The H₃ antagonist ABT-288 is tolerated at significantly higher exposures in subjects with schizophrenia than in healthy volunteers

Abstract

Aims

ABT-288 is a potent and selective H₃ receptor antagonist with procognitive effects in several preclinical models. In previous studies, 3 mg once daily was the maximal tolerated dose in healthy volunteers. This study characterized the safety, tolerability and pharmacokinetics of ABT-288 in stable subjects with schizophrenia.

Methods

This was a randomized, double-blind, placebo-controlled, dose-escalating study of ABT-288 (10 dose levels, from 1 to 60 mg once daily for 14 days) in stable subjects with schizophrenia treated with an atypical antipsychotic. In each dose group, five to seven and two to three participants were assigned to ABT-288 and placebo, respectively.

Results

Of the 67 participants enrolled, nine participants (on ABT-288) were prematurely discontinued, in seven of these due to adverse events. ABT-288 was generally safe and tolerated at doses up to 45 mg once daily. The most common adverse events, in decreasing frequency (from 31 to 5%), were abnormal dreams, headache, insomnia, dizziness, somnolence, dysgeusia, dry mouth, psychotic disorder, parosmia and tachycardia. Adverse events causing early termination were psychotic events (four) and increased creatine phosphokinase, pyrexia and insomnia (one each). The half-life of ABT-288 ranged from 28 to 51 h, and steady state was achieved by day 12 of dosing. At comparable multiple doses, ABT-288 exposure in subjects with schizophrenia was 45% lower than that previously observed in healthy subjects. At trough, ABT-288 cerebrospinal fluid concentrations were 40% of the total plasma concentrations.

Conclusions

ABT-288 was tolerated at a 15-fold higher dose and 12-fold higher exposures in subjects with schizophrenia than previously observed in healthy volunteers. The greater ABT-288 tolerability was not due to limited brain uptake.

WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT

• ABT-288 is a selective H₃ receptor antagonist with procognitive effects across several preclinical cognitive domains.

• Two phase 1 studies in healthy young adults and elderly subjects established 3 mg once daily as the maximally tolerated dose. ABT-288 at doses of 5 or 6 mg once daily was intolerable in healthy volunteers.

WHAT THIS STUDY ADDS

• We present in this report the safety, tolerability and pharmacokinetics of ABT-288 in subjects with stable schizophrenia receiving treatment with an atypical antipsychotic.

• ABT-288 was tolerated at a 15-fold higher dose and approximately 12-fold higher plasma exposures in subjects with schizophrenia than in healthy volunteers.

• This is the first report of improved tolerability of an H₃ antagonist in subjects with schizophrenia relative to healthy volunteers.

Introduction

Symptoms of schizophrenia have traditionally been classified into three domains, namely positive symptoms, negative symptoms and cognitive impairment 1,2. Currently available medications for schizophrenia treat positive symptoms adequately. To date, effective medications for the treatment of negative symptoms and cognitive impairment have not been developed and represent areas of unmet need for the treatment of schizophrenia 3–5.

The histamine H₃ receptor is a G protein-coupled receptor and one of four subtypes (H₁, H₂, H₃ and H₄) of the histamine receptor family 6. The H₃ receptors are highly expressed in brain regions, such as the cerebral cortex and hippocampus, that play an important role in attention, learning and memory. Blockade the H₃ receptors by selective antagonists has been shown to enhance the release of precognitive neurotransmitters, such as histamine, acetylcholine, dopamine and norepinephrine 6–8. In multiple nonhuman experimental models, selective H₃ antagonists have consistently improved performance across a number of cognitive domains 9–11. This suggested the potential utility of H₃ receptor antagonists for treatment of numerous neurocognitive disorders, including schizophrenia 6. The H₃ antagonists have also demonstrated clinical efficacy in the treatment of excessive sleep disorders 12,13.

ABT-288, or 2-[4′-((3aR,6aR)-5-methyl-hexahydro-pyrrolo[3,4-b]pyrrol-1-yl)-biphenyl-4-yl]-2H-pyridazin-3-one, is a potent and selective competitive H₃ receptor antagonist designed for symptomatic treatment of cognitive disorders 11. In humans, ABT-288 was previously evaluated in two phase 1 studies in healthy young adults and elderly volunteers 14. The safety, tolerability and pharmacokinetics of ABT-288 were comparable in young and elderly subjects. Multiple doses up to 3 mg once daily were tolerated in healthy subjects, whereas 5 or 6 mg once daily was intolerable. The most frequently reported adverse events were hot flush, headache, abnormal dreams, insomnia, nausea and dizziness, and the majority of them were believed to be histamine mediated 14. ABT-288 exposure (AUC) was dose proportional, and the elimination half-life ranged from 40 to 61 h in healthy subjects. Food did not have a clinically meaningful effect on ABT-288 exposure, and steady state was achieved by day 10 of once-daily dosing. Based on the phase 1 data, 1 and 3 mg once-daily doses of ABT-288 were advanced for evaluation in a phase 2 proof-of concept study in subjects with Alzheimer's disease.

