E X P E RT O P I N I O N
`
`Carbamazepine extended-release capsules
`in bipolar disorder
`
`Richard H Weisler
`Duke University, Durham, North
`Carolina and the University of North
`Carolina, Chapel Hill, NC, USA
`
`Correspondence: Richard H Weisler,
`700 Spring Forest Suite 125, Raleigh,
`NC 27609, USA
`Tel + 1 919 872 5900
`Fax + 1 919 878 0942
`Email rweisler@aol.com
`
`Abstract: Carbamazepine (CBZ) has long been a therapeutic option for bipolar disorder.
`Carbamazepine extended-release capsules (CBZ-ERC) are a recent formulation of CBZ
`approved by the US Food and Drug Administration in 2004 for the treatment of acute manic
`and mixed episodes associated with bipolar I disorder. This new formulation was developed
`to improve dosing convenience and decrease daily fl uctuations in serum CBZ concentration,
`thereby lowering the incidence of adverse events. Two randomized, double-blind,
`placebo-controlled trials and an open-label extension study have demonstrated that CBZ-ERC
`monotherapy is effi cacious in patients with bipolar I disorder experiencing either manic or
`mixed episodes. In these trials, CBZ-ERC was shown to be a safe and well-tolerated therapy.
`Retrospective chart reviews conducted in private practice settings have shown that clinical
`response to CBZ-ERC is independent of bipolar subtype, as patients with bipolar I depression
`and bipolar II disorder responded similarly to patients with bipolar I disorder either manic or
`mixed episodes. CBZ is currently considered a treatment alternative to lithium and valproate
`according to the American Psychiatric Association’s treatment guidelines for patients with
`bipolar disorder. Although further study is required, the clinical evidence presented in these
`studies may change the treatment paradigm.
`Keywords: bipolar disorder, carbamazepine, mania, extended-release
`
`Introduction
`Bipolar disorder is usually a chronic illness with an episodic and variable course
`(APA 2002). It is a leading cause of years lived with disability (WHO 2004) and
`is associated with poor health-related quality of life, high utilization of health care
`services, work impairment (Dean et al 2004), and signifi cant costs to society (Wyatt
`et al 1995).
`There are many forms of bipolar disorder, ranging from mild depression
`and brief hypomania to one of severe depression or mania with psychotic fea-
`tures (Muller-Oerlinghausen et al 2002). To help identify the various forms of
`the illness, 4 subtypes of bipolar disorder have been defi ned based on patients’
`clinical characteristics: bipolar I disorder, bipolar II disorder, cyclothymia,
`and bipolar disorder not otherwise specifi ed (APA 2000). Past estimates of the
`prevalence of the full spectrum of bipolar disorders have ranged from 3% to
`6.5%, with about 1% considered to be the bipolar I subtype (Angst 1998; Stimmel
`2004). Furthermore, new data on the prevalence of bipolar spectrum disorders
`have emerged from the recently completed US National Comorbidity Survey
`Replication, in which trained professionals interviewed a nationally representative
`population of 9282 English-speaking US residents aged ⬎18 years. National
`Comorbidity Survey probands were assessed for the presence of Diagnostic
`and Statistical Manual of Mental Disorders (4th edition) mental disorders in the
`12 months prior to the survey using the World Mental Health Survey Initiative
`
`Neuropsychiatric Disease and Treatment 2006:2(1) 3–11
`© 2006 Dove Medical Press Limited. All rights reserved
`
`3
`
`
`Page 1 of 10
`
`JAZZ EXHIBIT 2015
`Amneal Pharms. LLC (Petitioner) v. Jazz Pharms. Ireland LTD. (Patent Owner)
`Case IPR2016-00546
`
`

`
`Weisler
`
`version of the World Health Organization Composite
`International Diagnostic Interview, and it was found
`that the 12-month prevalence of bipolar disorder
`(subtype I or II) in the study sample was 2.6% (Kessler
`2005).
