Safety and Tolerability of Brexpiprazole in Participants with Agitation Associated with Dementia due to Alzheimer’s Disease: Pooled Analysis of Three Randomized Trials and an Extension Trial

Abstract

Background and Objective

Older adults with dementia are particularly vulnerable to antipsychotic side effects. Brexpiprazole, an atypical antipsychotic, is approved in a number of countries for the treatment of agitation associated with dementia due to Alzheimer’s disease. This pooled analysis aimed to evaluate the safety and tolerability of brexpiprazole in this patient population.

Methods

Data were included from three Phase 3, 12-week, randomized, double-blind, placebo-controlled trials and a Phase 3, 12-week, active-treatment extension trial in participants with agitation associated with dementia due to Alzheimer’s disease. Safety outcomes included treatment-emergent adverse events (TEAEs), weight change, suicidality, extrapyramidal symptoms, and cognitive dysfunction. Two datasets were considered: a 12-week dataset that pooled data from the three randomized trials for brexpiprazole 0.5–3 mg/day and placebo, and a 24-week dataset that combined data from the parent randomized trial and the extension trial for brexpiprazole 2–3 mg/day.

Results

Over 12 weeks, 335/655 (51.1%) participants on brexpiprazole and 178/388 (45.9%) participants on placebo reported ≥ 1 TEAE, which led to discontinuation in 41 (6.3%) and 13 (3.4%) participants, respectively. Headache was the only TEAE with incidence ≥ 5% (brexpiprazole, 50 [7.6%] participants; placebo, 36 [9.3%] participants). The incidences of cerebrovascular TEAEs (brexpiprazole, 0.5%; placebo, 0.3%), cardiovascular TEAEs (3.7%; 2.3%), extrapyramidal symptom-related TEAEs (5.3%; 3.1%), and somnolence/sedation TEAEs (3.7%; 1.8%) were generally similar between treatment groups. Six (0.9%) participants on brexpiprazole and 1 (0.3%) participant on placebo died; causes of death were not considered related to brexpiprazole and were generally in line with those expected in Alzheimer’s disease. Over 24 weeks, 110/226 (48.7%) participants on brexpiprazole reported ≥ 1 TEAE, which led to discontinuation in 19 (8.4%) participants. Headache was the only TEAE with incidence ≥ 5% (18 [8.0%] participants). No participants died during the extension trial. Over 12 and 24 weeks, mean changes in weight, suicidality, and extrapyramidal symptoms were minimal, with no worsening of cognition.

Conclusions

Considering pooled data from > 1000 participants on brexpiprazole or placebo, brexpiprazole appears to be generally well tolerated for up to 24 weeks in participants with agitation associated with dementia due to Alzheimer’s disease.

Study Registration

ClinicalTrials.gov identifiers: NCT01862640, NCT01922258, NCT03548584, NCT03594123.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40263-025-01200-9.

Key Points

In this extensive pooled analysis, brexpiprazole (up to 3 mg/day) was generally well tolerated with no specific safety concerns over up to 24 weeks for the treatment of agitation associated with dementia due to Alzheimer’s disease.

Adverse events/effects associated with other atypical antipsychotics—cerebrovascular, cardiovascular, venous thromboembolism, extrapyramidal symptoms, somnolence, falls, fractures, pneumonia, acute kidney injury/urinary tract infection, and cognitive decline—were generally reported at comparable levels to placebo.

Introduction

Agitation is a prevalent feature of Alzheimer’s disease/dementia, affecting 45% of individuals in community settings and 53% in care facilities [1, 2]. Together with other neuropsychiatric symptoms, agitation is a particularly challenging and stressful aspect of dementia care, being associated with increased morbidity and mortality, earlier nursing home placement, and greater caregiver burden [3]. Agitation can be exacerbated by reversible factors such as uncomfortable physical symptoms and pain, medications, delirium, or infection, and these factors should be investigated and addressed before considering agitation-specific interventions [4].

Agitation can sometimes be managed by nonpharmacological interventions, such as caregiver training to help understand and respond to behavioral disturbances [4, 5]. For individuals requiring pharmacotherapy, brexpiprazole, an atypical antipsychotic, is approved in a number of countries for the treatment of agitation associated with dementia due to Alzheimer’s disease [6–8]. The efficacy and safety of brexpiprazole in this indication were demonstrated in two international, Phase 3, randomized, fixed-dose clinical trials (supported by a post hoc analysis of a third trial with flexible dosing) and an active-treatment extension trial [9–11]. In these trials, efficacy was assessed using the Cohen-Mansfield Agitation Inventory (CMAI), which is a measure of the frequency of occurrence of 29 agitation behaviors [12, 13]. Over 12 weeks in the fixed-dose trials, brexpiprazole demonstrated statistically significant reduction in the frequency of agitation symptoms versus placebo when dosed at 2 mg/day (Cohen’s d effect size, 0.25; p = 0.040) and 2 or 3 mg/day (effect size, 0.35; p = 0.003) [9, 10]. Other atypical antipsychotics are sometimes used off-label to treat agitation, despite risk of harm to the patient [14–17].

Older adults are particularly vulnerable to antipsychotic side effects [17]. In the USA, all atypical antipsychotics carry a class boxed warning against use in “elderly patients with dementia-related psychosis,” which has been in place since 2005 [18]. This warning is supported by meta-analyses (conducted prior to the availability of brexpiprazole) showing an increased risk of mortality in older adults with dementia, predominantly due to cardio-/cerebrovascular events and infections [18–20]. For brexpiprazole, the boxed warning was amended in 2023 to state that brexpiprazole is “not approved for the treatment of patients with dementia-related psychosis without agitation associated with dementia due to Alzheimer’s disease” [6]. In addition to cardio-/cerebrovascular events, atypical antipsychotic use in people with dementia is associated with venous thromboembolism, extrapyramidal symptoms (EPS), somnolence, falls (in some studies), fractures, pneumonia, acute kidney injury/urinary tract infection, and cognitive decline [14, 15, 17, 21, 22]. Given these concerns with other antipsychotics, it is important to perform a comprehensive assessment of the safety of brexpiprazole in people with Alzheimer’s disease. The aim of this pooled analysis was to evaluate the safety and tolerability of brexpiprazole in participants with agitation associated with dementia due to Alzheimer’s disease, on the basis of data from three randomized trials and an extension trial.

