The TIME is composed of a diverse array of immune cell populations that infiltrate tumor tissues, including innate immune cells like macrophages, neutrophils, and dendritic cells, as well as adaptive immune cells including T and B lymphocytes. It also includes other cells such as hepatic stellate cells and cancer-associated fibroblasts. The interactions and functions of these cells collectively influence the role of TIME in cancer progression.30,31 HCC-derived exosomes further alter the TIME by modulating the polarization and function of these cellular components, thereby shifting immune responses from anti-tumorigenic to pro-tumorigenic, ultimately promoting tumor progression.
Innate immune cells
Macrophages
Macrophages are a crucial component of the innate immune system, exhibiting the capacity to be polarized by different microenvironment signals, enabling them to switch their phenotypes and functions. The polarization of these cells occurs through two major pathways: the classical pathway, leading to the predominance of M1 macrophages that exert anti-tumor activities, and the alternative pathway, resulting in M2 macrophages known for promoting tumorigenesis and tumor progression.32 Specifically, M1 macrophages facilitate tumor destruction by secreting cytotoxic molecules, including reactive oxygen species (ROS) and inducible nitric oxide synthase.33 Additionally, they support Th1 lymphocyte polarization and actively engage in the phagocytosis of tumor cells.34 In contrast, M2 macrophages suppress anti-tumor immune responses by expressing co-inhibitory molecules like PD-L1 and secreting cytokines with anti-inflammatory properties, such as IL-10, IL-6, and TNF-α,35,36 as well as by initiating the Th2 immune response. Additionally, M2 macrophages promote angiogenesis in HCC by secreting matrix metalloproteinases (MMPs) and vascular endothelial growth factor, facilitating metastasis through matrix remodeling and invasion.37 In the context of HCC, tumor-associated macrophages (TAMs) exhibit phenotypic shifts corresponding to different tumor progression stages, with early-stage M1 macrophages gradually giving way to M2 macrophages as the tumor advances, resulting in poorer patient outcomes.33,38 HCC-derived exosomes are critical in this transition, as they can induce TAMs to shift from M1 to M2 to create a tumor-friendly microenvironment.39
By triggering ferroptosis in M1 macrophages via exosomes, HCC cells can bypass surveillance by the immune system (Fig. 2). For example, miR-142-3p within exosomes derived from hepatitis B virus (HBV)-infected HCC cells triggers ferroptosis in M1 macrophages through increased levels of solute carrier family 3 member 2, potentially diminishing the immune response against tumors and aiding in tumor development.40 Given that HBV infection is a notable risk factor for HCC,41 this research contributes to understanding the pathogenesis of HBV-related HCC and offers fresh theoretical insights and potential treatment targets.
Apart from suppressing M1-type macrophages, exosomes originating from HCC dynamically modulate TAM polarization towards an M2 phenotype, thus facilitating tumor advancement (Fig. 2). Epigenetic modifications, which alter gene regulation by modifying DNA architecture, are increasingly recognized as key factors in cancer, including HCC. Exosomes carrying miRNAs, lncRNAs, and circRNAs significantly promote TAM differentiation toward the M2 polarized state.31,42 For instance, elevated levels of exosomal miR-452-5p present in HCC-originated exosomes encourage M2 polarization and HCC advancement by reducing TIMP3 levels, a typical anti-oncogene that prevents M2 macrophage polarization and triggers cell death in HCC cells.43,44 Similarly, a study by Yu and colleagues illustrated that highly expressed exosomal miR-21-5p originating from HCC attenuates MAPK signaling pathways in HCC cells through its attachment to the 3′-UTR of RhoB, which subsequently leads to an M2-like phenotype transformation in TAMs and promotes HCC progression.45
Furthermore, it has been demonstrated that lncRNAs and circRNAs within exosomes from HCC cells function as competitive endogenous RNA, preventing miRNA release from specific genes, thereby enabling macrophages to shift towards the M2 phenotype. Under hypoxic conditions, previous studies have shown that hypoxia-inducible factor (HIF)-1α enhances the transcription of the lncRNA hyaluronan-mediated motility receptor antisense RNA 1 (HMMR-AS1) and promotes exosome release by interacting with the regulatory region of lncRNA HMMR-AS1. The exosomal HMMR-AS1 is then internalized by macrophages, resulting in M2 macrophage polarization and accelerated HCC development through prevention of ARID3A degradation via miR-147a sequestration.46 Moreover, Ai et al. showed that exosome-packaged lncRNA HEIH, which is abundantly expressed in HCC, is transported to macrophages, triggering M2 macrophage polarization and subsequent enhancement of tumor progression through the HEIH/miR-98-5p/STAT3 pathway.47 Additionally, a study by Wang and colleagues demonstrated that elevated exosomal hsa_circ_0074854 can be internalized by TAMs to promote M2 polarization, leading to increased IL-10 expression and decreased TNF-α expression, which contributes to HCC progression. The study also found that hsa_circ_0074854 reduces the stability of the protein HuR and decreases ZEB1 expression, further promoting Epithelial-Mesenchymal Transition (EMT) in HCC cells.48 EMT is a fundamental cellular process in which epithelial cells lose their distinct traits and adopt a mesenchymal state that significantly boosts cancer cells’ capacity for migration and invasion.49
In contrast, reducing exosomes with growth-inhibiting cargo is a strategy for HCC to maintain proliferation. Gu et al. demonstrated that the quantity of exosomal miR-1-3p, which shows a marked reduction in HCC compared to the control group, can enter macrophages and regulate M2 macrophage polarization, thereby facilitating HCC progression. In a xenograft mouse model, tumors treated with miR-1-3p-loaded macrophages were significantly smaller and exhibited a marked increase in apoptosis, indicating the potential of miR-1-3p as a therapeutic target.50
Moreover, emerging evidence has confirmed that HCC-derived exosomes suppress T cell functions indirectly by inducing a phenotype shift in macrophages towards the M2 phenotype (Fig. 2). The mechanistic basis for this immunosuppressive effect is driven by increased expression of immune checkpoint proteins (TIM-3, CTLA4, PD-1) on T cells after macrophages are treated with HCC-derived exosomes, leading to T cell exhaustion and immunosuppression.51,52 Lysyl oxidase-like 4, an enzyme involved in extracellular matrix remodeling, exhibits high expression levels in HCC tissues. Secreted by HCC cells, exosomal lysyl oxidase-like 4 is predominantly internalized by hepatic macrophages, inducing PD-L1 expression by activating interferon (IFN)-mediated, STAT1-dependent pathways. This process leads to reduced cytotoxicity of CD8+ T cells against HCC cells, creating an immunosuppressive environment in HCC.53,54 Additionally, ER stress activation facilitated by exosomes has been identified in multiple tumor types, including HCC, and plays an essential role in promoting cancer advancement.55,56 Liu et al. discovered that under ER stress, exosomal miR-23a-3p is delivered from HCC cells to macrophages, causing