LEADING ARTICLE
`
`Drugs 2008; 68 (16): 2269-2292
`0012-6667/08/0016-2269/$53.45/0
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Upcoming Agents for the Treatment
`of Schizophrenia
`Mechanism of Action, Efficacy and Tolerability
`
`Delia Bishara and David Taylor
`Pharmacy Department, Maudsley Hospital, London, UK
`
`Abstract
`
`Since the introduction of a group of atypical antipsychotics in the 1990s, there
`has been a decline in the rate of new antipsychotics being introduced into clinical
`practice. However, with increasing safety and efficacy concerns over currently
`available drugs and a dearth of options available for atypical depot formulations,
`there is a considerable need for the development of new formulations and agents.
`This review examines the profile of seven antipsychotic drugs currently in the
`premarketing stage of development and summarizes their mechanism of action,
`clinical potential and safety.
`Asenapine is an antipsychotic with activity for multiple receptors and has
`potential to improve negative and cognitive symptoms of schizophrenia. Bifepru-
`nox is a partial dopamine D2 and serotonin 5-HT1A receptor agonist showing a
`less than convincing efficacy profile, but which may offer safety advantages over
`available agents by means of a reduced risk of metabolic complications. Iloper-
`idone is a D2 and 5-HT2A receptor antagonist requiring further studies to establish
`its effectiveness. It has a high affinity for α1-adrenoceptors, which can lead to
`associated haemodynamic adverse effects. Nemonapride is essentially a typical
`antipsychotic drug, similar in structure to sulpiride, which has been available for
`some time in Japan. It has efficacy against positive symptoms and has shown
`some antidepressant and anxiolytic properties, although efficacy data for it are
`somewhat limited. Norclozapine (N-desmethylclozapine) is a major metabolite of
`clozapine formed by its demethylation. Its partial agonist activity at D2 receptors
`has raised interest in it as an antipsychotic in its own right. In addition, it appears
`to have muscarinic agonist activity, which is believed to be responsible for the
`observed positive effects it has on cognition. It was envisaged to be effective as an
`adjunct to other agents or at high doses in the treatment of refractory schizo-
`phrenia, although a recent randomized, controlled study showed that it was no
`more effective than placebo in patients with schizophrenia experiencing an acute
`psychotic episode. Olanzapine pamoate depot injection has shown comparable
`efficacy to oral olanzapine in several studies. However, it has provoked considera-
`ble safety concerns by its association with inadvertent intravascular injection
`events in numerous patients. This accidental intravascular administration of
`olanzapine pamoate leads to excessive sedation, confusion, dizziness and altered
`speech. Post-injection observation periods and postmarketing surveillance are
`planned following the introduction of the depot. Paliperidone palmitate is the
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 001
`
`©
`

