Primary liver cancer is the sixth-most commonly occurring cancer and third leading cause of cancer death worldwide, of which hepatocellular carcinoma (HCC) accounts for 75–85% of the total liver cancer burden [1]. Due to the lack of distinguishing characteristics and effective screening methods in the early stage, the majority of HCC cases are diagnosed at the advanced stage, which leads to the overall survival of most HCC patients being less than 5 years ^[2][3][2,3]. Despite numerous measures in the field of HCC treatment, effective treatment options for advanced HCC are still limited [4]. With the deepening understanding of the tumor microenvironment (TME) in HCC, immunotherapy—especially immune checkpoint inhibitors (ICIs)—has brought new hope for improving the prognosis of advanced HCC. ICIs mainly include monoclonal antibodies against cytotoxic-T-lymphocyte-associated protein 4 (CTLA-4), programmed cell death protein 1 (PD-1), and its ligand-programmed cell death ligand 1 (PD-L1), which block these immune checkpoints, thereby promoting the T cell-mediated antitumoral immune responses ^[5][6][7][8][5,6,7,8]. Other immunotherapies, including cancer vaccines and adoptive cell therapy, have also shown great potential for clinical application to treat advanced HCC ^[9][10][9,10]. Nevertheless, numerous HCC patients do not benefit from immunotherapy, which has seriously restricted its application ^[11][12][11,12].

At present, the relationship between immunotherapy and exosomes is a new hotspot in HCC research that may improve theour understanding of the resistance mechanism of immunotherapy. Exosomes are small, single-membrane secreted organelles ranging in diameter from 30 to 150 nm [13]. They are formed by the invagination of the cell membrane of early sorting endosomes, which eventually grow into multivesicular bodies ^[14][15][14,15]. As a type of extracellular vesicle (EV), exosomes contain a variety of nucleic acids, proteins, lipids, and metabolites which can reflect, at least in part, those of their parental cells [16]. By transporting these components, exosomes participate in numerous biological processes, including antigen presentation, apoptosis, inflammation, and intercellular signaling, etc. [17]. Exosomes, particularly the proteins and noncoding RNA in them, play a crucial role in tumor growth, metastasis, angiogenesis, and invasion [18] and are closely linked with tumor immunotherapy [19]. Recent studies have demonstrated that tumor-cell-derived exosomes (TEXs) are also involved in TME remodeling, tumor progression, and drug resistance, which brings challenges for the effective treatment of HCC ^[20][21][20,21]. These complex biological roles enable exosomes to be applied for early diagnosis, treatment, and prognostic prediction of HCC ^[22][23][24][22,23,24], and research on exosomes is expected to help clinicians to identify suitable HCC patients for immunotherapy and improve the efficacy of immunotherapy.

2. Influence of Exosomes on Tumor Microenvironment of HCC

The HCC TME is an elaborate immunosuppressive microenvironment composited of different types of cells, growth factors, proteolytic enzymes, extracellular matrix proteins, and cytokines [25]. In HCC TME, not only immune cells but also non-immune cells are involved in tumorigenic processes, including tumor proliferation, invasion, and metastasis ^[26][27][26,27]. Recent studies have highlighted the critical role of exosomes for cell-to-cell communication in the HCC TME, which are recognized as a key factor in tumor progress and may be closely associated with resistance to immunotherapy in HCC patients ^[28][29][28,29].

