CELL INTELLIGENCE · 肿瘤细胞治疗研究

BACKGROUND: Non-small cell lung cancer (NSCLC) is the most common lung cancer subtype and a leading cause of cancer-related mortality due to asymptomatic early stages and poor 5-year survival, and tumor-infiltrating lymphocytes (TILs) play a pivotal but poorly defined role in NSCLC progression and prognosis within the tumor microenvironment (TME). METHODS: We analyzed single-cell RNA sequencing data from GEO (GSE148466), NSCLC-related data from TCGA, and gene expression data from GEO (GSE50081). A 9-gene TIL-related risk score model (HLA-DRA, NPC2, PRPF38B, PABPC1, ENO1, ANXA2, LGALS1, S100A10, SRGN) was constructed using LASSO regression, with its prognostic value assessed via Kaplan-Meier and ROC analyses and external validation. Immune infiltration and regulatory factors were analyzed using TISIDB, pathway associations via GSVA and GSEA, and functional roles of key genes (HLA-DRA, NPC2, PRPF38 B) validated using RT-qPCR, Western blotting, and co-culture assays. RESULTS: TILs were abundant in both smoking and non-smoking NSCLC samples. The 9-gene risk model effectively stratified patients by survival, with high-risk patients showing poorer outcomes and enriched tumor-promoting immune features and pathways. High-risk scores were associated with increased M0 macrophages, activated NK cells, and elevated immune checkpoints. Overexpression of HLA-DRA, NPC2, and PRPF38 B inhibited NSCLC cell proliferation, migration, and invasion, while enhancing apoptosis and suppressing M2 macrophage polarization. CONCLUSION: This study establishes a TIL-based prognostic risk model for NSCLC and identifies HLA-DRA, NPC2, and PRPF38 B as potential immune-regulatory therapeutic targets with tumor-suppressive functions, providing new insights for NSCLC precision immunotherapy.