.png){kind=icon}
.png){kind=icon}
Return
EMCN is associated with vascular-immune crosstalk and represents a potential biomarker in lung adenocarcinoma
.png){kind=icon}
OA
AI
Y
Q
K
F
X
Y
DOI：10.3389/fmolb.2026.1752442.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
BackgroundWhile MUC family genes have been established as prognostic biomarkers in gastric cancer; and GWAS studies link EMCN mutations to chemotherapy-induced myelosuppression in NSCLC; the systematic characterization of EMCN in lung adenocarcinoma (LUAD) remains elusive.MethodsThis multi-omics strategy combining bulk and single-cell transcriptomics study integrated differential expression analysis; WGCNA; and machine learning algorithms (LASSO/SVM-RFE/Random Forest) to identify EMCN as a diagnostic hub gene; followed by experimental validation using immunohistochemistry Western blot and qRT-PCR.ResultsEMCN (Endomucin) is a sialomucin-like glycoprotein predominantly expressed in vascular endothelial cells. Using bulk transcriptomic datasets and single-cell RNA-seq analysis; we found that EMCN expression was reduced in lung adenocarcinoma (LUAD) compared with non-tumor controls and was primarily localized to the endothelial compartment. Survival analysis using the median expression cutoff showed that high EMCN expression was associated with improved overall survival (Cox HR_high vs. low = 0.73; p = 0.04); indicating that low EMCN expression correlates with poorer prognosis. Machine learning-based feature selection (LASSO; Random Forest; and SVM) further prioritized EMCN among consensus candidate genes; supporting its potential relevance to the vascular-associated tumor microenvironment in LUAD. EMCN expression levels also showed a significant positive correlation with the degree of immune cell infiltration. Gene set enrichment analysis (GSEA) revealed that high EMCN expression in tumor tissues activates negative regulatory pathways associated with angiogenesis. Receiver operating characteristic (ROC) curve analysis highlights EMCN’s excellent diagnostic potential for LUAD; with an area under the curve (AUC) of 0.963. In vitro experiments confirm the downregulation of EMCN at both protein and mRNA levels; consistent with our bioinformatics predictions.ConclusionThis first comprehensive study establishes EMCN as a dual-functional regulator of vascular-immune crosstalk in LUAD; providing both a molecular diagnostic tool and therapeutic target for precision oncology.
Keywords:
machine learning
biomarker
lung adenocarcinoma
immune infiltration
EMCN
vascular-immune crosstalk
Journal
IF:
4
Papers:
6.0K
Citations:
2.0W
Organization
Cited Papers