Spatial transcriptomic-metabolic features of tumor foci and tumor capsule in microvascular invasion with hepatocellular carcinoma: A spatial multi-omics study

Figures Abstract Background Microvascular invasion (MVI) is closely related to the recurrence and metastasis of hepatocellular carcinoma (HCC), but the underlying cellular mechanism remains largely elusive. This study aims to elucidate the regional cellular discrepancy between MVI-positive (MVI+) and MVI-negative (MVI−) HCC by integrating Spatial transcriptomics (ST) and spatial metabolomics (SM). Methods and findings ST and SM were performed on six tissue samples from four patients (including 2 MVI+, 2 MVI−, and 2 paratumor tissues), with the integration of 79 public single-cell RNA sequencing datasets of HCC. Patient identity was used as a covariate in the linear equation for regional differentially expressed gene analysis with the ST data. Clinical validation was conducted through multiplex immunofluorescence staining in 79 patients, together with external validation in the cancer genome atlas (TCGA)-liver hepatocellular carcinoma (LIHC) cohort (n = 299) and an independent microarray dataset (n = 62). For cell-type-specific metabolic profiling, spatial transcriptomic-metabolic registration was performed. The functional roles of key metabolites were further validated in vitro using inflammatory cancer-associated fibroblasts (iCAFs) derived from hepatic stellate cells (HSCs) and primary CAFs through co-culture models and various functional assays assessing cell proliferation, migration, and invasion. In the tumor lesion, a malignant STMN1+HMGN2+GPC3+ cell subtype enriched in MVI+ HCC was identified, which exhibited enhanced proliferative activity and was associated with poor prognosis. This finding was further confirmed in a local cohort of 79 patients, where multiplex immunofluorescence staining for the three genes (STMN1, HMGN2, and GPC3) showed significantly higher expression in the MVI+ group than in the MVI− group (p = 0.046). Integrated SM analysis further revealed that this cell population underwent metabolic reprogramming characterized by suppressed glycerolipid metabolism. In the tumor capsule, iCAFs-related genes were downregulated in MVI+ cases, and iCAFs were located distally from the tumor boundary. Spatial metabolite mapping showed a strong correlation between taurine and iCAFs, and functional assays demonstrated that taurine promotes HCC proliferation and migration by suppressing iCAF activity. One limitation of this study is the small sample size of spatial omics data, which hinders a more complete molecular functional analysis of the STMN1+HMGN2+GPC3+ cell subtype and iCAFs in MVI+ HCC. Larger-scale ST cohorts are required to further validate and expand the findings of this study. Conclusions This integrative spatial atlas proposes a hypothesis that there exists a highly proliferative and metabolically reprogrammed malignant cell subtype in the tumor lesion of MVI+ HCC, and that taurine in the tumor capsule modulates iCAF activity to influence tumor progression. The exploratory results provide mechanistic insights into MVI-related HCC progression and offer potential avenues for targeted therapeutic intervention of MVI+ HCC. Author summary Why was this study done? - Microvascular invasion is a high tumor stage subtype of hepatocellular carcinoma (HCC) and is often correlated with tumor metastasis and recurrence. - HCC has clear tumor foci and capsule areas (rich in immune cells and stromal cells) in the tissue region, but the differences in cellular functions between the microvascular invasion subtype and the common subtype in these tissue regions are not clear. - This study aims to use spatial omics technologies to obtain the gene expression and metabolite distribution information in different tissue regions, and explore the changes in cells and their functions in the tumor area and capsule area of microvascular invasion-positive HCC. What did the researchers do and find? - Using spatial transcriptomic and metabolomic data, combined with single-cell RNA sequencing, RNA sequencing gene expression data, and multiplex immunofluorescence staining, we identified and validated a specific subpopulation of STMN1+HMGN2+GPC3+ cancer cells within the microvascular invasion HCC tumor region. We found that this subpopulation exhibited enhanced proliferation and decreased glyceride metabolism. - Using spatial omics data, RNA sequencing, and microarray data, we found the genes associated with inflammatory cancer-associated fibroblasts are downregulated in the microvascular invasion-positive HCC capsule. Spatial omics data demonstrated a positive correlation between the abundance of these cells and taurine concentration, while cell-based experiments confirmed that taurine can suppress the activity of these cells, thereby promoting cancer cell growth. What do these findings mean? - STMN1+HMGN2+GPC3+ cancer cells may play a key role in microvascular invasion of HCC and may serve as potential targets for precision therapy. - Inflammatory cancer-associated fibroblast (iCAF) cells may also serve as a potential therapeutic target for the microvascular invasive with HCC, and taurine may be a possible key intervention metabolite. - Due to the small sample size and sample differences, the potential functional characteristics of STMN1⁺HMGN2⁺GPC3⁺ cancer cells and iCAFs may have not been fully detected. Citation: Luo Z-H, Wang N, Zhao J, Long F, Wu S, Zhong W, et al. (2026) Spatial transcriptomic-metabolic features of tumor foci and tumor capsule in microvascular invasion with hepatocellular carcinoma: A spatial multi-omics study. PLoS Med 23(5): e1004703. https://doi.org/10.1371/journal.pmed.1004703 Academic Editor: Ricky W. Johnstone, Peter MacCallum Cancer Centre, AUSTRALIA Received: June 18, 2025; Accepted: April 8, 2026; Published: May 15, 2026 Copyright: © 2026 Luo et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability: The code for all analyses presented in this paper is available on GitHub (https://github.com/Jingwe-Zhao/HCC_Spatial_Multiomics). Spatial transcriptomics data generated in this study have been deposited in the Genome Sequence Archive of the National Genomics Data Center under accession number HRA011344 (Retrieved from: https://ngdc.cncb.ac.cn/gsa-human/); More information on accessing the data can be found at https://ngdc.cncb.ac.cn/gsa-human/browse/HRA011344. The imzML data of ADADESI-MS generated in this study have been deposited in the OMIX database of the National Genomics Data Center under accession number OMIX006935 (Retrieved from: https://ngdc.cncb.ac.cn/omix/releaseList); More information on accessing the data can be found at https://ngdc.cncb.ac.cn/omix/release/OMIX006935. The public 79 HCC scRNA-Seq data were downloaded from Gene Expression Omnibus (GSE149614 and GSE242889) and National Genomics Data Center (HRA001748). The bulk RNA-seq or microarray data for 1229 cases of liver cancer from five independent cohorts of cancerous and paracancerous samples were used in survival analysis. These included the TCGA-LIHC (n = 299), ICGC-LIHC (n = 232), GSE14520 (n = 488), GSE40873 (n = 49) and GSE54236 (n = 161). MVI validation cohorts were TCGA-LIHC and GSE10141. The public spatial transcriptomics data were downloaded from HRA000437. Details of how to access Gene Expression Omnibus, TCGA, and ICGC are available from https://www.ncbi.nlm.nih.gov/geo/, https://www.cancer.gov/ccg/research/genome-sequencing/tcga and https://www.icgc-argo.org/. Funding: This work was supported by the National Natural Science Foundation of China, grant number No.12375349 and No.12175167 to FBW, and No.32400555 to ZHL. JJZ received grants from the Department of Science and Technology of Hubei Province (No. 2022EHB035) and Hubei Talent Program (No. 1180011). