Integrative machine learning of hypoxia and centrosome-related gene signatures enables prognostic stratification and therapeutic insights in lung adenocarcinoma

Background: Lung adenocarcinoma (LUAD), a major subtype of non-small cell lung cancer (NSCLC), exhibits significant clinical heterogeneity and commonly observed therapeutic resistance. Although hypoxia-driven tumor adaptation and centrosome-mediated genomic instability are established microenvironmental drivers, their synergistic molecular contributions to LUAD progression remain poorly characterized. Therefore, this study aims to develop an integrative machine learning (ML) model based on hypoxia and centrosome-related genes to enable prognostic stratification and provide therapeutic insights for LUAD. Methods: We developed an integrative multi-omics framework that combines weighted gene co-expression network analysis (WGCNA) to identify key regulatory modules and single-sample gene set enrichment analysis (ssGSEA) for assessing the hypoxia and-centrosome pathway. Differential expression analysis of The Cancer Genome Atlas Lung Adenocarcinoma (TCGA-LUAD) cohorts identified hypoxia-centrosome-associated genes, which were refined via univariate Cox regression and ML to construct a prognostic signature. Clinical relevance was validated through nomogram development, tumor microenvironment (TME) profiling, mutational burden assessment, and therapeutic response prediction. Results: A 16-gene prognostic signature was established using 306 differentially expressed genes linked to hypoxia and centrosome dysregulation. Stratification of LUAD patients into high- and low-risk groups demonstrated longer overall survival (OS) in the low-risk cohort. High-risk patients demonstrated elevated tumor mutational burden (TMB) and immunosuppressive microenvironment features, including reduced infiltration of eosinophils, immature dendritic cells, and mast cells. Risk scores were correlated with sensitivity to targeted therapy and chemotherapy. Conclusions: Our integrative ML model uncovers hypoxia-centrosome crosstalk as a critical driver of LUAD progression. The hypoxia and centrosome score-related genes (HCSRGs) signature enables robust risk stratification and identifies actionable targets for precision oncology, providing a framework for personalized therapeutic strategies in LUAD.