Methods

Trial Designs and Participants

The trials in this analysis were conducted in accordance with the International Conference on Harmonisation Good Clinical Practice Guideline and local regulatory requirements. The trial protocols were approved by relevant institutional review boards and independent ethics committees. All participants and/or their legal representatives provided written or electronic informed consent prior to the start of the respective trial.

Randomized Trials

Data were included from three international, Phase 3, 12-week, multicenter, randomized, double-blind, placebo-controlled trials of brexpiprazole in participants with agitation associated with dementia due to Alzheimer’s disease: Trial 283 (ClinicalTrials.gov identifier: NCT01862640) [9], Trial 284 (NCT01922258) [9], and Trial 213 (NCT03548584) [10]. The trials were carried out in Europe, Russia, the USA, and Canada between 2013 and 2017 (Trials 283 and 284) and between 2018 and 2022 (Trial 213). The trials had similar designs, which have been previously published [9, 10], and are summarized here.

The trials enrolled participants aged 55–90 years with a diagnosis of probable Alzheimer’s disease by National Institute of Neurological and Communicative Disorders and Stroke—Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria [23] (other dementias were excluded) and a Mini-Mental State Examination (MMSE) [24] score of 5–22 (indicating cognitive impairment). Participants had a previous magnetic resonance imaging or computed tomography scan of the brain that had findings consistent with a diagnosis of Alzheimer’s disease and that indicated no other clinically significant central nervous system disease such as vascular changes (e.g., cortical stroke and multiple infarcts), space-occupying lesion (e.g., tumor), or other major structural brain disease. Participants were either living in a care facility or in a community-based setting provided they had a caregiver with sufficient contact to describe their symptoms and behaviors. Participants had agitation as indicated by a score ≥ 4 on the Neuropsychiatric Inventory (standard or Nursing Home version) Agitation/Aggression domain [25, 26] and, additionally in Trial 213, met International Psychogeriatric Association agitation criteria [27] and CMAI factor 1 (aggressive behaviors) positivity criteria [10, 12, 13]. Each participant’s agitation required pharmacotherapy based on investigator judgment after an evaluation for reversible factors and a trial of nonpharmacologic intervention. Participants with current clinically significant or uncontrolled medical conditions other than Alzheimer’s disease were excluded (whereas stable, controlled hypertension and diabetes were permitted). Participants with a positive drug screen for cocaine, cannabis (whether prescribed or not), or other illicit drugs were excluded.

The trials comprised a screening/washout period of up to 42 days, a 12-week, double-blind treatment period, and a 30-day post-treatment follow-up period (for participants who did not enroll in the extension trial). In the treatment period, participants were randomized to brexpiprazole (fixed or flexible doses ranging from 0.5 to 3 mg/day, depending on the trial) or placebo, as illustrated in Fig. 1. Stable background medications for the treatment of Alzheimer’s disease (e.g., acetylcholinesterase inhibitors, and memantine), most antidepressants (e.g., citalopram, escitalopram, mirtazapine, sertraline, and trazodone), buspirone, and medications to treat other medical conditions such as hypertension, hypercholesterolemia, and anti-platelet agents were permitted. Medications that are cytochrome P450 2D6 (CYP2D6) inhibitors or cytochrome P450 3A4 (CYP3A4) inhibitors and inducers, including certain antidepressants (e.g., fluoxetine, fluvoxamine), were prohibited. Benzodiazepines were permitted during the first 4 weeks only for agitation or insomnia. Antipsychotics, mood stabilizers (e.g., lithium, valproate, and carbamazepine), and anticonvulsants were prohibited. A full list of permitted and prohibited medications is published in the Trial 213 protocol (an online supplement to Lee et al. 2023 [10]). Visits occurred at 2-week intervals during the treatment period (and also at Day 3 for Trials 283 and 284).

Fig. 1.

Trial designs. ^aWith a 30-day post-treatment safety follow-up for participants who did not enter Trial 182. ^bThe trial initially included a fourth treatment arm, brexpiprazole 0.5 mg/day, which was removed (after 20 participants had been randomized to this arm) in response to new information indicating this dose was non-efficacious (the 20 participants were allowed to continue). ^cThis was a two-arm trial; however, within the brexpiprazole arm, participants were further randomized in a 1:2 ratio to 2 mg/day or 3 mg/day. ^dTrials 283 and 284: Days 1–3, 0.25 mg/day; Days 4–14, 0.5 mg/day; Days 15–28, 1 mg/day; and Day 29 onwards, assigned dose (flexible in 284). Trial 213: Days 1–7, 0.5 mg/day; Days 8–14, 1 mg/day; Days 15–28, 2 mg/day; and Day 29 onwards, assigned dose. ^eWith a 30-day post-treatment safety follow-up. ^fParticipants who received brexpiprazole in Trial 213 continued their previous dose; participants who received placebo in Trial 213 were titrated to brexpiprazole: Days 1–7, 0.5 mg/day; Days 8–14, 1 mg/day; and Day 15 onwards, 2 mg/day. One dose decrease and one dose increase within the 2–3 mg/day range were permitted. MMSE Mini-Mental State Examination, NPI-AA Neuropsychiatric Inventory Agitation/Aggression domain, NPI-NH-AA Neuropsychiatric Inventory−Nursing Home version, Agitation/Aggression domain

Extension Trial

This analysis also included data from a Phase 3, 12-week, multicenter, active-treatment extension trial: Trial 182 (NCT03594123) [11]. The trial was conducted in Europe and the USA between October 2018 and September 2022. The trial design has been previously published [11], and is summarized here.

Trial 182 enrolled participants who had completed 12 weeks of treatment in Trial 213 (termed the “parent” trial), regardless of whether participants had received brexpiprazole or placebo in the parent trial. As illustrated in Fig. 1, Trial 182 comprised a 12-week, active-treatment period and a 30-day post-treatment follow-up period (plus a mortality assessment by telephone 16 weeks after starting for participants who withdrew early). In the treatment period, all participants received brexpiprazole 2 or 3 mg/day. Participants remained blinded to the treatment they had received in the parent trial; participants previously receiving placebo were titrated in the extension trial, whereas participants previously receiving brexpiprazole continued on their same dose from the parent trial. Relative to baseline of the parent trial, visits occurred at Weeks 13, 14, 18, and 24.

