{"paper_id":"3eba008d-b7a7-4737-9a09-fd45cb9bd7f2","body_text":"Characteristics of Mechanically Stimulated Genes in Hepatocellular Carcinoma and Their Role as Prognostic Markers | 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 Research Article Characteristics of Mechanically Stimulated Genes in Hepatocellular Carcinoma and Their Role as Prognostic Markers peng cheng, xueyi feng, xianlu huang, JUNHONG JIN This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9047659/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 17 You are reading this latest preprint version Abstract Background: Hepatocellular carcinoma (HCC) is a highly prevalent malignant tumor with poor prognosis and high heterogeneity. Mechanical stimuli in the tumor microenvironment (TME) regulate HCC cell biological behavior via mechanosensitive-related genes (MSRGs), but their specific prognostic value and underlying mechanisms remain unclear. Methods: RNA-seq and clinical data of HCC patients were retrieved from The Cancer Genome Atlas (TCGA) database (accession: TCGA-LIHC, https://portal.gdc.cancer.gov/), and single-cell RNA-seq datasets (accession: GSE149614) were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). Univariate Cox regression analysis was used to screen MSRGs associated with overall survival (OS). Consensus clustering was performed to identify HCC molecular subtypes based on OS-related MSRGs. LASSO-Cox regression analysis was applied to construct an MSRG-based prognostic risk model, which was validated by Kaplan-Meier survival analysis and time-dependent ROC curves. Functional enrichment analysis, immune microenvironment characterization, genomic variation analysis and drug sensitivity prediction were conducted to explore the biological significance of the risk model. Single-cell RNA-seq analysis was used to map the risk signature to specific cell types, and SHAP analysis combined with in vitro experiments was performed to validate the key driver gene. Results: A total of 39 OS-related MSRGs were identified in HCC, and HCC samples were stratified into two molecular subtypes with distinct prognostic and immune microenvironment characteristics. A six-gene prognostic risk model (HPN, ENDOG, UCN, FYN, ETV1, KCNQ3) was constructed, which exhibited good prognostic discrimination (1-, 3-, 5-year AUC: 0.74, 0.74, 0.73). High-risk patients had shorter OS, a more immunosuppressive TME, and distinct genomic alteration patterns compared with low-risk patients. The two risk groups showed differential sensitivity to clinical targeted drugs (Axitinib, Erlotinib, Sorafenib, Sunitinib). Single-cell analysis revealed cell-type specificity of the risk signature, and KCNQ3 was identified as the key driver gene via SHAP analysis. In vitro experiments confirmed that KCNQ3 Knockdown significantly inhibited the proliferation and clonogenic ability of HCC Huh7 cells. Conclusion: MSRGs are closely associated with the prognosis and immune microenvironment of HCC. The constructed MSRG-based prognostic risk model has reliable predictive value for HCC patient survival, and KCNQ3 may serve as a potential prognostic biomarker and therapeutic target for HCC, providing new insights for personalized treatment of HCC. Hepatocellular carcinoma Mechanically stimulated genes Tumor microenvironment Prognostic marker KCNQ3 Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Research Background Hepatocellular carcinoma (HCC) ranks among the most prevalent malignant tumors globally, with persistently high incidence and mortality rates [1].According to World Health Organization (WHO) statistics, HCC ranks third among cancer-related deaths globally, following lung cancer and colorectal cancer [2]. Despite recent advances in HCC diagnosis and treatment—such as improved imaging techniques and the introduction of targeted and immunotherapies—patient prognosis remains poor, with a five-year survival rate below 20% [3]. The development and progression of liver cancer are closely associated with multiple factors, primarily including chronic hepatitis (such as hepatitis B and C), cirrhosis, metabolic syndrome (e.g., non-alcoholic fatty liver disease), and chronic alcohol consumption. These factors induce chronic inflammation and fibrosis in the liver, thereby promoting hepatocellular carcinogenesis [4,5].Furthermore, HCC exhibits high heterogeneity, with significant variations in gene expression, cellular composition, and microenvironmental characteristics among tumors from different patients, adding complexity and challenges to treatment. In recent years, the role of the tumor microenvironment (TME) in tumorigenesis, progression, and metastasis has garnered increasing attention [6]. The TME encompasses not only tumor cells themselves but also diverse non-tumor cells (e.g., immune cells, fibroblasts, endothelial cells) and the extracellular matrix (ECM). These components influence tumor biological behavior through complex interactions [7].Notably, mechanical stimuli, as a key factor within the TME, can influence tumor cell proliferation, migration, and invasion through ECM stiffness, cell-cell contacts, and cell-matrix interactions [8]. Studies indicate that liver cancer tissue exhibits significantly higher stiffness than normal liver tissue, primarily attributed to excessive deposition and cross-linking of the extracellular matrix [9]. Mechanical stimuli can modulate cellular biology through multiple mechanoreceptors (e.g., integrins, cadherins, GPCRs, Piezo channels) and downstream signaling pathways (e.g., YAP/TAZ pathway) [10-12].For instance, Piezo1 channels sense mechanical stimuli, enhancing HCC angiogenesis through calcium influx and downstream signaling pathways [13]. Furthermore, mechanical stress influences immune cell infiltration within the tumor microenvironment; highly rigid tumor tissues typically exhibit immunosuppressive features such as reduced T-cell infiltration and elevated expression of immune checkpoint molecules [14]. However, despite the emerging understanding of mechanical stimulation's role in the tumor microenvironment, its specific mechanisms in HCC remain unclear. Particularly, how Mechanical Stimulus-Related Genes (MSRGs) influence HCC initiation, progression, and prognosis has not been systematically investigated.Therefore, this study aims to identify MSRGs through integrated bulk RNA-Seq and single-cell RNA-Seq analyses and construct prognostic biomarkers based on these genes, thereby providing new insights for personalized treatment of HCC. Results Genetic Landscape of OS-Related MSRGs in Hepatocellular Carcinoma Through univariate Cox regression analysis, we identified 39 mechanosensitive-related genes (MSRGs) significantly associated with the overall survival (OS) of hepatocellular carcinoma (HCC) patients (Figure 1A). These included key genes such as TP53, CTNNB1, TTN, and MUC16 , most of which exhibited hazard ratios (HRs) greater than 1, suggesting that their high expression may be closely linked to poor patient prognosis . Among 371 HCC samples, 315 (84.91%) harbored at least one somatic mutation in the 39 OS-related MSRGs (Figure 1B). TP53 was the most frequently mutated gene . Further variant classification revealed that missense mutations were the dominant type (Figure 1C). A Circos plot clearly illustrated the chromosomal distribution of these 39 genes (Figure 1D). Gene expression correlation heatmap analysis uncovered extensive co-expression patterns or potential regulatory relationships among these genes (Figure 1E). Copy number variation (CNV) analysis demonstrated widespread copy number amplifications (gains) and deletions of these OS-related MSRGs in HCC samples (Figure 1F), indicating that genomic instability may influence their expression and thereby contribute to HCC progression . Identification of Molecular Subtypes Based on MSRG Expression Profiles Using consensus clustering analysis, we stratified HCC samples into two distinct molecular subtypes with significant heterogeneity based on the expression profiles of the 39 OS-related MSRGs. When the cluster number K=2, clustering stability was optimal, ultimately dividing all samples into two subgroups: Cluster 1 (n=199) and Cluster 2 (n=171) (Figure 2A). Principal component analysis (PCA) revealed clear separation between the two subtypes in a two-dimensional space formed by Dim1 (contribution 21.6%) and Dim2 (contribution 5.9%) (Figure 2B), indicating fundamental differences in their transcriptomic features. Survival analysis showed that Cluster 2 patients had significantly better overall survival than Cluster 1 (Log-rank test, P = 0.0012) (Figure 2C), suggesting that MSRG-based clustering has important prognostic value. To further elucidate the underlying mechanisms, we quantified immune cell infiltration patterns in the tumor microenvironment (TME) using the CIBERSORT algorithm. Results indicated that Cluster 2 exhibited an immune-\"hot\" phenotype, characterized by significant enrichment of effector immune cells such as CD8⁺ T cells (Figure 2D). Specifically, the TME of Cluster Construction and Validation of an MSRG-Based Prognostic Risk Model To build a robust prognostic model, we performed LASSO-Cox regression analysis on the previously identified 39 OS-related MSRGs to mitigate overfitting and identify key genes. The coefficient profile plot demonstrated a gradual shrinkage of regression coefficients toward zero as the regularization parameter λ increased (Figure 3A). The optimal λ value (lambda.min) was determined via cross-validation to be -3.288, achieving a minimal deviation while maintaining model parsimony (Figure 3B). Ultimately, the model identified several core genes, including HPN, ENDOG, UCN, FYN, ETV1, and KCNQ3 , which were used to calculate an individual risk score for each patient. Patients were stratified into high-risk (n=182) and low-risk (n=183) groups based on the median risk score.The survival time distribution plot revealed that patients in the high-risk group generally had shorter survival times, with a higher concentration of death events occurring in this group. In contrast, low-risk group patients exhibited longer survival times and were predominantly in a state of survival (Figure 3C). An integrated heatmap further indicated that the risk score was significantly correlated with multi-dimensional features, including tumor pathological stage and gene expression profiles (Figure 3D). Kaplan-Meier survival analysis confirmed that patients in the high-risk group had a significantly shorter overall survival than those in the low-risk group (Log-rank test, p < 0.0001) (Figure 3E), demonstrating the significant prognostic discriminatory power of the risk score.The model's predictive accuracy was evaluated by plotting ROC curves and calculating the area under the curve (AUC) for 1-, 3-, and 5-year overall survival. The results showed stable and good discriminatory ability at all three key time points, with AUC values of 0.74, 0.74, and 0.73, respectively (Figure 3F). This indicates that the model provides reliable predictive efficacy for both short-term and long-term survival outcomes in liver cancer patients.To clarify the clinical significance of the risk score, its relationship with tumor pathological stage (Stage I-IV) was analyzed. Violin plots showed significant differences in the distribution of risk scores among patients at different stages (Figures 2G-J). Specifically, patients with advanced-stage disease (Stage IV) had significantly lower risk scores compared to those with early-stage (Stage I, p < 0.0001) and intermediate-stage (Stage II, p < 0.01) disease. This result suggests that the MSRG-based risk score is closely associated with the progression of liver cancer and may serve as a potential indicator reflecting the degree of tumor malignancy. Functional Enrichment Analysis of Differentially Expressed Genes between High- and Low-Risk Groups Functional annotation of the differentially expressed genes (DEGs) between the high- and low-risk groups revealed key biological characteristics associated with the risk score. Gene Ontology (GO) enrichment analysis (Figure 4A, top) showed that the DEGs were significantly enriched in biological processes (BP) closely related to cell proliferation, such as \"organelle fission,\" \"nuclear division,\" and \"chromosome segregation.\" Regarding cellular components (CC) and molecular functions (MF), the DEGs were primarily enriched in terms related to structural functions like \"spindle\" and \"microtubule binding.\" Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis (Figure 4B) further indicated significant enrichment of DEGs in pathways such as \"Cell cycle,\" \"ECM-receptor interaction,\" and \"Protein digestion and absorption.\" These findings collectively suggest that high-risk tumors may possess stronger capabilities for cell proliferation and extracellular matrix remodeling.To quantify the activity changes in key biological pathways, we performed Gene Set Variation Analysis (GSVA) . A bar plot (Figure 4C) visually represents the gene sets with significantly differential activity between the risk groups. Pathways related to proliferation and survival, such as the G2M checkpoint, E2F targets, and PI3K-AKT-mTOR signaling , were significantly up-regulated in the high-risk group. In contrast, metabolism-related pathways like fatty acid metabolism and estrogen response were down-regulated. A correlation heatmap (Figure 4D) further elucidated the co-variation patterns among these differentially active gene sets and confirmed a clear association between their activity and the risk groups, with the high-risk group broadly linked to the activation of proliferation-related pathways. To confirm the clinical relevance of these pathway activities, patients were grouped based on the median activity of specific pathways, and Kaplan-Meier survival curves were plotted. The results showed that patients with high activity in the E2F targets and PI3K-AKT-mTOR signaling pathways had a significantly shorter overall survival (Figures E-F). Conversely, high activity in the fatty acid metabolism and late estrogen response pathways was significantly associated with longer survival (Figures G-H). This provides prognostic validation, from the opposite angle, for the GSVA results, indicating that these pathways are key biological processes driving the prognostic differences between the high- and low-risk groups. Comparison of Immune Microenvironment Characteristics Between High- and Low-Risk Groups Through quantitative analysis of immune cell infiltration levels between the high- and low-risk groups, we identified significant differences in the cellular composition of the tumor immune microenvironment (TME). Specifically, the low-risk group exhibited significantly higher infiltration proportions of activated CD4⁺ T cells, gamma delta T cells ( γδ T cells), and myeloid-derived suppressor cells (MDSCs) compared to the high-risk group. In contrast, the high-risk group was characterized by the enrichment of M0 macrophages and resting dendritic cells . This disparity suggests that the risk score based on MSRGs effectively reflects the heterogeneity of the immune status in hepatocellular carcinoma (HCC). Further assessment of the overall tumor microenvironment using the ESTIMATE algorithm revealed that the low-risk group had significantly higher ESTIMATE scores, Immune scores, and Stromal scores than the high-risk group, while its tumor purity was significantly lower . These results corroborate the immune cell infiltration analysis, indicating that the low-risk group possesses a richer infiltration of immune cells and stromal components, presenting an \"immune-activated\" microenvironment phenotype. To explore the potential regulatory role of the core genes in the risk model on the immune microenvironment, we analyzed the correlation between the expression of key MSRGs and immune cell infiltration. The results showed that the expression of KCNQ3 and ABHD12 was significantly positively correlated with M0 macrophage