Psychosis in Alzheimer Disease and Elevations in Disease-Relevant Biomarkers

Importance: The emergence of psychotic symptoms in Alzheimer disease (AD) is associated with accelerated cognitive and functional decline that may be related to disease pathology. Objective: To investigate the longitudinal dynamics of plasma tau phosphorylated at threonine 181 (p-tau181) and neurofilament light chain protein (NfL) levels in association with the emergence of psychotic symptoms (delusions and hallucinations) in the context of AD. Design, Setting, and Participants: This cohort study used longitudinal data from the Alzheimer Disease Neuroimaging Initiative (ADNI). Baseline analyses compared patients with mild cognitive impairment (MCI) and AD (both with psychosis [AD+P] and without psychosis [AD-P]) and participants who were cognitively unimpaired (CU). For the longitudinal analysis, participants with MCI and AD were subdivided into patients with evidence of psychosis at baseline (AD+P baseline) and patients free of psychosis at baseline who showed incidence of psychosis over the course of the study (AD+P incident). Study data were analyzed between June and November 2023. Exposures: Plasma p-tau181 and NfL measures in individuals with MCI and AD, both with and without psychosis. Main Outcomes and Measures: Plasma p-tau181 and NfL quantifications up to 48 months and concurrent assessments of presence or absence of delusions and hallucinations via the Neuropsychiatric Inventory (NPI) questionnaire. Results: The cohort included 752 participants with AD (mean [SD] age, 74.2 [7.7] years; 434 male [57.7%]). A total of 424 CU participants had a mean (SD) age of 75.4 (6.6) years of whom 222 were female (52.4%). In the longitudinal analysis of p-tau181 trajectories of the AD+P group, the group of patients who showed incidence of psychosis over the course of follow-up (AD+P incident) demonstrated an associated increase in plasma p-tau181 levels compared with the group of patients who had psychosis at baseline (AD+P baseline) and showed an associated decrease in plasma p-tau181 levels (F4, 117 = 3.24; P = .01). The mean slope of p-tau181 change was significantly different in AD+P incident and AD+P baseline groups (F5,746 = 86.76, P < .0001) and when only individuals with amyloid-β positivity (Aβ+), which was determined using positron emission tomography, were compared (F5,455 = 84.60, P < .001). Patients who experienced psychosis at any time had increased levels of NfL relative to those who never experienced psychosis. Conclusions and Relevance: Results of this cohort study suggest that the emergence of psychosis in AD was associated with elevations in plasma levels of p-tau181, highlighting the potential utility of plasma p-tau181 as a biomarker of neuropsychiatric illness in AD, which could have implications for predictive and treatment response strategies.