Background Brain metastasis (BM) is the most common intracranial malignancy causing significant mortality, and lung cancer is the most common origin of BM. However, the cellular origins and drivers of BM from lung adenocarcinoma (LUAD) have yet to be defined. Methods The cellular constitutions were characterized by single-cell transcriptomic profiles of 11 LUAD primary tumor (PT) and 10 BM samples (GSE131907). Copy number variation (CNV) and clonality analysis were applied to illustrate cellular origins of BM tumors. Brain metastasis-associated epithelial cells (BMAECs) were identified by pseudotime trajectory analysis. By using machine-learning algorithms, we developed the BM-index representing the relative abundance of BMAECs in the bulk RNA-seq data, indicating high risk of BM. Therapeutic drugs targeting BMAECs were predicted based on the drug sensitivity data of cancer cell lines. Results Differences in macrophages and T cells between PTs and BMs were investigated by single-cell RNA (scRNA) and immunohistochemistry and immunofluorescence data. CNV analysis demonstrated BM was derived from subclones of PT with a gain of chromosome 7. We then identified BMAECs and its biomarker, S100A9. Immunofluorescence indicated strong correlations of BMAECs with metastasis and prognosis evaluated by the paired PT and BM samples from Peking Union Medical College Hospital (PUMCH). We further evaluated the clinical significance of BM-index and identified 7 drugs that potentially target BMAECs. Conclusions This study clarified possible cellular origins and drivers of metastatic LUAD at single cell level, and laid a foundation for early detections of LUAD patients with a high risk of BM.

The immunosuppressive mechanisms of the surrounding microenvironment and distinct immunogenomic features in glioblastoma (GBM) have not been elucidated to date. To fill this gap, useful data were extracted from The Cancer Genome Atlas (TCGA), the Chinese Glioma Genome Atlas (CGGA), GSE16011, GSE43378, GSE23806, and GSE12907. With the ssGSEA method and the ESTIMATE and CIBERSORT algorithms, four microenvironmental signatures were used to identify glioma microenvironment genes, and the samples were reasonably classified into three immune phenotypes. The molecular and clinical features of these phenotypes were characterized via key gene set expression, tumor mutation burden, fraction of immune cell infiltration, and functional enrichment. Exhausted CD8+ T cell (GET) signature construction with the predictive response to commonly used antitumor drugs and peritumoral edema assisted in further characterizing the immune phenotype features. A total of 2,466 glioma samples with gene expression profiles were enrolled. Tumor purity, ESTIMATE, and immune and stromal scores served as the 4 microenvironment signatures used to classify gliomas into immune-high, immune-middle and immune-low groups, which had distinct immune heterogeneity and clinicopathological characteristics. The immune-H phenotype had higher expression of four immune signatures; however, most checkpoint molecules exhibited poor survival. Enriched pathways among the subtypes were related to immunity. The GET score was similar among the three phenotypes, while immune-L was more sensitive to bortezomib, cisplatin, docetaxel, lapatinib, and rapamycin prescriptions and displayed mild peritumor edema. The three novel immune phenotypes with distinct immunogenetic features could have utility for understanding glioma microenvironment regulation and determining prognosis. These results contribute to classifying glioma subtypes, remodeling the immunosuppressive microenvironment and informing novel cancer immunotherapy in the era of precision immuno-oncology.

No abstract