2.1. Exosomes on Cell-to-Cell Communication in HCC TME

The influence of exosomes on cell-to-cell communication in HCC TME is summarized in Figure 1. Tumor-cell-derived exosomes can be directly transported into CD8+ T cells, tumor-infiltrating T cells, or regulatory T cells to inhibit their anti-tumor function or modulate their cellular behaviors ^{[30][31][32][33]}[30,31,32,33]. Furthermore, HCC-derived exosomes can induce nature killer (NK) cell activation, attenuate NK cell cytotoxicity, and arrest them to secret cytokines ^[34][35][36][34,35,36]. HCC-derived exosomes can effectively activate dendritic cells (DCs) due to their ability to carry HCC antigens [37], while DC-derived exosomes (DEXs) can potentiate anti-tumor immune responses against HCC by contributing to the activation of CD8+ T cells and reshaping the TME ^[38][39][40][38,39,40]. Depending on the substances they carry, HCC-derived exosomes show apparently contradictory roles on macrophages. They can directly influence the expression of macrophage cell surface receptors or membrane-associated signaling molecules ^{[32][41][42][43][44]}[32,41,42,43,44] and can affect the differentiation of macrophages into two different directions: M1 (anti-tumorigenic) [45] or M2 (pro-tumorigenic) macrophages ^{[46][47][48][49]}[46,47,48,49]. Furthermore, macrophage-derived exosomes can also affect HCC development or other immune cell functions ^{[50][51][52][53][54][55][56][57]}[50,51,52,53,54,55,56,57]. High-mobility group box1 (HMGB1) is an evolutionarily conserved DNA-binding nuclear protein which has been demonstrated to be expressed on tumor-derived exosomal membranes. SRecent study has revealed that HCC-derived exosomal HGMB1 can foster HCC immune evasion by promoting TIM-1+ regulatory B cell expansion [58]. Exosomes derived from myeloid-derived suppressor cells, a type of special cell in tumor immunity, play important immunoregulatory roles in cancers, and tumor-derived exosomes also regulate the function of myeloid-derived suppressor cells ^[59][60][61][59,60,61]. Unfortunately, no relevant studies have been undertaken in HCC. There have also been numerous studies on the relationship between exosomes and non-immune cells in the HCC TME. HCC is a typical hypervascular tumor, HCC-derived exosomes can promote angiogenesis by targeting vascular endothelial cells ^[62][63][62,63], and exosomal CXCR4 has also been proved to promote HCC lymphangiogenesis [64]. Cancer-associated fibroblast-derived exosomes can affect the development, invasion, drug-resistance, and metabolism of HCC ^[65][66][67][65,66,67], and adipocyte-derived exosomes also have similar tumor-promoting effects ^[68][69][68,69]. Furthermore, HCC-derived exosomes can be transported to normal hepatocytes or other tumor cells to enhance motile ability ^[70][71][70,71], stimulate epithelial–mesenchymal transition of peripheral cells, and further promote HCC invasion ^[72][73][72,73]. In contrast, mesenchymal-stem-cell-derived exosomes can inhibit the malignant behavior of HCC to some degree by transporting some microRNAs, making them potentially useful for HCC treatment ^[74][75][74,75].

