary of efficacy or
`effectiveness relative to the
`adequate comparator’s
`point estimate of efficacy
`or effectiveness.
`
`In this Review we briefly examine the experimental
`and clinical strategies for AED discovery and develop-
`ment over the past few decades and discuss why these
`approaches may have failed to address unmet medical
`needs. We also outline the challenges for the pharma-
`ceutical industry that have had an impact on its attitude
`towards the discovery and development of AEDs. Given
`the serious unmet clinical needs in epilepsy treatment,
`we present new ideas on how to revitalize the pharma-
`cological and clinical development of better AEDs that
`could provide the foundation for a new, joint endeavour
`between academia and the industry.
`
`Previous AED discovery and development
`Until recently, the discovery and development of a new
`AED almost exclusively relied on preclinical testing in
`animal seizure models to establish anti-seizure efficacy
`prior to conducting clinical trials in humans16. This
`approach has been successful and crucially contributed
`to the development of numerous clinically effective
`AEDs4,17. Indeed, animal models with a similarly high
`predictive value do not exist for other central nervous
`system (CNS) disorders, such as bipolar disorders or
`migraine18.
`Since Merritt and Putnam19 first described the use
`of an electroshock seizure model to assess drugs for
`anti-seizure properties in 1937 (FIG. 1 (TIMELINE)), simple
`models of acute seizures — such as the MES seizure test
`and the subcutaneous pentylenetetrazole (PTZ) seizure
`test in mice and rats — have been widely used in AED
`discovery. These models were considered to be ideal
`for AED discovery, which necessitates the screening
`of large numbers of compounds; acute seizure models
`should therefore be easy to perform, time- and cost-
`efficient, and predictive of clinical activity. The rodent
`MES test created by Toman, Swinyard and Goodman20
`in 1946 is still the most commonly used first screen in
`the search for new AEDs and is quite effective in identi-
`fying drugs that block generalized tonic–clonic seizures
`in patients17. The MES test has also repeatedly been
`proposed to identify drugs that are active against partial
`seizures in patients, but this test failed to detect several
`AEDs (for example, levetiracetam and vigabatrin (Sabril;
`Lundbeck)) that are effective against partial seizures in
`patients; therefore, other models such as amygdala kindling
`are better for identifying anticonvulsant effects against
`partial seizures21.
`Following the report of Everett and Richards22 in
`1944 that the PTZ test can identify the anti-absence
`efficacy of AEDs, two simple animal models — the
`MES and PTZ tests — were thought to be sufficient
`for differentiating among AEDs with different clinical
`effects. This subsequently formed the basis for the pro-
`posal made by Swinyard and colleagues23,24 that the
`MES and subcutaneous PTZ tests in mice and rats be
`used as standard procedures for predicting the clinical
`anticonvulsant activity of investigational drugs (FIG. 1
`(TIMELINE)). However, because of false-positive and false-
`negative findings in these models, more complex chronic
`epilepsy models that were developed in the 1980s and
`1990s (FIG. 1 (TIMELINE)) have subsequently been included in
`
`later-stage screening to further characterize anti-epileptic
`efficacy — the most notable of these models being the
`kindling model and genetic models of epilepsy, such as
`the absence-epilepsy-prone GAERS rat. More recently, the
`6-Hz psychomotor seizure model in mice has been intro-
`duced for differentiating an investigational AED from
`existing AEDs. This model is resistant to some of the
`old AEDs and enables the screening of a large number of
`compounds17,25, which is not possible with more elaborate
`models such as the kindling model.
`
`Preclinical strategies. Three strategies have been used
`in AED discovery: first, the random, phenotypic
`screening of newly synthesized compounds of diverse
`structural categories with as yet unknown mechanisms;
`second, the structural variation of known AEDs; and
`third, hypothesis-driven, target-based drug design4,17,18.
`All three strategies have generated clinically useful
`AEDs but only very few AEDs have been identified
`by rational, target-based strategies. These have been
`based on previously presumed mechanisms of seizure
`generation: that is, impaired GABA (γ-aminobutyric
`acid)-ergic inhibition and increased glutamatergic
`excitation, resulting in AEDs that either potentiate
`GABA transmission (such as vigabatrin and tiagabine)
`or inhibit glutamate receptors (such as perampanel
`(Fycompa; Eisai))17. However, the old reductionistic
`view that seizures or epilepsy are due to an imbalance
`between GABAergic inhibition and glutamatergic exci-
`tation ignores the complexity of the alterations within
`these neurotransmitter systems in the brain of a patient
`suffer ing from epileptic seizures26.
