A new clustering: Angiogenesis‐associated molecular classification of lung adenocarcinoma

Lung adenocarcinoma (LUAD) is a heterogeneous disease with diverse molecular characteristics. In this study, we performed a comprehensive computer analysis of genomic data to identify angiogenesis‐associated molecular subtypes of LUAD. Using non‐negative matrix factorization (NMF) clustering on 1297 angiogenesis‐related genes, we classified LUAD into three distinct subtypes (C1, C2, and C3) with significant differences in overall survival (C1 and C2 showing better survival compared to C3, p < .001). Immune characterization revealed that C1 and C2 subtypes showed higher immune cell infiltration compared to the C3 subtype, including follicular helper T cells and NK CD56 bright cells. Additionally, C1 and C2 had higher activation scores for MYC and TP53 pathways. Evaluation using the ESTIMATE algorithm indicated that the C2 subtype had the highest immune score, while C3 had the highest stromal score. Functional enrichment analysis through gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) demonstrated that C1 and C2 subtypes were associated with cell cycle and hormone activity pathways. Drug sensitivity analysis revealed significant differences among subtypes in response to JNK Inhibitor VIII, AZD6244, and bicalutamide. Furthermore, five key genes ( FAM83A , FGB , S100A16 , TMEM125 , and TRIM29 ) were identified with significant impacts on patient survival ( p < .05). Single‐cell analysis uncovered coexpression patterns between these key genes and hypoxia‐related genes. Our findings offer a novel classification approach for LUAD, providing insights into its heterogeneity and potential for personalized antiangiogenic therapy.