Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 1/19
`
`

`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 2/19
`
`

`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 3/19
`
`

`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 4/19
`
`

`
`Clinical Neuropharmacology
`Vol. 26, No. 1, pp. 38–52
`© 2003 Lippincott Williams & Wilkins, Inc., Philadelphia
`
`Clinical Pharmacology of Antiepileptic Drugs
`
`Carl W. Bazil and Timothy A. Pedley
`
`The Comprehensive Epilepsy Center, The Neurological Institute of New York, Columbia University, New York, New York, USA
`
`Antiepileptic drugs (AEDs) are the mainstay of treat-
`ment for the majority of patients with epilepsy. How-
`ever, not all patients with seizures require treatment
`with AEDs, and many patients with some forms of epi-
`lepsy, especially children, need only 1 or 2 years of
`drug therapy (1). Thus, important questions for the
`treating physician to have in mind are “When to start
`AEDs?” and “When can AEDs be safely withdrawn?”
`Nonetheless, it is safe to say that, as a group, patients
`with epilepsy typically receive one or more drugs, usu-
`ally for many years. Therefore, it is essential that phy-
`sicians have a good understanding of the clinical phar-
`macology of these agents, including their mechanisms
`of action, utilization patterns, and potential for interac-
`tions with other therapeutic agents.
`
`SELECTION OF ANTIEPILEPTIC DRUGS
`
`Table 1 lists the drugs currently used to manage dif-
`ferent types of seizures. Table 2 gives the usual dosage,
`effective plasma concentration, half-life, and common
`side effects for the most commonly used of these agents
`(2).
`Carbamazepine, phenytoin, primidone, and pheno-
`barbital are equally effective for partial and secondarily
`generalized seizures, although one may be effective
`when another is not (3). Valproate is also as effective
`for secondarily generalized seizures but is somewhat
`less effective than carbamazepine (and, presumably,
`phenytoin) in managing simple or complex partial sei-
`zures (4). Despite relatively equal antiepileptic po-
`tency, however, these drugs differ substantially in
`terms of side effects, pharmacokinetic properties, and
`cost. Similar comparative data are generally lacking for
`
`AEDs introduced after 1994. Phenytoin, with its rela-
`tively long half-life, which usually allows the drug to
`be given once or twice daily after midadolescence, has
`traditionally been preferred to drugs with shorter half-
`lives (5). However, concern about phenytoin’s occa-
`sional undesirable cosmetic effects (gingival hypertro-
`phy, hirsutism, and coarsening of facial features) and
`other adverse consequences associated with long-term
`therapy have led to the preferential use of carbamaze-
`pine or one of the newer drugs (e.g., lamotrigine, topi-
`ramate, zonisamide) as initial treatment in many pa-
`tients today. Carbamazepine, despite its short half-life,
`can be given twice daily if an extended-release formu-
`lation (e.g., Tegretol-XR, Carbatrol) is used. Phenobar-
`bital and primidone are rarely used now, except in spe-
`cial circumstances, because of the high incidence of se-
`dation and cognitive side effects at therapeutic doses
`and plasma concentrations (3).
`As a group, generalized-onset seizures respond best
`to valproate (6). Valproate can be used effectively as
`monotherapy in about 80% of patients, even when sev-
`eral types of generalized-onset seizure coexist. La-
`motrigine and topiramate are appropriate alternatives
`when valproate fails (7). A growing number of neurolo-
`gists prefer lamotrigine to valproate because of a more
`favorable adverse effect profile. Phenytoin and carba-
`mazepine are also effective against generalized tonic–
`clonic seizures, but the response is less reliable than
`with valproate (2). However, they are ineffective
`against absence or myoclonic seizures, which com-
`monly coexist with generalized tonic–clonic seizures,
`which means another drug (e.g., ethosuximide, clonaz-
`epam) must be used. Furthermore, phenytoin, gabapen-
`tin, and especially carbamazepine can actually aggra-
`vate nonconvulsive generalized-onset seizures (7).
`
`Most of the material contained in this review was presented at the 54th
`Annual Meeting of the American Academy of Neurology and was in-
`cluded in a syllabus prepared for the Therapy in Neurology course given
`on April 15, 2002. It is reproduced here with the permission of the
`American Academy of Neurology.
`Address correspondence and reprint requests to Dr. Timothy A.
`Pedley, 710 W. 168th St., The Neurological Institute of New York,
`Columbia University, New York, NY 10032, USA; E-mail: tap2@
`columbia.edu.
`
`MONOTHERAPY OR POLYTHERAPY?
`
`Medical treatment of epilepsy in adults is generally
`less satisfactory than in children, especially when con-
`fronted with partial and secondarily generalized sei-
`zures. Fewer than 50% of such patients with recurrent
`seizures remain seizure free for more than 12 consecu-
`
`38
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 5/19
`
`

`
`CLINICAL PHARMACOLOGY OF ANTIEPILEPTIC DRUGS
`
`39
`
`TABLE 1. AEDs used in treating different types of seizures
`
`Seizure type
`
`Effective drugs
`
`BASIC PRINCIPLES OF CLINICAL
`PHARMACOLOGY
`
`Partial seizures (including
`secondarily generalized)
`
`Broad spectrum (all seizure
`types, including partial,
`absence, myoclonic, tonic, and
`both primary and secondarily
`generalized tonic-clonic)
`Absence only
`Infantile spasms
`
`* not available in the U.S.
`
`carbamazepine, phenytoin,
`gabapentin, tiagabine,
`oxcarbazepine, phenobarbital,
`primidone, levetiracetam,
`zonisamide, topiramate,
`lamotrigine, valproic acid,
`felbamate
`valproic acid, lamotrigine,
`topiramate, felbamate,
`zonisamide, levetiracetam (?)
`
`ethosuximide
`valproic acid, vigabatrin*,
`zonisamide
`
`tive months (8). In the nationwide collaborative VA
`Cooperative Study, monotherapy resulted in satisfac-
`tory control of seizures in about 60% of patients. Of
`those in whom the first drug was ineffective, about 55%
`responded to an alternative drug used alone. Of the re-
`maining patients, only half had better control of sei-
`zures with two drugs. Moreover, improved control of
`seizures with two or more drugs was usually accompa-
`nied by increased side effects. These were frequently
`the result of drug interactions, which were not neces-
`sarily reflected in “toxic” blood concentrations of ei-
`ther drug. A more recent study by Kwan and Brodie
`included newer antiseizure drugs (7,8). Forty-seven
`percent of patients treated with a first drug became sei-
`zure free without adverse effects. About one third of
`patients who failed to respond to treatment subse-
`quently became seizure free on a second drug if the first
`had been discontinued because of side effects or idio-
`syncratic reactions. Seizure-free rates were not differ-
`ent between add-on and substitution therapies. These
`results, like the earlier studies, also support trying a sec-
`ond drug in monotherapy even when the first drug fails.
`
`Pharmacodynamics
`
`Following absorption, a drug distributes between the
`plasma and various tissue compartments (Fig. 1). Be-
`cause the majority of AEDs are at least partially bound
`to serum proteins, equilibrium exists between the con-
`centration of protein-bound drug and the unbound
`(“free”) drug concentration in plasma. The concentra-
`tion of free drug in plasma is, in turn, in equilibrium
`with the drug concentration in extracellular fluid. Only
`free drug is capable of crossing the various lipoprotein
`membranes that surround receptor sites and, therefore,
`only that portion of the total drug concentration that is
`free is available to produce the desired effect.
`It is not possible to monitor drug concentrations at
`receptor sites directly in humans, but total plasma drug
`concentrations (the usual clinical measurement) reflect
`the equilibrium among concentrations in tissue, extra-
`cellular fluid, and plasma (including that portion which
`is protein bound). Plasma free levels reflect concentra-
`tions in the extracellular space (9).
`The mechanism of action of a drug refers to the ac-
`tual biochemical or physical process that is modified or
`elicited to produce a biologic response at a particular
`site (pharmacodynamic effect). In the case of AEDs,
`this refers to the action produced at a cellular level lead-
`ing to decreased epileptic excitability. However, the ac-
`tual clinical utility of any AED is affected not only by
`its mechanism of action but also by its utilization pat-
`tern, any active metabolites that are formed, its perme-
`ability across the blood–brain barrier, and its therapeu-
`tic index.
