ZHONGHUA YANGSHENG BAOJIAN ›› 2024, Vol. 42 ›› Issue (7): 4-9.

Previous Articles     Next Articles

Model of Pancreatic Cancer Prognosis Based on LncRNA Linked to Hypoxia

ZHENG Yan1,2,*, HAN Li-hong1,2, YAN Bin1,2, LIN Cui-ying1,2   

  1. 1. School of Basic Medicine, Putian University, Putian Fujian 351100, China;
    2. Key Laboratory of Translational Tumor Medicine in Fujian Province,Putian University, Putian Fujian 351100, China
  • Online:2024-04-01 Published:2024-03-20

Abstract: Objective The goal was to use hypoxia-associated LncRNAs to predict the prognosis of pancreatic cancer patients. Methods The TCGA database provided all of the transcriptome data and clinical information linked to pancreatic cancer. GSEA was used to identify hypoxia-associated genes, and co-expression was used to identify hypoxia-related LncRNAs.For the purpose of creating the prognostic model, prognostic-related hypoxic LncRNAs were filtered out using one-way and multifactorial Cox analyses. The predictive significance of hypoxia-associated LncRNA characteristics in pancreatic cancer was examined over time using survival curves and ROC curves. Ultimately, the expression of the distinctive LncRNAs was used to classify pancreatic cancer tumors, and the research confirmed that there was a strong correlation between the expression of these LncRNAs and pancreatic cancer. In order to identify the important hypoxia-associated LncRNAs with substantial prognostic value and research significance, the lasso regression model was further examined. Results A total of 29 of the differentially expressed hypoxia-associated LncRNAs were found to be potentially associated with prognosis by one-way Cox regression analysis. These 29 LncRNAs were then optimized by multifactorial Cox analysis to yield 6 distinctive LncRNAs influencing prognosis. The prognostic model survival curves created by the 6 LncRNAs also demonstrated that the low-risk group's prognosis was better than the high-risk group's, with a statistically significant difference between the two groups. statistically noteworthy. The prognostic model could be utilized as an independent prognostic factor, according to the independent prognostic analysis. Ultimately, four hypoxia-related LncRNAs with considerable predictive significance were obtained by further filtering and censoring the lasso regression model. Conclusion With the use of four hypoxia-related LncRNAs, we were able to effectively build a predictive model for pancreatic cancer in this work. This model is crucial for enhancing prognosis prediction and directing patients' individualized treatment. The four essential distinctive LncRNAs in this model could be unique markers for predicting pancreatic cancer prognosis.

Key words: pancreatic cancer, hypoxia related genes, LncRNA, prognosis

CLC Number: