Fibrosis Impairs cGAS-Mediated Responses to Immunotherapy in Advanced HCC

Fibrosis Impairs cGAS-Mediated Responses to Immunotherapy in Advanced HCC | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Fibrosis Impairs cGAS-Mediated Responses to Immunotherapy in Advanced HCC Ken Westover, Jianpeng Sheng, Lin Wang, Jinyuan Song, Junlei Zhang, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4536926/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Hepatocellular carcinoma (HCC) management is challenging due to its heterogeneous tumor microenvironment and poor treatment responses. To develop personalized approaches, we conducted a clinical trial of transarterial chemoembolization (TACE) combined with immune checkpoint blockade (ICB) where we collected tumor specimens for comprehensive multi-omic profiling. Treatment response was associated with increased infiltration of anti-tumor T cells, driven by cGAS-STING activation within immune-suppressive epithelial cells. However, fibrotic processes impaired these responses in some patients. Based on this insight, we conducted an animal trial in an HCC model where we added an anti-fibrotic agent to ICB and TACE mimic. This improved efficacy compared to ICB and TACE mimic alone. To identify patients who would benefit from such treatment, we constructed a predictive model using data from a limited immunohistochemistry panel and a group sparse learning algorithm. Our findings provide a blueprint for crafting personalized therapies for HCC and other cancers. Health sciences/Oncology/Cancer/Cancer therapy/Cancer immunotherapy Health sciences/Oncology/Cancer/Cancer microenvironment Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Significance We present a paradigm shift in HCC management by integrating multi-omics data with spatial information to guide personalized therapy. We believe these findings hold significant promise for improving patient outcomes in HCC and potentially other cancer types. Introduction Hepatocellular carcinoma (HCC) accounts for nearly 90% of primary liver cancer cases (1). Early detection of HCC is challenging and treatment options for inoperable patients are often ineffective, making HCC the third leading cause of cancer-related mortality in the world (2). Nevertheless, immunotherapy has improved the treatment landscape of HCC. Immune checkpoint blocking (ICB) therapy is a pillar of modern immunotherapy that functions to revers suppression of cytotoxic T lymphocytes (CTLs) by blocking immune checkpoint molecules such as programmed cell death protein 1 (PD-1). This enables the killing capacity of CTLs against tumors. In advanced HCC, combining ICB with anti-angiogenics such as bevacizumab or vascular endothelial growth factor receptor (VEGFR) inhibitors modifies tumor vasculature to bolster immune response, resulting in improved overall survival (3-6). In addition to systemic therapy, transarterial chemoembolization (TACE) is often used to improve loco-regional control for intermediate stage HCC with large or multinodular tumors without vascular invasion or extrahepatic spread (7, 8). TACE reduces tumor burden and can improve outcomes for these patients (9). One potential synergistic advantage of combining TACE with immunotherapy is that it modifies the immune microenvironment of HCC, including recruitment of antigen-specific T-cell and NK cells. This has motivated exploration of regimens that combine TACE with immune checkpoint inhibitors (10-12) and recently the EMERALD-1 trial, which evaluated a combination of TACE + durvalumab + VEGF inhibitor in unresectable HCC, showed an improvement in progression-free survival with the addition of ICB + VEGF inhibitor (13). Despite these advances, many patients still progress after therapy (14). We hypothesized that features of the tumor microenvironment (TME) may identify patients who respond well to dual therapies and may inform strategies to address non-responders. To discover such features, we evaluated specimens from a clinical trial (NCT04174781) of transarterial chemoembolization (DEB-TACE) and sintilimab (PD-1 antibody) in patients with HCC of BCLC stage A who exceeded the Milan criteria or BCLC stage B (15). We focused on extensive multi-omics analysis of tumor specimens before and after therapy. Using this approach, we were able to uncover regional response determinants, elucidate distinct mechanisms underlying regional responses and non-responses, optimize clinical trial strategy, and establish a predictive model to guide personalized treatment for HCC. Results Trial of TACE + ICB for Biomarker Discovery To evaluate the biological determinants of HCC responses following TACE with PD-1 antibody treatment, we evaluated specimens from patients enrolled in a clinical trial of DEB-TACE and sintilimab in locally advanced HCC (15). Patients had HCC of BCLC stage A who exceeded the Milan criteria or had BCLC stage B (Fig. 1A, Table S1-S2, NCT04174781). All patients had Child-Pugh A liver function and did not have vascular invasion, extrahepatic metastasis, or clinically significant portal hypertension. Sixty-one patients enrolled, all of whom underwent at least one treatment cycle combining DEB-TACE and sintilimab (Fig. 1A). The clinical characteristics of trial patients are summarized in Table S1. The majority of the patients were male (52 out of 61, representing 85%), with a median age of 58 years (ranging from 26 to 75 years old). Of these patients, 85% (52 individuals) had two or fewer lesions, while 56% (34 individuals) were diagnosed with BCLC stage B HCC. The median size of the target tumor was 7 cm (ranging from 2.1 to 14.1 cm). In total, 82% (50 out of 61) of the patients underwent only one treatment cycle with sintilimab combined with DEB-TACE, while 18% (11 patients) received two or three treatment cycles. At the data cutoff date, out of the 60 patients assessed for effectiveness, 37 (62%) exhibited an objective response to DEB-TACE combined with sintilimab; 3 patients had a complete response (5%), 34 a partial response (57%) and 20 experienced stable disease. Overall, the disease control rate was 95% (57/60) with a median follow up of 26 months. Patients showing complete and partial response were categorized as having an objective remission (OR), while patients showing steady (SD) were classified as such (Fig. 1B and Fig. S1A). Of the treated patients, adequate biopsy specimens were collected from 38, enabling analysis of biological correlates (Fig. 1C). Representative computer tomograph of OR (objective response) and SD (stable disease) patients were shown in Fig. 1D. Immune Suppressive Epithelial Cells Elevate in Responders We hypothesized that biological markers known to be active in modulating the HCC TME might be associated with treatment response. We generated a comprehensive spatial multi-omics dataset from patient samples. We excluded patients who underwent multiple rounds of TACE or did not have both pre- and post-treatment samples (Fig. 1C). As an initial evaluation, we devised a 40-marker IMC panel (39 protein markers and DNA) (Fig. 2A) based on a list compiled for the study of cellular neighborhoods in HCC (16). To ensure comprehensive coverage of each slide during IMC scanning, we performed scans across multiple regions (Fig. S1B). All antibodies were validated through immunohistochemistry (IHC) before heavy metal conjugation, and each metal-bound antibody was further verified by IMC (Fig. S2-4). The panel included markers for epithelial, endothelial, and stromal cells, various immune cells, as well as the cytokines IL-1β, TNF-α, and IL-6, the proliferation marker Ki-67, and the apoptotic indicator cleaved caspase-3 (Table S2). In total, mass cytometry analysis was done on 459 tumor ROIs from pre and post therapy samples (SD, n=15; OR, n=23). Marker expression levels in each cell were quantified and converted into a matrix. Following batch correction, approximately 1,895,221 cells were grouped into 19 cell meta-clusters identified by PhenoGraph (Fig. 2A to C). These meta-clusters were then annotated based on marker expression (Fig. 2C, Fig. S5A and B). Macrophage subsets, including resident macrophages (CD11b low CD68 + ) and infiltrating macrophages (CD11b + CD68 + CD14 + ) were separately identified (Fig. 2A). T cell subsets (CD4 + or CD8 + ), were further refined based on functional markers (Fig. 2A). Additionally, we observed epithelial cells (Pan-Cytokeratin + ) and immune-suppressive epithelial cells (PD-L1 + /B7H4 + Pan-Cytokeratin + ). Fibroblasts were further divided into Collagen I hi fibroblasts and α-SMA hi myofibroblasts (17) (Fig. 2A). Some clusters contained multiple cell types, such as epithelial cells and T cells, immune-suppressive epithelial cells and resident macrophages, and resident macrophages and fibroblasts, indicating close interactions of neighboring cells. One cluster from lineage-negative cells was also detected (Fig. 2A and Fig. S5). The frequencies of these cells in evaluated patients are shown in Fig. 2D. We observed a counterintuitive pattern in immune-suppressive epithelial cells, which increased after the treatment only within the OR group (Fig. 2E). In contrast, we identified opposing trends in α-SMA hi myofibroblasts and Collagen I hi fibroblasts following the combination treatment. The number of α-SMA hi myofibroblasts decreased, while Collagen I hi fibroblasts increased post-treatment. The increase in Collagen I hi fibroblasts was less pronounced in the OR group (Fig. 2E). The combination treatment also revealed opposing patterns in Tregs and CD4/8 + T cells, as Tregs experienced a decline while CD4/8 + T cells exhibited a rise. In addition, both resident and CD11c + macrophages demonstrated a frequency increase after the combined therapy (Fig. 2E). These observations suggest that therapy initiates an unexpected adaptive resistance mechanism. A decrease in α-SMA hi myofibroblasts along with an increase in Collagen I hi fibroblasts could reflect remodeling of the extracellular matrix that may either support or hinder an antitumor immune response, depending on the context. The observed decrease in Tregs alongside an increase in CD4/8 + T cells aligns with the intended immunostimulatory effects of PD-1 blockade, which aims to bolster antitumor immunity. Additionally, the rise in both resident and CD11c + macrophages after the combination therapy underscores the dynamic nature of innate immune responses. Treatment Response is Marked by Organizational Changes in the TME Beyond tissue composition, we hypothesized that cellular organization may also contribute to or predict for HCC responses to combined therapy. We utilized high-resolution spatial data to identify in situ alterations to cellular neighborhood organization after the combined therapy (18, 19), which we defined as the ten closest neighbor cells surrounding a central cell (Fig. 3A). We analyzed the cellular composition of the neighbors and annotated these neighborhoods based on the primary cellular clusters, resulting in various functional cellular neighborhood units (Fig. 3B). Additionally, we employed a Voronoi diagram, which was applied to each IMC image using the CN composition, to visualize the correlation between the defined CN functional units and the original IMC images (Fig. 3C) (20). The IMC topology map, depicted by the Voronoi diagram with different CNs represented by distinct colors (as shown in Fig. 3C), aligned well with the IMC image, confirming that cellular neighborhood assignments were correct. Furthermore, we adapted previously described methods to identify important regions within the HCC TME (21, 22). An expansive homotypic cellular cluster, as identified by uniform cellular phenotypes, was demarcated as a discrete spatial domain within the TME. Notable regions included the epithelial region, immune-suppressive epithelial region, fibroblast region, and the contiguous interfaces where immune-suppressive epithelial cells/immune cells, as well as regions of epithelial cells and fibroblast co-localization (Fig. 3C). Similar to the bulk frequency analysis, we found that the immune-suppressive epithelial region increased in the OR group after treatment, while the epithelial region decreased in the OR group. The fibroblast region showed an increase in both the SD and OR groups, although the magnitude of increase was less pronounced in the OR group. Additionally, the immune region experienced an increase in the OR group post-treatment. The epithelial/fibroblast intersection region and the immune suppressive epithelial/immune region displayed contrasting patterns in the OR group after treatment, with the former decreasing and the latter increasing (Fig. 3C and D). We performed interaction analysis to uncover potential mechanisms of resistance in the SD group. In that group, we observed a higher number of cellular interactions involving fibroblasts and epithelial/fibroblast interactions. Conversely, cellular interactions involving infiltrating macrophages, NK cells, and various T cells were notably absent in the epi-fibroblast intersection regions (Fig. 3E). On the other hand, in the immune suppressive-immune intersection regions, we observed higher cellular interactions involving immune suppressive epithelial cells and various T cells, while fibroblast interactions were comparatively reduced (Fig. 3F). This analysis suggests that distinct cellular subsets orchestrate environmental factors that give rise to contrasting regional features. Specifically, immune suppressive epithelial cells are in position to mediate cellular interactions with immune cells, particularly various T cells, in the OR group. In contrast, the SD group lacked these cellular interactions. These patterns suggest that immune-suppressive epithelial cells recruit T cells, whereas epithelial cells promote fibroblast enrichment, which may reduce the efficacy of therapeutic responses. Spatial and Single-Cell Transcriptomics Identify cGAS as a Regulator of the TME Our IMC analysis allowed us to visualize and quantify how cellular neighborhoods respond to TACE + ICB. To discover the biological processes that regulate these changes, we used single-cell RNA sequencing and spatial transcriptomics to evaluate samples (Fig. S6A). For this post-treatment samples from 6 patients in the OR and 4 patients in the SD cohorts were evaluable. We captured a total of 92,785 single cells from 6 OR and 4 SD patients post-surgery, with high sequencing quality. We identified 10 primary cell populations (Fig. 4A) based on specific markers (Fig. 4B), such as EpCAM + epithelial cells, VWF + endothelial cells (23) , and ACTA2 + fibroblasts (24) (Fig. 4B). We also identified various immune cell subsets, including CD3 + T cells, CD79B + B cells, IGHG1 + plasma B cells, CD68 + macrophages, GNLY + natural killer (NK) cells (25), and CPA3 + mast cells (26) (Fig. 4B). Additional clustering of T cells revealed 5 subsets (Fig. 4C and D): Granzyme A/K + CD8 + effector T cells, CD127 + CCR7 + central memory CD8 + T cells (CD8 + Tcm), Granzyme A/K + IL7R + effector memory CD8 + T cells (CD8 + Tem) (27) , CTLA4 + exhausted CD4 + T cells, and FOXP3 + regulatory T cells (Treg) (Fig. 4C and D). We identified two major subsets of macrophages: resident macrophages and infiltrating macrophages (Fig. 4E and F). Resident macrophages exhibited distinct markers, such as C1QA/B, which distinguished them from infiltrating macrophages (28). In contrast, infiltrating macrophages displayed higher levels of CCR2 and S100A8/9, than resident macrophages (Fig. 4E and F) Similar to the IMC analysis, we discerned two distinct subsets of epithelial cells within the TME: immune suppressive epithelial cells expressing PD-L1, and a separate population of epithelial cells devoid of PD-L1 expression (Fig. 4G and H). The PD-L1 population plays a