Huang Xiaoliang, Liu Jungang, Liu Haizhou, Mo Xianwei, Meng Yongsheng, Zhang Lihua, Deng Yuqing, Zhang Yawei, Tang Weizhong
Division of Colorectal & Anal Surgery, Department of Gastrointestinal Surgery, Guangxi Medical University Cancer Hospital, Nanning, China.
Guangxi Clinical Research Center for Colorectal Cancer, Nanning, China.
Front Oncol. 2021 Jan 15;10:595182. doi: 10.3389/fonc.2020.595182. eCollection 2020.
Epithelial mesenchymal transformation (EMT) and DNA repair status represent intrinsic features of colorectal cancer (CRC) and are associated with patient prognosis and treatment responsiveness. We sought to develop a combined EMT and DNA repair gene panel with potential application in patient classification and precise treatment.
We comprehensively evaluated the EMT and DNA repair patterns of 1,652 CRC patients from four datasets. Unsupervised clustering was used for classification. The clinical features, genetic mutation, tumor mutation load, and chemotherapy as well as immunotherapy sensitivity among different clusters were systematically compared. The least absolute shrinkage and selection operator regression method was used to develop the risk model.
Three distinct CRC clusters were determined. Clustet1 was characterized by down-regulated DNA repair pathways but active epithelial markers and metabolism pathway and had intermediate prognosis. Clustet2 was characterized by down-regulated both epithelial markers and DNA repair pathways and had poor outcome. Clustet3 presented with activation of DNA repair pathway and epithelial markers had favorable prognosis. Clustet1 might benefit form chemotherapy and Clustet3 had a higher response rate to immunotherapy. An EMT and DNA repair risk model related to prognosis and treatment response was developed.
This work developed and validated a combined EMT and DNA repair gene panel for CRC classification, which may be an effective tool for survival prediction and treatment guidance in CRC patients.
上皮间质转化(EMT)和DNA修复状态是结直肠癌(CRC)的内在特征,与患者预后和治疗反应性相关。我们试图开发一种结合EMT和DNA修复的基因panel,用于患者分类和精准治疗。
我们全面评估了来自四个数据集的1652例CRC患者的EMT和DNA修复模式。采用无监督聚类进行分类。系统比较了不同聚类间的临床特征、基因突变、肿瘤突变负荷以及化疗和免疫治疗敏感性。使用最小绝对收缩和选择算子回归方法建立风险模型。
确定了三个不同的CRC聚类。聚类1的特征是DNA修复途径下调,但上皮标志物和代谢途径活跃,预后中等。聚类2的特征是上皮标志物和DNA修复途径均下调,预后较差。聚类3表现为DNA修复途径激活且上皮标志物预后良好。聚类1可能从化疗中获益,聚类3对免疫治疗的反应率更高。建立了一个与预后和治疗反应相关的EMT和DNA修复风险模型。
本研究开发并验证了一种用于CRC分类的结合EMT和DNA修复的基因panel,这可能是预测CRC患者生存和指导治疗的有效工具。