| Primary liver cancer is the second leading cause of cancer-related deaths globally and one of the few tumors with steadily increasing incidence and mortality rates.Hepatocellular carcinoma(HCC)and intrahepatic cholangiocarcinoma(iCCA)are the most common types of primary liver cancer,accounting for 85%and 10%of malignant liver tumors,respectively.Although early-stage HCC and iCCA confined to the liver can be treated with surgical resection and ablation,the recurrence rate within 5 years after treatment can be as high as 40%-70%.Most patients are diagnosed at an advanced stage or with liver cirrhosis-related complications and are not eligible for curative surgery.Tumor microenvironment(TME)is a complex system composed of many different cell types,including tumor cells,immune cells,and other relevant cells.There are close and complex interactions and interrelationships between different cell subpopulations in TME.In primary liver cancer,multiple factors,including liver cirrhosis that can cause fibrosis and carcinogenesis,lead to synchronous and asynchronous tumor initiation events in different cell subpopulations throughout the liver,resulting in significant heterogeneity among patients and within tumors.The tumor microenvironment heterogeneity induces different progression patterns in primary liver cancer and results in different treatment responses among patients.Currently,combination therapies have been applied to the treatment of primary liver cancer.However,due to a lack of understanding of the heterogeneity of its tumor microenvironment and effective predictive biomarkers,a considerable portion of patients still cannot obtain effective clinical relief and survival benefits from these therapies.In recent years,transcriptome sequencing technology has made significant progress and has become the primary method for analyzing gene expression patterns and heterogeneity in malignant tumors.The development of second-generation sequencing,single-cell sequencing,and spatial transcriptomics has enabled us to combine the gene expression and spatial information of individual tumor cells,analyze the determinants of tumor microenvironment heterogeneity in primary liver cancer from multiple factors and dimensions,identify the molecular basis of tumor microenvironment formation,and discover potential cancer biomarkers and therapeutic targets by analyzing the information on cell-cell interactions.These findings provide new insights into diagnosing,treating,and preventing primary liver cancer.In the first part of the study,we focused on two high-incidence primary liver cancers,hepatocellular carcinoma,and intrahepatic cholangiocarcinoma.Based on a novel cell death mechanism,cuproptosis with second-generation sequencing data and single-cell sequencing data from TCGA,ICGC,and GEO databases,we developed a new prognostic scoring system for the two subtypes of primary liver cancer using machine learning methods to more accurately predict tumor characteristics and prognosis for patients with hepatocellular carcinoma and intrahepatic cholangiocarcinoma.We also experimentally validated the expression levels of related targets.In the second part of the study,we considered the increasing incidence of intrahepatic cholangiocarcinoma,the unclear pathogenic factors,and controversial treatment methods.Additionally,compared to other types of primary liver cancer,there is fewer transcriptome data available for intrahepatic cholangiocarcinoma,especially lacking spatial transcriptome data.To further reveal the tumor ecosystem heterogeneity of intrahepatic cholangiocarcinoma,we collected intrahepatic cholangiocarcinoma samples from the Department of General Surgery,Qilu Hospital of Shandong University,and used spatial transcriptomics(ST)and digital spatial profiler(DSP)technologies,combined with single-cell sequencing data,to reveal the spatial heterogeneity characteristics of intrahepatic cholangiocarcinoma at the high-resolution transcriptomic level.We also explored the molecular heterogeneity among tumor interstitial spaces,intratumoral spaces,and different members in the tumor microenvironment and their interactions.Part 1:Developing Prognostic Scoring Systems for Primary Liver Cancer Based on Copper Homeostasis and Cuproptosis to Predict Tumor Microenvironment and Immune Cell Communication FeaturesObjective:The maintenance of intracellular copper homeostasis requires a complex system,and the disruption of copper homeostasis can induce a novel form of cell death known as copper death.Given the crucial role of the liver in copper metabolism,copper-related physiological processes have great potential in developing new therapies for primary liver cancer.Existing evidence suggests that intervening in the tumor microenvironment(TME)by regulating copper homeostasis and inducing copper death has promising applications.However,for further development of targeted drugs and clinical applications,it is necessary to further explore copper homeostasis and copper death-related targets in primary liver cancer.Methods:RNA sequencing expression profiles,corresponding clinical and pathological data,and single-cell sequencing data of hepatocellular carcinoma(HCC)and intrahepatic cholangiocarcinoma(iCCA)patients were downloaded from TCGA,GEO,and ICGC databases.After performing Pearson correlation analyses,univariate Cox regression,least absolute shrinkage and selection operator(LASSO)regression,and multivariate Cox regression,prognostic genes