regulatory role in immune recognition, whereas the PD-L1 null population, represents a phenotypically different group of epithelial cells within the same tissue context. We classified fibroblasts into two subtypes based on the IMC results: Collagen I hi fibroblasts and α-SMA hi myofibroblast cells (Fig. 4I and J). Our scRNA-seq landscape encompassed the key cellular components of the HCC TME and are consistent with the IMC results. These findings provide a detailed accounting of the diverse cellular populations in the TME and frame their potential roles in tumor immune responses. To unravel the mechanisms underlying the spatial and cellular interactions that dictate the regional response to TACE and ICB treatment, we performed an integrative analysis of single-cell RNA sequencing (scRNA-seq) and corresponding spatial transcriptomics (29). We employed a robust cell type decomposition (RCTD) approach, a computational method that harnesses the cell type profiles acquired from scRNA-seq data to decompose cell type mixtures while accounting for discrepancies across sequencing technologies. We used RCTD to decompose the spots within the spatial transcriptomics of all tissue sections, based on the refined immune and non-immune cellular clusters. After single-cell spatial deconvolution, we evaluated each spot for critical cell types identified by the IMC analysis, including epithelial cells, immune suppressive epithelial cells, fibroblasts, and T cells (Fig. S6B-E). This approach let us to accurately assess the presence and distribution of these key cell types across the tissue sections, providing insights into the spatial organization of the TME. As Fig. S6B-E shows, epithelial cells and fibroblast spots were significantly enriched in the SD patients. By contrast, immune suppressive epithelial cells and T cell spots were heavily enriched in the OR patients. Notably, we observed double positive spots (green spots) indicating the co-occurrence of immune suppressive epithelial cells and T cells was prominently observed in the OR patients (Fig. 4K and Fig. S6F). Conversely, double positive spots for epithelial cells and fibroblasts were predominantly enriched in the SD patients (Fig. 4L and Fig. S6G). These spatial transcriptomic insights suggest that immune-suppressive epithelial cells recruit T cells, enhancing therapeutic efficacy, while epithelial cells promote fibrosis, impeding treatment response. To understand the biological determinants this organization, we conducted KEGG pathway analyses on the spots with dual positivity for immune suppressive epithelial cells and T cells (Fig S7A). In the OR group, immune activation pathways such as chemokine signaling, T cell differentiation/trans-endothelial migration, and NF kappa B were evident. We also observed immune checkpoint signaling following immune activation (Fig. S7A). We also noted activation of cytosolic DNA sensing (cGAS pathway) pathways in the top group (Fig. S7A). In contrast, in spots with dual positivity for epithelial cells and fibroblasts (Fig. S7B), we observed metabolism-related pathways and oxidative phosphorylation pathways. Particularly noteworthy was the identification of the PPAR signaling pathway, which plays a key role in fibrosis (30), in the SD group (Fig. S7B). To validate integrated single-cell and spatial transcriptomic findings, we performed bulk RNA transcriptomic analysis on post-treatment samples obtained from a separate cohort of 19 patients treated with the same regimen (Table S1). Among the top differential genes identified, one was cGAS, a major regulator of double stranded DNA recognition (31) (Fig. S7C). To demonstrate the role of cGAS-STING in regulating immune suppression in our patient samples, we conducted several analyses. First, we performed correlation analysis for all the pathways identified in Fig. S7A, achieved by employing gene signatures specific to each pathway for Gene Set Variation Analysis (GSVA). The gene signatures served as input for calculating the GSVA scores. Analysis focused on immune suppressive epithelial and T cell spots across all pathways outlined in Fig. S7A. Subsequently, these GSVA scores were utilized to compute the correlations among the pathways. We noticed the cGAS-STING pathway correlated positively with the most enriched molecular pathways, including those for T-cell recruitment and migration (Fig. S7D), with the exception of the pyroptosis pathway (Fig. S7D). Second, we used SCENIC (single-cell regulatory network inference and clustering) analysis to identify key transcription factors that form the gene regulatory network within immune suppressive epithelial cells (32). We detected cGAS-STING related transcription factors such as IRF3 and NF-κB (Fig. S7E) within immune suppressive epithelial cells. We validated the SCENIC results through Western blotting of individual patient tumor samples which showed elevated levels of phosphorylated versions of STING pathway proteins, TBK1, IRF3, STING, and RelA in the OR group but not the SD group (Fig. 4M). These findings suggest the STING pathway is activated in responsive samples, reinforcing the idea that STING participates in the treatment response. Consistently, mIHC staining also showed more cGAS signal and downstream RelA (33) signal in the OR group, while these signals were absent in the SD group (Fig. 4N). Epithelial cells in the OR group displayed high levels of PD-L1 expression (Pan-CK + ) compared to the absence of PD-L1 expression in epithelial cells in the SD group (Fig. 4O). Using mIHC we observed colocalization of PD-L1 and the cGAS expression (Fig. 4N). We note that this connection between cGAS-STING pathway and immune checkpoint expression has been observed previously (34). To further confirm the association between cGAS signaling and PD-L1 expression, we exposed a panel of three human HCC cell lines to the topoisomerase inhibitor epirubicin, which causes double-stranded DNA (dsDNA) breaks. This caused cGAS signaling and upregulation in PD-L1 expression (Fig. S8), consistent with a previous report (35). This may partially explain why single administration of STING agonists have not yielded satisfactory results in clinical trials since upregulation of PD-L1 could theoretically attenuate the anti-tumor immune response (36). Nevertheless, PD-L1 expression alone does not faithfully predict for immunotherapeutic response in hepatocellular carcinoma (37). We further analyzed the samples to find actionable hypotheses related to therapeutic vulnerabilities. Utilizing spatial transcriptomics to analyze samples from OR and SD cohorts, we delineated distinct spatial features unique to each group. In the OR cohort, regions of overlap between immune suppressive epithelial cells and T cells were pronounced, indicating a potential niche for immune modulation. Conversely, the SD cohort was characterized by a notable co-localization of epithelial cells and fibroblasts, suggesting a tissue composition of fibrosis. To further elucidate the cellular interactions underlying the recruitment of T cells by immune-suppressive epithelial cells and the promotion of fibrosis by epithelial cells, we performed a ligand-receptor analysis on both OR and SD patients, integrating data from both Spatial Transcriptomics (ST) and single-cell RNA sequencing (scRNA-seq). We used NicheNET, a tool that combines expression data from intercommunicating cells with pre-existing knowledge of signaling and gene regulatory networks, to predict potential ligand-target associations (38). First, we selected our genes of interest from the differentially expressed genes, which we calculated based on ST datasets for specific areas identified via IMC analysis. Next, we applied scRNA-seq for cellular interaction analysis via NicheNET. Our observations indicated that PD-L1 + epithelial cells could secrete multiple cytokines and chemokines, such as CCL5 (39), MIF (40), and LTB (41), thereby recruiting immune cells such as T cells (Fig. S9A). In contrast, in SD patients, epithelial cells transmitted numerous inhibitory signals for PPAR signaling pathways to fibroblasts (Fig. S9B), such as ATF3 (42), BCL6 (43), LCOR (44), EZH2 (45), NR0B2 (46), SIRT1 (47), GPS2 (48), and NRIP1 (49). Accordingly, based on ST analysis, we detected reduced PPAR signaling within the epithelial-fibroblast regions associated with SD patients (Fig. S9C), especially in areas showing dual-positive signals from both epithelial cells and fibroblasts. Our examination revealed downregulation of genes, such as PLIN2 (50), FABP6 (51), PLTP (52), UBC (53), and HMGCS1 (54) (Fig. S9C). These findings indicate a suppression of PPAR signaling in the non-responder (SD) patients. To validate NicheNET findings, we evaluated levels of double-stranded DNA (dsDNA) and subsequent activation of the cGAS signaling pathway using mIHC and found these were elevated in the OR group, along with expression of PD-L1. This activation was associated with increased recruitment and activation of T cells (CD45RO) (Fig. 4O and Fig. S9D). T cells also showed resident marker CD103 (Fig. S10). Notably fibrosis, as indicated by Collagen I staining, was relatively low in the OR group. Conversely, the SD group exhibited high levels of fibrosis, accompanied by low dsDNA and cGAS signaling. As a result, activated T cells were scarce in the SD group (Fig. 4O and Fig. S9D). These observations were consistent with the pathway analysis obtained through the integration of scRNA-seq and ST, which showed an interaction between dsDNA, cGAS signaling, fibrosis, and T cell activation. In conclusion, the above findings provide a model for how cGAS signaling can modulate responses to TACE + ICB in HCC. Activation of cGAS signaling by double-stranded DNA (dsDNA) results in increased recruitment and activation of T cells, as well as upregulation of PD-L1 expression in the responsive (OR) group. Conversely, non-responders (SD) group exhibit fibrosis, low dsDNA, and limited cGAS signaling, leading to reduced T cell recruitment and PD-L1 expression. These results highlight the significance of cGAS signaling and fibrosis in shaping the TME and suggest its potential as a therapeutic target in HCC (Fig. 4P). Addition of Anti-Fibrotic Agents Enhance Therapy for HCC The observation that fibrosis limited the efficacy of TACE+anti-PD-1 treatment suggests that anti-fibrotic agents might enhance therapy. We evaluated this concept in a mouse HCC model induced by cMyc and Nras expression (16) since in murine models of HCC, activation of both Myc and Ras oncogenes has been implicated in the promotion of hepatic fibrosis (55, 56). Treatments were conditioned starting on day 14 with Elafibranor, a PPAR agonist known to have anti-inflammatory and anti-fibrotic effects. Elafibranor has shown promising results in reducing liver inflammation, oxidative stress, and fibrosis in preclinical liver disease models (57). On day 21, we mimicked TACE by treating with the STING agonist cGAMP; we also administered anti-PD-1 therapy (Sintilimab). After 30 days, we harvested the liver with tumors for further analysis (Fig. 5A). Both Elafibranor and the dual combination therapy of TACE+anti-PD-1 demonstrated reductions in tumor burden, with the dual combination therapy showing a greater reduction. However, the triple combination therapy involving Elafibranor, TACE, and anti-PD-1 exhibited the least tumor burden (Fig. 5B and C). Treatment with Elafibranor alone resulted in reduced fibrosis. The dual combination therapy of TACE+anti-PD-1 showed an increase in CD8 + T cells. Notably, the triple combination therapy led to reduced fibrosis and increased CD8 + T cells (Fig. 5D and Fig. S11). Graphical Classification Based on Spatial Similarity (GCSS) Predicts Treatment Response An enticing clinical strategy would be to predict which patients are likely to have a poor response to therapy due to fibrosis, then add an anti-fibrotic agent for that population. We hypothesized that it might be possible to generate a predictive model using imaging mass cytometry (IMC) data obtained from needle biopsies. We used pre-treatment IMC images and associated clinical outcomes from the TACE+ICB trial as the training dataset for our model (Fig. 6A). The model was constructed by dividing segmented cells within each Region of Interest (ROI) into separate images (N=29), each reflecting a mask of specific cellular identity. Each image represents the spatial localization of a specific cell type. Subsequently, we calculate the pairwise spatial similarity between different cell types for each ROI, using a structural similarity matrix (SSIM) measurement (Fig. 6B). We then employed a network classification method (58), Graphical Classification based on Spatial Similarity (GCSS), to predict for association between spatial patterns of cells, including interactions, and the efficacy of therapy (Fig. 6A). The GCSS model considers both the spatial similarity information of different cell types and the network structure of the data in a computationally efficient way. To confirm the GCSS method was optimal, we conducted a comparison between the GCSS method and the classical support vector machine (SVM) and Least Absolute Shrinkage and Selection Operator (Lasso) methods. The results showed that GCSS exhibited higher sensitivity and accuracy compared to SVM and Lasso (Fig. 6C-D). Incidentally, GCSS also provided confirmatory mechanistic data regarding cellular interactions within each group. In the SD group, we observed interactions between epithelial/fibroblast and CD45RA + CD8 + T cells, effector CD8 + T cells, and epithelial/fibroblast cells, as well as neutrophil and epithelial cell interactions. In the OR group, we found interactions between NK cells and infiltrating macrophages, effector CD8 + T cells and immune suppressive epithelial cells, and GATA3 + CD4 + T cells and resident macrophages. These interactions are consistent with our integrated spatial multi-omics analysis and prior mechanistic exploration (Fig. 6E). We used dominant patterns in the model to create a minimized mIHC panel for predicting treatment response consisting of six markers, including CD4, CD8, CD68, Collagen I, Pan-CK, and PD-L1 (Fig. 6F). These markers represented the crucial cell interactions responsible for different treatment responses within the OR and SD groups (Fig. 6E). This approach is intended to enable clinical feasibility since it reduces assay costs and can be done more rapidly. To validate the predictive capability of the minimal panel, we collected 61 pre-treatment specimens from HCC patients who underwent dual combination treatment. Using mIHC input into the GCSS model, we achieved over 80% accuracy in predicting treatment response, similar to results on IMC data (compare Fig. 6C to 6G). These results suggest the mIHC panel will be effective for predicting treatment responses for HCC to TACE and ICB at minimal cost. This result was achieved despite a relatively small sample size, suggesting robustness in the method. Discussion Immunotherapy has shown that, on average, it can improve outcomes in unresectable HCC, but some patients do not respond. This study identified regional features within the TME of HCC that influence response to combined TACE and ICB therapy. Using spatial multi-omics, we revealed key interactions between Collagen I hi fibroblasts, immune-suppressive epithelial cells, and T cells. In responders, TACE treatment triggered dsDNA breaks, activating the cGAS pathway primarily in immune-suppressive epithelial cells. Conversely, non-responders displayed a diminished response to dsDNA sensing, accompanied by reduced cGAS activation in downstream effector cells. Additionally, we observed prominent fibrosis in non-responders, suggesting that triple combination therapy including an anti-fibrotic agent might be beneficial. This