Identification of novel biomarkers for Alzheimer’s disease: A deep learning omics-based approach to drug pair discovery and exploration of potential therapeutic targets

The understanding of Alzheimer's disease is shifting from a traditional focus on Aβ/tau pathology to an emerging consensus that positions immune dysregulation as a central synergistic driver in the early stages of the disease. However, the causal relationships between peripheral immune cells, plasma proteins, and Alzheimer's disease, as well as the mediating effects of plasma proteins on the disease, remain poorly understood. Moreover, there are no effective drug combination therapies targeting plasma proteins for Alzheimer's disease. This study investigated the causal associations between immune cells, plasma proteins, and Alzheimer's disease, with a focus on the role of Fc gamma receptor 3A in disease progression. Using a two-sample Mendelian randomization approach, we identified 59 plasma proteins and 65 immune cell types significantly associated with Alzheimer's disease. We performed data mining of a large Alzheimer's disease cohort and drug databases and established a biofactor-regulated neural network for rapidly screening and optimizing compound drug pairs. Among the immune cells, CD8 + T cells, particularly CD8 + CD28 + CD45RA. T cells, were found to have a protective effect against Alzheimer's disease. Furthermore, increased expression of Fc gamma receptor 3A (also known as CD16a, an activating receptors of NK cells) in plasma and the hippocampus correlated with enhanced CD8 + T-cell infiltration and accelerated Alzheimer's disease progression in 5×FAD mice. Mediation analysis revealed that Fc gamma receptor 3A mediates the effects of CD8 + T cells on Alzheimer's disease risk. Additionally, Fc gamma receptor 3A gene expression levels were significantly higher in patients with Alzheimer's disease compared with individuals with mild cognitive impairment and cognitively normal participants, as revealed by an analysis of the Alzheimer's Disease Neuroimaging Initiative database. These findings suggest that CD8 + T-cell infiltration and Fcγ receptor 3A expression play critical roles in the pathophysiology of Alzheimer's disease and may serve as therapeutic targets. Molecular docking analysis further identified 19 candidate drugs targeting Fcγ receptor 3A. This study proposes novel immune-based therapeutic strategies and introduces an omics-based intelligent drug discovery framework for repurposing existing drugs for the treatment of complex diseases. The key contributions of this study include the identification of potential immune-based therapeutic targets for Alzheimer's disease and demonstration of the utility of bioinformatics and drug repurposing approaches in addressing complex neurodegenerative diseases.