`
`Pharmacokinetics
`
`The plasma concentration that is achieved and main-
`tained following the administration of a fixed drug dose
`
`Drug
`
`Phenytoin (Dilantin)
`Carbamazepine (Tegretol, Carbatrol)
`Phenobarbital
`Valproate (Depakote, Depakene)
`Ethosuxamide (Zarontin)
`Felbamate (Felbatol)
`Gabapentin (Neurontin)
`Lamotrigine (Lamictal)
`Topiramate (Topamax)
`Tiagabine (Gabitril)
`Oxcarbazepine (Trileptal)
`Levetiracetam (Keppra)
`Zonisamide (Zonegran)
`
`TABLE 2. Pharmacokinetics of AEDs
`
`Usual
`adult dose
`(mg/day)
`
`300–400
`800–1600
`90–180
`1000–3000
`750–1500
`2400–3600
`1800–3600
`100–500
`200–400
`32–56
`600–1800
`1000–3000
`100–400
`
`Half-life
`(h)
`
`22
`8–22
`100
`15–20
`60
`14–23
`5–7
`12–60b
`19–25
`5–13
`8–10c
`6–8
`63
`
`Metabolism
`
`>90% hepatic with induction
`>90% hepatic with induction
`>90% hepatic with induction
`>95% hepatic with inhibition
`65% hepatic, no induction
`60% hepatic
`>95% renal
`>90% hepatic, no induction
`30% hepatic, no induction
`>90% hepatic, no induction
`>90% hepatic, mild induction
`>65% renal excretion
`70% hepatic, no induction
`
`Usual effective plasma
`concentration
`(µg/mL)
`
`Time to peak
`concentration
`(h)
`
`Bound
`fraction
`(%)
`
`10–20
`8–12
`15–40
`50–120
`40–100
`20–140a
`4–16a
`2–16a
`4–10
`NE
`10–35c
`5–45a
`10–40a
`
`3–8
`4–8
`2–8
`3–8
`3–7
`2–6
`2–3
`2–5
`2–4
`1
`3–13
`1
`2–6
`
`90–95
`75
`45
`80–90
`<5
`25
`<5
`55
`9–17
`95
`40c
`<10
`40
`
`Clin. Neuropharmacol., Vol. 26, No. 1, 2003
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 6/19
`
`

`
`40
`
`C. W. BAZIL AND T. A. PEDLEY
`
`FIG. 1. Principles of antiepileptic
`drug monitoring. From C.E. Pippenger
`and Ronald P. Lesser. Current Trends
`in Epilepsy–Unit II, copyright Epilepsy
`Foundation of America. Reproduced
`with permission of the authors and the
`Epilepsy Foundation.
`
`mechanisms become saturated. It is important to note
`than even when the initial portion of the dose–response
`curve seems to be linear (as in Fig. 2B), drug clearance
`is altered throughout the dosage range and at all plasma
`concentrations. Kinetics in this portion of the range
`does not truly parallel the kinetics observed in a first-
`order relationship.
`Fortunately, in clinical practice, only a few drugs ex-
`hibit zero-order kinetics. Phenytoin is the most com-
`mon and important example of an AED exhibiting
`zero-order kinetics.
`
`(A) Dose-response curve for drug observing first-order
`FIG. 2.
`kinetics (linear). (B) Dose-response curve for drug observing
`zero-order kinetics (non-linear or saturation).
`
`is a direct consequence of the interactions of a wide
`variety of interrelated processes (Fig. 1), which, in turn,
`can be affected by a variety of factors, including other
`drugs, many illnesses (especially those affecting gas-
`trointestinal absorption, the liver, and kidneys), and
`physiologic changes (e.g., age, pregnancy).
`The study of these interrelationships forms the basis
`of pharmacokinetics. Pharmacokinetics is the study of
`the time-course of drug and metabolite concentrations
`in different fluids, tissues, and body wastes, and of the
`mathematical relations that can be utilized to develop
`models for interpretation of the blood concentrations
`measured in a given patient.