hypothesis was confirmed in a mouse model. Finally, to identify patients who may benefit from this approach, we developed a fast and cost-effective assay based on a GCSS model to predict HCC responses to TACE and ICB treatments. While combination TACE therapies have shown promise in improving cancer outcomes, particularly with anti-angiogenic drugs like sorafenib, this study offers a more precise approach. Sorafenib disrupts blood vessel formation (angiogenesis) and cancer-promoting signals, demonstrating improved progression-free survival in HCC when combined with TACE (59). Given the distinct mechanisms of anti-angiogenic and anti-fibrotic drugs, we hypothesize that further combinations, including sorafenib or similar drugs alongside anti-fibrotic agents, might further optimize outcomes for specific patient groups. This study demonstrates that using direct measures of the composition and organization of the TME can lead to predictive models and mechanistic insights. This is important because prior efforts utilizing gene signatures, circulating tumor cells, high-dimensional flow cytometry, single-cell RNA sequencing, microbiome analysis, radiomics, and clinical markers, have yielded inconsistent or weak associations with treatment response (37). Similarly, studies evaluating neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) in HCC patients receiving anti-PD-1 antibodies, as well as high-dimensional immune profiling of peripheral blood mononuclear cells (PBMCs), have not identified reliable biomarkers for predicting ICI response in HCC patients (37, 60). This study also expands on our previous research that identified distinct regions within the HCC TME - normal, fibrotic, and tumor tissue (16). Here, we delve deeper, uncovering molecular mechanisms in play: excessive fibrosis hinders the transmission of dsDNA signals, thereby preventing cGAS activation. This aligns with Yao et al.'s recent findings on a tumor immune barrier (TIB) composed of SPP1 + macrophages and cancer-associated fibroblasts (CAFs) near the tumor boundary, which restricts immune cell infiltration and impacts the efficacy of immune checkpoint blockade (ICB) (61). We propose a more nuanced view – a moderate level of fibrosis might actually benefit chemotherapy and immunotherapy responses. Our data, integrating current and past studies, suggests the fibrotic region serves as an anchor for diverse immune cell populations, including various macrophage subsets. Since macrophages play a critical role in cGAS signal transduction, their presence underscores their importance in these therapies. Utilizing spatial multi-omics data integration, we discovered that features associated with response and non-response coexisted within the TME prior to treatment. Successful combination therapy appears to clear this "chaos" and enhance anti-tumor features via the cGAS pathway. Interestingly, traditionally immunosuppressive cells, such as neutrophils, PD-L1 + epithelial cells, and fibroblasts, were found to be crucial for the response to combined TACE and ICB therapy. Based on this mechanistic understanding, we have established a predictive model with potential to guide real-world HCC patient treatment decisions. Nevertheless, clinical trials incorporating the model will be necessary for validation. Methods Detailed protocols for clinical trial evaluation, IMC, other experiments, and data analysis are provided in Supplementary Information. Declarations AUTHOR CONTRIBUTION Jianpeng Sheng conceptualized the study, designed the experiments. Jianpeng Sheng, Zhiwei Zhou, Qing Li and Kenneth D. Westover drafted the manuscript. Lin Wang conducted data analysis, while Junlei Zhang and Jinyuan Song performed the experimental work. All three authors contributed equally to this work. Wei Shao and Qing Li developed the prediction model analysis and conducted imaging analysis. Yaxing Zhao provided assistance with bioinformatic analysis. Jianghui Tang, Yongtao Ji, and Taohong Li contributed to the experimental procedures. Tianxing Yan designed the graphical abstract and cover image. Tingbo Liang and Xueli Bai organized the clinical trial and obtained the clinical samples. Chengxiang Guo and Yiwen Chen assisted in coordinating the clinical trials. Qing Li, Zhiwei Zhou and Kenneth D. Westover contributed to proofreading the manuscript and reviewed the manuscript. Tingbo Liang, Xueli Bai, Jianpeng Sheng and Kenneth Westover supervised the project and secured funding. DATA AVAILABILITY The multi-omics data associated with this study have been deposited in various repositories managed by the Genome Sequence Archive of Beijing Institute of Genomics, Chinese Academy of Sciences. The scRNA-seq can be accessed via HRA004695 (Data were controlled access due to patients’ privacy. Reviewers can request data at https://ngdc.cncb.ac.cn/gsa-human/browse/HRA004695 and it will be released within three working days) and HRA007308 (https://ngdc.cncb.ac.cn/gsa-human/s/1hp5rUr4). Bulk transcriptomics data can be accessed via HRA007307 (https://ngdc.cncb.ac.cn/gsa-human/s/S1DGuE7C). Spatial transcriptomics data can be accessed via OMIX006713 (https://ngdc.cncb.ac.cn/omix/release/OMIX006713). The IMC data can be accessed via OMIX002774 (https://ngdc.cncb.ac.cn/omix/release/OMIX002774) and OMIX003997 (https://ngdc.cncb.ac.cn/omix/release/OMIX003997). The codes used in this analysis have been uploaded to Github at the following link: https://github.com/shaoweinuaa/GCSS and https://github.com/pyramidsnail/TACE. If further information is required for data reanalysis, please contact the lead researcher directly. CODE AVAILABILITY Please note that the single cell and spatial transcriptomics analysis in this study does not involve the original code. To access the original codes used, please refer to the methods section and respective references provided. The GCSS codes used in this analysis have been uploaded to Github at the following link: https://github.com/shaoweinuaa/GCSS. If you require any additional information to reanalyze the data presented in this paper, please contact the lead researcher directly. ACKNOWLEDGEMENT The authors would like to thank Hangzhou Yingfei Biotechnology for IMC and mIHC help. This work was supported by the National Key Research and Development Program of China (grant 2019YFA0803000 to J.S.), the Excellent Youth Foundation of Zhejiang Scientific (grant R22H1610037 to J.S.), the National Natural Science Foundation of China (grant 82173078 to J.S.), the Natural Science Foundation of Zhejiang Province (grant 2022C03037 to J.S.), the National Natural Science Foundation of China (grant 81871925 to X.B and grant 82188102 to T.L.), the National Key Research and Development Program (grant 2019YFC1316000 to T.L.), the National Natural Science Foundation Basic Science Centre of China (Study of Tumor Material and Energy Dynamics, 8218810) and Welch Foundation I-1829 (to K.D.W) References Chidambaranathan-Reghupaty S, Fisher PB, Sarkar D. Hepatocellular carcinoma (HCC): Epidemiology, etiology and molecular classification. Adv Cancer Res. 2021;149:1-61. Yang JD, Hainaut P, Gores GJ, Amadou A, Plymoth A, Roberts LR. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol. 2019;16(10):589-604. Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim T-Y, et al. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. New England Journal of Medicine. 2020;382(20):1894-905. Llovet JM, Castet F, Heikenwalder M, Maini MK, Mazzaferro V, Pinato DJ, et al. Immunotherapies for hepatocellular carcinoma. Nat Rev Clin Oncol. 2022;19(3):151-72. Jin H, Qin S, He J, Xiao J, Li Q, Mao Y, et al. New insights into checkpoint inhibitor immunotherapy and its combined therapies in hepatocellular carcinoma: from mechanisms to clinical trials. Int J Biol Sci. 2022;18(7):2775-94. Yap TA, Parkes EE, Peng W, Moyers JT, Curran MA, Tawbi HA. Development of Immunotherapy Combination Strategies in Cancer. Cancer Discov. 2021;11(6):1368-97. Raoul JL, Forner A, Bolondi L, Cheung TT, Kloeckner R, de Baere T. Updated use of TACE for hepatocellular carcinoma treatment: How and when to use it based on clinical evidence. Cancer Treat Rev. 2019;72:28-36. Han K, Kim JH. Transarterial chemoembolization in hepatocellular carcinoma treatment: Barcelona clinic liver cancer staging system. World J Gastroenterol. 2015;21(36):10327-35. Camma C, Schepis F, Orlando A, Albanese M, Shahied L, Trevisani F, et al. Transarterial chemoembolization for unresectable hepatocellular carcinoma: meta-analysis of randomized controlled trials. Radiology. 2002;224(1):47-54. Ayaru L, Pereira SP, Alisa A, Pathan AA, Williams R, Davidson B, et al. Unmasking of alpha-fetoprotein-specific CD4(+) T cell responses in hepatocellular carcinoma patients undergoing embolization. J Immunol. 2007;178(3):1914-22. Zerbini A, Pilli M, Laccabue D, Pelosi G, Molinari A, Negri E, et al. Radiofrequency thermal ablation for hepatocellular carcinoma stimulates autologous NK-cell response. Gastroenterology. 2010;138(5):1931-42. Llovet JM, De Baere T, Kulik L, Haber PK, Greten TF, Meyer T, et al. Locoregional therapies in the era of molecular and immune treatments for hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2021;18(5):293-313. Lencioni R, Kudo M, Erinjeri J, Qin S, Ren Z, Chan S, et al. EMERALD-1: A phase 3, randomized, placebo-controlled study of transarterial chemoembolization combined with durvalumab with or without bevacizumab in participants with unresectable hepatocellular carcinoma eligible for embolization. Journal of Clinical Oncology. 2024;42(3_suppl):LBA432-LBA. Lune PV, Aal AKA, Klimkowski S, Zarzour JG, Gunn AJ. Hepatocellular carcinoma: diagnosis, treatment algorithms, and imaging appearance after transarterial chemoembolization. Journal of Clinical and Translational Hepatology. 2018;6(2):175. Guo C, Zhang J, Huang X, Chen Y, Sheng J, Huang X, et al. Preoperative sintilimab plus transarterial chemoembolization for hepatocellular carcinoma exceeding the Milan criteria: A phase II trial. Hepatol Commun. 2023;7(3):e0054. Sheng J, Zhang J, Wang L, Tano V, Tang J, Wang X, et al. Topological analysis of hepatocellular carcinoma tumour microenvironment based on imaging mass cytometry reveals cellular neighbourhood regulated reversely by macrophages with different ontogeny. Gut. 2022;71(6):1176-91. Enzan H, Himeno H, Iwamura S, Saibara T, Onishi S, Yamamoto Y, et al. Immunohistochemical Identification of Ito Cells and Their Myofibroblastic Transformation in Adult Human Liver. Virchows Arch. 1994;424(3):249-56. Schürch CM, Bhate SS, Barlow GL, Phillips DJ, Noti L, Zlobec I, et al. Coordinated cellular neighborhoods orchestrate antitumoral immunity at the colorectal cancer invasive front. Cell. 2020;182(5):1341-59. e19. Schapiro D, Jackson HW, Raghuraman S, Fischer JR, Zanotelli VRT, Schulz D, et al. histoCAT: analysis of cell phenotypes and interactions in multiplex image cytometry data. Nat Methods. 2017;14(9):873-6. Schurch CM, Bhate SS, Barlow GL, Phillips DJ, Noti L, Zlobec I, et al. Coordinated Cellular Neighborhoods Orchestrate Antitumoral Immunity at the Colorectal Cancer Invasive Front. Cell. 2020;183(3):838. Hickey JW, Becker WR, Nevins SA, Horning A, Perez AE, Zhu C, et al. Organization of the human intestine at single-cell resolution. Nature. 2023;619(7970):572-84. Risom T, Glass DR, Averbukh I, Liu CC, Baranski A, Kagel A, et al. Transition to invasive breast cancer is associated with progressive changes in the structure and composition of tumor stroma. Cell. 2022;185(2):299-310 e18. Nightingale T, Cutler D. The secretion of von Willebrand factor from endothelial cells; an increasingly complicated story. J Thromb Haemost. 2013;11 Suppl 1(Suppl 1):192-201. Shinde AV, Humeres C, Frangogiannis NG. The role of alpha-smooth muscle actin in fibroblast-mediated matrix contraction and remodeling. Biochim Biophys Acta Mol Basis Dis. 2017;1863(1):298-309. Crespo AC, Mulik S, Dotiwala F, Ansara JA, Sen Santara S, Ingersoll K, et al. Decidual NK Cells Transfer Granulysin to Selectively Kill Bacteria in Trophoblasts. Cell. 2020;182(5):1125-39 e18. Xing W, Austen KF, Gurish MF, Jones TG. Protease phenotype of constitutive connective tissue and of induced mucosal mast cells in mice is regulated by the tissue. Proc Natl Acad Sci U S A. 2011;108(34):14210-5. De Rosa SC, Herzenberg LA, Herzenberg LA, Roederer M. 11-color, 13-parameter flow cytometry: identification of human naive T cells by phenotype, function, and T-cell receptor diversity. Nat Med. 2001;7(2):245-8. Purves-Tyson TD, Robinson K, Brown AM, Boerrigter D, Cai HQ, Weissleder C, et al. Increased Macrophages and C1qA, C3, C4 Transcripts in the Midbrain of People With Schizophrenia. Front Immunol. 2020;11:2002. 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. Blas-Garcia A, Apostolova N. Novel Therapeutic Approaches to Liver Fibrosis Based on Targeting Oxidative Stress. Antioxidants (Basel). 2023;12(8). Ablasser A, Chen ZJ. cGAS in action: Expanding roles in immunity and inflammation. Science. 2019;363(6431):eaat8657. Aibar S, Gonzalez-Blas CB, Moerman T, Huynh-Thu VA, Imrichova H, Hulselmans G, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods. 2017;14(11):1083-6. Kwon J, Bakhoum SF. The Cytosolic DNA-Sensing cGAS-STING Pathway in Cancer. Cancer Discov. 2020;10(1):26-39. Lee JJ, Kim SY, Kim SH, Choi S, Lee B, Shin JS. STING mediates nuclear PD-L1 targeting-induced senescence in cancer cells. Cell Death Dis. 2022;13(9):791. Du SS, Chen GW, Yang P, Chen YX, Hu Y, Zhao QQ, et al. Radiation Therapy Promotes Hepatocellular Carcinoma Immune Cloaking via PD-L1 Upregulation Induced by cGAS-STING Activation. Int J Radiat Oncol Biol Phys. 2022;112(5):1243-55. Le Naour J, Zitvogel L, Galluzzi L, Vacchelli E, Kroemer G. Trial watch: STING agonists in cancer therapy. Oncoimmunology. 2020;9(1):1777624. Greten TF, Villanueva A, Korangy F, Ruf B, Yarchoan M, Ma L, et al. Biomarkers for immunotherapy of hepatocellular carcinoma. Nature Reviews Clinical Oncology. 2023;20(11):780-98. Browaeys R, Saelens W, Saeys Y. NicheNet: modeling intercellular communication by linking ligands to target genes. Nat Methods. 2020;17(2):159-62. Chan O, Burke JD, Gao DF, Fish EN. The chemokine CCL5 regulates glucose uptake and AMP kinase signaling in activated T cells to facilitate chemotaxis. J Biol Chem. 2012;287(35):29406-16. Schober A, Bernhagen J, Weber C. Chemokine-like functions of MIF in atherosclerosis. J Mol Med (Berl). 2008;86(7):761-70. Abdulrahman Z, Santegoets SJ, Sturm G, Charoentong P, Ijsselsteijn ME, Somarakis A, et al. Tumor-specific T cells support chemokine-driven spatial organization of intratumoral immune microaggregates needed for long survival. J Immunother Cancer. 2022;10(2). Jang MK, Jung MH. ATF3 represses PPARgamma expression and inhibits adipocyte differentiation. Biochem Biophys Res Commun. 2014;454(1):58-64. Sommars MA, Ramachandran K, Senagolage MD, Futtner CR, Germain DM, Allred AL, et al. Dynamic repression by BCL6 controls the genome-wide liver response to fasting and steatosis. Elife. 2019;8:e43922. Stewart MD, Wong J. Nuclear receptor repression: regulatory mechanisms and physiological implications. Prog Mol Biol Transl Sci. 2009;87:235-59. Du Z, Liu M, Wang Z, Lin Z, Feng Y, Tian D, et al. EZH2-mediated inhibition of KLF14 expression promotes HSCs activation and liver fibrosis by downregulating PPARgamma. Cell Prolif. 2021;54(7):e13072. Chanda D, Xie YB, Choi HS. Transcriptional corepressor SHP recruits SIRT1 histone deacetylase to inhibit LRH-1 transactivation. Nucleic Acids Res. 2010;38(14):4607-19. Picard F, Kurtev M, Chung N, Topark-Ngarm A, Senawong T, Machado De Oliveira R, et al. Sirt1 promotes fat mobilization in white adipocytes by repressing PPAR-gamma. Nature. 