Research Paper Volume 12, Issue 21 pp 21559—21581

Single-cell transcriptome analysis demonstrates inter-patient and intra-tumor heterogeneity in primary and metastatic lung adenocarcinoma

Yafei Liu1, *, , Guanchao Ye1, *, , Lan Huang2, , Chunyang Zhang1, , Yinliang Sheng1, , Bin Wu1, , Lu Han1, , Chunli Wu1, , Bo Dong1, , Yu Qi1, ,

  • 1 Department of Thoracic Surgery, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China
  • 2 Biotherapy Center, The First Affiliated Hospital of Zhengzhou University, Zhengzhou 450052, China
* Equal contribution

Received: April 21, 2020       Accepted: August 8, 2020       Published: November 10, 2020
How to Cite

Copyright: © 2020 Liu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


In this study, we performed single-cell transcriptome data analysis of fifty primary and metastatic lung adenocarcinoma (LUAD) samples from the GSE123902 and GSE131907 datasets to determine the landscape of inter-patient and intra-tumoral heterogeneity. The gene expression profiles and copy number variations (CNV) showed significant heterogeneity in the primary and metastatic LUAD samples. We observed upregulation of pathways related to translational initiation, endoplasmic reticulum stress, exosomes, and unfolded protein response in the brain metastasis samples as compared to the primary tumor samples. Pathways related to exosomes, cell adhesion and metabolism were upregulated and the epithelial-to-mesenchymal-transition (EMT) pathway was downregulated in brain metastasis samples from chemotherapy-treated LUAD patients as compared to those from the untreated LUAD patients. Tumor cell subgroups in the brain metastasis samples showed differential expression of genes related to type II alveolar cells, chemoresistance, glycolysis and oxidative phosphorylation (metabolic reprogramming), and EMT. Thus, single-cell transcriptome analysis demonstrated intra-patient and intra-tumor heterogeneity in the regulation of pathways related to tumor progression, chemoresistance and metabolism in the primary and metastatic LUAD tissues. Moreover, our study demonstrates that single cell transcriptome analysis is a potentially useful tool for accurate diagnosis and personalized targeted treatment of LUAD patients.


LUAD: lung adenocarcinoma; NSCLC: non-small-cell lung carcinoma; LN: lymph node; CNVs: copy number variations; DEGs: differentially expressed genes; GO: gene ontology; KEGG: kyoto encyclopedia of genes and genomes pathway; TCGA: the cancer genome atlas; MF: molecular function; BP: biological process; CC: cellular components; GSVA: gene set variation analysis; EMT: epithelial-mesenchymal transition; PCA: principal component analysis; UMAP: uniformmanifold approximation and projection; MSigDB: molecular signatures database; MET: mesenchymal epithelial transition; OXPHOS: oxidative phosphorylation; tSNE: t-distributed stochastic neighbour embedding-based; ENCODE: the encyclopedia of DNA elements; logFC: logarithmic fold change.