Back

ELAVL1: A Novel Diagnostic and Prognostic Biomarker in Lung Adenocarcinoma Linked to Ferroptosis

Li, X.-h.; Hu, Y.; Shao, B.; Wang, J.

2025-11-25 oncology
10.1101/2025.11.23.25340837 medRxiv
Show abstract

BackgroundLung adenocarcinoma (LUAD) is the most common malignant neoplasm with a poor 5-year survival rate (<16%). Ferroptosis, an iron-dependent cell death mechanism, has emerged as a key regulator of tumor progression, yet its clinical relevance in LUAD remains unclear. This study aimed to explore ferroptosis-related genes (FRGs) as diagnostic and prognostic biomarkers for LUAD. MethodsDifferentially expressed genes (DEGs) were identified from TCGA (526 LUAD vs. 59 normal) and GEO (GSE31210, 226 LUAD) datasets. Hub FRGs were selected via intersection of DEGs and FerrDb V2 genes, followed by survival analysis. Diagnostic models were built using SVM with ten-fold cross-validation, while prognostic signatures were developed through LASSO Cox regression. Immune infiltration (CIBERSORT) and functional enrichment (GO/KEGG) analyses were performed. ELAVL1 and SLC7A11 were validated by IHC in tissues and siRNA knockdown in PC9 cells. ResultsThe two-gene diagnostic model (ELAVL1 + SLC7A11) achieved high accuracy (AUC = 0.9675). High-risk patients (elevated ELAVL1/SLC7A11) showed poorer overall survival (median OS: 32.1 vs. 48.6 months, HR = 1.78, P < 0.001). Functional enrichment revealed cell cycle dysregulation and immune activation in high-risk groups. ELAVL1 knockdown suppressed LUAD cell proliferation/migration and downregulated CDK4/6. ConclusionELAVL1 serves as a dual-functional biomarker for LUAD diagnosis and prognosis, mechanistically linking ferroptosis to immune evasion and cell cycle progression. The two-gene signature provides a robust tool for clinical risk stratification, highlighting ELAVL1 as a potential therapeutic target.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.