2004;429(6993):771-6. Guo C, Li Y, Gow CH, Wong M, Zha J, Yan C, et al. The optimal corepressor function of nuclear receptor corepressor (NCoR) for peroxisome proliferator-activated receptor gamma requires G protein pathway suppressor 2. J Biol Chem. 2015;290(6):3666-79. Viswakarma N, Jia Y, Bai L, Vluggens A, Borensztajn J, Xu J, et al. Coactivators in PPAR-Regulated Gene Expression. PPAR Res. 2010;2010. Shi H, Luo J, Zhu J, Li J, Sun Y, Lin X, et al. PPAR gamma Regulates Genes Involved in Triacylglycerol Synthesis and Secretion in Mammary Gland Epithelial Cells of Dairy Goats. PPAR Res. 2013;2013:310948. Zhang X, Yao J, Shi H, Gao B, Zhang L. LncRNA TINCR/microRNA-107/CD36 regulates cell proliferation and apoptosis in colorectal cancer via PPAR signaling pathway based on bioinformatics analysis. Biol Chem. 2019;400(5):663-75. Chehaibi K, Cedo L, Metso J, Palomer X, Santos D, Quesada H, et al. PPAR-beta/delta activation promotes phospholipid transfer protein expression. Biochem Pharmacol. 2015;94(2):101-8. Yang L, Zheng Y, Miao YM, Yan WX, Geng YZ, Dai Y, et al. Bergenin, a PPARgamma agonist, inhibits Th17 differentiation and subsequent neutrophilic asthma by preventing GLS1-dependent glutaminolysis. Acta Pharmacol Sin. 2022;43(4):963-76. Bionaz M, Thering BJ, Loor JJ. Fine metabolic regulation in ruminants via nutrient-gene interactions: saturated long-chain fatty acids increase expression of genes involved in lipid metabolism and immune response partly through PPAR-alpha activation. Br J Nutr. 2012;107(2):179-91. Nevzorova YA, Hu W, Cubero FJ, Haas U, Freimuth J, Tacke F, et al. Overexpression of c-myc in hepatocytes promotes activation of hepatic stellate cells and facilitates the onset of liver fibrosis. Biochim Biophys Acta. 2013;1832(10):1765-75. Foglia B, Cannito S, Bocca C, Parola M, Novo E. ERK Pathway in Activated, Myofibroblast-Like, Hepatic Stellate Cells: A Critical Signaling Crossroad Sustaining Liver Fibrosis. Int J Mol Sci. 2019;20(11). Westerouen Van Meeteren MJ, Drenth JPH, Tjwa E. Elafibranor: a potential drug for the treatment of nonalcoholic steatohepatitis (NASH). Expert Opin Investig Drugs. 2020;29(2):117-23. Relión JDA, Kessler D, Levina E, Taylor SF. Network classification with applications to brain connectomics. The annals of applied statistics. 2019;13(3):1648. Masatoshi K, Kazuomi U, Masafumi I, Takuji T, Nobukazu T, Hiroshi A, et al. Randomised, multicentre prospective trial of transarterial chemoembolisation (TACE) plus sorafenib as compared with TACE alone in patients with hepatocellular carcinoma: TACTICS trial. Gut. 2020;69(8):1492. Dharmapuri S, Özbek U, Lin JY, Sung M, Schwartz M, Branch AD, et al. Predictive value of neutrophil to lymphocyte ratio and platelet to lymphocyte ratio in advanced hepatocellular carcinoma patients treated with anti-PD-1 therapy. Cancer Med. 2020;9(14):4962-70. Liu Y, Xun Z, Ma K, Liang S, Li X, Zhou S, et al. Identification of a tumour immune barrier in the HCC microenvironment that determines the efficacy of immunotherapy. J Hepatol. 2023;78(4):770-82. Additional Declarations There is NO Competing Interest. Supplementary Files SMOcGASSTINGHCCSINaturecancer.docx Supplementary Figures Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4536926","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":319398828,"identity":"49bd2aa6-709d-4ab0-9e0a-804a13b6b767","order_by":0,"name":"Ken Westover","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0003-3653-5923","institution":"University of Texas Southwestern Medical Center","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Ken","middleName":"","lastName":"Westover","suffix":""},{"id":319398829,"identity":"c3505b33-4bda-431b-b245-4c3e934cb5dc","order_by":1,"name":"Jianpeng Sheng","email":"","orcid":"https://orcid.org/0000-0002-5535-5541","institution":"The First Affiliated Hospital, Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jianpeng","middleName":"","lastName":"Sheng","suffix":""},{"id":319398830,"identity":"49efcfba-8636-4d2a-b3de-d243c94226cf","order_by":2,"name":"Lin Wang","email":"","orcid":"","institution":"Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Lin","middleName":"","lastName":"Wang","suffix":""},{"id":319398831,"identity":"b3419fee-110d-4ab3-bccc-20145fd5dedd","order_by":3,"name":"Jinyuan Song","email":"","orcid":"","institution":"Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jinyuan","middleName":"","lastName":"Song","suffix":""},{"id":319398832,"identity":"63598bb6-d879-44cb-9539-796277fa62c5","order_by":4,"name":"Junlei Zhang","email":"","orcid":"","institution":"Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Junlei","middleName":"","lastName":"Zhang","suffix":""},{"id":319398833,"identity":"5c6209f4-e29e-4ef9-871a-05e467ccafa6","order_by":5,"name":"ZHIWEI ZHOU","email":"","orcid":"","institution":"University of Texas Southwestern Medical Center at Dallas","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"ZHIWEI","middleName":"","lastName":"ZHOU","suffix":""},{"id":319398834,"identity":"83524f07-4a38-448a-b5c9-e2d809d1638e","order_by":6,"name":"Yaxing Zhao","email":"","orcid":"","institution":"Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yaxing","middleName":"","lastName":"Zhao","suffix":""},{"id":319398835,"identity":"c1be1cef-9af7-4598-8396-fd5d92dde775","order_by":7,"name":"Chengxiang Guo","email":"","orcid":"","institution":"Zhejiang University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Chengxiang","middleName":"","lastName":"Guo","suffix":""},{"id":319398836,"identity":"dd4133b2-69d6-4460-99af-8fa9435dd454","order_by":8,"name":"Jianghui Tang","email":"","orcid":"","institution":"The First Affiliated Hospital, Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jianghui","middleName":"","lastName":"Tang","suffix":""},{"id":319398837,"identity":"d3d021d3-fb1f-4daa-a260-cde089f042b5","order_by":9,"name":"Yongtao Ji","email":"","orcid":"","institution":"Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Yongtao","middleName":"","lastName":"Ji","suffix":""},{"id":319398838,"identity":"2cb86ab0-9974-4cdf-94a0-c50dbc6e70dc","order_by":10,"name":"Taohong Li","email":"","orcid":"","institution":"Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Taohong","middleName":"","lastName":"Li","suffix":""},{"id":319398839,"identity":"b96d87b2-e06e-40c0-890f-dfd17a250069","order_by":11,"name":"Tianxing Yan","email":"","orcid":"","institution":"Peking University Health Science Center","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Tianxing","middleName":"","lastName":"Yan","suffix":""},{"id":319398840,"identity":"12c0478c-9f91-4222-a0d3-a4bf2c0bc8a1","order_by":12,"name":"Wei Shao","email":"","orcid":"","institution":"Nanjing University of Aeronautics and Astronautics","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Shao","suffix":""},{"id":319398841,"identity":"cc5ff1f3-5b30-4bdf-b20a-7fb5e30369e3","order_by":13,"name":"Xueli Bai","email":"","orcid":"https://orcid.org/0000-0002-2934-0880","institution":"The First Affiliated Hospital, Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xueli","middleName":"","lastName":"Bai","suffix":""},{"id":319398842,"identity":"67eb39a9-80e6-4400-81c1-475a86e3b99d","order_by":14,"name":"Qing Li","email":"","orcid":"","institution":"Chongqing University Cancer Hospital and School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Qing","middleName":"","lastName":"Li","suffix":""},{"id":319398843,"identity":"450681d8-df57-46ca-9f4a-14de7de692fd","order_by":15,"name":"Tingbo Liang","email":"","orcid":"https://orcid.org/0000-0003-0143-3353","institution":"The First Affiliated Hospital, Zhejiang University School of Medicine","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Tingbo","middleName":"","lastName":"Liang","suffix":""}],"badges":[],"createdAt":"2024-06-06 03:00:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4536926/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4536926/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":59285301,"identity":"820e3704-e7a8-48c8-b36c-1945c29c2e3a","added_by":"auto","created_at":"2024-06-28 16:20:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":431847,"visible":true,"origin":"","legend":"\u003cp\u003eClinical trial of TACE+ICI in advanced HCC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA\u003c/strong\u003e. Clinical trial design. Enrolled patients received sintilimab at a dose of 200 mg through intravenous infusion at the beginning of the DEB-TACE procedure on day 1 of the first treatment cycle and every 3 weeks thereafter. DEB-TACE procedures were repeated every 4-6 weeks based on tumor response, and sintilimab treatment was continued for a maximum of 3 cycles or until surgical resection, radiologic disease progression, unacceptable toxICBty, or withdrawal from the study. \u003cstrong\u003eB. \u003c/strong\u003eClinical characteristics of patients enrolled, including treatment response, TNM grading, HBV antigen, and liver function. \u003cstrong\u003eC.\u003c/strong\u003e Patient enrolment encompassed 60 patients who exceeded the Milan criteria for the clinical trial. However, 10 patients who underwent more than one cycle of combination treatment were excluded, along with three patients who lacked pre-operation biopsy. Additionally, nine patients who lacked post-operation tissues were also excluded from the study, thereby selecting 38 pairs of HCC tissues for IMC analysis. \u003cstrong\u003eD. \u003c/strong\u003eRepresentative computer tomograph of OR (objective response) and SD (stable disease) patients. Tumor is highlighted in red.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/65de634d39c52b755970ec5f.png"},{"id":59285302,"identity":"726f976d-9c2b-4999-bfbc-5d10baf8245b","added_by":"auto","created_at":"2024-06-28 16:20:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":969118,"visible":true,"origin":"","legend":"\u003cp\u003eIMC-derived spatial profiling of the HCC microenvironment before and after combination therapy demonstrates an upregulation of immunosuppressive epithelial cells in responders.\u003c/p\u003e\n\u003cp\u003eA. Heatmap showing the max-min normalized mean marker expression of 30 clusters. B. Segmented images of representative OR and SD regions of interest (ROIs). C. t-SNE plots based on the single-cell expression data extracted from IMC images. 18 major clusters of cells from pre and post-treatment HCC samples were defined according to their markers. D. Prevalence of the 18 major clusters as the proportion of total cells in SD-Pre, SD-Post, OR-Pre, and OR-Post respectively. E. The distribution of key cellular clusters based on IMC results (OR: 24 patients, SD: 14 patients). Statistical analysis was performed with a paired t-test (ns: p \u0026gt; 0.05, *: p \u0026lt;= 0.05, **: p \u0026lt;= 0.01, ***: p \u0026lt;= 0.001, ****: p \u0026lt;= 0.0001).\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/04ab7057dee2b958f48f5b81.png"},{"id":59285842,"identity":"91fb6059-322c-49b0-8e19-7c4adc59ccee","added_by":"auto","created_at":"2024-06-28 16:28:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1144816,"visible":true,"origin":"","legend":"\u003cp\u003eTherapeutic response characterization: co-localization of immunosuppressive epithelial cells with T cells in responders versus fibrosis in non-responders. A. Schematic representation of cellular neighborhoods (CNs), defined as 10 nearest neighbor cells relative to a center cell. B. 15 distinct CNs were identified based on the abundances of original cellular clusters within each CN. C. IMC images and corresponding Voronoi diagrams of the CNs, epithelial area, immune suppressive epithelial area, fibroblasts area, immune area, immune suppressive epithelial and immune area, fibroblast and epithelial area were shown for both representative OR and SD patients. Patients who exhibited an objective response (OR) demonstrated a pronounced enrichment of immune-suppressive territories, including regions characterized by both immune suppression and areas where immune cell populations intersect. Conversely, patients with stable disease (SD) showed a marked prevalence of fibroblast-centric zones, as well as locales where epithelial and fibroblast domains overlap. D. Box plots were presented to show the relative abundance of selected areas between OR and SD patients before and after treatments. In the cohort of individuals achieving an objective response (OR), there was a notable concentration of regions typified by immune suppression, including areas of confluence between immune-suppressive cells and other immune cell types. In contrast, individuals with stable disease (SD) predominantly exhibited regions dense with fibroblasts, as well as interfaces where epithelial and fibroblast cells co-localize. Statistical analysis was performed using paired t-tests (ns: p \u0026gt; 0.05, *: p \u0026lt;= 0.05, **: p \u0026lt;= 0.01, ***: p \u0026lt;= 0.001, ****: p \u0026lt;= 0.0001). E. Heatmap showing the average number of cell-cell interactions between different cell types in fibroblasts/epithelial positive areas. The Y-axis represents the central cell, while the X-axis denotes neighboring cells. Interaction counts between cells were initially quantified within the fibroblast/epithelial region, followed by a comparative analysis against the remaining areas. Interactions with |log2(FoldChange)|\u0026gt;1 and p-value\u0026lt;0.05 were labeled with larger size. F. Differential interaction analysis between Immune Suppressive Epithelial/Immune positive area and other regions.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/eb2980e2cfef0b14ca54a517.png"},{"id":59285306,"identity":"de101544-273e-4c84-8f16-a6d2b574641e","added_by":"auto","created_at":"2024-06-28 16:20:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":934078,"visible":true,"origin":"","legend":"\u003cp\u003ecGAS-STING pathway is the central regulator of therapeutic response.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA.\u003c/strong\u003e t-SNE plot based on scRNA-seq data, showing the identification of 10 major cell types. This plot visualizes the cellular heterogeneity within the TME, enabling the characterization of distinct cell populations. \u003cstrong\u003eB.\u003c/strong\u003e t-SNE plot displaying the key markers associated with each major cell type. By mapping the expression levels of these markers, we can determine the molecular signatures that define different cell populations within the tumor. \u003cstrong\u003eC \u003c/strong\u003e\u0026amp;\u003cstrong\u003e D.\u003c/strong\u003e t-SNE plots displaying the sub-clusters of T cell populations (D) and the key markers associated with each T cell sub-cluster (E). These plots provide a detailed view of the T cell heterogeneity, allowing us to identify distinct subsets of T cells with unique functional characteristics. \u003cstrong\u003eE \u003c/strong\u003e\u0026amp;\u003cstrong\u003e F.\u003c/strong\u003et-SNE plots displaying the sub-clusters of macrophage populations (F) and the key markers associated with each macrophage sub-cluster (G). These plots highlight the diversity of macrophages within the TME, revealing distinct functional states and activation profiles. \u003cstrong\u003eG \u003c/strong\u003e\u0026amp;\u003cstrong\u003e H.\u003c/strong\u003e t-SNE plots displaying the sub-clusters of epithelial cell populations (H) and the key markers associated with each epithelial cell sub-cluster (I). These plots uncover the heterogeneity of epithelial cells, providing insights into different subtypes and their potential roles in tumor development and progression. \u003cstrong\u003eI \u003c/strong\u003e\u0026amp;\u003cstrong\u003e J.\u003c/strong\u003e t-SNE plots displaying the sub-clusters of fibroblast populations (J) and the key markers associated with each fibroblast sub-cluster (K). These plots elucidate the complexity of fibroblasts in the TME, highlighting distinct subpopulations with unique functional properties. \u003cstrong\u003eK \u003c/strong\u003e\u0026amp;\u003cstrong\u003eL.