infiltration , and negatively correlated with immune cells such as resting mast cells . This suggests that these genes may influence the overall immune microenvironment state by regulating the recruitment or function of specific immune cells.In the analysis of immunotherapy targets, the expression of multiple inhibitory immune checkpoint molecules was significantly upregulated in the high-risk group, including CD276 (B7-H3), HAVCR1 (TIM-3), IDO2, TNFRSF18 (GITR), TNFSF18 (GITRL), and VTCN1 (B7-H4) . Concurrently, the expression of a series of chemokines such as CCL20, CCL28, CCL4, and XCL1 , which are closely associated with lymphocyte recruitment and inflammatory responses, was significantly upregulated in the low-risk group . This differential expression pattern indicates that the high-risk group may form a stronger immunosuppressive microenvironment, whereas the low-risk group demonstrates a greater capacity for immune cell recruitment. Genomic Variation Characteristics and Drug Sensitivity Analysis To investigate the molecular basis of the risk stratification, we analyzed the characteristics of somatic gene alterations in the high- and low-risk groups. Overall, the frequency of genetic alterations was higher in the high-risk group , with alterations occurring in 144 samples (81.36%); alterations were found in 135 samples (75.42%) in the low-risk group. The alteration spectrum in both groups was predominantly characterized by missense mutations . In the high-risk group, alterations in genes such as TP53 (highest mutation rate, >40%), TTN, and CTNNB1 were particularly frequent. The proportion of TP53 alterations was relatively lower in the low-risk group, while alterations in genes like ALB and PCLO were relatively more prominent . These differences suggest that different risk groups are associated with distinct patterns of driver gene alterations. To evaluate the potential value of the risk score model in guiding personalized therapy, we predicted the sensitivity (half-maximal inhibitory concentration, IC₅₀ values) of the high- and low-risk groups to four clinically relevant targeted drugs based on genomic data. The results showed:The predicted IC₅₀ values for Axitinib and Erlotinib were significantly lower in the high-risk group compared to the low-risk group (p < 0.0001), suggesting that high-risk tumors might be more sensitive to these two drugs.Conversely, Sorafenib and Sunitinib showed the opposite trend; the predicted IC₅₀ values were significantly higher in the high-risk group than in the low-risk group (p < 0.0001), indicating that low-risk tumors might be more sensitive to them .These findings suggest that this risk model could potentially guide the selection of targeted drugs for patients with hepatocellular carcinoma. Construction of Single-Cell Atlas and Mapping of Risk Signatures Through single-cell RNA sequencing (scRNA-seq) analysis , we constructed a cellular atlas of hepatocellular carcinoma (HCC) tissues, identifying six major cell types: T/NK cells, myeloid cells, plasma cells, fibroblasts, epithelial cells, and B cells (Figures 7A-C). UMAP visualization revealed that these cell populations formed distinct clusters in the reduced-dimensional space. We further mapped the MSRG-based risk score onto this atlas and found that high-risk scores were not uniformly distributed but were significantly enriched in specific cell types or subpopulations (Figures 7D-F), indicating that the risk signature exhibits cell-type specificity . To explore how different cell types contribute to the biological differences between risk groups, we performed differential expression analysis and pathway enrichment separately for epithelial cells and T/NK cells. In epithelial cells from the high-risk group, pathways such as \"Complement and coagulation cascades,\" \"Fatty acid degradation,\" and \"PPAR signaling pathway\" were significantly enriched (Figure 7H), suggesting these cells may be in a state of metabolic stress or remodeling. In T/NK cells from the high-risk group, pathways including \"Natural killer cell-mediated cytotoxicity,\" \"Th1/Th2/Th17 cell differentiation,\" and \"Antigen processing and presentation\" were significantly enriched (Figure 7G). Notably, these pathways involved key immune signaling genes such as FYN, PIK3R1, MAPK1, and CD247 , indicating substantial alterations in immune recognition and activation-related signaling in T/NK cells from the high-risk group. Through cell-cell interaction analysis , we systematically compared the communication networks between the high- and low-risk groups. Network diagrams revealed overall differences in the interaction patterns between the two groups (Figure 7J). A differential interaction heatmap further quantified these changes: compared to the low-risk group, the number of outgoing interactions from fibroblasts to other cells (e.g., epithelial cells, myeloid cells) was generally reduced in the high-risk group (Figure 7K). However, the strength of certain interactions, such as those between fibroblasts and endothelial cells, was increased (Figure 7J). Comparison of key signaling pathways showed that in the high-risk group, interactions related to immune regulation, inflammatory response, and angiogenesis—such as those mediated by MIF-(CD74+CXCR4), MIF-(CD74+CD44), VEGFA-VEGFR2, and SPP1-CD44 —were significantly enhanced (Figure 7L). Conversely, signaling pathways involved in extracellular matrix adhesion and epithelial integrity were generally weakened in the high-risk group. These included integrin-mediated signals like VTN-(ITGAV+ITGB1) and SPP1-(ITGAV+ITGB1) , as well as homotypic interactions such as OCLN-OCLN , which help maintain cell junctions (Figure 7L). Identification and Validation of Key Driver Genes in the Risk Model Based on SHAP Analysis To elucidate the decision-making mechanism of the constructed risk score model and identify key features, we employed the SHAP (SHapley Additive exPlanations) method for interpretability analysis. As shown in Figure 8A, the SHAP feature importance plot displays the mean absolute SHAP values for each gene. The wide distribution of these values indicates significant differences in the impact of different genes on the model's output. Genes such as FYN, KCNQ3, ENDOG, and ABHD12 exhibited large absolute SHAP values with broad distributions, and their feature values (expression levels) showed a significant positive correlation with the SHAP values. We further focused on KCNQ3 , a gene that showed particularly significant contribution in the SHAP analysis, for in-depth validation. The SHAP dependence plot (Figure 8B) clearly demonstrates a significant positive correlation between the expression level of KCNQ3 and its SHAP value. As KCNQ3 expression increases, its positive contribution to raising the risk score continuously increases, indicating that KCNQ3 is a stable and important negative predictor in the risk model.To validate its clinical significance, we stratified patients in the TCGA-LIHC cohort into high and low expression groups based on the median expression of KCNQ3 and performed survival analysis. The Kaplan-Meier curve (Figure 8C) revealed that patients with high KCNQ3 expression had a significantly shorter overall survival than those with low expression (p = 0.0035). This finding, from the perspective of the TCGA-LIHC cohort , confirms that high expression of KCNQ3 is an independent risk factor for poor prognosis in hepatocellular carcinoma patients , establishing it as a reliable biomarker.To investigate the role of KCNQ3 in HCC progression at the functional level, we conducted in vitro loss-of-function experiments in hepatocellular carcinoma cell lines. The CCK-8 cell proliferation assay (Figure 8D) results showed that knocking down KCNQ3 (shKCNQ3) significantly inhibited the proliferative capacity of HCC cells compared to the control group (shCtrl) (p < 0.001). The plate colony formation assay further confirmed that the number of cell colonies formed in the shKCNQ3 group was significantly lower than in the shCtrl group (p < 0.01) ( Figures 8E-F ). These results collectively demonstrate that KCNQ3 significantly promotes the proliferation and clonogenic ability of hepatocellular carcinoma cells*, exhibiting oncogene-like functions in the in vitro experiments. Methods Study Design and Data Collection This study integrated multi-omics data with experimental validation to systematically investigate prognostic models of hepatocellular carcinoma (HCC) and their underlying biological mechanisms. RNA sequencing data and clinical information from 375 HCC patients were obtained from The Cancer Genome Atlas (TCGA) database (TCGA-LIHC, https://portal.gdc.cancer.gov/). Single-cell RNA sequencing datasets (GSE149614)】 were downloaded from 【the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) for in-depth analysis of the tumor microenvironment. All statistical analyses were performed using R software (version 4.4.0), with false discovery rate (FDR) correction applied using the Benjamini-Hochberg method . Analysis Pipeline Identification of Molecular Subtypes : Consensus clustering was performed using the R package ConsensusClusterPlus . The K-means algorithm was applied with 1000 bootstrap iterations and a sampling ratio of 80%. The optimal number of clusters (k-value) was determined based on cumulative distribution functions and consensus matrix heatmaps, leading to the stratification of patients into distinct molecular subtypes . Single-Cell Transcriptomic Analysis : Quality control of single-cell data was conducted using the Seurat package (v5.0.3) . Cells with fewer than 200 genes or mitochondrial gene content exceeding 10% were filtered out. A total of 3000 highly variable genes were identified using the FindVariableFeatures function. Principal component analysis (PCA) was performed, and batch effects were corrected using the Harmony method. Cell clustering was accomplished via the FindNeighbors and FindClusters functions . Prognostic Model Construction and Validation : In the TCGA-LIHC cohort, genes significantly associated with overall survival (OS) were first screened using univariate Cox regression . LASSO regression was then employed for variable selection and shrinkage to construct a multi-gene risk score model. Patients were divided into high-risk and low-risk groups based on the median risk score. The prognostic performance of the model was evaluated using Kaplan-Meier survival analysis and time-dependent receiver operating characteristic (ROC) curves . Functional Enrichment and Tumor Microenvironment Characterization : Functional annotation of differentially expressed genes (DEGs) was performed through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses using the ClusterProfiler package (v4.8.3) . The infiltration abundance of 22 immune cell subtypes and immune/stromal scores were assessed via the CIBERSORT and ESTIMATE algorithms. Correlations between risk scores and immune features were examined using Spearman’s rank correlation . Differences in cell-cell communication networks between risk groups were analyzed with the CellChat package (v2.28.0) . Genomic Variation and Drug Sensitivity Prediction : Somatic mutation profiles were analyzed with the maftools package (v2.18.0) . Based on the GDSC2 database, the sensitivity of HCC patients to targeted therapies (half-maximal inhibitory concentration, IC₅₀) was predicted using the oncoPredict package , and differences between risk groups were compared . Cellular Experiments and Functional Validation Cell Culture : Human HCC Huh7 cell lines were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Cells were maintained at 37°C in a 5% CO₂ incubator and regularly passaged to maintain logarithmic growth. Loss-of-Function Experiments : Stable knockdown cell lines were generated via lentiviral transduction . Cell proliferation viability was assessed using the CCK-8 assay : cells were seeded at 5×10³ cells/well in 96-well plates. After 24, 48, and 96 hours of culture, CCK-8 reagent was added, and absorbance was measured at 450 nm. Colony Formation Assay : Cells were plated at 1×10³ cells/well in 6-well plates and cultured for 14 days. Colonies were fixed with 4% paraformaldehyde, stained with 0.5% crystal violet, and counted (considering colonies with diameters >0.5 mm as valid) to evaluate clonogenic ability . Statistical Analysis Comparisons between groups for continuous variables were conducted using the Wilcoxon rank-sum test . Survival differences were assessed with the log-rank test . Univariate and multivariate Cox regression models were applied to identify independent prognostic factors. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant. Discussion Hepatocellular carcinoma (HCC) development involves multiple etiologies, including obesity, alcohol abuse, viral hepatitis, and non-alcoholic fatty liver disease (NAFLD), with most causes leading to cirrhosis before its onset [ 15 ]. The mechanical stiffness of the extracellular matrix (ECM) plays a crucial role in tumorigenesis and progression [ 16 ].Through systematic bioinformatics analysis, this study first identified 39 mechanosensitive-related genes (MSRGs) significantly associated with overall survival (OS) in hepatocellular carcinoma (HCC) (Fig. 1 ). This discovery provides a crucial molecular foundation for understanding the mechanobiological mechanisms of HCC.TP53 and CTNNB1 (encoding β-catenin), as classic tumor driver genes, exhibit high mutation rates [ 16 , 17 ], suggesting deep cross-talk between mechanical signaling networks and classical oncogenic pathways.Specifically, TP53 not only functions as a guardian of the genome in response to DNA damage but also senses changes in extracellular matrix stiffness, influencing cell fate decisions by regulating cytoskeletal reorganization [ 18 ]. CTNNB1, a core effector molecule in the Wnt/β-catenin pathway, has been demonstrated to be activated by matrix stiffness, thereby promoting stem cell-like properties and chemotherapy resistance in HCC cells [ 19 ].Notably, most MSRGs in this study exhibited hazard ratios (HR) greater than 1 with statistical significance. This expression-prognosis association pattern indicates that high expression of these genes often promotes tumor progression through multiple mechanisms. From a biological function perspective, these genes may participate in regulating key malignant phenotypes such as cell cycle progression, epithelial-mesenchymal transition (EMT), and angiogenesis.For instance, TTN (troponin T), a crucial component of the cytoskeleton and a primary determinant of cardiac muscle cell stiffness, may alter cellular responsiveness to mechanical stress when abnormally expressed [ 20 ]. MUC16, a common diagnostic marker across multiple cancer types [ 21 ], potentially influences tumor invasiveness by regulating cell-matrix interactions. Based on MSRG characteristics, this study employed consensus clustering to classify hepatocellular carcinoma patients into two molecular subtypes with markedly divergent prognoses. Survival analysis revealed significantly superior overall survival in Cluster 2 patients compared to Cluster 1, demonstrating the prognostic value of mechanical sensitivity signaling in HCC.In-depth analysis of the immune microenvironment characteristics of the two subtypes revealed that Cluster 2 exhibited a typical \"hot\" immune phenotype, with significant enrichment of effector immune cells such as CD8⁺ T cells in its tumor microenvironment, while Cluster 1 presented a relatively \"cold\" immune state.Recent studies have demonstrated that alterations in ECM structure and stiffness lead to CD8⁺ T cell exhaustion [ 22 ]. Our findings further suggest that mechanosensitive signals regulate immune cell recruitment and function. Quantitative analysis using the CIBERSORT algorithm revealed significantly elevated infiltration levels of activated CD4⁺ T cells, γδ T cells, and myeloid-derived suppressor cells in Cluster 2, indicating a more active antitumor immune response in this subtype.In