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing interests: The authors have declared that no competing interests exist. Abbreviations: AFADESI-MSI, Airflow-Assisted Desorption Electrospray Ionization-Mass Spectrometry Imaging; AFP, alpha-fetoprotein; apCAFs, antigen-presenting CAFs; APOE⁺, apolipoprotein E positive; CAFs, cancer-associated fibroblasts; CCK8, cell counting kit-8; CI, confidence interval; CNV, copy number variations; CXCL14⁺, C-X-C Motif Chemokine Ligand 14 positive; DCN, decorin; DEGs, differentially expressed genes; DFS, disease-free survival; DG, diacylglycerols; EDU, 5-ethynyl-2′-deoxyuridine; FFPE, formalin-fixed, paraffin-embedded; GO, Gene Ontology; GSVA, Gene Set Variation Analysis; HCC, hepatocellular carcinoma; HMDB, Human Metabolome Database; HR, hazard ratio; HSCs, hepatic stellate cells; iCAFs, inflammatory cancer-associated fibroblasts; KEGG, Kyoto Encyclopedia of Genes and Genomes; LIHC, liver hepatocellular carcinoma; LPA, lysophosphatidic acid; mCAFs, matrix CAFs; MIF, multiplex immunofluorescence; MS, mass spectrometry; MVI, microvascular invasion; MVI+, MVI-positive; MVI−, MVI-negative; NK, natural killer; OS, overall survival; PA, phosphatidic acid; PCA, principal component analysis; ppm, parts per million; qPCR, quantitative real-time polymerase chain reaction; SA, Spatially Aware; SASA, Spatially Aware Structurally Adaptive; scRNA-seq, single-cell RNA sequencing; SM, spatial metabolomics; SSC, Spatial Shrunken Centroids; ST, spatial transcriptomics; TACE, transarterial chemoembolization; TG, triglycerides; TME, tumor microenvironment; UGs, upregulated genes; UMAP, Uniform Manifold Approximation and Projection; vCAFs, vascular CAFs Introduction Hepatocellular carcinoma (HCC) is the most prevalent primary liver tumor in the world, ranking sixth in global incidence and third in cancer-specific mortality [1]. The development of effective systemic therapies, such as liver transplantation, surgery, and immunotherapy, has substantially improved the outcomes for patients with HCC [2,3]. However, approximately 70% of the patients have disease recurrence within five years, which often results from intrahepatic metastasis [4]. Microvascular invasion (MVI), defined by neoplastic invasion of peritumoral microvasculature, is a critical determinant of its early recurrence and adverse survival outcomes [5,6]. By facilitating tumor dissemination via both intrahepatic and extrahepatic routes, MVI contributes to the initiation and progression of metastasis, resulting in low disease-free survival (DFS) rates in patients with MVI-positive (MVI+) HCC [7]. Notably, adjuvant hepatic arterial infusion chemotherapy with FOLFOX can significantly improve the DFS of patients with MVI+ HCC, underscoring the need for individualized therapeutic strategies specifically targeting MVI [8]. Therefore, elucidation of the intricate mechanisms underlying MVI can not only improve our understanding of HCC pathogenesis but also guide the development of potential targeted and precise interventions. MVI represents a complex pathobiological phenomenon involving multifaceted cellular interactions and microenvironmental dynamics, particularly the reciprocal crosstalk between HCC cells and their stromal components [9]. Single-cell RNA sequencing (scRNA-seq) of MVI⁺ tumors has delineated distinctive molecular programs, which are characterized by dysregulated lipid metabolism, augmented angiogenic signaling, and increased proliferative capacity [7,9,10]. Notably, TREM2 ⁺ macrophages are engaged in Midkine (MDK)-mediated intercellular communication with malignant cells, constituting a critical mechanism driving MVI development and tumor progression [10]. Furthermore, Apolipoprotein E positive (APOE⁺) macrophages and decorin (DCN)-expressing cancer-associated fibroblasts (CAFs) have been identified as another type of stromal contributors to MVI pathogenesis [7,11]. Spatial localization is a fundamental determinant of cellular functionality and interaction patterns, with distinct niche architectures demonstrating potent clinical relevance for patient stratification and therapeutic response prediction. For instance, Chen and colleagues identified an immune niche, which consists of antitumoral macrophages, CD8⁺ T cells, and natural killer (NK) T cells and is correlated with immunotherapy outcomes in small cell lung cancer [12], while C-X-C Motif Chemokine Ligand 14 positive (CXCL14⁺) CAFs at tumor-stroma boundary facilitate immune exclusion through extracellular matrix remodeling that impedes T-cell infiltration [13]. However, the spatial architecture of cellular ecosystems and molecular features that drive MVI formation remain poorly characterized at spatial resolution in HCC. This study is hypothesis-generating and exploratory, focusing on spatial multi-omics data of MVI+ HCC. Our hypothesis was that MVI+ and MVI-negative (MVI−) HCC exhibit distinct cellular functional differences in both the tumor core region and the capsular region. These differences may be associated with specific cell types, as well as spatial distribution patterns of cells. Accordingly, this study has two main objectives: The primary objective was to identify whether there exist malignant cell populations in the tumor core of MVI+ HCC that are distinct from those in MVI− HCC, and to characterize their major transcriptional and metabolic features. The second objective was to investigate whether functionally altered cell populations exist in the tumor capsule region of MVI+ HCC, and whether such functional changes are associated with specific metabolites. Methods Human HCC samples collection Six independent HCC surgically resected specimens (including 2 MVI+ tumor boundary tissues, 2 MVI− tumor boundary tissues, and 2 paratumor tissues) were collected from four patients at Zhongnan Hospital of Wuhan University for spatial transcriptomics (ST) and spatial metabolomics (SM) analyses. In addition, 79 formalin-fixed, paraffin-embedded (FFPE) HCC samples (21 MVI+, 58 MVI−) from the same hospital were obtained for multiplex immunofluorescence (MIF) validation. Ethics statement 10× CytAssist ST sequencing Freshly collected HCC and distant adjacent tissues were embedded in OCT and snap-frozen on dry ice to preserve their integrity. The tissues were sliced into 10 µm thick sections and then placed in the 10× Visium spatial slides (6.5 × 6.5 mm) capture area. Each capture area possessed 5,000 spots, with a diameter of approximately 55 µm. Tissue sections were subjected to methanol fixation, H&E staining, imaging, and destaining, following the 10× Genomics recommended procedure (CG000614). Following the 10× Genomics experimental workflow (CG000495), probe hybridization and probe release were performed. Briefly, fresh-frozen liver sections were first mounted onto blank slides, followed by the addition of a predefined array of capture probes, each with a unique spatial barcode to overlay the tissue. Following tissue permeabilization and hybridization, mRNA molecules from the tissue were captured in situ by these probes. Subsequently, reverse transcription and amplification steps were performed to generate cDNA libraries, which were then spatially indexed based on the corresponding barcodes. The probes were then transferred to the 10× Visium CytAssist slide, and library construction was carried out using the Visium CytAssist Spatial Gene Expression for FFPE kit (PN-1000520). The resulting DNA libraries were subjected to high-throughput sequencing on