\u003c/strong\u003e Representative plots showing overlapped spots in the spatial transcriptomic samples that correspond to the Immune Suppressive Epithelial \u0026amp; T Cells (K) and Epithelial \u0026amp; Fibroblasts (L) regions. By pinpointing the spatial locations of these specific cell populations, we can gain spatial context and understand their interactions within the TME. \u003cstrong\u003eM. \u003c/strong\u003eRepresentative Western Blot of P-STING, p-TBK1, p-IRF3 and p-RelA in OR and SD groups showing that cGAS signaling is activated in the OR group, but not the SD group. \u003cstrong\u003eN. \u003c/strong\u003eRepresentative mIHC images add evidence that cGAS is activated in OR patients. Expression of cGAS, PD-L1, RelA, and Pan-CK, markers of cGAS activity, in cells from tumor tissues are shown for the Objective Response (OR) and Stable Disease (SD) groups. \u003cstrong\u003eO.\u003c/strong\u003e Quantitative assessment of mIHC data showing the frequency of cGAS, PD-L1, dsDNA, Collagen I, and CD45RO in both tumor and fibroic areas of the OR and SD groups. \u003cstrong\u003eP. \u003c/strong\u003eModel of how fibrosis and cGAS/dsDNA impact immune response in the HCC TME.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/03c432e1c6705e3ccc458a2f.png"},{"id":59285304,"identity":"0125f9e1-b0e1-4b25-ae6d-6cec8c3fffba","added_by":"auto","created_at":"2024-06-28 16:20:48","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":642269,"visible":true,"origin":"","legend":"\u003cp\u003eThe antifibrotic elafibranor enhances combination therapy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eDesign of mouse model clinical trial. An orthotopic liver cancer model was established through tail veinplasmid injection. Elafibranor (antifibrotic), 2',3'-cGMP (STING agonist), and PD-1 monoclonal antibody were administered at specific doses and time points. Tumors were collected and analyzed at a designated time point to evaluate the efficacy of the triple combination therapy. \u003cstrong\u003eB. \u003c/strong\u003eTumor response in the orthotopic liver cancer model to control (Ctrl), Elafibranor, 2',3'-cGMP\u0026amp;PD-1, and Elafibranor combined with 2',3'-cGMP\u0026amp;PD-1. Qualitatively, combination regimens were more effective at reducing tumor nodules. \u003cstrong\u003eC. \u003c/strong\u003eQuantitation of liver surface tumors. Elafibranor + 2’,3’-cGMP\u0026amp;PD-1 showed the fewest tumors and was statistically better than all other treatments. \u003cstrong\u003eD.\u003c/strong\u003eRepresentative mIHC images showing the expression of CD8, Collagen I, and Pan-CK in tumor tissues of the different treatment groups. These images visualize the distribution of immune cells, extracellular matrix components, and tumor cells, providing insights into the TME and the effects of the treatments.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/d8f30da433b5c2d9ef7ce902.png"},{"id":59285843,"identity":"c6a03462-97b2-4d28-ab45-ea8669790cb8","added_by":"auto","created_at":"2024-06-28 16:28:48","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":760620,"visible":true,"origin":"","legend":"\u003cp\u003ePrediction of Combinatorial Treatment Outcomes Using Graphical Classification based on Spatial Similarity (GCSS) with a Restricted Marker Panel\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA. \u003c/strong\u003eWorkflow of the GCSS model. Here, the location of each identified cell cluster within the tissue sample is first precisely mapped. Following this, a spatial similarity matrix is generated based on the distribution of these cell clusters. This matrix quantifies the degree of spatial similarity between different clusters. Leveraging this spatial similarity data, the GCSS model then predicts whether a sample belongs to the responder (OR) or non-responder (SD) group. \u003cstrong\u003eB.\u003c/strong\u003e Spatial Similarity Matrix: This panel shows the spatial similarity matrix, which provides a comprehensive representation of the spatial relationships between different cellular clusters within the tissue sample. The matrix captures the level of similarity in the spatial distribution patterns, enabling the identification of spatially related clusters. \u003cstrong\u003eC.\u003c/strong\u003e Model Performance: This panel presents the Receiver Operating Characteristic (ROC) curve to evaluate the performance of the GCSS model (including a version based on 6 core markers - GCSS-1), alongside two other algorithms, Lasso and Support Vector Machine (SVM). The ROC curve illustrates the trade-off between the true positive rate and the false positive rate for different classification thresholds. \u003cstrong\u003eD.\u003c/strong\u003e AUC Values: This panel shows the Area Under the Curve (AUC) value calculated for each algorithm (GCSS, GCSS-6 marker, Lasso, and SVM). The AUC provides a measure of the model's ability to accurately distinguish between the OR and SD groups. \u003cstrong\u003eE.\u003c/strong\u003e Critical Cell-Cell Interactions: This panel presents critical cell-cell interaction edges identified by the GCSS model. These edges represent the important communication pathways between different cell types within the TME that are associated with the responder (OR) and non-responder (SD) groups. \u003cstrong\u003eF. \u003c/strong\u003eMinimized Marker Panel: Based on the critical cell-cell interaction edges identified in Fig. 5E, a minimized marker panel consisting of six markers (CD4, CD8, CD68, Pan-CK, and PD-L1) is presented here. Immunohistochemistry (IHC) staining was performed using these markers to visualize their expression in the tissue samples. \u003cstrong\u003eG.\u003c/strong\u003e Minimized Marker Panel Performance: This panel evaluates the performance of the minimized marker panel using an independent cohort of hepatocellular carcinoma (HCC) patients. The ROC curve is plotted to assess the predictive ability of the panel in distinguishing between responders and non-responders in this independent cohort.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/ba329b1b6d9103e7423a3c87.png"},{"id":62440602,"identity":"7244aefa-4620-4417-94db-58483ba2e3ad","added_by":"auto","created_at":"2024-08-14 08:48:47","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5701137,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/beb69e6c-f818-49af-a44c-efbbb5b3bf5f.pdf"},{"id":59285307,"identity":"ae1ce16a-6f03-44da-ba0a-fdb9aef87fc1","added_by":"auto","created_at":"2024-06-28 16:20:48","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12900982,"visible":true,"origin":"","legend":"Supplementary Figures","description":"","filename":"SMOcGASSTINGHCCSINaturecancer.docx","url":"https://assets-eu.researchsquare.com/files/rs-4536926/v1/b372a210b101404c81c76539.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Fibrosis Impairs cGAS-Mediated Responses to Immunotherapy in Advanced HCC","fulltext":[{"header":"Significance","content":"\u003cp\u003eWe present a paradigm shift in HCC management by integrating multi-omics data with spatial information to guide personalized therapy. We believe these findings hold significant promise for improving patient outcomes in HCC and potentially other cancer types.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eHepatocellular carcinoma (HCC) accounts for nearly 90% of primary liver cancer cases (1). Early detection of HCC is challenging and treatment options for inoperable patients are often ineffective, making HCC the third leading cause of cancer-related mortality in the world (2). Nevertheless, immunotherapy has improved the treatment landscape of HCC. Immune checkpoint blocking (ICB) therapy is a pillar of modern immunotherapy that functions to revers suppression of cytotoxic T lymphocytes (CTLs) by blocking immune checkpoint molecules such as programmed cell death protein 1 (PD-1). This enables the killing capacity of CTLs against tumors. In advanced HCC, combining ICB with anti-angiogenics such as bevacizumab or vascular endothelial growth factor receptor (VEGFR) inhibitors modifies tumor vasculature to bolster immune response, resulting in improved overall survival (3-6).\u003c/p\u003e\n\u003cp\u003eIn addition to systemic therapy, transarterial chemoembolization (TACE) is often used to improve loco-regional control for intermediate stage HCC with large or multinodular tumors without vascular invasion or extrahepatic spread (7, 8). TACE reduces tumor burden and can improve outcomes for these patients (9). One potential synergistic advantage of combining TACE with immunotherapy is that it modifies the immune microenvironment of HCC, including recruitment of antigen-specific T-cell and NK cells. This has motivated exploration of regimens that combine TACE with immune checkpoint inhibitors (10-12) and recently the EMERALD-1 trial, which evaluated a combination of TACE + durvalumab + VEGF inhibitor in unresectable HCC, showed an improvement in progression-free survival with the addition of ICB + VEGF inhibitor (13). Despite these advances, many patients still progress after therapy (14). We hypothesized that features of the tumor microenvironment (TME) may identify patients who respond well to dual therapies and may inform strategies to address non-responders.\u003c/p\u003e\n\u003cp\u003eTo discover such features, we evaluated specimens from a clinical trial (NCT04174781) of transarterial chemoembolization (DEB-TACE) and sintilimab (PD-1 antibody) in patients with HCC of BCLC stage A who exceeded the Milan criteria or BCLC stage B (15). We focused on extensive multi-omics analysis of tumor specimens before and after therapy. Using this approach, we were able to uncover regional response determinants, elucidate distinct mechanisms underlying regional responses and non-responses, optimize clinical trial strategy, and establish a predictive model to guide personalized treatment for HCC.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eTrial of TACE + ICB for Biomarker Discovery\u003c/h2\u003e\n\u003cp\u003eTo evaluate the biological determinants of HCC responses following TACE with PD-1 antibody treatment, we evaluated specimens from patients enrolled in a clinical trial of DEB-TACE and sintilimab in locally advanced HCC (15). Patients had HCC of BCLC stage A who exceeded the Milan criteria or had BCLC stage B (Fig. 1A, Table S1-S2, NCT04174781). All patients had Child-Pugh A liver function and did not have vascular invasion, extrahepatic metastasis, or clinically significant portal hypertension. Sixty-one patients enrolled, all of whom underwent at least one treatment cycle combining DEB-TACE and sintilimab (Fig. 1A). The clinical characteristics of trial patients are summarized in Table S1. The majority of the patients were male (52 out of 61, representing 85%), with a median age of 58 years (ranging from 26 to 75 years old). Of these patients, 85% (52 individuals) had two or fewer lesions, while 56% (34 individuals) were diagnosed with BCLC stage B HCC. The median size of the target tumor was 7 cm (ranging from 2.1 to 14.1 cm).\u003c/p\u003e\n\u003cp\u003eIn total, 82% (50 out of 61) of the patients underwent only one treatment cycle with sintilimab combined with DEB-TACE, while 18% (11 patients) received two or three treatment cycles. At the data cutoff date, out of the 60 patients assessed for effectiveness, 37 (62%) exhibited an objective response to DEB-TACE combined with sintilimab; 3 patients had a complete response (5%), 34 a partial response (57%) and 20 experienced stable disease. Overall, the disease control rate was 95% (57/60) with a median follow up of 26 months. Patients showing complete and partial response were categorized as having an objective remission (OR), while patients showing steady (SD) were classified as such (Fig. 1B and Fig. S1A). Of the treated patients, adequate biopsy specimens were collected from 38, enabling analysis of biological correlates (Fig. 1C). Representative computer tomograph of OR (objective response) and SD (stable disease) patients were shown in Fig. 1D.\u003c/p\u003e\n\u003ch2\u003eImmune Suppressive Epithelial Cells Elevate in Responders\u003c/h2\u003e\n\u003cp\u003eWe hypothesized that biological markers known to be active in modulating the HCC TME might be associated with treatment response. We generated a comprehensive spatial multi-omics dataset from patient samples. We excluded patients who underwent multiple rounds of TACE or did not have both pre- and post-treatment samples (Fig. 1C). As an initial evaluation, we devised a 40-marker IMC panel (39 protein markers and DNA) (Fig. 2A) based on a list compiled for the study of cellular neighborhoods in HCC (16). To ensure comprehensive coverage of each slide during IMC scanning, we performed scans across multiple regions (Fig. S1B). All antibodies were validated through immunohistochemistry (IHC) before heavy metal conjugation, and each metal-bound antibody was further verified by IMC (Fig. S2-4). The panel included markers for epithelial, endothelial, and stromal cells, various immune cells, as well as the cytokines IL-1\u0026beta;, TNF-\u0026alpha;, and IL-6, the proliferation marker Ki-67, and the apoptotic indicator cleaved caspase-3 (Table S2). In total, mass cytometry analysis was done on 459 tumor ROIs from pre and post therapy samples (SD, n=15; OR, n=23).\u003c/p\u003e\n\u003cp\u003eMarker expression levels in each cell were quantified and converted into a matrix. Following batch correction, approximately 1,895,221 cells were grouped into 19 cell meta-clusters identified by PhenoGraph (Fig. 2A to C). These meta-clusters were then annotated based on marker expression (Fig. 2C, Fig. S5A and B). Macrophage subsets, including resident macrophages (CD11b\u003csup\u003elow\u003c/sup\u003e CD68\u003csup\u003e+\u003c/sup\u003e) and infiltrating macrophages (CD11b\u003csup\u003e+\u003c/sup\u003e CD68\u003csup\u003e+\u003c/sup\u003e CD14\u003csup\u003e+\u003c/sup\u003e) were separately identified (Fig. 2A). T cell subsets (CD4\u003csup\u003e+\u003c/sup\u003e or CD8\u003csup\u003e+\u003c/sup\u003e), were further refined based on functional markers (Fig. 2A). Additionally, we observed epithelial cells (Pan-Cytokeratin\u003csup\u003e+\u003c/sup\u003e) and immune-suppressive epithelial cells (PD-L1\u003csup\u003e+\u003c/sup\u003e/B7H4\u003csup\u003e+\u003c/sup\u003e Pan-Cytokeratin\u003csup\u003e+\u003c/sup\u003e). Fibroblasts were further divided into Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts and \u0026alpha;-SMA\u003csup\u003ehi\u003c/sup\u003e myofibroblasts (17) (Fig. 2A). Some clusters contained multiple cell types, such as epithelial cells and T cells, immune-suppressive epithelial cells and resident macrophages, and resident macrophages and fibroblasts, indicating close interactions of neighboring cells. One cluster from lineage-negative cells was also detected (Fig. 2A and Fig. S5). The frequencies of these cells in evaluated patients are shown in Fig. 2D.