Exploring Hepatocellular Carcinoma Etiology through Multi-omics Bioinformatics

Exploring Hepatocellular Carcinoma Etiology through Multi-omics Bioinformatics | 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 Exploring Hepatocellular Carcinoma Etiology through Multi-omics Bioinformatics Yimin Chen, Zhiwen Wang, Wenjie Li, Jiahui Yang, Zheyu Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6188338/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Hepatocellular carcinoma (HCC) ranks as the sixth most common cancer globally, with increasing mortality and persistent therapeutic challenges. Emerging evidence suggests that the immune microenvironment, metabolic reprogramming, and gut microbiota dysbiosis play critical roles in HCC pathogenesis, though their causal effects are unclear. This study used Mendelian randomization (MR) to systematically assess these factors' causal relationships with HCC. Methods A two-sample MR analysis, integrated with meta-analysis, examined genetic data on 731 immune cell types, 91 immune factors, 1,400 metabolites, and 412 gut microbiota phenotypes. HCC outcome data were sourced from FinnGen (discovery) and UK Biobank (replication). Five MR methods—inverse-variance weighted, weighted median, MR-Egger, weighted mode, and simple mode—were applied, with rigorous sensitivity, heterogeneity, and reverse causation analyses to ensure validity. Results In the discovery stage, causal associations with HCC were identified for 4 immune cell phenotypes (3 protective, 1 pathogenic), 2 immune factors, 57 metabolites (24 pathogenic, 33 protective), and 105 gut microbiota phenotypes (51 pathogenic, 54 protective). Replication validated the metabolite 1-stearoyl-2-linoleoyl-GPC (18:0/18:2) and 6 gut microbiota phenotypes (4 protective, 2 oncogenic). Meta-analysis confirmed 2 protective immune cell phenotypes—CD20 on CD20- CD38- B cells and CD16-CD56 on NK cells—and 9 metabolites (3 protective, 6 oncogenic) as significant causal factors. Conclusion This study establishes causal links between specific immune cells, metabolites, and gut microbiota with HCC, revealing protective and oncogenic roles. These findings highlight potential biomarkers and therapeutic targets, advancing strategies for HCC prevention and personalized treatment. Hepatocellular Carcinoma Mendelian Randomization Immune Microenvironment Metabolic Reprogramming Gut Microbiota Biomarkers Figures Figure 1 Figure 2 Figure 3 1. Introduction Hepatocellular carcinoma (HCC), recognized as the sixth most prevalent cancer globally and the second leading cause of cancer-related mortality, continues to exhibit rising incidence and mortality rates driven by environmental factors—such as hepatitis virus infection and alcoholic liver disease—and metabolic disorders, including non-alcoholic fatty liver disease and diabetes [ 1 ]. Although recent advancements in targeted therapies and immune checkpoint inhibitors have improved outcomes for some patients, challenges such as low early diagnosis rates and robust treatment resistance persist, underscoring the urgent need for novel solutions [ 2 ]. Consequently, elucidating the biological mechanisms underlying HCC development, particularly in emerging areas such as the immune microenvironment, metabolic reprogramming, and gut microbiota dysbiosis, is imperative for the development of innovative diagnostic and therapeutic strategies. The roles of immune cells, notably B cells and NK cells, in HCC remain subjects of debate, with divergent findings shaping the current understanding. Certain studies suggest that tumor-infiltrating B cells may suppress tumor progression through mechanisms like antibody production or antigen presentation [ 3 ]; however, contrasting research indicates that B cells might accelerate HCC development by secreting pro-angiogenic factors [ 4 ]. Similarly, NK cells, pivotal effector cells in innate immune surveillance, have their functionality modulated by the tumor microenvironment, though their precise causal roles in HCC necessitate further validation [ 5 ]. Beyond immune dynamics, metabolic dysregulation—particularly in amino acid and lipid metabolism—has been robustly linked to HCC pathogenesis [ 6 ]. For instance, significant abnormalities in plasma levels of N-formylglycine, oxoglutaric acid, citrulline, and heptaethylene glycol have been observed in HCC patients, yet the directionality of these associations remains ambiguous [ 7 ]. Furthermore, the emerging concept of the gut-liver axis suggests that microbiota dysbiosis may promote liver fibrosis and carcinogenesis through metabolites like deoxycholic acid [ 8 ]. Nevertheless, much of the existing research relies on animal models or cross-sectional data, lacking the causal evidence derived from population-level studies [ 9 ]. While preliminary investigations in these domains offer valuable insights, their interpretability is often constrained by small sample sizes, confounding variables, and the risk of reverse causation, all of which hinder the establishment of definitive causal links between biomarkers and HCC. To address these limitations, Mendelian Randomization (MR), a methodology that leverages genetic variants to emulate randomized controlled trials, provides a promising avenue for mitigating biases inherent in traditional observational studies, thereby offering a novel perspective for causal inference in HCC research [ 10 ]. 2. Methods 2.1Study Design Our study employed a classical two-sample Mendelian Randomization (MR) framework combined with meta-analysis to investigate causal relationships [ 11 , 12 ]. MR leverages alleles strongly associated with a trait as proxies for that trait, capitalizing on their random allocation at conception—akin to randomization in controlled trials—to infer causality. By introducing instrumental variables (IVs), specifically single nucleotide polymorphisms (SNPs), MR substitutes exposure factors to assess their potential causal effects on outcomes [ 13 ]. The exposure data encompassed 731 immune cell types, 91 peripheral blood immune factors, 1,400 peripheral blood metabolites, and 412 gut microbiota profiles. Hepatocellular carcinoma (HCC) data from the FinnGen database served as the outcome for the discovery stage, while data from the UK Biobank (UKB) were utilized for the replication stage, enabling cross-validation of findings. Additionally, results excluding the 412 gut microbiota profiles underwent meta-analysis. Primary findings were defined as positive results replicated across both databases, with significant outcomes from the meta-analysis reported as secondary results. Throughout, we adhered strictly to the three core MR assumptions: IVs must be strongly associated with the exposure, independent of confounders, and affect the outcome solely through the exposure. To ensure robustness, we rigorously applied IV selection criteria, employed five MR analytical methods, and conducted sensitivity, heterogeneity, and reverse bias analyses. The study adhered to the STROBE-MR guidelines [ 14 ], with the complete workflow depicted in Fig. 1 . 2.2Data Sources The exposure dataset comprised 731 immune cell types, 91 peripheral blood immune factors, 1,400 peripheral blood metabolites, and 412 gut microbiota profiles. Data on 731 immune cell phenotypes were derived from the SardiNIA project [ 15 ], which analyzed 731 immune traits in 3,757 Sardinians via flow cytometry, including 118 absolute cell counts, 389 fluorescence intensities of surface antigens, 32 morphological parameters, and 192 relative counts, yielding 20,143,392 SNPs and 1,688,858 indels. The 91 inflammatory factors originated from Zhao JH’s study on circulating inflammatory proteins [ 16 ], involving 14,824 participants of European ancestry. Plasma protein levels were measured using the Olink Target Inflammation panel, normalized, and subjected to inverse rank normalization, followed by genome-wide association studies (GWAS) using linear regression with an additive genetic model, adjusting for population substructure via genetic principal components. This analysis evaluated over 10,000,000 genetic variants for associations with immune-related plasma proteins. Data on 1,400 peripheral blood metabolites were sourced from Shin SY et al.’s genomic atlas of human blood metabolites [ 17 ], which examined plasma or serum from 7,824 European adults using liquid and gas chromatography coupled with tandem mass spectrometry. This study identified genome-wide significant associations at 145 metabolic loci and biochemical links to over 400 metabolites, drawing from two cohorts: the TwinsUK cohort [ 18 ], comprising females aged 18–102 from the UK Adult Twin Registry (over 14,000 volunteers, with data from > 8,500 subjects collected over 27 years), and the KORA cohort [ 19 ], a population-based epidemiological study in Augsburg, Germany, including follow-up data from KORA F4 (2006–2008) with participants aged 32–77 (mean 61) and balanced gender representation. The 412 gut microbiota profiles were derived from a Dutch microbiome prospective study [ 20 ], encompassing 7,738 Dutch participants and analyzing 207 taxonomic groups and 205 pathways reflecting microbial composition and function. This study incorporated dietary variables into multivariate linear regression alongside principal component analysis (PCA) to enhance SNP effect estimates. For the discovery stage, outcome data were obtained from a FinnGen database study [ 21 ] of 314,693 individuals, including 500 HCC cases and 314,193 controls, with over 15 million SNPs. The replication stage utilized UKB data [ 22 ] from 456,441 Europeans, comprising 123 HCC cases and 456,225 controls, with over 9 million SNPs. Phenotype codes were C3_HEPATOCELLU_CARC_EXALLC (FinnGen R10) and GCST90041897 (UKB). 2.3Instrumental Variables Selection In selecting IVs, we first excluded ambiguous SNPs with unclear sequences or palindromic structures, setting a correlation threshold between IVs and exposures at 5e-5 [ 23 ]. We calculated SNP intervals, queried minor allele frequencies (MAF), and filtered out rare variants with MAF < 0.01. To mitigate linkage disequilibrium (LD) effects on SNP independence, we established a clustering distance window of 10,000 kb and computed LD correlation (R² = [2β² * EAF * (1 − EAF)] / [2β² * EAF * (1 − EAF) + 2N * EAF * (1 − EAF) * SE²]), excluding SNPs with R² < 0.001. Weak IVs were filtered by calculating F-statistics, discarding those with F < 10. To eliminate confounding, we removed SNPs associated with confounders (e.g., smoking, alcohol) identified in the GWAS Catalog. Finally, MR-PRESSO [ 24 ] addressed pleiotropy, and the Steiger test [ 25 ] excluded reverse bias, retaining exposures with at least three valid IVs for subsequent analyses. 2.4MR Analysis Our MR analysis employed five methods: Inverse-Variance Weighted (IVW) [ 26 ], Weighted Median [ 27 ], MR-Egger [ 28 ], Weighted Mode, and Simple Mode. Horizontal pleiotropy, where SNPs influence outcomes via non-exposure pathways, was a key consideration. IVW, the most robust method, assumes no pleiotropy and combines Wald estimates from each IV via meta-analysis for stable, accurate causal inference. MR-Egger relaxes this assumption, adjusting for pleiotropy by allowing a non-zero intercept, making it the most conservative approach. Weighted Median prioritizes stronger SNPs, weighting contributions by inverse variance of outcome associations. Weighted Mode and Simple Mode supplemented primary findings. Consistency in effect direction across all five methods bolstered result robustness. Statistical power was assessed using an online tool ( https://shiny.cnsgenomics.com ) [ 29 ], based on asymptotic theory, with results reported for power > 0.7. 2.5Sensitivity, Heterogeneity, and Reverse MR Sensitivity analyses utilized MR-Egger’s intercept method, MR-PRESSO, and Leave-One-Out (LOO) approaches. The Egger intercept assessed pleiotropy via lasso regression intercept (p < 0.05 indicating potential pleiotropy), while MR-PRESSO evaluated global pleiotropy through sampling (p 0.05 and no outliers indicated low pleiotropy risk. Heterogeneity was evaluated using Cochran’s Q test and funnel plots via IVW and MR-Egger methods, with p > 0.05 suggesting low heterogeneity risk. Reverse bias was examined through reverse MR, using HCC data from FinnGen and UKB as exposures and the 731 immune cells, 91 immune factors, 1,400 metabolites, and 412 microbiota as outcomes, following identical IV selection, MR analysis, and quality control steps. Reverse MR p-values > 0.05 and Steiger test p-values < 0.05 indicated low reverse bias risk. 2.6Meta-Analysis In the absence of significant pleiotropy, MR results from discovery and replication stages were combined via meta-analysis of IVW estimates, with a significance threshold of 0.01. Significant merged results were reported as secondary outcomes. 2.7Statistics Data analysis and visualization were conducted in R (version 4.3.1). IV extraction, five MR methods, sensitivity analyses, and result computation utilized the “TwoSampleMR” (v0.5.6) [ 30 ], “MendelianRandomization” (v0.9.0), and “MR-PRESSO” (v1.0) packages. Meta-analysis and heterogeneity assessments were performed using the “meta” package, with figures generated via “ggplot2”. 3. Results 3.1Instrumental SNPs Filtration Following the instrumental variable (IV) extraction process, we incorporated a total of 2,526 phenotypes in the discovery stage, comprising 719 immune cell phenotypes (with 9,570 independent IVs), 90 peripheral blood immune factor phenotypes (with 4,423 independent IVs), 1,347 peripheral blood metabolite phenotypes (with 15,958 independent IVs), and 370 gut microbiota phenotypes (with 10,706 independent IVs). In the replication stage, we included 2,575 phenotypes, encompassing 725 immune cell phenotypes (with 14,825 independent IVs), 90 peripheral blood immune factor phenotypes (with 5,065 independent IVs), 1,351 peripheral blood metabolite phenotypes (with 27,604 independent IVs), and 409 gut microbiota phenotypes (with 20,383 independent IVs).A summary of the MR results can be found in Fig. 2 3.2Peripheral Blood Immune Cells and HCC in Discovery and Replication Stages In the discovery stage, we identified 23 positive associations, including 4 immune cell phenotypes with potential oncogenic effects on HCC and 19 with protective effects. With a statistical power threshold > 0.7, four significant results emerged, three of which were protective: BAFF-R on IgD- CD27- B cells (p = 0.006, OR = 0.810, 95% CI = 0.696–0.942, power = 0.846), BAFF-R on naive-mature B cells (p = 0.011, OR = 0.834, 95% CI = 0.726–0.959, power = 0.813), and BAFF-R on IgD + B cells (p = 0.021, OR = 0.851, 95% CI = 0.742–0.976, power = 0.712). One phenotype exhibited a pathogenic effect: CD25 on resting CD4 regulatory T cells (p = 0.006, OR = 1.739, 95% CI = 1.168–2.588, power = 0.990). In the replication stage, none of the discovery stage findings for peripheral blood immune cell phenotypes were externally validated. However, 29 positive associations were identified, including 15 phenotypes with potential oncogenic effects and 14 with protective effects. With power > 0.7, two oncogenic phenotypes emerged: CD25 on IgD- CD27- B cells (p = 0.010, OR = 1.782, 95% CI = 1.148–2.765, power = 0.897) and CD38 on CD20- B cells (p = 0.028, OR = 1.996, 95% CI = 1.076–3.703, power = 0.897). Detailed results are presented in Supplementary Tables S1.1–S1.2. 