Assessments

Demographic information, including age, sex, and race, was collected at screening. Each participant’s baseline level of agitation was determined using the 29-item CMAI [12, 13] and the Clinical Global Impression–Severity of Illness (CGI-S) [28], specifically applied to agitation.

Adverse events were recorded by the investigator at all post-baseline visits. To elicit the reporting of adverse events, the participant, accompanied by the caregiver, was asked “How have you felt since your last visit?” Adverse events were coded using Medical Dictionary for Regulatory Activities (MedDRA) version 25.0 preferred terms. Treatment-emergent adverse events (TEAEs) were defined as adverse events that started after trial drug treatment or were continuous from baseline and worsened, became serious or trial drug related, or resulted in death, discontinuation, interruption, or reduction of trial therapy. A serious TEAE was defined as any event that resulted in death, was life-threatening, resulted in persistent or significant incapacity or disability or substantial disruption of the ability to conduct normal life functions, required inpatient hospitalization or prolonged hospitalization; a congenital anomaly or birth defect; or another medically significant event that, based upon appropriate medical judgment, may jeopardize the participant and may require medical or surgical intervention to prevent one of the outcomes listed above. TEAE severity was graded as follows: mild (discomfort noticed, but no disruption to daily activity), moderate (discomfort sufficient to reduce or affect normal daily activity), and severe (inability to work or perform normal daily activities). Selected TEAE categories of interest were also considered, as described in Supplementary Table S1.

Objective cognitive dysfunction was measured using the MMSE, an 11-item questionnaire that covers orientation, memory, attention, language skills, and ability to follow instructions [24]. MMSE scores range from 0 to 30, where lower scores indicate greater cognitive impairment [24]; various cut-offs are used to define severity, of which the brexpiprazole studies used mild (> 18), moderate (13–18), and severe (≤ 12) [9–11]. The MMSE was completed by participants at screening, baseline, Week 12, and Week 24.

Vital signs were measured at each visit. Body weight was measured at each visit in Trials 283 and 284; at screening, baseline, and Week 12 in Trial 213; and at Week 24 in Trial 182. Clinical laboratory tests (including glucose, cholesterol, and triglycerides) and electrocardiograms were completed periodically at different visits depending on the trial. Scales to measure EPS (Simpson–Angus Scale [29], Abnormal Involuntary Movement Scale [28], and Barnes Akathisia Rating Scale [30]) and the Sheehan Suicidality Tracking Scale (STS) [31] were also completed periodically. The Sheehan-STS assesses treatment-emergent suicidal ideation and behaviors across eight items, with total scores ranging from 0 (no suicidality) to 32 (extreme suicidality) [31].

Statistical Analysis

Two datasets were considered: (1) a 12-week dataset that pooled data from the three randomized trials (Trials 283, 284, and 213) and (2) a 24-week dataset that combined data from the parent randomized trial (Trial 213) and the extension trial (Trial 182).

In the 12-week analysis, data were pooled for all participants randomized to brexpiprazole, and separately for all participants randomized to placebo. In addition, subgroups were created for the following doses: (1) low fixed doses (0.5 or 1 mg/day, representing US Food and Drug Administration [FDA]-recommended starting and titration doses); (2) standard fixed doses (2 or 3 mg/day, representing FDA-recommended target and maximum doses); and (3) flexible doses (0.5–2 mg/day) [6].

The intention of the 24-week analysis was to generate data for up to 24 weeks of brexpiprazole treatment. A single brexpiprazole-treatment group was created by combining data across the parent and extension trials. Specifically, the analysis included all participants who were randomized to brexpiprazole in the parent trial; these participants had data for up to 12 or 24 weeks depending on whether or not they enrolled in the extension trial. The rationale for including all brexpiprazole-treated participants from the parent trial regardless of whether they enrolled in the extension trial was to capture TEAE data for all available brexpiprazole-treated participants, including participants who discontinued during the parent trial. No 24-week comparator group was created, because there was no placebo arm in the extension trial. Thus, participants who were randomized to placebo in the parent trial were not included in the 24-week analysis.

For both the 12-week and 24-week analyses, baseline was defined as the baseline visit of the randomized trial, i.e., the last evaluation prior to the first dose of trial medication (brexpiprazole or placebo). Baseline demographics were summarized using descriptive statistics in the randomized sample. Baseline clinical characteristics and all safety analyses were performed in the safety sample, comprising all randomized participants who received at least one dose of trial medication. All safety results were summarized using descriptive statistics for observed cases data. Analyses were performed using SAS version 9.4 (SAS Institute Inc; Cary, NC).

Results

Participants

The 12-Week Analysis

Across the randomized trials, 658 participants were randomized to brexpiprazole (all doses) and 389 to placebo (Table 1). Of these participants, 571 (86.8%) in the brexpiprazole group and 346 (88.9%) in the placebo group completed the trials. The most common reasons for discontinuation (≥ 2%) were an adverse event (brexpiprazole, 6.2%; placebo, 3.6%) and withdrawal by the participant (4.4%; 3.3%); a full list is provided in Table 1. Mean (standard deviation [SD]) exposure to brexpiprazole was 79.4 (14.7) days and to placebo was 80.0 (13.7) days.

Table 1.

