CELL INTELLIGENCE · 肿瘤细胞治疗研究

Lung Squamous Cell Carcinoma (LUSC), a common subtype of non-small cell lung cancer, primarily occurs in elderly individuals and is closely linked to senescence-related biological processes. Meanwhile, senescence reshapes the tumor microenvironment by promoting immune suppression and angiogenesis, and its underlying mechanisms remain unclear. Here, we identified three senescence-related molecular subtypes of LUSC, each with distinct clinical outcomes, genomic alterations, and immune features. Among them, Cluster 1 was associated with poor prognosis, increased genomic instability, and an immunosuppressive TME enriched with senescence signatures. To uncover the underlying mechanisms, we integrated bulk and single-cell transcriptomic data and found extensive senescence reprogramming across epithelial, myeloid, and T/NK cell compartments. Notably, senescent immune subsets-such as S100A9 + macrophages and TNFRSF4 + Tregs-were markedly enriched in older patients. Furthermore, cell communication analysis revealed enhanced intercellular signaling in senescent cells. Based on these findings, we constructed a robust machine learning-based prognostic model incorporating senescent gene expression and cell-type abundance to predict survival and immunotherapy response. Finally, functional validation demonstrated that UGT1A7 acts as a pro-proliferative driver that is significantly downregulated during the induction of cellular senescence. Experimental depletion of UGT1A7 phenocopies this senescence-associated decline, thereby triggering growth arrest and inhibiting tumor progression, suggesting its potential as a therapeutic target. Our study reveals how senescence-related alterations in cellular composition and transcriptional dysregulation reshape the tumor microenvironment, providing novel biomarkers and a theoretical basis for therapy.