Blood Biomarkers of Alzheimer Disease and Rates of Global and Domain-Specific Cognitive Decline.

Importance: While evidence is accumulating on blood-based biomarkers of Alzheimer disease (AD) and future cognitive decline, community-based studies that include phosphorylated tau (p-tau) 217 are still scarce. Moreover, there is limited knowledge regarding associations with domain-specific cognitive decline and modifying effects of key individual characteristics. Objective: To investigate associations of blood biomarkers of AD with global and domain-specific cognitive decline. Design, Setting, and Participants: This prospective cohort study included participants from the Swedish National Study on Aging and Care-Kungsholmen recruited between March 21, 2001, and August 30, 2004. Participants were aged 60 years or older, provided blood samples, and were free of dementia at baseline. Follow-up assessments were completed on December 19, 2019. The statistical analysis was conducted between April 1 and September 30, 2025. Main Outcomes and Measures: Baseline biomarkers, including amyloid β 42/40 ratio, p-tau217, p-tau181, total tau, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP), and cognitive performance (episodic memory, semantic memory, verbal fluency, perceptual speed) were assessed at baseline and across 15 years of follow-up. Linear mixed models were used to assess associations between biomarker quartiles and rates of change in specific cognitive domains and a global composite score. Interactions were explored for age, sex, and APOE ε4 status. Results: In the 2008 participants (mean [SD] age at baseline, 71.7 [10.1] years; 1215 female [60.5%]) included, associations with rate of change were observed across all cognitive domains for amyloid β 42/40 ratio (P for trend = .006), p-tau217, p-tau181, NfL, and GFAP (all P for trend < .001), but not total tau. Significant interactions were observed for APOE, for which higher levels of p-tau181 were associated with faster decline in ε4 carriers vs noncarriers (β = -0.037 [95% CI, -0.047 to -0.026] vs -0.020 [95% CI, -0.025 to -0.015]). In contrast, higher levels of NfL and GFAP were associated with faster cognitive decline only in non-ε4 carriers (β = -0.025 [95% CI, -0.031 to -0.020]) and -0.017 [95% CI, -0.022 to -0.012], respectively). Associations between NfL and GFAP and cognitive decline were exacerbated in females (β = -0.023 [95% CI, -0.029 to -0.018] and -0.021 [95% CI, -0.028 to -0.015], respectively). Conclusions and Relevance: This cohort study in a general older population found that AD biomarkers were associated with long-term rates of cognitive decline. The observed patterns varied across biomarkers and individual characteristics and suggest that sex and genetic risk of AD should be considered in the interpretation of these biomarkers.