Intrahepatic Platelets Orchestrate Microenvironmental Remodelling and Cellular Crosstalk in Liver Fibrosis

ABSTRACT

Platelets are increasingly recognized as active regulators of liver fibrosis, yet the relationship between systemic thrombocytopenia and intrahepatic platelet behaviour remains poorly understood. In selected etiological and stage‐specific contexts, systemic thrombocytopenia may coexist with intrahepatic platelet enrichment and activation, making circulating platelet counts insufficient to reflect local fibrotic activity. This review synthesizes current evidence to define platelet dynamics along the circulation–spleen–liver axis and their multicellular crosstalk within the fibrotic microenvironment, aiming to clarify how platelets integrate into a self‐reinforcing fibrogenic network and identify potential therapeutic strategies beyond correcting systemic thrombocytopenia. Systemic thrombocytopenia and intrahepatic platelet enrichment are not mutually exclusive; this paradox renders circulating platelet counts a poor surrogate for intrahepatic fibrotic activity. In the liver, platelets integrate thrombotic, inflammatory, angiogenic, and regenerative signals via mediator release (TGF‐β1, PDGF‐BB, CXCL4) and interactions with hepatocytes, Kupffer cells (KCs), liver sinusoidal endothelial cells (LSECs), and hepatic stellate cells (HSCs). Platelet‐hepatocyte crosstalk exerts opposing effects: injury amplification versus regeneration promotion. Platelet‐KC interactions are pro‐inflammatory and pro‐fibrotic, yet include CLEC4F‐mediated phagocytosis that limits excessive platelet accumulation. LSEC capillarization creates a pro‐thrombotic niche, where platelet adhesion activates NF‐κB, provides P‐selectin docking for leukocytes, and drives microthrombosis. Platelet‐HSC communication is directly pro‐fibrotic via coordinated TGF‐β1 and PDGF‐BB signalling, reinforced by bidirectional loops. Therapeutically, correction of peripheral thrombocytopenia should be distinguished from investigational strategies aimed at modulating pathogenic intrahepatic platelet activity.

Lay Summary

• Platelets are increasingly recognized as active regulators of liver fibrosis, not only through their classical role in haemostasis but also by coordinating inflammatory, thrombotic, angiogenic and regenerative responses within the liver microenvironment.

• Systemic thrombocytopenia does not necessarily indicate reduced platelet activity in the liver; instead, low circulating platelet counts may coexist with intrahepatic platelet enrichment and activation, making peripheral platelet counts an insufficient marker of local fibrotic activity.

• In the fibrotic liver, platelets interact with hepatocytes, KCs, LSECs, and HSCs to promote cellular crosstalk; therefore, future therapeutic strategies should distinguish correction of peripheral thrombocytopenia from targeted modulation of pathogenic intrahepatic platelet activity.

Abbreviations

BDL

bile duct ligation

CCl₄

carbon tetrachloride

CLD

chronic liver disease

CXCL4

C‐X‐C motif chemokine ligand 4

DAMPs

damage‐associated molecular patterns

ECM

extracellular matrix

GPIbα

glycoprotein Ibα

GPIIb/IIIa

glycoprotein IIb/IIIa

HSCs

hepatic stellate cells

IL‐6

interleukin‐6

KCs

Kupffer cells

LSECs

liver sinusoidal endothelial cells

MASH

metabolic dysfunction‐associated steatohepatitis

MASLD

metabolic dysfunction‐associated steatotic liver disease

NF‐κB

nuclear factor kappa B

PDGF

platelet‐derived growth factor

PDGF‐BB

platelet‐derived growth factor‐BB

PDGFR‐β

platelet‐derived growth factor receptor‐β

PI3K

phosphoinositide 3‐kinase

S1P

sphingosine 1‐phosphate

siRNA

small interfering RNA

SPOCK1

secreted protein acidic and rich in cysteine‐related modular calcium‐binding protein 1

TGF‐β1

transforming growth factor‐β1

THBS1

thrombospondin‐1

THBS2

thrombospondin‐2

TPO

thrombopoietin

VEGF

vascular endothelial growth factor

VWF

von Willebrand factor

1. Introduction

Liver fibrosis is a dynamic wound‐healing response to chronic liver injury, characterized by progressive extracellular matrix (ECM) accumulation, destruction of liver parenchyma, and eventual progression to cirrhosis and liver failure [1]. Clinically, fibrosis critically drives portal hypertension, hepatic decompensation, and hepatocellular carcinoma. The current lack of effective anti‐fibrotic therapies underscores the need to identify early regulatory mechanisms and actionable therapeutic targets.

Beyond their classical role in haemostasis, platelets have increasingly emerged as key modulators of immunity and inflammation [2]. In chronic liver disease (CLD), platelets undergo pronounced quantitative and qualitative remodelling, manifesting as thrombocytopenia, altered activation thresholds, and dysregulated interactions with vascular and immune cells [3, 4, 5]. Importantly, platelets are not merely passive biomarkers of disease severity but are active participants in multiple stages of liver pathology, including post‐injury inflammation, immune cell recruitment, hepatic stellate cell (HSC) activation, fibrogenesis, hepatocyte regeneration, and even tumour progression [6, 7].

Recent advances in in vivo imaging and single‐cell technologies have uncovered previously unrecognized interactions between platelets and key liver cell populations, including hepatocytes, Kupffer cells (KCs), liver sinusoidal endothelial cells (LSECs), and HSCs, reshaping our understanding of platelet behaviour within the hepatic microenvironment [8, 9, 10]. In selected experimental and clinical contexts, platelets may contribute to fibrogenesis through multiple mechanisms: release of diverse bioactive mediators (e.g., transforming growth factor‐β1 [TGF‐β1], platelet‐derived growth factor [PDGF], and chemokines), adhesive interactions with the sinusoidal endothelium, and dynamic crosstalk with immune and stromal cells [11]. However, two central questions persist: (i) What drives the phenotypic heterogeneity and spatial redistribution of platelets along the circulation‐spleen‐liver axis, creating the paradox of systemic thrombocytopenia alongside intrahepatic platelet enrichment? (ii) How do platelet‐centred interaction networks with hepatic cells orchestrate the balance between pro‐fibrotic remodelling and parenchymal repair during chronic injury?

A particularly important and distinctive concept emphasized in this review is the spatial redistribution model. Unlike many previous reviews that primarily considered platelets in relation to thrombocytopenia, bleeding risk, coagulation imbalance, liver regeneration, or systemic inflammation, the present review focuses on the compartmentalized behaviour of platelets within the injured liver (Table 1). In certain fibrotic contexts, platelets may be progressively trapped, aggregated, and activated within the hepatic microenvironment despite the presence of systemic thrombocytopenia. To address these issues, we discuss platelet dynamics along the circulation–spleen–liver axis, platelet recruitment and retention within the hepatic microenvironment, and platelet‐mediated interaction networks involving hepatocytes, KCs, LSECs, and HSCs. Through systematic stratification of existing evidence across diverse study types and disease contexts, this review aims to provide a balanced framework for understanding how platelet‐centred multicellular networks may regulate fibrotic remodelling and tissue repair. We also highlight platelet‐centred pathways that may guide future mechanistic and translational studies.

TABLE 1.

Summary of key findings from previous studies and comparison with the present review.

Authors	Main content	References
Hayashi et al.	Mechanisms and management of cirrhotic thrombocytopenia.	[12]
Mitchell et al.	Multifactorial pathophysiology of thrombocytopenia in CLD.	[13]
Peck‐Radosavljevic	Mechanisms and TPO‐based management of CLD‐related thrombocytopenia.	[14]
Lim et al.	Pathophysiology and periprocedural management of liver disease‐related thrombocytopenia.	[15]
Lisman et al.	Rebalanced haemostasis and bleeding/thrombosis management in cirrhosis.	[16]
Fernández‐Garibay et al.	Thrombocytopenia mechanisms in portal hypertension and CLD.	[17]
Scharf	Pathophysiology and haemostasis‐guided management in liver disease thrombocytopenia.	[18]
Saab and Brown Jr	Management strategies and emerging TPO receptor agonists for CLD thrombocytopenia.	[19]
Aiolfi et al.	Antiplatelet therapy, immune modulation, and reduced HBV‐related liver cancer risk.	[20]
Chauhan et al.	Dual platelet roles in liver inflammation, fibrosis, and regeneration.	[11]
Kurokawa et al.	Platelet regulation of liver inflammation, regeneration, fibrosis, and cancer.	[6]
Lisman et al.	Platelet modulation of liver inflammation, fibrosis, regeneration, and progression.	[5]
Mussbacher et al.	Platelet regulation of liver injury, regeneration, fibrosis, and cancer progression.	[21]
Nowatari et al.	Platelets modulate liver fibrosis, regeneration, and acute injury responses.	[22]
Ramadori et al.	Platelets as immune sentinels in liver injury, fibrosis, regeneration, and cancer.	[23]
Balaphas et al.	Platelets and extracellular vesicles in liver regeneration, inflammation, fibrosis, and progression.	[24]
Dalbeni et al.	Platelet activation promotes MASLD progression through inflammation and HSC activation.	[25]
Chen et al.	Context‐dependent platelet interactions in liver fibrosis, injury, regeneration, and cancer.	[26]
Casari et al.	Macrophages and platelets contributions to fibrosis and HCC progression.	[27]
Liang et al.	Platelet‐mediated regeneration and activation‐related graft injury in liver transplantation.	[28]
Lisman et al.	Platelet‐driven regeneration mechanisms and therapeutic risks.	[29]
Chen et al. (The present review)	Fibrosis‐centred platelet redistribution and intrahepatic platelet‐cell interactions in targeted anti‐fibrotic intervention.	—

2. Dynamic Changes of Platelet Levels in the Circulatory System and Spleen During Fibrosis

To properly dissect the complex behaviour of platelets in liver fibrosis, it is essential to first establish their physiological baseline. Traditionally, platelets are recognized primarily for their canonical roles in haemostasis and thrombosis, serving as frontline responders to vascular injury [30]. Under normal conditions, the spleen acts as the principal dynamic reservoir for circulating platelets, thereby maintaining platelet homeostasis [31]. This reservoir function is tightly regulated by sympathetic nervous system activity: adrenergic stimulation induces splenic contraction and promotes platelet release into the circulation. In response to stress (exercise, hypoxia, or acute inflammation), catecholamine‐driven splenic contraction rapidly increases circulating platelet counts [32, 33]. Beyond storage, emerging evidence suggests that the spleen can also actively produce platelets under inflammatory conditions. For example, in sepsis, sympathetic‐dependent extramedullary megakaryocyte generation in the spleen gives rise to platelet subsets with different immunophenotypes (e.g., high CD40 ligand expression), highlighting the spleen's previously underappreciated role in shaping platelet‐mediated immune responses [34].

During liver fibrosis progression, platelet number, activity, and functional status dynamically change in both the circulation and the spleen [35]. Multiple studies show that peripheral blood platelet counts progressively decline across CLDs of various etiologies [36, 37, 38], with thrombocytopenia commonly occurring in advanced fibrosis [39]. Concurrently, increased platelet pooling within the spleen is well documented in cirrhotic patients [40, 41]. The mechanisms of thrombocytopenia in cirrhosis have been extensively reviewed [13, 14, 17, 42]. Briefly, they involve impaired platelet production (reduced hepatic thrombopoietin [TPO] synthesis), increased spleen retention (hypersplenism due to portal hypertension), and enhanced peripheral destruction. Additional factors, including immune‐mediated clearance and myelosuppression from alcohol or viral infection, further drive persistent thrombocytopenia. Importantly, these mechanisms primarily reflect systemic alterations and do not explain the altered spatial distribution of platelets within the fibrotic liver—a distinction that has received insufficient attention.

Beyond quantitative changes, platelet function displays aetiology‐specific signatures—a nuance often overlooked. In hepatitis C virus‐related advanced liver fibrosis, platelet counts and aggregation are reduced, with impaired TRAP‐induced activation; yet global haemostasis assessed by thromboelastography often remains preserved, suggesting that increased per‐platelet activation may serve as a compensatory mechanism to maintain hemostatic balance [43]. In alcoholic liver disease, platelet count, activation, and aggregation are consistently reduced, with diminished expression of activation markers (activated glycoprotein IIb/IIIa [GPIIb/IIIa], P‐selectin, and CD63) indicating overt platelet dysfunction [44]. In stark contrast, patients with metabolic dysfunction‐associated steatotic liver disease (MASLD) show systemic platelet activation, evidenced by elevated urinary 11‐dehydro‐thromboxane B2 and plasma soluble P‐selectin levels, indicating enhanced adhesiveness and pro‐thrombotic activity [45]. Thus, during CLD progression, the number and functional status of circulating and splenic platelets provide a necessary but not sufficient baseline for understanding platelet abnormalities; one must consider aetiology‐specific signatures.