`
`Clinical strategies. Marketing approval of new AEDs for
`the treatment of epilepsy has been routinely obtained by
`adjunctive therapy placebo-controlled Phase III trials
`in adult patients with refractory seizures27. In the 1960s
`and 1970s, when few AEDs were available4, the enrol-
`ment of patients into these trials was straightforward
`and the use of placebo treatments was deemed acceptable
`given the lack of alternative treatment options28. This
`clinical strategy was very successful and has resulted in
`over 15 new AEDs entering the market since the 1980s
`(TABLE 1). Many AEDs that are marketed for adjunc-
`tive treatment are subsequently tested in monotherapy
`trials in patients with either refractory or previously
`untreated epilepsy. Because regulatory guidelines for
`monotherapy approval differ between Europe and the
`United States, sponsors need to pursue two separate and
`costly development programmes. The mono therapy
`development paradigm currently used in Europe for
`new-onset epilepsy is the non-inferiority trial design,
`which establishes a preset limit for the allowed differ-
`ence in outcome between the test drug and a standard
`AED27. In the United States, the preferred develop-
`ment path is conversion to monotherapy in refractory
`patients using historical controls. These designs have
`demonstrated that several AEDs are efficacious as
`monotherapies, including levetiracetam and zonis-
`amide (Zonegran; Eisai) in Europe and lamotrigine
`(Lamictal-XR; GlaxoSmithKline) in the United States28.
`
`NATURE REVIEWS | DRUG DISCOVERY
`
` VOLUME 12 | OCTOBER 2013 | 759
`
`© 2013 Macmillan Publishers Limited. All rights reserved
`
`

`
`R E V I E W S
`
`Timeline | Milestones in the development of animal models for AED discovery and development*
`
`Putnam and
`Merritt19; EST
`test (phenytoin)
`
`Swinyard23; MES plus
`PTZ tests (standard
`AED assay)
`
`Anticonvulsant Screening Project of
`the National Institute of Neurological
`Disorders and Stroke (NINDS) of the
`US National Institutes of Health
`(NIH)29,30; MES, PTZ and rotarod tests
`
`Vergnes et al.157;
`GAERS rat
`(spontaneous
`absences)
`
`Barton et al.25; 6-Hz
`model (first described
`by Toman in 1951)159
`
`1937
`
`1944
`
`1949
`
`1969
`
`1975
`
`1979
`
`1982
`
`1991
`
`2001
`
`Everett and
`Richards22; PTZ test
`(trimethadione)
`
`Goddard et al.155;
`kindling model
`(focal seizures)
`
`Ben-Ari et al.156;
`kainate-induced status
`epilepticus (SRS)
`
`Cavalheiro et al.158; pilocarpine-
`induced status epilepticus (SRS)
`
`Löscher and Rundfeldt148;
`phenytoin non-responders and
`responders in kindled rats
`
`AED, anti-epileptic drug; EST, electroshock threshold; GAERS, genetic absence epilepsy rat from Strasbourg; MES, maximal electroshock; PTZ, pentylenetetrazole;
`SRS, spontaneous recurrent seizures. *All animal models shown (except for the SRS models described by Ben-Ari et al.156,Vergnes et al.157 and Cavalheiro et al.158)
`are those in which seizures are electrically or chemically induced. All models, except for the EST method in cats described by Putnam and Merritt19, are still
`used in the development of new epilepsy therapies21. Various models are important for different purposes in epilepsy research21 and can be assigned to four
`major categories: first, acute seizure models in which single seizures are electrically or chemically induced in healthy, neurologically intact rodents, such as the
`MES, subcutaneous PTZ or 6-Hz tests; second, chronic seizure (or epilepsy) models in which single or multiple seizures are electrically or chemically induced in
`rodents with chronic brain alterations, such as amygdala kindling; third, genetic animal models with inborn chronic epilepsy, such as the GAERS rat (which is
`better suited than the PTZ test to identify drugs that are active against absence seizures); and fourth, chronic epilepsy models in which epilepsy with SRS is
`induced by brain insults, such as status epilepticus (for example, induced by pilocarpine or kainate) or traumatic brain injury21. The MES and subcutaneous PTZ
`tests, which were developed more than 60 years ago, have been widely used in the search for new AEDs but they obviously do not predict efficacy against
`difficult-to-treat (or pharmacoresistant) seizures4. Löscher and Rundfeldt148 were the first to describe a chronic model of pharmacoresistant seizures in which
`AED-resistant rats were selected from large groups of amygdala-kindled rats by repeated testing with phenytoin. Later, Löscher et al. also described the
`selection of AED-resistant subgroups of rats for post-status epilepticus models of temporal lobe epilepsy with SRS21,160.