3.3Peripheral Blood Immune Factors and HCC in Discovery and Replication Stages In the discovery stage, two significant associations were detected. One protective immune factor phenotype was identified: Delta and Notch-like epidermal growth factor-related receptor levels (p = 0.048, OR = 0.571, 95% CI = 0.328–0.994). One oncogenic phenotype with power > 0.7 was found: Interleukin-1-alpha levels (p = 0.026, OR = 1.712, 95% CI = 1.066–2.748, power = 0.857). In the replication stage, the discovery stage findings were not validated. However, one oncogenic immune factor was identified: Interleukin-10 receptor subunit alpha levels (p = 0.047, OR = 2.573, 95% CI = 1.014–6.530, power = 0.918). See Supplementary Tables S3.1–S3.2 for details. 3.4Peripheral Blood Metabolites and HCC in Discovery and Replication Stages In the discovery stage, 57 positive associations were observed, including 24 metabolite phenotypes with potential oncogenic effects, such as 1-stearoyl-2-linoleoyl-GPC (18:0/18:2) levels (p = 0.037, OR = 0.684, 95% CI = 0.478–0.977), and 33 with protective effects. With power > 0.7, 21 oncogenic phenotypes were identified, including Acetoacetate, Arachidonate (20:4n6) to paraxanthine ratio, Dimethylarginine (SDMA + ADMA), Docosatrienoate (22:3n6) levels, and others (see Supplementary Table S5.1 ). In the replication stage, the discovery stage finding of 1-stearoyl-2-linoleoyl-GPC (18:0/18:2) levels was validated (p = 0.004, OR = 0.381, 95% CI = 0.199–0.729), establishing it as a robust primary result. Additionally, 53 positive associations emerged, comprising 26 oncogenic and 27 protective metabolite phenotypes. With power > 0.7, 26 oncogenic phenotypes were identified, with Palmitoylcarnitine and Ornithine levels nearing a power of 1 (see Supplementary Table S5.2 ). 3.5Gut Microbiota and HCC in Discovery and Replication Stages In the discovery stage, 105 positive associations were identified, including 51 potentially oncogenic and 54 potentially protective gut microbiota phenotypes, with 59 exhibiting power > 0.7. Six phenotypes were validated in the replication stage, constituting primary results: two oncogenic phenotypes—Gut bacterial pathway abundance (p = 0.007, OR = 1.578, 95% CI = 1.135–2.196; p = 0.010, OR = 1.861, 95% CI = 1.160–2.986)—and four protective phenotypes—Gut bacterial pathway abundance (p < 0.001, OR = 0.588, 95% CI = 0.512–0.674), Gut microbiota abundance (p < 0.001, OR = 0.451, 95% CI = 0.321–0.633), Gut bacterial pathway abundance (p < 0.001, OR = 0.440, 95% CI = 0.297–0.650), and Gut bacterial pathway abundance (p = 0.008, OR = 0.563, 95% CI = 0.369–0.859). Notably, the replication stage yielded 105 positive associations, including 47 oncogenic and 58 protective phenotypes (see Supplementary Tables S7.1–S7.2). 3.6Combined MR Results via Meta-Analysis Dual validation across the discovery and replication stages identified seven primary results: five protective phenotypes—1-stearoyl-2-linoleoyl-GPC (18:0/18:2) levels (p = 0.004, OR = 0.381, 95% CI = 0.199–0.729), three Gut bacterial pathway abundances, and one Gut microbiota abundance—and two oncogenic phenotypes—both Gut bacterial pathway abundances. Given the prevalent pleiotropy in gut microbiota, we excluded these from meta-analysis to ensure reliability. Meta-analysis of immune cell, immune factor, and metabolite phenotypes revealed two significant secondary protective immune cell phenotypes: CD20 on CD20- CD38- B cells (random p = 0.002, OR = 0.703, 95% CI = 0.564–0.876) and CD16-CD56 on Natural Killer cells (random p = 0.006, OR = 0.830, 95% CI = 0.728–0.948). For metabolites, nine significant secondary results emerged, including three protective phenotypes—Hydroxypalmitoyl sphingomyelin (random p = 0.006, OR = 0.685, 95% CI = 0.522–0.899), X-13728 levels (random p = 0.008, OR = 0.549, 95% CI = 0.352–0.856), and Oleoyl-linoleoyl-glycerol (18:1/18:2) [ 2 ] to linoleoyl-arachidonoyl-glycerol (18:2/20:4) [ 2 ] ratio (random p = 0.008, OR = 0.760, 95% CI = 0.620–0.932)—and six oncogenic phenotypes, including Acetoacetate levels (random p = 0.002, OR = 1.661, 95% CI = 1.199–2.300)(Table 1)and others (see Supplementary Table S9). No significant immune factor results emerged post-meta-analysis. Table.1 The identified multi-omics phenotypes with causal effect on HCC(Primary results via cross validation). ID p-value OR 95%OR 1-stearoyl-2-linoleoyl-gpc GCST90200037 0.004 0.381 0.199–7.289 Gut bacterial pathway abundance (COBALSYN.PWY..adenosylcobalamin.salvage.from.cobinamide.I) GCST90027455 0.007 1.578 1.135–2.196 Gut bacterial pathway abundance (PWY.6121..5.aminoimidazole.ribonucleotide.biosynthesis.I) GCST90027568 0.010 1.861 1.160–2.986 Gut bacterial pathway abundance (PWY.7209..superpathway.of.pyrimidine.ribonucleosides.degradation) GCST90027612 <0.001 0.588 0.512–0.674 Gut microbiota abundance (k_Bacteria.p_Firmicutes.c_Clostridia.o_Clostridiales.f_Lachnospiraceae.g_Roseburia.s_Roseburia_unclassified) GCST90027857 <0.001 0.451 0.321–0.633 Gut bacterial pathway abundance (COA.PWY..coenzyme.A.biosynthesis.I) GCST90027454 <0.001 0.440 0.297–0.650 Gut bacterial pathway abundance (P461.PWY..hexitol.fermentation.to.lactate..formate..ethanol.and.acetate) GCST90027503 0.008 0.563 0.369-0859 3.7Sensitivity, Heterogeneity, and Reverse MR In two-sample MR, MR-Egger and MR-PRESSO p-values excluded pleiotropy (p ≥ 0.05), and phenotypes with p < 0.05 were excluded from meta-analysis. LOO analysis yielded negative results, with no outlier SNPs detected. Cochran’s Q test p-values exceeded 0.05, indicating low heterogeneity risk. Reverse MR showed no statistical significance, with Steiger test p-values < 0.05, suggesting minimal reverse bias risk. 4. Discussion Hepatocellular carcinoma (HCC), the predominant form of liver cancer, ranks as the sixth most common cancer globally and the second leading cause of cancer-related mortality. Its incidence and associated death rates are escalating worldwide, driven by environmental factors, immunization patterns, and shifts in lifestyle [ 31 ]. Given the limitations of conventional chemotherapy in HCC management, immunotherapeutic approaches have emerged, leveraging immune cells within and beyond the tumor microenvironment (TME) to selectively target and eradicate cancer cells with high specificity and minimal side effects [ 32 ]. Liu et al. demonstrated that selective recruitment of CXCR3(+) B cells bridges proinflammatory interleukin-17 responses and the polarization of tumor-promoting macrophages in the TME, suggesting that inhibiting CXCR3(+) B cell migration or function could mitigate HCC progression [ 33 ]. Similarly, Zhang Z et al. identified CD20 + B cells, naive B cells, and CD27 + isotype-switched memory B cells as independent prognostic factors for HCC survival, noting that intratumoral B cell infiltration is markedly impaired during HCC progression, with higher densities correlating with improved clinical outcomes [ 34 ]. Conversely, B cells may exert protumorigenic effects by producing cytokines that attract myeloid-derived suppressor cells (MDSCs) and promote angiogenesis [ 35 ]. He H et al. further reported that tumor-derived CCL20 interacts with CCR6-overexpressing CD19 + CD5 + B cells, potentially enhancing angiogenesis and fueling HCC development [ 36 ]. An inflammatory TME enriched with proliferative immune cells, such as T cells and CD56 + NK cells, has been linked to improved overall survival in HCC patients [ 37 ]. NK cells, critical antitumor effectors, mediate immune surveillance by releasing perforin and granzymes to induce apoptosis in malignant cells [ 38 , 39 ]. However, NK cell activator Poly(I:C) promotes HCC in HBs-Tg mice by inducing liver inflammation and hepatocyte damage, with increased epithelial-mesenchymal transition (EMT) reliant on NK cell presence and IFN-γ playing a pivotal role in HCC development [ 40 ]. When co-cultured with sorafenib-treated macrophages, cytotoxic NK cells are activated, triggering tumor cell death; moreover, sorafenib downregulates MHC class I expression on tumor cells, potentially reducing responsiveness to immune checkpoint therapies while enhancing NK cell activity [ 41 ]. Xing et al. highlighted the diagnostic utility of autoantibodies against tumor-associated antigens (TAAs) in HCC, designing a microarray based on key TAAs that achieved 69% sensitivity and 83% specificity with a 14-TAA panel—remarkably, approximately 50% of HCC patients with normal AFP levels were detectable, underscoring the predictive value of immune factors [ 42 ]. Our study identified two secondary immune cell results with protective causal effects on HCC: CD20 on CD20- CD38- B cells (meta-p = 0.002, OR = 0.703, 95% CI = 0.564–0.876) and CD16-CD56 on NK cells (meta-p = 0.006, OR = 0.830, 95% CI = 0.728–0.948). These findings suggest that elevated peripheral blood levels of these phenotypes correlate with reduced HCC risk, aligning with existing literature and supplementing prior data. Although no primary or secondary immune factor results emerged, single-database analyses revealed two oncogenic factors (Interleukin-1-alpha and Interleukin-10 receptor subunit alpha levels) and one protective factor (Delta and Notch-like epidermal growth factor-related receptor levels), warranting further investigation. Metabolic dysregulation is a primary driver of HCC pathogenesis [ 43 , 44 ]. Even in the presence of viral infections, metabolic disorders like diabetes independently elevate HCC risk [ 35 ]. Beyond diabetes, other aberrant metabolic processes may precede HCC onset, with dysregulated metabolites detectable in peripheral blood [ 45 ]. These disruptions span complex pathways, including carbohydrate, lipid, lipid derivative, amino acid, and amino acid derivative metabolism, yielding metabolites with significantly altered circulating concentrations that could serve as biomarkers for HCC diagnosis, treatment, or prognosis [ 46 ]. Fujiogi M et al. linked 1-stearoyl-2-linoleoyl-GPC to outcome risks in a study on bronchiolitis [ 47 ]. Although research on this metabolite is limited, our dual-database validation confirmed its positive association with HCC (p = 0.004, OR = 0.381, 95% CI = 0.199–0.729), indicating that higher peripheral blood levels increase HCC likelihood, thus deepening the understanding of its role. Our meta-analysis identified eight additional positive secondary metabolite results(Supplementary Table.1), categorized as amino acids, xenobiotics, lipids, and others. The liver, responsible for over 80% of protein synthesis (e.g., albumin, growth factors, and functional peptides), oxidizes protein degradation products into CO2 and H2O for ATP production while providing carbon skeletons for new protein, sugar, and fatty acid synthesis. Its unique urea cycle processes nitrogenous waste from amino acid metabolism, which can otherwise impair cellular function. In HCC, amino acid and glutamine metabolism are dysregulated, with genes and intermediates altered by activated oncogenes (e.g., mutated Kirsten rat sarcoma 2 viral oncogene homolog), aflatoxin B1, and non-coding RNAs [ 48 ]. Our findings identified four amino acids with oncogenic causal effects: Acetoacetate (meta-p = 0.002, OR = 1.661, 95% CI = 1.199–2.300), Dihomo-linolenoylcarnitine (C20:3n3 or 6) (meta-p = 0.004, OR = 1.415, 95% CI = 1.114–1.796), Arachidonoylcarnitine (C20:4) (meta-p = 0.008, OR = 1.302, 95% CI = 1.070–1.585), and 1-stearoyl-2-docosahexaenoyl-GPC (18:0/22:6) (meta-p = 0.009, OR = 1.400, 95% CI = 1.085–1.807). These results, indicating higher peripheral blood levels correlate with increased HCC risk, align with current knowledge, offering detailed insights into specific compounds and novel perspectives for precancerous screening. Xenobiotics, exogenous chemicals not naturally metabolized by organisms, can reach toxic levels without biotransformation, which targets them for delivery to tissues or excretion. Factors mediating xenobiotic transformation influence their function and toxicity, with short-term exposure inducing metabolism-related gene expression [ 49 ]. As a central metabolic organ, the liver is particularly exposed to reactive oxygen species (ROS) during both routine metabolism and xenobiotic biotransformation, disrupting redox balance, triggering oxidative stress, and modulating inflammation and disease [ 50 , 51 ]. Our study found Hydroxypalmitoyl sphingomyelin (meta-p = 0.006, OR = 0.685, 95% CI = 0.522–0.899) to have a protective effect, while Sulfate of piperine metabolite C16H19NO3 (meta-p = 0.006, OR = 1.585, 95% CI = 1.135–2.214) exhibited an oncogenic effect, suggesting that altered xenobiotic metabolism may induce genetic changes and oxidative stress—a finding supported by our data, despite limited prior compound-specific research. Hepatic lipid dysregulation, a key driver of HCC, is associated with obesity, diabetes, and non-alcoholic steatohepatitis (NASH) [ 52 ]. Fatty acids serve as signaling precursors, energy sources, and substrates for proliferation, survival, invasion, and angiogenesis in HCC, with fatty acid oxidation complementing glycolysis to meet energy demands in nutrient-scarce tumor cores [ 53 ]. Cholesterol, vital for membrane integrity and fluidity, is essential for proliferative cancer cells, including HCC [ 54 ]. Our result for Beta-hydroxyisovalerate (meta-p = 0.003, OR = 1.575, 95% CI = 1.170–2.121) confirmed its oncogenic causal effect, aligning with and reinforcing existing research while adding robust data support. Additionally, two other metabolites—X-13728 levels (meta-p = 0.008, OR = 0.549, 95% CI = 0.352–0.856) and Oleoyl-linoleoyl-glycerol (18:1/18:2) to linoleoyl-arachidonoyl-glycerol (18:2/20:4) ratio (meta-p = 0.008, OR = 0.760, 95% CI = 0.620–0.932)—demonstrated protective causal effects post-meta-analysis. The liver, receiving nutrient-rich blood from the gut, is the primary target of gut microbiota, microbial-associated molecular patterns (MAMPs), and microbial metabolites(Fig. 3 ), with MAMPs potentially eliciting inflammation via pattern recognition receptors (PRRs). A multilayered intestinal barrier minimizes proinflammatory MAMP exposure, yet its failure in chronic liver disease (CLD), coupled with microbiota dysbiosis, drives chronic inflammation and disease progression [ 55 ], heightening HCC risk as an end-stage outcome [ 56 – 58 ]. HCC is among the few cancers with established gut microbiome involvement. Yoshimoto S et al. reported