`
`Treatment options
`Although psychotherapy is a critical component of inter-
`vention in bipolar disorder, pharmacotherapy is essential in
`treating patients who are acutely symptomatic (Suppes et al
`2005). The most recent version of the Texas Implementa-
`tion of Medical Algorithms (TIMA) calls for the treatment
`of acute manic, depressive, or mixed bipolar symptoms
`to enable patients to return to normal psychosocial func-
`tioning. According to the TIMA recommendations, acute
`pharmacotherapy should be initiated for mild to severe
`bipolar symptoms, with decisions on whether to maintain
`stable treatment dosing, adjust treatment dosing, or select
`an alternative pharmacotherapeutic strategy being made on
`the basis of regular follow-up assessments. Upon symptom
`remission after treatment of acute bipolar mania, a 4- to
`6-month course of continuation therapy and subsequent
`lifetime maintenance therapy with mood stabilizers are
`recommended for the prevention of recurrences (Suppes
`et al 2005).
`Despite systematic guidelines for the treatment of
`bipolar disorder, the selection of appropriate therapeutic
`agents for use in conjunction with these guidelines can be
`challenging, because the symptoms of bipolar disorder
`may be very different in different phases of the illness.
`Several agents are currently available for the treatment
`of bipolar disorder, and the initial treatment depends
`on the type of episode the patient is experiencing at
`presentation. Lithium has been extensively studied in
`the treatment of acute mania and as maintenance therapy
`since the discovery of its mood-stabilizing properties
`more than 50 years ago (Goodwin et al 1969; Belmaker
`2004), and it is currently considered to be a fi rst-line
`treatment option in acute mania and prophylaxis in
`bipolar disorder. Nonetheless, despite its fi rst-line status
`and its demonstrated overall effi cacy in these settings,
`lithium therapy has been shown to be ineffective or poorly
`tolerated in a signifi cant proportion of patients (Bowden
`et al 1994). In addition, it has a narrow therapeutic range
`and requires regular blood level monitoring, as severe
`or toxic effects can occur at twice the therapeutic dose.
`Its narrow therapeutic range is an especially important
`consideration in older patients as their renal excretion
`
`becomes less effi cient, resulting in an increased risk of
`lithium-associated toxic effects (Goodwin 2003; Belmaker
`2004).
`Although lithium is widely used globally, the anticon-
`vulsants valproate and carbamazepine (CBZ) have also
`become established therapeutic options in the treatment
`of bipolar disorder (Mitchell et al 2002; Suppes et al
`2005). Extensive clinical research has shown that these
`agents have mood-stabilizing properties and are there-
`fore effective treatments in the management of bipolar
`disorder (Mitchell et al 2002). The current review will
`examine data for a twice-daily extended-release capsule
`formulation of CBZ (CBZ-ERC) (Shire Pharmaceuticals,
`Wayne, PA, USA) that contains three different types of
`beads (immediate release, extended release, and enteric
`release). This particular formulation has been approved
`by the US Food and Drug Administration (FDA) for the
`treatment of acute manic and mixed episodes associated
`with bipolar I disorder.
`
`Rationale for development of
`CBZ-ERC
`Clinical evidence of the effi cacy of CBZ in the treatment
`of bipolar disorder emerged in the early 1970s. At this
`time, several small studies reported the antimanic effects
`of CBZ as well as its prophylactic effects against the
`recurrence of manic and depressive episodes in patients
`with bipolar disorder (Okuma et al 1973). Over the next
`several decades, many double-blind, controlled trials
`demonstrated the effi cacy of CBZ in the treatment of acute
`mania in bipolar disorder, with response rates similar to
`those of lithium (McElroy et al 2000). Until recently, all
`controlled evaluations of CBZ in bipolar disorder have
`used immediate-release formulations of CBZ that require
`dosing 3 or 4 times daily to avoid potentially problematic
`serum drug fl uctuations. Studies have shown that the large
`fl uctuations in serum CBZ levels observed with immediate-
`release CBZ formulations are associated with intermittent
`adverse effects such as diplopia, drowsiness, and headache
`in patients with epilepsy (Hoppener et al 1980; Riva et al
`1984). The established correlation between fl uctuations
`in serum CBZ levels and intermittent side effects helped
`prompt the development of extended-release formula-
`tions of CBZ for use in epilepsy. The subsequent use of
`extended-release CBZ in patients with epilepsy provided
`clinical evidence that, compared with immediate-release
`formulations, extended-release CBZ was associated with
`lower peak serum CBZ concentrations, decreased circadian
`
`4
`
`Neuropsychiatric Disease and Treatment 2006:2(1)
`
`Page 2 of 10
`
`

`
`toxicity, and decreased central nervous system (CNS) side
`effects (Canger et al 1990; Haefeli et al 1994).