Illumina NovaSeq. ST data processing The fastq reads of ST data were mapped to GRCh38 by Space Ranger (https://www.10xgenomics.com/support/software/space-ranger/latest). Then, the gene expression matrix, cell barcode, cell spatial locations and corresponding H&E images were obtained. The expression data were normalized by the SCTransform [14], and integrated by SCTIntegration function in Seurat [15]. Then, principal component analysis (PCA) was performed by RunPCA function. The first 30 principal components were used in Findclusters function (shared nearest neighbor clustering) with resolution parameter ranging from 0.1 to 1. Considering both the accuracy and redundancy of clustering, we chose resolution 0.4 as the final result. Finally, 13 clusters were obtained for the integrated ST samples. ST clusters were firstly manually annotated by the marker genes of hepatocyte genes HAMP, MT1X, and MT1G [16], immune/stromal genes of CD3D, IGHG1, COL1A1, and ACTA2, and malignant cell genes of LAMB3, SERPINA12, and TUBB [17–19]. Then, the annotation of clusters was refined by the cell deconvolution results and copy number variations (CNV). Spatial co-localization analysis The cell percentages identified through deconvolution analysis were then used in the spatial co-localization analysis by mistyR [20]. Specifically, the cell percentage was split into six independent subsets based on the samples. Then they were built with an internal view model. Finally, the standardized importance median of all samples was summarized and defined as the cell type dependency relationship in different spatial contexts. Based on the internal view, a heatmap of cell type interactions was drawn with cutoff = 0. Differentially expressed genes analysis of ST According to cluster annotation, we separated tumor region into four sections: Tumor (T) region (cluster ST_C1, ST_C6, ST_C8, ST_C10, ST_C12, ST_C13), tumor capsule (TC) region (cluster ST_C3 and ST_C5), paratumoral (PT) region (cluster ST_C2, ST_C4, ST_C9 and ST_C11), and transition state (TR) region (cluster ST_C7). Cluster ST_C4 in patient P1 possessed CNV scores and was located inside the cancer lesion according to H&E staining images. Therefore, it was classified as the T region. Differentially expressed genes (DEGs) analysis was performed using the devil R-package [21] which applies a Bayesian Gamma–Poisson framework to restore patient-level independence and ensure statistically rigorous testing. For this analysis, spots from the T/TC region were considered. Raw counts were modeled with the fit_devil function, specifying MVI status (MVI+ or MVI−) as the main grouping variable in the design matrix. To address the issue of pseudo-replication inherent in spot-level analyses, patient identity was explicitly incorporated as the unit of replication by passing it to the test_de function through the “clusters” parameter. Differential expression between the MVI+ and MVI− groups was then assessed with test_de (setting max_lfc = 50), and statistical significance was adjusted using the Benjamini–Hochberg procedure. Genes were considered significantly differentially expressed if they met the criteria of adjusted P value 1. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were enriched for DEGs through clusterProfiler [22]. Triple-positive gene selection Firstly, we performed DEGs analysis between cluster SC_C6/SC_C7 and SC_C0/SC_C3/SC_C4/SC_C15 using Seurat. Among the top 30 DEGs, we selected the genes with high pct.1 (> 0.8). Then, we obtained STMN1, TUBA1B, H2AFZ, HMGN2, and TUBB as the dominant highly expression genes in cluster SC_6/SC_7. Secondly, we used the difference between pct.1 and pct.2 to measure the gene specificity in cluster SC_C6/SC_C7. We found that STMN1 was the gene with the highest specificity in cluster SC_C6/SC_C7 (S7 Table). Thirdly, we used the Gepia2 website (http://gepia2.cancer-pku.cn/#index) to search for genes with similar expression patterns to STMN1 in the cancer genome atlas (TCGA)-liver hepatocellular carcinoma (LIHC) data and found that HMGN2 ranked second in correlation with STMN1 among all genes. Therefore, we selected HMGN2 as an auxiliary gene of STMN1 to help us distinguish cluster SC_C6/SC_C7 from other subclusters. Fourth, considering that STMN1 and HGMN2 reflect proliferative and chromatin-remodeling programs that may also occur in non-malignant proliferative cells [23]. We incorporated GPC3 to enhance hepatocellular carcinoma specificity. GPC3 is a well-established HCC marker with minimal expression in non-tumor cells (Fig 5c). The final marker panel was therefore defined as STMN1⁺/HMGN2⁺/GPC3⁺ cells, integrating data-driven discovery (STMN1, HMGN2) with a priori tumor-specific knowledge (GPC3). a, Uniform Manifold Approximation and Projection (UMAP) visualization of spot clusters for integrated samples. b, Heatmap showing the average expression of cell markers in spot clusters. c–e, Overview of the spatial transcriptomic spot clusters for MVI− (c), MVI+ (d), and normal tissues (e). f, Stacked plot showing the proportion of each cluster in all samples. g, Scaled median cell-type compositions within each cluster. “*” represents a higher proportion of one cell type in a cluster compared with other clusters (Wilcoxon rank-sum test, p.adjust 0.05. MVI+, microvascular invasion positive; MVI−, microvascular invasion negative; SM, spatial metabolomics; T, tumor region; PT, paratumor region; TC, tumor capsule region; TR, transition state region; OR, odds ratio; CI, confidence interval. HCC tumor boundary and boundary distance For the ST data, each spot was spatially surrounded by six spots. For the spots inside the cancer lesion, the annotations of the six surrounding spots also belong to the TC area. For the spots at the boundary of the cancer lesion, at least one of the six spots around them did not belong to the TC area. We divided the spots in the TC area into tumor core spots and tumor boundary spots according to this rule. For non-tumor spots, we calculated the Euclidean distance between the spot and each boundary spot, and defined the shortest distance as the spatial distance from the spot to the boundary. This distance was then used in correlation calculations with the gene set scores. Gene signature score in ST data and TCGA RNA-Seq Diacylglycerol biosynthesis process and the triglyceride biosynthetic process gene sets were downloaded from Gene Ontology (GO) database (https://www.geneontology.org/). CAFs gene signatures were derived from previous studies [24], and all significant genes (log2 fold change >0.25 and P value 1, adjusted P value 0.25, P value 0, P value 0.25, adjusted P value 200 µm2 were excluded to minimize artifacts. For quantification, positive cells were identified by thresholding, and co-localization was assessed at the single-cell level based on overlapping fluorescence signals. The following metrics were calculated: - (1). Triple-positive ratio (primary endpoint): - (2). Double-positive ratios: Supplementary methods Detailed descriptions of additional analytical and experimental methods are provided in supplementary methods sections of S1 File. These include CNV inference, cell type deconvolution, cell interaction analysis, cell-type specific metabolites analysis, cell culture, fibroblast isolation, immunofluorescence, western blotting, RNA extraction and quantitative real-time polymerase chain reaction (qPCR) analysis, cell counting kit-8 (CCK8) assay, 5-Ethynyl-2′-deoxyuridine (EDU) assay, co-culture, migration and invasion assays, and enzyme-linked immunosorbent assay (ELISA). Results Spatial transcriptomic and metabolomic features of MVI+ and MVI− HCC HCC exhibits a distinct