Abstract
BACKGROUND: Lung cancer is a highly aggressive and prevalent disease worldwide. By the time it is first diagnosed, distant metastases have usually already occurred. Among them, the prognosis of patients with brain metastasis from lung cancer is very poor. Therefore, it is particularly important to identify the evolutionary status of tumor cells during lung cancer brain metastases and discover the underlying mechanisms of lung cancer brain metastases. METHODS: In this study, we analysed three types of data: single-cell RNA sequencing, bulk RNA sequencing, and spatial transcriptome. Firstly, we identified early metastatic epithelial cell clusters (EMEC) using CNV and trajectory analysis in scRNA-seq data. Secondly, we integrated scRNA-seq and spatial transcriptome data with the help of MIA (Multimodal intersection analysis) to explore the biological characteristics of EMEC. Finally, we used bulk RNA-seq data to validate the molecular characteristics of EMEC. RESULT: A total of 55,763 single cells were obtained and divided into 9 cell types. In brain metastasis, we found a significantly higher proportion of epithelial cells. In addition, we identified a specific subpopulation of epithelial cells, which was named as “early metastatic epithelial cell clusters (EMEC)”. It is enriched in oxidative phosphorylation, coagulation, complement. Moreover, we also found that EMEC underwent cellular communication with other immune cells through ligand-receptor pairs such as MIF-(CD74 + CXCR4) and MIF-(CD74 + CD44). Next, we validated that EMEC were associated with poor clinical prognosis using three independent external datasets. Finally, spatial transcriptome analysis revealed specificity in the spatial distribution of EMEC, which shifted from the peripheral regions to the central regions of the tumour as the depth of tumor invasion progressed. CONCLUSION: This study reveals the potential molecular mechanisms of lung cancer brain metastasis from both single-cell and spatial transcriptomic perspectives, providing biological insights and clinical reference value for detecting patients suffering from lung cancer brain metastasis.