`The introduction and subsequent widespread avail-
`ability of reliable therapeutic drug monitoring has
`made application of pharmacokinetic principles an ex-
`pected and realistic part of contemporary medical
`therapeutics. An awareness of some of the fundamental
`principles is essential to rational prescribing practice
`and use of AEDs (10).
`
`Drug Kinetics
`
`First-order kinetics (line A in Fig. 2) describes a lin-
`ear relation between plasma drug concentration and to-
`tal daily dose (mg/kg). In other words, a predictable
`increase or decrease in plasma drug concentration oc-
`curs in response to a change in daily dosage.
`Zero-order kinetics (curve B in Fig. 2) describes a
`drug utilization process that is nonlinear, so that at
`some point plasma concentration (and clearance)
`seems to be independent of drug dose. Zero-order ki-
`netics becomes apparent when enzyme or transport
`
`Clin. Neuropharmacol., Vol. 26, No. 1, 2003
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 7/19
`
`

`
`CLINICAL PHARMACOLOGY OF ANTIEPILEPTIC DRUGS
`
`41
`
`Drug Half-Life
`
`Drug half-life (or elimination half-time) is the time
`required for elimination of half the concentration of a
`drug present in the system, provided no additional drug
`is administered after a specified point in time. The half-
`life of a drug may differ when it is administered to a
`naive patient compared with a patient taking other
`drugs. Drug half-life reflects the net interaction of sev-
`eral different processes that regulate drug clearance, of
`which rates of drug metabolism and excretion are the
`most important. Drug half-life determines the dosing
`interval for AEDs.
`
`Steady State
`
`When long-term oral therapy is initiated with a drug,
`it will continue to accumulate within the body until
`such time as rate of clearance, which comprises all tis-
`sue distribution, metabolic, and excretion processes in-
`volved in drug disposition, equals the rate of adminis-
`tration. When equilibrium is achieved between drug
`clearance and intake, the system is said to be at steady
`state. That is, the amount of drug ingested in a 24-hour
`period is equal to the amount of drug eliminated in the
`same 24-hour period. It requires seven half-lives of
`drug administration before a true steady-state concen-
`tration is achieved. For practical purposes, however,
`steady state is practically complete (97%) within five
`half-lives. Time to steady state determines how fast
`dose increases may be made for any AED.
`
`Absorption and Clearance
`
`Following oral administration, the type of drug
`preparation, drug solubility, concomitant administra-
`tion of other agents, whether the drug is taken with
`meals, or any gastrointestinal illness can all alter the
`amount of drug that will be absorbed from the stomach
`or small intestine. Time of maximum plasma drug con-
`centration (Tmax) is dependent on the rate of drug ab-
`sorption and elimination. Maximum concentration
`(Cmax) is the concentration at Tmax. The trough level is
`the concentration immediately preceding onset of ab-
`sorption of the next dose. Transient symptoms of neu-
`rotoxicity occurring at Tmax or seizures associated with
`trough levels of an AED may be verified by measuring
`the drug’s concentration at those times, and, if neces-
`sary, adjustments can be made in the timing or amount
`of a drug dose.
`Parenteral administration permits rapid entrance of a
`drug into the circulation and avoids problems associ-
`ated with drug absorption following oral administra-
`tion. Thus, parenteral administration is used when
`
`therapeutic blood levels need to be achieved rapidly
`and reliably, as, for example, in status epileptics.
`Clearance is a measurement of the amount of a drug
`eliminated during a period of time at steady state. If
`clearance (and ingestion) does not change, AED levels
`are relatively stable over time. This means that compli-
`ance can be estimated by obtaining serial measure-
`ments of plasma drug concentrations. Variation in se-
`quential measurements by more than 25% indicates the
`probability of noncompliance, if one has eliminated the
`possibility of intercurrent illness and use of other drugs.