spatial structural organization. To explore the biological differences in tissue architecture between MVI+ and MVI− HCC, we employed spatial transcriptomic and metabolomic data to annotate their tissue structures and examine the basic distribution patterns and characteristics of cells and metabolites within these regions (S1 Fig). Spatial transcriptomic map of MVI+ and MVI− HCC To characterize the ST landscape of MVI+ and MVI− HCC tumors, we performed ST sequencing on four tumor tissues (two for MVI+, two for MVI−) from tumor boundary tissues and two distant peritumoral tissues (S1 Table). The analysis encompassed 21,920 spots with an average detection of 6,945 genes per sample. After uniform normalization and dimensionality reduction, 13 distinct transcriptional clusters were identified by the shared nearest neighbor clustering analysis (Fig 1a and 1b). These clusters demonstrated heterogeneous expression of some established hepatocyte markers (HAMP, MT1X, MT1G) [16], immune/stromal markers (CD3D, IGHG1, COL1A1, ACTA2), and HCC malignancy-associated markers (LAMB3, SERPINA12, TUBB) [17–19]. Annotation of the T regions in H&E images by pathology experts revealed distinct differences in the composition of clusters between MVI+ and MVI− HCC. The T region of MVI− contained ST_C1, ST_C6, and ST_C8, while that of MVI+ contained ST_C1, ST_C10, and ST_C12 (Fig 1c–1e). Spatial mapping demonstrated compartmentalized distribution patterns of canonical hepatocyte and tumor microenvironment (TME) markers (S2 Fig), reflecting high-degree cellular regionalization within HCC specimens. As each spot in ST contained multiple cells, it is necessary to clarify the cellular composition of these transcriptional clusters. Cell-type deconvolution was performed using public HCC scRNA-seq data [16] via the Robust Cell-Type Decomposition (RCTD) algorithm [34] (S1 File). Quantitative estimation of cellular proportions within the 13 transcriptional clusters revealed distinct distributions of hepatocytes, malignant hepatocytes, CAFs, endothelial cells, B lymphocytes, T/ NK cells, and myeloid populations in different clusters (S3 Fig). For example, cluster ST_C3 was primarily composed of stromal and immune cells, whereas cluster ST_C10 contained only cancer cells. Inter-sample analysis demonstrated substantial heterogeneity in cellular composition, with the P2 tumor specimen exhibiting obvious myeloid infiltration compared with other specimens (Figs 1f and S3), indicating patient-specific immunological variation. To explore the cell type bias of different clusters, hierarchical clustering was performed using the cellular components in the clusters. Clusters ST_C3 and ST_C5 exhibited heterogeneous cellular admixtures dominated by CAFs, accompanied by diverse endothelial and immune components (Figs 1g and S3). These two clusters were located in the TC region (Fig 1c–1e), indicating that this region is a mixed region of stromal cells and immune cells. Besides TC clusters, the hierarchical clustering revealed other two distinct cellular ecosystems, namely normal hepatocyte-enriched clusters (Clusters ST_C2, ST_C4, ST_C9, ST_C11) and tumor cell-dominant clusters (Clusters ST_C1, ST_C6, ST_C8, ST_C10, ST_C12, ST_C13) (Fig 1g). Furthermore, a spatial colocalization analysis was conducted to investigate the associations among cellular components within the tissue. The spatial dependencies between CAFs, endothelial/immune cells, and tumor cells suggested that tumor cell growth may be influenced by the surrounding microenvironment (Fig 1h). By integrating these findings with histomorphological information from H&E staining, the clusters could be classified into three pathologically defined regions for subsequent analyses, including the T region (Clusters ST_C1, ST_C6, ST_C8, ST_C10, ST_C12, ST_C13), the TC region (Clusters ST_C3, ST_C5), and the PT region (Clusters ST_C2, ST_C4, ST_C9, ST_C11) (Fig 1i). According to the above regional delineation, only cluster ST_C4 was present in both the T (in P1) and the PT (in P2, P3, P4) region. To examine the region annotation of these clusters, a CNV analysis was conducted across all clusters using the inferCNV [35] (S1 File). In Patient P1, clusters ST_C2 and ST_C9 demonstrated lower CNV scores than clusters ST_C1, ST_C4, ST_C6, and ST_C8 (S4a and S4e Fig). Notably, in Patients P2–P4, clusters ST_C2, ST_C4, and ST_C9 exhibited attenuated CNV scores relative to tumor-dominant clusters (Clusters ST_C1, ST_C6, ST_C8, ST_C10, ST_C12, ST_C13; S4b–S4e Fig). These differential CNV patterns revealed patient-specific malignant composition, where cluster ST_C4 belonged to the T region in patient P1, but was located in the PT region in other patients, while the division of other clusters was consistent between cellular deconvolution and CNV analysis. Therefore, we corrected the division of the T region and classified cluster ST_C4 in P1 into the T region. Cluster ST_C7, which was uniquely identified in Patient P3 (Fig 1d), displayed a gene expression pattern different from both the T and PT region. This cluster contained both normal hepatocytes and tumor cells (S4f Fig), suggesting its role as a transitional state between normal and malignant phenotypes. Given this ambiguous biological status, Cluster ST_C7 was excluded from subsequent comparative transcriptomic analyses between MVI+ and MVI− HCC. Normal hepatocyte identity was confirmed for clusters ST_C1, ST_C2, ST_C4, ST_C6-13 in control specimens N1 and N3. Finally, spatial mapping identified distinct regions in all samples with TC serving as a structural interface between the T and PT regions (Fig 1i). Spatial metabolomics to elucidate the regional distribution of metabolites Variations in transcriptional function are often accompanied by metabolic reprogramming. To systematically investigate the metabolic differences between MVI+ and MVI− HCC, SM was conducted on six tissue specimens. This high-resolution approach detected 719 positive and 408 negative metabolites with molecular weights ranging from 74.09 to 1189.99 m/z and 73.03 to 1041.81 m/z, respectively (S3 Table). To investigate the spatial distribution patterns of metabolites, a clustering analysis was conducted on the metabolite profiles. SSC clustering revealed 15 distinct spatial clusters, where both positive and negative metabolites could successfully reconstruct the tissue architecture as visualized by H&E-stained images (Figs 2a, 2b, S5a, and S5b). This spatial distribution pattern demonstrated the region-specific aggregation of metabolites in HCC. Hierarchical clustering of metabolites identified differential intensity patterns for positive metabolites across different clusters (Fig 2c; S4 Table; S1 File), whereas negative metabolites showed ambiguous inter-cluster variations (S5c Fig) and thus were excluded in further analyses. The above results indicated that the composition of metabolites varies substantially across different clusters. Spatial mapping revealed distinct metabolic regions: clusters SM_C2, SM_C10, SM_C13, and SM_C15 were preferentially localized in the T region of MVI– but in the PT region of MVI+; while clusters SM_C9 and SM_C11 occupied the TC region across both phenotypes (Fig 2b–2d). Given this spatial segregation, subsequent analysis would be focused on the T-region metabolic disparities and TC-region conserved signatures. The metabolites in clusters SM_C2, SM_C10, SM_C13, and SM_C15 showed significant enrichment (Fisher’s exact test, P 0.05. SM, spatial metabolomics; OR, odds ratio; CI, confidence interval; MVI+, microvascular invasion positive; MVI−, microvascular invasion negative. https://doi.org/10.1371/journal.pmed.1004703.s007 (TIF) S6 Fig. Multiplex immunofluorescence of marker genes. a, Correlation between STMN1 and HMGN2 genes. TPM, transcripts per million. b, Overall survival of STMN1 and HMGN2 genes score in HCC cohort. c, Violin plot of STMN1 and HMGN2 genes expression scores of tumor stages, d and e, Multiplex immunofluorescence of STMN1+GPC3+ (d) and HMGN2+GPC3+ (e) cells in MVI+ and MVI− samples. HR, hazard ratio; MVI−, microvascular invasion negative; MVI+, microvascular invasion positive. https://doi.org/10.1371/journal.pmed.1004703.s008 (TIF) S7 Fig. Ligand-receptor difference between STMN1+HMGN2+GPC3+ cell subtype and TME cells. a, Bubble Heatmap showing the mean interaction strength between STMN1+HMGN2+GPC3+ cell subtype and other tumor cells for ligand-receptor pairs. Dot color indicated the mean interaction strength levels. b, Interaction strength/weights bewteen STMN1+HMGN2+GPC3+ cell subtype, other tumor cells and microenvironment cells. c, CypA signaling strength between the sender and receiver cells. d, Spatial plots of PPIA and BSG genes expression in P3 tissue. T, tumor region; PT, paratumor region; TC, tumor capsule region; and TR, transition state region. PPIA, Peptidylprolyl Isomerase A; BSG, Basigin. e, Immunofluorescence images of PPIA and BSG expression in MVI− and MVI+ samples. f and g, The expression of PPIA (f) and BSG (g) in MVI+, MVI− and different tumor grades, and their relationship with patient prognosis. NK, natural killer; OS, overall survival; HR, hazard ratio; MVI−, microvascular invasion negative; MVI+, microvascular invasion positive. Statistical analysis was performed using the Student t test (f and g). https://doi.org/10.1371/journal.pmed.1004703.s009 (TIF) S8 Fig. TG biosynthesis decreased in STMN1+HMGN2+GPC3+ cell subtype. a–c, “triglyceride biosynthetic process” pathway activity comparison between MVI+ and MVI− tumors in ST T region (a), TCGA-LIHC RNA-Seq (b) data and STMN1+HMGN2+GPC3+ cell subtype clusters (c). d, Manual region division of spatial metabolome H&E images (left) and metabolites intensity boxplot between regions of P1 and P4 (right).ST, spatial transcriptomics; PA, phosphatidic acid; DG, diacylglycerol; TG, triglyceride; T, tumor region; PT, paratumor region; TC, tumor capsule region; MVI−, microvascular invasion negative; MVI+, microvascular invasion positive; TCGA, the cancer genome atlas; LIHC, liver hepatocellular carcinoma. Statistical analysis was performed using the Student t test (a–c). https://doi.org/10.1371/journal.pmed.1004703.s010 (TIF) S9 Fig. Spatial plots of CAFs subpopulations. a, Spatial plots of gene signature score for vCAF, mCAF, iCAFs and apCAFs in P1, P2 and P4 tissue (left) and spot distance to the tumor boundary (right). vCAF, vascular cancer-associated fibroblast; mCAF, myofibroblastic cancer-associated fibroblast; iCAF, inflammatory cancer-associated fibroblast; apCAF, antigen-presenting cancer-associated fibroblast; MVI−, microvascular invasion negative; MVI+, microvascular invasion positive. b, Spatial plots of gene signature scores for vCAF, mCAF, iCAFs and apCAFs in patient P3. c, Euclidean distance of spots in the TC region to the tumor boundary. T, tumor region; PT, paratumor region; TC, tumor capsule region; and TR, transition state region. d, Spearman correlation between CAFs signature scores and the distance of spots to the tumor boundary. https://doi.org/10.1371/journal.pmed.1004703.s011 (TIF) S10 Fig. Taurine promotes HCC growth by suppressing iCAFs. a, Volcano plot showing the 166 differential metabolites (DMs) between the TC region and other regions. TC, tumor capsule sections; ns, no significance. b, Heatmap of hierarchical clustering of DMs across four tumor samples. c, Bar plot of KEGG pathways enriched in 36 DMs. d, Heatmap showing the Pearson correlation between 36 metabolites and cell proportions of the TC region. DEM, differential expression metabolites; NK, Natural killer. e, Heatmaps showing the mean of Pearson correlation between taurine (Ta) intensity and signature scores of four CAF subpopulations across four tumor samples. CAFs, cancer-associated fibroblasts; vCAFs, vascular CAFs; mCAFs, matrix CAFs; iCAFs, inflammatory CAFs; apCAFs, antigen-presenting CAFs. f, Effects of taurine on iCAFHSC-NCCO proliferation evaluated using EDU assay. EDU, 5-ethynyl-2′-deoxyuridine; HSC, hepatic stellate cell; NCCO, non-contact co-culture. g, Transwell co-culture assay was used to detect the effect of iCAFHSC-NCCO with or without taurine on tumor cell migration and invasion. Statistical analysis was performed using the Student t test (f and g). https://doi.org/10.1371/journal.pmed.1004703.s012 (TIF) S11 Fig. Metabolites distribution and Schematic diagram of the image registration. a, Cluster distribution of Set 1 metabolites. b, spatial image of set 1 metabolite example. c, Cluster distribution of Set 2 metabolites. d, Schematic diagram of the image registration between ST and SM H&E images. ST, spatial transcriptomics; MSI, mass spectrometry imaging. SM, spatial metabolomics; H&E, hematoxylin and eosin staining; MVI+, microvascular invasion positive. https://doi.org/10.1371/journal.pmed.1004703.s013 (TIF) S12 Fig. Spatial correlation between metabolite intensity and cell proportion. a–d, The heatmap demonstrates the Pearson correlation between the intensity of 36 metabolites of Set 2 and the proportion of cell types in P1 (a), P2 (b), P3 (c), and P4 (d) samples. NK, natural killer; MVI−, microvascular invasion negative, MVI+, Microvascular invasion positive. https://doi.org/10.1371/journal.pmed.1004703.s014 (TIF) S13 Fig. TC region differential metabolites correlation with CAFs subpopulations in tumor samples. a, Heatmap of the mean of Pearson correlation between DEMs and CAFs subpopulation across 4 tumor samples. b, SM images of 4 CAFs related metabolites (C24H27ClN4O6, C7H9N2O, C4H5NO3S and C43H78NO7P) in all 6 samples. c, Taurine (C2H7NO3S) intensity distribution in SM images for all 6 samples. d, Pearson correlation between taurine intensity and the distance of spots to tumor boundary. DEM, differential expression metabolites; SM, spatial metabolomics; CAFS, cancer-associated fibroblasts; vCAF, vascular cancer-associated fibroblast; mCAF, myofibroblastic cancer-associated fibroblast; iCAF, inflammatory cancer-associated fibroblast; apCAF, antigen-presenting cancer-associated fibroblast; MVI−, microvascular invasion negative; MVI+, microvascular invasion positive. https://doi.org/10.1371/journal.pmed.1004703.s015 (TIF) S14 Fig. Induction of the iCAF Phenotype. a–c, Immunofluorescence (a) and western blot (b and c) analyses confirmed the expression of α-SMA and collagen 1 (COL1) in mCAFs induced by TGF-β, and in iCAFs induced through non-contact co-culture (NCCO) with HSCs. d, qPCR detected the expression of iCAF biomarkers IGF-1, CXCL2, and C7 following non-contact co-culture with HSCs. e, ELISA measured the concentrations of IGF-1, CXCL2, and C7 in the supernatant, further confirming iCAF induction by HSC co-culture. f and g, qPCR and ELISA assays were performed to measure the expression levels of iCAF biomarkers IGF-1, CXCL2, and C7 following NCCO induction of primary CAFs. α-SMA, α-smooth muscle actin; TGF-β, transforming growth factor beta; HSC, hepatic stellate cell; qPCR, quantitative polymerase chain reaction; IGF-1, insulin-like growth factor 1; CXCL2, C-X-C motif chemokine ligand 2; C7, complement component 7; ELISA, enzyme-linked immunosorbent assay; CAF, cancer-associated fibroblast; iCAF, inflammatory cancer-associated fibroblast; mCAF, myofibroblastic cancer-associated fibroblast; HSCs, hepatic stellate cells; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; pCAF, primary