strongly associated with copper metabolism or cuproptosis were screened,and prognostic models were constructed.ROC analysis and Kaplan-Meier survival analysis were performed to evaluate the effectiveness of the prognosis model and to analyze the correlation between clinical and pathological features of patients and risk scores.The nomogram was constructed using a combination of risk score and clinical pathological features to predict patient prognosis.Multiple bioinformatics methods were used to analyze functional enrichment,genomic changes,and immune landscapes of tumors in different risk groups and to predict potential effective therapeutic drugs for different risk group patients.Using single-cell sequencing data,functional enrichment of T cells,macrophages,and malignant tumor cells in different risk score patient groups was evaluated,in which relevant pathways and cell activity were evaluated.Besides,the interaction between tumor cells and immune cells was analyzed.Results:1.Prognostic model and pathological feature correlation:In HCC and iCCA,the scoring systems consisting of five copper homeostasis and cuproptosis-related genes were constructed.Core genes exhibit expression differences in both liver cancer cells and cancer-adjacent tissues,and their expression levels are significantly associated with overall survival of patients.Highrisk scores indicate poorer clinical prognosis and can serve as an independent predictor of HCC prognosis.The combination of risk scores and traditional pathological features in a line chart demonstrates good performance in predicting patient prognosis.In HCC,risk score is significantly correlated with pathological features.2.Functional enrichment analysis:In HCC,functional enrichment analysis indicates that pathways related to tumor activity are significantly enriched in the high-risk group compared to the low-risk group.The proportion of malignant cells is higher in the high-risk group,accompanied by a series of upregulated cancer-promoting pathways.Similarly,in the corresponding analysis of iCCA,multiple tumor activity-related pathways are significantly enriched in the high-risk group,with a higher proportion of malignant cells and multiple upregulated cancer-promoting pathways.The PD-1/CTLA4 targeted therapy may reverse the activity of some tumor-related pathways in high-risk patients.3.Genomic alteration analysis:In HCC,there are differences in tumor heterogeneity score and mutation rates of several key tumor suppressor genes.In addition,there were significant differences in the methylation levels of a series of oncogenes between the different risk groups.4.Immune landscape analysis:Different risk groups of HCC and iCCA patients exhibit significant differences in a series of immune-related pathways.In HCC,different risk groups have significant differences in immune infiltration and expression of immune checkpoint molecules.Single-cell sequencing data showed a significant decrease in the proportion of immune cells and other cells.Pathways associated with T cell activity were significantly more active in the low-risk group.In iCCA,the results showed that most immune checkpoint molecules show higher expression levels in the high-risk group.Single-cell sequencing data showed that T cells in the TME of the low-risk group of iCCA were more active,and targeted treatment for iCCA patients may reverse the T cell exhaustion state in the high-risk group.5.Cell communication analysis:Communication analysis between malignant tumor cells and other immune-related cells was performed based on single-cell sequencing data.Significant differences in signal strength and communication processes were observed in multiple pathways between different risk groups,and signal pathways with significant differences in total information flow between different risk groups were revealed.6.Drug relevance and treatment efficacy:In HCC,a total of 156 different potentially effective drugs were screened,with 127 drugs being more sensitive to the high-risk group and 29 drugs being more sensitive to the low-risk group.In iCCA,a total of 17 risk score-related drugs were screened,with 15 drugs being more sensitive to the high-risk group and 2 drugs being more sensitive to the low-risk group.HCC patients witlow-risksk scores had a poorer response to immune checkpoint blockade therapy(ICB)and shorter survival time after ICB treatment.7.Expression and function validation of model constructed genes:The qRT-PCR results of 21 HCC patients’ paired tumor and adjacent normal tissues showed differential expression of the model construction genes between HCC and normal tissues.The influence of model construction genes on cell proliferation ability was preliminarily verified through CCK8.Further studies on the model construction gene ZCRB1 showed that it positively regulated the proliferation,migration,invasion and clone formation of HCC cells.Conclusions:This study developed prognostic scoring models for HCC and iCCA based on copper homeostasis and cuproptosis-related genes respectively.The risk score reflects disease prognosis,tumor function,and immune-related status.Based on intercellular communication between different-risk patients,we proposed new receptor-ligand pairs as potential targets for immune and TME-related changes in the respective diseases and predicted targeted chemotherapy drugs with potential efficacy.Our study comprehensively demonstrates the overall potential risk of copper-related