In cirrhosis, thrombocytopenia is not merely a laboratory abnormality but carries important clinical implications. Severe thrombocytopenia may increase bleeding risk during invasive procedures, while in decompensated cirrhosis it often reflects disease severity [46]. However, patients with cirrhosis frequently exist in a state of ‘rebalanced haemostasis’, where thrombocytopenia coexists with remodelling of the coagulation/anticoagulation system, indicating that platelet count alone poorly predicts bleeding risk [42, 47]. Moreover, peripheral platelet count does not necessarily mirror intrahepatic platelet behaviour. In selected disease contexts, systemic thrombocytopenia may coexist with platelet trapping, aggregation, and activation within the hepatic microenvironment, consistent with a spatial redistribution model. This supports a context‐dependent spatial redistribution model rather than a universal feature of all CLDs. Clinically, two main questions therefore arise: (i) Does peripheral platelet count truly reflect intrahepatic platelet behaviour in a given disease aetiology and stage? (ii) When and how should thrombocytopenia be intervened upon? Current management strategies focus on elevating platelet counts in high‐risk settings, using platelet transfusion or TPO receptor agonists (e.g., avatrombopag, lusutrombopag) [19]. Crucially, these interventions primarily correct systemic thrombocytopenia and reduce bleeding risk, but do not directly address aberrant intrahepatic platelet enrichment and activation. Platelet transfusion and TPO receptor agonists may increase the availability of circulating platelets, but they do not directly target intrahepatic platelet sequestration within the hepatic microenvironment, platelet adhesion to LSECs, or platelet‐mediated interactions with KCs and HSCs. Potential thrombotic risks, including portal vein thrombosis, should also be considered, particularly in advanced liver disease [48]. In certain contexts, expanding the systemic platelet pool could theoretically provide more platelets available for intrahepatic recruitment; however, the clinical relevance of this possibility remains uncertain and warrants further investigation. Conceptually, this raises the possibility of combining strategies that increase circulating platelet availability with those that specifically restore intrahepatic platelet function. The latter would require a distinct therapeutic paradigm, potentially involving aetiology‐ and stage‐specific approaches targeting platelet‐liver sinusoidal endothelial cell adhesion, platelet activation and degranulation, platelet‐immune cell interactions, or platelet‐derived profibrotic signalling pathways.

Collectively, circulating platelet measures alone are insufficient to accurately reflect platelet behaviour in fibrotic liver disease. However, the relationship between systemic platelet counts and intrahepatic platelet accumulation appears to differ by aetiology and disease stage. In early‐stage disorders such as MASLD, circulating platelet counts and activity remain largely unchanged, whereas increased platelet recruitment and activation are observed within the liver [49]. This phenomenon is recapitulated in mouse models: in metabolic dysfunction‐associated steatohepatitis (MASH) models, enhanced intrahepatic platelet accumulation occurs despite normal peripheral platelet counts, underscoring that local microenvironmental cues dominate over systemic parameters [50]. These findings lead to a key conclusion: platelet dynamics during CLD are characterized by spatial redistribution rather than uniform systemic alterations (Figure 1). Therefore, a comprehensive assessment, integrating changes across the circulation, spleen, and liver as an interconnected process, is not merely helpful but essential. The available evidence supporting this spatial redistribution model is summarized according to disease aetiology, disease stage, and model system in Table 2.

FIGURE 1.

Context‐dependent platelet dynamics during CLD. Platelet homeostasis in CLD is governed by both systemic and local processes that may vary according to disease aetiology and stage. In the circulation and spleen, thrombocytopenia may result from reduced platelet production, splenic sequestration, and accelerated clearance, accompanied by aetiology‐dependent alterations in platelet function. Within the liver, normal sinusoidal patrolling may be progressively replaced by platelet adhesion, activation, and aggregation as fibrosis remodels the sinusoidal microenvironment. LSEC capillarization, basement membrane deposition, and reduced blood flow promote intrahepatic platelet retention, creating a fibrosis‐associated platelet trap. Thus, in selected disease contexts, systemic thrombocytopenia may coexist with intrahepatic platelet trapping and activation.

TABLE 2.

Stratified evidence for platelet alterations in chronic liver diseases according to disease context and model system.

Disease context/model	Disease stage	Species	Evidence source/study design	Principal platelet‐related findings	Principal limitation(s)	References
Multiple etiologies (viral hepatitis, MASLD, etc.)	Advanced fibrosis/Cirrhosis	Human	Human observational; retrospective database‐based nested case–control design	Decline in peripheral platelet count correlating with advanced fibrosis.	Selection bias; limited causal inference.	[36]
Chronic hepatitis C	Ishak fibrosis stage 0–6	Human	Human observational; retrospective longitudinal biomarker analysis	Reduced platelet counts and platelet‐derived growth factors during fibrosis progression.	Small sample size; circulating biomarkers only; limited mechanistic insight.	[37]
Chronic hepatitis C	Mild/Advanced fibrosis	Human	Human observational; prospective cohort study with ex vivo platelet function assessment	Reduced platelet counts and aggregation in advanced fibrosis.	Small sample size, short follow‐up, and ex vivo platelet function assays.	[51]
MASLD	F0‐F4	Human	Human observational; multicentre retrospective biopsy‐confirmed cohort study	Platelet count decline with increasing fibrosis severity.	Selection bias due to exclusion of patients with hepatic decompensation, limiting generalizability.	[38]
MASLD	F0‐F4	Human	Human observational; cross‐sectional platelet phenotyping study with ex vivo functional assays	Stage‐specific platelet dysfunction: early hyperactivity; advanced thrombocytopenia and impaired responsiveness	Platelet activation biomarkers are not entirely platelet‐specific.	[45]
MASLD	F0‐F4	Human	Human observational; single‐centre retrospective biopsy‐based study	Intrahepatic platelet aggregation despite unchanged peripheral platelet parameters.	Limited sample size; findings require validation in larger cohorts.	[49]
Alcohol‐related liver disease	Cirrhosis	Human	Human observational; case–control platelet function study	Decline in peripheral platelet count with impaired platelet function.	Small sample size; subgroup analyses were underpowered.	[44]
MASH	Liver fibrosis	Mouse	Animal in vivo mechanistic study with intravital imaging and genetic/pharmacologic platelet targeting; exploratory human tissue and pilot cohort evidence	Intrahepatic platelet activation despite unchanged peripheral platelet parameters.	Limited clinical validation of mechanistic findings.	[50]
CCl4	Liver fibrosis	Mouse	Animal in vivo mechanistic study with platelet TGF‐β1 deletion/depletion.	Activated platelet redistribution to injured hepatic tissue with TGF‐β1 release, promoting liver fibrosis.	Mouse CCl4 model only; no human validation and only partial reduction of fibrosis.	[52]
BDL	Liver fibrosis	Mouse	Animal in vivo mechanistic study with intravital microscopy and platelet/P‐selectin blockade.	BDL‐induced hepatic platelet aggregation/activation promotes leukocyte recruitment and microvascular dysfunction.	Preclinical BDL mouse data; no human validation; platelet activation was inferred, not directly measured.	[53]

