Department of Thoracic Surgery, 92323Zhongshan Hospital, Fudan University, Shanghai, China.
Technol Cancer Res Treat. 2022 Jan-Dec;21:15330338221117003. doi: 10.1177/15330338221117003.
The role of cancer stem cells in esophageal squamous cell carcinoma (ESCC) remains unclear. The mRNA stemness index (mRNAsi) of 179 ESCC patients (GSE53625) was calculated using a machine learning algorithm based on their mRNA expression. Stemness-related genes were identified by weighted correlation network analysis (WGCNA) and LASSO regression, whose associations with mutation status, immune cell infiltrations, and potential compounds were also analyzed. The role of these genes in proliferation and their expressions was assessed in ESCC cell lines and 112 samples from our center. The ESCC samples had significantly higher mRNAsi than the normal tissues. Patients with high mRNAsi exhibited higher worse OS. Seven stemness-related genes were identified by WGCNA and LASSO regression, based on which a risk-predicted score model was constructed. Among them, CST1, CILP, PITX2, F2RL2, and RIOX1 were favorable for OS, which were adverse for DPP4 and ZFHX4 in the GSE53625 dataset. However, RIOX1 was unfavorable for OS in patients from our center. In vitro assays showed that CST1, CILP, PITX2, F2RL2, and RIOX1 were pro-proliferated, which were opposite for DDP4 and ZFHX4. In addition, SMARCA4, NOTCH3, DNAH5, and KALRN were more mutated in the low-score group. The low-score group had significantly more memory B cells, monocytes, activated NK cells, and Tregs and less macrophages M2, resting mast cells, and resting dendritic cells. Seven stemness-related genes are significantly related to the prognosis, gene mutations, and immune cell infiltration of ESCC. Some potential anticancer compounds may be favorable for OS.
癌症干细胞在食管鳞状细胞癌（ESCC）中的作用尚不清楚。我们使用基于 mRNA 表达的机器学习算法计算了 179 名 ESCC 患者（GSE53625）的 mRNA 干性指数（mRNAsi）。通过加权相关网络分析（WGCNA）和 LASSO 回归确定与干性相关的基因，分析它们与突变状态、免疫细胞浸润和潜在化合物的关联。在 ESCC 细胞系和我们中心的 112 个样本中评估了这些基因在增殖中的作用及其表达。ESCC 样本的 mRNAsi 明显高于正常组织。mRNAsi 高的患者 OS 更差。通过 WGCNA 和 LASSO 回归鉴定了 7 个与干性相关的基因，基于这些基因构建了风险预测评分模型。其中 CST1、CILP、PITX2、F2RL2 和 RIOX1 有利于 OS，而在 GSE53625 数据集中，DPP4 和 ZFHX4 则不利于 OS。然而，在我们中心的患者中，RIOX1 不利于 OS。体外实验表明，CST1、CILP、PITX2、F2RL2 和 RIOX1 促进增殖，而 DDP4 和 ZFHX4 则相反。此外，SMARCA4、NOTCH3、DNAH5 和 KALRN 在低评分组中更多突变。低评分组的记忆 B 细胞、单核细胞、活化 NK 细胞和 Tregs 明显较多，而巨噬细胞 M2、静止肥大细胞和静止树突状细胞明显较少。7 个与干性相关的基因与 ESCC 的预后、基因突变和免疫细胞浸润显著相关。一些潜在的抗癌化合物可能有利于 OS。
