Multimodal Plasma Proteomic and T1-Radiomic Correlates of Memory Impairment in Biomarker-Defined Amyloid- Positive Late-Life Depression

Background Late-life depression (LLD) is associated with an increased risk of dementia and a biologically defined subgroup exhibits amyloid pathology associated with Alzheimer’s disease (AD). However, the molecular and neuroanatomical characteristics of this subgroup remain incompletely understood. Investigating the relationships among plasma proteomic profiles, T1-derived radiomic features, and cognition may help characterize biological heterogeneity at the intersection of LLD and the AD continuum. Methods This integrative proteomics–radiomics study included 116 LLD patients, whose AD biomarker status was defined using Core 1 biomarkers according to the 2024 Alzheimer’s Association revised criteria. A machine learning–derived radiomic signature distinguishing biomarker-defined amyloid-positive from amyloid-negative LLD was constructed. The key plasma proteomic module was identified using weighted gene co-expression network analysis, and its mechanistic role was evaluated using mediation analysis. Results The radiomic score (RadScore) discriminated biomarker-defined amyloid-positive LLD in a held-out internal test subset (AUC = 0.788), with influential features from the paracentral lobule and precuneus. It correlated with a neutrophil activation–enriched module (r = − 0.49, p < 0.001). The cross-sectional indirect-effect analysis indicated that the RadScore accounted for 52.5% of the modeled association with memory performance. Conclusions In biomarker-defined amyloid-positive LLD, T1-derived radiomic features were associated with a plasma protein module enriched for innate immune and neutrophil-related annotations and with memory performance. These cross-modal associations require independent longitudinal validation.

Authors

Publication Details

Journal
Research Square
Published
2026-11-18
DOI
https://doi.org/10.21203/rs.3.rs-10672734/v1
Primary Topic
Dementia and Cognitive Impairment Research
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Multimodal Plasma Proteomic and T1-Radiomic Correlates of Memory Impairment in Biomarker-Defined Amyloid- Positive Late-Life Depression

Haoye Tan, Yuping Ning, Zhidai Xiao, Danyan Xu et al.
2 links
Research Square
Dementia and Cognitive Impairment Research
preprint

Multimodal Plasma Proteomic and T1-Radiomic Correlates of Memory Impairment in Biomarker-Defined Amyloid- Positive Late-Life Depression

Haoye Tan, Yuping Ning, Zhidai Xiao, Danyan Xu, Pengbo Gao, Mingfeng Yang, Xiaomei Zhong, Ting Su, Youxuan Zheng, Huanmin Liu, Zhen Peng, Weilin Wang, Ben Chen, Wei He, Qiang Wang
preprint en

Abstract

Background Late-life depression (LLD) is associated with an increased risk of dementia and a biologically defined subgroup exhibits amyloid pathology associated with Alzheimer’s disease (AD). However, the molecular and neuroanatomical characteristics of this subgroup remain incompletely understood. Investigating the relationships among plasma proteomic profiles, T1-derived radiomic features, and cognition may help characterize biological heterogeneity at the intersection of LLD and the AD continuum. Methods This integrative proteomics–radiomics study included 116 LLD patients, whose AD biomarker status was defined using Core 1 biomarkers according to the 2024 Alzheimer’s Association revised criteria. A machine learning–derived radiomic signature distinguishing biomarker-defined amyloid-positive from amyloid-negative LLD was constructed. The key plasma proteomic module was identified using weighted gene co-expression network analysis, and its mechanistic role was evaluated using mediation analysis. Results The radiomic score (RadScore) discriminated biomarker-defined amyloid-positive LLD in a held-out internal test subset (AUC = 0.788), with influential features from the paracentral lobule and precuneus. It correlated with a neutrophil activation–enriched module (r = − 0.49, p < 0.001). The cross-sectional indirect-effect analysis indicated that the RadScore accounted for 52.5% of the modeled association with memory performance. Conclusions In biomarker-defined amyloid-positive LLD, T1-derived radiomic features were associated with a plasma protein module enriched for innate immune and neutrophil-related annotations and with memory performance. These cross-modal associations require independent longitudinal validation.

Research Square
Guangzhou Medical University
Good health and well-being
Dementia and Cognitive Impairment Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon