Brexpiprazole for the Treatment of Agitation Associated with Dementia due to Alzheimer’s Disease: Clinical Perspectives

Abstract

Dementia is a neuropsychiatric disorder that primarily affects the elderly, leading to a widespread decline in cognitive function and significant impairment of occupational, social, and personal functioning. In addition to cognitive deficits, dementia is frequently comorbid with behavioral and psychological symptoms of dementia (BPSD), such as agitation. When present, these secondary symptoms can exacerbate the clinical course of the disease, reduced treatment responsiveness, increased rates of admission to long-term care facilities, extended hospitalization, higher risk of personal injury and a substantial socioeconomic burden. Given these consequences, early management of BPSD—particularly agitation—is critical to mitigating these risks. Although antipsychotics are commonly prescribed to manage agitation, risperidone remains the only agent approved by regulatory authorities for this indication. Recently, however, brexpiprazole, a medication with a pharmacological profile distinct from that of risperidone, received U.S. FDA approval (on May 11, 2023) for the treatment of agitation associated with Alzheimer’s disease. Agitation is among the most prevalent BPSD manifestations, with symptoms ranging from verbal to physical aggression. Given its recent approval and unique pharmacodynamic properties, brexpiprazole may have strong potential as a therapeutic option for this population. This paper aims to review the pharmacological mechanisms, clinical evidence, and future perspectives of brexpiprazole as a novel therapeutic option for managing agitation in patients with Alzheimer’s disease.

INTRODUCTION

Dementia is a degenerative brain disease in which cognitive functions such as memory, language ability, judgment, and learning ability substantially decline, and it is a disease that significantly interferes with the patient’s ability to perform daily life. Dementia is not limited to a decline in cognitive function, but is accompanied by various changes in the patient’s behavior, emotions, and personality [1-4].

In particular, behavioral and psychological symptoms of dementia (BPSD) appears in more than 80% of dementia patients, and representative symptoms include agitation, hallucinations, delusions, anxiety, aggression, nervousness, sleep disorders, and depression. The overall frequency of BPSD increasing tendency by severity of dementia, 67% in mild dementia, 76% in moderate dementia and 88% in severe dementia [5]. These BPSDs not only seriously reduce the quality of life (QoL) of the patients themselves, but also cause extreme psychological and physical burdens to their families and caregivers [6].

BPSD is not simply secondary symptoms, but have a significant impact on the course of the disease, and early detection and appropriate intervention are mandatory. However, because the expression of BPSD is very diverse and varies from patient to patient, identifying the cause and establishing an effective treatment strategy remain challenging tasks. Among the BPSD in dementia patients, the prevalence of agitation reaches approximately 40% based on a recent meta-analysis in acute hospital setting [7]. It is also one of the symptoms that is particularly difficult to manage, and it appears in forms of verbal and psychical manifestations as well as comorbid with different symptoms such as aggression, restlessness, and anxiety. Agitation seriously impairs the QoL of dementia patients, increases the stress of caregivers, and is often a decisive factor in causing the need for early institutionalization or medication [6,8-10].

Cognitive enhancers such as donepezil, rivastigmine, galantamine, and memantine are known to help alleviate BPSD. However, there is currently a lack of clearly approved treatments for agitation, and off-label medication use is common in clinical settings. As a result, antidepressants, antiepileptic drugs, benzodiazepines and antipsychotics (APs) are commonly used to manage BPSD in routine practice [11]. In particular, most treatment guidelines recommend the limited use of APs due to the risk of various adverse events (AEs), such as extrapyramidal symptoms (EPS), sedation, falls, drowsiness, and others [11]. It is also recommended that APs should be used for the shortest duration at the lowest effective dose, given that traditional AEs of APs may impact clinical courses as well. Therefore, atypical antipsychotics (AAPs), such as risperidone, olanzapine, aripiprazole, and quetiapine, which have relatively fewer AEs and unique mechanisms of action compared to typical antipsychotics (tAPs), are primarily used and recommended in clinical practice [9,12]. The increased prescription of AAPs compared to tAPs in dementia patients has been consistently reported in many studies. For example, a study conducted in Japan showed that AAPs prescription accounted for approximately 50% of APs use in 2006, while such trends rose to 71% in 2012 and the use of tAPs decreased by 15% from 2006 to 2012 [13].

Given aforementioned, brexpiprazole (BRP) has recently been introduced as a new treatment option for agitation in patients with Alzheimer’s disease (AD). BRP is an AAP that acts as a partial agonist at both 5-HT1A and dopamine D2 receptors, a mechanism that distinguishes it from other currently available AAPs. Furthermore, BRP is also differentiated by its favorable tolerability and AEs profile, making it a potentially safer and more suitable therapeutic option compared with existing agents [14,15].

MARKET POSITION OF AAPS

The European Neurology and Dementia Agitation Guidelines (2020) state that AAPs can be used as a first-line treatment strategy for agitation and aggression symptoms in patients for whom non-pharmacological treatments are ineffective and that AAPs can be used individually according to the patient’s clinical profiles [16]. The administration of APs to dementia patients is reported in approximately 20−50% of cases, though this varies depending on the research methodology [17-19]. Among AAPs and tAPs, only risperidone and haloperidol have been approved for the treatment of BPSD [12], however, none of the APs have been approved by U.S. FDA for the treatment of agitation associated with AD. Therefore, most APs are prescribed off-label for the treatment of BPSD, proposing the unmet need for treatment options with better efficacy and fewer AEs in routine practice. Recently, BRP, which was previously approved for the treatment of schizophrenia and as adjunct therapy for major depressive disorder (MDD), was eventually approved by the U.S. FDA on May 11, 2023, for the treatment of agitation associated with AD.