`Some patients who have persistently low levels of an
`AED may have a genetic variation in hepatic metabo-
`lism (i.e., fast metabolizers). Low levels can also result
`from noncompliance (see preceding paragraph) or mal-
`absorption. Fast drug metabolism can be distinguished
`from malabsorption as a cause of persistently low
`plasma levels by measuring serial plasma drug concen-
`trations at given time intervals after parenteral admin-
`istration of the prescribed dose. If malabsorption is
`present, Cmax and observed drug half-life will be sig-
`nificantly higher following intravenous dosing than af-
`ter oral administration. On the other hand, if the patient
`problem is fast drug metabolism, there will be rela-
`tively little difference in plasma concentration or ob-
`served half-life regardless of the route of administra-
`tion. Fast drug metabolizers will require a larger daily
`dose (mg/kg) and often more frequent dosing to
`achieve the same serum concentration as a “normal”
`metabolizer. Conversely, someone who metabolizes
`drugs more slowly than usual will invariably exhibit
`drug toxicity at usual dosing schedules.
`
`Bioavailability
`
`Bioavailability is a measure of the amount of a dose
`absorbed by the body. Because of differences in manu-
`facturing processes, bioavailability of an active drug
`may vary from product to product. The majority of cur-
`rently available AEDs have narrow therapeutic ranges,
`and even small differences in bioavailability may be
`sufficient to produce toxicity or breakthrough seizures.
`Switching between dosage forms, especially in persons
`with poorly controlled seizures who have been titrated
`to their maximal tolerated dose, is usually not cost-
`effective because of the extra tests or clinic visits that
`are usually necessary.
`These and other factors that influence drug disposi-
`tion patterns in individual patients are listed in Table 3.
`Most of the factors influencing drug disposition and
`utilization in individual patients can be accounted for
`by appropriate therapeutic monitoring of AEDs. The
`indications for obtaining drug levels are summarized in
`Table 4.
`
`Clin. Neuropharmacol., Vol. 26, No. 1, 2003
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 8/19
`
`

`
`42
`
`C. W. BAZIL AND T. A. PEDLEY
`
`TABLE 3. Factors influencing drug disposition
`
`Patient compliance, including dosing error and wrong medication
`Absorption via route of administration
`Drug distribution
`Biotransformation
`Excretion
`Genetic variability
`Pathophysiologic factors (acute or chronic disease)
`Drug interactions
`Drug tolerance
`Inappropriate drug effects
`
`CHARACTERISTICS OF
`REPRESENTATIVE ANTIEPILEPTIC DRUGS
`
`Phenytoin
`
`Phenytoin (formerly diphenylhydantoin; Dilantin)
`was introduced clinically in 1938 following pioneering
`laboratory studies by Merritt and Putnam. For more
`than 60 years, phenytoin has been a drug of choice for
`management of partial and secondarily generalized
`tonic–clonic seizures.
`
`Administration and Bioavailability
`
`It is administered orally, preferably as the sodium
`salt in an extended release formulation. Approximately
`90% is absorbed after an oral dose, mainly from the
`duodenum, and peak concentrations are reached 6–12
`hours after administration. Maintenance dose is about
`4–6 mg/kg/day in adults; for children, doses of 8–10
`mg/kg/day are necessary. The common therapeutic
`range is 10–20 µg/mL.
`Parenteral phenytoin is available in a solution con-
`taining 50 mg/mL in 40% propylene glycol and 10%
`ethanol. The solution is highly alkaline (pH 12) and
`precipitates readily. It should be given by slow intrave-
`nous infusion, directly by the physician, at a rate not
`exceeding 50 mg/min. It should not be hung as an in-
`travenous solution, and Soluset or Volutrol dispensers
`should not be used. Phenytoin should never be given
`intramuscularly.
`
`TABLE 4.
`
`Indications for monitoring plasma drug levels
`
`To establish levels associated with optimal seizure control and/or
`development of toxicity for a drug with a narrow but well-defined
`therapeutic range (e.g., phenytoin, carbamazepine)
`When there is a discrepancy between the expected effect and the
`observed effect
`When non-compliance is suspected
`When interactions with other drugs is likely or suspected
`To compensate for changes in drug utilization caused by a secondary
`disease (e.g., uremia) or changing physiological state (e.g.
`pregnancy)
`When there is a need for medico-legal verification of treatment
`
`Clin. Neuropharmacol., Vol. 26, No. 1, 2003
`
`Fosphenytoin (Cerebyx) was introduced in 1996 as
`an alternative to parenteral phenytoin. Chemically, fos-
`phenytoin is a phosphate ester prodrug of phenytoin.