cancer-associated fibroblast. Statistical analysis was performed using the Student t test (d–g). https://doi.org/10.1371/journal.pmed.1004703.s016 (TIF) S15 Fig. Taurine suppressing the proliferation of iCAFs. a, The effects of taurine (Ta) on iCAFHSC-NCCO proliferation were evaluated using CCK-8 assay. b, western blot analysis of the effect of taurine on α-SMA and COL1 expression in iCAFsHSC-NCCO. c and d, qPCR (c), and ELISA assays (d) were utilized to determine the expression levels of iCAF markers IGF-1, CXCL2, and C7 following taurine treatment. iCAF, inflammatory cancer-associated fibroblast; HSC, hepatic stellate cell; NCCO, non-contact co-culture; CCK-8, cell counting Kit-8; qPCR, quantitative polymerase chain reaction; IGF-1, insulin-like growth factor 1; CXCL2, C-X-C motif chemokine ligand 2; C7, complement component 7; ELISA, enzyme-linked immunosorbent assay; Ctrl, control; OD, optical density; COL1, collagen I; α-SMA, α-smooth muscle actin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Statistical analysis was performed using the Student t test (a, c, and d). https://doi.org/10.1371/journal.pmed.1004703.s017 (TIF) S16 Fig. Parallel verification that taurine does not affect tumor cell proliferation. a–f, The influence of taurine (Ta) on the proliferation of Huh7 (a and b), HepG2 (c and d), and Hep3B (e and f) tumor cells was assessed using EDU and CCK-8 assay. CCK-8: Cell Counting Kit-8, EDU: 5-ethynyl-2′-deoxyuridine, Ctrl, control. Statistical analysis was performed using the Student t test (a–f). https://doi.org/10.1371/journal.pmed.1004703.s018 (TIF) S17 Fig. Primary CAFs verified that taurine promoted tumor cell migration and invasion by suppressing iCAFs. a, The impact of taurine (Ta) on iCAFpCAF-NCCO proliferation was determined using the EDU assay. b, western blot analysis of the effect of taurine on α-SMA and COL1 expression in iCAFs pCAF-NCCO. c and d, qPCR (c), and ELISA assays (d) confirmed the expression of IGF-1, CXCL2, and C7 in iCAFpCAF-NCCO after taurine treatment. e, Transwell co-culture assays examined the effect of iCAFpCAF-NCCO on tumor cells migration and invasion with or without taurine. CAF, cancer-associated fibroblast; iCAF, inflammatory cancer-associated fibroblast; pCAF, primary cancer-associated fibroblast; NCCO, non-contact co-culture; EDU, 5-ethynyl-2′-deoxyuridine; qPCR, quantitative polymerase chain reaction; IGF-1, insulin-like growth factor 1; CXCL2, C-X-C motif chemokine ligand 2; C7, complement component 7; ELISA, enzyme-linked immunosorbent assay; COL1, collagen I; α-SMA, α-smooth muscle actin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase. Statistical analysis was performed using the Student t test (a, c, d and e). https://doi.org/10.1371/journal.pmed.1004703.s019 (TIF) S1 Table. Clinical information for each patient. N1 tissue is from patient P1. N3 tissue is from patient P3. MVI, microvascular invasion; HCC, hepatocellular carcinoma. https://doi.org/10.1371/journal.pmed.1004703.s020 (XLSX) S2 Table. Exact P values corresponding to Fig 1g. https://doi.org/10.1371/journal.pmed.1004703.s021 (XLSX) S3 Table. Positive and negative ion metabolites annotation by m/z. ppm, part per million; HMDB, human metabolome database; KEGG, kyoto encyclopedia of genes and genomes; CAS, chemical abstracts service. https://doi.org/10.1371/journal.pmed.1004703.s022 (XLSX) S4 Table. Positive ion metabolites clusters identified by spatial shrunken centroids clustering analysis. The shrunken centroid scores indicate the distinctive strength or representativeness of each metabolite within the corresponding cluster. https://doi.org/10.1371/journal.pmed.1004703.s023 (XLSX) S5 Table. Differential gene expression analysis between MVI-positive and MVI-negative samples within the T region. Patient identity was used as a covariate in the linear equation for regional differentially expressed gene analysis. https://doi.org/10.1371/journal.pmed.1004703.s024 (XLSX) S6 Table. KEGG pathway enrichment for DEGs of T region between HCC MVI+ and MVI− tumors. https://doi.org/10.1371/journal.pmed.1004703.s025 (XLSX) S7 Table. DEGs of Scissor positive clusters SC_C6 and SC_C7 versus Scissor negative clusters in 79 scRNA-Seq malignant cells. https://doi.org/10.1371/journal.pmed.1004703.s026 (XLSX) S8 Table. Differential expression metabolites upregulated in the TC region. HMDB, human metabolome database; KEGG, kyoto encyclopedia of genes and genomes. https://doi.org/10.1371/journal.pmed.1004703.s027 (XLSX) S9 Table. Spatial correlation between DEMs in TC region and cell type proportion. DEM, differential expression metabolites. https://doi.org/10.1371/journal.pmed.1004703.s028 (XLSX) Acknowledgments We thank Professors Feng Li, Qin Pan, and Jianbo Tian for their insightful guidance and critical feedback throughout the project. We are grateful to the Hepatobiliary and Pancreatic Surgery of Zhongnan Hospital for their invaluable assistance with clinical sample collection and processing. We thank Professor Zuoxiong Liu for his valuable language polishing of this manuscript. We thank Shanghai OE Biotech Co., (Shanghai, China) for the AFADESI spatial resolved metabolomics used in this study. We also acknowledge Zhenyu Xu and Yuan Yuan for their important suggestions and valuable technical support. References - 1. Li Q, Ding C, Cao M, Yang F, Yan X, He S. Global epidemiology of liver cancer 2022: an emphasis on geographic disparities. Chin Med J. 2024;137(19):2334–42. pmid:39227359 - 2. Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet. 2018;391(10127):1301–14. pmid:29307467 - 3. Tabrizian P, Marino R, Chow PKH. Liver resection and transplantation in the era of checkpoint inhibitors. JHEP Rep Innov Hepatol. 2024;6(11):101181. pmid:39741696 - 4. Llovet JM, Pinyol R, Yarchoan M, Singal AG, Marron TU, Schwartz M, et al. Adjuvant and neoadjuvant immunotherapies in hepatocellular carcinoma. Nat Rev Clin Oncol. 2024;21(4):294–311. pmid:38424197 - 5. Lim K-C, Chow PK-H, Allen JC, Chia G-S, Lim M, Cheow P-C, et al. Microvascular invasion is a better predictor of tumor recurrence and overall survival following surgical resection for hepatocellular carcinoma compared to the Milan criteria. Ann Surg. 2011;254(1):108–13. pmid:21527845 - 6. Lee S, Kang TW, Song KD, Lee MW, Rhim H, Lim HK, et al. Effect of microvascular invasion risk on early recurrence of hepatocellular carcinoma after surgery and radiofrequency ablation. Ann Surg. 2021;273(3):564–71. pmid:31058694 - 7. Wang Y, Zhu G-Q, Yang R, Wang C, Qu W-F, Chu T-H, et al. Deciphering intratumoral heterogeneity of hepatocellular carcinoma with microvascular invasion with radiogenomic analysis. J Transl Med. 2023;21(1):734. pmid:37853415 - 8. Li S-H, Mei J, Cheng Y, Li Q, Wang Q-X, Fang C-K, et al. Postoperative adjuvant hepatic arterial infusion chemotherapy with FOLFOX in hepatocellular carcinoma with microvascular invasion: a multicenter, phase III, randomized study. J Clin Oncol. 2023;41(10):1898–908. pmid:36525610 - 9. Shi X, Zhao Y, Li K, Li Q, Cui Y, Sui Y, et al. Plasma proteomic signature for preoperative prediction of microvascular invasion in HCC. JHEP reports: innovation in hepatology. 2025;7(9):101481. https://doi.org/10.1016/j.jhepr.2025.101481 40823177. - 10. Li K, Zhang R, Wen F, Zhao Y, Meng F, Li Q, et al. Single-cell dissection of the multicellular ecosystem and molecular features underlying microvascular invasion in HCC. Hepatology. 