in Nature that obesity-induced microbiota alterations elevate deoxycholic acid—a microbial metabolite—promoting HCC via proinflammatory and protumorigenic modifications in hepatic stellate cells [ 59 , 60 ]. Dysbiosis acts as an infectious driver of liver disease progression [ 61 ]. In mice, high-fat diet-induced dysbiosis, marked by increased Gram-negative bacteria and a reduced Bacteroidetes-to-Firmicutes ratio, exacerbated liver injury and fibrosis when transplanted into control-diet mice post-bile duct ligation [ 62 ]. Probiotics, tested only in murine HCC models, lack human data; in DEN-induced rat HCC, VSL#3 administration mitigated dysbiosis, inflammation, and tumor growth [ 63 ]. Our study identified six primary gut microbiota results: four protective—Gut bacterial pathway abundance (PWY.7209) (p < 0.001, OR = 0.588, 95% CI = 0.512–0.674), Gut microbiota abundance (Roseburia_unclassified) (p < 0.001, OR = 0.451, 95% CI = 0.321–0.633), Gut bacterial pathway abundance (COA.PWY) (p < 0.001, OR = 0.440, 95% CI = 0.297–0.650), and Gut bacterial pathway abundance (P461.PWY) (p = 0.008, OR = 0.563, 95% CI = 0.369–0.859)—and two oncogenic—Gut bacterial pathway abundance (COBALSYN.PWY) (p = 0.007, OR = 1.578, 95% CI = 1.135–2.196) and Gut bacterial pathway abundance (PWY.6121) (p = 0.010, OR = 1.861, 95% CI = 1.160–2.986). These findings highlight the dual protective and oncogenic roles of gut microbiota in HCC, urging attention to gut ecology in liver disease patients and timely interventions for dysbiosis. This study employed Mendelian Randomization (MR) adhering to its three core principles, effectively reducing confounding and reverse causation biases. Dual-database validation bolstered result reliability, while meta-analysis elucidated total phenotypic effects. Leveraging extensive datasets, we analyzed numerous metabolite phenotypes, addressing research gaps. However, stringent IV selection criteria led to some phenotype loss, and the study lacked stratification by age or gender, focusing primarily on European populations—limiting generalizability. Future enhancements could include clinical trials for greater credibility, stratified analyses across diverse populations, and adjustments for age and gender to enhance rigor. 5. Conclusion This study employed a comprehensive two-sample Mendelian Randomization (MR) approach combined with meta-analysis to systematically explore the causal relationships between peripheral blood immune cells, immune factors, metabolites, and gut microbiota with hepatocellular carcinoma (HCC). Our findings pinpointed significant causal associations, including two protective immune cell phenotypes (CD20 on CD20- CD38- B cells and CD16-CD56 on NK cells), one validated metabolite (1-stearoyl-2-linoleoyl-GPC [18:0/18:2]), six gut microbiota phenotypes (four protective, two oncogenic), and nine additional metabolites (three protective, six oncogenic) identified through meta-analysis. These results underscore the intricate interplay of immune dynamics, metabolic reprogramming, and gut microbial ecology in HCC pathogenesis, offering robust evidence that specific peripheral blood and gut-derived factors may serve as potential biomarkers or therapeutic targets for HCC prevention and management. The identification of protective immune cell phenotypes and metabolites suggests avenues for enhancing immune surveillance and metabolic regulation to mitigate HCC risk, while the oncogenic factors highlight pathways that may exacerbate disease progression, warranting targeted inhibition. The dual role of gut microbiota—both protective and pathogenic—emphasizes the importance of maintaining gut-liver axis homeostasis, potentially through microbiota-modulating interventions. By leveraging large-scale genomic datasets and adhering to rigorous MR assumptions, this study overcomes limitations of traditional observational research, such as confounding and reverse causation, thereby strengthening the causal inference of these associations. Nevertheless, limitations remain, including the focus on European populations, which may limit generalizability, and the loss of some phenotypes due to stringent instrumental variable criteria. Future research should validate these findings in diverse populations, incorporate clinical trials to assess translational potential, and explore age- and gender-stratified effects to refine their applicability. Collectively, this work lays a foundation for integrating immune, metabolic, and microbial insights into HCC early diagnosis and personalized treatment strategies, paving the way for improved patient outcomes in this high-mortality malignancy. Declarations Ethics approval and consent to participate: Not applicable. Clinical trial number: not applicable Competing interests: The authors declare no competing interests. Funding: This work was supported by the Sichuan Provincial Science and Technology Department (Grant Number: 2023YFS0146 CXCL14). Author Contribution YC wrote the main manuscript text, ZW prepared figures 1-3, and WL and JY and ZC revised the manuscript. All authors reviewed the manuscript. References Devarbhavi H, Asrani SK, Arab JP, Nartey YA, Pose E, Kamath PS. Global burden of liver disease: 2023 update. J Hepatol. 2023 Aug;79(2):516-537. doi: 10.1016/j.jhep.2023.03.017. Epub 2023 Mar 27. PMID: 36990226. Liu X, Qin S. Immune Checkpoint Inhibitors in Hepatocellular Carcinoma: Opportunities and Challenges. Oncologist. 2019 Feb;24(Suppl 1):S3-S10. doi: 10.1634/theoncologist.2019-IO-S1-s01. PMID: 30819826; PMCID: PMC6394775. Zou J, Luo C, Xin H, Xue T, Xie X, Chen R, Zhang L. The role of tumor-infiltrating B cells in the tumor microenvironment of hepatocellular carcinoma and its prognostic value: a bioinformatics analysis. J Gastrointest Oncol. 2022 Aug;13(4):1959-1966. doi: 10.21037/jgo-22-717. PMID: 36092319; PMCID: PMC9459216. He H, Wu J, Zang M, Wang W, Chang X, Chen X, Wang R, Wu Z, Wang L, Wang D, Lu F, Sun Z, Qu C. CCR6+ B lymphocytes responding to tumor cell-derived CCL20 support hepatocellular carcinoma progression via enhancing angiogenesis. Am J Cancer Res. 2017 May 1;7(5):1151-1163. PMID: 28560063; PMCID: PMC5446480. Polidoro MA, Mikulak J, Cazzetta V, Lleo A, Mavilio D, Torzilli G, Donadon M. Tumor microenvironment in primary liver tumors: A challenging role of natural killer cells. World J Gastroenterol. 2020 Sep 7;26(33):4900-4918. doi: 10.3748/wjg.v26.i33.4900. PMID: 32952338; PMCID: PMC7476172. Mossmann D, Müller C, Park S, Ryback B, Colombi M, Ritter N, Weißenberger D, Dazert E, Coto-Llerena M, Nuciforo S, Blukacz L, Ercan C, Jimenez V, Piscuoglio S, Bosch F, Terracciano LM, Sauer U, Heim MH, Hall MN. Arginine reprograms metabolism in liver cancer via RBM39. Cell. 2023 Nov 9;186(23):5068-5083.e23. doi: 10.1016/j.cell.2023.09.011. Epub 2023 Oct 6. PMID: 37804830; PMCID: PMC10642370. Liu Z, Liu H, Chen Z, Deng C, Zhou L, Chen S, Kang J, Chen Y, He S, Zhou Z. Identification of a novel plasma metabolite panel as diagnostic biomarker for hepatocellular carcinoma. Clin Chim Acta. 2023 Mar 15;543:117302. doi: 10.1016/j.cca.2023.117302. Epub 2023 Mar 20. PMID: 36940842. Woodhouse CA, Patel VC, Singanayagam A, Shawcross DL. Review article: the gut microbiome as a therapeutic target in the pathogenesis and treatment of chronic liver disease. Aliment Pharmacol Ther. 2018 Jan;47(2):192-202. doi: 10.1111/apt.14397. Epub 2017 Oct 30. PMID: 29083037. Zhang Y, Yang FJ, Jiang QR, Gao HJ, Song X, Zhu HQ, Zhou X, Lu J. Association between gut microbiota and hepatocellular carcinoma and biliary tract cancer: A Mendelian randomization study. World J Clin Cases. 2024 Jun 26;12(18):3497-3504. doi: 10.12998/wjcc.v12.i18.3497. PMID: 38983434; PMCID: PMC11229907. Parisinos CA, Wilman HR, Thomas EL, Kelly M, Nicholls RC, McGonigle J, Neubauer S, Hingorani AD, Patel RS, Hemingway H, Bell JD, Banerjee R, Yaghootkar H. Genome-wide and Mendelian randomisation studies of liver MRI yield insights into the pathogenesis of steatohepatitis. J Hepatol. 2020 Aug;73(2):241-251. doi: 10.1016/j.jhep.2020.03.032. Epub 2020 Apr 2. Erratum in: J Hepatol. 2020 Dec;73(6):1594-1595. doi: 10.1016/j.jhep.2020.08.030. PMID: 32247823; PMCID: PMC7372222. Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925–1926. doi: 10.1001/jama.2017.17219. Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;362:k601. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014 Sep 15;23(R1):R89-98. doi: 10.1093/hmg/ddu328. Epub 2014 Jul 4. PMID: 25064373; PMCID: PMC4170722. Skrivankova VW, Richmond RC, Woolf BAR, Yarmolinsky J, Davies NM, Swanson SA, VanderWeele TJ, Higgins JPT, Timpson NJ, Dimou N, Langenberg C, Golub RM, Loder EW, Gallo V, Tybjaerg-Hansen A, Davey Smith G, Egger M, Richards JB. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA. 2021 Oct 26;326(16):1614-1621. doi: 10.1001/jama.2021.18236. PMID: 34698778. Orrù V, Steri M, Sidore C, Marongiu M, Serra V, Olla S, Sole G, Lai S, Dei M, Mulas A, Virdis F, Piras MG, Lobina M, Marongiu M, Pitzalis M, Deidda F, Loizedda A, Onano S, Zoledziewska M, Sawcer S, Devoto M, Gorospe M, Abecasis GR, Floris M, Pala M, Schlessinger D, Fiorillo E, Cucca F. Complex genetic signatures in immune cells underlie autoimmunity and inform therapy. Nat Genet. 2020 Oct;52(10):1036-1045. doi: 10.1038/s41588-020-0684-4. Epub 2020 Sep 14. Erratum in: Nat Genet. 2020 Sep 18;: PMID: 32929287; PMCID: PMC8517961. Zhao JH, Stacey D, Eriksson N, Macdonald-Dunlop E, Hedman ÅK, Kalnapenkis A, Enroth S, Cozzetto D, Digby-Bell J, Marten J, Folkersen L, Herder C, Jonsson L, Bergen SE, Gieger C, Needham EJ, Surendran P; Estonian Biobank Research Team; Paul DS, Polasek O, Thorand B, Grallert H, Roden M, Võsa U, Esko T, Hayward C, Johansson Å, Gyllensten U, Powell N, Hansson O, Mattsson-Carlgren N, Joshi PK, Danesh J, Padyukov L, Klareskog L, Landén M, Wilson JF, Siegbahn A, Wallentin L, Mälarstig A, Butterworth AS, Peters JE. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets. Nat Immunol. 2023 Sep;24(9):1540-1551. doi: 10.1038/s41590-023-01588-w. Epub 2023 Aug 10. Erratum in: Nat Immunol. 2023 Sep 7;: PMID: 37563310; PMCID: PMC10457199. Shin SY, Fauman EB, Petersen AK, Krumsiek J, Santos R, Huang J, Arnold M, Erte I, Forgetta V, Yang TP, Walter K, Menni C, Chen L, Vasquez L, Valdes AM, Hyde CL, Wang V, Ziemek D, Roberts P, Xi L, Grundberg E; Multiple Tissue Human Expression Resource (MuTHER) Consortium; Waldenberger M, Richards JB, Mohney RP, Milburn MV, John SL, Trimmer J, Theis FJ, Overington JP, Suhre K, Brosnan MJ, Gieger C, Kastenmüller G, Spector TD, Soranzo N. An atlas of genetic influences on human blood metabolites. Nat Genet. 2014 Jun;46(6):543-550. doi: 10.1038/ng.2982. Epub 2014 May 11. PMID: 24816252; PMCID: PMC4064254. Moayyeri A, Hammond CJ, Hart DJ, Spector TD. The UK Adult Twin Registry (TwinsUK Resource). Twin Res Hum Genet. 2013;16:144–149. Wichmann HE, Gieger C, Illig T. KORA-gen--resource for population genetics, controls and a broad spectrum of disease phenotypes. Gesundheitswesen. 2005;67(Suppl 1):S26–30. Lopera-Maya EA, Kurilshikov A, van der Graaf A, et al. Effect of host genetics on the gut microbiome in 7,738 participants of the Dutch Microbiome Project [published correction appears in Nat Genet. 2022 Sep;54(9):1448]. Nat Genet. 2022;54(2):143-151. doi:10.1038/s41588-021-00992-y Kurki, M.I., Karjalainen, J., Palta, P. et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613:508–518. https://doi.org/10.1038/s41586-022-05473-8 Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, Motyer A, Vukcevic D, Delaneau O, O’Connell J, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–9. Boddy S, Islam M, Moll T, Kurz J, Burrows D, McGown A, Bhargava A, Julian TH, Harvey C, Marshall JN, et al. Unbiased metabolome screen leads to personalized medicine strategy for amyotrophic lateral sclerosis. Brain Commun. 2022;4:fcac069. doi: 10.1093/braincomms/fcac069 Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50:693–8. Hemani G, Tilling K, Smith GD. Orienting the causal relationship between imprecisely measured traits using GWAS summary data. PLOS Genet. 2017;13:e1007081. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40:304–314. doi: 10.1002/gepi.21965 Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40:304–14. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44:512–25. Burgess, S.; Thompson, S.G. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32:377–389. Brion MJ, Shakhbazov K, Visscher PM. Calculating statistical power in Mendelian randomization studies. Int J Epidemiol. 2013 Oct;42(5):1497-501. doi: 10.1093/ije/dyt179. PMID: 24159078; PMCID: PMC3807619. Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife. 2018;7:e34408. doi: 10.7554/eLife.34408 Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer Incidence and Mortality Worldwide: Sources, Methods and Major Patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359–E386. doi: 10.1002/ijc.29210 Yost K. E., Satpathy A. T., Wells D. K., Qi Y., Wang C., Kageyama R., et al. Clonal Replacement of Tumor-specific T Cells Following PD-1 Blockade. Nat Med. 2019;25:1251–1259. doi: 10.1038/s41591-019-0522-3 Liu R.-X., Wei Y., Zeng Q.-H., Chan K.-W., Xiao X., Zhao X.-Y., et al. Chemokine (C-X-C Motif) Receptor 3-positive B Cells Link Interleukin-17 Inflammation to Protumorigenic Macrophage Polarization in Human Hepatocellular Carcinoma. Hepatology. 2015;62:1779–1790. doi: 10.1002/hep.28020 Zhang Z., Ma L., Goswami S., Ma J., Zheng B., Duan M., et al. Landscape of Infiltrating B Cells and Their Clinical Significance in Human Hepatocellular Carcinoma. Oncoimmunology. 2019;8:e1571388. doi: 10.1080/2162402x.2019.1571388 Tsou P., Katayama H., Ostrin E. J., Hanash S. M. The Emerging Role of B Cells in Tumor Immunity. Cancer Res. 2016;76:5597–5601. doi: 10.1158/0008-5472.can-16-0431 He H., Wu J., Zang M., Wang W., Chang X., Chen X., et al. CCR6+ B Lymphocytes Responding to Tumor Cell-Derived CCL20 Support Hepatocellular Carcinoma Progression via Enhancing Angiogenesis. Am J Cancer Res. 2017;7:1151–1163. Chew V., Tow C., Teo M., Wong H.L., Chan J., Gehring A., Loh M., Bolze A., Quek R., Lee V.K.M., et al. Inflammatory Tumour Microenvironment Is Associated with Superior Survival in Hepatocellular Carcinoma Patients. J Hepatol. 2010;52:370–379. doi: 10.1016/j.jhep.2009.07.013 Habif G., Crinier A., André P., Vivier E., Narni-Mancinelli E. Targeting Natural Killer Cells in Solid Tumors. Cell Mol Immunol. 2019;16:415–422. doi: 10.1038/s41423-019-0224-2 Guillerey C., Huntington N. D., Smyth M. J. Targeting Natural Killer Cells in Cancer Immunotherapy. Nat Immunol. 2016;17:1025–1036. doi: 10.1038/ni.3518 Chen Y., Hao X., Sun R., Wei H., Tian Z. Natural Killer Cell-Derived Interferon‐Gamma Promotes Hepatocellular Carcinoma through the Epithelial Cell Adhesion Molecule-Epithelial‐to‐Mesenchymal Transition Axis in Hepatitis B Virus Transgenic Mice. Hepatology. 