2.2. Exosomes on Immunotherapy Resistance of HCC

There are several vital exosome-associated mechanisms that may partly explain why many HCC patients show resistance to ICIs (Figure 2). ICIs, mainly against CTLA-4, PD-1, and PD-L1, not only invigorate T cells but also activate other cells involved in innate and adaptive immune response, all of which function together to inhibit tumors ^[76][77][76,77]. CD8+ T cells have great potential in ICI therapy and adoptive T cell therapy, but they are often dysfunctional in tumors ^[78][79][78,79]. In the HCC TME, exosomes can affect the function of CD8+ T cells through multiple pathways, which may lead to immunotherapy resistance. PD-1 is a vital co-inhibitory receptor which is primarily expressed on the surface of antigen-stimulated T cells. However, PD-L1 is upregulated on tumor cells and antigen-presenting cells in the TME and is bound to PD-1 on activated T cells and dampens anti-tumor immunity by counteracting T-cell-activating signals ^[80][81][82][80,81,82]. Exosomes may increase the expression of PD-1 or PD-L1, which augments the effective dose of ICIs. TEXs can directly impact the PD-1/PD-L1 expression of immune cells in the TME by carrying PD-L1 or indirectly impact this by regulating the PD-1/PD-L1 axis in adjacent immune cells, thereby affecting the efficacy of ICIs ^[83][84][83,84]. Spleen deficiency (SD) can also upregulate the expression of exosomal CTLA-4 and PD-1 [85]. ICIs can block PD-L1, but exosomes can convey cargos stably, which may help exosomal PD-L1 to avoid being blocked by ICIs [86]. Exosomal PD-L1 has the same membrane topology as PD-L1 on the surface of immune cell, which can directly suppress the function of T cells by binding to PD-1, thereby inducing HCC patients’ resistance to ICIs [83]. Exosomes containing PD-L1 are a major means by which HCC cells can have a dramatic influence on T cell function. Recently, Chen et al. demonstrated that Golgi membrane protein 1 in HCC can facilitate PD-L1 trafficking to exosomes by suppressing Rab27b in the trans-Golgi network (TGN) area [41]. Meanwhile, norcholic acid can elevate the expression of exosomal PD-L1 through inhibiting farnesoid X receptor signaling [87]. Furthermore, the HGMB1-driven RNA–RNA crosstalk network facilitated HCC cell glutamine metabolism, which could compromise the efficacy of immunotherapy through mTORC1-P70S6K-dependent PD-L1 production and PD-L1+ exosome activity [88]. These PD-L1-carrying exosomes can be taken up by macrophages and upregulate their PD-L1 expression, further inhibiting the anti-tumor function of CD8+ T cells [41]. Additionally, it is worth noting that in a study of melanoma treatment with pembrolizumab, an increase in the level of exosomal PD-L1 during the treatment may reinvigorate the IFN-γ to produce CD8+ T cells. However, the high pre-treatment levels of exosomal PD-L1 were associated with T cell exhaustion, meaning that it could not be reinvigorated by PD-1 inhibitors [89]. This is also a common way for HCC-derived exosomes to indirectly influence T cell function through surrounding cells. Endoplasmic reticulum stress contributes HCC cells releasing miR-23a-3p-enriched exosomes, which inhibit PTEN and in turn activate the AKT pathway, resulting in high PD-L1 expression in macrophages [42]. HCC-derived LOXL4, which is secreted through exosomes and primarily localized within macrophages, can strengthen PD-L1 expression in macrophages relying on interferon-mediated signal transducer and activator of transcription-dependent PD-L1 activation [43], thereby dampening the function of CD8+ T cells. Furthermore, HCC-derived exosomal circTMET181 sponges miR-488-3p and upregulates CD39 expression in macrophages. Macrophage-derived CD39 and HCC cell-derived CD73 synergistically activate the ATP–adenosine pathway, which in turn degrades extracellular ATP into adenosine. This process generates a large volume of adenosine, which impairs CD8+ T cell function and leads to anti-PD1 resistance [44]. In addition, the up-regulation of 14-3-3ζ protein can restrain the anti-tumor immunity of tumor-infiltrating T cells in TME, which may be due to CD8+ T cell exhaustion [30], and HCC cells can also release exosomal PCED1B-AS1 to inhibit recipient T cells [32]. Exosomes inhibit the anti-tumor function of T cells through the various above-mentioned mechanisms, thereby affecting the efficacy of ICIs. In addition, B cells are central to humoral immunity. HCC-derived exosomes promote TIM-1+ regulatory B cell expansion through exosomal HMGB1-TLR2/4-MAPK pathways and further lead to the overproduction of the immunosuppressive cytokine IL-10, which results in a marked inhibition of CD8+ T cell activity [58]. The effect of exosomes on NK cells, macrophages and other immune cells may also cause immunotherapy resistance, but these mechanisms require further in-depth research. In addition, the research on exosomes in other immunotherapies for HCC (such as tumor vaccines or adoptive cell therapies) is also limited. More related research will help researcherus to better understand the role of exosomes in immunotherapy resistance.