2024;79(6):1293–309. pmid:37972953 - 11. Zheng X, Wang P, Li L, Yu J, Yu C, Xu L, et al. Cancer-associated fibroblasts promote vascular invasion of hepatocellular carcinoma via downregulating decorin-integrin β1 signaling. Front Cell Dev Biol. 2021;9:678670. pmid:34504839 - 12. Chen H, Deng C, Gao J, Wang J, Fu F, Wang Y, et al. Integrative spatial analysis reveals tumor heterogeneity and immune colony niche related to clinical outcomes in small cell lung cancer. Cancer Cell. 2025;43(3):519-536.e5. pmid:39983726 - 13. Feng Y, Ma W, Zang Y, Guo Y, Li Y, Zhang Y, et al. Spatially organized tumor-stroma boundary determines the efficacy of immunotherapy in colorectal cancer patients. Nat Commun. 2024;15(1):10259. pmid:39592630 - 14. Choudhary S, Satija R. Comparison and evaluation of statistical error models for scRNA-seq. Genome Biol. 2022;23(1):27. pmid:35042561 - 15. Hao Y, Hao S, Andersen-Nissen E, Mauck WM 3rd, Zheng S, Butler A, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184(13):3573-3587.e29. pmid:34062119 - 16. Lu Y, Yang A, Quan C, Pan Y, Zhang H, Li Y, et al. A single-cell atlas of the multicellular ecosystem of primary and metastatic hepatocellular carcinoma. Nat Commun. 2022;13(1):4594. pmid:35933472 - 17. Pan T, Li J, Zhang O, Zhu Y, Zhou H, Ma M, et al. Knockdown of ribosome RNA processing protein 15 suppresses migration of hepatocellular carcinoma through inhibiting PATZ1-associated LAMC2/FAK pathway. BMC Cancer. 2024;24(1):334. pmid:38475740 - 18. Yu H, Zhou L, Loong JHC, Lam K-H, Wong T-L, Ng K-Y, et al. SERPINA12 promotes the tumorigenic capacity of HCC stem cells through hyperactivation of AKT/β-catenin signaling. Hepatology. 2023;78(6):1711–26. pmid:36630996 - 19. Wang X, Shi J, Huang M, Chen J, Dan J, Tang Y, et al. TUBB2B facilitates progression of hepatocellular carcinoma by regulating cholesterol metabolism through targeting HNF4A/CYP27A1. Cell Death Dis. 2023;14(3):179. pmid:36872411 - 20. Tanevski J, Flores ROR, Gabor A, Schapiro D, Saez-Rodriguez J. Explainable multiview framework for dissecting spatial relationships from highly multiplexed data. Genome Biol. 2022;23(1):97. pmid:35422018 - 21. Santacatterina G, Tosato N, Milite S, Davydzenka K, Insaghi E, Cozzini S. Scalable, fast and accurate differential gene expression testing from millions of cells of multiple patients. bioRxiv. 2025. - 22. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z. clusterProfiler 4.0: a universal enrichment tool for interpreting omics data. Innovation (Camb). 2021;2(3):100141. pmid:34557778 - 23. Xue R, Zhang Q, Cao Q, Kong R, Xiang X, Liu H, et al. Liver tumour immune microenvironment subtypes and neutrophil heterogeneity. Nature. 2022;612(7938):141–7. pmid:36352227 - 24. Cords L, Tietscher S, Anzeneder T, Langwieder C, Rees M, de Souza N, et al. Cancer-associated fibroblast classification in single-cell and spatial proteomics data. Nat Commun. 2023;14(1):4294. pmid:37463917 - 25. Sun D, Guan X, Moran AE, Wu L-Y, Qian DZ, Schedin P, et al. Identifying phenotype-associated subpopulations by integrating bulk and single-cell sequencing data. Nat Biotechnol. 2022;40(4):527–38. pmid:34764492 - 26. Roessler S, Jia H-L, Budhu A, Forgues M, Ye Q-H, Lee J-S, et al. A unique metastasis gene signature enables prediction of tumor relapse in early-stage hepatocellular carcinoma patients. Cancer Res. 2010;70(24):10202–12. pmid:21159642 - 27. Kudo A, Mogushi K, Takayama T, Matsumura S, Ban D, Irie T, et al. Mitochondrial metabolism in the noncancerous liver determine the occurrence of hepatocellular carcinoma: a prospective study. J Gastroenterol. 2014;49(3):502–10. pmid:23543312 - 28. Villa E, Critelli R, Lei B, Marzocchi G, Cammà C, Giannelli G, et al. Neoangiogenesis-related genes are hallmarks of fast-growing hepatocellular carcinomas and worst survival. Results from a prospective study. Gut. 2016;65(5):861–9. pmid:25666192 - 29. Luo Z, He J, Chen Y, He J, Gong T, Tang F, et al. Air flow-assisted ionization imaging mass spectrometry method for easy whole-body molecular imaging under ambient conditions. Anal Chem. 2013;85(5):2977–82. pmid:23384246 - 30. Race AM, Styles IB, Bunch J. Inclusive sharing of mass spectrometry imaging data requires a converter for all. J Proteomics. 2012;75(16):5111–2. pmid:22641155 - 31. Bemis KD, Harry A, Eberlin LS, Ferreira C, van de Ven SM, Mallick P, et al. Cardinal: an R package for statistical analysis of mass spectrometry-based imaging experiments. Bioinformatics. 2015;31(14):2418–20. pmid:25777525 - 32. Alexandrov T, Kobarg JH. Efficient spatial segmentation of large imaging mass spectrometry datasets with spatially aware clustering. Bioinformatics. 2011;27(13):i230-8. pmid:21685075 - 33. Tibshirani R, Hastie T, Narasimhan B, Chu G. Class prediction by nearest shrunken centroids, with applications to DNA microarrays. Statist Sci. 2003;18(1). - 34. Cable DM, Murray E, Zou LS, Goeva A, Macosko EZ, Chen F, et al. Robust decomposition of cell type mixtures in spatial transcriptomics. Nat Biotechnol. 2022;40(4):517–26. pmid:33603203 - 35. Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM, Wakimoto H, et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science. 2014;344(6190):1396–401. pmid:24925914 - 36. Erstad DJ, Tanabe KK. Prognostic and therapeutic implications of microvascular invasion in hepatocellular carcinoma. Ann Surg Oncol. 2019;26(5):1474–93. pmid:30788629 - 37. Wang W, Guo Y, Zhong J, Wang Q, Wang X, Wei H, et al. The clinical significance of microvascular invasion in the surgical planning and postoperative sequential treatment in hepatocellular carcinoma. Sci Rep. 2021;11(1):2415. pmid:33510294 - 38. Ma C, Yang C, Peng A, Sun T, Ji X, Mi J, et al. Pan-cancer spatially resolved single-cell analysis reveals the crosstalk between cancer-associated fibroblasts and tumor microenvironment. Mol Cancer. 2023;22(1):170. pmid:37833788 - 39. Lahan-Martins D, Perales SR, Gallani SK, da Costa LBE, Lago EAD, Boin I de FSF, et al. Microvascular invasion in hepatocellular carcinoma: is it predictable with quantitative computed tomography parameters?. Radiol Bras. 2019;52(5):287–92. pmid:31656344 - 40. Huang H, Wu F, Yu Y, Xu B, Chen D, Huo Y, et al. Multi-transcriptomics analysis of microvascular invasion-related malignant cells and development of a machine learning-based prognostic model in hepatocellular carcinoma. Front Immunol. 2024;15:1436131. pmid:39176099 - 41. Xiao J, Liu T, Liu Z, Xiao C, Du J, Zuo S, et al. A differentiation-related gene prognostic index contributes to prognosis and immunotherapy evaluation in patients with hepatocellular carcinoma. Cells. 2022;11(15):2302. pmid:35892599 - 42. Cai Y, Fu Y, Liu C, Wang X, You P, Li X, et al. Stathmin 1 is a biomarker for diagnosis of microvascular invasion to predict prognosis of early hepatocellular carcinoma. Cell Death Dis. 2022;13(2):176. pmid:35210426 - 43. Alvarez M, Benhammou JN, Darci-Maher N, French SW, Han SB, Sinsheimer JS, et al. Human liver single nucleus and single cell RNA sequencing identify a hepatocellular carcinoma-associated cell-type affecting survival. Genome Med. 2022;14(1):50. pmid:35581624 - 44. Filmus J, Selleck SB. Glypicans: proteoglycans with a surprise. J Clin Invest. 2001;108(4):497–501. pmid:11518720 - 45. De Cat B, David G. Developmental roles of the glypicans. Semin Cell Dev Biol. 2001;12(2):117–25. pmid:11292377 - 46. Filippi L, Frantellizzi V, Urso L, De Vincentis G, Urbano N. Targeting glypican-3 in liver cancer: groundbreaking preclinical and clinical insights. Biomedicines. 2025;13(7). pmid:40722645 - 47. Zheng X, Liu X, Lei Y, Wang G, Liu M. Glypican-3: a novel and promising target for the treatment of hepatocellular carcinoma. Front Oncol. 2022;12:824208. pmid:35251989 - 48. Chen F, Cai Y, Zhou J. Relationship between retinol metabolism and hepatocellular carcinoma: a comprehensive analysis of Mendelian randomization, prognostic characteristic and experiment. Discov Oncol. 