M2-type polarization and PD-L1 expression. This upregulation of PD-L1, driven by PTEN suppression and PI3K/AKT pathway activation, ultimately induces CD8+ T cell apoptosis and reduces IL-2 secretion, thereby lowering the CD8+ T cell ratio.57 Similarly, elevated levels of GOLM1 have been intricately linked to HCC development and metastasis.58 Research indicates that overexpression of GOLM1 promotes PD-L1 stability and facilitates its incorporation into exosomes through CSN5-mediated deubiquitination in HCC. These exosomes are then transferred to macrophages, resulting in upregulated PD-L1 on TAMs. The elevated PD-L1 expression on TAMs suppresses CD8+ T cell function, characterized by reduced production of effector cytokines such as IFN-γ and granzyme B, upregulation of inhibitory receptors PD-1 and TIM-3, and a higher rate of apoptosis indicated by activated caspase 3.29 Furthermore, Sal-like protein 4 (SALL4) is a transcription factor known to play a crucial role in the oncogenesis and progression of various malignancies, including HCC.59 In a study by Yin and colleagues, it was found that SALL4 overexpression enhances the incorporation of miR-146a-5p into exosomes, which are then delivered to macrophages, promoting their transition to the M2 phenotype via NF-κB pathway stimulation. These exosome-induced M2 macrophages subsequently cause T cell exhaustion and dysfunction by upregulating inhibitory receptors on T cells and reducing cytokine secretion, such as IL-2 and TNF-α. Thus, targeting SALL4 or exosomal miR-146a-5p may serve as a potential therapeutic strategy for HCC.52
A recent study reveals that the ATP-adenosine pathway weakens CD8+ T cell function, promotes immunosuppression within the TIME, and induces resistance to PD-1 therapy. Exosomes from HCC cells contain circTMEM181 (circular RNA transmembrane protein 181), which is absorbed by macrophages and sponges miR-488-3p, leading to elevated CD39 expression. CD39 converts extracellular ATP into ADP and AMP, which are further converted into adenosine by CD73 in HCC cells. This interplay between high CD39 levels in macrophages and CD73 in HCC cells activates the ATP-adenosine pathway, resulting in excess adenosine, impairing CD8+ T cell function and promoting PD-1 resistance.60 Elucidating the specific mechanisms of these exosomes will help assess their potential as therapeutic targets, thereby enhancing anti-tumor immunity and improving patient outcomes.
Dendritic cells
Dendritic cells (DCs) are vital elements of the immune system, named for their distinctive dendritic morphology. Derived from hematopoietic stem cells in the bone marrow and maturing through a complex differentiation process, DCs are competent antigen-presenting cells that effectively recognize, capture, process, and present antigens, contributing significantly to the initiation and modulation of immune responses.61 Meanwhile, the impact of tumor-derived exosomes (TDEs) on DCs within the TIME exhibits a range of variable outcomes that substantially affect anti-tumor immunity.62
On one hand, DCs actively participate in the immune response by efficiently internalizing Tumor-Derived Exosomes (TEXs) carrying a variety of tumor antigens. Once internalized, these antigens are processed by DCs and subsequently presented to T cells, marking an essential step in initiating an immune response (Fig. 2).63,64 Research conducted by Rao and colleagues demonstrated that, after DCs uptake exosomes derived from HCC cells, there is an increased expression of surface markers including CD11c, MHC classes I and II, alongside costimulatory molecules such as CD80, CD86, and intercellular adhesion molecule-1 (ICAM-1), suggesting that TDEs can induce DC maturation and activation.64 In addition, HCC-specific antigens such as alpha-fetoprotein, glypican-3, and heat-shock protein 70 are known to be present in these TEXs, enhancing the ability to stimulate T cells and inhibit cancer cell proliferation by stimulating DC proliferation and differentiation.65 Antigen presentation by TEX-primed DCs is critical for activating T cells, particularly CD8+ T cells. Additionally, TEX-primed DCs help modify the tumor microenvironment to favor immune-mediated tumor suppression, evidenced by increased T-cell infiltration and a shift in cytokine profiles toward higher levels of immunostimulatory cytokines like IFN-γ and lower levels of immune-inhibitory factors like IL-10 and TGF-β.64
Conversely, multiple studies have shown that TEXs may also contribute to tumor development by inhibiting the differentiation, maturation, and function of DCs, thereby compromising their ability to mount an effective immune response (Fig. 2).62 A previous study observed that exosomes derived from breast cancer can target and interact with CD11b+ myeloid precursor cells in the bone marrow. By inducing IL-6 production and activating the STAT3 signaling pathway, these exosomes significantly inhibit the differentiation of myeloid precursor cells into DCs. Even when some precursor cells manage to differentiate into DCs in the presence of these exosomes, they lose their ability to mature properly and fail to effectively activate T cells, significantly enhancing the tumor’s ability to evade the immune system and promoting tumor progression.61 Similarly, Ning et al. observed that DCs treated with exosomes derived from lung and breast cancer cells exhibit pronounced immunosuppressive properties, including a significant downregulation of surface markers such as CD80, MHC-II, and CD86, alongside upregulation of immunosuppressive markers like PD-L1, and a reduced expression of chemokine receptors such as CCR7, thereby inhibiting their migration to lymph nodes and diminishing their capacity to activate T cells effectively. These combined effects substantially diminish the proliferation, differentiation, and function of T cells, allowing tumor cells to evade immune surveillance.66 Additionally, TEXs enriched with S100A9 molecules have been shown to impede DC maturation, as indicated by the decreased expression of DC maturation markers.67 A recent study revealed that exosomes enriched with S100A9 are significantly present in blood specimens of HCC patients, indicating that HCC may inhibit DC maturation via S100A9-rich exosomes, thereby contributing to immune escape.68
Moreover, studies have shown that exosomes originating from the HCC cell line (Huh7) are rich in lipids, including glycolipids and fatty acids.69 These fatty acid-laden TEXs significantly increase cytoplasmic lipid concentrations and activate the metabolic switch peroxisome proliferator-activated receptor α, resulting in the accumulation of lipid droplets and enhanced fatty acid oxidation, which ultimately drives DC metabolism toward mitochondrial oxidative phosphorylation and severely impairs their immune function. Dysfunctional DCs are unable to effectively carry out antigen processing and activation of T cells, thereby weakening their ability to stimulate T cells and leading to immune dysfunction.70 Additionally, studies have found that bone marrow-derived dendritic cells exposed to antigen-free TDEs display impaired functionality, resulting in the suppression of cytotoxic T cell function and encouraging the development of immunosuppressive regulatory T cells (Tregs).70 Therefore, targeting peroxisome proliferator-activated receptor α presents a strategic approach to enhance DC-based therapies in cancer treatment.
In conclusion, DCs treated with TDEs exhibit the potential for dual roles in both promoting and suppressing tumors. Initially, TEXs stimulate DC activation and enhance anti-tumor immunity. However, as the tumor progresses, TEXs shift DCs to an immunosuppressive phenotype, reducing antigen presentation and increasing immunosuppressive cytokine secretion, which contributes to tumor immune evasion.71,72 While HCC exosomes can stimulate DCs, the overall immunosuppressive tumor microenvironment remains a significant challenge. A nuanced approach is needed to balance the therapeutic potential of TEXs with their role in immune escape.
Natural killer cells
Natural killer (NK) cells are a crucial component of the innate immune system, known for their ability to swiftly respond to tumors and virally infected cells without prior sensitization.73 They distinguish “self” from “non-self” by utilizing a complex system of surface receptors, which include both activating and inhibitory types. When activating receptors, such as NKG2D, encounter stress-induced ligands on tumor cells, they trigger the release of cytotoxic granules containing perforin and granzyme, ultimately leading to cell death.74,75 On the other hand, inhibitory receptors like KIR and CD94/NKG2A bind to MHC class I molecules on healthy cells, sending inhibitory signals that prevent NK cell activation and protect normal tissues. NK cell activity against cancer is regulated by a balance between activating and inhibitory signals, ensuring selective targeting of threats while preserving normal cells.74–76
NK cell function is often compromised by an immunosuppressive microenvironment that alters their phenotype and reduces cytotoxic capabilities. This phenomenon is associated with a decrease in activating receptors like NKG2D and an increase in exhaustion markers, such as PD-1, CD96, Tim-3, and TIGIT.74,77 Additionally, tumor cells and other cells in the tumor microenvironment, including Tregs, myeloid-derived suppressor cells, and tumor-associated fibroblasts, release soluble factors (such as TGF-β, IL-10, PGE2, and IDO) that further suppress NK cell activity.74,78 Notably, HCC-derived exosomes can be absorbed by NK cells, potentially compromising their function by modulating activating or inhibitory receptors, thereby enabling cancer cells to evade NK cell detection (Fig. 2).79,80 For example, exosomal miR-92b from HCC can downregulate CD69 expression on NK cells, decreasing their cytotoxicity and allowing the tumor to evade immune response.81 Furthermore, exosomal circUHRF1 from HCC cells exacerbates NK cell exhaustion by sponging miR-449c-5p, which raises TIM-3 levels and reduces the secretion of IFN-γ and TNF-α, further contributing to an immunosuppressive microenvironment.80
The human NKG2D receptor is a key activating receptor on NK cells, which exerts cytotoxic effects by recognizing specific stress-induced molecules, including MHC class I chain-related molecules A and B (MICA and MICB, collectively referred to as MIC), as well as six cytomegalovirus UL16-binding proteins. In healthy tissues, these ligands are typically absent or expressed at low levels, but their expression significantly increases during tumor transformation. Tumor cells rapidly induce the expression of these ligands in response to stressors such as genetic damage, metabolic abnormalities, and changes in the microenvironment. Elevated levels of MICA, MICB, and ULBP proteins act as “danger signals” that activate NKG2D receptors, initiating the cytotoxic response of NK cells.82 However, tumor cells can evade immune surveillance by reducing MHC class I molecule expression through proteolytic cleavage or exosome-mediated secretion, thus reducing NK cell recognition.83,84 Studies indicate that NKG2D receptors on NK cells can be bound by soluble NKG2D ligands, such as sMICA and sMICB, contained within exosomes produced by HCC, which further blocks NK cells from recognizing and destroying tumor cells, promoting immune evasion and impairing NK surveillance (Fig. 2).85–88 As exosomes accumulate in the tumor microenvironment, high levels of exosomal MICA/B inhibit NKG2D receptor expression and function, diminishing NKG2D-dependent tumor cell destruction, while decreased NKG2DL expression on tumor cells further reduces NK cell-mediated cytotoxic efficiency.89 Recent studies highlight the dual role of exosome-associated NKG2DLs within the TIME. Initially, exosomes carrying NKG2DLs can activate NK cells after short-term stimulation. However, prolonged exposure to these exosomes leads to a decrease in NKG2D receptor expression and a subsequent impairment in NK cell activity. This effect may result from the continuous release of exosomes within the TIME, which, over time, diminishes the ability of NKG2D to stimulate NK cell activation. Moreover, NK cells marked by exosomal NKG2DLs on their surface are targeted by other NK cells, leading to a perforin- and granzyme-mediated cytotoxic response that promotes self-destructive behavior in NK cells. This mechanism significantly impairs NK cells’ antitumor function and provides a novel explanation for tumor immune evasion.89
Neutrophils
Neutrophils, critical components of the innate immune system and members of the myeloid family, are pivotal in human immune defense due to their abundance and rapid response.90 Similar to their myeloid counterparts, macrophages, tumor-associated neutrophils (TANs) exhibit complex and diverse roles, contributing to both anti-tumor activity and tumor progression.91 The dual functionality of TANs is determined by their phenotype, with N1 TANs exhibiting tumor-suppressing properties and N2 TANs demonstrating tumorigenic effects7. The tumor-suppressing effect of N1 TANs is attributed to two primary mechanisms: direct cytotoxicity via the generation of ROS and the stimulation of various innate and adaptive immune cells.92 Conversely, N2 TANs enhance tumor proliferation by promoting angiogenesis, remodeling the extracellular matrix, increasing metastasis, and inducing immunosuppression, collectively reshaping the tumor microenvironment into a tumor-promoting state.93
TDEs have been observed to alter the TIME by inducing neutrophils to polarize into the N2 phenotype. Research has shown that exosomes originating from tumors increase the number of CD66b+ TANs and facilitate the polarization of TANs into the N2 phenotype, which supports tumor growth across various cancers, including HCC, lung, gastric, colon, and breast cancers. Inhibiting the release of these exosomes has been shown to reduce neutrophil infiltration and limit subsequent tumor cell expansion.94–97
Furthermore, upregulation of the TGF-β signaling pathway has been found to induce neutrophils to adopt the N2 phenotype, which is associated with pro-tumorigenic properties.98 A study by Yang et al. revealed a significant presence of TGF-β in exosomes derived from HCC cells, which can be delivered to recipient cells through membrane fusion, regulating the behavior of these cells (Fig. 2).99 The combined action of TGF-β and Axl induces the secretion of CXCL5, which promotes neutrophil infiltration into HCC tissues and drives the N2 polarization of neutrophils within the TME, thereby accelerating HCC progression.100 Similarly, studies in colorectal cancer have shown that CRC-derived exosomal circPACRGL is delivered to tumor-associated neutrophils, regulating TGF-β expression by sequestering miRNAs (miR-142-3p and miR-506-3p), which facilitates the transition of neutrophils from the N1 to the N2 subtype, thereby aiding tumor progression.96
Autophagy, a critical cellular process that degrades and recycles intracellular components, plays an essential role in maintaining cellular homeostasis and regulating immune responses. In the context of cancer, autophagy has been observed to enhance the survival of neutrophils, promote the production of pro-tumorigenic factors, and influence the tumor microenvironment by regulating various signaling pathways (Fig. 2).101 Research has demonstrated that tumor-derived exosomes can trigger autophagy, altering neutrophil phenotype and function to promote immune evasion and foster a tumor-promoting microenvironment.95 In a study by Li et al., exosomes from gastric cancer cells were found to contain high-mobility group box 1 (HMGB1), which interacts with TLR4 to activate the NF-κB pathway, triggering autophagy and pro-tumor activity in neutrophils. This activation results in the release of pro-inflammatory cytokines and MMP9, collectively enhancing the migration and invasion capabilities of gastric cancer cells.95 In HCC tissues, elevated autophagy in neutrophils has been shown to perpetuate their pro-tumorigenic activities, suggesting that investigating the impact of HCC-derived exosomes on neutrophil autophagy could provide valuable insights into the mechanisms underlying HCC progression.102 A substantial body of clinical evidence indicates that increased neutrophil infiltration is associated with poor prognosis and tumor progression.90 By elucidating the role of exosomes in mediating the behavior of TANs, researchers may be able to develop new cancer therapies that improve patient outcomes.
Adaptive immune cells
T lymphocytes
T lymphocytes comprise approximately 50% of all lymphocytic cells in a healthy liver and are crucial in mounting an immune response to HCC. These cells are primarily categorized into two subtypes: CD8+ T cells and CD4+ T cells. Known as cytotoxic T lymphocytes (CTLs), CD8+ T cells can effectively inhibit tumor growth by locating and directly destroying tumor cells through cell toxicity. CD4+ T cells support the effectiveness of other immune cells, ensuring a well-coordinated and robust defense against cancer cell proliferation.103,104 However, numerous tumor cells, including those found in HCC, adopt complex strategies to evade immune surveillance by utilizing exosomes. These exosomes disrupt T lymphocyte function by transferring signaling molecules, thereby providing a protective barrier for tumor cells, allowing them to evade the immune system.
CD8+T cells. CD8+ T cells are vital in the immune response against malignancy. Despite often becoming dysfunctional in the TIME, the presence of elevated CD8+ T cell levels is associated with higher survival rates in HCC patients.105 Researchers have postulated that the immunosuppressive tumor microenvironment of established solid tumors induced the dysfunction of these CD8+ T cells in the later stages of cancer progression.106 In this process, immune checkpoints serve as a primary means of tumors to evade immune responses by upregulating proteins such as PD-1, LAG-3, CTLA-4, and TIM-3.107 Notably, exosomes from tumor cells significantly impact the anti-tumor capabilities of CD8+ T cells by modulating immune checkpoint protein expression (Fig. 2). Wang et al. demonstrated that overexpressed 14-3-3ζ can be transferred from HCC cells to tumor-infiltrating lymphocytes via exosomes, which suppresses T cells’ anti-tumor activity and accelerates HCC progression. This suppression is primarily due to the elevated presence of PD-1 and TIM-3 on CD8+ T cells, ultimately leading to CD8+ T cell exhaustion.108 However, the specific mechanisms through which 14-3-3ζ influences the functionality of tumor-infiltrating lymphocytes are yet to be fully determined. Exosomes carrying epigenetic modifiers may also influence CD8+ T cell function, resulting in impaired immune responses. For instance, exosomal circCCAR1 from HCC can transfer to CD8+ T cells, where it stabilizes PD-1 expression and reduces the secretion of pro-inflammatory cytokines IFN-γ and TNF-α, leading to CD8+ T cell dysfunction and reduced efficacy of PD-1 therapy in HCC. In this process, exosomal circCCAR1 sequesters miR-127-5p and increases the levels of its target, WTAP. The m6A modification mediated by WTAP further stabilizes circCCAR1, creating a regulatory feedback cycle involving the circCCAR1/miR-127-5p/WTAP axis. Additionally, circCCAR1 interacts with the PD-1 protein, preventing its degradation by reducing ubiquitination, thereby further stabilizing PD-1.109 Zhang et al. discovered that elevated levels of circZMIZ1 in the serum of HCC patients promote KCNJ2 expression by sequestering miR-15a-5p, enhancing apoptosis in CD8+ T cells while simultaneously limiting their cytotoxic potential.110 Conversely, Zhang et al. identified that miR-15a-5p, which shows reduced expression in HCC tissues and cell lines, is encapsulated within exosomes and directly targets CD8+ T cells to inhibit PD-1 expression. Further in vitro experiments revealed that overexpressing miR-15a-5p decreased the aggressiveness of HepG2 cells, suggesting a novel target for HCC prevention and treatment.111
Moreover, previous research has demonstrated that tumor cells can upregulate PD-Ls, increase the incorporation of PD-L1 into exosomes, and transfer PD-L1 from PD-L1-positive to PD-L1-negative cells. This inhibits CD8+ T cell proliferation and function through the PD-1/PD-L1 interaction, potentially leading to broader immune escape (Fig. 2).112,113 Fan et al. discovered upregulated levels of PCED1B-AS1 in exosomes, which are secreted and absorbed by T cells or other HCC cells. PCED1B-AS1 acts as a microRNA sponge to inhibit the expression of miR-194-5p, leading to increased levels of PD-L1 and PD-L2 in recipient cells and thus suppressing both recipient T cell and macrophage activity.114 Wei et al. uncovered a sophisticated interplay between “RNA and RNA” in HCC, where HMGB1 enhances RICTOR mRNA expression by sequestering miR-429. This interaction promotes high PD-L1 expression and its incorporation into exosomes via the AKT-mTORC1-P70S6K pathway, which then binds to PD-1 receptors on T cells, impairing CD8+ T cell cytotoxicity and facilitating tumor immune evasion.115,116
Beyond the well-known inhibitory ligand PD-L1, cell surface proteins like ICAM-1 may be regulated or carried by tumor-derived exosomes, influencing the function of CD8+ T cells (Fig. 2). ICAM-1, an adhesion molecule produced on immune and endothelial cells, promotes T cell adhesion and motility by binding to lymphocyte function-associated antigen 1 (hereinafter referred to as LFA-1) on T cells.117 Zhang and colleagues discovered that ICAM-1 is essential for the PD-L1 exosome-mediated suppression of CD8+ T cells. In their research, IFN-γ and other inflammatory molecules greatly increased the co-expression of ICAM-1 and PD-L1 on tumor-derived exosomes. The interaction between exosomal PD-L1 and PD-1 on CD8+ T cells was significantly reduced in the absence of ICAM-1, suggesting that ICAM-1-LFA-1-mediated adhesion is necessary for exosomal PD-L1 to drive CD8+ T cell dysfunction.118 Furthermore, elevated exosome levels enriched with miR-222 lower ICAM-1 expression on tumor cell surfaces in patients with chronic hepatitis B and HCC,119,120 reducing the tumor cells’ susceptibility to CTL destruction.121 Blocking key adhesion molecules between tumor-derived exosomes and T cells may improve the efficacy of immune checkpoint therapies and patient outcomes. Further study of these mechanisms will deepen our understanding of HCC-derived exosomes in T cell regulation and their implications for immunotherapy.
CD4+T cells. There are numerous subtypes of CD4+ T cells, commonly known as “helper T cells,” which are categorized based on the cytokines and other secretions they produce during immune responses. Among these, Tregs are considered a crucial subset. Their pivotal role in regulating immune responses is essential, as they promote immune tolerance and maintain immune homeostasis within the body. In the TIME, Tregs exhibit strong immunosuppressive effects by inhibiting the activity of immune effector cells through multiple pathways, thus supporting immune suppression and facilitating cancer progression.122
Notably, tumor-secreted exosomes can stimulate the differentiation of immunosuppressive CD4+ T cell subtypes, particularly Tregs, thereby fostering an immunosuppressive tumor microenvironment (Fig. 2).123 Research has shown that overexpression of 14-3-3ζ inhibits the proliferation and function of CD3+ T cells in peripheral blood. Furthermore, it prompts naive T cells to diverge from becoming effector T cells, instead developing into regulatory T cells.108 Huang et al. made a significant discovery by identifying the transport of HCC-derived exosomal circGSE1 to CD4+ T cells, which promotes Treg proliferation and facilitates immune evasion in HCC by regulating the miR-324-5p/TGFβ receptor 1/SMAD family member 3 axis. In this mechanism, it functions as an absorbing molecule for miR-324-5p, activating the TGFβ receptor 1 and SMAD family member 3 signaling pathways, which, in turn, promote the proliferation of Tregs.124
Another distinct subgroup of CD4+ T helper cells is the Th17 cells, characterized by their ability to secrete IL-17. Under certain conditions, IL-17 released by Th17 cells can enhance the immune response against tumors.125 The imbalance between Th17 and Treg cells induced by TEXs represents another mechanism of immune evasion in HCC (Fig. 2). Guo et al. discovered that exosomal heat-shock protein 70 from heat-stressed tumor cells can convert immunosuppressive Tregs into Th17 cells through IL-6 and IL-17 secretion, thereby enhancing their anti-tumor efficacy.126 In contrast, Tian and colleagues demonstrated that under the acidic conditions of the HCC microenvironment, HIF-1α and HIF-2α are activated and bind to the promoter regions of miR-21 and miR-10b, significantly increasing the expression of these exosomal miRNAs. This elevation in exosomal miRNAs promotes the proliferation, migration, and invasion of HCC cells in both controlled laboratory and real-life settings.127 This effect likely occurs by suppressing the anti-cancer immune activity of Th17 cells and promoting the differentiation of immunosuppressive Treg cells, disrupting the balance of CD4+ T cell subtypes and establishing an immune-suppressive environment conducive to tumor progression.128,129
Moreover, PD-L1 overexpression in exosomes derived from HCC also impairs the functionality of CD4+ T cells (Fig. 2). Gong and colleagues revealed that norcholic acid significantly increases PD-L1 levels on the surface of HCC cells and their secreted exosomes by downregulating the farnesoid X receptor and the small heterodimer partner. This process upregulates the immune checkpoint proteins PD-1 and TIM3 on CD4+ T cells, thereby impairing CD4+ T cell function, facilitating tumor immune escape, and promoting HCC progression.27
B lymphocytes
B cells are integral to the adaptive immune system, playing multiple roles in cancer response through various mechanisms, including direct tumor cell killing via granzyme B release, antigen presentation, and cytokine release, such as IFN-γ and IL-12, to activate CTLs.130 Despite their relatively low abundance in the liver, B cells have a vital function in monitoring and responding to tumors, as well as in developing immunological memory.131 Nevertheless, B lymphocytes also have a significant function in encouraging tumor growth. Regulatory B (Breg) cells, a subtype of B cells, are closely associated with the progression of hepatocellular carcinoma.132 They can promote immunosuppression by upregulating cytokines such as IL-10 and PD-L1, which inhibit anti-tumor immune effector cells.132–134
Studies have demonstrated that exosomes from HCC can influence the expression and phenotype of Breg cells, thereby contributing to a tumor-supportive microenvironment that protects cancer cells from immune attack (Fig. 2). Ye et al. identified a novel regulatory B cell subset, termed TIM-1+ Breg cells, which is characterized by a distinct phenotype of CD5high, CD24−, CD27−/+, and CD38+/high, setting them apart from traditional peripheral regulatory B cells. Notably, these TIM-1+ Breg cells exhibit significantly higher expression within tumor tissues compared to surrounding tissues. Their study further demonstrated that HCC-derived exosomal HMGB1 promotes the proliferation of TIM-1+ Breg cells via the HMGB1-TLR2/4-MAPK signaling pathway. These TIM-1+ Breg cells secrete large amounts of IL-10 and TGF-β1, which inhibit the expansion and effector functions of CD8+ T cells and downregulate cytokines TNF-α and IFN-γ, thereby fostering an immunosuppressive microenvironment in HCC and promoting tumor progression.135 Additionally, Xiao et al. identified a PD-1hi B-cell phenotype in advanced HCC, marked by elevated PD-1 expression and a unique CD5hiCD24−/+CD27hi/+CD38dim signature. These cells, upon interaction with PD-L1, secrete substantial amounts of IL-10, impairing CD8+ T cell growth and activity.136 Although there is no direct evidence showing that exosomes derived from HCC can induce the PD-1hi B-cell phenotype, studies on esophageal squamous cell carcinoma suggest that exosomes from esophageal squamous cell carcinoma could facilitate the induction of PD-1hi Bregs.137 Therefore, it is highly plausible that HCC cells might also induce the PD-1hi B-cell phenotype through exosomal pathways, a hypothesis that warrants further investigation.
Other cells
Cancer-associated fibroblasts (CAFs)
CAFs are fundamental elements of the TIME and are extensively distributed within the stroma. They modulate tumor proliferation, angiogenesis, metastasis, and resistance to therapies.5,138 These functions are achieved through interactions with tumor-infiltrating immune cells and the release of signaling molecules such as cytokines, growth factors, and exosomes, which create an immunosuppressive environment that promotes immune escape and supports tumor growth.139
In HCC, CAFs contribute to immune suppression and tumor progression by producing immunosuppressive molecules such as IDO and IL-6, which impair NK cells,140 induce regulatory DCs,141 and promote pro-tumor immune cell polarization.139 Moreover, CAFs have been shown to upregulate the expression of negative immune regulators, including PD-L1 and PD-L2, leading to immune cell exhaustion and inactivation142,143 These actions create an environment conducive to cancer progression and resistance to immunotherapy. In addition to modulating immune responses, CAFs produce substantial amounts of extracellular matrix, which not only supports tumor structure and growth but also acts as a physical barrier, impeding immune cell migration and reducing the efficacy of therapeutic drugs.144–146
Research has demonstrated that TDEs play a crucial role in transforming normal fibroblasts into CAFs, as observed in various cancers, including head and neck squamous cell carcinoma and HCC (Fig. 2).147 Studies on head and neck squamous cell carcinoma have shown that TDEs can induce the transformation of normal fibroblasts into CAFs, creating an inflammatory TIME where T-cell proliferation is suppressed, and TAMs shift to an M2 phenotype, thus aiding tumor progression.148 Fang et al. discovered that highly metastatic HCC cells release miR-1247-3p within exosomes to normal fibroblasts, downregulating B4GALT3 levels and transforming them into CAFs through the β1 integrin/NF-κB signaling pathway. The activated CAFs then secrete cytokines such as IL-6 and IL-8, enhancing tumor cell proliferation, stem cell properties, EMT, and angiogenesis, thereby accelerating cancer progression and metastasis. Targeting this pathway or exosomal miR-1247-3p could provide novel strategies for preventing and treating HCC lung metastasis.149 Nonetheless, research on how exosomes derived from HCC activate CAFs remains sparse. Further exploration of this interaction may offer insights for developing novel therapeutic strategies against HCC.
Hepatic stellate cells
Hepatic stellate cells (HSCs), located within the Disse space of the liver, are crucial non-parenchymal cells known for their remarkable plasticity, enabling them to regulate various pathological processes.150 During chronic liver injury, HSCs transition from a quiescent state into an activated myofibroblast phenotype with enhanced proliferative and migratory capabilities, accompanied by excessive synthesis of the extracellular matrix.150 This transformation is a key driver of liver fibrosis, and if uncontrolled, chronic damage can progress to liver cirrhosis and potentially HCC.151 Within the TIME, activated HSCs also recruit immune-suppressing cells, including myeloid-derived suppressor cells and Tregs, which facilitate immune evasion and advance HCC progression.152
Extensive research has shown that exosomes significantly contribute to the development of precancerous liver conditions such as viral hepatitis, metabolism-related fatty liver disease, and liver fibrosis by activating HSCs (Fig. 2). Metabolic-associated fatty liver disease is increasingly recognized as a major risk factor for the development of liver fibrosis and HCC.153 X et al. demonstrated that upregulation of exosomal miR-1297 derived from hepatocytes under lipotoxic stress activates and proliferates HSCs by inhibiting the tumor suppressor gene PTEN, which subsequently activates the PI3K/AKT signaling pathway. Activation of this pathway is a critical driver of fibrogenesis and significantly exacerbates metabolic-associated fatty liver disease, facilitating the progression of the disease toward more severe forms of liver pathology, such as advanced fibrosis, cirrhosis, and HCC.154 In a similar study, Xu and colleagues showed that elevated expression of lncRNA CYTOR in exosomes derived from damaged liver cells can activate HSCs, leading to liver fibrosis, as evidenced by significantly increased levels of fibrosis-related markers such as α-SMA, type I collagen, and TGF-β. Mechanistically, lncRNA CYTOR functions as a competing endogenous RNA that binds to miR-125, relieving suppression on glial cell line-derived neurotrophic factor, allowing cell line-derived neurotrophic factor to activate LX2 cells—an effective mechanism for HSC activation.155 Another study revealed that exosomes derived from hepatocytes, following stimulation with carbon tetrachloride to induce liver injury, contain highly expressed H2AFJ, which promotes the migration and invasion of HSCs and exacerbates liver fibrosis through the activation of the MAPK/STMN1 signaling pathway.156 Moreover, acute viral hepatitis can lead to liver fibrosis, as Zhang et al. observed that miR-222 is significantly increased in exosomes from HBV-infected hepatocytes, promoting LX-2 cell activation by inhibiting TFRC and TFRC-induced ferroptosis, thus activating HSCs and promoting liver fibrosis.157 Based on the aforementioned research, exosomes play a crucial role in the initiation and progression of precancerous liver conditions by influencing HSC activation. This offers new therapeutic options for patients at risk of liver fibrosis and potential targets for the prevention of HCC.
Additionally, TDEs facilitate intercellular communication between tumor cells and HSCs, influencing the activation of HSCs and driving HCC progression (Fig. 2). Xia et al. discovered that when HCC cells are co-cultured with HSCs, the expression level of Smoothened (SMO) in quiescent HSCs is initially low but significantly increases in activated HSCs. Their research indicated that exosomes convey SMO from HCC cells to HSCs, leading to HSC activation, which enhances proliferation, EMT, and stemness, promoting HCC development. Mechanistically, SMO activates the Hedgehog signaling pathway, which enhances the transcriptional activity of Gli1 on MIRLET7BHG in activated HSCs. MIRLET7BHG then sequesters miR330-5p, boosting SMO levels and further activating HSCs and tumor growth in HCC.158 In addition, under the influence of tumor-derived exosomes, HSCs may serve as another source of CAFs (Fig. 2).159 Zhou and colleagues identified that the release of exosomal miRNA-21 from HCC cells can trigger the PDK1/AKT signaling pathway by reducing PTEN, transforming HSCs into CAFs. These activated CAFs then release angiogenic factors such as TGF-β, MMP2, bFGF, MMP9, and vascular endothelial growth factor, thereby promoting cancer progression.160 Understanding exosome-HSC interactions within the immune microenvironment is essential for comprehending the pathogenesis of precancerous liver diseases and HCC. Future research should focus on the mechanisms by which exosomes stimulate HSCs and drive liver disease development, with the goal of developing exosome-based treatments to prevent or reverse liver fibrosis and HCC. We have detailed the functions and pathways mediated by HCC-derived exosomes on immune cells and other cells within the HCC immune microenvironment (Table 1 and Fig. 2).27,29,40,43,45–48,50,52–54,57,60,64,70,80,81,89,95,96,99,100,108–111,114–116,118,124,127,135,137,149,154–158
