Systemic metabolic dysfunction drives platelet-mediated inflammation, fibrogenesis, and hepatocarcinogenesis in metabolic dysfunction-associated steatotic liver disease Authors: Bingjie Ye, Chengfu Xu Published: 2026 ABSTRACT AbstractMetabolic dysfunction-associated steatotic liver disease (MASLD) has emerged as the most prevalent chronic liver disease worldwide and is closely linked to systemic metabolic disorders. In addition to their classical role in hemostasis, platelets are increasingly recognized as active regulators of inflammation and immune responses, yet their contribution to MASLD pathogenesis remains incompletely defined. This review synthesizes current knowledge on how metabolic disturbances and gut microbiota dysbiosis trigger platelet hyperactivation and intrahepatic recruitment. We examined the mechanisms by which activated platelets exacerbate steatosis, amplify inflammation through interactions with immune cells, promote fibrogenic remodeling through hepatic stellate cell activation, and contribute to hepatocarcinogenesis. In the context of MASLD-associated hepatocellular carcinoma, platelet involvement may occur through both inflammation/fibrogenic remodeling–mediated and direct tumor-regulatory mechanisms. Furthermore, the therapeutic potential of antiplatelet agents, particularly aspirin, in attenuating disease progression has been evaluated. We conclude that targeting platelet-related pathways may represent a promising therapeutic strategy to interrupt the interplay between metabolic dysfunction and liver injury in MASLD.Keywords: Metabolic dysfunction, Platelets, MASLD, MASH, Aspirin INTRODUCTION Metabolic dysfunction-associated steatotic liver disease (MASLD), formerly termed nonalcoholic fatty liver disease, is the updated nomenclature proposed by the 2023 multisociety Delphi consensus [1]. In a cohort of more than 1,300 patients, more than 99.5% of individuals previously diagnosed with nonalcoholic fatty liver disease fulfilled the diagnostic criteria for MASLD [2]. Therefore, for clarity and terminological consistency, this review uses MASLD as the primary term throughout the manuscript. MASLD has become the predominant chronic liver disease globally, affecting approximately 38% of the adult population [3]. MASLD encompasses a pathological spectrum ranging from simple steatosis to metabolic dysfunction-associated steatohepatitis (MASH), cirrhosis, and hepatocellular carcinoma (HCC) [4,5]. MASH is characterized by hepatocyte injury, inflammation, and varying degrees of fibrosis, potentially progressing to cirrhosis and HCC [4]. Rather than being an isolated hepatic condition, MASLD is increasingly recognized as the hepatic manifestation of metabolic syndrome, intrinsically linked to obesity, type 2 diabetes mellitus (T2DM), hypertension, and dyslipidemia [6-8]. The pathophysiology of MASLD involves systemic metabolic dysfunction (including insulin resistance, dyslipidemia, and amino acid metabolism abnormalities), gut microbiota dysbiosis, and immune dysregulation [8-10]. Characterized by low-grade systemic inflammation, MASLD progression is promoted by inflammatory mediators such as tumor necrosis factor (TNF), interleukin (IL)-1β, and IL-6, as well as immune cells, which contribute to the establishment of chronic hepatic inflammation [9]. Paralleling the global epidemic of obesity and metabolic syndrome, the prevalence of MASLD is projected to increase over the next decade, imposing a substantial burden on public health systems [3,11]. While the precise pathogenesis of MASLD remains to be fully elucidated, accumulating evidence suggests a possible role for platelets as contributors to disease progression. Although traditionally defined by their hemostatic and thrombotic roles, platelets are now recognized as critical regulators of inflammation and immune responses [12]. In the context of MASLD, systemic metabolic disturbances create a milieu that precipitates platelet hyperactivation [13]. Activated platelets not only accumulate within the hepatic microvasculature—impairing microcirculation—but also may drive hepatic inflammation, fibrosis, and tumorigenesis through the release of proinflammatory mediators and direct interactions with liver sinusoidal endothelial cells (LSECs), Kupffer cells, and circulating leukocytes [14,15]. Collectively, the available evidence suggests that platelets may function as one link between metabolic dysfunction, inflammation, and hepatic pathology. While the contribution of platelets to MASLD progression has garnered increasing attention [16], a systematic review of the specific molecular mechanisms and their translational potential is lacking. This review aims to comprehensively explore the multifaceted roles of platelets in the pathogenesis of MASLD. We first analyze how systemic metabolic dysregulation triggers platelet activation and hyperreactivity, followed by a description of the molecular basis underlying platelet recruitment and adhesion within the liver. We subsequently examine how activated platelets exacerbate steatosis, drive immune cell infiltration, promote fibrogenic remodeling, and contribute to hepatocarcinogenesis. Finally, we summarize recent clinical evidence and controversies regarding antiplatelet therapy—particularly aspirin—in the management of MASLD and discuss future research directions. To improve transparency, the literature cited in this narrative review was identified through targeted PubMed searches from database inception to January 2026. The search terms included “MASLD”, “NAFLD”, “MAFLD”, “platelet”, “platelet activation”, “platelet–leukocyte aggregates”, “NETs”, “fibrosis”, “cirrhosis”, “hepatocellular carcinoma”, “aspirin”, and “antiplatelet therapy”. Priority was given to randomized clinical trials, cohort studies, and key mechanistic studies most relevant to platelet biology in steatotic liver disease, with additional articles identified through reference screening of relevant original articles and recent reviews. PLATELET HYPERREACTIVITY DRIVEN BY SYSTEMIC METABOLIC DYSFUNCTION Circulating platelets typically remain in a resting state but become activated under specific physiological or pathological conditions [17]. Activation is a complex, multistep event leading to diverse functional outcomes. Upon activation, platelets lose their discoid morphology and adopt a spherical conformation [18]. This remodeling is secondary to intracellular calcium release, which triggers reorganization of the actin-myosin cytoskeleton [18]. In clinical practice, this morphological change may manifest as an increase in mean platelet volume, a parameter reported to be elevated in patients with MASLD [19]. Following activation, platelets expose phosphatidylserine on their surface [20]. This phosphatidylserine exposure provides a catalytic platform for coagulation factor binding, promoting thrombin generation and thrombus formation [20]. The activation of the coagulation cascade and the subsequent increase in thrombin generation have been linked to the development of liver inflammation [21]. Activated platelets release their granular contents. Dense granules contain small molecules such as adenosine diphosphate (ADP), adenosine triphosphate (ATP), guanosine diphosphate (GDP), and serotonin (5-HT), whereas α-granules contain various cytokines, mitogens, and inflammatory modulators [22]. Degranulation is accompanied by increased membrane expression of P-selectin and CD40L, enabling activated platelets to interact with immune cells [23]. Certain proteins abnormally overexpressed in MASLD can directly trigger this activation. For example, Fetuin-A, a liver-derived plasma glycoprotein, enhances platelet activation via toll-like receptor (TLR)-4-dependent signaling pathways [24], whereas serum amyloid A1 (SAA1) may induce activation and aggregation via TLR2 [25]. Furthermore, within the context of MASLD, systemic metabolic dysfunction and gut microbiota dysbiosis create an environment that induces platelet activation (Fig. 1). Platelet hyperreactivity driven by systemic metabolic dysfunction. Under conditions of systemic metabolic dysfunction associated with MASLD, platelets circulate within an aberrant physiological environment. This leads to the activation of diverse downstream signaling pathways in platelets, collectively resulting in platelet activation. AGEs, advanced glycation end products; AGs, α-granules; ATP, adenosine triphosphate; CAP-PEs, carboxyalkylpyrrole-phosphatidylethanolamine derivatives; GLUT3, glucose transporter 3; GPVI, glycoprotein VI; GPCR, G protein-coupled receptors; IL-6, interleukin-6; KIV, α-ketoisovaleric acid; LPS, lipopolysaccharide; MASLD, metabolic dysfunction–associated steatotic liver disease; OxLDL, oxidized low-density lipoprotein; oxPCCD36, oxidized phospholipids, ligands for CD36; PAGln, phenylacetylglutamine; PMP, platelet-derived microparticle; ROS, reactive oxygen species; TLR, Toll-like receptor; TMAO, trimethylamine-N-oxide; TNF-α, tumor necrosis factor-α; TXA2, thromboxane A2. Created in BioRender. Ye, B. (2026) https://Bio-Render.com/bj8mkzf. Obesity, insulin resistance and hyperglycemia Obesity is closely associated with fatty liver disease [26,27]. Metabolic factors accompanying obesity—including insulin resistance, elevated free fatty acids, and alterations in inflammatory and adipokine profiles—contribute significantly to the development and progression of MASLD [28]. Obesity alters the platelet transcriptome, enriching transcripts related to MASLD pathways [29], whereas bariatric surgery has been shown to partially alleviate platelet activation [30]. Proteomic analysis revealed that platelets from obese patients presented upregulation of proteins associated with activation and aggregation, such as actin, fibrinogen, thrombospondin-1 (TSP-1), and integrin αIIb [31]. Furthermore, obese individuals exhibit elevated surface expression of glycoprotein VI, which is correlated with body mass index and plays a crucial role in collagen-induced activation [31,32]. Adipose tissue dysregulation drives platelet activation through altered adipokine secretion. Obesity is characterized by decreased levels of adiponectin and elevated levels of leptin, resistin, TNF-α, and IL-6 [33]. Adiponectin normally inhibits platelet activation [34], whereas leptin promotes activation by inducing Janus kinase 2 signaling, increasing intracellular Ca²⁺, and enhancing GPIIb/IIIa activity and thromboxane synthesis [35]. Resistin exacerbates insulin resistance and enhances thromboxane-dependent activation [36], whereas inflammatory mediators such as TNF-α and IL-6 directly stimulate platelets [37,38]. Insulin signaling is a key regulator of platelet function. Under physiological conditions, insulin suppresses platelet activation by binding to its receptors on the surface of platelets. In contrast, insulin resistance impairs this inhibitory pathway, leading to increased platelet activity and a greater propensity for aggregation [39]. In T2DM, adipokines such as resistin, leptin, plasminogen activator-1 and retinol-binding protein 4 interfere with insulin receptor substrate 1 activity in megakaryocytes, rendering platelets insulin resistant and prone to increased calcium mobilization and procoagulant activity [40]. Additionally, in T1DM, insulin resistance has been identified as an important regulator of platelet hyperresponsiveness [41]. Hyperglycemia acts as a potent predictor of platelet activation through mechanisms involving oxidative stress, calcium signaling, and metabolic reprogramming. Hyperglycemia activates the aldose reductase-polyol pathway to significantly increase the levels of platelet-derived reactive oxygen species (ROS), which increase collagen-induced surface expression of P-selectin and thromboxane A2 (TXA2) synthesis while activating the nuclear factor kappa B (NF-κB) signaling pathway to upregulate the P2Y12 receptor [42,43]. Furthermore, hyperglycemia promotes platelet activation and platelet-derived microparticle (PMP) release by increasing platelet surface protease-activated receptor 4 (PAR-4) receptor expression, increasing platelet-surface PAR-4 receptor expression-mediated calcium influx, and subsequently activating calpain proteases [44]. This process further exacerbates a vicious cycle of platelet activation and chronic inflammation by activating the NF-κB pathway via PMPs, thereby driving the secretion of the macrophage inflammatory cytokine IL-6 [44]. Metabolic reprogramming is essential for platelet hyperactivity. 45 Hyperglycemia drives activation by enhancing glucose metabolism via glucose transporter 3 [46]. Glucose transporter 3 facilitates glucose entry into α-granules, supporting intragranular glycolysis and ATP generation to complete full degranulation following glycogen depletion [47]. Activated platelets characteristically exhibit aerobic glycolysis (Warburg effect), which is regulated by dimeric pyruvate kinase M2 [48-50]. Under hyperglycemic conditions, dimeric pyruvate kinase M2 facilitates glucose uptake and lactate production via the PI3K/Akt/GSK3β axis, enhancing platelet activation [51]. Chronic hyperglycemia also leads to the accumulation of advanced glycation end products (AGEs) [52]. AGEs bind to platelet surface receptors, including RAGE and CD36, upregulate P-selectin, and activate MAPK/JNK signaling to increase sensitivity to low-dose agonists [53,54]. In addition, recent studies have revealed a novel endothelial-dependent paracrine mechanism through which hyperglycemia promotes platelet activation: hyperglycemia-induced nuclear translocation of endothelial yes-associated protein enhances the production of prostaglandin E2, which subsequently activates platelet-surface prostaglandin E receptor 3 to drive platelet hyperactivity [55]. Dyslipidemia Hypertriglyceridemia has been identified as an independent risk marker for platelet hyperactivation in patients with metabolic syndrome [56]. Dyslipidemia is intrinsically linked to oxidative stress and the generation of biologically active oxidized lipids, with plasma from hyperlipidemic patients containing multiple lipid oxidation products [57]. Oxidized low-density lipoprotein (oxLDL) binds to multiple scavenger receptors on the surface of platelets, including CD [36,58-60] lectin-like oxidized low-density lipoprotein receptor-1, scavenger receptor class B type 1, G protein-coupled lysophosphatidic acid receptor 1, and integrins [61]. Upon binding, these receptors trigger distinct signaling cascades that modulate key platelet functions, including secretion, adhesion, microvesicle release, procoagulant phosphatidylserine exposure, Ca²⁺ mobilization, and ROS production, lowering the threshold for activation and synergizing with physiological agonists [61,62]. Patients with metabolic syndrome exhibit significantly elevated levels of lipid peroxidation in LDL, which can activate platelets [63]. Furthermore, oxLDL and lipids induce ROS production and activate the NF-κB pathway, which upregulates platelet PPARα expression [64]. Ligand-bound PPARα subsequently promotes dense granule secretion via the p38/ROS/Akt axis, releasing agonists such as ADP [64]. ox-LDL also stimulates the production of PMPs, which in turn enhance platelet activation through CD36 binding [65]. Conversely, activated platelets contribute to LDL oxidation through phospholipase A2 (PLA2), proprotein convertase subtilisin/kexin type 9 (PCSK9), and NADPH oxidase 2-dependent mechanisms [66,67]. Additionally, oxLDL can increase the release of platelet CXCL12, promoting platelet uptake of oxLDL through the CXCL12-CXCR4-CXCR7 axis [68]. Therefore, a complex positive feedback loop exists between oxLDL and platelet activation. Different LDL oxidation pathways generate specific bioactive lipids. Oxidized choline glycerophospholipids (ox-PCCD36), present at sites of oxidative stress, induce the assembly of a CD36/TLR2/TLR6 complex on platelets and activate the downstream innate immune signaling pathway, ultimately activating platelet integrin signaling [69]. Moreover, oxPCCD36-mediated ROS production via NADPH oxidase 2 is essential for this activation process [70]. Similarly, in hyperlipidemic plasma, lipid peroxidation modifies phosphatidylethanolamine (PE) to form carboxyalkylpyrrole-PE derivatives (CAP-PEs) [71]. Like oxPCCD36, CAP-PEs activate platelets by inducing the formation of TLR2/TLR1 complexes and triggering the downstream innate immune cascade [71]. Moreover, while glucagon-like peptide-1 typically suppresses platelet reactivity by reducing oxidative stress and enhancing the NO/cGMP/PKG pathway while inhibiting PI3K and MAPK signaling [72], hypercholesterolemia confers resistance to these protective effects, resulting in sustained platelet hyperreactivity [73]. Amino acid metabolism Elevated plasma levels of branched-chain amino acids (BCAAs), including leucine, isoleucine, and valine, are correlated with distinct stages of MASLD, and defects in BCAA catabolism contribute to disease progression [10]. Valine is converted by branched-chain aminotransferase to form α-ketoisovaleric acid (KIV), which is subsequently metabolized by the branched-chain α-keto acid dehydrogenase complex and mitochondrial enzymes to yield propionyl-CoA [74]. Analyses of BCAA metabolism revealed that elevated KIV levels and a greater ratio of branched-chain α-keto acids to BCAAs are strongly associated with increased steatosis severity and MASH [75]. BCAA metabolic pathways may be involved in integrin αIIbβ3-mediated bidirectional signaling to promote platelet activation, with the valine metabolite KIV and the ultimate oxidation product propionyl-CoA exhibiting the strongest proplatelet activation effects [76]. Homocysteine, an intermediate of methionine metabolism, may also contribute to platelet hyperreactivity, with serum levels positively correlated with MASLD risk [77]. Hyperhomocysteinemia promotes platelet membrane phospholipid hydrolysis and phosphatidylserine externalization by activating cytosolic PLA2 while increasing the secretion of autotaxin, which directly binds to integrin β3. Together, these processes enhance integrin αIIbβ3 outside-in signaling [78]. Additionally, hyperhomocysteinemia induces N-homocystoylation of β-arrestin1/2 in platelets. This modification inhibits β-arrestin-mediated internalization and desensitization of G protein-coupled receptors—including PAR1/4, TP, P2Y1, and P2Y12—while biasing signaling toward G protein-dependent pathways [79]. Collectively, these mechanisms contribute to the heightened platelet reactivity observed in hyperhomocysteinemia. Gut microbiome and microbial metabolites The gut microbiota plays a pivotal role in the development of MASLD [80,81]. It drives disease progression by compromising the intestinal barrier, generating bioactive metabolites, and disrupting bile acid metabolism [81,82]. Impaired barrier integrity permits the translocation of bacterial components, such as lipopolysaccharide (LPS), into the systemic circulation [83]. Circulating LPS binds directly to TLR4 on the surface of platelets, increasing their responsiveness to low-dose agonists [84,85]. This interaction triggers downstream signaling pathways that generate eicosanoids, including