Exosomal Prognostic Score (EXPS)-based Prognostic Model in Pancreatic Tumours: Construction and Validation

Pancreatic adenocarcinoma (PAAD) is a highly lethal malignancy with limited therapeutic options and poor prognosis. Extracellular vesicles (EVs), essential mediators of intercellular communication, are emerging as non-invasive biomarkers for diagnosis and prognosis. Investigating exosomal gene expression offers a more accurate prognostic tool for PAAD patients.We analyzed four datasets from public databases: PAAD_ExoRbase, TcgaTargetGtex-PAAD, GSE62452_GPL6244, and GSE78229_GPL6244. Differentially expressed exosomal genes were identified, followed by univariate Cox regression to pinpoint prognosis-associated exogenes. Single-cell analysis and single-sample gene set enrichment analysis (ssGSEA) refined the gene selection, assessing their origins and predictive value for immune status. In vitro experiments and RNA-seq validated the regulatory role of the key gene KRT7.LASSO Cox regression identified five prognostic genes: KRT7, KRT19, CXCL5, TNFRSF13C, and GP1BA. The prognostic model accurately predicted overall survival in the TCGA-PAAD, GSE62452, and GSE78229 cohorts. Single-cell analysis revealed the gene signature originated from T-proliferative, malignant, and ductal cells. Immune cell infiltration and immune checkpoint analyses confirmed the model’s predictive strength. In vitro assays and RNA-seq demonstrated that KRT7 promotes PAAD cell proliferation, migration, and invasion.