`Non-adherence to medication is common among
`patients with bipolar disorder (Keck et al 1996; Svarstad
`et al 2001). Potential factors affecting compliance with
`medication include adverse effects and the demands of
`treatment, including frequent dosing regimens (Jamison
`et al 1983; Greenberg 1984). Since extended-release for-
`mulations of CBZ have been developed to decrease daily
`fl uctuations in serum CBZ concentrations and improve dos-
`ing convenience, several large clinical trials have recently
`been conducted to assess the effi cacy and tolerability of
`a novel, beaded, extended-release capsule formulation of
`CBZ in bipolar disorder (Weisler et al 2004c, 2005; Ketter
`et al 2004).
`
`Pharmacology
`Pharmacodynamics
`Although recent neurochemical and neuroimaging stud-
`ies have been promising, a specific pathophysiological
`abnormality in bipolar disorder has not been discovered
`(Belmaker 2004). Clinical and preclinical evidence suggests
`that second messenger systems and molecular “switches”
`known as G-proteins are involved in the underlying mecha-
`nisms that result in bipolar disorder (Gould et al 2002).
`A variety of medications used to treat bipolar disorder have
`widely varying mechanisms of action (MOA), which in
`many cases are also still not well understood. Alterations
`in postreceptor pathways, intracellular signaling, neural
`plasticity, and changes in gene expression are believed to
`play important roles in the therapeutic effects of these agents
`(Gould et al 2002).
`The MOA of CBZ in the treatment of bipolar disorder
`has not been clarifi ed. It has been well established that
`CBZ exerts its antiepileptic effects by inhibiting the
`high-frequency fi ring of sodium channels. This effect
`produces a functional blockage of voltage-gated sodium
`channels that could be associated with its mood-stabilizing
`properties (Gould et al 2004). In experimental studies,
`CBZ acts as an antagonist at adenosine receptors, which
`are generally G-protein-coupled receptors that modu-
`late neurotransmitter release and numerous behavioral
`and cognitive functions. Carbamazepine has also been
`reported to inhibit the enzyme adenylyl cyclase, which
`attenuates cyclic AMP-mediated signaling and may lead
`to inhibition of downstream activities with ion channels
`and transcription factors (Gould et al 2004). Additional
`fi ndings on the MOA of CBZ include reports of its effects
`
`CBZ-ERC in bipolar disorder
`
`on the voltage-gated ion channels in neurons where at
`therapeutic concentrations it acts as a calcium ion channel
`blocker (Ulrich et al 2003), and it exerts regulatory effects
`on receptor-mediated excitatory and inhibitory neurotrans-
`mission (Li et al 2002).
`
`Pharmacokinetics
`The extended-release formulation CBZ-ERC utilizes a drug
`delivery system consisting of a fi xed ratio of specialized
`beads to extend the release of CBZ beyond what can be
`achieved with conventional immediate-release formulations.
`Twenty-fi ve percent of these beads are designed to release
`drug immediately after swallowing for rapid absorption,
`40% of the beads are polymer-coated to dissolve gradually
`over 8–12 hours to achieve steady-state serum CBZ levels,
`and the remaining 35% are enteric-release beads with a
`pH-sensitive coating that releases CBZ slowly in the gut to
`maintain optimal blood levels. The timed release of CBZ is
`unaffected by variations in gastrointestinal (GI) transit time.
`Additionally, the capsule does not have to be taken with food;
`it can be opened and its contents sprinkled onto soft foods
`(McLean et al 2001).
`The pharmacokinetics of CBZ-ERC have been deter-
`mined following single and repeat dose administration.
`Following a single 200-mg dose of CBZ-ERC, peak plasma
`CBZ concentration was found to be 1.9 ± 0.3 μg/mL, and
`time to reach peak was 19 ± 7 hours. After repeated dose
`administration of CBZ-ERC 800 mg every 12 hours, the peak
`plasma CBZ concentration was 11.0 ± 2.5 μg/mL and time
`to reach peak was 5.9 ± 1.8 hours. The pharmacokinetics of
`CBZ-ERC are linear over a single-dose range of 200–800 mg
`(Shire 2005).