`

`2270
`
`Bishara & Taylor
`
`palmitate ester of paliperidone, the major metabolite of risperidone, and is
`formulated as a long-acting injection for intramuscular use. Its pharmacology is
`comparable to risperidone, having D2 and 5-HT2A receptor antagonist activity.
`Efficacy studies have shown positive results, and because paliperidone has no
`antagonistic activity at cholinergic receptors, it has low potential for anticholiner-
`gic adverse effects, including cognitive dysfunction. However, with higher doses,
`the frequency of extrapyramidal side effects and orthostatic hypotension have
`been shown to be greater than with placebo.
`
`such as weight gain[7,8] and impaired glucose meta-
`Schizophrenia is a severely debilitating psychiat-
`bolism.[9] Furthermore, similar to conventional anti-
`ric disorder observed worldwide. It often results in
`lengthy hospitalizations and is a considerable bur-
`psychotics, atypical antipsychotics have not proved
`den upon medical resources.[1] Prevalence amongst
`to be effective in treatment-refractory patients. Until
`adults tends to vary between studies but is usually
`recently, clozapine, discovered shortly after chlor-
`reported to be in the range of 0.5–1.5%.[2] Since the
`promazine, remained the only drug to have demon-
`introduction of chlorpromazine in the 1950s, the
`strated clinical superiority to other agents in treat-
`number of antipsychotic drugs available has notably ment-resistant schizophrenia[10] and in suicidali-
`increased. By the 1980s, several conventional or
`ty.[11] Attempts to develop new drugs based on its
`first-generation antipsychotics had been developed
`pharmacology in an effort to mimic its superior
`and were found to be effective in treating the posi-
`efficacy have so far largely been unsuccessful.
`tive symptoms of schizophrenia, such as delusions However, a recent paper reported that higher than
`and hallucinations. However, the negative symp-
`typically prescribed doses of olanzapine may be as
`toms of the illness, (emotional withdrawal, apathy,
`effective as conventional doses of clozapine in treat-
`avolition and cognitive dysfunction) were not effec- ment-resistant schizophrenia.[12]
`tively managed by these drugs. In addition, these
`In the last decade, antipsychotic drug develop-
`antipsychotics were found to produce a high burden
`ment has remained somewhat static, at least in terms
`of extrapyramidal side effects (EPS)[3] and adverse
`of new drug introductions. This may be because of
`effects related to elevation of serum prolactin.[4]
`poorly defined pathophysiology of the disorder and
`Moreover, their tendency to cause tardive dyskine-
`incomplete understanding of the pharmacology of
`sia[5] in the longer term made their continued use
`available drugs,[13] resulting in confusion as to the
`problematic.
`ideal mode of action required. In addition, failure of
`In an effort to develop novel agents that would
`some drugs in early clinical trials and the increased
`treat both the positive and negative symptoms of
`costs of drug development may have contributed to
`schizophrenia while affording a low propensity for
`the recent dearth of new agents being launched.
`movement disorders, the pharmaceutical industry
`Since many currently available atypical antipsychot-
`developed several antipsychotics in the 1990s, re-
`ics will soon lose patent protection, there is in-
`ferred to as second-generation antipsychotics, con-
`creased pressure on the pharmaceutical industry to
`sidered to be ‘atypical’. While these drugs have
`develop novel treatments, and so a renewed interest
`probably been of some benefit to patients, full ex-
`in drug development for schizophrenia has emer-
`pectations have not been realized. Despite having a
`ged.[13]
`somewhat improved (although debatable) efficacy
`This article reviews the antipsychotic agents that
`in treating the negative symptoms of schizophrenia
`have undergone extensive clinical development for
`as well as a lower propensity to cause movement
`the treatment of schizophrenia and reached the
`disorders, these benefits have been accompanied by
`other effects, including metabolic adverse effects[6]
`premarketing stage of development, and examines
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 002
`
`©
`