The present study evaluated safety, tolerability and pharmacokinetics of ABT-288 in stable subjects with schizophrenia receiving treatment with an atypical antipsychotic. The study was conducted to inform dose selection for a phase 2 study in the treatment of cognitive impairment associated with schizophrenia.

Methods

This was a single-centre, double-blind, randomized, placebo-controlled, phase 1 study of the safety, tolerability and pharmacokinetics of multiple doses of ABT-288 in subjects with stable schizophrenia. Ten oral dose levels of ABT-288 (1, 3, 6, 9, 12, 15, 20, 30, 60 and 45 mg or matching placebo) were tested in nine cohorts of participants, with each dose administered once daily for 14 days. In cohort 1 of the study, seven, seven and six participants were assigned to receive 1 mg ABT-288, 3 mg ABT-288 and placebo, respectively. In cohort 2, five and two participants were assigned to receive 6 mg ABT-288 and placebo, respectively. In each of cohorts 3–9, six participants were assigned to receive a dose level of ABT-288 (evaluated doses were 9, 12, 15, 20, 30, 60 or 45 mg once daily) and two participants were assigned to receive placebo. Overall, up to 83 adult male and female participants, who had a diagnosis of schizophrenia, who were stable and receiving treatment with an atypical antipsychotic, were planned to participate in the study. Participants were allowed to repeat enrolment in the study. Each dose of study drug was administered orally with water (240 ml) ∼30 min after the start of a standardized breakfast.

The study was conducted at California Clinical Trials (Glendale, CA, USA). The study protocol was reviewed and approved by the California IRB, Inc. (Pasadena, CA, USA) Institutional Review Board. The study was conducted in accordance with all applicable regulatory and Good Clinical Practice guidelines and following the ethical principles originating in the Declaration of Helsinki.

Participants

Participants underwent screening procedures within 7–28 days prior to initial study drug administration. Participants were considered eligible for enrolment if they were between 18 and 60 years old and met the following criteria: (i) they had a current diagnosis of schizophrenia based on the Diagnostic and Statistical Manual of Mental Disorders – Fourth Edition – Text Revision (DSM-IV-TR) 15; (ii) they remained clinically stable on the same single second-generation antipsychotic for 8 weeks prior to randomization, with no psychosis-related changes in dose of the current antipsychotic treatment for at least 4 weeks; and (iii) they had no more than ‘moderate’ severity rating on the following items from the Positive and Negative Syndrome Scale (PANSS) 16: hallucinatory behaviour, delusions, unusual thought content and/or negative symptom items (items N1–N7).

A subject was not eligible for enrolment if the subject had a substance-dependence disorder (by DSM-IV-TR criteria, excluding nicotine or tobacco products) that had not been in sustained remission for at least 1 year prior to screening; had a diagnosis of schizoaffective disorder, bipolar disorder, manic episode, dementia, obsessive-compulsive disorder or drug-induced psychosis or had a current major depressive episode; had previously been diagnosed with mental retardation; had been hospitalized for acute psychosis within 6 months prior to the screening visit; was, at the time of the study, being treated with clozapine (low-dose clozapine or other atypical antipsychotic as a hypnotic agent was acceptable); had current suicidal ideation or suicidal behaviour as assessed by the investigator; had body mass index ≥ 39 kg m⁻²; had current homicidal or other violent ideation in the opinion of the investigator; had a global parkinsonism score of ≥3 (moderate) on the Extrapyramidal Symptom Rating Scale – Abbreviated (ESRS-A); additionally, a subject was not eligible for randomization if he or she had a change in PANSS score of ≥25% from the screening visit to the baseline visit. Other inclusion and exclusion criteria typical for phase 1 evaluations were applied.

Participants signed a written informed consent prior to enrolment in the study. Twelve participants in cohorts 7, 8 and 9 agreed to an optional lumbar puncture procedure on study day 15 for the purpose of collecting cerebrospinal fluid (CSF) for assay of ABT-288. Participants who had a contraindication for lumbar puncture were not eligible to participate in the optional CSF sampling.

Concomitant medications

Given that many patients with schizophrenia are treated with antidepressants, mood stabilizers, anxiolytics and hypnotic agents, participants were permitted to take the following antidepressant medications on a routine basis during the study if they were at stable doses at least 4 weeks prior to confinement: sertraline, citalopram, escitalopram, venlafaxine, duloxetine, mirtazapine and trazodone. Use of any other antidepressants, including tricyclic antidepressants and monoamine oxidase inhibitors, either during the study or within the 2 week period prior to study drug administration, was prohibited.