2019;69:1735–1750. doi: 10.1002/hep.30317 Hage C., Hoves S., Strauss L., Bissinger S., Prinz Y., Pöschinger T., et al. Sorafenib Induces Pyroptosis in Macrophages and Triggers Natural Killer Cell-Mediated Cytotoxicity against Hepatocellular Carcinoma. Hepatology. 2019;70:1280–1297. doi: 10.1002/hep.30666 Xing M., Wang X., Kirken R., He L., Zhang J.-Y. Immunodiagnostic Biomarkers for Hepatocellular Carcinoma (HCC): The First Step in Detection and Treatment. Int J Mol Sci. 2021;22:6139. doi: 10.3390/ijms22116139 Wang P, Kang D, Cao W, Wang Y, Liu Z. Diabetes mellitus and risk of hepatocellular carcinoma: a systematic review and meta-analysis. Diabetes Metab Res Rev. 2012;28:109–122. doi: 10.1002/dmrr.1291 Mantovani A, Targher G. Type 2 diabetes mellitus and risk of hepatocellular carcinoma: spotlight on nonalcoholic fatty liver disease. Ann Transl Med. 2017;5:270. doi: 10.21037/atm.2017.04.41 Huang T-S, et al. Diabetes, hepatocellular carcinoma, and mortality in hepatitis C-infected patients: a population-based cohort study. J Gastroenter Hepatol. 2017;32:1355–1362. doi: 10.1111/jgh.13670 Xie D, Zhang G, Ma Y, Wu D, Jiang S, Zhou S, Jiang X. Circulating Metabolic Markers Related to the Diagnosis of Hepatocellular Carcinoma. J Oncol. 2022 Dec 9;2022:7840606. doi: 10.1155/2022/7840606. PMID: 36532884; PMCID: PMC9757943. Fujiogi M, Camargo CA Jr, Raita Y, Bochkov YA, Gern JE, Mansbach JM, Piedra PA, Hasegawa K. Respiratory viruses are associated with serum metabolome among infants hospitalized for bronchiolitis: A multicenter study. Pediatr Allergy Immunol. 2020 Oct;31(7):755-766. doi: 10.1111/pai.13296. Epub 2020 Jun 10. PMID: 32460384; PMCID: PMC7704725. Du D, Liu C, Qin M, Zhang X, Xi T, Yuan S, Hao H, Xiong J. Metabolic dysregulation and emerging therapeutical targets for hepatocellular carcinoma. Acta Pharm Sin B. 2022 Feb;12(2):558-580. doi: 10.1016/j.apsb.2021.09.019. Epub 2021 Sep 25. PMID: 35256934; PMCID: PMC8897153. Maurice CF, Haiser HJ, Turnbaugh PJ. Xenobiotics shape the physiology and gene expression of the active human gut microbiome. Cell. 2013;152(1–2):39–50. Björkholm B, Bok CM, Lundin A, Rafter J, Hibberd ML, Pettersson S. Intestinal microbiota regulate xenobiotic metabolism in the liver. PLoS One. 2009 Sep 9;4(9):e6958. doi: 10.1371/journal.pone.0006958. PMID: 19742318; PMCID: PMC2734986. Allameh A, Niayesh-Mehr R, Aliarab A, Sebastiani G, Pantopoulos K. Oxidative Stress in Liver Pathophysiology and Disease. Antioxidants (Basel). 2023 Aug 22;12(9):1653. doi: 10.3390/antiox12091653. PMID: 37759956; PMCID: PMC10525124. Grohmann M, Wiede F, Dodd GT, Gurzov EN, Ooi GJ, Butt T, et al. Obesity drives STAT-1-dependent NASH and STAT-3-dependent HCC. Cell. 2018;175:1289–1306.e20. Wang M, Han J, Xing H, Zhang H, Li Z, Liang L, et al. Dysregulated fatty acid metabolism in hepatocellular carcinoma. Hepat Oncol. 2016;3:241–251. Riscal R, Skuli N, Simon MC. Even cancer cells watch their cholesterol! Mol Cell. 2019;76:220–231. Schnabl B & Brenner DA. Interactions between the intestinal microbiome and liver diseases. Gastroenterology. 2014;146:1513–1524. doi: 10.1053/j.gastro.2014.01.020 Dapito DH et al. Promotion of hepatocellular carcinoma by the intestinal microbiota and TLR4. Cancer Cell. 2012;21:504–516. doi: 10.1016/j.ccr.2012.02.007 Yu LX et al. Endotoxin accumulation prevents carcinogen-induced apoptosis and promotes liver tumorigenesis in rodents. Hepatology (Baltimore, Md.). 2010;52:1322–1333. doi: 10.1002/hep.23845 Yoshimoto S et al. Obesity-induced gut microbial metabolite promotes liver cancer through senescence secretome. Nature. 2013;499:97–101. doi: 10.1038/nature12347 Yoshimoto S, Loo TM, Atarashi K, Kanda H, Sato S, Oyadomari S, Iwakura Y, Oshima K, Morita H, Hattori M, Honda K, Ishikawa Y, Hara E, Ohtani N. Obesity-induced gut microbial metabolite promotes liver cancer through senescence secretome. Nature. 2013;499:97–101. Coppé J-P, Desprez P-Y, Krtolica A, Campisi J. The senescence-associated secretory phenotype: The dark side of tumor suppression. Annu Rev Pathol. 2010;5:99–118. Henao-Mejia J et al. Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity. Nature. 2012;482:179–185. doi: 10.1038/nature10809 De Minicis S et al. Dysbiosis contributes to fibrogenesis in the course of chronic liver injury in mice. Hepatology (Baltimore, Md.). 2014;59:1738–1749. doi: 10.1002/hep.26695 Zhang HL et al. Profound impact of gut homeostasis on chemically-induced pro-tumorigenic inflammation and hepatocarcinogenesis in rats. J Hepatol. 2012;57:803–812. doi: 10.1016/j.jhep.2012.06.011 Additional Declarations No competing interests reported. Supplementary Files SupplementaryTable.1metasecondaryresult.xlsx SupplementaryTable.S3.Mainresultsofimmuefactors.xlsx SupplementaryTable.S7.Mainresultsofmicrobiota.xlsx SupplementaryTable.S9.Metaresultsmultiomics.xlsx SupplementaryTable.S1.Mainresultsofimmuecells.xlsx SupplementaryTable.S5.Mainresultsofmetabolites.xls Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6188338","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":445371568,"identity":"6ad99496-dd57-4f20-a8eb-01f3ed222d1c","order_by":0,"name":"Yimin Chen","email":"","orcid":"","institution":"2st Affiliated Hospital of Kunming Medical University","correspondingAuthor":false,"prefix":"","firstName":"Yimin","middleName":"","lastName":"Chen","suffix":""},{"id":445371569,"identity":"de495a4a-4ed6-467e-8701-d63ed7e73fc4","order_by":1,"name":"Zhiwen Wang","email":"","orcid":"","institution":"Chongqing medical university","correspondingAuthor":false,"prefix":"","firstName":"Zhiwen","middleName":"","lastName":"Wang","suffix":""},{"id":445371570,"identity":"7b164f7b-683d-4d71-b26c-fb50a1dcbeaa","order_by":2,"name":"Wenjie Li","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Wenjie","middleName":"","lastName":"Li","suffix":""},{"id":445371572,"identity":"f560722c-6af1-4d62-bc0f-b74c9c3459c0","order_by":3,"name":"Jiahui Yang","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Jiahui","middleName":"","lastName":"Yang","suffix":""},{"id":445371575,"identity":"2b771eab-a565-4663-90a9-23660dead860","order_by":4,"name":"Zheyu Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYDACCSBmbAAS7A1gPmMD8Vp4DpCsRSKBSC38s5uPPfy5wyZxw83nDz/zMNjIbjjA/OwBXkvuHEs3kDyTlrjhdo6xNA9DmvGGA2zmBvi0GEjkmEkYth0GaWFj5mEAMg7wsEng15L/TSKx7T/QYcefAbX8J0ZLDpvEwbYDiRtuMJgBtRwgrEXiRpqZZGNbsvHMMznGknMMgIzDbGZ4tfDPSH4m+bPNTrbv+PGHH95UgBjNz/BqgQHHBog7gZiZGPVAYE+kulEwCkbBKBiJAADYB0qf3wTe2wAAAABJRU5ErkJggg==","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":true,"prefix":"","firstName":"Zheyu","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2025-03-09 11:23:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6188338/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6188338/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":81039654,"identity":"182aa8d6-2fc5-4a9b-bc97-8f7395dcd55b","added_by":"auto","created_at":"2025-04-21 13:14:59","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4566311,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow of Mendelian Randomization Analysis for HCC\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/4dc8bfacef060d642c10d52b.jpg"},{"id":81039659,"identity":"b7612b38-a2db-4025-ad52-14f3c437a7d9","added_by":"auto","created_at":"2025-04-21 13:15:00","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":727387,"visible":true,"origin":"","legend":"\u003cp\u003eSummary of Causal Associations in HCC MR Analysis\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/5016f3f2f5682139cfdfad1d.jpg"},{"id":81039873,"identity":"b296d1c5-3a42-4af8-901f-47f5b27c6116","added_by":"auto","created_at":"2025-04-21 13:23:00","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":11810283,"visible":true,"origin":"","legend":"\u003cp\u003eInterplay of Metabolic, Immune, and Gut-Liver Axis Pathways in HCC Pathogenesis\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/f3282038a1adc3728f084c9e.png"},{"id":83985269,"identity":"8ddeb99d-3bb7-45ef-8e0f-545fa33cb2b0","added_by":"auto","created_at":"2025-06-05 11:02:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":17295756,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/4355dc02-9a1b-449a-96bc-423feef7ca56.pdf"},{"id":81039656,"identity":"0c219afb-d121-4eb6-9899-11a88aa4cdee","added_by":"auto","created_at":"2025-04-21 13:14:59","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":11416,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable.1metasecondaryresult.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/747e0c0d0d380df5cce118d6.xlsx"},{"id":81041273,"identity":"7b047f34-9bf9-47cd-b4d1-ec6d83a42cc3","added_by":"auto","created_at":"2025-04-21 13:39:00","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":63540,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable.S3.Mainresultsofimmuefactors.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/ad255b52d1919b68e9f28b5b.xlsx"},{"id":81041274,"identity":"cca62234-bb0d-4544-8f3e-5cddbdc2fa1b","added_by":"auto","created_at":"2025-04-21 13:39:00","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":240947,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable.S7.Mainresultsofmicrobiota.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/d2e3386ae7239318a48e6f8b.xlsx"},{"id":81039671,"identity":"73555d7b-16c4-4c07-8f93-70531bcc3309","added_by":"auto","created_at":"2025-04-21 13:15:00","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":214591,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable.S9.Metaresultsmultiomics.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/0f1b116e00f5072c3a5756dc.xlsx"},{"id":81039678,"identity":"fff0f4f5-8cb1-4131-abf5-69e5e0d1b66f","added_by":"auto","created_at":"2025-04-21 13:15:00","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":421839,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable.S1.Mainresultsofimmuecells.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/ca0e9aee17f34e2c7616ae08.xlsx"},{"id":81040552,"identity":"f6967e79-77a2-400d-9b41-33e3f9052776","added_by":"auto","created_at":"2025-04-21 13:31:00","extension":"xls","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":1830912,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTable.S5.Mainresultsofmetabolites.xls","url":"https://assets-eu.researchsquare.com/files/rs-6188338/v1/f9a32eb2189571df3c2d7548.xls"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploring Hepatocellular Carcinoma Etiology through Multi-omics Bioinformatics","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eHepatocellular carcinoma (HCC), recognized as the sixth most prevalent cancer globally and the second leading cause of cancer-related mortality, continues to exhibit rising incidence and mortality rates driven by environmental factors\u0026mdash;such as hepatitis virus infection and alcoholic liver disease\u0026mdash;and metabolic disorders, including non-alcoholic fatty liver disease and diabetes [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Although recent advancements in targeted therapies and immune checkpoint inhibitors have improved outcomes for some patients, challenges such as low early diagnosis rates and robust treatment resistance persist, underscoring the urgent need for novel solutions [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Consequently, elucidating the biological mechanisms underlying HCC development, particularly in emerging areas such as the immune microenvironment, metabolic reprogramming, and gut microbiota dysbiosis, is imperative for the development of innovative diagnostic and therapeutic strategies.\u003c/p\u003e \u003cp\u003eThe roles of immune cells, notably B cells and NK cells, in HCC remain subjects of debate, with divergent findings shaping the current understanding. Certain studies suggest that tumor-infiltrating B cells may suppress tumor progression through mechanisms like antibody production or antigen presentation [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]; however, contrasting research indicates that B cells might accelerate HCC development by secreting pro-angiogenic factors [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Similarly, NK cells, pivotal effector cells in innate immune surveillance, have their functionality modulated by the tumor microenvironment, though their precise causal roles in HCC necessitate further validation [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Beyond immune dynamics, metabolic dysregulation\u0026mdash;particularly in amino acid and lipid metabolism\u0026mdash;has been robustly linked to HCC pathogenesis [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. For instance, significant abnormalities in plasma levels of N-formylglycine, oxoglutaric acid, citrulline, and heptaethylene glycol have been observed in HCC patients, yet the directionality of these associations remains ambiguous [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFurthermore, the emerging concept of the gut-liver axis suggests that microbiota dysbiosis may promote liver fibrosis and carcinogenesis through metabolites like deoxycholic acid [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Nevertheless, much of the existing research relies on animal models or cross-sectional data, lacking the causal evidence derived from population-level studies [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. While preliminary investigations in these domains offer valuable insights, their interpretability is often constrained by small sample sizes, confounding variables, and the risk of reverse causation, all of which hinder the establishment of definitive causal links between biomarkers and HCC. To address these limitations, Mendelian Randomization (MR), a methodology that leverages genetic variants to emulate randomized controlled trials, provides a promising avenue for mitigating biases inherent in traditional observational studies, thereby offering a novel perspective for causal inference in HCC research [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1Study Design\u003c/h2\u003e \u003cp\u003eOur study employed a classical two-sample Mendelian Randomization (MR) framework combined with meta-analysis to investigate causal relationships [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. MR leverages alleles strongly associated with a trait as proxies for that trait, capitalizing on their random allocation at conception\u0026mdash;akin to randomization in controlled trials\u0026mdash;to infer causality. By introducing instrumental variables (IVs), specifically single nucleotide polymorphisms (SNPs), MR substitutes exposure factors to assess their potential causal effects on outcomes [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. The exposure data encompassed 731 immune cell types, 91 peripheral blood immune factors, 1,400 peripheral blood metabolites, and 412 gut microbiota profiles. Hepatocellular carcinoma (HCC) data from the FinnGen database served as the outcome for the discovery stage, while data from the UK Biobank (UKB) were utilized for the replication stage, enabling cross-validation of findings. Additionally, results excluding the 412 gut microbiota profiles underwent meta-analysis. Primary findings were defined as positive results replicated across both databases, with significant outcomes from the meta-analysis reported as secondary results. Throughout, we adhered strictly to the three core MR assumptions: IVs must be strongly associated with the exposure, independent of confounders, and affect the outcome solely through the exposure. To ensure robustness, we rigorously applied IV selection criteria, employed five MR analytical methods, and conducted sensitivity, heterogeneity, and reverse bias analyses. The study adhered to the STROBE-MR guidelines [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], with the complete workflow depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2Data Sources\u003c/h2\u003e \u003cp\u003eThe exposure dataset comprised 731 immune cell types, 91 peripheral blood immune factors, 1,400 peripheral blood metabolites, and 412 gut microbiota profiles. Data on 731 immune cell phenotypes were derived from the SardiNIA project [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], which analyzed 731 immune traits in 3,757 Sardinians via flow cytometry, including 118 absolute cell counts, 389 fluorescence intensities of surface antigens, 32 morphological parameters, and 192 relative counts, yielding 20,143,392 SNPs and 1,688,858 indels. The 91 inflammatory factors originated from Zhao JH\u0026rsquo;s study on circulating inflammatory proteins [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], involving 14,824 participants of European ancestry. Plasma protein levels were measured using the Olink Target Inflammation panel, normalized, and subjected to inverse rank normalization, followed by genome-wide association studies (GWAS) using linear regression with an additive genetic model, adjusting for population substructure via genetic principal components. This analysis evaluated over 10,000,000 genetic variants for associations with immune-related plasma proteins. Data on 1,400 peripheral blood metabolites were sourced from Shin SY et al.