contrast, Cluster 1 is characterized by enrichment of M0 macrophages and quiescent dendritic cells, indicating an immunosuppressive microenvironment. ESTIMATE algorithm further confirms that the better-prognosis Cluster 2 exhibits higher immune and stroma scores with relatively lower tumor purity, consistent with the pathological feature of extensive immune cell infiltration. The interaction between mechanical stress and the immune microenvironment is key to understanding this classification system. Physical factors within the tumor microenvironment—including extracellular matrix stiffness, solid stress, and fluid shear stress—can significantly influence immune cell behavior and function through mechanosensitive signaling pathways [ 23 ].In this study, high-risk group patients exhibited marked activation of ECM-receptor interaction pathways. This ECM remodeling has been reported to form physical barriers that impede effector T cell infiltration into the tumor parenchyma [ 24 ], thereby contributing to the development of an immune exclusion phenotype.Key inhibitory immune checkpoint molecules including CD276, HAVCR1, IDO2, and TNFRSF18 were significantly upregulated in the high-risk group [ 25 – 27 ], suggesting that mechanical stress may promote tumor immune escape by inducing an immunosuppressive microenvironment.Notably, chemokine expression profiles also exhibit significant differences between the two subtypes. CCL20, CCL28, CCL4, and XCL1—key lymphocyte chemokines [ 28 ]—show elevated expression in the low-risk group, potentially reflecting enhanced immune cell recruitment capacity. To develop a clinically valuable prognostic assessment tool, this study employed LASSO-Cox regression analysis to screen key variables from 39 mechanosensitive-related genes. The selected genes—HPN, ENDOG, UCN, FYN, ETV1, and KCNQ3—constituted the most predictive gene combination. The model's reliability was thoroughly validated through multiple statistical and clinical metrics.Kaplan-Meier survival analysis revealed significant differences in overall survival between high-risk and low-risk groups stratified by median risk score, with markedly increased mortality risk in the high-risk cohort. This finding remained consistent across both training and validation sets. Time-dependent ROC curve analysis further demonstrated the model's excellent discrimination and calibration.Notably, the risk score exhibited a unique inverse correlation with tumor pathological staging: patients with advanced disease had lower risk scores than those with early-stage disease. This phenomenon suggests the scoring system may emphasize reflecting the biological malignant potential of tumors rather than solely reflecting pathological progression stages. This prognostic value, independent of traditional TNM staging, provides a complementary biomarker for clinical risk assessment. Among the six core genes, potassium voltage-gated channel subfamily Q member 3 (KCNQ3) was identified through SHAP interpretability analysis as the key driver gene with the highest contribution to model prediction.The SHAP dependency plot clearly demonstrated a linear positive correlation between KCNQ3 expression levels and its contribution to risk scoring, indicating that high expression of this gene is a consistent factor elevating patient mortality risk. Survival analysis further confirmed that patients with high KCNQ3 expression exhibited significantly shorter overall survival, and this association was independent of other clinical and pathological parameters, establishing its reliability as a prognostic biomarker.The voltage-gated potassium channel encoded by KCNQ3 plays a central role in maintaining membrane potential homeostasis, with previous research primarily focused on neurological disorders [ 29 – 31 ]. In oncology, KCNQ3 has been reported to promote metastasis in esophageal adenocarcinoma and thyroid cancer [ 32 , 33 ]. This study first reveals its oncogenic function in the development of hepatocellular carcinoma.In vitro functional experiments provide direct evidence for KCNQ3's carcinogenic mechanism. In hepatocellular carcinoma cell lines, KCNQ3 expression was significantly suppressed via short hairpin RNA-mediated gene silencing. Subsequent CCK-8 cell proliferation assays and plate clonogenic assays demonstrated markedly reduced proliferation and clonogenic capacity in the KCNQ3-knockdown group compared to controls.This phenotypic change suggests KCNQ3 may promote malignant proliferation of hepatocellular carcinoma cells by sustaining survival signaling, consistent with the functional characteristics of classic oncogenes. We hypothesize that KCNQ3 may influence calcium influx by regulating membrane potential, thereby activating downstream survival signaling pathways such as PI3K-AKT-mTOR.Furthermore, analysis of the immune microenvironment revealed a significant positive correlation between KCNQ3 expression and M0 macrophage infiltration, suggesting this gene may indirectly promote tumor progression by recruiting tumor-associated macrophages to remodel the immunosuppressive microenvironment. These hypotheses require further experimental validation. However, this study has the following limitations: First, it is a retrospective analysis based on public databases, lacking prospective clinical validation, and the model's generalizability requires confirmation. Second, in vitro experiments only validated the function of KCNQ3, while the roles and detailed molecular mechanisms of the other core genes remain unexplored.Future research requires to conduct multicenter prospective studies to validate the model's clinical applicability; establish gene-interventional animal models to deeply analyze the molecular mechanisms of KCNQ3; and explore combined strategies of mechanical sensitivity signaling and immune checkpoint therapy. Declarations Funding This research received no external funding. Clinical Trial Number Clinical trial number: not applicable. Ethics, Consent to Participate, and Consent to Publish Ethics, Consent to Participate, and Consent to Publish declarations: not applicable. Data Availability Statement The data that support the findings of this study are available from the following public repositories: • Bulk RNA-seq and clinical data: The Cancer Genome Atlas (TCGA) Liver Hepatocellular Carcinoma (LIHC) dataset, accessible via accession number TCGA-LIHC at https://portal.gdc.cancer.gov/. • Single-cell RNA-seq data: Gene Expression Omnibus (GEO) dataset GSE149614, accessible at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE149614. These datasets were used for secondary data analysis as described in the Methods section. All accession numbers and repository links are also referenced in the main text and Methods section of the manuscript. Author Contribution # Author Contributions StatementJ.C. (Jin Hongjun) conceptualized the study, supervised the research design and implementation, and revised the manuscript critically for important intellectual content; P.C. (Cheng Peng) performed the bioinformatics analyses, conducted the in vitro experiments, and drafted the main manuscript text; X.Y.F. (Feng Xueyi) curated the multi-omics and clinical data, and validated the bioinformatic analysis results; X.L.H. (Huang Xianlu) prepared all the figures and tables, and assisted with the functional enrichment and statistical analyses. All authors read and approved the final version of the manuscript and agreed to the submission of the work to *Discover Oncology*. References Dopazo C, Søreide K, Rangelova E, et al. Hepatocellular carcinoma. Eur J Surg Oncol. 2024;50(1):107313. doi:10.1016/j.ejso.2023.107313 Nepomnyashchaya EM, Shaposhnikov AV, Yurieva EA. Hepatocellular carcinoma: new provisions of the WHO classification, 5th edition, 2019. Archives of Pathology. 2020;82(6):36-40. doi:10.17116/patol20208206136 Feng F, Zhao Y. Hepatocellular Carcinoma: Prevention, Diagnosis, and Treatment. Med Princ Pract. 2024;33(5):414-423. doi:10.1159/000539349 Tsui YM, Ho DW, Ng IO. Unraveling the tumor-initiating cells in hepatocellular carcinoma. Cancer Cell. 2024;42(12):1990-1993. doi:10.1016/j.ccell.2024.10.018 Sharma KK, Mohsin M, Mittal P, et al. Diagnosis of the Initial Stage of Hepatocellular Carcinoma: A Review. Current Pharmaceutical Design. 2024;30(22):1708-1724. doi:10.2174/0113816128298875240321073907 Safri F, Nguyen R, Zerehpooshnesfchi S, George J, Qiao L. Heterogeneity of hepatocellular carcinoma: from mechanisms to clinical implications. Cancer Gene Ther. 2024;31(8):1105-1112. doi:10.1038/s41417-024-00764-w Ma L, Hernandez MO, Zhao Y, et al. Tumor Cell Biodiversity Drives Microenvironmental Reprogramming in Liver Cancer. Cancer Cell. 2019;36(4):418-430.e6. doi:10.1016/j.ccell.2019.08.007 Tang H, Cheng X, Liang L, Chen BZ, Liu C, Wang Y. A stimulus responsive microneedle-based drug delivery system for cancer therapy. Biomater Sci. 2024;12(24):6274-6283. Published 2024 Dec 3. doi:10.1039/d4bm00741g Roy AM, Iyer R, Chakraborty S. The extracellular matrix in hepatocellular carcinoma: Mechanisms and therapeutic vulnerability. Cell Rep Med. 2023;4(9):101170. doi:10.1016/j.xcrm.2023.101170 Lv H, Li L, Sun M, et al. Mechanism of regulation of stem cell differentiation by matrix stiffness. Stem Cell Res Ther. 2015;6(1):103. Published 2015 May 27. doi:10.1186/s13287-015-0083-4 Miroshnikova YA, Manet S, Li X, Wickström SA, Faurobert E, Albiges-Rizo C. Calcium signaling mediates a biphasic mechanoadaptive response of endothelial cells to cyclic mechanical stretch. Mol Biol Cell. 2021;32(18):1724-1736. doi:10.1091/mbc.E21-03-0106 Abdollahiyan P, Oroojalian F, Baradaran B, de la Guardia M, Mokhtarzadeh A. Advanced mechanotherapy: Biotensegrity for governing metastatic tumor cell fate via modulating the extracellular matrix. J Control Release. 2021;335:596-618. doi:10.1016/j.jconrel.2021.06.002 Li M, Zhang X, Wang M, et al. Activation of Piezo1 contributes to matrix stiffness-induced angiogenesis in hepatocellular carcinoma. Cancer Commun (Lond). 2022;42(11):1162-1184. doi:10.1002/cac2.12364 Zhang J, Li J, Hou Y, et al. Osr2 functions as a biomechanical checkpoint to aggravate CD8+ T cell exhaustion in tumors. Cell. 2024;187(13):3409-3426.e24. doi:10.1016/j.cell.2024.04.023 Tarao K, Nozaki A, Ikeda T, Sato A, Komatsu H, Komatsu T, Taguri M, Tanaka K. Real impact of liver cirrhosis on the development of hepatocellular carcinoma in various liver diseases-meta-analytic assessment. Cancer Med. 2019 Mar;8(3):1054-1065. doi: 10.1002/cam4.1998. Epub 2019 Feb 21. PMID: 30791221; PMCID: PMC6434205. Giacomelli AO, Yang X, Lintner RE, McFarland JM, Duby M, Kim J, Howard TP, Takeda DY, Ly SH, Kim E, Gannon HS, Hurhula B, Sharpe T, Goodale A, Fritchman B, Steelman S, Vazquez F, Tsherniak A, Aguirre AJ, Doench JG, Piccioni F, Roberts CWM, Meyerson M, Getz G, Johannessen CM, Root DE, Hahn WC. Mutational processes shape the landscape of TP53 mutations in human cancer. Nat Genet. 2018 Oct;50(10):1381-1387. doi: 10.1038/s41588-018-0204-y. Epub 2018 Sep 17. PMID: 30224644; PMCID: PMC6168352. Li Y, Zhang F, Yang D. Comprehensive assessment and meta-analysis of the association between CTNNB1 polymorphisms and cancer risk. Biosci Rep. 2017 Nov 23;37(6):BSR20171121. doi: 10.1042/BSR20171121. PMID: 28963373; PMCID: PMC5700267. Tornesello ML. TP53 mutations in cancer: Molecular features and therapeutic opportunities (Review). Int J Mol Med. 2025 Jan;55(1):7. doi: 10.3892/ijmm.2024.5448. Epub 2024 Oct 25. PMID: 39450536; PMCID: PMC11554381. Yu F, Yu C, Li F, Zuo Y, Wang Y, Yao L, Wu C, Wang C, Ye L. Wnt/β-catenin signaling in cancers and targeted therapies. Signal Transduct Target Ther. 2021 Aug 30;6(1):307. doi: 10.1038/s41392-021-00701-5. PMID: 34456337; PMCID: PMC8403677. Loescher CM, Hobbach AJ, Linke WA. Titin (TTN): from molecule to modifications, mechanics, and medical significance. Cardiovasc Res. 2022 Nov 10;118(14):2903-2918. doi: 10.1093/cvr/cvab328. PMID: 34662387; PMCID: PMC9648829. Zhang XY, Hong LL, Ling ZQ. MUC16/CA125 in cancer: new advances. Clin Chim Acta. 2025 Jan 15;565:119981. doi: 10.1016/j.cca.2024.119981. Epub 2024 Oct 4. PMID: 39368688. Zhang J, Li J, Hou Y, Lin Y, Zhao H, Shi Y, Chen K, Nian C, Tang J, Pan L, Xing Y, Gao H, Yang B, Song Z, Cheng Y, Liu Y, Sun M, Linghu Y, Li J, Huang H, Lai Z, Zhou Z, Li Z, Sun X, Chen Q, Su D, Li W, Peng Z, Liu P, Chen W, Huang H, Chen Y, Xiao B, Ye L, Chen L, Zhou D. Osr2 functions as a biomechanical checkpoint to aggravate CD8+ T cell exhaustion in tumors. Cell. 2024 Jun 20;187(13):3409-3426.e24. doi: 10.1016/j.cell.2024.04.023. Epub 2024 May 13. PMID: 38744281. Ren T, Sun L, Zheng Y, Jiang Y, Guo Y, Ma J. Mechanical forces and immune cells in the tumor microenvironment: from regulation mechanisms to therapeutic strategies. Int J Surg. 2025 Aug 1;111(8):5420-5434. doi: 10.1097/JS9.0000000000002636. Epub 2025 Jun 5. PMID: 40478969. Zhu P, Lu H, Wang M, Chen K, Chen Z, Yang L. Targeted mechanical forces enhance the effects of tumor immunotherapy by regulating immune cells in the tumor microenvironment. Cancer Biol Med. 2023 Jan 12;20(1):44–55. doi: 10.20892/j.issn.2095-3941.2022.0491. PMID: 36647779; PMCID: PMC9843446. Liu S, Liang J, Liu Z, Zhang C, Wang Y, Watson AH, Zhou C, Zhang F, Wu K, Zhang F, Lu Y, Wang X. The Role of CD276 in Cancers. Front Oncol. 2021 Mar 26;11:654684. doi: 10.3389/fonc.2021.654684. PMID: 33842369; PMCID: PMC8032984. Bod L, Kye YC, Shi J, Torlai Triglia E, Schnell A, Fessler J, Ostrowski SM, Von-Franque MY, Kuchroo JR, Barilla RM, Zaghouani S, Christian E, Delorey TM, Mohib K, Xiao S, Slingerland N, Giuliano CJ, Ashenberg O, Li Z, Rothstein DM, Fisher DE, Rozenblatt-Rosen O, Sharpe AH, Quintana FJ, Apetoh L, Regev A, Kuchroo VK. B-cell-specific checkpoint molecules that regulate anti-tumor immunity. Nature. 2023 Jul;619(7969):348-356. doi: 10.1038/s41586-023-06231-0. Epub 2023 Jun 21. Erratum in: Nature. 2025 Apr;640(8057):E1. doi: 10.1038/s41586-025-08858-7. PMID: 37344597; PMCID: PMC10795478. Jia T, Guo Y, Cheng XM, Zhou Z, Xu X, Liu H, Yang X. Multi-omics profiling identifies TNFRSF18 as a novel marker of exhausted CD8⁺ T cells and reveals tumor-immune dynamics in colorectal cancer. Clin Transl Med. 2025 Aug;15(8):e70425. doi: 10.1002/ctm2.70425. PMID: 40770837; PMCID: PMC12328248. Yi M, Li T, Niu M, Zhang H, Wu Y, Wu K, Dai Z. Targeting cytokine and chemokine signaling pathways for cancer therapy. Signal Transduct Target Ther. 2024 Jul 22;9(1):176. doi: 10.1038/s41392-024-01868-3. PMID: 39034318; PMCID: PMC11275440. Miceli F, Soldovieri MV, Weckhuysen S, Cooper EC, Taglialatela M. KCNQ3-Related Disorders. 2014 May 22 [updated 2023 Sep 28]. In: Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. PMID: 24851285. Wang T, Hoekzema K, Vecchio D, Wu H, Sulovari A, Coe BP, Gillentine MA, Wilfert AB, Perez-Jurado LA, Kvarnung M, Sleyp Y, Earl RK, Rosenfeld JA, Geisheker MR, Han L, Du B, Barnett C, Thompson E, Shaw M, Carroll R, Friend K, Catford R, Palmer EE, Zou X, Ou J, Li H, Guo H, Gerdts J, Avola E, Calabrese G, Elia M, Greco D, Lindstrand A, Nordgren A, Anderlid BM, Vandeweyer G, Van Dijck A, Van der Aa N, McKenna B, Hancarova M, Bendova S, Havlovicova M, Malerba G, Bernardina BD, Muglia P, van Haeringen A, Hoffer MJV, Franke B, Cappuccio G, Delatycki M, Lockhart PJ, Manning MA, Liu P, Scheffer IE, Brunetti-Pierri N, Rommelse N, Amaral DG, Santen GWE, Trabetti E, Sedláček Z, Michaelson JJ, Pierce K, Courchesne E, Kooy RF; SPARK Consortium; Nordenskjöld M, Romano C, Peeters H, Bernier RA, Gecz J, Xia K, Eichler EE. Large-scale targeted sequencing identifies risk genes for neurodevelopmental disorders. Nat Commun. 2020 Oct 1;11(1):4932. doi: 10.1038/s41467-020-18723-y. Erratum in: Nat Commun. 