3. Platelet Aggregation and Activation in the Liver From Physiological Patrol to a Fibrosis‐Related ‘Platelet Trap’

Unlike the spleen, the liver has not traditionally been considered a platelet reservoir. This is largely due to the unique structure of LSECs: they are highly permeable, allowing circulating platelets to traverse the hepatic sinusoids freely, which prevents the formation of a stable resident platelet population [54]. Accordingly, mouse studies under normal conditions show that platelets typically patrol the hepatic sinusoids and transiently accumulate within intact hepatic structures without forming thrombotic aggregates [55]. This dynamic behaviour is thought to contribute to immune surveillance and sinusoidal homeostasis. Through transient interactions with LSECs and KCs, platelets sense vascular integrity, respond rapidly to microinjury, and help clear pathogens and circulating debris [56]. Moreover, platelet‐derived factors support endothelial function and modulate local immune responses, underscoring a broader role for platelets beyond haemostasis in the hepatic microenvironment [21]. Recent studies further show that platelets can actively interact with immune cells by forming integrin‐ and tetraspanin‐rich membrane junctions, promoting intercellular communication and regulating inflammatory responses [57]. Together, these findings emphasize that platelet patrol is not passive cycling but an active surveillance mechanism capable of shaping local immune dynamics.

In CLD, however, this physiological baseline can be disrupted, and hepatic platelet behaviour becomes considerably more complex. Accumulating evidence indicates that fibrotic microenvironmental remodelling actively drives platelet adhesion, aggregation, and activation within the hepatic sinusoids [26]. Key structural changes include sinusoidal capillarization, loss of endothelial fenestrae and deposition of a basement membrane, which disrupts the normal high‐permeability state of the sinusoids [58]. Under these pathophysiological conditions, sinusoidal blood flow becomes progressively sluggish and heterogeneous. This altered hemodynamic environment favours platelet adhesion to the capillarized endothelium, followed by aggregation into microthrombi that partially occlude the sinusoidal lumen [59]. Collectively, these local microvascular changes transform the fibrotic liver into an efficient ‘platelet trap’, characterized by regional platelet enrichment and activation.

This concept is strongly supported by recent animal studies; its broader applicability to human CLDs remains to be determined and is likely influenced by disease aetiology‐ and stage‐specific factors. In the classic carbon tetrachloride (CCl₄)‐induced liver fibrosis model, repetitive hepatocyte injury and ECM deposition disrupted LSECs and microcirculatory blood flow, facilitating early platelet recruitment and activation within the injured liver [60, 61]. Notably, increased intrahepatic platelet activation, as evidenced by enhanced co‐expression of CD41 with activation markers such as JON/A and P‐selectin, occurs at early injury time points, suggesting that platelet engagement is not merely a secondary event but may actively participate in fibrogenesis [52]. Similar phenomena are observed in bile duct ligation (BDL)‐induced cholestasis and MASH models, where sinusoidal capillarization, loss of fenestrae, and impaired perfusion promote intravascular platelet aggregation and microthrombus formation [62, 63]. Intravital microscopy studies have demonstrated markedly increased platelet adhesion in both hepatic sinusoids and postsinusoidal venules of BDL mice, with widespread intravascular platelet aggregates forming functional microthrombi that cause microcirculatory disturbances, potentially exacerbating fibrotic progression [53]. Taken together, these observations suggest that intrahepatic platelets are more likely active participants in fibrosis rather than passive responders.

Underlying these hemodynamic and structural changes, LSECs and hepatocytes upregulate adhesion molecules (e.g., von Willebrand factor [VWF] and P‐selectin) and coagulation factors (e.g., fibrinogen), further enhancing platelet activation and aggregation [64]. In selected inflammatory or fibrotic settings, platelet‐derived factors may promote local platelet recruitment and retention within the liver. Thrombospondin‐1 (THBS1) is a well‐established pro‐fibrotic matricellular protein that drives liver fibrosis progression through multiple mechanisms, including direct activation of the TGF‐β pathway and modulation of the hepatic immune and inflammatory microenvironment [65]. In MASH models, platelet‐derived Thbs1 deficiency markedly attenuates intrahepatic platelet accumulation and activation, as evidenced by reduced CD41‐positive staining, downregulation of P‐selectin expression, and decreased secretion of the platelet activation‐associated chemokine C‐X‐C motif chemokine ligand 4 (CXCL4/PF4), accompanied by diminished platelet‐macrophage interactions [66]. Notably, under physiological conditions, THBS1 deficiency does not impair coagulation or haemostasis, suggesting that THBS1 may preferentially regulate intrahepatic platelet aggregation and activation in the context of chronic liver inflammation [66]. Mechanistically, THBS1 promotes platelet adhesion, recruitment, and aggregation through CD36, CD47, integrin‐associated signalling, and protection of VWF from a disintegrin and metalloproteinase with thrombospondin motifs 13‐mediated cleavage. These observations establish THBS1 as a key platelet‐associated mediator that couples thrombotic platelet responses to pro‐fibrotic signalling, thereby amplifying pathological tissue remodelling [67].