PHARMACOLOGY OF BRP

BRP is pharmacologically similar enough to be considered a successor to aripiprazole. Its oral bioavailability is 95%, and the highest plasma concentration is reached within 4 hours after ingestion. A steady state is typically achieved within 10 to 12 days. BRP has a high protein-binding and is not significantly affected by kidney or liver impairment [20]. Pharmacodynamically, BRP acts as a partial agonist of the 5-HT1A receptor (5-HTR1A) as well as dopamine receptors D2 (DR2) and DR3. This action mechanisms can improve various neuropsychiatric symptoms and also provide additional benefits for cognition, anxiety, and mood stability [14,15]. BRP has lower intrinsic activity at DR2 compared to aripiprazole, making it more effective in reducing the risk of EPS. Additionally, it has a higher affinity for 5-HTR2, 5-HTR1A, and α1B/α2C adrenoreceptors than aripiprazole, offering an advantage in lower presentation of EPS, and care of anxiety, restlessness and agitation [14,15]. Antagonism at the 5-HTR7 receptor, combined with partial agonism at the 5-HTR1A receptor, helps improve cognitive function and produce antidepressant effects [14,15]. BRP has low sedation effect due to its low affinity and weak binding potential for histamine receptor 1 (HR1), proposing that the effect of BRP on agitation/aggression should be less related to its sedation effects [14,15,21,22]. The detailed receptor binding affinity is shown in Figure 1.

Fig. 1.

The action mechanism and binding affinity of brexpiprazole (unit: nM).

Modified from references [14,15,20-23].

Lower Ki values representing higher affinity. High binding affinity, Ki < 1 nM.

5-HT, serotonin; D, dopamine; α, alpha adrenergic.

RELEVANCE OF BRP FOR BPSD IN AD PATIENTS

The neurobiological causes of BPSD are not fully understood. However, it is believed that these symptoms are related to impaired regulation of key neurotransmitters, including dopamine (DA), serotonin (5-HT), and norepinephrine (NE). Traditionally, abnormalities in the DA system have been hypothesized as a primary etiological mechanism for BPSD, leading to the common use of tAPs for managing these symptoms. However, the limited efficacy and frequent AEs including EPS in more than 25% of patients treated with tAPs, have prompted the suggestion of a more complex deficit involving multiple neurotransmitter systems rather than solitary defects of DA. In fact, dysregulation of the dynamic balance between various neurotransmitter systems could impair the neural networks involved in mood instability, agitation, aggression, anxiety, acting out, and cognition [23-25].

Alterations in 5-HT have been consistently associated with the development of various neuropsychiatric symptoms, including agitation, in both animal and human studies [24]. A deficit in 5-HT may contribute to dysfunction within the prefrontal cortex (PFC), hippocampus, and amygdala—key brain regions where executive functions, memory, and emotional regulation are dynamically integrated. In addition, 5-HT system deficits have been consistently reported in AD patients with aggression compared to those without aggression [26,27].

5-HTR1A is crucially involved in the development of AD, as they are highly expressed in the human PFC, raphe nuclei, hippocampus and amygdala, a region critically involved in the regulation of cognition and emotion [24,28,29]. Furthermore, 5-HTR1A is involved in the regulation of several important neurotransmitters, including acetylcholine (ACh), NE, and DA, which play key roles in impulsivity, aggression, cognition, and emotional regulation [24,29]. Indeed 5-HTR1A alteration has been found to be directly related to aggression in AD, while dementia severity is more strongly related to the neurodegenerative process [30]. A previous postmortem study using 33 AD patients with prospective evaluation of cognition and behavioral symptoms along with 20 matched controls found that 5-HTR1A binding affinity and density were unchanged in the overall AD group compared with controls, while 5-HTR1A density in the temporal cortex was inversely correlated with aggression and dementia severity. Even after statistical correction, 5-HTR1A density remained the best predictor for aggression in the study [30]. The 5-HTR1A agonists (e.g., buspirone) have also been consistently reported to reduce social isolation-induced aggression in animal models [31] as well as in AD patients [32]. Alterations in 5-HTR1A receptors also serve as a gateway for controlling prefrontal and subcortical circuits [33], which aligns with the finding that frontal lobe dysfunction may predispose AD patients to agitation, triggered by underlying pathologies and other factors [34]. BRP has also been shown to improve phencyclidine-induced neuronal deficits in rodent through activation of 5-HTR1A [15,35].

5-HTR2 has also been found to be a part of responsibility in developing BPSD in AD patients, despite it is still controversial due to conflicting results [36]. Agitated AD patients showed significantly higher changes in prolactin levels compared to non-agitated AD patients, three hours after the administration of d,l-fenfluramine, a 5-HTR2 agonist [26]. Additionally, a significant positive correlation was found between the level of agitation, as assessed by the Cohen-Mansfield Agitation Inventory (CMAI), and the degree of change from baseline in prolactin levels at 3 hours in the study [26]. The 5-HTR2 gene polymorphism has also been found to be associated with various BPSD in AD patients [37,38].