\u0026rsquo;s genomic atlas of human blood metabolites [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], which examined plasma or serum from 7,824 European adults using liquid and gas chromatography coupled with tandem mass spectrometry. This study identified genome-wide significant associations at 145 metabolic loci and biochemical links to over 400 metabolites, drawing from two cohorts: the TwinsUK cohort [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], comprising females aged 18\u0026ndash;102 from the UK Adult Twin Registry (over 14,000 volunteers, with data from \u0026gt;\u0026thinsp;8,500 subjects collected over 27 years), and the KORA cohort [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], a population-based epidemiological study in Augsburg, Germany, including follow-up data from KORA F4 (2006\u0026ndash;2008) with participants aged 32\u0026ndash;77 (mean 61) and balanced gender representation. The 412 gut microbiota profiles were derived from a Dutch microbiome prospective study [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], encompassing 7,738 Dutch participants and analyzing 207 taxonomic groups and 205 pathways reflecting microbial composition and function. This study incorporated dietary variables into multivariate linear regression alongside principal component analysis (PCA) to enhance SNP effect estimates.\u003c/p\u003e \u003cp\u003eFor the discovery stage, outcome data were obtained from a FinnGen database study [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] of 314,693 individuals, including 500 HCC cases and 314,193 controls, with over 15\u0026nbsp;million SNPs. The replication stage utilized UKB data [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] from 456,441 Europeans, comprising 123 HCC cases and 456,225 controls, with over 9\u0026nbsp;million SNPs. Phenotype codes were C3_HEPATOCELLU_CARC_EXALLC (FinnGen R10) and GCST90041897 (UKB).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3Instrumental Variables Selection\u003c/h2\u003e \u003cp\u003eIn selecting IVs, we first excluded ambiguous SNPs with unclear sequences or palindromic structures, setting a correlation threshold between IVs and exposures at 5e-5 [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. We calculated SNP intervals, queried minor allele frequencies (MAF), and filtered out rare variants with MAF\u0026thinsp;\u0026lt;\u0026thinsp;0.01. To mitigate linkage disequilibrium (LD) effects on SNP independence, we established a clustering distance window of 10,000 kb and computed LD correlation (R\u0026sup2; = [2β\u0026sup2; * EAF * (1\u0026thinsp;\u0026minus;\u0026thinsp;EAF)] / [2β\u0026sup2; * EAF * (1\u0026thinsp;\u0026minus;\u0026thinsp;EAF)\u0026thinsp;+\u0026thinsp;2N * EAF * (1\u0026thinsp;\u0026minus;\u0026thinsp;EAF) * SE\u0026sup2;]), excluding SNPs with R\u0026sup2; \u0026lt; 0.001. Weak IVs were filtered by calculating F-statistics, discarding those with F\u0026thinsp;\u0026lt;\u0026thinsp;10. To eliminate confounding, we removed SNPs associated with confounders (e.g., smoking, alcohol) identified in the GWAS Catalog. Finally, MR-PRESSO [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] addressed pleiotropy, and the Steiger test [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] excluded reverse bias, retaining exposures with at least three valid IVs for subsequent analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4MR Analysis\u003c/h2\u003e \u003cp\u003eOur MR analysis employed five methods: Inverse-Variance Weighted (IVW) [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], Weighted Median [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e], MR-Egger [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], Weighted Mode, and Simple Mode. Horizontal pleiotropy, where SNPs influence outcomes via non-exposure pathways, was a key consideration. IVW, the most robust method, assumes no pleiotropy and combines Wald estimates from each IV via meta-analysis for stable, accurate causal inference. MR-Egger relaxes this assumption, adjusting for pleiotropy by allowing a non-zero intercept, making it the most conservative approach. Weighted Median prioritizes stronger SNPs, weighting contributions by inverse variance of outcome associations. Weighted Mode and Simple Mode supplemented primary findings. Consistency in effect direction across all five methods bolstered result robustness. Statistical power was assessed using an online tool (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://shiny.cnsgenomics.com\u003c/span\u003e\u003cspan address=\"https://shiny.cnsgenomics.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e], based on asymptotic theory, with results reported for power\u0026thinsp;\u0026gt;\u0026thinsp;0.7.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5Sensitivity, Heterogeneity, and Reverse MR\u003c/h2\u003e \u003cp\u003eSensitivity analyses utilized MR-Egger\u0026rsquo;s intercept method, MR-PRESSO, and Leave-One-Out (LOO) approaches. The Egger intercept assessed pleiotropy via lasso regression intercept (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 indicating potential pleiotropy), while MR-PRESSO evaluated global pleiotropy through sampling (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 suggesting pleiotropy). LOO analysis iteratively excluded each SNP to assess individual impacts, visualized via forest plots; p\u0026thinsp;\u0026gt;\u0026thinsp;0.05 and no outliers indicated low pleiotropy risk. Heterogeneity was evaluated using Cochran\u0026rsquo;s Q test and funnel plots via IVW and MR-Egger methods, with p\u0026thinsp;\u0026gt;\u0026thinsp;0.05 suggesting low heterogeneity risk. Reverse bias was examined through reverse MR, using HCC data from FinnGen and UKB as exposures and the 731 immune cells, 91 immune factors, 1,400 metabolites, and 412 microbiota as outcomes, following identical IV selection, MR analysis, and quality control steps. Reverse MR p-values\u0026thinsp;\u0026gt;\u0026thinsp;0.05 and Steiger test p-values\u0026thinsp;\u0026lt;\u0026thinsp;0.05 indicated low reverse bias risk.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6Meta-Analysis\u003c/h2\u003e \u003cp\u003eIn the absence of significant pleiotropy, MR results from discovery and replication stages were combined via meta-analysis of IVW estimates, with a significance threshold of 0.01. Significant merged results were reported as secondary outcomes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7Statistics\u003c/h2\u003e \u003cp\u003eData analysis and visualization were conducted in R (version 4.3.1). IV extraction, five MR methods, sensitivity analyses, and result computation utilized the \u0026ldquo;TwoSampleMR\u0026rdquo; (v0.5.6) [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], \u0026ldquo;MendelianRandomization\u0026rdquo; (v0.9.0), and \u0026ldquo;MR-PRESSO\u0026rdquo; (v1.0) packages. Meta-analysis and heterogeneity assessments were performed using the \u0026ldquo;meta\u0026rdquo; package, with figures generated via \u0026ldquo;ggplot2\u0026rdquo;.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.1Instrumental SNPs Filtration\u003c/h2\u003e \u003cp\u003eFollowing the instrumental variable (IV) extraction process, we incorporated a total of 2,526 phenotypes in the discovery stage, comprising 719 immune cell phenotypes (with 9,570 independent IVs), 90 peripheral blood immune factor phenotypes (with 4,423 independent IVs), 1,347 peripheral blood metabolite phenotypes (with 15,958 independent IVs), and 370 gut microbiota phenotypes (with 10,706 independent IVs). In the replication stage, we included 2,575 phenotypes, encompassing 725 immune cell phenotypes (with 14,825 independent IVs), 90 peripheral blood immune factor phenotypes (with 5,065 independent IVs), 1,351 peripheral blood metabolite phenotypes (with 27,604 independent IVs), and 409 gut microbiota phenotypes (with 20,383 independent IVs).A summary of the MR results can be found in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.2Peripheral Blood Immune Cells and HCC in Discovery and Replication Stages\u003c/h2\u003e \u003cp\u003eIn the discovery stage, we identified 23 positive associations, including 4 immune cell phenotypes with potential oncogenic effects on HCC and 19 with protective effects. With a statistical power threshold\u0026thinsp;\u0026gt;\u0026thinsp;0.7, four significant results emerged, three of which were protective: BAFF-R on IgD- CD27- B cells (p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;0.810, 95% CI\u0026thinsp;=\u0026thinsp;0.696\u0026ndash;0.942, power\u0026thinsp;=\u0026thinsp;0.846), BAFF-R on naive-mature B cells (p\u0026thinsp;=\u0026thinsp;0.011, OR\u0026thinsp;=\u0026thinsp;0.834, 95% CI\u0026thinsp;=\u0026thinsp;0.726\u0026ndash;0.959, power\u0026thinsp;=\u0026thinsp;0.813), and BAFF-R on IgD\u0026thinsp;+\u0026thinsp;B cells (p\u0026thinsp;=\u0026thinsp;0.021, OR\u0026thinsp;=\u0026thinsp;0.851, 95% CI\u0026thinsp;=\u0026thinsp;0.742\u0026ndash;0.976, power\u0026thinsp;=\u0026thinsp;0.712). One phenotype exhibited a pathogenic effect: CD25 on resting CD4 regulatory T cells (p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;1.739, 95% CI\u0026thinsp;=\u0026thinsp;1.168\u0026ndash;2.588, power\u0026thinsp;=\u0026thinsp;0.990).\u003c/p\u003e \u003cp\u003eIn the replication stage, none of the discovery stage findings for peripheral blood immune cell phenotypes were externally validated. However, 29 positive associations were identified, including 15 phenotypes with potential oncogenic effects and 14 with protective effects. With power\u0026thinsp;\u0026gt;\u0026thinsp;0.7, two oncogenic phenotypes emerged: CD25 on IgD- CD27- B cells (p\u0026thinsp;=\u0026thinsp;0.010, OR\u0026thinsp;=\u0026thinsp;1.782, 95% CI\u0026thinsp;=\u0026thinsp;1.148\u0026ndash;2.765, power\u0026thinsp;=\u0026thinsp;0.897) and CD38 on CD20- B cells (p\u0026thinsp;=\u0026thinsp;0.028, OR\u0026thinsp;=\u0026thinsp;1.996, 95% CI\u0026thinsp;=\u0026thinsp;1.076\u0026ndash;3.703, power\u0026thinsp;=\u0026thinsp;0.897). Detailed results are presented in Supplementary Tables S1.1\u0026ndash;S1.2.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.3Peripheral Blood Immune Factors and HCC in Discovery and Replication Stages\u003c/h2\u003e \u003cp\u003eIn the discovery stage, two significant associations were detected. One protective immune factor phenotype was identified: Delta and Notch-like epidermal growth factor-related receptor levels (p\u0026thinsp;=\u0026thinsp;0.048, OR\u0026thinsp;=\u0026thinsp;0.571, 95% CI\u0026thinsp;=\u0026thinsp;0.328\u0026ndash;0.994). One oncogenic phenotype with power\u0026thinsp;\u0026gt;\u0026thinsp;0.7 was found: Interleukin-1-alpha levels (p\u0026thinsp;=\u0026thinsp;0.026, OR\u0026thinsp;=\u0026thinsp;1.712, 95% CI\u0026thinsp;=\u0026thinsp;1.066\u0026ndash;2.748, power\u0026thinsp;=\u0026thinsp;0.857). In the replication stage, the discovery stage findings were not validated. However, one oncogenic immune factor was identified: Interleukin-10 receptor subunit alpha levels (p\u0026thinsp;=\u0026thinsp;0.047, OR\u0026thinsp;=\u0026thinsp;2.573, 95% CI\u0026thinsp;=\u0026thinsp;1.014\u0026ndash;6.530, power\u0026thinsp;=\u0026thinsp;0.918). See Supplementary Tables S3.1\u0026ndash;S3.2 for details.