Evaluation of the efficacy of pimavanserin in the treatment of agitation and aggression in patients with Alzheimer's disease psychosis: A post hoc analysis

Patients with Alzheimer's disease psychosis (ADP) commonly experience concomitant agitation and aggression. We investigated whether a reduction in ADP following pimavanserin treatment conferred a reduction in associated agitation and aggression.


| INTRODUCTION
More than 45 million people worldwide are living with dementia, the majority of whom have Alzheimer's disease (AD). 1 While AD is commonly thought of as a memory disorder, behavioral and psychological symptoms of dementia are nearly universal and include psychotic symptoms. [1][2][3] More than half of patients with AD will experience psychosis during their illness. [4][5][6] Hallucinations and delusions are frequently distressing to the individual and their caregivers and are associated with worse disease outcomes relative to AD patients without psychosis, 4 including accelerated cognitive decline, 7,8 more rapid progression of functional impairment, 9 increased hospital admissions, 10 earlier admission to institutional care, 8,9 and increased mortality. 9,11 In addition, psychotic symptoms are often antecedent to or comorbid with other neuropsychiatric symptoms like agitation, aggression, and depression, further adding to the impact on the individual and others making Alzheimer's disease psychosis (ADP) more difficult to treat. 2,5,12 Currently, no treatments have been approved for ADP. 13 The modest benefits provided by anxiolytics or typical/atypical antipsychotics that clinicians prescribe to mitigate hallucinations and delusions in ADP are associated with considerable adverse effects, such as accelerated cognitive decline and increased short-term mortality in elderly populations. [14][15][16] Disruptive behaviors and significant impairments in interpersonal relationships and social functioning make agitation and aggression one of the most difficult symptoms to treat. 17,18 However, the underlying causes and mechanisms of agitation and aggression are multifaceted and not well understood. 13,17 Agitation may be due to direct underlying biological or environmental factors (psychosocial stress or intercurrent medication) or may be a consequence of psychosis, major depressive disorder, or other mood or psychiatric disturbances. 16,17 It is unclear whether the symptoms of agitation and aggression associated with ADP are related to psychosis or are independent behavioral symptoms that need to be treated separately. To optimize the treatment of ADP, a greater understanding of how the treatment of psychosis affects agitation and aggression symptoms is needed.
Pimavanserin is currently approved by the United States Food and Drug Administration (FDA) for the treatment of hallucinations and delusions associated with Parkinson's disease psychosis (PDP). 19,20 The positive effect of pimavanserin in improving psychotic symptoms in ADP was also seen in a randomized, double-blind, placebo-controlled trial. 15,21 After 6 weeks of pimavanserin treatment, 51% of patients with ADP, treated with pimavanserin, showed ≥50% improvement compared with baseline in symptoms of psychosis (hallucinations and delusions) on the Neuropsychiatric Inventory-Nursing Home Version-Psychosis Score (NPI-NH-PS). 15 While the therapeutic benefit of pimavanserin in improving hallucinations and delusions is promising in patients with ADP, treating agitation and aggression symptoms in this group is also an important component of treatment. The goal of this post hoc analysis was to investigate whether decreases in the severity of a patient's hallucinations and delusions while receiving pimavanserin were associated with a reduction in the severity and frequency of the patient's agitation and aggression.

| Study design
Data were collected as a part of the study, ACP-103-019 (NCT02035553), a 12-week, randomized, double-blind, placebocontrolled phase 2 trial conducted in nursing home patients with ADP. Study design and primary results of the study as well as outcomes in patients with more severe psychosis symptoms have been previously published. 15,21 The primary objective of the study was to assess the efficacy of pimavanserin in attenuating ADP-related symptoms, as assessed by NPI-NH-PS, after 6 weeks of treatment. were then randomly assigned to receive either pimavanserin 34 mg or placebo administered orally once daily, with stratification by baseline Mini-Mental State Examination (MMSE) total score and NPI-NH-PS. 23 At baseline and after 2, 4, 6, 9, and 12 weeks of pimavanserin treatment or placebo, the NPI-NH 24

Key points
• Patients with ADP commonly experience concomitant agitation and aggression.
• Patients with ADP who respond to the therapeutic benefit of pimavanserin on psychosis symptoms also experience improvement in agitation and aggression compared with nonresponders.
• Patients with ADP and severe agitation and aggression at baseline who respond to the therapeutic benefit of pimavanserin on psychosis symptoms also show greater improvements in agitation and aggression than nonresponders.