The potential role of 5-HTR7 antagonists has consistently emerged as a promising target for treating mood instability, obsession and compulsion, stereotyped behaviors, and anxiety, as demonstrated by numerous animal and human studies [21,39]. Recent studies have found a strong relationship between alterations in 5-HTR7 and the development of neuropsychiatric symptoms in AD patients [40]. For example, decreased levels of 5-HT7 mRNA were observed in the PFC of AD patients, and a significant association was found between 5-HT7 levels in the PFC and Behavioral Pathology in Alzheimer’s Disease (BEHAVE-AD) cluster B symptoms (hallucinations) score. Furthermore, 5-HTR7 in the prelimbic areas was shown to regulate anxiety-like behaviors, affecting DA, 5-HT, and NE levels in the limbic system [40,41]. In addition close interaction between 5-HTR1A and 5-HTR7 has been found in cognition system [42].

A disruption of NE transmission has been strongly and consistently reported to increase impairment of executive control and emotional impulses, thereby increasing agitation and aggression [43-45]. In many studies, activation of α1-adrenoceptors and stimulation of LC terminals has been found to result in anxious and agitated behaviors [25]. Accordingly, α1-adrenoceptor antagonists (e.g., prazosin) have been reported to improve disruptive behaviors including agitation in patients with AD compared to placebo (PBO) [25,46,47]. Similarly, some adrenoreceptor antagonists, such as dexmedetomidine, have been proposed as potential treatments for acute agitation [25].

It is well-established that increased activation of the DA system is linked to agitation and aggression. Cerebrospinal fluid (CSF) levels of DOPAC were found to be associated with agitation and aggression. In a subgroup of dementia patients not receiving psychotropics, a strong correlation was also observed between CSF HVA and 5-HIAA [48]. Rodent studies have shown that aggression is associated with striatal DA release and activation of DR2, while antagonism of striatal DR2 reduces aggression [13,25]. Indeed, decreased DR2 density has also been observed in AD patients with BPSD [49]. Additionally, increased functional connectivity between the hippocampus and cerebellum has been linked to agitation, irritability, and disinhibition in AD patients, with both regions innervated by DA neurons from the ventral tegmental area (VTA) [50].

Cummings et al. [25] recently proposed that the relative preservation of DA signaling, combined with serotonergic system deficits, could lead to dysregulated DA release in the striatum, contributing to increased agitation and aggression in AD patients. Accordingly, Cummings et al. [25] addressed the plausible action mechanism of BRP in the treatment of agitation in AD as following; α1-adrenoreceptor antagonism and 5-HTR1A partial agonism can reduce the detrimental effects of elevated NE and restore 5-HT signaling, thereby supporting PFC function and reducing amygdala activity.

Meanwhile, DR2 partial agonistic action can modulate striatal DA activity, helping regulate DA release. In addition, BRP also activates DR1, which can enhance N-methyl-D-aspartic acid receptor (NMDAR)-mediated currents in the PFC and potentiate AMPAR-mediated neurotransmission [51]. BRP has also been shown to reverse cognitive impairments caused by NMDAR antagonists [52]. The NMDAR hypofunction is also implicated in AD [53,54].

Given the above findings, we may expect the reduction of agitation symptoms could work through intricate and complex interactions utilizing BRP’s unique multimodal effects as 5-HTR1A partial agonist, DR2 partial agonist, antagonist of 5-HTR2 and 5-HTR7 as well as α1-adrenoreceptors antagonist, resulting in dynamic balance of multiple neurotransmitters involving in the development of agitation in AD [52,55].

CLINICAL DATA OF BRP FOR TREATMENT OF AGITATION ASSOCIATED WITH AD

Pivotal Clinical Trials

BRP was evaluated for patients with agitation associated with Alzheimer’s disease (AAD) in three 12-week randomized controlled trials (RCTs), two of which were simultaneously published in Grossberg et al. (Study 1 and Study 2) [56,57]. In these studies, patients aged 55 to 90 years, living in either an inpatient facility or a community-based setting, and diagnosed with probable AD according to the diagnostic criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA), were eligible to participate. At the time of screening for the clinical trials, the Mental State Examination (MMSE) score must be 5 to 22 points and clinically significant agitation (agitation/aggression score of 4 or more on the Neuropsychiatric Inventory Nursing Home Version (NPI-NH) assessment tool) must be present for enrollment. The primary efficacy endpoint, assessed after the screening period, was based on the change in the total score of the CMAI, while the secondary efficacy endpoint was the change in the Clinical Global Impression Severity (CGI-S) agitation score.

Among the early studies, Study 1 [56] was a fixed-dose BRP trial in which 433 patients were randomly assigned to receive BRP at doses of 2 mg/day (n = 140), 1 mg/day (n = 137), 0.5 mg/day (n = 20), or PBO (n = 136). The 0.5 mg/day dose was discontinued due to new data on its effectiveness in elderly patients. After 12 weeks, BRP 2 mg/day showed a significant decrease in the change in the CMAI total score compared to PBO (adjusted mean difference [MD], −3.77), while BRP 1 mg/day showed no significant difference compared to PBO (adjusted MD, 0.23). In the secondary efficacy evaluation, BRP 2 mg/day showed a numerical decrease on the CGI-S, though this was not statistically significant, whereas BRP 1 mg/day showed minimal change. In the safety analysis, no clinically significant differences were observed between the BRP and PBO groups in terms of suicidality, EPS-related AEs, agitation, or changes in the QTc interval. No differences were noted in body weight, metabolic parameters, or cognitive impairment. In particular, in the case of EPS, only one patient was observed for BRP 0.5−1 mg/day and 2 mg/day, showing a very low distribution. In the case of BRP 2 mg/day, tremor was also observed in only one patient, similar to PBO. Parkinson’s symptoms were not observed at 0.5−1 mg/day and were observed in only one patient at 2 mg/day. Dyskinesia did not occur at both doses. The QTc interval tended to be better compared to PBO, and weight gain was observed to be very low at −0.3 kg in the PBO group, −0.1 kg in the BRP 0.5−1 mg/day group, and 0.2 kg in the BRP 2 mg/day group.