\u003c/p\u003e\n\u003cp\u003eWe observed a counterintuitive pattern in immune-suppressive epithelial cells, which increased after the treatment only within the OR group (Fig. 2E). In contrast, we identified opposing trends in \u0026alpha;-SMA\u003csup\u003ehi\u003c/sup\u003e myofibroblasts and Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts following the combination treatment. The number of \u0026alpha;-SMA\u003csup\u003ehi\u003c/sup\u003e myofibroblasts decreased, while Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts increased post-treatment. The increase in Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts was less pronounced in the OR group (Fig. 2E). The combination treatment also revealed opposing patterns in Tregs and CD4/8\u003csup\u003e+\u003c/sup\u003e T cells, as Tregs experienced a decline while CD4/8\u003csup\u003e+\u003c/sup\u003e T cells exhibited a rise. In addition, both resident and CD11c\u003csup\u003e+\u003c/sup\u003e macrophages demonstrated a frequency increase after the combined therapy (Fig. 2E). These observations suggest that therapy initiates an unexpected adaptive resistance mechanism. A decrease in \u0026alpha;-SMA\u003csup\u003ehi\u003c/sup\u003e myofibroblasts along with an increase in Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts could reflect remodeling of the extracellular matrix that may either support or hinder an antitumor immune response, depending on the context. The observed decrease in Tregs alongside an increase in CD4/8\u003csup\u003e+\u003c/sup\u003e T cells aligns with the intended immunostimulatory effects of PD-1 blockade, which aims to bolster antitumor immunity. Additionally, the rise in both resident and CD11c\u003csup\u003e+\u003c/sup\u003e macrophages after the combination therapy underscores the dynamic nature of innate immune responses.\u003c/p\u003e\n\u003ch2\u003eTreatment Response is Marked by Organizational Changes in the TME\u003c/h2\u003e\n\u003cp\u003eBeyond tissue composition, we hypothesized that cellular organization may also contribute to or predict for HCC responses to combined therapy. We utilized high-resolution spatial data to identify in situ alterations to cellular neighborhood organization after the combined therapy (18, 19), which we defined as the ten closest neighbor cells surrounding a central cell (Fig. 3A). We analyzed the cellular composition of the neighbors and annotated these neighborhoods based on the primary cellular clusters, resulting in various functional cellular neighborhood units (Fig. 3B). Additionally, we employed a Voronoi diagram, which was applied to each IMC image using the CN composition, to visualize the correlation between the defined CN functional units and the original IMC images (Fig. 3C) (20). The IMC topology map, depicted by the Voronoi diagram with different CNs represented by distinct colors (as shown in Fig. 3C), aligned well with the IMC image, confirming that cellular neighborhood assignments were correct.\u003c/p\u003e\n\u003cp\u003eFurthermore, we adapted previously described methods to identify important regions within the HCC TME (21, 22). An expansive homotypic cellular cluster, as identified by uniform cellular phenotypes, was demarcated as a discrete spatial domain within the TME. Notable regions included the epithelial region, immune-suppressive epithelial region, fibroblast region, and the contiguous interfaces where immune-suppressive epithelial cells/immune cells, as well as regions of epithelial cells and fibroblast co-localization (Fig. 3C). Similar to the bulk frequency analysis, we found that the immune-suppressive epithelial region increased in the OR group after treatment, while the epithelial region decreased in the OR group. The fibroblast region showed an increase in both the SD and OR groups, although the magnitude of increase was less pronounced in the OR group. Additionally, the immune region experienced an increase in the OR group post-treatment. The epithelial/fibroblast intersection region and the immune suppressive epithelial/immune region displayed contrasting patterns in the OR group after treatment, with the former decreasing and the latter increasing (Fig. 3C and D).\u003c/p\u003e\n\u003cp\u003eWe performed interaction analysis to uncover potential mechanisms of resistance in the SD group. In that group, we observed a higher number of cellular interactions involving fibroblasts and epithelial/fibroblast interactions. Conversely, cellular interactions involving infiltrating macrophages, NK cells, and various T cells were notably absent in the epi-fibroblast intersection regions (Fig. 3E). On the other hand, in the immune suppressive-immune intersection regions, we observed higher cellular interactions involving immune suppressive epithelial cells and various T cells, while fibroblast interactions were comparatively reduced (Fig. 3F). This analysis suggests that distinct cellular subsets orchestrate environmental factors that give rise to contrasting regional features. Specifically, immune suppressive epithelial cells are in position to mediate cellular interactions with immune cells, particularly various T cells, in the OR group. In contrast, the SD group lacked these cellular interactions. These patterns suggest that immune-suppressive epithelial cells recruit T cells, whereas epithelial cells promote fibroblast enrichment, which may reduce the efficacy of therapeutic responses.\u003c/p\u003e\n\u003ch2\u003eSpatial and Single-Cell Transcriptomics Identify cGAS as a Regulator of the TME\u003c/h2\u003e\n\u003cp\u003eOur IMC analysis allowed us to visualize and quantify how cellular neighborhoods respond to TACE + ICB. To discover the biological processes that regulate these changes, we used single-cell RNA sequencing and spatial transcriptomics to evaluate samples (Fig. S6A). For this post-treatment samples from 6 patients in the OR and 4 patients in the SD cohorts were evaluable. We captured a total of 92,785 single cells from 6 OR and 4 SD patients post-surgery, with high sequencing quality. \u0026nbsp;We identified 10 primary cell populations (Fig. 4A) based on specific markers (Fig. 4B), such as EpCAM\u003csup\u003e+\u003c/sup\u003e epithelial cells, VWF\u003csup\u003e+\u003c/sup\u003e endothelial cells (23) , and ACTA2\u003csup\u003e+\u003c/sup\u003e fibroblasts (24) (Fig. 4B). We also identified various immune cell subsets, including CD3\u003csup\u003e+\u003c/sup\u003e T cells, CD79B\u003csup\u003e+\u003c/sup\u003e B cells, IGHG1\u003csup\u003e+\u003c/sup\u003e plasma B cells, CD68\u003csup\u003e+\u003c/sup\u003e macrophages, GNLY\u003csup\u003e+\u003c/sup\u003e natural killer (NK) cells (25), and CPA3\u003csup\u003e+\u003c/sup\u003e mast cells (26) (Fig. 4B). Additional clustering of T cells revealed 5 subsets (Fig. 4C and D): Granzyme A/K\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e effector T cells, CD127\u003csup\u003e+\u003c/sup\u003e CCR7\u003csup\u003e+\u003c/sup\u003e central memory CD8\u003csup\u003e+\u003c/sup\u003e T cells (CD8\u003csup\u003e+\u003c/sup\u003e Tcm), Granzyme A/K\u003csup\u003e+\u003c/sup\u003e IL7R\u003csup\u003e+\u003c/sup\u003e effector memory CD8\u003csup\u003e+\u003c/sup\u003e T cells (CD8\u003csup\u003e+\u003c/sup\u003e Tem) (27) , CTLA4\u003csup\u003e+\u003c/sup\u003e exhausted CD4\u003csup\u003e+\u003c/sup\u003e T cells, and FOXP3\u003csup\u003e+\u003c/sup\u003e regulatory T cells (Treg) (Fig. 4C and D). We identified two major subsets of macrophages: resident macrophages and infiltrating macrophages (Fig. 4E and F). Resident macrophages exhibited distinct markers, such as C1QA/B, which distinguished them from infiltrating macrophages (28). In contrast, infiltrating macrophages displayed higher levels of CCR2 and S100A8/9, than resident macrophages (Fig. 4E and F)\u003c/p\u003e\n\u003cp\u003eSimilar to the IMC analysis, we discerned two distinct subsets of epithelial cells within the TME: immune suppressive epithelial cells expressing PD-L1, and a separate population of epithelial cells devoid of PD-L1 expression (Fig. 4G and H). The PD-L1 population plays a regulatory role in immune recognition, whereas the PD-L1 null population, represents a phenotypically different group of epithelial cells within the same tissue context. We classified fibroblasts into two subtypes based on the IMC results: Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts and \u0026alpha;-SMA\u003csup\u003ehi\u003c/sup\u003e myofibroblast cells (Fig. 4I and J). Our scRNA-seq landscape encompassed the key cellular components of the HCC TME and are consistent with the IMC results. These findings provide a detailed accounting of the diverse cellular populations in the TME and frame their potential roles in tumor immune responses.\u003c/p\u003e\n\u003cp\u003eTo unravel the mechanisms underlying the spatial and cellular interactions that dictate the regional response to TACE and ICB treatment, we performed an integrative analysis of single-cell RNA sequencing (scRNA-seq) and corresponding spatial transcriptomics (29). We employed a robust cell type decomposition (RCTD) approach, a computational method that harnesses the cell type profiles acquired from scRNA-seq data to decompose cell type mixtures while accounting for discrepancies across sequencing technologies. We used RCTD to decompose the spots within the spatial transcriptomics of all tissue sections, based on the refined immune and non-immune cellular clusters. After single-cell spatial deconvolution, we evaluated each spot for critical cell types identified by the IMC analysis, including epithelial cells, immune suppressive epithelial cells, fibroblasts, and T cells (Fig. S6B-E). This approach let us to accurately assess the presence and distribution of these key cell types across the tissue sections, providing insights into the spatial organization of the TME. As Fig. S6B-E shows, epithelial cells and fibroblast spots were significantly enriched in the SD patients. By contrast, immune suppressive epithelial cells and T cell spots were heavily enriched in the OR patients. Notably, we observed double positive spots (green spots) indicating the co-occurrence of immune suppressive epithelial cells and T cells was prominently observed in the OR patients (Fig. 4K and Fig. S6F). Conversely, double positive spots for epithelial cells and fibroblasts were predominantly enriched in the SD patients (Fig. 4L and Fig. S6G). These spatial transcriptomic insights suggest that immune-suppressive epithelial cells recruit T cells, enhancing therapeutic efficacy, while epithelial cells promote fibrosis, impeding treatment response.\u003c/p\u003e\n\u003cp\u003eTo understand the biological determinants this organization, we conducted KEGG pathway analyses on the spots with dual positivity for immune suppressive epithelial cells and T cells (Fig S7A). In the OR group, immune activation pathways such as chemokine signaling, T cell differentiation/trans-endothelial migration, and NF kappa B were evident. We also observed immune checkpoint signaling following immune activation (Fig. S7A). We also noted activation of cytosolic DNA sensing (cGAS pathway) pathways in the top group (Fig. S7A). In contrast, in spots with dual positivity for epithelial cells and fibroblasts (Fig. S7B), we observed metabolism-related pathways and oxidative phosphorylation pathways. Particularly noteworthy was the identification of the PPAR signaling pathway, which plays a key role in fibrosis (30), in the SD group (Fig. S7B).\u003c/p\u003e\n\u003cp\u003eTo validate integrated single-cell and spatial transcriptomic findings, we performed bulk RNA transcriptomic analysis on post-treatment samples obtained from a separate cohort of 19 patients treated with the same regimen (Table S1). Among the top differential genes identified, one was cGAS, a major regulator of double stranded DNA recognition (31) (Fig. S7C). To demonstrate the role of cGAS-STING in regulating immune suppression in our patient samples, we conducted several analyses. First, we performed correlation analysis for all the pathways identified in Fig. S7A, achieved by employing gene signatures specific to each pathway for Gene Set Variation Analysis (GSVA). The gene signatures served as input for calculating the GSVA scores. Analysis focused on immune suppressive epithelial and T cell spots across all pathways outlined in Fig. S7A. Subsequently, these GSVA scores were utilized to compute the correlations among the pathways. We noticed the cGAS-STING pathway correlated positively with the most enriched molecular pathways, including those for T-cell recruitment and migration (Fig. S7D), with the exception of the pyroptosis pathway (Fig. S7D). Second, we used SCENIC (single-cell regulatory network inference and clustering) analysis to identify key transcription factors that form the gene regulatory network within immune suppressive epithelial cells (32). We detected cGAS-STING related transcription factors such as IRF3 and NF-\u0026kappa;B (Fig. S7E) within immune suppressive epithelial cells. We validated the SCENIC results through Western blotting of individual patient tumor samples which showed elevated levels of phosphorylated versions of STING pathway proteins, TBK1, IRF3, STING, and RelA in the OR group but not the SD group (Fig. 4M). These findings suggest the STING pathway is activated in responsive samples, reinforcing the idea that STING participates in the treatment response. Consistently, mIHC staining also showed more cGAS signal and downstream RelA (33) signal in the OR group, while these signals were absent in the SD group (Fig. 4N). Epithelial cells in the OR group displayed high levels of PD-L1 expression (Pan-CK\u003csup\u003e+\u003c/sup\u003e) compared to the absence of PD-L1 expression in epithelial cells in the SD group (Fig. 4O). Using mIHC we observed colocalization of PD-L1 and the cGAS expression (Fig. 4N). We note that this connection between cGAS-STING pathway and immune checkpoint expression has been observed previously (34).\u003c/p\u003e\n\u003cp\u003eTo further confirm the association between cGAS signaling and PD-L1 expression, we exposed a panel of three human HCC cell lines to the topoisomerase inhibitor epirubicin, which causes double-stranded DNA (dsDNA) breaks. This caused cGAS signaling and upregulation in PD-L1 expression (Fig. S8), consistent with a previous report (35). This may partially explain why single administration of STING agonists have not yielded satisfactory results in clinical trials since upregulation of PD-L1 could theoretically attenuate the anti-tumor immune response (36). Nevertheless, PD-L1 expression alone does not faithfully predict for immunotherapeutic response in hepatocellular carcinoma (37).\u003c/p\u003e\n\u003cp\u003eWe further analyzed the samples to find actionable hypotheses related to therapeutic vulnerabilities. Utilizing spatial transcriptomics to analyze samples from OR and SD cohorts, we delineated distinct spatial features unique to each group. In the OR cohort, regions of overlap between immune suppressive epithelial cells and T cells were pronounced, indicating a potential niche for immune modulation. Conversely, the SD cohort was characterized by a notable co-localization of epithelial cells and fibroblasts, suggesting a tissue composition of fibrosis. To further elucidate the cellular interactions underlying the recruitment of T cells by immune-suppressive epithelial cells and the promotion of fibrosis by epithelial cells, we performed a ligand-receptor analysis on both OR and SD patients, integrating data from both Spatial Transcriptomics (ST) and single-cell RNA sequencing (scRNA-seq). We used NicheNET, a tool that combines expression data from intercommunicating cells with pre-existing knowledge of signaling and gene regulatory networks, to predict potential ligand-target associations (38). First, we selected our genes of interest from the differentially expressed genes, which we calculated based on ST datasets for specific areas identified via IMC analysis. Next, we applied scRNA-seq for cellular interaction analysis via NicheNET. Our observations indicated that PD-L1\u003csup\u003e+\u003c/sup\u003e epithelial cells could secrete multiple cytokines and chemokines, such as CCL5 (39), MIF (40), and LTB (41), thereby recruiting immune cells such as T cells (Fig. S9A). In contrast, in SD patients, epithelial cells transmitted numerous inhibitory signals for PPAR signaling pathways to fibroblasts (Fig. S9B), such as ATF3 (42), BCL6 (43), LCOR (44), EZH2 (45), NR0B2 (46), SIRT1 (47), GPS2 (48), and NRIP1 (49). Accordingly, based on ST analysis, we detected reduced PPAR signaling within the epithelial-fibroblast regions associated with SD patients (Fig. S9C), especially in areas showing dual-positive signals from both epithelial cells and fibroblasts. Our examination revealed downregulation of genes, such as PLIN2 (50), FABP6 (51), PLTP (52), UBC (53), and HMGCS1 (54) (Fig. S9C). These findings indicate a suppression of PPAR signaling in the non-responder (SD) patients.