biological processes in primary liver cancer,which can assist in the selection of treatment strategies for HCC and iCCA patients.Part 2:Spatial resolved transcriptomics reveals intra-tumor heterogeneity and microenvironment in intrahepatic cholangiocarcinomaObjective:The complex interactions between iCCA cells and various components of the tumor microenvironment at the spatial resolution level still lack corresponding evidence.Further research is needed to elucidate the potential molecular mechanisms driving iCCA pathogenesis.Deciphering the spatial structure of tissue cell components in iCCA can help understand the location background of gene expression that includes pathological changes and cellular heterogeneity information,identify new potential therapeutic targets,and provide new insights for the diagnosis,treatment,and prevention of intrahepatic cholangiocarcinoma.Methods:Surgical resected human iCCA samples were collected from the Department of General Surgery,Qilu Hospital of Shandong University.After quality control assessment,spatial transcriptomic amplification and library preparation were performed using the Visium Spatial Gene Expression Platform,capturing and analyzing approximately 3000 tissue domains on each frozen section.The final library was sequenced on the Illumina NovaSeq6000 system.The raw data were processed using Space Ranger to extract reads containing spatial barcodes and UMI sequences,which were then mapped to transcriptional information of the human genome reference GRCh38 using STAR.Each point was annotated with cell types based on pathology and SingleR,and manually adjusted.Seurat in R was used to identify differentially expressed genes in each annotated cluster,and GSVA was used to perform pathway enrichment analysis in different clusters.Digital Spatial Profiling was performed for quantitative gene expression(PanCK,CD3,CD68)using probe hybridization,UV cutting,and barcode collection,followed by PCR amplification and Illumina sequencing.Single-cell sequencing datasets were downloaded from the GEO database,and annotated single-cell data were combined with ST data.The Python package stereoscope was used to estimate the proportion of each cell type present at each capture location in ST.After manually selecting the areas where macrophages and malignant cells were in close proximity using the R package STUtility,CellPhoneDB was used to predict the signal crosstalk between soluble and membrane-bound factors produced by the above two cell types.Results:1.The spatial distribution characteristics of patient iCCA tissue in the spatial transcriptome cohort were identified.Two samples were identified as perihilar large duct iCCA(iCCAphl),and one sample was identified as peripheral small duct iCCA(iCCApps).Most of the cells in iCCA were malignant cells in the ST,while normal liver cells were wrapped around the slices,and the most infiltrated immune cell type was myeloid macrophages.The infiltrated T cells and B cells were very rare,indicating that iCCA is not an immunogenic malignancy.2.All iCCA patients showed a similar percentage of cell types,but each patient still exhibited a different gene expression profile and spatial structure.The differentially expressed genes of all cell types between different patients were revealed.3.Significant inter-patient heterogeneity of iCCA tumor cells was revealed.The behavior of malignant cell subpopulations was mainly driven by physical proximity,and it was influenced by the surrounding microenvironment.The spatial structure composed of tumor cells and other components was also revealed,and the expression of cancer-related genes and pathway activation characteristics of different malignant cell subpopulations were analyzed.4.The results of Digital Spatial Profiling showed substantial inter-patient heterogeneity in CD68-labeled macrophages and PanCK-labeled malignant epithelial cells among the three patients,while the cells within the same patient exhibited high homogeneity.CD3-labeled T cells showed high homogeneity both inter-patient and within the same patient.The T cell area was completely separated from the malignant epithelial cell area,demonstrating different transcriptomic characteristics.5.Combining iCCA single-cell sequencing data with ST data analysis,specific ligandreceptor complexes for tumor-associated macrophages(TAMs)were identified,including TNFRSF1A-GRN,TFF1-FGFR2,SPP1-PTGER4,and CEACAM6-CEACAM6.Conclusions:iCCA is a highly heterogeneous cancer,and the understanding of the iCCA tumor microenvironment is still limited.Our study reveals the microenvironment landscape of iCCA through spatial transcriptome sequencing,revealing the heterogeneity within and between iCCA tumors,and demonstrating the important role of tumor-associated macrophages(TAMs)in the iCCA immune-suppressive microenvironment.This study also discovered a variety of specific ligand-receptor complexes in the immune-related microenvironment,providing a theoretical basis for targeting the interaction between malignant cells and macrophages in iCCA to enhance the immune response in the tumor.Our study not only reveals the composition of malignant cells and the complex heterogeneity of the iCCA tumor microenvironment but also provides evidence for immune suppression and tumor progression in iCCA,which helps to further understand the mechanisms associated with iCCA progression and develop effective immunotherapeutic drugs for iCCA patients. |
PDF Full Text Request