Research Summary:

Depressive symptoms are common in older adults and have been associated with risk of Alzheimer’s disease and Alzheimer’s disease and related dementias (AD/ADRD), but the mechanisms and biomarkers underlying this relationship remain unclear. Prior studies have reported associations between depressive symptoms and AD/ADRD pathologies, sometimes before cognitive symptoms emerge, suggesting that late-life depressive symptoms might reflect early underlying disease processes. Blood-based biomarkers (BBMs) provide in vivo measures of AD/ADRD-related pathology, including amyloid beta (Aβ) 42/40 ratio, phosphorylated tau181 (p-tau181), neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP). Studies examining associations between amyloid or tau biomarkers and depressive symptoms have produced mixed findings, and neurodegeneration and neuroinflammation might represent shared biological pathways between depression and AD/ADRD. GFAP, an intermediate filament expressed by astrocytes, might reflect astrocytic damage, astrogliosis, or neuroinflammation, and plasma GFAP has previously been associated with clinical depression. This study by Bacci et al1 examined cross-sectional associations between AD/ADRD BBMs and depressive symptoms in a large, community-based cohort of older adults enrolled in the Aspirin in Reducing Events in the Elderly (ASPREE) clinical trial and evaluated whether these associations differed by sex or apolipoprotein E (APOE) ε4 carrier status.

Baseline data were analyzed from participants in ASPREE, a double-blind, randomized primary prevention trial that enrolled older adults from 2010 to 2014 to evaluate daily low-dose aspirin effects on dementia, persistent physical disability, and death. ASPREE excluded individuals with a history of cardiovascular disease events, diagnosed dementia or Modified Mini-Mental State Examination score <78, disability affecting activities of daily living, or conditions likely to cause death within 5 years. Of 19,114 participants enrolled at ASPREE baseline, 7,167 were excluded because of missing or incomplete data, leaving 11,947 participants with baseline AD/ADRD BBM and depressive symptom data for the current analysis. Baseline blood samples were collected within the first 12 months of the trial. Depressive symptoms were measured with the 10-item Center for Epidemiological Studies Depression Scale (CESD-10), with scores ranging from 0 to 30, with higher scores indicating greater depressive symptoms. Associations between BBMs and depressive symptoms were assessed using linear regression models. Model 1 was unadjusted; Model 2 adjusted for demographic, lifestyle, chronic condition, laboratory, and cognition variables; and Model 3 additionally adjusted for antidepressant use and history of depression. Sex- and APOE ε4-stratified analyses and interaction terms were used to evaluate effect modification, with P<0.05 as the significance threshold.

The 11,947 included participants had a median age of 73.87 years, 53.5% were women, and 98.6% were White. The median education duration was 12.00 years, 22.2% reported parental history of dementia, and 24.9% were APOE ε4 carriers. Median CESD-10 score was 2.00. Median BBM values were 0.06 for Aβ42/40 ratio, 31.46 pg/mL for p-tau181, 19.84 pg/mL for NfL, and 120.58 pg/mL for GFAP. In the unadjusted Model 1, lower p-tau181, higher NfL, and higher GFAP were associated with higher depressive symptoms, with beta values of −0.096 (standard error [SE]: 0.045; P=0.033), 0.183 (SE: 0.050; P<0.001), and 0.221 (SE: 0.044; P<0.001), respectively. Aβ42/40 ratio was not associated with depressive symptoms (beta: −0.034; SE: 0.086; P=0.690). In Model 2, after adjustment for demographic, lifestyle, clinical, laboratory, and cognition variables, higher NfL and higher GFAP remained associated with higher depressive symptoms, with beta values of 0.142 (SE: 0.060; P=0.018) and 0.123 (SE: 0.051; P=0.015), respectively; Aβ42/40 ratio and p-tau181 were not associated with depressive symptoms. In Model 3, after additional adjustment for antidepressant use and history of depression, only higher GFAP remained associated with higher depressive symptoms (beta: 0.270; SE: 0.088; P=0.002), while Aβ42/40 ratio, p-tau181, and NfL were not associated. Results were largely unchanged after correction for multiple comparisons. In sex-stratified analyses, higher GFAP was associated with higher depressive symptoms only among men, but formal interaction testing showed that no BBM-by-sex interaction term was statistically significant; the GFAP-sex interaction was shown to be trending, but not significant (P=0.075). In APOE ε4-stratified analyses, higher GFAP was associated with higher depressive symptoms only among APOE ε4 noncarriers, but this association did not remain significant after multiple-comparison correction, and no BBM-APOE ε4 interaction term was significant.

This study found that plasma GFAP, but not Aβ42/40 ratio, p-tau181, or NfL, was independently associated with depressive symptoms. Plasma Aβ42/40 ratio, p-tau181, and NfL were not independently associated with depressive symptoms in the fully adjusted model. These findings suggest that neuroinflammation, reflected by plasma GFAP, might be linked to depressive symptoms in older adults, but longitudinal studies in more diverse cohorts are needed to determine directionality and clarify whether this association differs across subgroups.


Reference:

  1. Bacci JR, Ryan J, Murray AM, et al. The association of Alzheimer’s disease and related dementias blood-based biomarkers with depressive symptoms. Alzheimers Dement. 2026;22(1):e71007.