TXA2 and isoprostanes, through oxidative stress, thereby inducing platelet activation [85]. In addition to bacterial structural components, metabolites derived from the gut microbiota also directly modulate platelet function. Trimethylamine-N-oxide (TMAO) is a key regulator that promotes MASLD [86,87]. Gut bacteria metabolize choline to produce trimethylamine, which diffuses into the bloodstream and is oxidized by the liver to form TMAO [88]. TMAO potentiates platelet activation in response to submaximal concentrations of multiple agonists by increasing IP3-mediated intracellular calcium release, thereby promoting a hyperreactive and prothrombotic platelet phenotype [89]. Phenylacetic acid, a gut microbiota metabolite linked to hepatic steatosis and lipid accumulation [90], is derived from the bacterial metabolism of phenylalanine. It is subsequently conjugated with glutamine by hepatic enzymes to form phenylacetylglutamine (PAGln) [91]. PAGln enhances platelet function by interacting with adrenergic receptors to activate downstream signaling and by amplifying stimulus-dependent intracellular calcium release [92]. Recent research has revealed that 2-methylbutyrylcarnitine, a short branched-chain acylcarnitine of microbial origin, binds directly to integrin α2β1 [93]. This binding enhances cytosolic PLA2 activation and promotes platelet hyperreactivity [93]. However, a direct association between 2-methylbutyrylcarnitine and the progression of MASLD has yet to be established. Collectively, these findings underscore that platelets function as systemic metabolic sensors, integrating signals from glucotoxicity, lipotoxicity, and microbial dysbiosis. The multifaceted nature of this activation suggests that in MASLD, platelet hyperreactivity is not driven by a single pathway but is a cumulative result of the systemic metabolic milieu. Consequently, relying on a single signaling blockade may be insufficient to fully normalize platelet function. Furthermore, the correlation between platelet activation markers (such as the mean platelet volume) and disease severity highlights the potential utility of incorporating platelet indices into noninvasive diagnostic algorithms [94], serving as a window into the patient’s systemic metabolic burden. MECHANISMS OF INTRAHEPATIC PLATELET RECRUITMENT AND ADHESION Platelets are the initial nonresident cell type that populates the liver within four weeks following the induction of a choline-deficient high-fat diet or a Western diet containing transfat [14]. LSECs exhibit a unique phenotype characterized by P-selectin nonexpression and low von Willebrand factor levels [15,95]. Combined with the relatively moderate shear rate within hepatic sinusoids, this provides a distinctive environment for platelet adhesion, with activated platelets exhibiting significantly enhanced adhesion [95]. Multiple mechanisms governing intrahepatic platelet recruitment and adhesion integrate signals from endothelial injury, innate immunity, and metabolic stress (Fig. 2). Mechanisms of intrahepatic platelet recruitment and adhesion. Activated platelets are recruited to the liver through multiple mechanisms. (A) Endothelial adhesion: In response to sterile injury, platelets adhere directly to liver sinusoidal endothelial cells (LSECs) via GPIIb/IIIa and GPIb, paving the way for neutrophils to enter the injured site for subsequent repair. (B) LPS-induced capture: LPS-induced systemic inflammation drives the initial adhesion of neutrophils and Kupffer cells in LSECs. These cells subsequently capture circulating platelets to form platelet–immune cell aggregates. (C) KC- and SAA1-mediated recruitment: Platelets are recruited through interactions with Kupffer cells via CD44-HA binding (early phase) and GPIbα-mediated binding (fibrosis stage). Additionally, serum amyloid A1 (SAA1) secreted by hepatocytes promotes platelet adhesion and aggregation within LSECs. GPIbα, glycoprotein Ib alpha; HA, hyaluronic acid; KC, Kupffer cell; LPS, lipopolysaccharide; SAA1, serum amyloid A1; TLR, toll-like receptor. Created in BioRender. Ye, B. (2026) https://BioRender.com/oo28oxk. In models of sterile liver injury, including diet-induced fatty liver disease, platelets adhere instantaneously to the sinusoidal endothelium adjacent to the injury site [96]. Early platelet recruitment is mediated by GPIIbIIIa (CD41), whereas sustained recruitment depends on the interaction of both GPIIbIIIa and GPIb (CD42B) with the sinusoidal endothelium [96]. This recruitment phase facilitates the subsequent entry of neutrophils into the liver to repair injured tissue [96]. Patients with MASLD, particularly those with MASH, exhibit significantly elevated serum levels of LPS, with marked LPS localization within hepatocytes [84]. Liver biopsies from these patients revealed that increased intrahepatic platelet aggregation is correlated with the formation of neutrophil extracellular traps (NETs), which function in immune defense [97]. These findings suggest that the pathological accumulation of gut-derived LPS via the portal vein may trigger platelet homing to the liver to assist in pathogen clearance. Experimental validation has shown that LPS-induced systemic inflammation drives the initial adhesion of neutrophils and Kupffer cells in liver sinusoids. These cells subsequently capture circulating platelets in a CD18-dependent manner to form dynamic platelet–immune cell aggregates, a mechanism distinct from direct platelet adhesion to the sinusoidal endothelium [98]. Kupffer cells appear to be important regulators of platelet recruitment during MASLD progression. Malehmir et al. [14] demonstrated that during the transition from steatosis to borderline MASH, hyaluronic acid released from lipotoxically injured hepatocytes is deposited on Kupffer cells. Platelets subsequently bind to this hyaluronic acid via CD44, facilitating their recruitment into hepatic sinusoids. In the stage of definitive MASH with fibrosis, platelet GPIbα is recruited by binding directly to unidentified receptors—distinct from von Willebrand factor, P-selectin, or macrophage-1 antigen (Mac-1, CD11b/CD18)—on Kupffer cells [14]. SAA1, a protein secreted in large quantities by hepatocytes during the acute-phase response [99], is also involved in this process. SAA1 is associated with lipid metabolism, bacterial clearance, and inflammation regulation [99]. The elevated SAA1 levels observed in MASH patients may be linked to increased inflammatory mediator levels and lipotoxicity [25,100]. Mechanistically, SAA1 directly drives abnormal platelet aggregation within hepatic sinusoids by enhancing platelet activation and adhesion, partially through activation of the TLR2 signaling pathway [25]. In vivo blockade of SAA1 expression inhibits platelet aggregation and reduces hepatic immune cell infiltration in MASLD [25]. Notably, platelet binding to the liver in MASLD often occurs through mechanisms distinct from classical hemostatic plugging, relying on specific interactions such as the CD44-hyaluronan axis and GPIbα-mediated binding to Kupffer cells. The identification of these specific nonhemostatic adhesion checkpoints provides a rationale for developing selective interventions that aim to prevent hepatic platelet accumulation without broadly impairing hemostasis. PLATELET ACTIVATION PROMOTES MASLD PROGRESSION Upon establishing intrahepatic recruitment and adhesion, these activated platelets act as important participants in the disease continuum, contributing to steatotic injury, the inflammatory transition to MASH, fibrogenic remodeling, and ultimately hepatocarcinogenesis through multiple mechanisms (Fig. 3). In this section, the mechanistic findings are discussed together with the available preclinical and human evidence. Possible mechanism through which activated platelets promote MASLD. Activated platelets release platelet-derived extracellular vesicles (pEVs), which deliver dysfunctional mitochondria to hepatocytes. The subsequent formation of dysfunctional lipid droplet–mitochondrial complexes (LDMs) disrupts hepatic lipid metabolism and increases mitochondrial ROS production. Platelets directly interact with neutrophils and monocytes via receptor–ligand pairs such as P-selectin/PSGL-1, promoting the formation of cellular aggregates and enhancing the proinflammatory responses of these immune cells. Platelets exert dual effects on liver fibrosis through the release of distinct factors. In hepatocarcinogenesis, platelets may contribute through two partially overlapping routes: an inflammation/fibrogenic remodeling– mediated route that promotes a protumorigenic hepatic microenvironment and a direct tumor-regulatory route that modulates antitumor immunity and tumor-cell behavior, including proliferation, migration, invasion, autophagy, and epithelial–mesenchymal transition. AG, α-granule; C3G, guanine nucleotide exchange factor for Ras-associated protein 1; CD40L, cluster of differentiation 40 ligand; CD8⁺ T, CD8-positive T cell; CXCL4, chemokine (C-X-C motif) ligand 4; DC, dendritic cell; EMT, epithelial–mesenchymal transition; HCC, hepatocellular carcinoma; HGF, hepatocyte growth factor; HSC, hepatic stellate cell; KLF6, Krüppel-like factor 6; LD, lipid droplet; LPS, lipopolysaccharide; LSEC, liver sinusoidal endothelial cell; MASH, metabolic dysfunction–associated steatohepatitis; MASLD, metabolic dysfunction–associated steatotic liver disease; Mo, monocyte; Mo-Mφ, monocyte-derived macrophage; mt, mitochondria; NET, neutrophil extracellular trap; NF-κB, nuclear factor kappa B; NKT, natural killer T cell; p38 MAPK, p38 mitogen-activated