International Journal of Molecular Medical Science, 2025, Vol.15, No.4, 175-184 http://medscipublisher.com/index.php/ijmms 181 6.2 The meanings of pathways related to neuroinflammation, apoptosis and synaptic function Pathway enrichment results often show that in AD, differentially expressed mirnas mainly affect processes such as neuroinflammation, cell death (apoptosis), and synaptic function. For instance, mirnas that regulate genes in the MAPK and tgf-β pathways are associated with brain inflammation and the survival of nerve cells; Mirnas that affect DNA damage responses and apoptotic pathways are associated with nerve cell death (Lu et al., 2012; Pereira et al., 2024). This indicates that abnormalities in miRNA may drive the development of AD by altering these key molecular processes. Furthermore, the enrichment in synaptic plasticity and neural signaling pathways indicates that miRNA is crucial for maintaining synaptic health and cognitive function. Mirnas regulating genes related to synaptic vesicles, cell adhesion and nerve fiber orientation were discovered, supporting their involvement in typical synaptic function impairment in AD (Lu et al., 2012; Backes et al., 2016; Tastsoglou et al., 2023). Therefore, pathway analysis not only explains how miRNA changes function but also points out potential therapeutic targets. 6.3 Disease progression prediction model based on miRNA combination and ROC evaluation To predict disease progression using a group of differentially expressed mirnas, machine learning methods such as logistic regression, support vector machines or random forests are often employed. These models are trained based on miRNA expression data with the aim of distinguishing disease stages or predicting progression. Which mirnas to incorporate into the model will refer to pathway analysis and biological significance (Pereira et al., 2024). The model effect is usually evaluated by receiver operating characteristic (ROC) curve analysis, which measures the recognition ability (sensitivity and specificity) under different judgment criteria. Studies have shown that mirNa-based models can diagnose AD or predict its development very accurately, and the area under the ROC curve (AUC) value is often greater than 0.85 (Pereira et al., 2024). Combining miRNA combinations with clinical information and brain scan results can further enhance predictive capabilities, supporting the use of exosome mirnas as non-invasive markers for the early detection and tracking of AD (Lu et al., 2012; Pereira et al., 2024). 7 Concluding Remarks Exosomal miRNA in the blood has great application value as a biomarker for the early detection and disease tracking of AD. They are very stable in body fluids, can cross the blood-brain barrier, and their expression patterns are related to diseases, so they are suitable for non-invasive detection. New types of sensors such as electrochemical ones are highly sensitive and specific in detecting exosomal miRNA, with an accuracy rate of over 90% in distinguishing AD patients from healthy individuals. Specific miRNA combinations (such as miR-125b, miR-342-3p, miR-30b-5p) have shown stable performance in studies, supporting their use for early detection and disease staging. The changes in these miRNA levels are associated with cognitive decline and known AD markers, which is beneficial for disease monitoring. However, challenges still exist: a unified approach is needed for exosome isolation, miRNA detection and data analysis to ensure comparable results. Technological progress is driving clinical application. Combining exosome miRNA information with brain scans (such as PET and MRI) can improve the accuracy of AD diagnosis and prediction of disease progression. This combination of multiple methods can simultaneously observe changes at the molecular level and in brain structure, providing a more comprehensive judgment on the state and development of diseases. The model that combines miRNA, cognitive scores and imaging data has a better diagnostic effect than a single method, and the AUC value of some studies exceeds 0.88. This can also help identify high-risk groups that may develop from mild memory problems to AD, facilitating early intervention. Formulating a unified multi-method diagnostic process is crucial for practical applications. Although the results are positive, large-scale multicenter studies are still needed to confirm the value of exosomal mirnas in different populations and hospitals. Current research is often limited by small sample sizes, being conducted only in a single hospital, or inconsistent methods, which affects the wide applicability of the results. To establish reliable and reproducible markers, there is an urgent need for unified sampling, separation and analysis
RkJQdWJsaXNoZXIy MjQ4ODYzNA==