`CBZ is 76% bound to plasma proteins and is primarily
`metabolized in the liver. Cytochrome P450 3A4 was identi-
`fi ed as the major isoform responsible for metabolizing CBZ
`to its active metabolite CBZ-10,11-epoxide. Since CBZ
`induces its own metabolism, the half-life is variable. The
`average half-life can range from 35 to 40 hours following
`a single dose of CBZ-ERC and 12 to 17 hours following
`repeated dosing of CBZ-ERC.
`In a pharmacokinetic study conducted in patients
`with epilepsy, twice-daily CBZ-ERC was shown to be
`bioequivalent to immediate-release formulations of CBZ
`given four times daily (Garnett et al 1998). Minimizing
`the fl uctuations in serum CBZ concentrations inherent in
`immediate-release formulations was an important factor in
`the development of CBZ-ERC. Low peak-to-trough blood
`level variability ensures that blood CBZ levels remain
`
`Neuropsychiatric Disease and Treatment 2006:2(1)
`
`5
`
`Page 3 of 10
`
`

`
`Weisler
`
`relatively stable (Stevens et al 1998). In studies of CBZ
`in patients with epilepsy, conversion from immediate- to
`extended-release CBZ resulted in marked reductions in
`common dose-related CNS side effects. Pharmacokinetic
`analysis showed marked variability in absorption and blood
`drug concentrations with immediate-release formulations
`of CBZ compared with CBZ-ERC. These fi ndings suggest
`that the improved CNS tolerability observed in patients
`treated with CBZ-ERC is a result of smoother drug deliv-
`ery and reduced variability in absorption provided by the
`extended-release formulation (Miller et al 2004b). These
`clinical fi ndings, in addition to twice daily dosing with
`CBZ-ERC, should improve patient compliance by pro-
`viding a more convenient dosing regimen and improved
`tolerance with fewer side effects.
`Clinical studies
`Effi cacy
`For more than 30 years, CBZ has been used to treat
`bipolar disorder. Early controlled trials of CBZ in the
`treatment of acute mania were conducted with conven-
`tional immediate-release formulations and confounded by
`coadministration with lithium or standard antipsychotics.
`In the fi rst double-blind, placebo-controlled study of CBZ
`in affectively ill patients (ie, bipolar, unipolar depression,
`schizoaffective illness), preliminary results showed that
`7 of the fi rst 10 patients treated with immediate-release
`CBZ responded favorably with an antimanic, an antidepres-
`sant, or a prophylactic response (Ballenger et al 1978). In
`comparative studies, immediate-release CBZ was found to
`be as effi cacious as lithium in the treatment of acute mania
`in bipolar disorder (Okuma et al 1990; Small et al 1991;
`Emilien et al 1996). In one study, a 4-week, double-blind,
`multicenter trial involving 105 patients aged 13–65 years
`with manic or mixed bipolar disorder, 62% of patients in
`the immediate-release CBZ group showed moderate to
`marked amelioration of manic symptoms (as judged by
`the treating physician) at fi nal assessment, compared with
`59% in the lithium group (not signifi cant) (Okuma et al
`1990). In a separate 4-week, comparative, double-blind
`study, immediate-release CBZ monotherapy (n = 15) and
`valproate monotherapy (n = 15) were found to be compa-
`rably effective in the treatment of acute mania in patients
`hospitalized with bipolar disorder (Vasudev et al 2000).
`The primary effi cacy analysis in that study revealed that
`the mean Young Mania Rating Scale (YMRS) score in the
`valproate group decreased signifi cantly more than in the
`immediate-release CBZ group, although rates of response
`
`(ⱖ50% decrease in total YMRS score from baseline) were
`not signifi cantly different between the two groups, as favor-
`able clinical responses were attained by 53% of patients
`in the immediate-release CBZ group and 73% of patients
`in the valproate group.
`While the effi cacy of CBZ in treating acute mania asso-
`ciated with bipolar I disorder has been well characterized,
`evidence suggests that clinical responses to this agent are
`independent of symptomatology or bipolar subtype. In
`particular, a 21-day, open-label study involving 36 patients
`with bipolar disorder revealed that immediate-release CBZ
`reduced Hamilton Depression Rating Scale (HDRS) scores
`by an average of 72.7% (from 32.6 to 8.9; p ⬍ 0.0001 for
`change relative to baseline) between baseline and end point
`in the subset of patients with bipolar depression (n = 27)
`and by an average of 50.4% (from 35.1 to 17.4; p = 0.0009
`for change relative to baseline) in the subset of patients
`with depressive mania (n = 9) (Dilsaver et al 1996). More
`recently, a retrospective chart review of patients treated in
`a private practice setting found that CBZ-ERC also appears
`to be effective in reducing the symptoms of bipolar II dis-
`order (Ginsberg 2004a). Of the 111 patients treated with
`CBZ-ERC in that review (85% of whom were classifi ed
`as being “markedly ill”, “severely ill”, or “extremely ill”
`at the time of treatment initiation), 82 (74%) experienced
`a response to treatment as indicated by the achievement
`of a Clinical Global Impression-Improvement (CGI-I)
`score ⱕ3.
`Aside from exhibiting effi cacy in the acute treatment of
`bipolar disorder, CBZ has also been found to be effective
`as maintenance therapy in ⱖ10 controlled or partially
`controlled studies, with a cumulative “marked or excel-
`lent” response rate of approximately 61% (Denicoff et al
`1997). Several studies have compared the prophylactic
`effi cacy of immediate-release CBZ with that of lithium,
`which, in a pooled analysis of maintenance studies averag-
`ing 1.5 years in length, was found to reduce the likelihood
`of manic recurrence by a factor of 2.5 and the likelihood
`of depressive recurrence by a factor of 1.8 (Goodwin
`and Jamison 1990; Baldessarini et al 1996). Comparison
`studies involving immediate-release CBZ and lithium
`have found overall response rates to be similar, with no
`trends indicating one agent’s superiority to the other in
`the maintenance setting. A meta-analysis of clinical data
`from 10 double-blind, randomized, comparative trials
`of immediate-release CBZ vs lithium as maintenance
`therapy found a relapse rate of 60% in lithium-treated
`patients vs 55% of CBZ-treated patients; the difference
`
`6
`
`Neuropsychiatric Disease and Treatment 2006:2(1)
`
`Page 4 of 10
`
`

`
`CBZ-ERC in bipolar disorder
`
`Baseline
`
`Day 7
`
`Day 14
`
`Day 21
`
`*
`
`*
`
`*
`
`Placebo
`n = 213
`Baseline = 27.6
`End point = 21.4
`
`Treatment visit
`
`CBZ-ERC
`n = 214
`Baseline = 27.6
`End point = 15.4
`
`0
`
`–2
`
`–4
`
`–6
`
`–8
`
`–10
`
`–12
`
`–14
`
`Mean change in YMRS
`
`score (LOCF)
`
`Figure 1 Mean change in YMRS total scores in pooled analysis of 3-week studies of
`CBZ-ERC in acute mania.
`*p ⬍ 0.0001 compared with placebo following analysis of covariance with baseline
`score as covariate.
`Abbreviations: YMRS, Young Mania Rating Scale; LOCF, last observation carried
`forward; CBZ-ERC, carbamazepine extended-release capsules.
`
`analysis completed the 21-day study period, and the rate
`of discontinuation due to lack of effi cacy was found to be
`higher in the placebo group (22%) than in the CBZ-ERC
`group (10%).
`The primary effi cacy endpoint was the YMRS total score
`at the end of the double-blind treatment period. As shown in
`Figure 1, using the last observation carried forward analysis,
`CBZ-ERC-treated patients in the combined patient popula-
`tion had signifi cantly greater decreases in mean YMRS scores
`from baseline than those in the placebo group at day 7, 14,
`and day 21, the primary endpoint (p ⬍ 0.0001 at all time
`points). By the end of the trial, 52% of CBZ-ERC-treated
`patients in the combined population had responded (ⱖ50%
`reduction in YMRS score) vs 26% of placebo-treated patients
`(p ⬍ 0.0001). Patients receiving CBZ-ERC also experienced
`signifi cantly greater improvements in symptoms ratings on
`both CGI-Improvement (CGI-I) and CGI-Severity (CGI-S)
`scales than those treated with placebo (p ⬍ 0.0001) (Weisler
`et al 2004a).