2020 Oct 21;11(1):5398. doi: 10.1038/s41467-020-19289-5. PMID: 33004838; PMCID: PMC7530681. Edmond MA, Hinojo-Perez A, Efrem M, Yi-Chun L, Shams I, Hayoz S, de la Cruz A, Perez Rodriguez ME, Diaz-Solares M, Dykxhoorn DM, Luo YL, Barro-Soria R. Lipophilic compounds restore function to neurodevelopmental-associated KCNQ3 mutations. Commun Biol. 2024 Sep 19;7(1):1181. doi: 10.1038/s42003-024-06873-4. PMID: 39300259; PMCID: PMC11413209. Foley K, Shorthouse D, Rahrmann E, Zhuang L, Devonshire G, Gilbertson RJ; OCCAMS consortium; Fitzgerald RC, Hall BA. SMAD4 and KCNQ3 alterations are associated with lymph node metastases in oesophageal adenocarcinoma. Biochim Biophys Acta Mol Basis Dis. 2024 Jan;1870(1):166867. doi: 10.1016/j.bbadis.2023.166867. Epub 2023 Aug 28. PMID: 37648039. Li Q, Liu M, Song X, Xie L, Liu D, Su T, Xu Y, Li G, Liang B, Huang D. Increased KCNQ3 expression in papillary thyroid cancer promotes proliferation and migration. Cancer Cell Int. 2025 Nov 17;25(1):406. doi: 10.1186/s12935-025-04049-6. PMID: 41250218; PMCID: PMC12625505. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 15 May, 2026 Reviews received at journal 19 Apr, 2026 Reviews received at journal 14 Apr, 2026 Reviews received at journal 13 Apr, 2026 Reviews received at journal 13 Apr, 2026 Reviewers agreed at journal 08 Apr, 2026 Reviewers agreed at journal 04 Apr, 2026 Reviewers agreed at journal 03 Apr, 2026 Reviews received at journal 03 Apr, 2026 Reviewers agreed at journal 02 Apr, 2026 Reviews received at journal 31 Mar, 2026 Reviewers agreed at journal 30 Mar, 2026 Reviewers agreed at journal 29 Mar, 2026 Reviewers invited by journal 27 Mar, 2026 Editor assigned by journal 14 Mar, 2026 Submission checks completed at journal 13 Mar, 2026 First submitted to journal 13 Mar, 2026 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-9047659\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Research Article\",\"associatedPublications\":[],\"authors\":[{\"id\":614028311,\"identity\":\"914ceadc-e3be-42df-a9fe-73d5e32f1c18\",\"order_by\":0,\"name\":\"peng cheng\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Graduate School of Bengbu Medical University / Department of Hepatobiliary Surgery, Lu'an People's Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"peng\",\"middleName\":\"\",\"lastName\":\"cheng\",\"suffix\":\"\"},{\"id\":614028312,\"identity\":\"448dacfa-a0c8-458a-a8e8-b8bdce203942\",\"order_by\":1,\"name\":\"xueyi feng\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Department of Hepatobiliary Surgery, Lu'an People's Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"xueyi\",\"middleName\":\"\",\"lastName\":\"feng\",\"suffix\":\"\"},{\"id\":614028313,\"identity\":\"6b677c08-2324-4d9e-852b-9ecb575649f4\",\"order_by\":2,\"name\":\"xianlu huang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Graduate School of Bengbu Medical University / Department of Hepatobiliary Surgery, Lu'an People's Hospital\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"xianlu\",\"middleName\":\"\",\"lastName\":\"huang\",\"suffix\":\"\"},{\"id\":614028314,\"identity\":\"4260d07e-e187-4034-b0ff-ff53354e9e5b\",\"order_by\":3,\"name\":\"JUNHONG JIN\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBklEQVRIie3RsUoDMRjA8e8I3C05uuZA2ycQchSKQ62vkhC4qYNjhyInBbsEXFMQnyHTrSYE6tIHcOy9gdDFwcEo56SpNwrmv4RAfgkfAYjF/mbMACYA2U1tXhZ+nwEkdS+CnbVq51f0O+kilXD5bQ9Cn1nrrk7ORwOYU5M/TPkjQrZVMBueBVyhGHMKk3JT76gpmopLlIpSgxhPzM9kQDzBmCTaSmrKxnmCJ8UeDG8CJO3IpXaYGn7fg3y9wvU2ZcbWHdFHSCH3n0RsJPJiW42l87MoGp6FPs3FIZfXF3ejdnV4W05P1+uVbeViNgwR/w8MEvn9qtDxjzJ/2euxA7FYLPbvewfGX2Hehf05wAAAAABJRU5ErkJggg==\",\"orcid\":\"\",\"institution\":\"Graduate School of Bengbu Medical University / Department of Hepatobiliary Surgery, Lu'an People's Hospital\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"JUNHONG\",\"middleName\":\"\",\"lastName\":\"JIN\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2026-03-06 07:55:26\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-9047659/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-9047659/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":105844074,\"identity\":\"64769bdf-7d37-4dcc-982c-c563dcdc14d1\",\"added_by\":\"auto\",\"created_at\":\"2026-03-31 17:30:40\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":707483,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eIdentification and Genomic Characterization of OS-Associated Mechanoresponsive Genes (MRGs) in Hepatocellular Carcinoma\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ Forest plot from univariate Cox regression analysis. Displays the hazard ratios (HRs) and their 95% confidence intervals for the 39 mechanoresponsive genes significantly associated with overall survival in HCC. An HR \\u0026gt; 1 indicates a risk gene.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Somatic mutation profile of the 39 OS-associated MRGs in the HCC cohort. The waterfall plot illustrates the mutation frequency and types for each gene.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Panoramic view of somatic mutations across 315 HCC samples. Each row represents a sample, each column represents a frequently mutated gene, and different colors denote different variant types.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ Circos plot showing the chromosomal distribution of the 39 OS-associated MRGs.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE.\\u003c/strong\\u003e​ Heatmap of expression correlation among the 39 OS-associated MRGs. Color intensity represents the Pearson correlation coefficient.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eF.\\u003c/strong\\u003e​ Frequency plot of copy number variations (CNVs) for the 39 OS-associated MRGs. Blue dots represent the frequency of copy number deletions, red dots represent the frequency of copy number amplifications. The x-axis shows the percentage CNV frequency.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/aedc404d0eeafbeff81d8741.png\"},{\"id\":105904911,\"identity\":\"e5361381-09ae-4aa9-85bd-2f6464a91761\",\"added_by\":\"auto\",\"created_at\":\"2026-04-01 10:11:02\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":575527,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eIdentification of Molecular Subtypes Based on OS-Associated MRG Expression Profiles and Their Characteristics\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ Consensus clustering heatmap. When the cluster number K=2, the consensus matrix showed the highest stability, dividing samples into Cluster 1 (n=199) and Cluster 2 (n=171).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Principal component analysis (PCA) plot. The two clusters show clear separation in the two-dimensional space defined by principal component 1 and principal component 2.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Kaplan-Meier survival curves for the two molecular subtypes. Patients in Cluster 2 had significantly better overall survival than those in Cluster 1 (Log-rank test, p = 0.0012).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ Stacked bar plot of immune cell infiltration proportions based on the CIBERSORT algorithm, showing the compositional differences of 22 immune cell subsets between the two subtypes.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE.\\u003c/strong\\u003e​ Box plot comparison of immune cell subsets with significant infiltration differences between the two subtypes.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eF.\\u003c/strong\\u003e​ Violin plots comparing ESTIMATE scores, Immune scores, Stromal scores, and tumor purity between the two subtypes. Cluster 2 had significantly higher ESTIMATE, Immune, and Stromal scores, and lower tumor purity.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/d94823938a9213a12976338a.png\"},{\"id\":105844079,\"identity\":\"0b220622-038d-404c-806b-71b691f8ffd4\",\"added_by\":\"auto\",\"created_at\":\"2026-03-31 17:30:40\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":549464,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eConstruction and Validation of the MRG-Based Prognostic Risk Model\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ Coefficient profile plot from the LASSO-Cox regression analysis. The regression coefficients of genes gradually shrink towards zero as the regularization parameter λ increases.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Cross-validation curve for the LASSO regression. The optimal λ value (lambda.min) was determined via ten-fold cross-validation, minimizing the partial likelihood deviation.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Distribution of patient risk scores, survival status, and survival time. Patients were divided into high-risk and low-risk groups based on the median risk score.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ Comprehensive heatmap integrating the expression of risk model genes, risk scores, and clinicopathological characteristics.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE.\\u003c/strong\\u003e​ Receiver operating characteristic (ROC) curves predicting patient 1-, 3-, and 5-year overall survival using the risk model. The areas under the curve (AUC) were 0.74, 0.74, and 0.73, respectively.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eF.\\u003c/strong\\u003e​ Kaplan-Meier survival curves for patients in the high-risk and low-risk groups. The high-risk group had significantly shorter overall survival (Log-rank test, p \\u0026lt; 0.0001).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eG-J.\\u003c/strong\\u003e​ Violin plots showing the distribution of risk scores among patients with different clinicopathological stages.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/c8ad0d0810b3158904b7cb26.png\"},{\"id\":105844076,\"identity\":\"464edf81-3a40-4b8e-ba48-7a12fbdb3efb\",\"added_by\":\"auto\",\"created_at\":\"2026-03-31 17:30:40\",\"extension\":\"png\",\"order_by\":4,\"title\":\"Figure 4\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":500117,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eFunctional Pathway Enrichment and Activity Analysis Between High- and Low-Risk Groups\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ Bubble plot of Gene Ontology (GO) enrichment analysis for differentially expressed genes (DEGs) between the high- and low-risk groups. Significantly enriched terms in Biological Process, Cellular Component, and Molecular Function categories are shown.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Bubble plot of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis for DEGs between the high- and low-risk groups.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Bar plot showing the differences in HALLMARK pathway activities between the high- and low-risk groups based on Gene Set Variation Analysis (GSVA).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ Heatmap of correlations between HALLMARK pathway GSVA scores and the risk scores.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE-H.\\u003c/strong\\u003e​ Kaplan-Meier survival curves for four key HALLMARK pathways, with patients grouped based on the median GSVA score for each pathway's activity.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture4.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/95fad7b54d585902830c7702.png\"},{\"id\":105904447,\"identity\":\"2f8464db-d9c4-4a7b-b73c-71d274946e76\",\"added_by\":\"auto\",\"created_at\":\"2026-04-01 10:08:40\",\"extension\":\"png\",\"order_by\":5,\"title\":\"Figure 5\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":577684,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eTumor Immune Microenvironment Characteristics of High- and Low-Risk Groups\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ Box plots comparing the infiltration abundance of 22 immune cell types between the high- and low-risk groups.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Violin plots comparing ESTIMATE scores, Immune scores, Stromal scores, and tumor purity between the high- and low-risk groups.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Heatmap of correlations between the expression of core genes in the risk model and the infiltration abundance of immune cells.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ Box plots comparing the expression levels of 22 immune checkpoint molecules between the high- and low-risk groups.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE.\\u003c/strong\\u003e​ Box plots comparing the expression levels of 17 chemokines and cytokines between the high- and low-risk groups.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture5.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/a123ff918879d3a3439d90b0.png\"},{\"id\":105844080,\"identity\":\"88a6e4bc-9462-4957-9249-956df9fb642a\",\"added_by\":\"auto\",\"created_at\":\"2026-03-31 17:30:40\",\"extension\":\"png\",\"order_by\":6,\"title\":\"Figure 6\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":187957,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eGenomic Alteration Landscape and Drug Sensitivity Prediction in High- and Low-Risk Groups\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA-B.\\u003c/strong\\u003e​ Waterfall plots of somatic mutations in the high-risk and low-risk groups, showing the top 20 most frequently mutated genes.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC-F.\\u003c/strong\\u003e​ Box plots comparing the predicted sensitivity (based on GDSC database) to four targeted drugs between the high- and low-risk groups. A lower IC50 value indicates higher drug sensitivity.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture6.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/28be6b808ad0d7f167ac6f3b.png\"},{\"id\":105844078,\"identity\":\"aabcb2df-9e63-434d-bb98-3e7204185d40\",\"added_by\":\"auto\",\"created_at\":\"2026-03-31 17:30:40\",\"extension\":\"png\",\"order_by\":7,\"title\":\"Figure 7\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":1193854,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eSingle-Cell Transcriptomic Profiling Reveals Tumor Microenvironment Heterogeneity in High- and Low-Risk Groups\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ UMAP projection of the single-cell transcriptome from HCC tissue, with colors representing major cell types identified by unsupervised clustering.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Dot plot showing the expression of marker genes used to define the major cell types.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Bar plot showing the number of cells per major cell type.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ UMAP projection visualizing the expression values of the risk score at the single-cell level.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE.\\u003c/strong\\u003e​ Dot plot showing the expression level of the risk score across different cell types.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eF.\\u003c/strong\\u003e​ UMAP plot colored by the patient's risk group, showing the distribution of cells derived from high-risk and low-risk patients within the microenvironment.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eG-I.\\u003c/strong\\u003e​ Bubble plots of KEGG pathway enrichment analysis for differentially expressed genes in epithelial cells, T/NK cells, and endothelial cells between the high- and low-risk groups.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eJ.\\u003c/strong\\u003e​ Heatmap displaying differences in the strength of cell-cell communication between the high- and low-risk groups.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eK.\\u003c/strong\\u003e​ Chord diagram illustrating the cell-cell communication networks in the high-risk group versus the low-risk group.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eL.\\u003c/strong\\u003e​ Dot plot showing ligand-receptor interaction signals specifically strengthened or weakened in the high-risk group.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture7.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/712631022cfed188687a762d.png\"},{\"id\":106401587,\"identity\":\"b3994a6f-49e8-46ff-9615-b3752219e9a4\",\"added_by\":\"auto\",\"created_at\":\"2026-04-08 09:07:41\",\"extension\":\"png\",\"order_by\":8,\"title\":\"Figure 8\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":356155,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ePrognostic Value and Functional Validation of KCNQ3, a Key Driver Gene in the Risk Model\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eA.