Intermittent use of short-acting hypnotic agents, such as zolpidem, and short-acting anxiolytics, such as lorazepam, was permitted. Concomitant nonpsychoactive medications were permitted, with the exception of known inhibitors or inducers of drug-metabolizing enzymes. Use of any known inducers of drug-metabolizing enzymes was not permitted within 1 month, and use of ‘moderate’ and ‘strong’ inhibitors was not permitted within five half-lives, prior to study drug administration.

Safety and tolerability monitoring

All observed or volunteered adverse events were recorded after administration of each dose with regard to their time of onset, severity, duration and possible relationship to the study drug. Safety and tolerability were evaluated based on assessments of adverse events, physical examinations, brief neurological examination, laboratory tests (haematology, clinical chemistry and urinalysis), vital signs, orthostatic vital signs and 12-lead electrocardograms.

A dose level was defined as not tolerated if 50% or more of the participants in a group experienced a moderate or severe adverse event that was possibly or probably related to the study drug (determined by the investigator) or if the dose resulted in a serious adverse event that was possibly or probably related to the study drug in any subject within a group. A dose was tolerated if fewer than 50% of the participants in a group experienced a moderate or severe adverse event that was possibly or probably related to the study drug and if there were no serious adverse events that were possibly or probably related to the study drug.

Blood and cerebrospinal fluid sampling

Blood samples for measurement of plasma ABT-288 concentrations were collected by venipuncture into 6 ml evacuated sodium-heparin-containing collection tubes. Blood samples were collected prior to dosing (0 h) and at 0.5, 1, 1.5, 2, 3, 4, 6, 9, 12, 16 and 24 h after dosing on study days 1, 7 and 14; prior to dosing (0 h) on study days 5, 6, 12 and 13; and at 36, 48, 72 and 120 h after dosing on study day 14. Participants in cohorts 4–9 (12–60 mg dose groups) had an additional blood sample collected at 168 h after dosing on study day 14. The 72–168 h samples following the last dose were collected on an outpatient basis. The blood samples were centrifuged within 30 min of collection using a refrigerated centrifuge to separate the plasma. The plasma samples were frozen within 1 h after collection and maintained at −70°C or colder.

For participants who agreed to the lumbar puncture to collect CSF for assay of ABT-288, at least 2 ml of CSF was collected on study day 15 (∼24 h following the dose on study day 14) into a 2 ml polypropylene tube that had been precooled on ice for at least 5 min. The CSF samples were maintained on ice and stored in the freezer within 1 h after collection and maintained at −20°C or colder. Plasma concentrations of ABT-288 were determined using a validated protein precipitation extraction with high-performance liquid chromatography–tandem mass spectrometric detection at AbbVie (North Chicago, IL, USA) as described previously 14. Cerebrospinal fluid concentrations of ABT-288 were determined using the same high-performance liquid chromatography–tandem mass spectrometric detection but with direct injection.

Pharmacokinetic analyses

The plasma pharmacokinetic parameters of ABT-288 were estimated using noncompartmental analysis with WinNonlin-Professional™ software (version 5.2.1; Pharsight Corporation, Mountain View, CA, USA). The estimated pharmacokinetic parameters included the following: the maximal observed plasma concentration (C_max) and time to C_max (T_max); the minimal observed plasma concentration on days 7 and 14 (C_min); the apparent terminal phase elimination rate constant (β); the terminal phase elimination half-life (t_1/2); the area under the plasma concentration–time curve (AUC) from time 0 to 24 h after dosing (AUC_0–24); the apparent oral clearance (CL/F); the accumulation ratio (R_ac; calculated as the ratio of ABT-288 AUC_0–24 on days 7 and 14 of dosing to AUC_0–24 on day 1); and the CSF-to-plasma concentration ratios.

Statistical analyses

Statistical analyses were performed using SAS software (version 8.2; SAS Institute Inc., Cary, NC, USA).

For participants who enrolled in the study twice and therefore had data for two dose levels, only the data from the higher dose level were included for the statistical analyses of pharmacokinetic parameters.

To assess attainment of steady state, two analyses were conducted. First, a test was performed on the change from study day 7 to study day 14 in ABT-288 log-transformed C_max, C_min and AUC in the framework of a one-way analysis of variance (anova), with participants classified by dose level. Second, given that the comparison between day 7 and day 14 led to the conclusion of a difference in exposure between these two days, a repeated-measures analysis, with effects for dose level, study day and the interaction of dose level and study day, was performed on the log-transformed dose-normalized predose concentrations of days 12, 13 and 14. The hypothesis of no difference between day 14 and each of days 12 and 13 was tested at a significance level of 0.05.