`Subgroup analyses of the pooled data from these two
`pivotal studies showed that CBZ-ERC reduced both manic
`and depressive symptoms. At endpoint, there were signifi -
`cant reductions in mean YMRS total scores in both manic
`(p ⬍ 0.0001) and mixed (p ⬍ 0.01) bipolar patients treated
`with CBZ-ERC. Importantly, CBZ-ERC treatment led to sig-
`nifi cant improvement in the severity of depressive symptoms
`in mixed bipolar I patients (p ⬍ 0.05) and in the combined
`patient population (p = 0.01) as evidenced by signifi cant
`
`between the groups was not signifi cant (Davis et al 1999).
`In one recent randomized, double-blind study comparing
`the prophylactic efficacy of immediate-release CBZ
`and lithium in bipolar disorder, lithium was superior to
`immediate-release CBZ in terms of 2.5-year hospitalization
`rates (lithium, 26%; immediate-release CBZ, 62%;
`p = 0.012) in patients with “classical” bipolar I disor-
`der (without mood-incongruent features and psychiatric
`comorbidity and without mixed states; n = 67), while
`immediate-release CBZ showed a trend toward being
`more effective (2.5-year hospitalization rates: lithium,
`44%; immediate-release CBZ, 31%; p = 0.34) in the
`more heterogeneous “nonclassical” subgroup of patients
`(n = 104).These findings suggest that patients with
`classical features benefi t more from prophylactic therapy
`with lithium, while those with nonclassical features may
`benefi t more from prophylaxis with CBZ, which appears
`to provide a broader spectrum of activity (Greil et al 1998;
`Kleindienst et al 2000).
`
`Effi cacy of CBZ-ERC in acute mania
`Recently, two large 3-week, double-blind, randomized,
`placebo-controlled, multicenter trials demonstrated that
`monotherapy with twice-daily CBZ-ERC was effective in
`the treatment of acute mania in patients with bipolar I dis-
`order experiencing manic or mixed episodes (Weisler et al
`2004c, 2005). To be eligible for enrollment in these studies,
`patients had to be ⱖ18 years old and meet DSM-IV criteria
`for bipolar I disorder with current episode manic or mixed
`and with manic symptoms severe enough to necessitate
`hospitalization. A history of ⱖ1 previous manic or mixed
`episode and a minimum baseline total score of 20 on the
`YMRS were required. The initial 400-mg dose of CBZ-ERC
`or placebo could be titrated by the investigators between
`200 mg and 1600 mg in daily increments of 200 mg.
`Effi cacy assessments were carried out weekly and included
`the YMRS, the Clinical Global Impression (CGI) scale,
`and the HDRS.
`To better elucidate the results from these two
`similar trials, the data were combined and the effi cacy
`and safety of CBZ-ERC were evaluated based on the
`combined study population. Pooled data from these
`two pivotal studies (which took place across a total of
`40 study sites – 34 in the United States and 6 in India)
`included a total of 443 patients, 223 of whom were
`randomized to double-blind treatment with CBZ-ERC
`and 220 of whom were randomized to receive placebo.
`A total of 240 randomized patients (54%) in the pooled
`
`Neuropsychiatric Disease and Treatment 2006:2(1)
`
`7
`
`Page 5 of 10
`
`

`
`18
`
`16
`
`14
`
`12
`
`10
`
`68
`
`Mean HDRS score (LOCF)
`
`–28 Days
`
`Day 1
`
`Month 1
`
`Month 2
`
`Month 4 Month 6
`
`Figure 3 HDRS total scores from the 6-month CBZ-ERC study including baseline
`score in the pooled analysis of the 3-week studies.
`*p ⬍0.0001, †p = 0.0004, ‡p = 0.0002 based on one sample t-test of mean change from
`baseline scores of pooled analysis from the 3-week studies.
`Notes: When scores were based on one sample t-test of mean change from baseline
`values at day 1 in 6-month study, no statistical signifi cance was detected.
`Abbreviations: HDRS, Hamilton Depression Rating Scale; LOCF, last observation
`carried forward; CBZ-ERC, carbamazepine extended-release capsules.