Beyond THBS1, its paralog thrombospondin‐2 (THBS2) remains much less defined [68]. THBS2 has also been implicated in liver fibrosis and is predominantly produced by activated HSCs and upregulated during fibrosis progression, where it promotes fibrogenesis through an autocrine toll‐like receptor 4‐focal adhesion kinase/TGF‐β signalling axis [69]. Kozumi et al. identified THBS2 as one of the most significantly upregulated molecules in advanced liver fibrosis among MASLD patients through transcriptome analysis [70]. These findings support a role for THBS2 as a fibrosis‐associated matrix mediator; however, its potential involvement in intrahepatic platelet recruitment, activation, or platelet‐mediated profibrotic signalling remains largely unexplored and mechanistically undefined. Given their structural similarities and enrichment in the fibrotic matrix, THBS2 may represent a candidate mediator worthy of investigation in the platelet‐fibrosis interface. However, this hypothesis requires further experimental validation. Of note, an earlier study reported that platelets themselves do not contain detectable THBS2, although THBS2 deficiency altered platelet function, implying that THBS2 may influence platelet biology indirectly rather than as a platelet‐derived effector [71]. Therefore, unlike THBS1, for which platelet‐related evidence is more established, THBS2 should currently be regarded as a candidate mediator and knowledge gap in the platelet‐fibrosis axis. Future studies should systematically compare THBS1 and THBS2 in the same fibrosis models to dissect their overlapping versus distinct roles in platelet regulation.

4. Multicellular Platelet Crosstalk in the Fibrotic Liver Microenvironment

4.1. Platelet‐Hepatocyte Interaction

In liver fibrosis, hepatocytes act both as initiators of injury and the key determinant of fibrotic and regenerative outcomes. Accumulating evidence indicates that platelet‐hepatocyte interactions exert microenvironment‐dependent effects. These interactions are largely indirect, mediated by platelet adhesion receptors (e.g., glycoprotein Ibα [GPIbα], CD44, glycoprotein VI) and their ligands in the liver microenvironment, with LSECs and KCs serving as critical intermediaries [64, 72]. Upon activation, platelets increase surface P‐selectin expression and engage glycoprotein receptors and integrin‐related pathways, which promotes their retention near injured parenchyma and amplifies local mediator release [23]. Notably, compared with the well‐characterized platelet‐HSC and platelet‐KC axes, direct platelet‐hepatocyte adhesion remains relatively understudied, a gap worth filling given that physical proximity alone may be sufficient to trigger functionally relevant hepatocyte signalling.

During chronic liver injury, platelet binding to hepatocytes contributes to amplification of hepatocellular stress and propagation of a pro‐fibrotic microenvironment. Damaged hepatocytes undergoing necrosis, apoptosis, or cellular stress release damage‐associated molecular patterns (DAMPs) and procoagulant signals, which in turn promote platelet adhesion and aggregation [73, 74]. This hepatocyte‐driven danger signal establishes a permissive environment for sustained platelet activation, reinforcing inflammatory signals, and supporting HSC activation [75]. Although direct platelet–hepatocyte binding has not been as extensively characterized as platelet interactions with other hepatic cell types, experiments show that physical proximity alone is sufficient to trigger hepatocyte signalling cascades through receptor engagement and enhance hepatocyte responsiveness to inflammatory cytokines, thereby impacting liver fibrosis progression [76, 77]. Thus, platelets act as amplifiers of hepatocyte injury, establishing a positive feedback loop that perpetuates both parenchymal damage and non‐parenchymal cell activation.

Conversely, platelets consistently promote hepatocyte survival and proliferation, especially during acute injury and liver regeneration (Figure 2). Mechanistically, platelet‐derived signals activate key pro‐survival and proliferative pathways in hepatocytes, notably the phosphoinositide 3‐kinase (PI3K)‐Akt and extracellular signal‐regulated kinase 1/2 pathways. Akt activation is a central anti‐apoptotic mechanism during liver regeneration, protecting hepatocytes from TGF‐β‐induced apoptosis [78]. Co‐culture studies demonstrate that platelets rapidly induce Akt and extracellular signal‐regulated kinase 1/2 activation and directly enhance hepatocyte DNA synthesis and proliferation [77]. In vivo, platelet transfusion or thrombocytosis accelerates hepatocyte proliferation, increases the liver‐to‐body weight ratio, and enhances regenerative signalling, accompanied by elevated hepatocyte growth factor, insulin‐like growth factor 1, and Akt phosphorylation [77, 79, 80, 81]. Moreover, platelet‐driven regeneration appears partially independent of KC‐derived signals, as thrombocytosis can rescue delayed liver regeneration following KC depletion [79]. Thus, platelets also act as potent drivers of hepatocyte regeneration, which may counterbalance fibrotic progression by restoring functional parenchyma. This functional duality—injury amplification versus regenerative support—likely depends on the injury context (chronic versus acute) and the spatiotemporal dynamics of platelet activation. We suggest that future studies directly compare these two faces of platelet‐hepatocyte crosstalk and examine whether the same platelet‐derived molecules elicit hepatoprotective or hepatotoxic effects depending on hepatocyte activation state, as this duality is critical for designing platelet targeted therapies.

FIGURE 2.

Platelet‐hepatocyte interactions in acute liver regeneration and chronic liver fibrosis. Platelet‐hepatocyte interactions exert context‐dependent effects in liver injury. In acute liver injury, platelet recruitment and close apposition to hepatocytes promote hepatocyte survival and proliferation via platelet‐derived mediators, including HGF and IGF‐1, activating the PI3K‐AKT and ERK1/2 signalling pathways. These signals support liver regeneration and functional recovery. In chronic liver injury, damaged hepatocytes release DAMPs and procoagulant signals, which enhance platelet adhesion and retention within the injured microenvironment. Platelet activation amplifies hepatocellular stress and sustains a positive feedback loop involving inflammatory and pro‐fibrotic signalling. This crosstalk is largely indirect, mediated by LSECs and KCs, and ultimately promotes HSC activation and liver fibrosis.

4.2. Platelet‐Macrophage Interaction

Under normal physiological conditions, the liver harbours the largest pool of macrophages in the body, and these cells play a crucial role in the initiation and progression of chronic liver injury [82]. In response to injury, distinct macrophage subsets are activated and recruited by injury‐associated signals, determining the nature, duration, and outcome (repair versus fibrosis) of the inflammatory response [83]. Platelets serve as important regulatory nodes in the dynamic process, albeit in context‐dependent and seemingly contradictory ways. On one hand, platelet‐derived CXCL4 and microparticles promote macrophage differentiation, thereby shaping the fibrotic microenvironment [84, 85, 86]. Concomitantly, the anatomical positioning of hepatic macrophages—resident KCs and recruited monocyte‐derived macrophages—within the sinusoidal lumen makes them among the first cells to sense platelet sequestration in the injured liver.