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.4Peripheral Blood Metabolites and HCC in Discovery and Replication Stages\u003c/h2\u003e \u003cp\u003eIn the discovery stage, 57 positive associations were observed, including 24 metabolite phenotypes with potential oncogenic effects, such as 1-stearoyl-2-linoleoyl-GPC (18:0/18:2) levels (p\u0026thinsp;=\u0026thinsp;0.037, OR\u0026thinsp;=\u0026thinsp;0.684, 95% CI\u0026thinsp;=\u0026thinsp;0.478\u0026ndash;0.977), and 33 with protective effects. With power\u0026thinsp;\u0026gt;\u0026thinsp;0.7, 21 oncogenic phenotypes were identified, including Acetoacetate, Arachidonate (20:4n6) to paraxanthine ratio, Dimethylarginine (SDMA\u0026thinsp;+\u0026thinsp;ADMA), Docosatrienoate (22:3n6) levels, and others (see Supplementary Table \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5.1\u003c/span\u003e). In the replication stage, the discovery stage finding of 1-stearoyl-2-linoleoyl-GPC (18:0/18:2) levels was validated (p\u0026thinsp;=\u0026thinsp;0.004, OR\u0026thinsp;=\u0026thinsp;0.381, 95% CI\u0026thinsp;=\u0026thinsp;0.199\u0026ndash;0.729), establishing it as a robust primary result. Additionally, 53 positive associations emerged, comprising 26 oncogenic and 27 protective metabolite phenotypes. With power\u0026thinsp;\u0026gt;\u0026thinsp;0.7, 26 oncogenic phenotypes were identified, with Palmitoylcarnitine and Ornithine levels nearing a power of 1 (see Supplementary Table \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5.2\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e3.5Gut Microbiota and HCC in Discovery and Replication Stages\u003c/h2\u003e \u003cp\u003eIn the discovery stage, 105 positive associations were identified, including 51 potentially oncogenic and 54 potentially protective gut microbiota phenotypes, with 59 exhibiting power\u0026thinsp;\u0026gt;\u0026thinsp;0.7. Six phenotypes were validated in the replication stage, constituting primary results: two oncogenic phenotypes\u0026mdash;Gut bacterial pathway abundance (p\u0026thinsp;=\u0026thinsp;0.007, OR\u0026thinsp;=\u0026thinsp;1.578, 95% CI\u0026thinsp;=\u0026thinsp;1.135\u0026ndash;2.196; p\u0026thinsp;=\u0026thinsp;0.010, OR\u0026thinsp;=\u0026thinsp;1.861, 95% CI\u0026thinsp;=\u0026thinsp;1.160\u0026ndash;2.986)\u0026mdash;and four protective phenotypes\u0026mdash;Gut bacterial pathway abundance (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, OR\u0026thinsp;=\u0026thinsp;0.588, 95% CI\u0026thinsp;=\u0026thinsp;0.512\u0026ndash;0.674), Gut microbiota abundance (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, OR\u0026thinsp;=\u0026thinsp;0.451, 95% CI\u0026thinsp;=\u0026thinsp;0.321\u0026ndash;0.633), Gut bacterial pathway abundance (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, OR\u0026thinsp;=\u0026thinsp;0.440, 95% CI\u0026thinsp;=\u0026thinsp;0.297\u0026ndash;0.650), and Gut bacterial pathway abundance (p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;0.563, 95% CI\u0026thinsp;=\u0026thinsp;0.369\u0026ndash;0.859). Notably, the replication stage yielded 105 positive associations, including 47 oncogenic and 58 protective phenotypes (see Supplementary Tables S7.1\u0026ndash;S7.2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e3.6Combined MR Results via Meta-Analysis\u003c/h2\u003e \u003cp\u003eDual validation across the discovery and replication stages identified seven primary results: five protective phenotypes\u0026mdash;1-stearoyl-2-linoleoyl-GPC (18:0/18:2) levels (p\u0026thinsp;=\u0026thinsp;0.004, OR\u0026thinsp;=\u0026thinsp;0.381, 95% CI\u0026thinsp;=\u0026thinsp;0.199\u0026ndash;0.729), three Gut bacterial pathway abundances, and one Gut microbiota abundance\u0026mdash;and two oncogenic phenotypes\u0026mdash;both Gut bacterial pathway abundances. Given the prevalent pleiotropy in gut microbiota, we excluded these from meta-analysis to ensure reliability. Meta-analysis of immune cell, immune factor, and metabolite phenotypes revealed two significant secondary protective immune cell phenotypes: CD20 on CD20- CD38- B cells (random p\u0026thinsp;=\u0026thinsp;0.002, OR\u0026thinsp;=\u0026thinsp;0.703, 95% CI\u0026thinsp;=\u0026thinsp;0.564\u0026ndash;0.876) and CD16-CD56 on Natural Killer cells (random p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;0.830, 95% CI\u0026thinsp;=\u0026thinsp;0.728\u0026ndash;0.948). For metabolites, nine significant secondary results emerged, including three protective phenotypes\u0026mdash;Hydroxypalmitoyl sphingomyelin (random p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;0.685, 95% CI\u0026thinsp;=\u0026thinsp;0.522\u0026ndash;0.899), X-13728 levels (random p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;0.549, 95% CI\u0026thinsp;=\u0026thinsp;0.352\u0026ndash;0.856), and Oleoyl-linoleoyl-glycerol (18:1/18:2) [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] to linoleoyl-arachidonoyl-glycerol (18:2/20:4) [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] ratio (random p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;0.760, 95% CI\u0026thinsp;=\u0026thinsp;0.620\u0026ndash;0.932)\u0026mdash;and six oncogenic phenotypes, including Acetoacetate levels (random p\u0026thinsp;=\u0026thinsp;0.002, OR\u0026thinsp;=\u0026thinsp;1.661, 95% CI\u0026thinsp;=\u0026thinsp;1.199\u0026ndash;2.300)(Table\u0026nbsp;1)and others (see Supplementary Table S9). No significant immune factor results emerged post-meta-analysis.\u003c/p\u003e \u003cp\u003eTable.1 The identified multi-omics phenotypes with causal effect on HCC(Primary results via cross validation).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eID\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOR\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e95%OR\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1-stearoyl-2-linoleoyl-gpc\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90200037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.381\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.199\u0026ndash;7.289\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGut bacterial pathway abundance\u003c/p\u003e \u003cp\u003e(COBALSYN.PWY..adenosylcobalamin.salvage.from.cobinamide.I)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90027455\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.007\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.578\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.135\u0026ndash;2.196\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGut bacterial pathway abundance\u003c/p\u003e \u003cp\u003e(PWY.6121..5.aminoimidazole.ribonucleotide.biosynthesis.I)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90027568\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.861\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1.160\u0026ndash;2.986\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGut bacterial pathway abundance\u003c/p\u003e \u003cp\u003e(PWY.7209..superpathway.of.pyrimidine.ribonucleosides.degradation)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90027612\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.588\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.512\u0026ndash;0.674\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGut microbiota abundance\u003c/p\u003e \u003cp\u003e(k_Bacteria.p_Firmicutes.c_Clostridia.o_Clostridiales.f_Lachnospiraceae.g_Roseburia.s_Roseburia_unclassified)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90027857\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.451\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.321\u0026ndash;0.633\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGut bacterial pathway abundance\u003c/p\u003e \u003cp\u003e(COA.PWY..coenzyme.A.biosynthesis.I)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90027454\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.440\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.297\u0026ndash;0.650\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGut bacterial pathway abundance\u003c/p\u003e \u003cp\u003e(P461.PWY..hexitol.fermentation.to.lactate..formate..ethanol.and.acetate)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGCST90027503\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.008\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.563\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.369-0859\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.7Sensitivity, Heterogeneity, and Reverse MR\u003c/h2\u003e \u003cp\u003eIn two-sample MR, MR-Egger and MR-PRESSO p-values excluded pleiotropy (p\u0026thinsp;\u0026ge;\u0026thinsp;0.05), and phenotypes with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were excluded from meta-analysis. LOO analysis yielded negative results, with no outlier SNPs detected. Cochran\u0026rsquo;s Q test p-values exceeded 0.05, indicating low heterogeneity risk. Reverse MR showed no statistical significance, with Steiger test p-values\u0026thinsp;\u0026lt;\u0026thinsp;0.05, suggesting minimal reverse bias risk.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eHepatocellular carcinoma (HCC), the predominant form of liver cancer, ranks as the sixth most common cancer globally and the second leading cause of cancer-related mortality. Its incidence and associated death rates are escalating worldwide, driven by environmental factors, immunization patterns, and shifts in lifestyle [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Given the limitations of conventional chemotherapy in HCC management, immunotherapeutic approaches have emerged, leveraging immune cells within and beyond the tumor microenvironment (TME) to selectively target and eradicate cancer cells with high specificity and minimal side effects [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Liu et al. demonstrated that selective recruitment of CXCR3(+) B cells bridges proinflammatory interleukin-17 responses and the polarization of tumor-promoting macrophages in the TME, suggesting that inhibiting CXCR3(+) B cell migration or function could mitigate HCC progression [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Similarly, Zhang Z et al. identified CD20\u0026thinsp;+\u0026thinsp;B cells, naive B cells, and CD27\u0026thinsp;+\u0026thinsp;isotype-switched memory B cells as independent prognostic factors for HCC survival, noting that intratumoral B cell infiltration is markedly impaired during HCC progression, with higher densities correlating with improved clinical outcomes [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Conversely, B cells may exert protumorigenic effects by producing cytokines that attract myeloid-derived suppressor cells (MDSCs) and promote angiogenesis [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. He H et al. further reported that tumor-derived CCL20 interacts with CCR6-overexpressing CD19\u0026thinsp;+\u0026thinsp;CD5\u0026thinsp;+\u0026thinsp;B cells, potentially enhancing angiogenesis and fueling HCC development [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. An inflammatory TME enriched with proliferative immune cells, such as T cells and CD56\u0026thinsp;+\u0026thinsp;NK cells, has been linked to improved overall survival in HCC patients [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. NK cells, critical antitumor effectors, mediate immune surveillance by releasing perforin and granzymes to induce apoptosis in malignant cells [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. However, NK cell activator Poly(I:C) promotes HCC in HBs-Tg mice by inducing liver inflammation and hepatocyte damage, with increased epithelial-mesenchymal transition (EMT) reliant on NK cell presence and IFN-γ playing a pivotal role in HCC development [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. When co-cultured with sorafenib-treated macrophages, cytotoxic NK cells are activated, triggering tumor cell death; moreover, sorafenib downregulates MHC class I expression on tumor cells, potentially reducing responsiveness to immune checkpoint therapies while enhancing NK cell activity [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Xing et al. highlighted the diagnostic utility of autoantibodies against tumor-associated antigens (TAAs) in HCC, designing a microarray based on key TAAs that achieved 69% sensitivity and 83% specificity with a 14-TAA panel\u0026mdash;remarkably, approximately 50% of HCC patients with normal AFP levels were detectable, underscoring the predictive value of immune factors [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Our study identified two secondary immune cell results with protective causal effects on HCC: CD20 on CD20- CD38- B cells (meta-p\u0026thinsp;=\u0026thinsp;0.002, OR\u0026thinsp;=\u0026thinsp;0.703, 95% CI\u0026thinsp;=\u0026thinsp;0.564\u0026ndash;0.876) and CD16-CD56 on NK cells (meta-p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;0.830, 95% CI\u0026thinsp;=\u0026thinsp;0.728\u0026ndash;0.948). These findings suggest that elevated peripheral blood levels of these phenotypes correlate with reduced HCC risk, aligning with existing literature and supplementing prior data. Although no primary or secondary immune factor results emerged, single-database analyses revealed two oncogenic factors (Interleukin-1-alpha and Interleukin-10 receptor subunit alpha levels) and one protective factor (Delta and Notch-like epidermal growth factor-related receptor levels), warranting further investigation.\u003c/p\u003e \u003cp\u003eMetabolic dysregulation is a primary driver of HCC pathogenesis [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Even in the presence of viral infections, metabolic disorders like diabetes independently elevate HCC risk [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Beyond diabetes, other aberrant metabolic processes may precede HCC onset, with dysregulated metabolites detectable in peripheral blood [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. These disruptions span complex pathways, including carbohydrate, lipid, lipid derivative, amino acid, and amino acid derivative metabolism, yielding metabolites with significantly altered circulating concentrations that could serve as biomarkers for HCC diagnosis, treatment, or prognosis [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Fujiogi M et al. linked 1-stearoyl-2-linoleoyl-GPC to outcome risks in a study on bronchiolitis [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. Although research on this metabolite is limited, our dual-database validation confirmed its positive association with HCC (p\u0026thinsp;=\u0026thinsp;0.004, OR\u0026thinsp;=\u0026thinsp;0.381, 95% CI\u0026thinsp;=\u0026thinsp;0.199\u0026ndash;0.729), indicating that higher peripheral blood levels increase HCC likelihood, thus deepening the understanding of its role. Our meta-analysis identified eight additional positive secondary metabolite results(Supplementary Table.1), categorized as amino acids, xenobiotics, lipids, and others. The liver, responsible for over 80% of protein synthesis (e.g., albumin, growth factors, and functional peptides), oxidizes protein degradation products into CO2 and H2O for ATP production while providing carbon skeletons for new protein, sugar, and fatty acid synthesis. Its unique urea cycle processes nitrogenous waste from amino acid metabolism, which can otherwise impair cellular function. In HCC, amino acid and glutamine metabolism are dysregulated, with genes and intermediates altered by activated oncogenes (e.g., mutated Kirsten rat sarcoma 2 viral oncogene homolog), aflatoxin B1, and non-coding RNAs [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. Our findings identified four amino acids with oncogenic causal effects: Acetoacetate (meta-p\u0026thinsp;=\u0026thinsp;0.002, OR\u0026thinsp;=\u0026thinsp;1.661, 95% CI\u0026thinsp;=\u0026thinsp;1.199\u0026ndash;2.300), Dihomo-linolenoylcarnitine (C20:3n3 or 6) (meta-p\u0026thinsp;=\u0026thinsp;0.004, OR\u0026thinsp;=\u0026thinsp;1.415, 95% CI\u0026thinsp;=\u0026thinsp;1.114\u0026ndash;1.796), Arachidonoylcarnitine (C20:4) (meta-p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;1.302, 95% CI\u0026thinsp;=\u0026thinsp;1.070\u0026ndash;1.585), and 1-stearoyl-2-docosahexaenoyl-GPC (18:0/22:6) (meta-p\u0026thinsp;=\u0026thinsp;0.009, OR\u0026thinsp;=\u0026thinsp;1.400, 95% CI\u0026thinsp;=\u0026thinsp;1.085\u0026ndash;1.807). These