`

`Upcoming Agents for Schizophrenia
`
`2271
`
`their pharmacology, clinical potential and tolerabili-
`ty.
`
`1.2 Atypical Antipsychotics
`
`1. Pharmacology of Currently
`Available Antipsychotics
`
`1.1 Conventional (Typical) Antipsychotics
`
`Because of the problems encountered with con-
`ventional antipsychotics, the atypical antipsychotics
`were developed based to a large extent on the com-
`plex pharmacology of clozapine. Clozapine has af-
`finity for a diverse range of receptors including D1,
`D2 and D4 dopaminergic, α1 and α2 adrenergic, H1
`histaminergic, muscarinic and various serotonin re-
`Since the dopamine hypothesis was first pro-
`ceptor subtypes.[21-23] Its supported clinical superior-
`posed in the 1960s, it has remained the central pivot
`ity[10] and near absence of the debilitating EPS has
`around which antipsychotic agents have been devel-
`fuelled an intense effort over the last 20 years to
`oped.[14] Conventional antipsychotic drugs all show
`develop similar agents.
`at least some affinity for the dopamine D2 receptors
`In the development of atypical antipsychotics,
`and there is a strong correlation between clinical
`efficacy of the drugs and their binding affinities.[15]
`researchers have tried to mimic the pharmacological
`action of clozapine while trying to avoid its own
`Positive symptoms of schizophrenia are believed to
`serious adverse effects, such as agranulocytosis.[24]
`result from dopaminergic hyperactivity,[16] while
`Like conventional antipsychotics, atypical antipsy-
`negative symptoms have been attributed to reduced
`chotics are also antagonists at D2 receptors. How-
`functioning in the prefrontal cortex, mainly resulting
`ever, they do show an additional range of binding
`from underactivity of prefrontal dopaminergic neu-
`activity at various other receptor sites. In particular,
`rons. Therefore, these symptoms are potentially im-
`their antagonist activity at serotonin receptors was
`proved by agents that reduce serotonergic function
`thought to account for the differences between the
`(by serotonin 5-HT1A receptor agonist activity),
`two classes of agents. Atypical drugs show a higher
`thus promoting increased dopamine activity in the
`affinity for 5-HT2A receptors compared with D2
`prefrontal cortex, and by drugs that block presynap-
`tic dopamine receptors.[17] Such an effect may also
`receptors, and this ratio of affinities has been hy-
`pothesized to account for their enhanced efficacy
`improve cognitive impairment because of the result-
`and fewer EPS.[25,26] Antagonism at 5-HT2A recep-
`ing stimulation of D1 receptors.[18] These concepts
`tors leading to an increase of dopamine activity in
`are consistent with the clinical profile of conven-
`the prefrontal cortex has also been suggested to
`tional antipsychotics, which are effective in treating
`account for the beneficial effects that atypical anti-
`positive symptoms, presumably by reducing overac-
`psychotics have against negative symptoms.[26]
`tivity in the mesolimbic pathways, but offer little
`benefit to the negative symptoms or cognitive defi- However, amisulpride has no affinity for 5-HT2A
`receptors[27] but clearly has atypical antipsychotic
`cits because of inadequate or adverse effects in the
`properties,[28] suggesting that this 5-HT2A/D2 recep-
`mesocortical pathways (where dopamine activity is
`already decreased).[19]
`tor hypothesis may not hold true for all drugs or that
`other receptor systems also play an important role in
`The high rates of movement disorders caused by
`the atypicality of antipsychotics.
`conventional antipsychotics are believed to arise
`It has also been suggested that atypical anti-
`from dopamine antagonism in the nigrostriatal path-
`ways. The subsequent reduced dopamine activity
`psychotic activity may be explained by differences
`leads to a relative increase in cholinergic activity,
`in the occupancy and dissociation of antipsychotics