Participant disposition

Participants, n (%)	12-week analysis					24-week analysis Brexpiprazole 2 or 3 mg
	Placebo	Brexpiprazole all doses	Brexpiprazole 0.5 or 1 mg fixed	Brexpiprazole 2 or 3 mg fixed	Brexpiprazole 0.5–2 mg flexible
Randomized	389	658	157	368	133	228
Received ≥ 1 dose of trial medication (safety sample)	388 (99.7)	655 (99.5)	157 (100.0)	366 (99.5)	132 (99.2)	226 (99.1)
Completed	346 (88.9)	571 (86.8)	134 (85.4)	320 (87.0)	117 (88.0)	142 (62.3)
Discontinued	43 (11.1)	87 (13.2)	23 (14.6)	48 (13.0)	16 (12.0)	51 (22.4)a
Adverse event	14 (3.6)	41 (6.2)	14 (8.9)	18 (4.9)	9 (6.8)	20 (8.8)
Participant withdrew consent	13 (3.3)	29 (4.4)	6 (3.8)	18 (4.9)	5 (3.8)	17 (7.5)
Participant met withdrawal criteria	5 (1.3)	4 (0.6)	2 (1.3)	2 (0.5)	0	0
Site terminated by sponsor	2 (0.5)	4 (0.6)	0	4 (1.1)	0	8 (3.5)
Withdrawn by investigator	5 (1.3)	3 (0.5)	1 (0.6)	1 (0.3)	1 (0.8)	0
Lack of efficacy	0	1 (0.2)	0	1 (0.3)	0	2 (0.9)
Lost to follow-up	2 (0.5)	1 (0.2)	0	0	1 (0.8)	0
Non-compliance with trial medication	0	1 (0.2)	0	1 (0.3)	0	1 (0.4)
Protocol deviation	0	1 (0.2)	0	1 (0.3)	0	0
Other	2 (0.5)	2 (0.3)	0	2 (0.5)	0	3 (1.3)

^aAnother 35/228 (15.4%) participants randomized to brexpiprazole completed the 12-week parent trial but did not enter the 12-week extension trial; these participants are not counted as discontinued

Baseline demographic and clinical characteristics were generally similar between treatment groups and subgroups (Table 2). Considering the total randomized sample (n = 1047), mean (SD) age was 73.9 (8.0) years, 604 (57.7%) participants were female, 443 (42.3%) were male, 1002 (95.7%) were White, and 45 (4.3%) were Black or African American, Asian, or other race. In the total safety sample, mean (range) MMSE score at baseline was 14.6 (5–25), and mean (range) CMAI total and CGI-S scores were 73.7 (35–140) and 4.6 (2–7), respectively (n = 1043).

Table 2.

Baseline demographic and clinical characteristics and concomitant medications

Characteristica	12-week analysis					24-week analysis Brexpiprazole 2 or 3 mg
	Placebo	Brexpiprazole all doses	Brexpiprazole 0.5 or 1 mg fixed	Brexpiprazole 2 or 3 mg fixed	Brexpiprazole 0.5–2 mg flexible
Randomized sample	(n = 389)	(n = 658)	(n = 157)	(n = 368)	(n = 133)	(n = 228)
Age (years)	73.7 (7.6)	74.0 (8.2)	73.8 (8.8)	74.2 (7.9)	73.5 (8.5)	74.5 (7.7)
Sex, n (%)	–	–	–	–	–	–
Female	218 (56.0)	386 (58.7)	90 (57.3)	214 (58.2)	82 (61.7)	135 (59.2)
Male	171 (44.0)	272 (41.3)	67 (42.7)	154 (41.8)	51 (38.3)	93 (40.8)
Race, n (%)	–	–	–	–	–	–
Asian	5 (1.3)	6 (0.9)	1 (0.6)	5 (1.4)	0	3 (1.3)
Black or African American	11 (2.8)	22 (3.3)	2 (1.3)	16 (4.3)	4 (3.0)	11 (4.8)
White	373 (95.9)	629 (95.6)	154 (98.1)	347 (94.3)	128 (96.2)	214 (93.9)
Other	0	1 (0.2)	0	0	1 (0.8)	0
Weight (kg)	69.6 (14.1)	69.3 (14.8)b	69.0 (14.3)	69.7 (15.0)	68.6 (14.8)c	70.5 (15.5)
BMI (kg/m2)	25.9 (4.8)	25.6 (4.5)b	25.3 (4.9)	25.8 (4.5)	25.2 (4.2)c	26.3 (4.7)
Time since Alzheimer’s disease diagnosis (months)	32.8 (31.6)	33.6 (34.7)	35.6 (39.1)	34.7 (34.6)	28.2 (28.3)	36.7 (36.9)
Time since onset of current episode of agitation associated with Alzheimer’s disease (months)	5.9 (8.4)d	8.2 (17.6)e	7.0 (17.3)f	9.7 (18.1)g	5.2 (16.5)c	10.0 (14.8)
Safety sample	(n = 388)	(n = 655)	(n = 157)	(n = 366)	(n = 132)	(n = 226)
MMSE scoreh	14.8 (4.1)	14.5 (3.9)	13.1 (3.8)	15.1 (3.7)	14.7 (4.1)	15.6 (3.7)i
CMAI total scorej	72.9 (17.6)	74.1 (17.3)	69.9 (16.8)	76.9 (17.2)	71.5 (16.8)	80.4 (16.7)i
CGI-S score as related to agitationk	4.6 (0.7)	4.6 (0.7)	4.5 (0.6)	4.6 (0.7)	4.5 (0.8)	4.7 (0.7)i
Took medication for Alzheimer’s disease during trial, n (%)	315 (81.2)	511 (78.0)	118 (75.2)	285 (77.9)	108 (81.8)	184 (81.4)
Memantine	200 (51.5)	325 (49.6)	74 (47.1)	185 (50.5)	66 (50.0)	119 (52.7)
Donepezil	128 (33.0)	204 (31.1)	41 (26.1)	123 (33.6)	40 (30.3)	85 (37.6)
Took concomitant medication for agitation during trial, n (%)	44 (11.3)	84 (12.8)	19 (12.1)	59 (16.1)	6 (4.5)	45 (19.9)

BMI body mass index, CGI-S Clinical Global Impression–Severity of Illness, CMAI Cohen-Mansfield Agitation Inventory, MMSE Mini-Mental State Examination

^aData are mean (standard deviation) unless otherwise stated

^bn = 657

^cn = 132

^dn = 388

^en = 653

^fn = 155

^gn = 366

^hRange 0–30; lower scores indicate greater cognitive impairment [24]

ⁱn = 228 (randomized sample)

^jRange 29–203; higher scores indicate higher frequency of agitation behaviors [12]

^kRange 1–7; higher scores indicate more severe agitation [28]

The 24-Week Analysis

For the 24-week analysis, 228 participants were allocated to brexpiprazole in the parent trial, of whom 226 (99.1%) were treated, 198 (86.8%) completed 12 weeks of treatment, 163 (71.5%) entered the extension trial, and 142 (62.3%) completed the extension trial (Table 1). The most common reasons for discontinuation (≥ 2%) were an adverse event (8.8%), withdrawal by the participant (7.5%), and the site being terminated by the sponsor (3.5%); a full list is provided in Table 1.