\u003c/p\u003e\n\u003cp\u003eTo validate NicheNET findings, we evaluated levels of double-stranded DNA (dsDNA) and subsequent activation of the cGAS signaling pathway using mIHC and found these were elevated in the OR group, along with expression of PD-L1. This activation was associated with increased recruitment and activation of T cells (CD45RO) (Fig. 4O and Fig. S9D). T cells also showed resident marker CD103 (Fig. S10). Notably fibrosis, as indicated by Collagen I staining, was relatively low in the OR group. Conversely, the SD group exhibited high levels of fibrosis, accompanied by low dsDNA and cGAS signaling. As a result, activated T cells were scarce in the SD group (Fig. 4O and Fig. S9D). These observations were consistent with the pathway analysis obtained through the integration of scRNA-seq and ST, which showed an interaction between dsDNA, cGAS signaling, fibrosis, and T cell activation.\u003c/p\u003e\n\u003cp\u003eIn conclusion, the above findings provide a model for how cGAS signaling can modulate responses to TACE + ICB in HCC. Activation of cGAS signaling by double-stranded DNA (dsDNA) results in increased recruitment and activation of T cells, as well as upregulation of PD-L1 expression in the responsive (OR) group. Conversely, non-responders (SD) group exhibit fibrosis, low dsDNA, and limited cGAS signaling, leading to reduced T cell recruitment and PD-L1 expression. These results highlight the significance of cGAS signaling and fibrosis in shaping the TME and suggest its potential as a therapeutic target in HCC (Fig. 4P).\u003c/p\u003e\n\u003ch2\u003eAddition of Anti-Fibrotic Agents Enhance Therapy for HCC\u003c/h2\u003e\n\u003cp\u003eThe observation that fibrosis limited the efficacy of TACE+anti-PD-1 treatment suggests that anti-fibrotic agents might enhance therapy. We evaluated this concept in a mouse HCC model induced by cMyc and Nras expression (16) since in murine models of HCC, activation of both Myc and Ras oncogenes has been implicated in the promotion of hepatic fibrosis (55, 56). Treatments were conditioned starting on day 14 with Elafibranor, a PPAR agonist known to have anti-inflammatory and anti-fibrotic effects. Elafibranor has shown promising results in reducing liver inflammation, oxidative stress, and fibrosis in preclinical liver disease models (57). On day 21, we mimicked TACE by treating with the STING agonist cGAMP; we also administered anti-PD-1 therapy (Sintilimab). After 30 days, we harvested the liver with tumors for further analysis (Fig. 5A).\u003c/p\u003e\n\u003cp\u003eBoth Elafibranor and the dual combination therapy of TACE+anti-PD-1 demonstrated reductions in tumor burden, with the dual combination therapy showing a greater reduction. However, the triple combination therapy involving Elafibranor, TACE, and anti-PD-1 exhibited the least tumor burden (Fig. 5B and C). Treatment with Elafibranor alone resulted in reduced fibrosis. The dual combination therapy of TACE+anti-PD-1 showed an increase in CD8\u003csup\u003e+\u003c/sup\u003e T cells. Notably, the triple combination therapy led to reduced fibrosis and increased CD8\u003csup\u003e+\u003c/sup\u003e T cells (Fig. 5D and Fig. S11).\u003c/p\u003e\n\u003ch2\u003eGraphical Classification Based on Spatial Similarity (GCSS) Predicts Treatment Response\u003c/h2\u003e\n\u003cp\u003eAn enticing clinical strategy would be to predict which patients are likely to have a poor response to therapy due to fibrosis, then add an anti-fibrotic agent for that population. We hypothesized that it might be possible to generate a predictive model using imaging mass cytometry (IMC) data obtained from needle biopsies. We used pre-treatment IMC images and associated clinical outcomes from the TACE+ICB trial as the training dataset for our model (Fig. 6A).\u003c/p\u003e\n\u003cp\u003eThe model was constructed by dividing segmented cells within each Region of Interest (ROI) into separate images (N=29), each reflecting a mask of specific cellular identity. Each image represents the spatial localization of a specific cell type. Subsequently, we calculate the pairwise spatial similarity between different cell types for each ROI, using a structural similarity matrix (SSIM) measurement (Fig. 6B). We then employed a network classification method (58), Graphical Classification based on Spatial Similarity (GCSS), to predict for association between spatial patterns of cells, including interactions, and the efficacy of therapy (Fig. 6A). The GCSS model considers both the spatial similarity information of different cell types and the network structure of the data in a computationally efficient way. To confirm the GCSS method was optimal, we conducted a comparison between the GCSS method and the classical support vector machine (SVM) and Least Absolute Shrinkage and Selection Operator (Lasso) methods. The results showed that GCSS exhibited higher sensitivity and accuracy compared to SVM and Lasso (Fig. 6C-D).\u003c/p\u003e\n\u003cp\u003eIncidentally, GCSS also provided confirmatory mechanistic data regarding cellular interactions within each group. In the SD group, we observed interactions between epithelial/fibroblast and CD45RA\u003csup\u003e+\u003c/sup\u003e CD8\u003csup\u003e+\u003c/sup\u003e T cells, effector CD8\u003csup\u003e+\u003c/sup\u003e T cells, and epithelial/fibroblast cells, as well as neutrophil and epithelial cell interactions. In the OR group, we found interactions between NK cells and infiltrating macrophages, effector CD8\u003csup\u003e+\u003c/sup\u003e T cells and immune suppressive epithelial cells, and GATA3\u003csup\u003e+\u003c/sup\u003e CD4\u003csup\u003e+\u003c/sup\u003e T cells and resident macrophages. These interactions are consistent with our integrated spatial multi-omics analysis and prior mechanistic exploration (Fig. 6E).\u003c/p\u003e\n\u003cp\u003eWe used dominant patterns in the model to create a minimized mIHC panel for predicting treatment response consisting of six markers, including CD4, CD8, CD68, Collagen I, Pan-CK, and PD-L1 (Fig. 6F). These markers represented the crucial cell interactions responsible for different treatment responses within the OR and SD groups (Fig. 6E). This approach is intended to enable clinical feasibility since it reduces assay costs and can be done more rapidly.\u003c/p\u003e\n\u003cp\u003eTo validate the predictive capability of the minimal panel, we collected 61 pre-treatment specimens from HCC patients who underwent dual combination treatment. Using mIHC input into the GCSS model, we achieved over 80% accuracy in predicting treatment response, similar to results on IMC data (compare Fig. 6C to 6G). These results suggest the mIHC panel will be effective for predicting treatment responses for HCC to TACE and ICB at minimal cost. This result was achieved despite a relatively small sample size, suggesting robustness in the method.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eImmunotherapy has shown that, on average, it can improve outcomes in unresectable HCC, but some patients do not respond. This study identified regional features within the TME of HCC that influence response to combined TACE and ICB therapy. Using spatial multi-omics, we revealed key interactions between Collagen I\u003csup\u003ehi\u003c/sup\u003e fibroblasts, immune-suppressive epithelial cells, and T cells. In responders, TACE treatment triggered dsDNA breaks, activating the cGAS pathway primarily in immune-suppressive epithelial cells. Conversely, non-responders displayed a diminished response to dsDNA sensing, accompanied by reduced cGAS activation in downstream effector cells. Additionally, we observed prominent fibrosis in non-responders, suggesting that triple combination therapy including an anti-fibrotic agent might be beneficial. This hypothesis was confirmed in a mouse model. Finally, to identify patients who may benefit from this approach, we developed a fast and cost-effective assay based on a GCSS model to predict HCC responses to TACE and ICB treatments.\u003c/p\u003e\n\u003cp\u003eWhile combination TACE therapies have shown promise in improving cancer outcomes, particularly with anti-angiogenic drugs like sorafenib, this study offers a more precise approach. Sorafenib disrupts blood vessel formation (angiogenesis) and cancer-promoting signals, demonstrating improved progression-free survival in HCC when combined with TACE (59). Given the distinct mechanisms of anti-angiogenic and anti-fibrotic drugs, we hypothesize that further combinations, including sorafenib or similar drugs alongside anti-fibrotic agents, might further optimize outcomes for specific patient groups.\u003c/p\u003e\n\u003cp\u003eThis study demonstrates that using direct measures of the composition and organization of the TME can lead to predictive models and mechanistic insights. \u0026nbsp; This is important because prior efforts utilizing gene signatures, circulating tumor cells, high-dimensional flow cytometry, single-cell RNA sequencing, microbiome analysis, radiomics, and clinical markers, have yielded inconsistent or weak associations with treatment response (37). Similarly, studies evaluating neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) in HCC patients receiving anti-PD-1 antibodies, as well as high-dimensional immune profiling of peripheral blood mononuclear cells (PBMCs), have not identified reliable biomarkers for predicting ICI response in HCC patients (37, 60).\u003c/p\u003e\n\u003cp\u003eThis study also expands on our previous research that identified distinct regions within the HCC TME - normal, fibrotic, and tumor tissue (16). Here, we delve deeper, uncovering molecular mechanisms in play: excessive fibrosis hinders the transmission of dsDNA signals, thereby preventing cGAS activation. This aligns with Yao et al.\u0026apos;s recent findings on a tumor immune barrier (TIB) composed of SPP1\u003csup\u003e+\u003c/sup\u003e macrophages and cancer-associated fibroblasts (CAFs) near the tumor boundary, which restricts immune cell infiltration and impacts the efficacy of immune checkpoint blockade (ICB) (61). We propose a more nuanced view \u0026ndash; a moderate level of fibrosis might actually benefit chemotherapy and immunotherapy responses. Our data, integrating current and past studies, suggests the fibrotic region serves as an anchor for diverse immune cell populations, including various macrophage subsets. Since macrophages play a critical role in cGAS signal transduction, their presence underscores their importance in these therapies.\u003c/p\u003e\n\u003cp\u003eUtilizing spatial multi-omics data integration, we discovered that features associated with response and non-response coexisted within the TME prior to treatment. Successful combination therapy appears to clear this \u0026quot;chaos\u0026quot; and enhance anti-tumor features via the cGAS pathway. Interestingly, traditionally immunosuppressive cells, such as neutrophils, PD-L1\u003csup\u003e+\u003c/sup\u003e epithelial cells, and fibroblasts, were found to be crucial for the response to combined TACE and ICB therapy. Based on this mechanistic understanding, we have established a predictive model with potential to guide real-world HCC patient treatment decisions. Nevertheless, clinical trials incorporating the model will be necessary for validation.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eDetailed protocols for clinical trial evaluation, IMC, other experiments, and data analysis are provided in Supplementary Information.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAUTHOR CONTRIBUTION\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJianpeng Sheng conceptualized the study, designed the experiments. Jianpeng Sheng, Zhiwei Zhou, Qing Li and Kenneth D. Westover drafted the manuscript. Lin Wang conducted data analysis, while Junlei Zhang and Jinyuan Song performed the experimental work. All three authors contributed equally to this work. Wei Shao and Qing Li developed the prediction model analysis and conducted imaging analysis. Yaxing Zhao provided assistance with bioinformatic analysis. Jianghui Tang, Yongtao Ji, and Taohong Li contributed to the experimental procedures. Tianxing Yan designed the graphical abstract and cover image. Tingbo Liang and Xueli Bai organized the clinical trial and obtained the clinical samples. Chengxiang Guo and Yiwen Chen assisted in coordinating the clinical trials. Qing Li, Zhiwei Zhou and Kenneth D. Westover contributed to proofreading the manuscript and reviewed the manuscript. Tingbo Liang, Xueli Bai, Jianpeng Sheng and Kenneth Westover supervised the project and secured funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDATA AVAILABILITY\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe multi-omics data associated with this study have been deposited in various repositories managed by the Genome Sequence Archive of Beijing Institute of Genomics, Chinese Academy of Sciences. The scRNA-seq can be accessed via HRA004695 (Data were controlled access due to patients\u0026rsquo; privacy. Reviewers can request data at https://ngdc.cncb.ac.cn/gsa-human/browse/HRA004695 and it will be released within three working days) and HRA007308 (https://ngdc.cncb.ac.cn/gsa-human/s/1hp5rUr4). Bulk transcriptomics data can be accessed via HRA007307 (https://ngdc.cncb.ac.cn/gsa-human/s/S1DGuE7C). Spatial transcriptomics data can be accessed via OMIX006713 (https://ngdc.cncb.ac.cn/omix/release/OMIX006713). The IMC data can be accessed via OMIX002774 (https://ngdc.cncb.ac.cn/omix/release/OMIX002774) and OMIX003997 (https://ngdc.cncb.ac.cn/omix/release/OMIX003997). The codes used in this analysis have been uploaded to Github at the following link: https://github.com/shaoweinuaa/GCSS and https://github.com/pyramidsnail/TACE. If further information is required for data reanalysis, please contact the lead researcher directly.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCODE AVAILABILITY\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePlease note that the single cell and spatial transcriptomics analysis in this study does not involve the original code. To access the original codes used, please refer to the methods section and respective references provided. The GCSS codes used in this analysis have been uploaded to Github at the following link: https://github.com/shaoweinuaa/GCSS. If you require any additional information to reanalyze the data presented in this paper, please contact the lead researcher directly.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors would like to thank Hangzhou Yingfei Biotechnology for IMC and mIHC help. This work was supported by the National Key Research and Development Program of China (grant 2019YFA0803000 to J.S.), the Excellent Youth Foundation of Zhejiang Scientific (grant R22H1610037 to J.S.), the National Natural Science Foundation of China (grant 82173078 to J.S.), the Natural Science Foundation of Zhejiang Province (grant 2022C03037 to J.S.), the National Natural Science Foundation of China (grant 81871925 to X.B and grant 82188102 to T.L.), the National Key Research and Development Program (grant 2019YFC1316000 to T.L.), the National Natural Science Foundation Basic Science Centre of China (Study of Tumor Material and Energy Dynamics, 8218810) and Welch Foundation I-1829 (to K.D.W)\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eChidambaranathan-Reghupaty S, Fisher PB, Sarkar D. Hepatocellular carcinoma (HCC): Epidemiology, etiology and molecular classification. Adv Cancer Res. 2021;149:1-61.