Background: Pancreatic cancer is one of the most lethal types of cancer, and immunotherapy has become a promising remedy with advancements in tumor immunology. However, predicting the clinical response to immunotherapy in pancreatic cancer remains a dilemma for clinicians. Methods: GEPIA database was used to analyze the differential expression of MMR and PD-L1 genes in 33 common cancer types including pancreatic cancer. The expression levels of MMR and PD-L1 genes were downloaded from the GEPIA and GEO databases to analyze the correlation between MMR genes and PD-L1, and the clinicopathological and survival information were downloaded from the TCGA databases to analyze the relationship between the expression of MMR, PD-L1 and clinicopathological characteristics, prognosis. Meanwhile, the tumor tissue samples of 41 patients with pancreatic cancer were collected, and the protein expression levels of MMR and PD-L1 were detected by immunohistochemical assay. Furthermore, we analyzed the correlation between MMR and PD-L1, and the correlation between the expression of MMR, PD-L1 and clinicopathological characteristics, prognosis of pancreatic cancer patients. Results: Bioinformatics analysis showed that MLH1, MLH3, MSH2, MSH3, and PMS2 were highly expressed in most cancer types including pancreatic cancer (P < 0.05). TCGA data revealed that MLH1 expression was related to gender (P = 0.012), clinical stage (I vs II: P = 0.016), MSH2 expression was related to clinical stage (P < 0.05), T stage (T3 vs T4: P = 0.039), and MSH3 expression was related to T stage (P < 0.05). Besides, both MSH2 expression (P < 0.001) and MSH6 (P = 0.044