\\u003c/strong\\u003e​ Dot plot of gene importance based on SHAP model interpretability analysis for the risk model genes. Dot color represents gene expression value, horizontal position represents the SHAP value.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eB.\\u003c/strong\\u003e​ Dependence plot of KCNQ3 gene expression level versus its SHAP value.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eC.\\u003c/strong\\u003e​ Kaplan-Meier survival curves for patients grouped by the median expression of KCNQ3. Patients with high expression had worse overall survival (p = 0.0035).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eD.\\u003c/strong\\u003e​ Cell proliferation curve measured by CCK-8 assay in Huh7 HCC cell lines after KCNQ3 knockdown (*** p \\u0026lt; 0.001).\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eE-F.\\u003c/strong\\u003e​ Representative images and quantitative bar plot from the plate colony formation assay, showing that KCNQ3 knockdown significantly inhibited the clonogenic ability of Huh7 cells (**​ p \\u0026lt; 0.01).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Picture8.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/2235643d9e476fb19ed4a92d.png\"},{\"id\":106406605,\"identity\":\"ad2ec933-1b07-4535-b82e-2dd3a92e7d3f\",\"added_by\":\"auto\",\"created_at\":\"2026-04-08 09:33:07\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":6954772,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-9047659/v1/36c4eab8-bf85-425b-aa2e-1f6ca4e7eb2f.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Characteristics of Mechanically Stimulated Genes in Hepatocellular Carcinoma and Their Role as Prognostic Markers\",\"fulltext\":[{\"header\":\"Research Background\",\"content\":\"\\u003cp\\u003eHepatocellular carcinoma (HCC) ranks among the most prevalent malignant tumors globally, with persistently high incidence and mortality rates [1].According to World Health Organization (WHO) statistics, HCC ranks third among cancer-related deaths globally, following lung cancer and colorectal cancer [2]. Despite recent advances in HCC diagnosis and treatment—such as improved imaging techniques and the introduction of targeted and immunotherapies—patient prognosis remains poor, with a five-year survival rate below 20% [3].\\u003c/p\\u003e\\n\\u003cp\\u003eThe development and progression of liver cancer are closely associated with multiple factors, primarily including chronic hepatitis (such as hepatitis B and C), cirrhosis, metabolic syndrome (e.g., non-alcoholic fatty liver disease), and chronic alcohol consumption. These factors induce chronic inflammation and fibrosis in the liver, thereby promoting hepatocellular carcinogenesis [4,5].Furthermore, HCC exhibits high heterogeneity, with significant variations in gene expression, cellular composition, and microenvironmental characteristics among tumors from different patients, adding complexity and challenges to treatment.\\u003c/p\\u003e\\n\\u003cp\\u003eIn recent years, the role of the tumor microenvironment (TME) in tumorigenesis, progression, and metastasis has garnered increasing attention [6]. The TME encompasses not only tumor cells themselves but also diverse non-tumor cells (e.g., immune cells, fibroblasts, endothelial cells) and the extracellular matrix (ECM). These components influence tumor biological behavior through complex interactions [7].Notably, mechanical stimuli, as a key factor within the TME, can influence tumor cell proliferation, migration, and invasion through ECM stiffness, cell-cell contacts, and cell-matrix interactions [8].\\u003c/p\\u003e\\n\\u003cp\\u003eStudies indicate that liver cancer tissue exhibits significantly higher stiffness than normal liver tissue, primarily attributed to excessive deposition and cross-linking of the extracellular matrix [9]. Mechanical stimuli can modulate cellular biology through multiple mechanoreceptors (e.g., integrins, cadherins, GPCRs, Piezo channels) and downstream signaling pathways (e.g., YAP/TAZ pathway) [10-12].For instance, Piezo1 channels sense mechanical stimuli, enhancing HCC angiogenesis through calcium influx and downstream signaling pathways [13]. Furthermore, mechanical stress influences immune cell infiltration within the tumor microenvironment; highly rigid tumor tissues typically exhibit immunosuppressive features such as reduced T-cell infiltration and elevated expression of immune checkpoint molecules [14].\\u003c/p\\u003e\\n\\u003cp\\u003eHowever, despite the emerging understanding of mechanical stimulation's role in the tumor microenvironment, its specific mechanisms in HCC remain unclear. Particularly, how Mechanical Stimulus-Related Genes (MSRGs) influence HCC initiation, progression, and prognosis has not been systematically investigated.Therefore, this study aims to identify MSRGs through integrated bulk RNA-Seq and single-cell RNA-Seq analyses and construct prognostic biomarkers based on these genes, thereby providing new insights for personalized treatment of HCC.\\u003c/p\\u003e\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\\n\"},{\"header\":\"Results\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eGenetic Landscape of OS-Related MSRGs in Hepatocellular Carcinoma\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eThrough univariate Cox regression analysis, we identified 39 mechanosensitive-related genes (MSRGs) significantly associated with the overall survival (OS) of hepatocellular carcinoma (HCC) patients (Figure 1A). These included key genes such as \\u003cstrong\\u003eTP53, CTNNB1, TTN, and MUC16\\u003c/strong\\u003e, most of which exhibited hazard ratios (HRs) greater than 1, suggesting that their high expression may be closely linked to poor patient prognosis . Among 371 HCC samples, 315 (84.91%) harbored at least one somatic mutation in the 39 OS-related MSRGs (Figure 1B). \\u003cstrong\\u003eTP53 was the most frequently mutated gene\\u003c/strong\\u003e. Further variant classification revealed that \\u003cstrong\\u003emissense mutations were the dominant type\\u003c/strong\\u003e (Figure 1C). A Circos plot clearly illustrated the chromosomal distribution of these 39 genes (Figure 1D). Gene expression correlation heatmap analysis uncovered extensive co-expression patterns or potential regulatory relationships among these genes (Figure 1E). Copy number variation (CNV) analysis demonstrated widespread \\u003cstrong\\u003ecopy number amplifications (gains) and deletions\\u003c/strong\\u003e of these OS-related MSRGs in HCC samples (Figure 1F), indicating that genomic instability may influence their expression and thereby contribute to HCC progression .\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eIdentification of Molecular Subtypes Based on MSRG Expression Profiles\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eUsing consensus clustering analysis, we stratified HCC samples into two distinct molecular subtypes with significant heterogeneity based on the expression profiles of the 39 OS-related MSRGs. When the cluster number K=2, clustering stability was optimal, ultimately dividing all samples into two subgroups: Cluster 1 (n=199) and Cluster 2 (n=171) (Figure 2A). Principal component analysis (PCA) revealed clear separation between the two subtypes in a two-dimensional space formed by Dim1 (contribution 21.6%) and Dim2 (contribution 5.9%) (Figure 2B), indicating fundamental differences in their transcriptomic features. Survival analysis showed that Cluster 2 patients had significantly better overall survival than Cluster 1 (Log-rank test, P = 0.0012) (Figure 2C), suggesting that MSRG-based clustering has important prognostic value. To further elucidate the underlying mechanisms, we quantified immune cell infiltration patterns in the tumor microenvironment (TME) using the CIBERSORT algorithm. Results indicated that Cluster 2 exhibited an immune-\\\"hot\\\" phenotype, characterized by significant enrichment of effector immune cells such as \\u003cstrong\\u003eCD8⁺ T cells\\u003c/strong\\u003e (Figure 2D). Specifically, the TME of Cluster\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eConstruction and Validation of an MSRG-Based Prognostic Risk Model\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eTo build a robust prognostic model, we performed \\u003cstrong\\u003eLASSO-Cox regression analysis\\u003c/strong\\u003e on the previously identified 39 OS-related MSRGs to mitigate overfitting and identify key genes. The coefficient profile plot demonstrated a gradual shrinkage of regression coefficients toward zero as the regularization parameter λ increased (Figure 3A). The optimal λ value (lambda.min) was determined via cross-validation to be -3.288, achieving a minimal deviation while maintaining model parsimony (Figure 3B). Ultimately, the model identified several core genes, including \\u003cstrong\\u003eHPN, ENDOG, UCN, FYN, ETV1, and KCNQ3\\u003c/strong\\u003e, which were used to calculate an individual risk score for each patient. Patients were stratified into \\u003cstrong\\u003ehigh-risk (n=182)\\u003c/strong\\u003e and \\u003cstrong\\u003elow-risk (n=183)\\u003c/strong\\u003e groups based on the median risk score.The survival time distribution plot revealed that patients in the high-risk group generally had shorter survival times, with a higher concentration of death events occurring in this group. In contrast, low-risk group patients exhibited longer survival times and were predominantly in a state of survival (Figure 3C). An integrated heatmap further indicated that the risk score was significantly correlated with multi-dimensional features, including tumor pathological stage and gene expression profiles (Figure 3D). \\u003cstrong\\u003eKaplan-Meier survival analysis\\u003c/strong\\u003e confirmed that patients in the high-risk group had a significantly shorter overall survival than those in the low-risk group (Log-rank test, p \\u0026lt; 0.0001) (Figure 3E), demonstrating the significant prognostic discriminatory power of the risk score.The model's predictive accuracy was evaluated by plotting \\u003cstrong\\u003eROC curves\\u003c/strong\\u003e and calculating the \\u003cstrong\\u003earea under the curve (AUC)\\u003c/strong\\u003e for 1-, 3-, and 5-year overall survival. The results showed stable and good discriminatory ability at all three key time points, with AUC values of 0.74, 0.74, and 0.73, respectively (Figure 3F). This indicates that the model provides reliable predictive efficacy for both short-term and long-term survival outcomes in liver cancer patients.To clarify the clinical significance of the risk score, its relationship with tumor pathological stage (Stage I-IV) was analyzed. Violin plots showed significant differences in the distribution of risk scores among patients at different stages (Figures 2G-J). Specifically, patients with advanced-stage disease (Stage IV) had significantly \\u003cstrong\\u003elower risk scores\\u003c/strong\\u003e compared to those with early-stage (Stage I, p \\u0026lt; 0.0001) and intermediate-stage (Stage II, p \\u0026lt; 0.01) disease. This result suggests that the MSRG-based risk score is closely associated with the progression of liver cancer and may serve as a potential indicator reflecting the degree of tumor malignancy.\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eFunctional Enrichment Analysis of Differentially Expressed Genes between High- and Low-Risk Groups\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eFunctional annotation of the differentially expressed genes (DEGs) between the high- and low-risk groups revealed key biological characteristics associated with the risk score. \\u003cstrong\\u003eGene Ontology (GO) enrichment analysis\\u003c/strong\\u003e (Figure 4A, top) showed that the DEGs were significantly enriched in biological processes (BP) closely related to cell proliferation, such as \\\"organelle fission,\\\" \\\"nuclear division,\\\" and \\\"chromosome segregation.\\\" Regarding cellular components (CC) and molecular functions (MF), the DEGs were primarily enriched in terms related to structural functions like \\\"spindle\\\" and \\\"microtubule binding.\\\" \\u003cstrong\\u003eKyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis\\u003c/strong\\u003e (Figure 4B) further indicated significant enrichment of DEGs in pathways such as \\\"Cell cycle,\\\" \\\"ECM-receptor interaction,\\\" and \\\"Protein digestion and absorption.\\\" These findings collectively suggest that high-risk tumors may possess stronger capabilities for cell proliferation and extracellular matrix remodeling.To quantify the activity changes in key biological pathways, we performed \\u003cstrong\\u003eGene Set Variation Analysis (GSVA)\\u003c/strong\\u003e. A bar plot (Figure 4C) visually represents the gene sets with significantly differential activity between the risk groups. Pathways related to proliferation and survival, such as the \\u003cstrong\\u003eG2M checkpoint, E2F targets, and PI3K-AKT-mTOR signaling\\u003c/strong\\u003e, were significantly up-regulated in the high-risk group. In contrast, metabolism-related pathways like \\u003cstrong\\u003efatty acid metabolism and estrogen response\\u003c/strong\\u003e were down-regulated. A correlation heatmap (Figure 4D) further elucidated the co-variation patterns among these differentially active gene sets and confirmed a clear association between their activity and the risk groups, with the high-risk group broadly linked to the activation of proliferation-related pathways.\\u003c/p\\u003e\\u003cp\\u003eTo confirm the clinical relevance of these pathway activities, patients were grouped based on the median activity of specific pathways, and Kaplan-Meier survival curves were plotted. The results showed that patients with high activity in the \\u003cstrong\\u003eE2F targets and PI3K-AKT-mTOR signaling pathways\\u003c/strong\\u003e had a significantly shorter overall survival (Figures E-F). Conversely, high activity in the \\u003cstrong\\u003efatty acid metabolism and late estrogen response pathways\\u003c/strong\\u003e was significantly associated with longer survival (Figures G-H). This provides prognostic validation, from the opposite angle, for the GSVA results, indicating that these pathways are key biological processes driving the prognostic differences between the high- and low-risk groups.\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eComparison of Immune Microenvironment Characteristics Between High- and Low-Risk Groups\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eThrough quantitative analysis of immune cell infiltration levels between the high- and low-risk groups, we identified significant differences in the cellular composition of the tumor immune microenvironment (TME). Specifically, the \\u003cstrong\\u003elow-risk group\\u003c/strong\\u003e exhibited significantly higher infiltration proportions of \\u003cstrong\\u003eactivated CD4⁺ T cells, gamma delta T cells (\\u003c/strong\\u003e\\u003cstrong\\u003eγδ\\u003c/strong\\u003e\\u003cstrong\\u003e\\u0026nbsp;T cells), and myeloid-derived suppressor cells (MDSCs)\\u003c/strong\\u003e compared to the high-risk group. In contrast, the \\u003cstrong\\u003ehigh-risk group\\u003c/strong\\u003e was characterized by the enrichment of \\u003cstrong\\u003eM0 macrophages and resting dendritic cells\\u003c/strong\\u003e . This disparity suggests that the risk score based on MSRGs effectively reflects the heterogeneity of the immune status in hepatocellular carcinoma (HCC).\\u003c/p\\u003e\\u003cp\\u003eFurther assessment of the overall tumor microenvironment using the \\u003cstrong\\u003eESTIMATE algorithm\\u003c/strong\\u003e revealed that the low-risk group had significantly higher \\u003cstrong\\u003eESTIMATE scores, Immune scores, and Stromal scores\\u003c/strong\\u003e than the high-risk group, while its \\u003cstrong\\u003etumor purity was significantly lower\\u003c/strong\\u003e . These results corroborate the immune cell infiltration analysis, indicating that the low-risk group possesses a richer infiltration of immune cells and stromal components, presenting an \\u003cstrong\\u003e\\\"immune-activated\\\"\\u003c/strong\\u003e microenvironment phenotype.