2025;16(1):513. pmid:40210831 - 49. Yan T, Lu L, Xie C, Chen J, Peng X, Zhu L, et al. Severely impaired and dysregulated cytochrome P450 expression and activities in hepatocellular carcinoma: implications for personalized treatment in patients. Mol Cancer Ther. 2015;14(12):2874–86. pmid:26516155 - 50. Yang C, Huang X, Liu Z, Qin W, Wang C. Metabolism-associated molecular classification of hepatocellular carcinoma. Mol Oncol. 2020;14(4):896–913. pmid:31955511 - 51. Dierge E, Debock E, Guilbaud C, Corbet C, Mignolet E, Mignard L, et al. Peroxidation of n-3 and n-6 polyunsaturated fatty acids in the acidic tumor environment leads to ferroptosis-mediated anticancer effects. Cell Metab. 2021;33(8):1701-1715.e5. pmid:34118189 - 52. Cai D, Yuan M, Frantz DF, Melendez PA, Hansen L, Lee J, et al. Local and systemic insulin resistance resulting from hepatic activation of IKK-beta and NF-kappaB. Nat Med. 2005;11(2):183–90. pmid:15685173 - 53. Michiel DF, Oppenheim JJ. Cytokines as positive and negative regulators of tumor promotion and progression. Semin Cancer Biol. 1992;3(1):3–15. pmid:1322742 - 54. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science. 1993;259(5091):87–91. pmid:7678183 - 55. McGillicuddy FC, Chiquoine EH, Hinkle CC, Kim RJ, Shah R, Roche HM, et al. Interferon gamma attenuates insulin signaling, lipid storage, and differentiation in human adipocytes via activation of the JAK/STAT pathway. J Biol Chem. 2009;284(46):31936–44. pmid:19776010 - 56. Chen X, Cheng B, Dai D, Wu Y, Feng Z, Tong C, et al. Heparanase induces necroptosis of microvascular endothelial cells to promote the metastasis of hepatocellular carcinoma. Cell Death Discov. 2021;7(1):33. pmid:33597510 - 57. Zhou Q, Tian W, Jiang Z, Huang T, Ge C, Liu T, et al. A Positive Feedback Loop of AKR1C3-Mediated Activation of NF-κB and STAT3 Facilitates Proliferation and Metastasis in Hepatocellular Carcinoma. Cancer Res. 2021;81(5):1361–74. pmid:33361392 - 58. Jin ZG, Melaragno MG, Liao DF, Yan C, Haendeler J, Suh YA, et al. Cyclophilin A is a secreted growth factor induced by oxidative stress. Circ Res. 2000;87(9):789–96. pmid:11055983 - 59. Sherry B, Yarlett N, Strupp A, Cerami A. Identification of cyclophilin as a proinflammatory secretory product of lipopolysaccharide-activated macrophages. Proc Natl Acad Sci U S A. 1992;89(8):3511–5. pmid:1565646 - 60. Li Z, Zhao X, Bai S, Wang Z, Chen L, Wei Y, et al. Proteomics identification of cyclophilin a as a potential prognostic factor and therapeutic target in endometrial carcinoma. Mol Cell Proteomics. 2008;7(10):1810–23. pmid:18421009 - 61. Chu MY, Huang HC, Li EM, Xu LY. CypA: A potential target of tumor radiotherapy and/or chemotherapy. Curr Med Chem. 2021;28(19):3787–802. pmid:33121398 - 62. Choi KJ, Piao YJ, Lim MJ, Kim JH, Ha J, Choe W, et al. Overexpressed cyclophilin A in cancer cells renders resistance to hypoxia- and cisplatin-induced cell death. Cancer Res. 2007;67(8):3654–62. pmid:17440077 - 63. Liao E-C, Law C-H, Chen H-Y, Wei Y-S, Tsai Y-T, Lin L-H, et al. PPIA enhances cell growth and metastasis through CD147 in oral cancer. Arch Biochem Biophys. 2025;765:110328. pmid:39921142 - 64. Kalinina A, Golubeva I, Kudryavtsev I, Khromova N, Antoshina E, Trukhanova L, et al. Cyclophilin A is a factor of antitumor defense in the early stages of tumor development. Int Immunopharmacol. 2021;94:107470. pmid:33640856 - 65. Meng Y, Fan X-Y, Yang L-J, Xu B-Q, He D, Xu Z, et al. Detachment Activated CyPA/CD147 Induces Cancer Stem Cell Potential in Non-stem Breast Cancer Cells. Front Cell Dev Biol. 2020;8:543856. pmid:33195186 - 66. Landras A, Reger de Moura C, Jouenne F, Lebbe C, Menashi S, Mourah S. CD147 Is a Promising Target of Tumor Progression and a Prognostic Biomarker. Cancers (Basel). 2019;11(11):1803. pmid:31744072 - 67. Wang X, Qu Y, Xu Q, Jiang Z, Wang H, Lin B, et al. NQO1 Triggers Neutrophil Recruitment and NET Formation to Drive Lung Metastasis of Invasive Breast Cancer. Cancer Res. 2024;84(21):3538–55. pmid:39073320 - 68. Wei C, Zhang X, Li H, Gu J, Xue F, Xie W, et al. STMN1 regulates the stemness of gastric cancer cells by binding to HN1L to activate the STAT3 signaling pathway. Discov Oncol. 2025;16(1):263. pmid:40035993 - 69. Xie Y, Li X, Ge J. STAT3-CyPA signaling pathway in endothelial cell apoptosis. Cell Signal. 2020;65:109413. pmid:31494257 - 70. Deng L-X, Wu G-X, Cao Y, Fan B, Gao X, Luo L, et al. The chromosomal protein HMGN2 mediates lipopolysaccharide-induced expression of β-defensins in A549 cells. FEBS J. 2011;278(12):2152–66. pmid:21518253 - 71. Xin S, Liu L, Li Y, Yang J, Zuo L, Cao P, et al. Cyclophilin A binds to AKT1 and facilitates the tumorigenicity of Epstein-Barr virus by mediating the activation of AKT/mTOR/NF-κB positive feedback loop. Virol Sin. 2022;37(6):913–21. pmid:36075565 - 72. Öhlund D, Handly-Santana A, Biffi G, Elyada E, Almeida AS, Ponz-Sarvise M, et al. Distinct populations of inflammatory fibroblasts and myofibroblasts in pancreatic cancer. J Exp Med. 2017;214(3):579–96. pmid:28232471 - 73. Cheng Y, Chen X, Feng L, Yang Z, Xiao L, Xiang B, et al. Stromal architecture and fibroblast subpopulations with opposing effects on outcomes in hepatocellular carcinoma. Cell Discov. 2025;11(1):1. pmid:39870619 - 74. Zhang Y, Ling L, Murad R, Maganti S, Manceau A, Hetrick HA, et al. Macropinocytosis maintains CAF subtype identity under metabolic stress in pancreatic cancer. Cancer Cell. 2025;43(9):1677-1696.e15. pmid:40712568 - 75. Villanueva A, Hoshida Y, Battiston C, Tovar V, Sia D, Alsinet C, et al. Combining clinical, pathology, and gene expression data to predict recurrence of hepatocellular carcinoma. Gastroenterology. 2011;140(5):1501-12.e2. pmid:21320499 - 76. Goldman MJ, Craft B, Hastie M, Repečka K, McDade F, Kamath A, et al. Visualizing and interpreting cancer genomics data via the Xena platform. Nat Biotechnol. 2020;38(6):675–8. pmid:32444850 - 77. Tu S, Zhang X, Luo D, Liu Z, Yang X, Wan H, et al. Effect of taurine on the proliferation and apoptosis of human hepatocellular carcinoma HepG2 cells. Exp Ther Med. 2015;10(1):193–200. pmid:26170934 - 78. Zhang X, Tu S, Wang Y, Xu B, Wan F. Mechanism of taurine-induced apoptosis in human colon cancer cells. Acta Biochim Biophys Sin (Shanghai). 2014;46(4):261–72. pmid:24610575 - 79. Tu S, Zhang X-L, Wan H-F, Xia Y-Q, Liu Z-Q, Yang X-H, et al. Effect of taurine on cell proliferation and apoptosis human lung cancer A549 cells. Oncol Lett. 2018;15(4):5473–80. pmid:29552188 - 80. Ping Y, Shan J, Liu Y, Liu F, Wang L, Liu Z, et al. Taurine enhances the antitumor efficacy of PD-1 antibody by boosting CD8+ T cell function. Cancer Immunol Immunother. 2023;72(4):1015–27. pmid:36261540 - 81. Xiao H, Du X, Tao Z, Jing N, Bao S, Gao W-Q, et al. Taurine inhibits ferroptosis mediated by the crosstalk between tumor cells and tumor-associated macrophages in prostate cancer. Adv Sci (Weinh). 2024;11(3):e2303894. pmid:38031260 - 82. Cao T, Zhang W, Wang Q, Wang C, Ma W, Zhang C, et al. Cancer SLC6A6-mediated taurine uptake transactivates immune checkpoint genes and induces exhaustion in CD8(+) T cells. Cell. 2024;187(9):2288–304.e27. pmid:38565142 - 83. Guo D-Z, Zhang X, Zhang S-Q, Zhang S-Y, Zhang X-Y, Yan J-Y, et al. Single-cell tumor heterogeneity landscape of hepatocellular carcinoma: unraveling the pro-metastatic subtype and its interaction loop with fibroblasts. Mol Cancer. 2024;23(1):157. pmid:39095854 - 84. Zhai W, Lai H, Kaya NA, Chen J, Yang H, Lu B, et al. Dynamic phenotypic heterogeneity and the evolution of multiple RNA subtypes in hepatocellular carcinoma: the PLANET study. Natl Sci Rev. 2021;9(3):nwab192. pmid:35382356