Logo image
The associations between major depressive disorder, plasma biomarkers of Alzheimer's disease, and mild behavior impairment among older adults
Journal article   Open access   Peer reviewed

The associations between major depressive disorder, plasma biomarkers of Alzheimer's disease, and mild behavior impairment among older adults

Yiqi Zhu, Jean‐Francois Trani, Ramkrishna Singh, Semere Bekena and Ganesh Babulal
Alzheimer's & dementia, Vol.21
12/01/2025

Abstract

Alzheimer's disease Antidepressants Behavior Depressive personality disorders Educational attainment Medical records Neuropsychiatric symptoms Physicians Plasma Politics Positive emotions Race Racial differences Symptoms Variables Adults African Americans Biological Markers Biomarkers Dementia Disease Medical Diagnosis Mental Depression Older People Psychiatry
Background Biomarkers and neuropsychiatric symptoms (NPS) are two early indicators for identifying Alzheimer's disease (AD) and related dementias (ADRD). Major depressive disorder (MDD) shares NPS with ADRD, complicating diagnoses. Plasma biomarkers, while non‐invasive, require further investigation to confirm their accuracy and consistency. Although both plasma biomarkers and mild behavior impairment, a new measure of NPS, are independently associated with ADRD risk, their interrelationship, especially in individuals with MDD, needs further exploration. This study explored the cross‐sectional associations between plasma biomarkers and MBI among participants with and without an MDD diagnosis. Methods Participants (n = 334) were community‐dwelling residents aged 65 and older and cognitively normal based on the Clinical Dementia Rating (CDR = 0). NPS were assessed using the MBI checklist. Plasma biomarkers including Aβ42/Aβ40, ptau181/nptau181, and ptau217/nptau217 were measured at C2N Diagnostics. MDD diagnoses were confirmed through medical records. Logistic regression models were used to evaluate the associations between MDD, plasma biomarkers, and MBI. The use of antidepressants and demographic variables were added to the models as control variables. Results Among the 334 participants, 236 were non‐depressed, and 98 were from the MDD cohort. 188 (56.3%) were female. The average age was 74.6 years (SD = 5.39), and the average year of education was 16.6 (SD = 2.35). Of the participants, 38 (11.4%) self‐identified as Black or African American. There was no significant difference in race or education level between participants with and without an MDD diagnosis. Compared to participants with neither a MDD diagnosis nor Aβ42/Aβ40 positivity, participants with MDD diagnoses and negative Aβ42/Aβ40 were 5.42 (95% CI = 2.69–11.57) times more likely to endorse MBI symptoms. Participants with both MDD and Aβ42/Aβ40 positivity were 7.23 (95% CI = 1.87–47.81) times more likely to endorse MBI symptoms. Conclusion MDD is associated with a worsening of MBI. Early identification and management of MDD may help reduce NPS severity. Differentiating ADRD from other neuropsychiatric conditions, such as MDD, remains a challenge. Further research is needed to clarify the distinct diagnostic and behavioral characteristics of ADRD and MDD.
url
https://doi.org/10.1002/alz70857_104024View
Published (Version of record) Open

Metrics

1 Record Views

Details

Logo image