protein kinase; PDGFB, platelet-derived growth factor-B; PMPs, platelet-derived microparticles; PSGL-1, P-selectin glycoprotein ligand-1; ROS, reactive oxygen species; S1P, sphingosine-1-phosphate; TGF-β, transforming growth factor-β; TLR, toll-like receptor; TSP1, thrombospondin 1. Created in BioRender. Ye, B. (2026) https://BioRender.com/wqsjhhs. From simple steatosis to MASH Disrupting metabolism and exacerbating steatosis MASLD is characterized by the aberrant accumulation of lipid droplets (LDs) within hepatocytes [101]. Dysregulation of LD dynamics—including accumulation, fusion, and degradation—drives disease progression by inducing endoplasmic reticulum stress, generating ROS, exacerbating lipotoxicity, and disrupting hepatic architecture [102]. In the pathological milieu of MASH, platelet activation triggers the release of extracellular vesicles [103]. These platelet-derived extracellular vesicles transfer dysfunctional mitochondria into hepatocytes, where they fuse with LDs to form lipid droplet‒mitochondrial complexes [104]. These dysfunctional lipid droplet‒mitochondrial complexes aggravate lipid metabolic disorders, promote further LD accumulation, and stimulate mitochondrial ROS production, thereby accelerating the onset and progression of MASH [104]. Platelet α-granules play important roles in MASH pathogenesis. TSP-1 derived from these granules negatively regulates the production of Neuregulin 4 in brown adipocytes [105]. Neuregulin 4 is an adipose-enriched endocrine factor known to protect hepatocyte integrity and counteract diet-induced MASH.106 However, there is conflicting evidence; Bai et al.107 reported that TSP-1 ameliorates hepatic steatosis in diet-induced obese mice. Furthermore, TSP-1 is also secreted by endothelial cells, smooth muscle cells, adipocytes, fibroblasts, and hepatic stellate cells (HSCs),107 and the specific contribution of platelet-derived TSP-1 requires further delineation. Nevertheless, genetic studies using Nbeal2⁻/⁻ mice, which lack α-granules, have demonstrated complete resistance to MASH development, suggesting that the α-granule cargo may serve as an important effector linking platelet activation to MASH pathology [14]. Driving immune cell recruitment and inflammation Liver immune cells undergo remodeling during MASH, potentially creating an uncontrolled inflammatory environment that promotes disease progression [108,109]. Platelets participate in the recruitment of liver immune cells during the development of MASLD and promote disease progression through interactions with immune cells. Research has shown that leukocyte recruitment in the liver does not conform to the classical paradigm of selectin-mediated rolling but instead adheres to liver sinusoids in a platelet-dependent manner [110]. Platelet binding to LSECs induces NF-κB activation and chemokine secretion, providing an adhesion substrate for circulating leukocytes to facilitate immune cell recruitment during hepatic inflammation [95]. Platelets can recruit immune cells to sites of liver inflammation by releasing inflammatory mediators and interacting with adhesion molecules. Activated platelet α-granules release a variety of bioactive molecules, such as chemokines, transforming growth factor-β (TGF-β), and TSP1. These molecules not only directly trigger hepatic inflammatory responses but also recruit multiple immune cells, including CD8⁺ T cells, NKT cells, and monocyte-derived macrophages, collectively contributing to subsequent hepatocyte injury progression [14,105]. Once recruited to the liver, the subsequent crosstalk between platelets and leukocytes relies on direct cell‒cell contact and the release of soluble mediators. The initial binding between platelets and leukocytes is mediated primarily by P-selectin on activated platelets interacting with P-selectin glycoprotein ligand-1 (PSGL-1) on leukocytes, and this interaction is further stabilized by the binding of platelet GPIbα or fibrinogen bridging via GPIIb/IIIa to Mac-1 on leukocytes [111]. Furthermore, soluble mediators released from platelet α-granules, such as CD40L, RANTES (CCL5), and CXCL4, play crucial roles in increasing leukocyte activation and stabilizing these aggregates [112,113]. In addition to their ability to recruit, activated platelets modulate neutrophil effector functions. Direct interactions and soluble mediators promote neutrophil phagocytosis and degranulation, leading to the release of proinflammatory cytokines [114,115]. Furthermore, platelets increase the oxidative burst in neutrophils, promoting the generation of ROS [116], which exacerbates damage during inflammation. Neutrophils are critical components of the host immune system and contribute to defense against diverse microorganisms through three primary mechanisms: degranulation, phagocytosis, and the release of NETs [117]. NETs are released primarily via a specific form of cell death known as NETosis. These structures are large extracellular weblike complexes composed of cytosolic and granule proteins assembled on a scaffold of decondensed chromatin [118]. They bind to and kill microorganisms, thereby playing a protective role. However, recent attention has increasingly focused on their pathogenic potential. Platelets are activated by the recognition of pathogen-associated molecular patterns (such as LPS) through their surface TLR4 receptors, which specifically bind to neutrophils already adhering to the vascular endothelium and induce the release of NETs [119]. This process is driven by P-selectin binding to PSGL-1, which activates ROS-dependent signaling pathways in neutrophils [120]. This interaction is further supported by platelet GPIbα, integrin αIIbβ3, and intercellular adhesion molecule 2 [121]. Activated platelets secrete soluble factors such as β-defensin 1, CD40L, PF4, and CCL5 and can induce NET formation by binding to neutrophil β2 integrins [121]. In summary, the mechanism of NET formation mediated by activated platelets depends on the inflammatory immune microenvironment, specific receptor‒ligand interactions, and soluble mediators that may serve as regulatory factors in this process. Notably, this interaction is bidirectional, as NETs may in turn feed back onto platelets and promote their activation and aggregation, thereby reinforcing thromboinflammatory signaling [122]. The formation of NETs is closely associated with the severity of MASH. In both animal models and human liver tissue, NETs are significantly positively correlated with the stages of inflammation and fibrosis [123,124]. However, no significant changes are observed in models in which only simple steatosis is induced, suggesting that NETs may represent an important effector mechanism in the progression of fatty liver disease to advanced inflammatory fibrosis [124]. NETs drive the progression of MASH through mechanisms such as recruiting monocyte-derived macrophages, promoting inflammatory cytokine production, inducing hepatocellular senescence to exacerbate lipotoxicity, triggering metabolic reprogramming of HSCs, and promoting Treg differentiation to suppress immune surveillance, thereby aggravating liver inflammation and fibrosis and ultimately leading to the occurrence of HCC [125-128]. A recent study on NETs in the mouse hepatic vasculature revealed that NET removal induces secondary inflammation (increased proinflammatory IL-1β and platelet aggregation), thereby generating new waves of NETs that may perpetuate a vicious cycle of liver injury [129]. In the circulation, platelets interact with monocytes primarily via the P-selectin/PSGL-1 and CD40L/CD40 axes, forming platelet‒monocyte aggregates under inflammatory conditions [130]. This interaction drives monocytes toward a proinflammatory and prothrombotic phenotype characterized by increased surface expression of CD40 and tissue factors, along with transcriptional upregulation of proinflammatory genes [131]. The binding of platelet P-selectin to PSGL-1 initiates a cross-presentation program, driving the rapid differentiation of monocytes into physiological dendritic cells that activate antigen-specific CD8+ T cells, thereby bridging innate and adaptive immunity [132]. Mechanistically, platelets can also modulate monocyte function through vesicle-mediated transfer of cytokine transcription regulators, such as NF-κB and p38 MAPK. This transcellular signaling enhances the proinflammatory cytokine response of human monocytes and helps prevent immunoparalysis [133]. In inflammatory liver disease, platelet-intrinsic TLR4 signaling has been shown to drive the formation of platelet–monocyte aggregates [134]. These aggregates bias circulating monocytes toward an inflammatory Ly6Chigh phenotype, thereby facilitating their interaction with cerebral endothelial cells and ultimately triggering microglial activation and sickness behavior [134]. Overall, the role of platelet‒monocyte‒macrophage interactions in the development of MASLD remains incompletely understood. Platelets amplify fibrogenesis across the MASLD continuum Bidirectional effects of platelets on fibrogenesis Current evidence does not support the notion that platelets are uniformly profibrotic or uniformly antifibrotic. On the profibrotic side, several platelet-derived mediators have been implicated in HSC activation. Platelet-derived TGF-β1, PDGF-β, S1P, and 5-HT have been reported to activate profibrotic programs in HSCs [135,136], and attenuation of platelet-specific TGF-β1 signaling or blockade of the PDGF-B/PDGFR-β axis reduces hepatic fibrosis in mouse models [137,138]. In addition, TSP1 and CXCL4, which are released from platelet