\u003c/li\u003e\n\u003cli\u003eYang JD, Hainaut P, Gores GJ, Amadou A, Plymoth A, Roberts LR. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol. 2019;16(10):589-604.\u003c/li\u003e\n\u003cli\u003eFinn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim T-Y, et al. Atezolizumab plus Bevacizumab in Unresectable Hepatocellular Carcinoma. New England Journal of Medicine. 2020;382(20):1894-905.\u003c/li\u003e\n\u003cli\u003eLlovet JM, Castet F, Heikenwalder M, Maini MK, Mazzaferro V, Pinato DJ, et al. Immunotherapies for hepatocellular carcinoma. Nat Rev Clin Oncol. 2022;19(3):151-72.\u003c/li\u003e\n\u003cli\u003eJin H, Qin S, He J, Xiao J, Li Q, Mao Y, et al. New insights into checkpoint inhibitor immunotherapy and its combined therapies in hepatocellular carcinoma: from mechanisms to clinical trials. Int J Biol Sci. 2022;18(7):2775-94.\u003c/li\u003e\n\u003cli\u003eYap TA, Parkes EE, Peng W, Moyers JT, Curran MA, Tawbi HA. Development of Immunotherapy Combination Strategies in Cancer. Cancer Discov. 2021;11(6):1368-97.\u003c/li\u003e\n\u003cli\u003eRaoul JL, Forner A, Bolondi L, Cheung TT, Kloeckner R, de Baere T. Updated use of TACE for hepatocellular carcinoma treatment: How and when to use it based on clinical evidence. Cancer Treat Rev. 2019;72:28-36.\u003c/li\u003e\n\u003cli\u003eHan K, Kim JH. Transarterial chemoembolization in hepatocellular carcinoma treatment: Barcelona clinic liver cancer staging system. World J Gastroenterol. 2015;21(36):10327-35.\u003c/li\u003e\n\u003cli\u003eCamma C, Schepis F, Orlando A, Albanese M, Shahied L, Trevisani F, et al. Transarterial chemoembolization for unresectable hepatocellular carcinoma: meta-analysis of randomized controlled trials. Radiology. 2002;224(1):47-54.\u003c/li\u003e\n\u003cli\u003eAyaru L, Pereira SP, Alisa A, Pathan AA, Williams R, Davidson B, et al. Unmasking of alpha-fetoprotein-specific CD4(+) T cell responses in hepatocellular carcinoma patients undergoing embolization. J Immunol. 2007;178(3):1914-22.\u003c/li\u003e\n\u003cli\u003eZerbini A, Pilli M, Laccabue D, Pelosi G, Molinari A, Negri E, et al. Radiofrequency thermal ablation for hepatocellular carcinoma stimulates autologous NK-cell response. Gastroenterology. 2010;138(5):1931-42.\u003c/li\u003e\n\u003cli\u003eLlovet JM, De Baere T, Kulik L, Haber PK, Greten TF, Meyer T, et al. Locoregional therapies in the era of molecular and immune treatments for hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2021;18(5):293-313.\u003c/li\u003e\n\u003cli\u003eLencioni R, Kudo M, Erinjeri J, Qin S, Ren Z, Chan S, et al. EMERALD-1: A phase 3, randomized, placebo-controlled study of transarterial chemoembolization combined with durvalumab with or without bevacizumab in participants with unresectable hepatocellular carcinoma eligible for embolization. Journal of Clinical Oncology. 2024;42(3_suppl):LBA432-LBA.\u003c/li\u003e\n\u003cli\u003eLune PV, Aal AKA, Klimkowski S, Zarzour JG, Gunn AJ. Hepatocellular carcinoma: diagnosis, treatment algorithms, and imaging appearance after transarterial chemoembolization. Journal of Clinical and Translational Hepatology. 2018;6(2):175.\u003c/li\u003e\n\u003cli\u003eGuo C, Zhang J, Huang X, Chen Y, Sheng J, Huang X, et al. Preoperative sintilimab plus transarterial chemoembolization for hepatocellular carcinoma exceeding the Milan criteria: A phase II trial. Hepatol Commun. 2023;7(3):e0054.\u003c/li\u003e\n\u003cli\u003eSheng J, Zhang J, Wang L, Tano V, Tang J, Wang X, et al. Topological analysis of hepatocellular carcinoma tumour microenvironment based on imaging mass cytometry reveals cellular neighbourhood regulated reversely by macrophages with different ontogeny. Gut. 2022;71(6):1176-91.\u003c/li\u003e\n\u003cli\u003eEnzan H, Himeno H, Iwamura S, Saibara T, Onishi S, Yamamoto Y, et al. Immunohistochemical Identification of Ito Cells and Their Myofibroblastic Transformation in Adult Human Liver. Virchows Arch. 1994;424(3):249-56.\u003c/li\u003e\n\u003cli\u003eSch\u0026uuml;rch CM, Bhate SS, Barlow GL, Phillips DJ, Noti L, Zlobec I, et al. Coordinated cellular neighborhoods orchestrate antitumoral immunity at the colorectal cancer invasive front. Cell. 2020;182(5):1341-59. e19.\u003c/li\u003e\n\u003cli\u003eSchapiro D, Jackson HW, Raghuraman S, Fischer JR, Zanotelli VRT, Schulz D, et al. histoCAT: analysis of cell phenotypes and interactions in multiplex image cytometry data. Nat Methods. 2017;14(9):873-6.\u003c/li\u003e\n\u003cli\u003eSchurch CM, Bhate SS, Barlow GL, Phillips DJ, Noti L, Zlobec I, et al. Coordinated Cellular Neighborhoods Orchestrate Antitumoral Immunity at the Colorectal Cancer Invasive Front. Cell. 2020;183(3):838.\u003c/li\u003e\n\u003cli\u003eHickey JW, Becker WR, Nevins SA, Horning A, Perez AE, Zhu C, et al. Organization of the human intestine at single-cell resolution. Nature. 2023;619(7970):572-84.\u003c/li\u003e\n\u003cli\u003eRisom T, Glass DR, Averbukh I, Liu CC, Baranski A, Kagel A, et al. Transition to invasive breast cancer is associated with progressive changes in the structure and composition of tumor stroma. Cell. 2022;185(2):299-310 e18.\u003c/li\u003e\n\u003cli\u003eNightingale T, Cutler D. The secretion of von Willebrand factor from endothelial cells; an increasingly complicated story. J Thromb Haemost. 2013;11 Suppl 1(Suppl 1):192-201.\u003c/li\u003e\n\u003cli\u003eShinde AV, Humeres C, Frangogiannis NG. The role of alpha-smooth muscle actin in fibroblast-mediated matrix contraction and remodeling. Biochim Biophys Acta Mol Basis Dis. 2017;1863(1):298-309.\u003c/li\u003e\n\u003cli\u003eCrespo AC, Mulik S, Dotiwala F, Ansara JA, Sen Santara S, Ingersoll K, et al. Decidual NK Cells Transfer Granulysin to Selectively Kill Bacteria in Trophoblasts. Cell. 2020;182(5):1125-39 e18.\u003c/li\u003e\n\u003cli\u003eXing W, Austen KF, Gurish MF, Jones TG. Protease phenotype of constitutive connective tissue and of induced mucosal mast cells in mice is regulated by the tissue. Proc Natl Acad Sci U S A. 2011;108(34):14210-5.\u003c/li\u003e\n\u003cli\u003eDe Rosa SC, Herzenberg LA, Herzenberg LA, Roederer M. 11-color, 13-parameter flow cytometry: identification of human naive T cells by phenotype, function, and T-cell receptor diversity. Nat Med. 2001;7(2):245-8.\u003c/li\u003e\n\u003cli\u003ePurves-Tyson TD, Robinson K, Brown AM, Boerrigter D, Cai HQ, Weissleder C, et al. Increased Macrophages and C1qA, C3, C4 Transcripts in the Midbrain of People With Schizophrenia. Front Immunol. 2020;11:2002.\u003c/li\u003e\n\u003cli\u003eCable 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.\u003c/li\u003e\n\u003cli\u003eBlas-Garcia A, Apostolova N. Novel Therapeutic Approaches to Liver Fibrosis Based on Targeting Oxidative Stress. Antioxidants (Basel). 2023;12(8).\u003c/li\u003e\n\u003cli\u003eAblasser A, Chen ZJ. cGAS in action: Expanding roles in immunity and inflammation. Science. 2019;363(6431):eaat8657.\u003c/li\u003e\n\u003cli\u003eAibar S, Gonzalez-Blas CB, Moerman T, Huynh-Thu VA, Imrichova H, Hulselmans G, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods. 2017;14(11):1083-6.\u003c/li\u003e\n\u003cli\u003eKwon J, Bakhoum SF. The Cytosolic DNA-Sensing cGAS-STING Pathway in Cancer. Cancer Discov. 2020;10(1):26-39.\u003c/li\u003e\n\u003cli\u003eLee JJ, Kim SY, Kim SH, Choi S, Lee B, Shin JS. STING mediates nuclear PD-L1 targeting-induced senescence in cancer cells. Cell Death Dis. 2022;13(9):791.\u003c/li\u003e\n\u003cli\u003eDu SS, Chen GW, Yang P, Chen YX, Hu Y, Zhao QQ, et al. Radiation Therapy Promotes Hepatocellular Carcinoma Immune Cloaking via PD-L1 Upregulation Induced by cGAS-STING Activation. Int J Radiat Oncol Biol Phys. 2022;112(5):1243-55.\u003c/li\u003e\n\u003cli\u003eLe Naour J, Zitvogel L, Galluzzi L, Vacchelli E, Kroemer G. Trial watch: STING agonists in cancer therapy. Oncoimmunology. 2020;9(1):1777624.\u003c/li\u003e\n\u003cli\u003eGreten TF, Villanueva A, Korangy F, Ruf B, Yarchoan M, Ma L, et al. Biomarkers for immunotherapy of hepatocellular carcinoma. Nature Reviews Clinical Oncology. 2023;20(11):780-98.\u003c/li\u003e\n\u003cli\u003eBrowaeys R, Saelens W, Saeys Y. NicheNet: modeling intercellular communication by linking ligands to target genes. Nat Methods. 2020;17(2):159-62.\u003c/li\u003e\n\u003cli\u003eChan O, Burke JD, Gao DF, Fish EN. The chemokine CCL5 regulates glucose uptake and AMP kinase signaling in activated T cells to facilitate chemotaxis. J Biol Chem. 2012;287(35):29406-16.\u003c/li\u003e\n\u003cli\u003eSchober A, Bernhagen J, Weber C. Chemokine-like functions of MIF in atherosclerosis. J Mol Med (Berl). 2008;86(7):761-70.\u003c/li\u003e\n\u003cli\u003eAbdulrahman Z, Santegoets SJ, Sturm G, Charoentong P, Ijsselsteijn ME, Somarakis A, et al. Tumor-specific T cells support chemokine-driven spatial organization of intratumoral immune microaggregates needed for long survival. J Immunother Cancer. 2022;10(2).\u003c/li\u003e\n\u003cli\u003eJang MK, Jung MH. ATF3 represses PPARgamma expression and inhibits adipocyte differentiation. Biochem Biophys Res Commun. 2014;454(1):58-64.\u003c/li\u003e\n\u003cli\u003eSommars MA, Ramachandran K, Senagolage MD, Futtner CR, Germain DM, Allred AL, et al. Dynamic repression by BCL6 controls the genome-wide liver response to fasting and steatosis. Elife. 2019;8:e43922.\u003c/li\u003e\n\u003cli\u003eStewart MD, Wong J. Nuclear receptor repression: regulatory mechanisms and physiological implications. Prog Mol Biol Transl Sci. 2009;87:235-59.\u003c/li\u003e\n\u003cli\u003eDu Z, Liu M, Wang Z, Lin Z, Feng Y, Tian D, et al. EZH2-mediated inhibition of KLF14 expression promotes HSCs activation and liver fibrosis by downregulating PPARgamma. Cell Prolif. 2021;54(7):e13072.\u003c/li\u003e\n\u003cli\u003eChanda D, Xie YB, Choi HS. Transcriptional corepressor SHP recruits SIRT1 histone deacetylase to inhibit LRH-1 transactivation. Nucleic Acids Res. 2010;38(14):4607-19.\u003c/li\u003e\n\u003cli\u003ePicard F, Kurtev M, Chung N, Topark-Ngarm A, Senawong T, Machado De Oliveira R, et al. Sirt1 promotes fat mobilization in white adipocytes by repressing PPAR-gamma. Nature. 2004;429(6993):771-6.\u003c/li\u003e\n\u003cli\u003eGuo C, Li Y, Gow CH, Wong M, Zha J, Yan C, et al. The optimal corepressor function of nuclear receptor corepressor (NCoR) for peroxisome proliferator-activated receptor gamma requires G protein pathway suppressor 2. J Biol Chem. 2015;290(6):3666-79.\u003c/li\u003e\n\u003cli\u003eViswakarma N, Jia Y, Bai L, Vluggens A, Borensztajn J, Xu J, et al. Coactivators in PPAR-Regulated Gene Expression. PPAR Res. 2010;2010.\u003c/li\u003e\n\u003cli\u003eShi H, Luo J, Zhu J, Li J, Sun Y, Lin X, et al. PPAR gamma Regulates Genes Involved in Triacylglycerol Synthesis and Secretion in Mammary Gland Epithelial Cells of Dairy Goats. PPAR Res. 2013;2013:310948.\u003c/li\u003e\n\u003cli\u003eZhang X, Yao J, Shi H, Gao B, Zhang L. LncRNA TINCR/microRNA-107/CD36 regulates cell proliferation and apoptosis in colorectal cancer via PPAR signaling pathway based on bioinformatics analysis. Biol Chem. 2019;400(5):663-75.\u003c/li\u003e\n\u003cli\u003eChehaibi K, Cedo L, Metso J, Palomer X, Santos D, Quesada H, et al. PPAR-beta/delta activation promotes phospholipid transfer protein expression. Biochem Pharmacol. 2015;94(2):101-8.\u003c/li\u003e\n\u003cli\u003eYang L, Zheng Y, Miao YM, Yan WX, Geng YZ, Dai Y, et al. Bergenin, a PPARgamma agonist, inhibits Th17 differentiation and subsequent neutrophilic asthma by preventing GLS1-dependent glutaminolysis. Acta Pharmacol Sin. 2022;43(4):963-76.\u003c/li\u003e\n\u003cli\u003eBionaz M, Thering BJ, Loor JJ. Fine metabolic regulation in ruminants via nutrient-gene interactions: saturated long-chain fatty acids increase expression of genes involved in lipid metabolism and immune response partly through PPAR-alpha activation. Br J Nutr. 2012;107(2):179-91.\u003c/li\u003e\n\u003cli\u003eNevzorova YA, Hu W, Cubero FJ, Haas U, Freimuth J, Tacke F, et al. Overexpression of c-myc in hepatocytes promotes activation of hepatic stellate cells and facilitates the onset of liver fibrosis. Biochim Biophys Acta. 2013;1832(10):1765-75.\u003c/li\u003e\n\u003cli\u003eFoglia B, Cannito S, Bocca C, Parola M, Novo E. ERK Pathway in Activated, Myofibroblast-Like, Hepatic Stellate Cells: A Critical Signaling Crossroad Sustaining Liver Fibrosis. Int J Mol Sci. 2019;20(11).\u003c/li\u003e\n\u003cli\u003eWesterouen Van Meeteren MJ, Drenth JPH, Tjwa E. Elafibranor: a potential drug for the treatment of nonalcoholic steatohepatitis (NASH). Expert Opin Investig Drugs. 2020;29(2):117-23.\u003c/li\u003e\n\u003cli\u003eReli\u0026oacute;n JDA, Kessler D, Levina E, Taylor SF. Network classification with applications to brain connectomics. The annals of applied statistics. 2019;13(3):1648.\u003c/li\u003e\n\u003cli\u003eMasatoshi K, Kazuomi U, Masafumi I, Takuji T, Nobukazu T, Hiroshi A, et al. Randomised, multicentre prospective trial of transarterial chemoembolisation (TACE) plus sorafenib as compared with TACE alone in patients with hepatocellular carcinoma: TACTICS trial. Gut. 2020;69(8):1492.\u003c/li\u003e\n\u003cli\u003eDharmapuri S, \u0026Ouml;zbek U, Lin JY, Sung M, Schwartz M, Branch AD, et al. Predictive value of neutrophil to lymphocyte ratio and platelet to lymphocyte ratio in advanced hepatocellular carcinoma patients treated with anti-PD-1 therapy. Cancer Med. 2020;9(14):4962-70.\u003c/li\u003e\n\u003cli\u003eLiu Y, Xun Z, Ma K, Liang S, Li X, Zhou S, et al. Identification of a tumour immune barrier in the HCC microenvironment that determines the efficacy of immunotherapy. J Hepatol. 2023;78(4):770-82.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4536926/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4536926/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Hepatocellular carcinoma (HCC) management is challenging due to its heterogeneous tumor microenvironment and poor treatment responses. To develop personalized approaches, we conducted a clinical trial of transarterial chemoembolization (TACE) combined with immune checkpoint blockade (ICB) where we collected tumor specimens for comprehensive multi-omic profiling. Treatment response was associated with increased infiltration of anti-tumor T cells, driven by cGAS-STING activation within immune-suppressive epithelial cells. However, fibrotic processes impaired these responses in some patients. Based on this insight, we conducted an animal trial in an HCC model where we added an anti-fibrotic agent to ICB and TACE mimic. This improved efficacy compared to ICB and TACE mimic alone. To identify patients who would benefit from such treatment, we constructed a predictive model using data from a limited immunohistochemistry panel and a group sparse learning algorithm. Our findings provide a blueprint for crafting personalized therapies for HCC and other cancers.","manuscriptTitle":"Fibrosis Impairs cGAS-Mediated Responses to Immunotherapy in Advanced HCC","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-28 16:20:43","doi":"10.21203/rs.3.rs-4536926/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"80a6d6fe-8dce-44ae-90d8-f86298a5d823","owner":[],"postedDate":"June 28th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":33771402,"name":"Health sciences/Oncology/Cancer/Cancer therapy/Cancer immunotherapy"},{"id":33771403,"name":"Health sciences/Oncology/Cancer/Cancer microenvironment"}],"tags":[],"updatedAt":"2024-08-14T08:40:38+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-28 16:20:43","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4536926","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4536926","identity":"rs-4536926","version":["v1"]},"buildId":"WrCJVZZCHTDjtuVLN7oU0","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}