CELL INTELLIGENCE · 肿瘤细胞治疗研究

BACKGROUNDS: Non-small cell lung cancer (NSCLC), the most common type of lung cancer, stands as a leading cause of cancer-related mortality worldwide. Immunogenic cell death (ICD )enhances cancer therapy efficacy by inducing immune responses and reshaping the TIME. While ICD increases cytotoxic T lymphocyte infiltration and reduces immunosuppressive elements, the specific TIME subtypes associated with ICD in lung adenocarcinoma (LUAD) remain undefined. METHODS: Publicly available datasets from The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) were utilized in this study to identify differentially expressed genes related to ICD. We applied consensus molecular clustering to integrate these genes with clinical phenotypes, revealing distinct NSCLC subtypes with varying prognostic outcomes. The tumor immune microenvironments of these clusters were characterized using the ‘estimate’ R package ‘and ‘CIBERSORT’.A prognostic model was established utilizing LASSO Cox regression and validated across independent cohorts. Functional validation involved RNA interference targeting the KCNN4 gene in PC-9 lung cancer cells, including quantitative PCR, cell proliferation assays, wound healing assays, and Transwell invasion assays. Additionally, molecular docking and molecular dynamics simulations were performed to identify and validate small-molecule drugs targeting KCNN4. RESULTS: Unsupervised clustering of ICD-related gene expressions delineated three novel NSCLC subtypes. Cluster 1, characterized by younger patients, exhibited enhanced immune activity with significant infiltration of activated CD4+ and CD8+ T cells, correlating with a favorable prognosis and responsiveness to immunotherapy. Cluster 2, predominantly female, displayed suppressed immune responses with reduced effector memory T cells and γδ T cells, associated with poorer outcomes. Cluster 3, linked to varying cancer stages, showed moderate immune activity with lower immune cell infiltration and higher tumor purity, indicating an intermediate prognosis. A comprehensive prognostic model combined the expression levels of five key ICD-related genes (CSF2RB, CD3D, ADA2, KCNN4, and AREG) with critical clinical factors. Targeted silencing of KCNN4 in PC-9 cells significantly reduced tumor aggressiveness, supporting its role as a therapeutic target. Molecular docking identified four promising small-molecule drugs, with ZINC000000001547 (Hydroxystilbamidine) showing stable interactions in molecular dynamics simulations. CONCLUSIONS: This study identifies three distinct ICD-related TIME subtypes in NSCLC with significant translational potential: Cluster 1 patients are ideal candidates for immune checkpoint inhibitors; Cluster 2 may benefit from combination therapies to overcome resistance; and Cluster 3 requires aggressive multimodal treatments. Additionally, the predictive model enhances risk stratification, improving patient outcomes in NSCLC.