\\u003c/p\\u003e\\u003cp\\u003eTo explore the potential regulatory role of the core genes in the risk model on the immune microenvironment, we analyzed the correlation between the expression of key MSRGs and immune cell infiltration. The results showed that the expression of \\u003cstrong\\u003eKCNQ3 and ABHD12\\u003c/strong\\u003e was significantly positively correlated with \\u003cstrong\\u003eM0 macrophage infiltration\\u003c/strong\\u003e, and negatively correlated with immune cells such as \\u003cstrong\\u003eresting mast cells\\u003c/strong\\u003e . This suggests that these genes may influence the overall immune microenvironment state by regulating the recruitment or function of specific immune cells.In the analysis of immunotherapy targets, the expression of multiple inhibitory \\u003cstrong\\u003eimmune checkpoint molecules\\u003c/strong\\u003e was significantly upregulated in the high-risk group, including \\u003cstrong\\u003eCD276 (B7-H3), HAVCR1 (TIM-3), IDO2, TNFRSF18 (GITR), TNFSF18 (GITRL), and VTCN1 (B7-H4)\\u003c/strong\\u003e. Concurrently, the expression of a series of \\u003cstrong\\u003echemokines\\u003c/strong\\u003e such as \\u003cstrong\\u003eCCL20, CCL28, CCL4, and XCL1\\u003c/strong\\u003e, which are closely associated with lymphocyte recruitment and inflammatory responses, was significantly upregulated in the low-risk group . This differential expression pattern indicates that the high-risk group may form a stronger immunosuppressive microenvironment, whereas the low-risk group demonstrates a greater capacity for immune cell recruitment.\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eGenomic Variation Characteristics and Drug Sensitivity Analysis\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eTo investigate the molecular basis of the risk stratification, we analyzed the characteristics of somatic gene alterations in the high- and low-risk groups. Overall, the \\u003cstrong\\u003efrequency of genetic alterations was higher in the high-risk group\\u003c/strong\\u003e, with alterations occurring in 144 samples (81.36%); alterations were found in 135 samples (75.42%) in the low-risk group. The alteration spectrum in both groups was predominantly characterized by \\u003cstrong\\u003emissense mutations\\u003c/strong\\u003e . In the high-risk group, alterations in genes such as \\u003cstrong\\u003eTP53 (highest mutation rate, \\u0026gt;40%), TTN, and CTNNB1\\u003c/strong\\u003e were particularly frequent. The proportion of TP53 alterations was relatively lower in the low-risk group, while alterations in genes like \\u003cstrong\\u003eALB and PCLO\\u003c/strong\\u003e were relatively more prominent . These differences suggest that different risk groups are associated with distinct patterns of driver gene alterations.\\u003c/p\\u003e\\u003cp\\u003eTo evaluate the potential value of the risk score model in guiding personalized therapy, we predicted the sensitivity (half-maximal inhibitory concentration, IC₅₀ values) of the high- and low-risk groups to four clinically relevant targeted drugs based on genomic data. The results showed:The predicted IC₅₀ values for \\u003cstrong\\u003eAxitinib\\u003c/strong\\u003e and \\u003cstrong\\u003eErlotinib\\u003c/strong\\u003e were significantly lower in the high-risk group compared to the low-risk group (p \\u0026lt; 0.0001), suggesting that high-risk tumors might be more sensitive to these two drugs.Conversely, \\u003cstrong\\u003eSorafenib\\u003c/strong\\u003e and \\u003cstrong\\u003eSunitinib\\u003c/strong\\u003e showed the opposite trend; the predicted IC₅₀ values were significantly higher in the high-risk group than in the low-risk group (p \\u0026lt; 0.0001), indicating that low-risk tumors might be more sensitive to them .These findings suggest that this risk model could potentially guide the selection of targeted drugs for patients with hepatocellular carcinoma.\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eConstruction of Single-Cell Atlas and Mapping of Risk Signatures\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eThrough \\u003cstrong\\u003esingle-cell RNA sequencing (scRNA-seq) analysis\\u003c/strong\\u003e, we constructed a cellular atlas of hepatocellular carcinoma (HCC) tissues, identifying six major cell types: \\u003cstrong\\u003eT/NK cells, myeloid cells, plasma cells, fibroblasts, epithelial cells, and B cells\\u003c/strong\\u003e (Figures 7A-C). UMAP visualization revealed that these cell populations formed distinct clusters in the reduced-dimensional space. We further mapped the MSRG-based risk score onto this atlas and found that high-risk scores were not uniformly distributed but were significantly enriched in specific cell types or subpopulations (Figures 7D-F), indicating that the risk signature exhibits \\u003cstrong\\u003ecell-type specificity\\u003c/strong\\u003e.\\u003c/p\\u003e\\u003cp\\u003eTo explore how different cell types contribute to the biological differences between risk groups, we performed \\u003cstrong\\u003edifferential expression analysis and pathway enrichment\\u003c/strong\\u003e separately for epithelial cells and T/NK cells. In epithelial cells from the high-risk group, pathways such as \\u003cstrong\\u003e\\\"Complement and coagulation cascades,\\\" \\\"Fatty acid degradation,\\\" and \\\"PPAR signaling pathway\\\"\\u003c/strong\\u003e were significantly enriched (Figure 7H), suggesting these cells may be in a state of metabolic stress or remodeling. In T/NK cells from the high-risk group, pathways including \\u003cstrong\\u003e\\\"Natural killer cell-mediated cytotoxicity,\\\" \\\"Th1/Th2/Th17 cell differentiation,\\\" and \\\"Antigen processing and presentation\\\"\\u003c/strong\\u003e were significantly enriched (Figure 7G). Notably, these pathways involved key immune signaling genes such as \\u003cstrong\\u003eFYN, PIK3R1, MAPK1, and CD247\\u003c/strong\\u003e, indicating substantial alterations in immune recognition and activation-related signaling in T/NK cells from the high-risk group.\\u003c/p\\u003e\\u003cp\\u003eThrough \\u003cstrong\\u003ecell-cell interaction analysis\\u003c/strong\\u003e, we systematically compared the communication networks between the high- and low-risk groups. Network diagrams revealed overall differences in the interaction patterns between the two groups (Figure 7J). A differential interaction heatmap further quantified these changes: compared to the low-risk group, the number of outgoing interactions from fibroblasts to other cells (e.g., epithelial cells, myeloid cells) was generally reduced in the high-risk group (Figure 7K). However, the strength of certain interactions, such as those between fibroblasts and endothelial cells, was increased (Figure 7J). Comparison of key signaling pathways showed that in the high-risk group, interactions related to immune regulation, inflammatory response, and angiogenesis—such as those mediated by \\u003cstrong\\u003eMIF-(CD74+CXCR4), MIF-(CD74+CD44), VEGFA-VEGFR2, and SPP1-CD44\\u003c/strong\\u003e—were significantly enhanced (Figure 7L). Conversely, signaling pathways involved in extracellular matrix adhesion and epithelial integrity were generally weakened in the high-risk group. These included integrin-mediated signals like \\u003cstrong\\u003eVTN-(ITGAV+ITGB1) and SPP1-(ITGAV+ITGB1)\\u003c/strong\\u003e, as well as homotypic interactions such as \\u003cstrong\\u003eOCLN-OCLN\\u003c/strong\\u003e, which help maintain cell junctions (Figure 7L).\\u003c/p\\u003e\\u003cp\\u003e\\u003cstrong\\u003eIdentification and Validation of Key Driver Genes in the Risk Model Based on SHAP Analysis\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\u003cp\\u003eTo elucidate the decision-making mechanism of the constructed risk score model and identify key features, we employed the \\u003cstrong\\u003eSHAP (SHapley Additive exPlanations)\\u0026nbsp;\\u003c/strong\\u003emethod for interpretability analysis. As shown in \\u003cstrong\\u003eFigure 8A,\\u003c/strong\\u003e the\\u003cstrong\\u003e\\u0026nbsp;SHAP feature importance plot\\u0026nbsp;\\u003c/strong\\u003edisplays the mean absolute SHAP values for each gene. The wide distribution of these values indicates significant differences in the impact of different genes on the model's output. Genes such as \\u003cstrong\\u003eFYN, KCNQ3, ENDOG, and ABHD12\\u003c/strong\\u003e exhibited large absolute SHAP values with broad distributions, and their feature values (expression levels) showed a significant positive correlation with the SHAP values. We further focused on \\u003cstrong\\u003eKCNQ3\\u003c/strong\\u003e, a gene that showed particularly significant contribution in the SHAP analysis, for in-depth validation. \\u003cstrong\\u003eThe SHAP dependence plot (Figure 8B)\\u003c/strong\\u003e clearly demonstrates a significant positive correlation between the expression level of KCNQ3 and its SHAP value. As KCNQ3 expression increases, its positive contribution to raising the risk score continuously increases, indicating that KCNQ3 is a stable and important negative predictor in the risk model.To validate its clinical significance, we stratified patients in the \\u003cstrong\\u003eTCGA-LIHC cohort\\u003c/strong\\u003e into high and low expression groups based on the median expression of KCNQ3 and performed survival analysis. The \\u003cstrong\\u003eKaplan-Meier curve (Figure 8C)\\u003c/strong\\u003e revealed that patients with high KCNQ3 expression had a significantly shorter overall survival than those with low expression (p = 0.0035). This finding, from the perspective of the \\u003cstrong\\u003eTCGA-LIHC cohort\\u003c/strong\\u003e, confirms that \\u003cstrong\\u003ehigh expression of KCNQ3 is an independent risk factor for poor prognosis in hepatocellular carcinoma patients\\u003c/strong\\u003e, establishing it as a reliable biomarker.To investigate the role of KCNQ3 in HCC progression at the functional level, we conducted in vitro loss-of-function experiments in hepatocellular carcinoma cell lines. \\u003cstrong\\u003eThe CCK-8 cell proliferation assay (Figure 8D)\\u0026nbsp;\\u003c/strong\\u003eresults showed that knocking down KCNQ3 (shKCNQ3) significantly inhibited the proliferative capacity of HCC cells compared to the control group (shCtrl) (p \\u0026lt; 0.001). The plate colony formation assay further confirmed that the number of cell colonies formed in the shKCNQ3 group was significantly lower than in the shCtrl group (p \\u0026lt; 0.01) (\\u003cstrong\\u003eFigures 8E-F\\u003c/strong\\u003e). These results collectively demonstrate that KCNQ3 significantly promotes the proliferation and clonogenic ability of hepatocellular carcinoma cells*, exhibiting oncogene-like functions in the in vitro experiments.\\u003c/p\\u003e\"},{\"header\":\"Methods\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eStudy Design and Data Collection\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis study integrated multi-omics data with experimental validation to systematically investigate prognostic models of hepatocellular carcinoma (HCC) and their underlying biological mechanisms. RNA sequencing data and clinical information from 375 HCC patients were obtained from The Cancer Genome Atlas (TCGA) database (TCGA-LIHC, https://portal.gdc.cancer.gov/). Single-cell RNA sequencing datasets (GSE149614)】 were downloaded from 【the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) for in-depth analysis of the tumor microenvironment. All statistical analyses were performed using R software (version 4.4.0), with false discovery rate (FDR) correction applied using the Benjamini-Hochberg method .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAnalysis Pipeline\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eIdentification of Molecular Subtypes\\u003c/strong\\u003e: Consensus clustering was performed using the R package \\u003cstrong\\u003eConsensusClusterPlus\\u003c/strong\\u003e. The \\u003cstrong\\u003eK-means algorithm\\u003c/strong\\u003e was applied with 1000 bootstrap iterations and a sampling ratio of 80%. The optimal number of clusters (k-value) was determined based on cumulative distribution functions and consensus matrix heatmaps, leading to the stratification of patients into distinct molecular subtypes .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eSingle-Cell Transcriptomic Analysis\\u003c/strong\\u003e: Quality control of single-cell data was conducted using the \\u003cstrong\\u003eSeurat package (v5.0.3)\\u003c/strong\\u003e. Cells with fewer than 200 genes or mitochondrial gene content exceeding 10% were filtered out. A total of 3000 highly variable genes were identified using the \\u003cstrong\\u003eFindVariableFeatures\\u003c/strong\\u003e function. Principal component analysis (PCA) was performed, and batch effects were corrected using the \\u003cstrong\\u003eHarmony\\u003c/strong\\u003e method. Cell clustering was accomplished via the \\u003cstrong\\u003eFindNeighbors\\u003c/strong\\u003e and \\u003cstrong\\u003eFindClusters\\u003c/strong\\u003e functions .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003ePrognostic Model Construction and Validation\\u003c/strong\\u003e: In the TCGA-LIHC cohort, genes significantly associated with overall survival (OS) were first screened using \\u003cstrong\\u003eunivariate Cox regression\\u003c/strong\\u003e. \\u003cstrong\\u003eLASSO regression\\u003c/strong\\u003e was then employed for variable selection and shrinkage to construct a multi-gene risk score model. Patients were divided into high-risk and low-risk groups based on the median risk score. The prognostic performance of the model was evaluated using \\u003cstrong\\u003eKaplan-Meier survival analysis\\u003c/strong\\u003e and \\u003cstrong\\u003etime-dependent receiver operating characteristic (ROC) curves\\u003c/strong\\u003e .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunctional Enrichment and Tumor Microenvironment Characterization\\u003c/strong\\u003e: Functional annotation of differentially expressed genes (DEGs) was performed through \\u003cstrong\\u003eGene Ontology (GO)\\u003c/strong\\u003e and \\u003cstrong\\u003eKyoto Encyclopedia of Genes and Genomes (KEGG)\\u003c/strong\\u003e pathway enrichment analyses using the \\u003cstrong\\u003eClusterProfiler package (v4.8.3)\\u003c/strong\\u003e. The infiltration abundance of 22 immune cell subtypes and immune/stromal scores were assessed via the \\u003cstrong\\u003eCIBERSORT\\u003c/strong\\u003e and \\u003cstrong\\u003eESTIMATE\\u003c/strong\\u003e algorithms. Correlations between risk scores and immune features were examined using \\u003cstrong\\u003eSpearman’s rank correlation\\u003c/strong\\u003e. Differences in cell-cell communication networks between risk groups were analyzed with the \\u003cstrong\\u003eCellChat package (v2.28.0)\\u003c/strong\\u003e .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eGenomic Variation and Drug Sensitivity Prediction\\u003c/strong\\u003e: Somatic mutation profiles were analyzed with the \\u003cstrong\\u003emaftools package (v2.18.0)\\u003c/strong\\u003e. Based on the GDSC2 database, the sensitivity of HCC patients to targeted therapies (half-maximal inhibitory concentration, IC₅₀) was predicted using the \\u003cstrong\\u003eoncoPredict package\\u003c/strong\\u003e, and differences between risk groups were compared .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCellular Experiments and Functional Validation\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCell Culture\\u003c/strong\\u003e: Human HCC Huh7 cell lines were cultured in \\u003cstrong\\u003eDulbecco’s Modified Eagle Medium (DMEM)\\u003c/strong\\u003e supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. Cells were maintained at 37°C in a 5% CO₂ incubator and regularly passaged to maintain logarithmic growth.