The Study On The Mechanism Of Poor Prognosis In Patients With Hepatocellular Carcinoma With Microvascular Invasion Based On Multi-Omics Technique

PartⅠA retrospective case-control study of prognostic factors in patients with microvascular invasion positive hepatocellular carcinomaObjective: Hepatocellular carcinoma(HCC)is one of the most common cancer all over the world,the morbidity and mortality rank front among all cancers,the 5-year survival rate after hepatectomy is less than 15%.as a marker of tumor progression,microvascular invasion(MVI)is one of the most important factors leading to poor postoperative prognosis of HCC patients.At present,most of the clinical studies on MVI and HCC focus on the preoperative predictors of MVI and the risk of MVI in HCC,while there are few studies on the prognostic factors of HCC patients with MVI.In this part,we first used our own data to continue to analyze the impact of MVI on the prognosis of HCC,and then focused on the factors influencing the postoperative disease free survival and overall survival of patients with MVI positived HCC.Methods: A retrospective study was conducted on 1202 patients with HCC admitted to the Guangxi Medical University Cancer Hospital from January 2012 to December 2017,including 432 patients with hepatocellular carcinoma with microvascular invasion(MVI-positive HCC)in the case group and 770 patients without HCC(MVI-negative HCC)in the control group.Retrospectively collected In this section,the patient’s information such as gender,age,white blood cell count,neutrophil count,lymphocyte count,neutrophil/lymphocyte ratio(NLR),platelet count,HBV infection,BCLC stage,AST,ALT,total bilirubin and AFP level,tumor size,tumor number,tumor capsular,liver cirrhosis,pathologic stage,macrovascular invasion,microvascular invasion.Survival analysis was performed using Kaplan-Meier method and Cox regression univariate and multivariate analyses were used to evaluate prognostic risk factors for disease-free survival and overall survival.In univariate analysis,Cox regression method was used to calculate the risk of each variable.The univariate risk variables which P value <0.05 were included in multivariate Cox regression model for multivariate analysis.Results: The disease-free survival and overall survival of MVI-positive HCC patients were significantly shorter than those of MVI-negative HCC patients(P < 0.001).Cox regression analysis showed that MVI was an independent risk factor for disease-free survival and overall survival in HCC patients.COX regression univariate analysis showed that NLR,BCLC stage(B + C),AST(> 40 U/L),AFP level(> 400 ng/m L),tumor number(multiple),tumor size(> 5 cm)are risk factors affecting the disease-free survival of MVI-positive HCC patients,multivariate analysis showed that tumor size(> 5 cm)is the independent risk factors influence the disease-free survival of MVI positive HCC patients.COX regression univariate analysis showed that BCLC stage(B + C),serum levels of white blood cell count,neutrophil count,NLR,tumor size(> 5 cm),incomplete tumor capsular are risk factors affecting the disease-free survival of MVI-positive HCC patients,multivariate analysis showed that NLR,BCLC stage(B + C),incomplete tumor capsular are the independent risk factors influence the overall survival of MVI positive HCC patients.Conclusion: Data from our center further confirmed that MVI was an independent risk factor for HCC disease-free survival and overall survival,and that the tumor size(> 5 cm)was an independent risk factor for disease-free survival of MVI-positive HCC patients,NLR,BCLC stage(B+C)and incomplete tumor capsule are independent risk factor for overall survival of MVI-positive HCC patients.Part Ⅱ Analysis of the significant enrichment pathway of MVI-positive HCC by RNA-seq techniqueObjective: We retrospectively studied the differences in clinicopathological data and prognosis between the MVI-positive and MVI-negative HCC patients in part I,as well as the influencing factors for the prognosis of MVI-positive HCC,providing some clues for the poor prognosis of MVI-positive HCC,however,its specific mechanism is not clear,in order to further explore the molecular mechanism affect the poor prognosis of MVI positive HCC,further studies are needed.In this part,the potential molecular mechanism of the poor prognosis of MVI-positive HCC is explored by RNA-seq technology.Methods: The cancer tissues of 93 HCC patients were divided into two groups: MVI-positive HCC(experimental group)(49 cases)and MVI-negative HCC(control group)(44 cases)according to the presence or absence of MVI.The transcriptome of HCC cancer tissues was analyzed by RNA-seq technique after RNA were extracted,then screening the significantly differential expressed genes(Differential expressed genes,DEGs),the selection standard for DEGs was the Fold change(|FC|≥2)or(|log2 FC| ≥1),P value < 0.05.GO(gene ontology)function enrichment analysis and KEGG(Kyoto encyclopedia of genes and genomes)signaling pathway enrichment analysis were performed for DEGs.In addition,we also performed ss GSEA and x CELL analysis on two groups of patients.Results: Our study found that a total of 1080 DEGs were screened between MVI-positive