Integrating multi-omics and machine learning to uncover CCL20 as a potential regulator of the immunosuppressive microenvironment in lung adenocarcinoma.

Background: The interplay between malignant cells and the immunosuppressive tumor microenvironment (TME) is pivotal for lung adenocarcinoma (LUAD) progression. Methods: This study employed single-cell RNA sequencing, spatial transcriptomics, multi-omics analysis, and ensemble machine learning, combined with in vitro and in vivo functional experiments, to investigate the role of CCL20 in the immunosuppressive TME of LUAD. Results: We identified a distinct, metastasis-enriched malignant epithelial subpopulation characterized by a pro-inflammatory signature and high CCL20 expression. Pseudotime and regulon analyses suggested enrichment of NF-κB/STAT signaling regulon activity during malignant evolution. Spatial transcriptomics and cellular communication inference demonstrated that CCL20-high tumor cells co-localized with and actively recruited regulatory T cells (Tregs) via the specific CCL20-CCR6 ligand-receptor pair. Multi-omics analysis confirmed that high CCL20 expression correlated with increased Treg infiltration and served as an independent prognostic biomarker. An ensemble machine learning model based on the CCL20-CCR6 axis effectively stratified high-risk patients across multiple validation datasets. Functionally, genetic ablation of CCL20 attenuated the proliferative, migratory, and invasive capacities of LUAD cells Conclusions: The primary causal evidence chain of this study centers on the functional validation of the CCL20-CCR6 axis in Treg chemotaxis; the upstream computational inference of NF-κB/STAT signaling regulation should be regarded as a hypothesis-generating exploration requiring further validation. CCL20 thus represents a potential prognostic biomarker and therapeutic target for LUAD.