`Because it is highly water soluble, fosphenytoin may be
`administered in aqueous solutions; therefore, the side
`effect profile is improved compared with parenteral
`phenytoin. Administration is also more rapid; fosphe-
`nytoin may be administered at 150 mg phenytoin
`equivalents (PE) as opposed to a maximum rate of 50
`mg/min intravenous phenytoin. The conversion half-
`life is 8–15 minutes, which is independent of age. At
`these rates of administration, the two preparations give
`bioequivalent plasma free phenytoin concentrations.
`Unlike phenytoin, fosphenytoin may be administered
`intramuscularly. Plasma phenytoin generally reaches
`maximum about 30 minutes following intramuscular
`administration of fosphenytoin (12) (13).
`Dilantin brand phenytoin is available in 30-, 60-, and
`100-mg capsules, and in a 50-mg Infatab. The latter is
`not bioequivalent to the capsules.
`
`Metabolism and Pharmacokinetics
`
`Phenytoin is metabolized in the liver and undergoes
`parahydroxylation by the cytochrome P-450 system. Its
`metabolite, p-5-(hydroxyphenyl)-5-phenylhydantoin
`(HPPH), is inactive and nontoxic; it is excreted in the
`urine and bile conjugated with glucuronic acid. Less
`than 5% of phenytoin is excreted unchanged.
`Phenytoin is highly protein bound (>90%). Sub-
`stances that compete with it for binding sites or condi-
`tions that reduce its binding will increase the concen-
`tration of free drug.
`Fosphenytoin is more tightly bound to plasma pro-
`teins than is phenytoin. Therefore, patients who take
`phenytoin and are given a bolus of fosphenytoin may
`have a transient increase in free phenytoin as a result of
`displacement by fosphenytoin. This is unlikely to be
`clinically significant, however, as these patients will
`typically have low phenytoin levels, and any change
`will resolve as the fosphenytoin is converted into phe-
`nytoin.
`Use of phenytoin is complicated by its zero-order ki-
`netics. It is one of the few commonly used drugs that
`have substantially nonlinear elimination characteristics
`at therapeutic doses. There are several practical conse-
`quences. First, time to steady state concentration may
`be considerably longer than rough estimates based on a
`dose-independent half-life would suggest. For ex-
`ample, at concentrations greater than 20 µg/mL, half-
`life may be 36 hours or longer. Second, phenytoin’s
`steady state concentration at one dose does not directly
`predict the steady state concentration at another dose.
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 9/19
`
`

`
`CLINICAL PHARMACOLOGY OF ANTIEPILEPTIC DRUGS
`
`43
`
`This means that if a physician attempts to increase or
`decrease the steady state concentration by simple linear
`extrapolation from a known plasma level, the result is
`often an unexpectedly high (and possibly toxic) or low
`(and subtherapeutic) concentration. Furthermore, in
`any patient with toxic serum levels, one cannot reliably
`predict the time that will be required for the phenytoin
`concentration to decrease into the therapeutic range. Fi-
`nally, phenytoin’s nonlinear pharmacokinetics exag-
`gerates differences in bioavailability between the brand
`and generic preparations.
`
`Toxicity
`
`Neurotoxicity is dose related. Nystagmus appears
`first and is usually clinically insignificant. Higher
`plasma levels are associated with ataxia, dysarthria,
`diplopia, nausea, and a feeling of dysequilibrium.
`Higher levels still (usually greater than 35 µg/mL) are
`accompanied by mental dulling, drowsiness, and,
`rarely, a paradoxical increase in seizure frequency. Re-
`versible movement disorders, including chorea, dys-
`tonic, myoclonus, and asterixis, can occur, most often
`in patients with underlying brain damage. Cerebellar
`atrophy occurs with long-term phenytoin use, but the
`existence of clinically important chronic cerebellar
`damage remains controversial. Long-term phenytoin
`use can cause a peripheral neuropathy, but this is usu-
`ally clinically evident only as minor findings on exami-
`nation, such as depressed stretch reflexes and de-
`creased vibratory sensation in the feet.