To assess dose proportionality, analysis of covariance (ancova) was performed on β (determined from the concentrations after the last dose of a regimen) and logarithms of dose-normalized day 14 C_max, C_min and AUC. Bodyweight, body surface area (BSA), age, sex, creatinine clearance, smoking status and the six most frequently used other medications in the study (quetiapine, risperdone, olanzapine, aripiprazole, lorazepam and simvastatin) were examined as covariates for ABT-288 C_max, AUC and C_min (at a significance level of 0.10). Within the framework of the ancova with the final selection of covariates, the hypothesis of no difference between the means of the highest dose (60 mg) and lowest dose (1 mg) was tested as planned. As there were data for only three participants for the highest dose on day 14, the comparisons between the second-highest dose (45 mg) and the lowest dose, as well as between the average of the two highest doses and the lowest dose, were also performed. In addition, a test was performed on a contrast in the dose level effects, with the contrast chosen to be sensitive to an approximately linear function of the logarithm of dose.

Results

Participant disposition

A total of 67 adult male and female participants enrolled in the study and, of those, 14 participants enrolled in the study twice (eight participants received ABT-288 at two dose levels and six participants received placebo in one cohort and ABT-288 in another cohort; the duration between the two treatments ranged from 129 to 330 days). Only one subject received placebo in each of cohorts 6 and 8 (15 and 60 mg once-daily dose cohorts) due to slow enrolment. Nine participants (all received ABT-288) were prematurely discontinued from the study prior to completing the 14 day regimen. In cohorts 7–9, 12 participants (six who received 30 mg ABT-288, three who received 45 mg ABT-288 and three who received placebo) agreed to optional CSF sampling.

A summary of the demographics of study participants and the background antipsychotics used by the study participants is presented in Table 1.

Table 1.

Demographic characteristics of study participants

	Mean ± SD (n = 67)	Range
Age (years)	42 ± 9	23–60
Weight (kg)	89 ± 17	53–121
Height (cm)	174 ± 8	159–196
Body surface area (m2)	2.0 ± 0.20	1.5–2.4

	Number (%)
Sex	52 males (78%), 15 females (22%)
Race	16 white (24%), 47 black (70%), 1 Asian (1%), 3 other (4%)
Smoking status	50 smokers (75%), 11 nonsmokers (16%), 6 ex-smokers (9%)
Background antipsychotic	32 quetiapine (48%), 19 risperdione (28%), 9 olanzapine (13%), 5 aripiprazole (7%), 2 ziprasidone (3%)

Any subject who participated twice in the study was counted only once. Demographics were comparable for subjects who received ABT-288 or placebo.

Safety and tolerability

ABT-288 appeared generally to be safe and tolerated by stable subjects with schizophrenia up to multiple doses of 45 mg once daily for 14 days. The majority of adverse events in ABT-288 and placebo groups were considered possibly or probably related to the study drug. All adverse events in the placebo group were mild, and none resulted in premature discontinuation from the study. The majority of adverse events in ABT-288 dose groups were mild or moderate. One adverse event of psychotic disorder (15 mg once-daily dose) was severe. Seven participants administered ABT-288 prematurely discontinued the study drug regimen due to adverse events.

Adverse events reported by at least three participants who received ABT-288 are presented in Table 2 by dose level. These most commonly reported adverse events, in order of decreasing frequency, were abnormal dreams, headache, insomnia, dizziness, somnolence, dysgeusia, dry mouth, psychotic disorder, parosmia, tachycardia, constipation, anxiety and middle insomnia.

Table 2.

Adverse events reported by at least three participants who received ABT-288

	Treatment group n (%)
	ABT-288
Adverse event	Placebo (n = 20)	1 mg (n = 7)	3 mg (n = 7)	6 mg (n = 5)	9 mg (n = 6)	12 mg (n = 6)	15 mg (n = 6)	20 mg (n = 6)	30 mg (n = 6)	45 mg (n = 6)	60 mg (n = 6)	Overall ABT-288 (n = 61)
Any adverse event	18 (90)	5 (71)	4 (57)	4 (80)	5 (83)	3 (50)	6 (100)	5 (83)	6 (100)	6 (100)	6 (100)	50 (82)
Abnormal dreams	7 (35)	0	0	2 (40)	4 (67)	0	2 (33)	3 (50)	4 (67)	2 (33)	2 (33)	19 (31)
Headache	6 (30)	1 (14)	2 (29)	3 (60)	1 (17)	2 (33)	0	2 (33)	2 (33)	1 (17)	0	14 (23)
Insomnia	2 (10)	0	1 (14)	0	1 (17)	1 (17)	3 (50)	0	1 (17)	2 (33)	5 (83)	14 (23)
Dizziness	2 (10)	1 (14)	1 (14)	0	1 (17)	2 (33)	1 (17)	2 (33)	0	1 (17)	2 (33)	11 (18)
Somnolence	7 (35)	2 (29)	2 (29)	0	0	1 (17)	0	0	3 (50)	1 (17)	0	9 (15)
Dysgeusia	1 (5)	0	0	0	0	1 (17)	0	3 (50)	0	0	4 (67)	8 (13)
Dry mouth	0	0	0	1 (20)	3 (50)	1 (17)	0	0	1 (17)	0	0	6 (10)
Psychotic disorder	1 (5)	0	0	0	1 (17)	0	2 (33)	0	0	2 (33)	1 (17)	6 (10)
Parosmia	0	0	0	0	0	1 (17)	0	0	1 (17)	0	2 (33)	4 (7)
Tachycardia	0	0	0	0	0	0	0	0	0	1 (17)	2 (33)	3 (5)
Constipation	0	0	1 (15)	0	0	0	0	1 (17)	0	1 (17)	0	3 (5)
Anxiety	1 (5)	0	0	0	0	1 (17)	1 (17)	0	0	0	1 (17)	3 (5)
Middle insomnia	0	0	0	0	0	0	1 (17)	0	2 (33)	0	0	3 (5)