Baseline demographic and clinical characteristics in the 24-week analysis aligned with those of the 12-week analysis (Table 2).

Adverse Events

The 12-Week Analysis

Overall, 335/655 (51.1%) participants treated with brexpiprazole (all doses) and 178/388 (45.9%) participants treated with placebo reported at least one TEAE. The overall incidence of TEAEs did not appear to increase with brexpiprazole dose (Table 3). Serious TEAEs were reported by 42 (6.4%) participants on brexpiprazole (all doses) and 16 (4.1%) participants on placebo; more participants reported serious TEAEs in the low-fixed-dose brexpiprazole subgroup (10.2%) compared with the other subgroups (5.2–5.3%). No specific serious TEAEs had incidence ≥ 1% in the brexpiprazole (all doses) group. Severe TEAEs were reported by 38 (5.8%) participants on brexpiprazole (all doses) and 16 (4.1%) participants on placebo, with similar incidence across brexpiprazole subgroups (5.2–6.8%). No specific severe TEAEs had incidence ≥ 1% in the brexpiprazole (all doses) group. TEAEs led to discontinuation in 41 (6.3%) participants on brexpiprazole (all doses) and 13 (3.4%) participants on placebo; again, this was more common in the low-fixed-dose brexpiprazole subgroup (8.9%) than in the other subgroups (4.9–6.8%). No specific TEAEs led to discontinuation in ≥ 1% of participants in the brexpiprazole (all doses) group.

Table 3.

Incidence of treatment-emergent adverse events (safety sample)

Event, n (%)	12-week analysis					24-week analysis Brexpiprazole 2 or 3 mg (n = 226)
	Placebo (n = 388)	Brexpiprazole all doses (n = 655)	Brexpiprazole 0.5 or 1 mg fixed (n = 157)	Brexpiprazole 2 or 3 mg fixed (n = 366)	Brexpiprazole 0.5–2 mg flexible (n = 132)
At least one TEAE	178 (45.9)	335 (51.1)	77 (49.0)	183 (50.0)	75 (56.8)	110 (48.7)
At least one serious TEAE	16 (4.1)	42 (6.4)	16 (10.2)	19 (5.2)	7 (5.3)	12 (5.3)
At least one severe TEAE	16 (4.1)	38 (5.8)	10 (6.4)	19 (5.2)	9 (6.8)	11 (4.9)
Discontinuation due to TEAE	13 (3.4)	41 (6.3)	14 (8.9)	18 (4.9)	9 (6.8)	19 (8.4)
Death	1 (0.3)	6 (0.9)	4 (2.5)	2 (0.5)	0	1 (0.4)a
TEAEs with an incidence ≥ 2% in the brexpiprazole (all doses) group and greater than placebo in the 12-week analysis
Insomnia	11 (2.8)	24 (3.7)	7 (4.5)	12 (3.3)	5 (3.8)	5 (2.2)
Somnolence	7 (1.8)	22 (3.4)	2 (1.3)	12 (3.3)	8 (6.1)	11 (4.9)
Nasopharyngitis	10 (2.6)	18 (2.7)	5 (3.2)	9 (2.5)	4 (3.0)	7 (3.1)
Urinary tract infection	6 (1.5)	17 (2.6)	3 (1.9)	12 (3.3)	2 (1.5)	7 (3.1)
Other TEAEs of interest
Headache	36 (9.3)	50 (7.6)	12 (7.6)	28 (7.7)	10 (7.6)	18 (8.0)
Dizziness	13 (3.4)	21 (3.2)	1 (0.6)	14 (3.8)	6 (4.5)	10 (4.4)
Fall	10 (2.6)	11 (1.7)	2 (1.3)	7 (1.9)	2 (1.5)	9 (4.0)
Electrocardiogram QT prolonged	2 (0.5)	8 (1.2)	4 (2.5)	3 (0.8)	1 (0.8)	1 (0.4)
Pneumonia	3 (0.8)	8 (1.2)	2 (1.3)	2 (0.5)	4 (3.0)	2 (0.9)
Akathisia	1 (0.3)	5 (0.8)	0	2 (0.5)	3 (2.3)	2 (0.9)
Extrapyramidal disorder	0	5 (0.8)	1 (0.6)	3 (0.8)	1 (0.8)	3 (1.3)
Orthostatic hypotension	2 (0.5)	3 (0.5)	1 (0.6)	1 (0.3)	1 (0.8)	0
Seizure	1 (0.3)	3 (0.5)	0	0	3 (2.3)	1 (0.4)
Acute kidney injury	0	2 (0.3)	0	2 (0.5)	0	2 (0.9)
Sedation	0	2 (0.3)	0	1 (0.3)	1 (0.8)	1 (0.4)
Aspiration pneumonia	0	1 (0.2)	1 (0.6)	0	0	0
Hip fracture	2 (0.5)	1 (0.2)	0	1 (0.3)	0	2 (0.9)
Syncope	3 (0.8)	1 (0.2)	1 (0.6)	0	0	1 (0.4)
Femur fracture	1 (0.3)	0	0	0	0	1 (0.4)

TEAE treatment-emergent adverse event

^aNo participants died in the extension trial; the one death in the 24-week analysis occurred in the parent trial

Headache was the only individual TEAE with incidence ≥ 5% in the brexpiprazole (all doses) group (50 participants; 7.6%) or the placebo group (36 participants; 9.3%). TEAEs with an incidence ≥ 2% in the brexpiprazole (all doses) group and greater than placebo were insomnia (brexpiprazole, 3.7%; placebo, 2.8%), somnolence (3.4%; 1.8%), nasopharyngitis (2.7%; 2.6%), and urinary tract infection (2.6%; 1.5%) (Table 3). There was no consistent relationship between brexpiprazole dose and the incidence of these TEAEs.

