Immune profiling and prognostic model of pancreatic cancer using quantitative pathology and single-cell RNA sequencing

BACKGROUND: Pancreatic ductal adenocarcinoma (PDAC) has a complex tumor immune microenvironment (TIME), the clinical value of which remains elusive. This study aimed to delineate the immune landscape of PDAC and determine the clinical value of immune features in TIME. METHODS: Univariable and multivariable Cox regression analyses were performed to evaluate the clinical value of immune features and establish a new prognostic model. We also conducted single-cell RNA sequencing (scRNA-seq) to further characterize the immune profiles of PDAC and explore cell-to-cell interactions. RESULTS: There was a significant difference in the immune profiles between PDAC and adjacent noncancerous tissues. Several novel immune features were captured by quantitative pathological analysis on multiplex immunohistochemistry (mIHC), some of which were significantly correlated with the prognosis of patients with PDAC. A risk score-based prognostic model was established based on these immune features. We also constructed a user-friendly nomogram plot to predict the overall survival (OS) of patients by combining the risk score and clinicopathological features. Both mIHC and scRNA-seq analysis revealed PD-L1 expression was low in PDAC. We found that PD1 + cells were distributed in different T cell subpopulations, and were not enriched in a specific subpopulation. In addition, there were other conserved receptor-ligand pairs (CCL5-SDC1/4) besides the PD1-PD-L1 interaction between PD1 + T cells and PD-L1 + tumor cells. CONCLUSION: Our findings reveal the immune landscape of PDAC and highlight the significant value of the combined application of mIHC and scRNA-seq for uncovering TIME, which might provide new clues for developing immunotherapy combination strategies. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12967-023-04051-4.