α-granules, may further promote fibrogenesis by enhancing HSC activation and sustaining proinflammatory myeloid-cell programs [105,139]. In parallel, platelet-mediated immune cell recruitment and NET formation may create a microenvironment favorable for fibrogenesis [123]. Moreover, a separate line of evidence suggests that platelets may exert antifibrotic or reparative effects under some conditions. Platelet-intrinsic C3G has been linked to an antifibrotic secretory profile, including increased levels of hepatocyte growth factor and IL-6 and reduced levels of profibrotic signaling [140-142], whereas platelet-rich plasma, platelet-derived extracellular vesicles, and platelet transfusion have been reported to improve fibrosis-related readouts and liver function indicators in animal models and in a small exploratory clinical study [143-145]. These observations indicate that platelet biology in fibrosis cannot be reduced to a single profibrotic axis and that the net effect of platelets likely depends on the biological setting in which they are studied. Stage-specific platelet phenotype and histologic associations from F0 to F4 To facilitate comparisons of the available human evidence across fibrosis stages, the major platelet-related variables discussed below are summarized in Table 1. Among them, the most consistent stage-linked change is a progressive decline in peripheral platelet count, which becomes most evident in cirrhosis [146]. In parallel, platelet-containing noninvasive fibrosis scores, including fibrosis-4 index (FIB-4) and AST-to-platelet ratio index (APRI), are higher in advanced fibrosis/cirrhosis, supporting their utility as platelet-containing surrogates of the fibrotic burden [147]. Table 1.Stage-linked platelet-related variables across fibrosis progression in MASLD patients and their associations with histological activity, fibrosis stage, and liver-related outcomesPlatelet-related variablesCase definition / PopulationF0–F2 (No/mild fibrosis)F3–F4 (Advanced fibrosis/cirrhosis)Association with histologic activity, fibrosis stage, or liver-related outcomesRefsPeripheral blood platelet count (×109/L)Biopsy-proven MASLD cohorts (N=1,048)F0: 248±68F3: 189±64Inversely associated with fibrosis stage; progressively decreases from F0 to F4[146]F1: 237±61F4: 124±44F2: 220±58Biopsy-proven MASLD cohorts (N=1,398)N=836 (93%) ≥192×109/LN=67 (7%) ≥192×109/LLower platelet count was associated with higher HCC risk; Compared with 192×109/ L, platelet count 100×109/L was associated with adjusted HR 11.2 (95% CI 3.81–32.7) for HCC development[154]N=338 (68%) <192×109/LN=157 (32%) <192×109/LPlatelet-containing noninvasive fibrosis scoresMASLD ≤F2 (N=82) and F3–F4 (N=48), transient elastography-basedFIB-4: 1.03 (0.78–1.38);FIB-4: 6.29 (3.81–8.74);Higher in F3–F4; Plateletcontaining surrogates of fibrosis burden[147]APRI: 0.29 (0.23–0.41)APRI: 1.35 (0.63–2.17)Circulating platelet activation biomarkersMASLD ≤F2 (N=82) and F3–F4 (N=48), transient elastography-basedsP-selectin: 136.5 (107.3–175.8) ng/mL;sP-selectin: 309.5 (263.8–356.0) ng/mL;Positively associated with advanced fibrosis; higher in F3–F4 than ≤F2[147]urinary 11-dh-TXB2: 1,908 (857–3,258) ng/g creatinineurinary 11-dh-TXB2: 9,581 (7,125–13,628) ng/g creatininePlatelet activity and functional changesMASLD ≤F2 (N=82) and F3–F4 (N=48), transient elastography-basedResting surface P-selectin: no significant difference;Resting surface P-selectin: no significant difference;Early adhesive/procoagulant phenotype;[147]TRAP-6-induced surface P-selectin: no significant reduction vs. controls;TRAP-6-induced surface P-selectin: significantly lower than controls;Later activated-but-hyporesponsive phenotypeResting αIIbβ3 activation: not increased vs. controls; lower than F3–F4;Resting αIIbβ3 activation: increased vs. controls and MASLD ≤F2;TRAP-6-induced αIIbβ3 activation: reduced vs. controls;TRAP-6-induced αIIbβ3 activation: reduced vs. controls;Resting Annexin V-positive platelets: increased vs. controls and F3–F4;Resting Annexin V-positive platelets: not increased as in ≤F2;TRAP-6-induced Annexin V positivity: highest among groups, consistent with a procoagulant platelet phenotype;TRAP-6-induced Annexin V positivity: lower than MASLD ≤F2;Adhesion under flow: increased vs. controlsAdhesion under flow: increased vs. controlsPlatelet–leukocyte aggregatesMASLD ≤F2 (N=82) and F3–F4 (N=48), transient elastography-basedPlatelet–leukocyte heteroaggregates not increasedPlatelet–monocyte aggregates lower at rest and after TRAP-6; platelet–neutrophil aggregates lower after stimulationNot positively associated with fibrosis severity overall; lower in F3–F4 than ≤F2[147]Intrahepatic platelet depositionplatelets/mm2Liver biopsy specimens from patients with MASH (n=17) vs. controls (n=14)N.A.N.A.Negative association with intrahepatic NET burden (P=0.008); no fibrosis-stage– resolved data available[123]platelet aggregates/mm2Liver biopsy specimens from patients with MASH (n=17) vs. controls (n=14)N.A.N.A.Negative association with intrahepatic NET burden (P=0.038); aggregates larger in MASH than controls; no fibrosis-stage–resolved data availableIntrahepatic platelet depositionMean platelet count/field median (range)Biopsy-proven MASLD (N=24)F0-F1: 23 (9.6–47.3)N.A.Not associated with fibrosis severity; positively associated with histologic activity and intrahepatic NET formation[97]F2: 20.5 (15.2–44.7)Platelet macroaggregates median (range)Biopsy-proven MASLD (N=24)F0-F1: 13 (0–90)N.A.F2: 33.5 (3–105)Intrahepatic NET burdenLiver biopsy specimens from patients with MASH (n=17) vs. controls (n=14)N.A.N.A.Present in 94.1% of MASH biopsies; positively associated with steatosis, ballooning, lobular/portal inflammation, NAS, and stage in univariate analysis; the association with stage did not remain independently significant after multivariable adjustment[123]Biopsy-proven MASLD (N=24)N.A.N.A.Positively associated with NAS, inflammation scores, and intrahepatic platelet accumulation, but not with fibrosis severity; no fibrosis-stage–resolved data reported[97]Intrahepatic recruitment site/mode*Experimental MASH (diet-induced mouse models)In steatosis/borderline MASH, hepatic platelet recruitment involves hyaluronan–CD44 binding; in later established MASH with fibrosis in the model, platelet GPIbα becomes a key mediator of platelet–Kupffer cell interaction and disease maintenanceStage-linked mechanism in experimental MASH; not human fibrosis-stage resolved[14]This table summarizes stage-linked platelet-related variables across liver fibrosis progression in MASLD, together with their reported associations with histologic severity, fibrosis stage, and liver-related outcomes. Variables are shown according to the original grouping strategy of each source study; therefore, the granularity of fibrosis staging differs across rows, and N.A. indicates that fibrosis-stage–resolved data were not available in the verified source study. In terms of intrahepatic platelet deposition, methodological differences between studies should be noted: In the study by Miele et al., single platelets and macroaggregates were quantified as CD41-positive dots per high-power field across 10 fields at 400× magnification, with macroaggregates defined as >2 positive dots; in the study by Arelaki et al., platelets and platelet aggregates were quantified as CD42b-positive elements per mm² across the whole biopsy surface. Because these definitions are not equivalent, direct numerical comparisons between the two studies should be interpreted with caution.11-dh-TXB2, urinary 11-dehydro-thromboxane B2; APRI, AST-to-platelet ratio index; CI, confidence interval; CD44, cluster of differentiation 44; FIB-4, fibrosis-4 index; GPIbα, glycoprotein Ib alpha; HCC, hepatocellular carcinoma; HR, hazard ratio; MASH, metabolic dysfunction-associated steatohepatitis; MASLD, metabolic dysfunction-associated steatotic liver disease; N.A., not available; NAS, Nonalcoholic Fatty Liver Disease activity score; NET, neutrophil extracellular trap; sP-selectin, soluble P-selectin; TRAP-6, thrombin receptor-activating peptide-6; αIIbβ3, integrin alpha IIb beta 3.*The row “Intrahepatic recruitment site/mode” summarizes mechanistic evidence derived mainly from experimental MASH models rather than human fibrosis-stage–resolved histologic data. Stage-linked platelet-related variables across fibrosis progression in MASLD patients and their associations with histological activity, fibrosis stage, and liver-related outcomes This table summarizes stage-linked platelet-related variables across liver fibrosis progression in MASLD, together with their reported associations with histologic severity, fibrosis stage, and liver-related outcomes. Variables are shown according to the original grouping strategy of each source study; therefore, the granularity of fibrosis staging differs across rows, and N.A. indicates that fibrosis-stage–resolved data were not available in the verified source study. In terms of intrahepatic platelet deposition, methodological differences between studies should be noted: In the study by Miele et al., single platelets and macroaggregates were quantified as CD41-positive dots per high-power field across 10 fields at 400× magnification, with macroaggregates defined as >2 positive dots; in the study by Arelaki et al., platelets and platelet aggregates were quantified as CD42b-positive elements per mm² across the whole biopsy surface. Because these definitions are not equivalent, direct numerical comparisons between the two studies should be interpreted with caution. 