Comparative analysis of single-cell transcriptome reveals heterogeneity in the tumor microenvironment of lung adenocarcinoma and brain metastases

Abstract Purpose Solid tumors such as lung adenocarcinoma include not only the tumor cells but also the microenvironment in which the tumor cells continuously interact with each other. An in-depth understanding of the oncological features and tumor microenvironment (TME) of lung adenocarcinoma and brain metastases at the single-cell level could provide new therapeutic strategies for brain metastases from lung adenocarcinoma. Methods To solve this problem, we performed single-cell RNA sequencing (scRNA-seq) analysis on 15 lung adenocarcinoma samples and 10 brain metastasis samples. Results A total of 86,282 single cells were obtained and divided into 8 cell types, including epithelial cells, endothelial cells, fibroblasts, oligodendrocytes, T/NK cells, B cells, mast cells, and macrophages. In brain metastases, we found a significantly lower proportion of T/NK cells and mast cells, and more severe immune dysregulation. In addition, we found a subpopulation of macrophages with high expression of metastasis-promoting-related genes enriched in brain metastatic tissues. Moreover, in brain metastases, we found a significantly increased proportion of myofibroblastic cancer-associated fibroblasts (myCAFs) and a higher angiogenic capacity of endothelial cells. Epithelial cells in brain metastases were more malignant and underwent genomic reprogramming. Next, we found that DNA damage-inducible transcript 4 (DDIT4) expression was upregulated in epithelial cells in brain metastases and was associated with poor prognosis. Finally, we experimentally validated that the downregulation of DDIT4 inhibited the proliferation, migration, and invasion of lung cancer cells. Conclusions This study depicts a single-cell atlas of lung adenocarcinoma and brain metastases by scRNA-seq and paves the way for the development of future therapeutic targets for brain metastases from lung cancer