Abstract
BACKGROUND: Psychosis, including delusions and hallucinations, is a significant neuropsychiatric symptom in Alzheimer's disease (AD) associated with poor prognosis. The relationship between psychosis and AD pathology remains controversial. This study investigates the role of AD pathology in mediating the association between psychosis and cognitive impairment. METHODS: Data were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI). We included individuals with a Clinical Dementia Rating (CDR) score of 0.5 or higher. Among a total of 833 individuals, 96 individuals with psychosis were matched to 192 individuals without psychosis using propensity scores based on age, sex, education level, and follow-up duration. Baseline cognitive performance was assessed using composite memory scores (ADNI-MEM) and executive function scores (ADNI-EF). AD pathology was measured using baseline cerebralspinal fluid (CSF) levels of β-amyloid(1-42) (Aβ(1-42)), hyperphosphorylated-tau(181) (p-tau(181)), and total tau. Logistic regression was performed to evaluate the association of psychosis with baseline cognitive performance and CSF biomarkers. Mediation analysis was conducted to assess whether AD biomarkers mediate the relationship between cognitive impairment and psychosis. RESULTS: Psychosis was significantly associated with worse ADNI MEM score (β = -0.622, p = 0.013) and worse ADNI EF score (β = -0.516, p = 0.003), and lower CSF Aβ(1-42) levels (β = -0.009, p = 0.007). No significant associations were found with p-tau(181) or total tau levels. Mediation analysis revealed that low CSF Aβ(1-42) levels mediated the relationship between cognitive impairment and psychosis. CONCLUSION: These findings suggest that amyloid pathology may mediate the effect of baseline cognitive impairment on psychosis during disease in AD, highlighting a potential pathological link between cognitive decline and psychotic symptoms.