Construction of a Novel Disulfidptosis-Related lncRNA Prognostic Signature in Pancreatic Cancer.
Faliang XingYi QinJin XuWei WangBo ZhangPublished in: Molecular biotechnology (2023)
Pancreatic cancer is a lethal, extremely aggressive gastrointestinal tumor with a poor prognosis and limited treatment alternatives. Disulfidptosis is a newly defined type of cell death with potential influence on cancer. Research on the association between disulfidptosis and pancreatic cancer is scarce. The expression data of disulfidptosis-related genes were downloaded from The Cancer Genome Atlas-Pancreatic Adenocarcinoma (TCGA). Disulfidptosis-related lncRNA signature (DRLS) was developed through the Cox and the least absolute shrinkage and selection operator (LASSO) analysis. Differences in enrichment functions, mutational landscape, immune microenvironment, and predicted therapeutic efficacy between high- and low-risk groups were assessed. Consensus clustering analysis was applied to identify the DRLS-related subtypes. Among 98 disulfidptosis-related lncRNAs, 5 lncRNAs were screened thus constructing a prognostic DRLS. DRLS showed high predictive accuracy and was an independent prognostic factor for pancreatic cancer. According to the risk scores calculated from the signature, samples were categorized into high- and low- risk groups. Overall, low-risk patients had a better prognosis, lower mutational occurrences, higher immune cell infiltration and more sensitivity to anti-tumor agents. The DRLS performed well in predicting prognosis and revealed intimate correlation with biological function, mutation status and immune infiltration landscape of pancreatic cancer, providing some insights for future research on the relationship between disulfidptosis and pancreatic cancer.
Keyphrases
- poor prognosis
- prognostic factors
- long non coding rna
- single cell
- cell death
- papillary thyroid
- ejection fraction
- long noncoding rna
- newly diagnosed
- chronic kidney disease
- machine learning
- young adults
- dna methylation
- deep learning
- risk assessment
- signaling pathway
- electronic health record
- big data
- genome wide
- network analysis
- clinical practice
- childhood cancer
- cell cycle arrest