HCC and MVI-negative HCC groups,among which 864 genes were and 216 genes were down-regulated.GO enrichment analysis showed that up-regulated genes of MVI positive HCC were mainly involved in cell differentiation,adhesion,extracellular matrix disassembly and other biological processes and molecular functions.KEGG signaling pathway analysis showed that the signaling pathways of up-regulation gene enrichment in MVI positive HCC group mainly included: extracellular matrix(ECM)receptor interaction,Cell adhesion molecules(CAMs),HIF-1 signaling pathway,PI3K-Akt signaling pathway,Wnt signaling pathway,MAPK signaling pathway,and Ras signaling pathway,These signaling pathways are closely related to the occurrence and development of tumors and can mediate the remodeling of tumor microenvironment by regulating immune function.By ss GSEA analysis,we found that the proportion of Mast cells(MCs)、Th2 cells、B cells were decreased significantly in MVI-positive HCC,the cytolytic activity and interferon-γ(IFN-γ)also known as type2 interferon response(Type_II_IFN_REPONSE)activity were also significantly reduced in MVI-positive HCC.The x CELL analysis showed that effector CD4+T cells(CD4+Tem)、central memory CD8+T cells(CD8+Tcm)、effective CD8+T cells(CD8+Tem)and conventional dendritic cell (cDC)were significantly decreased in MVI-positive HCC,and no significant difference was found in other immune cells.Conclusions: In our study,RNA-seq technology revealed that compared with MVI-negative HCC,MVI-positive HCC enriched more signaling pathways remolded by regulating tumor microenvironment,which was closely related to tumor progression,and we further found MVI-positive HCC tumors by ss GSEA and x CELL analysis The microenvironment may be in a low immune state,suggesting that MVI-positive HCC may have tumor microenvironment that is more conducive to cancer progression.Part Ⅲ Analysis of tumor microenvironment of MVI-positive HCC by CyTOFObjective: The RNA-Seq results in the second part of our study showed that,the significantly upregulated genes in the MVI-positive HCC group were mainly enriched in the pathways related to the tumor microenvironment(TME)compared with the MVI-negative HCC group,in addition,ss GSEA and x CELL analysis further showed that the TME of MVI-positive HCC may be in a low immune state.Therefore,we speculated that there may be different TME between MVI-positive HCC and MVI-negative HCC.TME is a complex and comprehensive environment in which tumor cells occur and develop,including surrounding immune cells,blood vessels,extracellular matrix,fibroblasts,lymphocytes,marrow derived immunosuppressive cells,inflammatory cytokines and chemokines.TME plays an important role in immune escape and therapeutic resistance in tumor progression.Further study of TME has important prognostic value and has guiding significance for tumor immunotherap.Therefore,in this part,we used cytometry by time-of-flight(CyTOF)to study the difference of TME between MVI-positive HCC and MVI-negative HCC groups,and to reveal the TME characteristics of MVI-positive HCC in depth.Methods:In this part of the study,cancer tissues from 68 HCC patients were collected and divided into two groups on the basis of the presence or absence of MVI: MVI-positive HCC(experimental group)(35 cases)and MVI-negative HCC(control group)(33 cases).After the cancer tissues were disintegrated and digested into single-cell suspension,CyTOF was used to analyze the differences in TME between the two groups.Results:Our study showed that compared with MVI-negative HCC,the ratio of immunosuppressive Treg cells to CD4+T cells in MVI-positive HCC patients was higher,and more immunosuppressive Treg cells were found in CD4+T cells in MVI-positive HCC patients.By analysis of the expression differences of immune exhaustion-related markers in each cell subpopulation,it was found that compared with MVI-negative HCC group,CD4+T cells expressed higher levels of TIM3,PD-1 and PD-L1,CD8+T cells expressed more LAG-3,PD-1 and PD-L1,TAMS expressed more TIM3,LAG-3 and PD-L1,and Treg expressed more LAG 3,PD-1 and PD-L1 in the MVI-positive HCC group.The expression of different immune markers in the two groups of HCC was further detected,and it was found that the expression of Ki67,IL-6,CCR6,TGF-β,TIM3,LAG-3,PD-1 and PD-L1 in MVI-positive HCC was significantly higher than that in MVI-negative HCC.Conclusions: Our CyTOF results showed that CD4+ T cells have much more number of the immunesuppressive Treg cells in MVI-positive HCC patients,CD4+ T cells、CD8+ T cells、Treg、TAMs and TANs expressed higher levels of immune exhaustion-related markers in MVI-positive HCC patients.In addition,MVI-positive HCC patients expressed higher levels of immunosuppressive markers in the TME,and were in a more active state of proliferation.In conclusion,there was an immunosuppressive TME in MVI-positive HCC,and may have a better response to immunotherapy.Part Ⅳ Analysis of tumor microenvironment of MVI-positive HCC by imaging mass spectrometryObjective: The results of the second part of our study revealed that compared with MVI-negative HCC group,extracellular matrix(ECM)receptor interaction pathways was significantly enriched in the MVI-positive