Alzheimer disease (AD), the most common cause of cognitive impairment, is characterized by β-amyloid (Aβ) plaques and tau neurofibrillary tangles, which lead to neuroinflammation, neurodegeneration, and cognitive decline.

Among various interventions, physical activity (PA) is widely recommended. Numerous studies suggest that PA not only slows cognitive decline but also improves specific cognitive functions. Potential mechanisms underlying the protective associations of PA with cognition include its influence on neuroplasticity, neuroinflammation, and cerebral blood flow. Additionally, several animal studies demonstrated that PA can lower brain Aβ and tau, although these findings were less compelling in human studies. These inconsistencies highlight the need for larger, more comprehensive studies to better understand the impact of PA on AD biomarkers.

Furthermore, although PA could potentially affect various pathophysiological mechanisms, few studies have investigated the entire spectrum of biomarkers in relation to PA. Thus, analyzing the relationships among PA, biomarkers for multiple pathomechanisms, and cognition could provide insights into how PA impacts cognitive health.

Recent advancements in plasma biomarkers have advanced diagnostic accuracy and the ability to estimate prognosis in neurodegenerative diseases. The Aβ42:40 ratio and phosphorylated-tau (p-tau) species are regarded as core biomarkers of AD pathology. Additionally, glial fibrillary acidic protein (GFAP) and neurofilament light chain (NfL) have been studied as promising plasma biomarkers reflecting neuroinflammation and neuronal injury, respectively. Research has shown that these plasma biomarkers are also associated with cognitive decline. Thus, plasma biomarkers could be potential therapeutic targets, and identifying modifiable factors, such as PA, that influence these biomarkers is crucial for developing management strategies to mitigate cognitive decline.

Objective was to investigate the associations among PA; plasma biomarkers, including β-amyloid 42/40 (Aβ42/40), phosphorylated-tau217 (ptau217), glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL); and cognition.

This cross-sectional study included participants with and without cognitive impairment recruited from multiple memory clinics in South Korea between May 2019 and May 2022. Data were analyzed from June to December 2024.

PA was assessed as metabolic equivalent task minutes per week using the International Physical Activity Questionnaire and categorized into quartiles from the lowest (Q1) to the highest (Q4).

Plasma Aβ42/40, ptau217, GFAP, and NfL were measured. Cognition was assessed using the Mini-Mental State Examination (MMSE) and Clinical Dementia Rating-Sum of Boxes (CDR-SB).

Among 1144 participants (mean [SD] age 70.9 [8.7] years; 744 [65.0%] female), the highest PA quartile showed significantly lower ptau217 (estimate [SE], −0.14 [0.06]; P = .01) and NfL (estimate [SE], −0.12 [0.05]; P = .01) compared with the lowest quartile. Higher PA quartiles were associated with higher MMSE scores (estimate [SE]: Q2, 0.93 [0.31]; P = .003; Q3, 0.82 [0.32]; P = .009; Q4, 0.94 [0.32]; P = .004) and lower CDR-SB scores (estimate [SE]: Q2, –0.33 [0.16]; P = .04; Q3, –0.37 [0.16]; P = .02; Q4, –0.55 [0.16]; P = .001) after adjusting for age, sex, education years, and β-amyloid uptake. In subgroup analyses according to age and cognitive status, the associations of PA and plasma biomarkers with cognition were more pronounced in the older (age ≥65 years) and cognitively impaired groups compared with the younger and cognitively unimpaired groups.

In this cross-sectional study of 1144 participants, including individuals with and without cognitive impairment, higher levels of PA were significantly associated with lower levels of plasma neurofilament light chain and phosphorylated tau-217 and better cognition. These associations were more pronounced in the cognitively impaired group and the group aged 65 years and older compared with the cognitively unimpaired and younger groups.

In conclusion, PA may help delay cognitive decline by modulating neurodegeneration and AD-specific tau pathologies. However, the cross-sectional design limits causal inference, and longitudinal studies are needed to confirm and clarify these