Abstract
BACKGROUND: We aimed to identify stromal cells associated with liver metastasis of gastric cancer (GC). METHODS: We downloaded single-cell RNA sequencing (scRNA-seq) datasets for GC tissues and adjacent normal tissues (NT), as well as liver metastases (LM) and healthy liver tissues (HL) from GEO. Additionally, we obtained RNA-Seq and clinical data from TCGA for the TCGA-STAD project. The datasets were analyzed using R software. Cell viability analysis were used to verify the potential drugs. RESULTS: We obtained 7 GC, 2 NT, 5 LM and 3 HL datasets, as well as 448 RNA-Seq files. We identified 13 cell types from a total of 107,875 cells, which were further re-clustered into 38 subclusters. Our analyses showed that both GC and LM exhibited increased angiogenesis, cancer-associated fibroblasts (CAFs) were enriched in primary tumors, CD8(+) T cells in both GC and LM showed a declined proportion and increased necroptosis, NK cells were reduced in LM, and M2 macrophages (Mφ) exhibited high activities in both GC and LM. We found immune cells in HL and LM expressed more transcripts, suggesting the differences in tumor microenvironment (TME) between primary and metastatic tumors. Additionally, we identified 3 potential drugs that could be effective against both primary and metastatic tumors. CONCLUSION: This study illustrated the heterogeneity of immune and stromal cells in TME of GC and LM, and identified several cell types associated with liver metastasis of GC. These cell types may serve as potential biomarkers or therapeutic targets in the future.