Open AccessCirculating tumor cell characterization of lung cancer brain metastases in the cerebrospinal fluid through single‐cell transcriptome analysis
Author(s) -
Ruan Haoyu,
Zhou Yihang,
Shen Jie,
Zhai Yue,
Xu Ying,
Pi Linyu,
Huang Ruofan,
Chen Kun,
Li Xiangyu,
Ma Weizhe,
Wu Zhiyuan,
Deng Xuan,
Wang Xu,
Zhang Chao,
Guan Ming
Publication year - 2020
Publication title -
clinical and translational medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.125
H-Index - 1
ISSN - 2001-1326
DOI - 10.1002/ctm2.246
Subject(s) - transcriptome , circulating tumor cell , lung cancer , cerebrospinal fluid , cancer , pathology , biology , adenocarcinoma , gene expression profiling , cell , cancer research , cancer cell , medicine , gene , gene expression , metastasis , genetics
Background Brain metastases explain the majority of mortality associated with lung cancer, which is the leading cause of cancer death. Cytology analysis of the cerebrospinal fluid (CSF) remains the diagnostic gold standard, however, the circulating tumor cells (CTCs) in CSF (CSF‐CTCs) are not well defined at the molecular and transcriptome levels. Methods We established an effective CSF‐CTCs collection procedure and isolated individual CSF cells from five lung adenocarcinoma leptomeningeal metastases (LUAD‐LM) patients and three controls. Three thousand seven hundred ninety‐two single‐cell transcriptomes were sequenced, and single‐cell RNA sequencing (scRNA‐seq) gene expression analysis was used to perform a comprehensive characterization of CSF cells. Results Through clustering and expression analysis, we defined CSF‐CTCs at the transcriptome level based on epithelial markers, proliferation markers, and genes with lung origin. The metastatic‐CTC signature genes are enriched for metabolic pathway and cell adhesion molecule categories, which are crucial for the survival and metastases of tumor cells. We discovered substantial heterogeneity in patient CSF‐CTCs. We quantified the degree of heterogeneity and found significantly greater among‐patient heterogeneity compared to among‐cell heterogeneity within a patient. This observation could be explained by spatial heterogeneity of metastatic sites, cell‐cycle gene, and cancer‐testis antigen (CTA) expression profiles as well as the proportion of CTCs displaying mesenchymal and cancer stem cell properties. In addition, our CSF‐CTCs transcriptome profiling allowed us to determine the biomarkers during the progression of an LM patient with cancer of unknown primary site (CUP). Conclusions Our results will provide candidate genes for an RNA‐based digital detection of CSF‐CTCs from LUAD‐LM and CUP‐LM cases, and shed light on the therapy and mechanism of LUAD‐LM.