CELL INTELLIGENCE · 肿瘤细胞治疗研究

Lung adenocarcinoma (LUAD) is a highly heterogeneous malignancy with poor clinical outcomes, underscoring the urgent need for robust prognostic biomarkers and therapeutically tractable regulatory molecules. Long non-coding RNAs (lncRNAs) have emerged as key modulators of tumor progression and immune regulation; however, the prognostic and functional significance of TMPO-AS1 in LUAD remains largely unexplored.

Firstly, the top upregulated lncRNAs in LUAD were identified using the lnc2cancer3.0 database, and their prognostic significance was evaluated with the Kaplan-Meier Plotter. Differential expression of the selected lncRNA candidate was validated using TCGA-based platforms, including UALCAN, ENCORI and R-based statistical packages. TMPO-AS1-associated miRNAs were then predicted using the miRNet database, and their expression correlations, prognostic relevance, and differential expressions were assessed using the ENCORI, KM Plotter databases and R-based packages, respectively. A miRNA-centered heterogeneous gene model was constructed using the CancerMIRNome database, and gene-miRNA correlations were validated via ENCORI, followed by their survival analysis using the KM Plotter. Finally, immune cell infiltration associated with the identified genes was analyzed using the GSCA dataset.

TMPO-AS1 was found to be significantly overexpressed in LUAD patients (HR = 2.16). The correlation analysis revealed hsa-let-7b-5p was significantly and negatively correlated with TMPO-AS1, and its overexpression was also linked with better prognosis. Survival analysis selectively highlighted TGFBR3, RNF144B, CD59, and MAT2B as the most significant positively associated genes, while AURKA and KIFC1 emerged as the most significant negatively associated genes. Immune infiltration analysis demonstrated that the miRNA-positively associated genes were negatively correlated with nTreg cells and positively correlated with NKT cells, whereas the miRNA-negatively associated genes showed an inverse correlation pattern.

The study concludes that TMPO-AS1 is overexpressed in cases of LUAD, highlighting its potential as a molecular classifier for LUAD and a prognostic biomarker associated with poor clinical outcomes.