The incidence of TEAEs in grouped categories of interest, including cerebrovascular, cardiovascular, QT prolongation, venous thromboembolism, EPS, somnolence, orthostatic hypotension/dizziness/syncope (including blood pressure abnormalities), accident/injury/fall, and seizure, are listed in Supplementary Table S2. Categories with an incidence ≥ 2% in the brexpiprazole (all doses) group and greater than placebo were any cardiovascular TEAE (brexpiprazole, 3.7%; placebo, 2.3%), any EPS-related TEAE (5.3%; 3.1%), and any somnolence/sedation TEAE (3.7%; 1.8%). The only individual EPS-related TEAE with incidence ≥ 1% was tremor (1.2%; 1.5%). Cerebrovascular TEAEs had a low incidence (brexpiprazole, 0.5%; placebo, 0.3%). TEAEs in the MedDRA metabolism and nutrition disorder system order class were reported by 23 (3.5%) participants in the brexpiprazole (all doses) group and 17 (4.4%) participants in the placebo group, and in the renal and urinary disorders system order class by 12 (1.8%) and 5 (1.3%), respectively. TEAEs of suicidal ideation or behavior were reported by 0 participants on brexpiprazole and 2 (0.5%) participants on placebo (1 intentional self-injury and 1 suicidal ideation).

Six (0.9%) participants who received brexpiprazole and 1 (0.3%) participant who received placebo died during the 12-week double-blind treatment period or protocol-specified 30-day post-treatment follow-up period (two additional deaths were reported after the 30-day follow-up period, both for the brexpiprazole 0.5–2 mg flexible-dose arm and considered not related to trial medication, and are not discussed further). Four of the participants who died were in the low-fixed-dose brexpiprazole subgroup. All of the deaths were considered by the investigator to be not related or unlikely related to trial medication. Brief narratives for the deaths are as follows: (1) In the brexpiprazole 0.5 mg subgroup (last dose: Day 8), an 87-year-old female participant had a serious TEAE of intracranial hemorrhage on Day 30, was hospitalized and received corrective treatment; the participant subsequently died from this event on Day 35. Autopsy revealed massive subarachnoid and parenchymal hemorrhage of the left anterior frontal and parietal lobes. The participant had a medical history of hypertension, subarachnoid hemorrhage, and congestive heart failure, and had experienced a mild fall on Day 8 with electrocardiogram abnormalities (corrective treatment included clopidogrel). (2) In the brexpiprazole 0.5 mg subgroup (last dose: Day 50), a 76-year-old male participant had a serious TEAE of encephalitis on Day 52. Despite corrective treatment, the participant died later the same day due to severe acute purulent meningoencephalitis. The participant had ongoing chronic cardiac failure, myocardial ischemia, and hypertensive heart disease for 18 months before their first dose of brexpiprazole. (3) In the brexpiprazole 1 mg subgroup (last dose: Day 65), a 78-year-old male participant had a serious TEAE of aspiration pneumonia on Day 63, was hospitalized, and received corrective treatment; on Day 73 the participant was transferred to hospice care for comfort care measures and subsequently died of aspiration pneumonia on Day 78. The participant’s medical history included chronic obstructive pulmonary disease and encephalopathy. (4) In the brexpiprazole 1 mg subgroup (last dose: Day 85; completed trial medication), a 66-year-old female participant living in a care home had a serious TEAE of obstructive airways disorder (choked on an orange) on Day 110. The participant was resuscitated, but entered a coma, and subsequently died on Day 152. (5) In the brexpiprazole 2 mg subgroup (last dose: Day 86; completed trial medication), an 86-year-old female participant had a serious event of Alzheimer’s dementia on Day 90 (the participant had been gradually declining in health and was refusing food, liquids, and medications) and died of end-stage Alzheimer’s dementia in hospice care on Day 95. (6) In the brexpiprazole 3 mg subgroup (last dose: Day 28 [2 mg, as still being titrated]), a 78-year-old male had a serious TEAE of cardiac failure on Day 51 and received corrective treatment; the participant died the same day. The participant also had pneumonia, and autopsy revealed coronary atherosclerosis. (7) In the placebo group (last dose: Day 74), an 86-year-old male participant who was bedbound and living in a care home was diagnosed with severe pneumonia on Day 72. Despite receiving antibiotic therapy, the participant’s respiratory function worsened, resulting in death from respiratory insufficiency due to pneumonia on Day 76.

The 24-Week Analysis

Overall, 110/226 (48.7%) participants who received brexpiprazole in the 24-week analysis reported at least one TEAE. Of these, 12 (5.3%) participants reported at least one serious TEAE, and 11 (4.9%) participants reported at least one severe TEAE. TEAEs led to discontinuation in 19 (8.4%) participants.

Headache was the only individual TEAE with an incidence ≥ 5% (8.0%). Additional TEAEs with an incidence ≥ 2% were somnolence (4.9%), dizziness (4.4%), fall (4.0%), asthenia (3.1%), diarrhea (3.1%), nasopharyngitis (3.1%), urinary tract infection (3.1%), blood pressure increased (2.2%), insomnia (2.2%), and respiratory disorder (2.2%). Other TEAEs of interest are listed in Table 3.

The incidence of TEAEs in grouped categories of interest are listed in Supplementary Table S2. Categories with incidence ≥ 2% were any EPS-related TEAE (6.6%), any somnolence/sedation TEAE (5.3%), any orthostatic hypotension/dizziness/syncope TEAE (4.9%), and any accident or injury TEAE including falls (4.4%). No individual EPS-related TEAE had an incidence ≥ 2%. TEAEs in the MedDRA metabolism and nutrition disorder system order class were reported by 3 (1.3%) participants, and in the renal and urinary disorders system order class by 5 (2.2%) participants. No participants reported TEAEs of suicidal ideation or behavior. No participants died during the extension trial.

Cognition

In the 12-week analysis, there was minimal change from baseline to Week 12 in mean (SD) MMSE score for participants in the brexpiprazole (all doses) group (+0.3 [2.4]; n = 561 at Week 12) and the placebo group (+0.1 [2.2]; n = 338 at Week 12). Corresponding changes in the brexpiprazole dose subgroups were also minimal, in the range –0.2 to +0.5. Similarly, in the 24-week analysis, there was minimal change from baseline to Week 24 in mean (SD) MMSE score among participants who received brexpiprazole (+0.4 [3.1]; n = 121 at Week 24) (Fig. 2).

