Liu Jinyang, Lan Yu, Tian Geng, Yang Jialiang
Department of Sciences, Geneis Beijing Co., Ltd., Beijing, China.
Department of Data Mining,Qingdao Geneis Institute of Big Data Mining and Precision Medicine, Qingdao, China.
Front Oncol. 2022 May 19;12:899156. doi: 10.3389/fonc.2022.899156. eCollection 2022.
As one of the most common cancers of the digestive system, colon cancer is a predominant cause of cancer-related deaths worldwide. To investigate prognostic genes in the tumor microenvironment of colon cancer, we collected 461 colon adenocarcinoma (COAD) and 172 rectal adenocarcinoma (READ) samples from The Cancer Genome Atlas (TCGA) database, and calculated the stromal and immune scores of each sample. We demonstrated that stromal and immune scores were significantly associated with colon cancer stages. By analyzing differentially expressed genes (DEGs) between two stromal and immune score groups, we identified 952 common DEGs. The significantly enriched Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) terms for these DEGs were associated with T-cell activation, immune receptor activity, and cytokine-cytokine receptor interaction. Through univariate Cox regression analysis, we identified 22 prognostic genes. Furthermore, nine key prognostic genes, namely, , , 2, , , , , , and , were identified using the LASSO Cox regression analysis. The risk score of each sample was calculated using the gene expression of the nine genes. Patients with high-risk scores had a poorer prognosis than those with low-risk scores. The prognostic model established with the nine-gene signature was able to effectively predict the outcome of colon cancer patients. Our findings may help in the clinical decisions and improve the prognosis for colon cancer.
作为消化系统最常见的癌症之一，结肠癌是全球癌症相关死亡的主要原因。为了研究结肠癌肿瘤微环境中的预后基因，我们从癌症基因组图谱（TCGA）数据库中收集了461例结肠腺癌（COAD）和172例直肠腺癌（READ）样本，并计算了每个样本的基质和免疫评分。我们证明基质和免疫评分与结肠癌分期显著相关。通过分析两个基质和免疫评分组之间的差异表达基因（DEG），我们鉴定出952个常见的DEG。这些DEG显著富集的基因本体（GO）和京都基因与基因组百科全书（KEGG）术语与T细胞活化、免疫受体活性和细胞因子-细胞因子受体相互作用有关。通过单变量Cox回归分析，我们鉴定出22个预后基因。此外，使用LASSO Cox回归分析鉴定出9个关键预后基因，即 、 、2、 、 、 、 、 和 。使用这9个基因的基因表达计算每个样本的风险评分。高风险评分的患者预后比低风险评分的患者差。用九基因特征建立的预后模型能够有效预测结肠癌患者的预后。我们的研究结果可能有助于临床决策并改善结肠癌的预后。