Intravital imaging reveals that even under basal conditions, platelets establish transient ‘touch‐and‐go’ interactions with KCs via platelet GPIbα and constitutively presented VWF on KCs; upon inflammatory challenge, these contacts stabilize and form platforms for immune activation and cellular recruitment [56]. In metabolism‐related liver injury, platelet enrichment in the hepatic sinusoid depends on KC‐mediated trapping. For instance, early ischemia–reperfusion studies show that platelets preferentially adhere and aggregate with KCs, amplifying local injury [87], and their interaction determines the severity of injury in steatotic livers [88]. Similarly, in MASLD (including MASH), platelet colonization relies on KCs via the hyaluronic acid‐CD44 axis [50]. Notably, while these findings consistently support a pro‐inflammatory and pro‐fibrotic role of platelet‐macrophage interactions, it is important to note that macrophages can also adopt reparative, fibrosis‐resolving phenotypes depending on their polarization state.

Despite this plasticity, when directly instructed by platelets, macrophages predominantly adopt a pro‐inflammatory and pro‐recruitment phenotype, characterized by enhanced cytokine/chemokine production that prolongs inflammatory pressure and promotes fibrosis progression [86, 89]. This persistent milieu continuously recruits circulating monocytes, bolsters the macrophage pool, and sustains HSC activation, thereby constraining fibrotic resolution. As ECM progressively accumulates, the altered mechanical and biochemical properties of the hepatic microenvironment further stabilize activated HSC phenotypes, establishing a self‐reinforcing inflammatory‐fibrotic loop [90]. Breaking this feed‐forward loop may require early interference with platelet‐macrophage crosstalk, suggesting a possible but as yet underexplored therapeutic avenue.

Beyond amplifying inflammation, platelets also engage in regulatory interactions with KCs that may produce opposing effects. Jiang et al. demonstrated that desialylated platelets are preferentially recognized and phagocytosed by KCs via CLEC4F, a clearance mechanism that limits excessive platelet accumulation [91]. In chronic liver injury and fibrosis, alterations in platelet glycosylation or activation status could modulate this KC‐platelet axis, reshaping the phagocytic burden and inflammatory signalling and indirectly influencing the pro‐fibrotic microenvironment. However, this hypothesis requires direct validation in fibrosis models. Given the conflicting roles of platelet‐KC interactions, future studies should directly examine how platelet desialylation or other post‐translational modifications tilt the balance between pro‐fibrotic and protective axes. Beyond clearance, KCs can serve as essential intermediaries in platelet‐driven hepatocyte signalling. Karakas et al. showed that KCs promote sinusoidal endothelial fenestrations and hepatocyte protrusions, thereby facilitating platelet GPIbα‐hepatocyte contact and hepatic TPO generation [92]. Together, these studies highlight the dual role of KCs as both effectors of platelet clearance and orchestrators of platelet‐dependent hepatic responses within the sinusoidal niche (Figure 3, left panel). Thus, targeting KC‐platelet interactions could offer context‐dependent therapeutic strategies for CLD—preserving homeostatic clearance and TPO‐generating functions while mitigating pathological inflammatory amplification.

FIGURE 3.

Platelet‐mediated crosstalk with macrophages and LSECs in CLD. Platelets interact with resident KCs through basal touch‐and‐go adhesion mediated by VWF‐GPIbα and, in disease contexts such as MASLD/MASH, via HA‐CD44‐dependent platelet sequestration. Platelet‐derived CXCL4 and microparticles promote macrophage differentiation toward a pro‐inflammatory phenotype, enhancing cytokine and chemokine production, monocyte recruitment, and sustained inflammatory pressure, ultimately contributing to LSEC capillarization, HSC activation, and fibrosis. In parallel, platelet‐LSEC interactions induce endothelial activation, NF‐κB signalling, and release of chemokines such as CCL2 and CXCL8, facilitating leukocyte recruitment and inflammation. Platelet‐derived S1P contributes to pathological angiogenesis. In terms of platelet homeostasis, KCs clear desialylated platelets via CLEC4F‐mediated phagocytosis and facilitate platelet‐hepatocyte contact, thereby supporting hepatic TPO generation.

4.3. Platelet‐LSEC Interaction

In the liver, LSECs line the low shear microcirculation channels and form a highly permeable vascular parenchymal interface, characterized by fenestrated and discontinuous vessel walls. This specialized structure maintains intrahepatic homeostasis and is involved in regulating inflammation and coagulation [93]. In this low‐shear setting, platelets avidly adhere to the sinusoidal endothelium in an integrin‐dependent manner, leading to platelet and endothelial activation and potentiating leukocyte recruitment [94]. During chronic liver injury, however, LSECs undergo ‘capillarization’ accompanied by inflammatory activation. They shift from a quiescent, anti‐adhesive phenotype toward a pro‐adhesive and prothrombotic state, marked by loss of fenestrae, basement membrane formation, upregulation of platelet‐adhesive pathways, and acquisition of vasoconstrictive functions [54]. This phenotypic switch echoes the endothelial dysfunction observed in other vascular beds (e.g., atherosclerosis), yet within the liver it establishes a distinctive thromboinflammatory niche. This switch promotes platelet recruitment and retention in the sinusoids, fostering a platelet‐rich microenvironment that supports local immunothrombosis. It may also help explain how intrahepatic platelet accumulation occurs even in the setting of systemic thrombocytopenia. In turn, platelet‐LSEC interactions amplify leukocyte recruitment, inflammatory signalling, and microvascular dysfunction, while disease‐associated angiocrine remodelling biases the vascular response toward pathological angiogenesis and fibrogenic progression [64, 95].

