Integrated single-cell and bulk transcriptomics reveals STAB1 as a novel therapeutic target for ovarian cancer

The immunosuppressive tumor microenvironment (TME) and intrinsic heterogeneity of ovarian cancer (OC) are primary drivers of therapeutic resistance and mortality. To deconvolute these complex dynamics and identify robust therapeutic targets, this study employed an integrative strategy combining ensemble machine learning algorithms with high-resolution single-cell transcriptomics and experimental validation. Through dual-feature selection (LASSO and SVM-RFE) applied to multi-cohort bulk transcriptomic data, we identified Stabilin-1 (STAB1) as a top-ranked prognostic determinant. Crucially, single-cell analysis of the OC ecosystem redefined the cellular localization of STAB1, revealing its predominant enrichment in LYVE1+ perivascular-like M2 macrophages and a hyper-aggressive, EMT-active tumor subpopulation. Validating these in silico insights, in vitro loss-of-function assays confirmed that STAB1 silencing in OC cell lines (A2780 and SK-OV-3) significantly suppressed cell proliferation, colony formation, and invasion. Collectively, our findings support STAB1 as a pivotal "dual-checkpoint" molecule that bridges the immunosuppressive stroma and the malignant epithelium, highlighting its potential as a novel therapeutic target for dismantling the ovarian cancer ecosystem.