results, indicating higher peripheral blood levels correlate with increased HCC risk, align with current knowledge, offering detailed insights into specific compounds and novel perspectives for precancerous screening. Xenobiotics, exogenous chemicals not naturally metabolized by organisms, can reach toxic levels without biotransformation, which targets them for delivery to tissues or excretion. Factors mediating xenobiotic transformation influence their function and toxicity, with short-term exposure inducing metabolism-related gene expression [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. As a central metabolic organ, the liver is particularly exposed to reactive oxygen species (ROS) during both routine metabolism and xenobiotic biotransformation, disrupting redox balance, triggering oxidative stress, and modulating inflammation and disease [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Our study found Hydroxypalmitoyl sphingomyelin (meta-p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;0.685, 95% CI\u0026thinsp;=\u0026thinsp;0.522\u0026ndash;0.899) to have a protective effect, while Sulfate of piperine metabolite C16H19NO3 (meta-p\u0026thinsp;=\u0026thinsp;0.006, OR\u0026thinsp;=\u0026thinsp;1.585, 95% CI\u0026thinsp;=\u0026thinsp;1.135\u0026ndash;2.214) exhibited an oncogenic effect, suggesting that altered xenobiotic metabolism may induce genetic changes and oxidative stress\u0026mdash;a finding supported by our data, despite limited prior compound-specific research. Hepatic lipid dysregulation, a key driver of HCC, is associated with obesity, diabetes, and non-alcoholic steatohepatitis (NASH) [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. Fatty acids serve as signaling precursors, energy sources, and substrates for proliferation, survival, invasion, and angiogenesis in HCC, with fatty acid oxidation complementing glycolysis to meet energy demands in nutrient-scarce tumor cores [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. Cholesterol, vital for membrane integrity and fluidity, is essential for proliferative cancer cells, including HCC [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. Our result for Beta-hydroxyisovalerate (meta-p\u0026thinsp;=\u0026thinsp;0.003, OR\u0026thinsp;=\u0026thinsp;1.575, 95% CI\u0026thinsp;=\u0026thinsp;1.170\u0026ndash;2.121) confirmed its oncogenic causal effect, aligning with and reinforcing existing research while adding robust data support. Additionally, two other metabolites\u0026mdash;X-13728 levels (meta-p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;0.549, 95% CI\u0026thinsp;=\u0026thinsp;0.352\u0026ndash;0.856) and Oleoyl-linoleoyl-glycerol (18:1/18:2) to linoleoyl-arachidonoyl-glycerol (18:2/20:4) ratio (meta-p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;0.760, 95% CI\u0026thinsp;=\u0026thinsp;0.620\u0026ndash;0.932)\u0026mdash;demonstrated protective causal effects post-meta-analysis.\u003c/p\u003e \u003cp\u003eThe liver, receiving nutrient-rich blood from the gut, is the primary target of gut microbiota, microbial-associated molecular patterns (MAMPs), and microbial metabolites(Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), with MAMPs potentially eliciting inflammation via pattern recognition receptors (PRRs). A multilayered intestinal barrier minimizes proinflammatory MAMP exposure, yet its failure in chronic liver disease (CLD), coupled with microbiota dysbiosis, drives chronic inflammation and disease progression [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e], heightening HCC risk as an end-stage outcome [\u003cspan additionalcitationids=\"CR57\" citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. HCC is among the few cancers with established gut microbiome involvement. Yoshimoto S et al. reported in Nature that obesity-induced microbiota alterations elevate deoxycholic acid\u0026mdash;a microbial metabolite\u0026mdash;promoting HCC via proinflammatory and protumorigenic modifications in hepatic stellate cells [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e]. Dysbiosis acts as an infectious driver of liver disease progression [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e]. In mice, high-fat diet-induced dysbiosis, marked by increased Gram-negative bacteria and a reduced Bacteroidetes-to-Firmicutes ratio, exacerbated liver injury and fibrosis when transplanted into control-diet mice post-bile duct ligation [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. Probiotics, tested only in murine HCC models, lack human data; in DEN-induced rat HCC, VSL#3 administration mitigated dysbiosis, inflammation, and tumor growth [\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e]. Our study identified six primary gut microbiota results: four protective\u0026mdash;Gut bacterial pathway abundance (PWY.7209) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, OR\u0026thinsp;=\u0026thinsp;0.588, 95% CI\u0026thinsp;=\u0026thinsp;0.512\u0026ndash;0.674), Gut microbiota abundance (Roseburia_unclassified) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, OR\u0026thinsp;=\u0026thinsp;0.451, 95% CI\u0026thinsp;=\u0026thinsp;0.321\u0026ndash;0.633), Gut bacterial pathway abundance (COA.PWY) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, OR\u0026thinsp;=\u0026thinsp;0.440, 95% CI\u0026thinsp;=\u0026thinsp;0.297\u0026ndash;0.650), and Gut bacterial pathway abundance (P461.PWY) (p\u0026thinsp;=\u0026thinsp;0.008, OR\u0026thinsp;=\u0026thinsp;0.563, 95% CI\u0026thinsp;=\u0026thinsp;0.369\u0026ndash;0.859)\u0026mdash;and two oncogenic\u0026mdash;Gut bacterial pathway abundance (COBALSYN.PWY) (p\u0026thinsp;=\u0026thinsp;0.007, OR\u0026thinsp;=\u0026thinsp;1.578, 95% CI\u0026thinsp;=\u0026thinsp;1.135\u0026ndash;2.196) and Gut bacterial pathway abundance (PWY.6121) (p\u0026thinsp;=\u0026thinsp;0.010, OR\u0026thinsp;=\u0026thinsp;1.861, 95% CI\u0026thinsp;=\u0026thinsp;1.160\u0026ndash;2.986). These findings highlight the dual protective and oncogenic roles of gut microbiota in HCC, urging attention to gut ecology in liver disease patients and timely interventions for dysbiosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThis study employed Mendelian Randomization (MR) adhering to its three core principles, effectively reducing confounding and reverse causation biases. Dual-database validation bolstered result reliability, while meta-analysis elucidated total phenotypic effects. Leveraging extensive datasets, we analyzed numerous metabolite phenotypes, addressing research gaps. However, stringent IV selection criteria led to some phenotype loss, and the study lacked stratification by age or gender, focusing primarily on European populations\u0026mdash;limiting generalizability. Future enhancements could include clinical trials for greater credibility, stratified analyses across diverse populations, and adjustments for age and gender to enhance rigor.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study employed a comprehensive two-sample Mendelian Randomization (MR) approach combined with meta-analysis to systematically explore the causal relationships between peripheral blood immune cells, immune factors, metabolites, and gut microbiota with hepatocellular carcinoma (HCC). Our findings pinpointed significant causal associations, including two protective immune cell phenotypes (CD20 on CD20- CD38- B cells and CD16-CD56 on NK cells), one validated metabolite (1-stearoyl-2-linoleoyl-GPC [18:0/18:2]), six gut microbiota phenotypes (four protective, two oncogenic), and nine additional metabolites (three protective, six oncogenic) identified through meta-analysis. These results underscore the intricate interplay of immune dynamics, metabolic reprogramming, and gut microbial ecology in HCC pathogenesis, offering robust evidence that specific peripheral blood and gut-derived factors may serve as potential biomarkers or therapeutic targets for HCC prevention and management.\u003c/p\u003e \u003cp\u003eThe identification of protective immune cell phenotypes and metabolites suggests avenues for enhancing immune surveillance and metabolic regulation to mitigate HCC risk, while the oncogenic factors highlight pathways that may exacerbate disease progression, warranting targeted inhibition. The dual role of gut microbiota\u0026mdash;both protective and pathogenic\u0026mdash;emphasizes the importance of maintaining gut-liver axis homeostasis, potentially through microbiota-modulating interventions. By leveraging large-scale genomic datasets and adhering to rigorous MR assumptions, this study overcomes limitations of traditional observational research, such as confounding and reverse causation, thereby strengthening the causal inference of these associations.\u003c/p\u003e \u003cp\u003eNevertheless, limitations remain, including the focus on European populations, which may limit generalizability, and the loss of some phenotypes due to stringent instrumental variable criteria. Future research should validate these findings in diverse populations, incorporate clinical trials to assess translational potential, and explore age- and gender-stratified effects to refine their applicability. Collectively, this work lays a foundation for integrating immune, metabolic, and microbial insights into HCC early diagnosis and personalized treatment strategies, paving the way for improved patient outcomes in this high-mortality malignancy.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics approval and consent to participate: Not applicable.\u003c/p\u003e\n\u003cp\u003eClinical trial number: not applicable\u003c/p\u003e\n\u003cp\u003eCompeting interests: The authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003eFunding: This work was supported by the Sichuan Provincial Science and Technology Department (Grant Number: 2023YFS0146 CXCL14).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eYC wrote the main manuscript text, ZW prepared figures 1-3, and WL and JY and ZC revised the manuscript. All authors reviewed the manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eDevarbhavi H, Asrani SK, Arab JP, Nartey YA, Pose E, Kamath PS. Global burden of liver disease: 2023 update. J Hepatol. 2023 Aug;79(2):516-537. doi: 10.1016/j.jhep.2023.03.017. Epub 2023 Mar 27. PMID: 36990226.\u003c/li\u003e\n\u003cli\u003eLiu X, Qin S. Immune Checkpoint Inhibitors in Hepatocellular Carcinoma: Opportunities and Challenges. Oncologist. 2019 Feb;24(Suppl 1):S3-S10. doi: 10.1634/theoncologist.2019-IO-S1-s01. PMID: 30819826; PMCID: PMC6394775.\u003c/li\u003e\n\u003cli\u003eZou J, Luo C, Xin H, Xue T, Xie X, Chen R, Zhang L. The role of tumor-infiltrating B cells in the tumor microenvironment of hepatocellular carcinoma and its prognostic value: a bioinformatics analysis. J Gastrointest Oncol. 2022 Aug;13(4):1959-1966. doi: 10.21037/jgo-22-717. PMID: 36092319; PMCID: PMC9459216.\u003c/li\u003e\n\u003cli\u003eHe H, Wu J, Zang M, Wang W, Chang X, Chen X, Wang R, Wu Z, Wang L, Wang D, Lu F, Sun Z, Qu C. CCR6+ B lymphocytes responding to tumor cell-derived CCL20 support hepatocellular carcinoma progression via enhancing angiogenesis. Am J Cancer Res. 2017 May 1;7(5):1151-1163. PMID: 28560063; PMCID: PMC5446480.\u003c/li\u003e\n\u003cli\u003ePolidoro MA, Mikulak J, Cazzetta V, Lleo A, Mavilio D, Torzilli G, Donadon M. Tumor microenvironment in primary liver tumors: A challenging role of natural killer cells. World J Gastroenterol. 2020 Sep 7;26(33):4900-4918. doi: 10.3748/wjg.v26.i33.4900. PMID: 32952338; PMCID: PMC7476172.\u003c/li\u003e\n\u003cli\u003eMossmann D, M\u0026uuml;ller C, Park S, Ryback B, Colombi M, Ritter N, Wei\u0026szlig;enberger D, Dazert E, Coto-Llerena M, Nuciforo S, Blukacz L, Ercan C, Jimenez V, Piscuoglio S, Bosch F, Terracciano LM, Sauer U, Heim MH, Hall MN. Arginine reprograms metabolism in liver cancer via RBM39. Cell. 2023 Nov 9;186(23):5068-5083.e23. doi: 10.1016/j.cell.2023.09.011. Epub 2023 Oct 6. PMID: 37804830; PMCID: PMC10642370.\u003c/li\u003e\n\u003cli\u003eLiu Z, Liu H, Chen Z, Deng C, Zhou L, Chen S, Kang J, Chen Y, He S, Zhou Z. Identification of a novel plasma metabolite panel as diagnostic biomarker for hepatocellular carcinoma. Clin Chim Acta. 2023 Mar 15;543:117302. doi: 10.1016/j.cca.2023.117302. Epub 2023 Mar 20. PMID: 36940842.\u003c/li\u003e\n\u003cli\u003eWoodhouse CA, Patel VC, Singanayagam A, Shawcross DL. Review article: the gut microbiome as a therapeutic target in the pathogenesis and treatment of chronic liver disease. Aliment Pharmacol Ther. 2018 Jan;47(2):192-202. doi: 10.1111/apt.14397. Epub 2017 Oct 30. PMID: 29083037.\u003c/li\u003e\n\u003cli\u003eZhang Y, Yang FJ, Jiang QR, Gao HJ, Song X, Zhu HQ, Zhou X, Lu J. Association between gut microbiota and hepatocellular carcinoma and biliary tract cancer: A Mendelian randomization study. World J Clin Cases. 2024 Jun 26;12(18):3497-3504. doi: 10.12998/wjcc.v12.i18.3497. PMID: 38983434; PMCID: PMC11229907.\u003c/li\u003e\n\u003cli\u003eParisinos CA, Wilman HR, Thomas EL, Kelly M, Nicholls RC, McGonigle J, Neubauer S, Hingorani AD, Patel RS, Hemingway H, Bell JD, Banerjee R, Yaghootkar H. Genome-wide and Mendelian randomisation studies of liver MRI yield insights into the pathogenesis of steatohepatitis. J Hepatol. 2020 Aug;73(2):241-251. doi: 10.1016/j.jhep.2020.03.032. Epub 2020 Apr 2. Erratum in: J Hepatol. 2020 Dec;73(6):1594-1595. doi: 10.1016/j.jhep.2020.08.030. PMID: 32247823; PMCID: PMC7372222.\u003c/li\u003e\n\u003cli\u003eEmdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925\u0026ndash;1926. doi: 10.1001/jama.2017.17219.\u003c/li\u003e\n\u003cli\u003eDavies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;362:k601.\u003c/li\u003e\n\u003cli\u003eDavey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014 Sep 15;23(R1):R89-98. doi: 10.1093/hmg/ddu328. Epub 2014 Jul 4. PMID: 25064373; PMCID: PMC4170722.\u003c/li\u003e\n\u003cli\u003eSkrivankova VW, Richmond RC, Woolf BAR, Yarmolinsky J, Davies NM, Swanson SA, VanderWeele TJ, Higgins JPT, Timpson NJ, Dimou N, Langenberg C, Golub RM, Loder EW, Gallo V, Tybjaerg-Hansen A, Davey Smith G, Egger M, Richards JB. Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization: The STROBE-MR Statement. JAMA. 2021 Oct 26;326(16):1614-1621. doi: 10.1001/jama.2021.18236. PMID: 34698778.\u003c/li\u003e\n\u003cli\u003eOrr\u0026ugrave; V, Steri M, Sidore C, Marongiu M, Serra V, Olla S, Sole G, Lai S, Dei M, Mulas A, Virdis F, Piras MG, Lobina M, Marongiu M, Pitzalis M, Deidda F, Loizedda A, Onano S, Zoledziewska M, Sawcer S, Devoto M, Gorospe M, Abecasis GR, Floris M, Pala M, Schlessinger D, Fiorillo E, Cucca F. Complex genetic signatures in immune cells underlie autoimmunity and inform therapy. Nat Genet. 2020 Oct;52(10):1036-1045. doi: 10.1038/s41588-020-0684-4. Epub 2020 Sep 14. Erratum in: Nat Genet. 2020 Sep 18;: PMID: 32929287; PMCID: PMC8517961.