Fig. 2.

MMSE score over time (24-week analysis; safety sample). Range 0–30; lower scores indicate greater cognitive impairment [24]. N-values indicate the number of participants with a measurement at that timepoint. MMSE Mini-Mental State Examination, SD standard deviation

Body Weight, Vital Signs, and Laboratory Tests

In the 12-week analysis, mean (SD) change in body weight was minimal from baseline to Week 12 for participants in the brexpiprazole (all doses) group (+0.2 [2.2] kg; n = 568 at Week 12) and the placebo group (−0.1 [1.8] kg; n = 345 at Week 12). Corresponding changes in the brexpiprazole dose subgroups were also minimal, in the range −0.2 to +0.3 kg. In the 24-week analysis, mean (SD) change in body weight was minimal from baseline to Week 24 among participants who received brexpiprazole (0.0 [4.6] kg; n = 122 at Week 24) (Fig. 3A).

Fig. 3.

Body weight over time, showing (A) change from baseline and (B) clinically relevant change at any visit (safety sample). N-values in A indicate the number of participants with a measurement at that timepoint and, in B, indicate the number of participants with a measurement at baseline and at least one post-baseline visit. SD standard deviation

The proportions of participants experiencing a ≥ 7% increase or decrease in body weight at any visit in the 12-week analysis were low and similar between the brexpiprazole (all doses) and placebo groups (Fig. 3B). In the 24-week analysis, 3.2% of participants experienced a ≥ 7% increase and 5.0% experienced a ≥ 7% decrease in body weight at any visit (Fig. 3B).

There were no meaningful differences between brexpiprazole and placebo in vital signs, laboratory parameters, or electrocardiograms in the 12-week analysis (data not shown). Over 24 weeks, changes in vital signs, laboratory parameters, and electrocardiograms were not considered to be clinically meaningful (data not shown; 24-week laboratory parameters and QT interval are previously published [11]). No participants had a new-onset QT corrected for heart rate by Fridericia’s formula (QTcF) interval value > 500 ms at any visit in the 12-week analysis or the extension trial. In the 12-week analysis, QTcF interval change of > 30–≤ 60 ms at any post-baseline visit occurred in 66/645 (10.2%) participants in the brexpiprazole (all doses) group and 44/382 (11.5%) participants in the placebo group; corresponding values for QTcF interval change of > 60 ms at any post-baseline visit were 2/645 (0.3%) and 6/382 (1.6%). In the extension trial, 7/145 (4.8%) participants (who received brexpiprazole in the parent trial) had QTcF interval change of > 30–≤ 60 ms at any post-baseline visit, and 1/145 (0.7%) had change > 60 ms at any post-baseline visit.

Suicidality and Extrapyramidal-Symptom Rating Scales

In the 12-week analysis, mean (SD) Sheehan-STS total score at baseline was 0.0 (0.2) (n = 655) in the brexpiprazole (all doses) group and 0.0 (0.1) (n = 388) in the placebo group. Mean (SD) change from baseline to Week 12 was 0.0 (0.3) (n = 568 at Week 12) in the brexpiprazole (all doses) group and 0.0 (0.1) (n = 345 at Week 12) in the placebo group. Corresponding changes in each of the brexpiprazole dose subgroups were also all 0.0. In the 24-week analysis, mean (SD) Sheehan-STS total score at baseline was 0.0 (0.2) (n = 226), and mean (SD) change from baseline to Week 24 was 0.0 (0.1) (n = 141 at Week 24).

Mean changes in EPS rating scales in the 12- and 24-week analyses were minimal, as shown in Supplementary Table S3.

Discussion

The findings of this analysis in a large sample of over 1000 participants from four Phase 3 clinical trials indicate that brexpiprazole at doses up to 3 mg/day is generally well tolerated in participants with agitation associated with dementia due to Alzheimer’s disease. The overall incidences of TEAEs (brexpiprazole, 51.1%; placebo, 45.9%) and discontinuation due to TEAEs (6.3%; 3.4%) over 12 weeks were broadly similar between brexpiprazole and placebo. Over 24 weeks, the incidence of TEAEs (48.7%) and discontinuation due to TEAEs (8.4%) on brexpiprazole were close to the 12-week values, supporting long-term tolerability. Most of the reported TEAEs were mild or moderate in severity. The incidence of TEAEs did not increase with standard therapeutic doses of brexpiprazole (2 or 3 mg/day) compared with low doses. Besides headache, which had a higher incidence with placebo, there were no specific TEAEs with an incidence ≥ 5% with brexpiprazole in the 12- or 24-week analyses.

Over 12 weeks, the incidence of death was numerically higher with brexpiprazole (6 participants; 0.9%) than placebo (1 participant; 0.3%). A prior meta-analysis of these three trials found no statistically significant difference in all-cause mortality between brexpiprazole and placebo [32]. Of note, four of the six deaths among participants receiving brexpiprazole occurred at 0.5 or 1 mg/day (below the target dose), and five of the six deaths occurred > 1 week after the participant’s last dose of brexpiprazole. None of the deaths were considered related to brexpiprazole by the investigator, and causes of death were generally in line with the types of events expected in people with Alzheimer’s disease (i.e., intracranial hemorrhage, pneumonia, choking, end-stage Alzheimer’s dementia, and heart failure). There were no deaths on brexpiprazole 2 or 3 mg/day during the extension trial.