`
`Tolerability and patient
`acceptability
`The pooled analysis of safety data from the two 3-week
`trials indicated that CBZ-ERC was generally well tolerated
`in both manic and mixed bipolar patients. The most fre-
`quently reported treatment-emergent adverse events among
`CBZ-ERC-treated patients (with an incidence ⱖ5% and at
`least twice the rate of placebo patients) included dizziness,
`somnolence, nausea, vomiting, ataxia, and pruritis (Weisler
`et al 2004b). The most common adverse events observed
`with long-term CBZ-ERC therapy were typical of CBZ and
`included headache, dizziness, and mild rash (Ketter et al
`2004).
`Aplastic anemia and agranulocytosis have been
`observed rarely with CBZ, and no incidences were reported
`with CBZ-ERC treatment in both the acute and long-term
`studies. There was no clinically signifi cant change in
`mean weight gain (ⱖ7% increase from baseline) among
`the pooled study population in 3-week studies or in the
`6-month trial, although a small percentage (5%) of patients
`did experience clinically signifi cant weight gain during
`the trials (Ketter et al 2004; Weisler et al 2004, 2005). No
`statistically signifi cant alterations in electrocardiogram
`variables (QTc) and no meaningful change in non-fasting
`blood glucose were reported with long-term CBZ-ERC
`therapy (Ketter et al 2004).
`Total plasma cholesterol (TC) concentrations have been
`shown to increase with CBZ-ERC treatment. In one of the
`
`Weisler
`
`decreases in HDRS total scores. The reduction in depressive
`symptoms experienced by patients treated with CBZ-ERC
`therapy in this trial suggests that CBZ may have true bimodal
`effects in bipolar disorder (Weisler et al 2004a).
`As little is known about the effi cacy of CBZ-ERC in
`long-term treatment, a 6-month open-label extension study
`was conducted among patients who had completed one of
`two previous 3-week, double-blind, placebo-controlled
`evaluations of CBZ-ERC for acute mania (Ketter et al
`2004). Ninety-two patients were enrolled in the study; 67%
`were mixed and 33% manic. Of the 77 patients treated with
`CBZ-ERC and evaluated in the intent-to-treat population,
`14.3% relapsed during the 6-month study (7 had received
`prior treatment with CBZ-ERC and 4 with placebo). The
`observed mean time to relapse was approximately 2 months.
`Interestingly, a relapse rate of 14% compares favorably
`with a previously reported relapse rate of 29% for lithium
`therapy in a meta-analysis of 19 double-blind, randomized
`controlled trials of lithium prophylaxis in 865 patients
`(Davis et al 1999). During this open-label treatment with
`CBZ-ERC, patients who had previously been treated with
`placebo had signifi cant improvements in YMRS and HDRS
`in the fi rst month of active therapy, which were maintained
`over the 6-month period (Figures 2 and 3). In addition,
`patients who had received prior CBZ-ERC therapy showed
`a trend toward continued improvement in YMRS scores
`(Ketter et al 2004).
`
`–28 Days
`
`Day 1
`
`Month 1
`
`Month 2
`
`Month 4
`
`Month 6
`
`†
`
`†
`
`*
`
`‡
`
`†
`
`0
`–2
`–4
`–6
`–8
`–10
`–12
`–14
`–16
`–18
`
`Mean change in YMRS score (LOCF)
`
`CBZ-ERC (aPlacebo lead-in)
`n = 40
`Baseline (–28 days) = 26.6
`Day 1 = 15.2
`End point = 10.0
`
`CBZ-ERC
`n = 37
`Baseline (–28 days) = 25.7
`Day 1 = 11.0
`End point = 9.0
`
`Figure 2 Mean change in YMRS total scores with long-term CBZ-ERC treatment in
`bipolar disorder patients with manic or mixed episodes.
`*p ⬍ 0.01; †p ⬍ 0.001; ‡p ⬍ 0.0001 vs baseline (one-sample t-test of mean change
`from day 1). aPatients who were previously placebo in the acute 3-week trials before
`entrance into the 6-month study.
`Abbreviations: YMRS, Young Mania Rating Scale; LOCF, last observation carried
`forward; CBZ-ERC, carbamazepine extended-release capsules.
