A transcriptomic landscape of cancer and TME in early-stage lungadenocarcinomaby single-cell sequencing.

33 Background: Immuno-checkpoint inhibition therapies has been revolutionizing cancer treatment. Yet only a fraction of patients show durable responses , whereas the majority of treated patients show relatively low clinical response. This difference is likely to be caused by the heterogeneity of both cancer cells and cells involved in the tumor microenvironment (TME). However, the extent of this heterogeneity, how it is shaped by other factors in the tumor and vice versa, remains poorly understood. Methods: To explore the heterogeneity of lung adenocarcinoma, we obtained tumor tissuesand peripheral blood froma cohort of 30 treatment-naive patients. For each sample, single-cell RNA sequencingand whole exome sequencing(WES) wereperformed. A graph-based unsupervised clusteringmethod was usedand cell types were assignedto clusters based on marker gene expression. The sub-types of both cancer cells and TME cells along with their gene expression patterns were comparedbetween different cancer stages and mutation status. Results: we presented a landscape of cancer and TME transcriptome in human lung adenocarcinoma at single-cell resolution. We found the expression of cancer cells exhibits distinct characteristics with different TNM stages. For instance, the tumor cells of a patient classified as T1aN2M0 highly express genes enriched in the cell migration pathway. By comparing TME cells among patients with different mutation status, we identified changes in various T cell subtypes and tumor-infiltrating myeloid cells. Conclusions: This study provided a detailed cell atlas of lung adenocarcinoma and identified gene expression patterns that are unique to specific TNM stages. Moreover, single-cell analyses offer valuable knowledgeof immune changes for each patient subgroup, providing ausefultool for the rational design of immunetherapies.