摘要
<strong>Background:</strong> In the present study, autophagy-related long non-coding RNAs (lncRNAs) in lung adenocarcinoma (LUAD) were screened for diagnosis and prognosis, and the molecular mechanisms of LUAD at the genetic level were investigated. <strong>Methods:</strong> From The Cancer Genome Atlas (TCGA) database, 497 gene expression data and 436 clinical data of LUAD cases were collected for analysis. In addition, 232 autophagy-related genes (ARGs) were extracted from the Human Autophagy Database (HADb). Spearman rank correlation test and the Akaike information criterion (AIC) were performed to screen the data. After filtering, a survival model including three autophagy-related lncRNAs was generated. Based on the following formula: risk score = ΣCoef gene i×Gene i expression, the risk score of all LUAD patients could be calculated. LUAD patients were divided into two groups based on risk score for survival curve using Kaplan-Meier survival analysis. Both univariate and multivariate survival analyses were used to determine whether the three lncRNAs were independent prognostic factors using the survival package in R. Furthermore, the receiver operating characteristic (ROC) curves of clinical data were created to assess the stability of the survival model. Finally, the Gene Set Enrichment Analysis (GSEA) was used for analysis of related pathways. <strong>Results:</strong> A prognostic model consisting of three lncRNAs (AC011477.2, AC099850.3, and TRG-AS1) was generated for analysis. The 5-year survival rate in the high-risk group was 26.51% (95% CI: 0.1842 - 0.382), which was statistically lower than in the low-risk group 41.6% (95% CI: 0.307 - 0.563, P < 0.05). The area under the ROC curve (AUC) of risk score was 0.700, indicating a higher diagnostic accuracy of risk score. The results of GSEA showed enrichment in 36 pathways, including pyrimidine metabolism, pentose phosphate pathway, citric acid cycle, and cell cycle in the high-risk group, and FC-EPSILON-RI signal pathway, intestinal immune network produced by IgA, and ABC transporters in the low-risk group. <strong>Conclusion:</strong> The prognosis model composed of autophagy-related lncRNAs, AC011477.2, AC099850.3, and TRG-AS1, in LUAD can be used to predict the prognosis of LUAD patients and is expected to improve clinical treatment.
<strong>Background:</strong> In the present study, autophagy-related long non-coding RNAs (lncRNAs) in lung adenocarcinoma (LUAD) were screened for diagnosis and prognosis, and the molecular mechanisms of LUAD at the genetic level were investigated. <strong>Methods:</strong> From The Cancer Genome Atlas (TCGA) database, 497 gene expression data and 436 clinical data of LUAD cases were collected for analysis. In addition, 232 autophagy-related genes (ARGs) were extracted from the Human Autophagy Database (HADb). Spearman rank correlation test and the Akaike information criterion (AIC) were performed to screen the data. After filtering, a survival model including three autophagy-related lncRNAs was generated. Based on the following formula: risk score = ΣCoef gene i×Gene i expression, the risk score of all LUAD patients could be calculated. LUAD patients were divided into two groups based on risk score for survival curve using Kaplan-Meier survival analysis. Both univariate and multivariate survival analyses were used to determine whether the three lncRNAs were independent prognostic factors using the survival package in R. Furthermore, the receiver operating characteristic (ROC) curves of clinical data were created to assess the stability of the survival model. Finally, the Gene Set Enrichment Analysis (GSEA) was used for analysis of related pathways. <strong>Results:</strong> A prognostic model consisting of three lncRNAs (AC011477.2, AC099850.3, and TRG-AS1) was generated for analysis. The 5-year survival rate in the high-risk group was 26.51% (95% CI: 0.1842 - 0.382), which was statistically lower than in the low-risk group 41.6% (95% CI: 0.307 - 0.563, P < 0.05). The area under the ROC curve (AUC) of risk score was 0.700, indicating a higher diagnostic accuracy of risk score. The results of GSEA showed enrichment in 36 pathways, including pyrimidine metabolism, pentose phosphate pathway, citric acid cycle, and cell cycle in the high-risk group, and FC-EPSILON-RI signal pathway, intestinal immune network produced by IgA, and ABC transporters in the low-risk group. <strong>Conclusion:</strong> The prognosis model composed of autophagy-related lncRNAs, AC011477.2, AC099850.3, and TRG-AS1, in LUAD can be used to predict the prognosis of LUAD patients and is expected to improve clinical treatment.
作者
Zhi-Peng Miao
Lei Liu
Dong-Juan Wang
Cai-Ling Jiang
Cui-Min Zhu
Hua-Chuan Zheng
Li Zhang
Zhi-Peng Miao;Lei Liu;Dong-Juan Wang;Cai-Ling Jiang;Cui-Min Zhu;Hua-Chuan Zheng;Li Zhang(Department of Oncology, The Affiliated Hospital of Chengde Medical University, Chengde, China;College of Basic Medicine, Chengde Medical University, Chengde, China;Department of Oncology and Experimental Center, The Affiliated Hospital of Chengde Medical University, Chengde, China)
