Machine learning identifies TIME subtypes linking EGFR mutations and immune states in lung adenocarcinoma.

Gong, Zetian; Du, Mingjun; Li, Ying; Ye, Bicheng; Huang, Yuming; Gong, Hui; Wang, Wei; Chen, Liang et al. · NPJ Digit Med · 2025

basic_science · Level V

Where this comes from

Abstract

Epidermal growth factor receptor (EGFR) mutation is a key oncogenic driver in lung adenocarcinoma (LUAD), but its impact on the tumor immune microenvironment (TIME) remains unclear. By integrating single-cell transcriptomes from 153 LUAD samples using machine learning, we generated an atlas of over one million cells that delineates immune heterogeneity. EGFR-mutant tumors exhibited enrichment of TIGIT<sup>+</sup>regulatory T cells, neutrophils, and macrophages, whereas wild-type tumors contained abundant ZNF683<sup>+</sup>CD8<sup>+</sup>tissue-resident memory T cells, diverse memory B cells, and FGFBP2<sup>+</sup>CD16<sup>high</sup> natural killer cells, reflecting an immune-active TIME. Non-negative matrix factorization defined five TIME subtypes, with EGFR-mutant patients clustering into immunosuppressive profiles linked to poor prognosis. Flow cytometry and mouse models confirmed the cytotoxic and PD-1 blockade-enhancing functions of FGFBP2<sup>+</sup>NK cells. These findings reveal distinct TIME landscapes in EGFR-mutant LUAD and illustrate the potential of machine learning-based immunogenomic analysis to inform precision immunotherapy.