Single-Cell Sequencing and Transcriptome Analysis Explored Changes in Midnolin-Related Immune Microenvironment and Constructed Combined Prognostic Model for Pancreatic Cancer

Background: Pancreatic cancer has one of the worst prognoses of any malignant tumor. The value of MIDN, midnolin-related genes and midnolin-related immune infiltrating cells (MICs) in the prognosis of pancreatic cancer remains unknown. Methods: Single-cell analysis were used to identify midnolin-related genes. Immune cell infiltration was obtained using CIBERSORT. The prognostic midnolin-related genes were identified through the utilization of Cox regression and the least absolute selection operator (LASSO) approach. The combined prognostic model was created using multifactorial Cox regression analysis. Survival analyses, immune microenvironment assessments, drug sensitivity checks were performed to evaluate the combined model performance. Finally, cellular experiments were carried out to confirm MIDN significance in pancreatic cancer. Results: The combined model was constructed based on MIDN expression, prognostic model of 10 midnolin-related genes and M1 cell infiltration. Most immune checkpoint-related genes were expressed at greater levels in the low-risk group, suggesting a greater chance of immunotherapy's benefits. The most significant model gene, MIDN, was shown to have a function by cellular tests. In pancreatic cancer, MIDN knockdown drastically decreased pancreatic cancer cell lines' activity, proliferation, and invasive potential. Conclusion: The combined model helped assess the prognosis of pancreatic cancer and offered fresh perspectives on immunotherapy in particular.