11-dh-TXB2, urinary 11-dehydro-thromboxane B2; APRI, AST-to-platelet ratio index; CI, confidence interval; CD44, cluster of differentiation 44; FIB-4, fibrosis-4 index; GPIbα, glycoprotein Ib alpha; HCC, hepatocellular carcinoma; HR, hazard ratio; MASH, metabolic dysfunction-associated steatohepatitis; MASLD, metabolic dysfunction-associated steatotic liver disease; N.A., not available; NAS, Nonalcoholic Fatty Liver Disease activity score; NET, neutrophil extracellular trap; sP-selectin, soluble P-selectin; TRAP-6, thrombin receptor-activating peptide-6; αIIbβ3, integrin alpha IIb beta 3. The row “Intrahepatic recruitment site/mode” summarizes mechanistic evidence derived mainly from experimental MASH models rather than human fibrosis-stage–resolved histologic data. Fibrosis progression is also accompanied by qualitative platelet remodeling rather than quantitative decline alone. In MASLD with low fibrosis (≤F2), platelets show increased adhesion under flow and a larger Annexin V-positive subpopulation, which is consistent with a more adhesive and procoagulant phenotype [147]. In advanced fibrosis/cirrhosis (F3–F4), circulating markers of in vivo platelet activation, including soluble P-selectin and urinary 11-dehydro-thromboxane B2, are increased, whereas agonist-induced responses are attenuated, including reduced TRAP-6-induced P-selectin upregulation and impaired αIIbβ3 activation [147]. Together, these findings support a shift from an early adhesive/procoagulant phenotype to a later activated-but-hyporesponsive phenotype. In the same cohort, platelet–leukocyte aggregates were not increased overall and were even lower in advanced fibrosis/cirrhosis, indicating that not all platelet-related variables change in parallel across disease progression [147]. In contrast, currently available human biopsy data do not support a parallel stage-dependent increase in intrahepatic platelet deposition [97]. In the study by Miele et al. [97], intrahepatic single platelets and platelet macroaggregates were positively associated with steatosis, lobular inflammation, ballooning, overall histological activity, and intrahepatic NET formation but not with fibrosis severity. These findings suggest that hepatic platelet accumulation is more closely linked to necroinflammatory activity than to fibrotic stage itself [97]. NET-related human evidence likewise appears to reflect inflammatory histology more strongly than fibrosis stage alone does. In human MASH liver tissue, intrahepatic NET burden was positively associated with steatosis, ballooning degeneration, portal and lobular inflammation, the Nonalcoholic Fatty Liver Disease activity score (NAS), and stage according to univariate analysis, but the association with stage did not remain independently significant after multivariable adjustment, likely reflecting collinearity between fibrosis stage and inflammatory histologic components [123]. Overall, the available human data suggest that some platelet-related variables, such as platelet count, FIB-4, APRI, soluble P-selectin, and urinary 11-dehydro-thromboxane B2, track advanced fibrosis more closely, whereas others, including intrahepatic platelet deposition, platelet–leukocyte aggregates, and NET burden, may better reflect inflammatory activity, local immune crosstalk, or stage-specific remodeling. When platelet-driven mechanisms may matter most during fibrosis progression A key unresolved question is when platelet-driven mechanisms are most likely to be active during fibrosis progression. Integrating the available human and preclinical evidence, platelet-driven mechanisms may be more relevant from inflammatory steatosis/borderline MASH to early or precirrhotic fibrogenesis when platelet counts are still relatively preserved and platelets retain adhesive and procoagulant features. This stage preference is also compatible with current human data suggesting that platelet-related signals are more prominent in inflammation-rich phenotypes, particularly those characterized by higher NAS, lobular/portal inflammation, and low-grade endotoxemia with platelet TLR4-related inflammatory activation [97]. However, direct human longitudinal evidence resolving the transitions from F0 to F4 remains limited. In contrast, in advanced fibrosis/cirrhosis (F3–F4), decreasing platelet counts together with qualitative platelet remodeling suggest that platelet-driven pathways may no longer operate in the same way as they do in earlier inflammatory stages. This pattern is more consistent with chronic stimulation and functional reprogramming in advanced disease than with platelet quiescence. In this context, platelets may remain biologically relevant, but thrombocytopenia and hyporesponsive signaling are likely to modify how strongly platelet-driven pathways contribute to the progression of fibrosis. Platelets and MASLD-associated hepatocarcinogenesis Mechanistic links between platelets and MASLD-associated HCC HCC is one of the most clinically consequential complications of MASLD. Importantly, MASLD-related HCC does not arise exclusively in patients with cirrhosis; a meaningful proportion of these patients have noncirrhotic disease [148,149]. Mechanistically, platelets may contribute to MASLD-associated hepatocarcinogenesis through at least two partially overlapping routes. One plausible route is indirect, through which platelets promote the development of HCC via the “fibrosis/cirrhosis” route. As discussed above, platelets may amplify steatohepatitis, fibrogenic remodeling, and the chronic inflammatory milieu from which HCC emerges. In experimental MASH, aspirin plus clopidogrel attenuated steatosis and necroinflammation and prevented progression to HCC, whereas platelet GPIbα was identified as a key mediator of MASH and subsequent liver cancer [14]. A second route is more direct, namely, that platelets and platelet-derived mediators may influence carcinogenic processes beyond fibrosis alone. In MASLD-associated liver cancer models, platelets restrained tumor growth through P2Y12-dependent CD40L release [150], indicating that platelet-mediated immunomodulation in MASLD-HCC is bidirectional rather than uniformly protumorigenic. In addition to MASLD-specific models, broader HCC studies have shown that platelet releasates, especially those of TGF-β, can promote HCC cell proliferation through the suppression of Krüppel-like factor 6 (KLF6) [151]; moreover, platelet-derived TGF-β1 can promote epithelial–mesenchymal transition and metastasis through HCC cell autophagy via the AMPK/mTOR pathway [152], and PMPs can increase the invasion and migration of HCC cells [153]. These latter studies are not MASLD-specific, but they support the plausibility that platelets may directly regulate tumor cell behavior as well as the inflammatory-fibrotic background in which HCC develops. Platelet count as a clinical risk stratification marker for MASLD-HCC Among currently available human data, platelet count provides a clinical signal for MASLD-related hepatocarcinogenesis. In the multicenter CLIONE subanalysis by Fujii et al. [154], 1,398 patients with biopsy-confirmed MASLD were followed for a median of 4.6 years, corresponding to 8,874 person-years, and 37 developed HCC. When a baseline platelet count cutoff of 192×10⁹/L was used, the lowercount group had a markedly higher HCC rate than the higher-count group (6.7% vs. 0.4%) [154]. Using a three-knot cubic-spline model, Fujii et al. [154] further suggested a nonlinear inverse association between platelet count and HCC risk. With 192×10⁹/L set as the reference, a platelet count of 100×10⁹/L corresponded to an unadjusted hazard ratio (HR) of 7.37, whereas 230×10⁹/L corresponded to an unadjusted HR of 0.66, indicating that the risk increased more sharply at lower platelet counts than it declined at higher counts [154]. This signal was not confined to overt cirrhosis. In the same cohort, 80% of F0–F2 HCC patients had platelet counts less than 192×109/L, supporting the value of the platelet count for identifying higher-risk patients even before advanced fibrosis is established [154]. Directionally consistent evidence is available from another Japanese MASLD cohort in which a platelet count <115×109/L was revealed as a risk factor for HCC development [155]. Taken together, these findings suggest that the platelet count is best interpreted as an integrated clinical marker that captures several processes relevant to hepatocarcinogenesis, including fibrosis burden, portal-hypertensive/thrombopoietic changes, and broader disease activity. ASPIRIN TREATMENT FOR MASLD Given the growing evidence linking platelet activation to inflammation, fibrosis and hepatocarcinogenesis in MASLD, antiplatelet agents, particularly aspirin (acetylsalicylic acid), have emerged as potential therapeutic strategies to attenuate MASLD progression (Table 2). Aspirin not only inhibits platelet activation by irreversibly inhibiting cyclooxygenase-1 activity and reducing TXA2 synthesis [156] but also has pleiotropic effects involving metabolic regulation and anti-inflammatory effects [157]. Table 2.Summary of preclinical and clinical evidence for aspirin/antiplatelet therapy in MASLDAuthor, yearStudy designTreatment regimenImpact on steatosisImpact on inflammation & fibrosisImpact on HCCPossible mechanismsKey findingsRefsMalehmir et al., 2019Preclinical studiesMouse models (CD-HFD; WD-HTF); Nbeal2−/− miceDAPT, aspirin + clopidogrel (Asp-Clo)Reversed steatosis; Prevented metabolic dysregulationAttenuated liver damage, inflammation, and fibrosis; Reduced immune cell infiltrationPrevented the transition from MASH to HCCInhibition of platelet