`
`8
`
`Neuropsychiatric Disease and Treatment 2006:2(1)
`
`Page 6 of 10
`
`

`
`3-week studies, TC, low-density lipoprotein cholesterol
`(p ⬍ 0.0001), and high-density lipoprotein-cholesterol
`(HDL-C) increased signifi cantly (p ⬍ 0.01) from base-
`line measurements in patients treated with CBZ-ERC
`(Weisler et al 2005). Despite these increases in cholesterol,
`TC to HDL-C ratio was 4 to 1 at baseline and endpoint,
`which is not thought to be problematic for cardiovascular
`disease. Importantly, results from a 6-month CBZ-ERC
`extension study showed no further increase in plasma
`cholesterol concentrations in patients who were previ-
`ously on CBZ-ERC in the 3-week studies (Ketter et al
`2004). This finding suggests that the increase in TC
`concentrations is not continuous, although no defi nitive
`conclusion on the effects of CBZ-ERC on cholesterol
`concentrations can be made without fasting cholesterol
`measurements.
`Studies have shown that patients with epilepsy tran-
`sitioned from immediate-release CBZ formulations to
`CBZ-ERC have reported fewer and less severe adverse
`events and greater patient satisfaction (Ficker 2004;
`Miller et al 2004a). In one 3-month, open-label study,
`patients with epilepsy who switched from their current
`immediate-release CBZ product to an equal total daily dose
`of CBZ-ERC experienced both a decreased occurrence and
`a decreased severity of adverse events, demonstrating the
`benefi ts of CBZ-ERC in terms of safety and tolerability
`(Ficker 2004). To further defi ne the improved tolerability
`of CBZ-ERC, a 3-month, multicenter, open-label study
`employed the Adverse Event Profi le (AEP) questionnaire
`to measure the adverse event changes associated with
`switching adult patients with epilepsy from immediate-
`release CBZ to equal daily doses of CBZ-ERC. Analysis
`of the AEP total scores at end point showed that patients
`converted to CBZ-ERC exhibited statistically signifi cant
`mean decreases in all common CNS side effects (sedation,
`vertigo, ataxia, diffi culty in coordination, confusion, and
`diplopia) as a result of transitioning from immediate-
`release CBZ to CBZ-ERC (p ⬍ 0.0001 for all side effects).
`In addition, there was a 46% decrease from baseline in the
`number of patients with “toxic” AEP scores (ⱖ45) (Miller
`et al 2004a). Overall, it appears that CBZ-ERC may be
`more tolerable than immediate-release CBZ, as transition
`to CBZ-ERC therapy signifi cantly reduces the frequency
`and severity of drug-induced side effects.
`
`Future needs
`While randomized clinical trials have defi nitively con-
`fi rmed the effi cacy of CBZ-ERC in the treatment of manic
`
`CBZ-ERC in bipolar disorder
`
`symptoms in patients with manic or mixed bipolar disorder,
`the full spectrum of benefi t provided by CBZ-ERC awaits
`comprehensive characterization in future studies. For
`example, CBZ-ERC has been linked to improvements
`in safety, tolerability, and patient satisfaction relative to
`what is seen in association with immediate-release CBZ
`in patients with epilepsy, and it has been hypothesized
`that CBZ-ERC provides similar benefi ts in patients with
`bipolar disorder. Nonetheless, because epilepsy and bipolar
`disorder are separate entities with distinct pathophysi-
`ological mechanisms, the potential merits of CBZ-ERC
`in comparison with immediate-release CBZ remain to be
`examined in a population specifi cally consisting of patients
`with bipolar disorder.
`Aside from the relative benefi ts of extended-release
`delivery of carbamazepine, another area that warrants
`further study is the effi cacy of CBZ-ERC in the prophylaxis
`of bipolar disorder. Results from a recent 6-month, open-
`label study were promising, with only 14% of patients
`experiencing recurrences during maintenance therapy with
`CBZ-ERC. However, because of the open-label nature of
`that study, and since the study involved only a relatively
`small number of patients, all of whom had completed a
`full course of treatment (CBZ-ERC or placebo) in one of
`two previous randomized trials, the generalised fi ndings
`that were made remain to be assessed in a double-blind
`trial involving a more inclusive study population. Another
`notable fi nding made in these recent clinical trials is
`that of significantly reduced depressive symptoms in
`patients receiv