HCC group,and ECM was mainly secreted by cancer-associated fibroblasts(CAFs),The results of the third part showed that there was immunosuppression in the TME of the MVI-positive HCC group,which was mainly manifested by T cells expressing more exhaustion-related markers,making T cells in the state of immune exhaustion,while CAFs could inhibit the tumor killing ability of T cells by interacting with T cells.In this part,Imaging Mass Cytometry(IMC)technique was used to explore the interaction between CAFs and T cells in order to further reveal the characteristics of TME in MVI-positive HCC group.Methods:In this part,cancer tissues from 18 HCC patients were divided into two groups on the basis of the presence or absence of MVI: MVI-positive HCC(experimental group)(8 cases)and MVI-negative HCC(control group)(10 cases).Paraffin-embedded tissue sections were analyzed by Imaging Mass Cytometry(IMC).IMC data analysis Raw images of each channel are exported from the MCD viewer.The sample image is cropped to allow for separate analysis of different scan areas.Each sample of the original image data import multichannel Histo CAT tissue imaging data analysis software(version 1.75),using Pheno Graph algorithm,based on each cell protein expression similarity clustering,to obtain a different phenotype of cells,we use bioinformatics software(ISAT)to analysis the distance between any two nuclei and nuclear.Results:Manual gating strategy was used to calculate the proportion of different CAFs.It was found that the proportion of α-SMA(+)CAFs and FAP+Vimentin+CAFs in MVI-positive HCC group was significantly higher than that in MVI-negative HCC group.T-SNE algorithm generated two-dimensional map to classify tumor cells,immune cells and mesenchymal cells of MVI-positive HCC and MVI-negative HCC groups,and all cells were divided into 19 clusters based on cell surface markers.They included one CD4+T cell cluster(cluster 10),one CD8+T cell cluster(cluster 16),three α-SMA+Collagen-1+ CAFs cluster(cluster 9,13,18),two FAP+Vimentin+ CAFs cluster(cluster 17,19)and two tumor cell clusters: cluster 1,2,4 and 7 belong to Ep CAM+He Par-1+Arginase + tumor cell cluster,cluster 3,5,6,8,11 and 15 belong to Ep CAM+Arginase+ tumor cell cluster,while cluster 12 and 14 belong to the other cells.Compared with MVI-negative HCC group,α-SMA+Collagen-1+CAFs and FAP+Vimentin+CAFs were significantly increased in MVI-positive HCC group,the distance between T cells and CAFs in MVI-positive HCC group was significantly shortened,and the expression of PD-L1,a marker of T cell immune exhaustion,was significantly higher than that in the MVI-negative HCC group.Conclusions: Compared with MVI-negative HCCs,the proportion of CAFs in MVI-positive HCC is significantly increased,the distance between T cells and CAFs is significantly shortened,and the expression of PD-L1,an immune exhaustion-related markers of T cell is higher.CAFs may inhibit T cell function through stronger interaction,leading to the existence of immunosuppressive TME in MVI-positive HCC.Part Ⅴ Analysis of the different lipids between MVI positive and MVI negative HCC group by lipidomicsObjective: The results in the second part of the transcriptome sequencing showed that the enrichment pathways in MVI positive HCC contained lipid metabolism related pathway : Ether lipid metabolism,phosphatidylcholine acyl-chain remodeling and phosphatidylglycerol acyl-chain remodeling.lipid metabolism is associated with tumor development and TME,in this part,lipids between MVI-positive HCC and MVI-negative HCC were analyzed by lipidomics,in order to further reveal the mechanism of poor prognosis of MVI-positive HCC from the level of lipidomics.Methods: In this part of the study,cancer tissues from 30 HCC patients were collected and divided into two groups on the basis of the presence or absence of MVI: MVI-positive HCC(experimental group)(16 cases)and MVI-negative HCC(control group)(14 cases).The lipids were extracted from cancer tissues and analyzed by lipidomics.Results: We detected a total of 28 kinds of lipid,most of which contained multiple subgroups.There were significant up-regulation of cholesterol,Acylcarnitine and its three subgroups: acylcarnitine-14:0,acylcarnitine-16:0,acylcarnitine-18:0,plasmalogen PC,Glu Cer and one of its subgroups: Glu Cer d18:1/24:1,Lac Cer and its five subgroups: Lac Cer d18:0/18:0,Lac Cer d18:1/22:0,Lac Cer d18:1/23:0,Lac Cer d18:1/24:1,SL and its seven subgroups: SL d18:1/22:0、SL d18:1/22:0h、SL d18:1/22:1、SL d18:1/24:0、SL d18:1/24:0h、SL d18:1/24:1、SL d18:1/24:1h,however Cardiolipins(CL)was significantly reduced in the MVI-positive HCC group.There was no significant difference among other lipids.Conclusions: Our lipid omics showed that there is also a significant difference between MVI-positive HCC and MVI-negative HCC group,compared with the MVI negative HCC,there are much more lipids in the MVI-positive HCC,and these elevated lipid are associated with the progression of HCC,especially the significantly elevated cholesterol in MVI-positive HCC group that can put T cells in a state of immune exhaustion,which further explains our previous experimental results that there is an immunosuppressive TME in MVI-positive HCC.