`
`Limitations of previous strategies
`Despite the development of various new AEDs since the
`early 1990s, the available evidence indicates that there has
`been a failure to deliver drugs with improved efficacy4.
`What are the reasons for this apparent failure to dis-
`cover drugs that can effectively control drug-refractory
`seizures and comorbidities as well as prevent or modify
`the disease?
`
`Problems with preclinical models. Simple seizure models
`such as the MES and PTZ tests in rodents have been
`instrumental in the identification of most AED candi-
`dates. The advantages of such acute seizure models are
`their technical simplicity and the ability to screen large
`numbers of compounds. A disadvantage is that the sei-
`zures do not mirror epilepsy (that is, spontaneous seizure
`occurrence) and occur in ‘normal’, non-epileptic brains.
`Furthermore, older AEDs provide complete seizure sup-
`pression in these tests, hampering the identification of
`new AED candidates with greater efficacy, including
`those that might be effective in patients who are resistant
`to the older drugs.
`More recently, large AED screening programmes
`such as the Anticonvulsant Screening Project (ASP) of
`the National Institute of Neurological Disorders and
`Stroke (NINDS) of the US National Institutes of Health
`(NIH), which was initiated in 1975 to stimulate the dis-
`covery and development of new chemical entities for
`the symptomatic treatment of human epilepsy29,30, have
`included models for pharmacoresistant partial seizures
`in drug screening. One particular model is the 6-Hz
`mouse test, which was also introduced to avoid missing
`
`the identification of compounds like levetiracetam;
`levetiracetam is ineffective in the MES and PTZ models
`but is among the most effective AEDs in the clinic16,25,31.
`However, although several novel AEDs — including bri-
`varacetam, retigabine (Potiga; Valeant Pharmaceuticals/
`GlaxoSmithKline) and carisbamate — are highly effective
`in the 6-Hz mouse model, they are not more effective in
`patients with pharmaco resistant partial seizures21.
`Thus, it seems that the simple acute seizure screening
`models used in the ASP and other programmes fail
`to differentiate between compounds with promising
`potential for efficacy against drug-resistant seizures and
`compounds that work through mechanisms that are not
`detected by these models. Importantly, chronic seizure
`models, such as the lamotrigine-resistant kindled rat32,
`in which seizures are induced in animals with chronic
`brain alterations, were therefore recently included in the
`ASP. However, none of the emerging models of therapy-
`resistant epilepsy (FIG. 1 (TIMELINE)) has actually been
`validated at predicting clinical success in the therapy-
`resistant patient population. Thus, it remains to be estab-
`lished whether the use of chronic models such as kindling
`or models with spontaneous recurrent seizures will lead
`to the identification of more effective anti-epileptic treat-
`ments, but we consider this approach to be much more
`viable than the exclusive use of simple acute seizure
`models, particularly when testing hypothesis-driven,
`target-based strategies of drug development21.
`
`Problems with broad-spectrum approaches. An impor-
`tant aim of previous research and development (R&D)
`efforts was to discover novel AEDs that exert a broad
`
`760 | OCTOBER 2013 | VOLUME 12
`
` www.nature.com/reviews/drugdisc
`
`© 2013 Macmillan Publishers Limited. All rights reserved
`
`

`
`R E V I E W S
`
`Table 1 | Characteristics of clinically approved AEDs*
`AED
`Companies
`Year of
`Presumed main
`approval
`mechanisms of
`action
`
`Approved indications Main utility
`
`Main limitations
`
`First-generation drugs
`Potassium
`Dow
`bromide
`
`1857‡
`
`GABA potentiation?