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eLoss-of-Function Experiments\\u003c/strong\\u003e: Stable knockdown cell lines were generated via \\u003cstrong\\u003elentiviral transduction\\u003c/strong\\u003e. Cell proliferation viability was assessed using the \\u003cstrong\\u003eCCK-8 assay\\u003c/strong\\u003e: cells were seeded at 5×10³ cells/well in 96-well plates. After 24, 48, and 96 hours of culture, CCK-8 reagent was added, and absorbance was measured at 450 nm.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eColony Formation Assay\\u003c/strong\\u003e: Cells were plated at 1×10³ cells/well in 6-well plates and cultured for 14 days. Colonies were fixed with 4% paraformaldehyde, stained with 0.5% crystal violet, and counted (considering colonies with diameters \\u0026gt;0.5 mm as valid) to evaluate clonogenic ability .\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eStatistical Analysis\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eComparisons between groups for continuous variables were conducted using the \\u003cstrong\\u003eWilcoxon rank-sum test\\u003c/strong\\u003e. Survival differences were assessed with the \\u003cstrong\\u003elog-rank test\\u003c/strong\\u003e. \\u003cstrong\\u003eUnivariate and multivariate Cox regression models\\u003c/strong\\u003e were applied to identify independent prognostic factors. All statistical tests were two-sided, and a \\u003cstrong\\u003ep-value \\u0026lt; 0.05\\u003c/strong\\u003e was considered statistically significant.\\u003c/p\\u003e\"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eHepatocellular carcinoma (HCC) development involves multiple etiologies, including obesity, alcohol abuse, viral hepatitis, and non-alcoholic fatty liver disease (NAFLD), with most causes leading to cirrhosis before its onset [\\u003cspan citationid=\\\"CR15\\\" class=\\\"CitationRef\\\"\\u003e15\\u003c/span\\u003e]. The mechanical stiffness of the extracellular matrix (ECM) plays a crucial role in tumorigenesis and progression [\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e].Through systematic bioinformatics analysis, this study first identified 39 mechanosensitive-related genes (MSRGs) significantly associated with overall survival (OS) in hepatocellular carcinoma (HCC) (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig8\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003e). This discovery provides a crucial molecular foundation for understanding the mechanobiological mechanisms of HCC.TP53 and CTNNB1 (encoding β-catenin), as classic tumor driver genes, exhibit high mutation rates [\\u003cspan citationid=\\\"CR16\\\" class=\\\"CitationRef\\\"\\u003e16\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR17\\\" class=\\\"CitationRef\\\"\\u003e17\\u003c/span\\u003e], suggesting deep cross-talk between mechanical signaling networks and classical oncogenic pathways.Specifically, TP53 not only functions as a guardian of the genome in response to DNA damage but also senses changes in extracellular matrix stiffness, influencing cell fate decisions by regulating cytoskeletal reorganization [\\u003cspan citationid=\\\"CR18\\\" class=\\\"CitationRef\\\"\\u003e18\\u003c/span\\u003e]. CTNNB1, a core effector molecule in the Wnt/β-catenin pathway, has been demonstrated to be activated by matrix stiffness, thereby promoting stem cell-like properties and chemotherapy resistance in HCC cells [\\u003cspan citationid=\\\"CR19\\\" class=\\\"CitationRef\\\"\\u003e19\\u003c/span\\u003e].Notably, most MSRGs in this study exhibited hazard ratios (HR) greater than 1 with statistical significance. This expression-prognosis association pattern indicates that high expression of these genes often promotes tumor progression through multiple mechanisms. From a biological function perspective, these genes may participate in regulating key malignant phenotypes such as cell cycle progression, epithelial-mesenchymal transition (EMT), and angiogenesis.For instance, TTN (troponin T), a crucial component of the cytoskeleton and a primary determinant of cardiac muscle cell stiffness, may alter cellular responsiveness to mechanical stress when abnormally expressed [\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e]. MUC16, a common diagnostic marker across multiple cancer types [\\u003cspan citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e], potentially influences tumor invasiveness by regulating cell-matrix interactions.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003cp\\u003eBased on MSRG characteristics, this study employed consensus clustering to classify hepatocellular carcinoma patients into two molecular subtypes with markedly divergent prognoses. Survival analysis revealed significantly superior overall survival in Cluster 2 patients compared to Cluster 1, demonstrating the prognostic value of mechanical sensitivity signaling in HCC.In-depth analysis of the immune microenvironment characteristics of the two subtypes revealed that Cluster 2 exhibited a typical \\\"hot\\\" immune phenotype, with significant enrichment of effector immune cells such as CD8⁺ T cells in its tumor microenvironment, while Cluster 1 presented a relatively \\\"cold\\\" immune state.Recent studies have demonstrated that alterations in ECM structure and stiffness lead to CD8⁺ T cell exhaustion [\\u003cspan citationid=\\\"CR22\\\" class=\\\"CitationRef\\\"\\u003e22\\u003c/span\\u003e]. Our findings further suggest that mechanosensitive signals regulate immune cell recruitment and function. Quantitative analysis using the CIBERSORT algorithm revealed significantly elevated infiltration levels of activated CD4⁺ T cells, γδ T cells, and myeloid-derived suppressor cells in Cluster 2, indicating a more active antitumor immune response in this subtype.In contrast, Cluster 1 is characterized by enrichment of M0 macrophages and quiescent dendritic cells, indicating an immunosuppressive microenvironment. ESTIMATE algorithm further confirms that the better-prognosis Cluster 2 exhibits higher immune and stroma scores with relatively lower tumor purity, consistent with the pathological feature of extensive immune cell infiltration.\\u003c/p\\u003e \\u003cp\\u003eThe interaction between mechanical stress and the immune microenvironment is key to understanding this classification system. Physical factors within the tumor microenvironment\\u0026mdash;including extracellular matrix stiffness, solid stress, and fluid shear stress\\u0026mdash;can significantly influence immune cell behavior and function through mechanosensitive signaling pathways [\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e].In this study, high-risk group patients exhibited marked activation of ECM-receptor interaction pathways. This ECM remodeling has been reported to form physical barriers that impede effector T cell infiltration into the tumor parenchyma [\\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e], thereby contributing to the development of an immune exclusion phenotype.Key inhibitory immune checkpoint molecules including CD276, HAVCR1, IDO2, and TNFRSF18 were significantly upregulated in the high-risk group [\\u003cspan additionalcitationids=\\\"CR26\\\" citationid=\\\"CR25\\\" class=\\\"CitationRef\\\"\\u003e25\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e], suggesting that mechanical stress may promote tumor immune escape by inducing an immunosuppressive microenvironment.Notably, chemokine expression profiles also exhibit significant differences between the two subtypes. CCL20, CCL28, CCL4, and XCL1\\u0026mdash;key lymphocyte chemokines [\\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e]\\u0026mdash;show elevated expression in the low-risk group, potentially reflecting enhanced immune cell recruitment capacity.\\u003c/p\\u003e \\u003cp\\u003eTo develop a clinically valuable prognostic assessment tool, this study employed LASSO-Cox regression analysis to screen key variables from 39 mechanosensitive-related genes. The selected genes\\u0026mdash;HPN, ENDOG, UCN, FYN, ETV1, and KCNQ3\\u0026mdash;constituted the most predictive gene combination. The model's reliability was thoroughly validated through multiple statistical and clinical metrics.Kaplan-Meier survival analysis revealed significant differences in overall survival between high-risk and low-risk groups stratified by median risk score, with markedly increased mortality risk in the high-risk cohort. This finding remained consistent across both training and validation sets. Time-dependent ROC curve analysis further demonstrated the model's excellent discrimination and calibration.Notably, the risk score exhibited a unique inverse correlation with tumor pathological staging: patients with advanced disease had lower risk scores than those with early-stage disease. This phenomenon suggests the scoring system may emphasize reflecting the biological malignant potential of tumors rather than solely reflecting pathological progression stages. This prognostic value, independent of traditional TNM staging, provides a complementary biomarker for clinical risk assessment.\\u003c/p\\u003e \\u003cp\\u003eAmong the six core genes, potassium voltage-gated channel subfamily Q member 3 (KCNQ3) was identified through SHAP interpretability analysis as the key driver gene with the highest contribution to model prediction.The SHAP dependency plot clearly demonstrated a linear positive correlation between KCNQ3 expression levels and its contribution to risk scoring, indicating that high expression of this gene is a consistent factor elevating patient mortality risk. Survival analysis further confirmed that patients with high KCNQ3 expression exhibited significantly shorter overall survival, and this association was independent of other clinical and pathological parameters, establishing its reliability as a prognostic biomarker.The voltage-gated potassium channel encoded by KCNQ3 plays a central role in maintaining membrane potential homeostasis, with previous research primarily focused on neurological disorders [\\u003cspan additionalcitationids=\\\"CR30\\\" citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR31\\\" class=\\\"CitationRef\\\"\\u003e31\\u003c/span\\u003e]. In oncology, KCNQ3 has been reported to promote metastasis in esophageal adenocarcinoma and thyroid cancer [\\u003cspan citationid=\\\"CR32\\\" class=\\\"CitationRef\\\"\\u003e32\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e]. This study first reveals its oncogenic function in the development of hepatocellular carcinoma.In vitro functional experiments provide direct evidence for KCNQ3's carcinogenic mechanism. In hepatocellular carcinoma cell lines, KCNQ3 expression was significantly suppressed via short hairpin RNA-mediated gene silencing. Subsequent CCK-8 cell proliferation assays and plate clonogenic assays demonstrated markedly reduced proliferation and clonogenic capacity in the KCNQ3-knockdown group compared to controls.This phenotypic change suggests KCNQ3 may promote malignant proliferation of hepatocellular carcinoma cells by sustaining survival signaling, consistent with the functional characteristics of classic oncogenes. We hypothesize that KCNQ3 may influence calcium influx by regulating membrane potential, thereby activating downstream survival signaling pathways such as PI3K-AKT-mTOR.Furthermore, analysis of the immune microenvironment revealed a significant positive correlation between KCNQ3 expression and M0 macrophage infiltration, suggesting this gene may indirectly promote tumor progression by recruiting tumor-associated macrophages to remodel the immunosuppressive microenvironment. These hypotheses require further experimental validation.\\u003c/p\\u003e \\u003cp\\u003eHowever, this study has the following limitations: First, it is a retrospective analysis based on public databases, lacking prospective clinical validation, and the model's generalizability requires confirmation. Second, in vitro experiments only validated the function of KCNQ3, while the roles and detailed molecular mechanisms of the other core genes remain unexplored.Future research requires to conduct multicenter prospective studies to validate the model's clinical applicability; establish gene-interventional animal models to deeply analyze the molecular mechanisms of KCNQ3; and explore combined strategies of mechanical sensitivity signaling and immune checkpoint therapy.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eFunding\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis research received no external funding.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eClinical Trial Number\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eClinical trial number: not applicable.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEthics, Consent to Participate, and Consent to Publish\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eEthics, Consent to Participate, and Consent to Publish declarations: not applicable.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eData Availability Statement\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe data that support the findings of this study are available from the following public repositories:\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026bull; Bulk RNA-seq and clinical data: The Cancer Genome Atlas (TCGA) Liver Hepatocellular Carcinoma (LIHC) dataset, accessible via accession number TCGA-LIHC at https://portal.gdc.cancer.gov/.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u0026bull; Single-cell RNA-seq data: Gene Expression Omnibus (GEO) dataset GSE149614, accessible at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE149614.\\u003c/p\\u003e\\n\\u003cp\\u003eThese datasets were used for secondary data analysis as described in the Methods section. All accession numbers and repository links are also referenced in the main text and Methods section of the manuscript.\\u003c/p\\u003e\\n\\u003ch2\\u003eAuthor Contribution\\u003c/h2\\u003e\\n\\u003cp\\u003e# Author Contributions StatementJ.C. (Jin Hongjun) conceptualized the study, supervised the research design and implementation, and revised the manuscript critically for important intellectual content; P.C. (Cheng Peng) performed the bioinformatics analyses, conducted the in vitro experiments, and drafted the main manuscript text; X.Y.F. (Feng Xueyi) curated the multi-omics and clinical data, and validated the bioinformatic analysis results; X.L.H. (Huang Xianlu) prepared all the figures and tables, and assisted with the functional enrichment and statistical analyses. All authors read and approved the final version of the manuscript and agreed to the submission of the work to *Discover Oncology*.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n\\u003cli\\u003eDopazo C, S\\u0026oslash;reide K, Rangelova E, et al. Hepatocellular carcinoma. Eur J Surg Oncol. 2024;50(1):107313. doi:10.1016/j.ejso.2023.107313\\u003c/li\\u003e\\n\\u003cli\\u003eNepomnyashchaya EM, Shaposhnikov AV, Yurieva EA. Hepatocellular carcinoma: new provisions of the WHO classification, 5th edition, 2019. Archives of Pathology. 2020;82(6):36-40. doi:10.17116/patol20208206136\\u003c/li\\u003e\\n\\u003cli\\u003eFeng F, Zhao Y. Hepatocellular Carcinoma: Prevention, Diagnosis, and Treatment. Med Princ Pract. 2024;33(5):414-423. doi:10.1159/000539349\\u003c/li\\u003e\\n\\u003cli\\u003eTsui YM, Ho DW, Ng IO. Unraveling the tumor-initiating cells in hepatocellular carcinoma. Cancer Cell. 2024;42(12):1990-1993. doi:10.1016/j.ccell.2024.10.018\\u003c/li\\u003e\\n\\u003cli\\u003eSharma KK, Mohsin M, Mittal P, et al. Diagnosis of the Initial Stage of Hepatocellular Carcinoma: A Review. Current Pharmaceutical Design. 