Six participants (two randomized to placebo and four randomized to ABT-288) were using a drug for treatment of insomnia prior to the study [zolpidem as needed or at bedtime (three), flurazepam once daily (one), hydroxazone twice daily (one) or trazadone once daily (one)] and continued its use during the study; of those, two participants randomized to ABT-288 [3 mg (one) and 15 mg (one)] reported the adverse event of insomnia during the study. Seven other participants initiated zolpidem (single dose to 8 days of use) to treat insomnia during the study; of those, two participants were randomized to placebo and five participants were randomized to ABT-288 [12 mg/60 mg (one), 15 mg (one), 30 mg (one), 45 mg (one) and 60 mg (one)].

Four participants (one randomized to placebo and three randomized to ABT-288) were using a drug for treatment of anxiety [lorazepam (three) and clonazepam (one)] prior to the study and continued its use during the study. Of these four participants, one participant (randomized to ABT-288 60 mg) reported the adverse event of anxiety during the study. Six other participants used lorazepam or clonazepam during the study (for treatment of anxiety or another disorder; one to several days of use); of these, one participant was randomized to placebo and five participants were randomized to ABT-288 [12 mg (one), 15 mg (two), 30 mg (one) and 60 mg (one)].

Five participants were using a drug for treatment of depression (citalopram, duloxetine, fuloxetine or mitrazapine) and continued its use during the study [placebo (one), 12 mg (one), 15 mg (one), 30 mg (one) and 45 mg (one)].

No deaths occurred during the study. One serious adverse event of increased psychosis (psychotic disorder) was reported in the 15 mg ABT-288 dose group. Seven participants who received ABT-288 [15 mg (one), 20 mg (one), 45 mg (two) and 60 mg once daily (three)] were discontinued due to adverse events that occurred during the study. The most frequent adverse event (four of seven cases) causing early termination was psychotic disorder [15 mg (one), 45 mg (two) and 60 mg (one)]. Additional adverse events that led to early termination were increased creatine phosphokinase [20 mg (one), 61 times greater than the upper limit of normal], pyrexia [60 mg (one)] and insomnia [60 mg (one)].

There were no clinically significant findings in measurements of individual vital signs or electrocardiograms other than the reported adverse events of tachycardia and autonomic nervous system imbalance in two participants who received ABT-288 45 mg once daily. Laboratory changes other than reported adverse events of increased creatine phosphokinase (one participant) and hypokalaemia (one participant) or laboratory changes associated with adverse events, such as increased white blood cell count and increased creatine phosphokinase (one participant), were generally sporadic, mild and not considered clinically significant.

Pharmacokinetics

Mean plasma concentration vs. time profiles after administration of multiple once-daily oral doses of ABT-288 to stable subjects with schizophrenia are presented in Figure 1. The pharmacokinetic parameters of ABT-288 are summarized in Table 3. Plasma concentrations of ABT-288 reached peak levels at approximately 2–4 h after dosing, on average. The harmonic mean terminal phase elimination half-life of ABT-288 ranged from 28 to 51 h. The median accumulation ratios of ABT-288 ranged from 2.0 to 3.3 on study day 7 and from 2.2 to 4.1 on study day 14 across dose groups.

Figure 1.

Mean ABT-288 plasma concentration vs. time profiles after administration of escalating multiple once-daily oral doses to subjects with schizophrenia. , 1 mg; , 3 mg; , 6 mg; , 9 mg; , 12 mg; , 15 mg; , 20 mg; , 30 mg; , 45 mg; , 60 mg

Table 3.

