Single-cell RNA Sequencing Dissects Cellular Heterogeneity In Cervical Adenosquamous Carcinoma

Background:Cervical cancer is the fourth most prevalent malignant tumor in women worldwide,mainly including squamous cell carcinoma,adenocarcinoma and adenosquamous carcinoma.Cervical adenosquamous carcinoma(CASC)is a special pathological type mixed with both squamous cell carcinoma and adenocarcinoma,which accounting for approximately 3%-5%of cervical cancer and indicating an especially poor prognosis.The complexity and diversity of CASC indicated the strong tumor heterogeneity,and different tumor cells showed significant differences in phenotypes such as proliferation rate,invasion ability and drug resistance.However,the tumor heterogeneity of CASC remains poorly defined,which is critical to the early diagnosis and optimal treatment options.Methods:Two adjacent tissue samples were obtained by surgery from the primary tumor of the patient,who did not receive any antitumor therapy before sample collection.Epithelial and immune cells were obtained by single-cell suspension preparing and flow sorting for the subsequent single-cell RNA-sequencing(scRNA-seq).ScRNA-seq data were generated by 10X Genomics and preprocessed using Cell Ranger.We performed uniform manifold approximation and projection analysis,copy number variants analysis,cell type determination,gene ontology and gene set variation analysis,and cell state scores defining to investigate the tumor heterogeneity of CASC.Results:We performed scRNA-seq on 18046 individual cells derived from two cervical adenosquamous carcinoma samples to analyze the transcriptional heterogeneity of both epithelial and immune constituents,identifying seven epithelial(Epil-7)and 11 immune subclusters.Based on expression of known cervical cancer markers,Epi1-2 primarily displayed features of adenocarcinoma,whereas Epi3-6 were instead characterized by features of squamous carcinoma.Our analyses also revealed that hypoxia and KRAS signaling were highly represented within Epi1;metabolic pathways mediating glycolysis and oxidative phosphorylation were enriched in Epi2-4;while Epi5 was enriched in p53 pathway components and features of epithelial-mesenchymal transition,Epi6 was enriched in IL6/JAK/STAT3 pathway.Moreover,in a cohort of 165 cervical cancer patients from The Cancer Genome Atlas database,survival analysis showed that the abundance of Epi2 was strongly associated with poor prognosis(p=0.015).In the tumor immune microenvironment,CD8+FGFBP2+ T cells and FGFBP2+natural killer cells were found to display high levels of cytotoxic effectors(GZMA,GZMB,GNLY,and PRF1)and low levels of inhibitory markers(PDCD1,TIGIT,and CTLA4),and the tumor immune infiltration by these populations was positively associated with patients’ survival(p=0.017 and 0.014,respectively).Conclusion:In this study,we found the epithelial cell clusters exhibited the significant differences in tumor and metabolism-related pathway activities,among which,Epi2 highly expressed the adenocarcinoma characteristics and was closely related to the poor patients’prognosis.We also revealed two key immune cell populations with tumor-suppressing functions,predicting better patients’ survival,providing prognostic markers and potential therapeutic targets for future studies.