activation, prevents platelet-GPIbα mediated recruitment of immune cells to the liverPlatelet cargo (α-granules) and GPIbα are critical for MASH/HCC; Asp-Clo effectively blocked hepatic immune cell influx and tumorigenesis[14]Fujita et al., 2008Preclinical studiesRat models (CDAA diet; HF/HC diet)SAPT, Aspirin (150 mg/ kg/day), Ticlopidine (100 mg/ kg/day) or Cilostazol (100 mg/kg/day) for 16/12 weeksAttenuated steatosis and liver triglyceride contentSuppressed liver inflammation and fibrosis (reduced collagen content)Not assessedSuppression of oxidative stress and downregulation of PDGF expressionAll three antiplatelet drugs attenuated liver steatosis, inflammation and fibrosis in the CDAA model; Cilostazol was the most effective agent in this study[161]Simon et al., 2024Clinical studies (RCT)Phase 2 RCT; Adults with MASLD without cirrhosis (N=80)SAPT, Aspirin 81 mg/day vs. Placebo for 6 monthsSignificant reduction in liver fat content (mean difference -10.2% by MRS; -3.7% by MRI-PDFF)Reduced ALT/AST and iron-corrected T1 scores; Reduced VCTE scoreNot assessedNot investigated6 months of low-dose aspirin significantly reduced hepatic fat quantity compared with placebo[166]Harm et al., 2025Clinical studies (prospective)Prospective cohort; MASLD patients (N=51)SAPT, Aspirin 100 mg/day for 6 months; DAPT, Aspirin 100 mg/day and a P2Y12 inhibitor (clopidogrel 75 mg/day, or ticagrelor 90 mg twice daily, or prasugrel 10 mg/day) for 6 monthsReduced liver volume, hepatic fat, and iron deposition (MRI)Reduced plasma levels of profibrotic/inflammatory chemokinesNot assessedSuppression of platelet reactivity and reduction in the release of platelet-derived proinflammatory chemokines, reduction of hepatic iron deposition and ferroptosis, and improvement of lipid metabolismReduced platelet aggregation correlates with attenuation of MASLD features; DAPT showed more pronounced effects than SAPT[162]Simon et al., 2019Clinical studies (prospective)Prospective cohort; Biopsy-proven MASLD (N=361)SAPT, daily aspirin use (self-reported & updated prospectively)Not assessedLower odds of MASH (aOR 0.68; 95% CI 0.37–0.89) and fibrosis (aOR 0.54; 95% CI 0.31–0.82), lower risk of incident advanced fibrosis (aHR 0.63; 95% CI 0.43–0.85)Not assessedNot investigatedAspirin use was associated with a duration-dependent (greatest benefit with ≥4 years use) reduction in the risk of fibrosis progression; Nonaspirin NSAIDs did not show similar protection[163]Schwarzkopf et al., 2018Clinical studies (prospective)Prospective cohort; Patients undergoing coronary angiography (N=505)SAPT, aspirin or P2Y12 receptor antagonist (clopidogrel, prasugrel, or ticagrelor); DAPT, aspirin + P2Y12 receptor antagonistNo significant difference in CAP (steatosis assessment)Inverse association between antiplatelet use and liver fibrosis (OR 0.67; 95% CI 0.51–0.89)Not assessedPlatelet-derived PDGF-β is a key driver of fibrosis; however, serum PDGF-β levels were not affected by antiplatelet agents in this studyUse of antiplatelet agents was independently associated with a lower prevalence of liver fibrosis in patients at high cardiovascular risk[164]Huang et al., 2025Clinical studies (prospective)Prospective cohort (Nurses’ Health Study II); US female nurses (N=53,490)SAPT, aspirinAspirin use associated with increased risk of incident MASLD (HR 1.17; 95% CI 1.05–1.29)Not assessedNot assessedNot investigatedLong-term aspirin use was associated with a modest increase in incident MASLD risk[171]Ahn et al., 2026Clinical studies (retrospective)Retrospective cohort (Korean National Health Insurance Service database); MASLD patients (N=1,723,435).SAPT, aspirin useNot assessedNot assessedReduced HCC risk in MASLD after propensity score matching (adjusted subdistribution HR 0.86; 95% CI 0.75–0.99); similar inverse association observed in UK Biobank genetic risk analysis (adjusted subdistribution HR 0.47; 95% CI 0.29–0.76)Not investigatedAspirin use was associated with lower HCC risk in MASLD in a nationwide cohort and a salicylic-acid genomic risk analysis[169]Complementary genetic risk analysis in UK BiobankAnson et al., 2024Clinical studies (retrospective)Retrospective cohort (TriNetX global database); MASLD patients (N=84,384 matched)Antiplatelet agents (including aspirin monotherapy and combination therapy) ≥1 yearNot assessedNot assessedReduced incidence of HCC (aspirin monotherapy, HR 0.46; 95% CI 0.32–0.64)Not investigatedAntiplatelet use was associated with reduced incidence of hepatic and extra-hepatic cancers in MASLD[168]Lee et al., 2023Clinical studies (retrospective)Retrospective cohort (Taiwan’s National Health Insurance Research Database); MASLD patients (N=89,027)SAPT, daily aspirin usage (≥90 days)Not assessedNot assessedReduced risk of HCC (aHR 0.48; 95% CI 0.37–0.63)Not investigatedBenefit was consistent in high-risk groups (older age, elevated ALT); Lower risk seen with ≥3 years of use[167]Huang et al., 2025Clinical studies (retrospective)Retrospective cohort (multi-institutional, Taiwan); MASLD patients (N=4,006 matched)SAPT, daily aspirin use (mean duration 4.59 years, with 97% on a daily dose of 75–100 mg)No significant improvement in non-invasive steatosis scores (HSI, NAFLD-LFS)No significant improvement in fibrosis scores (FIB-4, APRI)No significant reduction in HCC incidence was observed after 3 years of aspirin monotherapyNot investigatedDaily aspirin alone did not significantly reduce mortality or liver events over a 3-year period[172]Zeng et al., 2023Clinical studies (metaanalysis)Meta-analysis (11 studies for aspirin); 2,190,285 chronic liver disease patientsSAPT, aspirin useNot assessedNot assessedReduced HCC risk overall (HR 0.48; 95% CI 0.27–0.87)Not investigatedAspirin exhibited overall benefit in HCC, but not significant in studies accounting for concurrent statin and metformin use[170]Thongtan et al., 2022Clinical studies (metaanalysis)Meta-analysis (4 observational studies); MASLD patients (N=2,593)Aspirin and/or P2Y12 inhibitorsNot assessedAssociated with lower prevalence of advanced liver fibrosis (pooled OR 0.66; 95% CI 0.53–0.81)Not assessedNot investigatedSupports a protective association between antiplatelet therapy and advanced fibrosis; Limited by small number of studies[165]aHR, adjusted hazard ratio; ALT, alanine aminotransferase; aOR, adjusted odds ratio; APRI, aspartate aminotransferase to platelet ratio index; AST, aspartate aminotransferase; Asp-Clo, aspirin and clopidogrel; CAP, controlled attenuation parameter; CDAA, choline-deficient, L-amino acid-defined; CD-HFD, choline-deficient high-fat diet; CI, confidence interval; DAPT, dual antiplatelet therapy; FIB-4, fibrosis-4 index; GPIbα, glycoprotein Ib alpha; HCC, hepatocellular carcinoma; HF/HC diet, high-fat/high calorie diet; HR, hazard ratio; HSI, hepatic steatosis index; MASH, metabolic dysfunction–associated steatohepatitis; MASLD, metabolic dysfunction–associated steatotic liver disease; MRIPDFF, magnetic resonance imaging of the proton density fat fraction; MRS, magnetic resonance spectroscopy; N, number; NAFLD, nonalcoholic fatty liver disease; NAFLD-LFS, NAFLD-liver fat score; NSAID, nonsteroidal anti-inflammatory drug; OR, odds ratio; PDGF, platelet-derived growth factor; RCT, randomized clinical trial; SAPT, single antiplatelet therapy; VCTE, vibration-controlled transient elastography; WD-HTF, western-style diets with trans fat. Summary of preclinical and clinical evidence for aspirin/antiplatelet therapy in MASLD aHR, adjusted hazard ratio; ALT, alanine aminotransferase; aOR, adjusted odds ratio; APRI, aspartate aminotransferase to platelet ratio index; AST, aspartate aminotransferase; Asp-Clo, aspirin and clopidogrel; CAP, controlled attenuation parameter; CDAA, choline-deficient, L-amino acid-defined; CD-HFD, choline-deficient high-fat diet; CI, confidence interval; DAPT, dual antiplatelet therapy; FIB-4, fibrosis-4 index; GPIbα, glycoprotein Ib alpha; HCC, hepatocellular carcinoma; HF/HC diet, high-fat/high calorie diet; HR, hazard ratio; HSI, hepatic steatosis index; MASH, metabolic dysfunction–associated steatohepatitis; MASLD, metabolic dysfunction–associated steatotic liver disease; MRIPDFF, magnetic resonance imaging of the proton density fat fraction; MRS, magnetic resonance spectroscopy; N, number; NAFLD, nonalcoholic fatty liver disease; NAFLD-LFS, NAFLD-liver fat score; NSAID, nonsteroidal anti-inflammatory drug; OR, odds ratio; PDGF, platelet-derived growth factor; RCT, randomized clinical trial; SAPT, single antiplatelet therapy; VCTE, vibration-controlled transient elastography; WD-HTF, western-style diets with trans fat. Possible mechanisms Aspirin may influence MASLD through both platelet-dependent and platelet-independent mechanisms. On the platelet-dependent side, antiplatelet therapy (aspirin/clopidogrel) reduces intrahepatic platelet accumulation, decreases the frequency of platelet‒immune cell interactions, and suppresses the release of intrahepatic cytokines and chemokines, thereby limiting immune cell recruitment to the liver [14]. Mechanistic studies have also indicated that aspirin and P2Y12 inhibitors can reduce the formation of circulating platelet‒leukocyte aggregates, thereby inhibiting the proinflammatory phenotypic conversion and effector functions of monocytes and neutrophils [158]. Blocking the interaction between platelets and neutrophils can ameliorate the formation of NETs, thereby mitigating tissue damage [159]. In addition to the classical antiplatelet pathway, aspirin ameliorates MASLD through platelet-independent mechanisms. In vitro and in vivo studies have shown that aspirin upregulates