Study 2 [56] was a dose-adjustment study in which 270 patients were randomly assigned to either BRP 0.5−2 mg/day (n = 133) or PBO (n = 137), with the dosage titrated according to individual tolerability and efficacy. At week 12, there was no statistically significant difference in the change in the CMAI total score between the BRP group and the PBO group (adjusted MD, −2.34). However, in a post hoc analysis, BRP 2 mg/day showed a significant reduction in the total CMAI score compared to PBO at week 4 (adjusted MD, −5.06), whereas no significant difference was observed at the lower doses. Additionally, the BRP group showed a statistically significant decrease in CGI-S scores (adjusted MD, −0.31). According to a post hoc analysis based on CMAI Factor-1, BRP was found to be more effective in patients who exhibited frequent physical and verbal aggression. In the safety analysis, as like in Study 1 [56], there were no clinically significant differences between the BRP and PBO groups, including suicidality, EPS related AEs, agitation, and change of QTc interval. No differences were seen in body weight, metabolic parameters, and cognitive impairment. In particular, EPSs were observed in only one patient in the BRP group, and tremor occurred less frequently than in the PBO group (2.3% vs. 3.6%). Parkinson’s symptoms were reported in just one patient. Dyskinesia was reported in 1.5% (n = 2) of the BRP group and in none of the PBO group, while akathisia was reported in two patients in the BRP group and one patient in the PBO group. The QTc interval increased by 0.1 ms in the PBO group and 0.3 ms in the BRP group, showing a very slight increase in both groups. Weight gain was −0.1 kg in the PBO group and +0.2 kg in the BRP group. Furthermore, when AEs were analyzed by subgroup (titration up to 2 mg/day vs. no titration), no significant differences were found between the two dosing groups for any safety items, demonstrating the safety and tolerability of BRP 2 mg/day.

The most recent study [57] was a fixed-dose RCT in which 345 AAD patients aged 55−90 years, who met the diagnostic criteria of the International Psychogeriatric Association (IPA), were treated with BRP 2 mg/day, BRP 3 mg/day, or PBO for 12 weeks. The characteristic of this study is that its design was advanced by the interesting results that emerged from the post-hoc analysis of a previous study, specifically the analysis of CMAI Factor-1 [56]. CMAI Factor was included early in the study design, and a more rigorous screening evaluation was conducted, so the baseline CMAI score was assessed. The total score was higher compared to previous RCTs [56]. As an additional selection criterion at the time of study enrollment, CMAI Factor 1 (12 items of aggressive behavior, including hitting, kicking, scratching, grabbing, pushing, hurting oneself or others, throwing objects, swearing or using verbal abuse, spitting, tearing up or destroying property, screaming, and biting) was used. To meet the positive criteria for CMAI Factor 1, at least one aggressive behavior had to occur several times per week, or two or more aggressive behaviors had to occur at least once per week, or three or more aggressive behaviors had to occur less than once per week [57]. The results of the study showed that BRP 2 mg/day and 3 mg/day presented significantly greater improvement in the changes of CMAI total score compared to PBO, with a Cohen’s effect size (ES) of 0.35, the highest ES observed in any AAP study in AD to date. Indeed, the ES of AAPs was −0.21 in the Cochrane meta-analysis [58], which analyzed the overall effect of various AAPs (including olanzapine, quetiapine, risperidone, and aripiprazole) on agitation, incorporating the largest recent RCTs. In the safety analysis, the proportion of patients discontinuing treatment due to AEs was 12 of 226 (5.3%) in the BRP group and 5 of 116 (4.3%) in the PBO group, without significant differences between the two groups. The frequency of AEs did not differ between BRP doses either. Frequent AEs included the following: cardiovascular events, 2 (0.9%) in the BRP group and 1 (0.9%) in the PBO group; cerebrovascular events, 0 in both groups; EPSs, 8 (3.5%) in the BRP group and 0 in the PBO group; somnolence, 9 (4.0%) in the BRP group and 1 (0.9%) in the PBO group; falls and other injuries, 5 (2.2%) in the BRP group and 4 (3.4%) in the PBO group; and metabolic events, 3 (1.3%) in the BRP group and 2 (1.7%) in the PBO group. The results of efficacy and AEs from the three pivotal RCTs are summarized in Tables 1 and 2 [56,57].

Table 1.

Summary of efficacy in controlled clinical trials

Study	Outcomes	Drug
		PBO	BRP (1 mg/d)	BRP (2 mg/d)	Difference (Ad) or ES
Grossberg et al. [56]
Study 1	CMAI total	−17.8	−17.6	−21.6a	0.23 (1 mg), −3.77 (2 mg) ES of 2 mg = −0.25
	CGI-S	−1.1	−1.0	−1.3	0.09 (1 mg), −0.16 (2 mg) ES of 2 mg = −0.17
	NPI-NH agitation/aggression	−3.68	−3.78	−4.23	−0.1 (1 mg), −0.55 (2 mg) ES of 2 mg = −0.19
Study 2		PBO	BRP (0.5−2 mg/d)
	CMAI total	−16.5	−18.9		−2.34 / ES = −0.18
	CGI-S	−1.02	−1.32a		−0.31 / ES = −0.3
	NPI-NH agitation/aggression	−3.19	−4.07a		−0.87 / ES = 0.34
Lee et al. [57]		PBO	BRP (2 or 3 mg/d)
	CMAI totala	−17.3	−22.6		−5.32 / ES = −0.35
	CGI-Sa	−0.9	−1.2		−0.27 / ES = 0.31
	NPI-NH agitation/aggressiona	−12.7	−17.3		−4.6 / ES = 0.39

PBO, placebo; BRP, brexpiprazole; CMAI, Cohen-Mansfield Agitation Inventory; CGI-S, Clinical Global Impression Severity; NPI-NH, Neuropsychiatric Inventory Nursing Home Version; Ad, adjusted; ES, effect size.

^aStatistically significant difference vs. PBO.

Table 2.

Summary of AEs in controlled clinical trials (%)

AE/Study	Grossberg et al. [56] ≥ 5%							Lee et al. [57] ≥ 2% in the BRP 2 or 3 mg or reported when higher than frequency occurred in PBO treatment
	Study 1					Study 2
	PBO	BRP 0.5−1	BRP 2	All BRP		PBO	BPR 0.5−2	PBO	BPR 2	BPR3	All BPR
Any AE	45.9	49.0	65.0	56.6		58.4	56.8	31.0	38.4	41.8	40.7
SAEa	5.2	10.2	9.3	-		4.4	5.3	2.6	0.0	3.9	2.7
Dis d/t AE	5.2	8.9	4.3	6.7		0.7	6.8	4.3	1.4	7.2	5.3
Headache	8.1	7.6	9.3	8.4		12.4	7.6	-	-	-	-
Insomnia	4.4	4.5	5.7	5.1		-	-	-	-	-	-
UTI	1.5	1.9	5.0	3.4		-	-	0.9	0.0	3.3	2.2
Dizziness	3.0	0.6	5.7	3.0		5.1	4.5	1.7	1.4	3.3	2.7
Sleepiness	-	-	-	-		3.6	6.1	0.9	4.1	3.3	3.5
EPS	0.0	0.6	0.7	-		0.0	0.8	-	-	-	-
Movement disorder	0.7	0.0	0.0	-		0.0	1.5	-	-	-	-
Tremor	0.7	1.3	0.7	-		3.6	2.3	-	-	-	-
Akathisia	0.0	0.0	0.0	-		0.7	2.3	0.0	0.0	1.3	0.9
Fall down	-	-	-	-		-	-	1.7	2.7	1.3	1.8
QTcF (ms) change	2.7	−0.4	−1.3	-		0.1	0.3	-	-	-	-
Sheehan-STS total score change	−0.01	0.01	−0.02	-		0.01	0.04	-	-	-	-
Suicidal ideation	0.7	0.0	0.0	-		0.0	0.0	0.0	0.0	0.0	0.0
Weight change (kg)	−0.3	−0.1	0.2	-		−0.1	0.2	0.0	-	-	0.3
CVAE (n)	-	-	-	-		-	-	0	0	0	0
Death (n)	0	4	1	-		1	1b	0	0	1b	1b

PBO, placebo; BRP, brexpiprazole; AE, adverse event; d/t, due to; EPS, extrapyramidal symptoms; UTI, urinary tract infection; STS, Suicidality Tracking Scale; CVAE, cerebrovascular AE; SAE, serious AE; -, no report.

^aOne SAE (agitation, BRP 0.5 mg, possible association with drug) dose unit, mg/d; ^bdropout or 28 days after dropout.

Extension Trial

A 12-week, active-treatment (BRP 2 or 3 mg/day) extension trial was conducted in AD patients with agitation in a care facility/community-based setting who completed the RCT (n = 259 enrolled/analyzed for safety; 88.4% completed) [59]. The primary safety endpoint was the frequency and severity of AEs. Change in CMAI total score was an exploratory efficacy endpoint. Included patients were all from U.S. and Europe. The mean age was 74.3 years, 56.0% of the patients were female (n = 145), and 248 (95.8%) were Caucasian. AEs were reported in 25.9% of the patients (n = 67), and the most common AEs (incidence ≥ 2%) were headache (3.5%) and fall (2.3%), with mild or moderate in severity. No specific treatment emergent adverse event had an incidence ≥ 5% in the total sample. The mean change in body weight over 12 weeks was +0.1 kg, where the weight change was more prominent in patients who switched from PBO to BPR (+0.8 kg) than those who continued BRP (−0.3 kg). At week 24, the proportion of patients who reached weight gain ≥ 7% relative to week 12 was 3.1% (prior PBO, 4.7%; prior BRP, 2.1%). The serious AEs were developed in 5 patients (1.9%). Treatment discontinuation associated with AEs occurred in 12 patients (4.6%). There was no incidence of death during the study. The reduction of CMAI total score from week 12 to 24 was −9.1, where it was greater in patients who switched from PBO to BRP (−12.5), while it was −7.1 in patients who continued BRP. The reduction of CGI-S score from week 12 to 24 was −0.6, where it was greater in patients who switched from PBO to BRP (−0.8), while it was −0.5 in patients who continued BRP.

Asian Study (Short-term and Extension Trial)

In addition, there has been a recent phase II/III multicenter study in Japan [55] with the use of BRP 1 mg/day, BRP 2 mg/day, or PBO with a similar design to studies conducted globally [56,57]. BRP 1 mg/day was used to find differential effects between the two low doses in Asian population.