`Limiting side effects, apart from rare idiosyncratic
`reactions, including Stevens-Johnson syndrome, are
`cosmetic: gingival hyperplasia, hirsutism, and coarsen-
`ing of facial features. These seem to occur most often in
`children, adolescents, and young adults. Hypocalcemia
`and osteomalacia may be seen as a result of phenytoin-
`induced alterations in vitamin D metabolism. Megalo-
`blastic anemia occurs from folate deficiency.
`
`Mechanism of Action
`
`At clinically useful concentrations, phenytoin inhib-
`its high-frequency repetitive firing by a use-dependent
`blockade of the voltage-dependent sodium channel. At
`somewhat higher concentrations (but still partially
`within the clinical range), phenytoin also blocks volt-
`age-dependent calcium channels, but the relevance of
`this to phenytoin’s anticonvulsant action is controver-
`sial. Phenomenologically, phenytoin is especially ef-
`fective in preventing seizure spread from a focus.
`
`Drug Interactions
`
`These are summarized later in a separate section.
`
`Carbamazepine
`
`Carbamazepine (Tegretol) is an iminostilbene, a
`dibenzepine derivative that is chemically and pharma-
`cologically related to tricyclic antidepressant agents.
`Its spectrum of action, clinically and in animal seizure
`models, is similar to that of phenytoin. Carbamazepine
`was approved for management of seizures in 1974, al-
`though it had been introduced a decade earlier for man-
`agement of trigeminal neuralgia (13).
`
`Administration and Bioavailability
`
`Carbamazepine is even less soluble in water than
`phenytoin, a factor that has inhibited development of a
`parenteral solution for clinical use. Therefore, at pres-
`ent, carbamazepine is available only for oral adminis-
`tration. It is available in 200- and 100-mg tablets and as
`an oral suspension (100 mg/5 mL).
`An extended release formulation, Tegretol XR, be-
`came available in the United States in 1996. It appears
`as a standard coated tablet but contains an osmotically
`active core with an opening on one side for release of
`drug. It is available in 100-, 200-, and 400-mg tablets.
`Since the tablet is specifically formulated for slow re-
`lease, these may not be crushed or cut. Although the
`drug is released, the outer core may not dissolve and
`can sometimes be seen in the stool.
`Carbamazepine is absorbed slowly and somewhat
`erratically, probably because of the slow dissolution
`rates of the compound. Tegretol brand is somewhat
`more predictable than the generic forms. Tmax for a
`single dose ranges from 4 to 16 hours, with peaks as late
`as 24 hours being reported occasionally. With long-
`term administration, peak plasma levels occur at 1.5
`hours with the oral suspension, at 4–5 hours with con-
`ventional tablets, and at 3–12 hours with the extended
`release formulation. Because of the lack of a parenteral
`formulation, precise bioavailability data are lacking;
`however, the XR tablets demonstrate an 89% bioavail-
`ability compared with the oral suspension and are
`equivalent to the conventional tablets.
`Maintenance dosage is usually about 10 mg/kg/day
`in adults and 15–20 mg/kg/day in young children. The
`usual therapeutic concentration is 8–12 µg/mL.
`
`Metabolism and Pharmacokinetics
`
`Once absorbed, carbamazepine is distributed rapidly
`and fairly uniformly to all body tissues. On average,
`about 75% of absorbed carbamazepine is protein
`bound.
`Carbamazepine is eliminated by biotransformation.
`Its major metabolite is carbamazepine 10,11-epoxide,
`
`Clin. Neuropharmacol., Vol. 26, No. 1, 2003
`
`Argentum Pharm. v. Research Corp. Techs., IPR2016-00204
`RCT EX. 2082 - 10/19
`
`

`
`44
`
`C. W. BAZIL AND T. A. PEDLEY
`
`which itself undergoes further metabolism. The epox-
`ide metabolite has weak anticonvulsant properties and
`is likely responsible for some of carbamazepine’s neu-
`rotoxicity. Less than 2% of carbamazepine is excreted
`unchanged.
`Carbamazepine’s clinical pharmacokinetic proper-
`ties are complicated by the phenomenon of autoinduc-
`tion of its own metabolism by hepatic microsomal en-
`zymes. In naive patients, introduction of the drug is as-
`sociated with a half-life of about 30 hours. Within days
`to a few weeks, the half-life decreases to 10–20 hours.