Table 1The function and pathways regulated by HCC-derived exosomes in the HCC immune microenvironment
| Exosome component | Source cell | Recipient cell | Function | Pathway/Mechanism | Ref |
|---|
| miRNA | | | | | |
| miR-142-3p ↑ | HCC | Macrophage | Induce ferroptosis in HBV-infected M1-type macrophage | miR-142-3p/SLC3A2 axis | 40 |
| miR-452-5p ↑ | HCC | Macrophage | Induce M2 polarization of macrophage | miR-452-5p/TIMP3 axis | 43 |
| miR-146a-5p ↑ | HCC | Macrophage | Inhibit M2 polarization of macrophage | SALL4/miR-146a-5p axis | 52 |
| miR-21-5p ↑ | HCC | Macrophage | Induce M2 polarization of macrophage | miR-21-5p/RhoB axis | 45 |
| miR-1-3p ↓ | HCC | Macrophage | Inhibit M2 polarization of macrophage | \ | 50 |
| miR-23a-3p ↑ | HCC | Macrophage | Increase PD-L1 expression, promote M2-type polarization and induce CD8+ T cell apoptosis | miR-23a-3p/PTEN-PI3K-AKT axis | 57 |
| miR-146a-5p ↑ | HCC | Macrophage | Induce T cell exhaustion | SALL4/miR-146a-5p axis | 52 |
| miR-92b ↑ | HCC | NK cell | Impair the cytotoxicity of NK cells | miR-92b/CD69 axis | 81 |
| miR-15a-5p ↓ | HCC | CD8+ T cells | Inhibits PD-1 expression in CD8+ T cells | \ | 111 |
| miR-21, miR-10b ↑ | HCC | CD4+ T cells | Disrupt the balance of CD4+ T cell subtypes | \ | 127 |
| \ | Tumor cells | B cells | Induce PD-1hi B-cell | \ | 137 |
| miR-1247-3p ↑ | HCC | fibroblasts | Activate CAFs and stimulate the release of IL-6 and IL-8 | B4GALT3/β1-integrin/NF-κB axis | 149 |
| miR-1297 ↑ | Hepatocytes (lipotoxic) | HSCs | Promote the activation and proliferation of HSCs, accelerate the progression of MAFLD | PTEN/PI3K/AKT signaling pathway | 154 |
| miR-222 ↑ | Hepatocytes (infected with HBV) | HSCs | Promote HSC activation and exacerbate liver fibrosis | miR-222/TFRC axis | 157 |
| lncRNA | | | | | |
| HMMR-AS1 ↑ | HCC | Macrophage | Induce M2 polarization of macrophage | miR-147a/ARID3A axis | 46 |
| lncRNA (HEIH) ↑ | HCC | Macrophage | Induce M2 polarization of macrophage | HEIH/miR-98-5p/STAT3 axis | 47 |
| PCED1B-AS1 ↑ | HCC | HCC | Suppress recipient T cell and macrophage activity | Increase PD-Ls expression on T cells and HCC cells | 114 |
| lncRNA CYTOR ↑ | Hepatocytes (injured) | HSCs | Promote HSC activation and exacerbate liver fibrosis | CYTOR/miR-125/GDNF axis, TGF-β/Smad signaling pathway | 155 |
| circRNA | | | | | |
| hsa_circ_0074854 ↑ | HCC | Macrophage | Promote M2 polarization of macrophage and EMT in HCC cells | \ | 48 |
| circTMEM181 ↑ | HCC | Macrophage | Suppress the function of CD8+ T cell and enhance resistance to anti-PD-1 therapy | miR-488-3p/CD39/CD73/eATP-adenosine pathway | 60 |
| circUHRF1 ↑ | HCC | NK cell | Induce NK cell exhaustion and decreases IFN-γ and TNF-α production | miR-449c-5p/TIM-3 axis | 80 |
| circPACRGL ↑ | Tumor cells | Neutrophils | Promote the polarization of N2 Neutrophils | miR-142-3p, miR-506-3p/TGF-β1 axis | 96 |
| circCCAR1 ↑ | HCC | CD8+ T cells | Suppress the function of CD8+ T cell and enhance resistance to anti-PD-1 therapy | circCCAR1/miR-127-5p/WTAP axis | 109 |
| circZMIZ1 ↑ | HCC | CD8+ T cells | Induce apoptosis and decrease cytotoxicity in CD8+ T cells | miR-15a-5p/KCNJ2 axis | 110 |
| circGSE1 ↑ | HCC | CD4+ T cells | Promote the expansion of Tregs | miR-324-5p/TGFBR1/Smad3 axis | 124 |
| DNA | | | | | |
| H2AFJ ↑ | Hepatocytes (injured) | HSCs | Promote the migration and invasion of HSCs and exacerbate liver fibrosis | H2AFJ/MAPK/STMN1 axis | 156 |
| Protein | | | | | |
| LOXL4 ↑ | HCC | Macrophage | Enhance the expression of PD-L1 and inhibit the cytotoxicity of CD8+ T cells | IFN-STAT1 (STAT3)/PD-L1 axis | 53,54 |
| GOLM1 ↑ | HCC | Macrophage | Enhance the expression of PD-L1 and suppress the function of CD8+ T cells | CSN5-mediated deubiquitination | 29 |
| AFP, GPC3, HSP70 ↑ | HCC | DCs | Promote DC maturation and activation | \ | 64 |
| NKG2D ligands ↑ | HCC | NK cell | Regulate the functionality and induce self-destructive behaviors of NK cells | \ | 89 |
| TGF-β ↑ | HCC | Neutrophils | Promote the phenotypic transition of neutrophils to N2 | TGF-β/Axl/CXCL5 axis | 99,100 |
| HMGB1 ↑ | Tumor cells | Neutrophils | Induce autophagy and N2-like phenotype in neutrophils | HMGB1/TLR4/NF-κB pathway | 95 |
| 14-3-3ζ ↑ | HCC | T cells | Induce CD8+ T cell exhaustion and drive Treg cell differentiation | \ | 108 |
| PD-L1 ↑ | HCC | CD8+ T cells | Suppress the cytotoxicity of CD8+ T cells | Activate the HMGB1/RICTOR axis and AKT-mTORC1-P70S6K pathway | 115,116 |
| ICAM-1 ↑ | HCC | CD8+ T cells | Prerequisite for PD-L1-induced immune dysfunction in CD8+ T cells | \ | 118 |
| PD-L1 ↑ | HCC | CD4+ T cells | Impair the function of CD4+ T cells and promote the immune escape of HCC | NorCA/FXR/SHP/PD-L1 axis | 27 |
| HMGB1 ↑ | HCC | B cells | Induce TIM-1+ Breg cells | HMGB1-TLR2/4-MAPK pathway | 135 |
| SMO ↑ | HCC | HSCs | Promote activation proliferation, migration, invasion, and EMT in HSCs | MIRLET7BHG/SMO/Hedgehog signaling pathway | 158 |
| Lipids | | | | | |
| FAO, LD ↑ | Tumor cells | DCs | Cause DC immune dysfunction | FAs/PPARα/FAO axis | 70 |