Department of General Surgery, Shanghai General Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China.
Institute Pasteur of Shanghai, Chinese Academy of Sciences, Shanghai, China.
Cancer Med. 2023 Jan;12(2):2075-2088. doi: 10.1002/cam4.4975. Epub 2022 Jul 4.
Microsatellite instability-high (MSI-H) subgroup of gastric cancer (GC) is characterized by a high tumor mutational burden, increased lymphocytic infiltration, and enhanced inflammatory cytokines. GC patients with MSI-H status have a good response to immune checkpoint blockade management. However, heterogeneity within the subtype and the underlying mechanisms of shaping tumor microenvironments remain poorly understood.
RNA expression levels and clinical parameters of GC were obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. The data were analyzed using single-sample Gene Set Enrichment Analysis (ssGSEA), univariate Cox regression, multivariate Cox regression, and Least Absolute Shrinkage Selection Operator (LASSO) regression. In addition, multiplex immunohistochemistry (mIHC) was used in our clinical cohort for the tumor microenvironment study.
By ssGSEA and survival analysis, the EMT signaling pathway was identified as a representative pathway, which can stratify the patients with MSI-H GC with significant survival predictive power. Then, a novel representative EMT-related five-gene signature (namely CALU, PCOLCE2, PLOD2, SGCD, and THBS2) was established from EMT signaling gene set, which sensitivity and specificity were further validated in the independent GEO database (GSE62254) cohort for disease outcome prediction. Based on public single-cell data and in situ immunohistochemistry, we found that most of these five genes were abundantly expressed in cancer-associated fibroblasts. Furthermore, patients with high or low risk divided by this five-gene signature exhibited a strong correlation of the distinct patterns of tumor immune microenvironment. By mIHC staining of sections from 30 patients with MSI-H status, we showed that the patients with better prognoses had the increased infiltration of CD8 cells in the primary tumoral tissue.
Our study developed a simple five-gene signature for stratifying MSI-H GC patients with survival predictive power.
微卫星不稳定性高（MSI-H）亚组胃癌（GC）的特征是肿瘤突变负担高、淋巴细胞浸润增加和炎症细胞因子增强。MSI-H 状态的 GC 患者对免疫检查点阻断管理有良好的反应。然而，亚组内的异质性和塑造肿瘤微环境的潜在机制仍知之甚少。
从癌症基因组图谱（TCGA）和基因表达综合（GEO）数据库中获取 GC 的 RNA 表达水平和临床参数。使用单样本基因集富集分析（ssGSEA）、单因素 Cox 回归、多因素 Cox 回归和最小绝对收缩选择算子（LASSO）回归分析数据。此外，我们的临床队列还使用了多重免疫组化（mIHC）进行肿瘤微环境研究。
通过 ssGSEA 和生存分析，鉴定出 EMT 信号通路是一个有代表性的通路，它可以对 MSI-H GC 患者进行分层，具有显著的生存预测能力。然后，从 EMT 信号基因集中建立了一个新的 EMT 相关的五个基因特征（即 CALU、PCOLCE2、PLOD2、SGCD 和 THBS2），在独立的 GEO 数据库（GSE62254）队列中进一步验证了其对疾病结局预测的敏感性和特异性。基于公共单细胞数据和原位免疫组化，我们发现这五个基因中的大多数在癌相关成纤维细胞中大量表达。此外，通过这个五个基因特征划分的高危和低危患者，其肿瘤免疫微环境的模式具有很强的相关性。通过对 30 例 MSI-H 状态患者的组织切片进行 mIHC 染色，我们发现预后较好的患者在原发肿瘤组织中 CD8 细胞浸润增加。
我们的研究开发了一个简单的五个基因特征，用于分层具有生存预测能力的 MSI-H GC 患者。
