Spatiotemporal Transcriptional Dynamic Landscape of Lung Adenocarcinoma from Preneoplasia to Invasive Adenocarcinoma

Background: The invasive potential is regarded as a pivotal determinant in the prognosis of lung adenocarcinoma (LUAD). However, the cell ecology and spatial niche implicated in the dynamic and sequential process of LUAD from adenocarcinoma in situ (AIS) to microinvasive carcinoma (MIA) and subsequent invasive adenocarcinoma (IAC) are not elucidated yet.Methods: Droplet-based single-cell RNA sequencing (scRNA-seq) was performed to map the cellular ecology of cancer cells and tumour microenvironment (TME) in three stages of LUAD, and paired tissues were simultaneously examined by spatial transcriptomics (ST) to assess cell spatial distribution. To verify our results, we applied multiple methods: bulk RNA sequencing, immunofluorescence (IF) staining, analysis of The Cancer Genome Atlas (TCGA) data and basic experiment.Findings: We generated and integrated a large-scale, single-cell spatiotemporal multi-omics atlas of LUAD using scRNA-seq and ST to explore the invasion trajectory of LUAD, where we first discerned cell-type-specific transcriptional changes contributing to the invasive process of LUAD. UBE2C+ cancer cell subpopulation was found to be constantly increased during invasive process of LUAD with remarkable elevation in IAC. Furthermore, analysis of TME cell type subpopulation showed constant decrease of mast cells, monocytes and lymphatic endothelial cells were implicated in the whole process of invasive LUAD accompanying with increase of NK cells and MALT B cells from AIS to MIA, and increase of Treg, secretory B cells, and decreased of cancer-associated fibroblast (CAF) from MIA to IAC. Notably, communication and interaction between cancer cells and TME cells-induced constitutive activation of TGF-β signaling was only involved in the invasion of IAC. Finally, spatial localization and distribution of the candidate cancer and TME cells, and the spatial gene expression, were determined in individual LUAD tissue.Conclusions: Our results present the specific cellular information and spatial architecture from cancer cells and TME subpopulations, as well as cell interaction between them, which will facilitate the identification and development of precision medicine in each invasive process of LUAD from AIS to IAC.Funding Information: This work was funded by the National Natural Science Foundation of China (82002421), Discipline Innovation Development Plan Project of Tangdu Hospital (2021LCYJ005), Young Talent Program of Tangdu Hospital, and Special Fund for Elite Talents of Shaanxi Provincial People's Hospital(2021JY-20).Declaration of Interests: We declare that we have no conflicts of interest.Ethics Approval Statement: The studies involving human participants were reviewed and approved by the Ethics Committee of Air Force Medical University Affiliated Tangdu Hospital (K202107-19), and all recruited patients signed an informed consent form.