Certain TEAEs are of particular relevance to atypical antipsychotic use in people with dementia [14, 15, 17, 21], and are considered “adverse events of special interest” in clinical drug trials. For brexpiprazole, the incidence of any cerebrovascular TEAE was low (0.5%) and comparable to placebo (0.3%) over 12 weeks, and no participants experienced cerebrovascular TEAEs in the 24-week analysis. No participants had a stroke during the trials. These data support that the risk of cerebrovascular events was not increased during treatment with brexpiprazole. Similarly, the incidences of cardiovascular TEAEs (including myocardial infarction, cardiac failure, and ventricular arrhythmia), venous thromboembolism, EPS, somnolence, falls, fractures, pneumonia, and acute kidney injury/urinary tract infection were low and generally similar between brexpiprazole and placebo. The incidence of these events was also generally similar in the 24-week analysis compared with the 12-week analysis. With regard to cognition, the minimal clinically important difference for MMSE score has been estimated at 1.4–2 points [33], with positive changes indicating improvement [24]. In this analysis, the mean change in MMSE score from baseline to Weeks 12 and 24 was positive but less than 1 point among participants treated with brexpiprazole, which can be interpreted as no worsening of cognition. Mean changes on EPS rating scales were minimal, indicating no meaningful effect on akathisia or other EPS. There was no indication of suicidal ideation or behavior on brexpiprazole. Finally, across psychiatric diagnoses, most antipsychotic medications are associated with clinically significant weight gain [34], though the association is less certain in older adults with dementia [35]. The present analysis demonstrated minimal change in body weight over 24 weeks in participants who received brexpiprazole.

Prior publications show that the overall incidence of TEAEs over 12 weeks was higher in the first two trials (Trial 283: 56.6% for brexpiprazole 0.5, 1, or 2 mg/day fixed and 45.9% for placebo; Trial 284: 56.8% for brexpiprazole 0.5−2 mg/day flexible and 58.4% for placebo) than in the third trial (Trial 213: 40.7% for brexpiprazole 2 or 3 mg/day fixed and 31.0% for placebo), despite faster titration to 2 mg in the third trial [9, 10]. There is no obvious explanation for this difference; baseline characteristics were generally similar between trials, except for more severe agitation (CMAI total score) in Trial 213 due to the additional agitation-related inclusion criteria. This observation highlights the value of pooling data to obtain an overall picture of the safety of brexpiprazole beyond that shown in individual trials.

Strengths of this analysis are that it is one of the largest prospectively collected safety analyses for agitation associated with dementia due to Alzheimer’s disease, and that long-term analyses were conducted to generate 24 weeks of data. Furthermore, generalizability to clinical practice was increased by the trial protocols allowing participants with certain medically stable, chronic conditions (such as hypertension and diabetes), and allowing participants to take concomitant Alzheimer’s disease medications, antidepressants, and some other background medications. Limitations include a potential bias towards increased tolerability in the 24-week sample since 35/228 (15.4%) participants randomized to brexpiprazole who completed the parent trial did not enter the extension trial, and the reason for this decision or whether it related to adverse events is unknown. The 24-week analysis had no comparator, since all participants in the extension trial received brexpiprazole. While no participant had a stroke during the trials, the generalizability of this result is limited because the trials excluded individuals with vascular changes (based on prior magnetic resonance imaging or computed tomography scans), and no baseline data on cerebral microbleeds were collected. The nature of clinical trials, particularly the exclusion of people with concurrent clinically significant or uncontrolled medical conditions, may limit the generalizability of the results. Finally, although this analysis included over 1000 participants, even larger databases are needed to investigate rarer adverse events.

Conclusions

On the basis of this analysis of pooled data from over 1000 participants on brexpiprazole or placebo across three randomized trials and an extension trial, brexpiprazole at therapeutic doses appears to be generally well tolerated in participants with agitation associated with dementia due to Alzheimer’s disease. There were no specific safety concerns raised by this analysis of data accumulated over up to 24 weeks. Adverse events/effects associated with other atypical antipsychotics—cerebrovascular, cardiovascular, venous thromboembolism, EPS, somnolence, falls, fractures, pneumonia, acute kidney injury/urinary tract infection, and cognitive decline—were generally reported at comparable levels to placebo. Together with the demonstrated efficacy of brexpiprazole dosed at 2 or 3 mg/day [9, 10], the tolerability profile indicated by this extensive analysis supports the use of brexpiprazole as a treatment option for people with agitation associated with dementia due to Alzheimer’s disease.

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Funding

This work was supported by Otsuka Pharmaceutical Development & Commercialization Inc. (Princeton, NJ, USA) and H. Lundbeck A/S (Valby, Denmark). The sponsors were involved in the design of the research, the analysis and interpretation of data, and the writing and reviewing of this article. The sponsors paid the open access fee.

Conflicts of interest

Alpesh Shah, Dalei Chen, Denise Chang, Mary Slomkowski, Mary Hobart, and Malaak Brubaker are full-time employees of Otsuka Pharmaceutical Development & Commercialization Inc. Alvin Estilo, Daniel Lee, and Saloni Behl were full-time employees of Otsuka Pharmaceutical Development & Commercialization Inc. at the time of this work. Pamela L. Sheridan was a full-time employee of Lundbeck LLC at the time of this work. Uwa Kalu is a full-time employee of Lundbeck LLC. Nanco Hefting and Pedro Such are full-time employees of H. Lundbeck A/S. George T Grossberg has received consulting fees from Acadia, Abbott, Axsome, Biogen, BMS, BioXcel, Eisai, Karuna, Lundbeck, MapLight Therapeutics, Otsuka, and Takeda. George T. Grossberg is on the Editorial Board for CNS Drugs, Journal of the American Medical Directors Association, and International Journal of Alzheimer’s Disease; he was not involved in the selection of peer reviewers for this manuscript nor any of the subsequent editorial decisions. He is on Safety Monitoring Committees for Anavex, Quince, Merck, Newron, and Oligomerix.

Availability of data and material

To submit inquiries related to Otsuka clinical research, or to request access to individual participant data (IPD) associated with any Otsuka clinical trial, please visit https://clinical-trials.otsuka.com/. For all approved IPD access requests, Otsuka will share anonymized IPD on a remotely accessible data sharing platform.

Ethics approval

This analysis used pooled data from published clinical trials. The trials were approved by relevant institutional review boards and independent ethics committees and were performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments.

Code availability

Not applicable.

Consent to publish

Not applicable.

Consent to participate

This analysis used pooled data from published clinical trials. All participants and/or their legal representatives provided written or electronic informed consent prior to the start of the trials.

Author contributions

All authors contributed to the study conception and design. Data analysis was performed by Dalei Chen and Denise Chang. All authors critically reviewed the manuscript for intellectual content. All authors read and approved the final manuscript.