2643 - Identification of Autophagy-Related Biomarkers and Diagnostic Model for Radiation-Induced Pneumonia
Presenter(s)
C. Xia1,2, P. Wang1,2, J. Deng1,2, and W. Nian1,2; 1Chongqing Municipal Health Commission Key Laboratory for Translational Research in Integrated Chinese and Western Medicine Oncology Diagnosis and Treatment, Chongqing, Chongqing, China, 2Department of Oncology, Chongqing Hospital of Traditional Chinese Medicine, Chongqing, Chongqing, China
Purpose/Objective(s): Radiation-induced pneumonia (RP) is a common dose-limiting complication in thoracic radiotherapy with an elusive molecular pathogenesis. Since autophagy regulates radiation-induced apoptosis and inflammation, this study aimed to identify autophagy-related biomarkers for RP using bioinformatics methods and validate their expression in vitro and diagnostic potential.
Materials/Methods: Gene expression profiles were obtained from GSE242706 dataset. Differentially expressed genes (DEGs) were identified using R, and intersection analysis with autophagy-related genes from GeneCards was performed to screen for differentially expressed autophagy genes (DEAGs). GO/KEGG analyses elucidated biological functions. A protein-protein interaction (PPI) network was constructed using the STRING database to identify hub genes. A diagnostic model was developed using logistic regression and evaluated by ROC curves. Hub genes were validated by qRT-PCR in irradiated A549 cells and further verified in an external dataset GSE242840. Finally, potential miRNA targets were predicted using miRDB and validated via the HMDD database.
Results: A total of 750 DEGs and 22 DEAGs were identified. Enrichment analysis showed DEAGs are primarily involved in autophagy, positive regulation of IL-1ß production, cell surface receptor signaling pathway via STAT and the NOD-like receptor signaling pathway. PPI analysis identified ten core genes: IGF1, NLRP3, IFNB1, IL6, TP53, TLR4, PIK3CG, FLT3, MEFV, and BCL2. ROC analysis showed that NLRP3, IFNB1, IL6, TP53, TLR4 and FLT3 had AUC values > 0.7. qRT-PCR confirmed significant differential expression of NLRP3, IL6 and TLR4 in the RP cell model (P<0.05), which was consistent with results from the external dataset GSE242840. These three genes were selected as candidate biomarkers. The 3-gene diagnostic model demonstrated good predictive accuracy with an AUC of 0.805. Furthermore, 11 miRNAs closely related to RP were successfully predicted as potential therapeutic agents.
Conclusion: NLRP3, IL6 and TLR4 are robust diagnostic biomarkers and potential therapeutic targets for RP. The constructed diagnostic model provides high predictive value, offering new insights into the interaction between autophagy and radiation-induced lung injury and providing a theoretical basis for clinical diagnosis and targeted therapy.