2024;30(22):1708-1724. doi:10.2174/0113816128298875240321073907\\u003c/li\\u003e\\n\\u003cli\\u003eSafri F, Nguyen R, Zerehpooshnesfchi S, George J, Qiao L. Heterogeneity of hepatocellular carcinoma: from mechanisms to clinical implications. Cancer Gene Ther. 2024;31(8):1105-1112. doi:10.1038/s41417-024-00764-w\\u003c/li\\u003e\\n\\u003cli\\u003eMa L, Hernandez MO, Zhao Y, et al. Tumor Cell Biodiversity Drives Microenvironmental Reprogramming in Liver Cancer. Cancer Cell. 2019;36(4):418-430.e6. doi:10.1016/j.ccell.2019.08.007\\u003c/li\\u003e\\n\\u003cli\\u003eTang H, Cheng X, Liang L, Chen BZ, Liu C, Wang Y. A stimulus responsive microneedle-based drug delivery system for cancer therapy. Biomater Sci. 2024;12(24):6274-6283. Published 2024 Dec 3. doi:10.1039/d4bm00741g\\u003c/li\\u003e\\n\\u003cli\\u003eRoy AM, Iyer R, Chakraborty S. The extracellular matrix in hepatocellular carcinoma: Mechanisms and therapeutic vulnerability. Cell Rep Med. 2023;4(9):101170. doi:10.1016/j.xcrm.2023.101170\\u003c/li\\u003e\\n\\u003cli\\u003eLv H, Li L, Sun M, et al. Mechanism of regulation of stem cell differentiation by matrix stiffness. Stem Cell Res Ther. 2015;6(1):103. Published 2015 May 27. doi:10.1186/s13287-015-0083-4\\u003c/li\\u003e\\n\\u003cli\\u003eMiroshnikova YA, Manet S, Li X, Wickstr\\u0026ouml;m SA, Faurobert E, Albiges-Rizo C. Calcium signaling mediates a biphasic mechanoadaptive response of endothelial cells to cyclic mechanical stretch. Mol Biol Cell. 2021;32(18):1724-1736. doi:10.1091/mbc.E21-03-0106\\u003c/li\\u003e\\n\\u003cli\\u003eAbdollahiyan P, Oroojalian F, Baradaran B, de la Guardia M, Mokhtarzadeh A. Advanced mechanotherapy: Biotensegrity for governing metastatic tumor cell fate via modulating the extracellular matrix. J Control Release. 2021;335:596-618. doi:10.1016/j.jconrel.2021.06.002\\u003c/li\\u003e\\n\\u003cli\\u003eLi M, Zhang X, Wang M, et al. Activation of Piezo1 contributes to matrix stiffness-induced angiogenesis in hepatocellular carcinoma. Cancer Commun (Lond). 2022;42(11):1162-1184. doi:10.1002/cac2.12364\\u003c/li\\u003e\\n\\u003cli\\u003eZhang J, Li J, Hou Y, et al. Osr2 functions as a biomechanical checkpoint to aggravate CD8+ T cell exhaustion in tumors. Cell. 2024;187(13):3409-3426.e24. doi:10.1016/j.cell.2024.04.023\\u003c/li\\u003e\\n\\u003cli\\u003eTarao K, Nozaki A, Ikeda T, Sato A, Komatsu H, Komatsu T, Taguri M, Tanaka K. Real impact of liver cirrhosis on the development of hepatocellular carcinoma in various liver diseases-meta-analytic assessment. Cancer Med. 2019 Mar;8(3):1054-1065. doi: 10.1002/cam4.1998. Epub 2019 Feb 21. PMID: 30791221; PMCID: PMC6434205.\\u003c/li\\u003e\\n\\u003cli\\u003eGiacomelli AO, Yang X, Lintner RE, McFarland JM, Duby M, Kim J, Howard TP, Takeda DY, Ly SH, Kim E, Gannon HS, Hurhula B, Sharpe T, Goodale A, Fritchman B, Steelman S, Vazquez F, Tsherniak A, Aguirre AJ, Doench JG, Piccioni F, Roberts CWM, Meyerson M, Getz G, Johannessen CM, Root DE, Hahn WC. Mutational processes shape the landscape of TP53 mutations in human cancer. Nat Genet. 2018 Oct;50(10):1381-1387. doi: 10.1038/s41588-018-0204-y. Epub 2018 Sep 17. PMID: 30224644; PMCID: PMC6168352.\\u003c/li\\u003e\\n\\u003cli\\u003eLi Y, Zhang F, Yang D. Comprehensive assessment and meta-analysis of the association between CTNNB1 polymorphisms and cancer risk. Biosci Rep. 2017 Nov 23;37(6):BSR20171121. doi: 10.1042/BSR20171121. PMID: 28963373; PMCID: PMC5700267.\\u003c/li\\u003e\\n\\u003cli\\u003eTornesello ML. TP53 mutations in cancer: Molecular features and therapeutic opportunities (Review). Int J Mol Med. 2025 Jan;55(1):7. doi: 10.3892/ijmm.2024.5448. Epub 2024 Oct 25. PMID: 39450536; PMCID: PMC11554381.\\u003c/li\\u003e\\n\\u003cli\\u003eYu F, Yu C, Li F, Zuo Y, Wang Y, Yao L, Wu C, Wang C, Ye L. Wnt/\\u0026beta;-catenin signaling in cancers and targeted therapies. Signal Transduct Target Ther. 2021 Aug 30;6(1):307. doi: 10.1038/s41392-021-00701-5. PMID: 34456337; PMCID: PMC8403677.\\u003c/li\\u003e\\n\\u003cli\\u003eLoescher CM, Hobbach AJ, Linke WA. Titin (TTN): from molecule to modifications, mechanics, and medical significance. Cardiovasc Res. 2022 Nov 10;118(14):2903-2918. doi: 10.1093/cvr/cvab328. PMID: 34662387; PMCID: PMC9648829.\\u003c/li\\u003e\\n\\u003cli\\u003eZhang XY, Hong LL, Ling ZQ. MUC16/CA125 in cancer: new advances. Clin Chim Acta. 2025 Jan 15;565:119981. doi: 10.1016/j.cca.2024.119981. Epub 2024 Oct 4. PMID: 39368688.\\u003c/li\\u003e\\n\\u003cli\\u003eZhang J, Li J, Hou Y, Lin Y, Zhao H, Shi Y, Chen K, Nian C, Tang J, Pan L, Xing Y, Gao H, Yang B, Song Z, Cheng Y, Liu Y, Sun M, Linghu Y, Li J, Huang H, Lai Z, Zhou Z, Li Z, Sun X, Chen Q, Su D, Li W, Peng Z, Liu P, Chen W, Huang H, Chen Y, Xiao B, Ye L, Chen L, Zhou D. Osr2 functions as a biomechanical checkpoint to aggravate CD8+ T cell exhaustion in tumors. Cell. 2024 Jun 20;187(13):3409-3426.e24. doi: 10.1016/j.cell.2024.04.023. Epub 2024 May 13. PMID: 38744281.\\u003c/li\\u003e\\n\\u003cli\\u003eRen T, Sun L, Zheng Y, Jiang Y, Guo Y, Ma J. Mechanical forces and immune cells in the tumor microenvironment: from regulation mechanisms to therapeutic strategies. Int J Surg. 2025 Aug 1;111(8):5420-5434. doi: 10.1097/JS9.0000000000002636. Epub 2025 Jun 5. PMID: 40478969.\\u003c/li\\u003e\\n\\u003cli\\u003eZhu P, Lu H, Wang M, Chen K, Chen Z, Yang L. Targeted mechanical forces enhance the effects of tumor immunotherapy by regulating immune cells in the tumor microenvironment. Cancer Biol Med. 2023 Jan 12;20(1):44\\u0026ndash;55. doi: 10.20892/j.issn.2095-3941.2022.0491. PMID: 36647779; PMCID: PMC9843446.\\u003c/li\\u003e\\n\\u003cli\\u003eLiu S, Liang J, Liu Z, Zhang C, Wang Y, Watson AH, Zhou C, Zhang F, Wu K, Zhang F, Lu Y, Wang X. The Role of CD276 in Cancers. Front Oncol. 2021 Mar 26;11:654684. doi: 10.3389/fonc.2021.654684. PMID: 33842369; PMCID: PMC8032984.\\u003c/li\\u003e\\n\\u003cli\\u003eBod L, Kye YC, Shi J, Torlai Triglia E, Schnell A, Fessler J, Ostrowski SM, Von-Franque MY, Kuchroo JR, Barilla RM, Zaghouani S, Christian E, Delorey TM, Mohib K, Xiao S, Slingerland N, Giuliano CJ, Ashenberg O, Li Z, Rothstein DM, Fisher DE, Rozenblatt-Rosen O, Sharpe AH, Quintana FJ, Apetoh L, Regev A, Kuchroo VK. B-cell-specific checkpoint molecules that regulate anti-tumor immunity. Nature. 2023 Jul;619(7969):348-356. doi: 10.1038/s41586-023-06231-0. Epub 2023 Jun 21. Erratum in: Nature. 2025 Apr;640(8057):E1. doi: 10.1038/s41586-025-08858-7. PMID: 37344597; PMCID: PMC10795478.\\u003c/li\\u003e\\n\\u003cli\\u003eJia T, Guo Y, Cheng XM, Zhou Z, Xu X, Liu H, Yang X. Multi-omics profiling identifies TNFRSF18 as a novel marker of exhausted CD8⁺ T cells and reveals tumor-immune dynamics in colorectal cancer. Clin Transl Med. 2025 Aug;15(8):e70425. doi: 10.1002/ctm2.70425. PMID: 40770837; PMCID: PMC12328248.\\u003c/li\\u003e\\n\\u003cli\\u003eYi M, Li T, Niu M, Zhang H, Wu Y, Wu K, Dai Z. Targeting cytokine and chemokine signaling pathways for cancer therapy. Signal Transduct Target Ther. 2024 Jul 22;9(1):176. doi: 10.1038/s41392-024-01868-3. PMID: 39034318; PMCID: PMC11275440.\\u003c/li\\u003e\\n\\u003cli\\u003eMiceli F, Soldovieri MV, Weckhuysen S, Cooper EC, Taglialatela M. KCNQ3-Related Disorders. 2014 May 22 [updated 2023 Sep 28]. In: Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A, editors. GeneReviews\\u0026reg; [Internet]. Seattle (WA): University of Washington, Seattle; 1993\\u0026ndash;2026. PMID: 24851285.\\u003c/li\\u003e\\n\\u003cli\\u003eWang T, Hoekzema K, Vecchio D, Wu H, Sulovari A, Coe BP, Gillentine MA, Wilfert AB, Perez-Jurado LA, Kvarnung M, Sleyp Y, Earl RK, Rosenfeld JA, Geisheker MR, Han L, Du B, Barnett C, Thompson E, Shaw M, Carroll R, Friend K, Catford R, Palmer EE, Zou X, Ou J, Li H, Guo H, Gerdts J, Avola E, Calabrese G, Elia M, Greco D, Lindstrand A, Nordgren A, Anderlid BM, Vandeweyer G, Van Dijck A, Van der Aa N, McKenna B, Hancarova M, Bendova S, Havlovicova M, Malerba G, Bernardina BD, Muglia P, van Haeringen A, Hoffer MJV, Franke B, Cappuccio G, Delatycki M, Lockhart PJ, Manning MA, Liu P, Scheffer IE, Brunetti-Pierri N, Rommelse N, Amaral DG, Santen GWE, Trabetti E, Sedl\\u0026aacute;ček Z, Michaelson JJ, Pierce K, Courchesne E, Kooy RF; SPARK Consortium; Nordenskj\\u0026ouml;ld M, Romano C, Peeters H, Bernier RA, Gecz J, Xia K, Eichler EE. Large-scale targeted sequencing identifies risk genes for neurodevelopmental disorders. Nat Commun. 2020 Oct 1;11(1):4932. doi: 10.1038/s41467-020-18723-y. Erratum in: Nat Commun. 2020 Oct 21;11(1):5398. doi: 10.1038/s41467-020-19289-5. PMID: 33004838; PMCID: PMC7530681.\\u003c/li\\u003e\\n\\u003cli\\u003eEdmond MA, Hinojo-Perez A, Efrem M, Yi-Chun L, Shams I, Hayoz S, de la Cruz A, Perez Rodriguez ME, Diaz-Solares M, Dykxhoorn DM, Luo YL, Barro-Soria R. Lipophilic compounds restore function to neurodevelopmental-associated KCNQ3 mutations. Commun Biol. 2024 Sep 19;7(1):1181. doi: 10.1038/s42003-024-06873-4. PMID: 39300259; PMCID: PMC11413209.\\u003c/li\\u003e\\n\\u003cli\\u003eFoley K, Shorthouse D, Rahrmann E, Zhuang L, Devonshire G, Gilbertson RJ; OCCAMS consortium; Fitzgerald RC, Hall BA. SMAD4 and KCNQ3 alterations are associated with lymph node metastases in oesophageal adenocarcinoma. Biochim Biophys Acta Mol Basis Dis. 2024 Jan;1870(1):166867. doi: 10.1016/j.bbadis.2023.166867. Epub 2023 Aug 28. PMID: 37648039.\\u003c/li\\u003e\\n\\u003cli\\u003eLi Q, Liu M, Song X, Xie L, Liu D, Su T, Xu Y, Li G, Liang B, Huang D. Increased KCNQ3 expression in papillary thyroid cancer promotes proliferation and migration. Cancer Cell Int. 2025 Nov 17;25(1):406. doi: 10.1186/s12935-025-04049-6. PMID: 41250218; PMCID: PMC12625505.\\u003c/li\\u003e\\n\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":false,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"discover-oncology\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"dion\",\"sideBox\":\"Learn more about [Discover Oncology](https://www.springer.com/12672)\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Discover Oncology\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"Discover Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true},\"keywords\":\"Hepatocellular carcinoma, Mechanically stimulated genes, Tumor microenvironment, Prognostic marker, KCNQ3\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-9047659/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-9047659/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003e\\u003cstrong\\u003eBackground: \\u003c/strong\\u003eHepatocellular carcinoma (HCC) is a highly prevalent malignant tumor with poor prognosis and high heterogeneity. Mechanical stimuli in the tumor microenvironment (TME) regulate HCC cell biological behavior via mechanosensitive-related genes (MSRGs), but their specific prognostic value and underlying mechanisms remain unclear.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eMethods:\\u003c/strong\\u003e RNA-seq and clinical data of HCC patients were retrieved from The Cancer Genome Atlas (TCGA) database (accession: TCGA-LIHC, https://portal.gdc.cancer.gov/), and single-cell RNA-seq datasets (accession: GSE149614) were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/). Univariate Cox regression analysis was used to screen MSRGs associated with overall survival (OS). Consensus clustering was performed to identify HCC molecular subtypes based on OS-related MSRGs. LASSO-Cox regression analysis was applied to construct an MSRG-based prognostic risk model, which was validated by Kaplan-Meier survival analysis and time-dependent ROC curves. Functional enrichment analysis, immune microenvironment characterization, genomic variation analysis and drug sensitivity prediction were conducted to explore the biological significance of the risk model. Single-cell RNA-seq analysis was used to map the risk signature to specific cell types, and SHAP analysis combined with in vitro experiments was performed to validate the key driver gene.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eResults:\\u003c/strong\\u003e A total of 39 OS-related MSRGs were identified in HCC, and HCC samples were stratified into two molecular subtypes with distinct prognostic and immune microenvironment characteristics. A six-gene prognostic risk model (HPN, ENDOG, UCN, FYN, ETV1, KCNQ3) was constructed, which exhibited good prognostic discrimination (1-, 3-, 5-year AUC: 0.74, 0.74, 0.73). High-risk patients had shorter OS, a more immunosuppressive TME, and distinct genomic alteration patterns compared with low-risk patients. The two risk groups showed differential sensitivity to clinical targeted drugs (Axitinib, Erlotinib, Sorafenib, Sunitinib). Single-cell analysis revealed cell-type specificity of the risk signature, and KCNQ3 was identified as the key driver gene via SHAP analysis. In vitro experiments confirmed that KCNQ3 Knockdown significantly inhibited the proliferation and clonogenic ability of HCC Huh7 cells.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eConclusion: \\u003c/strong\\u003eMSRGs are closely associated with the prognosis and immune microenvironment of HCC. The constructed MSRG-based prognostic risk model has reliable predictive value for HCC patient survival, and KCNQ3 may serve as a potential prognostic biomarker and therapeutic target for HCC, providing new insights for personalized treatment of HCC.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Characteristics of Mechanically Stimulated Genes in Hepatocellular Carcinoma and Their Role as Prognostic Markers\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2026-03-31 17:30:30\",\"doi\":\"10.21203/rs.3.rs-9047659/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2026-05-15T07:49:28+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2026-04-19T05:05:24+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2026-04-14T13:20:52+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2026-04-14T03:38:12+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2026-04-13T09:05:55+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"45476952471446508302307328953009352192\",\"date\":\"2026-04-08T06:13:29+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"67025386894575509118457802541618785427\",\"date\":\"2026-04-04T08:03:30+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"95582928205513569182594003104269129771\",\"date\":\"2026-04-04T03:36:06+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2026-04-04T02:46:36+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"29325297902522382653747990706378359856\",\"date\":\"2026-04-02T08:06:54+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"editorInvitedReview\",\"content\":\"\",\"date\":\"2026-03-31T04:33:36+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"216001808651637417694622850719804450285\",\"date\":\"2026-03-30T06:40:25+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewerAgreed\",\"content\":\"285445774699043517496347227014293254600\",\"date\":\"2026-03-29T12:48:15+00:00\",\"index\":\"hide\",\"fulltext\":\"\"},{\"type\":\"reviewersInvited\",\"content\":\"\",\"date\":\"2026-03-27T06:25:24+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2026-03-14T05:30:38+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2026-03-13T20:34:29+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"Discover Oncology\",\"date\":\"2026-03-13T10:24:45+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"discover-oncology\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"dion\",\"sideBox\":\"Learn more about [Discover Oncology](https://www.springer.com/12672)\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Discover Oncology\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"stoa\",\"reportingPortfolio\":\"Discover Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"149cf470-5660-41f9-8780-af11ca88bd2b\",\"owner\":[],\"postedDate\":\"March 31st, 2026\",\"published\":true,\"recentEditorialEvents\":[{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2026-05-15T07:49:28+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"in-revision\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2026-05-15T07:56:42+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2026-03-31 17:30:30\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-9047659\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-9047659\",\"identity\":\"rs-9047659\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}