基于线粒体代谢相关基因构建和验证肺腺癌预后模型

Construction and validation of a prognostic model for predicting lung adenocarcinoma based on mitochondrial metabolism relat-edgenes

目的基于线粒体代谢相关基因(MMRGs)构建和验证肺腺癌(LUAD)预后模型,并探讨其临床应用价值。方法从癌症基因组图谱(TCGA)数据库中获取LUAD数据集,并按7∶3的比例随机分为训练集和验证集。首先获取LUAD特异性表达的MMRGs,然后在训练集中通过Cox回归分析、最小绝对收缩和选择算子回归筛选预后相关MMRGs,构建预后模型,计算风险评分,并以其中位值将患者分为低风险组和高风险组。在TCGA训练集、验证集、全集和GSE30219、GSE41271和GSE68465等3个测试集中分别绘制Kaplan-Meier曲线、ROC曲线,观察模型的稳健程度,通过Cox回归测试预后模型的预后性能;通过构建列线图模型进一步探讨预后模型与相关临床参数在预后监测中的作用,并探索预后模型与患者临床特征、驱动基因突变、免疫浸润等相关性。结果本研究构建了由4个MMRGs组成的预后模型。生存分析显示高风险组患者的预后明显劣于低风险组,风险评分可作为LUAD患者预后的独立危险因素;联合患者年龄、T分期、N分期和风险评分的列线图能更准确地预测LUAD患者的预后。患者的风险评分随T分期、N分期及TNM分期等增加而增加,且吸烟、鼠类肉瘤病毒癌基因(KRAS)突变型患者的风险评分高于不吸烟、KRAS野生型患者,而表皮生长因子受体(EGFR)突变型患者的风险评分低于EGFR野生型患者。高风险组患者中活化的肥大细胞、活化的CD4记忆T细胞、中性粒细胞、未活化的自然杀伤细胞、浆细胞、M1型巨噬细胞、M0型巨噬细胞浸润程度高。结论基于MMRGs构建的预后模型,对预测LUAD患者的预后评估具有较好的准确性和稳定性,且与患者的恶性表型、驱动基因突变和免疫浸润密切相关,可为LUAD的治疗及预后评估提供潜在依据。

Objective To construct and validate a prognostic model of lung adenocarcinoma(LUAD)based on mitochondrial metabolism related-genes(MMRGs)and to explore its clinical significance.Methods The LUAD cohort was obtained from The Cancer Genome Atlas(TCGA)and randomly divided into training and validation sets by 7∶3.The MMRGs specifically expressed in LUAD were obtained,and then the prognosis-related MMRGs were screened by Cox and LASSO regression in the training set to construct the prognostic model.The risk scores were calculated,and the patients were categorized into low-risk and high-risk groups by their median values.Kaplan-Meier curves and ROC were plotted in the training set,validation set,entire set of TCGA and three test sets,including GSE30219,GSE41271 and GSE68465,respectively to observe the robustness of the model.Cox regression analysis was applied to observe the performance and robustness of the prognostic model.A nomogram model was constructed to further explore the prognostic efficacy combining risk score and other independent prognostic factors.The correlation between prognostic models and patients'clinical features,driver gene mutations,and immune infiltration was explored.Results A prognostic model consisting of four MMRGs was constructed.Survival analysis showed that the prognosis of patients in the high-risk group was significantly worse than that of patients in the low-risk group,and multivariate Cox regression analysis indicated that risk score could be an independent factor for the prognosis of patients with LUAD.The nomogram involved age,T stage,N stage and risk score was able to more accurately predict the prognosis of patients.Risk score of patients increased with increasing T stage,N stage and TNM stage,which was higher in smoking or kirsten rats arcomaviral oncogene homolog(KRAS)mutation patients compared with that of non-smoking and KRAS wild type patients,but lower in epidermal growth factor receptor(EGFR)mutation patients compared with that of EGFR wild type patients.The infiltration degree of activated mast cells,activated CD4 memory T cells,neutrophils,unactivated nature killer cells,plasma cells,M1 macrophages and M0 macrophages was high among patients in the high-risk group.Conclusion The prognostic model constructed based on MMRGs has good accuracy and stability in predicting the prognosis of patients with LUAD,and is closely related to the malignant phenotypes of the patients,the mutations of the driver genes mutations,and immune infiltration,which could potentially provide a basis for the treatment of LUAD and the assessment of prognosis.