The primary efficacy was the change in CMAI total score after 10 weeks [55]. According to the results, both BRP 1 mg/day and BRP 2 mg/day were superior over PBO in terms of change in CMAI total score after 10 weeks of treatment (2 mg: MD −7.2, pvalue < 0.0001, 1 mg: −3.7, pvalue = 0.0175), however, BRP 1 mg/day showed later improvement compared to that of BRP 2 mg/day (1 mg/day: week 8, pvalue < 0.05 and 2 mg/day: week 4, pvalue < 0.01). By timepoint, both BRP groups showed greater decreases of mean changes in CGI-S scores from week 2 compared to that of PBO. The efficacy results from subgroup analysis for CMAI score were also similar to those from primary endpoint analysis. In all subscale scores, physically nonaggressive behavior, and verbally agitated behavior, improvements were observed in the BRP 1 mg/day and 2 mg vs. PBO (pvalue < 0.05 each), except the BRP 1 mg/day in physically nonaggressive behavior score (only numerical superiority). The treatment difference vs. PBO was greater in BRP 2 mg/day than in 1 mg/day. Therefore, the efficacy of BRP 1 mg/day was shown for the first time differing from previous global studies.

The most common AEs (incidence ≥ 5%) from treatments with BRP groups were insomnia (14.3%), somnolence (8.0%), muscle rigidity (8.0%), pyrexia (8.0%) and contusion (8.0%) for BRP 1 mg/day and somnolence (16.1%), bradykinesia (13.4%), insomnia (12.1%), hypersalivation (12.1%) and muscle rigidity (10.7%) for BRP 2 mg/day. BRP 1 mg/day and 2 mg/day were generally well tolerated with mild to moderate intensity and no new alerting safety data are not reported compared to previous global studies [56,57].

Recently a multicenter, open-label study [60] assessed the long-term safety and tolerability of BRP in Japanese patients who completed a preceding 10-week double-blind, PBO-controlled trial for 14-week extension period, with BRP doses of 1 or 2 mg/day. The primary endpoint was the incidence of AEs. Of 183 patients screened, 164 were treated. The overall incidence of AEs was high (90.2%) but most were mild or moderate in severity. Serious AEs occurred in 7.3% of patients, and the rate of discontinuation due to AEs was 20.1%. Common AEs included somnolence (12.2%), insomnia (10.4%), fall (9.8%), and sedation-related complication (9.8%). Other major AEs were as following: EPS-related AEs (30.5%), accident and injury-related AEs (25.6%), oversedation-related AEs (23.8%), cardiovascular-related AEs (4.3%), and CVA-related AEs (0.6%). The highest incidence of AEs was observed from 2 weeks to 3 weeks. One patient (0.6%, 1/164) treated with PBO in the parent study was dead due to metastatic pancreatic carcinoma, being assessed not related to current BRP treatment. There were no clinically significant findings requiring treatment intervention related to prolactin, lipid and glucose levels, or QT prolongation, including shifts to abnormal values. Efficacy was measured primarily by changes in the CMAI total score during the study and the CMAI score modestly decreased by 4 points (2.5 in prior BRP group and 6.4 in prior PBO) from baseline (49.9) to week 14 (45.9). Other efficacy measures also decreased from baseline to week 14 during the study; CGI-s (−0.2 in prior BRP and −0.7 in prior PBO).

Post-hoc Study of Pivotal Trials

The two 12-week BRP pivotal trials (n = 610) were combined for fixed-dose of BRP 2 or 3 mg/day and for PBO [61], to define any differential effects of BRP on different clinical manifestations in AD patients. Based on the results, the effect of BRP on symptoms of agitation was consistent across different patient subgroups, regardless of severity of dementia, presence or absence of co-occurring neuropsychiatric symptoms, use of concomitant treatments for dementia or psychiatric conditions, and care setting. Another post-hoc study [62] included the role of BRP on the same individual agitation behaviors, but specifically focusing on those patients who were frequently experiencing the behaviors at baseline (corresponding behavior occurred at least once per week [item score ≥ 3] at baseline). Fixed‐dose BRP 2 or 3 mg/day was associated with a numerically greater reduction in the frequency of most of these frequently occurring agitation behaviors vs. PBO.

Data for BRP 2 or 3 mg/day were obtained from two 12-week RCTs and a 12‐week, active‐treatment extension trial of patients with agitation associated with AD [63]. According to the results with two 12-week RCTs, the difference in proportion of patients who achieved clinically meaningful response (CMR, a 20‐point score reduction from baseline) was 17.0%, favoring BRP over PBO (hazard ratio, HR = 0.64, p = 0.006). The difference in corresponding proportion of patients achieving sustained CMR (SCMR, a 20‐point score reduction that was maintained to trial end) was 16.5%, favoring BRP over PBO (HR = 0.58, p = 0.004). Over 24 weeks, the difference in proportion of patients achieving CMR were 9.2% favoring BRP over PBO (HR = 0.63, p = 0.011), while the difference in corresponding proportion of patients achieving SCMR was 7.1%, favoring BRP over PBO (HR = 0.59, p = 0.010), indicating the substantial utility of BRP on the sustainability of efficacy in the treatment of agitation of AD patients.

