Single-cell RNA-seq analysis reveals microenvironmental infiltration of myeloid cells and pancreatic prognostic markers in PDAC.

Abstract

Background: Pancreatic ductal adenocarcinoma (PDAC) has a heterogeneous make-up of myeloid cells that influences the therapeutic response and prognosis. However, understanding the myeloid cell at both a genetic and cellular level remains a significant challenge.

Methods: Single-cell RNA sequencing (scRNA-seq) data were downloaded from t the Tumor Immune Single-cell Hub and gene expression data were retrieved from The Cancer Genome Atlas (TCGA) database and the Gene Expression Omnibus (GEO) database. Gene set variation analysis (GSVA) was used to estimate the relative proportions of each cell type based on the signatures identified by scRNA-seq or previous literature. Cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) was performed to evaluate the abundance of immune infiltrating cells. For further analysis, LASSO and Cox analyses were used to construct a risk model using univariate Cox regression.

Results: Using the scRNA-seq dataset, we identified 7 clusters of myeloid cells, and these clusters were assigned a cell type based on their marker genes. In addition, the results of the CellChat analysis and SCENIC analysis indicate that TAM-spp1 cells may promote the migration of pancreatic tumor cells on different levels. Moreover, the TAM-spp1 cell is most closely related to poor prognoses. An 8-gene risk model was constructed by using univariate Cox and LASSO analyses. In the GEO cohorts, this risk model demonstrated excellent predictive abilities for prognosis. Further, patients with high-risk scores had a lower likelihood of benefiting from immunotherapy.

Conclusion: Using bulk RNA-seq and single-cell RNA-seq, we analyzed myeloid heterogeneity at the single-cell level, and we developed an 8-gene model that predicts survival outcomes and immunotherapy response in PADC.