Abstract:Background and Aims Pancreatic cancer is a highly malignant tumor with a very poor prognosis, with a 5-year survival rate of about 11.5%. Nearly half of the patients have distant metastasis at the time of initial diagnosis, and liver metastasis accounts for 37% to 41.9% of them. Exploring new biomarkers for pancreatic cancer liver metastasis may help improve the treatment efficacy in patients. Therefore, this study was conducted to identify and validate key genes that play a critical role in the process of pancreatic cancer liver metastasis using bioinformatics approaches.Methods The high-throughput sequencing dataset GSE151580 for pancreatic ductal adenocarcinoma (PDAC) was downloaded from the GEO database, which included tissue samples from pancreatic cancer liver metastases and primary lesions. The differentially expressed genes between liver metastasis tissue samples and primary lesion tissue samples were screened using the R language limma package. The GO and KEGG functional enrichment analyses were performed on the differentially expressed genes. The protein-protein interaction networks were constructed using the STRING database, which were then visualized using Cytoscape. The top 10 genes were selected using the CytoHubba plugin based on the MCC topology analysis method, which were considered as the candidate core genes. Finally, the candidate core genes were validated using TCGA, GEPIA, UALCAN, and HPA databases.Results A total of 46 512 genes were included in the analysis, with 491 differentially expressed genes meeting the screening criteria, of which 162 were up-regulated and 329 were down-regulated. After selecting the top 10 genes with the highest MCC scores, validation of the candidate genes showed that the APOB gene was highly expressed in tumor tissues (P<0.05), with its expression product mainly located in the cytoplasm and cell membrane, and showing moderate positive staining in immunohistochemistry. APOB gene mutations were related to patients' M stage, with a higher proportion of M1 patients in the mutation group (P=0.022 1). However, the expression of this gene was not significantly associated with overall survival (OS) or disease-free survival (DFS) of the patients (both P>0.05). In addition, the expression product of the APOA4 gene was also mainly located in the cytoplasm and cell membrane, showing moderate positive staining in immunohistochemistry. APOA4 gene mutations were related to patients' TNM stage, with an earlier TNM stage in the mutation group (P=0.018 3). Patients with low expression of this gene had higher DFS (HR=1.75, P=0.025), but its expression was not related to OS (P>0.05).Conclusion The APOB gene may be associated with liver metastasis of pancreatic cancer and has the potential to serve as a molecular biomarker for early screening of pancreatic cancer liver metastasis. The APOA4 gene is associated with the DFS of pancreatic cancer patients and may become a new molecular biomarker for evaluating patient prognosis, monitoring tumor recurrence, or as a potential target for gene therapy.