Department of Urology, The Affiliated Hospital of Southwest Medical University, Luzhou, Sichuan 646000, China.
Sichuan Clinical Research Center for Nephropathy, Luzhou, Sichuan 646000, China.
Biomed Res Int. 2021 Mar 12;2021:9656947. doi: 10.1155/2021/9656947. eCollection 2021.
Gastric cancer (GC) is believed to be one of the most common digestive tract malignant tumors. The prognosis of GC remains poor due to its high malignancy, high incidence of metastasis and relapse, and lack of effective treatment. The constant progress in bioinformatics and molecular biology techniques has given rise to the discovery of biomarkers with clinical value to predict the GC patients' prognosis. However, the use of a single gene biomarker can hardly achieve the satisfactory specificity and sensitivity. Therefore, it is urgent to identify novel genetic markers to forecast the prognosis of patients with GC.
In our research, data mining was applied to perform expression profile analysis of mRNAs in the 443 GC patients from The Cancer Genome Atlas (TCGA) cohort. Genes associated with the overall survival (OS) of GC were identified using univariate analysis. The prognostic predictive value of the risk factors was determined using the Kaplan-Meier survival analysis and multivariate analysis. The risk scoring system was built in TCGA dataset and validated in an independent Gene Expression Omnibus (GEO) dataset comprising 300 GC patients. Based on the median of the risk score, GC patients were grouped into high-risk and low-risk groups.
We identified four genes (, , , and ) that were significantly correlated with GC patients' OS. The high-risk group showed poor prognosis, indicating that the risk score was an effective predictor for the prognosis of GC patients.
The signature consisting of four glycolysis-related genes could be used to forecast the GC patients' prognosis.
胃癌（GC）被认为是最常见的消化道恶性肿瘤之一。由于其高度恶性、高转移和复发率以及缺乏有效治疗，GC 的预后仍然很差。生物信息学和分子生物学技术的不断进步，促使人们发现了具有临床价值的生物标志物，以预测 GC 患者的预后。然而，单一基因生物标志物的使用很难达到令人满意的特异性和敏感性。因此，迫切需要识别新的遗传标记物来预测 GC 患者的预后。
在我们的研究中，应用数据挖掘对来自癌症基因组图谱（TCGA）队列的 443 名 GC 患者的 mRNA 表达谱进行分析。使用单因素分析确定与 GC 患者总生存期（OS）相关的基因。通过 Kaplan-Meier 生存分析和多因素分析确定风险因素的预后预测价值。在 TCGA 数据集构建风险评分系统，并在包含 300 名 GC 患者的独立基因表达综合数据集（GEO）中进行验证。根据风险评分的中位数，将 GC 患者分为高风险组和低风险组。
我们确定了四个与 GC 患者 OS 显著相关的基因（、、、和）。高风险组预后不良，表明风险评分是 GC 患者预后的有效预测因子。
由四个糖酵解相关基因组成的特征可用于预测 GC 患者的预后。
