Establishment and Analysis of a Prognostic Model of Pancreatic Ductal Adenocarcinomas Based on Nerve-Cancer Crosstalk-related Genes

Background Pancreatic ductal adenocarcinoma（PDAC）is the seventh leading cause of cancer-related death globally. It has an extremely poor prognosis, and its incidence rate and mortality are considered to increase sharply over the next few years. Combined chemotherapy after surgery is still the best treatment, and it is also the only curative option. Improvements in the understanding of genetic mutations, the tumor microenvironment (TME), and targeted therapies continue to improve the progression of PDAC. Nerves are an important component of the TME of pancreatic cancer. However, the role of cancer-nerve crosstalk-related genes in PDAC is still unclear. Results In our study, we constructed a prognostic model based on cancer-nerve crosstalk-related genes with prognostic value. The xCell algorithm was used to analyse the cell component enriched from samples in The Cancer Genome Atlas (TCGA) PDAC cohort. We found the neuronal content was significantly correlated with prognosis. Protein-protein interaction (PPI) network, GO, and KEGG enrichment analyses were used to identify nerve-related genes in PDAC. From the forty-two nerve-cancer crosstalk genes, we used K-M plotter to identify the canditate genes for LASSO regression. We constructed a three-gene prognostic model (CHRNB2, CHRNA10, NGFR). The model could effectively test and verify its 1-, 3- and 5-year survival rates in the TCGA-PDAC cohort. The effectiveness of the model was verified by external test data. The genes of this model were highly related to tumor progression and other immune microenvironments. Conclusion In summary, we used bioinformatics analysis to construct an effective prognostic model, explored the key mechanism of nerve-cancer cross-talk and revealed the potential of cancer-nerve cross-talk-related genes as prognostic biomarkers and therapeutic targets in PDAC.