Pharmacokinetic parameters of ABT-288 (arithmetic mean ± SD) on day 14 after administration of once-daily doses to subjects with schizophrenia

Pharmacokinetic parameters	ABT-288 dose
	1 mg (n = 7)	3 mg (n = 6)	6 mg (n = 5)	9 mg (n = 6)	12 mg (n = 5)	15 mg (n = 5)	20 mg (n = 5)	30 mg (n = 6)	45 mg (n = 4)	60 mg (n = 3)
Cmax* (ng ml−1)	0.95 ± 0.19 (0.94)	3.32 ± 1.33 (3.08)	5.74 ± 1.26 (5.62)	11.7 ± 2.24 (11.5)	16.2 ± 3.98 (15.8)	20.2 ± 10.5 (18.4)	40.0 ± 13.8 (38.1)	34.9 ± 12.0 (33.3)	65.0 ± 34.2 (58.0)	86.7 ± 29.7 (83.0)
Tmax†(h)	3.0 [2.0–9.0]	3.5 [2.0–6.0]	3.0 [2.0–6.0]	2.0 [1.5–3.0]	3.0 [2.0–4.0]	3.0 [0.5–6.0]	3.0 [2.0–12.0]	3.0 [2.0–4.0]	2.5 [2.0–4.0]	3.0 [1.5–3.0]
Cmin (ng ml−1)	0.44 ± 0.07	1.58 ± 0.65	2.50 ± 0.74	6.90 ± 2.05	7.95 ± 1.55	9.74 ± 6.37	20.5 ± 6.23	15.3 ± 8.99	32.2 ± 19.4	47.5 ± 22.0
AUC0–24* (ng h ml−1)	14.8 ± 2.50 (14.6)	52.8 ± 22.2 (48.7)	92.7 ± 24.0 (90.3)	202 ± 46.6 (198)	266 ± 61.8 (260)	330 ± 200 (290)	672 ± 187 (653)	538 ± 224 (500)	1060 ± 579 (939)	1480 ± 602 (1400)
t1/2β‡ (h)	28 ± 5.8	38 ± 11	35 ± 17	37 ± 14	37 ± 4.4	36 ± 8.6	36 ± 7.2	43 ± 16	34 ± 9.1	51 ± 13
CL/F (l h−1)	69 ± 12	67 ± 31	68 ± 17	47 ± 11	47 ± 11	58 ± 29	31 ± 7.5	65 ± 27	54 ± 30	46 ± 19
Accumulation ratio†	3.0 [2.2–3.8]	2.9 [1.6–3.9]	2.2 [1.8–3.1]	3.4 [2.9–4.6]	3.9 [3.2–4.0]	3.5 [2.0–4.2]	3.6 [3.4–5.3]	3.2 [1.7–4.9]	3.3 [2.0–3.7]	4.1 [2.7–4.8]

AUC_0-24, area under the plasma concentration-time curve for the dosing interval; C_max, maximum observed plasma concentration; C_min, minimum observed plasma concentration; CL/F, apparent oral clearance; T_max, time at which C_max is observed.

^*Geometric mean is also presented in parentheses.

^†Presented as median and range.

^‡Presented as harmonic mean ± pseudostandard deviation.

Values of C_max, C_min and AUC_0–24 for ABT-288 were statistically significantly higher (P < 0.05) on study day 14 than those values on study day 7 (the lower bounds of the 95% confidence interval of the ratio of study day 14 to study day 7 parameters were >1). However, the statistical significance of the difference in the pharmacokinetic parameters between study day 14 and study day 7 was not consistently reached for the individual dose levels. The overall mean of the dose-normalized predose concentrations for all multiple-dose regimens tested on study day 14 was not statistically significantly different (P > 0.05) from the overall mean on study day 12 or 13, indicating that a steady state was generally achieved for ABT-288 by day 12 of once-daily dosing.

Dose-normalized C_max and AUC₂₄ vs. dose relationships for ABT-288 are presented in Figure 2. Statistical comparison between the lowest ABT-288 dose (1 mg) and the two highest doses (45 and 60 mg) or the average of the two highest doses revealed no statistically significant differences (P > 0.05) for dose-normalized C_max, C_min or AUC_0–24 values. The 60 mg dose (and the average of the 45 and 60 mg doses) had a statistically significantly lower β value than that of the 1 mg dose (P < 0.05). There was no statistically significant (P > 0.05) linear trend of change in ABT-288 C_max, C_min or AUC_0–24 values with dose. Body surface area was a statistically significant covariate for ABT-288 C_max and AUC (P < 0.05) but did not reach significance for ABT-288 C_min (P = 0.13). Based on the statistical model estimates, a typical subject with BSA of 1.5 m² (lowest BSA of participants in the study) is estimated to have 31 and 36% higher ABT-288 C_max and AUC values, respectively, than a typical subject with BSA of 2.0 m² (the median BSA of participants in the study). A typical subject with BSA of 2.4 m² (highest BSA of participants in the study) is estimated to have 19 and 22% lower ABT-288 C_max and AUC values, respectively, than a typical subject with a BSA of 2.0 m². Other covariates, including smoking status and concomitant use of each of the six most frequently administered concomitant medications in the study, were not found to be significant covariates for ABT-288 exposure.

Figure 2.

