Gastric cancer (GC) is a highly fatal and common malignancy of the digestive system. Recent therapeutic advancements have significantly improved the clinical outcomes in GC, but due to the unavailability of suitable molecular targets, a large number of patients do not respond to the immune checkpoint inhibitors (ICI) therapy. To identify and validate potential therapeutic and prognostic targets of gastric cancer, we used the “inferCNV” R package for analyzing single-cell sequencing data (GSE112302) of GC and normal epithelial cells. First, by using LASSO, we screened genes that were highly correlated with copy number variations (CNVs). Therefrom, five gene signature (CPVL, DDC, GRTP1, ONECUT2, and PRSS21) was selected by cross-validating the prognosis and risk management with the GC RNA-seq data obtained from GEO and TCGA. Moreover, the correlation analyses between CNVs of these genes and immune cell infiltration in gastric cancer identified CPVL as a potential prognostic marker. Finally, CPVL showed high expression in gastric cancer samples and cell lines, then siRNA-mediated silencing of CPVL expression in gastric cancer cells showed significant proliferation arrest in MGC803 cells. Here, we conclude that CNVs are key regulators of the immune cells infiltration in gastric TME as well as cancer development, and CPVL could potentially be used as a prognostic and therapeutic marker in gastric cancer.

Colorectal cancer (COAD) is ranked as the third most common cancer and second in terms of cancer-related deaths worldwide. Due to its poor overall survival and prognosis, the incidents of COAD are significantly increasing. Although treatment methods have greatly been improved in the last decade, it is still not good enough to have satisfactory treatment outcomes. In recent years, immunotherapy has been successful to some extent in the treatment of many cancers but still, many patients do not respond to immunotherapy. Therefore, it is essential to have a deeper understanding of the immune characteristics of the tumor microenvironment and identify meaningful immune targets. In terms of immune targets, COAD has been poorly explored; thus, in the current study, based on the immune cell infiltration score and differentially expressed genes, COAD tumors were classified into hot and cold tumors. The Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis was used to identify hub genes, construct a prognostic model, and screen potential immune targets. In total, 12 genes (CLK3, CYSLTR2, GJA10, CYP4Z1, FAM185A, LINC00324, EEF1A1P34, EEF1B2P8, PTCSC3, MIR6780A, LINC01666, and RNU6.661P) differentially expressed between hot and cold tumors were screened out. Among them, CYSLTR2 was considered as a potential candidate gene, because it showed a significant positive correlation with immune cell infiltration and immune checkpoints (PDCD1, CD274, and CTLA4). Finally, we constructed and validated a new prognostic model for COAD showing 0.854 AUC for the ROC curve, and these results provide sufficient potential to choose CYSLTR2 as an important immune target for the prognosis of COAD.