`from D2 receptors.[29-31] Agents showing a relatively
`and the resulting imbalance accounts for these
`troublesome adverse effects.[3] Elevation of prolac-
`low D2 receptor affinity, such as clozapine,[32,33]
`quetiapine[34] and olanzapine,[35] and fast dissocia-
`tin caused by these drugs stems from their dopamine
`antagonist effects on the tuberoinfundibular path-
`tion from the receptor, have atypical properties. This
`way.[20]
`loose D2 receptor binding may also account for the
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 003
`
`©
`

`

`2272
`
`Bishara & Taylor
`
`Table I. New antipsychotic drugs
`
`Drug
`Asenapine
`Bifeprunox
`
`Iloperidone
`
`Norclozapine
`
`Nemonapride
`
`Mechanism of action
`Multiple receptors
`Partial D2 and 5-HT1A
`receptor agonist
`D2 and 5-HT2A receptor
`antagonist
`Partial D2 and muscarinic
`receptor agonist
`D2, D3 and D4 receptor
`antagonist
`D2 and 5-HT2A receptor
`antagonist
`Paliperidone palmitate D2 and 5-HT2A receptor
`antagonist
`
`Olanzapine pamoate
`
`Manufacturer
`Organon
`Lundbeck/Solvay
`
`Development status
`Phase III
`Phase III
`
`Expected launch date
`2009
`2010/2011
`
`Titan Pharmaceuticals
`
`Phase III
`
`ACADIA Pharmaceuticals
`
`Phase II
`
`Not known
`
`Not known
`
`Astellas
`
`Eli Lilly
`
`Janssen-Cilag
`
`Launched in Japan
`
`Phase III
`
`Phase III
`
`No plans for launch in
`US, UK or Europe
`Late 2008
`
`2009
`
`observed limbic selectivity observed for some drugs
`such as clozapine.[36]
`
`pathways, their propensity to cause EPS and elevat-
`ed prolactin levels may be reduced.[19]
`
`1.3 Dopamine Partial Agonists
`
`2. New Antipsychotics
`
`2.1 Asenapine
`
`Asenapine (figure 1) is a novel psychotropic
`agent being developed for the treatment of schizo-
`phrenia and bipolar disorder. Its chemical structure
`and pharmacological action are distinct from cur-
`rently available drugs.
`
`We are now approaching an exciting and chal-
`More recently, a new class of antipsychotics has
`lenging time for the development of new treatments
`been introduced, the dopamine partial agonists, of
`for schizophrenia. With continual, if somewhat
`which aripiprazole is the only one currently avail-
`slow, improvements in our understanding of the
`able in clinical practice. Aripiprazole is a potent
`pathophysiology of the disease and complex phar-
`partial agonist at D2 and 5-HT1A receptors and acts macology of the drugs, novel approaches for drug
`as an antagonist at 5-HT2A receptors.[37,38] While
`discovery are being studied. The following agents
`both typical and atypical antipsychotics act as full
`(table I) are presently in their developmental stages
`antagonists at dopamine receptors, schizophrenia as
`and are due to be introduced for clinical practice in
`outlined is thought to arise from a combination of
`the near future.
`over- and under-activity in different dopamine path-
`ways. Thus, blocking dopamine activity in all parts
`of the system may account for the problems encoun-
`tered with drug therapy already discussed. There-
`fore, in theory, the capacity for an agent to alter
`dopamine neurotransmission differently in separate
`areas of the dopaminergic system may have both
`therapeutic and safety advantages.
`Partial agonists are thought to exert their effects
`by acting effectively as dopamine antagonists in the
`mesolimbic pathway. However, in the mesocortical
`pathway, where reduced dopamine activity is
`thought to produce negative symptoms and cogni-
`tive impairment, they effectively act as dopamine
`agonists. Furthermore, because dopamine partial ag-
`onists do not produce complete dopamine activity
`blockade in the nigrostriatal and tuberoinfundibular
`
`O
`
`CI
`
`H
`
`H
`
`N
`Fig. 1. Structural formula of asenapine ((3aS,12bS)-5-chloro-2,3,3-
`a,12b-tetrahydro-2-methyl-1H-dibenz[2,3:6,7]
`oxepino[4,5-c]pyr-
`role).
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 004
`
`©
`