`
`GTCS, myoclonic
`seizures
`
`Broad use for focal
`and generalized
`seizures
`
`Phenobarbital
`
`Bayer
`
`1912‡
`
`GABA potentiation
`
`Phenytoin
`
`Parke-Davis/
`Pfizer
`
`1938
`
`Sodium channel
`blocker
`
`PGCS, sedation,
`anxiety disorders, sleep
`disorders
`PGCS
`
`Broad use for focal
`and generalized
`seizures
`First-line AED, i.v. use
`
`Trimethadione
`
`Abbott
`
`1946
`
`T-type calcium
`channel blocker
`
`Absence seizures
`
`Rare use for absence
`seizures
`
`Primidone
`
`Imperial
`Chemical
`Industries
`
`1954
`
`GABA potentiation
`
`PGCS
`
`Broad use for focal
`and generalized
`seizures
`
`Ethosuximide
`
`Parke-Davis/
`Pfizer
`
`1958
`
`T-type calcium
`channel blocker
`
`Absence seizures
`
`First-line AED, no skin
`hypersensitivity
`
`Second-generation drugs
`Diazepam
`Roche
`
`1963
`
`GABA potentiation
`
`Convulsive disorders,
`status epilepticus,
`anxiety, alcohol
`withdrawal
`
`Broad use for focal and
`generalized seizures,
`i.v. use, no clinical
`hepatotoxicity, no skin
`hypersensitivity
`
`Carbamazepine Novartis
`
`1964
`
`Sodium channel
`blockade
`
`PGCS, trigeminal pain,
`bipolar disorder
`
`First-line AED
`
`Valproate
`
`Sanofi/Abbott
`
`1967
`
`Clonazepam
`
`Roche
`
`1968
`
`Multiple (for example,
`GABA potentiation,
`glutamate (NMDA)
`inhibition, sodium
`channel and T-type
`calcium channel
`blockade)
`GABA potentiation
`
`Clobazam
`
`Hoechst Roussel/
`Lundbeck/Sanofi
`
`1975
`
`GABA potentiation
`
`PGCS, absence
`seizures, migraine
`prophylaxis, bipolar
`disorder
`
`Broad use for focal
`and generalized
`seizures, first-line
`AED, i.v. use, no skin
`hypersensitivity
`
`Lennox–Gastaut
`syndrome, myoclonic
`seizures, panic
`disorders
`
`Broad use for focal
`and generalized
`seizures, no clinical
`hepatotoxicity
`
`Lennox–Gastaut
`syndrome, anxiety
`disorders
`
`Broad use for focal
`and generalized
`seizures, no clinical
`hepatotoxicity
`
`Currently for
`adjunctive use only,
`not in wide use
`anymore; acts as a
`sedative
`Enzyme inducer;
`skin hypersensitivity;
`no absence seizures
`Enzyme
`inducer; skin
`hypersensitivity;
`NLPK; not useful
`for absence or
`myoclonic seizures
`Not in wide
`use anymore;
`teratogenic
`Enzyme
`inducer; skin
`hypersensitivity; no
`absence seizures;
`acts as a sedative
`Somnolence, loss
`of appetite, nausea,
`vomiting, singultus,
`depression,
`psychotic episodes,
`insomnia, rare
`aplastic anaemia
`
`Currently for
`adjunctive use
`only; emergency
`use only; acts as a
`sedative; leads to
`tolerance (loss of
`efficacy)
`Enzyme
`inducer; skin
`hypersensitivity;
`not useful for
`absence or
`myoclonic seizures
`Enzyme inhibitor;
`substantial
`teratogenicity;
`weight gain
`
`Currently for
`adjunctive use only;
`acts as a sedative;
`leads to tolerance
`(loss of efficacy)
`Currently for
`adjunctive use only;
`acts as a sedative;
`leads to tolerance
`(loss of efficacy)
`
`NATURE REVIEWS | DRUG DISCOVERY
`
` VOLUME 12 | OCTOBER 2013 | 761
`