Mechanistically, platelet‐LSEC crosstalk (Figure 3, right panel) is driven by low‐shear‐adapted adhesion and paracrine signalling. Even transient platelet adhesion rapidly activates nuclear factor kappa B (NF‐κB) signalling in LSECs, promoting p50 nuclear translocation and eliciting dose‐dependent secretion of chemokine ligand 2 and chemokine ligand 8. Notably, mature LSECs lack P‐selectin under both physiological and inflammatory conditions; therefore, adherent platelets act as an exogenous P‐selectin substrate, providing critical docking sites for neutrophil and lymphocyte attachment. This mechanism is functionally validated by the marked attenuation of platelet‐driven neutrophil recruitment upon P‐selectin inhibition [94]. Therapeutically, targeting P‐selectin signalling or the initial platelet‐LSEC adhesion step could represent a promising strategy to mitigate neutrophil infiltration and immunothrombosis, though this warrants further validation in CLD models. Collectively, these findings suggest that platelet‐LSEC crosstalk may function as a vascular ‘gate’ linking immunothrombosis with liver fibrosis and angiogenesis, providing a potential mechanistic basis for future investigation of targeted interventions.

Beyond adhesion, platelets actively reprogram LSECs. Co‐culture studies with human LSECs show that platelets increase LSEC proliferation and stimulate interleukin‐6 (IL‐6) and vascular endothelial growth factor (VEGF) production, with sphingosine 1‐phosphate (S1P) signalling contributing to IL‐6 induction [96]. In chronic disease, sustained IL‐6/VEGF signalling and angiogenesis are frequently linked to fibrogenesis, suggesting that platelet‐driven LSEC activation may indirectly reinforce HSC activation and scar expansion by fostering an inflammatory angiogenic microenvironment [97, 98, 99, 100]. This paracrine axis represents an amplification loop that is distinct from the direct platelet‐macrophage interactions discussed earlier. In principle, modulation of S1P signalling could affect both pathways and may therefore have therapeutic relevance.

Beyond these paracrine effects, platelet‐LSEC crosstalk also drives structural remodelling in the form of intrasinusoidal microthrombi. Capillarized LSECs acquire a prothrombotic phenotype that promotes platelet adhesion and aggregation into platelet‐rich microthrombi, resulting in regional perfusion defects that transform endothelial injury into progressive fibrosis [101]. Given that these microthrombi are a downstream consequence of LSEC capillarization, reversing or preventing capillarization—rather than simply anticoagulating—may be a more fundamental approach to breaking the self‐perpetuating fibrotic loop.

4.4. Platelet‐HSC Interaction

Within the fibrotic liver microenvironment, platelet accumulation and activation enable sustained crosstalk with non‐parenchymal cells, especially HSCs—the central effector cells of liver fibrosis (Figure 4). Under fibrotic conditions, platelets act as a principal exogenous source of HSC activation cues, chiefly via α‐granule‐derived mediators that drive HSC transdifferentiation, proliferation, migration, and matrix production [22, 75, 102, 103]. Unlike the transient platelet‐macrophage interactions discussed above, platelet‐HSC communication is more directly pro‐fibrotic, with less contextual variation—a distinction that may explain why targeting this axis has yielded more consistent anti‐fibrotic effects in preclinical models.

FIGURE 4.

Platelet‐HSC interactions in liver fibrosis. In the fibrotic liver microenvironment, activated platelets accumulate in the sinusoids and release α‐granule‐derived mediators that promote HSC activation and fibrogenesis. Platelet‐derived TGF‐β1 activates the TGF‐βR/Smad2/3 pathway, whereas PDGF‐BB engages PDGFR‐β/PI3K‐Akt‐FOXM1 signalling to induce SPOCK1 expression and downstream integrin α5β1/PI3K‐Akt signalling, thereby enhancing HSC activation, proliferation, migration, and collagen deposition. CXCL4 may contribute to microenvironmental remodelling, although its mechanism remains incompletely defined. Activated HSCs further reinforce platelet pro‐fibrotic activity through prorenin‐bearing extracellular vesicles that stimulate the platelet (pro)renin receptor, leading to tissue factor‐dependent thrombin generation and TGF‐β release, thus forming a self‐amplifying loop. TGF‐β/Smad inhibition is shown schematically as suppressing TGF‐β1‐induced Smad2/3 phosphorylation and downstream transcriptional activation. Key knowledge gaps highlighted in the schematic include the potential involvement of THBS2 in platelet‐HSC signalling and the unresolved mechanism of CXCL4.

Mechanistically, the pro‐fibrotic effect of platelets on HSCs is not mediated by a single factor but by a coordinated signalling network. Platelet‐derived TGF‐β1 appears to be a key upstream driver of HSC phenotypic transformation. Experimental evidence shows that genetic deletion of platelet‐derived TGF‐β1 in mice suppresses the TGF‐β/Smad pathway in HSCs, attenuating their transdifferentiation into collagen‐secreting myofibroblasts and alleviating liver fibrosis [52]. Notably, interrupting this pathway has been validated as a viable anti‐fibrotic strategy. For example, Arbidol blocks TGF‐β1/Smad signalling and inhibits HSC activation [104]. This mechanism appears conserved across organs: in renal fibrosis models, platelet‐mimetic nanoparticles loaded with TGF‐β1 small interfering RNA (siRNA) (PMVs@PLGA) significantly slowed fibrosis progression [105]. From a translational perspective, Arbidol‐based strategies and similar siRNA approaches may provide interesting opportunities for future research in liver fibrosis, but their clinical applicability remains to be determined.

Importantly, platelet‐HSC communication is not unidirectional. Activated HSCs can reinforce platelet pro‐fibrotic activity by releasing extracellular vesicles carrying prorenin on their surface. This prorenin binds to the platelet prorenin receptor, triggering tissue factor‐dependent thrombin generation, which then drives the release of pro‐fibrotic TGF‐β. This creates a self‐amplifying loop that sustains HSC activation and accelerates fibrosis progression [106]. Thus, rather than being passive recipients, HSCs actively participate in a reciprocal circuit that stabilizes the fibrogenic niche.

Within this reciprocal network, platelet‐derived growth factor‐BB (PDGF‐BB) is another potent factor promoting HSC proliferation and migration. PDGF‐BB activates platelet‐derived growth factor receptor‐β (PDGFR‐β) signalling, driving HSC expansion and migration to injured areas, thereby accelerating fibrosis [107, 108]. Clinical observations support this mechanism: in patients with hepatitis C cirrhosis, activated HSCs with elevated PDGFR‐β expression are frequently found adjacent to sequestered sinusoidal platelets [109]. At the intracellular level, PDGF‐BB induces secreted protein acidic and rich in cysteine‐related modular calcium‐binding protein 1 (SPOCK1) overexpression in primary HSCs via the PI3K‐Akt‐forkhead box M1 pathway, and SPOCK1 further activates integrin α5β1/PI3K/Akt signalling to enhance HSC activation and collagen deposition [110]. Moreover, platelet depletion in Mdr2‐null mice (a spontaneous fibrosis model) normalizes hepatic PDGF‐β levels and attenuates pro‐fibrotic gene expression [111]. Collectively, the PDGF‐BB axis may offer a more tractable therapeutic target than TGF‐β1, given the pleiotropic toxicity of systemic TGF‐β blockade.