| Post hoc analysis
The effect of pimavanserin on agitation and aggression symptoms was compared in patients responsive and nonresponsive to pimavanserin's treatment of psychotic symptoms. Psychosis responders were defined as patients who exhibited ≥50% improvement from baseline in NPI-NH-PS at week six; patients who received pimavanserin but did not meet these improvement criteria were considered nonresponders.
An additional analysis of psychosis responders vs nonresponders was conducted in the subgroup of pimavanserin-treated patients exhibiting more severe agitation and aggression symptoms at baseline, as demonstrated by an NPI-NH agitation/aggression subscale score higher than the midpoint (≥6).

| Statistical analysis
Baseline characteristics of week six psychosis responders and nonresponders were compared using a t-test for continuous parameters or chi-squared test for categorical parameters. the CMAI-SF scores at week six was also analyzed using a Spearman's correlation test. All analyses were conducted at the 5% level of significance using two-sided tests.

| Patients
In the study, 181 patients from 133 care homes across the United Kingdom (UK) were randomized to receive pimavanserin (n = 90) or placebo (n = 91). Efficacy analyses included all patients who received at least one dose of pimavanserin and completed the NPI-NH at baseline and at least one postbaseline study visit (pimavanserin, n = 87). At baseline, patients had an overall mean (standard deviation [SD]) score on the NPI-NH domain C of 4.7 (3.9) and an overall mean (SD) score on the CMAI-SF of 28.6 (8.8) ( Table 1). More severe baseline agitation/ aggression symptoms, defined by a score of ≥6 on the NPI-NH domain C subscale, were observed in 43 of 87 (49.4%) pimavanserin-treated patients and 42 of 91 (46.2%) placebo-treated patients.

| Overall changes in agitation and aggression symptoms
As previously reported, in the overall study population, pimavanserin treatment did not improve agitation and aggression symptoms (rela-

| Changes in symptoms of agitation and aggression in patients with more severe baseline symptoms
Among pimavanserin-treated patients who had a week 6 NPI-NH domain C response (n = 39), severe agitation and aggression (NPI-NH domain C score ≥ 6) were observed at baseline in 26 psychosis responders and 13 nonresponders (Table 1)

| DISCUSSION
This post hoc responder analysis suggests improvements in psychotic symptoms in patients receiving pimavanserin treatment were associated with an improvement in agitation and aggression symptoms.
Patients showing ≥50% improvement from baseline in psychotic symptoms after 6 weeks of pimavanserin treatment exhibited a greater improvement in agitation and aggression symptoms than nonresponders. This effect was observed on two distinct measures, the NPI-NH domain C score and the CMAI-SF score.
In patients with ADP, experiencing severe agitation and aggression symptoms at baseline (baseline NPI-NH domain C score ≥ 6), the therapeutic benefit of pimavanserin on psychosis was associated with an improvement in agitation and aggression symptoms on the NPI-NH domain C score. Psychosis responders also exhibited an improvement from baseline in CMAI-SF score at week six, although the difference between responders and nonresponders was not statistically significant. This may be explained by the relatively low patient numbers in the severe subgroups and by a slight improvement in CMAI-SF score for nonresponders between weeks four and six.
Improvement of psychosis was correlated with the improvement of agitation and aggression at the study's primary endpoint, in both the overall study population and those with severe agitation and aggression at baseline, supporting the relationship between these two types of symptoms in patients with AD. This correlation is consistent with the observation that patients with severe psychosis symptoms at baseline (NPI-NH psychosis score ≥ 12) in this study had higher agitation and aggression scores than the overall population 21 and with several studies that have reported significant correlations between psychosis and agitation and aggression in patients with dementia. 27 Collectively, results described here suggest that the improvement in hallucinations and delusions after pimavanserin treatment are related to improvements observed in agitation and aggression.
Mechanisms driving agitation and aggression in AD remain under investigation. However, evidence for unique patterns of underlying brain circuit dysregulation suggests that distinct types of agitation and aggression might require unique treatment approaches. 17,28 Consistent with this, AD patients in one study, exhibiting moderate agitation symptoms, responded dramatically to citalopram treatment for the management of agitation/aggression symptoms; however, in the same study, those exhibiting severe agitation symptoms showed no therapeutic benefit from the same treatment. 16 Our results suggest one potential underlying mechanism of agitation and aggression in AD is related to psychosis and that treating psychosis may provide a more precise approach to treating agitation and aggression in these individuals.
The current analysis was not prespecified and was conducted post hoc. The analysis is further limited by the small sample sizes in subgroups of pimavanserin-treated patients. The comparison of nonrandomized groups (psychosis responders and nonresponders) may be a source of bias unaccounted for by the statistical modeling. In addition, the study was limited to AD patients in nursing homes in the United Kingdom. Patients, in this study, were required to have psychosis symptoms at baseline but were not recruited for the presence of agitation and aggression, which may have limited the number of patients with a potential for improvement in agitation and aggression.