Identification of prognostic genes in the pancreatic adenocarcinoma immune microenvironment by integrated bioinformatics analysis

Purpose Pancreatic adenocarcinoma (PAAD) is one of the most common causes of death among solid tumors, and its pathogenesis remains to be clarified. This study aims to elucidate the value of immune/stromal-related genes in the prognosis of PAAD through comprehensive bioinformatics analysis based on the immune microenvironment and validated in Chinese pancreatic cancer patients. Methods Gene expression profiles of pancreatic cancer patients were obtained from TCGA database. Differentially expressed genes (DEGs) were identified based on the ESTIMATE algorithm. Gene co-expression networks were constructed using WGCNA. In the key module, survival analysis was used to reveal the prognostic value. Subsequently, we performed functional enrichment analysis to construct a protein–protein interaction (PPI) network. The relationship between tumor immune infiltration and hub genes was analyzed by TIMER and CIBERSORT. Finally, it was validated in the GEO database and in tissues of Chinese pancreatic cancer patients. Results In the TCGA pancreatic cancer cohort, a low immune/stromal score was associated with a good prognosis. After bioinformatic analysis, 57 genes were identified to be significantly associated with pancreatic cancer prognosis. Among them, up-regulation of four genes (COL6A3, PLAU, MMP11 and MMP14) indicated poor prognosis and was associated with multiple immune cell infiltration. IHC results showed that PLAU protein levels from Chinese pancreatic cancer tissues were significantly higher than those from adjacent non-tumor tissues and were also associated with tumor TNM stage and lymph node metastasis. Conclusion In conclusion, this study demonstrates that PLAU may serve as a new diagnostic and therapeutic target, which is highly expressed in Chinese pancreatic cancer tissues and associated with lymph node metastasis. Graphical abstract