In addition to directly inducing HSC activation, platelets may shape the fibrotic microenvironment through inflammatory chemokine signalling [7]. In mouse models, hepatic Cxcl4 expression correlates with fibrotic burden, and Cxcl4 ^−/− mice exhibit significantly attenuated fibrosis compared with wild‐type controls [112]. Consistently, serum and intrahepatic CXCL4 protein levels are elevated in patients with advanced liver fibrosis [112]. This chemokine‐driven pathway appears parallel to, rather than synergistic with, the TGF‐β/PDGF axes, but whether CXCL4 acts directly on HSCs or indirectly via immune cells remains an open question. Another unresolved question is whether THBS2 is involved in platelet‐HSC signalling. Direct evidence supporting this hypothesis is currently lacking. This knowledge gap should be interpreted cautiously: although THBS1 has a well‐established role in TGF‐β activation and platelet‐related responses, comparable evidence for THBS2 in platelet regulation or platelet‐HSC crosstalk is not yet available. Therefore, THBS2 should currently be regarded as a potential matrix‐associated candidate rather than a defined component of the platelet‐HSC fibrotic network. Future studies are needed to determine whether THBS2 has any direct or indirect role in the interaction between HSC activation and intrahepatic platelet recruitment or aggregation.

From a therapeutic perspective, these findings collectively suggest that platelet reprogramming may be a therapeutic strategy that targets fibrosis at its source. Recent studies show that aspirin can reprogram platelet signalling to reduce the production of pro‐fibrotic factors, inhibit ryanodine receptor 2‐driven HSC activation, and modulate the hepatic microenvironment, thereby attenuating inflammation and fibrosis in preclinical models of CLD [113]. Notably, aspirin's effects are pleiotropic and not limited to platelet inhibition; whether its potential anti‐fibrotic effects are primarily through platelets or directly on HSCs remains to be fully clarified. Its clinical application in patients with cirrhosis requires careful consideration. Individualized risk–benefit assessment is particularly important, taking into account portal hypertension, prior gastrointestinal bleeding, platelet counts, disease stage, hepatic and renal function, and the overall bleeding risk before initiating antiplatelet or anticoagulant therapy. Therefore, rather than broadly targeting systemic platelet availability or activity, future therapeutic strategies may need to focus more specifically on intrahepatic platelet dysregulation, aiming to modulate pathological platelet recruitment, activation, and platelet‐cell interactions within the hepatic microenvironment.

5. Summary and Perspectives

In summary, this review highlights a paradigm shift in our understanding of platelets in liver fibrosis: from passive bystanders reflecting portal hypertension‐associated thrombocytopenia to active regulators of the hepatic microenvironment. A central concept is the spatial redistribution model, in which systemic thrombocytopenia can coexist with intrahepatic platelet enrichment and activation, indicating that circulating platelet counts do not necessarily reflect local platelet activity within the fibrotic liver. Within the fibrotic niche, platelets function as central hubs integrating thrombotic, inflammatory, angiogenic, and regenerative signals. Through direct release of TGF‐β1, PDGF‐BB, and CXCL4, and through dynamic crosstalk with HSCs, KCs, LSECs, and hepatocytes, platelets orchestrate a feed‐forward network that sustains HSC activation, immunothrombosis, microvascular remodelling, and ECM deposition. At the same time, context‐dependent interactions with hepatocytes underscore their dual role in both injury and regeneration.

Future studies should move beyond descriptive platelet counts toward functional and spatial phenotyping, leveraging single‐cell multi‐omics and intravital imaging to dissect platelet heterogeneity along the liver–spleen–circulation axis. Mechanistically, clarifying how platelet subsets switch between pro‐fibrotic and pro‐regenerative programmes will be essential. From a therapeutic perspective, no approved platelet‐directed anti‐fibrotic therapy currently exists for liver fibrosis, and correction of peripheral thrombocytopenia should not be equated with modulation of pathogenic intrahepatic platelet activity. Platelet transfusions and TPO receptor agonists have clinical value in increasing circulating platelet counts and reducing bleeding risk in some patients. However, they do not directly address platelet enrichment in the liver during fibrosis, platelet–LSEC adhesion, or platelet‐mediated interactions with HSCs and KCs. Their use in chronic liver disease and cirrhosis also requires careful consideration of individual thrombotic and bleeding risks, including portal hypertension, variceal status, gastrointestinal bleeding history, platelet counts, renal function, and disease stage. Therefore, future strategies should move beyond simply increasing systemic platelet availability or broadly inhibiting platelet activity and instead focus on disease‐ and stage‐specific modulation of pathological intrahepatic platelet functions. An integrated strategy that bolsters circulating platelet counts while concurrently modulating aberrant intrahepatic platelet signalling and function may be a more promising direction. Further validation is required before microenvironmental mediators are considered therapeutic targets. Ultimately, a deeper understanding of platelet functional heterogeneity and the signals governing platelet behaviour may refine the framework of liver fibrosis and identify candidate pathways for future platelet‐oriented therapies.

Author Contributions

Z.C. drafted the manuscript with contributions from H.L., Q.J., R.G., W.Z., W.W., Z.P., H.M., C.W.; W.C. and H.Y. contributed to the study conception and design and provided critical revisions to the manuscript; all authors reviewed, edited, and approved the final version for submission.

Funding

This work was supported by National Natural Science Foundation of China (82470646 [to W.C.], 82430023 [to H.Y.]), Young Talent Funds of Chinese Institutes for Medical Research (CX23YQB08 [to W.C.]), and the Beijing High‐Level Innovation and Entrepreneurship Talent Support Program‐Young Top Talent Projects (G202522109 [to WC]).

Conflicts of Interest

The authors declare no conflicts of interest.

Chen Z., Li H., Ji Q., et al., “Intrahepatic Platelets Orchestrate Microenvironmental Remodelling and Cellular Crosstalk in Liver Fibrosis,” Liver International 46, no. 9 (2026): e70848, 10.1111/liv.70848.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.