Immune Subtyping For Pancreatic Cancer With Implication And Mechanisms In Clinical Outcomes And Improving Immune Checkpoint Blockade Treatment

Background:Pancreatic ductal adenocarcinoma(PDAC),with an increasing incidence and less than 10%overall survival rate,has been striking a heavy burden on human health and life safety.Over the past two decades,advances in chemotherapy has been achieved,yet the response rate is only about 40%.Patients with PD AC need more effective treatment.Anti-programmed death 1(PD-1)/programmed death ligand 1(PD-L1),as a promising immune checkpoint blockade(ICB)regimen,has yielded significant clinical efficacy in some tumors.However,low response rate and limited patients benefited from single-agent ICB were observed in PDAC,which may be attributed to the low immunogenicity and diverse immunosuppression mechanisms.Hence,to identify patients sensitive to ICB treatment,an specific immune subtyping model is needed.Also,a combinational therapy of ICB and other targets in the tumor microenvironment is considered to be a promising solution.Over the last decades,significant advances in molecular and genetic big data analysis have contributed to the understanding of the molecular pathogenesis of PDAC and provided clues for the design of more effective treatment regimens.However,to date,there is still lack of a molecular subtyping strategy based on immune characteristics to better understand the heterogeneity of the immune response and the reasons for ICB treatment's failure.Objective:(1)To establish a PDAC subtyping strategy based on the five immune features;(2)To depict the specific features of PDAC immune microenvironment,and analyze its correlation with patients' prognosis and ICB treatment effect;(3)To search for potential targets that can improve the efficacy of ICB,and to evaluate its correlation with the immune microenvironment features.(4)To explore the molecular mechanism of the synergistic target involved in the immune regulation and its impact on ICB treatment.Methods:(1)A total of 383 PDAC transcriptome data were downloaded from TCGA,ICGC and GEO public databases.Immune subtyping model of PDAC was established by enrichment analysis of immune-related gene sets(Macrophages,Lymphocyte,IFN-y,TGF-?,and Wound healing)and unsupervised clustering.(2)The R program and Xcell software were used to analyze the enrichment variation of immune cellular types and immunomodulator genes in each immune subtypes.Estimate software was used to calculate the immune and stroma scores of each immune subtypes and evaluate the characteristics of immune microenvironment.Cox regression and other methods were used to analyze the relationship between immune-related gene sets,immune cells,immunomodulators and PDAC patients'overall survival respectively,and C-index index was used to evaluate the accuracy of prognosis model.(3)The potential ICB synergistic therapeutic target was screened out by bioinformatic analysis,and the correlation between the target and clinical features was verified by PDAC tissue microarray.The correlation between the synergistic targets and immune cell infiltration and immunoregulatory gene enrichment was analyzed by Xcell software and R program.(4)The molecular regulation mechanism of PD-L1 expression in tumor cells was explored in vitro by plasmid construction,cell stable transmutation construction,Western Blotting and other methods.The synergistic effect of this target on ICB was explored via PDAC mice model.Results:(1)Via R software and clustering analysis,three immune subtypes of PDAC(Cluster 1,C1;Cluster 2,C2;and Cluster 3,C3)were established.Each subtype had distinct immune features and prognosis.Subtype C1,characterized by immunodeficiency,was mainly manifested by deficiency of immunomodulators.Subtype C2,characterized by immunosuppression,was mainly manifested by robust TGF-? pathway and stromal response,and patients within C2 have the worst prognosis.Subtype C3,characterized by hot immune response,was mainly manifested by abundant infiltration of immune cells and enrichment of immunomodulators.(2)The discrepancy in immune features(immune cells and immunomodulators)of each subtype indicated different suitability of anti-PD-1/PD-L1 therapy,as well as the potential sensitivity of each subtype to other immunotherapeutic modalities.Patients within the C3 group were more suitable for anti-PD-1/PD-L1 therapy.(3)TGM2,a promising target for improving anti-PD-1/PD-L1 therapy,was found based on the cellular and molecular features of immune subtypes.TGM2 was verified to be associated with poor prognosis of patients by tissue microarray,and the correlation between TGM2 and the immunosuppressive microenvironment was revealed.(4)The regulatory relationship between TGM2 and PD-L1 in PD AC cells was explored.It was proposed that TGM2 could regulate the expression of PD-L1 via STAT3 pathway and AKT/NF-kB pathway.In vivo study revealed that TGM2 silencing combined with PD-1 inhibitor can effectively slow down tumor growth,suggesting that TGM2 could be a potential synergistic target for anti-PD-1/PD-L1 therapy.Conclusion:(1)PD AC is highly heterogeneous.Based on differences in immune characteristics,PDAC can be divided into three different immune subtypes,characterized by immune cold,immunosuppression and immune hot,respectively(2)The differences in immune characteristics among immune subtypes are of great value for suggesting immunotherapy sensitivity and prognosis,and C3 subtype is more suitable for PD-1/PD-L1 treatment.(3)TGM2 is related to the characteristics of PDAC inhibitory immune microenvironment.(4)TGM2 can regulate the expression of PD-L1,and silencing TGM2 can improve the efficacy of ICB,which has a potential clinical application prospect.