\u003c/li\u003e\n\u003cli\u003eZhao JH, Stacey D, Eriksson N, Macdonald-Dunlop E, Hedman \u0026Aring;K, Kalnapenkis A, Enroth S, Cozzetto D, Digby-Bell J, Marten J, Folkersen L, Herder C, Jonsson L, Bergen SE, Gieger C, Needham EJ, Surendran P; Estonian Biobank Research Team; Paul DS, Polasek O, Thorand B, Grallert H, Roden M, V\u0026otilde;sa U, Esko T, Hayward C, Johansson \u0026Aring;, Gyllensten U, Powell N, Hansson O, Mattsson-Carlgren N, Joshi PK, Danesh J, Padyukov L, Klareskog L, Land\u0026eacute;n M, Wilson JF, Siegbahn A, Wallentin L, M\u0026auml;larstig A, Butterworth AS, Peters JE. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets. Nat Immunol. 2023 Sep;24(9):1540-1551. doi: 10.1038/s41590-023-01588-w. Epub 2023 Aug 10. Erratum in: Nat Immunol. 2023 Sep 7;: PMID: 37563310; PMCID: PMC10457199.\u003c/li\u003e\n\u003cli\u003eShin SY, Fauman EB, Petersen AK, Krumsiek J, Santos R, Huang J, Arnold M, Erte I, Forgetta V, Yang TP, Walter K, Menni C, Chen L, Vasquez L, Valdes AM, Hyde CL, Wang V, Ziemek D, Roberts P, Xi L, Grundberg E; Multiple Tissue Human Expression Resource (MuTHER) Consortium; Waldenberger M, Richards JB, Mohney RP, Milburn MV, John SL, Trimmer J, Theis FJ, Overington JP, Suhre K, Brosnan MJ, Gieger C, Kastenm\u0026uuml;ller G, Spector TD, Soranzo N. An atlas of genetic influences on human blood metabolites. Nat Genet. 2014 Jun;46(6):543-550. doi: 10.1038/ng.2982. Epub 2014 May 11. PMID: 24816252; PMCID: PMC4064254.\u003c/li\u003e\n\u003cli\u003eMoayyeri A, Hammond CJ, Hart DJ, Spector TD. The UK Adult Twin Registry (TwinsUK Resource). Twin Res Hum Genet. 2013;16:144\u0026ndash;149.\u003c/li\u003e\n\u003cli\u003eWichmann HE, Gieger C, Illig T. KORA-gen--resource for population genetics, controls and a broad spectrum of disease phenotypes. Gesundheitswesen. 2005;67(Suppl 1):S26\u0026ndash;30.\u003c/li\u003e\n\u003cli\u003eLopera-Maya EA, Kurilshikov A, van der Graaf A, et al. Effect of host genetics on the gut microbiome in 7,738 participants of the Dutch Microbiome Project [published correction appears in Nat Genet. 2022 Sep;54(9):1448]. Nat Genet. 2022;54(2):143-151. doi:10.1038/s41588-021-00992-y\u003c/li\u003e\n\u003cli\u003eKurki, M.I., Karjalainen, J., Palta, P. et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613:508\u0026ndash;518. https://doi.org/10.1038/s41586-022-05473-8\u003c/li\u003e\n\u003cli\u003eBycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, Motyer A, Vukcevic D, Delaneau O, O\u0026rsquo;Connell J, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203\u0026ndash;9.\u003c/li\u003e\n\u003cli\u003eBoddy S, Islam M, Moll T, Kurz J, Burrows D, McGown A, Bhargava A, Julian TH, Harvey C, Marshall JN, et al. Unbiased metabolome screen leads to personalized medicine strategy for amyotrophic lateral sclerosis. Brain Commun. 2022;4:fcac069. doi: 10.1093/braincomms/fcac069\u003c/li\u003e\n\u003cli\u003eVerbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50:693\u0026ndash;8.\u003c/li\u003e\n\u003cli\u003eHemani G, Tilling K, Smith GD. Orienting the causal relationship between imprecisely measured traits using GWAS summary data. PLOS Genet. 2017;13:e1007081.\u003c/li\u003e\n\u003cli\u003eBowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40:304\u0026ndash;314. doi: 10.1002/gepi.21965\u003c/li\u003e\n\u003cli\u003eBowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40:304\u0026ndash;14.\u003c/li\u003e\n\u003cli\u003eBowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44:512\u0026ndash;25. Burgess, S.; Thompson, S.G. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32:377\u0026ndash;389.\u003c/li\u003e\n\u003cli\u003eBrion MJ, Shakhbazov K, Visscher PM. Calculating statistical power in Mendelian randomization studies. Int J Epidemiol. 2013 Oct;42(5):1497-501. doi: 10.1093/ije/dyt179. PMID: 24159078; PMCID: PMC3807619.\u003c/li\u003e\n\u003cli\u003eHemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife. 2018;7:e34408. doi: 10.7554/eLife.34408\u003c/li\u003e\n\u003cli\u003eFerlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M, et al. Cancer Incidence and Mortality Worldwide: Sources, Methods and Major Patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359\u0026ndash;E386. doi: 10.1002/ijc.29210\u003c/li\u003e\n\u003cli\u003eYost K. E., Satpathy A. T., Wells D. K., Qi Y., Wang C., Kageyama R., et al. Clonal Replacement of Tumor-specific T Cells Following PD-1 Blockade. Nat Med. 2019;25:1251\u0026ndash;1259. doi: 10.1038/s41591-019-0522-3\u003c/li\u003e\n\u003cli\u003eLiu R.-X., Wei Y., Zeng Q.-H., Chan K.-W., Xiao X., Zhao X.-Y., et al. Chemokine (C-X-C Motif) Receptor 3-positive B Cells Link Interleukin-17 Inflammation to Protumorigenic Macrophage Polarization in Human Hepatocellular Carcinoma. Hepatology. 2015;62:1779\u0026ndash;1790. doi: 10.1002/hep.28020\u003c/li\u003e\n\u003cli\u003eZhang Z., Ma L., Goswami S., Ma J., Zheng B., Duan M., et al. Landscape of Infiltrating B Cells and Their Clinical Significance in Human Hepatocellular Carcinoma. Oncoimmunology. 2019;8:e1571388. doi: 10.1080/2162402x.2019.1571388\u003c/li\u003e\n\u003cli\u003eTsou P., Katayama H., Ostrin E. J., Hanash S. M. The Emerging Role of B Cells in Tumor Immunity. Cancer Res. 2016;76:5597\u0026ndash;5601. doi: 10.1158/0008-5472.can-16-0431\u003c/li\u003e\n\u003cli\u003eHe H., Wu J., Zang M., Wang W., Chang X., Chen X., et al. CCR6+ B Lymphocytes Responding to Tumor Cell-Derived CCL20 Support Hepatocellular Carcinoma Progression via Enhancing Angiogenesis. Am J Cancer Res. 2017;7:1151\u0026ndash;1163.\u003c/li\u003e\n\u003cli\u003eChew V., Tow C., Teo M., Wong H.L., Chan J., Gehring A., Loh M., Bolze A., Quek R., Lee V.K.M., et al. Inflammatory Tumour Microenvironment Is Associated with Superior Survival in Hepatocellular Carcinoma Patients. J Hepatol. 2010;52:370\u0026ndash;379. doi: 10.1016/j.jhep.2009.07.013\u003c/li\u003e\n\u003cli\u003eHabif G., Crinier A., Andr\u0026eacute; P., Vivier E., Narni-Mancinelli E. Targeting Natural Killer Cells in Solid Tumors. Cell Mol Immunol. 2019;16:415\u0026ndash;422. doi: 10.1038/s41423-019-0224-2\u003c/li\u003e\n\u003cli\u003eGuillerey C., Huntington N. D., Smyth M. J. Targeting Natural Killer Cells in Cancer Immunotherapy. Nat Immunol. 2016;17:1025\u0026ndash;1036. doi: 10.1038/ni.3518\u003c/li\u003e\n\u003cli\u003eChen Y., Hao X., Sun R., Wei H., Tian Z. Natural Killer Cell-Derived Interferon‐Gamma Promotes Hepatocellular Carcinoma through the Epithelial Cell Adhesion Molecule-Epithelial‐to‐Mesenchymal Transition Axis in Hepatitis B Virus Transgenic Mice. Hepatology. 2019;69:1735\u0026ndash;1750. doi: 10.1002/hep.30317\u003c/li\u003e\n\u003cli\u003eHage C., Hoves S., Strauss L., Bissinger S., Prinz Y., P\u0026ouml;schinger T., et al. Sorafenib Induces Pyroptosis in Macrophages and Triggers Natural Killer Cell-Mediated Cytotoxicity against Hepatocellular Carcinoma. Hepatology. 2019;70:1280\u0026ndash;1297. doi: 10.1002/hep.30666\u003c/li\u003e\n\u003cli\u003eXing M., Wang X., Kirken R., He L., Zhang J.-Y. Immunodiagnostic Biomarkers for Hepatocellular Carcinoma (HCC): The First Step in Detection and Treatment. Int J Mol Sci. 2021;22:6139. doi: 10.3390/ijms22116139\u003c/li\u003e\n\u003cli\u003eWang P, Kang D, Cao W, Wang Y, Liu Z. Diabetes mellitus and risk of hepatocellular carcinoma: a systematic review and meta-analysis. Diabetes Metab Res Rev. 2012;28:109\u0026ndash;122. doi: 10.1002/dmrr.1291\u003c/li\u003e\n\u003cli\u003eMantovani A, Targher G. Type 2 diabetes mellitus and risk of hepatocellular carcinoma: spotlight on nonalcoholic fatty liver disease. Ann Transl Med. 2017;5:270. doi: 10.21037/atm.2017.04.41\u003c/li\u003e\n\u003cli\u003eHuang T-S, et al. Diabetes, hepatocellular carcinoma, and mortality in hepatitis C-infected patients: a population-based cohort study. J Gastroenter Hepatol. 2017;32:1355\u0026ndash;1362. doi: 10.1111/jgh.13670\u003c/li\u003e\n\u003cli\u003eXie D, Zhang G, Ma Y, Wu D, Jiang S, Zhou S, Jiang X. Circulating Metabolic Markers Related to the Diagnosis of Hepatocellular Carcinoma. J Oncol. 2022 Dec 9;2022:7840606. doi: 10.1155/2022/7840606. PMID: 36532884; PMCID: PMC9757943.\u003c/li\u003e\n\u003cli\u003eFujiogi M, Camargo CA Jr, Raita Y, Bochkov YA, Gern JE, Mansbach JM, Piedra PA, Hasegawa K. Respiratory viruses are associated with serum metabolome among infants hospitalized for bronchiolitis: A multicenter study. Pediatr Allergy Immunol. 2020 Oct;31(7):755-766. doi: 10.1111/pai.13296. Epub 2020 Jun 10. PMID: 32460384; PMCID: PMC7704725.\u003c/li\u003e\n\u003cli\u003eDu D, Liu C, Qin M, Zhang X, Xi T, Yuan S, Hao H, Xiong J. Metabolic dysregulation and emerging therapeutical targets for hepatocellular carcinoma. Acta Pharm Sin B. 2022 Feb;12(2):558-580. doi: 10.1016/j.apsb.2021.09.019. Epub 2021 Sep 25. PMID: 35256934; PMCID: PMC8897153.\u003c/li\u003e\n\u003cli\u003eMaurice CF, Haiser HJ, Turnbaugh PJ. Xenobiotics shape the physiology and gene expression of the active human gut microbiome. Cell. 2013;152(1\u0026ndash;2):39\u0026ndash;50.\u003c/li\u003e\n\u003cli\u003eBj\u0026ouml;rkholm B, Bok CM, Lundin A, Rafter J, Hibberd ML, Pettersson S. Intestinal microbiota regulate xenobiotic metabolism in the liver. PLoS One. 2009 Sep 9;4(9):e6958. doi: 10.1371/journal.pone.0006958. PMID: 19742318; PMCID: PMC2734986.\u003c/li\u003e\n\u003cli\u003eAllameh A, Niayesh-Mehr R, Aliarab A, Sebastiani G, Pantopoulos K. Oxidative Stress in Liver Pathophysiology and Disease. Antioxidants (Basel). 2023 Aug 22;12(9):1653. doi: 10.3390/antiox12091653. PMID: 37759956; PMCID: PMC10525124.\u003c/li\u003e\n\u003cli\u003eGrohmann M, Wiede F, Dodd GT, Gurzov EN, Ooi GJ, Butt T, et al. Obesity drives STAT-1-dependent NASH and STAT-3-dependent HCC. Cell. 2018;175:1289\u0026ndash;1306.e20.\u003c/li\u003e\n\u003cli\u003eWang M, Han J, Xing H, Zhang H, Li Z, Liang L, et al. Dysregulated fatty acid metabolism in hepatocellular carcinoma. Hepat Oncol. 2016;3:241\u0026ndash;251.\u003c/li\u003e\n\u003cli\u003eRiscal R, Skuli N, Simon MC. Even cancer cells watch their cholesterol! Mol Cell. 2019;76:220\u0026ndash;231.\u003c/li\u003e\n\u003cli\u003eSchnabl B \u0026amp; Brenner DA. Interactions between the intestinal microbiome and liver diseases. Gastroenterology. 2014;146:1513\u0026ndash;1524. doi: 10.1053/j.gastro.2014.01.020\u003c/li\u003e\n\u003cli\u003eDapito DH et al. Promotion of hepatocellular carcinoma by the intestinal microbiota and TLR4. Cancer Cell. 2012;21:504\u0026ndash;516. doi: 10.1016/j.ccr.2012.02.007\u003c/li\u003e\n\u003cli\u003eYu LX et al. Endotoxin accumulation prevents carcinogen-induced apoptosis and promotes liver tumorigenesis in rodents. Hepatology (Baltimore, Md.). 2010;52:1322\u0026ndash;1333. doi: 10.1002/hep.23845\u003c/li\u003e\n\u003cli\u003eYoshimoto S et al. Obesity-induced gut microbial metabolite promotes liver cancer through senescence secretome. Nature. 2013;499:97\u0026ndash;101. doi: 10.1038/nature12347\u003c/li\u003e\n\u003cli\u003eYoshimoto S, Loo TM, Atarashi K, Kanda H, Sato S, Oyadomari S, Iwakura Y, Oshima K, Morita H, Hattori M, Honda K, Ishikawa Y, Hara E, Ohtani N. Obesity-induced gut microbial metabolite promotes liver cancer through senescence secretome. Nature. 2013;499:97\u0026ndash;101.\u003c/li\u003e\n\u003cli\u003eCopp\u0026eacute; J-P, Desprez P-Y, Krtolica A, Campisi J. The senescence-associated secretory phenotype: The dark side of tumor suppression. Annu Rev Pathol. 2010;5:99\u0026ndash;118.\u003c/li\u003e\n\u003cli\u003eHenao-Mejia J et al. Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity. Nature. 2012;482:179\u0026ndash;185. doi: 10.1038/nature10809\u003c/li\u003e\n\u003cli\u003eDe Minicis S et al. Dysbiosis contributes to fibrogenesis in the course of chronic liver injury in mice. Hepatology (Baltimore, Md.). 2014;59:1738\u0026ndash;1749. doi: 10.1002/hep.26695\u003c/li\u003e\n\u003cli\u003eZhang HL et al. Profound impact of gut homeostasis on chemically-induced pro-tumorigenic inflammation and hepatocarcinogenesis in rats. J Hepatol. 2012;57:803\u0026ndash;812. doi: 10.1016/j.jhep.2012.06.011\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Hepatocellular Carcinoma, Mendelian Randomization, Immune Microenvironment, Metabolic Reprogramming, Gut Microbiota, Biomarkers","lastPublishedDoi":"10.21203/rs.3.rs-6188338/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6188338/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eHepatocellular carcinoma (HCC) ranks as the sixth most common cancer globally, with increasing mortality and persistent therapeutic challenges. Emerging evidence suggests that the immune microenvironment, metabolic reprogramming, and gut microbiota dysbiosis play critical roles in HCC pathogenesis, though their causal effects are unclear. This study used Mendelian randomization (MR) to systematically assess these factors' causal relationships with HCC.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eA two-sample MR analysis, integrated with meta-analysis, examined genetic data on 731 immune cell types, 91 immune factors, 1,400 metabolites, and 412 gut microbiota phenotypes. HCC outcome data were sourced from FinnGen (discovery) and UK Biobank (replication). Five MR methods\u0026mdash;inverse-variance weighted, weighted median, MR-Egger, weighted mode, and simple mode\u0026mdash;were applied, with rigorous sensitivity, heterogeneity, and reverse causation analyses to ensure validity.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eIn the discovery stage, causal associations with HCC were identified for 4 immune cell phenotypes (3 protective, 1 pathogenic), 2 immune factors, 57 metabolites (24 pathogenic, 33 protective), and 105 gut microbiota phenotypes (51 pathogenic, 54 protective). Replication validated the metabolite 1-stearoyl-2-linoleoyl-GPC (18:0/18:2) and 6 gut microbiota phenotypes (4 protective, 2 oncogenic). Meta-analysis confirmed 2 protective immune cell phenotypes\u0026mdash;CD20 on CD20- CD38- B cells and CD16-CD56 on NK cells\u0026mdash;and 9 metabolites (3 protective, 6 oncogenic) as significant causal factors.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThis study establishes causal links between specific immune cells, metabolites, and gut microbiota with HCC, revealing protective and oncogenic roles. These findings highlight potential biomarkers and therapeutic targets, advancing strategies for HCC prevention and personalized treatment.\u003c/p\u003e","manuscriptTitle":"Exploring Hepatocellular Carcinoma Etiology through Multi-omics Bioinformatics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-21 13:14:55","doi":"10.21203/rs.3.rs-6188338/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"423e95f0-084f-446c-82e1-c568effe0dd7","owner":[],"postedDate":"April 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-05T10:53:51+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-21 13:14:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6188338","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6188338","identity":"rs-6188338","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}