Abstract
Pancreatic cancer (PC) is a malignant tumor of the digestive system that has a bad prognosis. N6-methyladenosine (m6A) is involved in a wide variety of biological activities due to the fact that it is the most common form of mRNA modification in mammals. Numerous research has accumulated evidence suggesting that a malfunction in the regulation of m6A RNA modification is associated with various illnesses, including cancers. However, its implications in PC remain poorly characterized. The methylation data, level 3 RNA sequencing data, and clinical information of PC patients were all retrieved from the TCGA datasets. Genes associated with m6A RNA methylation were compiled from the existing body of research and made available for download from the m6Avar database. The LASSO Cox regression method was used to construct a 4-gene methylation signature, which was then used to classify all PC patients included in the TCGA dataset into either a low- or high-risk group. In this study, based on the set criteria of |cor| > 0.4 and p value < 0.05. A total of 3507 gene methylation were identified to be regulated by m6A regulators. Based on the univariate Cox regression analysis and identified 3507 gene methylation, 858 gene methylation was significantly associated with the patient's prognosis. The multivariate Cox regression analysis identified four gene methylation (PCSK6, HSP90AA1, TPM3, and TTLL6) to construct a prognosis model. Survival assays indicated that the patients in the high-risk group tend to have a worse prognosis. ROC curves showed that our prognosis signature had a good prediction ability on patient survival. Immune assays suggested a different immune infiltration pattern in patients with high- and low-risk scores. Moreover, we found that two immune-related genes, CTLA4 and TIGIT, were downregulated in high-risk patients. We generated a unique methylation signature that is related to m6A regulators and is capable of accurately predicting the prognosis for patients with PC. The findings might prove useful for therapeutic customization and the process of making medical decisions.