Meta-analysis

The efficacy of BRP vs. PBO has also been shown in the treatment of agitation in AD patients in a recent meta-analysis [64], involving three 12-week RCTs [56,57] comprising 1,048 patients. Based on the results, any doses (MD −3.05) and 2 mg/day (MD −4.36) of BRP showed a superiority over PBO in the changes of CMAI total score from baseline during the study, which was replicated in the secondary endpoint as well (e.g., CGI-S score change, MD −0.20). There is no significant difference between BRP and PBO treatments in the incidence of at least one AEs (risk ratio [RR] 1.14) and all-cause mortality (RR 1.99) [64].

DISCUSSION

Based on a recent RCT from Japan [55], BRP can be titrated starting at 0.5 mg/day, with increases to 1 mg/day and 2 mg/day, showing efficacy without prominent safety concerns. Interestingly, the treatment effect with 2 mg/day was greater and occurred earlier than with 1 mg/day, a finding not demonstrated in previous global studies [56,57]. However, it’s important to note that Nakamura et al. [55]’s study was confined to a Japanese population, so these results may be more directly applicable to patients in adjacent East Asian countries, such as Korea and China, due to genetic similarities, including similarities in pharmacokinetics and CYP450 enzyme gene polymorphisms that could affect manifestations of AEs [65,66]. However, 1 mg/day was not effective in precedent study [56], while doses of 2−3 mg/day were effective compared to PBO in the treatment of agitation in previous global RCTs, with no treatment differences between 2 mg/day and 3 mg/day [56,57]. This discrepancy in dose effects warrants further investigation across different ethnic groups in future studies. Nonetheless, it is important to gradually escalate the dose and use the lowest effective dose of BRP in clinical practice, which is in line with treatment guidelines and also considering the increased susceptibility of older populations to AEs from AP treatments.

In addition, based on a recent large meta-analysis [67] including BRP, quetiapine, olanzapine, risperidone, and aripiprazole, BRP had a largest ES (SMD = −1.77 vs. PBO) in comparison with quetiapine (−0.44), risperidone (−0.62), olanzapine (−0.33) and aripiprazole (−0.16). Likewise, BRP showed overall 20−40% greater reduction in agitation symptoms (based on changes of CMAI scores) compared to PBO in the three pivotal trials, a range generally considered clinically meaningful in psychotropic trials [56,57,68,69]. Indeed, a minimum clinically important difference (MCID) defined as a 20% reduction serves as a practical benchmark for patients with moderately severe symptoms. Furthermore, both the Western and Asian extension studies have shown sustained and stable reductions in agitation among patients with AD, suggesting that BRP may be reasonably and cautiously employed on an as-needed basis for long-term use in clinical practice without concerns about diminished efficacy.

Depressive symptoms are also highly prevalent among the BPSD observed in clinical practice and represent a crucial clinical factor in the treatment of AD patients, alongside agitation symptoms. This is particularly significant, as numerous studies have consistently reported a strong correlation between depressive symptoms and behaviors such as agitation, aggression and impulsivity. The interaction between these symptoms emphasizes the importance of addressing depression in the comprehensive management of AD [70,71]. In addition, an increase of depressive symptoms is strongly correlated with the emergence of psychotic symptoms [72]. There is also a notable link between depressive symptoms and abusive behavior toward patients by care staff as well as concerns commonly observed in long-term nursing homes [72]. Therefore, managing depressive symptoms is also essential in the treatment of AD patients experiencing various behavioral symptoms [71]. A recent longitudinal study using Minimum Data Set (MDS) information from 2,032 residents of Dutch nursing homes found that depressive symptoms were present in approximately 51% of patients, while psychotic symptoms were seen in about 15%. Interestingly, two-thirds of patients with psychotic symptoms also had depressive symptoms [71]. The study revealed a strong correlation between depressive and psychotic symptoms, with pain also being linked to agitation. However, improvements in pain symptoms did not correlate with reductions in agitation. Improvement of depressive symptoms were also strongly associated with significant reductions in both agitation and psychotic symptoms [71]. This suggests that when APs with antidepressant properties may help indirectly alleviate agitation and psychotic symptoms, in addition to managing depression. Given BRP’s established antidepressant effects and its U.S. FDA approval for the treatment of MDD as an augmentation therapy, it should have substantial benefits in routine clinical practice in treating these complex symptoms in AD patients.

Dementia patients often suffer from co-morbid diseases such as diabetes, hypertension, and hyperlipidemia, which are also high-risk factors for developing dementia. These comorbidities can worsen cerebrovascular disease and accelerate the progression of not only vascular dementia but also AD [73-76]. Therefore, it is optimal to avoid interactions between concurrently administered medications or use AAPs with fewer metabolic effects alone. The frequency and susceptibility to metabolic AEs vary significantly among individual AAPs [77]. According to a network meta-analysis that examined recent large-scale clinical data (100 comparative clinical trials, n = 25,952 patients) [78], BRP exhibited relatively favorable profiles in metabolic AEs. This trend has been consistently reported in many studies, despite differences in research methodologies [77,79,80].

The pathophysiological mechanisms of EPS are not yet fully understood, but considered that it mainly occurs due to DA blockades in specific brain regions. However, researches consistently found that other neurotransmitters besides DA are also involved in the development of EPS, including 5-HT [12,81-84]. Given that BRP acts on various 5-HT receptors through intricate interactions, it is likely to be more advantageous in reducing the occurrence of EPS compared to existing APs. Not surprisingly, clinical trials of BRP for the treatment of agitation associated with AD have reported a lower incidence of EPS compared to other APs [56,57].

