| 摘要: |
| 目的 基于生物信息学方法构建并验证肝细胞癌（HCC）对索拉非尼敏感性相关基因的预后模型，探究该模型对HCC患者预后和索拉非尼敏感性的预测能力。方法 对基因表达汇编数据库GSE109211数据集、癌症基因组图谱（TCGA）数据库LIHC队列、国际肿瘤基因组协作组（ICGC）数据库LIRI队列进行差异基因分析，通过交集筛选出HCC索拉非尼敏感性相关基因，进行京都基因与基因组百科全书（KEGG）信号通路富集分析。利用单因素Cox回归与最小绝对收缩和选择算子（LASSO）回归构建预后模型，依据风险评分中位值将患者分为高、低风险组。通过Kaplan-Meier法和多因素Cox回归分析进行生存分析。通过癌症药物敏感性基因组学（GDSC）数据库分析索拉非尼的IC50并探索其与风险评分的关系。结果 筛选出365个HCC索拉非尼敏感性相关基因，KEGG富集分析显示存在与药物代谢相关的信号通路。通过单因素Cox回归分析获得221个与HCC预后相关的基因，通过LASSO回归构建了一个包括含TCP1伴侣蛋白亚单位3（CCT3）、红细胞生成素（erythropoietin，EPO）、甲酰转移酶环脱氨酶（FTCD）、葡萄糖-6-磷酸脱氢酶（G6PD）、驱动蛋白家族成员20A（KIF20A）、磷脂酰肌醇聚糖锚定生物合成U类基因（PIGU）、分泌型磷蛋白1（SPP1）7个关键基因的预后风险评分模型：风险评分＝ CCT3×0.032＋EPO×0.055＋FTCD×（－0.026）＋G6PD×0.083＋KIF20A×0.039＋PIGU×0.144＋SPP1×0.009。验证结果显示高风险组生存时间短于低风险组（P＜0.001），多因素Cox回归分析显示风险评分是独立预后因素。高风险组的索拉非尼IC50比低风险组低，提示高风险组对索拉非尼的治疗可能更敏感。结论 基于索拉非尼敏感性相关基因构建的HCC预后模型具有良好的预测价值。 |
| 关键词: 肝细胞癌 索拉非尼 药物敏感性 生物信息学 预后模型 |
| DOI：10.16781/j.CN31-2187/R.20230337 |
| 投稿时间：2023-06-15修订日期：2023-07-07 |
| 基金项目:国家自然科学基金面上项目（82072725），江苏省卫生健康委员会重点项目（ZD2021039）. |
|
| Construction of a prognostic prediction model for hepatocellular carcinoma based on sorafenib sensitivity-related genes and its clinical significance |
| JI Linlin1,HU Dingtao2,GAO Peng2,DING Jin2,CHU Xiaoyuan1* |
(1. Department of Oncology, Jinling Hospital, Medical School of Nanjing University, Nanjing 210000, Jiangsu, China;
2. Clinical Cancer Institute, Center of Translational Medicine, Naval Medical University(Second Military Medical University), Shanghai 200433, China
*Corresponding author) |
| Abstract: |
| Objective To construct and validate a prognostic model for sorafenib sensitivity-related genes in hepatocellular carcinoma (HCC) based on bioinformatics methods, and to explore the predictive ability of the model for prognosis and the efficacy for sorafenib treatment in HCC. Methods Differential gene analysis was performed on GSE109211 data set of Gene Expression Omnibus, LIHC cohort of The Cancer Genome Atlas (TCGA) and LIRI cohort of International Cancer Genome Consortium (ICGC). Sorafenib sensitivity-related genes in HCC were screened by intersection, and Kyoto Encyclopedia of Genes and Genomes (KEGG) signal pathway enrichment analysis was performed. A prognostic model was constructed using univariate Cox regression and least absolute shrinkage and selection operator (LASSO) regression, and the patients were divided into high-risk and low-risk groups based on the median risk score. Survival analysis was conducted using Kaplan-Meier method and multivariate Cox regression analysis. The half inhibitory concentration (IC50) of sorafenib was analyzed through the Genomics of Drug Sensitivity in Cancer (GDSC) and its relationship with risk score was explored. Results A total of 365 sorafenib sensitivity-related genes in HCC were screened, and the KEGG enrichment analysis revealed the presence of pathways associated with drug metabolism. Univariate Cox analysis identified 221 genes associated with prognosis, and a prognostic model containing 7 key genes (chaperonin containing TCP1 subunit 3［CCT3］, erythropoietin［EPO］, formimidoyltransferase cyclodeaminase［FTCD］, glucose-6-phosphate dehydrogenase［G6PD］, kinesin family member 20A［KIF20A］, phosphatidylinositol glycan anchor biosynthesis class U［PIGU］, and secreted phosphoprotein 1 ［SPP1］) was constructed by LASSO regression. Risk score＝CCT3×0.032＋EPO×0.055＋FTCD×(－0.026）＋G6PD×0.083＋KIF20A×0.039＋PIGU×0.144＋SPP1×0.009. The results showed that the high-risk group had a shorter survival time compared to the low-risk group (P＜0.001). Multivariate Cox analysis demonstrated that risk score was an independent prognostic factor. The IC50 of sorafenib in the high-risk group was lower than that in the low-risk group, suggesting that the high-risk group may be more sensitive to sorafenib. Conclusion The constructed prognostic model for HCC based on sorafenib sensitivity-related genes has good prognostic value for HCC. |
| Key words: hepatocellular carcinoma sorafenib drug susceptibility bioinformatics prognostic model |