`This process stabilizes at about 30 days. The clinical
`effect of autoinduction means that introducing the drug
`can be tricky. Therapy should be initiated with low
`doses (100 mg twice daily) and slowly increased every
`3–5 days. Another regimen that works in most patients
`is to give an initial dose of 6–8 mg/kg followed by 25%
`of that dose on day 2, 50% on days 3 and 4, and 75% on
`days 5–7. The initial dosing regimen recommended in
`the PDR is almost guaranteed to produce toxicity in the
`majority of patients. Once autoinduction is complete,
`the relatively short and variable half-life requires fre-
`quent daily dosing (at least three times daily, although
`many patients require or do better on four-times-daily
`dosing schedules). Tegretol XR and Carbatrol can be
`administered twice daily.
`
`Toxicity
`
`An acute dysequilibrium syndrome, fatigue, and
`drowsiness occur when treatment is started too rapidly.
`Dose-related, reversible neurotoxicity includes blurred
`or double vision, nystagmus, dizziness, headache, and
`incoordination. Mild symptoms may appear early in the
`course of treatment during dose adjustment, but toler-
`ance may develop. Similar symptoms can occur tran-
`siently with peak plasma concentrations following a
`dose (Cmax). Dystonia and chorea occur rarely, usually
`in the setting of polypharmacy and brain damage, but
`less often than with phenytoin.
`Skin rashes are more common with carbamazepine
`(15%) than with phenytoin (8%). Life-threatening id-
`iosyncratic reactions (e.g., Stevens-Johnson syndrome)
`are very rare.
`Mild elevations of hepatic enzymes (SGOT, SGPT)
`occur in 5–10% of treated patients and are of no clinical
`significance. Such mild abnormalities neither predict
`nor predispose to hepatitis, which is a hypersensitivity
`reaction presumably mediated by immune factors.
`Similarly, about 25–30% of patients treated with car-
`bamazepine develop transient leukopenia, which per-
`sists in about 10%. This benign leukopenia has no clini-
`cal significance and does not correlate with the rare
`cases (only 30 reported) of aplastic anemia.
`
`Clin. Neuropharmacol., Vol. 26, No. 1, 2003
`
`Carbamazepine has antidiuretic effects that result in
`hyponatremia and water retention. This may reflect re-
`nal and hypothalamic effects (anti-ADH). Mild to mod-
`erate hyponatremia is generally unimportant, although
`it sometimes relates to patient complaints of dizziness,
`headache, and nausea. Severe hyponatremia (<122
`mEq/L) and water retention can exacerbate seizures or
`contribute to congestive heart failure in patients with
`compromised cardiac function.
`
`Mechanism of Action
`
`Like phenytoin, carbamazepine inhibits voltage-
`dependent sodium channels at clinically relevant con-
`centrations. Carbamazepine also has high affinity for
`adenosine receptors, but it is unlikely that adenosine-
`receptor binding accounts for its anticonvulsant activ-
`ity.
`
`Drug Interactions
`
`These are summarized later in a separate section.
`
`Valproate
`
`Anticonvulsant properties of valproate (available as
`valproic acid [Depakene] or divalproex sodium [Depa-
`kote]) were discovered accidentally when it was being
`used as a solvent in tests of other compounds being
`screened for anticonvulsant activity. The FDA ap-
`proved it for treatment of absence seizures in 1978, al-
`though valproate had been used widely in Europe for a
`decade before that. It has a broad spectrum of anticon-
`vulsant activity, being the drug of choice for all forms
`of primary generalized epilepsy and is also useful for
`partial and secondarily generalized seizures (14).
`
`Administration and Bioavailability
`
`Absorption is rapid with valproic acid, slower and
`more sustained with the enteric-coated preparation
`(Depakote) that is commonly prescribed today. Peak
`times range from 3 to 8 hours. Food delays but does not
`decrease the extent of absorption. Divalproex sodium is
`available in 250-mg and 500-mg tablets. The therapeu-
`tic range for plasma concentration is usually given as
`50–100 µg/mL, but, in fact, this varies considerably de-
`pending on many factors. Many patients, especially
`those on pol