`

`Upcoming Agents for Schizophrenia
`
`2273
`
`9.4
`
`1.8
`
`0.31
`
`0.26
`
`0.16
`
`NEM
`
`4898
`
`0.4
`
`42.8
`
`5.6
`
`93.1
`
`25
`
`7.1
`
`21.4
`
`216
`ILO
`
`>>
`
`>>
`
`>>
`
`>>
`
`>>
`
`10.0
`
`1.6
`
`0.6
`
`3.2
`
`BIF
`
`>>
`
`1.0
`
`1.2
`
`0.034
`
`0.07
`
`2.7
`
`1.1
`
`0.42
`
`1.3
`
`1.4
`ASE
`
`>1500
`
`1.9
`
`>1000
`
`440
`
`6
`
`>10000
`
`45
`
`1100
`
`2
`
`3
`
`0.7
`
`210
`HAL
`
`8
`
`7
`
`6
`
`12
`
`770
`
`9–12
`
`473
`
`125
`
`85
`CLO
`
`50
`
`11
`
`1
`
`0.4
`
`32
`
`5
`
`7
`
`high Ki, therefore no appreciable affinity for receptor.
`Ki = dissociation constant; M = muscarinic; NEM = nemonapride; OLA = olanzapine; PAL = paliperidone; QUE = quetiapine; RIS = risperidone; ZIP = ziprasidone; >> indicates very
`5-HT = serotonergic; ARI = aripiprazole; ASE = asenapine; BIF = bifeprunox; CLO = clozapine; D = dopaminergic; H = histamine; HAL = haloperidol; ILO = iloperidone;
`
`3
`
`525
`ZIP
`
`120
`
`11
`
`7
`
`615
`
`220
`
`2450
`
`1600
`
`340
`
`160
`
`455
`QUE
`
`32
`
`1.0
`
`590
`
`30
`
`6.9
`
`4.8
`
`PAL
`
`>10000
`
`20
`
`0.7
`
`25
`
`0.5
`
`7
`
`2
`
`19
`
`11
`
`4
`
`210
`
`>1000
`
`9
`
`10
`
`4
`
`430
`RIS
`
`27
`
`49
`
`11
`
`31
`OLA
`
`>>
`
`61
`
`57
`
`15
`
`3.4
`
`1.7
`
`44
`
`0.8
`
`0.34
`
`M1
`
`H1
`α1-Adrenergic
`5-HT2C
`
`5-HT2A
`
`5-HT1A
`
`D4
`
`D3
`
`D2
`
`D1
`
`ARI
`Ki (nmol/L)
`
`Receptor
`
`Table II. Antipsychotic receptor-binding profiles[39-50] (adapted from Chou,[51] with permission)
`
`Asenapine has higher affinity for a variety of
`serotonergic (5-HT2A, 5-HT2C, 5-HT6, 5-HT7), nor-
`adrenergic (α2A, α2B, α2C) and dopaminergic (D3,
`D4) receptors than D2 receptors, but minimal affini-
`ty for muscarinic receptors (see table II for compari-
`son of binding affinities).[39]
`One of the mechanisms for alleviating negative
`symptoms is believed to be the blockade of 5-HT2A
`and 5-HT2C receptors. Asenapine binds to these
`receptors 19-fold and 38-fold, respectively, higher
`than D2 receptors, suggesting a potential for improv-
`ing negative symptoms.[39] Asenapine is thought to
`maintain adequate but not excessive blockade of D2
`receptors, and so it may allow control of positive
`symptoms without the resulting EPS and elevation
`of prolactin. Similarly, activity at α-adrenergic re-
`ceptors has been suggested to improve negative and
`cognitive symptoms via α2-receptor antagonism
`and positive symptoms via α1-adrenoceptor ant-
`agonism.[52] Asenapine appears to have relatively
`high affinity for adrenergic receptors, which may
`offer potential therapeutic advantages, although
`there is no such evidence as yet. In contrast, it has
`been shown to have very low affinity for muscarinic
`and other CNS receptors. The D2 receptor affinity of
`asenapine is approximately 6000-fold greater than
`the affinity for M1 receptors, thus minimizing the
`risk of antimuscarinic adverse effects that are seen
`with many other agents.[39]
`
`2.1.1 Preclinical Studies
`Results from preclinical studies using animal
`models have been consistent with the receptor
`profiles described in the previous section. Using the
`conditioned avoidance response (CAR) test in rats,
`the dose-response relationship for the antipsychotic-
`like effect of asenapine was determined. For appar-
`ently adequate antipsychotic effect (i.e. 80% sup-
`pression of CAR[53]), the dose needed was 0.1–0.2
`mg/kg (dose that produces a 50% effective response
`= 0.12 mg/kg).[54] When tested in the catalepsy test,
`asenapine 0.1 and 0.2 mg/kg did not reach a score of
`2 (where catalepsy is considered to begin[53]) at any
`time interval examined. These findings suggest that
`asenapine may exhibit a potent antipsychotic effect
`without inducing EPS.[53,54]
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 005
`
`©
`