the PPARδ-AMPK-PGC-1α signaling pathway, promoting fatty acid oxidation and inhibiting lipid biosynthesis, thereby improving hyperlipidemia and hepatic steatosis [160]. Furthermore, aspirin modulates adipose tissue inflammation, improves insulin resistance, and may ameliorate metabolic dysfunction through salicylate-induced mitochondrial biogenesis [157]. Preclinical evidence These findings provide preclinical support for the therapeutic potential of aspirin in MASLD. In a choline-deficient high-fat diet-induced MASH mouse model, treatment with aspirin combined with clopidogrel not only reversed steatosis and necroinflammation but also successfully prevented the transition from MASH to HCC by inhibiting GPIbα-mediated platelet activation [14]. However, because this was a dual-antiplatelet regimen, the observed benefit cannot be assigned to aspirin alone. Additionally, in a choline-deficient, L-amino acid-defined diet-induced rat model, multiple antiplatelet agents—including aspirin, ticlopidine, and cilostazol—attenuated hepatic steatosis, inflammation, and fibrosis, with cilostazol showing the strongest effect [161]. These findings suggest that the benefit observed in this model is more likely to reflect a broader antiplatelet class effect than a mechanism unique to aspirin. Taken together, the results of preclinical studies support biological plausibility but also indicate that the effects attributed to aspirin may arise from multiple overlapping mechanisms rather than platelet inhibition alone. Clinical evidence Multiple observational studies suggest a favorable association between aspirin use and MASLD-related outcomes, although causal interpretation should remain cautious. A prospective magnetic resonance imaging (MRI)-based study revealed that antiplatelet therapy was associated with decreases in liver volume, iron deposition, and hepatic fat content. This was accompanied by reduced levels of proinflammatory and profibrotic chemokines [162]. However, patients receiving antiplatelet therapy in that study differed in coronary disease burden and concomitant therapies [162]. A prospective cohort study of 361 patients with biopsy-confirmed MASLD revealed that daily aspirin use was associated with less severe histologic features and a lower risk of progression to advanced fibrosis during follow-up, with a stronger association after a longer duration of use [163]. In another prospective cohort of patients who underwent coronary angiography, antiplatelet agent use was independently associated with a lower incidence of liver fibrosis [164]. A subsequent meta-analysis also reported a lower pooled incidence of advanced fibrosis among antiplatelet users, although the authors emphasized the small number of included studies and the observational nature of the evidence [165]. A recently published phase II randomized clinical trial provided high-quality evidence for the efficacy of aspirin. In adults with MASLD without cirrhosis, compared with placebo, 6 months of daily low-dose aspirin reduced the hepatic fat content according to both magnetic resonance spectroscopy (MRS) and magnetic resonance imaging of the proton density fat fraction (MRI-PDFF) and increased the proportion of participants who achieved at least a 30% relative reduction in hepatic fat [166]. These data support a short-term anti-steatotic effect of aspirin. However, these steatosis-related outcomes are more plausibly explained, at least in part, by platelet-independent metabolic actions affecting fatty acid oxidation and lipogenesis. Observational studies suggest that aspirin use may be associated with a lower risk of MASLD-related HCC. Two large retrospective cohort studies from Taiwan and the USA both demonstrated that daily aspirin use was associated with a reduced risk of HCC development in patients with MASLD, with this protective effect being particularly pronounced in high-risk populations such as older adults and those with impaired liver function [167,168]. More recently, Ahn et al. [169] supported this association in two complementary datasets. In the Korean National Health Insurance Service cohort, aspirin use was associated with a lower risk of HCC in MASLD after propensity score matching (adjusted subdistribution HR 0.86; 95% confidence interval [CI] 0.75–0.99), and a similar signal was observed in the UK Biobank when a salicylic acid genomic risk score was used as a proxy for aspirin exposure (adjusted subdistribution HR 0.47; 95% CI 0.29–0.76) [169]. Although this design does not replace randomized evidence, the addition of a genetic proxy analysis makes the overall signal less dependent on treatment-selection patterns within a single observational cohort. Meta-analyses have further confirmed the association between aspirin and reduced risk of developing HCC [170]. Controversies and challenges Despite the substantial evidence supporting the benefits of aspirin, inconsistent reports exist. A prospective cohort analysis based on the Nurses’ Health Study II suggested that aspirin use was associated with a modestly increased risk of incident MASLD, which might be confounded by the more complex baseline comorbidities among aspirin users [171]. Additionally, another study indicated that while aspirin might reduce HCC risk, aspirin use alone did not significantly reduce all-cause mortality or liver-related events in MASLD patients over a three-year follow-up period [172]. These conflicting results suggest that the efficacy of aspirin may be influenced by factors such as dosage, duration of use and patient baseline characteristics. A study by Malehmir et al. [14] suggested that the key to platelet-driven MASH progression lies in their adhesion to immune cells and the release of granule cargo rather than mere platelet aggregation. Starting from the mechanism underlying the dependency of platelets on aspirin, future antiplatelet strategies can focus on platelet–immune cell interaction pathways that are selectively amplified within inflammatory microenvironments, including interference with the GPIbα–Mac-1 axis, neutralization of specific granule-derived mediators such as CXCL4 and TGF-β, modulation of NET formation [159,173,174], and inhibition of PMA formation [131]. Compared with the global suppression of platelet activation, such pathway- or interaction-selective interventions may be more effective at attenuating inflammatory cascades while relatively sparing essential hemostatic functions. The greatest challenge in applying antiplatelet therapy to MASLD patients is the risk of bleeding [175,176]. Patients with advanced liver disease often present with coagulopathy, thrombocytopenia, and esophageal/gastric varices, where the use of aspirin or P2Y12 inhibitors may increase the risk of gastrointestinal bleeding. Future clinical trials should focus on identifying subgroups that would benefit most patients (such as MASH patients with concurrent cardiovascular risk [164]) and exploring antiplatelet strategies that minimize the impact on hemostasis. CONCLUSIONS MASLD should be viewed not only as an isolated hepatic condition but also as the liver manifestation of a systemic metabolic disorder in which platelets may act as relevant pathogenic mediators. Activated platelets may translate systemic signals—including hyperglycemia, oxidized lipids, and gut-derived metabolites—into local hepatic inflammation and fibrosis. This creates a vicious cycle in which metabolic stress fuels platelet activation, which in turn exacerbates liver pathology. Therefore, future therapeutic paradigms must move beyond liver-centric approaches to integrate metabolic control with targeted antiplatelet interventions. Specifically, strategies that decouple thrombo-inflammation from hemostasis may represent a promising avenue to improve hepatic outcomes in patients with MASLD. Abbreviations serotonin adenosine diphosphate advanced glycation end products adenosine triphosphate branched-chain amino acids a guanine nucleotide exchange factor for Ras-associated protein 1 carboxyalkylpyrrole-phosphatidylethanolamine derivatives guanosine diphosphate hepatocellular carcinoma hepatocyte growth factor hepatic stellate cells α-ketoisovaleric acid lipid droplets lipopolysaccharide liver sinusoidal endothelial cells metabolic dysfunction-associated steatohepatitis metabolic dysfunction-associated steatotic liver disease neutrophil extracellular trap oxidized low-density lipoprotein oxidized choline glycerophospholipid phenylacetylglutamine protease-activated receptor 4 proprotein convertase subtilisin/kexin type 9 platelet-derived growth factor-β phospholipase A2 platelet-derived microparticles P-selectin glycoprotein ligand-1 reactive oxygen species sphingosine-1-phosphate serum amyloid A1 type 2 diabetes mellitus transforming growth factor-β toll-like receptor trimethylamine-N-oxide thrombospondin-1 thromboxane A2 Footnotes Authors’ contributions Bingjie Ye was responsible for the conception, design, and drafting of the manuscript. Chengfu Xu was responsible for the conception, provided critical revision of the manuscript and approved the final manuscript. Acknowledgements This work was supported by Noncommunicable Chronic Diseases-National Science and Technology Major Project (No. 2024ZD0530100), National Natural Science Foundation of China (82370574), and the Key Research and Development Program of Zhejiang Province (2024C03153). We thank all the previous studies for supporting the findings. We also acknowledge BioRender.com, which was used to create the figures. Conflicts of Interest The authors have no conflicts to disclose. REFERENCES