Systematic analysis of anoikis-related genes identifies SRPX2-FAK/AKT-IL-6 axis in the progression and peritoneal metastasis of gastric cancer

Introduction: Peritoneal metastasis (PM) remains a primary cause of poor prognosis in advanced gastric cancer (GC). While anoikis resistance enables detached tumor cells to survive and promotes invasion and metastasis, its specific mechanisms in GC related PM are not yet fully understood. Methods: Anoikis related differentially expressed genes were identified using GC samples from the TCGA and GEO databases. Molecular subtypes were characterized by non negative matrix factorization (NMF) clustering. Survival outcomes were assessed via Kaplan-Meier analysis, and functional profiles were evaluated through pathway enrichment scoring. A prognostic risk model was constructed by combining weighted gene co expression network analysis (WGCNA) with Lasso-Cox regression. For PM diagnosis, characteristic genes were screened using support vector machine (SVM) and random forest (RF) algorithms to build a diagnostic model. Experimental validation was performed to confirm the expression and functional role of SRPX2. Results: GC was stratified into two molecular subtypes. Subtype A showed marked enrichment of anoikis resistance related factors and was associated with significantly poorer prognosis. A six gene prognostic signature (HEYL, SRPX2, LBH, PLAT, ITGAV, HTRA1) was established and validated externally. A nine gene diagnostic panel (SLC30A9, ZFHX4, CYTB, NDFIP2, NMNAT2, SRPX2, TBC1D8B, CLEC3B, CHRDL2) was also constructed for PM. SRPX2 was identified as an independent prognostic marker and a PM associated biomarker, highly expressed in cancer associated fibroblasts (CAFs). Functional studies demonstrated that SRPX2 promotes GC progression and peritoneal metastasis by activating the FAK/AKT pathway and IL-6 paracrine signaling, thereby inducing anoikis resistance. Discussion: This study proposes an anoikis based molecular classification of GC and develops both prognostic and diagnostic models for PM. It further reveals that SRPX2 driven anoikis resistance, mediated through the FAK/AKT-IL-6 axis, facilitates peritoneal metastasis. These findings provide new directions for prognostic assessment and therapeutic strategies in gastric cancer.