`© 2013 Macmillan Publishers Limited. All rights reserved
`
`

`
`R E V I E W S
`
`Table 1 (cont.) | Characteristics of clinically approved AEDs*
`AED
`Companies
`Year of
`Presumed main
`approval
`mechanisms of
`action
`
`Approved
`indications
`
`Main utility
`
`Main limitations
`
`Third-generation drugs
`Progabide
`Synthelabo
`
`1985
`
`GABA potentiation
`
`Vigabatrin
`
`Sanofi/Lundbeck
`
`1989
`
`GABA potentiation
`
`Lamotrigine
`
`GlaxoSmithKline
`
`1990
`
`Oxcarbazepine Novartis
`
`1990
`
`Sodium channel
`blocker
`
`Sodium channel
`blocker
`
`PGCS, Lennox–Gastaut
`syndrome, myoclonic
`seizures, muscle
`hypertonia
`Infantile spasms,
`complex partial
`seizures (currently for
`adjunctive use only)
`
`Rarely used for focal
`seizures
`
`No clinical
`hepatotoxicity
`
`PGCS, Lennox–Gastaut
`syndrome, bipolar
`disorder
`Partial seizures
`
`Broad use for focal
`and generalized
`seizures, first-line AED
`First-line AED
`
`Felbamate
`
`Carter-Wallace/
`MedPointe
`Pharmaceuticals
`
`1993
`
`Multiple (GABA
`potentiation,
`glutamate (NMDA)
`inhibition, sodium
`and calcium channel
`blockade)
`
`PGCS, Lennox–Gastaut
`syndrome
`
`Broad use for focal
`and generalized
`seizures
`
`Gabapentin
`
`Parke-Davis/
`Pfizer
`
`1993
`
`Calcium channel
`blocker (α2δ subunit)
`
`PGCS, postherpetic
`and diabetic neuralgia,
`restless legs syndrome
`
`No clinical
`hepatotoxicity
`
`Topiramate
`
`Janssen/Johnson
`& Johnson
`
`1995
`
`Tiagabine
`
`Novo Nordisk
`
`1996
`
`Multiple (GABA
`potentiation,
`glutamate (AMPA)
`inhibition, sodium
`and calcium
`channel blockade)
`GABA potentiation
`
`PGCS, Lennox–Gastaut
`syndrome, migraine
`prophylaxis
`
`Partial seizures
`
`Levetiracetam
`
`UCB Pharma
`
`2000
`
`SV2A modulation
`
`PGCS, partial seizures,
`GTCS, JME
`
`Zonisamide
`
`Elan/Eisai
`
`2000
`
`Sodium channel
`blocker
`
`Partial seizures
`
`Stiripentol
`
`Biocodex
`
`2002
`
`Pregabalin
`
`Pfizer
`
`2004
`
`GABA potentiation,
`sodium channel
`blocker
`Calcium channel
`blocker (α2δ subunit)
`
`Dravet syndrome
`
`Partial seizures,
`neuropathic pain,
`generalized
`anxiety disorder,
`fibromyalgia
`
`Broad use for focal
`and generalized
`seizures, first-line
`AED, no clinical
`hepatotoxicity
`
`No clinical
`hepatotoxicity
`
`First-line AED, i.v.
`use, no clinical
`hepatotoxicity
`Broad use for focal
`and generalized
`seizures, no clinical
`hepatotoxicity
`No clinical
`hepatotoxicity
`
`No clinical
`hepatotoxicity
`
`Clinical
`hepatotoxicity,
`not in wide use
`anymore
`Not useful for
`absence or
`myoclonic seizures;
`vision loss;
`weight gain
`Enzyme
`inducer; skin
`hypersensitivity
`Enzyme
`inducer; skin
`hypersensitivity;
`not useful for
`absence or
`myoclonic seizures
`Currently for
`adjunctive use
`only; aplastic
`anaemia; clinical
`hepatotoxicity; skin
`hypersensitivity;
`clinical
`hepatotoxicity;
`not in wide use
`anymore
`Currently for
`adjunctive use only;
`weight gain; not
`useful for absence
`or myoclonic
`seizures
`Somno