BACKGROUND: We evaluated if plasma β-amyloid (Aβ) levels were associated with mortality risks in a subsample of the French Three-City (3C) prospective cohort study. METHODS: Analyses were based on 1254 participants randomly selected from the initial 3C cohort stratified by center, sex, and age in the context of a nested case-cohort study to investigate biological variables. Associations between plasma Aβ and mortality were assessed with the Cox regression model with delayed entry including various potential confounding factors and testing possible mediators. RESULTS: A relationship between high plasma Aβ1-40 concentrations and risk of mortality (hazards ratio, 1.15; 95% confidence interval, 1.01-1.31, P = .03) was unveiled independently of age, educational level, vascular risk factors, diet, physical activity, cognitive impairment, or frailty status. It was only modified when we included cystatin C levels. CONCLUSIONS: Further investigations are needed to determine precisely the pathophysiological roles of plasma Aβ1-40 and cystatin C and before envisioning any future clinical applications.