`

`2274
`
`Bishara & Taylor
`
`The behavioural effects of asenapine in rats, as more effective than placebo in treating negative
`assessed by in vivo microdialysis, have been found
`symptoms. This is not consistent with findings from
`to be accompanied by an increase in dopamine re-
`earlier trials showing risperidone 6 mg/day having
`lease in the three brain regions examined: the medial
`greater improvement on the PANSS negative sub-
`scale score compared with placebo.[58,59] Although
`prefrontal cortex, nucleus accumbens and stri-
`atum.[54] Asenapine induced a marked enhancement
`statistical comparisons between asenapine and ris-
`of dopamine efflux to a greater extent in the shell
`peridone were not performed, it is perhaps important
`region of the nucleus accumbens than in the core
`to note that discontinuations due to ineffectiveness
`region, thus sharing a similar profile to atypical
`of treatments were more common with risperidone
`drugs.[54] Microdialysis and electrophysiological as-
`than asenapine.[56]
`sessments showed that asenapine potentiated both
`The incidence of adverse effects was similar
`prefrontal dopaminergic and glutamatergic trans-
`across all three treatment groups, with the most
`mission. These effects may also contribute to im-
`frequent reports for asenapine being insomnia
`provement of negative symptoms and cognitive def-
`(11%), somnolence (11%), nausea (11%), anxiety
`icits.[54,55] Furthermore, the very low concentration
`(10%) and agitation (9%).[56] Asenapine showed a
`of asenapine required to facilitate glutamatergic
`placebo-equivalent risk of significant weight gain,
`transmission, even lower than that for clozapine, whereas risperidone was associated with a higher
`suggests an important cognitive-enhancing ac-
`incidence of clinically significant weight gain, con-
`tion.[54]
`sistent with previous reports,[60,61] along with a high
`incidence of hyperprolactinaemia.[56] Laboratory
`2.1.2 Clinical Studies
`findings in this study also showed that asenapine
`A randomized, double-blind, placebo- and risper- was not associated with metabolic disturbances or
`idone-controlled,
`fixed-dose, 6-week
`trial of
`adverse effects on cardiovascular function such as
`asenapine was carried out in the US.[56] Patients
`changes in blood pressure, heart rate or corrected
`were randomly assigned to receive sublingual QT interval prolongation.[56]
`asenapine 5 mg twice daily, placebo or oral risper-
`In summary, the receptor profile and combined
`idone 3 mg twice daily. Results for the primary
`data from preclinical and clinical studies predict that
`efficacy outcome measure, the Positive and Nega-
`asenapine has an ability to improve positive, nega-
`tive Syndrome Scale (PANSS) total score,[57] for the
`tive and cognitive symptoms of schizophrenia,
`intention-to-treat population showed mean changes while causing limited extrapyramidal, antimuscarin-
`at endpoint from baseline were –15.9 with asenapine
`ic and metabolic adverse effects. It may therefore
`and –5.3 with placebo (p < 0.005). The PANSS
`prove to be a useful option in patients with predomi-
`positive subscale score showed mean changes at
`nant or resistant negative symptoms. Further large-
`endpoint from baseline of –5.5 for asenapine com-
`scale studies will be needed in order to confirm these
`pared with –2.5 for placebo (p = 0.01) and –5.1 for
`findings and to distinguish it from other D2/5-HT2
`risperidone (also significant compared with placebo;
`receptor antagonists such as risperidone.
`p < 0.05). PANSS negative subscale score results
`showed mean changes at endpoint from a baseline of
`–3.2 for asenapine compared with –0.6 for placebo
`(p = 0.01).
`Bifeprunox (figure 2) is an antipsychotic show-
`The main efficacy findings from this study were
`ing partial agonist activity at D2, D4 and 5-HT1A
`that asenapine 5 mg twice daily was superior to
`receptors, and antagonism at D3 receptors. Unlike
`placebo in treating both the positive and negative
`symptoms of schizophrenia. Risperidone (3 mg many other antipsychotics, it shows no marked ac-
`twice daily), on the other hand, was superior to
`tivity at the 5-HT2A, 5-HT2C, noradrenergic, musca-
`rinic or histaminergic receptors.[46,62]
`placebo in treating positive symptoms, but it was not
`
`2.2 Bifeprunox
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 006
`
`©
`

