Machine learning-based signature of necroptosis-associated ln-cRNAs for prognostic and immunotherapy response prediction in pancreatic cancer
Jun Yang1
Aijuan Sheng1
Min Jia1
Haibin Yu1
Ran Xue2
1.Clinical Trial Institution Management Office,Beijing YouAn Hospital,Capital Medical University,No.8 Xi Tou Tiao,YouAn Men Wai Street,Fengtai District,Beijing,100069 China2.Department of Gastrointestinal Oncology,Key Laboratory of Carcinogenesis and Translational Research(Ministry of Education),Peking University Cancer Hospital and Institute,Fucheng Road 52,Haidian District,Beijing,100142 China
摘要：Objectives Pancreatic cancer is a leading cause of cancer-related deaths worldwide.Long non-coding RNAs(ln-cRNAs)have been identified as oncogenes,facilitating the development and progression of pancreatic cancer.Necro-ptosis is a novel form of programmed cell death that is involved in anti-tumor immunity and tumor prognosis.Until now,the clinical significance of necroptosis-associated lncRNAs in pancreatic cancer remains largely unexplored.
Methods Gene expression and clinical data were obtained from The Cancer Genome Atlas(TCGA)database.To identify lncRNAs,we conducted co-expression analysis using immune genes from the database.The risk model was constructed by employing univariate and multivariate Cox regression analyses,as well as Lasso penalized regression analysis.The patients were then divided into high-risk and low-risk groups.Subsequently,we assessed our signature across diverse clinical settings,encompassing clinicopathological characteristics,tumor-infiltrating immune cells,and checkpoint-related biomarkers.Prognostic prediction was achieved by integrating differentially expressed lncRNA signatures associated with necroptosis.We constructed a highly predictive nomogram by com-bining the necroptosis-related lncRNA signature with clinical features.
Results We generated lncRNA signatures by evaluating the expression variations of different lncRNAs.The area under the receiver operating characteristic curve value for the 5-year survival rate,which indicates the predictive ability of the signature,was 0.918.Further analysis demonstrated that our signature can effectively differentiate unfavorable survival outcomes,combine prognostic clinicopathological characteristics,and accurately determine tumor infiltration status.Significant correlations were observed between the low-risk group and high expression levels of immune checkpoint-related genes.
Conclusions An innovative necroptosis-associated lncRNA signature for pancreatic cancer was developed using the TCGA database,which demonstrated promising prognostic value.Our nomogram could serve as a novel tool to improve pancreatic clinical outcomes of pancreatic cancer.
机标关键词：necroptosispredictionsignatureimmunotherapycancerassociatedlearning-basedln-crnas
论文发表日期：2024-06-30
在线出版日期：2025-08-15（本平台首次上网日期，不代表文献的发表时间）
页数：17( 21-37 )
英文信息