Dose-normalized maximal observed ABT-288 plasma concentration (C_max; A; , day 1; , day 14) and area under the plasma concentration–time curve (AUC; B; , day 1; , day 14) values (mean ± SD) vs. dose plots

The median CSF-to-total plasma concentration ratio was 0.4 for ABT-288 at 24 h after administration of the last dose at the two dose levels evaluated (Table 4). Plasma protein binding was not measured in subjects with schizophrenia in the present study. Plasma protein binding of ABT-288 is ∼70% in human plasma and is not gender or concentration dependent (data on file). Using a protein binding value of 70%, the CSF-to-free ABT-288 steady-state trough concentration ratio is calculated to be ∼1.4.

Table 4.

Cerebrospinal fluid and total plasma concentrations of ABT-288 (means ± SD) 24 h after administration of the last dose

	30 mg ABT-288 (n = 6)	45 mg ABT-288 (n = 3)
Cerebrospinal fluid concentration (ng ml−1)	7.35 ± 3.14	15.4 ± 9.39
Plasma concentration (ng ml−1)	17.4 ± 7.52	37.3 ± 19.8
Cerebrospinal fluid/total plasma ratio*	0.42 [0.19–1.01]	0.40 [0.36–0.46]

^*Median and range are presented.

Discussion

This study was designed to assess the safety, tolerability and pharmacokinetics of multiple oral doses of ABT-288 in stable subjects with schizophrenia. Data from this study were used to select doses for a larger clinical study of the efficacy of ABT-288 as an adjunctive treatment for cognitive impairment in schizophrenia. Two previous studies evaluated ABT-288 safety, tolerability and pharmacokinetics in healthy young adults and elderly subjects. In both young and elderly subjects, the maximal tolerated ABT-288 dose was 3 mg once daily (mean C_max of ∼5 ng ml⁻¹ and AUC of ∼90 ng h ml⁻¹). Higher doses (5 mg in elderly subjects or 6 mg in young adults, C_max ≥ 7 ng ml⁻¹ and AUC ≥ 120 ng h ml⁻¹) were clearly intolerable 14. Dose-limiting effects appeared to be manifestations of excessive histamine, such as insomnia, abnormal dreams, hot flushes, nausea, dizziness, anxiety and palpitations. Therefore, the present study was initially designed to test doses up to 6 mg once daily in two cohorts. However, due to the observed improvement in tolerability, the protocol was amended, and cautious dose escalation continued up to 60 mg once daily. ABT-288 appeared generally to be safe and tolerated at multiple doses up to 45 mg once daily (mean plasma C_max of 65 ng ml⁻¹ and AUC of 1060 ng h ml⁻¹). As such, ABT-288 was tolerated at a 15-fold higher dose and approximately 12-fold higher plasma exposures in subjects with schizophrenia than previously observed in healthy young adults and elderly volunteers.

In subjects with schizophrenia, the adverse events profile was generally consistent with that in healthy subjects, but adverse events occurred at higher doses and plasma exposures. The increased incidence of some of the adverse events with ABT-288 dose, particularly abnormal dreams, insomnia and dysgeusia, suggest an association with ABT-288 (Table 2). In addition, the higher frequency of psychosis-related events and the greater rate of treatment-related discontinuations due to adverse events in the ABT-288 groups vs. placebo suggest that ABT-288 was activating.

To the best of our knowledge, this is the first report of greater tolerability of an H₃ antagonist in subjects with schizophrenia than in healthy subjects. It is well known that subjects with schizophrenia tolerate antipsychotics at higher doses than subjects without schizophrenia 17–19. For example, haloperidol and clozapine are tolerated at 200- and 36-fold higher doses, respectively, in subjects with schizophrenia than in healthy subjects 17. Features of the disease that may contribute to greater tolerability of antipsychotics include the following: (i) increased dopamine release; (ii) overexpression of the dopamine D₂ receptors in certain brain regions; and (iii) predominance of the monomeric form of the D₂ receptor in schizophrenia patients compared with control subjects 20,21. Subjects with schizophrenia also generally require larger doses of other psychotropic medications, but the underlying reasons are not well understood. Pharmacological effects of H₃ antagonists are believed to be mediated mainly by increases in histamine and acetylcholine, with lower contributions of dopamine and norepinephrine. Therefore, the greater tolerability of ABT-288 in subjects with schizophrenia was not fully anticipated. It is not clear whether the improved tolerability is a result of the pathophysiology of the disease or of a pharmacodynamic interaction between atypical antipsychotics and ABT-288. The use of hypnotics and anxiolytics in the present study was sporadic and was not believed to have significantly impacted the assessment of ABT-288 tolerability. Given the small incidence of adverse events in the present study, data from larger phase 2 trials will be more informative on the possible differential impact of concomitant medication use on expression of ABT-288-related adverse events.

Levels of ABT-288 in the cerebrospinal fluid were available from participants in the 30 and 45 mg once-daily dose groups. Plasma protein binding in subjects with schizophrenia was not characterized in the present study. Assuming the protein binding was not largely different from a previously characterized mean value of ∼70% in plasma of healthy volunteers, ABT-288 trough CSF concentrations in subjects with schizophrenia are not lower than (if they did not exceed) the free plasma concentrations. These data were consistent with the high in vitro permeability of ABT-288 and suggested that ABT-288 readily accesses its site of action in subjects with schizophrenia. Therefore, we do not believe that improved tolerability of ABT-288 in subjects with schizophrenia is due to limited brain uptake.

The variability in ABT-288 steady-state exposure in subjects with schizophrenia (% coefficient of variation for AUC up to 60%) was higher than that previously observed in healthy young and elderly subjects (% coefficient of variation for AUC up to 27%) 14. ABT-288 exposure was generally dose proportional in healthy subjects; however, the highest multiple dose evaluated in healthy subjects was 6 mg once daily. In the present study, there was no statistical evidence that ABT-288 exposure (C_max, C_min or AUC) deviated significantly from dose proportionality in subjects with schizophrenia over the 1–60 mg multiple-dose range. However, there was some apparent trend of increase in dose-normalized exposures in the 1–9 mg dose range (Figure 2) that did not reach statistical significance.

The mean elimination half-life of ABT-288 clustered around 34–38 h in seven of 10 dose groups (Table 2) in the present study. As such, ABT-288 elimination half-life appears to be shorter in subjects with schizophrenia than previously observed in healthy subjects (40–61 h). At the common multiple-dose range across populations (doses up to 5 or 6 mg once daily), the steady-state accumulation ratio of ABT-288 in subjects with schizophrenia (2.2–3.0) was lower than that in healthy subjects (3.4–4.2), and the dose-normalized exposure (AUC) of ABT-288 in subjects with schizophrenia was approximately 55% of the exposure in healthy subjects, on average 14. Several factors were considered in order to investigate the underlying reasons for the apparent difference in ABT-288 exposure and half-life between healthy subjects and subjects with schizophrenia. Body surface area was found to be a significant covariate for ABT-288 exposure in the present study. However, there was a small difference (∼5%) in the average BSA between participants of this study and the healthy young volunteers previously evaluated in a phase 1 study. As such, it is unlikely that body size contributed significantly to the difference in exposure between healthy subjects and subjects with schizophrenia (∼6% difference in ABT-288 exposure is calculated for a 5% difference in BSA). Other possible variables, such as smoking status, concomitant medication use or differences in race distribution across studies (the majority of participants were black in the present study vs. Caucasian in previous studies), were considered, but there was no evidence that any of these covariates correlated significantly with ABT-288 exposure.

Based on the results of the present study, 10 and 25 mg once-daily doses of ABT-288 were selected to test the hypothesis for treatment of cognitive impairment in subjects with schizophrenia adequately. Higher doses than 25 mg were thought to be at risk for intolerance in an unacceptable number of subjects in a phase 2 setting despite the overall tolerability up to 45 mg in the present study. Both the 10 and 25 mg doses result in plasma exposures that are higher than the efficacious levels in preclinical models of cognition 11. It is noteworthy that the 10 and 25 mg doses complement the dose levels that were selected for evaluation in subjects with Alzheimer's disease (1 and 3 mg once daily; based on the observed limits of tolerability in elderly subjects) 14. Overall, the selected ABT-288 doses across the Alzheimer's disease and schizophrenia phase 2 studies (1–25 mg once daily) should provide information on the possible procognitive effects at a wide range of exposure in humans. Detailed reports of the results of the phase 2 studies of ABT-288 are warranted.

In conclusion, the present study characterized the safety, tolerability and pharmacokinetics of ABT-288 in subjects with schizophrenia on stable doses of an atypical antipsychotic. At comparable multiple doses, ABT-288 exposure in subjects with schizophrenia was 45% lower than that previously observed in healthy subjects. ABT-288 was tolerated at a 15-fold higher dose and ∼12-fold higher plasma exposures in subjects with schizophrenia than previously characterized in healthy volunteers. Higher ABT-288 tolerability was not due to limited brain uptake. This is the first report of improved tolerability of an H₃ antagonist in subjects with schizophrenia relative to healthy subjects.

Competing Interests

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: the study presented in this manuscript was funded by AbbVie (AbbVie was previously part of Abbott), and AbbVie was responsible for data collection, monitoring and analysis of the data; the authors were responsible for study design, analysis and interpretation of data and writing the manuscript; Drs Othman, Haig, Florian, Locke and Dutta are employees and shareholders of AbbVie; Dr Gertsik is an employee of California Clinical Trials and was the investigator for the study. The authors declare no other relationships or activities that could appear to have influenced the submitted work.