`

`Upcoming Agents for Schizophrenia
`
`2275
`
`O
`
`HN
`
`O
`
`N
`
`N
`
`Fig. 2. Structural formula of bifeprunox (7-[4-[(3-phenylphenyl)-
`methyl]iperazin-1-yl]-3H-benzooxazol-2-one).
`
`phrenia. The efficacy, safety and tolerability of three
`fixed once-daily doses of bifeprunox (5, 10 and
`20 mg) were evaluated.
`In this study, only bifeprunox 20 mg showed a
`statistically significant reduction in PANSS total
`score[57] compared with placebo. In the last-observa-
`tion-carried-forward
`(LOCF)
`analysis, mean
`changes in PANSS total score were –11.3 and –5.3
`points for bifeprunox 20 mg and placebo, respec-
`2.2.1 Preclinical Studies
`tively (adjusted p = 0.031; treatment effect CI –11.1,
`In in vitro preclinical studies, bifeprunox demon-
`–0.4). Bifeprunox 5 and 10 mg produced mean
`strated partial dopamine agonist properties. Agonist
`changes of –9.7 and –5.0 points, respectively, but
`effects were seen to be induced when endogenous
`the difference compared with placebo was not statis-
`dopamine tone was low, for example in adenylate
`tically significant (adjusted p = 0.128, treatment
`cyclase activity assays in Chinese hamster ovary
`effect CI –9.2, 0.9 for 5 mg; p = 1.000, treatment
`cells and in rat striatal slices.[46] In contrast, in the
`effect CI –4.4, 5.5 for 10 mg). A statistically signif-
`presence of a full D2 receptor agonist (i.e. when
`icant –15.7-point change was observed in the risper-
`endogenous dopamine was high) bifeprunox acted
`idone arm (p < 0.0001 vs placebo; 95% CI –14.5,
`as a functional antagonist.[46,63] Results suggest a
`–6.5).[65] Bifeprunox 20 mg also produced statisti-
`distinct antipsychotic profile, showing functional
`cally significant changes versus placebo in the
`agonist activity in the prefrontal cortex region and a
`PANSS-positive (p = 0.037; treatment effect CI
`functional antagonist activity in brain regions such
`–2.9, –0.1) and PANSS-negative (p = 0.026; treat-
`as the nucleus accumbens where dopaminergic ment effect CI –2.6, –0.2) subscales.
`neurotransmission is thought to be increased in
`The incidence of patients withdrawing from the
`schizophrenia.[62]
`study because of adverse effects was similar in the
`In vivo preclinical studies on rats showed that
`bifeprunox and placebo groups. No statistically sig-
`bifeprunox, unlike conventional antipsychotics,
`nificant difference was observed at endpoint on any
`caused suppression of the CAR at near maximal D2 movement disorder scales between bifeprunox and
`receptor occupancy. This suppression was dose re-
`placebo. Treatment with bifeprunox at all doses was
`lated and, based on the relationship observed,
`associated with statistically significant weight de-
`bifeprunox was predicted to be clinically effective at
`creases, reduction in non-fasting total cholesterol
`doses ≥10 mg, producing >90% D2 receptor occu-
`and triglycerides, as well as decreases in prolactin
`pancy at these doses.[64] Receptor occupancy has
`levels compared with placebo.[65]
`also been studied in humans using positron emission
`tomography (PET). D2 receptor occupancy was also
`found to be dose related, with a plateau in occupan-
`cy observed at 90% for bifeprunox doses ≥10 mg
`(figure 3).[63]
`
`Observed
`Predicted
`
`100
`
`80
`
`60
`
`40
`
`20
`
`0
`
`receptor occupancy (%)
`
`Caudate nucleus,
`
`0
`
`5
`
`10
`15
`20
`Bifeprunox (ng/mL)
`Fig. 3. Bifeprunox plasma concentration at 2 hours after administra-
`tion and dopamine D2 receptor occupancy in caudate nucleus.[63]
`
`25
`
`30
`
`2.2.2 Clinical Studies
`The following four randomized clinical trials
`have examined the efficacy and safety of bifepru-
`nox. Casey and co-workers[65] conducted a multi-
`centre, 6-week, randomized, placebo-controlled, ris-
`peridone-referenced, dose-finding study including
`589 patients with acute exacerbation of schizo-
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 007
`
`©
`

