Introduction: The poorly differentiated pancreatic adenocarcinoma (PDAC) is an extremely lethal neoplasm without effective biomarkers for early detection and prognosis prediction, which is characteristically unresponsive to chemotherapeutic regimens. This study aims at searching for key genes which could be applied as novel prognostic biomarkers and therapeutic targets in PDAC. Methods: Clinical samples were collected and a comprehensive differential analysis of seven PDAC samples by integrating RNA-seq data of tumor tissues and matched normal tissues from both our cohort and gene expression profiling interactive analysis (GEPIA) were performed to discover potential prognostic genes in PDAC. Pathway enrichment analysis was carried out to determine the biological function of PDAC differentially expressed genes (DEGs), and protein-protein interaction (PPI) network was constructed for functional modules analysis. Real-time PCR was performed to validate expression of hub genes. Results: A total of 126 PDAC-specific expressed genes identified from seven PDAC samples were predominantly enriched in cell adhesion, integral component of membrane, signal transduction and chemical carcinogenesis, IL-17 signaling pathway, indicating that obtained genes might play a unique role in PDAC tumorigenesis. Furthermore, survival analysis revealed that five genes (CEACAM5, KRT6A, KRT6B, KRT7, KRT17) which exhibited high expression levels in tumor tissues were obviously correlated with the prognosis of PDAC patients and KRT7 was positively correlated with KRT6A, KRT6B, KRT17 expression. In addition, real-time PCR demonstrated that the expression level of the hub genes was consistent with RNA-seq analysis. Discussion: The current study suggested that CEACAM5, KRT6A, KRT6B, KRT7, and KRT17 may represent novel prognostic biomarkers as well as novel therapeutic targets for poorly differentiated PDAC.