BRP has been found to pose a very low susceptibility to hypersensitization of DR2 and 5-HTR2A, even when administered alone for an extended period or when substituted for other AAPs, such as risperidone [85]. Therefore, BRP is considered to have a lower risk of dyskinesia and rebound psychosis associated with D2 receptor supersensitivity.

AD Patients usually receive cognitive enhancers (e.g., donepezil, galantamine), which increase ACh and raise the risk of EPS by modulation of DA, especially when combined with APs or antidepressants like selective serotonin reuptake inhibitors (SSRIs) [12,86]. It is well-known that SSRIs can also develop EPS via stimulation of 5-HTR2, while 5-HTR1A partial agonists and 5-HTR2A/C antagonists have been shown to reduce EPS [12,81,84,87,88]. Hence, the selection of AAPs with unique action mechanisms, such as BRP, which modulates various 5-HT receptors, can be advantageous and beneficial for the treatment of AD patients on multiple medications.

A recent large meta-analysis [89] investigating the association between APs and falls/fractures in older population, found that AAPs were associated with the highest risk of falls compared with tAPs, anxiolytics, sedative-hypnotics, and antidepressants, whereas tAPs conferred the greatest risk of fracture. Additionally, meta-regression analysis revealed a statistically significant difference in fracture risk between patients aged > 65 years and those < 65 years, showing a positive association between age and fracture incidence.

In a recent meta-analysis including quetiapine, olanzapine, risperidone, BRP and aripiprazole [67], only aripiprazole (odds ratio [OR] = 0.72) was superior over PBO, there were no differences between BRP, risperidone, olanzapine, and quetiapine, while aripiprazole was superior over BRP (OR = 0.61) in terms of acceptability [67]. As for tolerability, olanzapine was worse than PBO (OR = 6.02), risperidone (OR = 3.67) and quetiapine (OR = 3.71), while aripiprazole was better than olanzapine (OR = 0.25) [67]. There were no differences between BRP, olanzapine, risperidone and quetiapine in tolerability. According to the surface under the cumulative ranking curve (SUCRA) analysis, BRP was most acceptable for AEs such as fall and sedation which should be extremely dangerous in older patients [67]. Indeed the incidence of falls was less on BRP (1.7%) than on PBO (2.6%) in three BRP trials [64]. Likewise, it was 6% in PBO vs. 8.8% in BRP in Asian study [55].

Older patients already have been with polypharmacy including various psychotropics. Based on a recent study exploring the prevalence of polypharmacy (concomitant use of ≥ 5 medications), hyperpolypharmacy (≥ 10 medications), and potentially inappropriate medication (PIM) use among older adults, hyperpolypharmacy, and PIM use were prevalent in 67.1%, 18.8%, and 23.9% of 3,912 participants [90]. Similar finding have been consistently replicated in consecutive studies, the pooled estimate of PIM and polypharmacy was 43% and 62% in a recent large meta-analysis [91]. It is well-known the risk of polypharmacy in older population, including drug-drug interaction (DDI), AEs, worsening of cognitive functions [90,92,93]. For instance, administration of two or more medications simultaneously had at least one potential DDI in 85.3% of patients in a recent cross-sectional study [92]. Hence, BRP should be also cautiously used in patients with polypharmacy or with medications affecting CYP3A4 and/or CYP2D6.

Finally, the FDA prescribing information for BRP includes a boxed warning, consistent with all APs, highlighting the increased risk of mortality in patients with dementia-related psychosis. It further specifies that BRP is not indicated for the treatment of dementia-related psychosis in the absence of agitation [94]. In pivotal BRP trials, 7 (0.9%) patients died in the BRP groups and 1 patient (0.3%) died in the PBO groups, with all causes of death, not confirmed to be related with BRP. Numerically, it was high mortality risk in comparison with that found in previous meta-analysis (OR, 2.22) [95]. However, two Western and Asian longer-term open-label safety extension studies showed no new safety signals or mortality [59].

BRP is the first medication approved by the U.S. FDA for the treatment of agitation associated with dementia. However, patients with dementia typically have multiple comorbidities and are exposed to numerous concomitant medications, making them particularly vulnerable to adverse effects. These factors require careful consideration for the use of BRP in clinical practice. In addition, there are several clinical uncertainties regarding the use of BRP in real-world practice, and comparative evaluations with alternative treatment options have not been conducted [96]. Following issues also remain to be clarified, including clinical information on the use of very low and high doses not studied in clinical trials; the precise duration after which discontinuation can be considered; differential effects on comorbid psychotic symptoms; strategies for tapering or discontinuation; management of potential worsening during withdrawal; data on adverse effects associated with long-term use, which may be necessary for certain patients; clinical variables that identify patients most suitable for treatment; and whether the statistically significant differences observed versus placebo in clinical trials translate into meaningful clinical benefit.

CONCLUSION

There is a shortage of pharmacological options for managing BPSD in AD, with most APs used off-label despite limited evidence for efficacy and tolerability. BRP, with its unique neurobiological mechanisms, is the first FDA-approved treatment for agitation in AD, providing clinicians with an approved pharmacological option.

However, important clinical uncertainties remain. These include its optimal dosing strategies, the appropriate duration of treatment (currently supported for up to six months), unidentified AEs associated with long-term use, the lack of direct comparisons with non-pharmacological treatment options, drug interactions, polypharmacy issues, and limited guidance on proper utilization.

Post-marketing surveillance and real-world data collection will be essential to address these unresolved issues. While BRP’s approval underscores the need to prioritize its use over off-label alternatives, clinicians should apply it carefully, considering its benefit– RR in routine practice.