`

`2276
`
`Bishara & Taylor
`
`doses of bifeprunox produced statistically signif-
`A very similar study was conducted by Rapaport
`and colleagues,[66] which evaluated the efficacy and
`icant decreases in prolactin levels (p = 0.0001),
`safety of bifeprunox at once-daily doses of 30 and whereas placebo and olanzapine were associated
`40 mg compared with placebo and with a reference with non-significant increases from baseline.[67]
`risperidone arm. In the LOCF analysis, bifeprunox
`In order to assess long-term efficacy and safety of
`30 mg produced a statistically significant difference
`bifeprunox, a 6-month, randomized, double-blind,
`from placebo in the change in PANSS total score.[57]
`parallel-group, placebo-controlled, fixed-dose study
`Mean changes were –13.5 and –7.7, respectively
`was conducted.[68] Patients received bifeprunox 20
`(p = 0.020; 95% CI –10.3, –1.4). Bifeprunox 40 mg,
`or 30 mg/day or placebo. The primary endpoint was
`on the other hand, produced a mean –10.3-point
`the time from randomization to deterioration, de-
`change, but the difference from placebo was not
`fined as a Clinical Global Impression (CGI)-Im-
`statistically significant (adjusted p = 0.156; 95% CI
`provement[69] score ≥5 or PANSS[57] item P7 (hostil-
`–7.7, 1.2).
`ity) and/or G8 (uncooperativeness) score ≥5 for
`In terms of safety assessments, the most frequent
`2 consecutive days or ≥20% increase in PANSS
`adverse events seen in bifeprunox 30 and 40 mg
`total score from baseline. Other efficacy and safety
`groups compared with placebo and risperidone were
`assessments were similar to the studies mentioned
`gastrointestinal in nature. For bifeprunox 30 mg,
`previously. In this study, treatment with bifeprunox
`they included constipation (13%), dyspepsia (11%),
`20 and 30 mg resulted in statistically significant
`nausea (18%), vomiting (12%) and dizziness (8%).
`(p = 0.008 and p = 0.006, respectively) longer time
`Similar results were seen with the higher dose.[66]
`to deterioration compared with placebo (LOCF).[68]
`Decreases in weight were greater with bifeprunox
`The proportion of patients who deteriorated was
`compared with placebo. Non-fasting cholesterol,
`40.5% in the bifeprunox 20 mg group, 38.4% in the
`glucose and triglyceride levels were also lower with
`30 mg group and 59.0% in the placebo group. In the
`bifeprunox 30 and 40 mg. Bifeprunox was also
`placebo group, the risk of deterioration was found to
`associated with a lower incidence of EPS than the
`be approximately 1.5-fold higher than the bifepru-
`active reference drug, risperidone.[66]
`nox 20 and 30 mg groups (hazard ratio 0.656 and
`Barbato and colleagues[67] used the same study
`0.653, respectively).
`design but examined once-daily doses of bifeprunox
`20 and 30 mg versus placebo, this time with olanza-
`pine 15 mg/day as the reference drug. In this study,
`no statistically significant change in PANSS total
`score[57] was observed with bifeprunox 20 or 30 mg
`compared with placebo, whereas the reference agent
`olanzapine did show significant improvement on
`measures of efficacy over placebo.
`In contrast with these efficacy findings, bifepru-
`nox showed some safety advantages over olanza-
`pine. At the 6-week study endpoint, the groups
`treated with bifeprunox 20 mg, bifeprunox 30 mg
`and placebo showed mean weight decreases of 1.05,
`0.5 and 0.59 kg, respectively, whereas olanzapine
`was associated with a weight increase of 0.26 kg
`(p < 0.0001 vs placebo).[67] Movement-related ad-
`verse effects occurred with equal frequency across
`groups and ECG results were also comparable. Both
`
`Long-term safety data showed that treatment-
`emergent adverse events were 72.3%, 83.1% and
`57.8% in bifeprunox 20 mg, 30 mg and placebo
`groups, respectively. Movement-related disorders
`occurred in 10%, 15% and 4% in the bifeprunox
`20 mg, 30 mg and placebo groups, respectively,
`although there were no clinically relevant changes in
`the Barnes Akathisia Rating Scale (BARS),[70]
`Simpson-Angus Scale (SAS)[71] and Abnormal In-
`voluntary Movement Scale (AIMS)[72] total scores at
`endpoint in all three treatment groups. Bifeprunox
`30 mg reached statistical significance in reduction in
`adjusted mean weight (p = 0.027) and triglycerides
`(p = 0.006) versus placebo, but the other dose and
`safety parameters did not reach statistical signifi-
`cance. There was no difference between the groups
`in incidence of abnormal ECG findings.[68]
`
` 2008 Adis Data Information BV. All rights reserved.
`
`Drugs 2008; 68 (16)
`
`Mylan v. Janssen (IPR2020-00440) Ex. 1038, p. 008
`
`©
`

`

`Upcoming Agents for Schizophrenia
`
`2277
`
`F
`
`N
`
`O
`
`O
`
`O
`
`O N
`Fig. 4. Structural formula of iloperidone (1-[4-[3-[4-(6-fluoro-1,2-benzisoxazol-3-yl)-1-piperidinyl]propoxy]-3-methoxyphenyl]ethanone).
`
`A pooled analysis from existing studies was per-
`formed in order to evaluate the metabolic effects of
`bifeprunox compared with placebo and the active
`reference agents used. Findings from this analysis
`reflected those shown in the studies discussed in this
`section. Bifeprunox treatment was associated with a
`reduction in weight, total cholesterol and triglycer-
`ide levels and minimal changes in plasma glucose
`levels.[73]
`
`and minimal activity at cholinergic receptors indi-
`cates that anticholinergic adverse effects may be
`avoided. Moreover, as a result of the low affinity for
`α2A adrenoceptors, iloperidone is not expected to
`induce convulsions.[47,76,77]
`
`2.3 Iloperidone
`
`2.3.1 Preclinical Studies
`In animal behavioural models, iloperidone exhib-
`ited potent, long-lasting ‘antipsychotic’ activity
`against the positive symptoms of schizophrenia as
`seen in the climbing mouse assay and CAR in rats.
`However, it was 300-fold less potent in causing
`Iloperidone (figure 4) is an atypical antipsychotic
`catalepsy and in inhibiting apomorphine-induced
`initially chosen for development because of its high
`stereotyped behaviour in rats. These two measures
`affinity for 5-HT2 receptors and moderate affinity
`evaluate activity